From: Theodore Ts'o <tytso@mit.edu>
Date: Mon, 15 Jul 2013 16:27:47 +0000 (-0400)
Subject: ext4: yield during large unlinks
X-Git-Tag: archive/raspbian/4.9.13-1+rpi1~10^2~8367^2~1
X-Git-Url: https://dgit.raspbian.org/%22http:/www.example.com/cgi/%22https:/www.github.com/%22bookmarks:///%22http:/www.example.com/cgi/%22https:/www.github.com/%22bookmarks:/?a=commitdiff_plain;h=76828c882630ced08b5ddce22cc0095b05de9bc5;p=linux-4.9.git

ext4: yield during large unlinks

During large unlink operations on files with extents, we can use a lot
of CPU time.  This adds a cond_resched() call when starting to examine
the next level of a multi-level extent tree.  Multi-level extent trees
are rare in the first place, and this should rarely be executed.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---

diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 593091537e76..cfdc51e30257 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -2835,6 +2835,9 @@ again:
 				err = -EIO;
 				break;
 			}
+			/* Yield here to deal with large extent trees.
+			 * Should be a no-op if we did IO above. */
+			cond_resched();
 			if (WARN_ON(i + 1 > depth)) {
 				err = -EIO;
 				break;