[v2,00/18] Venus QoL / maintainability fixes

Message ID 20230228-topic-venus-v2-0-d95d14949c79@linaro.org (mailing list archive)
Headers
Series Venus QoL / maintainability fixes |

Message

Konrad Dybcio May 4, 2023, 8 a.m. UTC
  v1 -> v2:
- Move "Write to VIDC_CTRL_INIT after unmasking interrupts" up and add
  a Fixes tag & Cc stable
- Reword the comment in "Correct IS_V6() checks"
- Move up "media: venus: Remap bufreq fields on HFI6XX", add Fixes and
  Cc stable
- Use better English in "Use newly-introduced hfi_buffer_requirements
  accessors" commit message
- Mention "Restrict writing SCIACMDARG3 to Venus V1/V2" doesn't seem to
  regress SM8250 in the commit message
- Pick up tags (note: I capitalized the R in Dikshita's 'reviewed-by'
  and removed one occurrence of random '**' to make sure review tools
  like b4 don't go crazy)
- Handle AR50_LITE in "Assign registers based on VPU version"
- Drop /* VPUn */ comments, they're invalid as explained by Vikash
- Take a different approach to the sys_idle problem in patch 1

v1: https://lore.kernel.org/r/20230228-topic-venus-v1-0-58c2c88384e9@linaro.org

Currently upstream assumes all (well, almost all - see 7280 or CrOS
specific checks) Venus implementations using the same version of the
Hardware Firmware Interface can be treated the same way. This is
however not the case.

This series tries to introduce the groundwork to start differentiating
them based on the VPU (Video Processing Unit) hardware type, fixes a
couple of issues that were an effect of that generalized assumption
and lays the foundation for supporting 8150 (IRIS1) and SM6115/QCM2290
(AR50 Lite), which will hopefully come soon.

Tested on 8250, but pretty please test it on your boards too!

Signed-off-by: Konrad Dybcio <konrad.dybcio@linaro.org>
---
Konrad Dybcio (18):
      media: venus: hfi_venus: Only consider sys_idle_indicator on V1
      media: venus: hfi_venus: Write to VIDC_CTRL_INIT after unmasking interrupts
      media: venus: Remap bufreq fields on HFI6XX
      media: venus: Introduce VPU version distinction
      media: venus: Add vpu_version to most SoCs
      media: venus: firmware: Leave a clue for homegrown porters
      media: venus: hfi_venus: Sanitize venus_boot_core() per-VPU-version
      media: venus: core: Assign registers based on VPU version
      media: venus: hfi_venus: Fix version checks in venus_halt_axi()
      media: venus: hfi_venus: Fix version checks in venus_isr()
      media: venus: hfi_venus: Fix version check in venus_cpu_and_video_core_idle()
      media: venus: hfi_venus: Fix version check in venus_cpu_idle_and_pc_ready()
      media: venus: firmware: Correct IS_V6() checks
      media: venus: hfi_platform: Check vpu_version instead of device compatible
      media: venus: vdec: Fix version check in vdec_set_work_route()
      media: venus: Introduce accessors for remapped hfi_buffer_reqs members
      media: venus: Use newly-introduced hfi_buffer_requirements accessors
      media: venus: hfi_venus: Restrict writing SCIACMDARG3 to Venus V1/V2

