AuxXxilium/linux_dsm_epyc7002
1
0
Fork 0
mirror of https://github.com/AuxXxilium/linux_dsm_epyc7002.git synced 2024-12-28 11:18:45 +07:00
Code Issues Actions Packages Projects Releases Wiki Activity
95e196337a
linux_dsm_epyc7002/drivers/net/ethernet/mellanox/mlx4/main.c

3982 lines
109 KiB
C
Raw Normal View History

IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
/*
* Copyright (c) 2004, 2005 Topspin Communications. All rights reserved.
* Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
mlx4: Update/add Mellanox Technologies copyright lines to mlx4 driver files Update existing Mellanox copyright lines to 2008, and add such lines to files where they are missing. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-07-26 00:32:52 +07:00
* Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. All rights reserved.
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
* Copyright (c) 2006, 2007 Cisco Systems, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#include <linux/module.h>
#include <linux/init.h>
#include <linux/errno.h>
#include <linux/pci.h>
#include <linux/dma-mapping.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 15:04:11 +07:00
#include <linux/slab.h>
mlx4: Add blue flame support for kernel consumers Using blue flame can improve latency by allowing the HW to more efficiently access the WQE. This patch presents two functions that are used to allocate or release HW resources for using blue flame; the caller need to supply a struct mlx4_bf object when allocating resources. Consumers that make use of this API should post doorbells to the UAR object pointed by the initialized struct mlx4_bf; Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:38:41 +07:00
#include <linux/io-mapping.h>
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
#include <linux/delay.h>
net/mlx4_core: Load higher level modules according to ports type Mellanox ConnectX architecture is: mlx4_core is the lower level PCI driver which register on the PCI id, and protocol specific drivers are depended on it: mlx4_en - for Ethernet and mlx4_ib for Infiniband. NIC could have multiple ports which can change their type dynamically. We use the request_module() call to load the relevant protocol driver when needed: on loading time or at port type change event. Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-15 21:55:24 +07:00
#include <linux/kmod.h>
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
#include <linux/mlx4/device.h>
#include <linux/mlx4/doorbell.h>
#include "mlx4.h"
#include "fw.h"
#include "icm.h"
MODULE_AUTHOR("Roland Dreier");
MODULE_DESCRIPTION("Mellanox ConnectX HCA low-level driver");
MODULE_LICENSE("Dual BSD/GPL");
MODULE_VERSION(DRV_VERSION);
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
struct workqueue_struct *mlx4_wq;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
#ifdef CONFIG_MLX4_DEBUG
int mlx4_debug_level = 0;
module_param_named(debug_level, mlx4_debug_level, int, 0644);
MODULE_PARM_DESC(debug_level, "Enable debug tracing if > 0");
#endif /* CONFIG_MLX4_DEBUG */
#ifdef CONFIG_PCI_MSI
mlx4_core: Enable MSI-X by default Recover from MSI-X errors by automatically falling back on regular interrupt, instead of asking the user to do this manually. This makes it possible to enable MSI-X by default, and will make it possible to get rid of the msi_x module option in the future. Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-07 20:08:28 +07:00
static int msi_x = 1;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
module_param(msi_x, int, 0444);
MODULE_PARM_DESC(msi_x, "attempt to use MSI-X if nonzero");
#else /* CONFIG_PCI_MSI */
#define msi_x (0)
#endif /* CONFIG_PCI_MSI */
net/mlx4: Adapt num_vfs/probed_vf params for single port VF A new syntax is added for the module parameters num_vfs and probe_vf. num_vfs=p1,p2,p1+p2 probe_bf=p1,p2,p1+p2 Where p1(2) is the number of VFs on / probed VFs for physical port1(2) and p1+p2 is the number of dual port VFs. Single port VFs are currently supported only when the link type for both ports of the device is Ethernet. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-19 23:11:53 +07:00
static uint8_t num_vfs[3] = {0, 0, 0};
net/mlx4_core: Allow not to specify probe_vf in SRIOV IB mode When the HCA is configured in SRIOV IB mode (that is, at least one of the ports is IB) and the probe_vf module param isn't specified, mlx4_init_one() failed because of the following condition: if (ib_ports && (num_vfs_argc > 1 || probe_vfs_argc > 1)) { ..... } The root cause for that is a mistake in the initialization of num_vfs_argc and probe_vfs_argc. When num_vfs / probe_vf aren't given, their argument count counterpart should be 0, fix that. Fixes: dd41cc3bb90e ('net/mlx4: Adapt num_vfs/probed_vf params for single port VF') Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-23 20:05:59 +07:00
static int num_vfs_argc;
net/mlx4: Adapt num_vfs/probed_vf params for single port VF A new syntax is added for the module parameters num_vfs and probe_vf. num_vfs=p1,p2,p1+p2 probe_bf=p1,p2,p1+p2 Where p1(2) is the number of VFs on / probed VFs for physical port1(2) and p1+p2 is the number of dual port VFs. Single port VFs are currently supported only when the link type for both ports of the device is Ethernet. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-19 23:11:53 +07:00
module_param_array(num_vfs, byte , &num_vfs_argc, 0444);
MODULE_PARM_DESC(num_vfs, "enable #num_vfs functions if num_vfs > 0\n"
"num_vfs=port1,port2,port1+2");
static uint8_t probe_vf[3] = {0, 0, 0};
net/mlx4_core: Allow not to specify probe_vf in SRIOV IB mode When the HCA is configured in SRIOV IB mode (that is, at least one of the ports is IB) and the probe_vf module param isn't specified, mlx4_init_one() failed because of the following condition: if (ib_ports && (num_vfs_argc > 1 || probe_vfs_argc > 1)) { ..... } The root cause for that is a mistake in the initialization of num_vfs_argc and probe_vfs_argc. When num_vfs / probe_vf aren't given, their argument count counterpart should be 0, fix that. Fixes: dd41cc3bb90e ('net/mlx4: Adapt num_vfs/probed_vf params for single port VF') Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-23 20:05:59 +07:00
static int probe_vfs_argc;
net/mlx4: Adapt num_vfs/probed_vf params for single port VF A new syntax is added for the module parameters num_vfs and probe_vf. num_vfs=p1,p2,p1+p2 probe_bf=p1,p2,p1+p2 Where p1(2) is the number of VFs on / probed VFs for physical port1(2) and p1+p2 is the number of dual port VFs. Single port VFs are currently supported only when the link type for both ports of the device is Ethernet. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-19 23:11:53 +07:00
module_param_array(probe_vf, byte, &probe_vfs_argc, 0444);
MODULE_PARM_DESC(probe_vf, "number of vfs to probe by pf driver (num_vfs > 0)\n"
"probe_vf=port1,port2,port1+2");
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
mlx4_core: Allow choosing flow steering mode Device managed flow steering will be enabled only under administrator directive provided through setting the existing module parameter log_num_mgm_entry_size to -1 (if the device actually supports flow steering). If flow steering isn't requested or not available, the driver will use the value of log_num_mgm_entry_size and B0 steering. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:12:00 +07:00
int mlx4_log_num_mgm_entry_size = MLX4_DEFAULT_MGM_LOG_ENTRY_SIZE;
mlx4: Traffic steering management support for SRIOV Let multicast/unicast attaching flow go through resource tracker. The PF is the one responsible for managing all the steering entries. Define and use module parameter that determines the number of qps per multicast group. Minor changes in function calls according to changed prototype. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:16:02 +07:00
module_param_named(log_num_mgm_entry_size,
mlx4_log_num_mgm_entry_size, int, 0444);
MODULE_PARM_DESC(log_num_mgm_entry_size, "log mgm size, that defines the num"
" of qp per mcg, for example:"
mlx4_core: Allow choosing flow steering mode Device managed flow steering will be enabled only under administrator directive provided through setting the existing module parameter log_num_mgm_entry_size to -1 (if the device actually supports flow steering). If flow steering isn't requested or not available, the driver will use the value of log_num_mgm_entry_size and B0 steering. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:12:00 +07:00
" 10 gives 248.range: 7 <="
{NET, IB}/mlx4: Add device managed flow steering firmware API The driver is modified to support three operation modes. If supported by firmware use the device managed flow steering API, that which we call device managed steering mode. Else, if the firmware supports the B0 steering mode use it, and finally, if none of the above, use the A0 steering mode. When the steering mode is device managed, the code is modified such that L2 based rules set by the mlx4_en driver for Ethernet unicast and multicast, and the IB stack multicast attach calls done through the mlx4_ib driver are all routed to use the device managed API. When attaching rule using device managed flow steering API, the firmware returns a 64 bit registration id, which is to be provided during detach. Currently the firmware is always programmed during HCA initialization to use standard L2 hashing. Future work should be done to allow configuring the flow-steering hash function with common, non proprietary means. Signed-off-by: Hadar Hen Zion <hadarh@mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-05 11:03:46 +07:00
" log_num_mgm_entry_size <= 12."
mlx4_core: Allow choosing flow steering mode Device managed flow steering will be enabled only under administrator directive provided through setting the existing module parameter log_num_mgm_entry_size to -1 (if the device actually supports flow steering). If flow steering isn't requested or not available, the driver will use the value of log_num_mgm_entry_size and B0 steering. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:12:00 +07:00
" To activate device managed"
" flow steering when available, set to -1");
mlx4: Traffic steering management support for SRIOV Let multicast/unicast attaching flow go through resource tracker. The PF is the one responsible for managing all the steering entries. Define and use module parameter that determines the number of qps per multicast group. Minor changes in function calls according to changed prototype. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:16:02 +07:00
net/mlx4_core: Set CQE/EQE size to 64B by default To achieve out of the box performance default is to use 64 byte CQE/EQE. In tests that we conduct in our labs, we achieved a performance improvement of twice the message rate. For older VF/libmlx4 support, enable_64b_cqe_eqe must be set to 0 (disabled). Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-20 02:20:15 +07:00
static bool enable_64b_cqe_eqe = true;
mlx4: 64-byte CQE/EQE support ConnectX-3 devices can use either 64- or 32-byte completion queue entries (CQEs) and event queue entries (EQEs). Using 64-byte EQEs/CQEs performs better because each entry is aligned to a complete cacheline. This patch queries the HCA's capabilities, and if it supports 64-byte CQEs and EQES the driver will configure the HW to work in 64-byte mode. The 32-byte vs 64-byte mode is global per HCA and not per CQ or EQ. Since this mode is global, userspace (libmlx4) must be updated to work with the configured CQE size, and guests using SR-IOV virtual functions need to know both EQE and CQE size. In case one of the 64-byte CQE/EQE capabilities is activated, the patch makes sure that older guest drivers that use the QUERY_DEV_FUNC command (e.g as done in mlx4_core of Linux 3.3..3.6) will notice that they need an update to be able to work with the PPF. This is done by changing the returned pf_context_behaviour not to be zero any more. In case none of these capabilities is activated that value remains zero and older guest drivers can run OK. The SRIOV related flow is as follows 1. the PPF does the detection of the new capabilities using QUERY_DEV_CAP command. 2. the PPF activates the new capabilities using INIT_HCA. 3. the VF detects if the PPF activated the capabilities using QUERY_HCA, and if this is the case activates them for itself too. Note that the VF detects that it must be aware to the new PF behaviour using QUERY_FUNC_CAP. Steps 1 and 2 apply also for native mode. User space notification is done through a new field introduced in struct mlx4_ib_ucontext which holds device capabilities for which user space must take action. This changes the binary interface so the ABI towards libmlx4 exposed through uverbs is bumped from 3 to 4 but only when **needed** i.e. only when the driver does use 64-byte CQEs or future device capabilities which must be in sync by user space. This practice allows to work with unmodified libmlx4 on older devices (e.g A0, B0) which don't support 64-byte CQEs. In order to keep existing systems functional when they update to a newer kernel that contains these changes in VF and userspace ABI, a module parameter enable_64b_cqe_eqe must be set to enable 64-byte mode; the default is currently false. Signed-off-by: Eli Cohen <eli@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-10-21 21:59:24 +07:00
module_param(enable_64b_cqe_eqe, bool, 0444);
MODULE_PARM_DESC(enable_64b_cqe_eqe,
net/mlx4_core: Set CQE/EQE size to 64B by default To achieve out of the box performance default is to use 64 byte CQE/EQE. In tests that we conduct in our labs, we achieved a performance improvement of twice the message rate. For older VF/libmlx4 support, enable_64b_cqe_eqe must be set to 0 (disabled). Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-20 02:20:15 +07:00
"Enable 64 byte CQEs/EQEs when the FW supports this (default: True)");
mlx4: 64-byte CQE/EQE support ConnectX-3 devices can use either 64- or 32-byte completion queue entries (CQEs) and event queue entries (EQEs). Using 64-byte EQEs/CQEs performs better because each entry is aligned to a complete cacheline. This patch queries the HCA's capabilities, and if it supports 64-byte CQEs and EQES the driver will configure the HW to work in 64-byte mode. The 32-byte vs 64-byte mode is global per HCA and not per CQ or EQ. Since this mode is global, userspace (libmlx4) must be updated to work with the configured CQE size, and guests using SR-IOV virtual functions need to know both EQE and CQE size. In case one of the 64-byte CQE/EQE capabilities is activated, the patch makes sure that older guest drivers that use the QUERY_DEV_FUNC command (e.g as done in mlx4_core of Linux 3.3..3.6) will notice that they need an update to be able to work with the PPF. This is done by changing the returned pf_context_behaviour not to be zero any more. In case none of these capabilities is activated that value remains zero and older guest drivers can run OK. The SRIOV related flow is as follows 1. the PPF does the detection of the new capabilities using QUERY_DEV_CAP command. 2. the PPF activates the new capabilities using INIT_HCA. 3. the VF detects if the PPF activated the capabilities using QUERY_HCA, and if this is the case activates them for itself too. Note that the VF detects that it must be aware to the new PF behaviour using QUERY_FUNC_CAP. Steps 1 and 2 apply also for native mode. User space notification is done through a new field introduced in struct mlx4_ib_ucontext which holds device capabilities for which user space must take action. This changes the binary interface so the ABI towards libmlx4 exposed through uverbs is bumped from 3 to 4 but only when **needed** i.e. only when the driver does use 64-byte CQEs or future device capabilities which must be in sync by user space. This practice allows to work with unmodified libmlx4 on older devices (e.g A0, B0) which don't support 64-byte CQEs. In order to keep existing systems functional when they update to a newer kernel that contains these changes in VF and userspace ABI, a module parameter enable_64b_cqe_eqe must be set to enable 64-byte mode; the default is currently false. Signed-off-by: Eli Cohen <eli@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-10-21 21:59:24 +07:00
net/mlx4_core: Enable CQE/EQE stride support This feature is intended for archs having cache line larger then 64B. Since our CQE/EQEs are generally 64B in those systems, HW will write twice to the same cache line consecutively, causing pipe locks due to he hazard prevention mechanism. For elements in a cyclic buffer, writes are consecutive, so entries smaller than a cache line should be avoided, especially if they are written at a high rate. Reduce consecutive writes to same cache line in CQs/EQs, by allowing the driver to increase the distance between entries so that each will reside in a different cache line. Until the introduction of this feature, there were two types of CQE/EQE: 1. 32B stride and context in the [0-31] segment 2. 64B stride and context in the [32-63] segment This feature introduces two additional types: 3. 128B stride and context in the [0-31] segment (128B cache line) 4. 256B stride and context in the [0-31] segment (256B cache line) Modify the mlx4_core driver to query the device for the CQE/EQE cache line stride capability and to enable that capability when the host cache line size is larger than 64 bytes (supported cache lines are 128B and 256B). The mlx4 IB driver and libmlx4 need not be aware of this change. The PF context behaviour is changed to require this change in VF drivers running on such archs. Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-18 15:50:59 +07:00
#define PF_CONTEXT_BEHAVIOUR_MASK (MLX4_FUNC_CAP_64B_EQE_CQE | \
net/mlx4: Add support for A0 steering Add the required firmware commands for A0 steering and a way to enable that. The firmware support focuses on INIT_HCA, QUERY_HCA, QUERY_PORT, QUERY_DEV_CAP and QUERY_FUNC_CAP commands. Those commands are used to configure and query the device. The different A0 DMFS (steering) modes are: Static - optimized performance, but flow steering rules are limited. This mode should be choosed explicitly by the user in order to be used. Dynamic - this mode should be explicitly choosed by the user. In this mode, the FW works in optimized steering mode as long as it can and afterwards automatically drops to classic (full) DMFS. Disable - this mode should be explicitly choosed by the user. The user instructs the system not to use optimized steering, even if the FW supports Dynamic A0 DMFS (and thus will be able to use optimized steering in Default A0 DMFS mode). Default - this mode is implicitly choosed. In this mode, if the FW supports Dynamic A0 DMFS, it'll work in this mode. Otherwise, it'll work at Disable A0 DMFS mode. Under SRIOV configuration, when the A0 steering mode is enabled, older guest VF drivers who aren't using the RX QP allocation flag (MLX4_RESERVE_A0_QP) will get a QP from the general range and fail when attempting to register a steering rule. To avoid that, the PF context behaviour is changed once on A0 static mode, to require support for the allocation flag in VF drivers too. In order to enable A0 steering, we use log_num_mgm_entry_size param. If the value of the parameter is not positive, we treat the absolute value of log_num_mgm_entry_size as a bit field. Setting bit 2 of this bit field enables static A0 steering. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:58:00 +07:00
MLX4_FUNC_CAP_EQE_CQE_STRIDE | \
MLX4_FUNC_CAP_DMFS_A0_STATIC)
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
#define RESET_PERSIST_MASK_FLAGS (MLX4_FLAG_SRIOV)
mlx4_core: remove __dev* attributes CONFIG_HOTPLUG is going away as an option. As result the __dev* markings will be going away. Remove use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit. Signed-off-by: Bill Pemberton <wfp5p@virginia.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-03 21:23:15 +07:00
static char mlx4_version[] =
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
DRV_NAME ": Mellanox ConnectX core driver v"
DRV_VERSION " (" DRV_RELDATE ")\n";
static struct mlx4_profile default_profile = {
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
.num_qp = 1 << 18,
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
.num_srq = 1 << 16,
IB/mlx4: Increase max outstanding RDMA reads as target Change the maximum number of outstanding RDMA reads allowed as a target from 4 to 16 to per QP. This allows RDMA read operations to pipeline better. Pointed out by Dotan Barak and Sagi Rotem. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-07-17 17:11:43 +07:00
.rdmarc_per_qp = 1 << 4,
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
.num_cq = 1 << 16,
.num_mcg = 1 << 13,
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
.num_mpt = 1 << 19,
mlx4_core: Fix mtt profile issue Num mtts from profile is really the number of mtt segments. Thus, in make profile, to get the proper number of MTT entries, must multiply num_mtts by mtts per segment. Signed-off-by: Marcel Apfelbaum <marcela@dev.mellanox.co.il> Reviewed-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-01-19 16:45:31 +07:00
.num_mtt = 1 << 20, /* It is really num mtt segements */
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
};
net/mlx4_core: Use low memory profile on kdump kernel When running in kdump kernel, reduce number of resources allocated for the hardware. This will enable the NIC to operate in this low memory environment at the expense of performance and some features not related to the basic NIC functionality. Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-22 19:44:11 +07:00
static struct mlx4_profile low_mem_profile = {
.num_qp = 1 << 17,
.num_srq = 1 << 6,
.rdmarc_per_qp = 1 << 4,
.num_cq = 1 << 8,
.num_mcg = 1 << 8,
.num_mpt = 1 << 9,
.num_mtt = 1 << 7,
};
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
static int log_num_mac = 7;
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
module_param_named(log_num_mac, log_num_mac, int, 0444);
MODULE_PARM_DESC(log_num_mac, "Log2 max number of MACs per ETH port (1-7)");
static int log_num_vlan;
module_param_named(log_num_vlan, log_num_vlan, int, 0444);
MODULE_PARM_DESC(log_num_vlan, "Log2 max number of VLANs per ETH port (0-7)");
mlx4_core: Deprecate log_num_vlan module param Enable the maximum size (128) supported by the device for the shadow vlans table, ignoring the module parameter that overrides it. This table is only used by the IBoE control plane for setting a vlan index into an RC/UC QP context or UD Address Handle. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-10-16 15:26:21 +07:00
/* Log2 max number of VLANs per ETH port (0-7) */
#define MLX4_LOG_NUM_VLANS 7
net/mlx4_core: Use low memory profile on kdump kernel When running in kdump kernel, reduce number of resources allocated for the hardware. This will enable the NIC to operate in this low memory environment at the expense of performance and some features not related to the basic NIC functionality. Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-22 19:44:11 +07:00
#define MLX4_MIN_LOG_NUM_VLANS 0
#define MLX4_MIN_LOG_NUM_MAC 1
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
module_param: make bool parameters really bool (net & drivers/net) module_param(bool) used to counter-intuitively take an int. In fddd5201 (mid-2009) we allowed bool or int/unsigned int using a messy trick. It's time to remove the int/unsigned int option. For this version it'll simply give a warning, but it'll break next kernel version. (Thanks to Joe Perches for suggesting coccinelle for 0/1 -> true/false). Cc: "David S. Miller" <davem@davemloft.net> Cc: netdev@vger.kernel.org Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 21:08:01 +07:00
static bool use_prio;
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
module_param_named(use_prio, use_prio, bool, 0444);
net/mlx4_core: Deprecate use_prio module parameter use_prio was added as part of an infrastructure for running FCoE in A0 mode. FCoE didn't get into Mellanox Upstream driver, and when it will, it won't be using A0 steering mode. Therefore we can safely deprecate this module parameter without hurting any existing user. CC: Carol Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-22 19:55:39 +07:00
MODULE_PARM_DESC(use_prio, "Enable steering by VLAN priority on ETH ports (deprecated)");
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
mlx4_core: mtts resources units changed to offset In the previous implementation mtts are managed by: 1. order - log(mtt segments), 'mtt segment' groups several mtts together. 2. first_seg - segment location relative to mtt table. In the current implementation: 1. order - log(mtts) rather than segments 2. offset - mtt index in mtt table Note: The actual mtt allocation is made in segments but it is transparent to callers. Rational: The mtt resource holders are not interested on how the allocation of mtt is done, but rather on how they will use it. Signed-off-by: Marcel Apfelbaum <marcela@dev.mellanox.co.il> Reviewed-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:16:56 +07:00
int log_mtts_per_seg = ilog2(MLX4_MTT_ENTRY_PER_SEG);
mlx4_core: Add module parameter for number of MTTs per segment The current MTT allocator uses kmalloc() to allocate a buffer for its buddy allocator, and thus is limited in the amount of MTT segments that it can control. As a result, the size of memory that can be registered is limited too. This patch uses a module parameter to control the number of MTT entries that each segment represents, allowing more memory to be registered with the same number of segments. Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-05-28 04:38:34 +07:00
module_param_named(log_mtts_per_seg, log_mtts_per_seg, int, 0444);
mlx4_core: Allow log_mtts_per_seg up to 7 As there are now machines with a lot more physical memory, we need to be able to register more memory. This patch lifts the upper limit of log_mtts_per_seg from 5 to 7, increasing the amount of memory that can be registered. Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-09-20 13:42:38 +07:00
MODULE_PARM_DESC(log_mtts_per_seg, "Log2 number of MTT entries per segment (1-7)");
mlx4_core: Add module parameter for number of MTTs per segment The current MTT allocator uses kmalloc() to allocate a buffer for its buddy allocator, and thus is limited in the amount of MTT segments that it can control. As a result, the size of memory that can be registered is limited too. This patch uses a module parameter to control the number of MTT entries that each segment represents, allowing more memory to be registered with the same number of segments. Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-05-28 04:38:34 +07:00
mlx4_core: Changing link sensing logic New FW can give clues to driver regarding default port type and whether or not we should default to link sensing on the port. 2 bits are added to QUERY_PORT command: 1. suggested_type: This bit gives a hint whether the default port type should be IB or Ethernet. The driver will use this hint in case the user didn't specify explicitly the link layer type he wants to set. 2. default_sense: If this bit is set, we would sense the port type on start-up and default the port to link sensing Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 11:00:34 +07:00
static int port_type_array[2] = {MLX4_PORT_TYPE_NONE, MLX4_PORT_TYPE_NONE};
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
static int arr_argc = 2;
module_param_array(port_type_array, int, &arr_argc, 0444);
mlx4_core: Changing link sensing logic New FW can give clues to driver regarding default port type and whether or not we should default to link sensing on the port. 2 bits are added to QUERY_PORT command: 1. suggested_type: This bit gives a hint whether the default port type should be IB or Ethernet. The driver will use this hint in case the user didn't specify explicitly the link layer type he wants to set. 2. default_sense: If this bit is set, we would sense the port type on start-up and default the port to link sensing Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 11:00:34 +07:00
MODULE_PARM_DESC(port_type_array, "Array of port types: HW_DEFAULT (0) is default "
"1 for IB, 2 for Ethernet");
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
struct mlx4_port_config {
struct list_head list;
enum mlx4_port_type port_type[MLX4_MAX_PORTS + 1];
struct pci_dev *pdev;
};
net/mlx4_core: mlx4_init_slave() shouldn't access comm channel before PF is ready Currently, the PF call to pci_enable_sriov from the PF probe function stalls for 10 seconds times the number of VFs probed on the host. This happens because the way for such VFs to determine of the PF initialization finished, is by attempting to issue reset on the comm-channel and get timeout (after 10s). The PF probe function is called from a kenernel workqueue, and therefore during that time, rcu lock is being held and kernel's workqueue is stalled. This blocks other processes that try to use the workqueue or rcu lock. For example, interface renaming which is calling rcu_synchronize is blocked, and timedout by systemd. Changed mlx4_init_slave() to allow VF probed on the host to immediatly detect that the PF is not ready, and return EPROBE_DEFER instantly. Only when the PF finishes the initialization, allow such VFs to access the comm channel. This issue and fix are relevant only for probed VFs on the hypervisor, there is no way to pass this information to a VM until comm channel is ready, so in a VM, if PF is not ready, the first command will be timedout after 10 seconds and return EPROBE_DEFER. Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-06 23:28:17 +07:00
static atomic_t pf_loading = ATOMIC_INIT(0);
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
int mlx4_check_port_params(struct mlx4_dev *dev,
enum mlx4_port_type *port_type)
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
{
int i;
mlx4_core: Check for DPDP violation only when DPDP is not supported Move check for DPDP out of the loop to make the code more readable. Signed-off-by: Yuval Shaia <yuval.shaia@oracle.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-12-14 01:18:40 +07:00
if (!(dev->caps.flags & MLX4_DEV_CAP_FLAG_DPDP)) {
for (i = 0; i < dev->caps.num_ports - 1; i++) {
if (port_type[i] != port_type[i + 1]) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Only same port types supported on this HCA, aborting\n");
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
return -EINVAL;
}
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
}
}
for (i = 0; i < dev->caps.num_ports; i++) {
if (!(port_type[i] & dev->caps.supported_type[i+1])) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Requested port type for port %d is not supported on this HCA\n",
i + 1);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
return -EINVAL;
}
}
return 0;
}
static void mlx4_set_port_mask(struct mlx4_dev *dev)
{
int i;
for (i = 1; i <= dev->caps.num_ports; ++i)
mlx4: Extanding port_mask functionality Port mask now has additional state. Port can be set as "none". In this case neither the mlx4_en or mlx4_ib drivers take ownership of the port. In multifunction mode there is an option to set the vfs as single ported devices. (in single function mode, both physical ports belong to same function) Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:10:41 +07:00
dev->caps.port_mask[i] = dev->caps.port_type[i];
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
}
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
enum {
MLX4_QUERY_FUNC_NUM_SYS_EQS = 1 << 0,
};
static int mlx4_query_func(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
{
int err = 0;
struct mlx4_func func;
if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_SYS_EQS) {
err = mlx4_QUERY_FUNC(dev, &func, 0);
if (err) {
mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting.\n");
return err;
}
dev_cap->max_eqs = func.max_eq;
dev_cap->reserved_eqs = func.rsvd_eqs;
dev_cap->reserved_uars = func.rsvd_uars;
err |= MLX4_QUERY_FUNC_NUM_SYS_EQS;
}
return err;
}
net/mlx4_core: Enable CQE/EQE stride support This feature is intended for archs having cache line larger then 64B. Since our CQE/EQEs are generally 64B in those systems, HW will write twice to the same cache line consecutively, causing pipe locks due to he hazard prevention mechanism. For elements in a cyclic buffer, writes are consecutive, so entries smaller than a cache line should be avoided, especially if they are written at a high rate. Reduce consecutive writes to same cache line in CQs/EQs, by allowing the driver to increase the distance between entries so that each will reside in a different cache line. Until the introduction of this feature, there were two types of CQE/EQE: 1. 32B stride and context in the [0-31] segment 2. 64B stride and context in the [32-63] segment This feature introduces two additional types: 3. 128B stride and context in the [0-31] segment (128B cache line) 4. 256B stride and context in the [0-31] segment (256B cache line) Modify the mlx4_core driver to query the device for the CQE/EQE cache line stride capability and to enable that capability when the host cache line size is larger than 64 bytes (supported cache lines are 128B and 256B). The mlx4 IB driver and libmlx4 need not be aware of this change. The PF context behaviour is changed to require this change in VF drivers running on such archs. Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-18 15:50:59 +07:00
static void mlx4_enable_cqe_eqe_stride(struct mlx4_dev *dev)
{
struct mlx4_caps *dev_cap = &dev->caps;
/* FW not supporting or cancelled by user */
if (!(dev_cap->flags2 & MLX4_DEV_CAP_FLAG2_EQE_STRIDE) ||
!(dev_cap->flags2 & MLX4_DEV_CAP_FLAG2_CQE_STRIDE))
return;
/* Must have 64B CQE_EQE enabled by FW to use bigger stride
* When FW has NCSI it may decide not to report 64B CQE/EQEs
*/
if (!(dev_cap->flags & MLX4_DEV_CAP_FLAG_64B_EQE) ||
!(dev_cap->flags & MLX4_DEV_CAP_FLAG_64B_CQE)) {
dev_cap->flags2 &= ~MLX4_DEV_CAP_FLAG2_CQE_STRIDE;
dev_cap->flags2 &= ~MLX4_DEV_CAP_FLAG2_EQE_STRIDE;
return;
}
if (cache_line_size() == 128 || cache_line_size() == 256) {
mlx4_dbg(dev, "Enabling CQE stride cacheLine supported\n");
/* Changing the real data inside CQE size to 32B */
dev_cap->flags &= ~MLX4_DEV_CAP_FLAG_64B_CQE;
dev_cap->flags &= ~MLX4_DEV_CAP_FLAG_64B_EQE;
if (mlx4_is_master(dev))
dev_cap->function_caps |= MLX4_FUNC_CAP_EQE_CQE_STRIDE;
} else {
net/mlx4_core: Fix misleading debug print on CQE stride support We do support cache line sizes of 32 and 64 bytes without activating the CQE stride feature. Fix a misleading print saying that these cache line sizes aren't supported. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-03 22:57:17 +07:00
if (cache_line_size() != 32 && cache_line_size() != 64)
mlx4_dbg(dev, "Disabling CQE stride, cacheLine size unsupported\n");
net/mlx4_core: Enable CQE/EQE stride support This feature is intended for archs having cache line larger then 64B. Since our CQE/EQEs are generally 64B in those systems, HW will write twice to the same cache line consecutively, causing pipe locks due to he hazard prevention mechanism. For elements in a cyclic buffer, writes are consecutive, so entries smaller than a cache line should be avoided, especially if they are written at a high rate. Reduce consecutive writes to same cache line in CQs/EQs, by allowing the driver to increase the distance between entries so that each will reside in a different cache line. Until the introduction of this feature, there were two types of CQE/EQE: 1. 32B stride and context in the [0-31] segment 2. 64B stride and context in the [32-63] segment This feature introduces two additional types: 3. 128B stride and context in the [0-31] segment (128B cache line) 4. 256B stride and context in the [0-31] segment (256B cache line) Modify the mlx4_core driver to query the device for the CQE/EQE cache line stride capability and to enable that capability when the host cache line size is larger than 64 bytes (supported cache lines are 128B and 256B). The mlx4 IB driver and libmlx4 need not be aware of this change. The PF context behaviour is changed to require this change in VF drivers running on such archs. Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-18 15:50:59 +07:00
dev_cap->flags2 &= ~MLX4_DEV_CAP_FLAG2_CQE_STRIDE;
dev_cap->flags2 &= ~MLX4_DEV_CAP_FLAG2_EQE_STRIDE;
}
}
net/mlx4: Refactor QUERY_PORT Currently QUERY_PORT is done as a part of QUERY_DEV_CAP firmware command. Since we would like to use it without querying all device capabilities, extract this part to be a function of its own. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:57:59 +07:00
static int _mlx4_dev_port(struct mlx4_dev *dev, int port,
struct mlx4_port_cap *port_cap)
{
dev->caps.vl_cap[port] = port_cap->max_vl;
dev->caps.ib_mtu_cap[port] = port_cap->ib_mtu;
dev->phys_caps.gid_phys_table_len[port] = port_cap->max_gids;
dev->phys_caps.pkey_phys_table_len[port] = port_cap->max_pkeys;
/* set gid and pkey table operating lengths by default
* to non-sriov values
*/
dev->caps.gid_table_len[port] = port_cap->max_gids;
dev->caps.pkey_table_len[port] = port_cap->max_pkeys;
dev->caps.port_width_cap[port] = port_cap->max_port_width;
dev->caps.eth_mtu_cap[port] = port_cap->eth_mtu;
dev->caps.def_mac[port] = port_cap->def_mac;
dev->caps.supported_type[port] = port_cap->supported_port_types;
dev->caps.suggested_type[port] = port_cap->suggested_type;
dev->caps.default_sense[port] = port_cap->default_sense;
dev->caps.trans_type[port] = port_cap->trans_type;
dev->caps.vendor_oui[port] = port_cap->vendor_oui;
dev->caps.wavelength[port] = port_cap->wavelength;
dev->caps.trans_code[port] = port_cap->trans_code;
return 0;
}
static int mlx4_dev_port(struct mlx4_dev *dev, int port,
struct mlx4_port_cap *port_cap)
{
int err = 0;
err = mlx4_QUERY_PORT(dev, port, port_cap);
if (err)
mlx4_err(dev, "QUERY_PORT command failed.\n");
return err;
}
net/mlx4_en: Add RX-ALL support Enabled when the device supports KEEP FCS and IGNORE FCS. When the flag is set, pass all received frames up the stack, even ones with invalid FCS, controlled by ethtool. Signed-off-by: Muhammad Mahajna <muhammadm@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-02 20:31:22 +07:00
static inline void mlx4_enable_ignore_fcs(struct mlx4_dev *dev)
{
if (!(dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_IGNORE_FCS))
return;
if (mlx4_is_mfunc(dev)) {
mlx4_dbg(dev, "SRIOV mode - Disabling Ignore FCS");
dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_IGNORE_FCS;
return;
}
if (!(dev->caps.flags & MLX4_DEV_CAP_FLAG_FCS_KEEP)) {
mlx4_dbg(dev,
"Keep FCS is not supported - Disabling Ignore FCS");
dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_IGNORE_FCS;
return;
}
}
net/mlx4: Refactor QUERY_PORT Currently QUERY_PORT is done as a part of QUERY_DEV_CAP firmware command. Since we would like to use it without querying all device capabilities, extract this part to be a function of its own. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:57:59 +07:00
#define MLX4_A0_STEERING_TABLE_SIZE 256
mlx4_core: Fix section mismatches Commit ee49bd93 ("mlx4_core: Reset device when internal error is detected") introduced some section mismatch problems when CONFIG_HOTPLUG=n, because the error recovery code tears down and reinitializes the device after everything is loaded, which ends up calling into lots of code marked __devinit and __devexit from regular .text. Fix this by getting rid of these now-incorrect section markers. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-10-11 05:43:54 +07:00
static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
{
int err;
IB/mlx4: Handle FW command interface rev 3 Upcoming firmware introduces command interface revision 3, which changes the way port capabilities are queried and set. Update the driver to handle both the new and old command interfaces by adding a new MLX4_FLAG_OLD_PORT_CMDS that it is set after querying the firmware interface revision and then using the correct interface based on the setting of the flag. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-06-18 22:15:02 +07:00
int i;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err = mlx4_QUERY_DEV_CAP(dev, dev_cap);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return err;
}
net/mlx4_core: Avoid double dumping of the PF device capabilities To support asymmetric EQ allocations, we should query the device capabilities prior to enabling SRIOV. As a side effect of adding that, we are dumping the PF device capabilities twice. Avoid that by moving the printing into a helper function which is called once. Fixes: 7ae0e400cd93 ('net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs') Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-14 21:18:05 +07:00
mlx4_dev_cap_dump(dev, dev_cap);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (dev_cap->min_page_sz > PAGE_SIZE) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "HCA minimum page size of %d bigger than kernel PAGE_SIZE of %ld, aborting\n",
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
dev_cap->min_page_sz, PAGE_SIZE);
return -ENODEV;
}
if (dev_cap->num_ports > MLX4_MAX_PORTS) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "HCA has %d ports, but we only support %d, aborting\n",
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
dev_cap->num_ports, MLX4_MAX_PORTS);
return -ENODEV;
}
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
if (dev_cap->uar_size > pci_resource_len(dev->persist->pdev, 2)) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "HCA reported UAR size of 0x%x bigger than PCI resource 2 size of 0x%llx, aborting\n",
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
dev_cap->uar_size,
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
(unsigned long long)
pci_resource_len(dev->persist->pdev, 2));
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return -ENODEV;
}
dev->caps.num_ports = dev_cap->num_ports;
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
dev->caps.num_sys_eqs = dev_cap->num_sys_eqs;
dev->phys_caps.num_phys_eqs = dev_cap->flags2 & MLX4_DEV_CAP_FLAG2_SYS_EQS ?
dev->caps.num_sys_eqs :
MLX4_MAX_EQ_NUM;
IB/mlx4: Handle FW command interface rev 3 Upcoming firmware introduces command interface revision 3, which changes the way port capabilities are queried and set. Update the driver to handle both the new and old command interfaces by adding a new MLX4_FLAG_OLD_PORT_CMDS that it is set after querying the firmware interface revision and then using the correct interface based on the setting of the flag. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-06-18 22:15:02 +07:00
for (i = 1; i <= dev->caps.num_ports; ++i) {
net/mlx4: Refactor QUERY_PORT Currently QUERY_PORT is done as a part of QUERY_DEV_CAP firmware command. Since we would like to use it without querying all device capabilities, extract this part to be a function of its own. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:57:59 +07:00
err = _mlx4_dev_port(dev, i, dev_cap->port_cap + i);
if (err) {
mlx4_err(dev, "QUERY_PORT command failed, aborting\n");
return err;
}
IB/mlx4: Handle FW command interface rev 3 Upcoming firmware introduces command interface revision 3, which changes the way port capabilities are queried and set. Update the driver to handle both the new and old command interfaces by adding a new MLX4_FLAG_OLD_PORT_CMDS that it is set after querying the firmware interface revision and then using the correct interface based on the setting of the flag. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-06-18 22:15:02 +07:00
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
dev->caps.uar_page_size = PAGE_SIZE;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
dev->caps.num_uars = dev_cap->uar_size / PAGE_SIZE;
dev->caps.local_ca_ack_delay = dev_cap->local_ca_ack_delay;
dev->caps.bf_reg_size = dev_cap->bf_reg_size;
dev->caps.bf_regs_per_page = dev_cap->bf_regs_per_page;
dev->caps.max_sq_sg = dev_cap->max_sq_sg;
dev->caps.max_rq_sg = dev_cap->max_rq_sg;
dev->caps.max_wqes = dev_cap->max_qp_sz;
dev->caps.max_qp_init_rdma = dev_cap->max_requester_per_qp;
dev->caps.max_srq_wqes = dev_cap->max_srq_sz;
dev->caps.max_srq_sge = dev_cap->max_rq_sg - 1;
dev->caps.reserved_srqs = dev_cap->reserved_srqs;
dev->caps.max_sq_desc_sz = dev_cap->max_sq_desc_sz;
dev->caps.max_rq_desc_sz = dev_cap->max_rq_desc_sz;
/*
* Subtract 1 from the limit because we need to allocate a
* spare CQE so the HCA HW can tell the difference between an
* empty CQ and a full CQ.
*/
dev->caps.max_cqes = dev_cap->max_cq_sz - 1;
dev->caps.reserved_cqs = dev_cap->reserved_cqs;
dev->caps.reserved_eqs = dev_cap->reserved_eqs;
mlx4_core: mtts resources units changed to offset In the previous implementation mtts are managed by: 1. order - log(mtt segments), 'mtt segment' groups several mtts together. 2. first_seg - segment location relative to mtt table. In the current implementation: 1. order - log(mtts) rather than segments 2. offset - mtt index in mtt table Note: The actual mtt allocation is made in segments but it is transparent to callers. Rational: The mtt resource holders are not interested on how the allocation of mtt is done, but rather on how they will use it. Signed-off-by: Marcel Apfelbaum <marcela@dev.mellanox.co.il> Reviewed-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:16:56 +07:00
dev->caps.reserved_mtts = dev_cap->reserved_mtts;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
dev->caps.reserved_mrws = dev_cap->reserved_mrws;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* The first 128 UARs are used for EQ doorbells */
dev->caps.reserved_uars = max_t(int, 128, dev_cap->reserved_uars);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
dev->caps.reserved_pds = dev_cap->reserved_pds;
IB/mlx4: Add support for XRC domains Support creating and destroying XRC domains. Any sharing of the XRCD is managed above the low-level driver. Signed-off-by: Sean Hefty <sean.hefty@intel.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-02 23:01:33 +07:00
dev->caps.reserved_xrcds = (dev->caps.flags & MLX4_DEV_CAP_FLAG_XRC) ?
dev_cap->reserved_xrcds : 0;
dev->caps.max_xrcds = (dev->caps.flags & MLX4_DEV_CAP_FLAG_XRC) ?
dev_cap->max_xrcds : 0;
mlx4_core: mtts resources units changed to offset In the previous implementation mtts are managed by: 1. order - log(mtt segments), 'mtt segment' groups several mtts together. 2. first_seg - segment location relative to mtt table. In the current implementation: 1. order - log(mtts) rather than segments 2. offset - mtt index in mtt table Note: The actual mtt allocation is made in segments but it is transparent to callers. Rational: The mtt resource holders are not interested on how the allocation of mtt is done, but rather on how they will use it. Signed-off-by: Marcel Apfelbaum <marcela@dev.mellanox.co.il> Reviewed-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:16:56 +07:00
dev->caps.mtt_entry_sz = dev_cap->mtt_entry_sz;
mlx4_core: Get the maximum message size from reported device capabilities Get the maximum message size from the device capabilities returned from the QUERY_DEV_CAP firmware command, rather than hard-coding 2 GB. Signed-off-by: Dotan Barak <dotanb@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-06-26 19:55:28 +07:00
dev->caps.max_msg_sz = dev_cap->max_msg_sz;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
dev->caps.page_size_cap = ~(u32) (dev_cap->min_page_sz - 1);
dev->caps.flags = dev_cap->flags;
mlx4_core: Add second capabilities flags field This patch adds a 64-bit flags2 features member to struct mlx4_dev to export further features of the hardware. The original flags field tracks features whose support bits are advertised by the firmware in offsets 0x40 and 0x44 of the query device capabilities command. flags2 will track features whose support bits are scattered at various offsets. RSS support is the first feature to be exported through flags2. RSS capabilities are located at offset 0x2e. The size of the RSS indirection table is also given in this offset. Signed-off-by: Shlomo Pongratz <shlomop@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-04-29 21:04:25 +07:00
dev->caps.flags2 = dev_cap->flags2;
IB/mlx4: Add support for memory management extensions and local DMA L_Key Add support for the following operations to mlx4 when device firmware supports them: - Send with invalidate and local invalidate send queue work requests; - Allocate/free fast register MRs; - Allocate/free fast register MR page lists; - Fast register MR send queue work requests; - Local DMA L_Key. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-07-23 22:12:26 +07:00
dev->caps.bmme_flags = dev_cap->bmme_flags;
dev->caps.reserved_lkey = dev_cap->reserved_lkey;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
dev->caps.stat_rate_support = dev_cap->stat_rate_support;
IB/mlx4: Add IPoIB LSO support Add TSO support to the mlx4_ib driver. Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-17 11:09:27 +07:00
dev->caps.max_gso_sz = dev_cap->max_gso_sz;
mlx4_core: Add second capabilities flags field This patch adds a 64-bit flags2 features member to struct mlx4_dev to export further features of the hardware. The original flags field tracks features whose support bits are advertised by the firmware in offsets 0x40 and 0x44 of the query device capabilities command. flags2 will track features whose support bits are scattered at various offsets. RSS support is the first feature to be exported through flags2. RSS capabilities are located at offset 0x2e. The size of the RSS indirection table is also given in this offset. Signed-off-by: Shlomo Pongratz <shlomop@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-04-29 21:04:25 +07:00
dev->caps.max_rss_tbl_sz = dev_cap->max_rss_tbl_sz;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
net/mlx4_core: Preparations for 802.1ad VLAN support mlx4_core preparation to support hardware accelerated 802.1ad VLAN device. To allow 802.1ad accelerated device, "packet has vlan" (phv) Firmware capability should be available. Firmware without the phv capability won't behave properly and can't support 802.1ad device acceleration. The driver checks the Firmware capability and sets the phv bit accordingly in SET_PORT command. Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-27 18:46:31 +07:00
if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_PHV_EN) {
struct mlx4_init_hca_param hca_param;
memset(&hca_param, 0, sizeof(hca_param));
err = mlx4_QUERY_HCA(dev, &hca_param);
/* Turn off PHV_EN flag in case phv_check_en is set.
* phv_check_en is a HW check that parse the packet and verify
* phv bit was reported correctly in the wqe. To allow QinQ
* PHV_EN flag should be set and phv_check_en must be cleared
* otherwise QinQ packets will be drop by the HW.
*/
if (err || hca_param.phv_check_en)
dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_PHV_EN;
}
mlx4_core: Clean up enabling of SENSE_PORT for older (ConnectX-1/-2) HCAs Instead of having a hard-coded "PCI device ID != 0x1003" (which obviously breaks as newer devices with ID != 0x1003 become available), instead let's set a flag in our PCI device table for the older devices where we're supposed to force using SENSE_PORT. This also avoids enabling SENSE_PORT for virtual functions by mistake. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:53:05 +07:00
/* Sense port always allowed on supported devices for ConnectX-1 and -2 */
if (mlx4_priv(dev)->pci_dev_data & MLX4_PCI_DEV_FORCE_SENSE_PORT)
mlx4: capability for link sensing For ConnectX3 devices, we allow link sensing only if FW explicitly reported it supports the feature. For older versions (ConnectX1 and 2), if the card supports both link layer types (Ethenet and Infiniband), link sensing is supported. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 11:00:26 +07:00
dev->caps.flags |= MLX4_DEV_CAP_FLAG_SENSE_SUPPORT;
mlx4_core: Disable SENSE_PORT for multifunction devices In the current driver, the SENSE_PORT firmware command is issued as a "wrapped" command, but the command handling code doesn't have a wrapper, so it will never do anything other than log an error message. The latest ConnectX-3 2.11.500 firmware reports the SENSE_PORT capability even in multi-function (SR-IOV) mode, so the driver will try to issue the command. At least until the driver has a proper wrapper for SENSE_PORT, make sure we disable the command for multi-function devices. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-28 00:01:19 +07:00
/* Don't do sense port on multifunction devices (for now at least) */
if (mlx4_is_mfunc(dev))
dev->caps.flags &= ~MLX4_DEV_CAP_FLAG_SENSE_SUPPORT;
mlx4: capability for link sensing For ConnectX3 devices, we allow link sensing only if FW explicitly reported it supports the feature. For older versions (ConnectX1 and 2), if the card supports both link layer types (Ethenet and Infiniband), link sensing is supported. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 11:00:26 +07:00
net/mlx4_core: Use low memory profile on kdump kernel When running in kdump kernel, reduce number of resources allocated for the hardware. This will enable the NIC to operate in this low memory environment at the expense of performance and some features not related to the basic NIC functionality. Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-22 19:44:11 +07:00
if (mlx4_low_memory_profile()) {
dev->caps.log_num_macs = MLX4_MIN_LOG_NUM_MAC;
dev->caps.log_num_vlans = MLX4_MIN_LOG_NUM_VLANS;
} else {
dev->caps.log_num_macs = log_num_mac;
dev->caps.log_num_vlans = MLX4_LOG_NUM_VLANS;
}
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
for (i = 1; i <= dev->caps.num_ports; ++i) {
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
dev->caps.port_type[i] = MLX4_PORT_TYPE_NONE;
if (dev->caps.supported_type[i]) {
/* if only ETH is supported - assign ETH */
if (dev->caps.supported_type[i] == MLX4_PORT_TYPE_ETH)
dev->caps.port_type[i] = MLX4_PORT_TYPE_ETH;
mlx4_core: Allow guests to have IB ports Modify mlx4_dev_cap to allow IB support when SR-IOV is active. Modify mlx4_slave_cap to set the "rdma-supported" bit in its flags area, and pass that to the guests (this is done in QUERY_FUNC_CAP and its wrapper). However, we don't activate IB support quite yet -- we leave the error return at the start of mlx4_ib_add in the mlx4_ib driver. In addition, set "protected fmr supported" bit to zero in the QUERY_FUNC_CAP wrapper. Finally, in the QUERY_FUNC_CAP wrapper, we needed to add code which checks for the port type (IB or Ethernet). Previously, this was not an issue, since only Ethernet ports were supported. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:43 +07:00
/* if only IB is supported, assign IB */
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
else if (dev->caps.supported_type[i] ==
mlx4_core: Allow guests to have IB ports Modify mlx4_dev_cap to allow IB support when SR-IOV is active. Modify mlx4_slave_cap to set the "rdma-supported" bit in its flags area, and pass that to the guests (this is done in QUERY_FUNC_CAP and its wrapper). However, we don't activate IB support quite yet -- we leave the error return at the start of mlx4_ib_add in the mlx4_ib driver. In addition, set "protected fmr supported" bit to zero in the QUERY_FUNC_CAP wrapper. Finally, in the QUERY_FUNC_CAP wrapper, we needed to add code which checks for the port type (IB or Ethernet). Previously, this was not an issue, since only Ethernet ports were supported. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:43 +07:00
MLX4_PORT_TYPE_IB)
dev->caps.port_type[i] = MLX4_PORT_TYPE_IB;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
else {
mlx4_core: Allow guests to have IB ports Modify mlx4_dev_cap to allow IB support when SR-IOV is active. Modify mlx4_slave_cap to set the "rdma-supported" bit in its flags area, and pass that to the guests (this is done in QUERY_FUNC_CAP and its wrapper). However, we don't activate IB support quite yet -- we leave the error return at the start of mlx4_ib_add in the mlx4_ib driver. In addition, set "protected fmr supported" bit to zero in the QUERY_FUNC_CAP wrapper. Finally, in the QUERY_FUNC_CAP wrapper, we needed to add code which checks for the port type (IB or Ethernet). Previously, this was not an issue, since only Ethernet ports were supported. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:43 +07:00
/* if IB and ETH are supported, we set the port
* type according to user selection of port type;
* if user selected none, take the FW hint */
if (port_type_array[i - 1] == MLX4_PORT_TYPE_NONE)
mlx4_core: Changing link sensing logic New FW can give clues to driver regarding default port type and whether or not we should default to link sensing on the port. 2 bits are added to QUERY_PORT command: 1. suggested_type: This bit gives a hint whether the default port type should be IB or Ethernet. The driver will use this hint in case the user didn't specify explicitly the link layer type he wants to set. 2. default_sense: If this bit is set, we would sense the port type on start-up and default the port to link sensing Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 11:00:34 +07:00
dev->caps.port_type[i] = dev->caps.suggested_type[i] ?
