AuxXxilium/linux_dsm_epyc7002
1
0
Fork 0
mirror of https://github.com/AuxXxilium/linux_dsm_epyc7002.git synced 2024-12-28 07:35:20 +07:00
Code Issues Actions Packages Projects Releases Wiki Activity
f6e51c354b
linux_dsm_epyc7002/drivers/ntb/hw/intel/ntb_hw_gen1.c

2074 lines
56 KiB
C
Raw Normal View History

PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
/*
* This file is provided under a dual BSD/GPLv2 license. When using or
* redistributing this file, you may do so under either license.
*
* GPL LICENSE SUMMARY
*
* Copyright(c) 2012 Intel Corporation. All rights reserved.
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
* Copyright (C) 2015 EMC Corporation. All Rights Reserved.
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
* Copyright (C) 2016 T-Platforms. All Rights Reserved.
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of version 2 of the GNU General Public License as
* published by the Free Software Foundation.
*
* BSD LICENSE
*
* Copyright(c) 2012 Intel Corporation. All rights reserved.
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
* Copyright (C) 2015 EMC Corporation. All Rights Reserved.
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
* Copyright (C) 2016 T-Platforms. All Rights Reserved.
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
*
* 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 copy
* notice, this list of conditions and the following disclaimer in
* the documentation and/or other materials provided with the
* distribution.
* * Neither the name of Intel Corporation nor the names of its
* contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
* Intel PCIe NTB Linux driver
*
* Contact Information:
* Jon Mason <jon.mason@intel.com>
*/
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
#include <linux/debugfs.h>
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
#include <linux/delay.h>
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
#include <linux/init.h>
#include <linux/interrupt.h>
#include <linux/module.h>
#include <linux/pci.h>
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
#include <linux/random.h>
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
#include <linux/slab.h>
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
#include <linux/ntb.h>
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
#include "ntb_hw_intel.h"
ntb: intel: header definitions refactor Break out the generation specific definitions to different headers Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:18 +07:00
#include "ntb_hw_gen1.h"
#include "ntb_hw_gen3.h"
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
#define NTB_NAME "ntb_hw_intel"
#define NTB_DESC "Intel(R) PCI-E Non-Transparent Bridge Driver"
#define NTB_VER "2.0"
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
MODULE_DESCRIPTION(NTB_DESC);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
MODULE_VERSION(NTB_VER);
MODULE_LICENSE("Dual BSD/GPL");
MODULE_AUTHOR("Intel Corporation");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
#define bar0_off(base, bar) ((base) + ((bar) << 2))
#define bar2_off(base, bar) bar0_off(base, (bar) - 2)
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static const struct intel_ntb_reg xeon_reg;
static const struct intel_ntb_alt_reg xeon_pri_reg;
static const struct intel_ntb_alt_reg xeon_sec_reg;
static const struct intel_ntb_alt_reg xeon_b2b_reg;
static const struct intel_ntb_xlat_reg xeon_pri_xlat;
static const struct intel_ntb_xlat_reg xeon_sec_xlat;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static const struct ntb_dev_ops intel_ntb_ops;
static const struct file_operations intel_ntb_debugfs_info;
NTB: Correct debugfs to work with more than 1 NTB Device Debugfs was setup in NTB to only have a single debugfs directory. This resulted in the leaking of debugfs directories and files when multiple NTB devices were present, due to each device stomping on the variables containing the previous device's values (thus preventing them from being freed on cleanup). Correct this by creating a secondary directory of the PCI BDF for each device present, and nesting the previously existing information in those directories. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-07-31 05:58:49 +07:00
static struct dentry *debugfs_dir;
NTB: Add parameters for Intel SNB B2B addresses Add module parameters for the addresses to be used in B2B topology. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-11 16:45:30 +07:00
static int b2b_mw_idx = -1;
module_param(b2b_mw_idx, int, 0644);
MODULE_PARM_DESC(b2b_mw_idx, "Use this mw idx to access the peer ntb. A "
"value of zero or positive starts from first mw idx, and a "
"negative value starts from last mw idx. Both sides MUST "
"set the same value here!");
static unsigned int b2b_mw_share;
module_param(b2b_mw_share, uint, 0644);
MODULE_PARM_DESC(b2b_mw_share, "If the b2b mw is large enough, configure the "
"ntb so that the peer ntb only occupies the first half of "
"the mw, so the second half can still be used as a mw. Both "
"sides MUST set the same value here!");
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
module_param_named(xeon_b2b_usd_bar2_addr64,
xeon_b2b_usd_addr.bar2_addr64, ullong, 0644);
MODULE_PARM_DESC(xeon_b2b_usd_bar2_addr64,
"XEON B2B USD BAR 2 64-bit address");
module_param_named(xeon_b2b_usd_bar4_addr64,
xeon_b2b_usd_addr.bar4_addr64, ullong, 0644);
NTB: ntb_hw_intel: Fix typo in module parameter descriptions Fix typo in module parameter descriptions. Signed-off-by: Wei Yongjun <weiyj.lk@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-08-08 16:48:42 +07:00
MODULE_PARM_DESC(xeon_b2b_usd_bar4_addr64,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
"XEON B2B USD BAR 4 64-bit address");
module_param_named(xeon_b2b_usd_bar4_addr32,
xeon_b2b_usd_addr.bar4_addr32, ullong, 0644);
NTB: ntb_hw_intel: Fix typo in module parameter descriptions Fix typo in module parameter descriptions. Signed-off-by: Wei Yongjun <weiyj.lk@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-08-08 16:48:42 +07:00
MODULE_PARM_DESC(xeon_b2b_usd_bar4_addr32,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
"XEON B2B USD split-BAR 4 32-bit address");
module_param_named(xeon_b2b_usd_bar5_addr32,
xeon_b2b_usd_addr.bar5_addr32, ullong, 0644);
NTB: ntb_hw_intel: Fix typo in module parameter descriptions Fix typo in module parameter descriptions. Signed-off-by: Wei Yongjun <weiyj.lk@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-08-08 16:48:42 +07:00
MODULE_PARM_DESC(xeon_b2b_usd_bar5_addr32,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
"XEON B2B USD split-BAR 5 32-bit address");
module_param_named(xeon_b2b_dsd_bar2_addr64,
xeon_b2b_dsd_addr.bar2_addr64, ullong, 0644);
MODULE_PARM_DESC(xeon_b2b_dsd_bar2_addr64,
"XEON B2B DSD BAR 2 64-bit address");
module_param_named(xeon_b2b_dsd_bar4_addr64,
xeon_b2b_dsd_addr.bar4_addr64, ullong, 0644);
NTB: ntb_hw_intel: Fix typo in module parameter descriptions Fix typo in module parameter descriptions. Signed-off-by: Wei Yongjun <weiyj.lk@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-08-08 16:48:42 +07:00
MODULE_PARM_DESC(xeon_b2b_dsd_bar4_addr64,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
"XEON B2B DSD BAR 4 64-bit address");
module_param_named(xeon_b2b_dsd_bar4_addr32,
xeon_b2b_dsd_addr.bar4_addr32, ullong, 0644);
NTB: ntb_hw_intel: Fix typo in module parameter descriptions Fix typo in module parameter descriptions. Signed-off-by: Wei Yongjun <weiyj.lk@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-08-08 16:48:42 +07:00
MODULE_PARM_DESC(xeon_b2b_dsd_bar4_addr32,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
"XEON B2B DSD split-BAR 4 32-bit address");
module_param_named(xeon_b2b_dsd_bar5_addr32,
xeon_b2b_dsd_addr.bar5_addr32, ullong, 0644);
NTB: ntb_hw_intel: Fix typo in module parameter descriptions Fix typo in module parameter descriptions. Signed-off-by: Wei Yongjun <weiyj.lk@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-08-08 16:48:42 +07:00
MODULE_PARM_DESC(xeon_b2b_dsd_bar5_addr32,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
"XEON B2B DSD split-BAR 5 32-bit address");
NTB: Add parameters for Intel SNB B2B addresses Add module parameters for the addresses to be used in B2B topology. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-11 16:45:30 +07:00
ntb: Adding Skylake Xeon NTB support The Skylake Xeon NTB hardware has made some changes to the register name, offset, and the way doorbells work. Adding driver support for the new hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-11-17 04:03:38 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
static int xeon_init_isr(struct intel_ntb_dev *ndev);
ntb: Adding Skylake Xeon NTB support The Skylake Xeon NTB hardware has made some changes to the register name, offset, and the way doorbells work. Adding driver support for the new hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-11-17 04:03:38 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static inline void ndev_reset_unsafe_flags(struct intel_ntb_dev *ndev)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->unsafe_flags = 0;
ndev->unsafe_flags_ignore = 0;
/* Only B2B has a workaround to avoid SDOORBELL */
if (ndev->hwerr_flags & NTB_HWERR_SDOORBELL_LOCKUP)
if (!ntb_topo_is_b2b(ndev->ntb.topo))
ndev->unsafe_flags |= NTB_UNSAFE_DB;
/* No low level workaround to avoid SB01BASE */
if (ndev->hwerr_flags & NTB_HWERR_SB01BASE_LOCKUP) {
ndev->unsafe_flags |= NTB_UNSAFE_DB;
ndev->unsafe_flags |= NTB_UNSAFE_SPAD;
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static inline int ndev_is_unsafe(struct intel_ntb_dev *ndev,
unsigned long flag)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return !!(flag & ndev->unsafe_flags & ~ndev->unsafe_flags_ignore);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static inline int ndev_ignore_unsafe(struct intel_ntb_dev *ndev,
unsigned long flag)
NTB: Disable interrupts and poll under high load Disable interrupts and poll under high load Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-04-19 07:59:44 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
flag &= ndev->unsafe_flags;
ndev->unsafe_flags_ignore |= flag;
NTB: Disable interrupts and poll under high load Disable interrupts and poll under high load Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-04-19 07:59:44 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return !!flag;
NTB: Disable interrupts and poll under high load Disable interrupts and poll under high load Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-04-19 07:59:44 +07:00
}
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int ndev_mw_to_bar(struct intel_ntb_dev *ndev, int idx)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Fix range check on memory window index The range check must exclude the upper bound. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-08-31 20:31:00 +07:00
if (idx < 0 || idx >= ndev->mw_count)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
return -EINVAL;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev->reg->mw_bar[idx];
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static inline int ndev_db_addr(struct intel_ntb_dev *ndev,
phys_addr_t *db_addr, resource_size_t *db_size,
phys_addr_t reg_addr, unsigned long reg)
{
NTB: Change WARN_ON_ONCE to pr_warn_once on unsafe The unsafe doorbell and scratchpad access should display reason when WARN is called. Otherwise we get a stack dump without any explanation. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-15 19:22:30 +07:00
if (ndev_is_unsafe(ndev, NTB_UNSAFE_DB))
pr_warn_once("%s: NTB unsafe doorbell access", __func__);
NTB: Disable interrupts and poll under high load Disable interrupts and poll under high load Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-04-19 07:59:44 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (db_addr) {
*db_addr = reg_addr + reg;
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&ndev->ntb.pdev->dev, "Peer db addr %llx\n", *db_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (db_size) {
*db_size = ndev->reg->db_size;
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&ndev->ntb.pdev->dev, "Peer db size %llx\n", *db_size);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
return 0;
}
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
u64 ndev_db_read(struct intel_ntb_dev *ndev,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
void __iomem *mmio)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Change WARN_ON_ONCE to pr_warn_once on unsafe The unsafe doorbell and scratchpad access should display reason when WARN is called. Otherwise we get a stack dump without any explanation. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-15 19:22:30 +07:00
if (ndev_is_unsafe(ndev, NTB_UNSAFE_DB))
pr_warn_once("%s: NTB unsafe doorbell access", __func__);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev->reg->db_ioread(mmio);
}
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int ndev_db_write(struct intel_ntb_dev *ndev, u64 db_bits,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
void __iomem *mmio)
{
NTB: Change WARN_ON_ONCE to pr_warn_once on unsafe The unsafe doorbell and scratchpad access should display reason when WARN is called. Otherwise we get a stack dump without any explanation. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-15 19:22:30 +07:00
if (ndev_is_unsafe(ndev, NTB_UNSAFE_DB))
pr_warn_once("%s: NTB unsafe doorbell access", __func__);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (db_bits & ~ndev->db_valid_mask)
return -EINVAL;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->reg->db_iowrite(db_bits, mmio);
NTB: Disable interrupts and poll under high load Disable interrupts and poll under high load Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-04-19 07:59:44 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return 0;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static inline int ndev_db_set_mask(struct intel_ntb_dev *ndev, u64 db_bits,
void __iomem *mmio)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
unsigned long irqflags;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Change WARN_ON_ONCE to pr_warn_once on unsafe The unsafe doorbell and scratchpad access should display reason when WARN is called. Otherwise we get a stack dump without any explanation. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-15 19:22:30 +07:00
if (ndev_is_unsafe(ndev, NTB_UNSAFE_DB))
pr_warn_once("%s: NTB unsafe doorbell access", __func__);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (db_bits & ~ndev->db_valid_mask)
return -EINVAL;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
spin_lock_irqsave(&ndev->db_mask_lock, irqflags);
{
ndev->db_mask |= db_bits;
ndev->reg->db_iowrite(ndev->db_mask, mmio);
}
spin_unlock_irqrestore(&ndev->db_mask_lock, irqflags);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return 0;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static inline int ndev_db_clear_mask(struct intel_ntb_dev *ndev, u64 db_bits,
void __iomem *mmio)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
unsigned long irqflags;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Change WARN_ON_ONCE to pr_warn_once on unsafe The unsafe doorbell and scratchpad access should display reason when WARN is called. Otherwise we get a stack dump without any explanation. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-15 19:22:30 +07:00
if (ndev_is_unsafe(ndev, NTB_UNSAFE_DB))
pr_warn_once("%s: NTB unsafe doorbell access", __func__);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (db_bits & ~ndev->db_valid_mask)
return -EINVAL;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
spin_lock_irqsave(&ndev->db_mask_lock, irqflags);
{
ndev->db_mask &= ~db_bits;
ndev->reg->db_iowrite(ndev->db_mask, mmio);
}
spin_unlock_irqrestore(&ndev->db_mask_lock, irqflags);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return 0;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static inline int ndev_vec_mask(struct intel_ntb_dev *ndev, int db_vector)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
u64 shift, mask;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
