On 2/2/21 2:03 PM, Vladimir Sementsov-Ogievskiy wrote: > 02.02.2021 21:59, Eric Blake wrote: >> Commit f93e19fb03b adjusted various iotest whitespace discrepancies, >> but missed a test that is not part of default 'make check'. > > > Actually, at point of f93e19fb03b there is no x-perf at all, so it's not > a miss in f93e19fb03b... > > Let me bisect it... > > Hmm, seems, it's wrong since introducing this line in 61623f82153788e > > Why was it wrongly added - it's a question.. Looks like something was > merged in a meantime, that changed how qmp log looks like.. And original > patch was made before that. But I don't want to dig more :)
Okay, I'll adjust my commit message; we definitely had a case of patches crossing one another, reintroducing the problem. > > for the patch itself: > > Reviewed-by: Vladimir Sementsov-Ogievskiy <[email protected]> Thanks; I'll include it in my NBD pull request, since I found it while using iotests for that. -- Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3226 Virtualization: qemu.org | libvirt.org
