On Mon, May 11, 2020 at 08:35:32AM -0700, Christoph Hellwig wrote: > On Sat, May 09, 2020 at 02:47:02PM -0500, Eric Sandeen wrote: > > So there's no real way to know that the grace period adjustment failed > > on an older kernel. We could consider that a bug and fix it, or > > consider it a change in behavior that we can't just make without > > at least some form of versioning. Thoughts? Well you /could/ roll this new functionality into a new revision of the ioctl that supports 64bit time values... > I'd consider a bug, as applications are very unlikely to rely on > these ignored calls (famous last words..). ...but OTOH that might be a better discussion to have when we start reviewing the bigtime series. In the meantime, seeing as xfs_quota never really let you push out soft grace period expiration anyway, I guess you could just teach userspace to try to make the change and declare success if an immediate re-query shows the value changed. --D