shift = ndev->db_vec_shift;
mask = BIT_ULL(shift) - 1;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return mask << (shift * db_vector);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static inline int ndev_spad_addr(struct intel_ntb_dev *ndev, int idx,
phys_addr_t *spad_addr, phys_addr_t reg_addr,
unsigned long reg)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Change WARN_ON_ONCE to pr_warn_once on unsafe The unsafe doorbell and scratchpad access should display reason when WARN is called. Otherwise we get a stack dump without any explanation. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-15 19:22:30 +07:00
if (ndev_is_unsafe(ndev, NTB_UNSAFE_SPAD))
pr_warn_once("%s: NTB unsafe scratchpad access", __func__);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (idx < 0 || idx >= ndev->spad_count)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
return -EINVAL;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (spad_addr) {
*spad_addr = reg_addr + reg + (idx << 2);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&ndev->ntb.pdev->dev, "Peer spad addr %llx\n",
*spad_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
return 0;
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static inline u32 ndev_spad_read(struct intel_ntb_dev *ndev, int idx,
void __iomem *mmio)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Change WARN_ON_ONCE to pr_warn_once on unsafe The unsafe doorbell and scratchpad access should display reason when WARN is called. Otherwise we get a stack dump without any explanation. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-15 19:22:30 +07:00
if (ndev_is_unsafe(ndev, NTB_UNSAFE_SPAD))
pr_warn_once("%s: NTB unsafe scratchpad access", __func__);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (idx < 0 || idx >= ndev->spad_count)
return 0;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ioread32(mmio + (idx << 2));
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static inline int ndev_spad_write(struct intel_ntb_dev *ndev, int idx, u32 val,
void __iomem *mmio)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Change WARN_ON_ONCE to pr_warn_once on unsafe The unsafe doorbell and scratchpad access should display reason when WARN is called. Otherwise we get a stack dump without any explanation. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-15 19:22:30 +07:00
if (ndev_is_unsafe(ndev, NTB_UNSAFE_SPAD))
pr_warn_once("%s: NTB unsafe scratchpad access", __func__);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (idx < 0 || idx >= ndev->spad_count)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
return -EINVAL;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
iowrite32(val, mmio + (idx << 2));
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
return 0;
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static irqreturn_t ndev_interrupt(struct intel_ntb_dev *ndev, int vec)
NTB: Use DMA Engine to Transmit and Receive Allocate and use a DMA engine channel to transmit and receive data over NTB. If none is allocated, fall back to using the CPU to transfer data. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Dan Williams <dan.j.williams@intel.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com>
2013-02-12 23:52:50 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
u64 vec_mask;
vec_mask = ndev_vec_mask(ndev, vec);
ntb: Adding Skylake Xeon NTB support The Skylake Xeon NTB hardware has made some changes to the register name, offset, and the way doorbells work. Adding driver support for the new hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-11-17 04:03:38 +07:00
if ((ndev->hwerr_flags & NTB_HWERR_MSIX_VECTOR32_BAD) && (vec == 31))
vec_mask |= ndev->db_link_mask;
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&ndev->ntb.pdev->dev, "vec %d vec_mask %llx\n", vec, vec_mask);
NTB: Use DMA Engine to Transmit and Receive Allocate and use a DMA engine channel to transmit and receive data over NTB. If none is allocated, fall back to using the CPU to transfer data. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Dan Williams <dan.j.williams@intel.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com>
2013-02-12 23:52:50 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->last_ts = jiffies;
if (vec_mask & ndev->db_link_mask) {
if (ndev->reg->poll_link(ndev))
ntb_link_event(&ndev->ntb);
}
if (vec_mask & ndev->db_valid_mask)
ntb_db_event(&ndev->ntb, vec);
return IRQ_HANDLED;
NTB: Use DMA Engine to Transmit and Receive Allocate and use a DMA engine channel to transmit and receive data over NTB. If none is allocated, fall back to using the CPU to transfer data. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Dan Williams <dan.j.williams@intel.com> Reviewed-by: Dave Jiang <dave.jiang@intel.com>
2013-02-12 23:52:50 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static irqreturn_t ndev_vec_isr(int irq, void *dev)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct intel_ntb_vec *nvec = dev;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&nvec->ndev->ntb.pdev->dev, "irq: %d nvec->num: %d\n",
ntb: Adding Skylake Xeon NTB support The Skylake Xeon NTB hardware has made some changes to the register name, offset, and the way doorbells work. Adding driver support for the new hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-11-17 04:03:38 +07:00
irq, nvec->num);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev_interrupt(nvec->ndev, nvec->num);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static irqreturn_t ndev_irq_isr(int irq, void *dev)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct intel_ntb_dev *ndev = dev;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
return ndev_interrupt(ndev, irq - ndev->ntb.pdev->irq);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int ndev_init_isr(struct intel_ntb_dev *ndev,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
int msix_min, int msix_max,
int msix_shift, int total_shift)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct pci_dev *pdev;
NTB: Use NUMA memory in Intel driver Allocate memory for the NUMA node of the NTB device. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-19 23:04:52 +07:00
int rc, i, msix_count, node;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
pdev = ndev->ntb.pdev;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Use NUMA memory in Intel driver Allocate memory for the NUMA node of the NTB device. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-19 23:04:52 +07:00
node = dev_to_node(&pdev->dev);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* Mask all doorbell interrupts */
ndev->db_mask = ndev->db_valid_mask;
ndev->reg->db_iowrite(ndev->db_mask,
ndev->self_mmio +
ndev->self_reg->db_mask);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* Try to set up msix irq */
NTB: Use NUMA memory in Intel driver Allocate memory for the NUMA node of the NTB device. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-19 23:04:52 +07:00
ndev->vec = kzalloc_node(msix_max * sizeof(*ndev->vec),
GFP_KERNEL, node);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (!ndev->vec)
goto err_msix_vec_alloc;
NTB: Use NUMA memory in Intel driver Allocate memory for the NUMA node of the NTB device. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-19 23:04:52 +07:00
ndev->msix = kzalloc_node(msix_max * sizeof(*ndev->msix),
GFP_KERNEL, node);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (!ndev->msix)
goto err_msix_alloc;
for (i = 0; i < msix_max; ++i)
ndev->msix[i].entry = i;
msix_count = pci_enable_msix_range(pdev, ndev->msix,
msix_min, msix_max);
if (msix_count < 0)
goto err_msix_enable;
for (i = 0; i < msix_count; ++i) {
ndev->vec[i].ndev = ndev;
ndev->vec[i].num = i;
rc = request_irq(ndev->msix[i].vector, ndev_vec_isr, 0,
"ndev_vec_isr", &ndev->vec[i]);
if (rc)
goto err_msix_request;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "Using %d msix interrupts\n", msix_count);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->db_vec_count = msix_count;
ndev->db_vec_shift = msix_shift;
return 0;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
err_msix_request:
while (i-- > 0)
NTB: Fix 'request_irq()' and 'free_irq()' inconsistancy 'request_irq()' and 'free_irq()' should have the same 'dev_id'. Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr> Acked-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-12-19 12:52:55 +07:00
free_irq(ndev->msix[i].vector, &ndev->vec[i]);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
pci_disable_msix(pdev);
err_msix_enable:
kfree(ndev->msix);
err_msix_alloc:
kfree(ndev->vec);
err_msix_vec_alloc:
ndev->msix = NULL;
ndev->vec = NULL;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* Try to set up msi irq */
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
rc = pci_enable_msi(pdev);
if (rc)
goto err_msi_enable;
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
rc = request_irq(pdev->irq, ndev_irq_isr, 0,
"ndev_irq_isr", ndev);
if (rc)
goto err_msi_request;
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "Using msi interrupts\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->db_vec_count = 1;
ndev->db_vec_shift = total_shift;
return 0;
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
err_msi_request:
pci_disable_msi(pdev);
err_msi_enable:
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* Try to set up intx irq */
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
pci_intx(pdev, 1);
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
rc = request_irq(pdev->irq, ndev_irq_isr, IRQF_SHARED,
"ndev_irq_isr", ndev);
if (rc)
goto err_intx_request;
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "Using intx interrupts\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->db_vec_count = 1;
ndev->db_vec_shift = total_shift;
return 0;
err_intx_request:
return rc;
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static void ndev_deinit_isr(struct intel_ntb_dev *ndev)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct pci_dev *pdev;
int i;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
pdev = ndev->ntb.pdev;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* Mask all doorbell interrupts */
ndev->db_mask = ndev->db_valid_mask;
ndev->reg->db_iowrite(ndev->db_mask,
ndev->self_mmio +
ndev->self_reg->db_mask);
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (ndev->msix) {
i = ndev->db_vec_count;
while (i--)
free_irq(ndev->msix[i].vector, &ndev->vec[i]);
pci_disable_msix(pdev);
kfree(ndev->msix);
kfree(ndev->vec);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
} else {
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
free_irq(pdev->irq, ndev);
if (pci_dev_msi_enabled(pdev))
pci_disable_msi(pdev);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
}
ntb: Adding Skylake Xeon NTB support The Skylake Xeon NTB hardware has made some changes to the register name, offset, and the way doorbells work. Adding driver support for the new hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-11-17 04:03:38 +07:00
static ssize_t ndev_ntb_debugfs_read(struct file *filp, char __user *ubuf,
size_t count, loff_t *offp)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct intel_ntb_dev *ndev;
NTB: ntb_hw_intel: show BAR size in debugfs info It will be useful to know the hardware configured BAR size to diagnose issues with NTB memory windows. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Acked-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-07-22 20:38:22 +07:00
struct pci_dev *pdev;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
void __iomem *mmio;
char *buf;
size_t buf_size;
ssize_t ret, off;
NTB: ntb_hw_intel: show BAR size in debugfs info It will be useful to know the hardware configured BAR size to diagnose issues with NTB memory windows. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Acked-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-07-22 20:38:22 +07:00
union { u64 v64; u32 v32; u16 v16; u8 v8; } u;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev = filp->private_data;
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
pdev = ndev->ntb.pdev;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
mmio = ndev->self_mmio;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
buf_size = min(count, 0x800ul);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
buf = kmalloc(buf_size, GFP_KERNEL);
if (!buf)
return -ENOMEM;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off = 0;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"NTB Device Information:\n");
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"Connection Topology -\t%s\n",
ntb_topo_string(ndev->ntb.topo));
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Improve index handling in B2B MW workaround Check that b2b_mw_idx is in range of the number of memory windows when initializing the device. The workaround is considered to be in effect only if the device b2b_idx is exactly UINT_MAX, instead of any index past the last memory window. Only print B2B MW workaround information in debugfs if the workaround is in effect. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-08-31 20:30:59 +07:00
if (ndev->b2b_idx != UINT_MAX) {
off += scnprintf(buf + off, buf_size - off,
"B2B MW Idx -\t\t%u\n", ndev->b2b_idx);
off += scnprintf(buf + off, buf_size - off,
"B2B Offset -\t\t%#lx\n", ndev->b2b_off);
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"BAR4 Split -\t\t%s\n",
ndev->bar4_split ? "yes" : "no");
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"NTB CTL -\t\t%#06x\n", ndev->ntb_ctl);
off += scnprintf(buf + off, buf_size - off,
"LNK STA -\t\t%#06x\n", ndev->lnk_sta);
if (!ndev->reg->link_is_up(ndev)) {
off += scnprintf(buf + off, buf_size - off,
"Link Status -\t\tDown\n");
} else {
off += scnprintf(buf + off, buf_size - off,
"Link Status -\t\tUp\n");
off += scnprintf(buf + off, buf_size - off,
"Link Speed -\t\tPCI-E Gen %u\n",
NTB_LNK_STA_SPEED(ndev->lnk_sta));
off += scnprintf(buf + off, buf_size - off,
"Link Width -\t\tx%u\n",
NTB_LNK_STA_WIDTH(ndev->lnk_sta));
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"Memory Window Count -\t%u\n", ndev->mw_count);
off += scnprintf(buf + off, buf_size - off,
"Scratchpad Count -\t%u\n", ndev->spad_count);
off += scnprintf(buf + off, buf_size - off,
"Doorbell Count -\t%u\n", ndev->db_count);
off += scnprintf(buf + off, buf_size - off,
"Doorbell Vector Count -\t%u\n", ndev->db_vec_count);
off += scnprintf(buf + off, buf_size - off,
"Doorbell Vector Shift -\t%u\n", ndev->db_vec_shift);
off += scnprintf(buf + off, buf_size - off,
"Doorbell Valid Mask -\t%#llx\n", ndev->db_valid_mask);
off += scnprintf(buf + off, buf_size - off,
"Doorbell Link Mask -\t%#llx\n", ndev->db_link_mask);
off += scnprintf(buf + off, buf_size - off,
"Doorbell Mask Cached -\t%#llx\n", ndev->db_mask);
u.v64 = ndev_db_read(ndev, mmio + ndev->self_reg->db_mask);
off += scnprintf(buf + off, buf_size - off,
"Doorbell Mask -\t\t%#llx\n", u.v64);
u.v64 = ndev_db_read(ndev, mmio + ndev->self_reg->db_bell);
off += scnprintf(buf + off, buf_size - off,
"Doorbell Bell -\t\t%#llx\n", u.v64);
NTB: ntb_hw_intel: show BAR size in debugfs info It will be useful to know the hardware configured BAR size to diagnose issues with NTB memory windows. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Acked-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-07-22 20:38:22 +07:00
off += scnprintf(buf + off, buf_size - off,
"\nNTB Window Size:\n");
pci_read_config_byte(pdev, XEON_PBAR23SZ_OFFSET, &u.v8);
off += scnprintf(buf + off, buf_size - off,
"PBAR23SZ %hhu\n", u.v8);
if (!ndev->bar4_split) {
pci_read_config_byte(pdev, XEON_PBAR45SZ_OFFSET, &u.v8);
off += scnprintf(buf + off, buf_size - off,
"PBAR45SZ %hhu\n", u.v8);
} else {
pci_read_config_byte(pdev, XEON_PBAR4SZ_OFFSET, &u.v8);
off += scnprintf(buf + off, buf_size - off,
"PBAR4SZ %hhu\n", u.v8);
pci_read_config_byte(pdev, XEON_PBAR5SZ_OFFSET, &u.v8);
off += scnprintf(buf + off, buf_size - off,
"PBAR5SZ %hhu\n", u.v8);
}
pci_read_config_byte(pdev, XEON_SBAR23SZ_OFFSET, &u.v8);
off += scnprintf(buf + off, buf_size - off,
"SBAR23SZ %hhu\n", u.v8);
if (!ndev->bar4_split) {
pci_read_config_byte(pdev, XEON_SBAR45SZ_OFFSET, &u.v8);
