On Tue, Sep 26, 2017 at 02:40:36PM +0000, Bart Van Assche wrote: > On Tue, 2017-09-26 at 16:13 +0800, Ming Lei wrote: > > I am pretty sure that suspend/resume can survive when resync in progress > > with my patchset applied on RAID1, without any MD change. > > The above shows that you do not understand how suspend and resume works. > In the documents in the directory Documentation/power it is explained clearly > that I/O must be stopped before the hibernation image is generation to avoid No, I don't use hibernation, and just use suspend/resume(s2r). > hard to repair filesystem corruption. Although md is not a filesystem I think > this also applies to md since md keeps some state information in RAM and some > state information on disk. It is essential that all that state information is > in consistent. > > > If your patchset depends on this MD change, something should be wrong > > in the following patches. Now I need to take a close look. > > The later patches that make SCSI quiesce and resume safe do not depend on > this change. Are you sure? If I remove the 1st patch, system suspend/resume will hang with all your other 6 patchset. -- Ming