MLX4_PORT_TYPE_ETH : MLX4_PORT_TYPE_IB;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
else
mlx4_core: Allow guests to have IB ports Modify mlx4_dev_cap to allow IB support when SR-IOV is active. Modify mlx4_slave_cap to set the "rdma-supported" bit in its flags area, and pass that to the guests (this is done in QUERY_FUNC_CAP and its wrapper). However, we don't activate IB support quite yet -- we leave the error return at the start of mlx4_ib_add in the mlx4_ib driver. In addition, set "protected fmr supported" bit to zero in the QUERY_FUNC_CAP wrapper. Finally, in the QUERY_FUNC_CAP wrapper, we needed to add code which checks for the port type (IB or Ethernet). Previously, this was not an issue, since only Ethernet ports were supported. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:43 +07:00
dev->caps.port_type[i] = port_type_array[i - 1];
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
}
mlx4_core: Changing link sensing logic New FW can give clues to driver regarding default port type and whether or not we should default to link sensing on the port. 2 bits are added to QUERY_PORT command: 1. suggested_type: This bit gives a hint whether the default port type should be IB or Ethernet. The driver will use this hint in case the user didn't specify explicitly the link layer type he wants to set. 2. default_sense: If this bit is set, we would sense the port type on start-up and default the port to link sensing Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 11:00:34 +07:00
/*
* Link sensing is allowed on the port if 3 conditions are true:
* 1. Both protocols are supported on the port.
* 2. Different types are supported on the port
* 3. FW declared that it supports link sensing
*/
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
mlx4_priv(dev)->sense.sense_allowed[i] =
mlx4: capability for link sensing For ConnectX3 devices, we allow link sensing only if FW explicitly reported it supports the feature. For older versions (ConnectX1 and 2), if the card supports both link layer types (Ethenet and Infiniband), link sensing is supported. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 11:00:26 +07:00
((dev->caps.supported_type[i] == MLX4_PORT_TYPE_AUTO) &&
mlx4_core: Changing link sensing logic New FW can give clues to driver regarding default port type and whether or not we should default to link sensing on the port. 2 bits are added to QUERY_PORT command: 1. suggested_type: This bit gives a hint whether the default port type should be IB or Ethernet. The driver will use this hint in case the user didn't specify explicitly the link layer type he wants to set. 2. default_sense: If this bit is set, we would sense the port type on start-up and default the port to link sensing Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 11:00:34 +07:00
(dev->caps.flags & MLX4_DEV_CAP_FLAG_DPDP) &&
mlx4: capability for link sensing For ConnectX3 devices, we allow link sensing only if FW explicitly reported it supports the feature. For older versions (ConnectX1 and 2), if the card supports both link layer types (Ethenet and Infiniband), link sensing is supported. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 11:00:26 +07:00
(dev->caps.flags & MLX4_DEV_CAP_FLAG_SENSE_SUPPORT));
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
mlx4_core: Changing link sensing logic New FW can give clues to driver regarding default port type and whether or not we should default to link sensing on the port. 2 bits are added to QUERY_PORT command: 1. suggested_type: This bit gives a hint whether the default port type should be IB or Ethernet. The driver will use this hint in case the user didn't specify explicitly the link layer type he wants to set. 2. default_sense: If this bit is set, we would sense the port type on start-up and default the port to link sensing Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 11:00:34 +07:00
/*
* If "default_sense" bit is set, we move the port to "AUTO" mode
* and perform sense_port FW command to try and set the correct
* port type from beginning
*/
mlx4_core: using array index for sense_allowed Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-29 14:42:34 +07:00
if (mlx4_priv(dev)->sense.sense_allowed[i] && dev->caps.default_sense[i]) {
mlx4_core: Changing link sensing logic New FW can give clues to driver regarding default port type and whether or not we should default to link sensing on the port. 2 bits are added to QUERY_PORT command: 1. suggested_type: This bit gives a hint whether the default port type should be IB or Ethernet. The driver will use this hint in case the user didn't specify explicitly the link layer type he wants to set. 2. default_sense: If this bit is set, we would sense the port type on start-up and default the port to link sensing Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 11:00:34 +07:00
enum mlx4_port_type sensed_port = MLX4_PORT_TYPE_NONE;
dev->caps.possible_type[i] = MLX4_PORT_TYPE_AUTO;
mlx4_SENSE_PORT(dev, i, &sensed_port);
if (sensed_port != MLX4_PORT_TYPE_NONE)
dev->caps.port_type[i] = sensed_port;
} else {
dev->caps.possible_type[i] = dev->caps.port_type[i];
}
net/mlx4: Refactor QUERY_PORT Currently QUERY_PORT is done as a part of QUERY_DEV_CAP firmware command. Since we would like to use it without querying all device capabilities, extract this part to be a function of its own. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:57:59 +07:00
if (dev->caps.log_num_macs > dev_cap->port_cap[i].log_max_macs) {
dev->caps.log_num_macs = dev_cap->port_cap[i].log_max_macs;
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_warn(dev, "Requested number of MACs is too much for port %d, reducing to %d\n",
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
i, 1 << dev->caps.log_num_macs);
}
net/mlx4: Refactor QUERY_PORT Currently QUERY_PORT is done as a part of QUERY_DEV_CAP firmware command. Since we would like to use it without querying all device capabilities, extract this part to be a function of its own. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:57:59 +07:00
if (dev->caps.log_num_vlans > dev_cap->port_cap[i].log_max_vlans) {
dev->caps.log_num_vlans = dev_cap->port_cap[i].log_max_vlans;
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_warn(dev, "Requested number of VLANs is too much for port %d, reducing to %d\n",
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
i, 1 << dev->caps.log_num_vlans);
}
}
net/mlx4_core: Disable Granular QoS per VF under IB/Eth VPI configuration Due to firmware bug, under VPI configuration when port1 = IB and port2 = Eth, Granular QoS per VF isn't working properly. More over, the whole QP0/QP1 Para-Virtualization in the mlx4 IB driver is broken on that config. Hence, we must disable Granular QoS per VF under that configuration till a fix is introduced. Once that happens, a new device capability will be used to mark the feature support on that specific configuration. Reported-by: Doug Ledford <dledford@redhat.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-14 21:13:06 +07:00
if (mlx4_is_master(dev) && (dev->caps.num_ports == 2) &&
(port_type_array[0] == MLX4_PORT_TYPE_IB) &&
(port_type_array[1] == MLX4_PORT_TYPE_ETH)) {
mlx4_warn(dev,
"Granular QoS per VF not supported with IB/Eth configuration\n");
dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_QOS_VPP;
}
net/mlx4_core: Add sink counter Reserve the last valid counter index for "sink" counter, when a new counter cannot be allocated, the driver will use this counter. In order to avoid allocating this counter on any other flow, fix the indices bitmap allocation range, and reserve the sink counter index. Add macro for the sink counter index and replace all appearences of the index with the macro. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:58:58 +07:00
dev->caps.max_counters = dev_cap->max_counters;
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW] = dev_cap->reserved_qps;
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_ETH_ADDR] =
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FC_ADDR] =
(1 << dev->caps.log_num_macs) *
(1 << dev->caps.log_num_vlans) *
dev->caps.num_ports;
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FC_EXCH] = MLX4_NUM_FEXCH;
net/mlx4: Add support for A0 steering Add the required firmware commands for A0 steering and a way to enable that. The firmware support focuses on INIT_HCA, QUERY_HCA, QUERY_PORT, QUERY_DEV_CAP and QUERY_FUNC_CAP commands. Those commands are used to configure and query the device. The different A0 DMFS (steering) modes are: Static - optimized performance, but flow steering rules are limited. This mode should be choosed explicitly by the user in order to be used. Dynamic - this mode should be explicitly choosed by the user. In this mode, the FW works in optimized steering mode as long as it can and afterwards automatically drops to classic (full) DMFS. Disable - this mode should be explicitly choosed by the user. The user instructs the system not to use optimized steering, even if the FW supports Dynamic A0 DMFS (and thus will be able to use optimized steering in Default A0 DMFS mode). Default - this mode is implicitly choosed. In this mode, if the FW supports Dynamic A0 DMFS, it'll work in this mode. Otherwise, it'll work at Disable A0 DMFS mode. Under SRIOV configuration, when the A0 steering mode is enabled, older guest VF drivers who aren't using the RX QP allocation flag (MLX4_RESERVE_A0_QP) will get a QP from the general range and fail when attempting to register a steering rule. To avoid that, the PF context behaviour is changed once on A0 static mode, to require support for the allocation flag in VF drivers too. In order to enable A0 steering, we use log_num_mgm_entry_size param. If the value of the parameter is not positive, we treat the absolute value of log_num_mgm_entry_size as a bit field. Setting bit 2 of this bit field enables static A0 steering. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:58:00 +07:00
if (dev_cap->dmfs_high_rate_qpn_base > 0 &&
dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_FS_EN)
dev->caps.dmfs_high_rate_qpn_base = dev_cap->dmfs_high_rate_qpn_base;
else
dev->caps.dmfs_high_rate_qpn_base =
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW];
if (dev_cap->dmfs_high_rate_qpn_range > 0 &&
dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_FS_EN) {
dev->caps.dmfs_high_rate_qpn_range = dev_cap->dmfs_high_rate_qpn_range;
dev->caps.dmfs_high_steer_mode = MLX4_STEERING_DMFS_A0_DEFAULT;
dev->caps.flags2 |= MLX4_DEV_CAP_FLAG2_FS_A0;
} else {
dev->caps.dmfs_high_steer_mode = MLX4_STEERING_DMFS_A0_NOT_SUPPORTED;
dev->caps.dmfs_high_rate_qpn_base =
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW];
dev->caps.dmfs_high_rate_qpn_range = MLX4_A0_STEERING_TABLE_SIZE;
}
net/mlx4_core: Add basic support for QP max-rate limiting Add the low-level device commands and definitions used for QP max-rate limiting. This is done through the following elements: - read rate-limit device caps in QUERY_DEV_CAP: number of different rates and the min/max rates in Kbs/Mbs/Gbs units - enhance the QP context struct to contain rate limit units and value - allow to do run time rate-limit setting to QPs through the update-qp firmware command - QP rate-limiting is disallowed for VFs Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-03-18 19:57:34 +07:00
dev->caps.rl_caps = dev_cap->rl_caps;
net/mlx4: Add A0 hybrid steering A0 hybrid steering is a form of high performance flow steering. By using this mode, mlx4 cards use a fast limited table based steering, in order to enable fast steering of unicast packets to a QP. In order to implement A0 hybrid steering we allocate resources from different zones: (1) General range (2) Special MAC-assigned QPs [RSS, Raw-Ethernet] each has its own region. When we create a rss QP or a raw ethernet (A0 steerable and BF ready) QP, we try hard to allocate the QP from range (2). Otherwise, we try hard not to allocate from this range. However, when the system is pushed to its limits and one needs every resource, the allocator uses every region it can. Meaning, when we run out of raw-eth qps, the allocator allocates from the general range (and the special-A0 area is no longer active). If we run out of RSS qps, the mechanism tries to allocate from the raw-eth QP zone. If that is also exhausted, the allocator will allocate from the general range (and the A0 region is no longer active). Note that if a raw-eth qp is allocated from the general range, it attempts to allocate the range such that bits 6 and 7 (blueflame bits) in the QP number are not set. When the feature is used in SRIOV, the VF has to notify the PF what kind of QP attributes it needs. In order to do that, along with the "Eth QP blueflame" bit, we reserve a new "A0 steerable QP". According to the combination of these bits, the PF tries to allocate a suitable QP. In order to maintain backward compatibility (with older PFs), the PF notifies which QP attributes it supports via QUERY_FUNC_CAP command. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:57:57 +07:00
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_RSS_RAW_ETH] =
net/mlx4: Add support for A0 steering Add the required firmware commands for A0 steering and a way to enable that. The firmware support focuses on INIT_HCA, QUERY_HCA, QUERY_PORT, QUERY_DEV_CAP and QUERY_FUNC_CAP commands. Those commands are used to configure and query the device. The different A0 DMFS (steering) modes are: Static - optimized performance, but flow steering rules are limited. This mode should be choosed explicitly by the user in order to be used. Dynamic - this mode should be explicitly choosed by the user. In this mode, the FW works in optimized steering mode as long as it can and afterwards automatically drops to classic (full) DMFS. Disable - this mode should be explicitly choosed by the user. The user instructs the system not to use optimized steering, even if the FW supports Dynamic A0 DMFS (and thus will be able to use optimized steering in Default A0 DMFS mode). Default - this mode is implicitly choosed. In this mode, if the FW supports Dynamic A0 DMFS, it'll work in this mode. Otherwise, it'll work at Disable A0 DMFS mode. Under SRIOV configuration, when the A0 steering mode is enabled, older guest VF drivers who aren't using the RX QP allocation flag (MLX4_RESERVE_A0_QP) will get a QP from the general range and fail when attempting to register a steering rule. To avoid that, the PF context behaviour is changed once on A0 static mode, to require support for the allocation flag in VF drivers too. In order to enable A0 steering, we use log_num_mgm_entry_size param. If the value of the parameter is not positive, we treat the absolute value of log_num_mgm_entry_size as a bit field. Setting bit 2 of this bit field enables static A0 steering. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:58:00 +07:00
dev->caps.dmfs_high_rate_qpn_range;
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
dev->caps.reserved_qps = dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW] +
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_ETH_ADDR] +
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FC_ADDR] +
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FC_EXCH];
mlx4_core: Add proxy and tunnel QPs to the reserved QP area In addition, pass the proxy and tunnel QP numbers to slaves so the driver can perform special QP paravirtualization. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:41 +07:00
dev->caps.sqp_demux = (mlx4_is_master(dev)) ? MLX4_MAX_NUM_SLAVES : 0;
mlx4: 64-byte CQE/EQE support ConnectX-3 devices can use either 64- or 32-byte completion queue entries (CQEs) and event queue entries (EQEs). Using 64-byte EQEs/CQEs performs better because each entry is aligned to a complete cacheline. This patch queries the HCA's capabilities, and if it supports 64-byte CQEs and EQES the driver will configure the HW to work in 64-byte mode. The 32-byte vs 64-byte mode is global per HCA and not per CQ or EQ. Since this mode is global, userspace (libmlx4) must be updated to work with the configured CQE size, and guests using SR-IOV virtual functions need to know both EQE and CQE size. In case one of the 64-byte CQE/EQE capabilities is activated, the patch makes sure that older guest drivers that use the QUERY_DEV_FUNC command (e.g as done in mlx4_core of Linux 3.3..3.6) will notice that they need an update to be able to work with the PPF. This is done by changing the returned pf_context_behaviour not to be zero any more. In case none of these capabilities is activated that value remains zero and older guest drivers can run OK. The SRIOV related flow is as follows 1. the PPF does the detection of the new capabilities using QUERY_DEV_CAP command. 2. the PPF activates the new capabilities using INIT_HCA. 3. the VF detects if the PPF activated the capabilities using QUERY_HCA, and if this is the case activates them for itself too. Note that the VF detects that it must be aware to the new PF behaviour using QUERY_FUNC_CAP. Steps 1 and 2 apply also for native mode. User space notification is done through a new field introduced in struct mlx4_ib_ucontext which holds device capabilities for which user space must take action. This changes the binary interface so the ABI towards libmlx4 exposed through uverbs is bumped from 3 to 4 but only when **needed** i.e. only when the driver does use 64-byte CQEs or future device capabilities which must be in sync by user space. This practice allows to work with unmodified libmlx4 on older devices (e.g A0, B0) which don't support 64-byte CQEs. In order to keep existing systems functional when they update to a newer kernel that contains these changes in VF and userspace ABI, a module parameter enable_64b_cqe_eqe must be set to enable 64-byte mode; the default is currently false. Signed-off-by: Eli Cohen <eli@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-10-21 21:59:24 +07:00
net/mlx4_core: VFs must ignore the enable_64b_cqe_eqe module param Slaves get the 64B CQE/EQE state from QUERY_HCA, not from the module parameter. If the parameter is set to zero, the slave outputs an incorrect/irrelevant warning message that 64B CQEs/EQEs are supported but not enabled (even if the hypervisor has enabled 64B CQEs/EQEs). Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-08-01 23:55:01 +07:00
if (!enable_64b_cqe_eqe && !mlx4_is_slave(dev)) {
mlx4: 64-byte CQE/EQE support ConnectX-3 devices can use either 64- or 32-byte completion queue entries (CQEs) and event queue entries (EQEs). Using 64-byte EQEs/CQEs performs better because each entry is aligned to a complete cacheline. This patch queries the HCA's capabilities, and if it supports 64-byte CQEs and EQES the driver will configure the HW to work in 64-byte mode. The 32-byte vs 64-byte mode is global per HCA and not per CQ or EQ. Since this mode is global, userspace (libmlx4) must be updated to work with the configured CQE size, and guests using SR-IOV virtual functions need to know both EQE and CQE size. In case one of the 64-byte CQE/EQE capabilities is activated, the patch makes sure that older guest drivers that use the QUERY_DEV_FUNC command (e.g as done in mlx4_core of Linux 3.3..3.6) will notice that they need an update to be able to work with the PPF. This is done by changing the returned pf_context_behaviour not to be zero any more. In case none of these capabilities is activated that value remains zero and older guest drivers can run OK. The SRIOV related flow is as follows 1. the PPF does the detection of the new capabilities using QUERY_DEV_CAP command. 2. the PPF activates the new capabilities using INIT_HCA. 3. the VF detects if the PPF activated the capabilities using QUERY_HCA, and if this is the case activates them for itself too. Note that the VF detects that it must be aware to the new PF behaviour using QUERY_FUNC_CAP. Steps 1 and 2 apply also for native mode. User space notification is done through a new field introduced in struct mlx4_ib_ucontext which holds device capabilities for which user space must take action. This changes the binary interface so the ABI towards libmlx4 exposed through uverbs is bumped from 3 to 4 but only when **needed** i.e. only when the driver does use 64-byte CQEs or future device capabilities which must be in sync by user space. This practice allows to work with unmodified libmlx4 on older devices (e.g A0, B0) which don't support 64-byte CQEs. In order to keep existing systems functional when they update to a newer kernel that contains these changes in VF and userspace ABI, a module parameter enable_64b_cqe_eqe must be set to enable 64-byte mode; the default is currently false. Signed-off-by: Eli Cohen <eli@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-10-21 21:59:24 +07:00
if (dev_cap->flags &
(MLX4_DEV_CAP_FLAG_64B_CQE | MLX4_DEV_CAP_FLAG_64B_EQE)) {
mlx4_warn(dev, "64B EQEs/CQEs supported by the device but not enabled\n");
dev->caps.flags &= ~MLX4_DEV_CAP_FLAG_64B_CQE;
dev->caps.flags &= ~MLX4_DEV_CAP_FLAG_64B_EQE;
}
net/mlx4_core: Enable CQE/EQE stride support This feature is intended for archs having cache line larger then 64B. Since our CQE/EQEs are generally 64B in those systems, HW will write twice to the same cache line consecutively, causing pipe locks due to he hazard prevention mechanism. For elements in a cyclic buffer, writes are consecutive, so entries smaller than a cache line should be avoided, especially if they are written at a high rate. Reduce consecutive writes to same cache line in CQs/EQs, by allowing the driver to increase the distance between entries so that each will reside in a different cache line. Until the introduction of this feature, there were two types of CQE/EQE: 1. 32B stride and context in the [0-31] segment 2. 64B stride and context in the [32-63] segment This feature introduces two additional types: 3. 128B stride and context in the [0-31] segment (128B cache line) 4. 256B stride and context in the [0-31] segment (256B cache line) Modify the mlx4_core driver to query the device for the CQE/EQE cache line stride capability and to enable that capability when the host cache line size is larger than 64 bytes (supported cache lines are 128B and 256B). The mlx4 IB driver and libmlx4 need not be aware of this change. The PF context behaviour is changed to require this change in VF drivers running on such archs. Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-18 15:50:59 +07:00
if (dev_cap->flags2 &
(MLX4_DEV_CAP_FLAG2_CQE_STRIDE |
MLX4_DEV_CAP_FLAG2_EQE_STRIDE)) {
mlx4_warn(dev, "Disabling EQE/CQE stride per user request\n");
dev_cap->flags2 &= ~MLX4_DEV_CAP_FLAG2_CQE_STRIDE;
dev_cap->flags2 &= ~MLX4_DEV_CAP_FLAG2_EQE_STRIDE;
}
mlx4: 64-byte CQE/EQE support ConnectX-3 devices can use either 64- or 32-byte completion queue entries (CQEs) and event queue entries (EQEs). Using 64-byte EQEs/CQEs performs better because each entry is aligned to a complete cacheline. This patch queries the HCA's capabilities, and if it supports 64-byte CQEs and EQES the driver will configure the HW to work in 64-byte mode. The 32-byte vs 64-byte mode is global per HCA and not per CQ or EQ. Since this mode is global, userspace (libmlx4) must be updated to work with the configured CQE size, and guests using SR-IOV virtual functions need to know both EQE and CQE size. In case one of the 64-byte CQE/EQE capabilities is activated, the patch makes sure that older guest drivers that use the QUERY_DEV_FUNC command (e.g as done in mlx4_core of Linux 3.3..3.6) will notice that they need an update to be able to work with the PPF. This is done by changing the returned pf_context_behaviour not to be zero any more. In case none of these capabilities is activated that value remains zero and older guest drivers can run OK. The SRIOV related flow is as follows 1. the PPF does the detection of the new capabilities using QUERY_DEV_CAP command. 2. the PPF activates the new capabilities using INIT_HCA. 3. the VF detects if the PPF activated the capabilities using QUERY_HCA, and if this is the case activates them for itself too. Note that the VF detects that it must be aware to the new PF behaviour using QUERY_FUNC_CAP. Steps 1 and 2 apply also for native mode. User space notification is done through a new field introduced in struct mlx4_ib_ucontext which holds device capabilities for which user space must take action. This changes the binary interface so the ABI towards libmlx4 exposed through uverbs is bumped from 3 to 4 but only when **needed** i.e. only when the driver does use 64-byte CQEs or future device capabilities which must be in sync by user space. This practice allows to work with unmodified libmlx4 on older devices (e.g A0, B0) which don't support 64-byte CQEs. In order to keep existing systems functional when they update to a newer kernel that contains these changes in VF and userspace ABI, a module parameter enable_64b_cqe_eqe must be set to enable 64-byte mode; the default is currently false. Signed-off-by: Eli Cohen <eli@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-10-21 21:59:24 +07:00
}
mlx4_core: Fix advertisement of wrong PF context behaviour Commit 08ff32352d6f ("mlx4: 64-byte CQE/EQE support") introduced a regression where older guest VF drivers failed to load even when 64-byte EQEs/CQEs are disabled, since the PF wrongly advertises the new context behaviour anyway. The failure looks like: mlx4_core 0000:00:07.0: Unknown pf context behaviour mlx4_core 0000:00:07.0: Failed to obtain slave caps mlx4_core: probe of 0000:00:07.0 failed with error -38 Fix this by basing this advertisement on dev->caps.flags, which is the operational capabilities used by the QUERY_FUNC_CAP command wrapper (dev_cap->flags holds the firmware capabilities). Reported-by: Alex Williamson <alex.williamson@redhat.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-01-10 22:18:35 +07:00
if ((dev->caps.flags &
mlx4: 64-byte CQE/EQE support ConnectX-3 devices can use either 64- or 32-byte completion queue entries (CQEs) and event queue entries (EQEs). Using 64-byte EQEs/CQEs performs better because each entry is aligned to a complete cacheline. This patch queries the HCA's capabilities, and if it supports 64-byte CQEs and EQES the driver will configure the HW to work in 64-byte mode. The 32-byte vs 64-byte mode is global per HCA and not per CQ or EQ. Since this mode is global, userspace (libmlx4) must be updated to work with the configured CQE size, and guests using SR-IOV virtual functions need to know both EQE and CQE size. In case one of the 64-byte CQE/EQE capabilities is activated, the patch makes sure that older guest drivers that use the QUERY_DEV_FUNC command (e.g as done in mlx4_core of Linux 3.3..3.6) will notice that they need an update to be able to work with the PPF. This is done by changing the returned pf_context_behaviour not to be zero any more. In case none of these capabilities is activated that value remains zero and older guest drivers can run OK. The SRIOV related flow is as follows 1. the PPF does the detection of the new capabilities using QUERY_DEV_CAP command. 2. the PPF activates the new capabilities using INIT_HCA. 3. the VF detects if the PPF activated the capabilities using QUERY_HCA, and if this is the case activates them for itself too. Note that the VF detects that it must be aware to the new PF behaviour using QUERY_FUNC_CAP. Steps 1 and 2 apply also for native mode. User space notification is done through a new field introduced in struct mlx4_ib_ucontext which holds device capabilities for which user space must take action. This changes the binary interface so the ABI towards libmlx4 exposed through uverbs is bumped from 3 to 4 but only when **needed** i.e. only when the driver does use 64-byte CQEs or future device capabilities which must be in sync by user space. This practice allows to work with unmodified libmlx4 on older devices (e.g A0, B0) which don't support 64-byte CQEs. In order to keep existing systems functional when they update to a newer kernel that contains these changes in VF and userspace ABI, a module parameter enable_64b_cqe_eqe must be set to enable 64-byte mode; the default is currently false. Signed-off-by: Eli Cohen <eli@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-10-21 21:59:24 +07:00
(MLX4_DEV_CAP_FLAG_64B_CQE | MLX4_DEV_CAP_FLAG_64B_EQE)) &&
mlx4_is_master(dev))
dev->caps.function_caps |= MLX4_FUNC_CAP_64B_EQE_CQE;
net/mlx4: Change QP allocation scheme When using BF (Blue-Flame), the QPN overrides the VLAN, CV, and SV fields in the WQE. Thus, BF may only be used for QPNs with bits 6,7 unset. The current Ethernet driver code reserves a Tx QP range with 256b alignment. This is wrong because if there are more than 64 Tx QPs in use, QPNs >= base + 65 will have bits 6/7 set. This problem is not specific for the Ethernet driver, any entity that tries to reserve more than 64 BF-enabled QPs should fail. Also, using ranges is not necessary here and is wasteful. The new mechanism introduced here will support reservation for "Eth QPs eligible for BF" for all drivers: bare-metal, multi-PF, and VFs (when hypervisors support WC in VMs). The flow we use is: 1. In mlx4_en, allocate Tx QPs one by one instead of a range allocation, and request "BF enabled QPs" if BF is supported for the function 2. In the ALLOC_RES FW command, change param1 to: a. param1[23:0] - number of QPs b. param1[31-24] - flags controlling QPs reservation Bit 31 refers to Eth blueflame supported QPs. Those QPs must have bits 6 and 7 unset in order to be used in Ethernet. Bits 24-30 of the flags are currently reserved. When a function tries to allocate a QP, it states the required attributes for this QP. Those attributes are considered "best-effort". If an attribute, such as Ethernet BF enabled QP, is a must-have attribute, the function has to check that attribute is supported before trying to do the allocation. In a lower layer of the code, mlx4_qp_reserve_range masks out the bits which are unsupported. If SRIOV is used, the PF validates those attributes and masks out unsupported attributes as well. In order to notify VFs which attributes are supported, the VF uses QUERY_FUNC_CAP command. This command's mailbox is filled by the PF, which notifies which QP allocation attributes it supports. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:57:54 +07:00
if (!mlx4_is_slave(dev)) {
net/mlx4_core: Enable CQE/EQE stride support This feature is intended for archs having cache line larger then 64B. Since our CQE/EQEs are generally 64B in those systems, HW will write twice to the same cache line consecutively, causing pipe locks due to he hazard prevention mechanism. For elements in a cyclic buffer, writes are consecutive, so entries smaller than a cache line should be avoided, especially if they are written at a high rate. Reduce consecutive writes to same cache line in CQs/EQs, by allowing the driver to increase the distance between entries so that each will reside in a different cache line. Until the introduction of this feature, there were two types of CQE/EQE: 1. 32B stride and context in the [0-31] segment 2. 64B stride and context in the [32-63] segment This feature introduces two additional types: 3. 128B stride and context in the [0-31] segment (128B cache line) 4. 256B stride and context in the [0-31] segment (256B cache line) Modify the mlx4_core driver to query the device for the CQE/EQE cache line stride capability and to enable that capability when the host cache line size is larger than 64 bytes (supported cache lines are 128B and 256B). The mlx4 IB driver and libmlx4 need not be aware of this change. The PF context behaviour is changed to require this change in VF drivers running on such archs. Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-18 15:50:59 +07:00
mlx4_enable_cqe_eqe_stride(dev);
net/mlx4: Change QP allocation scheme When using BF (Blue-Flame), the QPN overrides the VLAN, CV, and SV fields in the WQE. Thus, BF may only be used for QPNs with bits 6,7 unset. The current Ethernet driver code reserves a Tx QP range with 256b alignment. This is wrong because if there are more than 64 Tx QPs in use, QPNs >= base + 65 will have bits 6/7 set. This problem is not specific for the Ethernet driver, any entity that tries to reserve more than 64 BF-enabled QPs should fail. Also, using ranges is not necessary here and is wasteful. The new mechanism introduced here will support reservation for "Eth QPs eligible for BF" for all drivers: bare-metal, multi-PF, and VFs (when hypervisors support WC in VMs). The flow we use is: 1. In mlx4_en, allocate Tx QPs one by one instead of a range allocation, and request "BF enabled QPs" if BF is supported for the function 2. In the ALLOC_RES FW command, change param1 to: a. param1[23:0] - number of QPs b. param1[31-24] - flags controlling QPs reservation Bit 31 refers to Eth blueflame supported QPs. Those QPs must have bits 6 and 7 unset in order to be used in Ethernet. Bits 24-30 of the flags are currently reserved. When a function tries to allocate a QP, it states the required attributes for this QP. Those attributes are considered "best-effort". If an attribute, such as Ethernet BF enabled QP, is a must-have attribute, the function has to check that attribute is supported before trying to do the allocation. In a lower layer of the code, mlx4_qp_reserve_range masks out the bits which are unsupported. If SRIOV is used, the PF validates those attributes and masks out unsupported attributes as well. In order to notify VFs which attributes are supported, the VF uses QUERY_FUNC_CAP command. This command's mailbox is filled by the PF, which notifies which QP allocation attributes it supports. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:57:54 +07:00
dev->caps.alloc_res_qp_mask =
net/mlx4: Add A0 hybrid steering A0 hybrid steering is a form of high performance flow steering. By using this mode, mlx4 cards use a fast limited table based steering, in order to enable fast steering of unicast packets to a QP. In order to implement A0 hybrid steering we allocate resources from different zones: (1) General range (2) Special MAC-assigned QPs [RSS, Raw-Ethernet] each has its own region. When we create a rss QP or a raw ethernet (A0 steerable and BF ready) QP, we try hard to allocate the QP from range (2). Otherwise, we try hard not to allocate from this range. However, when the system is pushed to its limits and one needs every resource, the allocator uses every region it can. Meaning, when we run out of raw-eth qps, the allocator allocates from the general range (and the special-A0 area is no longer active). If we run out of RSS qps, the mechanism tries to allocate from the raw-eth QP zone. If that is also exhausted, the allocator will allocate from the general range (and the A0 region is no longer active). Note that if a raw-eth qp is allocated from the general range, it attempts to allocate the range such that bits 6 and 7 (blueflame bits) in the QP number are not set. When the feature is used in SRIOV, the VF has to notify the PF what kind of QP attributes it needs. In order to do that, along with the "Eth QP blueflame" bit, we reserve a new "A0 steerable QP". According to the combination of these bits, the PF tries to allocate a suitable QP. In order to maintain backward compatibility (with older PFs), the PF notifies which QP attributes it supports via QUERY_FUNC_CAP command. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:57:57 +07:00
(dev->caps.bf_reg_size ? MLX4_RESERVE_ETH_BF_QP : 0) |
MLX4_RESERVE_A0_QP;
net/mlx4: Warn users of depracated QoS Firmware A new capability bit was introduced in the past to to differ devices using the QoS ETS feature. The old was deprecated since then. If driver sees device which set only the old capabilty, it will print warning to user suggesting to upgrade the FW. Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-02 20:31:17 +07:00
if (!(dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ETS_CFG) &&
dev->caps.flags & MLX4_DEV_CAP_FLAG_SET_ETH_SCHED) {
mlx4_warn(dev, "Old device ETS support detected\n");
mlx4_warn(dev, "Consider upgrading device FW.\n");
dev->caps.flags2 |= MLX4_DEV_CAP_FLAG2_ETS_CFG;
}
net/mlx4: Change QP allocation scheme When using BF (Blue-Flame), the QPN overrides the VLAN, CV, and SV fields in the WQE. Thus, BF may only be used for QPNs with bits 6,7 unset. The current Ethernet driver code reserves a Tx QP range with 256b alignment. This is wrong because if there are more than 64 Tx QPs in use, QPNs >= base + 65 will have bits 6/7 set. This problem is not specific for the Ethernet driver, any entity that tries to reserve more than 64 BF-enabled QPs should fail. Also, using ranges is not necessary here and is wasteful. The new mechanism introduced here will support reservation for "Eth QPs eligible for BF" for all drivers: bare-metal, multi-PF, and VFs (when hypervisors support WC in VMs). The flow we use is: 1. In mlx4_en, allocate Tx QPs one by one instead of a range allocation, and request "BF enabled QPs" if BF is supported for the function 2. In the ALLOC_RES FW command, change param1 to: a. param1[23:0] - number of QPs b. param1[31-24] - flags controlling QPs reservation Bit 31 refers to Eth blueflame supported QPs. Those QPs must have bits 6 and 7 unset in order to be used in Ethernet. Bits 24-30 of the flags are currently reserved. When a function tries to allocate a QP, it states the required attributes for this QP. Those attributes are considered "best-effort". If an attribute, such as Ethernet BF enabled QP, is a must-have attribute, the function has to check that attribute is supported before trying to do the allocation. In a lower layer of the code, mlx4_qp_reserve_range masks out the bits which are unsupported. If SRIOV is used, the PF validates those attributes and masks out unsupported attributes as well. In order to notify VFs which attributes are supported, the VF uses QUERY_FUNC_CAP command. This command's mailbox is filled by the PF, which notifies which QP allocation attributes it supports. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:57:54 +07:00
} else {
dev->caps.alloc_res_qp_mask = 0;
}
net/mlx4_core: Enable CQE/EQE stride support This feature is intended for archs having cache line larger then 64B. Since our CQE/EQEs are generally 64B in those systems, HW will write twice to the same cache line consecutively, causing pipe locks due to he hazard prevention mechanism. For elements in a cyclic buffer, writes are consecutive, so entries smaller than a cache line should be avoided, especially if they are written at a high rate. Reduce consecutive writes to same cache line in CQs/EQs, by allowing the driver to increase the distance between entries so that each will reside in a different cache line. Until the introduction of this feature, there were two types of CQE/EQE: 1. 32B stride and context in the [0-31] segment 2. 64B stride and context in the [32-63] segment This feature introduces two additional types: 3. 128B stride and context in the [0-31] segment (128B cache line) 4. 256B stride and context in the [0-31] segment (256B cache line) Modify the mlx4_core driver to query the device for the CQE/EQE cache line stride capability and to enable that capability when the host cache line size is larger than 64 bytes (supported cache lines are 128B and 256B). The mlx4 IB driver and libmlx4 need not be aware of this change. The PF context behaviour is changed to require this change in VF drivers running on such archs. Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-18 15:50:59 +07:00
net/mlx4_en: Add RX-ALL support Enabled when the device supports KEEP FCS and IGNORE FCS. When the flag is set, pass all received frames up the stack, even ones with invalid FCS, controlled by ethtool. Signed-off-by: Muhammad Mahajna <muhammadm@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-02 20:31:22 +07:00
mlx4_enable_ignore_fcs(dev);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return 0;
}
net/mlx4_core: Warn if device doesn't have enough PCI bandwidth Check if the device get enough bandwidth from the entire PCI chain to satisfy its capabilities. This patch determines the PCIe device's bandwidth capabilities by reading its PCIe Link Capabilities registers and then call the pcie_get_minimum_link function to ensure that the adapter is hooked into a slot which is capable of providing the necessary bandwidth capabilities. Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-05 22:41:08 +07:00
static int mlx4_get_pcie_dev_link_caps(struct mlx4_dev *dev,
enum pci_bus_speed *speed,
enum pcie_link_width *width)
{
u32 lnkcap1, lnkcap2;
int err1, err2;
#define PCIE_MLW_CAP_SHIFT 4 /* start of MLW mask in link capabilities */
*speed = PCI_SPEED_UNKNOWN;
*width = PCIE_LNK_WIDTH_UNKNOWN;
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
err1 = pcie_capability_read_dword(dev->persist->pdev, PCI_EXP_LNKCAP,
&lnkcap1);
err2 = pcie_capability_read_dword(dev->persist->pdev, PCI_EXP_LNKCAP2,
&lnkcap2);
net/mlx4_core: Warn if device doesn't have enough PCI bandwidth Check if the device get enough bandwidth from the entire PCI chain to satisfy its capabilities. This patch determines the PCIe device's bandwidth capabilities by reading its PCIe Link Capabilities registers and then call the pcie_get_minimum_link function to ensure that the adapter is hooked into a slot which is capable of providing the necessary bandwidth capabilities. Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-05 22:41:08 +07:00
if (!err2 && lnkcap2) { /* PCIe r3.0-compliant */
if (lnkcap2 & PCI_EXP_LNKCAP2_SLS_8_0GB)
*speed = PCIE_SPEED_8_0GT;
else if (lnkcap2 & PCI_EXP_LNKCAP2_SLS_5_0GB)
*speed = PCIE_SPEED_5_0GT;
else if (lnkcap2 & PCI_EXP_LNKCAP2_SLS_2_5GB)
*speed = PCIE_SPEED_2_5GT;
}
if (!err1) {
*width = (lnkcap1 & PCI_EXP_LNKCAP_MLW) >> PCIE_MLW_CAP_SHIFT;
if (!lnkcap2) { /* pre-r3.0 */
if (lnkcap1 & PCI_EXP_LNKCAP_SLS_5_0GB)
*speed = PCIE_SPEED_5_0GT;
else if (lnkcap1 & PCI_EXP_LNKCAP_SLS_2_5GB)
*speed = PCIE_SPEED_2_5GT;
}
}
if (*speed == PCI_SPEED_UNKNOWN || *width == PCIE_LNK_WIDTH_UNKNOWN) {
return err1 ? err1 :
err2 ? err2 : -EINVAL;
}
return 0;
}
static void mlx4_check_pcie_caps(struct mlx4_dev *dev)
{
enum pcie_link_width width, width_cap;
enum pci_bus_speed speed, speed_cap;
int err;
#define PCIE_SPEED_STR(speed) \
(speed == PCIE_SPEED_8_0GT ? "8.0GT/s" : \
speed == PCIE_SPEED_5_0GT ? "5.0GT/s" : \
speed == PCIE_SPEED_2_5GT ? "2.5GT/s" : \
"Unknown")
err = mlx4_get_pcie_dev_link_caps(dev, &speed_cap, &width_cap);
if (err) {
mlx4_warn(dev,
"Unable to determine PCIe device BW capabilities\n");
return;
}
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
err = pcie_get_minimum_link(dev->persist->pdev, &speed, &width);
net/mlx4_core: Warn if device doesn't have enough PCI bandwidth Check if the device get enough bandwidth from the entire PCI chain to satisfy its capabilities. This patch determines the PCIe device's bandwidth capabilities by reading its PCIe Link Capabilities registers and then call the pcie_get_minimum_link function to ensure that the adapter is hooked into a slot which is capable of providing the necessary bandwidth capabilities. Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-05 22:41:08 +07:00
if (err || speed == PCI_SPEED_UNKNOWN ||
width == PCIE_LNK_WIDTH_UNKNOWN) {
mlx4_warn(dev,
"Unable to determine PCI device chain minimum BW\n");
return;
}
if (width != width_cap || speed != speed_cap)
mlx4_warn(dev,
"PCIe BW is different than device's capability\n");
mlx4_info(dev, "PCIe link speed is %s, device supports %s\n",
PCIE_SPEED_STR(speed), PCIE_SPEED_STR(speed_cap));
mlx4_info(dev, "PCIe link width is x%d, device supports x%d\n",
width, width_cap);
return;
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/*The function checks if there are live vf, return the num of them*/
static int mlx4_how_many_lives_vf(struct mlx4_dev *dev)
{
struct mlx4_priv *priv = mlx4_priv(dev);
struct mlx4_slave_state *s_state;
int i;
int ret = 0;
for (i = 1/*the ppf is 0*/; i < dev->num_slaves; ++i) {
s_state = &priv->mfunc.master.slave_state[i];
if (s_state->active && s_state->last_cmd !=
MLX4_COMM_CMD_RESET) {
mlx4_warn(dev, "%s: slave: %d is still active\n",
__func__, i);
ret++;
}
}
return ret;
}
mlx4_core: Implement mechanism for reserved Q_Keys The SR-IOV special QP tunneling mechanism uses proxy special QPs (instead of the real special QPs) for MADs on guests. These proxy QPs send their packets to a "tunnel" QP owned by the master. The master then forwards the MAD (after any required paravirtualization) to the real special QP, which sends out the MAD. For security reasons (i.e., to prevent guests from sending MADs to tunnel QPs belonging to other guests), each proxy-tunnel QP pair is assigned a unique, reserved, Q_Key. These Q_Keys are available only for proxy and tunnel QPs -- if the guest tries to use these Q_Keys with other QPs, it will fail. This patch introduces a mechanism for reserving a block of 64K Q_Keys for proxy/tunneling use. The patch introduces also two new fields into mlx4_dev: base_sqpn and base_tunnel_sqpn. In SR-IOV mode, the QP numbers for the "real," proxy, and tunnel sqps are added to the reserved QPN area (so that they will not change). There are 8 special QPs per port in the HCA, and each of them is assigned both a proxy and a tunnel QP, for each VF and for the PF as well in SR-IOV mode. The QPNs for these QPs are arranged as follows: 1. The real SQP numbers (8) 2. The proxy SQPs (8 * (max number of VFs + max number of PFs) 3. The tunnel SQPs (8 * (max number of VFs + max number of PFs) To support these QPs, two new fields are added to struct mlx4_dev: base_sqp: this is the QP number of the first of the real SQPs base_tunnel_sqp: this is the qp number of the first qp in the tunnel sqp region. (On guests, this is the first tunnel sqp of the 8 which are assigned to that guest). In addition, in SR-IOV mode, sqp_start is the number of the first proxy SQP in the proxy SQP region. (In guests, this is the first proxy SQP of the 8 which are assigned to that guest) Note that in non-SR-IOV mode, there are no proxies and no tunnels. In this case, sqp_start is set to sqp_base -- which minimizes code changes. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:42 +07:00
int mlx4_get_parav_qkey(struct mlx4_dev *dev, u32 qpn, u32 *qkey)
{
u32 qk = MLX4_RESERVED_QKEY_BASE;
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
if (qpn >= dev->phys_caps.base_tunnel_sqpn + 8 * MLX4_MFUNC_MAX ||
qpn < dev->phys_caps.base_proxy_sqpn)
mlx4_core: Implement mechanism for reserved Q_Keys The SR-IOV special QP tunneling mechanism uses proxy special QPs (instead of the real special QPs) for MADs on guests. These proxy QPs send their packets to a "tunnel" QP owned by the master. The master then forwards the MAD (after any required paravirtualization) to the real special QP, which sends out the MAD. For security reasons (i.e., to prevent guests from sending MADs to tunnel QPs belonging to other guests), each proxy-tunnel QP pair is assigned a unique, reserved, Q_Key. These Q_Keys are available only for proxy and tunnel QPs -- if the guest tries to use these Q_Keys with other QPs, it will fail. This patch introduces a mechanism for reserving a block of 64K Q_Keys for proxy/tunneling use. The patch introduces also two new fields into mlx4_dev: base_sqpn and base_tunnel_sqpn. In SR-IOV mode, the QP numbers for the "real," proxy, and tunnel sqps are added to the reserved QPN area (so that they will not change). There are 8 special QPs per port in the HCA, and each of them is assigned both a proxy and a tunnel QP, for each VF and for the PF as well in SR-IOV mode. The QPNs for these QPs are arranged as follows: 1. The real SQP numbers (8) 2. The proxy SQPs (8 * (max number of VFs + max number of PFs) 3. The tunnel SQPs (8 * (max number of VFs + max number of PFs) To support these QPs, two new fields are added to struct mlx4_dev: base_sqp: this is the QP number of the first of the real SQPs base_tunnel_sqp: this is the qp number of the first qp in the tunnel sqp region. (On guests, this is the first tunnel sqp of the 8 which are assigned to that guest). In addition, in SR-IOV mode, sqp_start is the number of the first proxy SQP in the proxy SQP region. (In guests, this is the first proxy SQP of the 8 which are assigned to that guest) Note that in non-SR-IOV mode, there are no proxies and no tunnels. In this case, sqp_start is set to sqp_base -- which minimizes code changes. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:42 +07:00
return -EINVAL;
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
if (qpn >= dev->phys_caps.base_tunnel_sqpn)
mlx4_core: Implement mechanism for reserved Q_Keys The SR-IOV special QP tunneling mechanism uses proxy special QPs (instead of the real special QPs) for MADs on guests. These proxy QPs send their packets to a "tunnel" QP owned by the master. The master then forwards the MAD (after any required paravirtualization) to the real special QP, which sends out the MAD. For security reasons (i.e., to prevent guests from sending MADs to tunnel QPs belonging to other guests), each proxy-tunnel QP pair is assigned a unique, reserved, Q_Key. These Q_Keys are available only for proxy and tunnel QPs -- if the guest tries to use these Q_Keys with other QPs, it will fail. This patch introduces a mechanism for reserving a block of 64K Q_Keys for proxy/tunneling use. The patch introduces also two new fields into mlx4_dev: base_sqpn and base_tunnel_sqpn. In SR-IOV mode, the QP numbers for the "real," proxy, and tunnel sqps are added to the reserved QPN area (so that they will not change). There are 8 special QPs per port in the HCA, and each of them is assigned both a proxy and a tunnel QP, for each VF and for the PF as well in SR-IOV mode. The QPNs for these QPs are arranged as follows: 1. The real SQP numbers (8) 2. The proxy SQPs (8 * (max number of VFs + max number of PFs) 3. The tunnel SQPs (8 * (max number of VFs + max number of PFs) To support these QPs, two new fields are added to struct mlx4_dev: base_sqp: this is the QP number of the first of the real SQPs base_tunnel_sqp: this is the qp number of the first qp in the tunnel sqp region. (On guests, this is the first tunnel sqp of the 8 which are assigned to that guest). In addition, in SR-IOV mode, sqp_start is the number of the first proxy SQP in the proxy SQP region. (In guests, this is the first proxy SQP of the 8 which are assigned to that guest) Note that in non-SR-IOV mode, there are no proxies and no tunnels. In this case, sqp_start is set to sqp_base -- which minimizes code changes. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:42 +07:00
/* tunnel qp */
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
qk += qpn - dev->phys_caps.base_tunnel_sqpn;
mlx4_core: Implement mechanism for reserved Q_Keys The SR-IOV special QP tunneling mechanism uses proxy special QPs (instead of the real special QPs) for MADs on guests. These proxy QPs send their packets to a "tunnel" QP owned by the master. The master then forwards the MAD (after any required paravirtualization) to the real special QP, which sends out the MAD. For security reasons (i.e., to prevent guests from sending MADs to tunnel QPs belonging to other guests), each proxy-tunnel QP pair is assigned a unique, reserved, Q_Key. These Q_Keys are available only for proxy and tunnel QPs -- if the guest tries to use these Q_Keys with other QPs, it will fail. This patch introduces a mechanism for reserving a block of 64K Q_Keys for proxy/tunneling use. The patch introduces also two new fields into mlx4_dev: base_sqpn and base_tunnel_sqpn. In SR-IOV mode, the QP numbers for the "real," proxy, and tunnel sqps are added to the reserved QPN area (so that they will not change). There are 8 special QPs per port in the HCA, and each of them is assigned both a proxy and a tunnel QP, for each VF and for the PF as well in SR-IOV mode. The QPNs for these QPs are arranged as follows: 1. The real SQP numbers (8) 2. The proxy SQPs (8 * (max number of VFs + max number of PFs) 3. The tunnel SQPs (8 * (max number of VFs + max number of PFs) To support these QPs, two new fields are added to struct mlx4_dev: base_sqp: this is the QP number of the first of the real SQPs base_tunnel_sqp: this is the qp number of the first qp in the tunnel sqp region. (On guests, this is the first tunnel sqp of the 8 which are assigned to that guest). In addition, in SR-IOV mode, sqp_start is the number of the first proxy SQP in the proxy SQP region. (In guests, this is the first proxy SQP of the 8 which are assigned to that guest) Note that in non-SR-IOV mode, there are no proxies and no tunnels. In this case, sqp_start is set to sqp_base -- which minimizes code changes. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:42 +07:00
