Audio: Volume: Clear peak meter channel max values in prepare() #9411
Conversation
singalsu
requested review from
lgirdwood,
plbossart,
mmaka1,
lbetlej,
dbaluta and
kv2019i
as code owners
|
@kv2019i pls review/merge |
src/audio/volume/volume_ipc4.c
Outdated
| @@ -147,6 +147,7 @@ int volume_init(struct processing_module *mod) | |||
| return -ENOMEM; | |||
| } | |||
|
|
|||
| memset(cd->peak_vol, 0, vol_size); | |||
There was a problem hiding this comment.
just rzalloc() above. BTW, is the .peak_vol allocation needed if CONFIG_COMP_PEAK_VOL is undefined?
There was a problem hiding this comment.
It's spread to nearly every volume source file, I'll add it to another commit.
There was a problem hiding this comment.
We have always for IPC4 systems CONFIG_COMP_PEAK_VOL=y, I don't think it makes sense to make a quite large patch in rush for 2.11. It would make sense if we would have another volume component for ChromeOS and Linux those do not use the peak values reporting from volume.
This patch avoids garbage values to be sent to mailbox by peak_vol_update() as very first peak volume measurement in stream start. The data structures cd->peak_vol and cd->peak_regs.peak_meter were not cleared. Signed-off-by: Seppo Ingalsuo <seppo.ingalsuo@linux.intel.com>
singalsu
force-pushed
the
peakvolume_clear_max_level_values
branch
from
9cb1ee4 to
09417f5
Compare
|
Ack @lgirdwood -- I couldn't get any PRs through on Friday (waiting on 9472) through Intel internal CI, so that's a potential problem for today as well. So in practise it depends on now 9472. We can't tag rc1 before that is in, so if we have other PRs ready before that, we can include. |
This patch avoids garbage values to be sent to mailbox by peak_vol_update() as very first peak volume measurement in stream start. The data structures cd->peak_vol and cd->peak_regs.peak_meter were not cleared.