off += scnprintf(buf + off, buf_size - off,
"SBAR45SZ %hhu\n", u.v8);
} else {
pci_read_config_byte(pdev, XEON_SBAR4SZ_OFFSET, &u.v8);
off += scnprintf(buf + off, buf_size - off,
"SBAR4SZ %hhu\n", u.v8);
pci_read_config_byte(pdev, XEON_SBAR5SZ_OFFSET, &u.v8);
off += scnprintf(buf + off, buf_size - off,
"SBAR5SZ %hhu\n", u.v8);
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"\nNTB Incoming XLAT:\n");
u.v64 = ioread64(mmio + bar2_off(ndev->xlat_reg->bar2_xlat, 2));
off += scnprintf(buf + off, buf_size - off,
"XLAT23 -\t\t%#018llx\n", u.v64);
NTB: Add split BAR output for debugfs stats When split BAR is enabled, the driver needs to dump out the split BAR registers rather than the original 64bit BAR registers. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-18 16:17:30 +07:00
if (ndev->bar4_split) {
u.v32 = ioread32(mmio + bar2_off(ndev->xlat_reg->bar2_xlat, 4));
off += scnprintf(buf + off, buf_size - off,
"XLAT4 -\t\t\t%#06x\n", u.v32);
u.v32 = ioread32(mmio + bar2_off(ndev->xlat_reg->bar2_xlat, 5));
off += scnprintf(buf + off, buf_size - off,
"XLAT5 -\t\t\t%#06x\n", u.v32);
} else {
u.v64 = ioread64(mmio + bar2_off(ndev->xlat_reg->bar2_xlat, 4));
off += scnprintf(buf + off, buf_size - off,
"XLAT45 -\t\t%#018llx\n", u.v64);
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
u.v64 = ioread64(mmio + bar2_off(ndev->xlat_reg->bar2_limit, 2));
off += scnprintf(buf + off, buf_size - off,
"LMT23 -\t\t\t%#018llx\n", u.v64);
NTB: Add split BAR output for debugfs stats When split BAR is enabled, the driver needs to dump out the split BAR registers rather than the original 64bit BAR registers. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-18 16:17:30 +07:00
if (ndev->bar4_split) {
u.v32 = ioread32(mmio + bar2_off(ndev->xlat_reg->bar2_limit, 4));
off += scnprintf(buf + off, buf_size - off,
"LMT4 -\t\t\t%#06x\n", u.v32);
u.v32 = ioread32(mmio + bar2_off(ndev->xlat_reg->bar2_limit, 5));
off += scnprintf(buf + off, buf_size - off,
"LMT5 -\t\t\t%#06x\n", u.v32);
} else {
u.v64 = ioread64(mmio + bar2_off(ndev->xlat_reg->bar2_limit, 4));
off += scnprintf(buf + off, buf_size - off,
"LMT45 -\t\t\t%#018llx\n", u.v64);
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
NTB: ntb_hw_intel: use local variable pdev Clean up duplicated expression by replacing it with the equivalent local variable pdev. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Acked-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-07-22 20:38:23 +07:00
if (pdev_is_xeon(pdev)) {
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (ntb_topo_is_b2b(ndev->ntb.topo)) {
off += scnprintf(buf + off, buf_size - off,
"\nNTB Outgoing B2B XLAT:\n");
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
u.v64 = ioread64(mmio + XEON_PBAR23XLAT_OFFSET);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"B2B XLAT23 -\t\t%#018llx\n", u.v64);
NTB: Add split BAR output for debugfs stats When split BAR is enabled, the driver needs to dump out the split BAR registers rather than the original 64bit BAR registers. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-18 16:17:30 +07:00
if (ndev->bar4_split) {
u.v32 = ioread32(mmio + XEON_PBAR4XLAT_OFFSET);
off += scnprintf(buf + off, buf_size - off,
"B2B XLAT4 -\t\t%#06x\n",
u.v32);
u.v32 = ioread32(mmio + XEON_PBAR5XLAT_OFFSET);
off += scnprintf(buf + off, buf_size - off,
"B2B XLAT5 -\t\t%#06x\n",
u.v32);
} else {
u.v64 = ioread64(mmio + XEON_PBAR45XLAT_OFFSET);
off += scnprintf(buf + off, buf_size - off,
"B2B XLAT45 -\t\t%#018llx\n",
u.v64);
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
u.v64 = ioread64(mmio + XEON_PBAR23LMT_OFFSET);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"B2B LMT23 -\t\t%#018llx\n", u.v64);
NTB: Add split BAR output for debugfs stats When split BAR is enabled, the driver needs to dump out the split BAR registers rather than the original 64bit BAR registers. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-18 16:17:30 +07:00
if (ndev->bar4_split) {
u.v32 = ioread32(mmio + XEON_PBAR4LMT_OFFSET);
off += scnprintf(buf + off, buf_size - off,
"B2B LMT4 -\t\t%#06x\n",
u.v32);
u.v32 = ioread32(mmio + XEON_PBAR5LMT_OFFSET);
off += scnprintf(buf + off, buf_size - off,
"B2B LMT5 -\t\t%#06x\n",
u.v32);
} else {
u.v64 = ioread64(mmio + XEON_PBAR45LMT_OFFSET);
off += scnprintf(buf + off, buf_size - off,
"B2B LMT45 -\t\t%#018llx\n",
u.v64);
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"\nNTB Secondary BAR:\n");
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
u.v64 = ioread64(mmio + XEON_SBAR0BASE_OFFSET);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"SBAR01 -\t\t%#018llx\n", u.v64);
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
u.v64 = ioread64(mmio + XEON_SBAR23BASE_OFFSET);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"SBAR23 -\t\t%#018llx\n", u.v64);
NTB: Add split BAR output for debugfs stats When split BAR is enabled, the driver needs to dump out the split BAR registers rather than the original 64bit BAR registers. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-18 16:17:30 +07:00
if (ndev->bar4_split) {
u.v32 = ioread32(mmio + XEON_SBAR4BASE_OFFSET);
off += scnprintf(buf + off, buf_size - off,
"SBAR4 -\t\t\t%#06x\n", u.v32);
u.v32 = ioread32(mmio + XEON_SBAR5BASE_OFFSET);
off += scnprintf(buf + off, buf_size - off,
"SBAR5 -\t\t\t%#06x\n", u.v32);
} else {
u.v64 = ioread64(mmio + XEON_SBAR45BASE_OFFSET);
off += scnprintf(buf + off, buf_size - off,
"SBAR45 -\t\t%#018llx\n",
u.v64);
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
off += scnprintf(buf + off, buf_size - off,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
"\nXEON NTB Statistics:\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
u.v16 = ioread16(mmio + XEON_USMEMMISS_OFFSET);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"Upstream Memory Miss -\t%u\n", u.v16);
off += scnprintf(buf + off, buf_size - off,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
"\nXEON NTB Hardware Errors:\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
NTB: ntb_hw_intel: use local variable pdev Clean up duplicated expression by replacing it with the equivalent local variable pdev. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Acked-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-07-22 20:38:23 +07:00
if (!pci_read_config_word(pdev,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
XEON_DEVSTS_OFFSET, &u.v16))
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"DEVSTS -\t\t%#06x\n", u.v16);
NTB: ntb_hw_intel: use local variable pdev Clean up duplicated expression by replacing it with the equivalent local variable pdev. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Acked-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-07-22 20:38:23 +07:00
if (!pci_read_config_word(pdev,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
XEON_LINK_STATUS_OFFSET, &u.v16))
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"LNKSTS -\t\t%#06x\n", u.v16);
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
NTB: ntb_hw_intel: use local variable pdev Clean up duplicated expression by replacing it with the equivalent local variable pdev. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Acked-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-07-22 20:38:23 +07:00
if (!pci_read_config_dword(pdev,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
XEON_UNCERRSTS_OFFSET, &u.v32))
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"UNCERRSTS -\t\t%#06x\n", u.v32);
NTB: ntb_hw_intel: use local variable pdev Clean up duplicated expression by replacing it with the equivalent local variable pdev. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Acked-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-07-22 20:38:23 +07:00
if (!pci_read_config_dword(pdev,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
XEON_CORERRSTS_OFFSET, &u.v32))
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
off += scnprintf(buf + off, buf_size - off,
"CORERRSTS -\t\t%#06x\n", u.v32);
}
ret = simple_read_from_buffer(ubuf, count, offp, buf, off);
kfree(buf);
return ret;
NTB: BWD Link Recovery The BWD NTB device will drop the link if an error is encountered on the point-to-point PCI bridge. The link will stay down until all errors are cleared and the link is re-established. On link down, check to see if the error is detected, if so do the necessary housekeeping to try and recover from the error and reestablish the link. There is a potential race between the 2 NTB devices recovering at the same time. If the times are synchronized, the link will not recover and the driver will be stuck in this loop forever. Add a random interval to the recovery time to prevent this race. Signed-off-by: Jon Mason <jon.mason@intel.com>
2012-11-17 08:52:57 +07:00
}
ntb: Adding Skylake Xeon NTB support The Skylake Xeon NTB hardware has made some changes to the register name, offset, and the way doorbells work. Adding driver support for the new hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-11-17 04:03:38 +07:00
static ssize_t ndev_debugfs_read(struct file *filp, char __user *ubuf,
size_t count, loff_t *offp)
{
struct intel_ntb_dev *ndev = filp->private_data;
ntb: remove Intel Atom NTB driver support Removing dead code since this is not being used. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-11-21 00:24:08 +07:00
if (pdev_is_xeon(ndev->ntb.pdev))
ntb: Adding Skylake Xeon NTB support The Skylake Xeon NTB hardware has made some changes to the register name, offset, and the way doorbells work. Adding driver support for the new hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-11-17 04:03:38 +07:00
return ndev_ntb_debugfs_read(filp, ubuf, count, offp);
else if (pdev_is_skx_xeon(ndev->ntb.pdev))
return ndev_ntb3_debugfs_read(filp, ubuf, count, offp);
return -ENXIO;
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static void ndev_init_debugfs(struct intel_ntb_dev *ndev)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (!debugfs_dir) {
ndev->debugfs_dir = NULL;
ndev->debugfs_info = NULL;
} else {
ndev->debugfs_dir =
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
debugfs_create_dir(pci_name(ndev->ntb.pdev),
debugfs_dir);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (!ndev->debugfs_dir)
ndev->debugfs_info = NULL;
else
ndev->debugfs_info =
debugfs_create_file("info", S_IRUSR,
ndev->debugfs_dir, ndev,
&intel_ntb_debugfs_info);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static void ndev_deinit_debugfs(struct intel_ntb_dev *ndev)
{
debugfs_remove_recursive(ndev->debugfs_dir);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_mw_count(struct ntb_dev *ntb, int pidx)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
if (pidx != NTB_DEF_PEER_IDX)
return -EINVAL;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ntb_ndev(ntb)->mw_count;
}
NTB: NTB-RP support Add support for Non-Transparent Bridge connected to a PCI-E Root Port on the remote system (also known as NTB-RP mode). This allows for a NTB enabled system to be connected to a non-NTB enabled system/slot. Modifications to the registers and BARs/MWs on the Secondary side by the remote system are reflected into registers on the Primary side for the local system. Similarly, modifications of registers and BARs/MWs on Primary side by the local system are reflected into registers on the Secondary side for the Remote System. This allows communication between the 2 sides via these registers and BARs/MWs. Note: there is not a fix for the Xeon Errata (that was already worked around in NTB-B2B mode) for NTB-RP mode. Due to this limitation, NTB-RP will not work on the Secondary side with the Xeon Errata workaround enabled. To get around this, disable the workaround via the xeon_errata_workaround=0 modparm. However, this can cause the hang described in the errata. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-07-16 06:43:54 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_mw_get_align(struct ntb_dev *ntb, int pidx, int idx,
resource_size_t *addr_align,
resource_size_t *size_align,
resource_size_t *size_max)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
resource_size_t bar_size, mw_size;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
int bar;
NTB: Xeon Errata Workaround There is a Xeon hardware errata related to writes to SDOORBELL or B2BDOORBELL in conjunction with inbound access to NTB MMIO Space, which may hang the system. To workaround this issue, use one of the memory windows to access the interrupt and scratch pad registers on the remote system. This bypasses the issue, but removes one of the memory windows from use by the transport. This reduction of MWs necessitates adding some logic to determine the number of available MWs. Since some NTB usage methodologies may have unidirectional traffic, the ability to disable the workaround via modparm has been added. See BF113 in http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-c5500-c3500-spec-update.pdf See BT119 in http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-e5-family-spec-update.pdf Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-04-19 07:07:36 +07:00
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
if (pidx != NTB_DEF_PEER_IDX)
return -EINVAL;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (idx >= ndev->b2b_idx && !ndev->b2b_off)
idx += 1;
ntb: Adding split BAR support for Haswell platforms On the Haswell platform, a split BAR option to allow creation of 2 32bit BARs (4 and 5) from the 64bit BAR 4. Adding support for this new option. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:23 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
bar = ndev_mw_to_bar(ndev, idx);
if (bar < 0)
return bar;
NTB: NTB-RP support Add support for Non-Transparent Bridge connected to a PCI-E Root Port on the remote system (also known as NTB-RP mode). This allows for a NTB enabled system to be connected to a non-NTB enabled system/slot. Modifications to the registers and BARs/MWs on the Secondary side by the remote system are reflected into registers on the Primary side for the local system. Similarly, modifications of registers and BARs/MWs on Primary side by the local system are reflected into registers on the Secondary side for the Remote System. This allows communication between the 2 sides via these registers and BARs/MWs. Note: there is not a fix for the Xeon Errata (that was already worked around in NTB-B2B mode) for NTB-RP mode. Due to this limitation, NTB-RP will not work on the Secondary side with the Xeon Errata workaround enabled. To get around this, disable the workaround via the xeon_errata_workaround=0 modparm. However, this can cause the hang described in the errata. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-07-16 06:43:54 +07:00
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
bar_size = pci_resource_len(ndev->ntb.pdev, bar);
ntb: use errata flag set via DID to implement workaround Instead of using a module parameter, we should detect the errata via PCI DID and then set an appropriate flag. This will be used for additional errata later on. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:18 +07:00
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
if (idx == ndev->b2b_idx)
mw_size = bar_size - ndev->b2b_off;
else
mw_size = bar_size;
if (addr_align)
*addr_align = pci_resource_len(ndev->ntb.pdev, bar);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
if (size_align)
*size_align = 1;
NTB: NTB-RP support Add support for Non-Transparent Bridge connected to a PCI-E Root Port on the remote system (also known as NTB-RP mode). This allows for a NTB enabled system to be connected to a non-NTB enabled system/slot. Modifications to the registers and BARs/MWs on the Secondary side by the remote system are reflected into registers on the Primary side for the local system. Similarly, modifications of registers and BARs/MWs on Primary side by the local system are reflected into registers on the Secondary side for the Remote System. This allows communication between the 2 sides via these registers and BARs/MWs. Note: there is not a fix for the Xeon Errata (that was already worked around in NTB-B2B mode) for NTB-RP mode. Due to this limitation, NTB-RP will not work on the Secondary side with the Xeon Errata workaround enabled. To get around this, disable the workaround via the xeon_errata_workaround=0 modparm. However, this can cause the hang described in the errata. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-07-16 06:43:54 +07:00