else
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
qk += qpn - dev->phys_caps.base_proxy_sqpn;
mlx4_core: Implement mechanism for reserved Q_Keys The SR-IOV special QP tunneling mechanism uses proxy special QPs (instead of the real special QPs) for MADs on guests. These proxy QPs send their packets to a "tunnel" QP owned by the master. The master then forwards the MAD (after any required paravirtualization) to the real special QP, which sends out the MAD. For security reasons (i.e., to prevent guests from sending MADs to tunnel QPs belonging to other guests), each proxy-tunnel QP pair is assigned a unique, reserved, Q_Key. These Q_Keys are available only for proxy and tunnel QPs -- if the guest tries to use these Q_Keys with other QPs, it will fail. This patch introduces a mechanism for reserving a block of 64K Q_Keys for proxy/tunneling use. The patch introduces also two new fields into mlx4_dev: base_sqpn and base_tunnel_sqpn. In SR-IOV mode, the QP numbers for the "real," proxy, and tunnel sqps are added to the reserved QPN area (so that they will not change). There are 8 special QPs per port in the HCA, and each of them is assigned both a proxy and a tunnel QP, for each VF and for the PF as well in SR-IOV mode. The QPNs for these QPs are arranged as follows: 1. The real SQP numbers (8) 2. The proxy SQPs (8 * (max number of VFs + max number of PFs) 3. The tunnel SQPs (8 * (max number of VFs + max number of PFs) To support these QPs, two new fields are added to struct mlx4_dev: base_sqp: this is the QP number of the first of the real SQPs base_tunnel_sqp: this is the qp number of the first qp in the tunnel sqp region. (On guests, this is the first tunnel sqp of the 8 which are assigned to that guest). In addition, in SR-IOV mode, sqp_start is the number of the first proxy SQP in the proxy SQP region. (In guests, this is the first proxy SQP of the 8 which are assigned to that guest) Note that in non-SR-IOV mode, there are no proxies and no tunnels. In this case, sqp_start is set to sqp_base -- which minimizes code changes. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:42 +07:00
*qkey = qk;
return 0;
}
EXPORT_SYMBOL(mlx4_get_parav_qkey);
mlx4: Implement QP paravirtualization and maintain phys_pkey_cache for smp_snoop This requires: 1. Replacing the paravirtualized P_Key index (inserted by the guest) with the real P_Key index. 2. For UD QPs, placing the guest's true source GID index in the address path structure mgid field, and setting the ud_force_mgid bit so that the mgid is taken from the QP context and not from the WQE when posting sends. 3. For UC and RC QPs, placing the guest's true source GID index in the address path structure mgid field. 4. For tunnel and proxy QPs, setting the Q_Key value reserved for that proxy/tunnel pair. Since not all the above adjustments occur in all the QP transitions, the QP transitions require separate wrapper functions. Secondly, initialize the P_Key virtualization table to its default values: Master virtualized table is 1-1 with the real P_Key table, guest virtualized table has P_Key index 0 mapped to the real P_Key index 0, and all the other P_Key indices mapped to the reserved (invalid) P_Key at index 127. Finally, add logic in smp_snoop for maintaining the phys_P_Key_cache. and generating events on the master only if a P_Key actually changed. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:43 +07:00
void mlx4_sync_pkey_table(struct mlx4_dev *dev, int slave, int port, int i, int val)
{
struct mlx4_priv *priv = container_of(dev, struct mlx4_priv, dev);
if (!mlx4_is_master(dev))
return;
priv->virt2phys_pkey[slave][port - 1][i] = val;
}
EXPORT_SYMBOL(mlx4_sync_pkey_table);
mlx4: Paravirtualize Node Guids for slaves This is necessary in order to support > 1 VF/PF in a VM for software that uses the node guid as a discriminator, such as librdmacm. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:56 +07:00
void mlx4_put_slave_node_guid(struct mlx4_dev *dev, int slave, __be64 guid)
{
struct mlx4_priv *priv = container_of(dev, struct mlx4_priv, dev);
if (!mlx4_is_master(dev))
return;
priv->slave_node_guids[slave] = guid;
}
EXPORT_SYMBOL(mlx4_put_slave_node_guid);
__be64 mlx4_get_slave_node_guid(struct mlx4_dev *dev, int slave)
{
struct mlx4_priv *priv = container_of(dev, struct mlx4_priv, dev);
if (!mlx4_is_master(dev))
return 0;
return priv->slave_node_guids[slave];
}
EXPORT_SYMBOL(mlx4_get_slave_node_guid);
mlx4_core: Fix one more static exported function Commit 22c8bff6face ("mlx4_core: Exported functions can't be static") fixed most of this up, but forgot about mlx4_is_slave_active(). Fix this one too. Cc: <stable@vger.kernel.org> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-02-26 16:48:12 +07:00
int mlx4_is_slave_active(struct mlx4_dev *dev, int slave)
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
{
struct mlx4_priv *priv = mlx4_priv(dev);
struct mlx4_slave_state *s_slave;
if (!mlx4_is_master(dev))
return 0;
s_slave = &priv->mfunc.master.slave_state[slave];
return !!s_slave->active;
}
EXPORT_SYMBOL(mlx4_is_slave_active);
mlx4_core: Adjustments to Flow Steering activation logic for SR-IOV Separate flow steering capability detection from the decision to activate. For the master (and for native), detect the flow steering capability in mlx4_dev_cap, but activate the appropriate steering type in a new function choose_flow_steering() based on detected data. For VFs, activate flow steering based on what was actually activated by the master, where that info is obtained via QUERY_HCA. This fixes the current VF detection which is wrongly based on QUERY_DEV_CAP. Also, for SR-IOV mode, if flow steering may be activated, do so only if the max number of QPs per rule is sufficient to satisfy one subscription per VF. If not, fall back to B0 mode. This is needed to serve registrations done by L2 network drivers such as mlx4_en and IPoIB when the network stack attempts to join to multicast groups such as all-hosts or the IPoIB broadcast group. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:11:59 +07:00
static void slave_adjust_steering_mode(struct mlx4_dev *dev,
struct mlx4_dev_cap *dev_cap,
struct mlx4_init_hca_param *hca_param)
{
dev->caps.steering_mode = hca_param->steering_mode;
if (dev->caps.steering_mode == MLX4_STEERING_MODE_DEVICE_MANAGED) {
dev->caps.num_qp_per_mgm = dev_cap->fs_max_num_qp_per_entry;
dev->caps.fs_log_max_ucast_qp_range_size =
dev_cap->fs_log_max_ucast_qp_range_size;
} else
dev->caps.num_qp_per_mgm =
4 * ((1 << hca_param->log_mc_entry_sz)/16 - 2);
mlx4_dbg(dev, "Steering mode is: %s\n",
mlx4_steering_mode_str(dev->caps.steering_mode));
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
static int mlx4_slave_cap(struct mlx4_dev *dev)
{
int err;
u32 page_size;
struct mlx4_dev_cap dev_cap;
struct mlx4_func_cap func_cap;
struct mlx4_init_hca_param hca_param;
net/mlx4_core: Use correct variable type for mlx4_slave_cap We've used an incorrect type for the loop counter and the mlx4_QUERY_FUNC_CAP function. The current input modifier is either a port or a boolean. Since the number of ports is always a positive value < 255, we should use u8 instead of an integer with casting. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:28 +07:00
u8 i;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
memset(&hca_param, 0, sizeof(hca_param));
err = mlx4_QUERY_HCA(dev, &hca_param);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "QUERY_HCA command failed, aborting\n");
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
return err;
}
net/mlx4_core: Removed unnecessary bit operation condition Fix the "warn: suspicious bitop condition" made by the smatch semantic checker on: drivers/net/ethernet/mellanox/mlx4/main.c:509 mlx4_slave_cap() Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-14 16:15:14 +07:00
/* fail if the hca has an unknown global capability
* at this time global_caps should be always zeroed
*/
if (hca_param.global_caps) {
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
mlx4_err(dev, "Unknown hca global capabilities\n");
return -ENOSYS;
}
mlx4_log_num_mgm_entry_size = hca_param.log_mc_entry_sz;
net/mlx4_core: Read HCA frequency and map internal clock Read HCA frequency, read PCI clock bar and offset, map internal clock to PCI bar. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-23 13:06:48 +07:00
dev->caps.hca_core_clock = hca_param.hca_core_clock;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
memset(&dev_cap, 0, sizeof(dev_cap));
net/mlx4_core: Fixes for VF / Guest startup flow - pass the following parameters: - firmware version (added QUERY_FW paravirtualization for that) - disable Blueflame on slaves. KVM disables write combining on guests, and we get better performance without BF in this case. (This requires QUERY_DEV_CAP paravirtualization, also in this commit) - max qp rdma as destination - get rid of a chunk of "if (0)" dead code Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-30 16:14:53 +07:00
dev->caps.max_qp_dest_rdma = 1 << hca_param.log_rd_per_qp;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
err = mlx4_dev_cap(dev, &dev_cap);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting\n");
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
return err;
}
net/mlx4_core: Fixes for VF / Guest startup flow - pass the following parameters: - firmware version (added QUERY_FW paravirtualization for that) - disable Blueflame on slaves. KVM disables write combining on guests, and we get better performance without BF in this case. (This requires QUERY_DEV_CAP paravirtualization, also in this commit) - max qp rdma as destination - get rid of a chunk of "if (0)" dead code Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-30 16:14:53 +07:00
err = mlx4_QUERY_FW(dev);
if (err)
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "QUERY_FW command failed: could not get FW version\n");
net/mlx4_core: Fixes for VF / Guest startup flow - pass the following parameters: - firmware version (added QUERY_FW paravirtualization for that) - disable Blueflame on slaves. KVM disables write combining on guests, and we get better performance without BF in this case. (This requires QUERY_DEV_CAP paravirtualization, also in this commit) - max qp rdma as destination - get rid of a chunk of "if (0)" dead code Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-30 16:14:53 +07:00
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
page_size = ~dev->caps.page_size_cap + 1;
mlx4_warn(dev, "HCA minimum page size:%d\n", page_size);
if (page_size > PAGE_SIZE) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "HCA minimum page size of %d bigger than kernel PAGE_SIZE of %ld, aborting\n",
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
page_size, PAGE_SIZE);
return -ENODEV;
}
/* slave gets uar page size from QUERY_HCA fw command */
dev->caps.uar_page_size = 1 << (hca_param.uar_page_sz + 12);
/* TODO: relax this assumption */
if (dev->caps.uar_page_size != PAGE_SIZE) {
mlx4_err(dev, "UAR size:%d != kernel PAGE_SIZE of %ld\n",
dev->caps.uar_page_size, PAGE_SIZE);
return -ENODEV;
}
memset(&func_cap, 0, sizeof(func_cap));
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
err = mlx4_QUERY_FUNC_CAP(dev, 0, &func_cap);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "QUERY_FUNC_CAP general command failed, aborting (%d)\n",
err);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
return err;
}
if ((func_cap.pf_context_behaviour | PF_CONTEXT_BEHAVIOUR_MASK) !=
PF_CONTEXT_BEHAVIOUR_MASK) {
net/mlx4: Add support for A0 steering Add the required firmware commands for A0 steering and a way to enable that. The firmware support focuses on INIT_HCA, QUERY_HCA, QUERY_PORT, QUERY_DEV_CAP and QUERY_FUNC_CAP commands. Those commands are used to configure and query the device. The different A0 DMFS (steering) modes are: Static - optimized performance, but flow steering rules are limited. This mode should be choosed explicitly by the user in order to be used. Dynamic - this mode should be explicitly choosed by the user. In this mode, the FW works in optimized steering mode as long as it can and afterwards automatically drops to classic (full) DMFS. Disable - this mode should be explicitly choosed by the user. The user instructs the system not to use optimized steering, even if the FW supports Dynamic A0 DMFS (and thus will be able to use optimized steering in Default A0 DMFS mode). Default - this mode is implicitly choosed. In this mode, if the FW supports Dynamic A0 DMFS, it'll work in this mode. Otherwise, it'll work at Disable A0 DMFS mode. Under SRIOV configuration, when the A0 steering mode is enabled, older guest VF drivers who aren't using the RX QP allocation flag (MLX4_RESERVE_A0_QP) will get a QP from the general range and fail when attempting to register a steering rule. To avoid that, the PF context behaviour is changed once on A0 static mode, to require support for the allocation flag in VF drivers too. In order to enable A0 steering, we use log_num_mgm_entry_size param. If the value of the parameter is not positive, we treat the absolute value of log_num_mgm_entry_size as a bit field. Setting bit 2 of this bit field enables static A0 steering. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:58:00 +07:00
mlx4_err(dev, "Unknown pf context behaviour %x known flags %x\n",
func_cap.pf_context_behaviour, PF_CONTEXT_BEHAVIOUR_MASK);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
return -ENOSYS;
}
dev->caps.num_ports = func_cap.num_ports;
mlx4: Structures and init/teardown for VF resource quotas This is step #1 for implementing SRIOV resource quotas for VFs. Quotas are implemented per resource type for VFs and the PF, to prevent any entity from simply grabbing all the resources for itself and leaving the other entities unable to obtain such resources. Resources which are allocated using quotas: QPs, CQs, SRQs, MPTs, MTTs, MAC, VLAN, and Counters. The quota system works as follows: Each entity (VF or PF) is given a max number of a given resource (its quota), and a guaranteed minimum number for each resource (starvation prevention). For QPs, CQs, SRQs, MPTs and MTTs: 50% of the available quantity for the resource is divided equally among the PF and all the active VFs (i.e., the number of VFs in the mlx4_core module parameter "num_vfs"). This 50% represents the "guaranteed minimum" pool. The other 50% is the "free pool", allocated on a first-come-first-serve basis. For each VF/PF, resources are first allocated from its "guaranteed-minimum" pool. When that pool is exhausted, the driver attempts to allocate from the resource "free-pool". The quota (i.e., max) for the VFs and the PF is: The free-pool amount (50% of the real max) + the guaranteed minimum For MACs: Guarantee 2 MACs per VF/PF per port. As a result, since we have only 128 MACs per port, reduce the allowable number of VFs from 64 to 63. Any remaining MACs are put into a free pool. For VLANs: For the PF, the per-port quota is 128 and guarantee is 64 (to allow the PF to register at least a VLAN per VF in VST mode). For the VFs, the per-port quota is 64 and the guarantee is 0. We assume that VGT VFs are trusted not to abuse the VLAN resource. For Counters: For all functions (PF and VFs), the quota is 128 and the guarantee is 0. In this patch, we define the needed structures, which are added to the resource-tracker struct. In addition, we do initialization for the resource quota, and adjust the query_device response to use quotas rather than resource maxima. As part of the implementation, we introduce a new field in mlx4_dev: quotas. This field holds the resource quotas used to report maxima to the upper layers (ib_core, via query_device). The HCA maxima of these values are passed to the VFs (via QUERY_HCA) so that they may continue to use these in handling QPs, CQs, SRQs and MPTs. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-03 15:03:23 +07:00
dev->quotas.qp = func_cap.qp_quota;
dev->quotas.srq = func_cap.srq_quota;
dev->quotas.cq = func_cap.cq_quota;
dev->quotas.mpt = func_cap.mpt_quota;
dev->quotas.mtt = func_cap.mtt_quota;
dev->caps.num_qps = 1 << hca_param.log_num_qps;
dev->caps.num_srqs = 1 << hca_param.log_num_srqs;
dev->caps.num_cqs = 1 << hca_param.log_num_cqs;
dev->caps.num_mpts = 1 << hca_param.log_mpt_sz;
dev->caps.num_eqs = func_cap.max_eq;
dev->caps.reserved_eqs = func_cap.reserved_eq;
net/mlx4_core: Add reserved lkey for VFs to QUERY_FUNC_CAP The reserved lKey is different for each VF. A base lkey value is returned in QUERY_DEV_CAP at offset 0x98. The reserved L_key value for a VF is: VF_lkey = base_lkey + (VF_number << 8). This VF L_key value should be returned in QUERY_FUNC_CAP (opcode-modifier = 0) at offset 0x48. To indicate that the lkey value at offset 0x48 is valid, the Hypervisor sets a flag bit in dword 0x0, offset 27 in the QUERY_FUNC_CAP wrapper function. When the VF calls QUERY_FUNC_CAP, it should check if this flag bit is set. If it is set, the VF should take the reserved lkey value at offset 0x48. If the bit is not set, the VF should not use a reserved lkey (i.e., should set its reserved lkey value to 0). Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-27 20:58:00 +07:00
dev->caps.reserved_lkey = func_cap.reserved_lkey;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
dev->caps.num_pds = MLX4_NUM_PDS;
dev->caps.num_mgms = 0;
dev->caps.num_amgms = 0;
if (dev->caps.num_ports > MLX4_MAX_PORTS) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "HCA has %d ports, but we only support %d, aborting\n",
dev->caps.num_ports, MLX4_MAX_PORTS);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
return -ENODEV;
}
mlx4: Add infrastructure for selecting VFs to enable QP0 via MLX proxy QPs This commit adds the infrastructure for enabling selected VFs to operate SMI (QP0) MADs without restriction. Additionally, for these enabled VFs, their QP0 proxy and tunnel QPs are MLX QPs. As such, they operate over VL15. Therefore, they are not affected by "credit" problems or changes in the VLArb table (which may shut down VL0). Non-enabled VFs may only create UD proxy QP0 qps (which are forced by the hypervisor to send packets using the q-key it assigns and places in the qp-context). Thus, non-enabled VFs will not pose a security risk. The hypervisor discards any privileged MADs it receives from these non-enabled VFs. By default, all VFs are NOT enabled, and must explicitly be enabled by the administrator. The sysfs interface which operates the VF enablement infrastructure is provided in the next commit. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-05-29 20:31:03 +07:00
dev->caps.qp0_qkey = kcalloc(dev->caps.num_ports, sizeof(u32), GFP_KERNEL);
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
dev->caps.qp0_tunnel = kcalloc(dev->caps.num_ports, sizeof (u32), GFP_KERNEL);
dev->caps.qp0_proxy = kcalloc(dev->caps.num_ports, sizeof (u32), GFP_KERNEL);
dev->caps.qp1_tunnel = kcalloc(dev->caps.num_ports, sizeof (u32), GFP_KERNEL);
dev->caps.qp1_proxy = kcalloc(dev->caps.num_ports, sizeof (u32), GFP_KERNEL);
if (!dev->caps.qp0_tunnel || !dev->caps.qp0_proxy ||
mlx4: Add infrastructure for selecting VFs to enable QP0 via MLX proxy QPs This commit adds the infrastructure for enabling selected VFs to operate SMI (QP0) MADs without restriction. Additionally, for these enabled VFs, their QP0 proxy and tunnel QPs are MLX QPs. As such, they operate over VL15. Therefore, they are not affected by "credit" problems or changes in the VLArb table (which may shut down VL0). Non-enabled VFs may only create UD proxy QP0 qps (which are forced by the hypervisor to send packets using the q-key it assigns and places in the qp-context). Thus, non-enabled VFs will not pose a security risk. The hypervisor discards any privileged MADs it receives from these non-enabled VFs. By default, all VFs are NOT enabled, and must explicitly be enabled by the administrator. The sysfs interface which operates the VF enablement infrastructure is provided in the next commit. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-05-29 20:31:03 +07:00
!dev->caps.qp1_tunnel || !dev->caps.qp1_proxy ||
!dev->caps.qp0_qkey) {
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
err = -ENOMEM;
goto err_mem;
}
mlx4: Put physical GID and P_Key table sizes in mlx4_phys_caps struct and paravirtualize them To allow easy paravirtualization of P_Key and GID table sizes, keep paravirtualized sizes in mlx4_dev->caps, but save the actual physical sizes from FW in struct: mlx4_dev->phys_cap. In addition, in SR-IOV mode, do the following: 1. Reduce reported P_Key table size by 1. This is done to reserve the highest P_Key index for internal use, for declaring an invalid P_Key in P_Key paravirtualization. We require a P_Key index which always contain an invalid P_Key value for this purpose (i.e., one which cannot be modified by the subnet manager). The way to do this is to reduce the P_Key table size reported to the subnet manager by 1, so that it will not attempt to access the P_Key at index #127. 2. Paravirtualize the GID table size to 1. Thus, each guest sees only a single GID (at its paravirtualized index 0). In addition, since we are paravirtualizing the GID table size to 1, we add paravirtualization of the master GID event here (i.e., we do not do ib_dispatch_event() for the GUID change event on the master, since its (only) GUID never changes). Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:44 +07:00
for (i = 1; i <= dev->caps.num_ports; ++i) {
net/mlx4_core: Use correct variable type for mlx4_slave_cap We've used an incorrect type for the loop counter and the mlx4_QUERY_FUNC_CAP function. The current input modifier is either a port or a boolean. Since the number of ports is always a positive value < 255, we should use u8 instead of an integer with casting. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:28 +07:00
err = mlx4_QUERY_FUNC_CAP(dev, i, &func_cap);
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "QUERY_FUNC_CAP port command failed for port %d, aborting (%d)\n",
i, err);
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
goto err_mem;
}
mlx4: Add infrastructure for selecting VFs to enable QP0 via MLX proxy QPs This commit adds the infrastructure for enabling selected VFs to operate SMI (QP0) MADs without restriction. Additionally, for these enabled VFs, their QP0 proxy and tunnel QPs are MLX QPs. As such, they operate over VL15. Therefore, they are not affected by "credit" problems or changes in the VLArb table (which may shut down VL0). Non-enabled VFs may only create UD proxy QP0 qps (which are forced by the hypervisor to send packets using the q-key it assigns and places in the qp-context). Thus, non-enabled VFs will not pose a security risk. The hypervisor discards any privileged MADs it receives from these non-enabled VFs. By default, all VFs are NOT enabled, and must explicitly be enabled by the administrator. The sysfs interface which operates the VF enablement infrastructure is provided in the next commit. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-05-29 20:31:03 +07:00
dev->caps.qp0_qkey[i - 1] = func_cap.qp0_qkey;
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
dev->caps.qp0_tunnel[i - 1] = func_cap.qp0_tunnel_qpn;
dev->caps.qp0_proxy[i - 1] = func_cap.qp0_proxy_qpn;
dev->caps.qp1_tunnel[i - 1] = func_cap.qp1_tunnel_qpn;
dev->caps.qp1_proxy[i - 1] = func_cap.qp1_proxy_qpn;
net/mlx4_core: Check port out-of-range before using in mlx4_slave_cap The range check was performed after using the port number. Reverse this to prevent a potential array overflow. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-30 16:14:54 +07:00
dev->caps.port_mask[i] = dev->caps.port_type[i];
net/mlx4_core: Expose physical port id as PF/VF capability Add the infrastructure needed to support ndo_get_phys_port_id which allows users to identify to which physical port a net-device is connected to by reading a unique port id. This will work for VFs and PFs. The driver uses a new device capability - phys_port_id, The PF driver reads the port phys_port_id from Firmware and stores it. The VF driver reads the port phys_port_id from the PF using QUERY_FUNC_CAP command. Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-20 02:20:12 +07:00
dev->caps.phys_port_id[i] = func_cap.phys_port_id;
mlx4: Put physical GID and P_Key table sizes in mlx4_phys_caps struct and paravirtualize them To allow easy paravirtualization of P_Key and GID table sizes, keep paravirtualized sizes in mlx4_dev->caps, but save the actual physical sizes from FW in struct: mlx4_dev->phys_cap. In addition, in SR-IOV mode, do the following: 1. Reduce reported P_Key table size by 1. This is done to reserve the highest P_Key index for internal use, for declaring an invalid P_Key in P_Key paravirtualization. We require a P_Key index which always contain an invalid P_Key value for this purpose (i.e., one which cannot be modified by the subnet manager). The way to do this is to reduce the P_Key table size reported to the subnet manager by 1, so that it will not attempt to access the P_Key at index #127. 2. Paravirtualize the GID table size to 1. Thus, each guest sees only a single GID (at its paravirtualized index 0). In addition, since we are paravirtualizing the GID table size to 1, we add paravirtualization of the master GID event here (i.e., we do not do ib_dispatch_event() for the GUID change event on the master, since its (only) GUID never changes). Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:44 +07:00
if (mlx4_get_slave_pkey_gid_tbl_len(dev, i,
&dev->caps.gid_table_len[i],
&dev->caps.pkey_table_len[i]))
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
goto err_mem;
mlx4: Put physical GID and P_Key table sizes in mlx4_phys_caps struct and paravirtualize them To allow easy paravirtualization of P_Key and GID table sizes, keep paravirtualized sizes in mlx4_dev->caps, but save the actual physical sizes from FW in struct: mlx4_dev->phys_cap. In addition, in SR-IOV mode, do the following: 1. Reduce reported P_Key table size by 1. This is done to reserve the highest P_Key index for internal use, for declaring an invalid P_Key in P_Key paravirtualization. We require a P_Key index which always contain an invalid P_Key value for this purpose (i.e., one which cannot be modified by the subnet manager). The way to do this is to reduce the P_Key table size reported to the subnet manager by 1, so that it will not attempt to access the P_Key at index #127. 2. Paravirtualize the GID table size to 1. Thus, each guest sees only a single GID (at its paravirtualized index 0). In addition, since we are paravirtualizing the GID table size to 1, we add paravirtualization of the master GID event here (i.e., we do not do ib_dispatch_event() for the GUID change event on the master, since its (only) GUID never changes). Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:44 +07:00
}
net/mlx4_core: Check port out-of-range before using in mlx4_slave_cap The range check was performed after using the port number. Reverse this to prevent a potential array overflow. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-30 16:14:54 +07:00
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if (dev->caps.uar_page_size * (dev->caps.num_uars -
dev->caps.reserved_uars) >
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
pci_resource_len(dev->persist->pdev,
2)) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "HCA reported UAR region size of 0x%x bigger than PCI resource 2 size of 0x%llx, aborting\n",
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
dev->caps.uar_page_size * dev->caps.num_uars,
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
(unsigned long long)
pci_resource_len(dev->persist->pdev, 2));
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
goto err_mem;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
mlx4: 64-byte CQE/EQE support ConnectX-3 devices can use either 64- or 32-byte completion queue entries (CQEs) and event queue entries (EQEs). Using 64-byte EQEs/CQEs performs better because each entry is aligned to a complete cacheline. This patch queries the HCA's capabilities, and if it supports 64-byte CQEs and EQES the driver will configure the HW to work in 64-byte mode. The 32-byte vs 64-byte mode is global per HCA and not per CQ or EQ. Since this mode is global, userspace (libmlx4) must be updated to work with the configured CQE size, and guests using SR-IOV virtual functions need to know both EQE and CQE size. In case one of the 64-byte CQE/EQE capabilities is activated, the patch makes sure that older guest drivers that use the QUERY_DEV_FUNC command (e.g as done in mlx4_core of Linux 3.3..3.6) will notice that they need an update to be able to work with the PPF. This is done by changing the returned pf_context_behaviour not to be zero any more. In case none of these capabilities is activated that value remains zero and older guest drivers can run OK. The SRIOV related flow is as follows 1. the PPF does the detection of the new capabilities using QUERY_DEV_CAP command. 2. the PPF activates the new capabilities using INIT_HCA. 3. the VF detects if the PPF activated the capabilities using QUERY_HCA, and if this is the case activates them for itself too. Note that the VF detects that it must be aware to the new PF behaviour using QUERY_FUNC_CAP. Steps 1 and 2 apply also for native mode. User space notification is done through a new field introduced in struct mlx4_ib_ucontext which holds device capabilities for which user space must take action. This changes the binary interface so the ABI towards libmlx4 exposed through uverbs is bumped from 3 to 4 but only when **needed** i.e. only when the driver does use 64-byte CQEs or future device capabilities which must be in sync by user space. This practice allows to work with unmodified libmlx4 on older devices (e.g A0, B0) which don't support 64-byte CQEs. In order to keep existing systems functional when they update to a newer kernel that contains these changes in VF and userspace ABI, a module parameter enable_64b_cqe_eqe must be set to enable 64-byte mode; the default is currently false. Signed-off-by: Eli Cohen <eli@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-10-21 21:59:24 +07:00
if (hca_param.dev_cap_enabled & MLX4_DEV_CAP_64B_EQE_ENABLED) {
dev->caps.eqe_size = 64;
dev->caps.eqe_factor = 1;
} else {
dev->caps.eqe_size = 32;
dev->caps.eqe_factor = 0;
}
if (hca_param.dev_cap_enabled & MLX4_DEV_CAP_64B_CQE_ENABLED) {
dev->caps.cqe_size = 64;
net/mlx4_core: Enable CQE/EQE stride support This feature is intended for archs having cache line larger then 64B. Since our CQE/EQEs are generally 64B in those systems, HW will write twice to the same cache line consecutively, causing pipe locks due to he hazard prevention mechanism. For elements in a cyclic buffer, writes are consecutive, so entries smaller than a cache line should be avoided, especially if they are written at a high rate. Reduce consecutive writes to same cache line in CQs/EQs, by allowing the driver to increase the distance between entries so that each will reside in a different cache line. Until the introduction of this feature, there were two types of CQE/EQE: 1. 32B stride and context in the [0-31] segment 2. 64B stride and context in the [32-63] segment This feature introduces two additional types: 3. 128B stride and context in the [0-31] segment (128B cache line) 4. 256B stride and context in the [0-31] segment (256B cache line) Modify the mlx4_core driver to query the device for the CQE/EQE cache line stride capability and to enable that capability when the host cache line size is larger than 64 bytes (supported cache lines are 128B and 256B). The mlx4 IB driver and libmlx4 need not be aware of this change. The PF context behaviour is changed to require this change in VF drivers running on such archs. Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-18 15:50:59 +07:00
dev->caps.userspace_caps |= MLX4_USER_DEV_CAP_LARGE_CQE;
mlx4: 64-byte CQE/EQE support ConnectX-3 devices can use either 64- or 32-byte completion queue entries (CQEs) and event queue entries (EQEs). Using 64-byte EQEs/CQEs performs better because each entry is aligned to a complete cacheline. This patch queries the HCA's capabilities, and if it supports 64-byte CQEs and EQES the driver will configure the HW to work in 64-byte mode. The 32-byte vs 64-byte mode is global per HCA and not per CQ or EQ. Since this mode is global, userspace (libmlx4) must be updated to work with the configured CQE size, and guests using SR-IOV virtual functions need to know both EQE and CQE size. In case one of the 64-byte CQE/EQE capabilities is activated, the patch makes sure that older guest drivers that use the QUERY_DEV_FUNC command (e.g as done in mlx4_core of Linux 3.3..3.6) will notice that they need an update to be able to work with the PPF. This is done by changing the returned pf_context_behaviour not to be zero any more. In case none of these capabilities is activated that value remains zero and older guest drivers can run OK. The SRIOV related flow is as follows 1. the PPF does the detection of the new capabilities using QUERY_DEV_CAP command. 2. the PPF activates the new capabilities using INIT_HCA. 3. the VF detects if the PPF activated the capabilities using QUERY_HCA, and if this is the case activates them for itself too. Note that the VF detects that it must be aware to the new PF behaviour using QUERY_FUNC_CAP. Steps 1 and 2 apply also for native mode. User space notification is done through a new field introduced in struct mlx4_ib_ucontext which holds device capabilities for which user space must take action. This changes the binary interface so the ABI towards libmlx4 exposed through uverbs is bumped from 3 to 4 but only when **needed** i.e. only when the driver does use 64-byte CQEs or future device capabilities which must be in sync by user space. This practice allows to work with unmodified libmlx4 on older devices (e.g A0, B0) which don't support 64-byte CQEs. In order to keep existing systems functional when they update to a newer kernel that contains these changes in VF and userspace ABI, a module parameter enable_64b_cqe_eqe must be set to enable 64-byte mode; the default is currently false. Signed-off-by: Eli Cohen <eli@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-10-21 21:59:24 +07:00
} else {
dev->caps.cqe_size = 32;
}
net/mlx4_core: Enable CQE/EQE stride support This feature is intended for archs having cache line larger then 64B. Since our CQE/EQEs are generally 64B in those systems, HW will write twice to the same cache line consecutively, causing pipe locks due to he hazard prevention mechanism. For elements in a cyclic buffer, writes are consecutive, so entries smaller than a cache line should be avoided, especially if they are written at a high rate. Reduce consecutive writes to same cache line in CQs/EQs, by allowing the driver to increase the distance between entries so that each will reside in a different cache line. Until the introduction of this feature, there were two types of CQE/EQE: 1. 32B stride and context in the [0-31] segment 2. 64B stride and context in the [32-63] segment This feature introduces two additional types: 3. 128B stride and context in the [0-31] segment (128B cache line) 4. 256B stride and context in the [0-31] segment (256B cache line) Modify the mlx4_core driver to query the device for the CQE/EQE cache line stride capability and to enable that capability when the host cache line size is larger than 64 bytes (supported cache lines are 128B and 256B). The mlx4 IB driver and libmlx4 need not be aware of this change. The PF context behaviour is changed to require this change in VF drivers running on such archs. Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-18 15:50:59 +07:00
if (hca_param.dev_cap_enabled & MLX4_DEV_CAP_EQE_STRIDE_ENABLED) {
dev->caps.eqe_size = hca_param.eqe_size;
dev->caps.eqe_factor = 0;
}
if (hca_param.dev_cap_enabled & MLX4_DEV_CAP_CQE_STRIDE_ENABLED) {
dev->caps.cqe_size = hca_param.cqe_size;
/* User still need to know when CQE > 32B */
dev->caps.userspace_caps |= MLX4_USER_DEV_CAP_LARGE_CQE;
}
net/mlx_en: Timestamping is not supported in slave mode Old hypervisors don't mask out timestamp capability for slave. Till slave support will be added, need to disable capability by slave. Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-20 18:58:10 +07:00
dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_TS;
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_warn(dev, "Timestamping is not supported in slave mode\n");
net/mlx_en: Timestamping is not supported in slave mode Old hypervisors don't mask out timestamp capability for slave. Till slave support will be added, need to disable capability by slave. Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-20 18:58:10 +07:00
mlx4_core: Adjustments to Flow Steering activation logic for SR-IOV Separate flow steering capability detection from the decision to activate. For the master (and for native), detect the flow steering capability in mlx4_dev_cap, but activate the appropriate steering type in a new function choose_flow_steering() based on detected data. For VFs, activate flow steering based on what was actually activated by the master, where that info is obtained via QUERY_HCA. This fixes the current VF detection which is wrongly based on QUERY_DEV_CAP. Also, for SR-IOV mode, if flow steering may be activated, do so only if the max number of QPs per rule is sufficient to satisfy one subscription per VF. If not, fall back to B0 mode. This is needed to serve registrations done by L2 network drivers such as mlx4_en and IPoIB when the network stack attempts to join to multicast groups such as all-hosts or the IPoIB broadcast group. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:11:59 +07:00
slave_adjust_steering_mode(dev, &dev_cap, &hca_param);
net/mlx4: Add RSS support for fragmented IP datagrams Enable RSS support for fragmented IP packets, when device supports it. Until now, fragmented IP packets were directed only to the default_qpn. Since IP fragments (datagram) have no upper protocols (L3 IP packets), hash is performed on 3-tuple - dst MAC, source IP and dest IP. The HW makes sure that this holds for the 1st fragment too, so all fragments go to the same QP. Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-04-02 20:31:06 +07:00
mlx4_dbg(dev, "RSS support for IP fragments is %s\n",
hca_param.rss_ip_frags ? "on" : "off");
mlx4_core: Adjustments to Flow Steering activation logic for SR-IOV Separate flow steering capability detection from the decision to activate. For the master (and for native), detect the flow steering capability in mlx4_dev_cap, but activate the appropriate steering type in a new function choose_flow_steering() based on detected data. For VFs, activate flow steering based on what was actually activated by the master, where that info is obtained via QUERY_HCA. This fixes the current VF detection which is wrongly based on QUERY_DEV_CAP. Also, for SR-IOV mode, if flow steering may be activated, do so only if the max number of QPs per rule is sufficient to satisfy one subscription per VF. If not, fall back to B0 mode. This is needed to serve registrations done by L2 network drivers such as mlx4_en and IPoIB when the network stack attempts to join to multicast groups such as all-hosts or the IPoIB broadcast group. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:11:59 +07:00
net/mlx4: Change QP allocation scheme When using BF (Blue-Flame), the QPN overrides the VLAN, CV, and SV fields in the WQE. Thus, BF may only be used for QPNs with bits 6,7 unset. The current Ethernet driver code reserves a Tx QP range with 256b alignment. This is wrong because if there are more than 64 Tx QPs in use, QPNs >= base + 65 will have bits 6/7 set. This problem is not specific for the Ethernet driver, any entity that tries to reserve more than 64 BF-enabled QPs should fail. Also, using ranges is not necessary here and is wasteful. The new mechanism introduced here will support reservation for "Eth QPs eligible for BF" for all drivers: bare-metal, multi-PF, and VFs (when hypervisors support WC in VMs). The flow we use is: 1. In mlx4_en, allocate Tx QPs one by one instead of a range allocation, and request "BF enabled QPs" if BF is supported for the function 2. In the ALLOC_RES FW command, change param1 to: a. param1[23:0] - number of QPs b. param1[31-24] - flags controlling QPs reservation Bit 31 refers to Eth blueflame supported QPs. Those QPs must have bits 6 and 7 unset in order to be used in Ethernet. Bits 24-30 of the flags are currently reserved. When a function tries to allocate a QP, it states the required attributes for this QP. Those attributes are considered "best-effort". If an attribute, such as Ethernet BF enabled QP, is a must-have attribute, the function has to check that attribute is supported before trying to do the allocation. In a lower layer of the code, mlx4_qp_reserve_range masks out the bits which are unsupported. If SRIOV is used, the PF validates those attributes and masks out unsupported attributes as well. In order to notify VFs which attributes are supported, the VF uses QUERY_FUNC_CAP command. This command's mailbox is filled by the PF, which notifies which QP allocation attributes it supports. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:57:54 +07:00
if (func_cap.extra_flags & MLX4_QUERY_FUNC_FLAGS_BF_RES_QP &&
dev->caps.bf_reg_size)
dev->caps.alloc_res_qp_mask |= MLX4_RESERVE_ETH_BF_QP;
net/mlx4: Add A0 hybrid steering A0 hybrid steering is a form of high performance flow steering. By using this mode, mlx4 cards use a fast limited table based steering, in order to enable fast steering of unicast packets to a QP. In order to implement A0 hybrid steering we allocate resources from different zones: (1) General range (2) Special MAC-assigned QPs [RSS, Raw-Ethernet] each has its own region. When we create a rss QP or a raw ethernet (A0 steerable and BF ready) QP, we try hard to allocate the QP from range (2). Otherwise, we try hard not to allocate from this range. However, when the system is pushed to its limits and one needs every resource, the allocator uses every region it can. Meaning, when we run out of raw-eth qps, the allocator allocates from the general range (and the special-A0 area is no longer active). If we run out of RSS qps, the mechanism tries to allocate from the raw-eth QP zone. If that is also exhausted, the allocator will allocate from the general range (and the A0 region is no longer active). Note that if a raw-eth qp is allocated from the general range, it attempts to allocate the range such that bits 6 and 7 (blueflame bits) in the QP number are not set. When the feature is used in SRIOV, the VF has to notify the PF what kind of QP attributes it needs. In order to do that, along with the "Eth QP blueflame" bit, we reserve a new "A0 steerable QP". According to the combination of these bits, the PF tries to allocate a suitable QP. In order to maintain backward compatibility (with older PFs), the PF notifies which QP attributes it supports via QUERY_FUNC_CAP command. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:57:57 +07:00
if (func_cap.extra_flags & MLX4_QUERY_FUNC_FLAGS_A0_RES_QP)
dev->caps.alloc_res_qp_mask |= MLX4_RESERVE_A0_QP;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
return 0;
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
err_mem:
mlx4: Add infrastructure for selecting VFs to enable QP0 via MLX proxy QPs This commit adds the infrastructure for enabling selected VFs to operate SMI (QP0) MADs without restriction. Additionally, for these enabled VFs, their QP0 proxy and tunnel QPs are MLX QPs. As such, they operate over VL15. Therefore, they are not affected by "credit" problems or changes in the VLArb table (which may shut down VL0). Non-enabled VFs may only create UD proxy QP0 qps (which are forced by the hypervisor to send packets using the q-key it assigns and places in the qp-context). Thus, non-enabled VFs will not pose a security risk. The hypervisor discards any privileged MADs it receives from these non-enabled VFs. By default, all VFs are NOT enabled, and must explicitly be enabled by the administrator. The sysfs interface which operates the VF enablement infrastructure is provided in the next commit. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-05-29 20:31:03 +07:00
kfree(dev->caps.qp0_qkey);
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
kfree(dev->caps.qp0_tunnel);
kfree(dev->caps.qp0_proxy);
kfree(dev->caps.qp1_tunnel);
kfree(dev->caps.qp1_proxy);
mlx4: Add infrastructure for selecting VFs to enable QP0 via MLX proxy QPs This commit adds the infrastructure for enabling selected VFs to operate SMI (QP0) MADs without restriction. Additionally, for these enabled VFs, their QP0 proxy and tunnel QPs are MLX QPs. As such, they operate over VL15. Therefore, they are not affected by "credit" problems or changes in the VLArb table (which may shut down VL0). Non-enabled VFs may only create UD proxy QP0 qps (which are forced by the hypervisor to send packets using the q-key it assigns and places in the qp-context). Thus, non-enabled VFs will not pose a security risk. The hypervisor discards any privileged MADs it receives from these non-enabled VFs. By default, all VFs are NOT enabled, and must explicitly be enabled by the administrator. The sysfs interface which operates the VF enablement infrastructure is provided in the next commit. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-05-29 20:31:03 +07:00
dev->caps.qp0_qkey = NULL;
dev->caps.qp0_tunnel = NULL;
dev->caps.qp0_proxy = NULL;
dev->caps.qp1_tunnel = NULL;
dev->caps.qp1_proxy = NULL;
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
return err;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
net/mlx4_core: Load higher level modules according to ports type Mellanox ConnectX architecture is: mlx4_core is the lower level PCI driver which register on the PCI id, and protocol specific drivers are depended on it: mlx4_en - for Ethernet and mlx4_ib for Infiniband. NIC could have multiple ports which can change their type dynamically. We use the request_module() call to load the relevant protocol driver when needed: on loading time or at port type change event. Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-15 21:55:24 +07:00
static void mlx4_request_modules(struct mlx4_dev *dev)
{
int port;
int has_ib_port = false;
int has_eth_port = false;
#define EN_DRV_NAME "mlx4_en"
#define IB_DRV_NAME "mlx4_ib"
for (port = 1; port <= dev->caps.num_ports; port++) {
if (dev->caps.port_type[port] == MLX4_PORT_TYPE_IB)
has_ib_port = true;
else if (dev->caps.port_type[port] == MLX4_PORT_TYPE_ETH)
has_eth_port = true;
}
if (has_eth_port)
request_module_nowait(EN_DRV_NAME);
net/mlx4_core: Load the Eth driver first When running in SRIOV mode, VM that is assigned with a non-provisioned Ethernet VFs get themselves a random mac when the Eth driver starts. In this case, if the IB driver startup code that deals with RoCE runs first, it will use a zero mac as the source mac for the Para-Virtual CM MADs which is buggy. To handle that, we change the order of loading. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-04 21:07:24 +07:00
if (has_ib_port || (dev->caps.flags & MLX4_DEV_CAP_FLAG_IBOE))
request_module_nowait(IB_DRV_NAME);
net/mlx4_core: Load higher level modules according to ports type Mellanox ConnectX architecture is: mlx4_core is the lower level PCI driver which register on the PCI id, and protocol specific drivers are depended on it: mlx4_en - for Ethernet and mlx4_ib for Infiniband. NIC could have multiple ports which can change their type dynamically. We use the request_module() call to load the relevant protocol driver when needed: on loading time or at port type change event. Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-15 21:55:24 +07:00
}
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
/*
* Change the port configuration of the device.
* Every user of this function must hold the port mutex.
*/
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
int mlx4_change_port_types(struct mlx4_dev *dev,
enum mlx4_port_type *port_types)
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
{
int err = 0;
int change = 0;
int port;
for (port = 0; port < dev->caps.num_ports; port++) {
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
/* Change the port type only if the new type is different
* from the current, and not set to Auto */
mlx4: Setting new port types after all interfaces unregistered In port type change flow, need to set the new port types only after all interfaces have finished the unregister process. Otherwise, during unregister, one of the interfaces might issue a SET_PORT command with wrong port types, it can cause bad FW behavior. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-21 10:41:07 +07:00
if (port_types[port] != dev->caps.port_type[port + 1])
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
change = 1;
}
if (change) {
mlx4_unregister_device(dev);
for (port = 1; port <= dev->caps.num_ports; port++) {
mlx4_CLOSE_PORT(dev, port);
mlx4_core: Fixing array indexes when setting port types Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-23 14:04:35 +07:00
dev->caps.port_type[port] = port_types[port - 1];
mlx4: Put physical GID and P_Key table sizes in mlx4_phys_caps struct and paravirtualize them To allow easy paravirtualization of P_Key and GID table sizes, keep paravirtualized sizes in mlx4_dev->caps, but save the actual physical sizes from FW in struct: mlx4_dev->phys_cap. In addition, in SR-IOV mode, do the following: 1. Reduce reported P_Key table size by 1. This is done to reserve the highest P_Key index for internal use, for declaring an invalid P_Key in P_Key paravirtualization. We require a P_Key index which always contain an invalid P_Key value for this purpose (i.e., one which cannot be modified by the subnet manager). The way to do this is to reduce the P_Key table size reported to the subnet manager by 1, so that it will not attempt to access the P_Key at index #127. 2. Paravirtualize the GID table size to 1. Thus, each guest sees only a single GID (at its paravirtualized index 0). In addition, since we are paravirtualizing the GID table size to 1, we add paravirtualization of the master GID event here (i.e., we do not do ib_dispatch_event() for the GUID change event on the master, since its (only) GUID never changes). Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:44 +07:00
err = mlx4_SET_PORT(dev, port, -1);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to set port %d, aborting\n",
port);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
goto out;
}
}
mlx4_set_port_mask(dev);
err = mlx4_register_device(dev);
net/mlx4_core: Load higher level modules according to ports type Mellanox ConnectX architecture is: mlx4_core is the lower level PCI driver which register on the PCI id, and protocol specific drivers are depended on it: mlx4_en - for Ethernet and mlx4_ib for Infiniband. NIC could have multiple ports which can change their type dynamically. We use the request_module() call to load the relevant protocol driver when needed: on loading time or at port type change event. Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-15 21:55:24 +07:00
if (err) {
mlx4_err(dev, "Failed to register device\n");
goto out;
}
mlx4_request_modules(dev);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
}
out:
return err;
}
static ssize_t show_port_type(struct device *dev,
struct device_attribute *attr,
char *buf)
{
struct mlx4_port_info *info = container_of(attr, struct mlx4_port_info,
port_attr);
struct mlx4_dev *mdev = info->dev;
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
char type[8];
sprintf(type, "%s",
(mdev->caps.port_type[info->port] == MLX4_PORT_TYPE_IB) ?
