On 01/28/2013 04:41 PM, Carl Worth wrote:
Kenneth Graunke <[email protected]> writes:
Previously, BLORP only handled BlitFramebuffer.  This patch adds an
additional frontend for doing CopyTexSubImage.  It also makes it the
default.

Reading through the patch, it looks like there are two independent
things happening here---extending blorp to support a copytexsubimage
operation and then preferring to use that first ti implement
CopyTexSubImage for i965.

And sure enough your commit message has "also" in it, (and even "and"
in the one-line summary).

I recommend splitting this into two commits.

I didn't want to add dead code, so I felt obliged to use the new method somewhere in the order. Putting it below meta doesn't make much sense---hardware paths below software paths? I could put it below BLT and move it to #1 in the next commit, but...that's still a change in the behavior. It's not clear to me that it's better.

How would you prefer it split?

No Piglit regressions on Ivybridge.  I have not tested Sandybridge.

I just tested on Sandybridge---no piglit regressions.

Reviewed-by: Carl Worth <[email protected]>

-Carl

Thanks Carl!
_______________________________________________
mesa-dev mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/mesa-dev

Reply via email to