On Thu, 2018-12-06 at 10:00 -0700, Jens Axboe wrote: +AD4 On 12/6/18 9:47 AM, Bart Van Assche wrote: +AD4 +AD4 If I merge Jens' for-next branch with Linus' master branch, boot the +AD4 +AD4 resulting kernel in a VM and run blktests/tests/srp/002 then that test +AD4 +AD4 never finishes. The same test passes against Linus' master branch. I +AD4 +AD4 think this is a regression. The following appears in the system log if +AD4 +AD4 I run that test: +AD4 +AD4 You are running that test on a dm device? Can you shed some light on +AD4 how that dm device is setup? Hi Jens, The dm device referred to in my e-mail is a dm-mpath device created by test srp/002. All parameters of that device are under control of that test script. >From the srp/002 test script: DESCRIPTION+AD0AIg-File I/O on top of multipath concurrently with logout and login (mq)+ACI >From srp/multipath.conf: defaults +AHs find+AF8-multipaths no user+AF8-friendly+AF8-names yes queue+AF8-without+AF8-daemon no +AH0 devices +AHs device +AHs vendor +ACI-LIO-ORG+AHw-SCST+AF8-BIO+AHw-FUSIONIO+ACI product +ACI.+ACoAIg features +ACI-1 queue+AF8-if+AF8-no+AF8-path+ACI path+AF8-checker tur +AH0 +AH0 Bart.
