On Tue, Sep 18, 2018 at 10:27:20AM -0500, Eric Sandeen wrote: > (this may be a bit uninformed on my part, Darrick reminds me that jbd2's > careful use of cache flushing & FUA probably means that it won't get burned > by a partial 4k metadata update if the power fails.) cache flushing and FUA is never going to help you with torn writes. > > I'll stop for now and see if Dave wants to chime in on xfs's reliance on > the actual atomic IO size for metadata IO. ;) What helps todays XFS (and probably ext4) is checksums on all metadata blocks, as that way we can check if we wrote the full "block". This doesn't help applications that rely on the sector size unless they have similar protections of their own.
