On 05/12/2011 09:05 PM, Eric Anholt wrote:
Improves glbenchmark egypt performance 0.6% +/- 0.4% (n=6).
...
+   bool header_present = false;
...
     if (ir->shadow_comparitor) {
-      mlen = MAX2(mlen, 1 + 4 * reg_width);
+      mlen = MAX2(mlen, header_present + 4 * reg_width);

Relying on boolean "true" being automatically promoted to 1 (and not some other non-zero value) always makes me nervous for some reason...but I looked it up, and it does always work that way.

Reviewed-by: Kenneth Graunke <kenn...@whitecape.org>
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Reply via email to