On 20 September 2011 18:28, Kenneth Graunke <<a href="mailto:kenneth@whitecape.org">kenneth@whitecape.org</a>> wrote: <div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"> Here's a few cleanups to our compiler's IR that I've been wanting to do for a very long time. With this series, expression trees are finally guaranteed to be pure; you can no longer put calls with side effects in them. ir_call is now a statement, rather than a confused hybrid. I apologize for the size of patch 3. I wasn't sure how to do the conversion in smaller steps without introducing intermediate breakage. Apparently no regressions in piglit/glean/ES2 on Sandybridge. _______________________________________________ mesa-dev mailing list <a href="mailto:mesa-dev@lists.freedesktop.org">mesa-dev@lists.freedesktop.org</a> <a href="http://lists.freedesktop.org/mailman/listinfo/mesa-dev" target="_blank">http://lists.freedesktop.org/mailman/listinfo/mesa-dev</a> </blockquote></div> I made comments on patches 1/5 and 3/5. The other patches are Reviewed-by: Paul Berry <<a href="mailto:stereotype441@gmail.com">stereotype441@gmail.com</a>> FYI, I suspect my clip distance lowering pass will have to be updated to reflect these changes. That patch series hasn't landed yet, and I'm not sure which patch series will land first. I'll ping you on IRC when I'm ready to land the clip distance lowering pass so we can coordinate.