FGP_NOFS is legacy; filesystems should be using memalloc_nofs_save/restore
instead.  We have it here in iomap because it was buried in
grab_cache_page_write_begin() and we didn't want to change this behaviour
as part of the folio transition.

I have tested this with XFS and see no issues.  Other filesystems (cc'd)
may need to make adjustments.  Please test with lockdep enabled.

Cc: "Darrick J. Wong" <djwong@kernel.org> (iomap)
Cc: Jens Axboe <axboe@kernel.dk> (block)
Cc: Namjae Jeon <linkinjeon@kernel.org> (exfat, ntfs)
Cc: Sungjong Seo <sj1557.seo@samsung.com> (exfat)
Cc: Yuezhang Mo <yuezhang.mo@sony.com> (exfat)
Cc: Miklos Szeredi <miklos@szeredi.hu> (fuse)
Cc: Andreas Gruenbacher <agruenba@redhat.com> (gfs2)
Cc: Hyunchul Lee <hyc.lee@gmail.com> (ntfs)
Cc: Konstantin Komarov <almaz.alexandrovich@paragon-software.com> (ntfs3)
Cc: Carlos Maiolino <cem@kernel.org> (xfs)
Cc: Damien Le Moal <dlemoal@kernel.org> (zonefs)
Cc: Naohiro Aota <naohiro.aota@wdc.com> (zonefs)
Cc: Johannes Thumshirn <jth@kernel.org> (zonefs)
Cc: linux-xfs@vger.kernel.org
Cc: linux-fsdevel@vger.kernel.org
Cc: linux-block@vger.kernel.org
Cc: fuse-devel@lists.linux.dev
Cc: gfs2@lists.linux.dev
Cc: ntfs3@lists.linux.dev
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 fs/iomap/buffered-io.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c
index 8d4806dc46d4..27bc2455a98d 100644
--- a/fs/iomap/buffered-io.c
+++ b/fs/iomap/buffered-io.c
@@ -768,7 +768,7 @@ EXPORT_SYMBOL_GPL(iomap_is_partially_uptodate);
  */
 struct folio *iomap_get_folio(struct iomap_iter *iter, loff_t pos, size_t len)
 {
-	fgf_t fgp = FGP_WRITEBEGIN | FGP_NOFS;
+	fgf_t fgp = FGP_WRITEBEGIN;
 
 	if (iter->flags & IOMAP_NOWAIT)
 		fgp |= FGP_NOWAIT;
-- 
2.47.3