Netdev List
 help / color / mirror / Atom feed
* [PATCH net V2 0/2] net/mlx5: Fix HWS and DR steering memory safety bugs
@ 2026-11-18 17:32 Tariq Toukan
  2026-11-18 17:32 ` [PATCH net V2 1/2] net/mlx5: HWS, Handle NULL in vport destination action creation Tariq Toukan
                   ` (2 more replies)
  0 siblings, 3 replies; 6+ messages in thread
From: Tariq Toukan @ 2026-11-18 17:32 UTC (permalink / raw)
  To: Andrew Lunn, David S. Miller, Eric Dumazet, Jakub Kicinski,
	netdev, Paolo Abeni
  Cc: Alexandre Cassen, Alex Vesker, Erez Shitrit, Gal Pressman,
	Kees Cook, Leon Romanovsky, open list, linux-rdma, Mark Bloch,
	Moshe Shemesh, Prathamesh Deshpande, Saeed Mahameed, Simon Horman,
	Tariq Toukan, William Theesfeld, Yevgeny Kliteynik

Hi,

This series by Yevgeny fixes two memory safety bugs in the mlx5 software
and hardware steering code paths.

Both are missing-check or wrong-handling problems that lead to a NULL
pointer being inserted into an xarray, or an out-of-bounds write that
corrupts the kernel heap.

Regards,
Tariq

V2:
Took patch #2 out of the series.
Patch #1: Remove empty line. Updated commit message.
Patch #3 (now #2): Update commit message.
Not allocating chunks in error state.
Print error once when putting pool into error state

V1:
https://lore.kernel.org/all/20261021074949.3471704-1-tariqt@nvidia.com/

Yevgeny Kliteynik (2):
  net/mlx5: HWS, Handle NULL in vport destination action creation
  net/mlx5: DR, Fix out-of-bounds write when sync to HW fails

 .../mellanox/mlx5/core/steering/hws/fs_hws.c  |  2 ++
 .../mlx5/core/steering/sws/dr_icm_pool.c      | 32 +++++++++++++++++--
 2 files changed, 31 insertions(+), 3 deletions(-)


base-commit: aa40afa831ebfe45e40a23721711059f2ac584d4
-- 
2.44.0


^ permalink raw reply	[flat|nested] 6+ messages in thread

* [PATCH net V2 1/2] net/mlx5: HWS, Handle NULL in vport destination action creation
  2026-11-18 17:32 [PATCH net V2 0/2] net/mlx5: Fix HWS and DR steering memory safety bugs Tariq Toukan
@ 2026-11-18 17:32 ` Tariq Toukan
  2026-11-20  5:35   ` netdev-bot+sashiko
  2026-11-18 17:32 ` [PATCH net V2 2/2] net/mlx5: DR, Fix out-of-bounds write when sync to HW fails Tariq Toukan
  2026-11-18 17:37 ` [PATCH net V2 0/2] net/mlx5: Fix HWS and DR steering memory safety bugs netdev-bot+sinfo
  2 siblings, 1 reply; 6+ messages in thread
From: Tariq Toukan @ 2026-11-18 17:32 UTC (permalink / raw)
  To: Andrew Lunn, David S. Miller, Eric Dumazet, Jakub Kicinski,
	netdev, Paolo Abeni
  Cc: Alexandre Cassen, Alex Vesker, Erez Shitrit, Gal Pressman,
	Kees Cook, Leon Romanovsky, open list, linux-rdma, Mark Bloch,
	Moshe Shemesh, Prathamesh Deshpande, Saeed Mahameed, Simon Horman,
	Tariq Toukan, William Theesfeld, Yevgeny Kliteynik

From: Yevgeny Kliteynik <kliteyn@nvidia.com>

mlx5hws_action_create_dest_vport() can return NULL. If this
happens, xa_insert() will insert a NULL entry into the xarray.
The next mlx5hws_action_create_dest_vport() failure will try to
insert NULL again, but xa_insert() will return -EBUSY, because
we already have NULL entry. In such case the code will go into
error flow of action creation, and eventually will cause NULL
dereference at hws_action_destroy_hws().

The issue was discovered by code inspection.

