[PATCH 0/2] Don't allocate ib/iw cm workqueues with WQ_MEM_RECLAIM

[PATCH 0/2] Don't allocate ib/iw cm workqueues with WQ_MEM_RECLAIM

[Sagi Grimberg]

It's not correct due to the fact that such workqueues cannot allocate
memory in their work execution by definition, which is definitly not
the case for cm workqueues.

Sagi Grimberg (2):
  cm: Don't allocate ib_cm workqueue with WQ_MEM_RECLAIM
  iwcm: Don't allocate iwcm workqueue with WQ_MEM_RECLAIM

 drivers/infiniband/core/cm.c   | 2 +-
 drivers/infiniband/core/iwcm.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

-- 
2.7.4

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 1/2] cm: Don't allocate ib_cm workqueue with WQ_MEM_RECLAIM

[Sagi Grimberg]

create_workqueue always creates the worqeueu with WQ_MEM_RECLAIM
and silences a flush dependancy warn for WQ_LEGACY. Instead, we
want to keep the warn in case the allocator tries to flush the
cm workqueue because its very likely that cm work execution will
yield memory allocations (for example cm connection requests) .

Reported-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
Signed-off-by: Sagi Grimberg <sagi-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org>
---
 drivers/infiniband/core/cm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c
index 2b4d613a3474..838e3507eadc 100644
--- a/drivers/infiniband/core/cm.c
+++ b/drivers/infiniband/core/cm.c
@@ -4201,7 +4201,7 @@ static int __init ib_cm_init(void)
 		goto error1;
 	}
 
-	cm.wq = create_workqueue("ib_cm");
+	cm.wq = alloc_workqueue("ib_cm", 0, 1);
 	if (!cm.wq) {
 		ret = -ENOMEM;
 		goto error2;
-- 
2.7.4

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

RE: [PATCH 1/2] cm: Don't allocate ib_cm workqueue with WQ_MEM_RECLAIM

[Steve Wise]

> create_workqueue always creates the worqeueu with WQ_MEM_RECLAIM

Typo "worqeueu" -> "workqueue"

> and silences a flush dependancy warn for WQ_LEGACY. Instead, we
> want to keep the warn in case the allocator tries to flush the
> cm workqueue because its very likely that cm work execution will
> yield memory allocations (for example cm connection requests) .
> 
> Reported-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
> Signed-off-by: Sagi Grimberg <sagi-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org>

The change looks good.

Reviewed-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>


--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 1/2] cm: Don't allocate ib_cm workqueue with WQ_MEM_RECLAIM

[Leon Romanovsky]

[-- Attachment #1: Type: text/plain, Size: 1070 bytes --]

On Tue, Aug 15, 2017 at 12:30:12PM -0500, Steve Wise wrote:
> > create_workqueue always creates the worqeueu with WQ_MEM_RECLAIM
>
> Typo "worqeueu" -> "workqueue"
>
> > and silences a flush dependancy warn for WQ_LEGACY. Instead, we
> > want to keep the warn in case the allocator tries to flush the
> > cm workqueue because its very likely that cm work execution will
> > yield memory allocations (for example cm connection requests) .

And extra space before the last dot.

Reviewed-by: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>


> >
> > Reported-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
> > Signed-off-by: Sagi Grimberg <sagi-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org>
>
> The change looks good.
>
> Reviewed-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

[PATCH 2/2] iwcm: Don't allocate iwcm workqueue with WQ_MEM_RECLAIM

[Sagi Grimberg]

Its very likely that iwcm work execution will yield memory
allocations (for example cm connection request).

Reported-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
Signed-off-by: Sagi Grimberg <sagi-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org>
---
 drivers/infiniband/core/iwcm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/core/iwcm.c b/drivers/infiniband/core/iwcm.c
index 31661b5c1743..ff6d7bc44c1f 100644
--- a/drivers/infiniband/core/iwcm.c
+++ b/drivers/infiniband/core/iwcm.c
@@ -1181,7 +1181,7 @@ static int __init iw_cm_init(void)
 	if (ret)
 		pr_err("iw_cm: couldn't register netlink callbacks\n");
 
-	iwcm_wq = alloc_ordered_workqueue("iw_cm_wq", WQ_MEM_RECLAIM);
+	iwcm_wq = alloc_ordered_workqueue("iw_cm_wq", 0);
 	if (!iwcm_wq)
 		return -ENOMEM;
 
-- 
2.7.4

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

RE: [PATCH 2/2] iwcm: Don't allocate iwcm workqueue with WQ_MEM_RECLAIM

[Steve Wise]

> 
> Its very likely that iwcm work execution will yield memory
> allocations (for example cm connection request).
> 
> Reported-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>
> Signed-off-by: Sagi Grimberg <sagi-NQWnxTmZq1alnMjI0IkVqw@public.gmane.org>
> ---
>  drivers/infiniband/core/iwcm.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Looks good.

Reviewed-by: Steve Wise <swise-7bPotxP6k4+P2YhJcF5u+vpXobYPEAuW@public.gmane.org>


--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html