On Mon, Aug 12, 2024 at 8:06 PM Christoph Hellwig <hch@xxxxxxxxxxxxx> wrote: > > > Apart from generic/622, all other shutdown-related cases can pass. > > > > generic/622 fails the test after the shutdown ioctl implementation, but > > when it's not implemented, this case will be skipped. > > > > This case designed to test the lazytime mount option, based on the test > > results, it appears that the atime and ctime of files cannot be > > synchronized to the disk through interfaces such as sync or fsync. > > It seems that it has little to do with the implementation of shutdown > > itself. > > > > If you need detailed information about generic/622, I can upload it. > > generic/622 tests that file systems implement at least lazytime > semantics. If exfat fails that it probably has timestamp handling > issue which are unrelated to the shutdown support, but which are only > exposed by this test that requires shutdown support. It would be great > if someone could look into that, but that's not a precondition for > your patch. > Thank you for your help. I will pay attention to the timestamp issue later.