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
if (size_max)
*size_max = mw_size;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
return 0;
}
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
static int intel_ntb_mw_set_trans(struct ntb_dev *ntb, int pidx, int idx,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
dma_addr_t addr, resource_size_t size)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
unsigned long base_reg, xlat_reg, limit_reg;
resource_size_t bar_size, mw_size;
void __iomem *mmio;
u64 base, limit, reg_val;
int bar;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
if (pidx != NTB_DEF_PEER_IDX)
return -EINVAL;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (idx >= ndev->b2b_idx && !ndev->b2b_off)
idx += 1;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
bar = ndev_mw_to_bar(ndev, idx);
if (bar < 0)
return bar;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
bar_size = pci_resource_len(ndev->ntb.pdev, bar);
if (idx == ndev->b2b_idx)
mw_size = bar_size - ndev->b2b_off;
else
mw_size = bar_size;
/* hardware requires that addr is aligned to bar size */
if (addr & (bar_size - 1))
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
return -EINVAL;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* make sure the range fits in the usable mw size */
if (size > mw_size)
return -EINVAL;
mmio = ndev->self_mmio;
base_reg = bar0_off(ndev->xlat_reg->bar0_base, bar);
xlat_reg = bar2_off(ndev->xlat_reg->bar2_xlat, bar);
limit_reg = bar2_off(ndev->xlat_reg->bar2_limit, bar);
if (bar < 4 || !ndev->bar4_split) {
NTB: Clear property bits in BAR value The lower bits read from a BAR register will contain property bits that we do not care about. Clear those so that we can use the BAR values for limit and xlat registers. Reported-by: Conrad Meyer <cem@freebsd.org> Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-11-20 04:00:54 +07:00
base = ioread64(mmio + base_reg) & NTB_BAR_MASK_64;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* Set the limit if supported, if size is not mw_size */
if (limit_reg && size != mw_size)
limit = base + size;
else
limit = 0;
/* set and verify setting the translation address */
iowrite64(addr, mmio + xlat_reg);
reg_val = ioread64(mmio + xlat_reg);
if (reg_val != addr) {
iowrite64(0, mmio + xlat_reg);
return -EIO;
}
/* set and verify setting the limit */
iowrite64(limit, mmio + limit_reg);
reg_val = ioread64(mmio + limit_reg);
if (reg_val != limit) {
iowrite64(base, mmio + limit_reg);
iowrite64(0, mmio + xlat_reg);
return -EIO;
}
} else {
/* split bar addr range must all be 32 bit */
if (addr & (~0ull << 32))
return -EINVAL;
if ((addr + size) & (~0ull << 32))
return -EINVAL;
NTB: Clear property bits in BAR value The lower bits read from a BAR register will contain property bits that we do not care about. Clear those so that we can use the BAR values for limit and xlat registers. Reported-by: Conrad Meyer <cem@freebsd.org> Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-11-20 04:00:54 +07:00
base = ioread32(mmio + base_reg) & NTB_BAR_MASK_32;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* Set the limit if supported, if size is not mw_size */
if (limit_reg && size != mw_size)
limit = base + size;
else
limit = 0;
/* set and verify setting the translation address */
iowrite32(addr, mmio + xlat_reg);
reg_val = ioread32(mmio + xlat_reg);
if (reg_val != addr) {
iowrite32(0, mmio + xlat_reg);
return -EIO;
}
/* set and verify setting the limit */
iowrite32(limit, mmio + limit_reg);
reg_val = ioread32(mmio + limit_reg);
if (reg_val != limit) {
iowrite32(base, mmio + limit_reg);
iowrite32(0, mmio + xlat_reg);
return -EIO;
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return 0;
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
u64 intel_ntb_link_is_up(struct ntb_dev *ntb, enum ntb_speed *speed,
enum ntb_width *width)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (ndev->reg->link_is_up(ndev)) {
if (speed)
*speed = NTB_LNK_STA_SPEED(ndev->lnk_sta);
if (width)
*width = NTB_LNK_STA_WIDTH(ndev->lnk_sta);
return 1;
} else {
/* TODO MAYBE: is it possible to observe the link speed and
* width while link is training? */
if (speed)
*speed = NTB_SPEED_NONE;
if (width)
*width = NTB_WIDTH_NONE;
return 0;
}
}
static int intel_ntb_link_enable(struct ntb_dev *ntb,
enum ntb_speed max_speed,
enum ntb_width max_width)
{
struct intel_ntb_dev *ndev;
u32 ntb_ctl;
ndev = container_of(ntb, struct intel_ntb_dev, ntb);
if (ndev->ntb.topo == NTB_TOPO_SEC)
return -EINVAL;
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&ntb->pdev->dev,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
"Enabling link with max_speed %d max_width %d\n",
max_speed, max_width);
if (max_speed != NTB_SPEED_AUTO)
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&ntb->pdev->dev, "ignoring max_speed %d\n", max_speed);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (max_width != NTB_WIDTH_AUTO)
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&ntb->pdev->dev, "ignoring max_width %d\n", max_width);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ntb_ctl = ioread32(ndev->self_mmio + ndev->reg->ntb_ctl);
ntb_ctl &= ~(NTB_CTL_DISABLE | NTB_CTL_CFG_LOCK);
ntb_ctl |= NTB_CTL_P2S_BAR2_SNOOP | NTB_CTL_S2P_BAR2_SNOOP;
ntb_ctl |= NTB_CTL_P2S_BAR4_SNOOP | NTB_CTL_S2P_BAR4_SNOOP;
if (ndev->bar4_split)
ntb_ctl |= NTB_CTL_P2S_BAR5_SNOOP | NTB_CTL_S2P_BAR5_SNOOP;
iowrite32(ntb_ctl, ndev->self_mmio + ndev->reg->ntb_ctl);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
return 0;
}
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_link_disable(struct ntb_dev *ntb)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct intel_ntb_dev *ndev;
u32 ntb_cntl;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev = container_of(ntb, struct intel_ntb_dev, ntb);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (ndev->ntb.topo == NTB_TOPO_SEC)
return -EINVAL;
NTB: Add Error Handling in ntb_device_setup If an error is encountered in ntb_device_setup, it is possible that the spci_cmd isn't populated. Writes to the offset can result in a NULL pointer dereference. This issue is easily encountered by running in NTB-RP mode, as it currently is not supported and will generate an error. To get around this issue, return if an error is encountered prior to attempting to write to the spci_cmd offset. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-07-16 03:23:47 +07:00
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&ntb->pdev->dev, "Disabling link\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* Bring NTB link down */
ntb_cntl = ioread32(ndev->self_mmio + ndev->reg->ntb_ctl);
ntb_cntl &= ~(NTB_CTL_P2S_BAR2_SNOOP | NTB_CTL_S2P_BAR2_SNOOP);
ntb_cntl &= ~(NTB_CTL_P2S_BAR4_SNOOP | NTB_CTL_S2P_BAR4_SNOOP);
if (ndev->bar4_split)
ntb_cntl &= ~(NTB_CTL_P2S_BAR5_SNOOP | NTB_CTL_S2P_BAR5_SNOOP);
ntb_cntl |= NTB_CTL_DISABLE | NTB_CTL_CFG_LOCK;
iowrite32(ntb_cntl, ndev->self_mmio + ndev->reg->ntb_ctl);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Add Error Handling in ntb_device_setup If an error is encountered in ntb_device_setup, it is possible that the spci_cmd isn't populated. Writes to the offset can result in a NULL pointer dereference. This issue is easily encountered by running in NTB-RP mode, as it currently is not supported and will generate an error. To get around this issue, return if an error is encountered prior to attempting to write to the spci_cmd offset. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-07-16 03:23:47 +07:00
return 0;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_peer_mw_count(struct ntb_dev *ntb)
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
{
/* Numbers of inbound and outbound memory windows match */
return ntb_ndev(ntb)->mw_count;
}
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_peer_mw_get_addr(struct ntb_dev *ntb, int idx,
phys_addr_t *base, resource_size_t *size)
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
{
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
int bar;
if (idx >= ndev->b2b_idx && !ndev->b2b_off)
idx += 1;
bar = ndev_mw_to_bar(ndev, idx);
if (bar < 0)
return bar;
if (base)
*base = pci_resource_start(ndev->ntb.pdev, bar) +
(idx == ndev->b2b_idx ? ndev->b2b_off : 0);
if (size)
*size = pci_resource_len(ndev->ntb.pdev, bar) -
(idx == ndev->b2b_idx ? ndev->b2b_off : 0);
return 0;
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static int intel_ntb_db_is_unsafe(struct ntb_dev *ntb)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev_ignore_unsafe(ntb_ndev(ntb), NTB_UNSAFE_DB);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
u64 intel_ntb_db_valid_mask(struct ntb_dev *ntb)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ntb_ndev(ntb)->db_valid_mask;
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_db_vector_count(struct ntb_dev *ntb)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
struct intel_ntb_dev *ndev;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev = container_of(ntb, struct intel_ntb_dev, ntb);
NTB: Disable interrupts and poll under high load Disable interrupts and poll under high load Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-04-19 07:59:44 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev->db_vec_count;
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
u64 intel_ntb_db_vector_mask(struct ntb_dev *ntb, int db_vector)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (db_vector < 0 || db_vector > ndev->db_vec_count)
return 0;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev->db_valid_mask & ndev_vec_mask(ndev, db_vector);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static u64 intel_ntb_db_read(struct ntb_dev *ntb)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev_db_read(ndev,
ndev->self_mmio +
ndev->self_reg->db_bell);
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static int intel_ntb_db_clear(struct ntb_dev *ntb, u64 db_bits)
{
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
NTB: Disable interrupts and poll under high load Disable interrupts and poll under high load Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-04-19 07:59:44 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev_db_write(ndev, db_bits,
ndev->self_mmio +
ndev->self_reg->db_bell);
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_db_set_mask(struct ntb_dev *ntb, u64 db_bits)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev_db_set_mask(ndev, db_bits,
ndev->self_mmio +
ndev->self_reg->db_mask);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_db_clear_mask(struct ntb_dev *ntb, u64 db_bits)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev_db_clear_mask(ndev, db_bits,
ndev->self_mmio +
ndev->self_reg->db_mask);
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_peer_db_addr(struct ntb_dev *ntb, phys_addr_t *db_addr,
resource_size_t *db_size)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev_db_addr(ndev, db_addr, db_size, ndev->peer_addr,
ndev->peer_reg->db_bell);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static int intel_ntb_peer_db_set(struct ntb_dev *ntb, u64 db_bits)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev_db_write(ndev, db_bits,
ndev->peer_mmio +
ndev->peer_reg->db_bell);
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_spad_is_unsafe(struct ntb_dev *ntb)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
return ndev_ignore_unsafe(ntb_ndev(ntb), NTB_UNSAFE_SPAD);
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_spad_count(struct ntb_dev *ntb)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
struct intel_ntb_dev *ndev;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev = container_of(ntb, struct intel_ntb_dev, ntb);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev->spad_count;
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
u32 intel_ntb_spad_read(struct ntb_dev *ntb, int idx)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev_spad_read(ndev, idx,
ndev->self_mmio +
ndev->self_reg->spad);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_spad_write(struct ntb_dev *ntb, int idx, u32 val)
ntb: Split ntb_setup_msix() into separate BWD/SNB routines This is an cleanup effort to make ntb_setup_msix() more readable - use ntb_setup_bwd_msix() to init MSI-Xs on BWD hardware and ntb_setup_snb_msix() - on SNB hardware. Function ntb_setup_snb_msix() also initializes MSI-Xs the way it should has been done - looping pci_enable_msix() until success or failure. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-03-11 23:00:22 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
ntb: Split ntb_setup_msix() into separate BWD/SNB routines This is an cleanup effort to make ntb_setup_msix() more readable - use ntb_setup_bwd_msix() to init MSI-Xs on BWD hardware and ntb_setup_snb_msix() - on SNB hardware. Function ntb_setup_snb_msix() also initializes MSI-Xs the way it should has been done - looping pci_enable_msix() until success or failure. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-03-11 23:00:22 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return ndev_spad_write(ndev, idx, val,
ndev->self_mmio +
ndev->self_reg->spad);
}
ntb: Split ntb_setup_msix() into separate BWD/SNB routines This is an cleanup effort to make ntb_setup_msix() more readable - use ntb_setup_bwd_msix() to init MSI-Xs on BWD hardware and ntb_setup_snb_msix() - on SNB hardware. Function ntb_setup_snb_msix() also initializes MSI-Xs the way it should has been done - looping pci_enable_msix() until success or failure. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-03-11 23:00:22 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_peer_spad_addr(struct ntb_dev *ntb, int pidx, int sidx,
phys_addr_t *spad_addr)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
ntb: Split ntb_setup_msix() into separate BWD/SNB routines This is an cleanup effort to make ntb_setup_msix() more readable - use ntb_setup_bwd_msix() to init MSI-Xs on BWD hardware and ntb_setup_snb_msix() - on SNB hardware. Function ntb_setup_snb_msix() also initializes MSI-Xs the way it should has been done - looping pci_enable_msix() until success or failure. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-03-11 23:00:22 +07:00
NTB: Alter Scratchpads API to support multi-ports devices Even though there is no any real NTB hardware, which would have both more than two ports and Scratchpad registers, it is logically correct to have Scratchpad API accepting a peer port index as well. Intel/AMD drivers utilize Primary and Secondary topology to split Scratchpad between connected root devices. Since port-index API introduced, Intel/AMD NTB hardware drivers can use device port to determine which Scratchpad registers actually belong to local and peer devices. The same approach can be used if some potential hardware in future will be multi-port and have some set of Scratchpads. Here are the brief of changes in the API: ntb_spad_count() - return number of Scratchpads per each port ntb_peer_spad_addr(pidx, sidx) - address of Scratchpad register of the peer device with pidx-index ntb_peer_spad_read(pidx, sidx) - read specified Scratchpad register of the peer with pidx-index ntb_peer_spad_write(pidx, sidx) - write data to Scratchpad register of the peer with pidx-index Since there is hardware which doesn't support Scratchpad registers, the corresponding API methods are now made optional. Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:13:20 +07:00
return ndev_spad_addr(ndev, sidx, spad_addr, ndev->peer_addr,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->peer_reg->spad);
}
ntb: Split ntb_setup_msix() into separate BWD/SNB routines This is an cleanup effort to make ntb_setup_msix() more readable - use ntb_setup_bwd_msix() to init MSI-Xs on BWD hardware and ntb_setup_snb_msix() - on SNB hardware. Function ntb_setup_snb_msix() also initializes MSI-Xs the way it should has been done - looping pci_enable_msix() until success or failure. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-03-11 23:00:22 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