Fixes: 8e2e08a6d1e0 ("net/mlx5: fs, add support for dest vport HWS action")
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Alex Vesker <valex@nvidia.com>
Reviewed-by: Erez Shitrit <erezsh@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c
index cbcbc2b62809..de22e7279a31 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c
@@ -441,6 +441,8 @@ mlx5_fs_get_dest_action_vport(struct mlx5_fs_hws_context *fs_ctx,
 
 	dest = mlx5hws_action_create_dest_vport(ctx, vport_num,	vhca_id_valid,
 						dest_attr->vport.vhca_id, flags);
+	if (!dest)
+		return NULL;
 
 	err = xa_insert(dests_xa, idx, dest, GFP_KERNEL);
 	if (err) {
-- 
2.44.0


^ permalink raw reply related	[flat|nested] 6+ messages in thread

* [PATCH net V2 2/2] net/mlx5: DR, Fix out-of-bounds write when sync to HW fails
  2026-11-18 17:32 [PATCH net V2 0/2] net/mlx5: Fix HWS and DR steering memory safety bugs Tariq Toukan
  2026-11-18 17:32 ` [PATCH net V2 1/2] net/mlx5: HWS, Handle NULL in vport destination action creation Tariq Toukan
@ 2026-11-18 17:32 ` Tariq Toukan
  2026-11-20  5:35   ` netdev-bot+sashiko
  2026-11-18 17:37 ` [PATCH net V2 0/2] net/mlx5: Fix HWS and DR steering memory safety bugs netdev-bot+sinfo
  2 siblings, 1 reply; 6+ messages in thread
From: Tariq Toukan @ 2026-11-18 17:32 UTC (permalink / raw)
  To: Andrew Lunn, David S. Miller, Eric Dumazet, Jakub Kicinski,
	netdev, Paolo Abeni
  Cc: Alexandre Cassen, Alex Vesker, Erez Shitrit, Gal Pressman,
	Kees Cook, Leon Romanovsky, open list, linux-rdma, Mark Bloch,
	Moshe Shemesh, Prathamesh Deshpande, Saeed Mahameed, Simon Horman,
	Tariq Toukan, William Theesfeld, Yevgeny Kliteynik

From: Yevgeny Kliteynik <kliteyn@nvidia.com>

mlx5dr_icm_free_chunk() moves a freed chunk to the "hot" chunks array by
writing it to hot_chunks_arr and, once the accumulated hot memory
crosses a threshold, triggers a sync to HW that flushes the array.  The
array is large enough to hold the chunks of a single accumulation cycle,
and relies on the sync path to flush it.

However, if sync to HW fails, hot chunks array is not cleared, and every
subsequent mlx5dr_icm_free_chunk will write past the end of
hot_chunks_arr, resulting in an out-of-bounds write and kernel heap
corruption.

A failed sync to HW also means it is no longer safe to recycle ICM: a
chunk's memory may only be returned to the buddy allocator (and thus be
reused) after a sync has confirmed that HW no longer references it.

The issue was discovered by a fuzzer that repeatedly triggered sync
failure, at which point the out-of-bounds write was reported by KASAN:

   KASAN log excerpt: BUG:
     slab-out-of-bounds in mlx5dr_icm_free_chunk+0x4cd at call 2
     (hot_chunks_num=3 > cap=2).

Fix this by putting the pool into an error state on the first failed
sync to HW, reporting the failure once. The sync is not retried, and
while the pool is in this state:

- mlx5dr_icm_free_chunk() doesn't queue freed chunks; the segment is
  left reserved and is reclaimed when the pool is destroyed.

- mlx5dr_icm_alloc_chunk() doesn't serve new chunks.

Fixes: 4519fc45beeb ("net/mlx5: DR, Keep track of hot ICM chunks in an array instead of list")
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
---
 .../mlx5/core/steering/sws/dr_icm_pool.c      | 32 +++++++++++++++++--
 1 file changed, 29 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_icm_pool.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_icm_pool.c
index 1b928b40686e..a906a72601c4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_icm_pool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_icm_pool.c
@@ -33,6 +33,11 @@ struct mlx5dr_icm_pool {
 	u64 hot_memory_size;
 	/* hot memory size threshold for triggering sync */
 	u64 th;
+	/* Set when a sync to HW fails - no new chunks are served, freed
+	 * chunks are no longer queued, and their ICM is not returned to
+	 * the buddy allocator.
+	 */
+	bool error_state;
 };
 
