This patch improves memory utilization and therefore the packets rate
for special MTU's. Instead of setting the frag_stride to the maximal
hard coded frag_size, use the actual frag_size that is set according to
the MTU, when setting the stride of the last frag.
So, for example, for MTU 1600, where the frag_size of the 2nd frag is
86, the frag_size is set to 128 instead of 4096. See below:
Before:
frag:0 - size:1536 prefix:0 stride:1536
frag:1 - size:86 prefix:1536 stride:4096
frag 0 allocator: - size:32768 frags:21
frag 1 allocator: - size:32768 frags:8
After:
frag:0 - size:1536 prefix:0 stride:1536
frag:1 - size:86 prefix:1536 stride:128
frag 0 allocator: - size:32768 frags:21
frag 1 allocator: - size:32768 frags:256
Signed-off-by: Ido Shamay <idos@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After Initialization of page_alloc, print actual allocated page
size and number of frags it contains. prints is done only when drv
message level is set on the interface.
Signed-off-by: Ido Shamay <idos@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We do support cache line sizes of 32 and 64 bytes without activating the
CQE stride feature. Fix a misleading print saying that these cache line
sizes aren't supported.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add Initialization to struct config_dev before filling and using it.
Fix to warning:
warning: config_dev.rx_checksum_val may be used uninitialized in this function
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
a) Previously, mlx4_mr_rereg_write filled the MPT's start
and length with the old MPT's values.
Fixing the initialization to take the new start and length.
b) In addition access flags in mpt_status were initialized instead of
status due to bad boolean operation. Fixing the operation.
c) Initialization of pd_slave caused a protection error.
Fix - removing this initialization.
d) In resource_tracker.c: Fixing vf encoding to be one-based.
Fixes: e630664c ('mlx4_core: Add helper functions to support MR re-registration')
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Capture NETDEV events generated by the bonding driver and based on that
make decisions of how to configure port aggregation in the mlx4 core driver.
This includes setting the V2P port table and re-creating the interested
interfaces in bonded/non-bonded mode.
Signed-off-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Supply interface functions to bond and unbond ports of a mlx4 internal
interfaces. Example for such an interface is the one registered by the
mlx4 IB driver under RoCE.
There are
1. Functions to go in/out to/from bonded mode
2. Function to remap virtual ports to physical ports
The bond_mutex prevents simultaneous access to data that keep status of
the device in bonded mode.
The upper mlx4 interface marks to the mlx4 core module that they
want to be subject for such bonding by setting the MLX4_INTFF_BONDING
flag. Interface which goes to/from bonded mode is re-created.
The mlx4 Ethernet driver does not set this flag when registering the
interface, the IB driver does.
Signed-off-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implement the hardware interface required for port aggregation.
1. Disable RX port check on receive - don't perform a validity check
that matches to QP's port and the port where the packet is received.
2. Virtual to physical port remap - configure virtual to physical port
mapping. Port remap capability for virtual functions.
Signed-off-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit de966c5928 (net/mlx4_core: Support more than 64 VFs) was meant to
allow up to 126 VFs. However, due to leaving MLX4_MFUNC_MAX too low, using
more than 80 VFs resulted in memory corruptions (and Oopses) when more than
80 VFs were requested. In addition, the number of slaves was left too high.
This commit fixes these issues.
Fixes: de966c5928 ("net/mlx4_core: Support more than 64 VFs")
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The firmware might change the hca core clock frequency after the driver
issues the INIT_PORT command. Therefore we need to query the new
value again and save in to the cached dev caps.
Fixes: ddd8a6c1 ('net/mlx4_core: Read HCA frequency and map internal clock')
Signed-off-by: Majd Dibbiny <majd@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We are dumping device capabilities which are supported both by the
firmware and the driver. Align the array that holds the capability
strings with this practice.
Reported-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix a memory corruption at mlx4_MAD_IFC_wrapper.
A table of size dev->caps.pkey_table_len[port]*sizeof(*table)
was allocated, but get_full_pkey_table() assumes that the number
of entries in the table is a multiplication of 32 (which isn't always
correct).
Fixes: 0a9a018 ('mlx4: MAD_IFC paravirtualization')
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use cmd->autoneg as a user hint to decide what to set in ethtool set settings callback.
When cmd->autoneg == AUTONEG_ENABLE set according to ethtool->advertise otherwise,
set according to ethtool->speed.
Usage:
- ethtool -s eth<x> speed 56000 autoneg off
- ethtool -s eth<x> advertise 0x800000 autoneg on
While we're here:
- Move proto_admin masking outcome check to be adjacent to the operation.
- Move en_warn("port reset..") print to "port reset" block.
Fixes: 312df74 ("net/mlx4_en: mlx4_en_set_settings() always fails when autoneg is set")
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
mlx4_bf_alloc had an unnecessary/duplicate code line. Did no harm,
but not good practice.
Reported by the Mellanox Beijing team.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Struct mlx4_vhcr was implicitly padded by the gcc compiler on 64-bit
architectures.
This commit makes that padding explicit, to prevent issues with
changing compilers and with incompatibilities between 32-bit architecture
implicit padding and 64-bit architecture implicit padding.
This structure is used in virtualization for communication between
the Host and its Guests. The explicit padding allows 64-bit Hosts
(old and new) to continue to interoperate with 64-bit Guests (old and new).
However, without this fix, 64-bit Hosts could not interoperate with 32-bit
Guests (since these did not insert the padding dword). With this fix,
32-bit Guests will be able to interoperate with 64-bit Hosts (since
the structure offsets will be identical on both).
Reported-by: Alexander Schmidt <alexs@linux.vnet.ibm.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The driver incorrectly assigned an out-mailbox to this command,
and used an opcode modifier = 0, which is a reserved value (it
should use opcode modifier = 1).
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The firmware spec states that the timeout for all commands should be 60 seconds.
In the past, the spec indicated that there were several classes of timeout
(short, medium, and long). The driver has these different timeout classes.
We leave the class differentiation in the driver as-is (to protect against any
future spec changes), but set the timeout for all classes to be 60 seconds.
In addition, we fix a few commands which had hard-coded numeric timeouts specified.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Structs allocated for the resource tracker must be freed in
the error flow.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The reserved lKey is different for each VF.
A base lkey value is returned in QUERY_DEV_CAP at offset 0x98.
The reserved L_key value for a VF is:
VF_lkey = base_lkey + (VF_number << 8).
This VF L_key value should be returned in QUERY_FUNC_CAP
(opcode-modifier = 0) at offset 0x48.
To indicate that the lkey value at offset 0x48 is valid, the Hypervisor
sets a flag bit in dword 0x0, offset 27 in the QUERY_FUNC_CAP wrapper
function.
When the VF calls QUERY_FUNC_CAP, it should check if this flag bit is set.
If it is set, the VF should take the reserved lkey value at offset 0x48.
If the bit is not set, the VF should not use a reserved lkey
(i.e., should set its reserved lkey value to 0).
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If the firmware can detect a bad cable, allow it to generate an
event, and print the problem in the log.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
arch/arm/boot/dts/imx6sx-sdb.dts
net/sched/cls_bpf.c
Two simple sets of overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
When SRIOV commands are executed over the comm-channel and get
a fatal error (e.g. timeout, closing command failure) the VF enters
into error state and reset flow is activated.
To be able to recognize whether the failure was on a closing command, the
operational code for the given VHCR command is used. Once the device entered
into an error state we prevent redundant error messages from being printed.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In SRIOV, both the PF and the VF may attempt device recovery whenever they
assume that the device is not functioning. When the PF driver resets the
device, the VF should detect this and attempt to reinitialize itself.
The VF must be able to reset itself under all circumstances, even
if the PF is not responsive.
The VF shall reset itself in the following cases:
1. Commands are not processed within reasonable time over the communication channel.
This is done considering device state and the correct return code based on
the command as was done in the native mode, done in the next patch.
2. The VF driver receives an internal error event reported by the PF on the
communication channel. This occurs when the PF driver resets the device or
when VF is out of sync with the PF.
Add 'VF reset' capability, which allows the VF to reinitialize itself even when the
PF is not responsive.
As PF and VF may run their reset flow simulantanisly, there are several cases
that are handled:
- Prevent freeing VF resources upon FLR, when PF is in its unloading stage.
- Prevent PF getting VF commands before it has finished initializing its resources.
- Upon VF startup, check that comm-channel is online before sending
commands to the PF and getting timed-out.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix AER callbacks to work properly, it includes:
- Refractoring AER to be aligned with Reset flow support.
- Sync with concurrent catas flow.
In addition, fix the shutdown PCI callback to sync with
concurrent catas flow.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We need to manage interface state to sync between reset flow and some other
relative cases such as remove_one. This has to be done to prevent certain
races. For example in case software stack is down as a result of unload call,
the remove_one should skip the unload phase.
Implement the remove_one case, handling AER and other cases comes next.
The interface can be up/down, upon remove_one, the state will include an extra
bit indicating that the device is cleaned-up, forcing other tasks to finish
before the final cleanup.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We activate reset flow upon command fatal errors, when the device enters an
erroneous state, and must be reset.
The cases below are assumed to be fatal: FW command timed-out, an error from FW
on closing commands, pci is offline when posting/pending a command.
In those cases we place the device into an error state: chip is reset, pending
commands are awakened and completed immediately. Subsequent commands will
return immediately.
The return code in the above cases will depend on the command. Commands which
free and close resources will return success (because the chip was reset, so
callers may safely free their kernel resources). Other commands will return -EIO.
Since the device's state was marked as error, the catas poller will
detect this and restart the device's software stack (as is done when a FW
internal error is directly detected). The device state is protected by a
persistent mutex lives on its mlx4_dev, as such no need any more for the
hcr_mutex which is removed.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This includes:
- resetting the chip when a fatal error is detected (the current code
does not do this).
- exposing the ability to enter error state from outside the catas code
by calling its functionality. (E.g. FW Command timeout, AER error).
- managing a persistent device state. This is needed to sync between
reset flow cases.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Using a WQ per device instead of a single global WQ, this allows
independent reset handling per device even when SRIOV is used.
This comes as a pre-patch for supporting chip reset
for both native and SRIOV.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When an HCA enters an internal error state, this is detected by the driver.
The driver then should reset the HCA and restart the software stack.
Keep ports information and some SRIOV configuration in a persistent area
to have it valid across reset.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Maintain a persistent memory that should survive reset flow/PCI error.
This comes as a preparation for coming series to support above flows.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Except for VXLAN steering rules, all offloads should work as they were
under plain DMFS mode. Fix that by enabling all the offloads under
DMFS-A0 mode, except for VXLAN steering rules.
