On Fri, Jan 23, 2026 at 06:12:16AM +0100, Christoph Hellwig wrote: > On Thu, Jan 22, 2026 at 01:27:00PM -0800, Darrick J. Wong wrote: > > Nice hoist, though I wonder -- as an exported fs function, should we be > > checking that the returned folio doesn't cover EOF? Not that any of the > > users actually check that returned merkle tree folios fit that > > criterion. > > As in past i_size because this is verity metadata? I think per the > last discussion that's only guranteed to be true, not the folio. It > might be useful to assert this, but it might be better for combine > this with the work to use different on-disk vs in-memory offset > and to consolidate all the offset magic. Which is worthwhile, > but І don't really want to add that in this series.
<nod> You're probably right that adding such a check would be better off in whatever series formally defines the post-eof pagecache offset and whatnot. --D _______________________________________________ Linux-f2fs-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