"ib" : "eth");
if (mdev->caps.possible_type[info->port] == MLX4_PORT_TYPE_AUTO)
sprintf(buf, "auto (%s)\n", type);
else
sprintf(buf, "%s\n", type);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
return strlen(buf);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
}
static ssize_t set_port_type(struct device *dev,
struct device_attribute *attr,
const char *buf, size_t count)
{
struct mlx4_port_info *info = container_of(attr, struct mlx4_port_info,
port_attr);
struct mlx4_dev *mdev = info->dev;
struct mlx4_priv *priv = mlx4_priv(mdev);
enum mlx4_port_type types[MLX4_MAX_PORTS];
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
enum mlx4_port_type new_types[MLX4_MAX_PORTS];
net/mlx4_core: Protect port type setting by mutex We need to protect set_port_type() for concurrency, as the sysfs code could call it from mutliple contexts in parallel. The port_mutex is not enough because we need to protect from concurrent modification of 'info' and stopping of the port sensing work. Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-02 21:26:14 +07:00
static DEFINE_MUTEX(set_port_type_mutex);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
int i;
int err = 0;
net/mlx4_core: Protect port type setting by mutex We need to protect set_port_type() for concurrency, as the sysfs code could call it from mutliple contexts in parallel. The port_mutex is not enough because we need to protect from concurrent modification of 'info' and stopping of the port sensing work. Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-02 21:26:14 +07:00
mutex_lock(&set_port_type_mutex);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
if (!strcmp(buf, "ib\n"))
info->tmp_type = MLX4_PORT_TYPE_IB;
else if (!strcmp(buf, "eth\n"))
info->tmp_type = MLX4_PORT_TYPE_ETH;
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
else if (!strcmp(buf, "auto\n"))
info->tmp_type = MLX4_PORT_TYPE_AUTO;
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
else {
mlx4_err(mdev, "%s is not supported port type\n", buf);
net/mlx4_core: Protect port type setting by mutex We need to protect set_port_type() for concurrency, as the sysfs code could call it from mutliple contexts in parallel. The port_mutex is not enough because we need to protect from concurrent modification of 'info' and stopping of the port sensing work. Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-02 21:26:14 +07:00
err = -EINVAL;
goto err_out;
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
}
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
mlx4_stop_sense(mdev);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
mutex_lock(&priv->port_mutex);
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
/* Possible type is always the one that was delivered */
mdev->caps.possible_type[info->port] = info->tmp_type;
for (i = 0; i < mdev->caps.num_ports; i++) {
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
types[i] = priv->port[i+1].tmp_type ? priv->port[i+1].tmp_type :
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
mdev->caps.possible_type[i+1];
if (types[i] == MLX4_PORT_TYPE_AUTO)
types[i] = mdev->caps.port_type[i+1];
}
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
mlx4: capability for link sensing For ConnectX3 devices, we allow link sensing only if FW explicitly reported it supports the feature. For older versions (ConnectX1 and 2), if the card supports both link layer types (Ethenet and Infiniband), link sensing is supported. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-19 11:00:26 +07:00
if (!(mdev->caps.flags & MLX4_DEV_CAP_FLAG_DPDP) &&
!(mdev->caps.flags & MLX4_DEV_CAP_FLAG_SENSE_SUPPORT)) {
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
for (i = 1; i <= mdev->caps.num_ports; i++) {
if (mdev->caps.possible_type[i] == MLX4_PORT_TYPE_AUTO) {
mdev->caps.possible_type[i] = mdev->caps.port_type[i];
err = -EINVAL;
}
}
}
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(mdev, "Auto sensing is not supported on this HCA. Set only 'eth' or 'ib' for both ports (should be the same)\n");
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
goto out;
}
mlx4_do_sense_ports(mdev, new_types, types);
err = mlx4_check_port_params(mdev, new_types);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
if (err)
goto out;
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
/* We are about to apply the changes after the configuration
* was verified, no need to remember the temporary types
* any more */
for (i = 0; i < mdev->caps.num_ports; i++)
priv->port[i + 1].tmp_type = 0;
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
err = mlx4_change_port_types(mdev, new_types);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
out:
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
mlx4_start_sense(mdev);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
mutex_unlock(&priv->port_mutex);
net/mlx4_core: Protect port type setting by mutex We need to protect set_port_type() for concurrency, as the sysfs code could call it from mutliple contexts in parallel. The port_mutex is not enough because we need to protect from concurrent modification of 'info' and stopping of the port sensing work. Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-02 21:26:14 +07:00
err_out:
mutex_unlock(&set_port_type_mutex);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
return err ? err : count;
}
mlx4_core: Allow dynamic MTU configuration for IB ports Set the MTU for IB ports in the driver instead of using the firmware default of 2KB (the driver defaults to 4KB). Allow for dynamic mtu configuration through a new, per-port sysfs entry. Since there's a dependency between the port MTU and the max number of HW VLs the port can support, apply a mim/max approach, using a loop that goes down from the highest possible number of VLs to the lowest, using the firmware return status to know whether the requested number of VLs is possible with a given MTU. For now, as with the dynamic link type change / VPI support, the sysfs entry to change the mtu is exposed only when NOT running in SR-IOV mode. To allow changing the MTU for the master in SR-IOV mode, primary-function-initiated FLR (Function Level Reset) needs to be implemented. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-01-12 00:02:17 +07:00
enum ibta_mtu {
IB_MTU_256 = 1,
IB_MTU_512 = 2,
IB_MTU_1024 = 3,
IB_MTU_2048 = 4,
IB_MTU_4096 = 5
};
static inline int int_to_ibta_mtu(int mtu)
{
switch (mtu) {
case 256: return IB_MTU_256;
case 512: return IB_MTU_512;
case 1024: return IB_MTU_1024;
case 2048: return IB_MTU_2048;
case 4096: return IB_MTU_4096;
default: return -1;
}
}
static inline int ibta_mtu_to_int(enum ibta_mtu mtu)
{
switch (mtu) {
case IB_MTU_256: return 256;
case IB_MTU_512: return 512;
case IB_MTU_1024: return 1024;
case IB_MTU_2048: return 2048;
case IB_MTU_4096: return 4096;
default: return -1;
}
}
static ssize_t show_port_ib_mtu(struct device *dev,
struct device_attribute *attr,
char *buf)
{
struct mlx4_port_info *info = container_of(attr, struct mlx4_port_info,
port_mtu_attr);
struct mlx4_dev *mdev = info->dev;
if (mdev->caps.port_type[info->port] == MLX4_PORT_TYPE_ETH)
mlx4_warn(mdev, "port level mtu is only used for IB ports\n");
sprintf(buf, "%d\n",
ibta_mtu_to_int(mdev->caps.port_ib_mtu[info->port]));
return strlen(buf);
}
static ssize_t set_port_ib_mtu(struct device *dev,
struct device_attribute *attr,
const char *buf, size_t count)
{
struct mlx4_port_info *info = container_of(attr, struct mlx4_port_info,
port_mtu_attr);
struct mlx4_dev *mdev = info->dev;
struct mlx4_priv *priv = mlx4_priv(mdev);
int err, port, mtu, ibta_mtu = -1;
if (mdev->caps.port_type[info->port] == MLX4_PORT_TYPE_ETH) {
mlx4_warn(mdev, "port level mtu is only used for IB ports\n");
return -EINVAL;
}
net/mlx4_core: Replace sscanf() with kstrtoint() It is not safe to use sscanf. Signed-off-by: Dotan Barak <dotanb@dev.mellanox.com> Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-25 16:09:36 +07:00
err = kstrtoint(buf, 0, &mtu);
if (!err)
mlx4_core: Allow dynamic MTU configuration for IB ports Set the MTU for IB ports in the driver instead of using the firmware default of 2KB (the driver defaults to 4KB). Allow for dynamic mtu configuration through a new, per-port sysfs entry. Since there's a dependency between the port MTU and the max number of HW VLs the port can support, apply a mim/max approach, using a loop that goes down from the highest possible number of VLs to the lowest, using the firmware return status to know whether the requested number of VLs is possible with a given MTU. For now, as with the dynamic link type change / VPI support, the sysfs entry to change the mtu is exposed only when NOT running in SR-IOV mode. To allow changing the MTU for the master in SR-IOV mode, primary-function-initiated FLR (Function Level Reset) needs to be implemented. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-01-12 00:02:17 +07:00
ibta_mtu = int_to_ibta_mtu(mtu);
net/mlx4_core: Replace sscanf() with kstrtoint() It is not safe to use sscanf. Signed-off-by: Dotan Barak <dotanb@dev.mellanox.com> Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-25 16:09:36 +07:00
if (err || ibta_mtu < 0) {
mlx4_core: Allow dynamic MTU configuration for IB ports Set the MTU for IB ports in the driver instead of using the firmware default of 2KB (the driver defaults to 4KB). Allow for dynamic mtu configuration through a new, per-port sysfs entry. Since there's a dependency between the port MTU and the max number of HW VLs the port can support, apply a mim/max approach, using a loop that goes down from the highest possible number of VLs to the lowest, using the firmware return status to know whether the requested number of VLs is possible with a given MTU. For now, as with the dynamic link type change / VPI support, the sysfs entry to change the mtu is exposed only when NOT running in SR-IOV mode. To allow changing the MTU for the master in SR-IOV mode, primary-function-initiated FLR (Function Level Reset) needs to be implemented. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-01-12 00:02:17 +07:00
mlx4_err(mdev, "%s is invalid IBTA mtu\n", buf);
return -EINVAL;
}
mdev->caps.port_ib_mtu[info->port] = ibta_mtu;
mlx4_stop_sense(mdev);
mutex_lock(&priv->port_mutex);
mlx4_unregister_device(mdev);
for (port = 1; port <= mdev->caps.num_ports; port++) {
mlx4_CLOSE_PORT(mdev, port);
mlx4: Put physical GID and P_Key table sizes in mlx4_phys_caps struct and paravirtualize them To allow easy paravirtualization of P_Key and GID table sizes, keep paravirtualized sizes in mlx4_dev->caps, but save the actual physical sizes from FW in struct: mlx4_dev->phys_cap. In addition, in SR-IOV mode, do the following: 1. Reduce reported P_Key table size by 1. This is done to reserve the highest P_Key index for internal use, for declaring an invalid P_Key in P_Key paravirtualization. We require a P_Key index which always contain an invalid P_Key value for this purpose (i.e., one which cannot be modified by the subnet manager). The way to do this is to reduce the P_Key table size reported to the subnet manager by 1, so that it will not attempt to access the P_Key at index #127. 2. Paravirtualize the GID table size to 1. Thus, each guest sees only a single GID (at its paravirtualized index 0). In addition, since we are paravirtualizing the GID table size to 1, we add paravirtualization of the master GID event here (i.e., we do not do ib_dispatch_event() for the GUID change event on the master, since its (only) GUID never changes). Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:44 +07:00
err = mlx4_SET_PORT(mdev, port, -1);
mlx4_core: Allow dynamic MTU configuration for IB ports Set the MTU for IB ports in the driver instead of using the firmware default of 2KB (the driver defaults to 4KB). Allow for dynamic mtu configuration through a new, per-port sysfs entry. Since there's a dependency between the port MTU and the max number of HW VLs the port can support, apply a mim/max approach, using a loop that goes down from the highest possible number of VLs to the lowest, using the firmware return status to know whether the requested number of VLs is possible with a given MTU. For now, as with the dynamic link type change / VPI support, the sysfs entry to change the mtu is exposed only when NOT running in SR-IOV mode. To allow changing the MTU for the master in SR-IOV mode, primary-function-initiated FLR (Function Level Reset) needs to be implemented. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-01-12 00:02:17 +07:00
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(mdev, "Failed to set port %d, aborting\n",
port);
mlx4_core: Allow dynamic MTU configuration for IB ports Set the MTU for IB ports in the driver instead of using the firmware default of 2KB (the driver defaults to 4KB). Allow for dynamic mtu configuration through a new, per-port sysfs entry. Since there's a dependency between the port MTU and the max number of HW VLs the port can support, apply a mim/max approach, using a loop that goes down from the highest possible number of VLs to the lowest, using the firmware return status to know whether the requested number of VLs is possible with a given MTU. For now, as with the dynamic link type change / VPI support, the sysfs entry to change the mtu is exposed only when NOT running in SR-IOV mode. To allow changing the MTU for the master in SR-IOV mode, primary-function-initiated FLR (Function Level Reset) needs to be implemented. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-01-12 00:02:17 +07:00
goto err_set_port;
}
}
err = mlx4_register_device(mdev);
err_set_port:
mutex_unlock(&priv->port_mutex);
mlx4_start_sense(mdev);
return err ? err : count;
}
net/mlx4_core: Port aggregation upper layer interface Supply interface functions to bond and unbond ports of a mlx4 internal interfaces. Example for such an interface is the one registered by the mlx4 IB driver under RoCE. There are 1. Functions to go in/out to/from bonded mode 2. Function to remap virtual ports to physical ports The bond_mutex prevents simultaneous access to data that keep status of the device in bonded mode. The upper mlx4 interface marks to the mlx4 core module that they want to be subject for such bonding by setting the MLX4_INTFF_BONDING flag. Interface which goes to/from bonded mode is re-created. The mlx4 Ethernet driver does not set this flag when registering the interface, the IB driver does. Signed-off-by: Moni Shoua <monis@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-03 21:48:33 +07:00
int mlx4_bond(struct mlx4_dev *dev)
{
int ret = 0;
struct mlx4_priv *priv = mlx4_priv(dev);
mutex_lock(&priv->bond_mutex);
if (!mlx4_is_bonded(dev))
ret = mlx4_do_bond(dev, true);
else
ret = 0;
mutex_unlock(&priv->bond_mutex);
if (ret)
mlx4_err(dev, "Failed to bond device: %d\n", ret);
else
mlx4_dbg(dev, "Device is bonded\n");
return ret;
}
EXPORT_SYMBOL_GPL(mlx4_bond);
int mlx4_unbond(struct mlx4_dev *dev)
{
int ret = 0;
struct mlx4_priv *priv = mlx4_priv(dev);
mutex_lock(&priv->bond_mutex);
if (mlx4_is_bonded(dev))
ret = mlx4_do_bond(dev, false);
mutex_unlock(&priv->bond_mutex);
if (ret)
mlx4_err(dev, "Failed to unbond device: %d\n", ret);
else
mlx4_dbg(dev, "Device is unbonded\n");
return ret;
}
EXPORT_SYMBOL_GPL(mlx4_unbond);
int mlx4_port_map_set(struct mlx4_dev *dev, struct mlx4_port_map *v2p)
{
u8 port1 = v2p->port1;
u8 port2 = v2p->port2;
struct mlx4_priv *priv = mlx4_priv(dev);
int err;
if (!(dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_PORT_REMAP))
return -ENOTSUPP;
mutex_lock(&priv->bond_mutex);
/* zero means keep current mapping for this port */
if (port1 == 0)
port1 = priv->v2p.port1;
if (port2 == 0)
port2 = priv->v2p.port2;
if ((port1 < 1) || (port1 > MLX4_MAX_PORTS) ||
(port2 < 1) || (port2 > MLX4_MAX_PORTS) ||
(port1 == 2 && port2 == 1)) {
/* besides boundary checks cross mapping makes
* no sense and therefore not allowed */
err = -EINVAL;
} else if ((port1 == priv->v2p.port1) &&
(port2 == priv->v2p.port2)) {
err = 0;
} else {
err = mlx4_virt2phy_port_map(dev, port1, port2);
if (!err) {
mlx4_dbg(dev, "port map changed: [%d][%d]\n",
port1, port2);
priv->v2p.port1 = port1;
priv->v2p.port2 = port2;
} else {
mlx4_err(dev, "Failed to change port mape: %d\n", err);
}
}
mutex_unlock(&priv->bond_mutex);
return err;
}
EXPORT_SYMBOL_GPL(mlx4_port_map_set);
mlx4_core: Fix more section mismatches Commit 3d73c288 ("mlx4_core: Fix section mismatches") fixed some of the section mismatches introduced when error recovery was added, but there were still more cases of errory recovery code calling into __devinit code from regular .text. Fix this by getting rid of the now-incorrect __devinit annotations. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-02-05 11:20:41 +07:00
static int mlx4_load_fw(struct mlx4_dev *dev)
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
{
struct mlx4_priv *priv = mlx4_priv(dev);
int err;
priv->fw.fw_icm = mlx4_alloc_icm(dev, priv->fw.fw_pages,
mlx4_core: Support ICM tables in coherent memory Enable having ICM tables in coherent memory, and use coherent memory for the dMPT table. This will allow writing MPT entries for MRs both via the SW2HW_MPT command and also directly by the driver for FMR remapping without needing to flush or worry about cacheline boundaries. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-01 16:28:20 +07:00
GFP_HIGHUSER | __GFP_NOWARN, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (!priv->fw.fw_icm) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Couldn't allocate FW area, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return -ENOMEM;
}
err = mlx4_MAP_FA(dev, priv->fw.fw_icm);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "MAP_FA command failed, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_free;
}
err = mlx4_RUN_FW(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "RUN_FW command failed, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_unmap_fa;
}
return 0;
err_unmap_fa:
mlx4_UNMAP_FA(dev);
err_free:
mlx4_core: Support ICM tables in coherent memory Enable having ICM tables in coherent memory, and use coherent memory for the dMPT table. This will allow writing MPT entries for MRs both via the SW2HW_MPT command and also directly by the driver for FMR remapping without needing to flush or worry about cacheline boundaries. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-01 16:28:20 +07:00
mlx4_free_icm(dev, priv->fw.fw_icm, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return err;
}
mlx4_core: Fix more section mismatches Commit 3d73c288 ("mlx4_core: Fix section mismatches") fixed some of the section mismatches introduced when error recovery was added, but there were still more cases of errory recovery code calling into __devinit code from regular .text. Fix this by getting rid of the now-incorrect __devinit annotations. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-02-05 11:20:41 +07:00
static int mlx4_init_cmpt_table(struct mlx4_dev *dev, u64 cmpt_base,
int cmpt_entry_sz)
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
{
struct mlx4_priv *priv = mlx4_priv(dev);
int err;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
int num_eqs;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err = mlx4_init_icm_table(dev, &priv->qp_table.cmpt_table,
cmpt_base +
((u64) (MLX4_CMPT_TYPE_QP *
cmpt_entry_sz) << MLX4_CMPT_SHIFT),
cmpt_entry_sz, dev->caps.num_qps,
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW],
0, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err)
goto err;
err = mlx4_init_icm_table(dev, &priv->srq_table.cmpt_table,
cmpt_base +
((u64) (MLX4_CMPT_TYPE_SRQ *
cmpt_entry_sz) << MLX4_CMPT_SHIFT),
cmpt_entry_sz, dev->caps.num_srqs,
mlx4_core: Support ICM tables in coherent memory Enable having ICM tables in coherent memory, and use coherent memory for the dMPT table. This will allow writing MPT entries for MRs both via the SW2HW_MPT command and also directly by the driver for FMR remapping without needing to flush or worry about cacheline boundaries. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-01 16:28:20 +07:00
dev->caps.reserved_srqs, 0, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err)
goto err_qp;
err = mlx4_init_icm_table(dev, &priv->cq_table.cmpt_table,
cmpt_base +
((u64) (MLX4_CMPT_TYPE_CQ *
cmpt_entry_sz) << MLX4_CMPT_SHIFT),
cmpt_entry_sz, dev->caps.num_cqs,
mlx4_core: Support ICM tables in coherent memory Enable having ICM tables in coherent memory, and use coherent memory for the dMPT table. This will allow writing MPT entries for MRs both via the SW2HW_MPT command and also directly by the driver for FMR remapping without needing to flush or worry about cacheline boundaries. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-01 16:28:20 +07:00
dev->caps.reserved_cqs, 0, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err)
goto err_srq;
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
num_eqs = dev->phys_caps.num_phys_eqs;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err = mlx4_init_icm_table(dev, &priv->eq_table.cmpt_table,
cmpt_base +
((u64) (MLX4_CMPT_TYPE_EQ *
cmpt_entry_sz) << MLX4_CMPT_SHIFT),
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
cmpt_entry_sz, num_eqs, num_eqs, 0, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err)
goto err_cq;
return 0;
err_cq:
mlx4_cleanup_icm_table(dev, &priv->cq_table.cmpt_table);
err_srq:
mlx4_cleanup_icm_table(dev, &priv->srq_table.cmpt_table);
err_qp:
mlx4_cleanup_icm_table(dev, &priv->qp_table.cmpt_table);
err:
return err;
}
mlx4_core: Fix section mismatches Commit ee49bd93 ("mlx4_core: Reset device when internal error is detected") introduced some section mismatch problems when CONFIG_HOTPLUG=n, because the error recovery code tears down and reinitializes the device after everything is loaded, which ends up calling into lots of code marked __devinit and __devexit from regular .text. Fix this by getting rid of these now-incorrect section markers. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-10-11 05:43:54 +07:00
static int mlx4_init_icm(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
struct mlx4_init_hca_param *init_hca, u64 icm_size)
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
{
struct mlx4_priv *priv = mlx4_priv(dev);
u64 aux_pages;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
int num_eqs;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
int err;
err = mlx4_SET_ICM_SIZE(dev, icm_size, &aux_pages);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "SET_ICM_SIZE command failed, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return err;
}
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_dbg(dev, "%lld KB of HCA context requires %lld KB aux memory\n",
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
(unsigned long long) icm_size >> 10,
(unsigned long long) aux_pages << 2);
priv->fw.aux_icm = mlx4_alloc_icm(dev, aux_pages,
mlx4_core: Support ICM tables in coherent memory Enable having ICM tables in coherent memory, and use coherent memory for the dMPT table. This will allow writing MPT entries for MRs both via the SW2HW_MPT command and also directly by the driver for FMR remapping without needing to flush or worry about cacheline boundaries. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-01 16:28:20 +07:00
GFP_HIGHUSER | __GFP_NOWARN, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (!priv->fw.aux_icm) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Couldn't allocate aux memory, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return -ENOMEM;
}
err = mlx4_MAP_ICM_AUX(dev, priv->fw.aux_icm);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "MAP_ICM_AUX command failed, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_free_aux;
}
err = mlx4_init_cmpt_table(dev, init_hca->cmpt_base, dev_cap->cmpt_entry_sz);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to map cMPT context memory, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_unmap_aux;
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
num_eqs = dev->phys_caps.num_phys_eqs;
mlx4_core: Allocate and map sufficient ICM memory for EQ context The current implementation allocates a single host page for EQ context memory, which was OK when we only allocated a few EQs. However, since we now allocate an EQ for each CPU core, this patch removes the hard-coded limit (which we exceed with 4 KB pages and 128 byte EQ context entries with 32 CPUs) and uses the same ICM table code as all other context tables, which ends up simplifying the code quite a bit while fixing the problem. This problem was actually hit in practice on a dual-socket Nehalem box with 16 real hardware threads and sufficiently odd ACPI tables that it shows on boot SMP: Allowing 32 CPUs, 16 hotplug CPUs so num_possible_cpus() ends up 32, and mlx4 ends up creating 33 MSI-X interrupts and 33 EQs. This mlx4 bug means that mlx4 can't even initialize at all on this quite mainstream system. Cc: <stable@kernel.org> Reported-by: Eli Cohen <eli@mellanox.co.il> Tested-by: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-09-06 10:24:49 +07:00
err = mlx4_init_icm_table(dev, &priv->eq_table.table,
init_hca->eqc_base, dev_cap->eqc_entry_sz,
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
num_eqs, num_eqs, 0, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to map EQ context memory, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_unmap_cmpt;
}
mlx4_core: Write MTTs from CPU instead with of WRITE_MTT FW command Write MTT entries directly to ICM from the driver (eliminating use of WRITE_MTT command). This reduces the number of FW commands needed to register an MR by at least a factor of 2 and speeds up memory registration significantly. This code will also be used to implement FMRs. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-01 16:28:53 +07:00
/*
* Reserved MTT entries must be aligned up to a cacheline
* boundary, since the FW will write to them, while the driver
* writes to all other MTT entries. (The variable
* dev->caps.mtt_entry_sz below is really the MTT segment
* size, not the raw entry size)
*/
dev->caps.reserved_mtts =
ALIGN(dev->caps.reserved_mtts * dev->caps.mtt_entry_sz,
dma_get_cache_alignment()) / dev->caps.mtt_entry_sz;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err = mlx4_init_icm_table(dev, &priv->mr_table.mtt_table,
init_hca->mtt_base,
dev->caps.mtt_entry_sz,
mlx4_core: mtts resources units changed to offset In the previous implementation mtts are managed by: 1. order - log(mtt segments), 'mtt segment' groups several mtts together. 2. first_seg - segment location relative to mtt table. In the current implementation: 1. order - log(mtts) rather than segments 2. offset - mtt index in mtt table Note: The actual mtt allocation is made in segments but it is transparent to callers. Rational: The mtt resource holders are not interested on how the allocation of mtt is done, but rather on how they will use it. Signed-off-by: Marcel Apfelbaum <marcela@dev.mellanox.co.il> Reviewed-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:16:56 +07:00
dev->caps.num_mtts,
mlx4_core: Support ICM tables in coherent memory Enable having ICM tables in coherent memory, and use coherent memory for the dMPT table. This will allow writing MPT entries for MRs both via the SW2HW_MPT command and also directly by the driver for FMR remapping without needing to flush or worry about cacheline boundaries. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-01 16:28:20 +07:00
dev->caps.reserved_mtts, 1, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to map MTT context memory, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_unmap_eq;
}
err = mlx4_init_icm_table(dev, &priv->mr_table.dmpt_table,
init_hca->dmpt_base,
dev_cap->dmpt_entry_sz,
dev->caps.num_mpts,
mlx4_core: Support ICM tables in coherent memory Enable having ICM tables in coherent memory, and use coherent memory for the dMPT table. This will allow writing MPT entries for MRs both via the SW2HW_MPT command and also directly by the driver for FMR remapping without needing to flush or worry about cacheline boundaries. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-01 16:28:20 +07:00
dev->caps.reserved_mrws, 1, 1);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to map dMPT context memory, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_unmap_mtt;
}
err = mlx4_init_icm_table(dev, &priv->qp_table.qp_table,
init_hca->qpc_base,
dev_cap->qpc_entry_sz,
dev->caps.num_qps,
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW],
0, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to map QP context memory, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_unmap_dmpt;
}
err = mlx4_init_icm_table(dev, &priv->qp_table.auxc_table,
init_hca->auxc_base,
dev_cap->aux_entry_sz,
dev->caps.num_qps,
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW],
0, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to map AUXC context memory, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_unmap_qp;
}
err = mlx4_init_icm_table(dev, &priv->qp_table.altc_table,
init_hca->altc_base,
dev_cap->altc_entry_sz,
dev->caps.num_qps,
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW],
0, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to map ALTC context memory, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_unmap_auxc;
}
err = mlx4_init_icm_table(dev, &priv->qp_table.rdmarc_table,
init_hca->rdmarc_base,
dev_cap->rdmarc_entry_sz << priv->qp_table.rdmarc_shift,
dev->caps.num_qps,
mlx4_core: Support multiple pre-reserved QP regions For ethernet support, we need to reserve QPs for the ethernet and fibre channel driver. The QPs are reserved at the end of the QP table. (This way we assure that they are aligned to their size) We need to consider these reserved ranges in bitmap creation, so we extend the mlx4 bitmap utility functions to allow reserved ranges at both the bottom and the top of the range. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 00:25:29 +07:00
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW],
0, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err) {
mlx4_err(dev, "Failed to map RDMARC context memory, aborting\n");
goto err_unmap_altc;
}
err = mlx4_init_icm_table(dev, &priv->cq_table.table,
init_hca->cqc_base,
dev_cap->cqc_entry_sz,
dev->caps.num_cqs,
mlx4_core: Support ICM tables in coherent memory Enable having ICM tables in coherent memory, and use coherent memory for the dMPT table. This will allow writing MPT entries for MRs both via the SW2HW_MPT command and also directly by the driver for FMR remapping without needing to flush or worry about cacheline boundaries. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-01 16:28:20 +07:00
dev->caps.reserved_cqs, 0, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to map CQ context memory, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_unmap_rdmarc;
}
err = mlx4_init_icm_table(dev, &priv->srq_table.table,
init_hca->srqc_base,
dev_cap->srq_entry_sz,
dev->caps.num_srqs,
mlx4_core: Support ICM tables in coherent memory Enable having ICM tables in coherent memory, and use coherent memory for the dMPT table. This will allow writing MPT entries for MRs both via the SW2HW_MPT command and also directly by the driver for FMR remapping without needing to flush or worry about cacheline boundaries. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-01 16:28:20 +07:00
dev->caps.reserved_srqs, 0, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to map SRQ context memory, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_unmap_cq;
}
/*
{NET, IB}/mlx4: Add device managed flow steering firmware API The driver is modified to support three operation modes. If supported by firmware use the device managed flow steering API, that which we call device managed steering mode. Else, if the firmware supports the B0 steering mode use it, and finally, if none of the above, use the A0 steering mode. When the steering mode is device managed, the code is modified such that L2 based rules set by the mlx4_en driver for Ethernet unicast and multicast, and the IB stack multicast attach calls done through the mlx4_ib driver are all routed to use the device managed API. When attaching rule using device managed flow steering API, the firmware returns a 64 bit registration id, which is to be provided during detach. Currently the firmware is always programmed during HCA initialization to use standard L2 hashing. Future work should be done to allow configuring the flow-steering hash function with common, non proprietary means. Signed-off-by: Hadar Hen Zion <hadarh@mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-05 11:03:46 +07:00
* For flow steering device managed mode it is required to use
* mlx4_init_icm_table. For B0 steering mode it's not strictly
* required, but for simplicity just map the whole multicast
* group table now. The table isn't very big and it's a lot
* easier than trying to track ref counts.
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
*/
err = mlx4_init_icm_table(dev, &priv->mcg_table.table,
mlx4: Traffic steering management support for SRIOV Let multicast/unicast attaching flow go through resource tracker. The PF is the one responsible for managing all the steering entries. Define and use module parameter that determines the number of qps per multicast group. Minor changes in function calls according to changed prototype. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:16:02 +07:00
init_hca->mc_base,
mlx4_get_mgm_entry_size(dev),
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
dev->caps.num_mgms + dev->caps.num_amgms,
dev->caps.num_mgms + dev->caps.num_amgms,
mlx4_core: Support ICM tables in coherent memory Enable having ICM tables in coherent memory, and use coherent memory for the dMPT table. This will allow writing MPT entries for MRs both via the SW2HW_MPT command and also directly by the driver for FMR remapping without needing to flush or worry about cacheline boundaries. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-01 16:28:20 +07:00
0, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to map MCG context memory, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_unmap_srq;
}
return 0;
err_unmap_srq:
mlx4_cleanup_icm_table(dev, &priv->srq_table.table);
err_unmap_cq:
mlx4_cleanup_icm_table(dev, &priv->cq_table.table);
err_unmap_rdmarc:
mlx4_cleanup_icm_table(dev, &priv->qp_table.rdmarc_table);
err_unmap_altc:
mlx4_cleanup_icm_table(dev, &priv->qp_table.altc_table);
err_unmap_auxc:
mlx4_cleanup_icm_table(dev, &priv->qp_table.auxc_table);
err_unmap_qp:
mlx4_cleanup_icm_table(dev, &priv->qp_table.qp_table);
err_unmap_dmpt:
mlx4_cleanup_icm_table(dev, &priv->mr_table.dmpt_table);
err_unmap_mtt:
mlx4_cleanup_icm_table(dev, &priv->mr_table.mtt_table);
err_unmap_eq:
mlx4_core: Allocate and map sufficient ICM memory for EQ context The current implementation allocates a single host page for EQ context memory, which was OK when we only allocated a few EQs. However, since we now allocate an EQ for each CPU core, this patch removes the hard-coded limit (which we exceed with 4 KB pages and 128 byte EQ context entries with 32 CPUs) and uses the same ICM table code as all other context tables, which ends up simplifying the code quite a bit while fixing the problem. This problem was actually hit in practice on a dual-socket Nehalem box with 16 real hardware threads and sufficiently odd ACPI tables that it shows on boot SMP: Allowing 32 CPUs, 16 hotplug CPUs so num_possible_cpus() ends up 32, and mlx4 ends up creating 33 MSI-X interrupts and 33 EQs. This mlx4 bug means that mlx4 can't even initialize at all on this quite mainstream system. Cc: <stable@kernel.org> Reported-by: Eli Cohen <eli@mellanox.co.il> Tested-by: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-09-06 10:24:49 +07:00
mlx4_cleanup_icm_table(dev, &priv->eq_table.table);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err_unmap_cmpt:
mlx4_cleanup_icm_table(dev, &priv->eq_table.cmpt_table);
mlx4_cleanup_icm_table(dev, &priv->cq_table.cmpt_table);
mlx4_cleanup_icm_table(dev, &priv->srq_table.cmpt_table);
mlx4_cleanup_icm_table(dev, &priv->qp_table.cmpt_table);
err_unmap_aux:
mlx4_UNMAP_ICM_AUX(dev);
err_free_aux:
mlx4_core: Support ICM tables in coherent memory Enable having ICM tables in coherent memory, and use coherent memory for the dMPT table. This will allow writing MPT entries for MRs both via the SW2HW_MPT command and also directly by the driver for FMR remapping without needing to flush or worry about cacheline boundaries. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-01 16:28:20 +07:00
mlx4_free_icm(dev, priv->fw.aux_icm, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return err;
}
static void mlx4_free_icms(struct mlx4_dev *dev)
{
struct mlx4_priv *priv = mlx4_priv(dev);
mlx4_cleanup_icm_table(dev, &priv->mcg_table.table);
mlx4_cleanup_icm_table(dev, &priv->srq_table.table);
mlx4_cleanup_icm_table(dev, &priv->cq_table.table);
mlx4_cleanup_icm_table(dev, &priv->qp_table.rdmarc_table);
mlx4_cleanup_icm_table(dev, &priv->qp_table.altc_table);
mlx4_cleanup_icm_table(dev, &priv->qp_table.auxc_table);
mlx4_cleanup_icm_table(dev, &priv->qp_table.qp_table);
mlx4_cleanup_icm_table(dev, &priv->mr_table.dmpt_table);
mlx4_cleanup_icm_table(dev, &priv->mr_table.mtt_table);
mlx4_core: Allocate and map sufficient ICM memory for EQ context The current implementation allocates a single host page for EQ context memory, which was OK when we only allocated a few EQs. However, since we now allocate an EQ for each CPU core, this patch removes the hard-coded limit (which we exceed with 4 KB pages and 128 byte EQ context entries with 32 CPUs) and uses the same ICM table code as all other context tables, which ends up simplifying the code quite a bit while fixing the problem. This problem was actually hit in practice on a dual-socket Nehalem box with 16 real hardware threads and sufficiently odd ACPI tables that it shows on boot SMP: Allowing 32 CPUs, 16 hotplug CPUs so num_possible_cpus() ends up 32, and mlx4 ends up creating 33 MSI-X interrupts and 33 EQs. This mlx4 bug means that mlx4 can't even initialize at all on this quite mainstream system. Cc: <stable@kernel.org> Reported-by: Eli Cohen <eli@mellanox.co.il> Tested-by: Christoph Lameter <cl@linux-foundation.org> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-09-06 10:24:49 +07:00
mlx4_cleanup_icm_table(dev, &priv->eq_table.table);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_cleanup_icm_table(dev, &priv->eq_table.cmpt_table);
mlx4_cleanup_icm_table(dev, &priv->cq_table.cmpt_table);
mlx4_cleanup_icm_table(dev, &priv->srq_table.cmpt_table);
mlx4_cleanup_icm_table(dev, &priv->qp_table.cmpt_table);
mlx4_UNMAP_ICM_AUX(dev);
mlx4_core: Support ICM tables in coherent memory Enable having ICM tables in coherent memory, and use coherent memory for the dMPT table. This will allow writing MPT entries for MRs both via the SW2HW_MPT command and also directly by the driver for FMR remapping without needing to flush or worry about cacheline boundaries. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-01 16:28:20 +07:00
mlx4_free_icm(dev, priv->fw.aux_icm, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
static void mlx4_slave_exit(struct mlx4_dev *dev)
{
struct mlx4_priv *priv = mlx4_priv(dev);
mlx4_core: Fix crash on uninitialized priv->cmd.slave_sem On an SR-IOV master device, __mlx4_init_one() calls mlx4_init_hca() before mlx4_multi_func_init(). However, for unlucky configurations, mlx4_init_hca() might call mlx4_SENSE_PORT() (via mlx4_dev_cap()), and that calls mlx4_cmd_imm() with MLX4_CMD_WRAPPED set. However, on a multifunction device with MLX4_CMD_WRAPPED, __mlx4_cmd() calls into mlx4_slave_cmd(), and that immediately tries to do down(&priv->cmd.slave_sem); but priv->cmd.slave_sem isn't initialized until mlx4_multi_func_init() (which we haven't called yet). The next thing it tries to do is access priv->mfunc.vhcr, but that hasn't been allocated yet. Fix this by moving the initialization of slave_sem and vhcr up into mlx4_cmd_init(). Also, since slave_sem is really just being used as a mutex, convert it into a slave_cmd_mutex. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-26 11:24:07 +07:00
mutex_lock(&priv->cmd.slave_cmd_mutex);
net/mlx4_core: Reset flow activation upon SRIOV fatal command cases When SRIOV commands are executed over the comm-channel and get a fatal error (e.g. timeout, closing command failure) the VF enters into error state and reset flow is activated. To be able to recognize whether the failure was on a closing command, the operational code for the given VHCR command is used. Once the device entered into an error state we prevent redundant error messages from being printed. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:43 +07:00
if (mlx4_comm_cmd(dev, MLX4_COMM_CMD_RESET, 0, MLX4_COMM_CMD_NA_OP,
MLX4_COMM_TIME))
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_warn(dev, "Failed to close slave function\n");
mlx4_core: Fix crash on uninitialized priv->cmd.slave_sem On an SR-IOV master device, __mlx4_init_one() calls mlx4_init_hca() before mlx4_multi_func_init(). However, for unlucky configurations, mlx4_init_hca() might call mlx4_SENSE_PORT() (via mlx4_dev_cap()), and that calls mlx4_cmd_imm() with MLX4_CMD_WRAPPED set. However, on a multifunction device with MLX4_CMD_WRAPPED, __mlx4_cmd() calls into mlx4_slave_cmd(), and that immediately tries to do down(&priv->cmd.slave_sem); but priv->cmd.slave_sem isn't initialized until mlx4_multi_func_init() (which we haven't called yet). The next thing it tries to do is access priv->mfunc.vhcr, but that hasn't been allocated yet. Fix this by moving the initialization of slave_sem and vhcr up into mlx4_cmd_init(). Also, since slave_sem is really just being used as a mutex, convert it into a slave_cmd_mutex. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-26 11:24:07 +07:00
mutex_unlock(&priv->cmd.slave_cmd_mutex);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
mlx4: Add blue flame support for kernel consumers Using blue flame can improve latency by allowing the HW to more efficiently access the WQE. This patch presents two functions that are used to allocate or release HW resources for using blue flame; the caller need to supply a struct mlx4_bf object when allocating resources. Consumers that make use of this API should post doorbells to the UAR object pointed by the initialized struct mlx4_bf; Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:38:41 +07:00
static int map_bf_area(struct mlx4_dev *dev)
{
struct mlx4_priv *priv = mlx4_priv(dev);
resource_size_t bf_start;
resource_size_t bf_len;
int err = 0;
mlx4_core: Do not map BF area if capability is 0 BF can be disabled in some cases, the capability field, bf_reg_size is set to zero in this case. Don't map the BF area in this case, it would cause failures. In addition, leaving the BF area unmapped also alerts the ETH driver to not use BF. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-20 04:38:52 +07:00
if (!dev->caps.bf_reg_size)
return -ENXIO;
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
bf_start = pci_resource_start(dev->persist->pdev, 2) +
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
(dev->caps.num_uars << PAGE_SHIFT);
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
bf_len = pci_resource_len(dev->persist->pdev, 2) -
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
(dev->caps.num_uars << PAGE_SHIFT);
mlx4: Add blue flame support for kernel consumers Using blue flame can improve latency by allowing the HW to more efficiently access the WQE. This patch presents two functions that are used to allocate or release HW resources for using blue flame; the caller need to supply a struct mlx4_bf object when allocating resources. Consumers that make use of this API should post doorbells to the UAR object pointed by the initialized struct mlx4_bf; Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:38:41 +07:00
priv->bf_mapping = io_mapping_create_wc(bf_start, bf_len);
if (!priv->bf_mapping)
err = -ENOMEM;
return err;
}
static void unmap_bf_area(struct mlx4_dev *dev)
{
if (mlx4_priv(dev)->bf_mapping)
io_mapping_free(mlx4_priv(dev)->bf_mapping);
}
net/mlx4_en: Add HW timestamping (TS) support The patch allows to enable/disable HW timestamping for incoming and/or outgoing packets. It adds and initializes all structs and callbacks needed by kernel TS API. To enable/disable HW timestamping appropriate ioctl should be used. Currently HWTSTAMP_FILTER_ALL/NONE and HWTSAMP_TX_ON/OFF only are supported. When enabling TS on receive flow - VLAN stripping will be disabled. Also were made all relevant changes in RX/TX flows to consider TS request and plant HW timestamps into relevant structures. mlx4_ib was fixed to compile with new mlx4_cq_alloc() signature. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-23 13:06:49 +07:00
cycle_t mlx4_read_clock(struct mlx4_dev *dev)
{
u32 clockhi, clocklo, clockhi1;
cycle_t cycles;
int i;
struct mlx4_priv *priv = mlx4_priv(dev);
for (i = 0; i < 10; i++) {
clockhi = swab32(readl(priv->clock_mapping));
clocklo = swab32(readl(priv->clock_mapping + 4));
clockhi1 = swab32(readl(priv->clock_mapping));
if (clockhi == clockhi1)
break;
}
cycles = (u64) clockhi << 32 | (u64) clocklo;
return cycles;
}
EXPORT_SYMBOL_GPL(mlx4_read_clock);
net/mlx4_core: Read HCA frequency and map internal clock Read HCA frequency, read PCI clock bar and offset, map internal clock to PCI bar. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-23 13:06:48 +07:00
static int map_internal_clock(struct mlx4_dev *dev)
{
struct mlx4_priv *priv = mlx4_priv(dev);
priv->clock_mapping =
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
ioremap(pci_resource_start(dev->persist->pdev,
priv->fw.clock_bar) +
net/mlx4_core: Read HCA frequency and map internal clock Read HCA frequency, read PCI clock bar and offset, map internal clock to PCI bar. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-23 13:06:48 +07:00
priv->fw.clock_offset, MLX4_CLOCK_SIZE);
if (!priv->clock_mapping)
return -ENOMEM;
return 0;
}
IB/mlx4: Add mmap call to map the hardware clock In order to read the HCA's cycle counter efficiently in user space, we need to map the HCA's register. This is done through mmap call. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2015-06-11 20:35:26 +07:00
int mlx4_get_internal_clock_params(struct mlx4_dev *dev,
struct mlx4_clock_params *params)
{
struct mlx4_priv *priv = mlx4_priv(dev);
if (mlx4_is_slave(dev))
return -ENOTSUPP;
if (!params)
return -EINVAL;
params->bar = priv->fw.clock_bar;
params->offset = priv->fw.clock_offset;
params->size = MLX4_CLOCK_SIZE;
return 0;
}
EXPORT_SYMBOL_GPL(mlx4_get_internal_clock_params);
net/mlx4_core: Read HCA frequency and map internal clock Read HCA frequency, read PCI clock bar and offset, map internal clock to PCI bar. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-23 13:06:48 +07:00
static void unmap_internal_clock(struct mlx4_dev *dev)
{
struct mlx4_priv *priv = mlx4_priv(dev);
if (priv->clock_mapping)
iounmap(priv->clock_mapping);
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
static void mlx4_close_hca(struct mlx4_dev *dev)
{
net/mlx4_core: Read HCA frequency and map internal clock Read HCA frequency, read PCI clock bar and offset, map internal clock to PCI bar. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-23 13:06:48 +07:00
unmap_internal_clock(dev);
mlx4: Add blue flame support for kernel consumers Using blue flame can improve latency by allowing the HW to more efficiently access the WQE. This patch presents two functions that are used to allocate or release HW resources for using blue flame; the caller need to supply a struct mlx4_bf object when allocating resources. Consumers that make use of this API should post doorbells to the UAR object pointed by the initialized struct mlx4_bf; Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:38:41 +07:00
unmap_bf_area(dev);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if (mlx4_is_slave(dev))
mlx4_slave_exit(dev);
else {
mlx4_CLOSE_HCA(dev, 0);
mlx4_free_icms(dev);
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
}
}
static void mlx4_close_fw(struct mlx4_dev *dev)
{
if (!mlx4_is_slave(dev)) {
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
mlx4_UNMAP_FA(dev);
mlx4_free_icm(dev, mlx4_priv(dev)->fw.fw_icm, 0);
}
}
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
static int mlx4_comm_check_offline(struct mlx4_dev *dev)
{
#define COMM_CHAN_OFFLINE_OFFSET 0x09
u32 comm_flags;
u32 offline_bit;
unsigned long end;
struct mlx4_priv *priv = mlx4_priv(dev);
end = msecs_to_jiffies(MLX4_COMM_OFFLINE_TIME_OUT) + jiffies;
while (time_before(jiffies, end)) {
comm_flags = swab32(readl((__iomem char *)priv->mfunc.comm +
MLX4_COMM_CHAN_FLAGS));
offline_bit = (comm_flags &
(u32)(1 << COMM_CHAN_OFFLINE_OFFSET));
if (!offline_bit)
return 0;
/* There are cases as part of AER/Reset flow that PF needs
* around 100 msec to load. We therefore sleep for 100 msec
* to allow other tasks to make use of that CPU during this
* time interval.