Fixes: d57febe1a4 "net/mlx4: Add A0 hybrid steering"
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The same macros are used for rx as well. So rename it.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
This driver uses the function, clocksource_khz2mult, and so it really must
include clocksource.h.
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We shouldn't call UNMAP_FA here, this is done in mlx4_load_one.
If mlx4_query_func fails, we need to invoke CLOSE_HCA for both
native and master.
Fixes: a0eacca948 ('net/mlx4_core: Refactor mlx4_load_one')
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Previously, mlx4_mt_rereg_write filled the MPT's entity_size with the
old MTT's page shift, which could result in using an incorrect offset.
Fix the initialization to be after we calculate the new MTT offset.
In addition, assign mtt order to -1 after calling mlx4_mtt_cleanup. This
is necessary in order to mark the MTT as invalid and avoid freeing it later.
Fixes: e630664 ('mlx4_core: Add helper functions to support MR re-registration')
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current timecounter implementation will drop a variable amount
of resolution, depending on the magnitude of the time delta. In
other words, reading the clock too often or too close to a time
stamp conversion will introduce errors into the time values. This
patch fixes the issue by introducing a fractional nanosecond field
that accumulates the low order bits.
Reported-by: Janusz Użycki <j.uzycki@elproma.com.pl>
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch changes the driver to use the new and improved method
for adjusting the offset of a timecounter.
Compile tested only.
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
GSO isn't the only offload feature with restrictions that
potentially can't be expressed with the current features mechanism.
Checksum is another although it's a general issue that could in
theory apply to anything. Even if it may be possible to
implement these restrictions in other ways, it can result in
duplicate code or inefficient per-packet behavior.
This generalizes ndo_gso_check so that drivers can remove any
features that don't make sense for a given packet, similar to
netif_skb_features(). It also converts existing driver
restrictions to the new format, completing the work that was
done to support tunnel protocols since the issues apply to
checksums as well.
By actually removing features from the set that are used to do
offloading, it solves another problem with the existing
interface. In these cases, GSO would run with the original set
of features and not do anything because it appears that
segmentation is not required.
CC: Tom Herbert <therbert@google.com>
CC: Joe Stringer <joestringer@nicira.com>
CC: Eric Dumazet <edumazet@google.com>
CC: Hayes Wang <hayeswang@realtek.com>
Signed-off-by: Jesse Gross <jesse@nicira.com>
Acked-by: Tom Herbert <therbert@google.com>
Fixes: 04ffcb255f ("net: Add ndo_gso_check")
Tested-by: Hayes Wang <hayeswang@realtek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
iowrite32() will byteswap it's argument on big endian archs.
iowrite32be() will byteswap on little endian archs.
Since we don't want to do this unnecessary byteswap on the fast path,
doorbell is stored in the NIC's native endianness. Using the right
iowrite() according to the arch endianness.
CC: Wei Yang <weiyang@linux.vnet.ibm.com>
CC: David Laight <david.laight@aculab.com>
Fixes: 6a4e812 ("net/mlx4_en: Avoid calling bswap in tx fast path")
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- On-demand paging support in core midlayer and mlx5 driver. This lets
userspace create non-pinned memory regions and have the adapter HW
trigger page faults.
- iSER and IPoIB updates and fixes.
- Low-level HW driver updates for cxgb4, mlx4 and ocrdma.
- Other miscellaneous fixes.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=MCO9
-----END PGP SIGNATURE-----
Merge tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband
Pull infiniband updates from Roland Dreier:
"Main batch of InfiniBand/RDMA changes for 3.19:
- On-demand paging support in core midlayer and mlx5 driver. This
lets userspace create non-pinned memory regions and have the
adapter HW trigger page faults.
- iSER and IPoIB updates and fixes.
- Low-level HW driver updates for cxgb4, mlx4 and ocrdma.
- Other miscellaneous fixes"
* tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (56 commits)
IB/mlx5: Implement on demand paging by adding support for MMU notifiers
IB/mlx5: Add support for RDMA read/write responder page faults
IB/mlx5: Handle page faults
IB/mlx5: Page faults handling infrastructure
IB/mlx5: Add mlx5_ib_update_mtt to update page tables after creation
IB/mlx5: Changes in memory region creation to support on-demand paging
IB/mlx5: Implement the ODP capability query verb
mlx5_core: Add support for page faults events and low level handling
mlx5_core: Re-add MLX5_DEV_CAP_FLAG_ON_DMND_PG flag
IB/srp: Allow newline separator for connection string
IB/core: Implement support for MMU notifiers regarding on demand paging regions
IB/core: Add support for on demand paging regions
IB/core: Add flags for on demand paging support
IB/core: Add support for extended query device caps
IB/mlx5: Add function to read WQE from user-space
IB/core: Add umem function to read data from user-space
IB/core: Replace ib_umem's offset field with a full address
IB/mlx5: Enhance UMR support to allow partial page table update
IB/mlx5: Remove per-MR pas and dma pointers
RDMA/ocrdma: Always resolve destination mac from GRH for UD QPs
...
This commit contains 2 fixes for the 128B CQE/EQE stride feaure.
Wei found that mlx4_QUERY_HCA function marked the wrong capability
in flags (64B CQE/EQE), when CQE/EQE stride feature was enabled.
Also added small fix in initial CQE ownership bit assignment, when CQE
is size is not default 32B.
Fixes: 77507aa24 (net/mlx4: Enable CQE/EQE stride support)
Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com>
Signed-off-by: Ido Shamay <idos@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move check for DPDP out of the loop to make the code more readable.
Signed-off-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
To support asymmetric EQ allocations, we should query the device
capabilities prior to enabling SRIOV. As a side effect of adding that,
we are dumping the PF device capabilities twice. Avoid that by moving
the printing into a helper function which is called once.
Fixes: 7ae0e400cd ('net/mlx4_core: Flexible (asymmetric) allocation of
EQs and MSI-X vectors for PF/VFs')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current mlx4_load_one has a memory leak as it always allocates
dev_cap, but frees it only on error.
In addition, even if VFs exist when mlx4_load_one is called,
we still need to notify probed VFs that we're loading (by
incrementing pf_loading).
Fixes: a0eacca948 ('net/mlx4_core: Refactor mlx4_load_one')
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add the required firmware commands for A0 steering and a way to enable
that. The firmware support focuses on INIT_HCA, QUERY_HCA, QUERY_PORT,
QUERY_DEV_CAP and QUERY_FUNC_CAP commands. Those commands are used
to configure and query the device.
The different A0 DMFS (steering) modes are:
Static - optimized performance, but flow steering rules are
limited. This mode should be choosed explicitly by the user
in order to be used.
Dynamic - this mode should be explicitly choosed by the user.
In this mode, the FW works in optimized steering mode as long as
it can and afterwards automatically drops to classic (full) DMFS.
Disable - this mode should be explicitly choosed by the user.
The user instructs the system not to use optimized steering, even if
the FW supports Dynamic A0 DMFS (and thus will be able to use optimized
steering in Default A0 DMFS mode).
Default - this mode is implicitly choosed. In this mode, if the FW
supports Dynamic A0 DMFS, it'll work in this mode. Otherwise, it'll
work at Disable A0 DMFS mode.
Under SRIOV configuration, when the A0 steering mode is enabled,
older guest VF drivers who aren't using the RX QP allocation flag
(MLX4_RESERVE_A0_QP) will get a QP from the general range and
fail when attempting to register a steering rule. To avoid that,
the PF context behaviour is changed once on A0 static mode, to
require support for the allocation flag in VF drivers too.
In order to enable A0 steering, we use log_num_mgm_entry_size param.
If the value of the parameter is not positive, we treat the absolute
value of log_num_mgm_entry_size as a bit field. Setting bit 2 of this
bit field enables static A0 steering.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently QUERY_PORT is done as a part of QUERY_DEV_CAP firmware command.
Since we would like to use it without querying all device capabilities,
extract this part to be a function of its own.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a given flow steering rule is invalid in respect to the current
steering configuration, print the correct error message to the system log.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A0 hybrid steering is a form of high performance flow steering.
By using this mode, mlx4 cards use a fast limited table based steering,
in order to enable fast steering of unicast packets to a QP.
In order to implement A0 hybrid steering we allocate resources
from different zones:
(1) General range
(2) Special MAC-assigned QPs [RSS, Raw-Ethernet] each has its own region.
When we create a rss QP or a raw ethernet (A0 steerable and BF ready) QP,
we try hard to allocate the QP from range (2). Otherwise, we try hard not
to allocate from this range. However, when the system is pushed to its
limits and one needs every resource, the allocator uses every region it can.
Meaning, when we run out of raw-eth qps, the allocator allocates from the
general range (and the special-A0 area is no longer active). If we run out
of RSS qps, the mechanism tries to allocate from the raw-eth QP zone. If that
is also exhausted, the allocator will allocate from the general range
(and the A0 region is no longer active).
Note that if a raw-eth qp is allocated from the general range, it attempts
to allocate the range such that bits 6 and 7 (blueflame bits) in the
QP number are not set.
When the feature is used in SRIOV, the VF has to notify the PF what
kind of QP attributes it needs. In order to do that, along with the
"Eth QP blueflame" bit, we reserve a new "A0 steerable QP". According
to the combination of these bits, the PF tries to allocate a suitable QP.
In order to maintain backward compatibility (with older PFs), the PF
notifies which QP attributes it supports via QUERY_FUNC_CAP command.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The zone allocator is a mechanism which manages a few mlx4_bitmaps.
When allocating a resource, the user indicates the desired zone of
which this resource will be allocated from. If possible, the resource
will be allocated from this zone. Otherwise, the resource will be
allocated from a less-than, equal-to, higher-than priority zone,
according to the desired zone's properties with that respective
allocation order.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The number of reserved QPs is affected both from the firmware and
from the driver's requirements. This patch adds a check that
validates that this number is indeed feasable.
Signed-off-by: Dotan Barak <dotanb@dev.mellanox.co.il>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When using BF (Blue-Flame), the QPN overrides the VLAN, CV, and SV fields
in the WQE. Thus, BF may only be used for QPNs with bits 6,7 unset.
The current Ethernet driver code reserves a Tx QP range with 256b alignment.
This is wrong because if there are more than 64 Tx QPs in use,
QPNs >= base + 65 will have bits 6/7 set.
This problem is not specific for the Ethernet driver, any entity that
tries to reserve more than 64 BF-enabled QPs should fail. Also, using
ranges is not necessary here and is wasteful.
