<div class="gmail_quote">On Sat, Apr 14, 2012 at 22:41, Ben Widawsky <span dir="ltr"><<a href="mailto:ben@bwidawsk.net" target="_blank">ben@bwidawsk.net</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
This originally started as a patch from Bernard as a way of simply<br>
setting the VS scheduler. After submitting the RFC patch, we decided to<br>
also modify the DS scheduler. To be most explicit, I've made the patch<br>
explicitly set all scheduler modes, and included the defines for other<br>
modes (in case someone feels frisky later).<br>
<br>
The rest of the story gets a bit weird. The first version of the patch<br>
showed an almost unbelievable performance improvement. Since rebasing my<br>
branch it appears the performance improvement has gone, unfortunately.<br>
But setting these bits seem to be the right thing to do given that the<br>
docs describe corruption that can occur with the default settings.<br>
<br>
In summary, I am seeing no more perf improvements (or regressions) in my<br>
limited testing, but we believe this should be set to prevent rendering<br>
corruption, therefore cc stable.<br>
<br>
v1: Clear bit 4 also (Ken + Eugeni)<br>
Do a full clear + set of the bits we want (Me).<br>
<br>
Cc: Bernard Kilarski <<a href="mailto:bernard.r.kilarski@intel.com" target="_blank">bernard.r.kilarski@intel.com</a>><br>
Cc: stable <<a href="mailto:stable@vger.kernel.org" target="_blank">stable@vger.kernel.org</a>><br>
Reviewed-by (RFC): Kenneth Graunke <<a href="mailto:kenneth@whitecape.org" target="_blank">kenneth@whitecape.org</a>><br>
Signed-off-by: Ben Widawsky <<a href="mailto:benjamin.widawsky@intel.com" target="_blank">benjamin.widawsky@intel.com</a>><br></blockquote><div><br></div><div>Very nice!</div><div><br></div><div>I also suspect that maybe the initial performance improvement you've seen with previous testing could be related to the occasional turbo disabling we've been seeing in other cases as well (e.g., <a href="https://bugs.freedesktop.org/show_bug.cgi?id=44006" target="_blank">https://bugs.freedesktop.org/show_bug.cgi?id=44006</a>).</div>
<div><br></div><div>But as for this patch, I have just one comment/suggestion below, but other than that:</div><div><br></div><div>Reviewed-by: Eugeni Dodonov <<a href="mailto:eugeni.dodonov@intel.com" target="_blank">eugeni.dodonov@intel.com</a>></div>
<div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">+static void gen7_setup_fixed_func_scheduler(struct drm_i915_private *dev_priv)<br></blockquote><div><br>
</div><div>Perhaps this functions should be named ivybridge_setup_fixed_func_scheduler instead?</div><div><br></div><div>Even if those bits are not ivy bridge-exclusive, this specific explicit setup applies to ivb only..</div>
<div><br></div></div>-- <br>Eugeni Dodonov<a href="http://eugeni.dodonov.net/" target="_blank"><br></a><br>