* [PATCH for-rc v2] RDMA/siw: add sendpage_ok() check to disable MSG_SPLICE_PAGES
@ 2024-10-07 12:58 Showrya M N
2024-10-11 17:00 ` Jason Gunthorpe
0 siblings, 1 reply; 2+ messages in thread
From: Showrya M N @ 2024-10-07 12:58 UTC (permalink / raw)
To: jgg, leonro, bmt; +Cc: linux-rdma, Showrya M N, Potnuri Bharat Teja
While running ISER over SIW, the initiator machine encounters a warning
from skb_splice_from_iter() indicating that a slab page is being used
in send_page. To address this, it is better to add a sendpage_ok() check
within the driver itself, and if it returns 0, then MSG_SPLICE_PAGES flag
should be disabled before entering the network stack.
A similar issue has been discussed for NVMe in this thread:
https://lore.kernel.org/all/20240530142417.146696-1-ofir.gal@volumez.com/
stack trace:
...
[ 2157.532917] WARNING: CPU: 0 PID: 5342 at net/core/skbuff.c:7140 skb_splice_from_iter+0x173/0x320
Call Trace:
[ 2157.533064] Call Trace:
[ 2157.533069] ? __warn+0x84/0x130
[ 2157.533073] ? skb_splice_from_iter+0x173/0x320
[ 2157.533075] ? report_bug+0xfc/0x1e0
[ 2157.533081] ? handle_bug+0x3f/0x70
[ 2157.533085] ? exc_invalid_op+0x17/0x70
[ 2157.533088] ? asm_exc_invalid_op+0x1a/0x20
[ 2157.533096] ? skb_splice_from_iter+0x173/0x320
[ 2157.533101] tcp_sendmsg_locked+0x368/0xe40
[ 2157.533111] siw_tx_hdt+0x695/0xa40 [siw]
[ 2157.533134] ? sched_balance_find_src_group+0x44/0x4f0
[ 2157.533143] ? __update_load_avg_cfs_rq+0x272/0x300
[ 2157.533152] ? place_entity+0x19/0xf0
[ 2157.533157] ? enqueue_entity+0xdb/0x3d0
[ 2157.533162] ? pick_eevdf+0xe2/0x120
[ 2157.533169] ? check_preempt_wakeup_fair+0x161/0x1f0
[ 2157.533174] ? wakeup_preempt+0x61/0x70
[ 2157.533177] ? ttwu_do_activate+0x5d/0x1e0
[ 2157.533183] ? try_to_wake_up+0x78/0x610
[ 2157.533188] ? xas_load+0xd/0xb0
[ 2157.533193] ? xa_load+0x80/0xb0
[ 2157.533200] siw_qp_sq_process+0x102/0xb00 [siw]
[ 2157.533213] ? __pfx_siw_run_sq+0x10/0x10 [siw]
[ 2157.533224] siw_sq_resume+0x39/0x110 [siw]
[ 2157.533236] siw_run_sq+0x74/0x160 [siw]
[ 2157.533246] ? __pfx_autoremove_wake_function+0x10/0x10
[ 2157.533252] kthread+0xd2/0x100
[ 2157.533257] ? __pfx_kthread+0x10/0x10
[ 2157.533261] ret_from_fork+0x34/0x40
[ 2157.533266] ? __pfx_kthread+0x10/0x10
[ 2157.533269] ret_from_fork_asm+0x1a/0x30
.
[ 2157.533301] iser: iser_qp_event_callback: qp event QP request error (2)
[ 2157.533307] iser: iser_qp_event_callback: qp event send queue drained (5)
[ 2157.533348] connection26:0: detected conn error (1011)
Signed-off-by: Showrya M N <showrya@chelsio.com>
Signed-off-by: Potnuri Bharat Teja <bharat@chelsio.com>
---
V1 -> V2: addressed review comment
https://lore.kernel.org/all/BN8PR15MB2513C8CC78CBEADA83117A5599712@BN8PR15MB2513.namprd15.prod.outlook.com/
---
drivers/infiniband/sw/siw/siw_qp_tx.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/infiniband/sw/siw/siw_qp_tx.c b/drivers/infiniband/sw/siw/siw_qp_tx.c
index 64ad9e0895bd..a034264c5669 100644
--- a/drivers/infiniband/sw/siw/siw_qp_tx.c
+++ b/drivers/infiniband/sw/siw/siw_qp_tx.c
@@ -331,6 +331,8 @@ static int siw_tcp_sendpages(struct socket *s, struct page **page, int offset,
msg.msg_flags &= ~MSG_MORE;
tcp_rate_check_app_limited(sk);
+ if (!sendpage_ok(page[i]))
+ msg.msg_flags &= ~MSG_SPLICE_PAGES;
bvec_set_page(&bvec, page[i], bytes, offset);
iov_iter_bvec(&msg.msg_iter, ITER_SOURCE, &bvec, 1, size);
--
2.39.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH for-rc v2] RDMA/siw: add sendpage_ok() check to disable MSG_SPLICE_PAGES
2024-10-07 12:58 [PATCH for-rc v2] RDMA/siw: add sendpage_ok() check to disable MSG_SPLICE_PAGES Showrya M N
@ 2024-10-11 17:00 ` Jason Gunthorpe
0 siblings, 0 replies; 2+ messages in thread
From: Jason Gunthorpe @ 2024-10-11 17:00 UTC (permalink / raw)
To: Showrya M N; +Cc: leonro, bmt, linux-rdma, Potnuri Bharat Teja
On Mon, Oct 07, 2024 at 06:28:36PM +0530, Showrya M N wrote:
> While running ISER over SIW, the initiator machine encounters a warning
> from skb_splice_from_iter() indicating that a slab page is being used
> in send_page. To address this, it is better to add a sendpage_ok() check
> within the driver itself, and if it returns 0, then MSG_SPLICE_PAGES flag
> should be disabled before entering the network stack.