The new mechanism introduced here will support reservation for
"Eth QPs eligible for BF" for all drivers: bare-metal, multi-PF, and VFs
(when hypervisors support WC in VMs). The flow we use is:
1. In mlx4_en, allocate Tx QPs one by one instead of a range allocation,
and request "BF enabled QPs" if BF is supported for the function
2. In the ALLOC_RES FW command, change param1 to:
a. param1[23:0] - number of QPs
b. param1[31-24] - flags controlling QPs reservation
Bit 31 refers to Eth blueflame supported QPs. Those QPs must have
bits 6 and 7 unset in order to be used in Ethernet.
Bits 24-30 of the flags are currently reserved.
When a function tries to allocate a QP, it states the required attributes
for this QP. Those attributes are considered "best-effort". If an attribute,
such as Ethernet BF enabled QP, is a must-have attribute, the function has
to check that attribute is supported before trying to do the allocation.
In a lower layer of the code, mlx4_qp_reserve_range masks out the bits
which are unsupported. If SRIOV is used, the PF validates those attributes
and masks out unsupported attributes as well. In order to notify VFs which
attributes are supported, the VF uses QUERY_FUNC_CAP command. This command's
mailbox is filled by the PF, which notifies which QP allocation attributes
it supports.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Previously, we've fired all our completion callbacks straight from our ISR.
Some of those callbacks were lightweight (for example, mlx4_en's and
IPoIB napi callbacks), but some of them did more work (for example,
the user-space RDMA stack uverbs' completion handler). Besides that,
doing more than the minimal work in ISR is generally considered wrong,
it could even lead to a hard lockup of the system. Since when a lot
of completion events are generated by the hardware, the loop over those
events could be so long, that we'll get into a hard lockup by the system
watchdog.
In order to avoid that, add a new way of invoking completion events
callbacks. In the interrupt itself, we add the CQs which receive completion
event to a per-EQ list and schedule a tasklet. In the tasklet context
we loop over all the CQs in the list and invoke the user callback.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When VFs (guests in this context) issue the QUERY_DEV_CAP command, they
need not be told that host side virtualization features such as VST, FSM
(MAC anti-spoofing) and running > 80 VFs are supported by the device.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This was dropped by mistake for the napi_gro_frags flow, fix that.
Fixes: dd65beac48 ('net/mlx4_en: Extend usage of napi_gro_frags')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ConnectX HW is capable of using one of the following hash functions:
Toeplitz and an XOR hash function. This patch extends the implementation
of the mlx4_en driver set/get_rxfh callbacks to support getting and
setting the RSS hash function used by the device.
Signed-off-by: Eyal Perry <eyalpe@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch extends the set/get_rxfh ethtool-options for getting or
setting the RSS hash function.
It modifies drivers implementation of set/get_rxfh accordingly.
This change also delegates the responsibility of checking whether a
modification to a certain RX flow hash parameter is supported to the
driver implementation of set_rxfh.
User-kernel API is done through the new hfunc bitmask field in the
ethtool_rxfh struct. A bit set in the hfunc field is corresponding to an
index in the new string-set ETH_SS_RSS_HASH_FUNCS.
Got approval from most of the relevant driver maintainers that their
driver is using Toeplitz, and for the few that didn't answered, also
assumed it is Toeplitz.
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Ariel Elior <ariel.elior@qlogic.com>
Cc: Prashant Sreedharan <prashant@broadcom.com>
Cc: Michael Chan <mchan@broadcom.com>
Cc: Hariprasad S <hariprasad@chelsio.com>
Cc: Sathya Perla <sathya.perla@emulex.com>
Cc: Subbu Seetharaman <subbu.seetharaman@emulex.com>
Cc: Ajit Khaparde <ajit.khaparde@emulex.com>
Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Cc: Bruce Allan <bruce.w.allan@intel.com>
Cc: Carolyn Wyborny <carolyn.wyborny@intel.com>
Cc: Don Skidmore <donald.c.skidmore@intel.com>
Cc: Greg Rose <gregory.v.rose@intel.com>
Cc: Matthew Vick <matthew.vick@intel.com>
Cc: John Ronciak <john.ronciak@intel.com>
Cc: Mitch Williams <mitch.a.williams@intel.com>
Cc: Amir Vadai <amirv@mellanox.com>
Cc: Solarflare linux maintainers <linux-net-drivers@solarflare.com>
Cc: Shradha Shah <sshah@solarflare.com>
Cc: Shreyas Bhatewara <sbhatewara@vmware.com>
Cc: "VMware, Inc." <pv-drivers@vmware.com>
Cc: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: Eyal Perry <eyalpe@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
So this can be reused for identification of other "items" as well.
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Reviewed-by: Thomas Graf <tgraf@suug.ch>
Acked-by: John Fastabend <john.r.fastabend@intel.com>
Acked-by: Andy Gospodarek <gospo@cumulusnetworks.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Some VF drivers use the upper byte of "param1" (the qp count field)
in mlx4_qp_reserve_range() to pass flags which are used to optimize
the range allocation.
Under the current code, if any of these flags are set, the 32-bit
count field yields a count greater than 2^24, which is out of range,
and this VF fails.
As these flags represent a "best-effort" allocation hint anyway, they may
safely be ignored. Therefore, the PF driver may simply mask out the bits.
Fixes: c82e9aa0a8 "mlx4_core: resource tracking for HCA resources used by guests"
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
drivers/net/ieee802154/fakehard.c
A bug fix went into 'net' for ieee802154/fakehard.c, which is removed
in 'net-next'.
Add build fix into the merge from Stephen Rothwell in openvswitch, the
logging macros take a new initial 'log' argument, a new call was added
in 'net' so when we merge that in here we have to explicitly add the
new 'log' arg to it else the build fails.
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix ethtool set settings to not check AUTONEG_ENABLE
mlx4_en_set_settings should not check if cmd->autoneg == AUTONEG_ENABLE,
cmd->autoneg can be enabled by default and this check will fail other settings requests.
mlx4_en driver doesn't support changing autoneg value, but shouldn't fail the request
in case cmd->autoneg was set.
Fixes: d48b3ab ("net/mlx4_en: Use PTYS register to set ethtool settings (Speed)")
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is currently missing, which results in a crash when one attempts
to set VXLAN tunnel over the mlx4_en when acting as PF.
[ 2408.785472] BUG: unable to handle kernel NULL pointer dereference at (null)
[...]
[ 2408.994104] Call Trace:
[ 2408.996584] [<ffffffffa021f7f5>] ? vxlan_get_rx_port+0xd6/0x103 [vxlan]
[ 2409.003316] [<ffffffffa021f71f>] ? vxlan_lowerdev_event+0xf2/0xf2 [vxlan]
[ 2409.010225] [<ffffffffa0630358>] mlx4_en_start_port+0x862/0x96a [mlx4_en]
[ 2409.017132] [<ffffffffa063070f>] mlx4_en_open+0x17f/0x1b8 [mlx4_en]
While here, make sure to invoke vxlan_get_rx_port() only when VXLAN
offloads are actually enabled and not when they are only supported.
Reported-by: Ido Shamay <idos@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use of well known RSS key increases attack surface.
Switch to a random one, using generic helper so that all
ports share a common key.
Also provide ethtool -x support to fetch RSS key
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use vxlan_gso_check() to advertise offload support for this NIC.
Signed-off-by: Joe Stringer <joestringer@nicira.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
drivers/net/ethernet/chelsio/cxgb4vf/sge.c
drivers/net/ethernet/intel/ixgbe/ixgbe_phy.c
sge.c was overlapping two changes, one to use the new
__dev_alloc_page() in net-next, and one to use s->fl_pg_order in net.
ixgbe_phy.c was a set of overlapping whitespace changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
We now allow up to 126 VFs. Note though that certain firmware
versions only allow up to 80 VFs. Moreover, old HCAs only support 64 VFs.
In these cases, we limit the maximum number of VFs to 64.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Previously, the driver queried the firmware in order to get the number
of supported EQs. Under SRIOV, since this was done before the driver
notified the firmware how many VFs it actually needs, the firmware had
to take into account a worst case scenario and always allocated four EQs
per VF, where one was used for events while the others were used for completions.
Now, when the firmware supports the asymmetric allocation scheme, denoted
by exposing num_sys_eqs > 0 (--> MLX4_DEV_CAP_FLAG2_SYS_EQS), we use the
QUERY_FUNC command to query the firmware before enabling SRIOV. Thus we
can get more EQs and MSI-X vectors per function.
Moreover, when running in the new firmware/driver mode, the limitation
that the number of EQs should be a power of two is lifted.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
QUERY_FUNC firmware command could be used in order to query the
number of EQs, reserved EQs, etc for a specific function.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Refactor mlx4_load_one, as a preparation step for a new and
more complicated load function. The goal is to support both
newer firmware that required init_hca to be done before
enable_sriov and legacy firmwares that requires things to
be done the other way around.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Refactoring mlx4_cmd_init and mlx4_cmd_cleanup such that partial init
and cleanup are possible. After this refactoring, calling mlx4_cmd_init
several times is safe.
This is necessary in the VF init flow when mlx4_init_hca returns -EACCESS,
we need to issue cleanup and re-attempt to call it with the slave flag.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We've used an incorrect type for the loop counter and the
mlx4_QUERY_FUNC_CAP function. The current input modifier
is either a port or a boolean.
Since the number of ports is always a positive value < 255,
we should use u8 instead of an integer with casting.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We mistakenly read the reserved_eqs field as a standard
numeric value rather than a log2 value.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currenly we only support Large-Send and TX checksum offloads for
encapsulated traffic of type VXLAN. We must make sure to advertize
these offloads up to the stack only when VXLAN tunnel is set.
Failing to do so, would mislead the the networking stack to assume
that the driver can offload the internal TX checksum for GRE packets
and other buggy schemes.
Reported-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When processing received traffic, pass CHECKSUM_COMPLETE status to the
stack, with calculated checksum for non TCP/UDP packets (such
as GRE or ICMP).
Although the stack expects checksum which doesn't include the pseudo
header, the HW adds it. To address that, we are subtracting the pseudo
header checksum from the checksum value provided by the HW.
In the IPv6 case, we also compute/add the IP header checksum which
is not added by the HW for such packets.
Cc: Jerry Chu <hkchu@google.com>
Signed-off-by: Shani Michaeli <shanim@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We can call napi_gro_frags for all the received traffic regardless
of the checksum status. Specifically, received packets whose status
is CHECKSUM_NONE (and soon to be added CHECKSUM_COMPLETE)
are eligible for napi_gro_frags as well.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Shani Michaeli <shanim@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After commit 1a28817282 ("mlx4: use napi_complete_done()") we ended up
calling napi_complete_done() in the case NAPI poll consumed all its
budget.
