Bard Liao
Bard Liao
> Pretty much every race condition I ever looked at is affected by compiler changes. Yes, it could be a race condition.
@plbossart Currently, we set regcache_cache_only false on resume. Is it possible that the codec is unattached and attached, but no resume occurs? For example, the codec is unattached after resume....
@plbossart Adding the default value can ensure that driver will not try to access the register and can get the cached value properly when the regmap is in cache only...
> @bardliao would anyone from Realtek be able to suggest the correct verbs here ? Kailang and Pshou may help.
> > > @bardliao would anyone from Realtek be able to suggest the correct verbs here ? > > > > > > Kailang [[email protected]](mailto:[email protected]) and Pshou [[email protected]](mailto:[email protected]) may help....
@plbossart @libinyang I checked the dmesg and it seems the main issue is that codecs are probed AFTER machine driver is created.
@plbossart It looks like the issue has been fixed. The log of 12111 and 12176 are the same as this issue. But the log of 12202 and 12215 is ```...
@keqiaozhang Have you seen this issue on other devices? It is odd to me if it is SoundWire related.
@plbossart Should we add a new file in the drivers/clk/x86 folder? Does it make sense to reuse skl-ssp-clk.c?
@plbossart Do we need to change the terms of device property, too? For example, "mipi-sdw-master-count", "mipi-sdw-slave-channelprepare-timeout"?