A couple of frames in I get: ElementalDemo: intel_copy_image.c:83: copy_image_with_blitter: Assertion `src_width % bw == 0' failed. mesa commit 79bff488bc Haswell gpu: 00:02.0 VGA compatible controller [0300]: Intel Corporation Xeon E3-1200 v3/4th Gen Core Processor Integrated Graphics Controller [8086:0412] (rev 06) DRI3-enabled DDX (in case it matters). Demo downloadable from https://wiki.unrealengine.com/Linux_Demos Run like this: ./ElementalDemo ResX=640 ResY=480
I wonder if this is related to my ARB_copy_image fix: commit 912babba7bf1abd3caa49f6372d581ae1afe7e84 Author: Ilia Mirkin <imirkin@alum.mit.edu> Date: Sun Nov 8 04:46:38 2015 -0500 mesa/copyimage: allow width/height to not be multiples of block
(In reply to Ilia Mirkin from comment #1) > I wonder if this is related to my ARB_copy_image fix: > > commit 912babba7bf1abd3caa49f6372d581ae1afe7e84 > Author: Ilia Mirkin <imirkin@alum.mit.edu> > Date: Sun Nov 8 04:46:38 2015 -0500 > > mesa/copyimage: allow width/height to not be multiples of block Quite possibly. Mind actually bisecting (or just checking either side of that commit) or are you going to make us do it? :-) Looking at that assert, we probably just need to remove the width and height asserts and use DIV_ROUND_UP instead of regular divide.
(In reply to Jason Ekstrand from comment #2) > (In reply to Ilia Mirkin from comment #1) > > I wonder if this is related to my ARB_copy_image fix: > > > > commit 912babba7bf1abd3caa49f6372d581ae1afe7e84 > > Author: Ilia Mirkin <imirkin@alum.mit.edu> > > Date: Sun Nov 8 04:46:38 2015 -0500 > > > > mesa/copyimage: allow width/height to not be multiples of block > > Quite possibly. Mind actually bisecting (or just checking either side of > that commit) or are you going to make us do it? :-) I was hoping you guys could work it out. Not sure when I'll be in front of the HSW machine next... probably Monday. I was mostly testing that demo on an NVIDIA GK208 board (where it used to have artifacts, all good now - commit 7752bbc44e fixed it right up), and ran it against Intel "just to see what happens". This happened, figured it was of interest.
(In reply to Ilia Mirkin from comment #3) > I was hoping you guys could work it out. Not sure when I'll be in front of > the HSW machine next... probably Monday. I was mostly testing that demo on > an NVIDIA GK208 board (where it used to have artifacts, all good now - > commit 7752bbc44e fixed it right up), and ran it against Intel "just to see > what happens". This happened, figured it was of interest. That's fine. Thanks for the bug report none-the-less. Like I said above, I'm sure it's an easy fix. Just need to find time to fix it.
commit 088280e022bca10c3712dfa6f8e0ff343a37ce77 Author: Ben Widawsky <benjamin.widawsky@intel.com> Date: Sat Feb 6 18:11:21 2016 -0800 i965: Make sure we blit a full compressed block
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.