Hi Linus,
There is only the amdgpu runtime pm regression fix in here. I'm going to be away next week, I'll probably pre-send you the drm-next main PR before I go, if I can get it done today!
Dave.
drm-fixes-2022-01-07: drm fixes for 5.16 final
amdgpu: - suspend/resume fix - fix runtime PM regression The following changes since commit c9e6606c7fe92b50a02ce51dda82586ebdf99b48:
Linux 5.16-rc8 (2022-01-02 14:23:25 -0800)
are available in the Git repository at:
git://anongit.freedesktop.org/drm/drm tags/drm-fixes-2022-01-07
for you to fetch changes up to 936a93775b7c4f2293f651f64c4139c82e19a164:
Merge tag 'amd-drm-fixes-5.16-2021-12-31' of ssh://gitlab.freedesktop.org/agd5f/linux into drm-fixes (2022-01-07 06:46:08 +1000)
---------------------------------------------------------------- drm fixes for 5.16 final
amdgpu: - suspend/resume fix - fix runtime PM regression
---------------------------------------------------------------- Alex Deucher (2): fbdev: fbmem: add a helper to determine if an aperture is used by a fw fb drm/amdgpu: disable runpm if we are the primary adapter
Dave Airlie (1): Merge tag 'amd-drm-fixes-5.16-2021-12-31' of ssh://gitlab.freedesktop.org/agd5f/linux into drm-fixes
Evan Quan (1): drm/amd/pm: keep the BACO feature enabled for suspend
drivers/gpu/drm/amd/amdgpu/amdgpu.h | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 28 ++++++++++++++++++ drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c | 6 ++++ drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c | 8 +++++- drivers/video/fbdev/core/fbmem.c | 47 +++++++++++++++++++++++++++++++ include/linux/fb.h | 1 + 6 files changed, 90 insertions(+), 1 deletion(-)
The pull request you sent on Fri, 7 Jan 2022 13:23:45 +1000:
git://anongit.freedesktop.org/drm/drm tags/drm-fixes-2022-01-07
has been merged into torvalds/linux.git: https://git.kernel.org/torvalds/c/7a6043cc2e863ab45016622c30879e555523ee13
Thank you!
On Thu, Jan 6, 2022 at 7:23 PM Dave Airlie airlied@gmail.com wrote:
There is only the amdgpu runtime pm regression fix in here.
Thanks, from a quick test it works for me - the backlight actually does eventually go away.
It does so only on the second time the monitors say "no signal, going to power save", but that has been true before too.
So I think there's still some confusion in this area, but it might be elsewhere - who knows what Wayland and friends do. At least it doesn't look like a regression to me any more.
Thanks, Linus
On Fri, Jan 7, 2022 at 6:42 PM Linus Torvalds torvalds@linux-foundation.org wrote:
On Thu, Jan 6, 2022 at 7:23 PM Dave Airlie airlied@gmail.com wrote:
There is only the amdgpu runtime pm regression fix in here.
Thanks, from a quick test it works for me - the backlight actually does eventually go away.
It does so only on the second time the monitors say "no signal, going to power save", but that has been true before too.
So I think there's still some confusion in this area, but it might be elsewhere - who knows what Wayland and friends do. At least it doesn't look like a regression to me any more.
Well it's not a true fix, just a "go back to exact old behaviour, but limited to relevant gpus for amdgpu only" so that i915 doesn't regress. I think there's some more debug to do here and Alex/Harry&team can look at leisure now :-)
Cheers, Daniel
dri-devel@lists.freedesktop.org