Script 'mail_helper' called by obssrc
Hello community,

here is the log from the commit of package Mesa for openSUSE:Factory checked in 
at 2022-11-03 19:13:43
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Comparing /work/SRC/openSUSE:Factory/Mesa (Old)
 and      /work/SRC/openSUSE:Factory/.Mesa.new.2275 (New)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Package is "Mesa"

Thu Nov  3 19:13:43 2022 rev:456 rq:1032970 version:unknown

Changes:
--------
--- /work/SRC/openSUSE:Factory/Mesa/Mesa.changes        2022-10-22 
14:13:15.116761446 +0200
+++ /work/SRC/openSUSE:Factory/.Mesa.new.2275/Mesa.changes      2022-11-03 
19:13:53.871906386 +0100
@@ -1,0 +2,13 @@
+Wed Nov  2 16:56:33 UTC 2022 - Stefan Dirsch <sndir...@suse.com>
+
+- u_nouveau-corrupted-colors-boo1203949.patch
+  * fixes corrupted colors in videos on nouveau with Kepler in
+    Firefox (boo#1203949, issue#7416)
+
+-------------------------------------------------------------------
+Wed Nov  2 08:53:22 UTC 2022 - Stefan Dirsch <sndir...@suse.com>
+
+- moved drirc.d config snippets from Mesa to Mea-dri package;
+  radv driver specific conf was missing completely (boo#1204866)
+
+-------------------------------------------------------------------

New:
----
  u_nouveau-corrupted-colors-boo1203949.patch

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Other differences:
------------------
++++++ Mesa.spec ++++++
--- /var/tmp/diff_new_pack.au8M2w/_old  2022-11-03 19:13:55.243914454 +0100
+++ /var/tmp/diff_new_pack.au8M2w/_new  2022-11-03 19:13:55.251914501 +0100
@@ -143,6 +143,7 @@
 Patch300:       n_buildfix-21.3.0.patch
 Patch400:       n_no-sse2-on-ix86-except-for-intel-drivers.patch
 Patch500:       n_stop-iris-flicker.patch
+Patch600:       u_nouveau-corrupted-colors-boo1203949.patch
 %ifarch %{ix86} x86_64
 BuildRequires:  DirectX-Headers
 %endif
@@ -767,6 +768,7 @@
 %patch400 -p1
 %endif
 %patch500 -p1
+%patch600 -p1
 
 # Remove requires to vulkan libs from baselibs.conf on platforms
 # where vulkan build is disabled; ugly ...
@@ -889,9 +891,6 @@
 # Delete things that we do not package in the Mesa-drivers variant, but can
 # not disable from buildling and installing.
 
-# in Mesa
-rm -rf %{buildroot}/%{_datadir}/drirc.d
-
 rm -f %{buildroot}/%{_libdir}/libEGL.so*
 # in Mesa-libEGL-devel
 rm %{buildroot}/%{_includedir}/EGL/egl.h
@@ -926,6 +925,8 @@
 rm -f %{buildroot}/%{_libdir}/vdpau/libvdpau_gallium.so
 
 %else
+# package in Mesa-dri
+rm -rf %{buildroot}/%{_datadir}/drirc.d
 
 rm -f %{buildroot}/%{_libdir}/dri/*_dri.so
 
@@ -991,8 +992,6 @@
 %files
 %license docs/license.rst
 %doc docs/README*
-%dir %{_datadir}/drirc.d
-%config %{_datadir}/drirc.d/*
 
 %files libEGL1
 %{_libdir}/libEGL_mesa.so*
@@ -1107,6 +1106,8 @@
 
 %if "%{flavor}" == "drivers"
 %files -n Mesa-dri
+%dir %{_datadir}/drirc.d
+%config %{_datadir}/drirc.d/*
 %dir %{_libdir}/dri
 %{_libdir}/dri/*_dri.so
 %ifarch %{ix86} x86_64 aarch64 %{arm} ppc64 ppc64le riscv64



++++++ u_nouveau-corrupted-colors-boo1203949.patch ++++++
>From 6014a642ae24e37c287adbf41d615efa18d0da3d Mon Sep 17 00:00:00 2001
From: Joan Bruguera <joanbrugue...@gmail.com>
Date: Tue, 1 Nov 2022 23:01:50 +0100
Subject: [PATCH] nv50/ir/nir: ignore sampler for TXF/TXQ ops.

Recently, a regression was reported where videos in Firefox had shifted/
glitched colors on certain Kepler hardware. This was bisected to
bf02bffe156214dad942f56ee68c380d1968704f, however, the issue already
existed but didn't hit users until TGSI was switched to NIR as default.

The issue was traced to a YUV-to-RGB fragment shader used by Firefox,
which uses three samplers for the Y/U/V components. The Y component was
handled correctly, but the U/V components were bogus, causing the issue.

After analysis, it appears the TXF/TXQ ops. should only handle the texture
(r) but not the sampler (s), see 63b850403c90f33c295d3ad6be4ad749d4ea6274
and 346ce0b98832e33d5411200002571b3edea9e2bb.
Similarly, handleTXQ/handleTXF on nv50_ir_from_tgsi always sets s=0.
Only Kepler was affected because other hardware ignores s at codegen.

Always set s=0 on NIR for TXF/TXQ, to keep TGSI behavior and fix the
regression.

Thanks: Karol Herbst and M Henning for help diagnosing the issue.
Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/7416
Cc: mesa-stable
Suggested-by: Karol Herbst <kher...@redhat.com>
Reviewed-by: Karol Herbst <kher...@redhat.com>
Reviewed-by: M Henning <dra...@darkrefraction.com>
Signed-off-by: Joan Bruguera <joanbrugue...@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/19453>
---
 src/nouveau/codegen/nv50_ir_from_nir.cpp | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/src/nouveau/codegen/nv50_ir_from_nir.cpp 
b/src/nouveau/codegen/nv50_ir_from_nir.cpp
index cbefa93d25fa..4550acbf180d 100644
--- a/src/nouveau/codegen/nv50_ir_from_nir.cpp
+++ b/src/nouveau/codegen/nv50_ir_from_nir.cpp
@@ -3133,6 +3133,8 @@ Converter::visit(nir_tex_instr *insn)
 
       r = bindless ? 0xff : insn->texture_index;
       s = bindless ? 0x1f : insn->sampler_index;
+      if (op == OP_TXF || op == OP_TXQ)
+         s = 0;
 
       defs.resize(newDefs.size());
       for (uint8_t d = 0u; d < newDefs.size(); ++d) {
-- 
GitLab

Reply via email to