[prev in list] [next in list] [prev in thread] [next in thread] 

List:       openvz-devel
Subject:    [Devel] [PATCH rh7] sctp: suppress warnings on high order memory allocation on sctp chunk creation
From:       Konstantin Khorenko <khorenko () virtuozzo ! com>
Date:       2019-10-31 10:08:34
Message-ID: 20191031100834.16101-1-khorenko () virtuozzo ! com
[Download RAW message or body]

_sctp_make_chunk() could allocate skbs up to 65K in size
(4th order page), so we get complains about high order page
allocations from our detection mechanizm while running sctp
testsuite.

It's not evident that substituting alloc_skb() with
alloc_skb_with_frags() speeds up the sender performance.
On another hand if the node's memory is highly fragmented
and allocation of high order pages has noticable latency,
it influences the sender most, not the node as a whole.

And taking into account that sctp is not widely used,
let's just shut up the warning in this place.

https://jira.sw.ru/browse/PSBM-93798

Signed-off-by: Konstantin Khorenko <khorenko@virtuozzo.com>
---
 net/sctp/sm_make_chunk.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
index 8fef389dd2fa..4611e85ace9f 100644
--- a/net/sctp/sm_make_chunk.c
+++ b/net/sctp/sm_make_chunk.c
@@ -1408,7 +1408,7 @@ static struct sctp_chunk *_sctp_make_chunk(const struct sctp_association *asoc,
 		goto nodata;
 
 	/* No need to allocate LL here, as this is only a chunk. */
-	skb = alloc_skb(chunklen, gfp);
+	skb = alloc_skb(chunklen, gfp | __GFP_ORDER_NOWARN);
 	if (!skb)
 		goto nodata;
 
-- 
2.15.1

_______________________________________________
Devel mailing list
Devel@openvz.org
https://lists.openvz.org/mailman/listinfo/devel
[prev in list] [next in list] [prev in thread] [next in thread]