Bug#1059041: Xorg segfault when unlocking from Xscreensaver while video playback
On Tue, 19 Dec 2023 20:22:43 +0100 Eduard Bloch wrote:
#7 0x7fb14945a510 __restore_rt (libc.so.6 + 0x3c510)
#8 0x7fb149186702 n/a (amdgpu_drv.so + 0x16702)
#9 0x7fb149186c96 n/a (amdgpu_drv.so + 0x16c96)
#10 0x5577d8e75a6b xf86DPMSSet (Xorg + 0xd3a6b)
Hello,
I am not maintainer of xserver-xorg-video-amdgpu,
just tried to collect some more information.
The interesting lines above in the stacktrace translate to this:
Stack trace of thread 1011552:
...
#7 0x7fb14945a510 __restore_rt (libc.so.6 + 0x3c510) |
#8 0x7fb149186702 n/a (amdgpu_drv.so + 0x16702) | in drmmode_set_mode at
../../src/drmmode_display.c:1267
#9 0x7fb149186c96 n/a (amdgpu_drv.so + 0x16c96) | in
drmmode_set_mode_major at ../../src/drmmode_display.c:1371
#10 0x5577d8e75a6b xf86DPMSSet (Xorg + 0xd3a6b) | in xf86DPMSSet
...
This leads to this [1] upstream issue and merge request [2].
Unfortunately this is not yet part of an upstream release.
A workaround could be a local build of xserver-xorg-video-amdgpu
containing the two commits. They are the first commits after the 23.0.0 release.
Kind regards,
Bernhard
[1] https://gitlab.freedesktop.org/xorg/driver/xf86-video-amdgpu/-/issues/70
[2]
https://gitlab.freedesktop.org/xorg/driver/xf86-video-amdgpu/-/merge_requests/85
# 2024-05-04 Trixie/testing amd64 qemu VM
apt install gdb xserver-xorg xserver-xorg-video-amdgpu xserver-xorg-core-dbgsym
xserver-xorg-video-amdgpu-dbgsym
apt build-dep xserver-xorg-video-amdgpu
mkdir /home/benutzer/source/xserver-xorg-video-amdgpu/orig -p
cd/home/benutzer/source/xserver-xorg-video-amdgpu/orig
apt source xserver-xorg-video-amdgpu
gdb -q
set width 0
set pagination off
directory
/home/benutzer/source/xserver-xorg-video-amdgpu/orig/xserver-xorg-video-amdgpu-23.0.0/debian/patches
file /usr/lib/xorg/Xorg
tb main
run
call dlopen("/usr/lib/xorg/modules/drivers/amdgpu_drv.so",0x101)
print (int)getpid()
shell cat /proc/2535/maps | grep amdgpu_drv | head -n1
b *(0x774f9000 + 0x16c96)
b *(0x774f9000 + 0x16702)
info b
pipe disassemble drmmode_set_mode_major | grep -i c96 -C3
pipe disassemble drmmode_set_mode | grep -i 702 -C3
list drmmode_display.c:1371
list drmmode_display.c:1267
benutzer@debian:~$ gdb -q
(gdb) set width 0
(gdb) set pagination off
(gdb) directory
/home/benutzer/source/xserver-xorg-video-amdgpu/orig/xserver-xorg-video-amdgpu-23.0.0/debian/patches
Source directories searched:
/home/benutzer/source/xserver-xorg-video-amdgpu/orig/xserver-xorg-video-amdgpu-23.0.0/debian/patches:$cdir:$cwd
(gdb) file /usr/lib/xorg/Xorg
Reading symbols from /usr/lib/xorg/Xorg...
Reading symbols from
/usr/lib/debug/.build-id/77/6f662cfdbd2d0952921614575518e7c1b90261.debug...
(gdb) tb main
Temporary breakpoint 1 at 0x4d260: file ../../../../dix/stubmain.c, line 33.
(gdb) run
Starting program: /usr/lib/xorg/Xorg
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
Temporary breakpoint 1, main (argc=1, argv=0x7fffe478, envp=0x7fffe488)
at ../../../../dix/stubmain.c:33
33 ../../../../dix/stubmain.c: Datei oder Verzeichnis nicht gefunden.
(gdb) call dlopen("/usr/lib/xorg/modules/drivers/amdgpu_drv.so",0x101)
$1 = (void *) 0x55803a80
(gdb) info inferior
Num Description Connection Executable
* 1process 2535 1 (native) /usr/lib/xorg/Xorg
(gdb) print (int)getpid()
$2 = 2535
(gdb) shell cat /proc/2535/maps | grep amdgpu_drv | head -n1
774f9000-7750 r--p 08:12 681046
/usr/lib/xorg/modules/drivers/amdgpu_drv.so
(gdb) b *(0x774f9000 + 0x16c96)
Breakpoint 2 at 0x7750fc96: file ../../src/drmmode_display.c, line 1371.
