<html> <head> <base href="https://bugs.freedesktop.org/" /> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW --- - i965: Support instruction compaction on Gen < 6" href="https://bugs.freedesktop.org/show_bug.cgi?id=77742#c2">Comment # 2</a> on <a class="bz_bug_link bz_status_NEW " title="NEW --- - i965: Support instruction compaction on Gen < 6" href="https://bugs.freedesktop.org/show_bug.cgi?id=77742">bug 77742</a> from <a class="email" href="mailto:mattst88@gmail.com" title="Matt Turner <mattst88@gmail.com>"> Matt Turner</a> <pre>(In reply to <a href="show_bug.cgi?id=77742#c1">comment #1</a>) > Created <a href="attachment.cgi?id=98113" name="attach_98113" title="Patch">attachment 98113</a> <a href="attachment.cgi?id=98113&action=edit" title="Patch">[details]</a> <a href='page.cgi?id=splinter.html&bug=77742&attachment=98113'>[review]</a> [review] > Patch > > This patch adds support for instruction compaction to ILK and previous > architectures (G45 and i965). > > I have access to a testing machine with ILK. I have tested it with a simple > program which runs this new code without any visual difference. However I > don't know this is enough or if there is another method to check whether > this compaction code is correctly working or not by using something like a > piglit test or other. Great. Thanks! That was quick. It's unclear from the i965 docs whether it actually supports instruction compaction (the docs in those days often shipped after changes started being made for the next chip). Eric's initial instruction compaction commit says GM45 and newer support it. We should probably change the gen4_ prefix to gm45_ and then do two small fix ups to only allow compaction on gen4 iff brw->is_g4x is true. With those changes and a passing piglit run on ILK, I think the patch would be good to go. Feel free to send to the list. We'll find a GM45 system to test around the office. >From there we should test whether the original gen4 supports instruction compaction.</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the QA Contact for the bug.</li> </ul> </body> </html>