On Fri, Jun 14, 2024 at 10:04:28AM -0600, Jens Axboe wrote: > > That sounds like you didn't apply the above mentioned > > "convert the SCSI ULDs to the atomic queue limits API v2" series before? > > That might indeed explain it... Surprising it applied without. Also as mentioned a couple weeks ago and again earlier this week, can we please have a shared branch for the SCSI tree to pull in for the limits conversions (including the flags series I need to resend next week) so that Martin can pull it in?