*/
msleep(100);
}
mlx4_err(dev, "Communication channel is offline.\n");
return -EIO;
}
static void mlx4_reset_vf_support(struct mlx4_dev *dev)
{
#define COMM_CHAN_RST_OFFSET 0x1e
struct mlx4_priv *priv = mlx4_priv(dev);
u32 comm_rst;
u32 comm_caps;
comm_caps = swab32(readl((__iomem char *)priv->mfunc.comm +
MLX4_COMM_CHAN_CAPS));
comm_rst = (comm_caps & (u32)(1 << COMM_CHAN_RST_OFFSET));
if (comm_rst)
dev->caps.vf_caps |= MLX4_VF_CAP_FLAG_RESET;
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
static int mlx4_init_slave(struct mlx4_dev *dev)
{
struct mlx4_priv *priv = mlx4_priv(dev);
u64 dma = (u64) priv->mfunc.vhcr_dma;
int ret_from_reset = 0;
u32 slave_read;
u32 cmd_channel_ver;
net/mlx4_core: mlx4_init_slave() shouldn't access comm channel before PF is ready Currently, the PF call to pci_enable_sriov from the PF probe function stalls for 10 seconds times the number of VFs probed on the host. This happens because the way for such VFs to determine of the PF initialization finished, is by attempting to issue reset on the comm-channel and get timeout (after 10s). The PF probe function is called from a kenernel workqueue, and therefore during that time, rcu lock is being held and kernel's workqueue is stalled. This blocks other processes that try to use the workqueue or rcu lock. For example, interface renaming which is calling rcu_synchronize is blocked, and timedout by systemd. Changed mlx4_init_slave() to allow VF probed on the host to immediatly detect that the PF is not ready, and return EPROBE_DEFER instantly. Only when the PF finishes the initialization, allow such VFs to access the comm channel. This issue and fix are relevant only for probed VFs on the hypervisor, there is no way to pass this information to a VM until comm channel is ready, so in a VM, if PF is not ready, the first command will be timedout after 10 seconds and return EPROBE_DEFER. Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-06 23:28:17 +07:00
if (atomic_read(&pf_loading)) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_warn(dev, "PF is not ready - Deferring probe\n");
net/mlx4_core: mlx4_init_slave() shouldn't access comm channel before PF is ready Currently, the PF call to pci_enable_sriov from the PF probe function stalls for 10 seconds times the number of VFs probed on the host. This happens because the way for such VFs to determine of the PF initialization finished, is by attempting to issue reset on the comm-channel and get timeout (after 10s). The PF probe function is called from a kenernel workqueue, and therefore during that time, rcu lock is being held and kernel's workqueue is stalled. This blocks other processes that try to use the workqueue or rcu lock. For example, interface renaming which is calling rcu_synchronize is blocked, and timedout by systemd. Changed mlx4_init_slave() to allow VF probed on the host to immediatly detect that the PF is not ready, and return EPROBE_DEFER instantly. Only when the PF finishes the initialization, allow such VFs to access the comm channel. This issue and fix are relevant only for probed VFs on the hypervisor, there is no way to pass this information to a VM until comm channel is ready, so in a VM, if PF is not ready, the first command will be timedout after 10 seconds and return EPROBE_DEFER. Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-06 23:28:17 +07:00
return -EPROBE_DEFER;
}
mlx4_core: Fix crash on uninitialized priv->cmd.slave_sem On an SR-IOV master device, __mlx4_init_one() calls mlx4_init_hca() before mlx4_multi_func_init(). However, for unlucky configurations, mlx4_init_hca() might call mlx4_SENSE_PORT() (via mlx4_dev_cap()), and that calls mlx4_cmd_imm() with MLX4_CMD_WRAPPED set. However, on a multifunction device with MLX4_CMD_WRAPPED, __mlx4_cmd() calls into mlx4_slave_cmd(), and that immediately tries to do down(&priv->cmd.slave_sem); but priv->cmd.slave_sem isn't initialized until mlx4_multi_func_init() (which we haven't called yet). The next thing it tries to do is access priv->mfunc.vhcr, but that hasn't been allocated yet. Fix this by moving the initialization of slave_sem and vhcr up into mlx4_cmd_init(). Also, since slave_sem is really just being used as a mutex, convert it into a slave_cmd_mutex. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-26 11:24:07 +07:00
mutex_lock(&priv->cmd.slave_cmd_mutex);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
priv->cmd.max_cmds = 1;
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
if (mlx4_comm_check_offline(dev)) {
mlx4_err(dev, "PF is not responsive, skipping initialization\n");
goto err_offline;
}
mlx4_reset_vf_support(dev);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
mlx4_warn(dev, "Sending reset\n");
ret_from_reset = mlx4_comm_cmd(dev, MLX4_COMM_CMD_RESET, 0,
net/mlx4_core: Reset flow activation upon SRIOV fatal command cases When SRIOV commands are executed over the comm-channel and get a fatal error (e.g. timeout, closing command failure) the VF enters into error state and reset flow is activated. To be able to recognize whether the failure was on a closing command, the operational code for the given VHCR command is used. Once the device entered into an error state we prevent redundant error messages from being printed. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:43 +07:00
MLX4_COMM_CMD_NA_OP, MLX4_COMM_TIME);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* if we are in the middle of flr the slave will try
* NUM_OF_RESET_RETRIES times before leaving.*/
if (ret_from_reset) {
if (MLX4_DELAY_RESET_SLAVE == ret_from_reset) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_warn(dev, "slave is currently in the middle of FLR - Deferring probe\n");
net/mlx4_core: Return -EPROBE_DEFER when a VF is probed before PF is sufficiently initialized In the PF initialization, SRIOV is enabled before the PF is fully initialized. This allows the kernel to probe the newly-exposed VFs before the PF is ready to handle them (nested probes). Have the probe method return the -EPROBE_DEFER value in this situation (instead of the VF probe method retrying its initialization in a loop, and returning -EIO on failure). When -EPROBE_DEFER is returned by the VF probe method, the kernel itself will retry the probe after a suitable delay. Based upon a suggestion by Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-04 12:13:27 +07:00
mutex_unlock(&priv->cmd.slave_cmd_mutex);
return -EPROBE_DEFER;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
} else
goto err;
}
/* check the driver version - the slave I/F revision
* must match the master's */
slave_read = swab32(readl(&priv->mfunc.comm->slave_read));
cmd_channel_ver = mlx4_comm_get_version();
if (MLX4_COMM_GET_IF_REV(cmd_channel_ver) !=
MLX4_COMM_GET_IF_REV(slave_read)) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "slave driver version is not supported by the master\n");
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
goto err;
}
mlx4_warn(dev, "Sending vhcr0\n");
if (mlx4_comm_cmd(dev, MLX4_COMM_CMD_VHCR0, dma >> 48,
net/mlx4_core: Reset flow activation upon SRIOV fatal command cases When SRIOV commands are executed over the comm-channel and get a fatal error (e.g. timeout, closing command failure) the VF enters into error state and reset flow is activated. To be able to recognize whether the failure was on a closing command, the operational code for the given VHCR command is used. Once the device entered into an error state we prevent redundant error messages from being printed. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:43 +07:00
MLX4_COMM_CMD_NA_OP, MLX4_COMM_TIME))
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
goto err;
if (mlx4_comm_cmd(dev, MLX4_COMM_CMD_VHCR1, dma >> 32,
net/mlx4_core: Reset flow activation upon SRIOV fatal command cases When SRIOV commands are executed over the comm-channel and get a fatal error (e.g. timeout, closing command failure) the VF enters into error state and reset flow is activated. To be able to recognize whether the failure was on a closing command, the operational code for the given VHCR command is used. Once the device entered into an error state we prevent redundant error messages from being printed. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:43 +07:00
MLX4_COMM_CMD_NA_OP, MLX4_COMM_TIME))
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
goto err;
if (mlx4_comm_cmd(dev, MLX4_COMM_CMD_VHCR2, dma >> 16,
net/mlx4_core: Reset flow activation upon SRIOV fatal command cases When SRIOV commands are executed over the comm-channel and get a fatal error (e.g. timeout, closing command failure) the VF enters into error state and reset flow is activated. To be able to recognize whether the failure was on a closing command, the operational code for the given VHCR command is used. Once the device entered into an error state we prevent redundant error messages from being printed. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:43 +07:00
MLX4_COMM_CMD_NA_OP, MLX4_COMM_TIME))
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
goto err;
net/mlx4_core: Reset flow activation upon SRIOV fatal command cases When SRIOV commands are executed over the comm-channel and get a fatal error (e.g. timeout, closing command failure) the VF enters into error state and reset flow is activated. To be able to recognize whether the failure was on a closing command, the operational code for the given VHCR command is used. Once the device entered into an error state we prevent redundant error messages from being printed. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:43 +07:00
if (mlx4_comm_cmd(dev, MLX4_COMM_CMD_VHCR_EN, dma,
MLX4_COMM_CMD_NA_OP, MLX4_COMM_TIME))
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
goto err;
mlx4_core: Fix crash on uninitialized priv->cmd.slave_sem On an SR-IOV master device, __mlx4_init_one() calls mlx4_init_hca() before mlx4_multi_func_init(). However, for unlucky configurations, mlx4_init_hca() might call mlx4_SENSE_PORT() (via mlx4_dev_cap()), and that calls mlx4_cmd_imm() with MLX4_CMD_WRAPPED set. However, on a multifunction device with MLX4_CMD_WRAPPED, __mlx4_cmd() calls into mlx4_slave_cmd(), and that immediately tries to do down(&priv->cmd.slave_sem); but priv->cmd.slave_sem isn't initialized until mlx4_multi_func_init() (which we haven't called yet). The next thing it tries to do is access priv->mfunc.vhcr, but that hasn't been allocated yet. Fix this by moving the initialization of slave_sem and vhcr up into mlx4_cmd_init(). Also, since slave_sem is really just being used as a mutex, convert it into a slave_cmd_mutex. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-26 11:24:07 +07:00
mutex_unlock(&priv->cmd.slave_cmd_mutex);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
return 0;
err:
net/mlx4_core: Reset flow activation upon SRIOV fatal command cases When SRIOV commands are executed over the comm-channel and get a fatal error (e.g. timeout, closing command failure) the VF enters into error state and reset flow is activated. To be able to recognize whether the failure was on a closing command, the operational code for the given VHCR command is used. Once the device entered into an error state we prevent redundant error messages from being printed. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:43 +07:00
mlx4_comm_cmd(dev, MLX4_COMM_CMD_RESET, 0, MLX4_COMM_CMD_NA_OP, 0);
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
err_offline:
mlx4_core: Fix crash on uninitialized priv->cmd.slave_sem On an SR-IOV master device, __mlx4_init_one() calls mlx4_init_hca() before mlx4_multi_func_init(). However, for unlucky configurations, mlx4_init_hca() might call mlx4_SENSE_PORT() (via mlx4_dev_cap()), and that calls mlx4_cmd_imm() with MLX4_CMD_WRAPPED set. However, on a multifunction device with MLX4_CMD_WRAPPED, __mlx4_cmd() calls into mlx4_slave_cmd(), and that immediately tries to do down(&priv->cmd.slave_sem); but priv->cmd.slave_sem isn't initialized until mlx4_multi_func_init() (which we haven't called yet). The next thing it tries to do is access priv->mfunc.vhcr, but that hasn't been allocated yet. Fix this by moving the initialization of slave_sem and vhcr up into mlx4_cmd_init(). Also, since slave_sem is really just being used as a mutex, convert it into a slave_cmd_mutex. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-26 11:24:07 +07:00
mutex_unlock(&priv->cmd.slave_cmd_mutex);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
return -EIO;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
}
mlx4: Put physical GID and P_Key table sizes in mlx4_phys_caps struct and paravirtualize them To allow easy paravirtualization of P_Key and GID table sizes, keep paravirtualized sizes in mlx4_dev->caps, but save the actual physical sizes from FW in struct: mlx4_dev->phys_cap. In addition, in SR-IOV mode, do the following: 1. Reduce reported P_Key table size by 1. This is done to reserve the highest P_Key index for internal use, for declaring an invalid P_Key in P_Key paravirtualization. We require a P_Key index which always contain an invalid P_Key value for this purpose (i.e., one which cannot be modified by the subnet manager). The way to do this is to reduce the P_Key table size reported to the subnet manager by 1, so that it will not attempt to access the P_Key at index #127. 2. Paravirtualize the GID table size to 1. Thus, each guest sees only a single GID (at its paravirtualized index 0). In addition, since we are paravirtualizing the GID table size to 1, we add paravirtualization of the master GID event here (i.e., we do not do ib_dispatch_event() for the GUID change event on the master, since its (only) GUID never changes). Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:44 +07:00
static void mlx4_parav_master_pf_caps(struct mlx4_dev *dev)
{
int i;
for (i = 1; i <= dev->caps.num_ports; i++) {
mlx4: In RoCE allow guests to have multiple GIDS The GIDs are statically distributed, as follows: PF: gets 16 GIDs VFs: Remaining GIDS are divided evenly between VFs activated by the driver. If the division is not even, lower-numbered VFs get an extra GID. For an IB interface, the number of gids per guest remains as before: one gid per guest. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-12 17:00:39 +07:00
if (dev->caps.port_type[i] == MLX4_PORT_TYPE_ETH)
dev->caps.gid_table_len[i] =
net/mlx4: Adapt code for N-Port VF Adds support for N-Port VFs, this includes: 1. Adding support in the wrapped FW command In wrapped commands, we need to verify and convert the slave's port into the real physical port. Furthermore, when sending the response back to the slave, a reverse conversion should be made. 2. Adjusting sqpn for QP1 para-virtualization The slave assumes that sqpn is used for QP1 communication. If the slave is assigned to a port != (first port), we need to adjust the sqpn that will direct its QP1 packets into the correct endpoint. 3. Adjusting gid[5] to modify the port for raw ethernet In B0 steering, gid[5] contains the port. It needs to be adjusted into the physical port. 4. Adjusting number of ports in the query / ports caps in the FW commands When a slave queries the hardware, it needs to view only the physical ports it's assigned to. 5. Adjusting the sched_qp according to the port number The QP port is encoded in the sched_qp, thus in modify_qp we need to encode the correct port in sched_qp. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-19 23:11:52 +07:00
mlx4_get_slave_num_gids(dev, 0, i);
mlx4: In RoCE allow guests to have multiple GIDS The GIDs are statically distributed, as follows: PF: gets 16 GIDs VFs: Remaining GIDS are divided evenly between VFs activated by the driver. If the division is not even, lower-numbered VFs get an extra GID. For an IB interface, the number of gids per guest remains as before: one gid per guest. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-12 17:00:39 +07:00
else
dev->caps.gid_table_len[i] = 1;
mlx4: Put physical GID and P_Key table sizes in mlx4_phys_caps struct and paravirtualize them To allow easy paravirtualization of P_Key and GID table sizes, keep paravirtualized sizes in mlx4_dev->caps, but save the actual physical sizes from FW in struct: mlx4_dev->phys_cap. In addition, in SR-IOV mode, do the following: 1. Reduce reported P_Key table size by 1. This is done to reserve the highest P_Key index for internal use, for declaring an invalid P_Key in P_Key paravirtualization. We require a P_Key index which always contain an invalid P_Key value for this purpose (i.e., one which cannot be modified by the subnet manager). The way to do this is to reduce the P_Key table size reported to the subnet manager by 1, so that it will not attempt to access the P_Key at index #127. 2. Paravirtualize the GID table size to 1. Thus, each guest sees only a single GID (at its paravirtualized index 0). In addition, since we are paravirtualizing the GID table size to 1, we add paravirtualization of the master GID event here (i.e., we do not do ib_dispatch_event() for the GUID change event on the master, since its (only) GUID never changes). Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:44 +07:00
dev->caps.pkey_table_len[i] =
dev->phys_caps.pkey_phys_table_len[i] - 1;
}
}
mlx4_core: Allow choosing flow steering mode Device managed flow steering will be enabled only under administrator directive provided through setting the existing module parameter log_num_mgm_entry_size to -1 (if the device actually supports flow steering). If flow steering isn't requested or not available, the driver will use the value of log_num_mgm_entry_size and B0 steering. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:12:00 +07:00
static int choose_log_fs_mgm_entry_size(int qp_per_entry)
{
int i = MLX4_MIN_MGM_LOG_ENTRY_SIZE;
for (i = MLX4_MIN_MGM_LOG_ENTRY_SIZE; i <= MLX4_MAX_MGM_LOG_ENTRY_SIZE;
i++) {
if (qp_per_entry <= 4 * ((1 << i) / 16 - 2))
break;
}
return (i <= MLX4_MAX_MGM_LOG_ENTRY_SIZE) ? i : -1;
}
net/mlx4: Add support for A0 steering Add the required firmware commands for A0 steering and a way to enable that. The firmware support focuses on INIT_HCA, QUERY_HCA, QUERY_PORT, QUERY_DEV_CAP and QUERY_FUNC_CAP commands. Those commands are used to configure and query the device. The different A0 DMFS (steering) modes are: Static - optimized performance, but flow steering rules are limited. This mode should be choosed explicitly by the user in order to be used. Dynamic - this mode should be explicitly choosed by the user. In this mode, the FW works in optimized steering mode as long as it can and afterwards automatically drops to classic (full) DMFS. Disable - this mode should be explicitly choosed by the user. The user instructs the system not to use optimized steering, even if the FW supports Dynamic A0 DMFS (and thus will be able to use optimized steering in Default A0 DMFS mode). Default - this mode is implicitly choosed. In this mode, if the FW supports Dynamic A0 DMFS, it'll work in this mode. Otherwise, it'll work at Disable A0 DMFS mode. Under SRIOV configuration, when the A0 steering mode is enabled, older guest VF drivers who aren't using the RX QP allocation flag (MLX4_RESERVE_A0_QP) will get a QP from the general range and fail when attempting to register a steering rule. To avoid that, the PF context behaviour is changed once on A0 static mode, to require support for the allocation flag in VF drivers too. In order to enable A0 steering, we use log_num_mgm_entry_size param. If the value of the parameter is not positive, we treat the absolute value of log_num_mgm_entry_size as a bit field. Setting bit 2 of this bit field enables static A0 steering. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:58:00 +07:00
static const char *dmfs_high_rate_steering_mode_str(int dmfs_high_steer_mode)
{
switch (dmfs_high_steer_mode) {
case MLX4_STEERING_DMFS_A0_DEFAULT:
return "default performance";
case MLX4_STEERING_DMFS_A0_DYNAMIC:
return "dynamic hybrid mode";
case MLX4_STEERING_DMFS_A0_STATIC:
return "performance optimized for limited rule configuration (static)";
case MLX4_STEERING_DMFS_A0_DISABLE:
return "disabled performance optimized steering";
case MLX4_STEERING_DMFS_A0_NOT_SUPPORTED:
return "performance optimized steering not supported";
default:
return "Unrecognized mode";
}
}
#define MLX4_DMFS_A0_STEERING (1UL << 2)
mlx4_core: Adjustments to Flow Steering activation logic for SR-IOV Separate flow steering capability detection from the decision to activate. For the master (and for native), detect the flow steering capability in mlx4_dev_cap, but activate the appropriate steering type in a new function choose_flow_steering() based on detected data. For VFs, activate flow steering based on what was actually activated by the master, where that info is obtained via QUERY_HCA. This fixes the current VF detection which is wrongly based on QUERY_DEV_CAP. Also, for SR-IOV mode, if flow steering may be activated, do so only if the max number of QPs per rule is sufficient to satisfy one subscription per VF. If not, fall back to B0 mode. This is needed to serve registrations done by L2 network drivers such as mlx4_en and IPoIB when the network stack attempts to join to multicast groups such as all-hosts or the IPoIB broadcast group. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:11:59 +07:00
static void choose_steering_mode(struct mlx4_dev *dev,
struct mlx4_dev_cap *dev_cap)
{
net/mlx4: Add support for A0 steering Add the required firmware commands for A0 steering and a way to enable that. The firmware support focuses on INIT_HCA, QUERY_HCA, QUERY_PORT, QUERY_DEV_CAP and QUERY_FUNC_CAP commands. Those commands are used to configure and query the device. The different A0 DMFS (steering) modes are: Static - optimized performance, but flow steering rules are limited. This mode should be choosed explicitly by the user in order to be used. Dynamic - this mode should be explicitly choosed by the user. In this mode, the FW works in optimized steering mode as long as it can and afterwards automatically drops to classic (full) DMFS. Disable - this mode should be explicitly choosed by the user. The user instructs the system not to use optimized steering, even if the FW supports Dynamic A0 DMFS (and thus will be able to use optimized steering in Default A0 DMFS mode). Default - this mode is implicitly choosed. In this mode, if the FW supports Dynamic A0 DMFS, it'll work in this mode. Otherwise, it'll work at Disable A0 DMFS mode. Under SRIOV configuration, when the A0 steering mode is enabled, older guest VF drivers who aren't using the RX QP allocation flag (MLX4_RESERVE_A0_QP) will get a QP from the general range and fail when attempting to register a steering rule. To avoid that, the PF context behaviour is changed once on A0 static mode, to require support for the allocation flag in VF drivers too. In order to enable A0 steering, we use log_num_mgm_entry_size param. If the value of the parameter is not positive, we treat the absolute value of log_num_mgm_entry_size as a bit field. Setting bit 2 of this bit field enables static A0 steering. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:58:00 +07:00
if (mlx4_log_num_mgm_entry_size <= 0) {
if ((-mlx4_log_num_mgm_entry_size) & MLX4_DMFS_A0_STEERING) {
if (dev->caps.dmfs_high_steer_mode ==
MLX4_STEERING_DMFS_A0_NOT_SUPPORTED)
mlx4_err(dev, "DMFS high rate mode not supported\n");
else
dev->caps.dmfs_high_steer_mode =
MLX4_STEERING_DMFS_A0_STATIC;
}
}
if (mlx4_log_num_mgm_entry_size <= 0 &&
mlx4_core: Allow choosing flow steering mode Device managed flow steering will be enabled only under administrator directive provided through setting the existing module parameter log_num_mgm_entry_size to -1 (if the device actually supports flow steering). If flow steering isn't requested or not available, the driver will use the value of log_num_mgm_entry_size and B0 steering. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:12:00 +07:00
dev_cap->flags2 & MLX4_DEV_CAP_FLAG2_FS_EN &&
mlx4_core: Adjustments to Flow Steering activation logic for SR-IOV Separate flow steering capability detection from the decision to activate. For the master (and for native), detect the flow steering capability in mlx4_dev_cap, but activate the appropriate steering type in a new function choose_flow_steering() based on detected data. For VFs, activate flow steering based on what was actually activated by the master, where that info is obtained via QUERY_HCA. This fixes the current VF detection which is wrongly based on QUERY_DEV_CAP. Also, for SR-IOV mode, if flow steering may be activated, do so only if the max number of QPs per rule is sufficient to satisfy one subscription per VF. If not, fall back to B0 mode. This is needed to serve registrations done by L2 network drivers such as mlx4_en and IPoIB when the network stack attempts to join to multicast groups such as all-hosts or the IPoIB broadcast group. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:11:59 +07:00
(!mlx4_is_mfunc(dev) ||
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
(dev_cap->fs_max_num_qp_per_entry >=
(dev->persist->num_vfs + 1))) &&
mlx4_core: Allow choosing flow steering mode Device managed flow steering will be enabled only under administrator directive provided through setting the existing module parameter log_num_mgm_entry_size to -1 (if the device actually supports flow steering). If flow steering isn't requested or not available, the driver will use the value of log_num_mgm_entry_size and B0 steering. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:12:00 +07:00
choose_log_fs_mgm_entry_size(dev_cap->fs_max_num_qp_per_entry) >=
MLX4_MIN_MGM_LOG_ENTRY_SIZE) {
dev->oper_log_mgm_entry_size =
choose_log_fs_mgm_entry_size(dev_cap->fs_max_num_qp_per_entry);
mlx4_core: Adjustments to Flow Steering activation logic for SR-IOV Separate flow steering capability detection from the decision to activate. For the master (and for native), detect the flow steering capability in mlx4_dev_cap, but activate the appropriate steering type in a new function choose_flow_steering() based on detected data. For VFs, activate flow steering based on what was actually activated by the master, where that info is obtained via QUERY_HCA. This fixes the current VF detection which is wrongly based on QUERY_DEV_CAP. Also, for SR-IOV mode, if flow steering may be activated, do so only if the max number of QPs per rule is sufficient to satisfy one subscription per VF. If not, fall back to B0 mode. This is needed to serve registrations done by L2 network drivers such as mlx4_en and IPoIB when the network stack attempts to join to multicast groups such as all-hosts or the IPoIB broadcast group. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:11:59 +07:00
dev->caps.steering_mode = MLX4_STEERING_MODE_DEVICE_MANAGED;
dev->caps.num_qp_per_mgm = dev_cap->fs_max_num_qp_per_entry;
dev->caps.fs_log_max_ucast_qp_range_size =
dev_cap->fs_log_max_ucast_qp_range_size;
} else {
net/mlx4: Add support for A0 steering Add the required firmware commands for A0 steering and a way to enable that. The firmware support focuses on INIT_HCA, QUERY_HCA, QUERY_PORT, QUERY_DEV_CAP and QUERY_FUNC_CAP commands. Those commands are used to configure and query the device. The different A0 DMFS (steering) modes are: Static - optimized performance, but flow steering rules are limited. This mode should be choosed explicitly by the user in order to be used. Dynamic - this mode should be explicitly choosed by the user. In this mode, the FW works in optimized steering mode as long as it can and afterwards automatically drops to classic (full) DMFS. Disable - this mode should be explicitly choosed by the user. The user instructs the system not to use optimized steering, even if the FW supports Dynamic A0 DMFS (and thus will be able to use optimized steering in Default A0 DMFS mode). Default - this mode is implicitly choosed. In this mode, if the FW supports Dynamic A0 DMFS, it'll work in this mode. Otherwise, it'll work at Disable A0 DMFS mode. Under SRIOV configuration, when the A0 steering mode is enabled, older guest VF drivers who aren't using the RX QP allocation flag (MLX4_RESERVE_A0_QP) will get a QP from the general range and fail when attempting to register a steering rule. To avoid that, the PF context behaviour is changed once on A0 static mode, to require support for the allocation flag in VF drivers too. In order to enable A0 steering, we use log_num_mgm_entry_size param. If the value of the parameter is not positive, we treat the absolute value of log_num_mgm_entry_size as a bit field. Setting bit 2 of this bit field enables static A0 steering. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:58:00 +07:00
if (dev->caps.dmfs_high_steer_mode !=
MLX4_STEERING_DMFS_A0_NOT_SUPPORTED)
dev->caps.dmfs_high_steer_mode = MLX4_STEERING_DMFS_A0_DISABLE;
mlx4_core: Adjustments to Flow Steering activation logic for SR-IOV Separate flow steering capability detection from the decision to activate. For the master (and for native), detect the flow steering capability in mlx4_dev_cap, but activate the appropriate steering type in a new function choose_flow_steering() based on detected data. For VFs, activate flow steering based on what was actually activated by the master, where that info is obtained via QUERY_HCA. This fixes the current VF detection which is wrongly based on QUERY_DEV_CAP. Also, for SR-IOV mode, if flow steering may be activated, do so only if the max number of QPs per rule is sufficient to satisfy one subscription per VF. If not, fall back to B0 mode. This is needed to serve registrations done by L2 network drivers such as mlx4_en and IPoIB when the network stack attempts to join to multicast groups such as all-hosts or the IPoIB broadcast group. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:11:59 +07:00
if (dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_UC_STEER &&
dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_MC_STEER)
dev->caps.steering_mode = MLX4_STEERING_MODE_B0;
else {
dev->caps.steering_mode = MLX4_STEERING_MODE_A0;
if (dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_UC_STEER ||
dev->caps.flags & MLX4_DEV_CAP_FLAG_VEP_MC_STEER)
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_warn(dev, "Must have both UC_STEER and MC_STEER flags set to use B0 steering - falling back to A0 steering mode\n");
mlx4_core: Adjustments to Flow Steering activation logic for SR-IOV Separate flow steering capability detection from the decision to activate. For the master (and for native), detect the flow steering capability in mlx4_dev_cap, but activate the appropriate steering type in a new function choose_flow_steering() based on detected data. For VFs, activate flow steering based on what was actually activated by the master, where that info is obtained via QUERY_HCA. This fixes the current VF detection which is wrongly based on QUERY_DEV_CAP. Also, for SR-IOV mode, if flow steering may be activated, do so only if the max number of QPs per rule is sufficient to satisfy one subscription per VF. If not, fall back to B0 mode. This is needed to serve registrations done by L2 network drivers such as mlx4_en and IPoIB when the network stack attempts to join to multicast groups such as all-hosts or the IPoIB broadcast group. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:11:59 +07:00
}
mlx4_core: Allow choosing flow steering mode Device managed flow steering will be enabled only under administrator directive provided through setting the existing module parameter log_num_mgm_entry_size to -1 (if the device actually supports flow steering). If flow steering isn't requested or not available, the driver will use the value of log_num_mgm_entry_size and B0 steering. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:12:00 +07:00
dev->oper_log_mgm_entry_size =
mlx4_log_num_mgm_entry_size > 0 ?
mlx4_log_num_mgm_entry_size :
MLX4_DEFAULT_MGM_LOG_ENTRY_SIZE;
mlx4_core: Adjustments to Flow Steering activation logic for SR-IOV Separate flow steering capability detection from the decision to activate. For the master (and for native), detect the flow steering capability in mlx4_dev_cap, but activate the appropriate steering type in a new function choose_flow_steering() based on detected data. For VFs, activate flow steering based on what was actually activated by the master, where that info is obtained via QUERY_HCA. This fixes the current VF detection which is wrongly based on QUERY_DEV_CAP. Also, for SR-IOV mode, if flow steering may be activated, do so only if the max number of QPs per rule is sufficient to satisfy one subscription per VF. If not, fall back to B0 mode. This is needed to serve registrations done by L2 network drivers such as mlx4_en and IPoIB when the network stack attempts to join to multicast groups such as all-hosts or the IPoIB broadcast group. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:11:59 +07:00
dev->caps.num_qp_per_mgm = mlx4_get_qp_per_mgm(dev);
}
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_dbg(dev, "Steering mode is: %s, oper_log_mgm_entry_size = %d, modparam log_num_mgm_entry_size = %d\n",
mlx4_core: Allow choosing flow steering mode Device managed flow steering will be enabled only under administrator directive provided through setting the existing module parameter log_num_mgm_entry_size to -1 (if the device actually supports flow steering). If flow steering isn't requested or not available, the driver will use the value of log_num_mgm_entry_size and B0 steering. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:12:00 +07:00
mlx4_steering_mode_str(dev->caps.steering_mode),
dev->oper_log_mgm_entry_size,
mlx4_log_num_mgm_entry_size);
mlx4_core: Adjustments to Flow Steering activation logic for SR-IOV Separate flow steering capability detection from the decision to activate. For the master (and for native), detect the flow steering capability in mlx4_dev_cap, but activate the appropriate steering type in a new function choose_flow_steering() based on detected data. For VFs, activate flow steering based on what was actually activated by the master, where that info is obtained via QUERY_HCA. This fixes the current VF detection which is wrongly based on QUERY_DEV_CAP. Also, for SR-IOV mode, if flow steering may be activated, do so only if the max number of QPs per rule is sufficient to satisfy one subscription per VF. If not, fall back to B0 mode. This is needed to serve registrations done by L2 network drivers such as mlx4_en and IPoIB when the network stack attempts to join to multicast groups such as all-hosts or the IPoIB broadcast group. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:11:59 +07:00
}
net/mlx4_core: Add basic support for TCP/IP offloads under tunneling Add the low-level device commands and definitions used for TCP/IP HW offloads of tunneled/vxlan traffic which are supported by the ConnectX3-pro NIC. This is done through the following elements: - read tunneling device caps in QUERY_DEV_CAP - add helper function to do SET_PORT for tunneling - add DMFS VXLAN steering rule definitions - add CQE and WQE checksum offload field definitions Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-23 21:09:43 +07:00
static void choose_tunnel_offload_mode(struct mlx4_dev *dev,
struct mlx4_dev_cap *dev_cap)
{
if (dev->caps.steering_mode == MLX4_STEERING_MODE_DEVICE_MANAGED &&
net/mlx4: Don't disable vxlan offloads under DMFS-A0 optimized steering Except for VXLAN steering rules, all offloads should work as they were under plain DMFS mode. Fix that by enabling all the offloads under DMFS-A0 mode, except for VXLAN steering rules. Fixes: d57febe1a478 "net/mlx4: Add A0 hybrid steering" Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-15 20:28:54 +07:00
dev_cap->flags2 & MLX4_DEV_CAP_FLAG2_VXLAN_OFFLOADS)
net/mlx4_core: Add basic support for TCP/IP offloads under tunneling Add the low-level device commands and definitions used for TCP/IP HW offloads of tunneled/vxlan traffic which are supported by the ConnectX3-pro NIC. This is done through the following elements: - read tunneling device caps in QUERY_DEV_CAP - add helper function to do SET_PORT for tunneling - add DMFS VXLAN steering rule definitions - add CQE and WQE checksum offload field definitions Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-23 21:09:43 +07:00
dev->caps.tunnel_offload_mode = MLX4_TUNNEL_OFFLOAD_MODE_VXLAN;
else
dev->caps.tunnel_offload_mode = MLX4_TUNNEL_OFFLOAD_MODE_NONE;
mlx4_dbg(dev, "Tunneling offload mode is: %s\n", (dev->caps.tunnel_offload_mode
== MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) ? "vxlan" : "none");
}
net/mlx4: Add support for A0 steering Add the required firmware commands for A0 steering and a way to enable that. The firmware support focuses on INIT_HCA, QUERY_HCA, QUERY_PORT, QUERY_DEV_CAP and QUERY_FUNC_CAP commands. Those commands are used to configure and query the device. The different A0 DMFS (steering) modes are: Static - optimized performance, but flow steering rules are limited. This mode should be choosed explicitly by the user in order to be used. Dynamic - this mode should be explicitly choosed by the user. In this mode, the FW works in optimized steering mode as long as it can and afterwards automatically drops to classic (full) DMFS. Disable - this mode should be explicitly choosed by the user. The user instructs the system not to use optimized steering, even if the FW supports Dynamic A0 DMFS (and thus will be able to use optimized steering in Default A0 DMFS mode). Default - this mode is implicitly choosed. In this mode, if the FW supports Dynamic A0 DMFS, it'll work in this mode. Otherwise, it'll work at Disable A0 DMFS mode. Under SRIOV configuration, when the A0 steering mode is enabled, older guest VF drivers who aren't using the RX QP allocation flag (MLX4_RESERVE_A0_QP) will get a QP from the general range and fail when attempting to register a steering rule. To avoid that, the PF context behaviour is changed once on A0 static mode, to require support for the allocation flag in VF drivers too. In order to enable A0 steering, we use log_num_mgm_entry_size param. If the value of the parameter is not positive, we treat the absolute value of log_num_mgm_entry_size as a bit field. Setting bit 2 of this bit field enables static A0 steering. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:58:00 +07:00
static int mlx4_validate_optimized_steering(struct mlx4_dev *dev)
{
int i;
struct mlx4_port_cap port_cap;
if (dev->caps.dmfs_high_steer_mode == MLX4_STEERING_DMFS_A0_NOT_SUPPORTED)
return -EINVAL;
for (i = 1; i <= dev->caps.num_ports; i++) {
if (mlx4_dev_port(dev, i, &port_cap)) {
mlx4_err(dev,
"QUERY_DEV_CAP command failed, can't veify DMFS high rate steering.\n");
} else if ((dev->caps.dmfs_high_steer_mode !=
MLX4_STEERING_DMFS_A0_DEFAULT) &&
(port_cap.dmfs_optimized_state ==
!!(dev->caps.dmfs_high_steer_mode ==
MLX4_STEERING_DMFS_A0_DISABLE))) {
mlx4_err(dev,
"DMFS high rate steer mode differ, driver requested %s but %s in FW.\n",
dmfs_high_rate_steering_mode_str(
dev->caps.dmfs_high_steer_mode),
(port_cap.dmfs_optimized_state ?
"enabled" : "disabled"));
}
}
return 0;
}
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
static int mlx4_init_fw(struct mlx4_dev *dev)
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
{
mlx4_core: Use MOD_STAT_CFG command to get minimal page size There was a bug in some versions of the mlx4 driver in mlx4_alloc_fmr(), which hardcoded the minimum acceptable page_shift to be 12. However, new ConnectX firmware can support a minimum page_shift of 9 (log_pg_sz of 9 returned by QUERY_DEV_LIM) -- so with old drivers, ib_fmr_alloc() would fail for ULPs using the device minimum when creating FMRs. To preserve firmware compatibility with released mlx4 drivers, the firmware will continue to return 12 as before for log_page_sz in QUERY_DEV_CAP for these drivers. However, to enable new drivers to take advantage of the available smaller page size, the mlx4 driver now first sets the log_pg_sz to the device minimum by setting a log_page_sz value to 0 via the MOD_STAT_CFG command and then reading the real minimum via QUERY_DEV_CAP. Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il> Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-07-15 13:48:53 +07:00
struct mlx4_mod_stat_cfg mlx4_cfg;
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
int err = 0;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if (!mlx4_is_slave(dev)) {
err = mlx4_QUERY_FW(dev);
if (err) {
if (err == -EACCES)
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_info(dev, "non-primary physical function, skipping\n");
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
else
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "QUERY_FW command failed, aborting\n");
net/mlx4_core: Fixing error flow in case of QUERY_FW failure The order of operations was wrong on the teardown flow. Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-06 05:50:51 +07:00
return err;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
err = mlx4_load_fw(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to start FW, aborting\n");
net/mlx4_core: Fixing error flow in case of QUERY_FW failure The order of operations was wrong on the teardown flow. Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-06 05:50:51 +07:00
return err;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
mlx4_cfg.log_pg_sz_m = 1;
mlx4_cfg.log_pg_sz = 0;
err = mlx4_MOD_STAT_CFG(dev, &mlx4_cfg);
if (err)
mlx4_warn(dev, "Failed to override log_pg_sz parameter\n");
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
}
mlx4_core: Use MOD_STAT_CFG command to get minimal page size There was a bug in some versions of the mlx4 driver in mlx4_alloc_fmr(), which hardcoded the minimum acceptable page_shift to be 12. However, new ConnectX firmware can support a minimum page_shift of 9 (log_pg_sz of 9 returned by QUERY_DEV_LIM) -- so with old drivers, ib_fmr_alloc() would fail for ULPs using the device minimum when creating FMRs. To preserve firmware compatibility with released mlx4 drivers, the firmware will continue to return 12 as before for log_page_sz in QUERY_DEV_CAP for these drivers. However, to enable new drivers to take advantage of the available smaller page size, the mlx4 driver now first sets the log_pg_sz to the device minimum by setting a log_page_sz value to 0 via the MOD_STAT_CFG command and then reading the real minimum via QUERY_DEV_CAP. Signed-off-by: Jack Morgenstein <jackm@mellanox.co.il> Signed-off-by: Vladimir Sokolovsky <vlad@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-07-15 13:48:53 +07:00
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
return err;
}
static int mlx4_init_hca(struct mlx4_dev *dev)
{
struct mlx4_priv *priv = mlx4_priv(dev);
struct mlx4_adapter adapter;
struct mlx4_dev_cap dev_cap;
struct mlx4_profile profile;
struct mlx4_init_hca_param init_hca;
u64 icm_size;
struct mlx4_config_dev_params params;
int err;
if (!mlx4_is_slave(dev)) {
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
err = mlx4_dev_cap(dev, &dev_cap);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting\n");
net/mlx4_core: Fix error flow in mlx4_init_hca() We shouldn't call UNMAP_FA here, this is done in mlx4_load_one. If mlx4_query_func fails, we need to invoke CLOSE_HCA for both native and master. Fixes: a0eacca948d2 ('net/mlx4_core: Refactor mlx4_load_one') Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-30 16:59:50 +07:00
return err;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_core: Adjustments to Flow Steering activation logic for SR-IOV Separate flow steering capability detection from the decision to activate. For the master (and for native), detect the flow steering capability in mlx4_dev_cap, but activate the appropriate steering type in a new function choose_flow_steering() based on detected data. For VFs, activate flow steering based on what was actually activated by the master, where that info is obtained via QUERY_HCA. This fixes the current VF detection which is wrongly based on QUERY_DEV_CAP. Also, for SR-IOV mode, if flow steering may be activated, do so only if the max number of QPs per rule is sufficient to satisfy one subscription per VF. If not, fall back to B0 mode. This is needed to serve registrations done by L2 network drivers such as mlx4_en and IPoIB when the network stack attempts to join to multicast groups such as all-hosts or the IPoIB broadcast group. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:11:59 +07:00
choose_steering_mode(dev, &dev_cap);
net/mlx4_core: Add basic support for TCP/IP offloads under tunneling Add the low-level device commands and definitions used for TCP/IP HW offloads of tunneled/vxlan traffic which are supported by the ConnectX3-pro NIC. This is done through the following elements: - read tunneling device caps in QUERY_DEV_CAP - add helper function to do SET_PORT for tunneling - add DMFS VXLAN steering rule definitions - add CQE and WQE checksum offload field definitions Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-23 21:09:43 +07:00
choose_tunnel_offload_mode(dev, &dev_cap);
mlx4_core: Adjustments to Flow Steering activation logic for SR-IOV Separate flow steering capability detection from the decision to activate. For the master (and for native), detect the flow steering capability in mlx4_dev_cap, but activate the appropriate steering type in a new function choose_flow_steering() based on detected data. For VFs, activate flow steering based on what was actually activated by the master, where that info is obtained via QUERY_HCA. This fixes the current VF detection which is wrongly based on QUERY_DEV_CAP. Also, for SR-IOV mode, if flow steering may be activated, do so only if the max number of QPs per rule is sufficient to satisfy one subscription per VF. If not, fall back to B0 mode. This is needed to serve registrations done by L2 network drivers such as mlx4_en and IPoIB when the network stack attempts to join to multicast groups such as all-hosts or the IPoIB broadcast group. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:11:59 +07:00
net/mlx4: Add support for A0 steering Add the required firmware commands for A0 steering and a way to enable that. The firmware support focuses on INIT_HCA, QUERY_HCA, QUERY_PORT, QUERY_DEV_CAP and QUERY_FUNC_CAP commands. Those commands are used to configure and query the device. The different A0 DMFS (steering) modes are: Static - optimized performance, but flow steering rules are limited. This mode should be choosed explicitly by the user in order to be used. Dynamic - this mode should be explicitly choosed by the user. In this mode, the FW works in optimized steering mode as long as it can and afterwards automatically drops to classic (full) DMFS. Disable - this mode should be explicitly choosed by the user. The user instructs the system not to use optimized steering, even if the FW supports Dynamic A0 DMFS (and thus will be able to use optimized steering in Default A0 DMFS mode). Default - this mode is implicitly choosed. In this mode, if the FW supports Dynamic A0 DMFS, it'll work in this mode. Otherwise, it'll work at Disable A0 DMFS mode. Under SRIOV configuration, when the A0 steering mode is enabled, older guest VF drivers who aren't using the RX QP allocation flag (MLX4_RESERVE_A0_QP) will get a QP from the general range and fail when attempting to register a steering rule. To avoid that, the PF context behaviour is changed once on A0 static mode, to require support for the allocation flag in VF drivers too. In order to enable A0 steering, we use log_num_mgm_entry_size param. If the value of the parameter is not positive, we treat the absolute value of log_num_mgm_entry_size as a bit field. Setting bit 2 of this bit field enables static A0 steering. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:58:00 +07:00
if (dev->caps.dmfs_high_steer_mode == MLX4_STEERING_DMFS_A0_STATIC &&
mlx4_is_master(dev))
dev->caps.function_caps |= MLX4_FUNC_CAP_DMFS_A0_STATIC;
net/mlx4_core: Expose physical port id as PF/VF capability Add the infrastructure needed to support ndo_get_phys_port_id which allows users to identify to which physical port a net-device is connected to by reading a unique port id. This will work for VFs and PFs. The driver uses a new device capability - phys_port_id, The PF driver reads the port phys_port_id from Firmware and stores it. The VF driver reads the port phys_port_id from the PF using QUERY_FUNC_CAP command. Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-20 02:20:12 +07:00
err = mlx4_get_phys_port_id(dev);
if (err)
mlx4_err(dev, "Fail to get physical port id\n");
mlx4: Put physical GID and P_Key table sizes in mlx4_phys_caps struct and paravirtualize them To allow easy paravirtualization of P_Key and GID table sizes, keep paravirtualized sizes in mlx4_dev->caps, but save the actual physical sizes from FW in struct: mlx4_dev->phys_cap. In addition, in SR-IOV mode, do the following: 1. Reduce reported P_Key table size by 1. This is done to reserve the highest P_Key index for internal use, for declaring an invalid P_Key in P_Key paravirtualization. We require a P_Key index which always contain an invalid P_Key value for this purpose (i.e., one which cannot be modified by the subnet manager). The way to do this is to reduce the P_Key table size reported to the subnet manager by 1, so that it will not attempt to access the P_Key at index #127. 2. Paravirtualize the GID table size to 1. Thus, each guest sees only a single GID (at its paravirtualized index 0). In addition, since we are paravirtualizing the GID table size to 1, we add paravirtualization of the master GID event here (i.e., we do not do ib_dispatch_event() for the GUID change event on the master, since its (only) GUID never changes). Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:44 +07:00
if (mlx4_is_master(dev))
mlx4_parav_master_pf_caps(dev);
net/mlx4_core: Use low memory profile on kdump kernel When running in kdump kernel, reduce number of resources allocated for the hardware. This will enable the NIC to operate in this low memory environment at the expense of performance and some features not related to the basic NIC functionality. Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-07-22 19:44:11 +07:00
if (mlx4_low_memory_profile()) {
mlx4_info(dev, "Running from within kdump kernel. Using low memory profile\n");
profile = low_mem_profile;
} else {
profile = default_profile;
}
{NET, IB}/mlx4: Add device managed flow steering firmware API The driver is modified to support three operation modes. If supported by firmware use the device managed flow steering API, that which we call device managed steering mode. Else, if the firmware supports the B0 steering mode use it, and finally, if none of the above, use the A0 steering mode. When the steering mode is device managed, the code is modified such that L2 based rules set by the mlx4_en driver for Ethernet unicast and multicast, and the IB stack multicast attach calls done through the mlx4_ib driver are all routed to use the device managed API. When attaching rule using device managed flow steering API, the firmware returns a 64 bit registration id, which is to be provided during detach. Currently the firmware is always programmed during HCA initialization to use standard L2 hashing. Future work should be done to allow configuring the flow-steering hash function with common, non proprietary means. Signed-off-by: Hadar Hen Zion <hadarh@mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-05 11:03:46 +07:00
if (dev->caps.steering_mode ==
MLX4_STEERING_MODE_DEVICE_MANAGED)
profile.num_mcg = MLX4_FS_NUM_MCG;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
icm_size = mlx4_make_profile(dev, &profile, &dev_cap,
&init_hca);
if ((long long) icm_size < 0) {
err = icm_size;
net/mlx4_core: Fix error flow in mlx4_init_hca() We shouldn't call UNMAP_FA here, this is done in mlx4_load_one. If mlx4_query_func fails, we need to invoke CLOSE_HCA for both native and master. Fixes: a0eacca948d2 ('net/mlx4_core: Refactor mlx4_load_one') Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-30 16:59:50 +07:00
return err;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4: Enforce device max FMR maps in FMR alloc ConnectX devices have a limit on the number of mappings that can be done on an FMR before having to call sync_tpt. The current mlx4_ib driver reports the limit correctly in max_map_per_fmr in .query_device(), but mlx4_core doesn't check it when actually allocating FMRs. Add a max_fmr_maps field to struct mlx4_caps and enforce this maximum value on FMR allocations. Signed-off-by: Eli Cohen <eli@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-02-09 23:10:06 +07:00
dev->caps.max_fmr_maps = (1 << (32 - ilog2(dev->caps.num_mpts))) - 1;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
init_hca.log_uar_sz = ilog2(dev->caps.num_uars);
init_hca.uar_page_sz = PAGE_SHIFT - 12;
mlx4_core: Enable memory windows in {INIT, QUERY}_HCA Add memory windows-related code to INIT_HCA and QUERY_HCA. Signed-off-by: Haggai Eran <haggaie@mellanox.com> Signed-off-by: Shani Michaeli <shanim@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-02-06 23:19:11 +07:00
init_hca.mw_enabled = 0;
if (dev->caps.flags & MLX4_DEV_CAP_FLAG_MEM_WINDOW ||
dev->caps.bmme_flags & MLX4_BMME_FLAG_TYPE_2_WIN)
init_hca.mw_enabled = INIT_HCA_TPT_MW_ENABLE;
mlx4: Add blue flame support for kernel consumers Using blue flame can improve latency by allowing the HW to more efficiently access the WQE. This patch presents two functions that are used to allocate or release HW resources for using blue flame; the caller need to supply a struct mlx4_bf object when allocating resources. Consumers that make use of this API should post doorbells to the UAR object pointed by the initialized struct mlx4_bf; Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:38:41 +07:00
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
err = mlx4_init_icm(dev, &dev_cap, &init_hca, icm_size);
if (err)
net/mlx4_core: Fix error flow in mlx4_init_hca() We shouldn't call UNMAP_FA here, this is done in mlx4_load_one. If mlx4_query_func fails, we need to invoke CLOSE_HCA for both native and master. Fixes: a0eacca948d2 ('net/mlx4_core: Refactor mlx4_load_one') Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-30 16:59:50 +07:00
return err;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
err = mlx4_INIT_HCA(dev, &init_hca);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "INIT_HCA command failed, aborting\n");
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
goto err_free_icm;
}
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
if (dev_cap.flags2 & MLX4_DEV_CAP_FLAG2_SYS_EQS) {
err = mlx4_query_func(dev, &dev_cap);
if (err < 0) {
mlx4_err(dev, "QUERY_FUNC command failed, aborting.\n");
net/mlx4_core: Fix error flow in mlx4_init_hca() We shouldn't call UNMAP_FA here, this is done in mlx4_load_one. If mlx4_query_func fails, we need to invoke CLOSE_HCA for both native and master. Fixes: a0eacca948d2 ('net/mlx4_core: Refactor mlx4_load_one') Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-30 16:59:50 +07:00
goto err_close;
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
} else if (err & MLX4_QUERY_FUNC_NUM_SYS_EQS) {
dev->caps.num_eqs = dev_cap.max_eqs;
dev->caps.reserved_eqs = dev_cap.reserved_eqs;
dev->caps.reserved_uars = dev_cap.reserved_uars;
}
}
net/mlx4_core: Read HCA frequency and map internal clock Read HCA frequency, read PCI clock bar and offset, map internal clock to PCI bar. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-23 13:06:48 +07:00
/*
* If TS is supported by FW
* read HCA frequency by QUERY_HCA command
*/
if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_TS) {
memset(&init_hca, 0, sizeof(init_hca));
err = mlx4_QUERY_HCA(dev, &init_hca);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "QUERY_HCA command failed, disable timestamp\n");
net/mlx4_core: Read HCA frequency and map internal clock Read HCA frequency, read PCI clock bar and offset, map internal clock to PCI bar. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-23 13:06:48 +07:00
dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_TS;
} else {
dev->caps.hca_core_clock =
init_hca.hca_core_clock;
}
/* In case we got HCA frequency 0 - disable timestamping
* to avoid dividing by zero
*/
if (!dev->caps.hca_core_clock) {
dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_TS;
mlx4_err(dev,
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
"HCA frequency is 0 - timestamping is not supported\n");
net/mlx4_core: Read HCA frequency and map internal clock Read HCA frequency, read PCI clock bar and offset, map internal clock to PCI bar. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-23 13:06:48 +07:00
} else if (map_internal_clock(dev)) {
/*
* Map internal clock,
* in case of failure disable timestamping
*/
dev->caps.flags2 &= ~MLX4_DEV_CAP_FLAG2_TS;
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to map internal clock. Timestamping is not supported\n");
net/mlx4_core: Read HCA frequency and map internal clock Read HCA frequency, read PCI clock bar and offset, map internal clock to PCI bar. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-23 13:06:48 +07:00
}
}
net/mlx4: Add support for A0 steering Add the required firmware commands for A0 steering and a way to enable that. The firmware support focuses on INIT_HCA, QUERY_HCA, QUERY_PORT, QUERY_DEV_CAP and QUERY_FUNC_CAP commands. Those commands are used to configure and query the device. The different A0 DMFS (steering) modes are: Static - optimized performance, but flow steering rules are limited. This mode should be choosed explicitly by the user in order to be used. Dynamic - this mode should be explicitly choosed by the user. In this mode, the FW works in optimized steering mode as long as it can and afterwards automatically drops to classic (full) DMFS. Disable - this mode should be explicitly choosed by the user. The user instructs the system not to use optimized steering, even if the FW supports Dynamic A0 DMFS (and thus will be able to use optimized steering in Default A0 DMFS mode). Default - this mode is implicitly choosed. In this mode, if the FW supports Dynamic A0 DMFS, it'll work in this mode. Otherwise, it'll work at Disable A0 DMFS mode. Under SRIOV configuration, when the A0 steering mode is enabled, older guest VF drivers who aren't using the RX QP allocation flag (MLX4_RESERVE_A0_QP) will get a QP from the general range and fail when attempting to register a steering rule. To avoid that, the PF context behaviour is changed once on A0 static mode, to require support for the allocation flag in VF drivers too. In order to enable A0 steering, we use log_num_mgm_entry_size param. If the value of the parameter is not positive, we treat the absolute value of log_num_mgm_entry_size as a bit field. Setting bit 2 of this bit field enables static A0 steering. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:58:00 +07:00
if (dev->caps.dmfs_high_steer_mode !=
MLX4_STEERING_DMFS_A0_NOT_SUPPORTED) {
if (mlx4_validate_optimized_steering(dev))
mlx4_warn(dev, "Optimized steering validation failed\n");
if (dev->caps.dmfs_high_steer_mode ==
MLX4_STEERING_DMFS_A0_DISABLE) {
dev->caps.dmfs_high_rate_qpn_base =
dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW];
dev->caps.dmfs_high_rate_qpn_range =
MLX4_A0_STEERING_TABLE_SIZE;
}
mlx4_dbg(dev, "DMFS high rate steer mode is: %s\n",
dmfs_high_rate_steering_mode_str(
dev->caps.dmfs_high_steer_mode));
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
} else {
err = mlx4_init_slave(dev);
if (err) {
net/mlx4_core: Return -EPROBE_DEFER when a VF is probed before PF is sufficiently initialized In the PF initialization, SRIOV is enabled before the PF is fully initialized. This allows the kernel to probe the newly-exposed VFs before the PF is ready to handle them (nested probes). Have the probe method return the -EPROBE_DEFER value in this situation (instead of the VF probe method retrying its initialization in a loop, and returning -EIO on failure). When -EPROBE_DEFER is returned by the VF probe method, the kernel itself will retry the probe after a suitable delay. Based upon a suggestion by Ben Hutchings <bhutchings@solarflare.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-04 12:13:27 +07:00
if (err != -EPROBE_DEFER)
mlx4_err(dev, "Failed to initialize slave\n");
net/mlx4_core: Fixing error flow in case of QUERY_FW failure The order of operations was wrong on the teardown flow. Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-06 05:50:51 +07:00
return err;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
err = mlx4_slave_cap(dev);
if (err) {
mlx4_err(dev, "Failed to obtain slave caps\n");
goto err_close;
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if (map_bf_area(dev))
mlx4_dbg(dev, "Failed to map blue flame area\n");
/*Only the master set the ports, all the rest got it from it.*/
if (!mlx4_is_slave(dev))
mlx4_set_port_mask(dev);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err = mlx4_QUERY_ADAPTER(dev, &adapter);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "QUERY_ADAPTER command failed, aborting\n");
net/mlx4_core: Fixing error flow in case of QUERY_FW failure The order of operations was wrong on the teardown flow. Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-06 05:50:51 +07:00
goto unmap_bf;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
}
net/mlx4_en: Extend checksum offloading by CHECKSUM COMPLETE When processing received traffic, pass CHECKSUM_COMPLETE status to the stack, with calculated checksum for non TCP/UDP packets (such as GRE or ICMP). Although the stack expects checksum which doesn't include the pseudo header, the HW adds it. To address that, we are subtracting the pseudo header checksum from the checksum value provided by the HW. In the IPv6 case, we also compute/add the IP header checksum which is not added by the HW for such packets. Cc: Jerry Chu <hkchu@google.com> Signed-off-by: Shani Michaeli <shanim@mellanox.com> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-09 18:51:53 +07:00
/* Query CONFIG_DEV parameters */
err = mlx4_config_dev_retrieval(dev, &params);
if (err && err != -ENOTSUPP) {
mlx4_err(dev, "Failed to query CONFIG_DEV parameters\n");
} else if (!err) {
dev->caps.rx_checksum_flags_port[1] = params.rx_csum_flags_port_1;
dev->caps.rx_checksum_flags_port[2] = params.rx_csum_flags_port_2;
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
priv->eq_table.inta_pin = adapter.inta_pin;
IB/mlx4: Display misc device information under /sys/class/infiniband/ display the following device information under /sys/class/infiniband/mlx4_X: board_id, fw_ver, hw_rev, hca_type. This patch makes this information available to userspace utilities such as ibstat and ibv_devinfo. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-09-18 14:14:18 +07:00
memcpy(dev->board_id, adapter.board_id, sizeof dev->board_id);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return 0;
net/mlx4_core: Fixing error flow in case of QUERY_FW failure The order of operations was wrong on the teardown flow. Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-06 05:50:51 +07:00
unmap_bf:
net/mlx4_core: Read HCA frequency and map internal clock Read HCA frequency, read PCI clock bar and offset, map internal clock to PCI bar. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-04-23 13:06:48 +07:00
unmap_internal_clock(dev);
net/mlx4_core: Fixing error flow in case of QUERY_FW failure The order of operations was wrong on the teardown flow. Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-06 05:50:51 +07:00
unmap_bf_area(dev);
mlx4_core: Fix memory leaks in SR-IOV error paths Fix a few memory leaks that happen if errors happen in SR-IOV mode. Signed-off-by: Dotan Barak <dotanb@dev.mellanox.co.il> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-05-29 20:30:59 +07:00
if (mlx4_is_slave(dev)) {
mlx4: Add infrastructure for selecting VFs to enable QP0 via MLX proxy QPs This commit adds the infrastructure for enabling selected VFs to operate SMI (QP0) MADs without restriction. Additionally, for these enabled VFs, their QP0 proxy and tunnel QPs are MLX QPs. As such, they operate over VL15. Therefore, they are not affected by "credit" problems or changes in the VLArb table (which may shut down VL0). Non-enabled VFs may only create UD proxy QP0 qps (which are forced by the hypervisor to send packets using the q-key it assigns and places in the qp-context). Thus, non-enabled VFs will not pose a security risk. The hypervisor discards any privileged MADs it receives from these non-enabled VFs. By default, all VFs are NOT enabled, and must explicitly be enabled by the administrator. The sysfs interface which operates the VF enablement infrastructure is provided in the next commit. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-05-29 20:31:03 +07:00
kfree(dev->caps.qp0_qkey);
mlx4_core: Fix memory leaks in SR-IOV error paths Fix a few memory leaks that happen if errors happen in SR-IOV mode. Signed-off-by: Dotan Barak <dotanb@dev.mellanox.co.il> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-05-29 20:30:59 +07:00
kfree(dev->caps.qp0_tunnel);
kfree(dev->caps.qp0_proxy);
kfree(dev->caps.qp1_tunnel);
kfree(dev->caps.qp1_proxy);
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err_close:
mlx4_core: Perform correct resource cleanup if mlx4_QUERY_ADAPTER() fails Fixed the resource cleanup to act correctly and prevent a kernel oops when mlx4_QUERY_ADAPTER() fails. Signed-off-by: Dotan Barak <dotanb@dev.mellanox.co.il> Reviewed-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-10-21 21:59:23 +07:00
if (mlx4_is_slave(dev))
mlx4_slave_exit(dev);
else
mlx4_CLOSE_HCA(dev, 0);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err_free_icm:
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if (!mlx4_is_slave(dev))
mlx4_free_icms(dev);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return err;
}
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
static int mlx4_init_counters_table(struct mlx4_dev *dev)
{
struct mlx4_priv *priv = mlx4_priv(dev);
net/mlx4_core: Add sink counter Reserve the last valid counter index for "sink" counter, when a new counter cannot be allocated, the driver will use this counter. In order to avoid allocating this counter on any other flow, fix the indices bitmap allocation range, and reserve the sink counter index. Add macro for the sink counter index and replace all appearences of the index with the macro. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:58:58 +07:00
int nent_pow2;
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
if (!(dev->caps.flags & MLX4_DEV_CAP_FLAG_COUNTERS))
return -ENOENT;
net/mlx4_core: Remove counters table allocation from VF flow Since virtual functions get their counters indices allocation from the PF, allocate counters indices bitmap only in case the function isn't virtual. Also, check that the device has counters to allocate before creating the indices bitmap table. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:58:59 +07:00
if (!dev->caps.max_counters)
return -ENOSPC;
net/mlx4_core: Add sink counter Reserve the last valid counter index for "sink" counter, when a new counter cannot be allocated, the driver will use this counter. In order to avoid allocating this counter on any other flow, fix the indices bitmap allocation range, and reserve the sink counter index. Add macro for the sink counter index and replace all appearences of the index with the macro. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:58:58 +07:00
nent_pow2 = roundup_pow_of_two(dev->caps.max_counters);
/* reserve last counter index for sink counter */
return mlx4_bitmap_init(&priv->counters_bitmap, nent_pow2,
nent_pow2 - 1, 0,
nent_pow2 - dev->caps.max_counters + 1);
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
}
static void mlx4_cleanup_counters_table(struct mlx4_dev *dev)
{
net/mlx4_core: Check before cleaning counters bitmap If counters are not supported by the device. The indices bitmap table is not allocated during initialization. Add the symmetrical check before cleaning the counters bitmap table or freeing a counter. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:58:56 +07:00
if (!(dev->caps.flags & MLX4_DEV_CAP_FLAG_COUNTERS))
return;
net/mlx4_core: Remove counters table allocation from VF flow Since virtual functions get their counters indices allocation from the PF, allocate counters indices bitmap only in case the function isn't virtual. Also, check that the device has counters to allocate before creating the indices bitmap table. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:58:59 +07:00
if (!dev->caps.max_counters)
return;
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
mlx4_bitmap_cleanup(&mlx4_priv(dev)->counters_bitmap);
}
net/mlx4_core: Allocate default counter per port Default counter per port will be allocated at the mlx4 core driver load. Every QP opened by the Ethernet driver will be attached to the port's default counter. This is an infrastructure step to collect VF statistics from the PF. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:59:02 +07:00
static void mlx4_cleanup_default_counters(struct mlx4_dev *dev)
{
struct mlx4_priv *priv = mlx4_priv(dev);
int port;
for (port = 0; port < dev->caps.num_ports; port++)
if (priv->def_counter[port] != -1)
mlx4_counter_free(dev, priv->def_counter[port]);
}
static int mlx4_allocate_default_counters(struct mlx4_dev *dev)
{
struct mlx4_priv *priv = mlx4_priv(dev);
int port, err = 0;
u32 idx;
for (port = 0; port < dev->caps.num_ports; port++)
priv->def_counter[port] = -1;
for (port = 0; port < dev->caps.num_ports; port++) {
err = mlx4_counter_alloc(dev, &idx);
if (!err || err == -ENOSPC) {
priv->def_counter[port] = idx;
} else if (err == -ENOENT) {
err = 0;
continue;
net/mlx4_core: Use sink counter for the VF default as fallback Some old PF drivers don't let VFs allocate counters, in that case, use the sink counter so the VF can load and operate properly. Fixes: 6de5f7f6a1fa ('net/mlx4_core: Allocate default counter per port') Reported-by: Sebastian Ott <sebott@linux.vnet.ibm.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-22 20:53:46 +07:00
} else if (mlx4_is_slave(dev) && err == -EINVAL) {
priv->def_counter[port] = MLX4_SINK_COUNTER_INDEX(dev);
mlx4_warn(dev, "can't allocate counter from old PF driver, using index %d\n",
MLX4_SINK_COUNTER_INDEX(dev));
err = 0;
