Re: [PATCH v3 10/11] zram: use rw_semaphore for decompression streams

[Sergey Senozhatsky <senozhatsky@chromium.org>]

Hi,

On (26/10/16 11:10), Barry Song wrote:
> > > I applied the kmalloc+memcpy variant on-device and re-ran the same
> > > Android A/B. It removes the exclusive path entirely in our workload.
> > >
> > >                       9+10       9+10 + kmalloc fastpath
> > >   read contended      ~1,959     0
> > >   read wait_ns total  ~12 s      0
> > >   read wait_ns max    72-96 ms   0
> > >   shared              76.7%      100%
> > >   exclusive           23.3%      0%
> > >   dpin_contended      0          0
> > >
> > > Every decompression now takes the shared (down_read) path, including the
> > > cross-page objects that previously had to go exclusive. The ~1,959
> > > residual contentions on the exclusive path are gone, and pin contention
> > > stays at zero, so the full read side runs concurrently with no stream
> > > contention.
> > >
> > > Read-side contention end to end:
> > >
> > >                  baseline   9+10      9+10 + kmalloc
> > >   read contended 24,749     ~1,959    0
> > >   shared         0%         76.7%     100%
> >
> > Thank you.  I'll include kmalloc patch to v4 of the series.
> 
> Hi Sergey, Bo,
> 
> I assume when Sergey sent v4, could tag for the patchset with
> 
> Tested-by: Bo Zhang <zhangbo56@xiaomi.com>

That would be great.

> also, Bo's data can be put in cover-letter?

Sure, I can include it, along with my numbers.