On Thursday, November 20, 2014 11:14:48 AM Ian Romanick wrote:
This little series is just a bit of house keeping in the valid-to-render
path. There is a follow-up series in progress that tries to optimize
some of these paths for CPU-bound workloads.
The whole series has been reordered several times. Orignally patches 4
and 5 of this series were farther apart. As it stands now, they should
probably be squashed (taking the commit message from patch 5).
I would not squash them. One is purely stylistic, and the other changes
assert to unreachable. I don't feel strongly though.
Patches 2-7 are:
Reviewed-by: Kenneth Graunke kenn...@whitecape.org
FWIW... I was not able to measure any performance delta on any test on
any platform at even 80% confidence across this series.
These changes are so small I would expect them to be basically unmeasurable.
They're worthwhile anyway since they clean up the codebase.
--Ken
signature.asc
Description: This is a digitally signed message part.
___
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
