Currently, before the qp is created, a page size needs to be calculated
for the base address table to store all base addresses in the mtr. As a
result, the parameter configuration of the mtr is complex. So integrate
the process of calculating the base table page size into the hem related
interface to simplify the process of using mtr.
Link: https://lore.kernel.org/r/1583839084-31579-5-git-send-email-liweihang@huawei.com
Signed-off-by: Xi Wang <wangxi11@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
These judgments were used to keep the compatibility with older versions of
userspace that don't have the field named "cap_flags" in structure
hns_roce_ib_create_cq_resp. But it will be wrong to compare outlen with
the size of resp if another new field were added in resp. oulen should be
compared with the end offset of cap_flags in resp.
Fixes: 4f8f0d5e33 ("RDMA/hns: Package the flow of creating cq")
Link: https://lore.kernel.org/r/1583845569-47257-1-git-send-email-liweihang@huawei.com
Signed-off-by: Weihang Li <liweihang@huawei.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>
Accesses to these members needs to be locked. There is no reason not to
hold a spinlock while calling queue_work(), so move the tests into a
helper and always call it under lock.
The helper should be called when available_mrs is adjusted.
Link: https://lore.kernel.org/r/20200310082238.239865-10-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The size_write function is supposed to adjust the total_mr's to match the
user's request, but lacks locking and safety checking.
total_mrs can only be adjusted by at most available_mrs. mrs already
assigned to users cannot be revoked. Ensure that the user provides a
target value within the range of available_mrs and within the high/low
water mark.
limit_write has confusing and wrong sanity checking, and doesn't have the
ability to deallocate on limit reduction.
Since both functions use the same algorithm to adjust the available_mrs,
consolidate it into one function and write it correctly. Fix the locking
and by holding the spinlock for all accesses to ent->X.
Always fail if the user provides a malformed string.
Fixes: e126ba97db ("mlx5: Add driver for Mellanox Connect-IB adapters")
Link: https://lore.kernel.org/r/20200310082238.239865-9-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The cache bucket tracks the total number of MRs that exists, both inside
and outside of the cache. Removing a MR from the cache (by setting
cache_ent to NULL) without updating total_mrs will cause the tracking to
leak and be inflated.
Further fix the rereg_mr path to always destroy the MR. reg_create will
always overwrite all the MR data in mlx5_ib_mr, so the MR must be
completely destroyed, in all cases, before this function can be
called. Detach the MR from the cache and unconditionally destroy it to
avoid leaking HW mkeys.
Fixes: afd1417404 ("IB/mlx5: Use direct mkey destroy command upon UMR unreg failure")
Fixes: 56e11d628c ("IB/mlx5: Added support for re-registration of MRs")
Link: https://lore.kernel.org/r/20200310082238.239865-8-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>
As mlx5_ib is the only user of the mlx5_core_create_mkey_cb, move the
logic inside mlx5_ib and cleanup the code in mlx5_core.
Signed-off-by: Michael Guralnik <michaelgur@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@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>
On reg_mr_callback() mlx5_ib is recalculating the mkey variant which is
wrong and will lead to using a different key variant than the one
submitted to firmware on create mkey command invocation.
To fix this, we store the mkey variant before invoking the firmware
command and use it later on completion (reg_mr_callback).
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Reviewed-by: Eli Cohen <eli@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Until now the flex parser capability was used in ib_query_device() to
indicate tunnel_offloads_caps support for mpls_over_gre/mpls_over_udp.
Newer devices and firmware will have configurations with the flexparser
but without mpls support.
Testing for the flex parser capability was a mistake, the tunnel_stateless
capability was intended for detecting mpls and was introduced at the same
time as the flex parser capability.
Otherwise userspace will be incorrectly informed that a future device
supports MPLS when it does not.
Link: https://lore.kernel.org/r/20200305123841.196086-1-leon@kernel.org
Cc: <stable@vger.kernel.org> # 4.17
Fixes: e818e255a5 ("IB/mlx5: Expose MPLS related tunneling offloads")
Signed-off-by: Alex Vesker <valex@mellanox.com>
Reviewed-by: Ariel Levkovich <lariel@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@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>
Commit f164be8c03 ("IB/mlx5: Extend caps stage to handle VAR
capabilities") introduced a straight "/" division of the u64 variable
"bar_size".
This was fixed with commit 685eff5131 ("IB/mlx5: Use div64_u64 for
num_var_hw_entries calculation"). However, div64_u64() is redundant here
as mlx5_var_table::stride_size is of type u32. Make the actual code way
more optimized on 32-bit kernels using div_u64() and fix 80 chars
break-through by the way.