 struct mlx5dr_icm_dm {
@@ -368,7 +373,15 @@ static int dr_icm_pool_sync_all_buddy_pools(struct mlx5dr_icm_pool *pool)
 
 	err = mlx5dr_cmd_sync_steering(pool->dmn->mdev);
 	if (err) {
-		mlx5dr_err(pool->dmn, "Failed to sync to HW (err: %d)\n", err);
+		/* The hot chunks could not be flushed. Put the pool in an
+		 * error state so that no further chunks are queued (which
+		 * would overflow hot_chunks_arr), and no new chunks are
+		 * served. The sync is not retried.
+		 */
+		mlx5dr_err(pool->dmn,
+			   "Failed to sync to HW (err: %d), disabling ICM pool\n",
+			   err);
+		pool->error_state = true;
 		return err;
 	}
 
@@ -443,6 +456,10 @@ mlx5dr_icm_alloc_chunk(struct mlx5dr_icm_pool *pool,
 		return NULL;
 
 	mutex_lock(&pool->mutex);
+
+	if (pool->error_state)
+		goto out;
+
 	/* find mem, get back the relevant buddy pool and seg in that mem */
 	ret = dr_icm_handle_buddies_get_mem(pool, chunk_size, &buddy, &seg);
 	if (ret)
@@ -475,6 +492,14 @@ void mlx5dr_icm_free_chunk(struct mlx5dr_icm_chunk *chunk)
 	/* move the chunk to the waiting chunks array, AKA "hot" memory */
 	mutex_lock(&pool->mutex);
 
+	/* If the pool is in an error state, the chunk must not be queued.
+	 * ICM memory is not returned to the buddy allocator - it is
+	 * reclaimed only when the pool is destroyed.
+	 * Just free the chunk's bookkeeping struct.
+	 */
+	if (pool->error_state)
+		goto out;
+
 	pool->hot_memory_size += mlx5dr_icm_pool_get_chunk_byte_size(chunk);
 
 	hot_chunk = &pool->hot_chunks_arr[pool->hot_chunks_num++];
@@ -482,12 +507,13 @@ void mlx5dr_icm_free_chunk(struct mlx5dr_icm_chunk *chunk)
 	hot_chunk->seg = chunk->seg;
 	hot_chunk->size = chunk->size;
 
-	kmem_cache_free(chunks_cache, chunk);
-
 	/* Check if we have chunks that are waiting for sync-ste */
 	if (dr_icm_pool_is_sync_required(pool))
 		dr_icm_pool_sync_all_buddy_pools(pool);
 
+out:
+	kmem_cache_free(chunks_cache, chunk);
+
 	mutex_unlock(&pool->mutex);
 }
 
-- 
2.44.0


^ permalink raw reply related	[flat|nested] 6+ messages in thread

* Re: [PATCH net V2 0/2] net/mlx5: Fix HWS and DR steering memory safety bugs
  2026-11-18 17:32 [PATCH net V2 0/2] net/mlx5: Fix HWS and DR steering memory safety bugs Tariq Toukan
  2026-11-18 17:32 ` [PATCH net V2 1/2] net/mlx5: HWS, Handle NULL in vport destination action creation Tariq Toukan
  2026-11-18 17:32 ` [PATCH net V2 2/2] net/mlx5: DR, Fix out-of-bounds write when sync to HW fails Tariq Toukan
@ 2026-11-18 17:37 ` netdev-bot+sinfo
  2 siblings, 0 replies; 6+ messages in thread
From: netdev-bot+sinfo @ 2026-11-18 17:37 UTC (permalink / raw)
  To: Tariq Toukan
  Cc: Andrew Lunn, David S. Miller, Eric Dumazet, Jakub Kicinski,
	netdev, Paolo Abeni, Alexandre Cassen, Alex Vesker, Erez Shitrit,
	Gal Pressman, Kees Cook, Leon Romanovsky, open list, linux-rdma,
	Mark Bloch, Moshe Shemesh, Prathamesh Deshpande, Saeed Mahameed,
	Simon Horman, William Theesfeld, Yevgeny Kliteynik

Hi!

This is an automated message. This series looks like a fix, but its
commit messages seem to be missing some information:

 - What hardware the change was tested on. For driver fixes please
   mention the device (and if relevant firmware version) used for
   testing, or say that the change was not tested on real hardware.

Please do not repost the series just to address the above. Instead,
reply to this email with the missing information, so that reviewers
can take it into account.

