| Message ID | 20231016132812.63703-3-wuyun.abel@bytedance.com (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | [net-next,v2,1/3] sock: Code cleanup on __sk_mem_raise_allocated() | expand |
On Mon, Oct 16, 2023 at 6:28 AM Abel Wu <wuyun.abel@bytedance.com> wrote: > > Before sockets became aware of net-memcg's memory pressure since > commit e1aab161e013 ("socket: initial cgroup code."), the memory > usage would be granted to raise if below average even when under > protocol's pressure. This provides fairness among the sockets of > same protocol. > > That commit changes this because the heuristic will also be > effective when only memcg is under pressure which makes no sense. > Fix this by reverting to the behavior before that commit. > > After this fix, __sk_mem_raise_allocated() no longer considers > memcg's pressure. As memcgs are isolated from each other w.r.t. > memory accounting, consuming one's budget won't affect others. > So except the places where buffer sizes are needed to be tuned, > allow workloads to use the memory they are provisioned. > > Fixes: e1aab161e013 ("socket: initial cgroup code.") > Signed-off-by: Abel Wu <wuyun.abel@bytedance.com> Acked-by: Shakeel Butt <shakeelb@google.com>
On Mon, 2023-10-16 at 21:28 +0800, Abel Wu wrote: > Before sockets became aware of net-memcg's memory pressure since > commit e1aab161e013 ("socket: initial cgroup code."), the memory > usage would be granted to raise if below average even when under > protocol's pressure. This provides fairness among the sockets of > same protocol. > > That commit changes this because the heuristic will also be > effective when only memcg is under pressure which makes no sense. > Fix this by reverting to the behavior before that commit. > > After this fix, __sk_mem_raise_allocated() no longer considers > memcg's pressure. As memcgs are isolated from each other w.r.t. > memory accounting, consuming one's budget won't affect others. > So except the places where buffer sizes are needed to be tuned, > allow workloads to use the memory they are provisioned. > > Fixes: e1aab161e013 ("socket: initial cgroup code.") > Signed-off-by: Abel Wu <wuyun.abel@bytedance.com> > --- > v2: > - Ignore memcg pressure when raising memory allocated. > --- > net/core/sock.c | 14 ++++++++++++-- > 1 file changed, 12 insertions(+), 2 deletions(-) > > diff --git a/net/core/sock.c b/net/core/sock.c > index 9f969e3c2ddf..1d28e3e87970 100644 > --- a/net/core/sock.c > +++ b/net/core/sock.c > @@ -3035,7 +3035,13 @@ EXPORT_SYMBOL(sk_wait_data); > * @amt: pages to allocate > * @kind: allocation type > * > - * Similar to __sk_mem_schedule(), but does not update sk_forward_alloc > + * Similar to __sk_mem_schedule(), but does not update sk_forward_alloc. > + * > + * Unlike the globally shared limits among the sockets under same protocol, > + * consuming the budget of a memcg won't have direct effect on other ones. > + * So be optimistic about memcg's tolerance, and leave the callers to decide > + * whether or not to raise allocated through sk_under_memory_pressure() or > + * its variants. > */ > int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind) > { > @@ -3093,7 +3099,11 @@ int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind) > if (sk_has_memory_pressure(sk)) { > u64 alloc; > > - if (!sk_under_memory_pressure(sk)) > + /* The following 'average' heuristic is within the > + * scope of global accounting, so it only makes > + * sense for global memory pressure. > + */ > + if (!sk_under_global_memory_pressure(sk)) > return 1; Since the whole logic is fairly non trivial I'd like to explicitly note (for my own future memory) that I think this is the correct approach. The memcg granted the current allocation via the mem_cgroup_charge_skmem() call above, the heuristic to eventually suppress the allocation should be outside the memcg scope. LGTM, thanks! Paolo
On Mon, 2023-10-16 at 21:28 +0800, Abel Wu wrote: > Before sockets became aware of net-memcg's memory pressure since > commit e1aab161e013 ("socket: initial cgroup code."), the memory > usage would be granted to raise if below average even when under > protocol's pressure. This provides fairness among the sockets of > same protocol. > > That commit changes this because the heuristic will also be > effective when only memcg is under pressure which makes no sense. > Fix this by reverting to the behavior before that commit. > > After this fix, __sk_mem_raise_allocated() no longer considers > memcg's pressure. As memcgs are isolated from each other w.r.t. > memory accounting, consuming one's budget won't affect others. > So except the places where buffer sizes are needed to be tuned, > allow workloads to use the memory they are provisioned. > > Fixes: e1aab161e013 ("socket: initial cgroup code.") I think it's better to drop this fixes tag. This is a functional change and with such tag on at this point of the cycle, will land soon into every stable tree. That feels not appropriate. Please repost without such tag, thanks! You can send the change to stables trees later, if needed. Paolo
On 10/16/23 11:52 PM, Shakeel Butt Wrote: > On Mon, Oct 16, 2023 at 6:28 AM Abel Wu <wuyun.abel@bytedance.com> wrote: >> >> Before sockets became aware of net-memcg's memory pressure since >> commit e1aab161e013 ("socket: initial cgroup code."), the memory >> usage would be granted to raise if below average even when under >> protocol's pressure. This provides fairness among the sockets of >> same protocol. >> >> That commit changes this because the heuristic will also be >> effective when only memcg is under pressure which makes no sense. >> Fix this by reverting to the behavior before that commit. >> >> After this fix, __sk_mem_raise_allocated() no longer considers >> memcg's pressure. As memcgs are isolated from each other w.r.t. >> memory accounting, consuming one's budget won't affect others. >> So except the places where buffer sizes are needed to be tuned, >> allow workloads to use the memory they are provisioned. >> >> Fixes: e1aab161e013 ("socket: initial cgroup code.") >> Signed-off-by: Abel Wu <wuyun.abel@bytedance.com> > > Acked-by: Shakeel Butt <shakeelb@google.com> Thanks!
