> On Jul 27, 2021, at 5:20 PM, Marciniszyn, Mike <mike.marciniszyn@xxxxxxxxxxxxxxxxxxxx> wrote: > >>> Haakon's original analysis said that this was an INIT->INIT >>> transition, so I'm a bit confused why we lost a RESET->INIT transition >>> in the end? >> >> Perhaps the patch should have removed the ib_modify_qp() from >> cma_modify_qp_rtr() instead. > > I think that will work. Implemented and tested. It doesn't work. :-) The conclusion I draw is that there are still spots that depend on one or the other of those state transitions remaining where they are. -- Chuck Lever
