The mlx5 VF driver doesn't set QP tx port affinity because it doesn't know
if the lag is active or not, since the "lag_active" works only for PF
interfaces. In this case for VF interfaces only one lag is used which
brings performance issue.
Add a lag_tx_port_affinity CAP bit; When it is enabled and
"num_lag_ports > 1", then driver always set QP tx affinity, regardless
of lag state.
Link: https://lore.kernel.org/r/20200527055014.355093-1-leon@kernel.org
Signed-off-by: Mark Zhang <markz@mellanox.com>
Reviewed-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The patch sets the lag tx affinity of the data QPs and the GSI QPs
according to the LAG xmit slave.
For GSI QPs, in case the link layer is Ethenet (RoCE) we create two GSI
QPs, one for each physical port. When the driver selects the GSI QP, it
will consider the port affinity result. For connected QPs, the driver
sets the affinity of the xmit slave.
The above, ensures that RC QP and it's corresponding GSI QP will transmit
from the same physical port.
Link: https://lore.kernel.org/r/20200430192146.12863-17-maorg@mellanox.com
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Instead of keeping special field to separate kernel/user create/destroy
flows, rely on existence of udata pointer. All allocation flows are
using kzalloc() and leave uninitialized pointers as NULL which makes
MLX5_QP_EMPTY and MLX5_QP_KERNEL flows to be the same.
Link: https://lore.kernel.org/r/20200427154636.381474-25-leon@kernel.org
Reviewed-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
QP type is stored in the IB/core QP struct, but it doesn't have all the
needed information, like internal QP type used in the driver itself.
Update mlx5_ib to have cached QP type which includes both IBTA and
Mellanox specific one.
Such change allows us to make even further cleanup of QP creation flow.
Link: https://lore.kernel.org/r/20200427154636.381474-21-leon@kernel.org
Reviewed-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The mlx5_core doesn't need any functionality coded in qp.c, so move
that file to drivers/infiniband/ be under mlx5_ib responsibility.
Reviewed-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Remove mlx5_ib implementation of Q counter allocation logic
together with cleaning boolean which controlled validity of the
counter. It is not needed, because counter_id == 0 means that
counter is not valid.
Reviewed-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Pull rdma updates from Jason Gunthorpe:
"The majority of the patches are cleanups, refactorings and clarity
improvements.
This cycle saw some more activity from Syzkaller, I think we are now
clean on all but one of those bugs, including the long standing and
obnoxious rdma_cm locking design defect. Continue to see many drivers
getting cleanups, with a few new user visible features.
Summary:
- Various driver updates for siw, bnxt_re, rxe, efa, mlx5, hfi1
- Lots of cleanup patches for hns
- Convert more places to use refcount
- Aggressively lock the RDMA CM code that syzkaller says isn't
working
- Work to clarify ib_cm
- Use the new ib_device lifecycle model in bnxt_re
- Fix mlx5's MR cache which seems to be failing more often with the
new ODP code
- mlx5 'dynamic uar' and 'tx steering' user interfaces"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (144 commits)
RDMA/bnxt_re: make bnxt_re_ib_init static
IB/qib: Delete struct qib_ivdev.qp_rnd
RDMA/hns: Fix uninitialized variable bug
RDMA/hns: Modify the mask of QP number for CQE of hip08
RDMA/hns: Reduce the maximum number of extend SGE per WQE
RDMA/hns: Reduce PFC frames in congestion scenarios
RDMA/mlx5: Add support for RDMA TX flow table
net/mlx5: Add support for RDMA TX steering
IB/hfi1: Call kobject_put() when kobject_init_and_add() fails
IB/hfi1: Fix memory leaks in sysfs registration and unregistration
IB/mlx5: Move to fully dynamic UAR mode once user space supports it
IB/mlx5: Limit the scope of struct mlx5_bfreg_info to mlx5_ib
IB/mlx5: Extend QP creation to get uar page index from user space
IB/mlx5: Extend CQ creation to get uar page index from user space
IB/mlx5: Expose UAR object and its alloc/destroy commands
IB/hfi1: Get rid of a warning
RDMA/hns: Remove redundant judgment of qp_type
RDMA/hns: Remove redundant assignment of wc->smac when polling cq
RDMA/hns: Remove redundant qpc setup operations
RDMA/hns: Remove meaningless prints
...
* 'mlx5-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux:
mlx5: Remove uninitialized use of key in mlx5_core_create_mkey
{IB,net}/mlx5: Move asynchronous mkey creation to mlx5_ib
{IB,net}/mlx5: Assign mkey variant in mlx5_ib only
{IB,net}/mlx5: Setup mkey variant before mr create command invocation
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Leon Romanovsky says:
====================
Those two patches from Michael extends mlx5_core and mlx5_ib flow steering
to support RDMA TX in similar way to already supported RDMA RX.
