On Wed, Sep 26, 2018 at 3:30 AM Gert Wollny <[email protected]> wrote: > > Am Dienstag, den 25.09.2018, 10:20 -0400 schrieb Ilia Mirkin: > > I haven't double-checked yet, but doesn't this result in a reduction > > of functionality for pre-independent-blend GPUs (like the early > > NVIDIA > > Tesla series)? Configuring blending for an integer RT does nothing on > > NVIDIA hardware, so it all works out there just fine... > Unfortunately I can't test this ... > > > > > Perhaps both patches should just be reverted and keep things as they > > were, and let drivers worry about this? > Do as you see fit, my only concern is that there are no regressions, > and if the two patches together still result in regressions then it is > indeed better to revert, > > best, > Gert
I had a look at this in more detail. Ken's change basically made handling of integer buffers be the same as buffers whose colormask was set to empty, or whose blending had been disabled. This should be supportable by all GPUs which support EXT_draw_buffers2. The additional bit is independent blend *functions* (ARB_draw_buffers_blend), which are only supported on GT215+ (on the NVIDIA front). A quick look through the various GPU capabilities, it seems like pre-G80, there is no EXT_draw_buffers2 support, but on the bright side, no integer RTs either, so all is well. So these changes should be fine, although they do force us onto the independent blend path unnecessarily. I suspect this happens ~never in practice. Cheers, -ilia _______________________________________________ mesa-dev mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/mesa-dev