u32 intel_ntb_peer_spad_read(struct ntb_dev *ntb, int pidx, int sidx)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
ntb: Split ntb_setup_msix() into separate BWD/SNB routines This is an cleanup effort to make ntb_setup_msix() more readable - use ntb_setup_bwd_msix() to init MSI-Xs on BWD hardware and ntb_setup_snb_msix() - on SNB hardware. Function ntb_setup_snb_msix() also initializes MSI-Xs the way it should has been done - looping pci_enable_msix() until success or failure. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-03-11 23:00:22 +07:00
NTB: Alter Scratchpads API to support multi-ports devices Even though there is no any real NTB hardware, which would have both more than two ports and Scratchpad registers, it is logically correct to have Scratchpad API accepting a peer port index as well. Intel/AMD drivers utilize Primary and Secondary topology to split Scratchpad between connected root devices. Since port-index API introduced, Intel/AMD NTB hardware drivers can use device port to determine which Scratchpad registers actually belong to local and peer devices. The same approach can be used if some potential hardware in future will be multi-port and have some set of Scratchpads. Here are the brief of changes in the API: ntb_spad_count() - return number of Scratchpads per each port ntb_peer_spad_addr(pidx, sidx) - address of Scratchpad register of the peer device with pidx-index ntb_peer_spad_read(pidx, sidx) - read specified Scratchpad register of the peer with pidx-index ntb_peer_spad_write(pidx, sidx) - write data to Scratchpad register of the peer with pidx-index Since there is hardware which doesn't support Scratchpad registers, the corresponding API methods are now made optional. Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:13:20 +07:00
return ndev_spad_read(ndev, sidx,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->peer_mmio +
ndev->peer_reg->spad);
}
ntb: Split ntb_setup_msix() into separate BWD/SNB routines This is an cleanup effort to make ntb_setup_msix() more readable - use ntb_setup_bwd_msix() to init MSI-Xs on BWD hardware and ntb_setup_snb_msix() - on SNB hardware. Function ntb_setup_snb_msix() also initializes MSI-Xs the way it should has been done - looping pci_enable_msix() until success or failure. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-03-11 23:00:22 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int intel_ntb_peer_spad_write(struct ntb_dev *ntb, int pidx, int sidx,
u32 val)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
struct intel_ntb_dev *ndev = ntb_ndev(ntb);
ntb: Split ntb_setup_msix() into separate BWD/SNB routines This is an cleanup effort to make ntb_setup_msix() more readable - use ntb_setup_bwd_msix() to init MSI-Xs on BWD hardware and ntb_setup_snb_msix() - on SNB hardware. Function ntb_setup_snb_msix() also initializes MSI-Xs the way it should has been done - looping pci_enable_msix() until success or failure. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-03-11 23:00:22 +07:00
NTB: Alter Scratchpads API to support multi-ports devices Even though there is no any real NTB hardware, which would have both more than two ports and Scratchpad registers, it is logically correct to have Scratchpad API accepting a peer port index as well. Intel/AMD drivers utilize Primary and Secondary topology to split Scratchpad between connected root devices. Since port-index API introduced, Intel/AMD NTB hardware drivers can use device port to determine which Scratchpad registers actually belong to local and peer devices. The same approach can be used if some potential hardware in future will be multi-port and have some set of Scratchpads. Here are the brief of changes in the API: ntb_spad_count() - return number of Scratchpads per each port ntb_peer_spad_addr(pidx, sidx) - address of Scratchpad register of the peer device with pidx-index ntb_peer_spad_read(pidx, sidx) - read specified Scratchpad register of the peer with pidx-index ntb_peer_spad_write(pidx, sidx) - write data to Scratchpad register of the peer with pidx-index Since there is hardware which doesn't support Scratchpad registers, the corresponding API methods are now made optional. Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:13:20 +07:00
return ndev_spad_write(ndev, sidx, val,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->peer_mmio +
ndev->peer_reg->spad);
}
ntb: Split ntb_setup_msix() into separate BWD/SNB routines This is an cleanup effort to make ntb_setup_msix() more readable - use ntb_setup_bwd_msix() to init MSI-Xs on BWD hardware and ntb_setup_snb_msix() - on SNB hardware. Function ntb_setup_snb_msix() also initializes MSI-Xs the way it should has been done - looping pci_enable_msix() until success or failure. Signed-off-by: Alexander Gordeev <agordeev@redhat.com> Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-03-11 23:00:22 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static u64 xeon_db_ioread(void __iomem *mmio)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
return (u64)ioread16(mmio);
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static void xeon_db_iowrite(u64 bits, void __iomem *mmio)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
iowrite16((u16)bits, mmio);
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static int xeon_poll_link(struct intel_ntb_dev *ndev)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
u16 reg_val;
int rc;
ndev->reg->db_iowrite(ndev->db_link_mask,
ndev->self_mmio +
ndev->self_reg->db_bell);
rc = pci_read_config_word(ndev->ntb.pdev,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
XEON_LINK_STATUS_OFFSET, &reg_val);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (rc)
return 0;
if (reg_val == ndev->lnk_sta)
return 0;
ndev->lnk_sta = reg_val;
return 1;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
int xeon_link_is_up(struct intel_ntb_dev *ndev)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Enable link for Intel root port mode in probe Link training should be enabled in the driver probe for root port mode. We should not have to wait for transport to be loaded for this to happen. Otherwise the ntb device will not show up on the transparent bridge side of the link. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 03:59:34 +07:00
if (ndev->ntb.topo == NTB_TOPO_SEC)
return 1;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return NTB_LNK_STA_ACTIVE(ndev->lnk_sta);
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
enum ntb_topo xeon_ppd_topo(struct intel_ntb_dev *ndev, u8 ppd)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
switch (ppd & XEON_PPD_TOPO_MASK) {
case XEON_PPD_TOPO_B2B_USD:
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return NTB_TOPO_B2B_USD;
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
case XEON_PPD_TOPO_B2B_DSD:
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return NTB_TOPO_B2B_DSD;
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
case XEON_PPD_TOPO_PRI_USD:
case XEON_PPD_TOPO_PRI_DSD: /* accept bogus PRI_DSD */
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return NTB_TOPO_PRI;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
case XEON_PPD_TOPO_SEC_USD:
case XEON_PPD_TOPO_SEC_DSD: /* accept bogus SEC_DSD */
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return NTB_TOPO_SEC;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return NTB_TOPO_NONE;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static inline int xeon_ppd_bar4_split(struct intel_ntb_dev *ndev, u8 ppd)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
if (ppd & XEON_PPD_SPLIT_BAR_MASK) {
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&ndev->ntb.pdev->dev, "PPD %d split bar\n", ppd);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return 1;
}
return 0;
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static int xeon_init_isr(struct intel_ntb_dev *ndev)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
return ndev_init_isr(ndev, XEON_DB_MSIX_VECTOR_COUNT,
XEON_DB_MSIX_VECTOR_COUNT,
XEON_DB_MSIX_VECTOR_SHIFT,
XEON_DB_TOTAL_SHIFT);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static void xeon_deinit_isr(struct intel_ntb_dev *ndev)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
ndev_deinit_isr(ndev);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static int xeon_setup_b2b_mw(struct intel_ntb_dev *ndev,
const struct intel_b2b_addr *addr,
const struct intel_b2b_addr *peer_addr)
NTB: debugfs device entry Create a debugfs entry for the NTB device to log the basic device info, as well as display the error count on a number of registers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-04-08 00:55:47 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct pci_dev *pdev;
void __iomem *mmio;
resource_size_t bar_size;
phys_addr_t bar_addr;
int b2b_bar;
u8 bar_sz;
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
pdev = ndev->ntb.pdev;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
mmio = ndev->self_mmio;
NTB: Improve index handling in B2B MW workaround Check that b2b_mw_idx is in range of the number of memory windows when initializing the device. The workaround is considered to be in effect only if the device b2b_idx is exactly UINT_MAX, instead of any index past the last memory window. Only print B2B MW workaround information in debugfs if the workaround is in effect. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-08-31 20:30:59 +07:00
if (ndev->b2b_idx == UINT_MAX) {
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "not using b2b mw\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
b2b_bar = 0;
ndev->b2b_off = 0;
} else {
b2b_bar = ndev_mw_to_bar(ndev, ndev->b2b_idx);
if (b2b_bar < 0)
return -EIO;
NTB: debugfs device entry Create a debugfs entry for the NTB device to log the basic device info, as well as display the error count on a number of registers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-04-08 00:55:47 +07:00
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "using b2b mw bar %d\n", b2b_bar);
NTB: debugfs device entry Create a debugfs entry for the NTB device to log the basic device info, as well as display the error count on a number of registers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-04-08 00:55:47 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
bar_size = pci_resource_len(ndev->ntb.pdev, b2b_bar);
NTB: debugfs device entry Create a debugfs entry for the NTB device to log the basic device info, as well as display the error count on a number of registers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-04-08 00:55:47 +07:00
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "b2b bar size %#llx\n", bar_size);
NTB: debugfs device entry Create a debugfs entry for the NTB device to log the basic device info, as well as display the error count on a number of registers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-04-08 00:55:47 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
if (b2b_mw_share && XEON_B2B_MIN_SIZE <= bar_size >> 1) {
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "b2b using first half of bar\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->b2b_off = bar_size >> 1;
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
} else if (XEON_B2B_MIN_SIZE <= bar_size) {
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "b2b using whole bar\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->b2b_off = 0;
--ndev->mw_count;
} else {
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "b2b bar size is too small\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return -EIO;
}
NTB: debugfs device entry Create a debugfs entry for the NTB device to log the basic device info, as well as display the error count on a number of registers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-04-08 00:55:47 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* Reset the secondary bar sizes to match the primary bar sizes,
* except disable or halve the size of the b2b secondary bar.
*
* Note: code for each specific bar size register, because the register
* offsets are not in a consistent order (bar5sz comes after ppd, odd).
*/
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
pci_read_config_byte(pdev, XEON_PBAR23SZ_OFFSET, &bar_sz);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "PBAR23SZ %#x\n", bar_sz);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (b2b_bar == 2) {
if (ndev->b2b_off)
bar_sz -= 1;
else
bar_sz = 0;
}
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
pci_write_config_byte(pdev, XEON_SBAR23SZ_OFFSET, bar_sz);
pci_read_config_byte(pdev, XEON_SBAR23SZ_OFFSET, &bar_sz);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "SBAR23SZ %#x\n", bar_sz);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (!ndev->bar4_split) {
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
pci_read_config_byte(pdev, XEON_PBAR45SZ_OFFSET, &bar_sz);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "PBAR45SZ %#x\n", bar_sz);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (b2b_bar == 4) {
if (ndev->b2b_off)
bar_sz -= 1;
else
bar_sz = 0;
}
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
pci_write_config_byte(pdev, XEON_SBAR45SZ_OFFSET, bar_sz);
pci_read_config_byte(pdev, XEON_SBAR45SZ_OFFSET, &bar_sz);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "SBAR45SZ %#x\n", bar_sz);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
} else {
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
pci_read_config_byte(pdev, XEON_PBAR4SZ_OFFSET, &bar_sz);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "PBAR4SZ %#x\n", bar_sz);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (b2b_bar == 4) {
if (ndev->b2b_off)
bar_sz -= 1;
else
bar_sz = 0;
}
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
pci_write_config_byte(pdev, XEON_SBAR4SZ_OFFSET, bar_sz);
pci_read_config_byte(pdev, XEON_SBAR4SZ_OFFSET, &bar_sz);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "SBAR4SZ %#x\n", bar_sz);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
pci_read_config_byte(pdev, XEON_PBAR5SZ_OFFSET, &bar_sz);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "PBAR5SZ %#x\n", bar_sz);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (b2b_bar == 5) {
if (ndev->b2b_off)
bar_sz -= 1;
else
bar_sz = 0;
}
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
pci_write_config_byte(pdev, XEON_SBAR5SZ_OFFSET, bar_sz);
pci_read_config_byte(pdev, XEON_SBAR5SZ_OFFSET, &bar_sz);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "SBAR5SZ %#x\n", bar_sz);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
NTB: debugfs device entry Create a debugfs entry for the NTB device to log the basic device info, as well as display the error count on a number of registers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-04-08 00:55:47 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* SBAR01 hit by first part of the b2b bar */
if (b2b_bar == 0)
bar_addr = addr->bar0_addr;
else if (b2b_bar == 2)
bar_addr = addr->bar2_addr64;
else if (b2b_bar == 4 && !ndev->bar4_split)
bar_addr = addr->bar4_addr64;
else if (b2b_bar == 4)
bar_addr = addr->bar4_addr32;
else if (b2b_bar == 5)
bar_addr = addr->bar5_addr32;
else
return -EIO;
NTB: debugfs device entry Create a debugfs entry for the NTB device to log the basic device info, as well as display the error count on a number of registers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-04-08 00:55:47 +07:00
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "SBAR01 %#018llx\n", bar_addr);
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite64(bar_addr, mmio + XEON_SBAR0BASE_OFFSET);
NTB: debugfs device entry Create a debugfs entry for the NTB device to log the basic device info, as well as display the error count on a number of registers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-04-08 00:55:47 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* Other SBAR are normally hit by the PBAR xlat, except for b2b bar.
* The b2b bar is either disabled above, or configured half-size, and
* it starts at the PBAR xlat + offset.
