| Summary: | Artifacts in form of wrong pixels appearing on some surfaces | ||
|---|---|---|---|
| Product: | Mesa | Reporter: | Fabian Maurer <dark.shadow4> |
| Component: | Mesa core | Assignee: | mesa-dev |
| Status: | RESOLVED MOVED | QA Contact: | mesa-dev |
| Severity: | normal | ||
| Priority: | medium | ||
| Version: | git | ||
| Hardware: | x86-64 (AMD64) | ||
| OS: | Linux (All) | ||
| Whiteboard: | |||
| i915 platform: | i915 features: | ||
| Attachments: |
Minecraft trage file showing the corruption
Screenshot with artifacts (mesa) Screenshot how it should look like (windows closed amd driver) |
||
Created attachment 132769 [details]
Screenshot with artifacts (mesa)
Created attachment 132770 [details]
Screenshot how it should look like (windows closed amd driver)
A quick look at the trace shows a bunch of errors messages due to the game expecting the GL_NV_fog_distance extensions functionality to be available. Yes, the fog is buggy too. But is this relevant to the pixel errors? It used to work, after all. (In reply to Fabian Maurer from comment #4) > Yes, the fog is buggy too. But is this relevant to the pixel errors? It used > to work, after all. Ok I see what you are talking about now. If it used to work it would be very helpful if you could do a git bisect of a working commit vs master. (In reply to Timothy Arceri from comment #3) > A quick look at the trace shows a bunch of errors messages due to the game > expecting the GL_NV_fog_distance extensions functionality to be available. Sorry this error message was happening on i965 which doesn't have the extension enable for some reason, these errors are not reported on gallium drivers. Also there are no artifacts when running on i965 so it would seem this is a gallium issue rather than a core mesa problem. Sorry, look like I was wrong. There seem to always have been artifacts, but since a newer Minecraft version they're colorful instead of black, making them way more noticeable. "LIBGL_ALWAYS_SOFTWARE=1" reduces the artifacts massively, and "GALLIUM_DRIVER=swr LIBGL_ALWAYS_SOFTWARE=1" removes them. Now, since I can't do a regression test, is there another way I could help? -- GitLab Migration Automatic Message -- This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/mesa/mesa/issues/1013. |
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.
Created attachment 132768 [details] Minecraft trage file showing the corruption Title says it all, will attach screenshots in the following comments. Apitrace attached. Images both taken from frame 600 of the trace file: -One on linux -One on win10 with latest closed source AMD drivers. LIBGL_ALWAYS_SOFTWARE doesn't make a difference. System the bug was tested on: - Arch Linux 64bit - Linux 4.12.2, AMDGPU driver - Mesa 17.2.0-devel (git-80a0c9745c) / Mesa 17.1.4 - Radeon R9 285