 drivers/media/platform/qcom/venus/core.c           | 11 ++--
 drivers/media/platform/qcom/venus/core.h           | 15 ++++++
 drivers/media/platform/qcom/venus/firmware.c       | 19 +++++--
 drivers/media/platform/qcom/venus/helpers.c        |  7 +--
 drivers/media/platform/qcom/venus/hfi_helper.h     | 61 +++++++++++++++++++---
 drivers/media/platform/qcom/venus/hfi_msgs.c       |  2 +-
 .../media/platform/qcom/venus/hfi_plat_bufs_v6.c   | 22 ++++----
 drivers/media/platform/qcom/venus/hfi_platform.c   |  2 +-
 drivers/media/platform/qcom/venus/hfi_venus.c      | 45 ++++++++--------
 drivers/media/platform/qcom/venus/vdec.c           | 10 ++--
 drivers/media/platform/qcom/venus/vdec_ctrls.c     |  2 +-
 drivers/media/platform/qcom/venus/venc.c           |  4 +-
 drivers/media/platform/qcom/venus/venc_ctrls.c     |  2 +-
 13 files changed, 138 insertions(+), 64 deletions(-)
---
base-commit: 92e815cf07ed24ee1c51b122f24ffcf2964b4b13
change-id: 20230228-topic-venus-70ea3bc76688

Best regards,
  

Comments

Bryan O'Donoghue May 12, 2023, 3:01 a.m. UTC | #1
On 04/05/2023 09:00, Konrad Dybcio wrote:
> Tested on 8250, but pretty please test it on your boards too!

What's the definition of test here ?

I ran this

ffplay -codec:video h264_v4l2m2m FantasticFour-ROTSS.mp4

and this

ffplay -codec:video vp8_v4l2m2m /mnt/big-buck-bunny_trailer.webm

on db410c with no errors. Then again I applied and disapplied the 8x8 
264 fix to that branch and saw no discernable difference so I'm not very 
confident we have good coverage.

@Stan @Vikash could you give some suggested tests for coverage here ?

@Konrad - get a db410c !

My superficial first-pass on this series looks good but, before giving a 
Tested-by here, I think we should define a set of coverage tests, run 
them - the upper end on sm8250 and lower end msm8916 "makes sense to me"

20? different gstreamer tests at different formats and different sizes 
on our selected platforms db410c, rb5, rb3 I have - also an 820 I 
haven't booted and an enforce sdm660.

Which tests will we use to validate this series and subsequent series to 
ensure we don't have more regressions ?

---
bod
  
Vikash Garodia May 16, 2023, 12:57 p.m. UTC | #2
On 5/12/2023 8:31 AM, Bryan O'Donoghue wrote:
> On 04/05/2023 09:00, Konrad Dybcio wrote:
>> Tested on 8250, but pretty please test it on your boards too!
> 
> What's the definition of test here ?
> 
> I ran this
> 
> ffplay -codec:video h264_v4l2m2m FantasticFour-ROTSS.mp4
> 
> and this
> 
> ffplay -codec:video vp8_v4l2m2m /mnt/big-buck-bunny_trailer.webm
> 
> on db410c with no errors. Then again I applied and disapplied the 8x8 264 fix to
> that branch and saw no discernable difference so I'm not very confident we have
> good coverage.
> 
> @Stan @Vikash could you give some suggested tests for coverage here ?

I could think of below test aspects for this series
1. Suspend Resume
2. Concurrency test
3. Module load -> video usecase -> module unload -> module load -> video
usecase. This would ensure video firmware is reloaded and functional.
4. Video playback and encode for all supported resolution and codecs.
5. In general, video playback with more test content.

I would be testing the series with stability test suite on CrOS. That would be
either on sc7180 or sc7280 setup.

Konrad, you can post the new version as one patch needs to be dropped. Test can
be done on the new version. There are few patches in the series pending review,
which can be done in parallel.

