[PATCH] RDMA/irdma: Fix erroneous -ENOMEM for large MR registration

[PATCH] RDMA/irdma: Fix erroneous -ENOMEM for large MR registration

[Jacob Moroni]

Commit cc8997c94bf3 ("RDMA/irdma: Refactor PBLE functions")
replaced the explicit level1_only boolean with a bit mask
but did not properly convert the "if (level1_only)" check
in irdma_get_pble() to be "lvl == PBLE_LEVEL_1" and instead
left it as "if (lvl)".

User MR registration passes "lvl = PBLE_LEVEL_1 | PBLE_LEVEL_2"
so this check always evaluates to true and causes the
loop to bail after grabbing a single SD even if the region
actually requires more than one SD (as is the case for
regions larger than 1 GiB with 4k pages). As a result, these
registrations can fail with -ENOMEM even if there are
sufficient PBLE resources.

Fixes: cc8997c94bf3 ("RDMA/irdma: Refactor PBLE functions")
Signed-off-by: Jacob Moroni <jmoroni@google.com>
---
 drivers/infiniband/hw/irdma/pble.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/irdma/pble.c b/drivers/infiniband/hw/irdma/pble.c
index 28dfad7f940c..14318af97d58 100644
--- a/drivers/infiniband/hw/irdma/pble.c
+++ b/drivers/infiniband/hw/irdma/pble.c
@@ -482,7 +482,7 @@ int irdma_get_pble(struct irdma_hmc_pble_rsrc *pble_rsrc,
 
 		status = get_lvl1_lvl2_pble(pble_rsrc, palloc, lvl);
 		/* if level1_only, only go through it once */
-		if (!status || lvl)
+		if (!status || lvl == PBLE_LEVEL_1)
 			break;
 	}
 
-- 
2.55.0.1032.g73a4cd73de-goog

Re: [PATCH] RDMA/irdma: Fix erroneous -ENOMEM for large MR registration

[sashiko-bot]

> Commit cc8997c94bf3 ("RDMA/irdma: Refactor PBLE functions")
> replaced the explicit level1_only boolean with a bit mask
> but did not properly convert the "if (level1_only)" check
> in irdma_get_pble() to be "lvl == PBLE_LEVEL_1" and instead
> left it as "if (lvl)".
> 
> User MR registration passes "lvl = PBLE_LEVEL_1 | PBLE_LEVEL_2"
> so this check always evaluates to true and causes the
> loop to bail after grabbing a single SD even if the region
> actually requires more than one SD (as is the case for
> regions larger than 1 GiB with 4k pages). As a result, these
> registrations can fail with -ENOMEM even if there are
> sufficient PBLE resources.
> 
> Fixes: cc8997c94bf3 ("RDMA/irdma: Refactor PBLE functions")
> Signed-off-by: Jacob Moroni <jmoroni@google.com>

Sashiko has reviewed this patch and found no issues. It looks great!

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260915170317.2901672-1-jmoroni@google.com?part=1

Re: [PATCH] RDMA/irdma: Fix erroneous -ENOMEM for large MR registration

[Leon Romanovsky]

On Tue, 15 Sep 2026 17:03:17 +0000, Jacob Moroni wrote:
> Commit cc8997c94bf3 ("RDMA/irdma: Refactor PBLE functions")
> replaced the explicit level1_only boolean with a bit mask
> but did not properly convert the "if (level1_only)" check
> in irdma_get_pble() to be "lvl == PBLE_LEVEL_1" and instead
> left it as "if (lvl)".
> 
> User MR registration passes "lvl = PBLE_LEVEL_1 | PBLE_LEVEL_2"
> so this check always evaluates to true and causes the
> loop to bail after grabbing a single SD even if the region
> actually requires more than one SD (as is the case for
> regions larger than 1 GiB with 4k pages). As a result, these
> registrations can fail with -ENOMEM even if there are
> sufficient PBLE resources.
> 
> [...]

Applied, thanks!

[1/1] RDMA/irdma: Fix erroneous -ENOMEM for large MR registration
      https://git.kernel.org/rdma/rdma/c/5bb710151753fe

Best regards,
-- 
Leon Romanovsky <leon@kernel.org>