<html>
<head>
<base href="https://bugs.freedesktop.org/" />
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_ASSIGNED "
title="ASSIGNED --- - [HSW] DOTA world lighting has no effect"
href="https://bugs.freedesktop.org/show_bug.cgi?id=76749#c7">Comment # 7</a>
on <a class="bz_bug_link
bz_status_ASSIGNED "
title="ASSIGNED --- - [HSW] DOTA world lighting has no effect"
href="https://bugs.freedesktop.org/show_bug.cgi?id=76749">bug 76749</a>
from <span class="vcard"><a class="email" href="mailto:olvaffe@gmail.com" title="Chia-I Wu <olvaffe@gmail.com>"> <span class="fn">Chia-I Wu</span></a>
</span></b>
<pre>(In reply to <a href="show_bug.cgi?id=76749#c6">comment #6</a>)
<span class="quote">> (In reply to <a href="show_bug.cgi?id=76749#c3">comment #3</a>)
> > This is also reproducible on SNB. Several other (probably unrelated) issues
> > are noticed on SNB when replaying the trace file from which the snapshots
> > were taken.
> >
> > commit 9630ba6c6e754b438cf67c7d76ec1c99488df3ba
> > Author: Matt Turner <<a href="mailto:mattst88@gmail.com">mattst88@gmail.com</a>>
> > Date: Tue Mar 11 13:14:08 2014 -0700
> >
> > i965/vec4: Let dead code eliminate trim dead channels.
> >
> > causes an assertion failure:
> >
> > glretrace: brw_vec4_generator.cpp:184: void
> > brw::vec4_generator::generate_math1_gen6(brw::vec4_instruction*, brw_reg,
> > brw_reg): Assertion `dst.dw1.bits.writemask == 0xf' failed.
> >
> > commit dc0f5099fa3cb564c25eb892fde93cacd29df8f1
> > Author: Matt Turner <<a href="mailto:mattst88@gmail.com">mattst88@gmail.com</a>>
> > Date: Tue Mar 11 13:06:20 2014 -0700
> >
> > i965/vec4: Track live ranges per-channel, not per vgrf.
> >
> > causes another assertion failure:
> >
> > glretrace: brw_vec4.cpp:346: bool brw::vec4_visitor::dead_code_eliminate():
> > Assertion `this->virtual_grf_end[inst->dst.reg] >= pc' failed.
>
> There are bugs for these that I'm working on now.
>
> > Finally,
> >
> > commit 947c828d5cbffe9640ac63103a6223112eeff27f
> > Author: Matt Turner <<a href="mailto:mattst88@gmail.com">mattst88@gmail.com</a>>
> > Date: Thu Dec 12 00:30:16 2013 -0800
> >
> > i965/fs: Add a saturation propagation optimization pass.
> >
> > causes more lights to have no effect.
>
> Ken made a fix to this pass, commit 4d2e79269a. Does that fix this bit?</span >
Nope. With mesa master, the spotlight on the heroes is still missing unless I
comment out fs_visitor::opt_saturate_propagation.
The other commit, 9cd51bb0c4608258199c69bc7738e72f055799d2, causes the ground
to be rendered with a random pattern. It looks like the commit incorrectly
eliminates some writes whose results are actually read.
Those comments are notes to myself while looking into this specific bug. I
planned to revisit them after this bug is fixed. It would be great if you are
interested too. The trace file in question can be found at
people.freedesktop.org:~cworth/dota2-jordan-trim.trace. I can file bug reports
if that is preferred.</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>