From: David Howells <dhowells@redhat.com>
Date: Mon, 31 Jan 2022 17:54:43 +0000
Subject: cifs: Fix the readahead conversion to manage the batch when reading
 from cache
Git-commit: 46f5cbdef7d4fbb0f857a3caddec6799d0b5bb2f
Patch-mainline: v5.17-rc3
References: jsc#SES-1880

Fix the readahead conversion to correctly manage the last batch skipping
when reading from cache.  This involves a readahead batch of one page or
one folio, so set the batch size according to the number of constituent
pages (should be 1 for a filesystem that doesn't do multipage folios yet).

Signed-off-by: David Howells <dhowells@redhat.com>
cc: Steve French <smfrench@gmail.com>
Reviewed-by: Rohith Surabattula <rohiths.msft@gmail.com>
Reviewed-by: Shyam Prasad N <nspmangalore@gmail.com>
cc: Jeff Layton <jlayton@kernel.org>
cc: linux-cifs@vger.kernel.org
Signed-off-by: Steve French <stfrench@microsoft.com>
Acked-by: Luis Henriques <lhenriques@suse.de>
---
 fs/cifs/file.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index a50912674915..e7af802dcfa6 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -4446,7 +4446,7 @@ static void cifs_readahead(struct readahead_control *ractl)
 				 * by the cache.
 				 */
 				page = readahead_page(ractl);
-
+				last_batch_size = 1 << thp_order(page);
 				if (cifs_readpage_from_fscache(ractl->mapping->host,
 							       page) < 0) {
 					/*