*IMPORTANT*: If you do repost (for other reasons) - please include
the relevant info in the cover letter or commit messages.
-- 
The evaluation is done by an LLM so it may be wrong, if you think
that is the case please reply and explain.

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH net V2 1/2] net/mlx5: HWS, Handle NULL in vport destination action creation
  2026-11-18 17:32 ` [PATCH net V2 1/2] net/mlx5: HWS, Handle NULL in vport destination action creation Tariq Toukan
@ 2026-11-20  5:35   ` netdev-bot+sashiko
  0 siblings, 0 replies; 6+ messages in thread
From: netdev-bot+sashiko @ 2026-11-20  5:35 UTC (permalink / raw)
  To: tariqt
  Cc: andrew+netdev, davem, edumazet, kuba, netdev, pabeni, acassen,
	valex, erezsh, gal, kees+treewide, leon, linux-kernel, linux-rdma,
	mbloch, moshe, prathameshdeshpande7, saeedm, horms, william,
	kliteyn

Thank you for your contribution! Sashiko AI review found more severe
issue(s) in other patches of this series. This mail is only an FYI, in case
you want to address the nit picks below when you respin.

Low priority / nit picks for your consideration:

- [Low] The commit message for "net/mlx5: HWS, Handle NULL in vport
  destination action creation" describes only part of what the bug did.

Pre-existing issues:
- [High] This problem wasn't introduced by this patch, but the same
  caller, mlx5_fs_fte_get_hws_actions() in…

--

net/mlx5: HWS, Handle NULL in vport destination action creation

This patch makes mlx5_fs_get_dest_action_vport() return NULL as soon as
mlx5hws_action_create_dest_vport() fails. A NULL entry is therefore never
inserted into the vport destination xarray. Before this change, inserting
NULL could later cause a NULL dereference in hws_action_destroy_hws().

> The next mlx5hws_action_create_dest_vport() failure will try to
> insert NULL again, but xa_insert() will return -EBUSY, because
> we already have NULL entry. In such case the code will go into
> error flow of action creation, and eventually will cause NULL
> dereference at hws_action_destroy_hws().

[Severity: Low]
This isn't a bug, but does this describe the whole impact? The text covers
a later creation for the same idx failing again. What about a later
creation that succeeds?

When the entry is NULL, __xa_insert() stores XA_ZERO_ENTRY instead:

lib/xarray.c:__xa_insert() {
    ...
	if (!entry)
		entry = XA_ZERO_ENTRY;
    ...
}

xa_load() turns that back into NULL. The slot looks empty but stays
reserved.

Suppose a later mlx5hws_action_create_dest_vport() call succeeds, for
example once a transient kzalloc or STC pool failure has cleared. The old
code then appears to loop:

mlx5_fs_get_dest_action_vport()
  dest_load:
    xa_load() returns NULL
    mlx5hws_action_create_dest_vport() returns a valid action
    xa_insert() returns -EBUSY
    mlx5hws_action_destroy(dest)
    goto dest_load

Wouldn't this spin forever and hang rule insertion for that vport?

The fix covers both outcomes, since NULL is never inserted now. Could the
commit message also mention the livelock?

> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c
> index cbcbc2b62809d..de22e7279a312 100644
> --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c
> +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c
> @@ -441,6 +441,8 @@ mlx5_fs_get_dest_action_vport(struct mlx5_fs_hws_context *fs_ctx,
>  
>  	dest = mlx5hws_action_create_dest_vport(ctx, vport_num,	vhca_id_valid,
>  						dest_attr->vport.vhca_id, flags);
> +	if (!dest)
> +		return NULL;
>  
>  	err = xa_insert(dests_xa, idx, dest, GFP_KERNEL);
>  	if (err) {

[Severity: High]
This is a pre-existing issue, but can the same caller,
mlx5_fs_fte_get_hws_actions(), still oops on a NULL action in other
destination cases?

