-----Original Message----- From: Bart Van Assche <bvanassche@xxxxxxx> Sent: Saturday, May 7, 2022 11:14 PM To: Bob Pearson <rpearsonhpe@xxxxxxxxx>; Yanjun Zhu <yanjun.zhu@xxxxxxxxx>; Jason Gunthorpe <jgg@xxxxxxxxxx> Cc: Zhu Yanjun <zyjzyj2000@xxxxxxxxx>; linux-rdma@xxxxxxxxxxxxxxx; Bernard Metzler <bmt@xxxxxxxxxxxxxx> Subject: Re: Apparent regression in blktests since 5.18-rc1+ On 5/7/22 06:43, Bob Pearson wrote: > The hang I was showing was for siw not rxe. rxe also shows similar hangs on my system. > The siw run was vanilla rdma-linux without any patches. Hi Bob, If I run the SRP tests against the rdma for-next branch then I can reproduce the hang mentioned in a previous email. If I merge Linus' master branch into the rdma for-next branch then the SRP tests pass with the Soft-iWARP driver and also with the soft-RoCE driver. I think this shows that the root cause of the hang is in the rdma for-next branch and not in the SRP tests. Maybe a patch from the master branch is missing from the rdma for-next branch? Thanks, Bart. Thanks for this. (I'm on a road trip for the next couple of days.) This should help get this solved. Bob