*/
NTB: Correct debugfs to work with more than 1 NTB Device Debugfs was setup in NTB to only have a single debugfs directory. This resulted in the leaking of debugfs directories and files when multiple NTB devices were present, due to each device stomping on the variables containing the previous device's values (thus preventing them from being freed on cleanup). Correct this by creating a secondary directory of the PCI BDF for each device present, and nesting the previously existing information in those directories. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-07-31 05:58:49 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
bar_addr = addr->bar2_addr64 + (b2b_bar == 2 ? ndev->b2b_off : 0);
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite64(bar_addr, mmio + XEON_SBAR23BASE_OFFSET);
bar_addr = ioread64(mmio + XEON_SBAR23BASE_OFFSET);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "SBAR23 %#018llx\n", bar_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (!ndev->bar4_split) {
bar_addr = addr->bar4_addr64 +
(b2b_bar == 4 ? ndev->b2b_off : 0);
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite64(bar_addr, mmio + XEON_SBAR45BASE_OFFSET);
bar_addr = ioread64(mmio + XEON_SBAR45BASE_OFFSET);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "SBAR45 %#018llx\n", bar_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
} else {
bar_addr = addr->bar4_addr32 +
(b2b_bar == 4 ? ndev->b2b_off : 0);
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite32(bar_addr, mmio + XEON_SBAR4BASE_OFFSET);
bar_addr = ioread32(mmio + XEON_SBAR4BASE_OFFSET);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "SBAR4 %#010llx\n", bar_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
bar_addr = addr->bar5_addr32 +
(b2b_bar == 5 ? ndev->b2b_off : 0);
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite32(bar_addr, mmio + XEON_SBAR5BASE_OFFSET);
bar_addr = ioread32(mmio + XEON_SBAR5BASE_OFFSET);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "SBAR5 %#010llx\n", bar_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
NTB: Correct debugfs to work with more than 1 NTB Device Debugfs was setup in NTB to only have a single debugfs directory. This resulted in the leaking of debugfs directories and files when multiple NTB devices were present, due to each device stomping on the variables containing the previous device's values (thus preventing them from being freed on cleanup). Correct this by creating a secondary directory of the PCI BDF for each device present, and nesting the previously existing information in those directories. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-07-31 05:58:49 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* setup incoming bar limits == base addrs (zero length windows) */
NTB: Correct debugfs to work with more than 1 NTB Device Debugfs was setup in NTB to only have a single debugfs directory. This resulted in the leaking of debugfs directories and files when multiple NTB devices were present, due to each device stomping on the variables containing the previous device's values (thus preventing them from being freed on cleanup). Correct this by creating a secondary directory of the PCI BDF for each device present, and nesting the previously existing information in those directories. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-07-31 05:58:49 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
bar_addr = addr->bar2_addr64 + (b2b_bar == 2 ? ndev->b2b_off : 0);
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite64(bar_addr, mmio + XEON_SBAR23LMT_OFFSET);
bar_addr = ioread64(mmio + XEON_SBAR23LMT_OFFSET);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "SBAR23LMT %#018llx\n", bar_addr);
NTB: Correct debugfs to work with more than 1 NTB Device Debugfs was setup in NTB to only have a single debugfs directory. This resulted in the leaking of debugfs directories and files when multiple NTB devices were present, due to each device stomping on the variables containing the previous device's values (thus preventing them from being freed on cleanup). Correct this by creating a secondary directory of the PCI BDF for each device present, and nesting the previously existing information in those directories. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-07-31 05:58:49 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (!ndev->bar4_split) {
bar_addr = addr->bar4_addr64 +
(b2b_bar == 4 ? ndev->b2b_off : 0);
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite64(bar_addr, mmio + XEON_SBAR45LMT_OFFSET);
bar_addr = ioread64(mmio + XEON_SBAR45LMT_OFFSET);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "SBAR45LMT %#018llx\n", bar_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
} else {
bar_addr = addr->bar4_addr32 +
(b2b_bar == 4 ? ndev->b2b_off : 0);
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite32(bar_addr, mmio + XEON_SBAR4LMT_OFFSET);
bar_addr = ioread32(mmio + XEON_SBAR4LMT_OFFSET);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "SBAR4LMT %#010llx\n", bar_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
bar_addr = addr->bar5_addr32 +
(b2b_bar == 5 ? ndev->b2b_off : 0);
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite32(bar_addr, mmio + XEON_SBAR5LMT_OFFSET);
bar_addr = ioread32(mmio + XEON_SBAR5LMT_OFFSET);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "SBAR5LMT %#05llx\n", bar_addr);
NTB: Correct debugfs to work with more than 1 NTB Device Debugfs was setup in NTB to only have a single debugfs directory. This resulted in the leaking of debugfs directories and files when multiple NTB devices were present, due to each device stomping on the variables containing the previous device's values (thus preventing them from being freed on cleanup). Correct this by creating a secondary directory of the PCI BDF for each device present, and nesting the previously existing information in those directories. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-07-31 05:58:49 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* zero incoming translation addrs */
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite64(0, mmio + XEON_SBAR23XLAT_OFFSET);
NTB: Enable Snoop on Primary Side Enable Snoop from Primary to Secondary side on BAR23 and BAR45 on all TLPs. Previously, Snoop was only enabled from Secondary to Primary side. This can have a performance improvement on some workloads. Also, make the code more obvious about how the link is being enabled. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-10-04 07:24:03 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (!ndev->bar4_split) {
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite64(0, mmio + XEON_SBAR45XLAT_OFFSET);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
} else {
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite32(0, mmio + XEON_SBAR4XLAT_OFFSET);
iowrite32(0, mmio + XEON_SBAR5XLAT_OFFSET);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
ntb: Adding split BAR support for Haswell platforms On the Haswell platform, a split BAR option to allow creation of 2 32bit BARs (4 and 5) from the 64bit BAR 4. Adding support for this new option. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:23 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* zero outgoing translation limits (whole bar size windows) */
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite64(0, mmio + XEON_PBAR23LMT_OFFSET);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (!ndev->bar4_split) {
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite64(0, mmio + XEON_PBAR45LMT_OFFSET);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
} else {
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite32(0, mmio + XEON_PBAR4LMT_OFFSET);
iowrite32(0, mmio + XEON_PBAR5LMT_OFFSET);
NTB: Enable Snoop on Primary Side Enable Snoop from Primary to Secondary side on BAR23 and BAR45 on all TLPs. Previously, Snoop was only enabled from Secondary to Primary side. This can have a performance improvement on some workloads. Also, make the code more obvious about how the link is being enabled. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-10-04 07:24:03 +07:00
}
NTB: Fix NTB-RP Link Up The Xeon NTB-RP setup, the transparent side does not get a link up/down interrupt. Since the presence of a NTB device on the transparent side means that we have a NTB link up, we can work around the lack of an interrupt by simply calling the link up function to notify the upper layers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-09-14 07:05:23 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* set outgoing translation offsets */
bar_addr = peer_addr->bar2_addr64;
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite64(bar_addr, mmio + XEON_PBAR23XLAT_OFFSET);
bar_addr = ioread64(mmio + XEON_PBAR23XLAT_OFFSET);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "PBAR23XLAT %#018llx\n", bar_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (!ndev->bar4_split) {
bar_addr = peer_addr->bar4_addr64;
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite64(bar_addr, mmio + XEON_PBAR45XLAT_OFFSET);
bar_addr = ioread64(mmio + XEON_PBAR45XLAT_OFFSET);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "PBAR45XLAT %#018llx\n", bar_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
} else {
bar_addr = peer_addr->bar4_addr32;
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite32(bar_addr, mmio + XEON_PBAR4XLAT_OFFSET);
bar_addr = ioread32(mmio + XEON_PBAR4XLAT_OFFSET);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "PBAR4XLAT %#010llx\n", bar_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
bar_addr = peer_addr->bar5_addr32;
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite32(bar_addr, mmio + XEON_PBAR5XLAT_OFFSET);
bar_addr = ioread32(mmio + XEON_PBAR5XLAT_OFFSET);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "PBAR5XLAT %#010llx\n", bar_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
NTB: Fix NTB-RP Link Up The Xeon NTB-RP setup, the transparent side does not get a link up/down interrupt. Since the presence of a NTB device on the transparent side means that we have a NTB link up, we can work around the lack of an interrupt by simply calling the link up function to notify the upper layers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-09-14 07:05:23 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* set the translation offset for b2b registers */
if (b2b_bar == 0)
bar_addr = peer_addr->bar0_addr;
else if (b2b_bar == 2)
bar_addr = peer_addr->bar2_addr64;
else if (b2b_bar == 4 && !ndev->bar4_split)
bar_addr = peer_addr->bar4_addr64;
else if (b2b_bar == 4)
bar_addr = peer_addr->bar4_addr32;
else if (b2b_bar == 5)
bar_addr = peer_addr->bar5_addr32;
else
return -EIO;
/* B2B_XLAT_OFFSET is 64bit, but can only take 32bit writes */
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "B2BXLAT %#018llx\n", bar_addr);
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
iowrite32(bar_addr, mmio + XEON_B2B_XLAT_OFFSETL);
iowrite32(bar_addr >> 32, mmio + XEON_B2B_XLAT_OFFSETU);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (b2b_bar) {
/* map peer ntb mmio config space registers */
ndev->peer_mmio = pci_iomap(pdev, b2b_bar,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
XEON_B2B_MIN_SIZE);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (!ndev->peer_mmio)
return -EIO;
ntb: ntb_hw_intel: init peer_addr in struct intel_ntb_dev The peer_addr member of intel_ntb_dev is not set, therefore when acquiring ntb_peer_db and ntb_peer_spad we only get the offset rather than the actual physical address. Adding fix to correct that. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-10-28 01:06:44 +07:00
ndev->peer_addr = pci_resource_start(pdev, b2b_bar);
NTB: Fix NTB-RP Link Up The Xeon NTB-RP setup, the transparent side does not get a link up/down interrupt. Since the presence of a NTB device on the transparent side means that we have a NTB link up, we can work around the lack of an interrupt by simply calling the link up function to notify the upper layers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-09-14 07:05:23 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return 0;
NTB: Fix NTB-RP Link Up The Xeon NTB-RP setup, the transparent side does not get a link up/down interrupt. Since the presence of a NTB device on the transparent side means that we have a NTB link up, we can work around the lack of an interrupt by simply calling the link up function to notify the upper layers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2013-09-14 07:05:23 +07:00
}
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static int xeon_init_ntb(struct intel_ntb_dev *ndev)
ntb: Adding split BAR support for Haswell platforms On the Haswell platform, a split BAR option to allow creation of 2 32bit BARs (4 and 5) from the 64bit BAR 4. Adding support for this new option. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:23 +07:00
{
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
struct device *dev = &ndev->ntb.pdev->dev;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
int rc;
NTB: Enable link for Intel root port mode in probe Link training should be enabled in the driver probe for root port mode. We should not have to wait for transport to be loaded for this to happen. Otherwise the ntb device will not show up on the transparent bridge side of the link. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 03:59:34 +07:00
u32 ntb_ctl;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (ndev->bar4_split)
ndev->mw_count = HSX_SPLIT_BAR_MW_COUNT;
ntb: Adding split BAR support for Haswell platforms On the Haswell platform, a split BAR option to allow creation of 2 32bit BARs (4 and 5) from the 64bit BAR 4. Adding support for this new option. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:23 +07:00
else
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
ndev->mw_count = XEON_MW_COUNT;
ntb: Adding split BAR support for Haswell platforms On the Haswell platform, a split BAR option to allow creation of 2 32bit BARs (4 and 5) from the 64bit BAR 4. Adding support for this new option. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:23 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
ndev->spad_count = XEON_SPAD_COUNT;
ndev->db_count = XEON_DB_COUNT;
ndev->db_link_mask = XEON_DB_LINK_BIT;
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
switch (ndev->ntb.topo) {
case NTB_TOPO_PRI:
if (ndev->hwerr_flags & NTB_HWERR_SDOORBELL_LOCKUP) {
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_err(dev, "NTB Primary config disabled\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return -EINVAL;
}
NTB: Enable link for Intel root port mode in probe Link training should be enabled in the driver probe for root port mode. We should not have to wait for transport to be loaded for this to happen. Otherwise the ntb device will not show up on the transparent bridge side of the link. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 03:59:34 +07:00
/* enable link to allow secondary side device to appear */
ntb_ctl = ioread32(ndev->self_mmio + ndev->reg->ntb_ctl);
ntb_ctl &= ~NTB_CTL_DISABLE;
iowrite32(ntb_ctl, ndev->self_mmio + ndev->reg->ntb_ctl);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* use half the spads for the peer */
ndev->spad_count >>= 1;
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
ndev->self_reg = &xeon_pri_reg;
ndev->peer_reg = &xeon_sec_reg;
ndev->xlat_reg = &xeon_sec_xlat;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
break;
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
case NTB_TOPO_SEC:
if (ndev->hwerr_flags & NTB_HWERR_SDOORBELL_LOCKUP) {
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_err(dev, "NTB Secondary config disabled\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return -EINVAL;
}
/* use half the spads for the peer */
ndev->spad_count >>= 1;
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
ndev->self_reg = &xeon_sec_reg;
ndev->peer_reg = &xeon_pri_reg;
ndev->xlat_reg = &xeon_pri_xlat;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
break;
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
case NTB_TOPO_B2B_USD:
case NTB_TOPO_B2B_DSD:
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
ndev->self_reg = &xeon_pri_reg;
ndev->peer_reg = &xeon_b2b_reg;
ndev->xlat_reg = &xeon_sec_xlat;
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (ndev->hwerr_flags & NTB_HWERR_SDOORBELL_LOCKUP) {
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
ndev->peer_reg = &xeon_pri_reg;
ntb: Adding split BAR support for Haswell platforms On the Haswell platform, a split BAR option to allow creation of 2 32bit BARs (4 and 5) from the 64bit BAR 4. Adding support for this new option. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:23 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (b2b_mw_idx < 0)
ndev->b2b_idx = b2b_mw_idx + ndev->mw_count;
else
ndev->b2b_idx = b2b_mw_idx;
ntb: Adding split BAR support for Haswell platforms On the Haswell platform, a split BAR option to allow creation of 2 32bit BARs (4 and 5) from the 64bit BAR 4. Adding support for this new option. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:23 +07:00
NTB: Improve index handling in B2B MW workaround Check that b2b_mw_idx is in range of the number of memory windows when initializing the device. The workaround is considered to be in effect only if the device b2b_idx is exactly UINT_MAX, instead of any index past the last memory window. Only print B2B MW workaround information in debugfs if the workaround is in effect. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-08-31 20:30:59 +07:00
if (ndev->b2b_idx >= ndev->mw_count) {
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(dev,
NTB: Improve index handling in B2B MW workaround Check that b2b_mw_idx is in range of the number of memory windows when initializing the device. The workaround is considered to be in effect only if the device b2b_idx is exactly UINT_MAX, instead of any index past the last memory window. Only print B2B MW workaround information in debugfs if the workaround is in effect. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-08-31 20:30:59 +07:00
"b2b_mw_idx %d invalid for mw_count %u\n",
b2b_mw_idx, ndev->mw_count);
return -EINVAL;
}
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(dev, "setting up b2b mw idx %d means %d\n",
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
b2b_mw_idx, ndev->b2b_idx);
} else if (ndev->hwerr_flags & NTB_HWERR_B2BDOORBELL_BIT14) {
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_warn(dev, "Reduce doorbell count by 1\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->db_count -= 1;
}
if (ndev->ntb.topo == NTB_TOPO_B2B_USD) {
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
rc = xeon_setup_b2b_mw(ndev,
&xeon_b2b_dsd_addr,
&xeon_b2b_usd_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
} else {
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
rc = xeon_setup_b2b_mw(ndev,
&xeon_b2b_usd_addr,
&xeon_b2b_dsd_addr);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
if (rc)
return rc;
/* Enable Bus Master and Memory Space on the secondary side */
iowrite16(PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER,
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
ndev->self_mmio + XEON_SPCICMD_OFFSET);
ntb: Adding split BAR support for Haswell platforms On the Haswell platform, a split BAR option to allow creation of 2 32bit BARs (4 and 5) from the 64bit BAR 4. Adding support for this new option. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:23 +07:00
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
break;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
default:
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return -EINVAL;
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->db_valid_mask = BIT_ULL(ndev->db_count) - 1;
ndev->reg->db_iowrite(ndev->db_valid_mask,
ndev->self_mmio +
ndev->self_reg->db_mask);
ntb: Adding split BAR support for Haswell platforms On the Haswell platform, a split BAR option to allow creation of 2 32bit BARs (4 and 5) from the 64bit BAR 4. Adding support for this new option. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:23 +07:00
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
return 0;
}
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static int xeon_init_dev(struct intel_ntb_dev *ndev)
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct pci_dev *pdev;
u8 ppd;
int rc, mem;
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
pdev = ndev->ntb.pdev;
NTB: Check the device ID to set errata flags Set errata flags for the specific device IDs to which they apply, instead of the whole Xeon hardware class. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-08 23:24:40 +07:00
switch (pdev->device) {
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* There is a Xeon hardware errata related to writes to SDOORBELL or
* B2BDOORBELL in conjunction with inbound access to NTB MMIO Space,
* which may hang the system. To workaround this use the second memory
* window to access the interrupt and scratch pad registers on the
* remote system.
