* [PATCH net-next v2 0/2] ipv6: Avoid rt6_probe() taking writer lock in the fast path @ 2015-07-24 16:57 Martin KaFai Lau 2015-07-24 16:57 ` [PATCH net-next v2 1/2] ipv6: Re-arrange code in rt6_probe() Martin KaFai Lau ` (2 more replies) 0 siblings, 3 replies; 6+ messages in thread From: Martin KaFai Lau @ 2015-07-24 16:57 UTC (permalink / raw) To: netdev; +Cc: Kernel Team v1 -> v2: 1. Separate the code re-arrangement into another patch 2. Fix style ^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH net-next v2 1/2] ipv6: Re-arrange code in rt6_probe() 2015-07-24 16:57 [PATCH net-next v2 0/2] ipv6: Avoid rt6_probe() taking writer lock in the fast path Martin KaFai Lau @ 2015-07-24 16:57 ` Martin KaFai Lau 2015-07-28 0:07 ` YOSHIFUJI Hideaki/吉藤英明 2015-07-24 16:57 ` [PATCH net-next v2 2/2] ipv6: Avoid rt6_probe() taking writer lock in the fast path Martin KaFai Lau 2015-07-27 8:08 ` [PATCH net-next v2 0/2] " David Miller 2 siblings, 1 reply; 6+ messages in thread From: Martin KaFai Lau @ 2015-07-24 16:57 UTC (permalink / raw) To: netdev Cc: Kernel Team, Hannes Frederic Sowa, Julian Anastasov, YOSHIFUJI Hideaki It is a prep work for the next patch to remove write_lock from rt6_probe(). 1. Reduce the number of if(neigh) check. From 4 to 1. 2. Bring the write_(un)lock() closer to the operations that the lock is protecting. Hopefully, the above make rt6_probe() more readable. Signed-off-by: Martin KaFai Lau <kafai@fb.com> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> Cc: Julian Anastasov <ja@ssi.bg> Cc: YOSHIFUJI Hideaki <hideaki.yoshifuji@miraclelinux.com> --- net/ipv6/route.c | 44 ++++++++++++++++++++------------------------ 1 file changed, 20 insertions(+), 24 deletions(-) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 7f2214f..6d503db 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -545,6 +545,7 @@ static void rt6_probe_deferred(struct work_struct *w) static void rt6_probe(struct rt6_info *rt) { + struct __rt6_probe_work *work; struct neighbour *neigh; /* * Okay, this does not seem to be appropriate @@ -559,34 +560,29 @@ static void rt6_probe(struct rt6_info *rt) rcu_read_lock_bh(); neigh = __ipv6_neigh_lookup_noref(rt->dst.dev, &rt->rt6i_gateway); if (neigh) { + work = NULL; write_lock(&neigh->lock); - if (neigh->nud_state & NUD_VALID) - goto out; - } - - if (!neigh || - time_after(jiffies, neigh->updated + rt->rt6i_idev->cnf.rtr_probe_interval)) { - struct __rt6_probe_work *work; - - work = kmalloc(sizeof(*work), GFP_ATOMIC); - - if (neigh && work) - __neigh_set_probe_once(neigh); - - if (neigh) - write_unlock(&neigh->lock); - - if (work) { - INIT_WORK(&work->work, rt6_probe_deferred); - work->target = rt->rt6i_gateway; - dev_hold(rt->dst.dev); - work->dev = rt->dst.dev; - schedule_work(&work->work); + if (!(neigh->nud_state & NUD_VALID) && + time_after(jiffies, + neigh->updated + + rt->rt6i_idev->cnf.rtr_probe_interval)) { + work = kmalloc(sizeof(*work), GFP_ATOMIC); + if (work) + __neigh_set_probe_once(neigh); } - } else { -out: write_unlock(&neigh->lock); + } else { + work = kmalloc(sizeof(*work), GFP_ATOMIC); + } + + if (work) { + INIT_WORK(&work->work, rt6_probe_deferred); + work->target = rt->rt6i_gateway; + dev_hold(rt->dst.dev); + work->dev = rt->dst.dev; + schedule_work(&work->work); } + rcu_read_unlock_bh(); } #else -- 1.8.1 ^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH net-next v2 1/2] ipv6: Re-arrange code in rt6_probe() 2015-07-24 16:57 ` [PATCH net-next v2 1/2] ipv6: Re-arrange code in rt6_probe() Martin KaFai Lau @ 2015-07-28 0:07 ` YOSHIFUJI Hideaki/吉藤英明 0 siblings, 0 replies; 6+ messages in thread From: YOSHIFUJI Hideaki/吉藤英明 @ 2015-07-28 0:07 UTC (permalink / raw) To: Martin KaFai Lau, netdev Cc: hideaki.yoshifuji, Kernel Team, Hannes Frederic Sowa, Julian Anastasov Martin KaFai Lau wrote: > It is a prep