[PATCH 3.16.y-ckt 071/126] vfs: Avoid softlockups with sendfile(2)

Luis Henriques luis.henriques at canonical.com
Wed Jan 6 10:36:40 UTC 2016


3.16.7-ckt22 -stable review patch.  If anyone has any objections, please let me know.

------------------

From: Jan Kara <jack at suse.cz>

commit c2489e07c0a71a56fb2c84bc0ee66cddfca7d068 upstream.

The following test program from Dmitry can cause softlockups or RCU
stalls as it copies 1GB from tmpfs into eventfd and we don't have any
scheduling point at that path in sendfile(2) implementation:

        int r1 = eventfd(0, 0);
        int r2 = memfd_create("", 0);
        unsigned long n = 1<<30;
        fallocate(r2, 0, 0, n);
        sendfile(r1, r2, 0, n);

Add cond_resched() into __splice_from_pipe() to fix the problem.

CC: Dmitry Vyukov <dvyukov at google.com>
Signed-off-by: Jan Kara <jack at suse.cz>
Signed-off-by: Al Viro <viro at zeniv.linux.org.uk>
Signed-off-by: Luis Henriques <luis.henriques at canonical.com>
---
 fs/splice.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fs/splice.c b/fs/splice.c
index 31a66a3278cb..28f8d9dd92e3 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -888,6 +888,7 @@ ssize_t __splice_from_pipe(struct pipe_inode_info *pipe, struct splice_desc *sd,
 
 	splice_from_pipe_begin(sd);
 	do {
+		cond_resched();
 		ret = splice_from_pipe_next(pipe, sd);
 		if (ret > 0)
 			ret = splice_from_pipe_feed(pipe, sd, actor);




More information about the kernel-team mailing list