]> git.baikalelectronics.ru Git - kernel.git/commit
ALSA: hda: Allow HDA to be runtime suspended when dGPU is not bound to a driver
authorKai-Heng Feng <kai.heng.feng@canonical.com>
Fri, 18 Oct 2019 07:38:48 +0000 (15:38 +0800)
committerTakashi Iwai <tiwai@suse.de>
Mon, 21 Oct 2019 13:14:07 +0000 (15:14 +0200)
commit1e2890c5007e0ddd8f3a5fa772667ca3b91d9dc7
treed6b51d9e20607810a49ee9e61fc7bb46830ffb79
parent17eb09130fd6bd6f80b42afb274b6e7e3ca2efa3
ALSA: hda: Allow HDA to be runtime suspended when dGPU is not bound to a driver

Nvidia proprietary driver doesn't support runtime power management, so
when a user only wants to use the integrated GPU, it's a common practice
to let dGPU not to bind any driver, and let its upstream port to be
runtime suspended. At the end of runtime suspension the port uses
platform power management to disable power through _OFF method of power
resource, which is listed by _PR3.

After commit 60dd6d0417fc ("PCI: Enable NVIDIA HDA controllers"), when
the dGPU comes with an HDA function, the HDA won't be suspended if the
dGPU is unbound, so the power resource can't be turned off by its
upstream port driver.

Commit 18812b1fc2f6 ("ALSA: hda - Enable runtime PM only for
discrete GPU") only allows HDA to be runtime suspended once GPU is
bound, to keep APU's HDA working.

However, HDA on dGPU isn't that useful if dGPU is not bound to any
driver.  So let's relax the runtime suspend requirement for dGPU's HDA
function, to disable the power source to save lots of power.

BugLink: https://bugs.launchpad.net/bugs/1840835
Fixes: 60dd6d0417fc ("PCI: Enable NVIDIA HDA controllers")
Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
Link: https://lore.kernel.org/r/20191018073848.14590-2-kai.heng.feng@canonical.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
sound/pci/hda/hda_intel.c