So far happened only when watching videos (vlc, web), most usually in full- screen mode, but there have been a few non-full-screen occurrences as well. Blackouts last around 0.5-1.5s and seem to be happening quite randomly: sometimes I'm able to watch a 2 hours video without a single incident, then when rewatching the same video it happens every 2 minutes. After each blackout there *always* appears an entry in logs like the below: kernel: i915 0000:00:02.0: [drm] *ERROR* CPU pipe X FIFO underrun ...where 'X' is either 'A' or 'B'. I switched from Ubuntu-18.04 on the same hardware few months ago and on the previous system this problem has never happened. I've installed fully upgraded Ubuntu-24.04 on a separate drive and there I could not reproduce the problem (although I watched just 2 45-minute videos, so given the randomness of occurrences, this does not completely exclude the bug being present there as well). I've also installed fully upgraded Debian Bookworm on a separate drive and there the problem DID occur. I usually use an external display (please find xrandr.txt file attached with the output of `xrandr --verbose`), so today I've tried to watch few videos on my laptop's internal display and I was not able to reproduce the problem (but again given the randomness it should be taken with a grain of salt). The problem has occurred on all versions of Trixie kernel that I've run so far: 6.7.12, 6.9.7, 6.9.8, 6.9.9. Attached please find the output of `last reboot` and `journalctl |grep i915` in files last-reboot.txt and i915.log respectively.
Just in case I've just verified that the issue occurs both when watching videos with codecs that are and are not hardware-supportted by my GPU.
Nevertheless in case it matters I have the following video-related packages installed:
intel-media-va-driver:amd64 24.2.5+dfsg1-1
intel-media-va-driver:i386 24.2.5+dfsg1-1
i965-va-driver:amd64 2.4.1+dfsg1-1
i965-va-driver:i386 2.4.1+dfsg1-1
firmware-intel-graphics 20240610-1
the following packages are NOT installed:
intel-media-va-driver-non-free
i965-va-driver-shaders
LIBVA_DRIVER_NAME env var is NOT defined
vlc has "Hardware-accelerated decoding" set to "Automatic"
When playing videos with H264 codec `intel_gpu_top` shows that the video engine is being used, for all other codecs it is idle.
Firefox config has been modified according to https://wiki.debian.org/Firefox#Hardware_Video_Acceleration instructions:
launcher is defined as the below:
Exec=env MOZ_X11_EGL=1 /usr/lib/firefox-esr/firefox-esr %u
related config variables are defined as follows:
media.ffmpeg.vaapi.enabled true
media.ffvpx.enabled false
media.av1.enabled false
I have verified with `intel_gpu_top` that the video engine is being used when watching videos in Firefox.
When I was trying to reproduce the bug on Ubuntu-24.04 I modified the config the same way and also verified with intel_gpu_top that the video engine was active.