This added extra interrupt pressure, this patch restores proper
behavior.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Fixes: 1a28817282 ("mlx4: use napi_complete_done()")
Signed-off-by: David S. Miller <davem@davemloft.net>
To enable gro_flush_timeout, a driver has to use napi_complete_done()
instead of napi_complete().
Tested:
Ran 200 netperf TCP_STREAM from A to B (10Gbe mlx4 link, 8 RX queues)
Without this feature, we send back about 305,000 ACK per second.
GRO aggregation ratio is low (811/305 = 2.65 segments per GRO packet)
Setting a timer of 2000 nsec is enough to increase GRO packet sizes
and reduce number of ACK packets. (811/19.2 = 42)
Receiver performs less calls to upper stacks, less wakes up.
This also reduces cpu usage on the sender, as it receives less ACK
packets.
Note that reducing number of wakes up increases cpu efficiency, but can
decrease QPS, as applications wont have the chance to warmup cpu caches
doing a partial read of RPC requests/answers if they fit in one skb.
B:~# sar -n DEV 1 10 | grep eth0 | tail -1
Average: eth0 811269.80 305732.30 1199462.57 19705.72 0.00
0.00 0.50
B:~# echo 2000 >/sys/class/net/eth0/gro_flush_timeout
B:~# sar -n DEV 1 10 | grep eth0 | tail -1
Average: eth0 811577.30 19230.80 1199916.51 1239.80 0.00
0.00 0.50
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add code to issue CONFIG_DEV "get" firmware command.
This command is used in order to obtain certain parameters used for
supporting various RX checksumming options and vxlan UDP port.
The GET operation is allowed for VFs too.
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Shani Michaeli <shanim@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Needed in order to get cache cold pages (L3 flushed) for HW scatter.
Otherwise memory may flush those entries when the packet comes from
PCI, causing back pressure resulting in BW decrease.
Signed-off-by: Ido Shamay <idos@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When IP_ALIGN has a non zero value, hardware will write to a non aligned
address. The only reader from this address is when copying the header
from the first frag into the linear buffer (further access to the IP
address will be from the linear buffer, in which the headers are
aligned). Since the penalty of non align access by the hardware is
greater than the software memcpy, changing the frag_align to always be 0.
Signed-off-by: Ido Shamay <idos@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We need to protect set_port_type() for concurrency, as the sysfs code could
call it from mutliple contexts in parallel.
The port_mutex is not enough because we need to protect from concurrent
modification of 'info' and stopping of the port sensing work.
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Added wrapper to the ACCESS_REG command for handling guest HW
registers access, preventing write operations, but do allow reads.
This will prevent SRIOV guests to change port PTYS configuration,
such as speed/advertised link modes.
Fixes: adbc7ac5c1 ('net/mlx4_core: Introduce ACCESS_REG CMD [...]')
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If mlx4_ib_create_flow() attempts to create > 1 rules with the
firmware, and one of these registrations fail, we leaked the
already created flow rules.
One example of the leak is when the registration of the VXLAN ghost
steering rule fails, we didn't unregister the original rule requested
by the user, introduced in commit d2fce8a906 "mlx4: Set
user-space raw Ethernet QPs to properly handle VXLAN traffic".
While here, add dump of the VXLAN portion of steering rules
so it can actually be seen when flow creation fails.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
For VXLAN/NVGRE encapsulation, the current HW doesn't support offloading
both the outer UDP TX checksum and the inner TCP/UDP TX checksum.
The driver doesn't advertize SKB_GSO_UDP_TUNNEL_CSUM, however we are wrongly
telling the HW to offload the outer UDP checksum for encapsulated packets,
fix that.
Fixes: 837052d0cc ('net/mlx4_en: Add netdev support for TCP/IP
offloads of vxlan tunneling')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
mlx4_en_rx_irq() and mlx4_en_tx_irq() run from hard interrupt context.
They can use napi_schedule_irqoff() instead of napi_schedule()
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-By: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Hardware requires the number of rings in indirection table to be a power
of 2. When setting number of channels to a non power of 2 number,
indirection table is using only the closest power of 2 rings.
Report this number in 'ethtool -x' and not the total number of rx rings.
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Upon failures, destroy_netdev is called, and spinlocks/works must be
initialized before calling it. Otherwise kernel panic may occur.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is instead of calling the actual implementation of
napi_synchronize, for better encapsulation.
Signed-off-by: Ido Shamay <idos@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
clang flagged the following. All are actually cosmetic cleanups, not really bugs:
drivers/net/ethernet/mellanox/mlx4/en_main.c:233:3: warning: Value stored to 'err' is never read
err = -ENOMEM;
^ ~~~~~~~
drivers/net/ethernet/mellanox/mlx4/en_main.c:293:3: warning: Value stored to 'err' is never read
err = -ENOMEM;
drivers/net/ethernet/mellanox/mlx4/en_netdev.c:648:16: warning: Assigned value is garbage or undefined
entry->reg_id = reg_id;
^ ~~~~~~
drivers/net/ethernet/mellanox/mlx4/en_netdev.c:659:2: warning: Function call argument is an uninitialized value
mlx4_en_uc_steer_release(priv, priv->dev->dev_addr, *qpn, reg_id);
(NOTE: reg_id is only used in the device-managed flow steering path, in which is it always initialized.
This is not a bug. Cleanup here is therefore cosmetic only).
drivers/net/ethernet/mellanox/mlx4/en_rx.c:122:3: warning: Value stored to 'frag_info' is never read
frag_info = &priv->frag_info[i];
^ ~~~~~~~~~~~~~~~~~~~
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move mlx4_en_reset_config to en_netdev.c as it now serves more general purpose.
Add support for turning OFF/ON the rx/tx vlan offlad.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rename mlx4_en_timestamp_config to mlx4_en_reset_config and extend it to support
choosing RX vlan offload configuration.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Added Support to set speed or advertised link modes via ethtool:
ethtool -s <ifname> [speed <speed>] [advertise <link modes>]
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- If dev cap MLX4_DEV_CAP_FLAG2_ETH_PROT_CTRL is ON, query PTYS register to fill ethtool settings.
else use default values.
- Use autoneg port cap and dev backplane autoneg cap to reprort autoneg interface capbilities.
- Fix typo in mlx4_en_port_state struct field (transciver to transceiver).
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Added 100M, 20G and 56G ethtool speed reporting support.
Update mlx4_en_test_speed self test with the new speeds.
Defined new link speeds in include/uapi/linux/ethtool.h:
+#define SPEED_20000 20000
+#define SPEED_40000 40000
+#define SPEED_56000 56000
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Adding ACCESS REG mlx4 command and use it to implement Query method for
PTYS (Port Type and Speed Register).
Query and store eth_prot_ctrl dev cap.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Added support for get_module_info/get_module_eeprom ethtool support for cable info reading.
Added new cable types enum in include/uapi/linux/ethtool.h for ethtool use.
+#define ETH_MODULE_SFF_8636 0x3
+#define ETH_MODULE_SFF_8636_LEN 256
+#define ETH_MODULE_SFF_8436 0x4
+#define ETH_MODULE_SFF_8436_LEN 256
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Added new MAD_IFC command to read cable module info with attribute id (0xFF60).
Update include/linux/mlx4/device.h with function declaration (mlx4_get_module_info)
and the needed defines/enums for future use.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
After moving the EQ ownership to software effectively destroying it, call
synchronize_irq() to ensure that any handler routines running on other CPU
cores finish execution. Only then free the EQ buffer.
The same thing is done when we destroy a CQ which is one of the sources
generating interrupts. In the case of CQ we want to avoid completion handlers
on a CQ that was destroyed. In the case we do the same to avoid receiving
asynchronous events after the EQ has been destroyed and its buffers freed.
Signed-off-by: Eli Cohen <eli@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is illegal to use atomic_set(&page->_count, ...) even if we 'own'
the page. Other entities in the kernel need to use get_page_unless_zero()
to get a reference to the page before testing page properties, so we could
loose a refcount increment.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull networking updates from David Miller:
"Most notable changes in here:
1) By far the biggest accomplishment, thanks to a large range of
contributors, is the addition of multi-send for transmit. This is
the result of discussions back in Chicago, and the hard work of
several individuals.
Now, when the ->ndo_start_xmit() method of a driver sees
skb->xmit_more as true, it can choose to defer the doorbell
telling the driver to start processing the new TX queue entires.
skb->xmit_more means that the generic networking is guaranteed to
call the driver immediately with another SKB to send.
There is logic added to the qdisc layer to dequeue multiple
packets at a time, and the handling mis-predicted offloads in
software is now done with no locks held.
Finally, pktgen is extended to have a "burst" parameter that can
be used to test a multi-send implementation.
Several drivers have xmit_more support: i40e, igb, ixgbe, mlx4,
virtio_net
Adding support is almost trivial, so export more drivers to
support this optimization soon.
I want to thank, in no particular or implied order, Jesper
Dangaard Brouer, Eric Dumazet, Alexander Duyck, Tom Herbert, Jamal
Hadi Salim, John Fastabend, Florian Westphal, Daniel Borkmann,
David Tat, Hannes Frederic Sowa, and Rusty Russell.
2) PTP and timestamping support in bnx2x, from Michal Kalderon.
3) Allow adjusting the rx_copybreak threshold for a driver via
ethtool, and add rx_copybreak support to enic driver. From
Govindarajulu Varadarajan.
4) Significant enhancements to the generic PHY layer and the bcm7xxx
driver in particular (EEE support, auto power down, etc.) from
Florian Fainelli.
5) Allow raw buffers to be used for flow dissection, allowing drivers
to determine the optimal "linear pull" size for devices that DMA
into pools of pages. The objective is to get exactly the
necessary amount of headers into the linear SKB area pre-pulled,
but no more. The new interface drivers use is eth_get_headlen().
From WANG Cong, with driver conversions (several had their own
by-hand duplicated implementations) by Alexander Duyck and Eric
Dumazet.
6) Support checksumming more smoothly and efficiently for
encapsulations, and add "foo over UDP" facility. From Tom
Herbert.
7) Add Broadcom SF2 switch driver to DSA layer, from Florian
Fainelli.
8) eBPF now can load programs via a system call and has an extensive
testsuite. Alexei Starovoitov and Daniel Borkmann.
9) Major overhaul of the packet scheduler to use RCU in several major
areas such as the classifiers and rate estimators. From John
Fastabend.
10) Add driver for Intel FM10000 Ethernet Switch, from Alexander
Duyck.