(gdb) b *(0x774f9000 + 0x16702)
Breakpoint 3 at 0x7750f702: file ../../src/drmmode_display.c, line 1267.
(gdb) info b
Num Type Disp Enb AddressWhat
2 breakpoint keep y 0x7750fc96 in drmmode_set_mode_major at
../../src/drmmode_display.c:1371
3 breakpoint keep y 0x7750f702 in drmmode_set_mode at
../../src/drmmode_display.c:1267
(gdb) pipe disassemble drmmode_set_mode_major | grep -i c96 -C3
0x7750fc8b <+955>: mov%r14,%rsi
0x7750fc8e <+958>: mov%rbx,%rdi
0x7750fc91 <+961>: call 0x7750f650
0x7750fc96 <+966>: test %eax,%eax
0x7750fc98 <+968>: jne0x7750fef8
0x7750fc9e <+974>: mov0x30(%rsp),%rdi
0x7750fca3 <+979>: lea0x60(%rsp),%rsi
(gdb) pipe disassemble drmmode_set_mode | grep -i 702 -C3
0x7750f6f8 <+168>: jne0x7750f710
0x7750f6fa <+170>: mov0x78(%rsi),%rsi
0x7750f6fe <+174>: mov0x10(%rsi),%rsi
0x7750f702 <+178>: mov(%rsi),%r11d
0x7750f705 <+181>: movslq %ecx,%rsi
0x7750f70
[Git][xorg-team/xorg][debian-unstable] close a bug
Timo Aaltonen pushed to branch debian-unstable at X Strike Force / xorg Commits: f04cfbe7 by Timo Aaltonen at 2024-05-04T18:58:24+03:00 close a bug - - - - - 1 changed file: - debian/changelog Changes: = debian/changelog = @@ -18,7 +18,7 @@ xorg (1:7.7+24) UNRELEASED; urgency=medium duplicates source. [ Timo Aaltonen ] - * control: Drop libdmx-dev from xorg-dev depends. + * control: Drop libdmx-dev from xorg-dev depends. (Closes: #1035474) -- Debian Janitor Thu, 13 Oct 2022 08:40:16 - View it on GitLab: https://salsa.debian.org/xorg-team/xorg/-/commit/f04cfbe79f1ac129d8f71e010dc007c4e5f09a9b -- View it on GitLab: https://salsa.debian.org/xorg-team/xorg/-/commit/f04cfbe79f1ac129d8f71e010dc007c4e5f09a9b You're receiving this email because of your account on salsa.debian.org.
mesa_24.1.0~rc2-1_source.changes ACCEPTED into experimental
Thank you for your contribution to Debian. Accepted: -BEGIN PGP SIGNED MESSAGE- Hash: SHA512 Format: 1.8 Date: Sat, 04 May 2024 19:00:30 +0300 Source: mesa Built-For-Profiles: noudeb Architecture: source Version: 24.1.0~rc2-1 Distribution: experimental Urgency: medium Maintainer: Debian X Strike Force Changed-By: Timo Aaltonen Changes: mesa (24.1.0~rc2-1) experimental; urgency=medium . * New upstream release candidate. Checksums-Sha1: 2cbbdc3353874848ab0196f698f786b1fe02f4bb 5657 mesa_24.1.0~rc2-1.dsc acb5931bcdfc46ad5f5f4f6c07e7a780827d9fea 46840795 mesa_24.1.0~rc2.orig.tar.gz 4a1aa525f09685c74e9cbfe5fe30d9a2a5529c94 115936 mesa_24.1.0~rc2-1.diff.gz 7fe6b85f5e51749ed24b1af3670728a3f3794ce9 10604 mesa_24.1.0~rc2-1_source.buildinfo Checksums-Sha256: 3c983b7e53f0179ca1232a5b36d439e33a7cb0cc7a7f1d548ea16a8e7c2a1502 5657 mesa_24.1.0~rc2-1.dsc c32224c1622aa6c28ae8c494b7e18386f7ce6f77ade6a303ec93b5fa9216a5bd 46840795 mesa_24.1.0~rc2.orig.tar.gz e106f6d82c78c23af217755d6ad5711d403e508215371c099534db79847fb325 