On Tue, 2018-08-07 at 20:03 +-0530, Sreekanth Reddy wrote: +AD4- The main intention of this patch to reset the smid to zero after +AD4- resetting the corresponding smid entry's chain+AF8-offset to zero. While +AD4- posting +ACI-mpt3sas: Fix calltrace observed while running IO +ACY- reset+ACI- +AD4- patch I have done manual mistake that I reset the smid to zero before +AD4- resetting the chain+AF8-offset which is wrong. Due to which driver +AD4- always's reset the chain+AF8-offset of zero th entry of chain+AF8-lookup+AFsAXQ- +AD4- table which is wrong and hence I am correcting it with this patch. +AD4- +AD4- After that you asked me to add memory barriers and hence I have added +AD4- memory barriers in subsequent versions of the patch. Hello Sreekanth, Please answer the questions I already asked you twice instead of sidestepping these. Thanks, Bart.
