From: Vlastimil Babka <vbabka@suse.cz> To: Shaohua Li <shli@fb.com>, netdev@vger.kernel.org Cc: davem@davemloft.net, Kernel-team@fb.com, clm@fb.com, linux-mm@kvack.org, dbavatar@gmail.com, Eric Dumazet <edumazet@google.com> Subject: Re: [RFC V3] net: don't wait for order-3 page allocation Date: Fri, 12 Jun 2015 11:36:52 +0200 [thread overview] Message-ID: <557AA834.8070503@suse.cz> (raw) In-Reply-To: <0099265406c32b9b9057de100404a4148d602cdd.1434066549.git.shli@fb.com> On 06/12/2015 01:50 AM, Shaohua Li wrote: > 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@google.com> > Cc: Chris Mason <clm@fb.com> > Cc: Debabrata Banerjee <dbavatar@gmail.com> > Signed-off-by: Shaohua Li <shli@fb.com> Acked-by: Vlastimil Babka <vbabka@suse.cz> > --- > net/core/skbuff.c | 2 +- > net/core/sock.c | 2 +- > 2 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/net/core/skbuff.c b/net/core/skbuff.c > index 3cfff2a..41ec022 100644 > --- a/net/core/skbuff.c > +++ b/net/core/skbuff.c > @@ -4398,7 +4398,7 @@ struct sk_buff *alloc_skb_with_frags(unsigned long header_len, > > while (order) { > if (npages >= 1 << order) { > - page = alloc_pages(gfp_mask | > + page = alloc_pages((gfp_mask & ~__GFP_WAIT) | > __GFP_COMP | > __GFP_NOWARN | > __GFP_NORETRY, Note that __GFP_NORETRY is weaker than ~__GFP_WAIT and thus redundant. But it won't hurt anything leaving it there. And you might consider __GFP_NO_KSWAPD instead, as I said in the other thread. > diff --git a/net/core/sock.c b/net/core/sock.c > index 292f422..e9855a4 100644 > --- a/net/core/sock.c > +++ b/net/core/sock.c > @@ -1883,7 +1883,7 @@ bool skb_page_frag_refill(unsigned int sz, struct page_frag *pfrag, gfp_t gfp) > > pfrag->offset = 0; > if (SKB_FRAG_PAGE_ORDER) { > - pfrag->page = alloc_pages(gfp | __GFP_COMP | > + pfrag->page = alloc_pages((gfp & ~__GFP_WAIT) | __GFP_COMP | > __GFP_NOWARN | __GFP_NORETRY, > SKB_FRAG_PAGE_ORDER); > if (likely(pfrag->page)) { > -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
WARNING: multiple messages have this Message-ID (diff)
From: Vlastimil Babka <vbabka@suse.cz> To: Shaohua Li <shli@fb.com>, netdev@vger.kernel.org Cc: davem@davemloft.net, Kernel-team@fb.com, clm@fb.com, linux-mm@kvack.org, dbavatar@gmail.com, Eric Dumazet <edumazet@google.com> Subject: Re: [RFC V3] net: don't wait for order-3 page allocation Date: Fri, 12 Jun 2015 11:36:52 +0200 [thread overview] Message-ID: <557AA834.8070503@suse.cz> (raw) In-Reply-To: <0099265406c32b9b9057de100404a4148d602cdd.1434066549.git.shli@fb.com> On 06/12/2015 01:50 AM, Shaohua Li wrote: > 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@google.com> > Cc: Chris Mason <clm@fb.com> > Cc: Debabrata Banerjee <dbavatar@gmail.com> > Signed-off-by: Shaohua Li <shli@fb.com> Acked-by: Vlastimil Babka <vbabka@suse.cz> > --- > net/core/skbuff.c | 2 +- > net/core/sock.c | 2 +- > 2 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/net/core/skbuff.c b/net/core/skbuff.c > index 3cfff2a..41ec022 100644 > --- a/net/core/skbuff.c > +++ b/net/core/skbuff.c > @@ -4398,7 +4398,7 @@ struct sk_buff *alloc_skb_with_frags(unsigned long header_len, > > while (order) { > if (npages >= 1 << order) { > - page = alloc_pages(gfp_mask | > + page = alloc_pages((gfp_mask & ~__GFP_WAIT) | > __GFP_COMP | > __GFP_NOWARN | > __GFP_NORETRY, Note that __GFP_NORETRY is weaker than ~__GFP_WAIT and thus redundant. But it won't hurt anything leaving it there. And you might consider __GFP_NO_KSWAPD instead, as I said in the other thread. > diff --git a/net/core/sock.c b/net/core/sock.c > index 292f422..e9855a4 100644 > --- a/net/core/sock.c > +++ b/net/core/sock.c > @@ -1883,7 +1883,7 @@ bool skb_page_frag_refill(unsigned int sz, struct page_frag *pfrag, gfp_t gfp) > > pfrag->offset = 0; > if (SKB_FRAG_PAGE_ORDER) { > - pfrag->page = alloc_pages(gfp | __GFP_COMP | > + pfrag->page = alloc_pages((gfp & ~__GFP_WAIT) | __GFP_COMP | > __GFP_NOWARN | __GFP_NORETRY, > SKB_FRAG_PAGE_ORDER); > if (likely(pfrag->page)) { >
next prev parent reply other threads:[~2015-06-12 9:36 UTC|newest] Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top 2015-06-11 23:50 [RFC V3] net: don't wait for order-3 page allocation Shaohua Li 2015-06-11 23:50 ` Shaohua Li 2015-06-12 0:02 ` Eric Dumazet 2015-06-12 0:02 ` Eric Dumazet 2015-06-12 0:34 ` David Miller 2015-06-12 0:34 ` David Miller 2015-06-12 9:36 ` Vlastimil Babka [this message] 2015-06-12 9:36 ` Vlastimil Babka 2015-06-17 23:02 ` David Rientjes 2015-06-17 23:02 ` David Rientjes 2015-06-18 14:30 ` Michal Hocko 2015-06-18 14:30 ` Michal Hocko 2015-06-18 14:35 ` Eric Dumazet 2015-06-18 14:43 ` Michal Hocko 2015-06-18 14:43 ` Michal Hocko 2015-06-18 15:22 ` Vlastimil Babka 2015-06-18 15:47 ` Michal Hocko 2015-06-18 15:47 ` Michal Hocko 2015-06-30 23:49 ` David Rientjes 2015-06-30 23:49 ` David Rientjes
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=557AA834.8070503@suse.cz \ --to=vbabka@suse.cz \ --cc=Kernel-team@fb.com \ --cc=clm@fb.com \ --cc=davem@davemloft.net \ --cc=dbavatar@gmail.com \ --cc=edumazet@google.com \ --cc=linux-mm@kvack.org \ --cc=netdev@vger.kernel.org \ --cc=shli@fb.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.