115936 mesa_24.1.0~rc2-1.diff.gz 386384015117eafea2793384f944ac8387e923e5394146c9fab537275f09b92d 10604 mesa_24.1.0~rc2-1_source.buildinfo Files: 67b4234039da080f3e84254e0a865144 5657 graphics optional mesa_24.1.0~rc2-1.dsc 9eec7e18951ae1cc79fb0c2daf6a7e6c 46840795 graphics optional mesa_24.1.0~rc2.orig.tar.gz c27daf9fc0700ee623b26828d05b4880 115936 graphics optional mesa_24.1.0~rc2-1.diff.gz 1489230a4a660f8024d36ee980ae3949 10604 graphics optional mesa_24.1.0~rc2-1_source.buildinfo -BEGIN PGP SIGNATURE- iQIzBAEBCgAdFiEEdS3ifE3rFwGbS2Yjy3AxZaiJhNwFAmY2W7EACgkQy3AxZaiJ hNwRuhAArJ4sQ0bSmJXqJp/f5BpDAQ12rpjRhhboSRScizNn2GtCugnFaxBK0K7Q S28pQZepgCOYsrbbzC5byZoOTPvngTpitH9JBvPGXRsSeaYY3ntjZFr/aeJUot8S TGxcaYJhB3R7oEDKySf5B+f0TSqPVFVio0e9Xq2/oY7S8o/OqEmnJWZbUFhVk+QP spig+v5E/b3g4364I4DcOnOWXH8t5fMp9bZvnce75TNbUU5j2878rgPdWrO/TEnX jbHYXv5/DkmSrVcOofryYiVvm7xm0w67uHWdt2ovu/A/UyIox+Fs6vEW4VGuZGTQ ctThNIoBQcxWWTFwsBGKhWyy0JhkzxzBRv5/wSZ/aqtgnrgfmg6bYm5xaXASKQ8X csBDKfNlMoP0yd1ko8vGLcpvEBob/loZkbYM9ZjFMFpkTaLygcCnbCQsuePfO2+r MDQuuQRjVvvX5LfOubqa6p3RwFmbus/G74rRhw8u8AM/B5iW2ywsUOskIXGpVMoq i/RuPkoqGPrZvlHrf1M2+Wkfchp3jwDRj05mPG8HL/wFphSv6Fg3NR0d13lK+opP cnomTqOS+3lPypz8VNr+SJ14RPXQGIS8WAx+HQyou+pEnjBT1sP29SUa+JPzPAYH Re+11bt8klNEwzrVplo47RcwX+wIzyU/g2ng2GV9jEfx0MFy168= =M2XA -END PGP SIGNATURE- pgpNGRCT3gruh.pgp Description: PGP signature
[Git][xorg-team/lib/mesa] Pushed new tag mesa-24.1.0-rc2-1
Timo Aaltonen pushed new tag mesa-24.1.0-rc2-1 at X Strike Force / lib / mesa -- View it on GitLab: https://salsa.debian.org/xorg-team/lib/mesa/-/tree/mesa-24.1.0-rc2-1 You're receiving this email because of your account on salsa.debian.org.
[Git][xorg-team/lib/mesa][debian-experimental] 53 commits: .pick_status.json: Update to 4660ee1deaace6457bf5fbf3fc8810e4a2453cb5
Timo Aaltonen pushed to branch debian-experimental at X Strike Force / lib /
mesa
Commits:
418e19a6 by Eric Engestrom at 2024-04-25T16:16:21+02:00
.pick_status.json: Update to 4660ee1deaace6457bf5fbf3fc8810e4a2453cb5
- - - - -
1d8901f3 by Eric Engestrom at 2024-04-25T16:48:54+02:00
ci: fix container rules on release branches and tags
Fixes: 2487e18d4e25c56a7063 ("ci: bare-metal: poe: Create strutured
logs")
Fixes: 7c0b19a6071dc8fa0cd8 ("ci: run python-test automatically only in
merge pipelines")
Part-of: ;
(cherry picked from commit 0666a715c7210558017ce717f6b0b947c679a68e)
- - - - -
69278233 by Eric Engestrom at 2024-04-25T16:50:22+02:00
.pick_status.json: Update to 84632dce93f44e8d88cda47648cfd4cc0958918f
- - - - -
99627c61 by Samuel Pitoiset at 2024-04-25T16:50:26+02:00
radv: fix image format properties with fragment shading rate usage
This was missing and this caused test failures for formats different
than VK_FORMAT_R8_UINT which is the only one supported for FSR.