====================
Based on the mlx5-next branch at
git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux
Due to dependencies
* branch 'mlx5_tx_steering':
RDMA/mlx5: Add support for RDMA TX flow table
net/mlx5: Add support for RDMA TX steering
Expose UAR object and its alloc/destroy commands to be used over the ioctl
interface by user space applications.
This API supports both BF & NC modes and enables a dynamic allocation of
UARs once really needed.
As the number of driver objects were limited by the core ones when the
merged tree is prepared, had to decrease the number of core objects to
enable the new UAR object usage.
Link: https://lore.kernel.org/r/20200324060143.1569116-2-leon@kernel.org
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Reviewed-by: Michael Guralnik <michaelgur@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Leon Romanovsky says:
====================
This series fixes various corner cases in the mlx5_ib MR cache
implementation, see specific commit messages for more information.
====================
Based on the mlx5-next branch at
git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux
Due to dependencies
* branch 'mlx5_mr-cache':
RDMA/mlx5: Allow MRs to be created in the cache synchronously
RDMA/mlx5: Revise how the hysteresis scheme works for cache filling
RDMA/mlx5: Fix locking in MR cache work queue
RDMA/mlx5: Lock access to ent->available_mrs/limit when doing queue_work
RDMA/mlx5: Fix MR cache size and limit debugfs
RDMA/mlx5: Always remove MRs from the cache before destroying them
RDMA/mlx5: Simplify how the MR cache bucket is located
RDMA/mlx5: Rename the tracking variables for the MR cache
RDMA/mlx5: Replace spinlock protected write with atomic var
{IB,net}/mlx5: Move asynchronous mkey creation to mlx5_ib
{IB,net}/mlx5: Assign mkey variant in mlx5_ib only
{IB,net}/mlx5: Setup mkey variant before mr create command invocation
If the cache is completely out of MRs, and we are running in cache mode,
then directly, and synchronously, create an MR that is compatible with the
cache bucket using a sleeping mailbox command. This ensures that the
thread that is waiting for the MR absolutely will get one.
When a MR allocated in this way becomes freed then it is compatible with
the cache bucket and will be recycled back into it.
Deletes the very buggy ent->compl scheme to create a synchronous MR
allocation.
Link: https://lore.kernel.org/r/20200310082238.239865-13-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Currently if the work queue is running then it is in 'hysteresis' mode and
will fill until the cache reaches the high water mark. This implicit state
is very tricky and doesn't interact with pending very well.
Instead of self re-scheduling the work queue after the add_keys() has
started to create the new MR, have the queue scheduled from
reg_mr_callback() only after the requested MR has been added.
This avoids the bad design of an in-rush of queue'd work doing back to
back add_keys() until EAGAIN then sleeping. The add_keys() will be paced
one at a time as they complete, slowly filling up the cache.
Also, fix pending to be only manipulated under lock.
Link: https://lore.kernel.org/r/20200310082238.239865-12-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
All of the members of mlx5_cache_ent must be accessed while holding the
spinlock, add the missing spinlock in the __cache_work_func().
Using cache->stopped and flush_workqueue() is an inherently racy way to
shutdown self-scheduling work on a queue. Replace it with ent->disabled
under lock, and always check disabled before queuing any new work. Use
cancel_work_sync() to shutdown the queue.
Use READ_ONCE/WRITE_ONCE for dev->last_add to manage concurrency as
coherency is less important here.
Split fill_delay from the bitfield. C bitfield updates are not atomic and
this is just a mess. Use READ_ONCE/WRITE_ONCE, but this could also use
test_bit()/set_bit().
Link: https://lore.kernel.org/r/20200310082238.239865-11-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
There are many bad APIs here that are accepting a cache bucket index
instead of a bucket pointer. Many of the callers already have a bucket
pointer, so this results in a lot of confusing uses of order2idx().
Pass the struct mlx5_cache_ent into add_keys(), remove_keys(), and
alloc_cached_mr().
Once the MR is in the cache, store the cache bucket pointer directly in
the MR, replacing the 'bool allocated_from cache'.
In the end there is only one place that needs to form index from order,
alloc_mr_from_cache(). Increase the safety of this function by disallowing
it from accessing cache entries in the ODP special area.
Link: https://lore.kernel.org/r/20200310082238.239865-7-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
mkey variant is not required for mlx5_core use, move the mkey variant
counter to mlx5_ib.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Yishai Hadas Says:
====================
Expose raw packet pacing APIs to be used by DEVX based applications. The
existing code was refactored to have a single flow with the new raw APIs.
====================
Based on the mlx5-next branch at
git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux
Due to dependencies
* branch 'mlx5_packet_pacing':
IB/mlx5: Introduce UAPIs to manage packet pacing
net/mlx5: Expose raw packet pacing APIs
Add two debugfs parameters described below.
np_min_time_between_cnps - Minimum time between sending CNPs from the
port.
Unit = microseconds.
