[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#983860: marked as done (segfaults because tries to use non-free kernels)



Your message dated Wed, 04 Aug 2021 22:48:38 +0000
with message-id <E1mBPgk-0004UE-R5@fasolo.debian.org>
and subject line Bug#983860: fixed in intel-media-driver 21.3.0+dfsg1-1
has caused the Debian Bug report #983860,
regarding segfaults because tries to use non-free kernels
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
983860: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=983860
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message ---
Package: intel-media-va-driver
Version: 21.1.1+dfsg1-1
Severity: important

On my CPU iHD_drv_video.so segfaults when decoding an MP4 movie. I can for
example reproduce it by installing gstreamer1.0-vaapi and running:

$ gst-launch-1.0 playbin uri=https://test-
videos.co.uk/vids/bigbuckbunny/mp4/h264/360/Big_Buck_Bunny_360_10s_1MB.mp4

Gdb shows this backtrace (truncated, because it's more than 100 frames,
and the interesting ones are at the top):

#0  KernelDll_AllocateStates(void*, uint32_t, void*, uint32_t, Kdll_RuleEntry
const*, void (*)(PKdll_State)) (pKernelBin=pKernelBin@entry=0x7fffd42e59f0,
uKernelSize=0x0, pFcPatchCache=pFcPatchCache@entry=0x0,
uFcPatchCacheSize=<optimized out>, pDefaultRules=0x0,
ModifyFunctionPointers=0x0) at
./media_driver/agnostic/common/vp/kdll/hal_kerneldll.c:3350
#1  0x00007fffd296bfdb in VphalRenderer::Initialize(VphalSettings const*, bool)
(this=0x7fffd4327e30, pSettings=0x7ffff1745ac0, isApoEnabled=<optimized out>)
at ./media_driver/agnostic/common/vp/hal/vphal_renderer.cpp:1414
#2  0x00007fffd2952e15 in VphalState::Allocate(VphalSettings const*)
(pVpHalSettings=0x7ffff1745ac0, this=0x7fffd42e23a0) at
./media_driver/agnostic/common/vp/hal/vphal.cpp:146
#3  VphalState::Allocate(VphalSettings const*) (this=0x7fffd42e23a0,
pVpHalSettings=0x7ffff1745ac0) at
./media_driver/agnostic/common/vp/hal/vphal.cpp:75
#4  0x00007fffd2b56105 in DdiVp_InitVpHal(DDI_VP_CONTEXT*)
(pVpCtx=0x7fffd42c1680) at
./media_driver/linux/common/vp/ddi/media_libva_vp.c:1800
#5  0x00007fffd2b5a59f in DdiVp_InitCtx(VADriverContext*, DDI_VP_CONTEXT*)
(pVaDrvCtx=<optimized out>, pVpCtx=0x7fffd42c1680) at
./media_driver/linux/common/vp/ddi/media_libva_vp.c:1660
#6  0x00007fffd2b5a99e in DdiVp_CreateContext(VADriverContext*, unsigned int,
int, int, int, unsigned int*, int, unsigned int*)
(pVaDrvCtx=pVaDrvCtx@entry=0x7fffd4265240, vaConfigID=vaConfigID@entry=0x0,
iWidth=iWidth@entry=0x0, iHeight=iHeight@entry=0x0, iFlag=iFlag@entry=0x0,
vaSurfIDs=vaSurfIDs@entry=0x0, iNumSurfs=0x0, pVaCtxID=0x7ffff1745dec) at
./media_driver/linux/common/vp/ddi/media_libva_vp.c:3109
#7  0x00007fffd2b21be2 in DdiMedia_PutImage(VADriverContext*, unsigned int,
unsigned int, int, int, unsigned int, unsigned int, int, int, unsigned int,
unsigned int) (ctx=0x7fffd4265240, surface=0x0, image=<optimized out>,
