| Summary: | [radeonsi] Blue-ish textures in Shadow of Mordor | ||
|---|---|---|---|
| Product: | Mesa | Reporter: | Jan Ziak (http://atom-symbol.net) <0xe2.0x9a.0x9b> |
| Component: | Drivers/Gallium/radeonsi | Assignee: | Default DRI bug account <dri-devel> |
| Status: | RESOLVED FIXED | QA Contact: | Default DRI bug account <dri-devel> |
| Severity: | normal | ||
| Priority: | medium | CC: | 0xe2.0x9a.0x9b |
| Version: | git | ||
| Hardware: | All | ||
| OS: | Linux (All) | ||
| Whiteboard: | |||
| i915 platform: | i915 features: | ||
| Attachments: |
Screenshot: AMD OpenGL driver
Screenshot: Mesa radeonsi OpenGL Screenshot: Mesa radeonsi OpenGL - patched |
||
|
Description
Jan Ziak (http://atom-symbol.net)
2016-03-23 10:33:10 UTC
Created attachment 122494 [details]
Screenshot: AMD OpenGL driver
Created attachment 122495 [details]
Screenshot: Mesa radeonsi OpenGL
Further information: - Linux kernel 4.5.0 - Kernel module: radeon $ glxinfo | grep OpenGL OpenGL vendor string: X.Org OpenGL renderer string: Gallium 0.4 on AMD HAWAII (DRM 2.43.0, LLVM 3.8.0) OpenGL core profile version string: 4.1 (Core Profile) Mesa 11.3.0-devel (git-530593d) I saw this too with Amdgpu Fury. Thank you for the report. I suspect some of the BGR <-> RGB changes for shader image bitcasts are responsible. Could one of you please provide an apitrace that shows the problem? That would be very helpful. (In reply to Nicolai Hähnle from comment #5) > Thank you for the report. > > I suspect some of the BGR <-> RGB changes for shader image bitcasts are > responsible. Could one of you please provide an apitrace that shows the > problem? That would be very helpful. Running Shadow of Mordor under apitrace 6.1 is very slow. I didn't run the benchmark because the ingame menu screen wasn't responding. However, the menu screen already has the blue-ish tint that shouldn't be there. The trace is quite large: 392 MB uncompressed, 219 MB compressed with xz. Do you still want the trace despite its size? If not, do you want me to run apitrace on AMD's Opengl library and then perform "apitrace diff ShadowOfMordor-radeonsi.trace ShadowOfMordor-amd.trace"? Hi Jan! First, please try this patch which fixes a similar bug: https://patchwork.freedesktop.org/patch/78118/ If that doesn't fix the problem: I'd appreciate if you could upload the apitrace somewhere (Dropbox, Google Drive, etc.). A diff is not going to be useful. But first, try if the problem is already fixed with the patch above :) (In reply to Nicolai Hähnle from comment #7) > Hi Jan! First, please try this patch which fixes a similar bug: > https://patchwork.freedesktop.org/patch/78118/ > > If that doesn't fix the problem: I'd appreciate if you could upload the > apitrace somewhere (Dropbox, Google Drive, etc.). A diff is not going to be > useful. > > But first, try if the problem is already fixed with the patch above :) The patch works! Created attachment 122511 [details]
Screenshot: Mesa radeonsi OpenGL - patched
Fixed in Mesa master, commit 6b763c0 |
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.