11) Rearrange TCP_SKB_CB() to reduce cache line misses, from Eric
Dumazet.
12) Add Datacenter TCP congestion control algorithm support, From
Florian Westphal.
13) Reorganize sk_buff so that __copy_skb_header() is significantly
faster. From Eric Dumazet"
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1558 commits)
netlabel: directly return netlbl_unlabel_genl_init()
net: add netdev_txq_bql_{enqueue, complete}_prefetchw() helpers
net: description of dma_cookie cause make xmldocs warning
cxgb4: clean up a type issue
cxgb4: potential shift wrapping bug
i40e: skb->xmit_more support
net: fs_enet: Add NAPI TX
net: fs_enet: Remove non NAPI RX
r8169:add support for RTL8168EP
net_sched: copy exts->type in tcf_exts_change()
wimax: convert printk to pr_foo()
af_unix: remove 0 assignment on static
ipv6: Do not warn for informational ICMP messages, regardless of type.
Update Intel Ethernet Driver maintainers list
bridge: Save frag_max_size between PRE_ROUTING and POST_ROUTING
tipc: fix bug in multicast congestion handling
net: better IFF_XMIT_DST_RELEASE support
net/mlx4_en: remove NETDEV_TX_BUSY
3c59x: fix bad split of cpu_to_le32(pci_map_single())
net: bcmgenet: fix Tx ring priority programming
...
Add two helpers so that drivers do not have to care of BQL being
available or not.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Jim Davis <jim.epost@gmail.com>
Fixes: 29d40c9032 ("net/mlx4_en: Use prefetch in tx path")
Signed-off-by: David S. Miller <davem@davemloft.net>
Drivers should avoid NETDEV_TX_BUSY as much as possible.
They should stop the tx queue before qdisc even tries to push another
packet, to avoid requeues.
For a driver supporting skb->xmit_more, this is likely to be a prereq
anyway, otherwise we could have a tx deadlock : We need to force a
doorbell if TX ring is full.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of setting inline threshold using module parameter only on
driver load, use set_tunable() to set it dynamically.
No need to store the threshold per ring, using instead the netdev global
priv->prof->inline_thold
Initial value still is set using the module parameter, therefore
backward compatability is kept.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reorganize code to call is_inline() once, so compiler can inline it
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Properly clear tx_info->ts_requested
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Access skb_headlen() once in tx flow
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Acces skb_shinfo(skb) once in tx flow.
Also, rename @i variable to @i_frag to avoid confusion, as the "goto
tx_drop_unmap;" relied on this @i variable.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
mlx4_en_process_tx_cq() carefully fetches and writes ring->last_nr_txbb
and ring->cons only one time to avoid false sharing
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
mlx4_en_free_tx_desc() uses a prefetchw(&skb->users) to speed up
consume_skb()
prefetchw(&ring->tx_queue->dql) to speed up BQL update
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add frag0_dma/frag0_byte_count into mlx4_en_tx_info to avoid a cache
line miss in TX completion for frames having one dma element. (We avoid
reading back the tx descriptor)
Note this could be extended to 2/3 dma elements later, as we have free
room in mlx4_en_tx_info
Also, mlx4_en_free_tx_desc() no longer accesses skb_shinfo(). We use a
new nr_maps fields in mlx4_en_tx_info to avoid 2 or 3 cache misses.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Try to allocate using kmalloc_node() first, only on failure use
vmalloc()
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- doorbell_qpn is stored in the cpu_to_be32() way to avoid bswap() in fast
path.
- mdev->mr.key stored in ring->mr_key to also avoid bswap() and access to
cold cache line.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reorganize struct mlx4_en_tx_ring to have:
- One cache line containing last_nr_txbb & cons & wake_queue, used by tx
completion.
- One cache line containing fields dirtied by mlx4_en_xmit()
- Following part is read mostly and shared by cpus.
Align struct mlx4_en_tx_info to a cache line
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- Remove unused variable ring->poll_cnt
- No need to set some fields if using blueflame
- Add missing const's
- Use unlikely
- Remove unneeded new line
- Make some comments more precise
- struct mlx4_bf @offset field reduced to unsigned int to save space
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ethtool -S reports a new counter, tracking number of time doorbell
was not triggered, because skb->xmit_more was set.
$ ethtool -S eth0 | egrep "tx_packet|xmit_more"
tx_packets: 2413288400
xmit_more: 666121277
I merged the tso_packet false sharing avoidance in this patch as well.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
drivers/net/usb/r8152.c
net/netfilter/nfnetlink.c
Both r8152 and nfnetlink conflicts were simple overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
ConnectX2 HCAs have max_mtu=4k and max_vl=8 vls. However, if you specify
a 4K mtu, the max_vl supported for 4K is 4 vls. The driver at startup
attempts to set a 4K mtu using the max_vl value obtained from QUERY_PORT.
Since the max_vl value is 8 vls (which is supported up to 2K mtu size),
the first attempt to set the mtl/vl port value will fail, generating
the following error message in the log:
mlx4_core 0000:06:00.0: command 0xc failed: fw status = 0x40
The driver then tries again, using mtu=4k, vls=4, and this succeeds.
Since we do not want to have this error message always displayed at driver
start when there are ConnectX2 HCAs on the host, we deprecate the error
message for this specific command/input_modifier/opcode_modifier/fw-status
to be debug.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The function mlx4_QUERY_PORT_wrapper implements only the
QUERY_PORT "general" case (opcode modifier = 0).
Verify that the opcode modifier is zero, and also that the
input modifier contains only the port number in bits 0..7
(all other bits should be zero).
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the new flow, we separate the pci initialization and teardown
from the initialization and teardown of the other resources.
__mlx4_init_one handles the pci resources initialization. It then
calls mlx4_load_one to initialize the remainder of the resources.
When removing a device, mlx4_remove_one is invoked. However, now
mlx4_remove_one calls mlx4_unload_one to free all the resources except the pci
resources. When mlx4_unload_one returns, mlx4_remove_one then frees the
pci resources.
The above separation will allow us to implement 'reset flow' in the future.
It will also enable more EQs for VFs and is a pre-step to the modern API to
enable/disable SRIOV.
Also added nvfs; an integer array of size MLX4_MAX_PORTS + 1; to the mlx4_dev
struct. This new field is used to avoid parsing the num_vfs module parameter
each time the mlx4_restart_one is called.
Signed-off-by: Majd Dibbiny <majd@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When unloading the host driver while there are VFs active on VMs,
the PF driver disabled sriov anyway, causing kernel crashes.
We now leave SRIOV enabled, to avoid that.
When the driver is reloaded, __mlx4_init_one is invoked on the PF.
It now checks to see if SRIOV is already enabled on the PF -- and
if so does not enable sriov again.
Signed-off-by: Tal Alon <talal@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
skb->xmit_more tells us if another skb is coming next.
We need to send doorbell when : xmit_more is not set,
or txqueue is stopped (preventing next skb to come immediately)
Tested with a modified pktgen version, I got a 40% increase of
throughput.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When the HCA is configured in SRIOV IB mode (that is, at least one of
the ports is IB) and the probe_vf module param isn't specified,
mlx4_init_one() failed because of the following condition:
if (ib_ports && (num_vfs_argc > 1 || probe_vfs_argc > 1)) {
.....
}
The root cause for that is a mistake in the initialization of num_vfs_argc
and probe_vfs_argc. When num_vfs / probe_vf aren't given, their argument
count counterpart should be 0, fix that.
Fixes: dd41cc3bb9 ('net/mlx4: Adapt num_vfs/probed_vf params for single port VF')
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
No caller or macro uses the return value so make it void.
Signed-off-by: Joe Perches <joe@perches.com>
Acked-By: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- Fixes for the new memory region re-registration support
- iSER initiator error path fixes
- Grab bag of small fixes for the qib and ocrdma hardware drivers
- Larger set of fixes for mlx4, especially in RoCE mode
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=Kpm1
-----END PGP SIGNATURE-----
Merge tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband
Pull infiniband/rdma fixes from Roland Dreier:
"Last late set of InfiniBand/RDMA fixes for 3.17:
- fixes for the new memory region re-registration support
- iSER initiator error path fixes
- grab bag of small fixes for the qib and ocrdma hardware drivers
- larger set of fixes for mlx4, especially in RoCE mode"
* tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (26 commits)
IB/mlx4: Fix VF mac handling in RoCE
IB/mlx4: Do not allow APM under RoCE
IB/mlx4: Don't update QP1 in native mode
IB/mlx4: Avoid accessing netdevice when building RoCE qp1 header
mlx4: Fix mlx4 reg/unreg mac to work properly with 0-mac addresses
IB/core: When marshaling uverbs path, clear unused fields
IB/mlx4: Avoid executing gid task when device is being removed
IB/mlx4: Fix lockdep splat for the iboe lock
IB/mlx4: Get upper dev addresses as RoCE GIDs when port comes up
IB/mlx4: Reorder steps in RoCE GID table initialization
IB/mlx4: Don't duplicate the default RoCE GID
IB/mlx4: Avoid null pointer dereference in mlx4_ib_scan_netdevs()
IB/iser: Bump version to 1.4.1
IB/iser: Allow bind only when connection state is UP
IB/iser: Fix RX/TX CQ resource leak on error flow
RDMA/ocrdma: Use right macro in query AH
RDMA/ocrdma: Resolve L2 address when creating user AH
mlx4: Correct error flows in rereg_mr
IB/qib: Correct reference counting in debugfs qp_stats
IPoIB: Remove unnecessary port query
...
Conflicts:
arch/mips/net/bpf_jit.c
drivers/net/can/flexcan.c
Both the flexcan and MIPS bpf_jit conflicts were cases of simple
overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
There is a chance that the VF mlx4 RoCE driver (mlx4_ib) may see a 0-mac
as the current default MAC address when a RoCE interface first comes up.
In this case, the RoCE driver registers the 0-mac to get its MAC index --
used in the INIT2RTR transition when it creates its proxy Q1 qp's.
If we do not allow QP1 to be created, the RoCE driver will not come up.
If we do not register the 0-mac, but simply use a random mac-index,
QP1 will attempt to send packets with an someone's else source MAC which
will get the system into more troubled.
Since a 0-mac was previously used to indicate a free slot, this leads to
errors, both when the 0-mac is registered and when it is unregistered.
The required fix is to check in addition that the slot containing the
0-mac has a reference count of zero.
Additionally, when comparing MAC addresses, need to mask out the 2 MSBs
of the u64 mac on both sides of the comparison.