net/mlx4_core: Allocate default counter per port Default counter per port will be allocated at the mlx4 core driver load. Every QP opened by the Ethernet driver will be attached to the port's default counter. This is an infrastructure step to collect VF statistics from the PF. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:59:02 +07:00
} else {
mlx4_err(dev, "%s: failed to allocate default counter port %d err %d\n",
__func__, port + 1, err);
mlx4_cleanup_default_counters(dev);
return err;
}
mlx4_dbg(dev, "%s: default counter index %d for port %d\n",
__func__, priv->def_counter[port], port + 1);
}
return err;
}
net/mlx4_core: Add XRC domains and counters to resource tracker Add missing resource tracking for XRC domains and complete the tracking for HCA network flow counters. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-15 17:35:03 +07:00
int __mlx4_counter_alloc(struct mlx4_dev *dev, u32 *idx)
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
{
struct mlx4_priv *priv = mlx4_priv(dev);
if (!(dev->caps.flags & MLX4_DEV_CAP_FLAG_COUNTERS))
return -ENOENT;
*idx = mlx4_bitmap_alloc(&priv->counters_bitmap);
net/mlx4_core: Allocate default counter per port Default counter per port will be allocated at the mlx4 core driver load. Every QP opened by the Ethernet driver will be attached to the port's default counter. This is an infrastructure step to collect VF statistics from the PF. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:59:02 +07:00
if (*idx == -1) {
*idx = MLX4_SINK_COUNTER_INDEX(dev);
return -ENOSPC;
}
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
return 0;
}
net/mlx4_core: Add XRC domains and counters to resource tracker Add missing resource tracking for XRC domains and complete the tracking for HCA network flow counters. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-15 17:35:03 +07:00
int mlx4_counter_alloc(struct mlx4_dev *dev, u32 *idx)
{
u64 out_param;
int err;
if (mlx4_is_mfunc(dev)) {
err = mlx4_cmd_imm(dev, 0, &out_param, RES_COUNTER,
RES_OP_RESERVE, MLX4_CMD_ALLOC_RES,
MLX4_CMD_TIME_CLASS_A, MLX4_CMD_WRAPPED);
if (!err)
*idx = get_param_l(&out_param);
return err;
}
return __mlx4_counter_alloc(dev, idx);
}
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
EXPORT_SYMBOL_GPL(mlx4_counter_alloc);
net/mlx4_core: Reset counters data when freed Add resetting the counter data to the free counter flow, so the counter's data won't be accessible anymore if querying the counter. Also, on next counter allocation (to another VM for example), it will be fresh and clear. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:58:57 +07:00
static int __mlx4_clear_if_stat(struct mlx4_dev *dev,
u8 counter_index)
{
struct mlx4_cmd_mailbox *if_stat_mailbox;
int err;
u32 if_stat_in_mod = (counter_index & 0xff) | MLX4_QUERY_IF_STAT_RESET;
if_stat_mailbox = mlx4_alloc_cmd_mailbox(dev);
if (IS_ERR(if_stat_mailbox))
return PTR_ERR(if_stat_mailbox);
err = mlx4_cmd_box(dev, 0, if_stat_mailbox->dma, if_stat_in_mod, 0,
MLX4_CMD_QUERY_IF_STAT, MLX4_CMD_TIME_CLASS_C,
MLX4_CMD_NATIVE);
mlx4_free_cmd_mailbox(dev, if_stat_mailbox);
return err;
}
net/mlx4_core: Add XRC domains and counters to resource tracker Add missing resource tracking for XRC domains and complete the tracking for HCA network flow counters. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-15 17:35:03 +07:00
void __mlx4_counter_free(struct mlx4_dev *dev, u32 idx)
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
{
net/mlx4_core: Check before cleaning counters bitmap If counters are not supported by the device. The indices bitmap table is not allocated during initialization. Add the symmetrical check before cleaning the counters bitmap table or freeing a counter. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:58:56 +07:00
if (!(dev->caps.flags & MLX4_DEV_CAP_FLAG_COUNTERS))
return;
net/mlx4_core: Allocate default counter per port Default counter per port will be allocated at the mlx4 core driver load. Every QP opened by the Ethernet driver will be attached to the port's default counter. This is an infrastructure step to collect VF statistics from the PF. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:59:02 +07:00
if (idx == MLX4_SINK_COUNTER_INDEX(dev))
return;
net/mlx4_core: Reset counters data when freed Add resetting the counter data to the free counter flow, so the counter's data won't be accessible anymore if querying the counter. Also, on next counter allocation (to another VM for example), it will be fresh and clear. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:58:57 +07:00
__mlx4_clear_if_stat(dev, idx);
mlx4_core: Roll back round robin bitmap allocation commit for CQs, SRQs, and MPTs Commit f4ec9e9 "mlx4_core: Change bitmap allocator to work in round-robin fashion" introduced round-robin allocation (via bitmap) for all resources which allocate via a bitmap. Round robin allocation is desirable for mcgs, counters, pd's, UARs, and xrcds. These are simply numbers, with no involvement of ICM memory mapping. Round robin is required for QPs, since we had a problem with immediate reuse of a 24-bit QP number (commit f4ec9e9). However, for other resources which use the bitmap allocator and involve mapping ICM memory -- MPTs, CQs, SRQs -- round-robin is not desirable. What happens in these cases is the following: ICM memory is allocated and mapped in chunks of 256K. Since the resource allocation index goes up monotonically, the allocator will eventually require mapping a new chunk. Now, chunks are also unmapped when their reference count goes back to zero. Thus, if a single app is running and starts/exits frequently we will have the following situation: When the app starts, a new chunk must be allocated and mapped. When the app exits, the chunk reference count goes back to zero, and the chunk is unmapped and freed. Therefore, the app must pay the cost of allocation and mapping of ICM memory each time it runs (although the price is paid only when allocating the initial entry in the new chunk). For apps which allocate MPTs/SRQs/CQs and which operate as described above, this presented a performance problem. We therefore roll back the round-robin allocator modification for MPTs, CQs, SRQs. Reported-by: Matthew Finlay <matt@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-08 21:50:17 +07:00
mlx4_bitmap_free(&mlx4_priv(dev)->counters_bitmap, idx, MLX4_USE_RR);
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
return;
}
net/mlx4_core: Add XRC domains and counters to resource tracker Add missing resource tracking for XRC domains and complete the tracking for HCA network flow counters. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-15 17:35:03 +07:00
void mlx4_counter_free(struct mlx4_dev *dev, u32 idx)
{
net/mlx4_core: Fix endianness bug in set_param_l The set_param_l function assumes casting a u64 pointer to a u32 pointer allows to access the lower 32bits, but it results in writing the upper 32 bits on big endian systems. The fixed function reads the upper 32 bits of the 64 argument, and or's them with the 32 bits of the 32-bit value passed to the function. Since this is now a "read-modify-write" operation, we got many "unintialized variable" warnings which needed to be fixed as well. Reported-by: Alexander Schmidt <alexschm@de.ibm.com>. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-03-07 10:46:54 +07:00
u64 in_param = 0;
net/mlx4_core: Add XRC domains and counters to resource tracker Add missing resource tracking for XRC domains and complete the tracking for HCA network flow counters. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-05-15 17:35:03 +07:00
if (mlx4_is_mfunc(dev)) {
set_param_l(&in_param, idx);
mlx4_cmd(dev, in_param, RES_COUNTER, RES_OP_RESERVE,
MLX4_CMD_FREE_RES, MLX4_CMD_TIME_CLASS_A,
MLX4_CMD_WRAPPED);
return;
}
__mlx4_counter_free(dev, idx);
}
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
EXPORT_SYMBOL_GPL(mlx4_counter_free);
net/mlx4_core: Allocate default counter per port Default counter per port will be allocated at the mlx4 core driver load. Every QP opened by the Ethernet driver will be attached to the port's default counter. This is an infrastructure step to collect VF statistics from the PF. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:59:02 +07:00
int mlx4_get_default_counter_index(struct mlx4_dev *dev, int port)
{
struct mlx4_priv *priv = mlx4_priv(dev);
return priv->def_counter[port - 1];
}
EXPORT_SYMBOL_GPL(mlx4_get_default_counter_index);
net/mlx4_core: Manage alias GUID per VF Manages alias GUIDs per VF per port in the core layer. This is a pre-step for managing alias GUIDs in a mode that the admin GUID is returned via ib_query_gid() regardless of whether the SM has approved it or not. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2015-03-03 15:54:48 +07:00
void mlx4_set_admin_guid(struct mlx4_dev *dev, __be64 guid, int entry, int port)
{
struct mlx4_priv *priv = mlx4_priv(dev);
priv->mfunc.master.vf_admin[entry].vport[port].guid = guid;
}
EXPORT_SYMBOL_GPL(mlx4_set_admin_guid);
__be64 mlx4_get_admin_guid(struct mlx4_dev *dev, int entry, int port)
{
struct mlx4_priv *priv = mlx4_priv(dev);
return priv->mfunc.master.vf_admin[entry].vport[port].guid;
}
EXPORT_SYMBOL_GPL(mlx4_get_admin_guid);
net/mlx4_core: Set initial admin GUIDs for VFs To have out of the box experience, the PF generates random GUIDs who serve as the initial admin values. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Doug Ledford <dledford@redhat.com>
2015-03-03 16:23:32 +07:00
void mlx4_set_random_admin_guid(struct mlx4_dev *dev, int entry, int port)
{
struct mlx4_priv *priv = mlx4_priv(dev);
__be64 guid;
/* hw GUID */
if (entry == 0)
return;
get_random_bytes((char *)&guid, sizeof(guid));
guid &= ~(cpu_to_be64(1ULL << 56));
guid |= cpu_to_be64(1ULL << 57);
priv->mfunc.master.vf_admin[entry].vport[port].guid = guid;
}
mlx4_core: Fix section mismatches Commit ee49bd93 ("mlx4_core: Reset device when internal error is detected") introduced some section mismatch problems when CONFIG_HOTPLUG=n, because the error recovery code tears down and reinitializes the device after everything is loaded, which ends up calling into lots of code marked __devinit and __devexit from regular .text. Fix this by getting rid of these now-incorrect section markers. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-10-11 05:43:54 +07:00
static int mlx4_setup_hca(struct mlx4_dev *dev)
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
{
struct mlx4_priv *priv = mlx4_priv(dev);
int err;
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
int port;
mlx4_core: Save/restore default port IB capability mask Commit 7ff93f8b ("mlx4_core: Multiple port type support") introduced support for different port types. As part of that support, SET_PORT is invoked to set the port type during driver startup. However, as a side-effect, for IB ports the invocation of this command also sets the port's capability mask to zero (losing the default value set by FW). To fix this, get the default ib port capabilities (via a MAD_IFC Port Info query) during driver startup, and save them for use in the mlx4_SET_PORT command when setting the port-type to Infiniband. This patch fixes problems with subnet manager (SM) failover such as <https://bugs.openfabrics.org/show_bug.cgi?id=1183>, which occurred because the IsTrapSupported bit in the capability mask was zeroed. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-11-29 12:29:46 +07:00
__be32 ib_port_default_caps;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err = mlx4_init_uar_table(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to initialize user access region table, aborting\n");
return err;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
}
err = mlx4_uar_alloc(dev, &priv->driver_uar);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to allocate driver access region, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_uar_table_free;
}
mlx4_{core, ib, en}: Fix driver when sizeof (phys_addr_t) > sizeof (long) Some systems have PCI addresses that don't fit in unsigned long (eg some 32-bit PowerPC 440 systems have 36-bit bus addresses). Fix up mlx4 drivers by using phys_addr_t where appropriate, so we don't truncate any PCI resource addresses before ioremapping them. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2011-01-13 00:50:36 +07:00
priv->kar = ioremap((phys_addr_t) priv->driver_uar.pfn << PAGE_SHIFT, PAGE_SIZE);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (!priv->kar) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Couldn't map kernel access region, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err = -ENOMEM;
goto err_uar_free;
}
err = mlx4_init_pd_table(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to initialize protection domain table, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_kar_unmap;
}
IB/mlx4: Add support for XRC domains Support creating and destroying XRC domains. Any sharing of the XRCD is managed above the low-level driver. Signed-off-by: Sean Hefty <sean.hefty@intel.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-02 23:01:33 +07:00
err = mlx4_init_xrcd_table(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to initialize reliable connection domain table, aborting\n");
IB/mlx4: Add support for XRC domains Support creating and destroying XRC domains. Any sharing of the XRCD is managed above the low-level driver. Signed-off-by: Sean Hefty <sean.hefty@intel.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-02 23:01:33 +07:00
goto err_pd_table_free;
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err = mlx4_init_mr_table(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to initialize memory region table, aborting\n");
IB/mlx4: Add support for XRC domains Support creating and destroying XRC domains. Any sharing of the XRCD is managed above the low-level driver. Signed-off-by: Sean Hefty <sean.hefty@intel.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-02 23:01:33 +07:00
goto err_xrcd_table_free;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
}
net/mlx4_core: Respond to operation request by firmware This commit adds new firmware command and new firmware event. The firmware raises the MLX4_EVENT_TYPE_OP_REQUIRED event in order to signal the driver it needs to perform an administrative operation throughout the MLX4_CMD_GET_OP_REQ command. At the moment the supported operation is adding/removing multicast entries which are used by the firmware for handling NCSI traffic in B0 steering mode. Also, had to swap the order of mlx4_init_mcg_table() and mlx4_init_eq_table() to make sure that driver will get events only after resources are initialized to handle it. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.com> Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-07-28 22:54:21 +07:00
if (!mlx4_is_slave(dev)) {
err = mlx4_init_mcg_table(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to initialize multicast group table, aborting\n");
net/mlx4_core: Respond to operation request by firmware This commit adds new firmware command and new firmware event. The firmware raises the MLX4_EVENT_TYPE_OP_REQUIRED event in order to signal the driver it needs to perform an administrative operation throughout the MLX4_CMD_GET_OP_REQ command. At the moment the supported operation is adding/removing multicast entries which are used by the firmware for handling NCSI traffic in B0 steering mode. Also, had to swap the order of mlx4_init_mcg_table() and mlx4_init_eq_table() to make sure that driver will get events only after resources are initialized to handle it. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.com> Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-07-28 22:54:21 +07:00
goto err_mr_table_free;
}
mlx4_core: Add support for secure-host and SMP firewall Secure-host is the general term for the capability of a device to protect itself and the subnet from malicious host software. This is achieved by: 1. Not allowing un-trusted entities to access device configuration registers, directly (through pci_cr or pci_conf) and indirectly (through MADs). 2. Hiding M_Key from untrusted entities. 3. Preventing the modification of GUID0 by un-trusted entities 4. Not allowing drivers on untrusted hosts to receive nor to transmit packets over QP0 (SMP Firewall). The secure-host capability depends on firmware handling all QP0 packets, and not passing these packets up to the driver. Any information required by the driver for proper operation (e.g., SM lid) is passed via events generated by the firmware while processing QP0 MADs. Driver support mainly requires using the MAD_DEMUX FW command at startup, where the feature is enabled/disabled through a procedure described in the Mellanox HCA tools package. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> [ Fix error path in mlx4_setup_hca to go to err_mcg_table_free. - Roland ] Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-06-01 15:53:50 +07:00
err = mlx4_config_mad_demux(dev);
if (err) {
mlx4_err(dev, "Failed in config_mad_demux, aborting\n");
goto err_mcg_table_free;
}
net/mlx4_core: Respond to operation request by firmware This commit adds new firmware command and new firmware event. The firmware raises the MLX4_EVENT_TYPE_OP_REQUIRED event in order to signal the driver it needs to perform an administrative operation throughout the MLX4_CMD_GET_OP_REQ command. At the moment the supported operation is adding/removing multicast entries which are used by the firmware for handling NCSI traffic in B0 steering mode. Also, had to swap the order of mlx4_init_mcg_table() and mlx4_init_eq_table() to make sure that driver will get events only after resources are initialized to handle it. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.com> Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-07-28 22:54:21 +07:00
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err = mlx4_init_eq_table(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to initialize event queue table, aborting\n");
net/mlx4_core: Respond to operation request by firmware This commit adds new firmware command and new firmware event. The firmware raises the MLX4_EVENT_TYPE_OP_REQUIRED event in order to signal the driver it needs to perform an administrative operation throughout the MLX4_CMD_GET_OP_REQ command. At the moment the supported operation is adding/removing multicast entries which are used by the firmware for handling NCSI traffic in B0 steering mode. Also, had to swap the order of mlx4_init_mcg_table() and mlx4_init_eq_table() to make sure that driver will get events only after resources are initialized to handle it. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.com> Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-07-28 22:54:21 +07:00
goto err_mcg_table_free;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
}
err = mlx4_cmd_use_events(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to switch to event-driven firmware commands, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_eq_table_free;
}
err = mlx4_NOP(dev);
if (err) {
mlx4_core: Enable MSI-X by default Recover from MSI-X errors by automatically falling back on regular interrupt, instead of asking the user to do this manually. This makes it possible to enable MSI-X by default, and will make it possible to get rid of the msi_x module option in the future. Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-07 20:08:28 +07:00
if (dev->flags & MLX4_FLAG_MSI_X) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_warn(dev, "NOP command failed to generate MSI-X interrupt IRQ %d)\n",
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
priv->eq_table.eq[MLX4_EQ_ASYNC].irq);
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_warn(dev, "Trying again without MSI-X\n");
mlx4_core: Enable MSI-X by default Recover from MSI-X errors by automatically falling back on regular interrupt, instead of asking the user to do this manually. This makes it possible to enable MSI-X by default, and will make it possible to get rid of the msi_x module option in the future. Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-07 20:08:28 +07:00
} else {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "NOP command failed to generate interrupt (IRQ %d), aborting\n",
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
priv->eq_table.eq[MLX4_EQ_ASYNC].irq);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_err(dev, "BIOS or ACPI interrupt routing problem?\n");
mlx4_core: Enable MSI-X by default Recover from MSI-X errors by automatically falling back on regular interrupt, instead of asking the user to do this manually. This makes it possible to enable MSI-X by default, and will make it possible to get rid of the msi_x module option in the future. Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-07 20:08:28 +07:00
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_cmd_poll;
}
mlx4_dbg(dev, "NOP command IRQ test passed\n");
err = mlx4_init_cq_table(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to initialize completion queue table, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_cmd_poll;
}
err = mlx4_init_srq_table(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to initialize shared receive queue table, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_cq_table_free;
}
err = mlx4_init_qp_table(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to initialize queue pair table, aborting\n");
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto err_srq_table_free;
}
net/mlx4_core: Remove counters table allocation from VF flow Since virtual functions get their counters indices allocation from the PF, allocate counters indices bitmap only in case the function isn't virtual. Also, check that the device has counters to allocate before creating the indices bitmap table. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:58:59 +07:00
if (!mlx4_is_slave(dev)) {
err = mlx4_init_counters_table(dev);
if (err && err != -ENOENT) {
mlx4_err(dev, "Failed to initialize counters table, aborting\n");
goto err_qp_table_free;
}
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
}
net/mlx4_core: Allocate default counter per port Default counter per port will be allocated at the mlx4 core driver load. Every QP opened by the Ethernet driver will be attached to the port's default counter. This is an infrastructure step to collect VF statistics from the PF. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:59:02 +07:00
err = mlx4_allocate_default_counters(dev);
if (err) {
mlx4_err(dev, "Failed to allocate default counters, aborting\n");
goto err_counters_table_free;
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if (!mlx4_is_slave(dev)) {
for (port = 1; port <= dev->caps.num_ports; port++) {
ib_port_default_caps = 0;
err = mlx4_get_port_ib_caps(dev, port,
&ib_port_default_caps);
if (err)
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_warn(dev, "failed to get port %d default ib capabilities (%d). Continuing with caps = 0\n",
port, err);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
dev->caps.ib_port_def_cap[port] = ib_port_default_caps;
net/mlx4_core: Initialize IB port capabilities for all slaves With IB SR-IOV, each slave has its own separate copy of the port capabilities flags. For example, the master can run a subnet manager (which causes the IsSM bit to be set in the master's port capabilities) without affecting the port capabilities seen by the slaves (the IsSM bit will be seen as cleared in the slaves). Also add a static inline mlx4_master_func_num() to enhance readability of the code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:41 +07:00
/* initialize per-slave default ib port capabilities */
if (mlx4_is_master(dev)) {
int i;
for (i = 0; i < dev->num_slaves; i++) {
if (i == mlx4_master_func_num(dev))
continue;
priv->mfunc.master.slave_state[i].ib_cap_mask[port] =
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
ib_port_default_caps;
net/mlx4_core: Initialize IB port capabilities for all slaves With IB SR-IOV, each slave has its own separate copy of the port capabilities flags. For example, the master can run a subnet manager (which causes the IsSM bit to be set in the master's port capabilities) without affecting the port capabilities seen by the slaves (the IsSM bit will be seen as cleared in the slaves). Also add a static inline mlx4_master_func_num() to enhance readability of the code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:41 +07:00
}
}
mlx4_core: Allow dynamic MTU configuration for IB ports Set the MTU for IB ports in the driver instead of using the firmware default of 2KB (the driver defaults to 4KB). Allow for dynamic mtu configuration through a new, per-port sysfs entry. Since there's a dependency between the port MTU and the max number of HW VLs the port can support, apply a mim/max approach, using a loop that goes down from the highest possible number of VLs to the lowest, using the firmware return status to know whether the requested number of VLs is possible with a given MTU. For now, as with the dynamic link type change / VPI support, the sysfs entry to change the mtu is exposed only when NOT running in SR-IOV mode. To allow changing the MTU for the master in SR-IOV mode, primary-function-initiated FLR (Function Level Reset) needs to be implemented. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-01-12 00:02:17 +07:00
if (mlx4_is_mfunc(dev))
dev->caps.port_ib_mtu[port] = IB_MTU_2048;
else
dev->caps.port_ib_mtu[port] = IB_MTU_4096;
mlx4_core: Add extended port capabilities support An Extended Port Info packet is sent to each hw port during HCA init. If it returns without error, we assume the port supports extended port capabilities. Signed-off-by: Marcel Apfelbaum <marcela@dev.mellanox.co.il> Reviewed-by: Jack Morgenstein <jackm@dev.mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-10-24 16:02:34 +07:00
mlx4: Put physical GID and P_Key table sizes in mlx4_phys_caps struct and paravirtualize them To allow easy paravirtualization of P_Key and GID table sizes, keep paravirtualized sizes in mlx4_dev->caps, but save the actual physical sizes from FW in struct: mlx4_dev->phys_cap. In addition, in SR-IOV mode, do the following: 1. Reduce reported P_Key table size by 1. This is done to reserve the highest P_Key index for internal use, for declaring an invalid P_Key in P_Key paravirtualization. We require a P_Key index which always contain an invalid P_Key value for this purpose (i.e., one which cannot be modified by the subnet manager). The way to do this is to reduce the P_Key table size reported to the subnet manager by 1, so that it will not attempt to access the P_Key at index #127. 2. Paravirtualize the GID table size to 1. Thus, each guest sees only a single GID (at its paravirtualized index 0). In addition, since we are paravirtualizing the GID table size to 1, we add paravirtualization of the master GID event here (i.e., we do not do ib_dispatch_event() for the GUID change event on the master, since its (only) GUID never changes). Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-06-19 15:21:44 +07:00
err = mlx4_SET_PORT(dev, port, mlx4_is_master(dev) ?
dev->caps.pkey_table_len[port] : -1);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if (err) {
mlx4_err(dev, "Failed to set port %d, aborting\n",
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
port);
net/mlx4_core: Allocate default counter per port Default counter per port will be allocated at the mlx4 core driver load. Every QP opened by the Ethernet driver will be attached to the port's default counter. This is an infrastructure step to collect VF statistics from the PF. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:59:02 +07:00
goto err_default_countes_free;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
}
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return 0;
net/mlx4_core: Allocate default counter per port Default counter per port will be allocated at the mlx4 core driver load. Every QP opened by the Ethernet driver will be attached to the port's default counter. This is an infrastructure step to collect VF statistics from the PF. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:59:02 +07:00
err_default_countes_free:
mlx4_cleanup_default_counters(dev);
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
err_counters_table_free:
net/mlx4_core: Remove counters table allocation from VF flow Since virtual functions get their counters indices allocation from the PF, allocate counters indices bitmap only in case the function isn't virtual. Also, check that the device has counters to allocate before creating the indices bitmap table. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:58:59 +07:00
if (!mlx4_is_slave(dev))
mlx4_cleanup_counters_table(dev);
mlx4_core: Add network flow counters ConnectX devices support a set of flow counters that can be attached to a set containing one or more QPs. Each such counter tracks receive and transmit packets and bytes of these QPs. This patch queries the device to check support for counters, handles initialization of the HCA to enable counters, and initializes a bitmap allocator to control counter allocations. Derived from patch by Eli Cohen <eli@mellanox.co.il>. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-15 21:47:14 +07:00
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err_qp_table_free:
mlx4_cleanup_qp_table(dev);
err_srq_table_free:
mlx4_cleanup_srq_table(dev);
err_cq_table_free:
mlx4_cleanup_cq_table(dev);
err_cmd_poll:
mlx4_cmd_use_polling(dev);
err_eq_table_free:
mlx4_cleanup_eq_table(dev);
net/mlx4_core: Respond to operation request by firmware This commit adds new firmware command and new firmware event. The firmware raises the MLX4_EVENT_TYPE_OP_REQUIRED event in order to signal the driver it needs to perform an administrative operation throughout the MLX4_CMD_GET_OP_REQ command. At the moment the supported operation is adding/removing multicast entries which are used by the firmware for handling NCSI traffic in B0 steering mode. Also, had to swap the order of mlx4_init_mcg_table() and mlx4_init_eq_table() to make sure that driver will get events only after resources are initialized to handle it. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.com> Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-07-28 22:54:21 +07:00
err_mcg_table_free:
if (!mlx4_is_slave(dev))
mlx4_cleanup_mcg_table(dev);
mlx4_core: Reset device when internal error is detected Reset the device when an internal error is detected. Also, detect errors by polling the error buffer rather than using interrupts. This is more robust and doesn't depend on MSI-X. Remove the old interrupt handler entirely, since we don't want to support two mechanisms for detecting internal errors. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-07-12 21:50:45 +07:00
err_mr_table_free:
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_cleanup_mr_table(dev);
IB/mlx4: Add support for XRC domains Support creating and destroying XRC domains. Any sharing of the XRCD is managed above the low-level driver. Signed-off-by: Sean Hefty <sean.hefty@intel.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-02 23:01:33 +07:00
err_xrcd_table_free:
mlx4_cleanup_xrcd_table(dev);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err_pd_table_free:
mlx4_cleanup_pd_table(dev);
err_kar_unmap:
iounmap(priv->kar);
err_uar_free:
mlx4_uar_free(dev, &priv->driver_uar);
err_uar_table_free:
mlx4_cleanup_uar_table(dev);
return err;
}
net/mlx4_core: Move affinity hints to mlx4_core ownership Now that EQs management is in the sole responsibility of mlx4_core, the IRQ affinity hints configuration should be in its hands as well. request_irq is called only once by the first consumer (maybe mlx4_ib), so mlx4_en passes the affinity mask too late. We also need to request vectors according to the cores we want to run on. mlx4_core distribution of IRQs to cores is straight forward, EQ(i)->IRQ will set affinity hint to core i. Consumers need to request EQ vectors, according to their cores considerations (NUMA). Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:17 +07:00
static int mlx4_init_affinity_hint(struct mlx4_dev *dev, int port, int eqn)
{
int requested_cpu = 0;
struct mlx4_priv *priv = mlx4_priv(dev);
struct mlx4_eq *eq;
int off = 0;
int i;
if (eqn > dev->caps.num_comp_vectors)
return -EINVAL;
for (i = 1; i < port; i++)
off += mlx4_get_eqs_per_port(dev, i);
requested_cpu = eqn - off - !!(eqn > MLX4_EQ_ASYNC);
/* Meaning EQs are shared, and this call comes from the second port */
if (requested_cpu < 0)
return 0;
eq = &priv->eq_table.eq[eqn];
if (!zalloc_cpumask_var(&eq->affinity_mask, GFP_KERNEL))
return -ENOMEM;
cpumask_set_cpu(requested_cpu, eq->affinity_mask);
return 0;
}
mlx4_core: Fix more section mismatches Commit 3d73c288 ("mlx4_core: Fix section mismatches") fixed some of the section mismatches introduced when error recovery was added, but there were still more cases of errory recovery code calling into __devinit code from regular .text. Fix this by getting rid of the now-incorrect __devinit annotations. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-02-05 11:20:41 +07:00
static void mlx4_enable_msi_x(struct mlx4_dev *dev)
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
{
struct mlx4_priv *priv = mlx4_priv(dev);
mlx4_core: Add support for multiple completion event vectors When using MSI-X mode, create a completion event queue for each CPU. Report the number of completion EQs in a new struct mlx4_caps member, num_comp_vectors, and extend the mlx4_cq_alloc() interface with a vector parameter so that consumers can specify which completion EQ should be used to report events for the CQ being created. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-12-22 22:15:03 +07:00
struct msix_entry *entries;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
int i;
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
int port = 0;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (msi_x) {
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
int nreq = dev->caps.num_ports * num_online_cpus() + 1;
net/mlx4_core: Capping number of requested MSIXs to MAX_MSIX We currently manage IRQs in pool_bm which is a bit field of MAX_MSIX bits. Thus, allocating more than MAX_MSIX interrupts can't be managed in pool_bm. Fixing this by capping number of requested MSIXs to MAX_MSIX. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-28 02:43:25 +07:00
bool shared_ports = false;
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
net/mlx4_core: Set number of msix vectors under SRIOV mode to firmware defaults The lines if (mlx4_is_mfunc(dev)) { nreq = 2; } else { which hard code the number of requested msi-x vectors under multi-function mode to two can be removed completely, since the firmware sets num_eqs and reserved_eqs appropriately Thus, the code line: nreq = min_t(int, dev->caps.num_eqs - dev->caps.reserved_eqs, nreq); is by itself sufficient and correct for all cases. Currently, for mfunc mode num_eqs = 32 and reserved_eqs = 28, hence four vectors will be enabled. This triples (one vector is used for the async events and commands EQ) the horse power provided for processing of incoming packets on netdev RSS scheme, IO initiators/targets commands processing flows, etc. Reviewed-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-01-17 12:30:43 +07:00
nreq = min_t(int, dev->caps.num_eqs - dev->caps.reserved_eqs,
nreq);
net/mlx4_core: Capping number of requested MSIXs to MAX_MSIX We currently manage IRQs in pool_bm which is a bit field of MAX_MSIX bits. Thus, allocating more than MAX_MSIX interrupts can't be managed in pool_bm. Fixing this by capping number of requested MSIXs to MAX_MSIX. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-28 02:43:25 +07:00
if (nreq > MAX_MSIX) {
nreq = MAX_MSIX;
shared_ports = true;
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
mlx4_core: Add support for multiple completion event vectors When using MSI-X mode, create a completion event queue for each CPU. Report the number of completion EQs in a new struct mlx4_caps member, num_comp_vectors, and extend the mlx4_cq_alloc() interface with a vector parameter so that consumers can specify which completion EQ should be used to report events for the CQ being created. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-12-22 22:15:03 +07:00
entries = kcalloc(nreq, sizeof *entries, GFP_KERNEL);
if (!entries)
goto no_msi;
for (i = 0; i < nreq; ++i)
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
entries[i].entry = i;
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
nreq = pci_enable_msix_range(dev->persist->pdev, entries, 2,
nreq);
mlx4: Use pci_enable_msix_range() instead of pci_enable_msix() As result of deprecation of MSI-X/MSI enablement functions pci_enable_msix() and pci_enable_msi_block() all drivers using these two interfaces need to be updated to use the new pci_enable_msi_range() and pci_enable_msix_range() interfaces. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Cc: "David S. Miller" <davem@davemloft.net> Cc: Amir Vadai <amirv@mellanox.com> Cc: netdev@vger.kernel.org Cc: linux-pci@vger.kernel.org Acked-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-02-18 17:11:47 +07:00
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
if (nreq < 0 || nreq < MLX4_EQ_ASYNC) {
mlx4_core: Fix memory leak in mlx4_enable_msi_x() When the msi_x option is enabled but pci_enable_msix() fails (not enough vectors are available etc), the entries array was not freed on the error path. Signed-off-by: Nicolas Morey-Chaisemartin <nicolas.morey-chaisemartin@ext.bull.net> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-04-22 00:11:06 +07:00
kfree(entries);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
goto no_msi;
mlx4: Changing interrupt scheme Adding a pool of MSI-X vectors and EQs that can be used explicitly by mlx4_core customers (mlx4_ib, mlx4_en). The consumers will assign their own names to the interrupt vectors. Those vectors are not opened at mlx4 device initialization, opened by demand. Changed the max number of possible EQs according to the new scheme, no longer relies on on number of cores. The new functionality is exposed through mlx4_assign_eq() and mlx4_release_eq(). Customers that do not use the new API will get completion vectors as before. Signed-off-by: Markuze Alex <markuze@mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:37:47 +07:00
}
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
/* 1 is reserved for events (asyncrounous EQ) */
dev->caps.num_comp_vectors = nreq - 1;
priv->eq_table.eq[MLX4_EQ_ASYNC].irq = entries[0].vector;
bitmap_zero(priv->eq_table.eq[MLX4_EQ_ASYNC].actv_ports.ports,
dev->caps.num_ports);
net/mlx4_core: Capping number of requested MSIXs to MAX_MSIX We currently manage IRQs in pool_bm which is a bit field of MAX_MSIX bits. Thus, allocating more than MAX_MSIX interrupts can't be managed in pool_bm. Fixing this by capping number of requested MSIXs to MAX_MSIX. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-28 02:43:25 +07:00
if (MLX4_IS_LEGACY_EQ_MODE(dev->caps))
shared_ports = true;
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
for (i = 0; i < dev->caps.num_comp_vectors + 1; i++) {
if (i == MLX4_EQ_ASYNC)
continue;
priv->eq_table.eq[i].irq =
entries[i + 1 - !!(i > MLX4_EQ_ASYNC)].vector;
net/mlx4_core: Capping number of requested MSIXs to MAX_MSIX We currently manage IRQs in pool_bm which is a bit field of MAX_MSIX bits. Thus, allocating more than MAX_MSIX interrupts can't be managed in pool_bm. Fixing this by capping number of requested MSIXs to MAX_MSIX. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-08-28 02:43:25 +07:00
if (shared_ports) {
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
bitmap_fill(priv->eq_table.eq[i].actv_ports.ports,
dev->caps.num_ports);
net/mlx4_core: Move affinity hints to mlx4_core ownership Now that EQs management is in the sole responsibility of mlx4_core, the IRQ affinity hints configuration should be in its hands as well. request_irq is called only once by the first consumer (maybe mlx4_ib), so mlx4_en passes the affinity mask too late. We also need to request vectors according to the cores we want to run on. mlx4_core distribution of IRQs to cores is straight forward, EQ(i)->IRQ will set affinity hint to core i. Consumers need to request EQ vectors, according to their cores considerations (NUMA). Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:17 +07:00
/* We don't set affinity hint when there
* aren't enough EQs
*/
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
} else {
set_bit(port,
priv->eq_table.eq[i].actv_ports.ports);
net/mlx4_core: Move affinity hints to mlx4_core ownership Now that EQs management is in the sole responsibility of mlx4_core, the IRQ affinity hints configuration should be in its hands as well. request_irq is called only once by the first consumer (maybe mlx4_ib), so mlx4_en passes the affinity mask too late. We also need to request vectors according to the cores we want to run on. mlx4_core distribution of IRQs to cores is straight forward, EQ(i)->IRQ will set affinity hint to core i. Consumers need to request EQ vectors, according to their cores considerations (NUMA). Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:17 +07:00
if (mlx4_init_affinity_hint(dev, port + 1, i))
mlx4_warn(dev, "Couldn't init hint cpumask for EQ %d\n",
i);
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
}
/* We divide the Eqs evenly between the two ports.
* (dev->caps.num_comp_vectors / dev->caps.num_ports)
* refers to the number of Eqs per port
* (i.e eqs_per_port). Theoretically, we would like to
* write something like (i + 1) % eqs_per_port == 0.
* However, since there's an asynchronous Eq, we have
* to skip over it by comparing this condition to
* !!((i + 1) > MLX4_EQ_ASYNC).
*/
if ((dev->caps.num_comp_vectors > dev->caps.num_ports) &&
((i + 1) %
(dev->caps.num_comp_vectors / dev->caps.num_ports)) ==
!!((i + 1) > MLX4_EQ_ASYNC))
/* If dev->caps.num_comp_vectors < dev->caps.num_ports,
* everything is shared anyway.
*/
port++;
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
dev->flags |= MLX4_FLAG_MSI_X;
mlx4_core: Add support for multiple completion event vectors When using MSI-X mode, create a completion event queue for each CPU. Report the number of completion EQs in a new struct mlx4_caps member, num_comp_vectors, and extend the mlx4_cq_alloc() interface with a vector parameter so that consumers can specify which completion EQ should be used to report events for the CQ being created. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-12-22 22:15:03 +07:00
kfree(entries);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return;
}
no_msi:
mlx4_core: Add support for multiple completion event vectors When using MSI-X mode, create a completion event queue for each CPU. Report the number of completion EQs in a new struct mlx4_caps member, num_comp_vectors, and extend the mlx4_cq_alloc() interface with a vector parameter so that consumers can specify which completion EQ should be used to report events for the CQ being created. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-12-22 22:15:03 +07:00
dev->caps.num_comp_vectors = 1;
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
BUG_ON(MLX4_EQ_ASYNC >= 2);
for (i = 0; i < 2; ++i) {
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
priv->eq_table.eq[i].irq = dev->persist->pdev->irq;
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
if (i != MLX4_EQ_ASYNC) {
bitmap_fill(priv->eq_table.eq[i].actv_ports.ports,
dev->caps.num_ports);
}
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
}
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
static int mlx4_init_port_info(struct mlx4_dev *dev, int port)
mlx4_core: Ethernet MAC/VLAN management Add support for managing MAC and VLAN filters for each port. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Oren Duer <oren@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 01:44:46 +07:00
{
struct mlx4_port_info *info = &mlx4_priv(dev)->port[port];
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
int err = 0;
mlx4_core: Ethernet MAC/VLAN management Add support for managing MAC and VLAN filters for each port. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Oren Duer <oren@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 01:44:46 +07:00
info->dev = dev;
info->port = port;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if (!mlx4_is_slave(dev)) {
mlx4_init_mac_table(dev, &info->mac_table);
mlx4_init_vlan_table(dev, &info->vlan_table);
net/mlx4_core: Reset RoCE VF gids when guest driver goes down Reset the GIDs assigned to a VF in the port RoCE GID table when that guest goes down (either crashes or goes down cleanly). As part of this fix, we refactor the RoCE gid table driver copy, moving it to the mlx4_port_info structure (together with the MAC and VLAN tables). As with the MAC and VLAN tables, we now use a mutex per port for the GID table so that modifying the driver copy and modifying the firmware copy of a port GID table becomes an atomic operation (thus avoiding driver-copy/FW-copy mismatches). Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-27 13:26:38 +07:00
mlx4_init_roce_gid_table(dev, &info->gid_table);
net/mlx4: Move Ethernet related functionality from mlx4_core to mlx4_en Move low level code that deals with management of Ethernet MACs and QPs from mlx4_core to mlx4_en. Also convert the new functions to deal with MACs in form of char array instead of u64. Actual functions moved: mlx4_replace_mac mlx4_get_eth_qp mlx4_put_eth_qp To conduct this change, some functionality had to be exported from the core, the following functions were added: mlx4_get_base_qp __mlx4_replace_mac (low level function for CX1/A0 compatibility) Signed-off-by: Yan Burman <yanb@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-02-07 09:25:22 +07:00
info->base_qpn = mlx4_get_base_qpn(dev, port);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
sprintf(info->dev_name, "mlx4_port%d", port);
info->port_attr.attr.name = info->dev_name;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if (mlx4_is_mfunc(dev))
info->port_attr.attr.mode = S_IRUGO;
else {
info->port_attr.attr.mode = S_IRUGO | S_IWUSR;
info->port_attr.store = set_port_type;
}
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
info->port_attr.show = show_port_type;
sysfs: fix sysfs lockdep warning in mlx4 code This fixes a sysfs lockdep warning in the mlx4 code. Cc: Yinghai Lu <yinghai@kernel.org> Cc: Eric Biederman <ebiederm@xmission.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2010-03-16 04:01:55 +07:00
sysfs_attr_init(&info->port_attr.attr);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
err = device_create_file(&dev->persist->pdev->dev, &info->port_attr);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
if (err) {
mlx4_err(dev, "Failed to create file for port %d\n", port);
info->port = -1;
}
mlx4_core: Allow dynamic MTU configuration for IB ports Set the MTU for IB ports in the driver instead of using the firmware default of 2KB (the driver defaults to 4KB). Allow for dynamic mtu configuration through a new, per-port sysfs entry. Since there's a dependency between the port MTU and the max number of HW VLs the port can support, apply a mim/max approach, using a loop that goes down from the highest possible number of VLs to the lowest, using the firmware return status to know whether the requested number of VLs is possible with a given MTU. For now, as with the dynamic link type change / VPI support, the sysfs entry to change the mtu is exposed only when NOT running in SR-IOV mode. To allow changing the MTU for the master in SR-IOV mode, primary-function-initiated FLR (Function Level Reset) needs to be implemented. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-01-12 00:02:17 +07:00
sprintf(info->dev_mtu_name, "mlx4_port%d_mtu", port);
info->port_mtu_attr.attr.name = info->dev_mtu_name;
if (mlx4_is_mfunc(dev))
info->port_mtu_attr.attr.mode = S_IRUGO;
else {
info->port_mtu_attr.attr.mode = S_IRUGO | S_IWUSR;
info->port_mtu_attr.store = set_port_ib_mtu;
}
info->port_mtu_attr.show = show_port_ib_mtu;
sysfs_attr_init(&info->port_mtu_attr.attr);
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
err = device_create_file(&dev->persist->pdev->dev,
&info->port_mtu_attr);
mlx4_core: Allow dynamic MTU configuration for IB ports Set the MTU for IB ports in the driver instead of using the firmware default of 2KB (the driver defaults to 4KB). Allow for dynamic mtu configuration through a new, per-port sysfs entry. Since there's a dependency between the port MTU and the max number of HW VLs the port can support, apply a mim/max approach, using a loop that goes down from the highest possible number of VLs to the lowest, using the firmware return status to know whether the requested number of VLs is possible with a given MTU. For now, as with the dynamic link type change / VPI support, the sysfs entry to change the mtu is exposed only when NOT running in SR-IOV mode. To allow changing the MTU for the master in SR-IOV mode, primary-function-initiated FLR (Function Level Reset) needs to be implemented. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-01-12 00:02:17 +07:00
if (err) {
mlx4_err(dev, "Failed to create mtu file for port %d\n", port);
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
device_remove_file(&info->dev->persist->pdev->dev,
&info->port_attr);
mlx4_core: Allow dynamic MTU configuration for IB ports Set the MTU for IB ports in the driver instead of using the firmware default of 2KB (the driver defaults to 4KB). Allow for dynamic mtu configuration through a new, per-port sysfs entry. Since there's a dependency between the port MTU and the max number of HW VLs the port can support, apply a mim/max approach, using a loop that goes down from the highest possible number of VLs to the lowest, using the firmware return status to know whether the requested number of VLs is possible with a given MTU. For now, as with the dynamic link type change / VPI support, the sysfs entry to change the mtu is exposed only when NOT running in SR-IOV mode. To allow changing the MTU for the master in SR-IOV mode, primary-function-initiated FLR (Function Level Reset) needs to be implemented. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-01-12 00:02:17 +07:00
info->port = -1;
}
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
return err;
}
static void mlx4_cleanup_port_info(struct mlx4_port_info *info)
{
if (info->port < 0)
return;
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
device_remove_file(&info->dev->persist->pdev->dev, &info->port_attr);
device_remove_file(&info->dev->persist->pdev->dev,
&info->port_mtu_attr);
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
#ifdef CONFIG_RFS_ACCEL
free_irq_cpu_rmap(info->rmap);
info->rmap = NULL;
#endif
mlx4_core: Ethernet MAC/VLAN management Add support for managing MAC and VLAN filters for each port. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Oren Duer <oren@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 01:44:46 +07:00
}
mlx4: Add support for promiscuous mode in the new steering model. For Ethernet mode only, When we want to register QP as promiscuous, it must be added to all the existing steering entries and also to the default one. The promiscuous QP might also be on of "real" QPs, which means we need to monitor every entry to avoid duplicates and ensure we close an entry when all it has is promiscuous QPs. Same mechanism both for unicast and multicast. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:38:24 +07:00
static int mlx4_init_steering(struct mlx4_dev *dev)
{
struct mlx4_priv *priv = mlx4_priv(dev);
int num_entries = dev->caps.num_ports;
int i, j;
priv->steer = kzalloc(sizeof(struct mlx4_steer) * num_entries, GFP_KERNEL);
if (!priv->steer)
return -ENOMEM;
mlx4: remove unused field high_prios Remove unnecessary field high_prios from mlx4_steer struct and initialization Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il> Reviewed-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-14 13:37:41 +07:00
for (i = 0; i < num_entries; i++)
mlx4: Add support for promiscuous mode in the new steering model. For Ethernet mode only, When we want to register QP as promiscuous, it must be added to all the existing steering entries and also to the default one. The promiscuous QP might also be on of "real" QPs, which means we need to monitor every entry to avoid duplicates and ensure we close an entry when all it has is promiscuous QPs. Same mechanism both for unicast and multicast. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:38:24 +07:00
for (j = 0; j < MLX4_NUM_STEERS; j++) {
INIT_LIST_HEAD(&priv->steer[i].promisc_qps[j]);
INIT_LIST_HEAD(&priv->steer[i].steer_entries[j]);
}
return 0;
}
static void mlx4_clear_steering(struct mlx4_dev *dev)
{
struct mlx4_priv *priv = mlx4_priv(dev);
struct mlx4_steer_index *entry, *tmp_entry;
struct mlx4_promisc_qp *pqp, *tmp_pqp;
int num_entries = dev->caps.num_ports;
int i, j;
for (i = 0; i < num_entries; i++) {
for (j = 0; j < MLX4_NUM_STEERS; j++) {
list_for_each_entry_safe(pqp, tmp_pqp,
&priv->steer[i].promisc_qps[j],
list) {
list_del(&pqp->list);
kfree(pqp);
}
list_for_each_entry_safe(entry, tmp_entry,
&priv->steer[i].steer_entries[j],
list) {
list_del(&entry->list);
list_for_each_entry_safe(pqp, tmp_pqp,
&entry->duplicates,
list) {
list_del(&pqp->list);
kfree(pqp);
}
kfree(entry);
}
}
}
kfree(priv->steer);
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
static int extended_func_num(struct pci_dev *pdev)
{
return PCI_SLOT(pdev->devfn) * 8 + PCI_FUNC(pdev->devfn);
}
#define MLX4_OWNER_BASE 0x8069c
#define MLX4_OWNER_SIZE 4
static int mlx4_get_ownership(struct mlx4_dev *dev)
{
void __iomem *owner;
u32 ret;
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
if (pci_channel_offline(dev->persist->pdev))
mlx4: Add support for EEH error recovery Currently the mlx4 drivers don't have the necessary callbacks to implement EEH errors detection and recovery, so the PCI layer uses the probe and remove callbacks to try to recover the device after an error on the bus. However, these callbacks have race conditions with the internal catastrophic error recovery functions, which will also detect the error and this can cause the system to crash if both EEH and catas functions try to reset the device. This patch adds the necessary error recovery callbacks and makes sure that the internal catastrophic error functions will not try to reset the device in such scenarios. It also adds some calls to pci_channel_offline() to suppress reads/writes on the bus when the slot cannot accept I/O operations so we prevent unnecessary accesses to the bus and speed up the device removal. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Shlomo Pongratz <shlomop@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-20 16:55:43 +07:00
return -EIO;
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
owner = ioremap(pci_resource_start(dev->persist->pdev, 0) +
MLX4_OWNER_BASE,
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
MLX4_OWNER_SIZE);
if (!owner) {
mlx4_err(dev, "Failed to obtain ownership bit\n");
return -ENOMEM;
}
ret = readl(owner);
iounmap(owner);
return (int) !!ret;
}
static void mlx4_free_ownership(struct mlx4_dev *dev)
{
void __iomem *owner;
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
if (pci_channel_offline(dev->persist->pdev))
mlx4: Add support for EEH error recovery Currently the mlx4 drivers don't have the necessary callbacks to implement EEH errors detection and recovery, so the PCI layer uses the probe and remove callbacks to try to recover the device after an error on the bus. However, these callbacks have race conditions with the internal catastrophic error recovery functions, which will also detect the error and this can cause the system to crash if both EEH and catas functions try to reset the device. This patch adds the necessary error recovery callbacks and makes sure that the internal catastrophic error functions will not try to reset the device in such scenarios. It also adds some calls to pci_channel_offline() to suppress reads/writes on the bus when the slot cannot accept I/O operations so we prevent unnecessary accesses to the bus and speed up the device removal. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Shlomo Pongratz <shlomop@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-20 16:55:43 +07:00
return;
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
owner = ioremap(pci_resource_start(dev->persist->pdev, 0) +
MLX4_OWNER_BASE,
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
MLX4_OWNER_SIZE);
if (!owner) {
mlx4_err(dev, "Failed to obtain ownership bit\n");
return;
}
writel(0, owner);
msleep(1000);
iounmap(owner);
}
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
#define SRIOV_VALID_STATE(flags) (!!((flags) & MLX4_FLAG_SRIOV) ==\
!!((flags) & MLX4_FLAG_MASTER))
static u64 mlx4_enable_sriov(struct mlx4_dev *dev, struct pci_dev *pdev,
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
u8 total_vfs, int existing_vfs, int reset_flow)
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
{
u64 dev_flags = dev->flags;
net/mlx4_core: Fixed memory leak and incorrect refcount in mlx4_load_one The current mlx4_load_one has a memory leak as it always allocates dev_cap, but frees it only on error. In addition, even if VFs exist when mlx4_load_one is called, we still need to notify probed VFs that we're loading (by incrementing pf_loading). Fixes: a0eacca948d2 ('net/mlx4_core: Refactor mlx4_load_one') Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-14 21:18:04 +07:00
int err = 0;
net/mlx4_core: Add extra check for total vfs for SRIOV Add extra check for total vfs for SRIOV to check if that value is bigger than total vfs in pci SRIOV capabalities. Fix a check and print of the number of maximum vfs that hw can handle. Fix a check and print of the number of maximum vfs per port that driver can handle. Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-06 21:20:19 +07:00
int fw_enabled_sriov_vfs = min(pci_sriov_get_totalvfs(pdev),
MLX4_MAX_NUM_VF);
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
if (reset_flow) {
dev->dev_vfs = kcalloc(total_vfs, sizeof(*dev->dev_vfs),
GFP_KERNEL);
if (!dev->dev_vfs)
goto free_mem;
return dev_flags;
}
net/mlx4_core: Fixed memory leak and incorrect refcount in mlx4_load_one The current mlx4_load_one has a memory leak as it always allocates dev_cap, but frees it only on error. In addition, even if VFs exist when mlx4_load_one is called, we still need to notify probed VFs that we're loading (by incrementing pf_loading). Fixes: a0eacca948d2 ('net/mlx4_core: Refactor mlx4_load_one') Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-14 21:18:04 +07:00
atomic_inc(&pf_loading);
if (dev->flags & MLX4_FLAG_SRIOV) {
if (existing_vfs != total_vfs) {
mlx4_err(dev, "SR-IOV was already enabled, but with num_vfs (%d) different than requested (%d)\n",
existing_vfs, total_vfs);
total_vfs = existing_vfs;
}
}
dev->dev_vfs = kzalloc(total_vfs * sizeof(*dev->dev_vfs), GFP_KERNEL);
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
if (NULL == dev->dev_vfs) {
mlx4_err(dev, "Failed to allocate memory for VFs\n");
goto disable_sriov;
net/mlx4_core: Fixed memory leak and incorrect refcount in mlx4_load_one The current mlx4_load_one has a memory leak as it always allocates dev_cap, but frees it only on error. In addition, even if VFs exist when mlx4_load_one is called, we still need to notify probed VFs that we're loading (by incrementing pf_loading). Fixes: a0eacca948d2 ('net/mlx4_core: Refactor mlx4_load_one') Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-14 21:18:04 +07:00
}
if (!(dev->flags & MLX4_FLAG_SRIOV)) {
net/mlx4_core: Add extra check for total vfs for SRIOV Add extra check for total vfs for SRIOV to check if that value is bigger than total vfs in pci SRIOV capabalities. Fix a check and print of the number of maximum vfs that hw can handle. Fix a check and print of the number of maximum vfs per port that driver can handle. Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-06 21:20:19 +07:00
if (total_vfs > fw_enabled_sriov_vfs) {
mlx4_err(dev, "requested vfs (%d) > available vfs (%d). Continuing without SR_IOV\n",
total_vfs, fw_enabled_sriov_vfs);
err = -ENOMEM;
goto disable_sriov;
}
net/mlx4_core: Fixed memory leak and incorrect refcount in mlx4_load_one The current mlx4_load_one has a memory leak as it always allocates dev_cap, but frees it only on error. In addition, even if VFs exist when mlx4_load_one is called, we still need to notify probed VFs that we're loading (by incrementing pf_loading). Fixes: a0eacca948d2 ('net/mlx4_core: Refactor mlx4_load_one') Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-14 21:18:04 +07:00
mlx4_warn(dev, "Enabling SR-IOV with %d VFs\n", total_vfs);
err = pci_enable_sriov(pdev, total_vfs);
}
if (err) {
mlx4_err(dev, "Failed to enable SR-IOV, continuing without SR-IOV (err = %d)\n",
err);
goto disable_sriov;
} else {
mlx4_warn(dev, "Running in master mode\n");
dev_flags |= MLX4_FLAG_SRIOV |
MLX4_FLAG_MASTER;
dev_flags &= ~MLX4_FLAG_SLAVE;
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
dev->persist->num_vfs = total_vfs;
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
}
return dev_flags;
disable_sriov:
net/mlx4_core: Fixed memory leak and incorrect refcount in mlx4_load_one The current mlx4_load_one has a memory leak as it always allocates dev_cap, but frees it only on error. In addition, even if VFs exist when mlx4_load_one is called, we still need to notify probed VFs that we're loading (by incrementing pf_loading). Fixes: a0eacca948d2 ('net/mlx4_core: Refactor mlx4_load_one') Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-14 21:18:04 +07:00
atomic_dec(&pf_loading);
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
free_mem:
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
dev->persist->num_vfs = 0;
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
kfree(dev->dev_vfs);
net/mlx4_core: double free of dev_vfs If user loads mlx4_core with num_vfs greater than supported then variable dev->dev_vfs is freed 2 times after unloading the driver. Acked-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-03 04:07:23 +07:00
dev->dev_vfs = NULL;
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
return dev_flags & ~MLX4_FLAG_MASTER;
}
net/mlx4_core: Support more than 64 VFs We now allow up to 126 VFs. Note though that certain firmware versions only allow up to 80 VFs. Moreover, old HCAs only support 64 VFs. In these cases, we limit the maximum number of VFs to 64. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:33 +07:00
enum {
MLX4_DEV_CAP_CHECK_NUM_VFS_ABOVE_64 = -1,
};
static int mlx4_check_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap,
int *nvfs)
{
int requested_vfs = nvfs[0] + nvfs[1] + nvfs[2];
/* Checking for 64 VFs as a limitation of CX2 */
if (!(dev_cap->flags2 & MLX4_DEV_CAP_FLAG2_80_VFS) &&
requested_vfs >= 64) {
mlx4_err(dev, "Requested %d VFs, but FW does not support more than 64\n",
requested_vfs);
return MLX4_DEV_CAP_CHECK_NUM_VFS_ABOVE_64;
}
return 0;
}
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
static int mlx4_load_one(struct pci_dev *pdev, int pci_dev_data,
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
int total_vfs, int *nvfs, struct mlx4_priv *priv,
int reset_flow)
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
{
struct mlx4_dev *dev;
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
unsigned sum = 0;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
int err;
mlx4_core: Ethernet MAC/VLAN management Add support for managing MAC and VLAN filters for each port. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Oren Duer <oren@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 01:44:46 +07:00
int port;
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
int i;
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
struct mlx4_dev_cap *dev_cap = NULL;
net/mlx4_core: Don't disable SRIOV if there are active VFs When unloading the host driver while there are VFs active on VMs, the PF driver disabled sriov anyway, causing kernel crashes. We now leave SRIOV enabled, to avoid that. When the driver is reloaded, __mlx4_init_one is invoked on the PF. It now checks to see if SRIOV is already enabled on the PF -- and if so does not enable sriov again. Signed-off-by: Tal Alon <talal@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:47 +07:00
int existing_vfs = 0;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
dev = &priv->dev;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_core: Initialize ctx_list and ctx_lock earlier We may call mlx4_dispatch_event() before mlx4_register_device() is called for a device, because for example a catastrophic error happens immediately after we enable interrupts. Therefore priv->ctx_list and priv->ctx_lock need to be initialized earlier. This bug was actually exposed by the MSI-X bug that returned IRQ numbers to drivers in reverse order, so that the first FW command interrupt looked to mlx4 like a catastrophic error. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-06-08 01:51:58 +07:00
INIT_LIST_HEAD(&priv->ctx_list);
spin_lock_init(&priv->ctx_lock);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
mutex_init(&priv->port_mutex);
net/mlx4_core: Port aggregation upper layer interface Supply interface functions to bond and unbond ports of a mlx4 internal interfaces. Example for such an interface is the one registered by the mlx4 IB driver under RoCE. There are 1. Functions to go in/out to/from bonded mode 2. Function to remap virtual ports to physical ports The bond_mutex prevents simultaneous access to data that keep status of the device in bonded mode. The upper mlx4 interface marks to the mlx4 core module that they want to be subject for such bonding by setting the MLX4_INTFF_BONDING flag. Interface which goes to/from bonded mode is re-created. The mlx4 Ethernet driver does not set this flag when registering the interface, the IB driver does. Signed-off-by: Moni Shoua <monis@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-03 21:48:33 +07:00
mutex_init(&priv->bond_mutex);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
mlx4_core: Move kernel doorbell management into core In addition to mlx4_ib, there will be ethernet and FC consumers of mlx4_core, so move the code for managing kernel doorbells into the core module to avoid having to duplicate this multiple times. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-04-24 01:55:45 +07:00
INIT_LIST_HEAD(&priv->pgdir_list);
mutex_init(&priv->pgdir_mutex);
mlx4: Add blue flame support for kernel consumers Using blue flame can improve latency by allowing the HW to more efficiently access the WQE. This patch presents two functions that are used to allocate or release HW resources for using blue flame; the caller need to supply a struct mlx4_bf object when allocating resources. Consumers that make use of this API should post doorbells to the UAR object pointed by the initialized struct mlx4_bf; Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:38:41 +07:00
INIT_LIST_HEAD(&priv->bf_list);
mutex_init(&priv->bf_mutex);
mlx4: use pci_dev->revision Commit 725c89997e03d71b09ea3c17c997da0712b9d835 (mlx4_en: Reporting HW revision in ethtool -i) added code to read the revision ID from the PCI configuration register while it's already stored by PCI subsystem in the 'revision' field of 'struct pci_dev'... While at it, move the code being changed a bit in order to not break the initialization sequence. Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-06-23 11:44:30 +07:00
dev->rev_id = pdev->revision;
net/mlx4_core: ICM pages are allocated on device NUMA node This is done to optimize FW/HW access to host memory. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.com> Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Reviewed-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-07 17:19:53 +07:00
dev->numa_node = dev_to_node(&pdev->dev);
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* Detect if this device is a virtual function */
mlx4_core: Stash PCI ID driver_data in mlx4_priv structure That way we can check flags later on, when we've finished with the pci_device_id structure. Also convert the "is VF" flag to an enum: "Never do in the preprocessor what can be done in C." Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:23:41 +07:00
if (pci_dev_data & MLX4_PCI_DEV_IS_VF) {
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
mlx4_warn(dev, "Detected virtual function - running in slave mode\n");
dev->flags |= MLX4_FLAG_SLAVE;
} else {
/* We reset the device and enable SRIOV only for physical
* devices. Try to claim ownership on the device;
* if already taken, skip -- do not allow multiple PFs */
err = mlx4_get_ownership(dev);
if (err) {
if (err < 0)
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
return err;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
else {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_warn(dev, "Multiple PFs not yet supported - Skipping PF\n");
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
return -EINVAL;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
}
mlx4: use pci_dev->revision Commit 725c89997e03d71b09ea3c17c997da0712b9d835 (mlx4_en: Reporting HW revision in ethtool -i) added code to read the revision ID from the PCI configuration register while it's already stored by PCI subsystem in the 'revision' field of 'struct pci_dev'... While at it, move the code being changed a bit in order to not break the initialization sequence. Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-06-23 11:44:30 +07:00
net/mlx4_core: Respond to operation request by firmware This commit adds new firmware command and new firmware event. The firmware raises the MLX4_EVENT_TYPE_OP_REQUIRED event in order to signal the driver it needs to perform an administrative operation throughout the MLX4_CMD_GET_OP_REQ command. At the moment the supported operation is adding/removing multicast entries which are used by the firmware for handling NCSI traffic in B0 steering mode. Also, had to swap the order of mlx4_init_mcg_table() and mlx4_init_eq_table() to make sure that driver will get events only after resources are initialized to handle it. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.com> Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-07-28 22:54:21 +07:00
atomic_set(&priv->opreq_count, 0);
INIT_WORK(&priv->opreq_task, mlx4_opreq_action);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/*
* Now reset the HCA before we touch the PCI capabilities or
* attempt a firmware command, since a boot ROM may have left
* the HCA in an undefined state.