src_x=0x0, src_y=0x0, src_width=0x40, src_height=0x40, dest_x=0x0, dest_y=0x0,
dest_width=0x40, dest_height=0x40) at
./media_driver/linux/common/ddi/media_libva.cpp:5407
#8  0x00007ffff006a17c in vaPutImage () at /usr/lib/x86_64-linux-gnu/libva.so.2
#9  0x00007ffff00efb2c in gst_vaapi_surface_put_image
(surface=surface@entry=0x7fffd4252d60, image=image@entry=0x7fffd42742d0) at
../gst-libs/gst/vaapi/gstvaapisurface.c:761
#10 0x00007ffff00ae476 in extract_allowed_surface_formats
(img_formats=0x7fffd425c290, display=0x7fffec0046c0
[GstVaapiDisplay|vaapidisplayglx0]) at ../gst/vaapi/gstvaapipluginbase.c:1451
#11 ensure_allowed_raw_caps (plugin=0x7fffd424e910) at
../gst/vaapi/gstvaapipluginbase.c:1483
#12 gst_vaapi_plugin_base_get_allowed_sinkpad_raw_caps
(plugin=plugin@entry=0x7fffd424e910) at ../gst/vaapi/gstvaapipluginbase.c:1515
#13 0x00007ffff00b567e in ensure_allowed_sinkpad_caps (postproc=0x7fffd424e910)
at ../gst/vaapi/gstvaapipostproc.c:1312
#14 gst_vaapipostproc_transform_caps_impl (direction=<optimized out>,
trans=0x7fffd424e910 [GstBaseTransform|vaapipostproc0]) at
../gst/vaapi/gstvaapipostproc.c:1422
#15 gst_vaapipostproc_transform_caps (trans=0x7fffd424e910
[GstBaseTransform|vaapipostproc0], direction=<optimized out>,
caps=0x7fffd4264ad0, filter=0x7fffd4264c50) at
../gst/vaapi/gstvaapipostproc.c:1445
#16 0x00007ffff67154b3 in gst_base_transform_transform_caps
(trans=trans@entry=0x7fffd424e910 [GstBaseTransform|vaapipostproc0],
direction=GST_PAD_SRC, caps=caps@entry=0x7fffd4264ad0,
filter=filter@entry=0x7fffd4264c50) at ../libs/gst/base/gstbasetransform.c:474
#17 0x00007ffff6718e6d in gst_base_transform_query_caps (filter=0x7fffd4264c50,
pad=0x7fffd41f9840 [GstPad|sink], trans=0x7fffd424e910
[GstBaseTransform|vaapipostproc0]) at ../libs/gst/base/gstbasetransform.c:698
#18 gst_base_transform_default_query (trans=0x7fffd424e910
[GstBaseTransform|vaapipostproc0], direction=<optimized out>,
query=0x7fffd4264ca0) at ../libs/gst/base/gstbasetransform.c:1597
#19 0x00007ffff7eace58 in gst_pad_query (pad=pad@entry=0x7fffd41f9840
[GstPad|sink], query=query@entry=0x7fffd4264ca0) at ../gst/gstpad.c:4144
#20 0x00007ffff7ead5bb in gst_pad_peer_query (pad=pad@entry=0x7fffd41f95f0
[GstPad|src], query=query@entry=0x7fffd4264ca0) at ../gst/gstpad.c:4276

Debugging more in depth, I discovered that pKernelBin and uKernelSize
(parameters
of KernelDll_AllocateStates, frame #0) are not initialized by the caller. The
caller
(VphalRenderer::Initialize in
media_driver/agnostic/common/vp/hal/vphal_renderer.cpp)
should initialize the parameters from VphalRenderer members pcKernelBin and
dwKernelBinSize, but these members are in turn never changed from their default
zero value. Initialization should happen (on my CPU, at least) in
VphalRendererG9::InitKdllParam, but the assignment is disabled because the
macro
_FULL_OPEN_SOURCE is defined.