Note that when the EN driver (mlx4_en) comes up, it set itself a proper
mac --> the RoCE driver gets to be notified on that and further handing
is done with the update qp command, as was added by commit 9433c18891
("IB/mlx4: Invoke UPDATE_QP for proxy QP1 on MAC changes").
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
This patch addresses feedback from Sagi Grimberg on the rereg_mr
implementation of mlx4. The following are fixed:
1. Set the correct pd_flags
2. Make sure we change the iova and size MR fields only after
successful write and allocation of the MTTs.
3. Make the error checking more robust
Fixes: e630664c83 ("mlx4_core: Add helper functions to support MR re-registration")
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
This function derives the base address of the CQE from the CQE size,
and calculates the real CQE context segment in it from the factor
(this is like before). Before this change the code used the factor to
calculate the base address of the CQE as well.
The factor indicates in which segment of the cqe stride the cqe information
is located. For 32-byte strides, the segment is 0, and for 64 byte strides,
the segment is 1 (bytes 32..63). Using the factor was ok as long as we had
only 32 and 64 byte strides. However, with larger strides, the factor is zero,
and so cannot be used to calculate the base of the CQE.
The helper uses the same method of CQE buffer pulling made by all other
components that reads the CQE buffer (mlx4_ib driver and libmlx4).
Signed-off-by: Ido Shamay <idos@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Enable mlx4 interrupt handler to work with EQE stride feature,
The feature may be enabled when cache line is bigger than 64B.
The EQE size will then be the cache line size, and the context
segment resides in [0-31] offset.
Signed-off-by: Ido Shamay <idos@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This feature is intended for archs having cache line larger then 64B.
Since our CQE/EQEs are generally 64B in those systems, HW will write
twice to the same cache line consecutively, causing pipe locks due to
he hazard prevention mechanism. For elements in a cyclic buffer, writes
are consecutive, so entries smaller than a cache line should be
avoided, especially if they are written at a high rate.
Reduce consecutive writes to same cache line in CQs/EQs, by allowing the
driver to increase the distance between entries so that each will reside
in a different cache line. Until the introduction of this feature, there
were two types of CQE/EQE:
1. 32B stride and context in the [0-31] segment
2. 64B stride and context in the [32-63] segment
This feature introduces two additional types:
3. 128B stride and context in the [0-31] segment (128B cache line)
4. 256B stride and context in the [0-31] segment (256B cache line)
Modify the mlx4_core driver to query the device for the CQE/EQE cache
line stride capability and to enable that capability when the host
cache line size is larger than 64 bytes (supported cache lines are
128B and 256B).
The mlx4 IB driver and libmlx4 need not be aware of this change. The PF
context behaviour is changed to require this change in VF drivers
running on such archs.
Signed-off-by: Ido Shamay <idos@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Changing the vlan stripping policy of the QP isn't supported by older
firmware versions for the INIT2RTR command. Nevertheless, we've used it.
Fix that by doing this policy change using INIT2RTR only if the firmware
supports it, otherwise, we call UPDATE_QP command to do the task.
Fixes: 7677fc9 ('net/mlx4: Strengthen VLAN tags/priorities enforcement in VST mode')
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The current wrapper implementation of the UPDATE_QP command tries to get
the MAC index, even if MAC wasn't set by the VF. Fix it up to only handle
the MAC field if it's valid.
Fixes: ce8d9e0 ('net/mlx4_core: Add UPDATE_QP SRIOV wrapper support')
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When doing VGT->VST->VGT state changes, we used an incorrect mask
for the vlan-stripping-disable (VSD) flag, hence the vlan related policy
for user-space Raw Ethernet QPs open by VFs wasn't really applied.
Fix that, by using the correct mask.
Fixes: f0f829b ('net/mlx4_core: Add immediate activate for VGT->VST->VGT')
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Single port VFs are seen PCI wise on both ports of the PF (we don't have
single port PFs with ConnectX). With this in mind, it's possible for
virtualization tools to try and configure a single ported VF through
the "wrong" PF port.
To handle that, we use the PF driver mapping of single port VFs to NIC
ports and adjust the port value before calling into the low level
code that does the actual VF configuration
Fixes: 449fc48 ('net/mlx4: Adapt code for N-Port VF')
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The driver does not support pause autonegotiation so it should return
-EINVAL when the function is called with non-zero autoneg.
Cc: Amir Vadai <amirv@mellanox.com>
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use the new fancy eth_get_headlen() to pull exactly the headers
into skb->head.
This speeds up GRE traffic (or more generally tunneled traffuc),
as GRO can aggregate up to 17 MSS per GRO packet instead of 8.
(Pulling too much data was forcing GRO to keep 2 frags per MSS)
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It would appear the mlx4_en driver was still making a call to
dev_kfree_skb_any() where dev_consume_skb_any() would be more
appropriate. This should make dropped packet profiling/tracking
easier/better over a NIC driven by mlx4_en.
Signed-off-by: Rick Jones <rick.jones2@hp.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Set skb->csum_level instead of skb->encapsulation when indicating
CHECKSUM_UNNECESSARY for an encapsulated checksum.
Signed-off-by: Tom Herbert <therbert@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move the function which we use to set VXLAN DMFS (flow-steering) rules
from mlx4_en to mlx4_core. This refactoring will allow the mlx4_ib driver
to call the helper for the use case of user-space RAW Ethernet QPs, such
that they can serve VXLAN traffic too.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Miscellaneous
- Remove DEFINE_PCI_DEVICE_TABLE macro use (Benoit Taine)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=jJGu
-----END PGP SIGNATURE-----
Merge tag 'pci-v3.17-changes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Pull DEFINE_PCI_DEVICE_TABLE removal from Bjorn Helgaas:
"Part two of the PCI changes for v3.17:
- Remove DEFINE_PCI_DEVICE_TABLE macro use (Benoit Taine)
It's a mechanical change that removes uses of the
DEFINE_PCI_DEVICE_TABLE macro. I waited until later in the merge
window to reduce conflicts, but it's possible you'll still see a few"
* tag 'pci-v3.17-changes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
PCI: Remove DEFINE_PCI_DEVICE_TABLE macro use
- MR reregistration support
- MAD support for RMPP in userspace
- iSER and SRP initiator updates
- ocrdma hardware driver updates
- other fixes...
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=wOm0
-----END PGP SIGNATURE-----
Merge tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband
Pull infiniband/rdma updates from Roland Dreier:
"Main set of InfiniBand/RDMA updates for 3.17 merge window:
- MR reregistration support
- MAD support for RMPP in userspace
- iSER and SRP initiator updates
- ocrdma hardware driver updates
- other fixes..."
* tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (52 commits)
IB/srp: Fix return value check in srp_init_module()
RDMA/ocrdma: report asic-id in query device
RDMA/ocrdma: Update sli data structure for endianness
RDMA/ocrdma: Obtain SL from device structure
RDMA/uapi: Include socket.h in rdma_user_cm.h
IB/srpt: Handle GID change events
IB/mlx5: Use ARRAY_SIZE instead of sizeof/sizeof[0]
IB/mlx4: Use ARRAY_SIZE instead of sizeof/sizeof[0]
RDMA/amso1100: Check for integer overflow in c2_alloc_cq_buf()
IPoIB: Remove unnecessary test for NULL before debugfs_remove()
IB/mad: Add user space RMPP support
IB/mad: add new ioctl to ABI to support new registration options
IB/mad: Add dev_notice messages for various umad/mad registration failures
IB/mad: Update module to [pr|dev]_* style print messages
IB/ipoib: Avoid multicast join attempts with invalid P_key
IB/umad: Update module to [pr|dev]_* style print messages
IB/ipoib: Avoid flushing the workqueue from worker context
IB/ipoib: Use P_Key change event instead of P_Key polling mechanism
IB/ipath: Add P_Key change event support
mlx4_core: Add support for secure-host and SMP firewall
...
We should prefer `struct pci_device_id` over `DEFINE_PCI_DEVICE_TABLE` to
meet kernel coding style guidelines. This issue was reported by checkpatch.
A simplified version of the semantic patch that makes this change is as
follows (http://coccinelle.lip6.fr/):
// <smpl>
@@
identifier i;
declarer name DEFINE_PCI_DEVICE_TABLE;
initializer z;
@@
- DEFINE_PCI_DEVICE_TABLE(i)
+ const struct pci_device_id i[]
= z;
// </smpl>
[bhelgaas: add semantic patch]
Signed-off-by: Benoit Taine <benoit.taine@lip6.fr>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Secure-host is the general term for the capability of a device
to protect itself and the subnet from malicious host software.
This is achieved by:
1. Not allowing un-trusted entities to access device configuration
registers, directly (through pci_cr or pci_conf) and indirectly
(through MADs).
2. Hiding M_Key from untrusted entities.
3. Preventing the modification of GUID0 by un-trusted entities
4. Not allowing drivers on untrusted hosts to receive nor to transmit
packets over QP0 (SMP Firewall).
The secure-host capability depends on firmware handling all QP0
packets, and not passing these packets up to the driver. Any information
required by the driver for proper operation (e.g., SM lid) is passed
via events generated by the firmware while processing QP0 MADs.
Driver support mainly requires using the MAD_DEMUX FW command at startup,
where the feature is enabled/disabled through a procedure described in
the Mellanox HCA tools package.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
[ Fix error path in mlx4_setup_hca to go to err_mcg_table_free. - Roland ]
Signed-off-by: Roland Dreier <roland@purestorage.com>
Add few helper functions to support a mechanism of getting an MPT,
modifying it and updating the HCA with the modified object.
The code takes 2 paths, one for directly changing the MPT (and
sometimes its related MTTs) and another one which queries the MPT and
updates the HCA via fw command SW2HW_MPT. The first path is used in
native mode; the second path is slower and is used only in SRIOV.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Fixes sparse warning intrduced by commit 0fef9d0 ("net/mlx4_en: Disable
blueflame using ethtool private flags")
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When memory is limited, reduce number of rx and tx rings.
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When running in kdump kernel, reduce number of resources allocated for
the hardware. This will enable the NIC to operate in this low memory
environment at the expense of performance and some features not related
to the basic NIC functionality.
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Enable the user to turn off the hardware feature called BlueFlame.
Since it is something specific to mlx4_en hardware, we control
the feature via ethtool private flags.
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When port is down dev_addr is changed (e.g. by bonding) but current_mac
is not touched. When port is up again, hash_mac is updated to dev_addr,
but current_mac isn't. This leads to inconsistency between current_mac
and mac_hash. Because of that, mlx4_en_replace_mac() fails to find
current_mac in mac_hash.
Fix is to reset current_mac to dev_addr when port is up - as we do for
mac_hash.
