> Something I have brought up before: what does the host lock protect in the UFS > driver? Rather than reworking the code that acquires and releases the host lock, > all uses of the host lock should be eliminated from the UFS driver. The host lock > should be replaced with new locks of which it is clearly documented what > member variables these new locks protect. Then either this series is pre-mature and needs to be dropped for now, Or the changes you are planning can take place regardless? Thanks, Avri > > Thanks, > > Bart.