Of course I am not contesting that non-free kernels are not provided in the
free
version of the package; but if they are not provided, neither should they be
expected by another piece of the code.

Thanks, Giovanni.


-- System Information:
Debian Release: bullseye/sid
  APT prefers unstable-debug
  APT policy: (500, 'unstable-debug'), (500, 'unstable'), (500, 'testing'), (1,
'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 5.10.0-3-amd64 (SMP w/12 CPU threads)
Kernel taint flags: TAINT_PROPRIETARY_MODULE, TAINT_OOT_MODULE,
TAINT_UNSIGNED_MODULE
Locale: LANG=it_IT.utf8, LC_CTYPE=it_IT.utf8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages intel-media-va-driver depends on:
ii  libc6                           2.31-9
ii  libgcc-s1                       10.2.1-6
ii  libigdgmm11                     20.4.1+ds1-1
ii  libstdc++6                      10.2.1-6
ii  libva2 [libva-driver-abi-1.10]  2.10.0-1

intel-media-va-driver recommends no packages.

intel-media-va-driver suggests no packages.

--- End Message ---
--- Begin Message ---
Source: intel-media-driver
Source-Version: 21.3.0+dfsg1-1
Done: Sebastian Ramacher <sramacher@debian.org>

We believe that the bug you reported is fixed in the latest version of
intel-media-driver, which is due to be installed in the Debian FTP archive.

A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed.  If you
have further comments please address them to 983860@bugs.debian.org,
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Sebastian Ramacher <sramacher@debian.org> (supplier of updated intel-media-driver package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing ftpmaster@ftp-master.debian.org)


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

Format: 1.8
Date: Wed, 04 Aug 2021 23:13:46 +0200
Source: intel-media-driver
Architecture: source
Version: 21.3.0+dfsg1-1
Distribution: experimental
Urgency: medium
Maintainer: Debian Multimedia Maintainers <debian-multimedia@lists.debian.org>
Changed-By: Sebastian Ramacher <sramacher@debian.org>
Closes: 983860
Changes:
 intel-media-driver (21.3.0+dfsg1-1) experimental; urgency=medium
 .
   * New upstream release
     - Don't crash if non-free kernels are not available (Closes: #983860)
Checksums-Sha1:
 3ac83a97914494f29af405447365a0cd4fb6b5b9 2359 intel-media-driver_21.3.0+dfsg1-1.dsc
 ca51fd5a7b8371e82c93170d929a36ef90fc894a 6422564 intel-media-driver_21.3.0+dfsg1.orig.tar.xz
 c1e6e84995a06eaefa27fda06c1ada3b87ee3f30 5924 intel-media-driver_21.3.0+dfsg1-1.debian.tar.xz
Checksums-Sha256:
 33ec36391d4aa224747d26bec7805f7b035e8636ce32e14e4f8508fa43385da1 2359 intel-media-driver_21.3.0+dfsg1-1.dsc
 7fb72ddd9eacb379103d62e1580e3025d17527470f6c78cbf4ab543f605638ea 6422564 intel-media-driver_21.3.0+dfsg1.orig.tar.xz
 5947caa00fa2d8cc6ae1c90cc3a38ad6498f89c433a0ae08ea5c62d45e72366c 5924 intel-media-driver_21.3.0+dfsg1-1.debian.tar.xz
Files:
 7e24dcc5d17870a5eeb0f865ccaf0cd9 2359 video optional intel-media-driver_21.3.0+dfsg1-1.dsc
 5c8eaaf5e66519b2725c155799623a12 6422564 video optional intel-media-driver_21.3.0+dfsg1.orig.tar.xz
 e5d3052e7979e5b08c128ab38ae8b7f2 5924 video optional intel-media-driver_21.3.0+dfsg1-1.debian.tar.xz

-----BEGIN PGP SIGNATURE-----
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=u4JT
-----END PGP SIGNATURE-----

--- End Message ---

Reply to: