On Fri, May 19, 2023 at 06:11:36AM +0200, Christoph Hellwig wrote: > On Thu, May 18, 2023 at 10:00:12PM -0400, Theodore Ts'o wrote: > > On Thu, May 18, 2023 at 06:23:09AM +0200, Christoph Hellwig wrote: > > > Hi all, > > > > > > this series fixes the long standing problem that we never had a good way > > > to communicate block device events to the user of the block device. > > > > > > It fixes this by introducing a new set of holder ops registered at > > > blkdev_get_by_* time for the exclusive holder, and then wire that up > > > to a shutdown super operation to report the block device remove to the > > > file systems. > > > > Thanks for working on this! Is there going to be an fstest which > > simulates a device removal while we're running fsstress or some such, > > so we can exercise full device removal path? > > > So the problem with xfstests is that there isn't really any generic > way to remove a block device, and even less so to put it back. > > xfstests has some scsi_debug based tests, maybe I can cook something up > for that. My testing has been with nvme, so another option would be > to add nvme-loop support to xfstests and use that. I'll see what I can > do. Could you make dm-error accept a 'message' telling it to invoke all these bdev removal actions? There's already a bunch of helpers in fstests to make that less awful for test authors. --D