In three cases the result goes into fs_actions[] and is counted before
the NULL check. These are FLOW_TABLE_NUM (when the action has to be
created), RANGE and FLOW_SAMPLER:

drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c:mlx5_fs_fte_get_hws_actions() {
    ...
				dest_action = mlx5_fs_create_dest_action_table_num(fs_ctx,
										   dst);
				fs_actions[num_fs_actions++].action = dest_action;
				break;
			case MLX5_FLOW_DESTINATION_TYPE_RANGE:
				dest_action = mlx5_fs_create_dest_action_range(ctx, dst);
				fs_actions[num_fs_actions++].action = dest_action;
				break;
    ...
			case MLX5_FLOW_DESTINATION_TYPE_FLOW_SAMPLER:
				dest_action =
					mlx5_fs_get_dest_action_sampler(fs_ctx,
									dst);
				fs_actions[num_fs_actions].action = dest_action;
				fs_actions[num_fs_actions++].sampler_id =
							dst->dest_attr.sampler_id;
				break;
    ...
			if (!dest_action) {
				err = -ENOMEM;
				goto free_actions;
			}
    ...
}

Each of these can return NULL:

  mlx5hws_action_create_dest_table_num(): allocation failure or STC pool
  exhaustion

  mlx5hws_action_create_dest_match_range(): invalid range parameters, or
  allocation, definer or STC failure

  mlx5_fs_get_dest_action_sampler(): GFP_ATOMIC allocation or
  __xa_insert() failure, or failure to create the sampler action

The free_actions path then calls teardown on every counted entry:

mlx5_fs_destroy_fs_actions()
  mlx5_fs_destroy_fs_action()
    mlx5hws_action_get_type(fs_action->action)
      return action->type;   <-- NULL dereference

The other patch in this series, "net/mlx5: DR, Fix out-of-bounds write
when sync to HW fails", only touches sws/dr_icm_pool.c and does not fix
this.

Could dest_action be checked before it goes into fs_actions[]?
Alternatively, mlx5_fs_destroy_fs_actions() could skip NULL entries.

-- 
Sashiko AI review · https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20261118173227.1957670-1-tariqt%40nvidia.com

^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH net V2 2/2] net/mlx5: DR, Fix out-of-bounds write when sync to HW fails
  2026-11-18 17:32 ` [PATCH net V2 2/2] net/mlx5: DR, Fix out-of-bounds write when sync to HW fails Tariq Toukan
@ 2026-11-20  5:35   ` netdev-bot+sashiko
  0 siblings, 0 replies; 6+ messages in thread
From: netdev-bot+sashiko @ 2026-11-20  5:35 UTC (permalink / raw)
  To: tariqt
  Cc: andrew+netdev, davem, edumazet, kuba, netdev, pabeni, acassen,
	valex, erezsh, gal, kees+treewide, leon, linux-kernel, linux-rdma,
	mbloch, moshe, prathameshdeshpande7, saeedm, horms, william,
	kliteyn

Thank you for your contribution! Sashiko AI review found 2 potential
issue(s) to consider.

Critical: 0 · High: 0 · Medium: 1 · Low: 1

If you believe none of the issues are legitimate, please reply to each
issue of Medium severity or higher and explain why the AI is wrong.
If any of them are legitimate and you plan to respin, please let patchwork
know by including "pw-bot: cr" as a separate line at the end of your reply
(one such reply per series is enough).

- [Medium] One SYNC_STEERING failure now disables the ICM pool
  permanently.
- [Low] The 'KASAN log excerpt' in the commit message is not real KASAN
  output.

--

net/mlx5: DR, Fix out-of-bounds write when sync to HW fails

This patch adds an error_state flag to struct mlx5dr_icm_pool. The flag
is set when SYNC_STEERING fails in dr_icm_pool_sync_all_buddy_pools().
While it is set, mlx5dr_icm_free_chunk() stops queuing chunks into
hot_chunks_arr, which prevents the out-of-bounds write, and
mlx5dr_icm_alloc_chunk() returns NULL.

> The issue was discovered by a fuzzer that repeatedly triggered sync
> failure, at which point the out-of-bounds write was reported by KASAN:
>
>    KASAN log excerpt: BUG:
>      slab-out-of-bounds in mlx5dr_icm_free_chunk+0x4cd at call 2
>      (hot_chunks_num=3 > cap=2).

[Severity: Low]
Is this excerpt from an actual KASAN report? A real splat starts with
"BUG: KASAN: slab-out-of-bounds in func+0xoff/0xsize". It is followed
by "Write of size N at addr ... by task ..." and a stack trace. KASAN
doesn't print "at call N" or notes like "(hot_chunks_num=3 > cap=2)".