-Vikash

> 
> @Konrad - get a db410c !
> 
> My superficial first-pass on this series looks good but, before giving a
> Tested-by here, I think we should define a set of coverage tests, run them - the
> upper end on sm8250 and lower end msm8916 "makes sense to me"
> 
> 20? different gstreamer tests at different formats and different sizes on our
> selected platforms db410c, rb5, rb3 I have - also an 820 I haven't booted and an
> enforce sdm660.
> 
> Which tests will we use to validate this series and subsequent series to ensure
> we don't have more regressions ?
> 
> ---
> bod
  
Konrad Dybcio May 17, 2023, 8:25 p.m. UTC | #3
On 12.05.2023 05:01, Bryan O'Donoghue wrote:
> On 04/05/2023 09:00, Konrad Dybcio wrote:
>> Tested on 8250, but pretty please test it on your boards too!
> 
> What's the definition of test here ?
> 
> I ran this
> 
> ffplay -codec:video h264_v4l2m2m FantasticFour-ROTSS.mp4
> 
> and this
> 
> ffplay -codec:video vp8_v4l2m2m /mnt/big-buck-bunny_trailer.webm
> 
> on db410c with no errors. Then again I applied and disapplied the 8x8 264 fix to that branch and saw no discernable difference so I'm not very confident we have good coverage.
I don't think we have any coverage at all, especially considering
there were 1 or 2 patches fixing vdec not working at all in the past
few months.. Maybe CrOS has some internal pipelines for this.

> 
> @Stan @Vikash could you give some suggested tests for coverage here ?
> 
> @Konrad - get a db410c !
Don't think they're even made anymore!

> 
> My superficial first-pass on this series looks good but, before giving a Tested-by here, I think we should define a set of coverage tests, run them - the upper end on sm8250 and lower end msm8916 "makes sense to me"
> 
> 20? different gstreamer tests at different formats and different sizes on our selected platforms db410c, rb5, rb3 I have - also an 820 I haven't booted and an enforce sdm660.
> 
> Which tests will we use to validate this series and subsequent series to ensure we don't have more regressions ?
Personally I've done:

- boot and check if venus still probes and doesn't shout in dmesg
- decode and re-encode a H264 file

which is far from perfect..

Konrad
> 
> ---
> bod
  
Konrad Dybcio May 17, 2023, 8:25 p.m. UTC | #4
On 16.05.2023 14:57, Vikash Garodia wrote:
> 
> On 5/12/2023 8:31 AM, Bryan O'Donoghue wrote:
>> On 04/05/2023 09:00, Konrad Dybcio wrote:
>>> Tested on 8250, but pretty please test it on your boards too!
>>
>> What's the definition of test here ?
>>
>> I ran this
>>
>> ffplay -codec:video h264_v4l2m2m FantasticFour-ROTSS.mp4
>>
>> and this
>>
>> ffplay -codec:video vp8_v4l2m2m /mnt/big-buck-bunny_trailer.webm
>>
>> on db410c with no errors. Then again I applied and disapplied the 8x8 264 fix to
>> that branch and saw no discernable difference so I'm not very confident we have
>> good coverage.
>>
>> @Stan @Vikash could you give some suggested tests for coverage here ?
> 
> I could think of below test aspects for this series
> 1. Suspend Resume
> 2. Concurrency test
> 3. Module load -> video usecase -> module unload -> module load -> video
> usecase. This would ensure video firmware is reloaded and functional.
> 4. Video playback and encode for all supported resolution and codecs.
> 5. In general, video playback with more test content.
> 
> I would be testing the series with stability test suite on CrOS. That would be
> either on sc7180 or sc7280 setup.
> 
> Konrad, you can post the new version as one patch needs to be dropped. Test can
> be done on the new version. There are few patches in the series pending review,
> which can be done in parallel.
Sure, working on it!

Konrad
> 
> -Vikash
> 
>>
>> @Konrad - get a db410c !
>>
>> My superficial first-pass on this series looks good but, before giving a
>> Tested-by here, I think we should define a set of coverage tests, run them - the
>> upper end on sm8250 and lower end msm8916 "makes sense to me"
>>
>> 20? different gstreamer tests at different formats and different sizes on our
>> selected platforms db410c, rb5, rb3 I have - also an 820 I haven't booted and an
>> enforce sdm660.
>>
>> Which tests will we use to validate this series and subsequent series to ensure
>> we don't have more regressions ?
>>
>> ---
>> bod