Hi,

I am trying to understand the suspend/resume mechanism in mmc.

The kernel ( in hibernation as well as in suspend to ram) first calls suspend_thaw_processes() and then calls pm_notifier_call_chain(PM_POST_SUSPEND/HIBERNATION). The mmc driver schedules mmc_rescan on receiving the notification of PM_POST_SUSPEND/HIBERNATION. Thus before mmc_rescan() gets called, the mmc card is not ready to be used.

So if any processes before hibernation/suspend were using mmc device, then on resume when these processes would be unfrozen( from thaw_processes ) and if it tries to access the mmc device( before mmc_rescan gets scheduled), then it would hit a problem.

Please let me know if my understanding is correct or not.

Regards, Ayan Kumar Halder