work for the next patch to remove write_lock > from rt6_probe(). > > 1. Reduce the number of if(neigh) check. From 4 to 1. > 2. Bring the write_(un)lock() closer to the operations that the > lock is protecting. > > Hopefully, the above make rt6_probe() more readable. > > Signed-off-by: Martin KaFai Lau <kafai@fb.com> > Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> > Cc: Julian Anastasov <ja@ssi.bg> > Cc: YOSHIFUJI Hideaki <hideaki.yoshifuji@miraclelinux.com> Acked-by: YOSHIFUJI Hideaki <hideaki.yoshifuji@miraclelinux.com> --yoshfuji > --- > net/ipv6/route.c | 44 ++++++++++++++++++++------------------------ > 1 file changed, 20 insertions(+), 24 deletions(-) > > diff --git a/net/ipv6/route.c b/net/ipv6/route.c > index 7f2214f..6d503db 100644 > --- a/net/ipv6/route.c > +++ b/net/ipv6/route.c > @@ -545,6 +545,7 @@ static void rt6_probe_deferred(struct work_struct *w) > > static void rt6_probe(struct rt6_info *rt) > { > + struct __rt6_probe_work *work; > struct neighbour *neigh; > /* > * Okay, this does not seem to be appropriate > @@ -559,34 +560,29 @@ static void rt6_probe(struct rt6_info *rt) > rcu_read_lock_bh(); > neigh = __ipv6_neigh_lookup_noref(rt->dst.dev, &rt->rt6i_gateway); > if (neigh) { > + work = NULL; > write_lock(&neigh->lock); > - if (neigh->nud_state & NUD_VALID) > - goto out; > - } > - > - if (!neigh || > - time_after(jiffies, neigh->updated + rt->rt6i_idev->cnf.rtr_probe_interval)) { > - struct __rt6_probe_work *work; > - > - work = kmalloc(sizeof(*work), GFP_ATOMIC); > - > - if (neigh && work) > - __neigh_set_probe_once(neigh); > - > - if (neigh) > - write_unlock(&neigh->lock); > - > - if (work) { > - INIT_WORK(&work->work, rt6_probe_deferred); > - work->target = rt->rt6i_gateway; > - dev_hold(rt->dst.dev); > - work->dev = rt->dst.dev; > - schedule_work(&work->work); > + if (!(neigh->nud_state & NUD_VALID) && > + time_after(jiffies, > + neigh->updated + > + rt->rt6i_idev->cnf.rtr_probe_interval)) { > + work = kmalloc(sizeof(*work), GFP_ATOMIC); > + if (work) > + __neigh_set_probe_once(neigh); > } > - } else { > -out: > write_unlock(&neigh->lock); > + } else { > + work = kmalloc(sizeof(*work), GFP_ATOMIC); > + } > + > + if (work) { > + INIT_WORK(&work->work, rt6_probe_deferred); > + work->target = rt->rt6i_gateway; > + dev_hold(rt->dst.dev); > + work->dev = rt->dst.dev; > + schedule_work(&work->work); > } > + > rcu_read_unlock_bh(); > } > #else > -- 吉藤英明 <hideaki.yoshifuji@miraclelinux.com> ミラクル・リナックス株式会社 技術本部 サポート部 ^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH net-next v2 2/2] ipv6: Avoid rt6_probe() taking writer lock in the fast path 2015-07-24 16:57 [PATCH net-next v2 0/2] ipv6: Avoid rt6_probe() taking writer lock in the fast path Martin KaFai Lau 2015-07-24 16:57 ` [PATCH net-next v2 1/2] ipv6: Re-arrange code in rt6_probe() Martin KaFai Lau @ 2015-07-24 16:57 ` Martin KaFai Lau 2015-07-28 0:17 ` YOSHIFUJI Hideaki 2015-07-27 8:08 ` [PATCH net-next v2 0/2] " David Miller 2 siblings, 1 reply; 6+ messages in thread From: Martin KaFai Lau @ 2015-07-24 16:57 UTC (permalink / raw) To: netdev Cc: Kernel Team, Hannes Frederic Sowa, Julian Anastasov, YOSHIFUJI Hideaki The patch checks neigh->nud_state before acquiring the writer lock. Note that rt6_probe() is only used in CONFIG_IPV6_ROUTER_PREF. 40 udpflood processes and a /64 gateway route are used. The gateway has NUD_PERMANENT. Each of them is run for 30s. At the end, the total number of finished sendto(): Before: 55M After: 95M Signed-off-by: Martin KaFai Lau <kafai@fb.com> Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> CC: Julian Anastasov <ja@ssi.bg> CC: YOSHIFUJI Hideaki <hideaki.yoshifuji@miraclelinux.com> --- net/ipv6/route.