*/
err = mlx4_reset(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to reset HCA, aborting\n");
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
goto err_sriov;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
if (total_vfs) {
dev->flags = MLX4_FLAG_MASTER;
net/mlx4_core: Fixed memory leak and incorrect refcount in mlx4_load_one The current mlx4_load_one has a memory leak as it always allocates dev_cap, but frees it only on error. In addition, even if VFs exist when mlx4_load_one is called, we still need to notify probed VFs that we're loading (by incrementing pf_loading). Fixes: a0eacca948d2 ('net/mlx4_core: Refactor mlx4_load_one') Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-14 21:18:04 +07:00
existing_vfs = pci_num_vf(pdev);
if (existing_vfs)
dev->flags |= MLX4_FLAG_SRIOV;
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
dev->persist->num_vfs = total_vfs;
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
}
net/mlx4_core: Enhance the catas flow to support device reset This includes: - resetting the chip when a fatal error is detected (the current code does not do this). - exposing the ability to enter error state from outside the catas code by calling its functionality. (E.g. FW Command timeout, AER error). - managing a persistent device state. This is needed to sync between reset flow cases. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:38 +07:00
/* on load remove any previous indication of internal error,
* device is up.
*/
dev->persist->state = MLX4_DEVICE_STATE_UP;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
slave_start:
net/mlx4_core: Return the error value in case of command initialization failure If mlx4_cmd_init() failed, the init_one function returned success, although no resources were opened. Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-06 05:50:52 +07:00
err = mlx4_cmd_init(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to init command interface, aborting\n");
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
goto err_sriov;
}
/* In slave functions, the communication channel must be initialized
* before posting commands. Also, init num_slaves before calling
* mlx4_init_hca */
if (mlx4_is_mfunc(dev)) {
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
if (mlx4_is_master(dev)) {
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
dev->num_slaves = MLX4_MAX_NUM_SLAVES;
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
} else {
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
dev->num_slaves = 0;
net/mlx4_core: Return proper error code when __mlx4_add_one fails Returning 0 (success) when in fact we are aborting the load, leads to kernel panic when unloading the module. Fix that by returning the actual error code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-01-24 08:54:17 +07:00
err = mlx4_multi_func_init(dev);
if (err) {
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "Failed to init slave mfunc interface, aborting\n");
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
goto err_cmd;
}
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
}
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
err = mlx4_init_fw(dev);
if (err) {
mlx4_err(dev, "Failed to init fw, aborting.\n");
goto err_mfunc;
}
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
if (mlx4_is_master(dev)) {
net/mlx4_core: Fixed memory leak and incorrect refcount in mlx4_load_one The current mlx4_load_one has a memory leak as it always allocates dev_cap, but frees it only on error. In addition, even if VFs exist when mlx4_load_one is called, we still need to notify probed VFs that we're loading (by incrementing pf_loading). Fixes: a0eacca948d2 ('net/mlx4_core: Refactor mlx4_load_one') Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-14 21:18:04 +07:00
/* when we hit the goto slave_start below, dev_cap already initialized */
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
if (!dev_cap) {
dev_cap = kzalloc(sizeof(*dev_cap), GFP_KERNEL);
if (!dev_cap) {
err = -ENOMEM;
goto err_fw;
}
err = mlx4_QUERY_DEV_CAP(dev, dev_cap);
if (err) {
mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting.\n");
goto err_fw;
}
net/mlx4_core: Support more than 64 VFs We now allow up to 126 VFs. Note though that certain firmware versions only allow up to 80 VFs. Moreover, old HCAs only support 64 VFs. In these cases, we limit the maximum number of VFs to 64. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:33 +07:00
if (mlx4_check_dev_cap(dev, dev_cap, nvfs))
goto err_fw;
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
if (!(dev_cap->flags2 & MLX4_DEV_CAP_FLAG2_SYS_EQS)) {
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
u64 dev_flags = mlx4_enable_sriov(dev, pdev,
total_vfs,
existing_vfs,
reset_flow);
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
net/mlx4_core: need to call close fw if alloc icm is called twice If mlx4_enable_sriov is called by adapter without this feature MLX4_DEV_CAP_FLAG2_SYS_EQS then during this path the function alloc icm is called twice without freeing the structures from the first time. Acked-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-03 04:07:24 +07:00
mlx4_close_fw(dev);
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
mlx4_cmd_cleanup(dev, MLX4_CMD_CLEANUP_ALL);
dev->flags = dev_flags;
if (!SRIOV_VALID_STATE(dev->flags)) {
mlx4_err(dev, "Invalid SRIOV state\n");
goto err_sriov;
}
err = mlx4_reset(dev);
if (err) {
mlx4_err(dev, "Failed to reset HCA, aborting.\n");
goto err_sriov;
}
goto slave_start;
}
} else {
/* Legacy mode FW requires SRIOV to be enabled before
* doing QUERY_DEV_CAP, since max_eq's value is different if
* SRIOV is enabled.
*/
memset(dev_cap, 0, sizeof(*dev_cap));
err = mlx4_QUERY_DEV_CAP(dev, dev_cap);
if (err) {
mlx4_err(dev, "QUERY_DEV_CAP command failed, aborting.\n");
goto err_fw;
}
net/mlx4_core: Support more than 64 VFs We now allow up to 126 VFs. Note though that certain firmware versions only allow up to 80 VFs. Moreover, old HCAs only support 64 VFs. In these cases, we limit the maximum number of VFs to 64. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:33 +07:00
if (mlx4_check_dev_cap(dev, dev_cap, nvfs))
goto err_fw;
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
}
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err = mlx4_init_hca(dev);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if (err) {
if (err == -EACCES) {
/* Not primary Physical function
* Running in slave mode */
net/mlx4_core: Refactor mlx4_cmd_init and mlx4_cmd_cleanup Refactoring mlx4_cmd_init and mlx4_cmd_cleanup such that partial init and cleanup are possible. After this refactoring, calling mlx4_cmd_init several times is safe. This is necessary in the VF init flow when mlx4_init_hca returns -EACCESS, we need to issue cleanup and re-attempt to call it with the slave flag. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:29 +07:00
mlx4_cmd_cleanup(dev, MLX4_CMD_CLEANUP_ALL);
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
/* We're not a PF */
if (dev->flags & MLX4_FLAG_SRIOV) {
if (!existing_vfs)
pci_disable_sriov(pdev);
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
if (mlx4_is_master(dev) && !reset_flow)
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
atomic_dec(&pf_loading);
dev->flags &= ~MLX4_FLAG_SRIOV;
}
if (!mlx4_is_slave(dev))
mlx4_free_ownership(dev);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
dev->flags |= MLX4_FLAG_SLAVE;
dev->flags &= ~MLX4_FLAG_MASTER;
goto slave_start;
} else
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
goto err_fw;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
if (mlx4_is_master(dev) && (dev_cap->flags2 & MLX4_DEV_CAP_FLAG2_SYS_EQS)) {
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
u64 dev_flags = mlx4_enable_sriov(dev, pdev, total_vfs,
existing_vfs, reset_flow);
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
if ((dev->flags ^ dev_flags) & (MLX4_FLAG_MASTER | MLX4_FLAG_SLAVE)) {
mlx4_cmd_cleanup(dev, MLX4_CMD_CLEANUP_VHCR);
dev->flags = dev_flags;
err = mlx4_cmd_init(dev);
if (err) {
/* Only VHCR is cleaned up, so could still
* send FW commands
*/
mlx4_err(dev, "Failed to init VHCR command interface, aborting\n");
goto err_close;
}
} else {
dev->flags = dev_flags;
}
if (!SRIOV_VALID_STATE(dev->flags)) {
mlx4_err(dev, "Invalid SRIOV state\n");
goto err_close;
}
}
net/mlx4_core: Warn if device doesn't have enough PCI bandwidth Check if the device get enough bandwidth from the entire PCI chain to satisfy its capabilities. This patch determines the PCIe device's bandwidth capabilities by reading its PCIe Link Capabilities registers and then call the pcie_get_minimum_link function to ensure that the adapter is hooked into a slot which is capable of providing the necessary bandwidth capabilities. Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-05 22:41:08 +07:00
/* check if the device is functioning at its maximum possible speed.
* No return code for this call, just warn the user in case of PCI
* express device capabilities are under-satisfied by the bus.
*/
net/mlx4_core: Don't issue PCIe speed/width checks for VFs Carrying out PCI speed/width checks through pcie_get_minimum_link() on VFs yield wrong results, so remove them. Fixes: b912b2f ('net/mlx4_core: Warn if device doesn't have enough PCI bandwidth') Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-04 21:07:25 +07:00
if (!mlx4_is_slave(dev))
mlx4_check_pcie_caps(dev);
net/mlx4_core: Warn if device doesn't have enough PCI bandwidth Check if the device get enough bandwidth from the entire PCI chain to satisfy its capabilities. This patch determines the PCIe device's bandwidth capabilities by reading its PCIe Link Capabilities registers and then call the pcie_get_minimum_link function to ensure that the adapter is hooked into a slot which is capable of providing the necessary bandwidth capabilities. Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-05 22:41:08 +07:00
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* In master functions, the communication channel must be initialized
* after obtaining its address from fw */
if (mlx4_is_master(dev)) {
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
if (dev->caps.num_ports < 2 &&
num_vfs_argc > 1) {
err = -EINVAL;
mlx4_err(dev,
"Error: Trying to configure VFs on port 2, but HCA has only %d physical ports\n",
dev->caps.num_ports);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
goto err_close;
}
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
memcpy(dev->persist->nvfs, nvfs, sizeof(dev->persist->nvfs));
net/mlx4: Adapt num_vfs/probed_vf params for single port VF A new syntax is added for the module parameters num_vfs and probe_vf. num_vfs=p1,p2,p1+p2 probe_bf=p1,p2,p1+p2 Where p1(2) is the number of VFs on / probed VFs for physical port1(2) and p1+p2 is the number of dual port VFs. Single port VFs are currently supported only when the link type for both ports of the device is Ethernet. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-19 23:11:53 +07:00
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
for (i = 0;
i < sizeof(dev->persist->nvfs)/
sizeof(dev->persist->nvfs[0]); i++) {
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
unsigned j;
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
for (j = 0; j < dev->persist->nvfs[i]; ++sum, ++j) {
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
dev->dev_vfs[sum].min_port = i < 2 ? i + 1 : 1;
dev->dev_vfs[sum].n_ports = i < 2 ? 1 :
dev->caps.num_ports;
net/mlx4: Add data structures to support N-Ports per VF Adds the required data structures to support VFs with N (1 or 2) ports instead of always using the number of physical ports. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-19 23:11:50 +07:00
}
}
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
/* In master functions, the communication channel
* must be initialized after obtaining its address from fw
*/
err = mlx4_multi_func_init(dev);
if (err) {
mlx4_err(dev, "Failed to init master mfunc interface, aborting.\n");
goto err_close;
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_core: Add support for multiple completion event vectors When using MSI-X mode, create a completion event queue for each CPU. Report the number of completion EQs in a new struct mlx4_caps member, num_comp_vectors, and extend the mlx4_cq_alloc() interface with a vector parameter so that consumers can specify which completion EQ should be used to report events for the CQ being created. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-12-22 22:15:03 +07:00
err = mlx4_alloc_eq_table(dev);
if (err)
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
goto err_master_mfunc;
mlx4_core: Add support for multiple completion event vectors When using MSI-X mode, create a completion event queue for each CPU. Report the number of completion EQs in a new struct mlx4_caps member, num_comp_vectors, and extend the mlx4_cq_alloc() interface with a vector parameter so that consumers can specify which completion EQ should be used to report events for the CQ being created. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-12-22 22:15:03 +07:00
net/mlx4: Add EQ pool Previously, mlx4_en allocated EQs and used them exclusively. This affected RoCE performance, as applications which are events sensitive were limited to use only the legacy EQs. Change that by introducing an EQ pool. This pool is managed by mlx4_core. EQs are assigned to ports (when there are limited number of EQs, multiple ports could be assigned to the same EQs). An exception to this rule is the ASYNC EQ which handles various events. Legacy EQs are completely removed as all EQs could be shared. When a consumer (mlx4_ib/mlx4_en) requests an EQ, it asks for EQ serving on a specific port. The core driver calculates which EQ should be assigned to that request. Because IRQs are shared between IB and Ethernet modules, their names only include the PCI device BDF address. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Ido Shamay <idos@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-05-31 13:30:16 +07:00
bitmap_zero(priv->msix_ctl.pool_bm, MAX_MSIX);
mlx4: Replacing pool_lock with mutex Under the spinlock we call request_irq(), which allocates memory with GFP_KERNEL, This causes the following trace when DEBUG_SPINLOCK is enabled, it can cause the following trace: BUG: spinlock wrong CPU on CPU#2, ethtool/2595 lock: ffff8801f9cbc2b0, .magic: dead4ead, .owner: ethtool/2595, .owner_cpu: 0 Pid: 2595, comm: ethtool Not tainted 3.0.18 #2 Call Trace: spin_bug+0xa2/0xf0 do_raw_spin_unlock+0x71/0xa0 _raw_spin_unlock+0xe/0x10 mlx4_assign_eq+0x12b/0x190 [mlx4_core] mlx4_en_activate_cq+0x252/0x2d0 [mlx4_en] ? mlx4_en_activate_rx_rings+0x227/0x370 [mlx4_en] mlx4_en_start_port+0x189/0xb90 [mlx4_en] mlx4_en_set_ringparam+0x29a/0x340 [mlx4_en] dev_ethtool+0x816/0xb10 ? dev_get_by_name_rcu+0xa4/0xe0 dev_ioctl+0x2b5/0x470 handle_mm_fault+0x1cd/0x2d0 sock_do_ioctl+0x5d/0x70 sock_ioctl+0x79/0x2f0 do_vfs_ioctl+0x8c/0x340 sys_ioctl+0xa1/0xb0 system_call_fastpath+0x16/0x1b Replacing with mutex, which is enough in this case. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-02-21 10:39:32 +07:00
mutex_init(&priv->msix_ctl.pool_lock);
mlx4: Changing interrupt scheme Adding a pool of MSI-X vectors and EQs that can be used explicitly by mlx4_core customers (mlx4_ib, mlx4_en). The consumers will assign their own names to the interrupt vectors. Those vectors are not opened at mlx4 device initialization, opened by demand. Changed the max number of possible EQs according to the new scheme, no longer relies on on number of cores. The new functionality is exposed through mlx4_assign_eq() and mlx4_release_eq(). Customers that do not use the new API will get completion vectors as before. Signed-off-by: Markuze Alex <markuze@mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:37:47 +07:00
mlx4_core: Enable MSI-X by default Recover from MSI-X errors by automatically falling back on regular interrupt, instead of asking the user to do this manually. This makes it possible to enable MSI-X by default, and will make it possible to get rid of the msi_x module option in the future. Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-07 20:08:28 +07:00
mlx4_enable_msi_x(dev);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if ((mlx4_is_mfunc(dev)) &&
!(dev->flags & MLX4_FLAG_MSI_X)) {
net/mlx4_core: Return proper error code when __mlx4_add_one fails Returning 0 (success) when in fact we are aborting the load, leads to kernel panic when unloading the module. Fix that by returning the actual error code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-01-24 08:54:17 +07:00
err = -ENOSYS;
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_err(dev, "INTx is not supported in multi-function mode, aborting\n");
mlx4: Add support for promiscuous mode in the new steering model. For Ethernet mode only, When we want to register QP as promiscuous, it must be added to all the existing steering entries and also to the default one. The promiscuous QP might also be on of "real" QPs, which means we need to monitor every entry to avoid duplicates and ensure we close an entry when all it has is promiscuous QPs. Same mechanism both for unicast and multicast. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:38:24 +07:00
goto err_free_eq;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
if (!mlx4_is_slave(dev)) {
err = mlx4_init_steering(dev);
if (err)
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
goto err_disable_msix;
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
}
mlx4: Add support for promiscuous mode in the new steering model. For Ethernet mode only, When we want to register QP as promiscuous, it must be added to all the existing steering entries and also to the default one. The promiscuous QP might also be on of "real" QPs, which means we need to monitor every entry to avoid duplicates and ensure we close an entry when all it has is promiscuous QPs. Same mechanism both for unicast and multicast. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:38:24 +07:00
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err = mlx4_setup_hca(dev);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if (err == -EBUSY && (dev->flags & MLX4_FLAG_MSI_X) &&
!mlx4_is_mfunc(dev)) {
mlx4_core: Enable MSI-X by default Recover from MSI-X errors by automatically falling back on regular interrupt, instead of asking the user to do this manually. This makes it possible to enable MSI-X by default, and will make it possible to get rid of the msi_x module option in the future. Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-07 20:08:28 +07:00
dev->flags &= ~MLX4_FLAG_MSI_X;
net/mlx4: Use single completion vector after NOP failure Fix a crash at the error flow of NOP command which caused the driver to try and use a completion vector which wasn't allocated. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-25 07:24:12 +07:00
dev->caps.num_comp_vectors = 1;
mlx4_core: Enable MSI-X by default Recover from MSI-X errors by automatically falling back on regular interrupt, instead of asking the user to do this manually. This makes it possible to enable MSI-X by default, and will make it possible to get rid of the msi_x module option in the future. Signed-off-by: Michael S. Tsirkin <mst@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-08-07 20:08:28 +07:00
pci_disable_msix(pdev);
err = mlx4_setup_hca(dev);
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
if (err)
mlx4: Add support for promiscuous mode in the new steering model. For Ethernet mode only, When we want to register QP as promiscuous, it must be added to all the existing steering entries and also to the default one. The promiscuous QP might also be on of "real" QPs, which means we need to monitor every entry to avoid duplicates and ensure we close an entry when all it has is promiscuous QPs. Same mechanism both for unicast and multicast. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:38:24 +07:00
goto err_steer;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4: Structures and init/teardown for VF resource quotas This is step #1 for implementing SRIOV resource quotas for VFs. Quotas are implemented per resource type for VFs and the PF, to prevent any entity from simply grabbing all the resources for itself and leaving the other entities unable to obtain such resources. Resources which are allocated using quotas: QPs, CQs, SRQs, MPTs, MTTs, MAC, VLAN, and Counters. The quota system works as follows: Each entity (VF or PF) is given a max number of a given resource (its quota), and a guaranteed minimum number for each resource (starvation prevention). For QPs, CQs, SRQs, MPTs and MTTs: 50% of the available quantity for the resource is divided equally among the PF and all the active VFs (i.e., the number of VFs in the mlx4_core module parameter "num_vfs"). This 50% represents the "guaranteed minimum" pool. The other 50% is the "free pool", allocated on a first-come-first-serve basis. For each VF/PF, resources are first allocated from its "guaranteed-minimum" pool. When that pool is exhausted, the driver attempts to allocate from the resource "free-pool". The quota (i.e., max) for the VFs and the PF is: The free-pool amount (50% of the real max) + the guaranteed minimum For MACs: Guarantee 2 MACs per VF/PF per port. As a result, since we have only 128 MACs per port, reduce the allowable number of VFs from 64 to 63. Any remaining MACs are put into a free pool. For VLANs: For the PF, the per-port quota is 128 and guarantee is 64 (to allow the PF to register at least a VLAN per VF in VST mode). For the VFs, the per-port quota is 64 and the guarantee is 0. We assume that VGT VFs are trusted not to abuse the VLAN resource. For Counters: For all functions (PF and VFs), the quota is 128 and the guarantee is 0. In this patch, we define the needed structures, which are added to the resource-tracker struct. In addition, we do initialization for the resource quota, and adjust the query_device response to use quotas rather than resource maxima. As part of the implementation, we introduce a new field in mlx4_dev: quotas. This field holds the resource quotas used to report maxima to the upper layers (ib_core, via query_device). The HCA maxima of these values are passed to the VFs (via QUERY_HCA) so that they may continue to use these in handling QPs, CQs, SRQs and MPTs. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-03 15:03:23 +07:00
mlx4_init_quotas(dev);
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
/* When PF resources are ready arm its comm channel to enable
* getting commands
*/
if (mlx4_is_master(dev)) {
err = mlx4_ARM_COMM_CHANNEL(dev);
if (err) {
mlx4_err(dev, " Failed to arm comm channel eq: %x\n",
err);
goto err_steer;
}
}
mlx4: Structures and init/teardown for VF resource quotas This is step #1 for implementing SRIOV resource quotas for VFs. Quotas are implemented per resource type for VFs and the PF, to prevent any entity from simply grabbing all the resources for itself and leaving the other entities unable to obtain such resources. Resources which are allocated using quotas: QPs, CQs, SRQs, MPTs, MTTs, MAC, VLAN, and Counters. The quota system works as follows: Each entity (VF or PF) is given a max number of a given resource (its quota), and a guaranteed minimum number for each resource (starvation prevention). For QPs, CQs, SRQs, MPTs and MTTs: 50% of the available quantity for the resource is divided equally among the PF and all the active VFs (i.e., the number of VFs in the mlx4_core module parameter "num_vfs"). This 50% represents the "guaranteed minimum" pool. The other 50% is the "free pool", allocated on a first-come-first-serve basis. For each VF/PF, resources are first allocated from its "guaranteed-minimum" pool. When that pool is exhausted, the driver attempts to allocate from the resource "free-pool". The quota (i.e., max) for the VFs and the PF is: The free-pool amount (50% of the real max) + the guaranteed minimum For MACs: Guarantee 2 MACs per VF/PF per port. As a result, since we have only 128 MACs per port, reduce the allowable number of VFs from 64 to 63. Any remaining MACs are put into a free pool. For VLANs: For the PF, the per-port quota is 128 and guarantee is 64 (to allow the PF to register at least a VLAN per VF in VST mode). For the VFs, the per-port quota is 64 and the guarantee is 0. We assume that VGT VFs are trusted not to abuse the VLAN resource. For Counters: For all functions (PF and VFs), the quota is 128 and the guarantee is 0. In this patch, we define the needed structures, which are added to the resource-tracker struct. In addition, we do initialization for the resource quota, and adjust the query_device response to use quotas rather than resource maxima. As part of the implementation, we introduce a new field in mlx4_dev: quotas. This field holds the resource quotas used to report maxima to the upper layers (ib_core, via query_device). The HCA maxima of these values are passed to the VFs (via QUERY_HCA) so that they may continue to use these in handling QPs, CQs, SRQs and MPTs. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-03 15:03:23 +07:00
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
for (port = 1; port <= dev->caps.num_ports; port++) {
err = mlx4_init_port_info(dev, port);
if (err)
goto err_port;
}
mlx4_core: Ethernet MAC/VLAN management Add support for managing MAC and VLAN filters for each port. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Oren Duer <oren@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 01:44:46 +07:00
net/mlx4_core: Port aggregation upper layer interface Supply interface functions to bond and unbond ports of a mlx4 internal interfaces. Example for such an interface is the one registered by the mlx4 IB driver under RoCE. There are 1. Functions to go in/out to/from bonded mode 2. Function to remap virtual ports to physical ports The bond_mutex prevents simultaneous access to data that keep status of the device in bonded mode. The upper mlx4 interface marks to the mlx4 core module that they want to be subject for such bonding by setting the MLX4_INTFF_BONDING flag. Interface which goes to/from bonded mode is re-created. The mlx4 Ethernet driver does not set this flag when registering the interface, the IB driver does. Signed-off-by: Moni Shoua <monis@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-02-03 21:48:33 +07:00
priv->v2p.port1 = 1;
priv->v2p.port2 = 2;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err = mlx4_register_device(dev);
if (err)
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
goto err_port;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
net/mlx4_core: Load higher level modules according to ports type Mellanox ConnectX architecture is: mlx4_core is the lower level PCI driver which register on the PCI id, and protocol specific drivers are depended on it: mlx4_en - for Ethernet and mlx4_ib for Infiniband. NIC could have multiple ports which can change their type dynamically. We use the request_module() call to load the relevant protocol driver when needed: on loading time or at port type change event. Signed-off-by: Eyal Perry <eyalpe@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-10-15 21:55:24 +07:00
mlx4_request_modules(dev);
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
mlx4_sense_init(dev);
mlx4_start_sense(dev);
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
priv->removed = 0;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
if (mlx4_is_master(dev) && dev->persist->num_vfs && !reset_flow)
net/mlx4_core: Defer VF initialization till PF is fully initialized Fix in commit [1] is not sufficient since a deferred VF initialization could happen after pci_enable_sriov() is finished, but before the PF is fully initialized. Need to prevent VFs from initializing till the PF is fully ready and comm channel is operational. [1] - 9798935 "net/mlx4_core: mlx4_init_slave() shouldn't access comm channel before PF is ready" CC: Stuart Hayes <Stuart_Hayes@Dell.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 15:17:22 +07:00
atomic_dec(&pf_loading);
net/mlx4_core: Fixed memory leak and incorrect refcount in mlx4_load_one The current mlx4_load_one has a memory leak as it always allocates dev_cap, but frees it only on error. In addition, even if VFs exist when mlx4_load_one is called, we still need to notify probed VFs that we're loading (by incrementing pf_loading). Fixes: a0eacca948d2 ('net/mlx4_core: Refactor mlx4_load_one') Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-14 21:18:04 +07:00
kfree(dev_cap);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return 0;
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
err_port:
mlx4_core: Fix cleanup in __mlx4_init_one() error path If mlx4_init_port_info() fails, cleanup the initialized ports only. Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-01-07 03:54:39 +07:00
for (--port; port >= 1; --port)
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
mlx4_cleanup_port_info(&priv->port[port]);
net/mlx4_core: Allocate default counter per port Default counter per port will be allocated at the mlx4 core driver load. Every QP opened by the Ethernet driver will be attached to the port's default counter. This is an infrastructure step to collect VF statistics from the PF. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:59:02 +07:00
mlx4_cleanup_default_counters(dev);
net/mlx4_core: Remove counters table allocation from VF flow Since virtual functions get their counters indices allocation from the PF, allocate counters indices bitmap only in case the function isn't virtual. Also, check that the device has counters to allocate before creating the indices bitmap table. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:58:59 +07:00
if (!mlx4_is_slave(dev))
mlx4_cleanup_counters_table(dev);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_cleanup_qp_table(dev);
mlx4_cleanup_srq_table(dev);
mlx4_cleanup_cq_table(dev);
mlx4_cmd_use_polling(dev);
mlx4_cleanup_eq_table(dev);
net/mlx4_core: Respond to operation request by firmware This commit adds new firmware command and new firmware event. The firmware raises the MLX4_EVENT_TYPE_OP_REQUIRED event in order to signal the driver it needs to perform an administrative operation throughout the MLX4_CMD_GET_OP_REQ command. At the moment the supported operation is adding/removing multicast entries which are used by the firmware for handling NCSI traffic in B0 steering mode. Also, had to swap the order of mlx4_init_mcg_table() and mlx4_init_eq_table() to make sure that driver will get events only after resources are initialized to handle it. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.com> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.com> Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-07-28 22:54:21 +07:00
mlx4_cleanup_mcg_table(dev);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_cleanup_mr_table(dev);
IB/mlx4: Add support for XRC domains Support creating and destroying XRC domains. Any sharing of the XRCD is managed above the low-level driver. Signed-off-by: Sean Hefty <sean.hefty@intel.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-06-02 23:01:33 +07:00
mlx4_cleanup_xrcd_table(dev);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_cleanup_pd_table(dev);
mlx4_cleanup_uar_table(dev);
mlx4: Add support for promiscuous mode in the new steering model. For Ethernet mode only, When we want to register QP as promiscuous, it must be added to all the existing steering entries and also to the default one. The promiscuous QP might also be on of "real" QPs, which means we need to monitor every entry to avoid duplicates and ensure we close an entry when all it has is promiscuous QPs. Same mechanism both for unicast and multicast. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:38:24 +07:00
err_steer:
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
if (!mlx4_is_slave(dev))
mlx4_clear_steering(dev);
mlx4: Add support for promiscuous mode in the new steering model. For Ethernet mode only, When we want to register QP as promiscuous, it must be added to all the existing steering entries and also to the default one. The promiscuous QP might also be on of "real" QPs, which means we need to monitor every entry to avoid duplicates and ensure we close an entry when all it has is promiscuous QPs. Same mechanism both for unicast and multicast. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-03-23 05:38:24 +07:00
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
err_disable_msix:
if (dev->flags & MLX4_FLAG_MSI_X)
pci_disable_msix(pdev);
mlx4_core: Add support for multiple completion event vectors When using MSI-X mode, create a completion event queue for each CPU. Report the number of completion EQs in a new struct mlx4_caps member, num_comp_vectors, and extend the mlx4_cq_alloc() interface with a vector parameter so that consumers can specify which completion EQ should be used to report events for the CQ being created. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-12-22 22:15:03 +07:00
err_free_eq:
mlx4_free_eq_table(dev);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
err_master_mfunc:
net/mlx4_core: Fix mem leak in SRIOV mlx4_init_one error flow Structs allocated for the resource tracker must be freed in the error flow. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-27 20:58:01 +07:00
if (mlx4_is_master(dev)) {
mlx4_free_resource_tracker(dev, RES_TR_FREE_STRUCTS_ONLY);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
mlx4_multi_func_cleanup(dev);
net/mlx4_core: Fix mem leak in SRIOV mlx4_init_one error flow Structs allocated for the resource tracker must be freed in the error flow. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-27 20:58:01 +07:00
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
mlx4_core: Fix memory leaks in SR-IOV error paths Fix a few memory leaks that happen if errors happen in SR-IOV mode. Signed-off-by: Dotan Barak <dotanb@dev.mellanox.co.il> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-05-29 20:30:59 +07:00
if (mlx4_is_slave(dev)) {
mlx4: Add infrastructure for selecting VFs to enable QP0 via MLX proxy QPs This commit adds the infrastructure for enabling selected VFs to operate SMI (QP0) MADs without restriction. Additionally, for these enabled VFs, their QP0 proxy and tunnel QPs are MLX QPs. As such, they operate over VL15. Therefore, they are not affected by "credit" problems or changes in the VLArb table (which may shut down VL0). Non-enabled VFs may only create UD proxy QP0 qps (which are forced by the hypervisor to send packets using the q-key it assigns and places in the qp-context). Thus, non-enabled VFs will not pose a security risk. The hypervisor discards any privileged MADs it receives from these non-enabled VFs. By default, all VFs are NOT enabled, and must explicitly be enabled by the administrator. The sysfs interface which operates the VF enablement infrastructure is provided in the next commit. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-05-29 20:31:03 +07:00
kfree(dev->caps.qp0_qkey);
mlx4_core: Fix memory leaks in SR-IOV error paths Fix a few memory leaks that happen if errors happen in SR-IOV mode. Signed-off-by: Dotan Barak <dotanb@dev.mellanox.co.il> Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-05-29 20:30:59 +07:00
kfree(dev->caps.qp0_tunnel);
kfree(dev->caps.qp0_proxy);
kfree(dev->caps.qp1_tunnel);
kfree(dev->caps.qp1_proxy);
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err_close:
mlx4_close_hca(dev);
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
err_fw:
mlx4_close_fw(dev);
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
err_mfunc:
if (mlx4_is_slave(dev))
mlx4_multi_func_cleanup(dev);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err_cmd:
net/mlx4_core: Refactor mlx4_cmd_init and mlx4_cmd_cleanup Refactoring mlx4_cmd_init and mlx4_cmd_cleanup such that partial init and cleanup are possible. After this refactoring, calling mlx4_cmd_init several times is safe. This is necessary in the VF init flow when mlx4_init_hca returns -EACCESS, we need to issue cleanup and re-attempt to call it with the slave flag. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:29 +07:00
mlx4_cmd_cleanup(dev, MLX4_CMD_CLEANUP_ALL);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
err_sriov:
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
if (dev->flags & MLX4_FLAG_SRIOV && !existing_vfs) {
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
pci_disable_sriov(pdev);
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
dev->flags &= ~MLX4_FLAG_SRIOV;
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
if (mlx4_is_master(dev) && dev->persist->num_vfs && !reset_flow)
net/mlx4_core: Defer VF initialization till PF is fully initialized Fix in commit [1] is not sufficient since a deferred VF initialization could happen after pci_enable_sriov() is finished, but before the PF is fully initialized. Need to prevent VFs from initializing till the PF is fully ready and comm channel is operational. [1] - 9798935 "net/mlx4_core: mlx4_init_slave() shouldn't access comm channel before PF is ready" CC: Stuart Hayes <Stuart_Hayes@Dell.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 15:17:22 +07:00
atomic_dec(&pf_loading);
net/mlx4: Add data structures to support N-Ports per VF Adds the required data structures to support VFs with N (1 or 2) ports instead of always using the number of physical ports. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-19 23:11:50 +07:00
kfree(priv->dev.dev_vfs);
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
if (!mlx4_is_slave(dev))
mlx4_free_ownership(dev);
net/mlx4_core: Flexible (asymmetric) allocation of EQs and MSI-X vectors for PF/VFs Previously, the driver queried the firmware in order to get the number of supported EQs. Under SRIOV, since this was done before the driver notified the firmware how many VFs it actually needs, the firmware had to take into account a worst case scenario and always allocated four EQs per VF, where one was used for events while the others were used for completions. Now, when the firmware supports the asymmetric allocation scheme, denoted by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we can get more EQs and MSI-X vectors per function. Moreover, when running in the new firmware/driver mode, the limitation that the number of EQs should be a power of two is lifted. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:32 +07:00
kfree(dev_cap);
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
return err;
}
static int __mlx4_init_one(struct pci_dev *pdev, int pci_dev_data,
struct mlx4_priv *priv)
{
int err;
int nvfs[MLX4_MAX_PORTS + 1] = {0, 0, 0};
int prb_vf[MLX4_MAX_PORTS + 1] = {0, 0, 0};
const int param_map[MLX4_MAX_PORTS + 1][MLX4_MAX_PORTS + 1] = {
{2, 0, 0}, {0, 1, 2}, {0, 1, 2} };
unsigned total_vfs = 0;
unsigned int i;
pr_info(DRV_NAME ": Initializing %s\n", pci_name(pdev));
err = pci_enable_device(pdev);
if (err) {
dev_err(&pdev->dev, "Cannot enable PCI device, aborting\n");
return err;
}
/* Due to requirement that all VFs and the PF are *guaranteed* 2 MACS
* per port, we must limit the number of VFs to 63 (since their are
* 128 MACs)
*/
for (i = 0; i < sizeof(nvfs)/sizeof(nvfs[0]) && i < num_vfs_argc;
total_vfs += nvfs[param_map[num_vfs_argc - 1][i]], i++) {
nvfs[param_map[num_vfs_argc - 1][i]] = num_vfs[i];
if (nvfs[i] < 0) {
dev_err(&pdev->dev, "num_vfs module parameter cannot be negative\n");
err = -EINVAL;
goto err_disable_pdev;
}
}
for (i = 0; i < sizeof(prb_vf)/sizeof(prb_vf[0]) && i < probe_vfs_argc;
i++) {
prb_vf[param_map[probe_vfs_argc - 1][i]] = probe_vf[i];
if (prb_vf[i] < 0 || prb_vf[i] > nvfs[i]) {
dev_err(&pdev->dev, "probe_vf module parameter cannot be negative or greater than num_vfs\n");
err = -EINVAL;
goto err_disable_pdev;
}
}
net/mlx4_core: Add extra check for total vfs for SRIOV Add extra check for total vfs for SRIOV to check if that value is bigger than total vfs in pci SRIOV capabalities. Fix a check and print of the number of maximum vfs that hw can handle. Fix a check and print of the number of maximum vfs per port that driver can handle. Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-06 21:20:19 +07:00
if (total_vfs > MLX4_MAX_NUM_VF) {
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
dev_err(&pdev->dev,
net/mlx4_core: Add extra check for total vfs for SRIOV Add extra check for total vfs for SRIOV to check if that value is bigger than total vfs in pci SRIOV capabalities. Fix a check and print of the number of maximum vfs that hw can handle. Fix a check and print of the number of maximum vfs per port that driver can handle. Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-06 21:20:19 +07:00
"Requested more VF's (%d) than allowed by hw (%d)\n",
total_vfs, MLX4_MAX_NUM_VF);
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
err = -EINVAL;
goto err_disable_pdev;
}
for (i = 0; i < MLX4_MAX_PORTS; i++) {
net/mlx4_core: Add extra check for total vfs for SRIOV Add extra check for total vfs for SRIOV to check if that value is bigger than total vfs in pci SRIOV capabalities. Fix a check and print of the number of maximum vfs that hw can handle. Fix a check and print of the number of maximum vfs per port that driver can handle. Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-06 21:20:19 +07:00
if (nvfs[i] + nvfs[2] > MLX4_MAX_NUM_VF_P_PORT) {
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
dev_err(&pdev->dev,
net/mlx4_core: Add extra check for total vfs for SRIOV Add extra check for total vfs for SRIOV to check if that value is bigger than total vfs in pci SRIOV capabalities. Fix a check and print of the number of maximum vfs that hw can handle. Fix a check and print of the number of maximum vfs per port that driver can handle. Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-06 21:20:19 +07:00
"Requested more VF's (%d) for port (%d) than allowed by driver (%d)\n",
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
nvfs[i] + nvfs[2], i + 1,
net/mlx4_core: Add extra check for total vfs for SRIOV Add extra check for total vfs for SRIOV to check if that value is bigger than total vfs in pci SRIOV capabalities. Fix a check and print of the number of maximum vfs that hw can handle. Fix a check and print of the number of maximum vfs per port that driver can handle. Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-07-06 21:20:19 +07:00
MLX4_MAX_NUM_VF_P_PORT);
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
err = -EINVAL;
goto err_disable_pdev;
}
}
/* Check for BARs. */
if (!(pci_dev_data & MLX4_PCI_DEV_IS_VF) &&
!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM)) {
dev_err(&pdev->dev, "Missing DCS, aborting (driver_data: 0x%x, pci_resource_flags(pdev, 0):0x%lx)\n",
pci_dev_data, pci_resource_flags(pdev, 0));
err = -ENODEV;
goto err_disable_pdev;
}
if (!(pci_resource_flags(pdev, 2) & IORESOURCE_MEM)) {
dev_err(&pdev->dev, "Missing UAR, aborting\n");
err = -ENODEV;
goto err_disable_pdev;
}
err = pci_request_regions(pdev, DRV_NAME);
if (err) {
dev_err(&pdev->dev, "Couldn't get PCI resources, aborting\n");
goto err_disable_pdev;
}
pci_set_master(pdev);
err = pci_set_dma_mask(pdev, DMA_BIT_MASK(64));
if (err) {
dev_warn(&pdev->dev, "Warning: couldn't set 64-bit PCI DMA mask\n");
err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
if (err) {
dev_err(&pdev->dev, "Can't set PCI DMA mask, aborting\n");
goto err_release_regions;
}
}
err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
if (err) {
dev_warn(&pdev->dev, "Warning: couldn't set 64-bit consistent PCI DMA mask\n");
err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32));
if (err) {
dev_err(&pdev->dev, "Can't set consistent PCI DMA mask, aborting\n");
goto err_release_regions;
}
}
/* Allow large DMA segments, up to the firmware limit of 1 GB */
dma_set_max_seg_size(&pdev->dev, 1024 * 1024 * 1024);
/* Detect if this device is a virtual function */
if (pci_dev_data & MLX4_PCI_DEV_IS_VF) {
/* When acting as pf, we normally skip vfs unless explicitly
* requested to probe them.
