On Tue, Apr 28, 2015 at 6:17 PM, Rogovin, Kevin <kevin.rogo...@intel.com> wrote: > Hello, > > >> No, because the non-shared code is (by your own admission) untested and/or >> dead code. Untested code is broken code. I would personally be ok with a >> lot > of the changes that just replace fb->Width with >> _mesa_geometric_width(fb) since it's effectively just replacing a direct >> access with a getter. However, almost half of the patch is updating the >> upload_sf_vp > function which is only used for gen <= 5. A comment or >> assert there would be sufficient rather than reworking it. > > Fair enough. Would the following be good: > - keep all those that replace fb->whatever with _mesa_geomety_whatever, > - instead of the ick I have done to upload_sf_vp, place a big comment warning
Yes, I think that would be sufficient. > I would be happy with the above as it addresses my main concern and the > dead-is-broken code concern as well. If I had physical access to a Gen4 and 5 > box I would test it and if it worked, enable the extension on those platforms > as well. I don't think there's any good reason to turn it on for Gen5 or older. However, we should still test it because it does touch code that hits those platforms. Testing across all the hardware can be done fairly easily by pushing to our Jenkins system. I know that Martin and Topi (and probably Curro) have accounts and could run it easily enough. _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