Fixes recent
dEQP-VK.api.info.unsupported_image_usage.*.fragment_shading_rate_attachment.*.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset
Part-of: ;
(cherry picked from commit e8d94536d21a6f2087b21eecc9f15dcac0a362e8)
- - - - -
cc5da81e by Kenneth Graunke at 2024-04-25T16:50:27+02:00
isl: Set MOCS to uncached for Gfx12.0 blitter sources/destinations
We were accidentally leaving XY_BLOCK_COPY_BLT's Source and Destination
MOCS fields set to 0 (Error: Reserved for Non-Use) on Gfx12.0 systems.
This was causing assert fails in debug builds, since we try to ensure
that we don't do that. In theory, MOCS 0 is supposed to be equivalent
to MOCS 2 (all the caching), but...we probably ought to use MOCS 3
(uncached). Every Gfx12.5+ platform requires it, so although there
isn't a note about Gfx12.0 needing that, it's possible that it does.
We're currently only using the blitter for DRI PRIME blits on Gfx12.0,
anyway, and I think we're flushing all the caches regardless.
This bug was somewhat obscure to hit:
- You need a hybrid graphics system with Gfx12.0 and some other GPU
- You have to be using "reverse PRIME", i.e. rendering on the
integrated
GPU and displaying on the discrete one. This is not the common case.
- You have to be using a debug build.
No observable performance delta in GfxBench5 Car Chase (an arbitrary
program) when rendering on Alderlake GT1 and displaying on an Arc A770.
Fixes: 194afe84163 ("anv/iris/blorp: use the right MOCS values for each
engine")
Reviewed-by: Lionel Landwerlin
Reviewed-by: Rohan Garg
Part-of: ;
(cherry picked from commit e6fb3ba03798fc2550bdb5ec6651690a34ac3509)
- - - - -
e08d22ae by Lionel Landwerlin at 2024-04-25T16:50:27+02:00
anv: disable dual source blending state if not used in shader
Fixing some simulation issues on Gfx9/11 with zink on anv running dual
source blending piglit tests like :
./bin/arb_blend_func_extended-dual-src-blending-discard-without-src1 -auto
-fbo
Signed-off-by: Lionel Landwerlin
Cc: mesa-stable
Reviewed-by: Tapani Pälli
Part-of: ;
(cherry picked from commit 68dfe17abcca6299951734df3f9758c401286b57)
- - - - -
01f45e49 by Mary Guillemard at 2024-04-25T16:50:28+02:00
nak: Pass has_mod to all form of src2 requiring it
This was missing from the original changes and was causing HFMA2 to
misbehave with an immediate value.
Also fix inverted value passed around for cbuf and ureg forms.
Fixes: bad23ddb484 ("nak: Add F16 and F16v2 sources")
Signed-off-by: Mary Guillemard
Part-of: ;
(cherry picked from commit 40422927dcb1249cdb1444557aca671387140428)
- - - - -
15466d7d by Constantine Shablia at 2024-04-25T16:50:28+02:00
pan/bi: fix 1D array tex coord lowering
We were erroneously specifying Y for 1D arrays
Cc: mesa-stable
Suggested-by: Erik Faye-Lund
Reviewed-by: Erik Faye-Lund
Reviewed-by: Boris Brezillon
Reviewed-by: Mary Guillemard
Part-of: ;
(cherry picked from commit 3139f8f62384901b02d1fdc516ff3989f8a1f339)
- - - - -
30f53526 by Mike Blumenkrantz at 2024-04-25T16:50:35+02:00
glthread: check for invalid primitive modes in DrawElementsBaseVertex
fixes KHR-GLESEXT.draw_elements_base_vertex_tests.invalid_mode_argument
cc: mesa-stable
Reviewed-by: Marek Olšák
Part-of: ;
(cherry picked from commit
[Git][xorg-team/lib/mesa][upstream-experimental] 51 commits: .pick_status.json: Update to 4660ee1deaace6457bf5fbf3fc8810e4a2453cb5
Timo Aaltonen pushed to branch upstream-experimental at X Strike Force / lib /
mesa
Commits:
418e19a6 by Eric Engestrom at 2024-04-25T16:16:21+02:00
.pick_status.json: Update to 4660ee1deaace6457bf5fbf3fc8810e4a2453cb5
- - - - -
1d8901f3 by Eric Engestrom at 2024-04-25T16:48:54+02:00
ci: fix container rules on release branches and tags
Fixes: 2487e18d4e25c56a7063 ("ci: bare-metal: poe: Create strutured
logs")
Fixes: 7c0b19a6071dc8fa0cd8 ("ci: run python-test automatically only in
merge pipelines")
Part-of: ;
(cherry picked from commit 0666a715c7210558017ce717f6b0b947c679a68e)
- - - - -
69278233 by Eric Engestrom at 2024-04-25T16:50:22+02:00
.pick_status.json: Update to 84632dce93f44e8d88cda47648cfd4cc0958918f
- - - - -
99627c61 by Samuel Pitoiset at 2024-04-25T16:50:26+02:00
radv: fix image format properties with fragment shading rate usage
This was missing and this caused test failures for formats different
than VK_FORMAT_R8_UINT which is the only one supported for FSR.
