[Bug 88561] [radeonsi][regression, bisected] Depth test/buffer issues in Portal

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Sun Jan 18 09:34:27 PST 2015


https://bugs.freedesktop.org/show_bug.cgi?id=88561

            Bug ID: 88561
           Summary: [radeonsi][regression,bisected] Depth test/buffer
                    issues in Portal
           Product: Mesa
           Version: git
          Hardware: Other
                OS: All
            Status: NEW
          Severity: normal
          Priority: medium
         Component: Drivers/Gallium/radeonsi
          Assignee: dri-devel at lists.freedesktop.org
          Reporter: daniel at constexpr.org

Created attachment 112428
  --> https://bugs.freedesktop.org/attachment.cgi?id=112428&action=edit
Bad frame

With current Mesa master (commit 461103ef64858e9d81073fad1bd8222b70b2754b),
(some) background geometry will be randomly drawn above foreground for some
frames in Portal (also observed in the Portal mod Rexaura). It does not happen
often and the same scene may be fine depending on what was loaded before or
just randomness.

Bisecting blames commit 02ba7334d35cf8182048c17a149b16f18104c6bf
Author: Marek Olšák <marek.olsak at amd.com>
Date:   Sun Jan 4 22:16:53 2015 +0100

    radeonsi: don't use TC L2 for updating descriptors on SI

    It's causing problems, because we mix uncached CP DMA with cached
WRITE_DATA
    when updating the same memory.

    The solution for SI is to use uncached access here, because CP DMA doesn't
    support cached access.

    CIK will be handled in the next patch.

    Reviewed-by: Michel Dänzer <michel.daenzer at amd.com>

Simply reverting this commit on master causes other rendering issues.

GPU: Radeon HD 7950 (OpenGL renderer string: Gallium 0.4 on AMD TAHITI)
LLVM compiled from svn last week (r225584)
GCC: 4.8.4
Kernel: Linux 3.18.2-gentoo
Distro: Gentoo

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/dri-devel/attachments/20150118/8f0ef636/attachment-0001.html>


More information about the dri-devel mailing list