vainfo output:
libva info: VA-API version 1.22.0
libva info: Trying to open /usr/lib/x86_64-linux-gnu/dri/iHD_drv_video.so
libva info: Found init function __vaDriverInit_1_22
libva info: va_openDriver() returns 0
vainfo: VA-API version: 1.22 (libva 2.12.0)
vainfo: Driver version: Intel iHD driver for Intel(R) Gen Graphics - 24.2.5 ()
vainfo: Supported profile and entrypoints
VAProfileMPEG2Simple : VAEntrypointVLD
VAProfileMPEG2Main : VAEntrypointVLD
VAProfileH264Main : VAEntrypointVLD
VAProfileH264Main : VAEntrypointEncSliceLP
VAProfileH264High : VAEntrypointVLD
VAProfileH264High : VAEntrypointEncSliceLP
VAProfileJPEGBaseline : VAEntrypointVLD
VAProfileJPEGBaseline : VAEntrypointEncPicture
VAProfileH264ConstrainedBaseline: VAEntrypointVLD
VAProfileH264ConstrainedBaseline: VAEntrypointEncSliceLP
VAProfileVP8Version0_3 : VAEntrypointVLD
VAProfileHEVCMain : VAEntrypointVLD
VAProfileHEVCMain10 : VAEntrypointVLD
VAProfileVP9Profile0 : VAEntrypointVLD
VAProfileVP9Profile2 : VAEntrypointVLD
vdpauinfo output:
display: :0 screen: 0
libva info: VA-API version 1.22.0
libva info: Trying to open /usr/lib/x86_64-linux-gnu/dri/iHD_drv_video.so
libva info: Found init function __vaDriverInit_1_22
libva info: va_openDriver() returns 0
API version: 1
Information string: OpenGL/VAAPI backend for VDPAU
Video surface:
name width height types
-------------------------------------------
420 4096 4096 NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8 NV24 YV24 P010 P016 Y_U_V_444_16
422 4096 4096 NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8 NV24 YV24 P010 P016 Y_U_V_444_16
444 4096 4096 NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8 NV24 YV24 P010 P016 Y_U_V_444_16
420_16 4096 4096 NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8 NV24 YV24 P010 P016 Y_U_V_444_16
422_16 4096 4096 NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8 NV24 YV24 P010 P016 Y_U_V_444_16
444_16 4096 4096 NV12 YV12 UYVY YUYV Y8U8V8A8 V8U8Y8A8 NV24 YV24 P010 P016 Y_U_V_444_16
Decoder capabilities:
name level macbs width height
----------------------------------------------------
MPEG1 --- not supported ---
MPEG2_SIMPLE --- not supported ---
MPEG2_MAIN --- not supported ---
H264_BASELINE 51 16384 2048 2048
H264_MAIN 51 16384 2048 2048
H264_HIGH 51 16384 2048 2048
VC1_SIMPLE --- not supported ---
VC1_MAIN --- not supported ---
VC1_ADVANCED --- not supported ---
MPEG4_PART2_SP --- not supported ---
MPEG4_PART2_ASP --- not supported ---
DIVX4_QMOBILE --- not supported ---
DIVX4_MOBILE --- not supported ---
DIVX4_HOME_THEATER --- not supported ---
DIVX4_HD_1080P --- not supported ---
DIVX5_QMOBILE --- not supported ---
DIVX5_MOBILE --- not supported ---
DIVX5_HOME_THEATER --- not supported ---
DIVX5_HD_1080P --- not supported ---
H264_CONSTRAINED_BASELINE 51 16384 2048 2048
H264_EXTENDED --- not supported ---
H264_PROGRESSIVE_HIGH --- not supported ---
H264_CONSTRAINED_HIGH --- not supported ---
H264_HIGH_444_PREDICTIVE --- not supported ---
VP9_PROFILE_0 --- not supported ---
VP9_PROFILE_1 --- not supported ---
VP9_PROFILE_2 --- not supported ---
VP9_PROFILE_3 --- not supported ---
HEVC_MAIN --- not supported ---
HEVC_MAIN_10 --- not supported ---
HEVC_MAIN_STILL --- not supported ---
HEVC_MAIN_12 --- not supported ---
HEVC_MAIN_444 --- not supported ---
HEVC_MAIN_444_10 --- not supported ---
HEVC_MAIN_444_12 --- not supported ---
AV1_MAIN --- not supported ---
AV1_HIGH --- not supported ---
AV1_PROFESSIONAL --- not supported ---
Output surface:
name width height nat types
----------------------------------------------------
B8G8R8A8 16384 16384 y
R8G8B8A8 16384 16384 y
R10G10B10A2 16384 16384 y
B10G10R10A2 16384 16384 y
A8 16384 16384 y
Bitmap surface:
name width height
------------------------------
B8G8R8A8 16384 16384
R8G8B8A8 16384 16384
R10G10B10A2 16384 16384
B10G10R10A2 16384 16384
A8 16384 16384
Video mixer:
feature name sup
------------------------------------
DEINTERLACE_TEMPORAL -
DEINTERLACE_TEMPORAL_SPATIAL -
INVERSE_TELECINE -