Signed-off-by: Eyal Perry <eyalpe@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
drivers/infiniband/hw/cxgb4/device.c
The cxgb4 conflict was simply overlapping changes.
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix a regression introduced by commit 35f6f45 ("net/mlx4_en: Don't use
irq_affinity_notifier to track changes in IRQ affinity map").
When core is started in legacy EQ's (number of IRQ's < rx rings), cq->irq_desc
was NULL. This caused a kernel crash under heavy traffic - when having more
than rx NAPI budget completions.
Fixed to have it set for both EQ modes.
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In B0 steering mode if promiscuous QP asks to be detached from MCG entry,
and it is the only one in this entry then the entry will never be deleted.
This is a wrong behavior since we don't want to keep those entries after
the promiscuous QP becomes non-promiscuous. Therefore remove steering
entry containing only promiscuous QP.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In current situation host is adding the promiscuous QP to all steering
entries and the default entry as well. In this case when having PV
and SR-IOV on the same setup bridge will receive all traffic that is
targeted to the other VMs. This is bad.
Solution: In SR-IOV mode host can add promiscuous QP to default entry only.
The above problem and fix are relevant for B0 steering mode only.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In B0 steering mode when adding QPs to the default MCG entry need
to check that maximal number of QPs per MCG entry was not exceeded.
Signed-off-by: Alexander Guller <alexg@mellanox.com>
Reviewed-by: Aviad Yehezkel <aviadye@mellanox.co.il>
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
To make sure that the array index isn't used in the code with
negative value, we stop using the for loop integer iterator
outside of it.
>From now on use members count to swap the last QP with removed one.
Fix also the second occurrence of this flow in mlx4_qp_detach_common().
In mlx4_qp_detach_common() use members_count instead of
loop iterator outside of the for loop.
Signed-off-by: Dotan Barak <dotanb@dev.mellanox.co.il>
Reviewed-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When removing multicast address in B0 steering mode there is
a bug in cases where there is a single QP registered for the address,
and this QP is also promiscuous. In such cases the entry wouldn't be
deleted from the SW structure representing all Ethernet MCG entries,
but would be removed in HW. This way when driver goes to remove it
from SW and HW structures the HW deletion fails.
Moreover the same index could later be used for registering
different address, which can be Infiniband.
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Napi id was not marked for gro_skb, this will lead rx busy loop won't
work correctly since they stack never try to call low latency receive
method because of a zero socket napi id. Fix this by marking napi id
for gro_skb.
The transaction rate of 1 byte netperf tcp_rr gets about 50% increased
(from 20531.68 to 30610.88).
Cc: Amir Vadai <amirv@mellanox.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It is recommended that TX work not count against the quota.
The cost of TX packet liberation is a minute percentage of what it costs to
process an RX frame. Furthermore, that SKB freeing makes memory available for
other paths in the stack.
Give the TX a larger budget and be more aggressive about cleaning up the Tx
descriptors this budget could be changed using ethtool:
$ ethtool -C eth1 tx-frames-irq <budget>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Using a local copy of dev_addr in mlx4_en_set_mac() to prevent dev_addr
from being modified during error flow or when dev_addr is modified in
another context (which is another problem that is being discussed over
the mailing list [1]).
Also fixing bad naming of priv->prev_mac into priv->current_mac.
[1] - http://patchwork.ozlabs.org/patch/351489/
Reviewed-by: Eyal Perry <eyalpe@mellanox.com>
Signed-off-by: Noa Osherovich <noaos@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
LLC/SNAP 8 bytes should not be added as part of header calculation.
If used, payload will be decreased accordingly. For MTU of 1500
we'll set 1522 instead of 1523.
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Reviewed-by: Liran Liss <liranl@mellanox.com>
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Promiscous mode is only for MACs.
Should not disable/enable VLAN filter when entering/leaving promisuous mode.
Signed-off-by: Aviad Yehezkel <aviadye@mellanox.co.il>
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Verify port number to avoid crashes if port number is outside the range.
Signed-off-by: Eli Cohen <eli@mellanox.com>
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Loopback can't work when port is down.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In 40GE we can't use the default bw units for set ratelimit (100 Mbps)
since the max is 255*100 Mbps = 25 Gbps (not suited for 40GE), thus we need 1 Gbps units.
But for 10GE 1 Gbps units might be too bruit so we use the following solution.
For user set ratelimit <= 25 Gbps:
use 100 Mbps units * user_ratelimit (* 10).
For user set ratelimit > 25 Gbps:
use 1 Gbps units * user_ratelimit.
For user set unlimited ratelimit (0 Gbps):
use 1 Gbps units * MAX_RATELIMIT_DEFAULT (57)
Note: any value > 58 will damage the FW ratelimit computation, so we allow
a max and any higher value will be pulled down to 57.
Signed-off-by: Sagi Grimberg <sagig@mellanox.com>
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The add_vxlan_port ndo driver code was wrongly testing whether HW vxlan offloads
are supported by the device instead of checking if they are currently enabled.
This causes the driver to configure the HW parser to conduct matching for vxlan
packets but since no steering rules were set, vxlan packets are dropped on RX.
Fix that by doing the right test, as done in the del_vxlan_port ndo handler.
Fixes: 1b136de ('net/mlx4: Implement vxlan ndo calls')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Need to remove affinity hint at mlx4_en_deactivate_cq() and not at
mlx4_en_destroy_cq() - since affinity_mask might be free'd while still
being used by procfs.
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
IRQ affinity notifier can only have a single notifier - cpu_rmap
notifier. Can't use it to track changes in IRQ affinity map.
Detect IRQ affinity changes by comparing CPU to current IRQ affinity map
during NAPI poll thread.
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Ben Hutchings <ben@decadent.org.uk>
Fixes: 2eacc23 ("net/mlx4_core: Enforce irq affinity changes immediatly")
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Single ported VF are currently not supported on configurations where
one or both ports are IB. When we hit this case, the relevant flow in
the driver didn't return error and jumped to the wrong label. Fix that.
Fixes: dd41cc3 ('net/mlx4: Adapt num_vfs/probed_vf params for single port VF')
Reported-by: Shirley Ma <shirley.ma@oracle.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull networking updates from David Miller:
1) Seccomp BPF filters can now be JIT'd, from Alexei Starovoitov.
2) Multiqueue support in xen-netback and xen-netfront, from Andrew J
Benniston.
3) Allow tweaking of aggregation settings in cdc_ncm driver, from Bjørn
Mork.
4) BPF now has a "random" opcode, from Chema Gonzalez.
5) Add more BPF documentation and improve test framework, from Daniel
Borkmann.
6) Support TCP fastopen over ipv6, from Daniel Lee.
7) Add software TSO helper functions and use them to support software
TSO in mvneta and mv643xx_eth drivers. From Ezequiel Garcia.
8) Support software TSO in fec driver too, from Nimrod Andy.
9) Add Broadcom SYSTEMPORT driver, from Florian Fainelli.
10) Handle broadcasts more gracefully over macvlan when there are large
numbers of interfaces configured, from Herbert Xu.
11) Allow more control over fwmark used for non-socket based responses,
from Lorenzo Colitti.
12) Do TCP congestion window limiting based upon measurements, from Neal
Cardwell.
13) Support busy polling in SCTP, from Neal Horman.
14) Allow RSS key to be configured via ethtool, from Venkata Duvvuru.
15) Bridge promisc mode handling improvements from Vlad Yasevich.
16) Don't use inetpeer entries to implement ID generation any more, it
performs poorly, from Eric Dumazet.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1522 commits)
rtnetlink: fix userspace API breakage for iproute2 < v3.9.0
tcp: fixing TLP's FIN recovery
net: fec: Add software TSO support
net: fec: Add Scatter/gather support
net: fec: Increase buffer descriptor entry number
net: fec: Factorize feature setting
net: fec: Enable IP header hardware checksum
net: fec: Factorize the .xmit transmit function
bridge: fix compile error when compiling without IPv6 support
bridge: fix smatch warning / potential null pointer dereference
via-rhine: fix full-duplex with autoneg disable
bnx2x: Enlarge the dorq threshold for VFs
bnx2x: Check for UNDI in uncommon branch
bnx2x: Fix 1G-baseT link
bnx2x: Fix link for KR with swapped polarity lane
sctp: Fix sk_ack_backlog wrap-around problem
net/core: Add VF link state control policy
net/fsl: xgmac_mdio is dependent on OF_MDIO
net/fsl: Make xgmac_mdio read error message useful
net_sched: drr: warn when qdisc is not work conserving
...
The “affinity hint” mechanism is used by the user space
daemon, irqbalancer, to indicate a preferred CPU mask for irqs.
Irqbalancer can use this hint to balance the irqs between the
cpus indicated by the mask.
We wish the HCA to preferentially map the IRQs it uses to numa cores
close to it. To accomplish this, we use cpumask_set_cpu_local_first(), that
sets the affinity hint according the following policy:
First it maps IRQs to “close” numa cores. If these are exhausted, the
remaining IRQs are mapped to “far” numa cores.
Signed-off-by: Yuval Atias <yuvala@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Following commit befdf89 "net/mlx4_core: Preserve pci_dev_data after
__mlx4_remove_one()", there are two mlx4 pci callbacks which will
attempt to release the mlx4_priv object -- .shutdown and .remove.
This leads to a use-after-free access to the already freed mlx4_priv
instance and trigger a "Kernel access of bad area" crash when both
.shutdown and .remove are called.
During reboot or kexec, .shutdown is called, with the VFs probed to
the host going through shutdown first and then the PF. Later, the PF
will trigger VFs' .remove since VFs still have driver attached.
Fix that by keeping only one driver entry which releases mlx4_priv.
Fixes: befdf89 ('net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one()')
CC: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The Hypervisor driver tracks free slots and reserved slots at the global level
and tracks allocated slots and guaranteed slots per VF.
Guaranteed slots are treated as reserved by the driver, so the total
reserved slots is the sum of all guaranteed slots over all the VFs.
As VFs allocate resources, free (global) is decremented and allocated (per VF)
is incremented for those resources. However, reserved (global) is never changed.
This means that effectively, when a VF allocates a resource from its
guaranteed pool, it is actually reducing that resource's free pool (since
the global reserved count was not also reduced).
The fix for this problem is the following: For each resource, as long as a
VF's allocated count is <= its guaranteed number, when allocating for that
VF, the reserved count (global) should be reduced by the allocation as well.
When the global reserved count reaches zero, the remaining global free count
is still accessible as the free pool for that resource.
