<html>
<head>
<base href="https://bugs.freedesktop.org/">
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - [GEN9+] 14-24% perf drop in SynMark2 v7 CSDof"
href="https://bugs.freedesktop.org/show_bug.cgi?id=109517#c5">Comment # 5</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - [GEN9+] 14-24% perf drop in SynMark2 v7 CSDof"
href="https://bugs.freedesktop.org/show_bug.cgi?id=109517">bug 109517</a>
from <span class="vcard"><a class="email" href="mailto:eero.t.tamminen@intel.com" title="Eero Tamminen <eero.t.tamminen@intel.com>"> <span class="fn">Eero Tamminen</span></a>
</span></b>
<pre>(In reply to Jason Ekstrand from <a href="show_bug.cgi?id=109517#c4">comment #4</a>)
<span class="quote">> Ugh... I'm not really sure what we should do about this one. Mark's bisect
> is exactly correct. I've looked at the shaders, and there seems to be two
> issues:
>
> 1) There's one SIMD8 shader which schedules massively differently for no
> apparent reason.</span >
>
<span class="quote">> 2) There's a SIMD16 shader which starts spilling way more than it was before</span >
Based on your comment below I assume that SIMD8 shader also got worse, but does
it also spill? I.e. is the bad behavior limited to spilling shaders?
<span class="quote">> In both cases, I have no idea why it's happening beyond the fact that our
> current RA and scheduling has rather random behaviour at times. Using SENDS
> should only ever decrease register pressure and increase RA freedom because
> it no longer has to build the message into a single hunk and can just send
> the two bits (address and data) separately.
>
> As I said in the commit message I have another (unfortunately not public
> yet) customer workload where the opposite happens and using SENDS decreases
> spilling and improves performance by 2x.</span >
SENDS support is needed performance feature. If the current implementation
improves things more than it regresses, and especially if the improving cases
are more important like here, I think letting regression in for the release is
fine.
There could be some meta-bug about the RA / scheduler related issues which this
(and e.g. bugs about bad sampler fetch scheduling) would link to though.</pre>
</div>
</p>
<hr>
<span>You are receiving this mail because:</span>
<ul>
<li>You are the QA Contact for the bug.</li>
</ul>
</body>
</html>