NOISE_REDUCTION -
SHARPNESS -
LUMA_KEY -
HIGH QUALITY SCALING - L1 -
HIGH QUALITY SCALING - L2 -
HIGH QUALITY SCALING - L3 -
HIGH QUALITY SCALING - L4 -
HIGH QUALITY SCALING - L5 -
HIGH QUALITY SCALING - L6 -
HIGH QUALITY SCALING - L7 -
HIGH QUALITY SCALING - L8 -
HIGH QUALITY SCALING - L9 -
parameter name sup min max
-----------------------------------------------------
VIDEO_SURFACE_WIDTH -
VIDEO_SURFACE_HEIGHT -
CHROMA_TYPE -
LAYERS -
attribute name sup min max
-----------------------------------------------------
BACKGROUND_COLOR -
CSC_MATRIX -
NOISE_REDUCTION_LEVEL -
SHARPNESS_LEVEL -
LUMA_KEY_MIN_LUMA -
LUMA_KEY_MAX_LUMA -
I've just replicated this bug on Ubuntu-24.04 (kernel 6.8.0-39-generic), so it seems to be an upstream bug rather than a Debian specific.
This seems to be the same or a very similar issue as posted here: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/5455 I will be posting further updates there.
Hi This bug was filed for a (very) old kernel or the bug is old itself without resolution. Maybe it was for a feature enablement which nobody acted on. We are sorry we were not able to timely deal with this issue. There are many open bugs for the src:linux package and thus we are closing older bugs where it's unclear if they still occur in newer versions and are still relevant to the reporter. For an overview see: https://bugs.debian.org/src:linux . If you can reproduce your issue with - the current version in unstable/testing - the latest kernel from backports or, if it was a feature addition/wishlist and still consider it relevant, then: Please reopen the bug, see https://www.debian.org/Bugs/server-control for details. Please try to provide as much fresh details including kernel logs where relevant. In particular were an issue is coupled with specific hardware we might ask you to do additional debugging on your side as the owner of the hardware. Regards, Salvatore
Hi This bug was filed for a (very) old kernel or the bug is old itself without resolution. Maybe it was for a feature enablement which nobody acted on. We are sorry we were not able to timely deal with this issue. There are many open bugs for the src:linux package and thus we are closing older bugs where it's unclear if they still occur in newer versions and are still relevant to the reporter. For an overview see: https://bugs.debian.org/src:linux . If you can reproduce your issue with - the current version in unstable/testing - the latest kernel from backports or, if it was a feature addition/wishlist and still consider it relevant, then: Please reopen the bug, see https://www.debian.org/Bugs/server-control for details. Please try to provide as much fresh details including kernel logs where relevant. In particular were an issue is coupled with specific hardware we might ask you to do additional debugging on your side as the owner of the hardware. Regards, Salvatore
Hi and sorry for the delay: I was traveling until recently and didn't have a full access to the hardware in question. This bug is still present in kernel 6.12 and 6.14 (I've experienced it during last few days when using 6.12.25-1, 6.12.27-1, 6.14.5-1~exp1 and 6.14.6-1~exp1 on trixie). Generally please refer to the mentioned upstream bug https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/5455 : this bug should probably not be closed until the upstream one is. I hope to find some time this week to try to replicate it using drm-tip kernel as requested by the upstream maintainers. Cheers!
Hello Piotr, This upstream bug was closed recently as it's not reproducible for the original reporter on 6.15.4. Can you please test this (or a newer) kernel version an confirm? 6.15.4 is in experimental which should make this easy. Best regards Uwe
Note that in the meantime the bug was reopened again. This doesn't make retesting on 6.15 useless, as that would at least result in a bump for upstream that the issue is still relevant for more than one person. Best regards Uwe