When the VF frees resources, the reverse happens: the global reserved count
for a resource is incremented only once the VFs allocated number falls below
its guaranteed number.
This fix was developed by Rick Kready <kready@us.ibm.com>
Reported-by: Rick Kready <kready@us.ibm.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- Add iWARP port mapper to avoid conflicts between RDMA and normal
stack TCP connections.
- Fixes for i386 / x86-64 structure padding differences (ABI
compatibility for 32-on-64) from Yann Droneaud.
- A pile of SRP initiator fixes from Bart Van Assche.
- Fixes for a writeback / memory allocation deadlock with NFS over
IPoIB connected mode from Jiri Kosina.
- The usual fixes and cleanups to mlx4, mlx5, cxgb4 and other
low-level drivers.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQIcBAABCAAGBQJTlzyEAAoJEENa44ZhAt0h9yoP/1UeXlejOpCJyiNdtJZ+ilcU
cb0PEzsjzqACyDqcoQ0EpQM3/3emccVIC3uUXK12mzlTIXOFYTeRLays/TbxZDLt
FK5D/NrMmmJmciPt1ZRgUX82kFFRGScEfpkXYs7jxtRaNT7CW5KwSNQr6aFXskUz
1gpdK1ARCN5rWcGl2HJx5o9C4c/Fa/Vov8lOsAkUZXD1SuPNT/fFN0u1pRzU68g0
k3oj81XnZq5ejOBQKXEHImcmjXwaJ2yjmzxhSsKebqDWDdXuS/F9e4taKneHTZmr
AdwJaLLJPWmAGi/vYYhkuLKpzIDpzMCqwr39lEabmjWvznYOlnjfVUXwUTE2nwNC
DIXuHOLFrSvF2cNxh8ZeEYKS8AV+PjAOahPC5whkWkY256Q67uB7cy9ilWAK+7xS
QcQ5Inr6iXvxIGYA4hNwUo8aK0NuKFwhkVVFEbkPaurbQZPqiKwyVE3w2FOws/Qp
0kLLCVvpRQYjKzkxyof2tb1AcNuVNKXHrYk6RaBDJ9mjxHbhvY4OSt4CBxAAXBu6
zoedUydN1Nz1UgAB1jDsBdyE2QQnXockA1+JJKNq6gM5Dz0DUdAylzQ2NqY9tnYz
RTzihEPYIiQUkV3B8ErbqsuO6z7M830AXO5AR6bLZn1zgJ0cbMLBaKLA8LRufJI/
qxNVwL32Uv1PjKZ+yX1x
=Wcdc
-----END PGP SIGNATURE-----
Merge tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband
Pull main InfiniBand/RDMA updates from Roland Dreier:
- add iWARP port mapper to avoid conflicts between RDMA and normal
stack TCP connections.
- fixes for i386 / x86-64 structure padding differences (ABI
compatibility for 32-on-64) from Yann Droneaud.
- a pile of SRP initiator fixes from Bart Van Assche.
- fixes for a writeback / memory allocation deadlock with NFS over
IPoIB connected mode from Jiri Kosina.
- the usual fixes and cleanups to mlx4, mlx5, cxgb4 and other low-level
drivers.
* tag 'rdma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband: (61 commits)
RDMA/cxgb4: Add support for iWARP Port Mapper user space service
RDMA/nes: Add support for iWARP Port Mapper user space service
RDMA/core: Add support for iWARP Port Mapper user space service
IB/mlx4: Fix gfp passing in create_qp_common()
IB/umad: Fix use-after-free on close
IB/core: Fix kobject leak on device register error flow
RDMA/cxgb4: add missing padding at end of struct c4iw_alloc_ucontext_resp
mlx4_core: Fix GFP flags parameters to be gfp_t
IB/core: Fix port kobject deletion during error flow
IB/core: Remove unneeded kobject_get/put calls
IB/core: Fix sparse warnings about redeclared functions
IB/mad: Fix sparse warning about gfp_t use
IB/mlx4: Implement IB_QP_CREATE_USE_GFP_NOIO
IB: Add a QP creation flag to use GFP_NOIO allocations
IB: Return error for unsupported QP creation flags
IB: Allow build of hw/ and ulp/ subdirectories independently
mlx4_core: Move handling of MLX4_QP_ST_MLX to proper switch statement
RDMA/cxgb4: Add missing padding at end of struct c4iw_create_cq_resp
IB/srp: Avoid problems if a header uses pr_fmt
IB/umad: Fix error handling
...
Otherwise sparse gives a bunch of warnings like
drivers/net/ethernet/mellanox/mlx4/srq.c:110:66: sparse: incorrect type in argument 4 (different base types)
drivers/net/ethernet/mellanox/mlx4/srq.c:110:66: expected int [signed] gfp
drivers/net/ethernet/mellanox/mlx4/srq.c:110:66: got restricted gfp_t
Signed-off-by: Roland Dreier <roland@purestorage.com>
Conflicts:
include/net/inetpeer.h
net/ipv6/output_core.c
Changes in net were fixing bugs in code removed in net-next.
Signed-off-by: David S. Miller <davem@davemloft.net>
Ben Hutchings says:
====================
Pull request: Fixes for new ethtool RSS commands
This addresses several problems I previously identified with the new
ETHTOOL_{G,S}RSSH commands:
1. Missing validation of reserved parameters
2. Vague documentation
3. Use of unnamed magic number
4. No consolidation with existing driver operations
I don't currently have access to suitable network hardware, but have
tested these changes with a dummy driver that can support various
combinations of operations and sizes, together with (a) Debian's ethtool
3.13 (b) ethtool 3.14 with the submitted patch to use ETHTOOL_{G,S}RSSH
and minor adjustment for fixes 1 and 3.
v2: Update RSS operations in vmxnet3 too
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
ETHTOOL_{G,S}RXFHINDIR and ETHTOOL_{G,S}RSSH should work for drivers
regardless of whether they expose the hash key, unless you try to
set a hash key for a driver that doesn't expose it.
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Modify the various routines used to allocate memory resources which
serve QPs in mlx4 to get an input GFP directive. Have the Ethernet
driver to use GFP_KERNEL in it's QP allocations as done prior to this
commit, and the IB driver to use GFP_NOIO when the IB verbs
IB_QP_CREATE_USE_GFP_NOIO QP creation flag is provided.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
The “affinity hint” mechanism is used by the user space
daemon, irqbalancer, to indicate a preferred CPU mask for irqs.
Irqbalancer can use this hint to balance the irqs between the
cpus indicated by the mask.
We wish the HCA to preferentially map the IRQs it uses to numa cores
close to it. To accomplish this, we use cpumask_set_cpu_local_first(), that
sets the affinity hint according the following policy:
First it maps IRQs to “close” numa cores. If these are exhausted, the
remaining IRQs are mapped to “far” numa cores.
Signed-off-by: Yuval Atias <yuvala@mellanox.com>
Signed-off-by: Amir Vadai <amirv@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Reset the GIDs assigned to a VF in the port RoCE GID table when
that guest goes down (either crashes or goes down cleanly).
As part of this fix, we refactor the RoCE gid table driver copy,
moving it to the mlx4_port_info structure (together with the MAC
and VLAN tables).
As with the MAC and VLAN tables, we now use a mutex per port
for the GID table so that modifying the driver copy and
modifying the firmware copy of a port GID table becomes an
atomic operation (thus avoiding driver-copy/FW-copy mismatches).
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The handling of MLX4_QP_ST_MLX in verify_qp_parameters() was
accidentally put inside the inner switch statement (that handles which
transition of RC/UC/XRC QPs is happening). Fix this by moving the case
to the outer switch statement.
The compiler pointed this out with:
drivers/net/ethernet/mellanox/mlx4/resource_tracker.c: In function 'verify_qp_parameters':
>> drivers/net/ethernet/mellanox/mlx4/resource_tracker.c:2875:3: warning: case value '7' not in enumerated type 'enum qp_transition' [-Wswitch]
case MLX4_QP_ST_MLX:
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Fixes: 99ec41d0a4 ("mlx4: Add infrastructure for selecting VFs to enable QP0 via MLX proxy QPs")
Signed-off-by: Roland Dreier <roland@purestorage.com>
This commit adds the sysfs interface for enabling QP0 on VFs for
selected VF/port.
By default, no VFs are enabled for QP0 operation.
To enable QP0 operation on a VF/port, under
/sys/class/infiniband/mlx4_x/iov/<b:d:f>/ports/x there are two new entries:
- smi_enabled (read-only). Indicates whether smi is currently
enabled for the indicated VF/port
- enable_smi_admin (rw). Used by the admin to request that smi
capability be enabled or disabled for the indicated VF/port.
0 = disable, 1 = enable.
The requested enablement will occur at the next reset of the
VF (e.g. driver restart on the VM which owns the VF).
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
This commit adds the infrastructure for enabling selected VFs to
operate SMI (QP0) MADs without restriction.
Additionally, for these enabled VFs, their QP0 proxy and tunnel QPs
are MLX QPs. As such, they operate over VL15. Therefore, they are
not affected by "credit" problems or changes in the VLArb table (which
may shut down VL0).
Non-enabled VFs may only create UD proxy QP0 qps (which are forced by
the hypervisor to send packets using the q-key it assigns and places
in the qp-context). Thus, non-enabled VFs will not pose a security
risk. The hypervisor discards any privileged MADs it receives from
these non-enabled VFs.
By default, all VFs are NOT enabled, and must explicitly be enabled
by the administrator.
The sysfs interface which operates the VF enablement infrastructure
is provided in the next commit.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Currently, VFs in SRIOV VFs are denied QP0 access. The main reason
for this decision is security, since Subnet Management Datagrams
(SMPs) are not restricted by network partitioning and may affect the
physical network topology. Moreover, even the SM may be denied access
from portions of the network by setting management keys unknown to the
SM.
However, it is desirable to grant SMI access to certain privileged
VFs, so that certain network management activities may be conducted
within virtual machines instead of the hypervisor.
This commit does the following:
1. Create QP0 tunnel QPs for all VFs.
2. Discard SMI mads sent-from/received-for non-privileged VFs in the
hypervisor MAD multiplex/demultiplex logic. SMI mads from/for
privileged VFs are allowed to pass.
3. MAD_IFC wrapper changes/fixes. For non-privileged VFs, only
host-view MAD_IFC commands are allowed, and only for SMI LID-Routed
GET mads. For privileged VFs, there are no restrictions.
This commit does not allow privileged VFs as yet. To determine if a VF
is privileged, it calls function mlx4_vf_smi_enabled(). This function
returns 0 unconditionally for now.
The next two commits allow defining and activating privileged VFs.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