*/
if (total_vfs) {
unsigned vfs_offset = 0;
for (i = 0; i < sizeof(nvfs)/sizeof(nvfs[0]) &&
vfs_offset + nvfs[i] < extended_func_num(pdev);
vfs_offset += nvfs[i], i++)
;
if (i == sizeof(nvfs)/sizeof(nvfs[0])) {
err = -ENODEV;
goto err_release_regions;
}
if ((extended_func_num(pdev) - vfs_offset)
> prb_vf[i]) {
dev_warn(&pdev->dev, "Skipping virtual function:%d\n",
extended_func_num(pdev));
err = -ENODEV;
goto err_release_regions;
}
}
}
net/mlx4_core: Refactor the catas flow to work per device Using a WQ per device instead of a single global WQ, this allows independent reset handling per device even when SRIOV is used. This comes as a pre-patch for supporting chip reset for both native and SRIOV. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:37 +07:00
err = mlx4_catas_init(&priv->dev);
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
if (err)
goto err_release_regions;
net/mlx4_core: Refactor the catas flow to work per device Using a WQ per device instead of a single global WQ, this allows independent reset handling per device even when SRIOV is used. This comes as a pre-patch for supporting chip reset for both native and SRIOV. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:37 +07:00
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
err = mlx4_load_one(pdev, pci_dev_data, total_vfs, nvfs, priv, 0);
net/mlx4_core: Refactor the catas flow to work per device Using a WQ per device instead of a single global WQ, this allows independent reset handling per device even when SRIOV is used. This comes as a pre-patch for supporting chip reset for both native and SRIOV. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:37 +07:00
if (err)
goto err_catas;
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
return 0;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
net/mlx4_core: Refactor the catas flow to work per device Using a WQ per device instead of a single global WQ, this allows independent reset handling per device even when SRIOV is used. This comes as a pre-patch for supporting chip reset for both native and SRIOV. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:37 +07:00
err_catas:
mlx4_catas_end(&priv->dev);
mlx4_core: Use pci_request_regions() The old code used two calls to pci_request_region() to get the two BARs for the mlx4 device, for no particularly good reason. Clean up the code a little by converting this to a single call to pci_request_regions(). Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-09-06 10:24:48 +07:00
err_release_regions:
pci_release_regions(pdev);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
err_disable_pdev:
pci_disable_device(pdev);
pci_set_drvdata(pdev, NULL);
return err;
}
drivers/net: fix up function prototypes after __dev* removals The __dev* removal patches for the network drivers ended up messing up the function prototypes for a bunch of drivers. This patch fixes all of them back up to be properly aligned. Bonus is that this almost removes 100 lines of code, always a nice surprise. Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-12-06 21:30:56 +07:00
static int mlx4_init_one(struct pci_dev *pdev, const struct pci_device_id *id)
mlx4_core: Fix section mismatches Commit ee49bd93 ("mlx4_core: Reset device when internal error is detected") introduced some section mismatch problems when CONFIG_HOTPLUG=n, because the error recovery code tears down and reinitializes the device after everything is loaded, which ends up calling into lots of code marked __devinit and __devexit from regular .text. Fix this by getting rid of these now-incorrect section markers. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-10-11 05:43:54 +07:00
{
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
struct mlx4_priv *priv;
struct mlx4_dev *dev;
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
int ret;
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
drivers/net/mlx4: Use %pV, pr_<level>, printk_once Remove near duplication of format string constants by using the newly introduced vsprintf extention %pV to reduce text by 20k or so. $ size drivers/net/mlx4/built-in.o* text data bss dec hex filename 161367 1866 48784 212017 33c31 drivers/net/mlx4/built-in.o 142621 1866 46248 190735 2e90f drivers/net/mlx4/built-in.o.new Use printk_once as appropriate. Convert printks to pr_<level>, some bare printks now use pr_cont. Remove now unused #define PFX. Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-07-10 14:22:46 +07:00
printk_once(KERN_INFO "%s", mlx4_version);
mlx4_core: Fix section mismatches Commit ee49bd93 ("mlx4_core: Reset device when internal error is detected") introduced some section mismatch problems when CONFIG_HOTPLUG=n, because the error recovery code tears down and reinitializes the device after everything is loaded, which ends up calling into lots of code marked __devinit and __devexit from regular .text. Fix this by getting rid of these now-incorrect section markers. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-10-11 05:43:54 +07:00
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
priv = kzalloc(sizeof(*priv), GFP_KERNEL);
if (!priv)
return -ENOMEM;
dev = &priv->dev;
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
dev->persist = kzalloc(sizeof(*dev->persist), GFP_KERNEL);
if (!dev->persist) {
kfree(priv);
return -ENOMEM;
}
dev->persist->pdev = pdev;
dev->persist->dev = dev;
pci_set_drvdata(pdev, dev->persist);
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
priv->pci_dev_data = id->driver_data;
net/mlx4_core: Enhance the catas flow to support device reset This includes: - resetting the chip when a fatal error is detected (the current code does not do this). - exposing the ability to enter error state from outside the catas code by calling its functionality. (E.g. FW Command timeout, AER error). - managing a persistent device state. This is needed to sync between reset flow cases. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:38 +07:00
mutex_init(&dev->persist->device_state_mutex);
net/mlx4_core: Manage interface state for Reset flow cases We need to manage interface state to sync between reset flow and some other relative cases such as remove_one. This has to be done to prevent certain races. For example in case software stack is down as a result of unload call, the remove_one should skip the unload phase. Implement the remove_one case, handling AER and other cases comes next. The interface can be up/down, upon remove_one, the state will include an extra bit indicating that the device is cleaned-up, forcing other tasks to finish before the final cleanup. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:40 +07:00
mutex_init(&dev->persist->interface_state_mutex);
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
ret = __mlx4_init_one(pdev, id->driver_data, priv);
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
if (ret) {
kfree(dev->persist);
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
kfree(priv);
net/mlx4_core: Handle AER flow properly Fix AER callbacks to work properly, it includes: - Refractoring AER to be aligned with Reset flow support. - Sync with concurrent catas flow. In addition, fix the shutdown PCI callback to sync with concurrent catas flow. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:41 +07:00
} else {
pci_save_state(pdev);
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
}
net/mlx4_core: Handle AER flow properly Fix AER callbacks to work properly, it includes: - Refractoring AER to be aligned with Reset flow support. - Sync with concurrent catas flow. In addition, fix the shutdown PCI callback to sync with concurrent catas flow. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:41 +07:00
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
return ret;
mlx4_core: Fix section mismatches Commit ee49bd93 ("mlx4_core: Reset device when internal error is detected") introduced some section mismatch problems when CONFIG_HOTPLUG=n, because the error recovery code tears down and reinitializes the device after everything is loaded, which ends up calling into lots of code marked __devinit and __devexit from regular .text. Fix this by getting rid of these now-incorrect section markers. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-10-11 05:43:54 +07:00
}
net/mlx4_core: Set device configuration data to be persistent across reset When an HCA enters an internal error state, this is detected by the driver. The driver then should reset the HCA and restart the software stack. Keep ports information and some SRIOV configuration in a persistent area to have it valid across reset. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:36 +07:00
static void mlx4_clean_dev(struct mlx4_dev *dev)
{
struct mlx4_dev_persistent *persist = dev->persist;
struct mlx4_priv *priv = mlx4_priv(dev);
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
unsigned long flags = (dev->flags & RESET_PERSIST_MASK_FLAGS);
net/mlx4_core: Set device configuration data to be persistent across reset When an HCA enters an internal error state, this is detected by the driver. The driver then should reset the HCA and restart the software stack. Keep ports information and some SRIOV configuration in a persistent area to have it valid across reset. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:36 +07:00
memset(priv, 0, sizeof(*priv));
priv->dev.persist = persist;
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
priv->dev.flags = flags;
net/mlx4_core: Set device configuration data to be persistent across reset When an HCA enters an internal error state, this is detected by the driver. The driver then should reset the HCA and restart the software stack. Keep ports information and some SRIOV configuration in a persistent area to have it valid across reset. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:36 +07:00
}
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
static void mlx4_unload_one(struct pci_dev *pdev)
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
{
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
struct mlx4_dev_persistent *persist = pci_get_drvdata(pdev);
struct mlx4_dev *dev = persist->dev;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
struct mlx4_priv *priv = mlx4_priv(dev);
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
int pci_dev_data;
net/mlx4_core: Set device configuration data to be persistent across reset When an HCA enters an internal error state, this is detected by the driver. The driver then should reset the HCA and restart the software stack. Keep ports information and some SRIOV configuration in a persistent area to have it valid across reset. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:36 +07:00
int p, i;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
if (priv->removed)
return;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
net/mlx4_core: Set device configuration data to be persistent across reset When an HCA enters an internal error state, this is detected by the driver. The driver then should reset the HCA and restart the software stack. Keep ports information and some SRIOV configuration in a persistent area to have it valid across reset. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:36 +07:00
/* saving current ports type for further use */
for (i = 0; i < dev->caps.num_ports; i++) {
dev->persist->curr_port_type[i] = dev->caps.port_type[i + 1];
dev->persist->curr_port_poss_type[i] = dev->caps.
possible_type[i + 1];
}
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
pci_dev_data = priv->pci_dev_data;
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
mlx4_stop_sense(dev);
mlx4_unregister_device(dev);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
for (p = 1; p <= dev->caps.num_ports; p++) {
mlx4_cleanup_port_info(&priv->port[p]);
mlx4_CLOSE_PORT(dev, p);
}
if (mlx4_is_master(dev))
mlx4_free_resource_tracker(dev,
RES_TR_FREE_SLAVES_ONLY);
net/mlx4_core: Allocate default counter per port Default counter per port will be allocated at the mlx4 core driver load. Every QP opened by the Ethernet driver will be attached to the port's default counter. This is an infrastructure step to collect VF statistics from the PF. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:59:02 +07:00
mlx4_cleanup_default_counters(dev);
net/mlx4_core: Remove counters table allocation from VF flow Since virtual functions get their counters indices allocation from the PF, allocate counters indices bitmap only in case the function isn't virtual. Also, check that the device has counters to allocate before creating the indices bitmap table. Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-06-15 21:58:59 +07:00
if (!mlx4_is_slave(dev))
mlx4_cleanup_counters_table(dev);
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
mlx4_cleanup_qp_table(dev);
mlx4_cleanup_srq_table(dev);
mlx4_cleanup_cq_table(dev);
mlx4_cmd_use_polling(dev);
mlx4_cleanup_eq_table(dev);
mlx4_cleanup_mcg_table(dev);
mlx4_cleanup_mr_table(dev);
mlx4_cleanup_xrcd_table(dev);
mlx4_cleanup_pd_table(dev);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
if (mlx4_is_master(dev))
mlx4_free_resource_tracker(dev,
RES_TR_FREE_STRUCTS_ONLY);
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
iounmap(priv->kar);
mlx4_uar_free(dev, &priv->driver_uar);
mlx4_cleanup_uar_table(dev);
if (!mlx4_is_slave(dev))
mlx4_clear_steering(dev);
mlx4_free_eq_table(dev);
if (mlx4_is_master(dev))
mlx4_multi_func_cleanup(dev);
mlx4_close_hca(dev);
net/mlx4_core: Refactor mlx4_load_one Refactor mlx4_load_one, as a preparation step for a new and more complicated load function. The goal is to support both newer firmware that required init_hca to be done before enable_sriov and legacy firmwares that requires things to be done the other way around. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:30 +07:00
mlx4_close_fw(dev);
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
if (mlx4_is_slave(dev))
mlx4_multi_func_cleanup(dev);
net/mlx4_core: Refactor mlx4_cmd_init and mlx4_cmd_cleanup Refactoring mlx4_cmd_init and mlx4_cmd_cleanup such that partial init and cleanup are possible. After this refactoring, calling mlx4_cmd_init several times is safe. This is necessary in the VF init flow when mlx4_init_hca returns -EACCESS, we need to issue cleanup and re-attempt to call it with the slave flag. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-11-13 19:45:29 +07:00
mlx4_cmd_cleanup(dev, MLX4_CMD_CLEANUP_ALL);
mlx4: Modify proxy/tunnel QP mechanism so that guests do no calculations Previously, the structure of a guest's proxy QPs followed the structure of the PPF special qps (qp0 port 1, qp0 port 2, qp1 port 1, qp1 port 2, ...). The guest then did offset calculations on the sqp_base qp number that the PPF passed to it in QUERY_FUNC_CAP(). This is now changed so that the guest does no offset calculations regarding proxy or tunnel QPs to use. This change frees the PPF from needing to adhere to a specific order in allocating proxy and tunnel QPs. Now QUERY_FUNC_CAP provides each port individually with its proxy qp0, proxy qp1, tunnel qp0, and tunnel qp1 QP numbers, and these are used directly where required (with no offset calculations). To accomplish this change, several fields were added to the phys_caps structure for use by the PPF and by non-SR-IOV mode: base_sqpn -- in non-sriov mode, this was formerly sqp_start. base_proxy_sqpn -- the first physical proxy qp number -- used by PPF base_tunnel_sqpn -- the first physical tunnel qp number -- used by PPF. The current code in the PPF still adheres to the previous layout of sqps, proxy-sqps and tunnel-sqps. However, the PPF can change this layout without affecting VF or (paravirtualized) PF code. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-08-03 15:40:57 +07:00
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
if (dev->flags & MLX4_FLAG_MSI_X)
pci_disable_msix(pdev);
if (!mlx4_is_slave(dev))
mlx4_free_ownership(dev);
mlx4: Add infrastructure for selecting VFs to enable QP0 via MLX proxy QPs This commit adds the infrastructure for enabling selected VFs to operate SMI (QP0) MADs without restriction. Additionally, for these enabled VFs, their QP0 proxy and tunnel QPs are MLX QPs. As such, they operate over VL15. Therefore, they are not affected by "credit" problems or changes in the VLArb table (which may shut down VL0). Non-enabled VFs may only create UD proxy QP0 qps (which are forced by the hypervisor to send packets using the q-key it assigns and places in the qp-context). Thus, non-enabled VFs will not pose a security risk. The hypervisor discards any privileged MADs it receives from these non-enabled VFs. By default, all VFs are NOT enabled, and must explicitly be enabled by the administrator. The sysfs interface which operates the VF enablement infrastructure is provided in the next commit. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-05-29 20:31:03 +07:00
kfree(dev->caps.qp0_qkey);
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
kfree(dev->caps.qp0_tunnel);
kfree(dev->caps.qp0_proxy);
kfree(dev->caps.qp1_tunnel);
kfree(dev->caps.qp1_proxy);
kfree(dev->dev_vfs);
net/mlx4_core: Set device configuration data to be persistent across reset When an HCA enters an internal error state, this is detected by the driver. The driver then should reset the HCA and restart the software stack. Keep ports information and some SRIOV configuration in a persistent area to have it valid across reset. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:36 +07:00
mlx4_clean_dev(dev);
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
priv->pci_dev_data = pci_dev_data;
priv->removed = 1;
}
static void mlx4_remove_one(struct pci_dev *pdev)
{
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
struct mlx4_dev_persistent *persist = pci_get_drvdata(pdev);
struct mlx4_dev *dev = persist->dev;
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
struct mlx4_priv *priv = mlx4_priv(dev);
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
int active_vfs = 0;
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
net/mlx4_core: Manage interface state for Reset flow cases We need to manage interface state to sync between reset flow and some other relative cases such as remove_one. This has to be done to prevent certain races. For example in case software stack is down as a result of unload call, the remove_one should skip the unload phase. Implement the remove_one case, handling AER and other cases comes next. The interface can be up/down, upon remove_one, the state will include an extra bit indicating that the device is cleaned-up, forcing other tasks to finish before the final cleanup. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:40 +07:00
mutex_lock(&persist->interface_state_mutex);
persist->interface_state |= MLX4_INTERFACE_STATE_DELETION;
mutex_unlock(&persist->interface_state_mutex);
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
/* Disabling SR-IOV is not allowed while there are active vf's */
if (mlx4_is_master(dev) && dev->flags & MLX4_FLAG_SRIOV) {
active_vfs = mlx4_how_many_lives_vf(dev);
if (active_vfs) {
pr_warn("Removing PF when there are active VF's !!\n");
pr_warn("Will not disable SR-IOV.\n");
}
}
net/mlx4_core: Manage interface state for Reset flow cases We need to manage interface state to sync between reset flow and some other relative cases such as remove_one. This has to be done to prevent certain races. For example in case software stack is down as a result of unload call, the remove_one should skip the unload phase. Implement the remove_one case, handling AER and other cases comes next. The interface can be up/down, upon remove_one, the state will include an extra bit indicating that the device is cleaned-up, forcing other tasks to finish before the final cleanup. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:40 +07:00
/* device marked to be under deletion running now without the lock
* letting other tasks to be terminated
*/
if (persist->interface_state & MLX4_INTERFACE_STATE_UP)
mlx4_unload_one(pdev);
else
mlx4_info(dev, "%s: interface is down\n", __func__);
net/mlx4_core: Refactor the catas flow to work per device Using a WQ per device instead of a single global WQ, this allows independent reset handling per device even when SRIOV is used. This comes as a pre-patch for supporting chip reset for both native and SRIOV. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:37 +07:00
mlx4_catas_end(dev);
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
if (dev->flags & MLX4_FLAG_SRIOV && !active_vfs) {
mlx4_warn(dev, "Disabling SR-IOV\n");
pci_disable_sriov(pdev);
}
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
pci_release_regions(pdev);
pci_disable_device(pdev);
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
kfree(dev->persist);
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
kfree(priv);
pci_set_drvdata(pdev, NULL);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
}
net/mlx4_core: Set device configuration data to be persistent across reset When an HCA enters an internal error state, this is detected by the driver. The driver then should reset the HCA and restart the software stack. Keep ports information and some SRIOV configuration in a persistent area to have it valid across reset. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:36 +07:00
static int restore_current_port_types(struct mlx4_dev *dev,
enum mlx4_port_type *types,
enum mlx4_port_type *poss_types)
{
struct mlx4_priv *priv = mlx4_priv(dev);
int err, i;
mlx4_stop_sense(dev);
mutex_lock(&priv->port_mutex);
for (i = 0; i < dev->caps.num_ports; i++)
dev->caps.possible_type[i + 1] = poss_types[i];
err = mlx4_change_port_types(dev, types);
mlx4_start_sense(dev);
mutex_unlock(&priv->port_mutex);
return err;
}
mlx4_core: Reset device when internal error is detected Reset the device when an internal error is detected. Also, detect errors by polling the error buffer rather than using interrupts. This is more robust and doesn't depend on MSI-X. Remove the old interrupt handler entirely, since we don't want to support two mechanisms for detecting internal errors. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-07-12 21:50:45 +07:00
int mlx4_restart_one(struct pci_dev *pdev)
{
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
struct mlx4_dev_persistent *persist = pci_get_drvdata(pdev);
struct mlx4_dev *dev = persist->dev;
mlx4_core: Stash PCI ID driver_data in mlx4_priv structure That way we can check flags later on, when we've finished with the pci_device_id structure. Also convert the "is VF" flag to an enum: "Never do in the preprocessor what can be done in C." Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:23:41 +07:00
struct mlx4_priv *priv = mlx4_priv(dev);
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
int nvfs[MLX4_MAX_PORTS + 1] = {0, 0, 0};
int pci_dev_data, err, total_vfs;
mlx4_core: Stash PCI ID driver_data in mlx4_priv structure That way we can check flags later on, when we've finished with the pci_device_id structure. Also convert the "is VF" flag to an enum: "Never do in the preprocessor what can be done in C." Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:23:41 +07:00
pci_dev_data = priv->pci_dev_data;
net/mlx4_core: Maintain a persistent memory for mlx4 device Maintain a persistent memory that should survive reset flow/PCI error. This comes as a preparation for coming series to support above flows. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:35 +07:00
total_vfs = dev->persist->num_vfs;
memcpy(nvfs, dev->persist->nvfs, sizeof(dev->persist->nvfs));
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
mlx4_unload_one(pdev);
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
err = mlx4_load_one(pdev, pci_dev_data, total_vfs, nvfs, priv, 1);
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
if (err) {
mlx4_err(dev, "%s: ERROR: mlx4_load_one failed, pci_name=%s, err=%d\n",
__func__, pci_name(pdev), err);
return err;
}
net/mlx4_core: Set device configuration data to be persistent across reset When an HCA enters an internal error state, this is detected by the driver. The driver then should reset the HCA and restart the software stack. Keep ports information and some SRIOV configuration in a persistent area to have it valid across reset. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:36 +07:00
err = restore_current_port_types(dev, dev->persist->curr_port_type,
dev->persist->curr_port_poss_type);
if (err)
mlx4_err(dev, "could not restore original port types (%d)\n",
err);
net/mlx4_core: New init and exit flow for mlx4_core In the new flow, we separate the pci initialization and teardown from the initialization and teardown of the other resources. __mlx4_init_one handles the pci resources initialization. It then calls mlx4_load_one to initialize the remainder of the resources. When removing a device, mlx4_remove_one is invoked. However, now mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci resources. When mlx4_unload_one returns, mlx4_remove_one then frees the pci resources. The above separation will allow us to implement 'reset flow' in the future. It will also enable more EQs for VFs and is a pre-step to the modern API to enable/disable SRIOV. Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev struct. This new field is used to avoid parsing the num_vfs module parameter each time the mlx4_restart_one is called. Signed-off-by: Majd Dibbiny <majd@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-09-30 16:03:48 +07:00
return err;
mlx4_core: Reset device when internal error is detected Reset the device when an internal error is detected. Also, detect errors by polling the error buffer rather than using interrupts. This is more robust and doesn't depend on MSI-X. Remove the old interrupt handler entirely, since we don't want to support two mechanisms for detecting internal errors. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-07-12 21:50:45 +07:00
}
PCI: Remove DEFINE_PCI_DEVICE_TABLE macro use We should prefer `struct pci_device_id` over `DEFINE_PCI_DEVICE_TABLE` to meet kernel coding style guidelines. This issue was reported by checkpatch. A simplified version of the semantic patch that makes this change is as follows (http://coccinelle.lip6.fr/): // <smpl> @@ identifier i; declarer name DEFINE_PCI_DEVICE_TABLE; initializer z; @@ - DEFINE_PCI_DEVICE_TABLE(i) + const struct pci_device_id i[] = z; // </smpl> [bhelgaas: add semantic patch] Signed-off-by: Benoit Taine <benoit.taine@lip6.fr> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2014-08-08 20:56:03 +07:00
static const struct pci_device_id mlx4_pci_table[] = {
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT25408 "Hermon" SDR */
mlx4_core: Clean up enabling of SENSE_PORT for older (ConnectX-1/-2) HCAs Instead of having a hard-coded "PCI device ID != 0x1003" (which obviously breaks as newer devices with ID != 0x1003 become available), instead let's set a flag in our PCI device table for the older devices where we're supposed to force using SENSE_PORT. This also avoids enabling SENSE_PORT for virtual functions by mistake. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:53:05 +07:00
{ PCI_VDEVICE(MELLANOX, 0x6340), MLX4_PCI_DEV_FORCE_SENSE_PORT },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT25408 "Hermon" DDR */
mlx4_core: Clean up enabling of SENSE_PORT for older (ConnectX-1/-2) HCAs Instead of having a hard-coded "PCI device ID != 0x1003" (which obviously breaks as newer devices with ID != 0x1003 become available), instead let's set a flag in our PCI device table for the older devices where we're supposed to force using SENSE_PORT. This also avoids enabling SENSE_PORT for virtual functions by mistake. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:53:05 +07:00
{ PCI_VDEVICE(MELLANOX, 0x634a), MLX4_PCI_DEV_FORCE_SENSE_PORT },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT25408 "Hermon" QDR */
mlx4_core: Clean up enabling of SENSE_PORT for older (ConnectX-1/-2) HCAs Instead of having a hard-coded "PCI device ID != 0x1003" (which obviously breaks as newer devices with ID != 0x1003 become available), instead let's set a flag in our PCI device table for the older devices where we're supposed to force using SENSE_PORT. This also avoids enabling SENSE_PORT for virtual functions by mistake. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:53:05 +07:00
{ PCI_VDEVICE(MELLANOX, 0x6354), MLX4_PCI_DEV_FORCE_SENSE_PORT },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT25408 "Hermon" DDR PCIe gen2 */
mlx4_core: Clean up enabling of SENSE_PORT for older (ConnectX-1/-2) HCAs Instead of having a hard-coded "PCI device ID != 0x1003" (which obviously breaks as newer devices with ID != 0x1003 become available), instead let's set a flag in our PCI device table for the older devices where we're supposed to force using SENSE_PORT. This also avoids enabling SENSE_PORT for virtual functions by mistake. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:53:05 +07:00
{ PCI_VDEVICE(MELLANOX, 0x6732), MLX4_PCI_DEV_FORCE_SENSE_PORT },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT25408 "Hermon" QDR PCIe gen2 */
mlx4_core: Clean up enabling of SENSE_PORT for older (ConnectX-1/-2) HCAs Instead of having a hard-coded "PCI device ID != 0x1003" (which obviously breaks as newer devices with ID != 0x1003 become available), instead let's set a flag in our PCI device table for the older devices where we're supposed to force using SENSE_PORT. This also avoids enabling SENSE_PORT for virtual functions by mistake. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:53:05 +07:00
{ PCI_VDEVICE(MELLANOX, 0x673c), MLX4_PCI_DEV_FORCE_SENSE_PORT },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT25408 "Hermon" EN 10GigE */
mlx4_core: Clean up enabling of SENSE_PORT for older (ConnectX-1/-2) HCAs Instead of having a hard-coded "PCI device ID != 0x1003" (which obviously breaks as newer devices with ID != 0x1003 become available), instead let's set a flag in our PCI device table for the older devices where we're supposed to force using SENSE_PORT. This also avoids enabling SENSE_PORT for virtual functions by mistake. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:53:05 +07:00
{ PCI_VDEVICE(MELLANOX, 0x6368), MLX4_PCI_DEV_FORCE_SENSE_PORT },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT25408 "Hermon" EN 10GigE PCIe gen2 */
mlx4_core: Clean up enabling of SENSE_PORT for older (ConnectX-1/-2) HCAs Instead of having a hard-coded "PCI device ID != 0x1003" (which obviously breaks as newer devices with ID != 0x1003 become available), instead let's set a flag in our PCI device table for the older devices where we're supposed to force using SENSE_PORT. This also avoids enabling SENSE_PORT for virtual functions by mistake. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:53:05 +07:00
{ PCI_VDEVICE(MELLANOX, 0x6750), MLX4_PCI_DEV_FORCE_SENSE_PORT },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT25458 ConnectX EN 10GBASE-T 10GigE */
mlx4_core: Clean up enabling of SENSE_PORT for older (ConnectX-1/-2) HCAs Instead of having a hard-coded "PCI device ID != 0x1003" (which obviously breaks as newer devices with ID != 0x1003 become available), instead let's set a flag in our PCI device table for the older devices where we're supposed to force using SENSE_PORT. This also avoids enabling SENSE_PORT for virtual functions by mistake. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:53:05 +07:00
{ PCI_VDEVICE(MELLANOX, 0x6372), MLX4_PCI_DEV_FORCE_SENSE_PORT },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT25458 ConnectX EN 10GBASE-T+Gen2 10GigE */
mlx4_core: Clean up enabling of SENSE_PORT for older (ConnectX-1/-2) HCAs Instead of having a hard-coded "PCI device ID != 0x1003" (which obviously breaks as newer devices with ID != 0x1003 become available), instead let's set a flag in our PCI device table for the older devices where we're supposed to force using SENSE_PORT. This also avoids enabling SENSE_PORT for virtual functions by mistake. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:53:05 +07:00
{ PCI_VDEVICE(MELLANOX, 0x675a), MLX4_PCI_DEV_FORCE_SENSE_PORT },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT26468 ConnectX EN 10GigE PCIe gen2*/
mlx4_core: Clean up enabling of SENSE_PORT for older (ConnectX-1/-2) HCAs Instead of having a hard-coded "PCI device ID != 0x1003" (which obviously breaks as newer devices with ID != 0x1003 become available), instead let's set a flag in our PCI device table for the older devices where we're supposed to force using SENSE_PORT. This also avoids enabling SENSE_PORT for virtual functions by mistake. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:53:05 +07:00
{ PCI_VDEVICE(MELLANOX, 0x6764), MLX4_PCI_DEV_FORCE_SENSE_PORT },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT26438 ConnectX EN 40GigE PCIe gen2 5GT/s */
mlx4_core: Clean up enabling of SENSE_PORT for older (ConnectX-1/-2) HCAs Instead of having a hard-coded "PCI device ID != 0x1003" (which obviously breaks as newer devices with ID != 0x1003 become available), instead let's set a flag in our PCI device table for the older devices where we're supposed to force using SENSE_PORT. This also avoids enabling SENSE_PORT for virtual functions by mistake. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:53:05 +07:00
{ PCI_VDEVICE(MELLANOX, 0x6746), MLX4_PCI_DEV_FORCE_SENSE_PORT },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT26478 ConnectX2 40GigE PCIe gen2 */
mlx4_core: Clean up enabling of SENSE_PORT for older (ConnectX-1/-2) HCAs Instead of having a hard-coded "PCI device ID != 0x1003" (which obviously breaks as newer devices with ID != 0x1003 become available), instead let's set a flag in our PCI device table for the older devices where we're supposed to force using SENSE_PORT. This also avoids enabling SENSE_PORT for virtual functions by mistake. Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:53:05 +07:00
{ PCI_VDEVICE(MELLANOX, 0x676e), MLX4_PCI_DEV_FORCE_SENSE_PORT },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT25400 Family [ConnectX-2 Virtual Function] */
mlx4_core: Stash PCI ID driver_data in mlx4_priv structure That way we can check flags later on, when we've finished with the pci_device_id structure. Also convert the "is VF" flag to an enum: "Never do in the preprocessor what can be done in C." Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:23:41 +07:00
{ PCI_VDEVICE(MELLANOX, 0x1002), MLX4_PCI_DEV_IS_VF },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* MT27500 Family [ConnectX-3] */
{ PCI_VDEVICE(MELLANOX, 0x1003), 0 },
/* MT27500 Family [ConnectX-3 Virtual Function] */
mlx4_core: Stash PCI ID driver_data in mlx4_priv structure That way we can check flags later on, when we've finished with the pci_device_id structure. Also convert the "is VF" flag to an enum: "Never do in the preprocessor what can be done in C." Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-09-27 23:23:41 +07:00
{ PCI_VDEVICE(MELLANOX, 0x1004), MLX4_PCI_DEV_IS_VF },
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
{ PCI_VDEVICE(MELLANOX, 0x1005), 0 }, /* MT27510 Family */
{ PCI_VDEVICE(MELLANOX, 0x1006), 0 }, /* MT27511 Family */
{ PCI_VDEVICE(MELLANOX, 0x1007), 0 }, /* MT27520 Family */
{ PCI_VDEVICE(MELLANOX, 0x1008), 0 }, /* MT27521 Family */
{ PCI_VDEVICE(MELLANOX, 0x1009), 0 }, /* MT27530 Family */
{ PCI_VDEVICE(MELLANOX, 0x100a), 0 }, /* MT27531 Family */
{ PCI_VDEVICE(MELLANOX, 0x100b), 0 }, /* MT27540 Family */
{ PCI_VDEVICE(MELLANOX, 0x100c), 0 }, /* MT27541 Family */
{ PCI_VDEVICE(MELLANOX, 0x100d), 0 }, /* MT27550 Family */
{ PCI_VDEVICE(MELLANOX, 0x100e), 0 }, /* MT27551 Family */
{ PCI_VDEVICE(MELLANOX, 0x100f), 0 }, /* MT27560 Family */
{ PCI_VDEVICE(MELLANOX, 0x1010), 0 }, /* MT27561 Family */
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
{ 0, }
};
MODULE_DEVICE_TABLE(pci, mlx4_pci_table);
mlx4: Add support for EEH error recovery Currently the mlx4 drivers don't have the necessary callbacks to implement EEH errors detection and recovery, so the PCI layer uses the probe and remove callbacks to try to recover the device after an error on the bus. However, these callbacks have race conditions with the internal catastrophic error recovery functions, which will also detect the error and this can cause the system to crash if both EEH and catas functions try to reset the device. This patch adds the necessary error recovery callbacks and makes sure that the internal catastrophic error functions will not try to reset the device in such scenarios. It also adds some calls to pci_channel_offline() to suppress reads/writes on the bus when the slot cannot accept I/O operations so we prevent unnecessary accesses to the bus and speed up the device removal. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Shlomo Pongratz <shlomop@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-20 16:55:43 +07:00
static pci_ers_result_t mlx4_pci_err_detected(struct pci_dev *pdev,
pci_channel_state_t state)
{
net/mlx4_core: Handle AER flow properly Fix AER callbacks to work properly, it includes: - Refractoring AER to be aligned with Reset flow support. - Sync with concurrent catas flow. In addition, fix the shutdown PCI callback to sync with concurrent catas flow. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:41 +07:00
struct mlx4_dev_persistent *persist = pci_get_drvdata(pdev);
mlx4_err(persist->dev, "mlx4_pci_err_detected was called\n");
mlx4_enter_error_state(persist);
mlx4: Add support for EEH error recovery Currently the mlx4 drivers don't have the necessary callbacks to implement EEH errors detection and recovery, so the PCI layer uses the probe and remove callbacks to try to recover the device after an error on the bus. However, these callbacks have race conditions with the internal catastrophic error recovery functions, which will also detect the error and this can cause the system to crash if both EEH and catas functions try to reset the device. This patch adds the necessary error recovery callbacks and makes sure that the internal catastrophic error functions will not try to reset the device in such scenarios. It also adds some calls to pci_channel_offline() to suppress reads/writes on the bus when the slot cannot accept I/O operations so we prevent unnecessary accesses to the bus and speed up the device removal. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Shlomo Pongratz <shlomop@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-20 16:55:43 +07:00
net/mlx4_core: Handle AER flow properly Fix AER callbacks to work properly, it includes: - Refractoring AER to be aligned with Reset flow support. - Sync with concurrent catas flow. In addition, fix the shutdown PCI callback to sync with concurrent catas flow. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:41 +07:00
mutex_lock(&persist->interface_state_mutex);
if (persist->interface_state & MLX4_INTERFACE_STATE_UP)
mlx4_unload_one(pdev);
mutex_unlock(&persist->interface_state_mutex);
if (state == pci_channel_io_perm_failure)
return PCI_ERS_RESULT_DISCONNECT;
pci_disable_device(pdev);
return PCI_ERS_RESULT_NEED_RESET;
mlx4: Add support for EEH error recovery Currently the mlx4 drivers don't have the necessary callbacks to implement EEH errors detection and recovery, so the PCI layer uses the probe and remove callbacks to try to recover the device after an error on the bus. However, these callbacks have race conditions with the internal catastrophic error recovery functions, which will also detect the error and this can cause the system to crash if both EEH and catas functions try to reset the device. This patch adds the necessary error recovery callbacks and makes sure that the internal catastrophic error functions will not try to reset the device in such scenarios. It also adds some calls to pci_channel_offline() to suppress reads/writes on the bus when the slot cannot accept I/O operations so we prevent unnecessary accesses to the bus and speed up the device removal. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Shlomo Pongratz <shlomop@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-20 16:55:43 +07:00
}
static pci_ers_result_t mlx4_pci_slot_reset(struct pci_dev *pdev)
{
net/mlx4_core: Handle AER flow properly Fix AER callbacks to work properly, it includes: - Refractoring AER to be aligned with Reset flow support. - Sync with concurrent catas flow. In addition, fix the shutdown PCI callback to sync with concurrent catas flow. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:41 +07:00
struct mlx4_dev_persistent *persist = pci_get_drvdata(pdev);
struct mlx4_dev *dev = persist->dev;
net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one() pci_match_id() just match the static pci_device_id, which may return NULL if someone binds the driver to a device manually using /sys/bus/pci/drivers/.../new_id. This patch wrap up a helper function __mlx4_remove_one() which does the tear down function but preserve the drv_data. Functions like mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out releasing drvdata. Fixes: 97a5221 "net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset". CC: Bjorn Helgaas <bhelgaas@google.com> CC: Amir Vadai <amirv@mellanox.com> CC: Jack Morgenstein <jackm@dev.mellanox.co.il> CC: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-04-14 08:51:19 +07:00
struct mlx4_priv *priv = mlx4_priv(dev);
int ret;
net/mlx4_core: Handle AER flow properly Fix AER callbacks to work properly, it includes: - Refractoring AER to be aligned with Reset flow support. - Sync with concurrent catas flow. In addition, fix the shutdown PCI callback to sync with concurrent catas flow. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:41 +07:00
int nvfs[MLX4_MAX_PORTS + 1] = {0, 0, 0};
int total_vfs;
net/mlx4_core: pass pci_device_id.driver_data to __mlx4_init_one during reset The second parameter of __mlx4_init_one() is used to identify whether the pci_dev is a PF or VF. Currently, when it is invoked in mlx4_pci_slot_reset() this information is missed. This patch match the pci_dev with mlx4_pci_table and passes the pci_device_id.driver_data to __mlx4_init_one() in mlx4_pci_slot_reset(). Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-03-27 08:28:31 +07:00
net/mlx4_core: Handle AER flow properly Fix AER callbacks to work properly, it includes: - Refractoring AER to be aligned with Reset flow support. - Sync with concurrent catas flow. In addition, fix the shutdown PCI callback to sync with concurrent catas flow. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:41 +07:00
mlx4_err(dev, "mlx4_pci_slot_reset was called\n");
ret = pci_enable_device(pdev);
if (ret) {
mlx4_err(dev, "Can not re-enable device, ret=%d\n", ret);
return PCI_ERS_RESULT_DISCONNECT;
}
pci_set_master(pdev);
pci_restore_state(pdev);
pci_save_state(pdev);
total_vfs = dev->persist->num_vfs;
memcpy(nvfs, dev->persist->nvfs, sizeof(dev->persist->nvfs));
mutex_lock(&persist->interface_state_mutex);
if (!(persist->interface_state & MLX4_INTERFACE_STATE_UP)) {
ret = mlx4_load_one(pdev, priv->pci_dev_data, total_vfs, nvfs,
net/mlx4_core: Enable device recovery flow with SRIOV In SRIOV, both the PF and the VF may attempt device recovery whenever they assume that the device is not functioning. When the PF driver resets the device, the VF should detect this and attempt to reinitialize itself. The VF must be able to reset itself under all circumstances, even if the PF is not responsive. The VF shall reset itself in the following cases: 1. Commands are not processed within reasonable time over the communication channel. This is done considering device state and the correct return code based on the command as was done in the native mode, done in the next patch. 2. The VF driver receives an internal error event reported by the PF on the communication channel. This occurs when the PF driver resets the device or when VF is out of sync with the PF. Add 'VF reset' capability, which allows the VF to reinitialize itself even when the PF is not responsive. As PF and VF may run their reset flow simulantanisly, there are several cases that are handled: - Prevent freeing VF resources upon FLR, when PF is in its unloading stage. - Prevent PF getting VF commands before it has finished initializing its resources. - Upon VF startup, check that comm-channel is online before sending commands to the PF and getting timed-out. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:42 +07:00
priv, 1);
net/mlx4_core: Handle AER flow properly Fix AER callbacks to work properly, it includes: - Refractoring AER to be aligned with Reset flow support. - Sync with concurrent catas flow. In addition, fix the shutdown PCI callback to sync with concurrent catas flow. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:41 +07:00
if (ret) {
mlx4_err(dev, "%s: mlx4_load_one failed, ret=%d\n",
__func__, ret);
goto end;
}
ret = restore_current_port_types(dev, dev->persist->
curr_port_type, dev->persist->
curr_port_poss_type);
if (ret)
mlx4_err(dev, "could not restore original port types (%d)\n", ret);
}
end:
mutex_unlock(&persist->interface_state_mutex);
mlx4: Add support for EEH error recovery Currently the mlx4 drivers don't have the necessary callbacks to implement EEH errors detection and recovery, so the PCI layer uses the probe and remove callbacks to try to recover the device after an error on the bus. However, these callbacks have race conditions with the internal catastrophic error recovery functions, which will also detect the error and this can cause the system to crash if both EEH and catas functions try to reset the device. This patch adds the necessary error recovery callbacks and makes sure that the internal catastrophic error functions will not try to reset the device in such scenarios. It also adds some calls to pci_channel_offline() to suppress reads/writes on the bus when the slot cannot accept I/O operations so we prevent unnecessary accesses to the bus and speed up the device removal. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Shlomo Pongratz <shlomop@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-20 16:55:43 +07:00
return ret ? PCI_ERS_RESULT_DISCONNECT : PCI_ERS_RESULT_RECOVERED;
}
net/mlx4_core: Handle AER flow properly Fix AER callbacks to work properly, it includes: - Refractoring AER to be aligned with Reset flow support. - Sync with concurrent catas flow. In addition, fix the shutdown PCI callback to sync with concurrent catas flow. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:41 +07:00
static void mlx4_shutdown(struct pci_dev *pdev)
{
struct mlx4_dev_persistent *persist = pci_get_drvdata(pdev);
mlx4_info(persist->dev, "mlx4_shutdown was called\n");
mutex_lock(&persist->interface_state_mutex);
if (persist->interface_state & MLX4_INTERFACE_STATE_UP)
mlx4_unload_one(pdev);
mutex_unlock(&persist->interface_state_mutex);
}
netdev: make pci_error_handlers const Signed-off-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
2012-09-07 23:33:15 +07:00
static const struct pci_error_handlers mlx4_err_handler = {
mlx4: Add support for EEH error recovery Currently the mlx4 drivers don't have the necessary callbacks to implement EEH errors detection and recovery, so the PCI layer uses the probe and remove callbacks to try to recover the device after an error on the bus. However, these callbacks have race conditions with the internal catastrophic error recovery functions, which will also detect the error and this can cause the system to crash if both EEH and catas functions try to reset the device. This patch adds the necessary error recovery callbacks and makes sure that the internal catastrophic error functions will not try to reset the device in such scenarios. It also adds some calls to pci_channel_offline() to suppress reads/writes on the bus when the slot cannot accept I/O operations so we prevent unnecessary accesses to the bus and speed up the device removal. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Shlomo Pongratz <shlomop@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-20 16:55:43 +07:00
.error_detected = mlx4_pci_err_detected,
.slot_reset = mlx4_pci_slot_reset,
};
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
static struct pci_driver mlx4_driver = {
.name = DRV_NAME,
.id_table = mlx4_pci_table,
.probe = mlx4_init_one,
net/mlx4_core: Handle AER flow properly Fix AER callbacks to work properly, it includes: - Refractoring AER to be aligned with Reset flow support. - Sync with concurrent catas flow. In addition, fix the shutdown PCI callback to sync with concurrent catas flow. Signed-off-by: Yishai Hadas <yishaih@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-25 21:59:41 +07:00
.shutdown = mlx4_shutdown,
mlx4_core: remove __dev* attributes CONFIG_HOTPLUG is going away as an option. As result the __dev* markings will be going away. Remove use of __devinit, __devexit_p, __devinitdata, __devinitconst, and __devexit. Signed-off-by: Bill Pemberton <wfp5p@virginia.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-12-03 21:23:15 +07:00
.remove = mlx4_remove_one,
mlx4: Add support for EEH error recovery Currently the mlx4 drivers don't have the necessary callbacks to implement EEH errors detection and recovery, so the PCI layer uses the probe and remove callbacks to try to recover the device after an error on the bus. However, these callbacks have race conditions with the internal catastrophic error recovery functions, which will also detect the error and this can cause the system to crash if both EEH and catas functions try to reset the device. This patch adds the necessary error recovery callbacks and makes sure that the internal catastrophic error functions will not try to reset the device in such scenarios. It also adds some calls to pci_channel_offline() to suppress reads/writes on the bus when the slot cannot accept I/O operations so we prevent unnecessary accesses to the bus and speed up the device removal. Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com> Acked-by: Shlomo Pongratz <shlomop@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-07-20 16:55:43 +07:00
.err_handler = &mlx4_err_handler,
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
};
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
static int __init mlx4_verify_params(void)
{
if ((log_num_mac < 0) || (log_num_mac > 7)) {
net/mlx4_core: Replace pr_warning() with pr_warn() As checkpatch suggests. Also changed some printk's into pr_* Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-22 19:55:40 +07:00
pr_warn("mlx4_core: bad num_mac: %d\n", log_num_mac);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
return -1;
}
mlx4_core: Deprecate log_num_vlan module param Enable the maximum size (128) supported by the device for the shadow vlans table, ignoring the module parameter that overrides it. This table is only used by the IBoE control plane for setting a vlan index into an RC/UC QP context or UD Address Handle. Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2011-10-16 15:26:21 +07:00
if (log_num_vlan != 0)
net/mlx4_core: Replace pr_warning() with pr_warn() As checkpatch suggests. Also changed some printk's into pr_* Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-22 19:55:40 +07:00
pr_warn("mlx4_core: log_num_vlan - obsolete module param, using %d\n",
MLX4_LOG_NUM_VLANS);
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
net/mlx4_core: Deprecate use_prio module parameter use_prio was added as part of an infrastructure for running FCoE in A0 mode. FCoE didn't get into Mellanox Upstream driver, and when it will, it won't be using A0 steering mode. Therefore we can safely deprecate this module parameter without hurting any existing user. CC: Carol Soto <clsoto@linux.vnet.ibm.com> Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-22 19:55:39 +07:00
if (use_prio != 0)
pr_warn("mlx4_core: use_prio - obsolete module param, ignored\n");
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
mlx4_core: Allow log_mtts_per_seg up to 7 As there are now machines with a lot more physical memory, we need to be able to register more memory. This patch lifts the upper limit of log_mtts_per_seg from 5 to 7, increasing the amount of memory that can be registered. Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2010-09-20 13:42:38 +07:00
if ((log_mtts_per_seg < 1) || (log_mtts_per_seg > 7)) {
net/mlx4_core: Replace pr_warning() with pr_warn() As checkpatch suggests. Also changed some printk's into pr_* Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-22 19:55:40 +07:00
pr_warn("mlx4_core: bad log_mtts_per_seg: %d\n",
log_mtts_per_seg);
mlx4_core: Add module parameter for number of MTTs per segment The current MTT allocator uses kmalloc() to allocate a buffer for its buddy allocator, and thus is limited in the amount of MTT segments that it can control. As a result, the size of memory that can be registered is limited too. This patch uses a module parameter to control the number of MTT entries that each segment represents, allowing more memory to be registered with the same number of segments. Signed-off-by: Eli Cohen <eli@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-05-28 04:38:34 +07:00
return -1;
}
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
/* Check if module param for ports type has legal combination */
if (port_type_array[0] == false && port_type_array[1] == true) {
net/mlx4_core: Replace pr_warning() with pr_warn() As checkpatch suggests. Also changed some printk's into pr_* Signed-off-by: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-22 19:55:40 +07:00
pr_warn("Module parameter configuration ETH/IB is not supported. Switching to default configuration IB/IB\n");
mlx4_core: Modify driver initialization flow to accommodate SRIOV for Ethernet 1. Added module parameters sr_iov and probe_vf for controlling enablement of SRIOV mode. 2. Increased default max num-qps, num-mpts and log_num_macs to accomodate SRIOV mode 3. Added port_type_array as a module parameter to allow driver startup with ports configured as desired. In SRIOV mode, only ETH is supported, and this array is ignored; otherwise, for the case where the FW supports both port types (ETH and IB), the port_type_array parameter is used. By default, the port_type_array is set to configure both ports as IB. 4. When running in sriov mode, the master needs to initialize the ICM eq table to hold the eq's for itself and also for all the slaves. 5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap. 6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca). VFs obtain their startup information from the PF (master) device via the comm channel. 7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver aborts loading the device. 8. Do not allow setting port type via sysfs when running in SRIOV mode. 9. mlx4_get_ownership: Currently, only one PF is supported by the driver. If the HCA is burned with FW which enables more than one PF, only one of the PFs is allowed to run. The first one up grabs a FW ownership semaphone -- all other PFs will find that semaphore taken, and the driver will not allow them to run. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Liran Liss <liranl@mellanox.co.il> Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il> Signed-off-by: David S. Miller <davem@davemloft.net>
2011-12-13 11:18:30 +07:00
port_type_array[0] = true;
}
net/mlx4: Add support for A0 steering Add the required firmware commands for A0 steering and a way to enable that. The firmware support focuses on INIT_HCA, QUERY_HCA, QUERY_PORT, QUERY_DEV_CAP and QUERY_FUNC_CAP commands. Those commands are used to configure and query the device. The different A0 DMFS (steering) modes are: Static - optimized performance, but flow steering rules are limited. This mode should be choosed explicitly by the user in order to be used. Dynamic - this mode should be explicitly choosed by the user. In this mode, the FW works in optimized steering mode as long as it can and afterwards automatically drops to classic (full) DMFS. Disable - this mode should be explicitly choosed by the user. The user instructs the system not to use optimized steering, even if the FW supports Dynamic A0 DMFS (and thus will be able to use optimized steering in Default A0 DMFS mode). Default - this mode is implicitly choosed. In this mode, if the FW supports Dynamic A0 DMFS, it'll work in this mode. Otherwise, it'll work at Disable A0 DMFS mode. Under SRIOV configuration, when the A0 steering mode is enabled, older guest VF drivers who aren't using the RX QP allocation flag (MLX4_RESERVE_A0_QP) will get a QP from the general range and fail when attempting to register a steering rule. To avoid that, the PF context behaviour is changed once on A0 static mode, to require support for the allocation flag in VF drivers too. In order to enable A0 steering, we use log_num_mgm_entry_size param. If the value of the parameter is not positive, we treat the absolute value of log_num_mgm_entry_size as a bit field. Setting bit 2 of this bit field enables static A0 steering. Signed-off-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-12-11 15:58:00 +07:00
if (mlx4_log_num_mgm_entry_size < -7 ||
(mlx4_log_num_mgm_entry_size > 0 &&
(mlx4_log_num_mgm_entry_size < MLX4_MIN_MGM_LOG_ENTRY_SIZE ||
mlx4_log_num_mgm_entry_size > MLX4_MAX_MGM_LOG_ENTRY_SIZE))) {
pr_warn("mlx4_core: mlx4_log_num_mgm_entry_size (%d) not in legal range (-7..0 or %d..%d)\n",
mellanox: Logging message cleanups Use a more current logging style. o Coalesce formats o Add missing spaces for coalesced formats o Align arguments for modified formats o Add missing newlines for some logging messages o Use DRV_NAME as part of format instead of %s, DRV_NAME to reduce overall text. o Use ..., ##__VA_ARGS__ instead of args... in macros o Correct a few format typos o Use a single line message where appropriate Signed-off-by: Joe Perches <joe@perches.com> Acked-By: Amir Vadai <amirv@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-05-08 02:52:57 +07:00
mlx4_log_num_mgm_entry_size,
MLX4_MIN_MGM_LOG_ENTRY_SIZE,
MLX4_MAX_MGM_LOG_ENTRY_SIZE);
mlx4_core: Allow choosing flow steering mode Device managed flow steering will be enabled only under administrator directive provided through setting the existing module parameter log_num_mgm_entry_size to -1 (if the device actually supports flow steering). If flow steering isn't requested or not available, the driver will use the value of log_num_mgm_entry_size and B0 steering. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2012-12-07 00:12:00 +07:00
return -1;
}
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
return 0;
}
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
static int __init mlx4_init(void)
{
int ret;
mlx4_core: Multiple port type support Multi-protocol adapters support different port types. Each consumer of mlx4_core queries for supported port types; in particular mlx4_ib can no longer assume that all physical ports belong to it. Port type is configured through a sysfs interface. When the type of a port is changed, all mlx4 interfaces are unregistered, and then registered again with the new port types. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2008-10-23 05:38:42 +07:00
if (mlx4_verify_params())
return -EINVAL;
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
mlx4_wq = create_singlethread_workqueue("mlx4");
if (!mlx4_wq)
return -ENOMEM;
mlx4_core: Reset device when internal error is detected Reset the device when an internal error is detected. Also, detect errors by polling the error buffer rather than using interrupts. This is more robust and doesn't depend on MSI-X. Remove the old interrupt handler entirely, since we don't want to support two mechanisms for detecting internal errors. Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-07-12 21:50:45 +07:00
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
ret = pci_register_driver(&mlx4_driver);
net/mlx4_core: destroy workqueue when driver fails to register When driver registration fails, we need to clean up the resources allocated before. mlx4_core missed destroying the workqueue allocated. This patch destroys the workqueue when registration fails. Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-12-03 09:04:10 +07:00
if (ret < 0)
destroy_workqueue(mlx4_wq);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
return ret < 0 ? ret : 0;
}
static void __exit mlx4_cleanup(void)
{
pci_unregister_driver(&mlx4_driver);
mlx4_core: Add link type autosensing When a port's link is down (except to driver restart) and the port is configured for auto sensing, we try to sense port link type (Ethernet or InfiniBand) in order to determine how to initialize the port. If the port type needs to be changed, all mlx4 for the device interfaces are unregistered and then registered again with the new port types. Sensing is done with intervals of 3 seconds. Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il> Signed-off-by: Roland Dreier <rolandd@cisco.com>
2009-03-19 09:45:11 +07:00
destroy_workqueue(mlx4_wq);
IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters Add an InfiniBand driver for Mellanox ConnectX adapters. Because these adapters can also be used as ethernet NICs and Fibre Channel HBAs, the driver is split into two modules: mlx4_core: Handles low-level things like device initialization and processing firmware commands. Also controls resource allocation so that the InfiniBand, ethernet and FC functions can share a device without stepping on each other. mlx4_ib: Handles InfiniBand-specific things; plugs into the InfiniBand midlayer. Signed-off-by: Roland Dreier <rolandd@cisco.com>
2007-05-09 08:00:38 +07:00
}
module_init(mlx4_init);
module_exit(mlx4_cleanup);
Reference in New Issue Copy Permalink