[3.16.y-ckt stable] Patch "net: don't wait for order-3 page allocation" has been added to staging queue
Luis Henriques
luis.henriques at canonical.com
Mon Jul 13 09:20:03 UTC 2015
This is a note to let you know that I have just added a patch titled
net: don't wait for order-3 page allocation
to the linux-3.16.y-queue branch of the 3.16.y-ckt extended stable tree
which can be found at:
http://kernel.ubuntu.com/git/ubuntu/linux.git/log/?h=linux-3.16.y-queue
This patch is scheduled to be released in version 3.16.7-ckt15.
If you, or anyone else, feels it should not be added to this tree, please
reply to this email.
For more information about the 3.16.y-ckt tree, see
https://wiki.ubuntu.com/Kernel/Dev/ExtendedStable
Thanks.
-Luis
------
>From c93e1ae4ce72ade36a43f8ba55bcce13ca4d2ef0 Mon Sep 17 00:00:00 2001
From: Shaohua Li <shli at fb.com>
Date: Thu, 11 Jun 2015 16:50:48 -0700
Subject: net: don't wait for order-3 page allocation
commit fb05e7a89f500cfc06ae277bdc911b281928995d upstream.
We saw excessive direct memory compaction triggered by skb_page_frag_refill.
This causes performance issues and add latency. Commit 5640f7685831e0
introduces the order-3 allocation. According to the changelog, the order-3
allocation isn't a must-have but to improve performance. But direct memory
compaction has high overhead. The benefit of order-3 allocation can't
compensate the overhead of direct memory compaction.
This patch makes the order-3 page allocation atomic. If there is no memory
pressure and memory isn't fragmented, the alloction will still success, so we
don't sacrifice the order-3 benefit here. If the atomic allocation fails,
direct memory compaction will not be triggered, skb_page_frag_refill will
fallback to order-0 immediately, hence the direct memory compaction overhead is
avoided. In the allocation failure case, kswapd is waken up and doing
compaction, so chances are allocation could success next time.
alloc_skb_with_frags is the same.
The mellanox driver does similar thing, if this is accepted, we must fix
the driver too.
V3: fix the same issue in alloc_skb_with_frags as pointed out by Eric
V2: make the changelog clearer
Cc: Eric Dumazet <edumazet at google.com>
Cc: Chris Mason <clm at fb.com>
Cc: Debabrata Banerjee <dbavatar at gmail.com>
Signed-off-by: Shaohua Li <shli at fb.com>
Acked-by: Eric Dumazet <edumazet at google.com>
Signed-off-by: David S. Miller <davem at davemloft.net>
[ luis: backported to 3.16: used davem's backport to 3.14 ]
Signed-off-by: Luis Henriques <luis.henriques at canonical.com>
---
net/core/skbuff.c | 4 +++-
net/core/sock.c | 4 +++-
2 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index dc27721ece4d..167a92c896b9 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -368,9 +368,11 @@ refill:
for (order = NETDEV_FRAG_PAGE_MAX_ORDER; ;) {
gfp_t gfp = gfp_mask;
- if (order)
+ if (order) {
gfp |= __GFP_COMP | __GFP_NOWARN |
__GFP_NOMEMALLOC;
+ gfp &= ~__GFP_WAIT;
+ }
nc->frag.page = alloc_pages(gfp, order);
if (likely(nc->frag.page))
break;
diff --git a/net/core/sock.c b/net/core/sock.c
index a6ddd4ada315..9956e854d0e6 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1914,8 +1914,10 @@ bool skb_page_frag_refill(unsigned int sz, struct page_frag *pfrag, gfp_t prio)
do {
gfp_t gfp = prio;
- if (order)
+ if (order) {
gfp |= __GFP_COMP | __GFP_NOWARN | __GFP_NORETRY;
+ gfp &= ~__GFP_WAIT;
+ }
pfrag->page = alloc_pages(gfp, order);
if (likely(pfrag->page)) {
pfrag->offset = 0;
More information about the kernel-team
mailing list