FYI,
you might have seen this problem, but for the record I am reporting here,
Following test cases have been failing for a while on the mainline kernel.
Please find crash log and git short between bad and good commit.
List of failed test cases:
v4l2-compliance
* VIDIOC_ENUM_FMT-FRAMESIZES-FRAMEINTERVALS
* VIDIOC_G_FMT
* VIDIOC_S_FMT
* VIDIOC_TRY_FMT
test output log:
-------------------
Format ioctls (Input 0):
fail: ../../../v4l-utils-1.18.1/utils/v4l2-compliance/v4l2-test-formats.cpp(265):
unknown flag 000001e0 returned
test VIDIOC_ENUM_FMT/FRAMESIZES/FRAMEINTERVALS: FAIL
warn: ../../../v4l-utils-1.18.1/utils/v4l2-compliance/v4l2-test-formats.cpp(1320):
S_PARM is supported for buftype 1, but not for ENUM_FRAMEINTERVALS
test VIDIOC_G/S_PARM: OK
test VIDIOC_G_FBUF: OK
fail: ../../../v4l-utils-1.18.1/utils/v4l2-compliance/v4l2-test-formats.cpp(444):
pixelformat 56595559 (YUYV) for buftype 1 not reported by ENUM_FMT
test VIDIOC_G_FMT: FAIL
fail: ../../../v4l-utils-1.18.1/utils/v4l2-compliance/v4l2-test-formats.cpp(444):
pixelformat 56595559 (YUYV) for buftype 1 not reported by ENUM_FMT
test VIDIOC_TRY_FMT: FAIL
fail: ../../../v4l-utils-1.18.1/utils/v4l2-compliance/v4l2-test-formats.cpp(444):
pixelformat 56595559 (YUYV) for buftype 1 not reported by ENUM_FMT
test VIDIOC_S_FMT: FAIL
Reported-by: Naresh Kamboju <[email protected]>
# git log --oneline dfef313e9990..c4439713e82a -- drivers/media/v4l2-core/
1efe3c28eba1 media: v4l2-mem2mem: Fix spurious v4l2_m2m_buf_done
463c43fcd97e Merge tag 'v5.9-rc7' into patchwork
1698a7f15112 media: v4l2-mem2mem: simplify poll logic
566463afdbc4 media: v4l2-mem2mem: always consider OUTPUT queue during poll
5823833c9ada media: v4l2-ctrl: Add VP9 codec levels
32b6e400c34d media: v4l2-fwnode: Use debug level for printing link frequencies
d13ee586e026 Merge tag 'v5.9-rc4' into patchwork
1faa39e0f3bc media: videobuf-dma-sg: number of pages should be unsigned long
44f5b2fffc32 media: v4l2-ctrl: Add frame-skip std encoder control
4ad1b0d410c8 media: v4l2-ctrls: Add encoder constant quality control
d9358563179a media: uapi: h264: Clean slice invariants syntax elements
f9879eb37829 media: uapi: h264: Increase size of DPB entry pic_num
c02ff21952a6 media: uapi: h264: Clean DPB entry interface
eb44c6c9c236 media: uapi: h264: Split prediction weight parameters
911ea8ec42de media: v4l2-mem2mem: add v4l2_m2m_suspend, v4l2_m2m_resume
38df0b850613 media: v4l2-subdev: Introduce [get|set]_mbus_config pad ops
Full test log,
https://lkft.validation.linaro.org/scheduler/job/1987206#L1664
--
Linaro LKFT
https://lkft.linaro.org
On 27/11/2020 15:00, Naresh Kamboju wrote:
> FYI,
> you might have seen this problem, but for the record I am reporting here,
> Following test cases have been failing for a while on the mainline kernel.
> Please find crash log and git short between bad and good commit.
>
> List of failed test cases:
> v4l2-compliance
> * VIDIOC_ENUM_FMT-FRAMESIZES-FRAMEINTERVALS
> * VIDIOC_G_FMT
> * VIDIOC_S_FMT
> * VIDIOC_TRY_FMT
>
> test output log:
> -------------------
> Format ioctls (Input 0):
> fail: ../../../v4l-utils-1.18.1/utils/v4l2-compliance/v4l2-test-formats.cpp(265):
1.18.1 is much too old.
If you want reliable output from v4l2-compliance, then always compile from the
git repository (git://linuxtv.org/v4l-utils.git).
The version distributed by distros is much too infrequently updated, and shouldn't
be used.
If you still have issues with the version from the git repo, then let me know!
Regards,
Hans
> unknown flag 000001e0 returned
> test VIDIOC_ENUM_FMT/FRAMESIZES/FRAMEINTERVALS: FAIL
> warn: ../../../v4l-utils-1.18.1/utils/v4l2-compliance/v4l2-test-formats.cpp(1320):
> S_PARM is supported for buftype 1, but not for ENUM_FRAMEINTERVALS
> test VIDIOC_G/S_PARM: OK
> test VIDIOC_G_FBUF: OK
> fail: ../../../v4l-utils-1.18.1/utils/v4l2-compliance/v4l2-test-formats.cpp(444):
> pixelformat 56595559 (YUYV) for buftype 1 not reported by ENUM_FMT
> test VIDIOC_G_FMT: FAIL
> fail: ../../../v4l-utils-1.18.1/utils/v4l2-compliance/v4l2-test-formats.cpp(444):
> pixelformat 56595559 (YUYV) for buftype 1 not reported by ENUM_FMT
> test VIDIOC_TRY_FMT: FAIL
> fail: ../../../v4l-utils-1.18.1/utils/v4l2-compliance/v4l2-test-formats.cpp(444):
> pixelformat 56595559 (YUYV) for buftype 1 not reported by ENUM_FMT
> test VIDIOC_S_FMT: FAIL
>
> Reported-by: Naresh Kamboju <[email protected]>
>
> # git log --oneline dfef313e9990..c4439713e82a -- drivers/media/v4l2-core/
> 1efe3c28eba1 media: v4l2-mem2mem: Fix spurious v4l2_m2m_buf_done
> 463c43fcd97e Merge tag 'v5.9-rc7' into patchwork
> 1698a7f15112 media: v4l2-mem2mem: simplify poll logic
> 566463afdbc4 media: v4l2-mem2mem: always consider OUTPUT queue during poll
> 5823833c9ada media: v4l2-ctrl: Add VP9 codec levels
> 32b6e400c34d media: v4l2-fwnode: Use debug level for printing link frequencies
> d13ee586e026 Merge tag 'v5.9-rc4' into patchwork
> 1faa39e0f3bc media: videobuf-dma-sg: number of pages should be unsigned long
> 44f5b2fffc32 media: v4l2-ctrl: Add frame-skip std encoder control
> 4ad1b0d410c8 media: v4l2-ctrls: Add encoder constant quality control
> d9358563179a media: uapi: h264: Clean slice invariants syntax elements
> f9879eb37829 media: uapi: h264: Increase size of DPB entry pic_num
> c02ff21952a6 media: uapi: h264: Clean DPB entry interface
> eb44c6c9c236 media: uapi: h264: Split prediction weight parameters
> 911ea8ec42de media: v4l2-mem2mem: add v4l2_m2m_suspend, v4l2_m2m_resume
> 38df0b850613 media: v4l2-subdev: Introduce [get|set]_mbus_config pad ops
>
>
> Full test log,
> https://lkft.validation.linaro.org/scheduler/job/1987206#L1664
>