afs: Don't use folio->private to record partial modification
authorDavid Howells <dhowells@redhat.com>
Fri, 2 Dec 2022 14:12:41 +0000 (14:12 +0000)
committerDavid Howells <dhowells@redhat.com>
Sun, 24 Dec 2023 15:08:51 +0000 (15:08 +0000)
commita34847d4b73c3a98b565b1d1cc6e1b70c661e18b
tree7d8a291a082ef312e523092cadec63abeb65b082
parent5f5ce7ba15e7e6a6539ac8e1f845757aaebecf0d
afs: Don't use folio->private to record partial modification

AFS currently uses folio->private to store the range of bytes within a
folio that have been modified - the idea being that if we have, say, a 2MiB
folio and someone writes a single byte, we only have to write back that
single page and not the whole 2MiB folio - thereby saving on network
bandwidth.

Remove this, at least for now, and accept the extra network load (which
doesn't matter in the common case of writing a whole file at a time from
beginning to end).

This makes folio->private available for netfslib to use.

Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
cc: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
cc: linux-cachefs@redhat.com
cc: linux-fsdevel@vger.kernel.org
cc: linux-mm@kvack.org
fs/afs/file.c
fs/afs/internal.h
fs/afs/write.c
include/trace/events/afs.h