Default = 0 (no min wait time; generated
based on incoming ECN marked packets).
rp_max_rate - Maximum rate at which reaction point node can transmit.
Once this limit is reached, RP is no longer rate limited.
Unit = Mbits/sec
Default = 0 (full speed)
Link: https://lore.kernel.org/r/20200227125246.99472-1-leon@kernel.org
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Following race may occur because of the call_srcu and the placement of
the synchronize_srcu vs the xa_erase.
CPU0 CPU1
mlx5_ib_free_implicit_mr: destroy_unused_implicit_child_mr:
xa_erase(odp_mkeys)
synchronize_srcu()
xa_lock(implicit_children)
if (still in xarray)
atomic_inc()
call_srcu()
xa_unlock(implicit_children)
xa_erase(implicit_children):
xa_lock(implicit_children)
__xa_erase()
xa_unlock(implicit_children)
flush_workqueue()
[..]
free_implicit_child_mr_rcu:
(via call_srcu)
queue_work()
WARN_ON(atomic_read())
[..]
free_implicit_child_mr_work:
(via wq)
free_implicit_child_mr()
mlx5_mr_cache_invalidate()
mlx5_ib_update_xlt() <-- UMR QP fail
atomic_dec()
The wait_event() solves the race because it blocks until
free_implicit_child_mr_work() completes.
Fixes: 5256edcb98 ("RDMA/mlx5: Rework implicit ODP destroy")
Link: https://lore.kernel.org/r/20200227113918.94432-1-leon@kernel.org
Signed-off-by: Artemy Kovalyov <artemyko@mellanox.com>
Reviewed-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Relaxed ordering is not supported in UMR so we are disabling UMR usage
when user passes relaxed ordering access flag.
Enable using UMR when user requested relaxed ordering but there are no
relaxed ordering capabilities.
This will prevent user from unnecessarily registering a new mkey.
Fixes: d6de0bb185 ("RDMA/mlx5: Set relaxed ordering when requested")
Link: https://lore.kernel.org/r/20200227113834.94233-1-leon@kernel.org
Signed-off-by: Michael Guralnik <michaelgur@mellanox.com>
Reviewed-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Compilation of mlx5 driver without CONFIG_INFINIBAND_USER_ACCESS generates
the following error.
on x86_64:
ld: drivers/infiniband/hw/mlx5/main.o: in function `mlx5_ib_handler_MLX5_IB_METHOD_VAR_OBJ_ALLOC':
main.c:(.text+0x186d): undefined reference to `ib_uverbs_get_ucontext_file'
ld: drivers/infiniband/hw/mlx5/main.o:(.rodata+0x2480): undefined reference to `uverbs_idr_class'
ld: drivers/infiniband/hw/mlx5/main.o:(.rodata+0x24d8): undefined reference to `uverbs_destroy_def_handler'
This is happening because some parts of the UAPI description are not
static. This is a hold over from earlier code that relied on struct
pointers to refer to object types, now object types are referenced by
number. Remove the unused globals and add statics to the remaining UAPI
description elements.
Remove the redundent #ifdefs around mlx5_ib_*defs and obsolete
mlx5_ib_get_devx_tree().
The compiler now trims alot more unused code, including the above
problematic definitions when !CONFIG_INFINIBAND_USER_ACCESS.
Fixes: 7be76bef32 ("IB/mlx5: Introduce VAR object and its alloc/destroy methods")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
From https://git.kernel.org/pub/scm/linux/kernel/git/leon/linux-rdma
Leon Romanovsky says:
====================
Use ODP MRs for kernel ULPs
The following series extends MR creation routines to allow creation of
user MRs through kernel ULPs as a proxy. The immediate use case is to
allow RDS to work over FS-DAX, which requires ODP (on-demand-paging)
MRs to be created and such MRs were not possible to create prior this
series.
The first part of this patchset extends RDMA to have special verb
ib_reg_user_mr(). The common use case that uses this function is a
userspace application that allocates memory for HCA access but the
responsibility to register the memory at the HCA is on an kernel ULP.
This ULP acts as an agent for the userspace application.
The second part provides advise MR functionality for ULPs. This is
integral part of ODP flows and used to trigger pagefaults in advance
to prepare memory before running working set.
The third part is actual user of those in-kernel APIs.
====================
* tag 'rds-odp-for-5.5':
net/rds: Use prefetch for On-Demand-Paging MR
net/rds: Handle ODP mr registration/unregistration
net/rds: Detect need of On-Demand-Paging memory registration
RDMA/mlx5: Fix handling of IOVA != user_va in ODP paths
IB/mlx5: Mask out unsupported ODP capabilities for kernel QPs
RDMA/mlx5: Don't fake udata for kernel path
IB/mlx5: Add ODP WQE handlers for kernel QPs
IB/core: Add interface to advise_mr for kernel users
IB/core: Introduce ib_reg_user_mr
IB: Allow calls to ib_umem_get from kernel ULPs
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