Fixes: 685eff5131 ("IB/mlx5: Use div64_u64 for num_var_hw_entries calculation")
Link: https://lore.kernel.org/r/20200217073629.8051-1-alobakin@dlink.ru
Signed-off-by: Alexander Lobakin <alobakin@dlink.ru>
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>
Fixes gcc '-Wunused-but-set-variable' warning:
drivers/infiniband/hw/bnxt_re/qplib_rcfw.c: In function '__send_message':
drivers/infiniband/hw/bnxt_re/qplib_rcfw.c:101:10: warning:
variable 'idx' set but not used [-Wunused-but-set-variable]
drivers/infiniband/hw/bnxt_re/qplib_rcfw.c:101:6: warning:
variable 'pg' set but not used [-Wunused-but-set-variable]
commit cee0c7bba4 ("RDMA/bnxt_re: Refactor command queue management
code") involved this, but not used.
Link: https://lore.kernel.org/r/20200227064900.92255-1-yuehaibing@huawei.com
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Fixes gcc '-Wunused-but-set-variable' warning:
drivers/infiniband/hw/bnxt_re/ib_verbs.c: In function 'bnxt_re_create_gsi_qp':
drivers/infiniband/hw/bnxt_re/ib_verbs.c:1283:30: warning:
variable 'dev_attr' set but not used [-Wunused-but-set-variable]
commit 8dae419f9e ("RDMA/bnxt_re: Refactor queue pair creation code")
involved this, but not used, so remove it.
Link: https://lore.kernel.org/r/20200227064542.91205-1-yuehaibing@huawei.com
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Fixes gcc '-Wunused-but-set-variable' warning:
drivers/infiniband/hw/bnxt_re/qplib_res.c: In function '__alloc_pbl':
drivers/infiniband/hw/bnxt_re/qplib_res.c:109:13: warning:
variable 'pg_size' set but not used [-Wunused-but-set-variable]
commit 0c4dcd6028 ("RDMA/bnxt_re: Refactor hardware queue memory
allocation") involved this, but not used, so remove it.
Link: https://lore.kernel.org/r/20200227064209.87893-1-yuehaibing@huawei.com
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Using the new unregister APIs provided by the core. Provide the
dealloc_driver hook for the core to callback at the time of device
un-registration.
bnxt_re VF resources are created by the corresponding PF driver. During
ib_unregister_driver, PF might get removed before VF and this could cause
failure when VFs are removed. Driver is explicitly queuing the removal of
VF devices before calling ib_unregister_driver.
Link: https://lore.kernel.org/r/1582731932-26574-3-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
- bnxt_re_ib_reg() handles two main functionalities - initializing the
device and registering with the IB stack. Split it into 2 functions
i.e. bnxt_re_dev_init() and bnxt_re_ib_init() to account for the same
thereby improve modularity. Do the same for
bnxt_re_ib_unreg()i.e. split into two functions - bnxt_re_dev_uninit()
and bnxt_re_ib_uninit().
- Simplify the code by combining the different steps to add and remove
the device into two functions.
- Report correct netdev link state during device register
Link: https://lore.kernel.org/r/1582731932-26574-2-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The packet handling function, specifically the iteration of the qp list
for mad packet processing misses locking RCU before running through the
list. Not only is this incorrect, but the list_for_each_entry_rcu() call
can not be called with a conditional check for lock dependency. Remedy
this by invoking the rcu lock and unlock around the critical section.
This brings MAD packet processing in line with what is done for non-MAD
packets.
Fixes: 7724105686 ("IB/hfi1: add driver files")
Link: https://lore.kernel.org/r/20200225195445.140896.41873.stgit@awfm-01.aw.intel.com
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
When destroying a DMA mmapped object, there is no need to artificially
delay the freeing of the pages to the mmap entry removal. Since the vma
keeps a reference count on these pages, free_pages_exact can be called on
the destroy verb as it won't really free the pages until the reference
count is cleared (in case the user hasn't called munmap yet).
Remove the special handling of DMA pages and call free_pages_exact on
destroy_qp/cq. The mmap entry removal is moved to the beginning of the
destroy flows, so the driver can safely free the pages.
Link: https://lore.kernel.org/r/20200225114010.21790-4-galpress@amazon.com
Reviewed-by: Firas JahJah <firasj@amazon.com>
Reviewed-by: Yossi Leybovich <sleybo@amazon.com>
Signed-off-by: Gal Pressman <galpress@amazon.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