Fixes recent
dEQP-VK.api.info.unsupported_image_usage.*.fragment_shading_rate_attachment.*.
Cc: mesa-stable
Signed-off-by: Samuel Pitoiset
Part-of: ;
(cherry picked from commit e8d94536d21a6f2087b21eecc9f15dcac0a362e8)
- - - - -
cc5da81e by Kenneth Graunke at 2024-04-25T16:50:27+02:00
isl: Set MOCS to uncached for Gfx12.0 blitter sources/destinations
We were accidentally leaving XY_BLOCK_COPY_BLT's Source and Destination
MOCS fields set to 0 (Error: Reserved for Non-Use) on Gfx12.0 systems.
This was causing assert fails in debug builds, since we try to ensure
that we don't do that. In theory, MOCS 0 is supposed to be equivalent
to MOCS 2 (all the caching), but...we probably ought to use MOCS 3
(uncached). Every Gfx12.5+ platform requires it, so although there
isn't a note about Gfx12.0 needing that, it's possible that it does.
We're currently only using the blitter for DRI PRIME blits on Gfx12.0,
anyway, and I think we're flushing all the caches regardless.
This bug was somewhat obscure to hit:
- You need a hybrid graphics system with Gfx12.0 and some other GPU
- You have to be using "reverse PRIME", i.e. rendering on the
integrated
GPU and displaying on the discrete one. This is not the common case.
- You have to be using a debug build.
No observable performance delta in GfxBench5 Car Chase (an arbitrary
program) when rendering on Alderlake GT1 and displaying on an Arc A770.
Fixes: 194afe84163 ("anv/iris/blorp: use the right MOCS values for each
engine")
Reviewed-by: Lionel Landwerlin
Reviewed-by: Rohan Garg
Part-of: ;
(cherry picked from commit e6fb3ba03798fc2550bdb5ec6651690a34ac3509)
- - - - -
e08d22ae by Lionel Landwerlin at 2024-04-25T16:50:27+02:00
anv: disable dual source blending state if not used in shader
Fixing some simulation issues on Gfx9/11 with zink on anv running dual
source blending piglit tests like :
./bin/arb_blend_func_extended-dual-src-blending-discard-without-src1 -auto
-fbo
Signed-off-by: Lionel Landwerlin
Cc: mesa-stable
Reviewed-by: Tapani Pälli
Part-of: ;
(cherry picked from commit 68dfe17abcca6299951734df3f9758c401286b57)
- - - - -
01f45e49 by Mary Guillemard at 2024-04-25T16:50:28+02:00
nak: Pass has_mod to all form of src2 requiring it
This was missing from the original changes and was causing HFMA2 to
misbehave with an immediate value.
Also fix inverted value passed around for cbuf and ureg forms.
Fixes: bad23ddb484 ("nak: Add F16 and F16v2 sources")
Signed-off-by: Mary Guillemard
Part-of: ;
(cherry picked from commit 40422927dcb1249cdb1444557aca671387140428)
- - - - -
15466d7d by Constantine Shablia at 2024-04-25T16:50:28+02:00
pan/bi: fix 1D array tex coord lowering
We were erroneously specifying Y for 1D arrays
Cc: mesa-stable
Suggested-by: Erik Faye-Lund
Reviewed-by: Erik Faye-Lund
Reviewed-by: Boris Brezillon
Reviewed-by: Mary Guillemard
Part-of: ;
(cherry picked from commit 3139f8f62384901b02d1fdc516ff3989f8a1f339)
- - - - -
30f53526 by Mike Blumenkrantz at 2024-04-25T16:50:35+02:00
glthread: check for invalid primitive modes in DrawElementsBaseVertex
fixes KHR-GLESEXT.draw_elements_base_vertex_tests.invalid_mode_argument
cc: mesa-stable
Reviewed-by: Marek Olšák
Part-of: ;
(cherry picked from comm
Processing of mesa_24.1.0~rc2-1_source.changes
mesa_24.1.0~rc2-1_source.changes uploaded successfully to localhost along with the files: mesa_24.1.0~rc2-1.dsc mesa_24.1.0~rc2.orig.tar.gz mesa_24.1.0~rc2-1.diff.gz mesa_24.1.0~rc2-1_source.buildinfo Greetings, Your Debian queue daemon (running on host usper.debian.org)
