On Thu 24 Jan 2019 at 17:21, Dennis Zhou wrote:
> Hi Vlad and Eric,
>
> On Tue, Jan 22, 2019 at 09:33:10AM -0800, Eric Dumazet wrote:
>> On Mon, Jan 21, 2019 at 3:24 AM Vlad Buslov wrote:
>> >
>> > Hi Eric,
>> >
>> > I've been invest
Hi Vlad and Eric,
On Tue, Jan 22, 2019 at 09:33:10AM -0800, Eric Dumazet wrote:
> On Mon, Jan 21, 2019 at 3:24 AM Vlad Buslov wrote:
> >
> > Hi Eric,
> >
> > I've been investigating significant tc filter insertion rate degradation
> > and it seems it is ca
On Tue, Jan 22, 2019 at 1:18 PM Tejun Heo wrote:
>
> Hello,
>
> Percpu storage is expensive and cache line sharing tends to be less of
> a problem (cuz they're per-cpu), so it is useful to support custom
> alignments for tighter packing.
>
We have BPF percpu maps of two 8-byte counters (packet
Hello,
On Tue, Jan 22, 2019 at 09:33:10AM -0800, Eric Dumazet wrote:
> > 1) Before:
> >
> > Samples: 63K of event 'cycles:ppp', Event count (approx.): 48796480071
> > Children Self Co Shared Object Symbol
> > + 21.19% 3.38% tc [kernel.vmlinux] [k] pcpu_alloc
> > +3.45%
On Mon, Jan 21, 2019 at 3:24 AM Vlad Buslov wrote:
>
> Hi Eric,
>
> I've been investigating significant tc filter insertion rate degradation
> and it seems it is caused by your commit 001c96db0181 ("net: align
> gnet_stats_basic_cpu struct"). With this commit inse
Hi Eric,
I've been investigating significant tc filter insertion rate degradation
and it seems it is caused by your commit 001c96db0181 ("net: align
gnet_stats_basic_cpu struct"). With this commit insertion rate is
reduced from ~65k rules/sec to ~43k rules/sec when inserting 1m rul
