ASoC: SOF: topology: do not power down primary core during topology r… #3005
Conversation
ranj063
force-pushed
the
fix/core_power_down
branch
from
f707445 to
1ce2f15
Compare
hmm that would be this one: Should I add it? |
|
It's more the commit below, before this we didn't have a notion of PRIMARY_CORE. |
…emoval When removing the topology components, do not power down the primary core. Doing so will result in an IPC timeout when the SOF PCI device runtime suspends. Fixes: 0dcdf84 ("ASoC: SOF: add a "core" parameter to widget loading functions") Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
done. Thanks! |
ranj063
force-pushed
the
fix/core_power_down
branch
from
1ce2f15 to
9f2e75d
Compare
| if (pipeline->core == SOF_DSP_PRIMARY_CORE) | ||
| break; | ||
|
|
||
| ret = snd_sof_dsp_core_power_down(sdev, 1 << pipeline->core); |
There was a problem hiding this comment.
how about the case that the same secondary core is used by another active pipeline? we should add ref_count to DSP cores, then we don't need to treat this Primary core explicitly.
There was a problem hiding this comment.
Yes, but until then we need this change
| * CTX_SAVE IPC can succeed during runtime suspend. | ||
| */ | ||
| if (pipeline->core == SOF_DSP_PRIMARY_CORE) | ||
| break; |
There was a problem hiding this comment.
This wil help in debugging as well.
4 participants
…emoval
When removing the topology components, do not power down
the primary core. Doing so will result in an IPC timeout
when the SOF PCI device runtime suspends.
Signed-off-by: Ranjani Sridharan ranjani.sridharan@linux.intel.com