Trace compression improvements

[José Fonseca]

Thanks for the figures. They sound good.

I finally started testing this.

I noticed one problem though -- the handling of exceptions: it is
imperative that when there is a crash inside a GL call, all pending calls
up to the crash get written to the trace file. But this is not happening
with threaded-file branch.

I made a test for this, at
https://github.com/apitrace/apitrace-tests/commit/c8219f827e6aa4dcad2e94c6b05976b58c266ddc
,
and tracing it should produce something like

  $ apitrace dump -v exception.trace
  [...]
  11 glGetIntegerv(pname = GL_VIEWPORT, params = ?) // incomplete


Commit "Support for threaded compession" introduces abstractions for
platform-specific thread functions which duplicate the existing ones.
Please rewrite the code to use the existing abstractions, which are based
off http://en.cppreference.com/w/cpp/thread/thread .

I believe we can proceed with LZ4 format however.

Jose

On Tue, Jun 4, 2013 at 9:14 PM, Eugene Velesevich
<eugvelesevich at gmail.com>wrote:

> The best improvements we've seen were on an arm/android board, where
> the threaded approach eliminated very noticeable periodic stutters
> from compression; however, it's hard to provide quantitative data on
> that. On a modern x86 system, we're seeing 7-8% better fps rates with
> ipers, even with lz4hc compression that consumes 30-40% cpu in its
> thread.
>
> With regards to compression ratio the LZ4 compressed trace of ipers is
> larger by 11-12% than the snappy one, but LZ4HC compresses better by
> 35-36% than snappy (you can check it using "apitrace repack")
>
> On Sun, Jun 2, 2013 at 5:09 PM, José Fonseca <jose.r.fonseca at gmail.com>
> wrote:
> > I haven't looked at the changes in detail yet -- I'll do it as soon as I
> > find the time -- but sounds good in principle. Indeed trying out LZ4 has
> > been in the to do for some time, so thanks for doing this.
> >
> > Did you get any figures (speed r& compression ratio) on how it compares
> with
> > snappy? A good benchmark is "ipers" demo, part of mesa demos. It is what
> > Zack used when he was improving the compression speed w/ snappy.
> >
> > Jose
> >
> >
> >
> > On Sat, Jun 1, 2013 at 2:39 PM, evel <evel at ispras.ru> wrote:
> >>
> >> Hello,
> >>
> >> This patch series improves trace compression by implementing threaded
> >> compression offloading and providing alternative compression methods.
>  The
> >> main difference from previously implemented threaded compression is that
> >> locking overhead is significantly reduced thanks to using a
> >> double-buffered
> >> output buffer instead of a ring buffer that was locked on each call dump
> >> operation.
> >>
> >> https://github.com/Testiki/apitrace/tree/threaded-file
> >> _______________________________________________
> >> apitrace mailing list
> >> apitrace at lists.freedesktop.org
> >> http://lists.freedesktop.org/mailman/listinfo/apitrace
> >
> >
> >
> > _______________________________________________
> > apitrace mailing list
> > apitrace at lists.freedesktop.org
> > http://lists.freedesktop.org/mailman/listinfo/apitrace
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/apitrace/attachments/20130615/74527d88/attachment.html>