*/
NTB: Check the device ID to set errata flags Set errata flags for the specific device IDs to which they apply, instead of the whole Xeon hardware class. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-08 23:24:40 +07:00
case PCI_DEVICE_ID_INTEL_NTB_SS_JSF:
case PCI_DEVICE_ID_INTEL_NTB_PS_JSF:
case PCI_DEVICE_ID_INTEL_NTB_B2B_JSF:
case PCI_DEVICE_ID_INTEL_NTB_SS_SNB:
case PCI_DEVICE_ID_INTEL_NTB_PS_SNB:
case PCI_DEVICE_ID_INTEL_NTB_B2B_SNB:
case PCI_DEVICE_ID_INTEL_NTB_SS_IVT:
case PCI_DEVICE_ID_INTEL_NTB_PS_IVT:
case PCI_DEVICE_ID_INTEL_NTB_B2B_IVT:
case PCI_DEVICE_ID_INTEL_NTB_SS_HSX:
case PCI_DEVICE_ID_INTEL_NTB_PS_HSX:
case PCI_DEVICE_ID_INTEL_NTB_B2B_HSX:
NTB: Add PCI Device IDs for Broadwell Xeon Adding PCI Device IDs for B2B (back to back), RP (root port, primary), and TB (transparent bridge, secondary) devices. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-07-13 19:07:18 +07:00
case PCI_DEVICE_ID_INTEL_NTB_SS_BDX:
case PCI_DEVICE_ID_INTEL_NTB_PS_BDX:
case PCI_DEVICE_ID_INTEL_NTB_B2B_BDX:
NTB: Check the device ID to set errata flags Set errata flags for the specific device IDs to which they apply, instead of the whole Xeon hardware class. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-08 23:24:40 +07:00
ndev->hwerr_flags |= NTB_HWERR_SDOORBELL_LOCKUP;
break;
}
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
NTB: Check the device ID to set errata flags Set errata flags for the specific device IDs to which they apply, instead of the whole Xeon hardware class. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-08 23:24:40 +07:00
switch (pdev->device) {
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* There is a hardware errata related to accessing any register in
* SB01BASE in the presence of bidirectional traffic crossing the NTB.
*/
NTB: Check the device ID to set errata flags Set errata flags for the specific device IDs to which they apply, instead of the whole Xeon hardware class. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-08 23:24:40 +07:00
case PCI_DEVICE_ID_INTEL_NTB_SS_IVT:
case PCI_DEVICE_ID_INTEL_NTB_PS_IVT:
case PCI_DEVICE_ID_INTEL_NTB_B2B_IVT:
case PCI_DEVICE_ID_INTEL_NTB_SS_HSX:
case PCI_DEVICE_ID_INTEL_NTB_PS_HSX:
case PCI_DEVICE_ID_INTEL_NTB_B2B_HSX:
NTB: Add PCI Device IDs for Broadwell Xeon Adding PCI Device IDs for B2B (back to back), RP (root port, primary), and TB (transparent bridge, secondary) devices. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-07-13 19:07:18 +07:00
case PCI_DEVICE_ID_INTEL_NTB_SS_BDX:
case PCI_DEVICE_ID_INTEL_NTB_PS_BDX:
case PCI_DEVICE_ID_INTEL_NTB_B2B_BDX:
NTB: Check the device ID to set errata flags Set errata flags for the specific device IDs to which they apply, instead of the whole Xeon hardware class. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-08 23:24:40 +07:00
ndev->hwerr_flags |= NTB_HWERR_SB01BASE_LOCKUP;
break;
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
NTB: Check the device ID to set errata flags Set errata flags for the specific device IDs to which they apply, instead of the whole Xeon hardware class. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-08 23:24:40 +07:00
switch (pdev->device) {
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* HW Errata on bit 14 of b2bdoorbell register. Writes will not be
* mirrored to the remote system. Shrink the number of bits by one,
* since bit 14 is the last bit.
*/
NTB: Check the device ID to set errata flags Set errata flags for the specific device IDs to which they apply, instead of the whole Xeon hardware class. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-08 23:24:40 +07:00
case PCI_DEVICE_ID_INTEL_NTB_SS_JSF:
case PCI_DEVICE_ID_INTEL_NTB_PS_JSF:
case PCI_DEVICE_ID_INTEL_NTB_B2B_JSF:
case PCI_DEVICE_ID_INTEL_NTB_SS_SNB:
case PCI_DEVICE_ID_INTEL_NTB_PS_SNB:
case PCI_DEVICE_ID_INTEL_NTB_B2B_SNB:
case PCI_DEVICE_ID_INTEL_NTB_SS_IVT:
case PCI_DEVICE_ID_INTEL_NTB_PS_IVT:
case PCI_DEVICE_ID_INTEL_NTB_B2B_IVT:
case PCI_DEVICE_ID_INTEL_NTB_SS_HSX:
case PCI_DEVICE_ID_INTEL_NTB_PS_HSX:
case PCI_DEVICE_ID_INTEL_NTB_B2B_HSX:
NTB: Add PCI Device IDs for Broadwell Xeon Adding PCI Device IDs for B2B (back to back), RP (root port, primary), and TB (transparent bridge, secondary) devices. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-07-13 19:07:18 +07:00
case PCI_DEVICE_ID_INTEL_NTB_SS_BDX:
case PCI_DEVICE_ID_INTEL_NTB_PS_BDX:
case PCI_DEVICE_ID_INTEL_NTB_B2B_BDX:
NTB: Check the device ID to set errata flags Set errata flags for the specific device IDs to which they apply, instead of the whole Xeon hardware class. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-08 23:24:40 +07:00
ndev->hwerr_flags |= NTB_HWERR_B2BDOORBELL_BIT14;
break;
}
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
ndev->reg = &xeon_reg;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
rc = pci_read_config_byte(pdev, XEON_PPD_OFFSET, &ppd);
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
if (rc)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
return -EIO;
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
ndev->ntb.topo = xeon_ppd_topo(ndev, ppd);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "ppd %#x topo %s\n", ppd,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ntb_topo_string(ndev->ntb.topo));
if (ndev->ntb.topo == NTB_TOPO_NONE)
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
return -EINVAL;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (ndev->ntb.topo != NTB_TOPO_SEC) {
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
ndev->bar4_split = xeon_ppd_bar4_split(ndev, ppd);
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "ppd %#x bar4_split %d\n",
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ppd, ndev->bar4_split);
} else {
/* This is a way for transparent BAR to figure out if we are
* doing split BAR or not. There is no way for the hw on the
* transparent side to know and set the PPD.
*/
mem = pci_select_bars(pdev, IORESOURCE_MEM);
ndev->bar4_split = hweight32(mem) ==
HSX_SPLIT_BAR_MW_COUNT + 1;
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_dbg(&pdev->dev, "mem %#x bar4_split %d\n",
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
mem, ndev->bar4_split);
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
}
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
rc = xeon_init_ntb(ndev);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (rc)
return rc;
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
return xeon_init_isr(ndev);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static void xeon_deinit_dev(struct intel_ntb_dev *ndev)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
xeon_deinit_isr(ndev);
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static int intel_ntb_init_pci(struct intel_ntb_dev *ndev, struct pci_dev *pdev)
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
{
int rc;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
pci_set_drvdata(pdev, ndev);
rc = pci_enable_device(pdev);
if (rc)
goto err_pci_enable;
rc = pci_request_regions(pdev, NTB_NAME);
if (rc)
goto err_pci_regions;
pci_set_master(pdev);
rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(64));
if (rc) {
rc = pci_set_dma_mask(pdev, DMA_BIT_MASK(32));
if (rc)
goto err_dma_mask;
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_warn(&pdev->dev, "Cannot DMA highmem\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
rc = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64));
if (rc) {
rc = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(32));
if (rc)
goto err_dma_mask;
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
dev_warn(&pdev->dev, "Cannot DMA consistent highmem\n");
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
}
NTB: Set dma mask and dma coherent mask to NTB devices The dma_mask and dma_coherent_mask fields of the NTB struct device weren't initialized in hardware drivers. In fact it should be done instead of PCIe interface usage, since NTB clients are supposed to use NTB API and left unaware of real hardware implementation. In addition to that ntb_device_register() method shouldn't clear the passed ntb_dev structure, since it dma_mask is initialized by hardware drivers. Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-12-06 21:31:53 +07:00
rc = dma_coerce_mask_and_coherent(&ndev->ntb.dev,
dma_get_mask(&pdev->dev));
if (rc)
goto err_dma_mask;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->self_mmio = pci_iomap(pdev, 0, 0);
if (!ndev->self_mmio) {
rc = -EIO;
goto err_mmio;
}
ndev->peer_mmio = ndev->self_mmio;
ntb: ntb_hw_intel: init peer_addr in struct intel_ntb_dev The peer_addr member of intel_ntb_dev is not set, therefore when acquiring ntb_peer_db and ntb_peer_spad we only get the offset rather than the actual physical address. Adding fix to correct that. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-10-28 01:06:44 +07:00
ndev->peer_addr = pci_resource_start(pdev, 0);
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
return 0;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
err_mmio:
err_dma_mask:
pci_clear_master(pdev);
pci_release_regions(pdev);
err_pci_regions:
pci_disable_device(pdev);
err_pci_enable:
pci_set_drvdata(pdev, NULL);
return rc;
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static void intel_ntb_deinit_pci(struct intel_ntb_dev *ndev)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
ntb_hw_intel: Style fixes: open code macros that just obfuscate code As per a comments in [1] by Greg Kroah-Hartman, the ndev_* macros should be cleaned up. This makes it more clear what's actually going on when reading the code. [1] http://www.spinics.net/lists/linux-pci/msg56904.html Signed-off-by: Logan Gunthorpe <logang@deltatee.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:33:37 +07:00
struct pci_dev *pdev = ndev->ntb.pdev;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (ndev->peer_mmio && ndev->peer_mmio != ndev->self_mmio)
pci_iounmap(pdev, ndev->peer_mmio);
pci_iounmap(pdev, ndev->self_mmio);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
pci_clear_master(pdev);
pci_release_regions(pdev);
pci_disable_device(pdev);
pci_set_drvdata(pdev, NULL);
}
ntb: use errata flag set via DID to implement workaround Instead of using a module parameter, we should detect the errata via PCI DID and then set an appropriate flag. This will be used for additional errata later on. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:18 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static inline void ndev_init_struct(struct intel_ntb_dev *ndev,
struct pci_dev *pdev)
{
ndev->ntb.pdev = pdev;
ndev->ntb.topo = NTB_TOPO_NONE;
ndev->ntb.ops = &intel_ntb_ops;
ntb: use errata flag set via DID to implement workaround Instead of using a module parameter, we should detect the errata via PCI DID and then set an appropriate flag. This will be used for additional errata later on. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:18 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->b2b_off = 0;
NTB: Improve index handling in B2B MW workaround Check that b2b_mw_idx is in range of the number of memory windows when initializing the device. The workaround is considered to be in effect only if the device b2b_idx is exactly UINT_MAX, instead of any index past the last memory window. Only print B2B MW workaround information in debugfs if the workaround is in effect. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-08-31 20:30:59 +07:00
ndev->b2b_idx = UINT_MAX;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->bar4_split = 0;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->mw_count = 0;
ndev->spad_count = 0;
ndev->db_count = 0;
ndev->db_vec_count = 0;
ndev->db_vec_shift = 0;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->ntb_ctl = 0;
ndev->lnk_sta = 0;
ntb: conslidate reading of PPD to move platform detection earlier To simplify some of the platform detection code. Move the platform detection to a function to be called earlier. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:13 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->db_valid_mask = 0;
ndev->db_link_mask = 0;
ndev->db_mask = 0;
ntb: Adding split BAR support for Haswell platforms On the Haswell platform, a split BAR option to allow creation of 2 32bit BARs (4 and 5) from the 64bit BAR 4. Adding support for this new option. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:23 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
spin_lock_init(&ndev->db_mask_lock);
}
ntb: Adding split BAR support for Haswell platforms On the Haswell platform, a split BAR option to allow creation of 2 32bit BARs (4 and 5) from the 64bit BAR 4. Adding support for this new option. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2014-08-29 03:53:23 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static int intel_ntb_pci_probe(struct pci_dev *pdev,
const struct pci_device_id *id)
{
struct intel_ntb_dev *ndev;
NTB: Use NUMA memory in Intel driver Allocate memory for the NUMA node of the NTB device. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-19 23:04:52 +07:00
int rc, node;
node = dev_to_node(&pdev->dev);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb: remove Intel Atom NTB driver support Removing dead code since this is not being used. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-11-21 00:24:08 +07:00
if (pdev_is_xeon(pdev)) {
NTB: Use NUMA memory in Intel driver Allocate memory for the NUMA node of the NTB device. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-19 23:04:52 +07:00
ndev = kzalloc_node(sizeof(*ndev), GFP_KERNEL, node);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (!ndev) {
rc = -ENOMEM;
goto err_ndev;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev_init_struct(ndev, pdev);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
rc = intel_ntb_init_pci(ndev, pdev);
if (rc)
goto err_init_pci;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
rc = xeon_init_dev(ndev);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
if (rc)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
goto err_init_dev;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
ntb: Adding Skylake Xeon NTB support The Skylake Xeon NTB hardware has made some changes to the register name, offset, and the way doorbells work. Adding driver support for the new hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-11-17 04:03:38 +07:00
} else if (pdev_is_skx_xeon(pdev)) {
ndev = kzalloc_node(sizeof(*ndev), GFP_KERNEL, node);
if (!ndev) {
rc = -ENOMEM;
goto err_ndev;
}
ndev_init_struct(ndev, pdev);
ndev->ntb.ops = &intel_ntb3_ops;
rc = intel_ntb_init_pci(ndev, pdev);
if (rc)
goto err_init_pci;
rc = skx_init_dev(ndev);
if (rc)
goto err_init_dev;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
} else {
rc = -EINVAL;
goto err_ndev;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev_reset_unsafe_flags(ndev);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev->reg->poll_link(ndev);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
