ACK: [PATCH 0/2][SRU][Noble] Fix power consumption while using HW accelerated video decode on AMD platforms

Aaron Jauregui aaron.jauregui at canonical.com
Thu Aug 8 00:19:40 UTC 2024


On Tue, Aug 06, 2024 at 11:23:30AM +0800, You-Sheng Yang wrote:
> BugLink: https://bugs.launchpad.net/bugs/2073282
> 
> [Impact]
> 
> Power consumption measurements are worse than Windows while using HW
> accelerated video decoding.
> 
> [Fix]
> 
> Two commits from upstream v6.11-rc1 to add a unified queue to VCN for
> more power saving.
> 
> [Test Case]
> 
> This involves attaching probes on mainboard and can only be performed by
> AMD/ODM.
> 
> [Where problems could occur]
> 
> No.
> 
> [Other Info]
> 
> To apply to all kernels >= 6.8. Orauclar/Noble, OEM-6.8, OEM-6.11 are
> nominated for fix.
> 
> Boyuan Zhang (2):
>   drm/amdgpu/vcn: identify unified queue in sw init
>   drm/amdgpu/vcn: not pause dpg for unified queue
> 
>  drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c | 53 ++++++++++++-------------
>  drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h |  1 +
>  2 files changed, 27 insertions(+), 27 deletions(-)

Acked-by: Aaron Jauregui <aaron.jauregui at canonical.com>



More information about the kernel-team mailing list