Repository navigation
[TEST] ASoC: rt700: disable jack detection when the codec suspends - #2949
shumingfan wants to merge 1 commit into
Conversation
This patch tries to avoid the unexpected interrupt to do the jack detection after the codec suspends. Signed-off-by: Shuming Fan <shumingf@realtek.com>
| dev_dbg(dev, "in %s jd enable\n", __func__); | ||
|
|
||
| regmap_write(rt700->regmap, | ||
| RT700_SET_AUDIO_POWER_STATE, AC_PWRST_D3); |
There was a problem hiding this comment.
I guess this is correct somehow and has been tested, but just seems a bit odd... Is it really switching to D3 during resume?
There was a problem hiding this comment.
It is just the power-saving mode for the codec. The codec driver should go to power-saving mode before going to the suspend state. So, I just set back to power-saving mode during the resume function.
There was a problem hiding this comment.
ok, but we're resuming for playback, the codec goes to D3 here and then presumably it's switched back to D0 immediately, right? Is that common or has some specific reason? Naively I'd expect .resume() to bring the system to a fully operational mode to be able to begin operation immediately. Although bringing the hardware to some intermediate state, waiting for the actual operation to begin, should work too...
|
@shumingfan @bardliao I am not sure if we want this 'eature'. The Intel parts are capable of detecting a SoundWire wake while in pm_runtime suspend. I am not sure why we would disable this capability at the codec level. It's fine for a test, but it doesn't look like a good move. |
@plbossart This patch just is used to debug and test for issue #2943. After the testing, this patch can't fix the issue and the problem should be why the codec driver goes to pm_runtime suspend from the resume so quick. |
3 participants
This patch tries to avoid the unexpected interrupt to do the jack detection after the codec suspends.
TEST only for #2943
Signed-off-by: Shuming Fan shumingf@realtek.com