ndev_init_debugfs(ndev);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
rc = ntb_register_device(&ndev->ntb);
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
if (rc)
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
goto err_register;
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Print driver name and version in module init Printouts driver name and version to indicate what is being loaded. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-15 19:21:33 +07:00
dev_info(&pdev->dev, "NTB device registered.\n");
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
return 0;
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
err_register:
ndev_deinit_debugfs(ndev);
ntb: remove Intel Atom NTB driver support Removing dead code since this is not being used. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-11-21 00:24:08 +07:00
if (pdev_is_xeon(pdev) || pdev_is_skx_xeon(pdev))
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
xeon_deinit_dev(ndev);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
err_init_dev:
intel_ntb_deinit_pci(ndev);
err_init_pci:
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
kfree(ndev);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
err_ndev:
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
return rc;
}
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static void intel_ntb_pci_remove(struct pci_dev *pdev)
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
{
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
struct intel_ntb_dev *ndev = pci_get_drvdata(pdev);
ntb_unregister_device(&ndev->ntb);
ndev_deinit_debugfs(ndev);
ntb: remove Intel Atom NTB driver support Removing dead code since this is not being used. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-11-21 00:24:08 +07:00
if (pdev_is_xeon(pdev) || pdev_is_skx_xeon(pdev))
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
xeon_deinit_dev(ndev);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
intel_ntb_deinit_pci(ndev);
kfree(ndev);
}
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static const struct intel_ntb_reg xeon_reg = {
.poll_link = xeon_poll_link,
.link_is_up = xeon_link_is_up,
.db_ioread = xeon_db_ioread,
.db_iowrite = xeon_db_iowrite,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
.db_size = sizeof(u32),
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
.ntb_ctl = XEON_NTBCNTL_OFFSET,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
.mw_bar = {2, 4, 5},
};
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static const struct intel_ntb_alt_reg xeon_pri_reg = {
.db_bell = XEON_PDOORBELL_OFFSET,
.db_mask = XEON_PDBMSK_OFFSET,
.spad = XEON_SPAD_OFFSET,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
};
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static const struct intel_ntb_alt_reg xeon_sec_reg = {
.db_bell = XEON_SDOORBELL_OFFSET,
.db_mask = XEON_SDBMSK_OFFSET,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* second half of the scratchpads */
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
.spad = XEON_SPAD_OFFSET + (XEON_SPAD_COUNT << 1),
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
};
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static const struct intel_ntb_alt_reg xeon_b2b_reg = {
.db_bell = XEON_B2B_DOORBELL_OFFSET,
.spad = XEON_B2B_SPAD_OFFSET,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
};
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static const struct intel_ntb_xlat_reg xeon_pri_xlat = {
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
/* Note: no primary .bar0_base visible to the secondary side.
*
* The secondary side cannot get the base address stored in primary
* bars. The base address is necessary to set the limit register to
* any value other than zero, or unlimited.
*
* WITHOUT THE BASE ADDRESS, THE SECONDARY SIDE CANNOT DISABLE the
* window by setting the limit equal to base, nor can it limit the size
* of the memory window by setting the limit to base + size.
*/
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
.bar2_limit = XEON_PBAR23LMT_OFFSET,
.bar2_xlat = XEON_PBAR23XLAT_OFFSET,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
};
NTB: Rename Intel code names to platform names Instead of using the platform code names, use the correct platform names to identify the respective Intel NTB hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-05-20 23:55:47 +07:00
static const struct intel_ntb_xlat_reg xeon_sec_xlat = {
.bar0_base = XEON_SBAR0BASE_OFFSET,
.bar2_limit = XEON_SBAR23LMT_OFFSET,
.bar2_xlat = XEON_SBAR23XLAT_OFFSET,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
};
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
struct intel_b2b_addr xeon_b2b_usd_addr = {
NTB: unify translation addresses There is no need for the upstream and downstream addresses to be different for the NTB configs. Go to using a single set of address. It is still possible to configure them differently using module parameter override however. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked and Tested-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-09-25 03:03:05 +07:00
.bar2_addr64 = XEON_B2B_BAR2_ADDR64,
.bar4_addr64 = XEON_B2B_BAR4_ADDR64,
.bar4_addr32 = XEON_B2B_BAR4_ADDR32,
.bar5_addr32 = XEON_B2B_BAR5_ADDR32,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
};
ntb: intel: split out the gen3 code Move the Intel hw gen3 code to its own source file. The ntb_hw_intel.c was getting too large and makes it hard to maintain with future hardware changes. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2018-01-30 03:22:24 +07:00
struct intel_b2b_addr xeon_b2b_dsd_addr = {
NTB: unify translation addresses There is no need for the upstream and downstream addresses to be different for the NTB configs. Go to using a single set of address. It is still possible to configure them differently using module parameter override however. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked and Tested-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-09-25 03:03:05 +07:00
.bar2_addr64 = XEON_B2B_BAR2_ADDR64,
.bar4_addr64 = XEON_B2B_BAR4_ADDR64,
.bar4_addr32 = XEON_B2B_BAR4_ADDR32,
.bar5_addr32 = XEON_B2B_BAR5_ADDR32,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
};
/* operations for primary side of local ntb */
static const struct ntb_dev_ops intel_ntb_ops = {
.mw_count = intel_ntb_mw_count,
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
.mw_get_align = intel_ntb_mw_get_align,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
.mw_set_trans = intel_ntb_mw_set_trans,
NTB: Alter MW API to support multi-ports devices Multi-port NTB devices permit to share a memory between all accessible peers. Memory Windows API is altered to correspondingly initialize and map memory windows for such devices: ntb_mw_count(pidx); - number of inbound memory windows, which can be allocated for shared buffer with specified peer device. ntb_mw_get_align(pidx, widx); - get alignment and size restriction parameters to properly allocate inbound memory region. ntb_peer_mw_count(); - get number of outbound memory windows. ntb_peer_mw_get_addr(widx); - get mapping address of an outbound memory window If hardware supports inbound translation configured on the local ntb port: ntb_mw_set_trans(pidx, widx); - set translation address of allocated inbound memory window so a peer device could access it. ntb_mw_clear_trans(pidx, widx); - clear the translation address of an inbound memory window. If hardware supports outbound translation configured on the peer ntb port: ntb_peer_mw_set_trans(pidx, widx); - set translation address of a memory window retrieved from a peer device ntb_peer_mw_clear_trans(pidx, widx); - clear the translation address of an outbound memory window Signed-off-by: Serge Semin <fancer.lancer@gmail.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2017-01-11 07:11:33 +07:00
.peer_mw_count = intel_ntb_peer_mw_count,
.peer_mw_get_addr = intel_ntb_peer_mw_get_addr,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
.link_is_up = intel_ntb_link_is_up,
.link_enable = intel_ntb_link_enable,
.link_disable = intel_ntb_link_disable,
.db_is_unsafe = intel_ntb_db_is_unsafe,
.db_valid_mask = intel_ntb_db_valid_mask,
.db_vector_count = intel_ntb_db_vector_count,
.db_vector_mask = intel_ntb_db_vector_mask,
.db_read = intel_ntb_db_read,
.db_clear = intel_ntb_db_clear,
.db_set_mask = intel_ntb_db_set_mask,
.db_clear_mask = intel_ntb_db_clear_mask,
.peer_db_addr = intel_ntb_peer_db_addr,
.peer_db_set = intel_ntb_peer_db_set,
.spad_is_unsafe = intel_ntb_spad_is_unsafe,
ntb: Adding Skylake Xeon NTB support The Skylake Xeon NTB hardware has made some changes to the register name, offset, and the way doorbells work. Adding driver support for the new hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-11-17 04:03:38 +07:00
.spad_count = intel_ntb_spad_count,
.spad_read = intel_ntb_spad_read,
.spad_write = intel_ntb_spad_write,
.peer_spad_addr = intel_ntb_peer_spad_addr,
.peer_spad_read = intel_ntb_peer_spad_read,
.peer_spad_write = intel_ntb_peer_spad_write,
};
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static const struct file_operations intel_ntb_debugfs_info = {
.owner = THIS_MODULE,
.open = simple_open,
.read = ndev_debugfs_read,
};
static const struct pci_device_id intel_ntb_pci_tbl[] = {
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_B2B_JSF)},
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_B2B_SNB)},
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_B2B_IVT)},
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_B2B_HSX)},
NTB: Add PCI Device IDs for Broadwell Xeon Adding PCI Device IDs for B2B (back to back), RP (root port, primary), and TB (transparent bridge, secondary) devices. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-07-13 19:07:18 +07:00
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_B2B_BDX)},
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_PS_JSF)},
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_PS_SNB)},
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_PS_IVT)},
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_PS_HSX)},
NTB: Add PCI Device IDs for Broadwell Xeon Adding PCI Device IDs for B2B (back to back), RP (root port, primary), and TB (transparent bridge, secondary) devices. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-07-13 19:07:18 +07:00
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_PS_BDX)},
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_SS_JSF)},
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_SS_SNB)},
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_SS_IVT)},
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_SS_HSX)},
NTB: Add PCI Device IDs for Broadwell Xeon Adding PCI Device IDs for B2B (back to back), RP (root port, primary), and TB (transparent bridge, secondary) devices. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-07-13 19:07:18 +07:00
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_SS_BDX)},
ntb: Adding Skylake Xeon NTB support The Skylake Xeon NTB hardware has made some changes to the register name, offset, and the way doorbells work. Adding driver support for the new hardware. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Acked-by: Allen Hubbe <Allen.Hubbe@dell.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2016-11-17 04:03:38 +07:00
{PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_NTB_B2B_SKX)},
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
{0}
};
MODULE_DEVICE_TABLE(pci, intel_ntb_pci_tbl);
static struct pci_driver intel_ntb_pci_driver = {
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
.name = KBUILD_MODNAME,
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
.id_table = intel_ntb_pci_tbl,
.probe = intel_ntb_pci_probe,
.remove = intel_ntb_pci_remove,
PCI-Express Non-Transparent Bridge Support A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus connecting 2 systems, providing electrical isolation between the two subsystems. A non-transparent bridge is functionally similar to a transparent bridge except that both sides of the bridge have their own independent address domains. The host on one side of the bridge will not have the visibility of the complete memory or I/O space on the other side of the bridge. To communicate across the non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to the local system. Writes to these apertures are mirrored to memory on the remote system. Communications can also occur through the use of doorbell registers that initiate interrupts to the alternate domain, and scratch-pad registers accessible from both sides. The NTB device driver is needed to configure these memory windows, doorbell, and scratch-pad registers as well as use them in such a way as they can be turned into a viable communication channel to the remote system. ntb_hw.[ch] determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away the underlying hardware to provide access and a common interface to the doorbell registers, scratch pads, and memory windows. These hardware interfaces are exported so that other, non-mainlined kernel drivers can access these. ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a communication channel(s) and provide a reliable way of transferring data from one side to the other, which it then exports so that "client" drivers can access them. These client drivers are used to provide a standard kernel interface (i.e., Ethernet device) to NTB, such that Linux can transfer data from one system to the other in a standard way. Signed-off-by: Jon Mason <jon.mason@intel.com> Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-11-17 09:27:12 +07:00
};
NTB: debugfs device entry Create a debugfs entry for the NTB device to log the basic device info, as well as display the error count on a number of registers. Signed-off-by: Jon Mason <jon.mason@intel.com>
2014-04-08 00:55:47 +07:00
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
static int __init intel_ntb_pci_driver_init(void)
{
NTB: Print driver name and version in module init Printouts driver name and version to indicate what is being loaded. Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-06-15 19:21:33 +07:00
pr_info("%s %s\n", NTB_DESC, NTB_VER);
NTB: Split ntb_hw_intel and ntb_transport drivers Change ntb_hw_intel to use the new NTB hardware abstraction layer. Split ntb_transport into its own driver. Change it to use the new NTB hardware abstraction layer. Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
2015-04-09 21:33:20 +07:00
if (debugfs_initialized())
debugfs_dir = debugfs_create_dir(KBUILD_MODNAME, NULL);
return pci_register_driver(&intel_ntb_pci_driver);
}
module_init(intel_ntb_pci_driver_init);
static void __exit intel_ntb_pci_driver_exit(void)
{
pci_unregister_driver(&intel_ntb_pci_driver);
debugfs_remove_recursive(debugfs_dir);
}
module_exit(intel_ntb_pci_driver_exit);
Reference in New Issue Copy Permalink