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 6d503db..76dcff8 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -560,6 +560,9 @@ static void rt6_probe(struct rt6_info *rt) rcu_read_lock_bh(); neigh = __ipv6_neigh_lookup_noref(rt->dst.dev, &rt->rt6i_gateway); if (neigh) { + if (neigh->nud_state & NUD_VALID) + goto out; + work = NULL; write_lock(&neigh->lock); if (!(neigh->nud_state & NUD_VALID) && @@ -583,6 +586,7 @@ static void rt6_probe(struct rt6_info *rt) schedule_work(&work->work); } +out: rcu_read_unlock_bh(); } #else -- 1.8.1 ^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH net-next v2 2/2] ipv6: Avoid rt6_probe() taking writer lock in the fast path 2015-07-24 16:57 ` [PATCH net-next v2 2/2] ipv6: Avoid rt6_probe() taking writer lock in the fast path Martin KaFai Lau @ 2015-07-28 0:17 ` YOSHIFUJI Hideaki 0 siblings, 0 replies; 6+ messages in thread From: YOSHIFUJI Hideaki @ 2015-07-28 0:17 UTC (permalink / raw) To: Martin KaFai Lau, netdev Cc: hideaki.yoshifuji, Kernel Team, Hannes Frederic Sowa, Julian Anastasov Hi, Martin KaFai Lau wrote: > The patch checks neigh->nud_state before acquiring the writer lock. > Note that rt6_probe() is only used in CONFIG_IPV6_ROUTER_PREF. > > 40 udpflood processes and a /64 gateway route are used. > The gateway has NUD_PERMANENT. Each of them is run for 30s. > At the end, the total number of finished sendto(): > > Before: 55M > After: 95M I think it is better to describe why it is okay without any locks. --yoshfuji > > Signed-off-by: Martin KaFai Lau <kafai@fb.com> > Cc: Hannes Frederic Sowa <hannes@stressinduktion.org> > CC: Julian Anastasov <ja@ssi.bg> > CC: YOSHIFUJI Hideaki <hideaki.yoshifuji@miraclelinux.com> > --- > net/ipv6/route.c | 4 ++++ > 1 file changed, 4 insertions(+) > > diff --git a/net/ipv6/route.c b/net/ipv6/route.c > index 6d503db..76dcff8 100644 > --- a/net/ipv6/route.c > +++ b/net/ipv6/route.c > @@ -560,6 +560,9 @@ static void rt6_probe(struct rt6_info *rt) > rcu_read_lock_bh(); > neigh = __ipv6_neigh_lookup_noref(rt->dst.dev, &rt->rt6i_gateway); > if (neigh) { > + if (neigh->nud_state & NUD_VALID) > + goto out; > + > work = NULL; > write_lock(&neigh->lock); > if (!(neigh->nud_state & NUD_VALID) && > @@ -583,6 +586,7 @@ static void rt6_probe(struct rt6_info *rt) > schedule_work(&work->work); > } > > +out: > rcu_read_unlock_bh(); > } > #else > -- Hideaki Yoshifuji <hideaki.yoshifuji@miraclelinux.com> Technical Division, MIRACLE LINUX CORPORATION ^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH net-next v2 0/2] ipv6: Avoid rt6_probe() taking writer lock in the fast path 2015-07-24 16:57 [PATCH net-next v2 0/2] ipv6: Avoid rt6_probe() taking writer lock in the fast path Martin KaFai Lau 2015-07-24 16:57 ` [PATCH net-next v2 1/2] ipv6: Re-arrange code in rt6_probe() Martin KaFai Lau 2015-07-24 16:57 ` [PATCH net-next v2 2/2] ipv6: Avoid rt6_probe() taking writer lock in the fast path Martin KaFai Lau @ 2015-07-27 8:08 ` David Miller 2 siblings, 0 replies; 6+ messages in thread From: David Miller @ 2015-07-27 8:08 UTC (permalink / raw) To: kafai; +Cc: netdev, kernel-team From: Martin KaFai Lau <kafai@fb.com> Date: Fri, 24 Jul 2015 09:57:41 -0700 > v1 -> v2: > 1. Separate the code re-arrangement into another patch > 2. Fix style Looks good, series applied. ^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2015-07-28 0:17 UTC | newest] Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2015-07-24 16:57 [PATCH net-next v2 0/2] ipv6: Avoid rt6_probe() taking writer lock in the fast path Martin KaFai Lau 2015-07-24 16:57 ` [PATCH net-next v2 1/2] ipv6: Re-arrange code in rt6_probe() Martin KaFai Lau 2015-07-28 0:07 ` YOSHIFUJI Hideaki/吉藤英明 2015-07-24 16:57 ` [PATCH net-next v2 2/2] ipv6: Avoid rt6_probe() taking writer lock in the fast path Martin KaFai Lau 2015-07-28 0:17 ` YOSHIFUJI Hideaki 2015-07-27 8:08 ` [PATCH net-next v2 0/2] " David Miller
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).