<div dir="ltr">Here goes results on BDW with pure today's nightly (with idle_frame=1)<div><br></div><div># First run</div><div><br></div><div><div>IGT-Version: 1.7-gd4b43f0 (x86_64) (Linux: 3.17.0-rc2+ x86_64)</div><div>Subtest primary_page_flip: SUCCESS</div><div>Subtest primary_mmap_gtt: SUCCESS</div><div>Test assertion failure function test_crc, file kms_psr_sink_crc.c:307:</div><div>Failed assertion: strcmp(ref_crc, CRC_GREEN) != 0</div><div>Subtest primary_mmap_gtt_waiting: FAIL</div><div>Subtest primary_mmap_cpu: SUCCESS</div><div>Subtest primary_blt: SUCCESS</div><div>Subtest primary_render: SUCCESS</div><div>Subtest sprite_mmap_gtt: SUCCESS</div><div>Waiting 10s...</div><div>Subtest sprite_mmap_gtt_waiting: SUCCESS</div><div>Subtest sprite_mmap_cpu: SUCCESS</div><div>Subtest sprite_blt: SUCCESS</div><div>Subtest sprite_render: SUCCESS</div><div>Subtest sprite_plane_move: SUCCESS</div><div>Subtest sprite_plane_onoff: SUCCESS</div><div>Subtest cursor_mmap_gtt: SUCCESS</div><div>Waiting 10s...</div><div>Subtest cursor_mmap_gtt_waiting: SUCCESS</div><div>Subtest cursor_mmap_cpu: SUCCESS</div><div>Subtest cursor_blt: SUCCESS</div><div>Subtest cursor_render: SUCCESS</div><div>Subtest cursor_plane_move: SUCCESS</div><div>Subtest cursor_plane_onoff: SUCCESS</div></div><div><br></div><div>#second run:</div><div><br></div><div><div>IGT-Version: 1.7-gd4b43f0 (x86_64) (Linux: 3.17.0-rc2+ x86_64)</div><div>Subtest primary_page_flip: SUCCESS</div><div>Subtest primary_mmap_gtt: SUCCESS</div><div>Waiting 10s...</div><div>Subtest primary_mmap_gtt_waiting: SUCCESS</div><div>Subtest primary_mmap_cpu: SUCCESS</div><div>Subtest primary_blt: SUCCESS</div><div>Subtest primary_render: SUCCESS</div><div>Subtest sprite_mmap_gtt: SUCCESS</div><div>Waiting 10s...</div><div>Subtest sprite_mmap_gtt_waiting: SUCCESS</div><div>Subtest sprite_mmap_cpu: SUCCESS</div><div>Subtest sprite_blt: SUCCESS</div><div>Test assertion failure function test_crc, file kms_psr_sink_crc.c:307:</div><div>Failed assertion: strcmp(ref_crc, CRC_GREEN) != 0</div><div>Subtest sprite_render: FAIL</div><div>Subtest sprite_plane_move: SUCCESS</div><div>Subtest sprite_plane_onoff: SUCCESS</div><div>Subtest cursor_mmap_gtt: SUCCESS</div><div>Waiting 10s...</div><div>Subtest cursor_mmap_gtt_waiting: SUCCESS</div><div>Subtest cursor_mmap_cpu: SUCCESS</div><div>Subtest cursor_blt: SUCCESS</div><div>Subtest cursor_render: SUCCESS</div><div>Subtest cursor_plane_move: SUCCESS</div><div>Subtest cursor_plane_onoff: SUCCESS</div></div><div><br></div><div>random failures! but better than hsw at least.</div><div><br></div><div>However the hardcoded color is indeed a mistake... Green on this panel is different from the green on the other panel.</div><div>But I'm also not sure about drawing the color with cairo... How can we be sure what is there is what we want anyway.</div><div><br></div><div>So maybe it is better to fall back to old scheme where we just check if changed when we want it changes... without checking for colors.</div><div>What do you think?</div><div><br></div><div>Thanks,</div><div>Rodrigo.</div><div><br></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Thu, Sep 4, 2014 at 1:20 PM, Daniel Vetter <span dir="ltr"><<a href="mailto:daniel@ffwll.ch" target="_blank">daniel@ffwll.ch</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="">On Thu, Sep 4, 2014 at 9:16 PM, Rodrigo Vivi <<a href="mailto:rodrigo.vivi@gmail.com">rodrigo.vivi@gmail.com</a>> wrote:<br> >> Hm, I don't see a pattern at all. And that sprites seem to work best<br> >> also looks funky. Are the results stable when you randomize them<br> >> (piglit can do that for you)? Can you add a residency checks in the<br> >> testcase (i.e. before the update and after the update) so that we know<br> >> it's really psr and not something else funny going on? I assume this<br> >> is on bdw, any difference in results on bdw?<br> ><br> > Yes, this is really strange. I couldn't find a pattern at all as well.<br> > Do you have any example of piglit helping randomization?<br> ><br> > The bad things with residency check is that vlv/chv doesn't have performance<br> > counters.<br> > So this test would just work on hsw/bdw/<br> ><br> > I did this work on a HSW. Test on BDW is one of tests I should do next here.<br> <br> </span>There's no shuffling in piglit unfortunately. But piglit is good for<br> comparing results, so I wonder whether they are stable or change. And<br> hsw is even more suprising since sprite tests pass more often in your<br> example, but sprite also has now hw invalidate on hsw. Confusing.<br> <div class="HOEnZb"><div class="h5">-Daniel<br> --<br> Daniel Vetter<br> Software Engineer, Intel Corporation<br> <a href="tel:%2B41%20%280%29%2079%20365%2057%2048" value="+41793655748">+41 (0) 79 365 57 48</a> - <a href="http://blog.ffwll.ch" target="_blank">http://blog.ffwll.ch</a><br> </div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div>Rodrigo Vivi</div><div>Blog: <a href="http://blog.vivi.eng.br" target="_blank">http://blog.vivi.eng.br</a></div><div> </div> </div>