On 10/19/23 4:53 PM, Paolo Abeni Wrote: > On Mon, 2023-10-16 at 21:28 +0800, Abel Wu wrote: >> Before sockets became aware of net-memcg's memory pressure since >> commit e1aab161e013 ("socket: initial cgroup code."), the memory >> usage would be granted to raise if below average even when under >> protocol's pressure. This provides fairness among the sockets of >> same protocol. >> >> That commit changes this because the heuristic will also be >> effective when only memcg is under pressure which makes no sense. >> Fix this by reverting to the behavior before that commit. >> >> After this fix, __sk_mem_raise_allocated() no longer considers >> memcg's pressure. As memcgs are isolated from each other w.r.t. >> memory accounting, consuming one's budget won't affect others. >> So except the places where buffer sizes are needed to be tuned, >> allow workloads to use the memory they are provisioned. >> >> Fixes: e1aab161e013 ("socket: initial cgroup code.") > > I think it's better to drop this fixes tag. This is a functional change > and with such tag on at this point of the cycle, will land soon into > every stable tree. That feels not appropriate. > > Please repost without such tag, thanks! > > You can send the change to stables trees later, if needed. OK. Shall I add a Acked-by tag for you? Thanks! Abel
On Thu, 2023-10-19 at 19:23 +0800, Abel Wu wrote: > On 10/19/23 4:53 PM, Paolo Abeni Wrote: > > On Mon, 2023-10-16 at 21:28 +0800, Abel Wu wrote: > > > Before sockets became aware of net-memcg's memory pressure since > > > commit e1aab161e013 ("socket: initial cgroup code."), the memory > > > usage would be granted to raise if below average even when under > > > protocol's pressure. This provides fairness among the sockets of > > > same protocol. > > > > > > That commit changes this because the heuristic will also be > > > effective when only memcg is under pressure which makes no sense. > > > Fix this by reverting to the behavior before that commit. > > > > > > After this fix, __sk_mem_raise_allocated() no longer considers > > > memcg's pressure. As memcgs are isolated from each other w.r.t. > > > memory accounting, consuming one's budget won't affect others. > > > So except the places where buffer sizes are needed to be tuned, > > > allow workloads to use the memory they are provisioned. > > > > > > Fixes: e1aab161e013 ("socket: initial cgroup code.") > > > > I think it's better to drop this fixes tag. This is a functional change > > and with such tag on at this point of the cycle, will land soon into > > every stable tree. That feels not appropriate. > > > > Please repost without such tag, thanks! > > > > You can send the change to stables trees later, if needed. > > OK. Shall I add a Acked-by tag for you? Let's be formal: Acked-by: Paolo Abeni <pabeni@redhat.com> /P
diff --git a/net/core/sock.c b/net/core/sock.c index 9f969e3c2ddf..1d28e3e87970 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3035,7 +3035,13 @@ EXPORT_SYMBOL(sk_wait_data); * @amt: pages to allocate * @kind: allocation type * - * Similar to __sk_mem_schedule(), but does not update sk_forward_alloc + * Similar to __sk_mem_schedule(), but does not update sk_forward_alloc. + * + * Unlike the globally shared limits among the sockets under same protocol, + * consuming the budget of a memcg won't have direct effect on other ones. + * So be optimistic about memcg's tolerance, and leave the callers to decide + * whether or not to raise allocated through sk_under_memory_pressure() or + * its variants. */ int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind) { @@ -3093,7 +3099,11 @@ int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind) if (sk_has_memory_pressure(sk)) { u64 alloc; - if (!sk_under_memory_pressure(sk)) + /* The following 'average' heuristic is within the + * scope of global accounting, so it only makes + * sense for global memory pressure. + */ + if (!sk_under_global_memory_pressure(sk)) return 1; /* Try to be fair among all the sockets under global
Before sockets became aware of net-memcg's memory pressure since commit e1aab161e013 ("socket: initial cgroup code."), the memory usage would be granted to raise if below average even when under protocol's pressure. This provides fairness among the sockets of same protocol. That commit changes this because the heuristic will also be effective when only memcg is under pressure which makes no sense. Fix this by reverting to the behavior before that commit. After this fix, __sk_mem_raise_allocated() no longer considers memcg's pressure. As memcgs are isolated from each other w.r.t. memory accounting, consuming one's budget won't affect others. So except the places where buffer sizes are needed to be tuned, allow workloads to use the memory they are provisioned. Fixes: e1aab161e013 ("socket: initial cgroup code.") Signed-off-by: Abel Wu <wuyun.abel@bytedance.com> --- v2: - Ignore memcg pressure when raising memory allocated. --- net/core/sock.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-)