>
> A similar issue has been discussed for NVMe in this thread:
> https://lore.kernel.org/all/20240530142417.146696-1-ofir.gal@volumez.com/
>
> stack trace:
> ...
> [ 2157.532917] WARNING: CPU: 0 PID: 5342 at net/core/skbuff.c:7140 skb_splice_from_iter+0x173/0x320
> Call Trace:
> [ 2157.533064] Call Trace:
> [ 2157.533069] ? __warn+0x84/0x130
> [ 2157.533073] ? skb_splice_from_iter+0x173/0x320
> [ 2157.533075] ? report_bug+0xfc/0x1e0
> [ 2157.533081] ? handle_bug+0x3f/0x70
> [ 2157.533085] ? exc_invalid_op+0x17/0x70
> [ 2157.533088] ? asm_exc_invalid_op+0x1a/0x20
> [ 2157.533096] ? skb_splice_from_iter+0x173/0x320
> [ 2157.533101] tcp_sendmsg_locked+0x368/0xe40
> [ 2157.533111] siw_tx_hdt+0x695/0xa40 [siw]
> [ 2157.533134] ? sched_balance_find_src_group+0x44/0x4f0
> [ 2157.533143] ? __update_load_avg_cfs_rq+0x272/0x300
> [ 2157.533152] ? place_entity+0x19/0xf0
> [ 2157.533157] ? enqueue_entity+0xdb/0x3d0
> [ 2157.533162] ? pick_eevdf+0xe2/0x120
> [ 2157.533169] ? check_preempt_wakeup_fair+0x161/0x1f0
> [ 2157.533174] ? wakeup_preempt+0x61/0x70
> [ 2157.533177] ? ttwu_do_activate+0x5d/0x1e0
> [ 2157.533183] ? try_to_wake_up+0x78/0x610
> [ 2157.533188] ? xas_load+0xd/0xb0
> [ 2157.533193] ? xa_load+0x80/0xb0
> [ 2157.533200] siw_qp_sq_process+0x102/0xb00 [siw]
> [ 2157.533213] ? __pfx_siw_run_sq+0x10/0x10 [siw]
> [ 2157.533224] siw_sq_resume+0x39/0x110 [siw]
> [ 2157.533236] siw_run_sq+0x74/0x160 [siw]
> [ 2157.533246] ? __pfx_autoremove_wake_function+0x10/0x10
> [ 2157.533252] kthread+0xd2/0x100
> [ 2157.533257] ? __pfx_kthread+0x10/0x10
> [ 2157.533261] ret_from_fork+0x34/0x40
> [ 2157.533266] ? __pfx_kthread+0x10/0x10
> [ 2157.533269] ret_from_fork_asm+0x1a/0x30
> .
> [ 2157.533301] iser: iser_qp_event_callback: qp event QP request error (2)
> [ 2157.533307] iser: iser_qp_event_callback: qp event send queue drained (5)
> [ 2157.533348] connection26:0: detected conn error (1011)
>
> Signed-off-by: Showrya M N <showrya@chelsio.com>
> Signed-off-by: Potnuri Bharat Teja <bharat@chelsio.com>
> ---
> V1 -> V2: addressed review comment
> https://lore.kernel.org/all/BN8PR15MB2513C8CC78CBEADA83117A5599712@BN8PR15MB2513.namprd15.prod.outlook.com/
> ---
>
> drivers/infiniband/sw/siw/siw_qp_tx.c | 2 ++
> 1 file changed, 2 insertions(+)
Applied to for-rc, thanks
Jason
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2024-10-11 17:00 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-10-07 12:58 [PATCH for-rc v2] RDMA/siw: add sendpage_ok() check to disable MSG_SPLICE_PAGES Showrya M N
2024-10-11 17:00 ` Jason Gunthorpe
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).