A capacity of 2 also looks hard to reach on real hardware.
mlx5dr_icm_pool_create() sizes the array like this:

	num_of_chunks = DIV_ROUND_UP(max_hot_size, entry_size) + 1;
	pool->th = max_hot_size;

Here max_hot_size is a percentage of the byte size of the largest chunk.
A capacity of 2 would need max_hot_size <= entry_size. Real device caps
make the array thousands of entries long.

The overflow itself looks real. Before this patch, a failed sync left
hot_chunks_num and hot_memory_size unchanged.

Could the commit message include the real KASAN splat? Or could it say
plainly that the overflow was found by injecting sync failures in a
test harness, and describe how they were injected?

> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_icm_pool.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_icm_pool.c
> index 1b928b40686e5..a906a72601c4f 100644
> --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_icm_pool.c
> +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/sws/dr_icm_pool.c

[ ... ]

> @@ -368,7 +373,15 @@ static int dr_icm_pool_sync_all_buddy_pools(struct mlx5dr_icm_pool *pool)
>  
>  	err = mlx5dr_cmd_sync_steering(pool->dmn->mdev);
>  	if (err) {
> -		mlx5dr_err(pool->dmn, "Failed to sync to HW (err: %d)\n", err);
> +		/* The hot chunks could not be flushed. Put the pool in an
> +		 * error state so that no further chunks are queued (which
> +		 * would overflow hot_chunks_arr), and no new chunks are
> +		 * served. The sync is not retried.
> +		 */
> +		mlx5dr_err(pool->dmn,
> +			   "Failed to sync to HW (err: %d), disabling ICM pool\n",
> +			   err);
> +		pool->error_state = true;
                ^^^^

[Severity: Medium]
Does one SYNC_STEERING failure now disable this pool for good? Only
mlx5dr_icm_pool_destroy() clears error_state, and the sync is never
retried.

After that, every mlx5dr_icm_alloc_chunk() call returns NULL. STE hash
table allocation in mlx5dr_ste_htbl_alloc(), modify action allocation
and pattern allocation in dr_ptrn.c all fail. No new steering rules can
be inserted until the domain is torn down by an eswitch mode change,
driver reload or health recovery. Each later mlx5dr_icm_free_chunk()
also leaves its segment reserved in the buddy.

mlx5dr_cmd_sync_steering() only skips the device internal error state:

	if (mdev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR)
		return 0;
	...
	return mlx5_cmd_exec_in(mdev, sync_steering, in);

That means transient host-side errors also set the flag. For example,
the SYNC_STEERING inbox is at most 16 bytes, so alloc_msg() in cmd.c
skips the message cache:

	if (in_size <= 16)
		goto cache_miss;

It then allocates with GFP_KERNEL, which can fail with -ENOMEM under
memory pressure. Command timeouts and the -ENXIO paths are passed up
the same way.

Also, error_state is per pool (STE, MODIFY_ACTION, MODIFY_HDR_PTRN), but
the sync is device-wide. A later successful sync from another pool won't
re-enable this one.

The commit message says a chunk may only return to the buddy after a
sync confirms HW no longer references it. That explains not recycling
the queued hot chunks. Does it also justify refusing to allocate from
segments that are already in the buddy free lists? Those were either
freed after an earlier successful sync or never used.

Would it work to stop queuing only when hot_chunks_num reaches the array
capacity, leaving those segments reserved, and to retry the sync on
later frees? That avoids the overflow and keeps the no-recycle rule
without disabling the pool permanently.

>  		return err;
>  	}
>  

[ ... ]

-- 
Sashiko AI review · https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20261118173227.1957670-1-tariqt%40nvidia.com

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2026-11-20  5:35 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-18 17:32 [PATCH net V2 0/2] net/mlx5: Fix HWS and DR steering memory safety bugs Tariq Toukan
2026-11-18 17:32 ` [PATCH net V2 1/2] net/mlx5: HWS, Handle NULL in vport destination action creation Tariq Toukan
2026-11-20  5:35   ` netdev-bot+sashiko
2026-11-18 17:32 ` [PATCH net V2 2/2] net/mlx5: DR, Fix out-of-bounds write when sync to HW fails Tariq Toukan
2026-11-20  5:35   ` netdev-bot+sashiko
2026-11-18 17:37 ` [PATCH net V2 0/2] net/mlx5: Fix HWS and DR steering memory safety bugs netdev-bot+sinfo

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox