[PATCH 4.4 10/38] fs: direct-io: fix missing sdio->boundary

From: Greg Kroah-Hartman
Date: Thu Apr 15 2021 - 10:48:49 EST


From: Jack Qiu <jack.qiu@xxxxxxxxxx>

commit df41872b68601059dd4a84858952dcae58acd331 upstream.

I encountered a hung task issue, but not a performance one. I run DIO
on a device (need lba continuous, for example open channel ssd), maybe
hungtask in below case:

DIO: Checkpoint:
get addr A(at boundary), merge into BIO,
no submit because boundary missing
flush dirty data(get addr A+1), wait IO(A+1)
writeback timeout, because DIO(A) didn't submit
get addr A+2 fail, because checkpoint is doing

dio_send_cur_page() may clear sdio->boundary, so prevent it from missing
a boundary.

Link: https://lkml.kernel.org/r/20210322042253.38312-1-jack.qiu@xxxxxxxxxx
Fixes: b1058b981272 ("direct-io: submit bio after boundary buffer is added to it")
Signed-off-by: Jack Qiu <jack.qiu@xxxxxxxxxx>
Reviewed-by: Jan Kara <jack@xxxxxxx>
Cc: <stable@xxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
fs/direct-io.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)

--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -780,6 +780,7 @@ submit_page_section(struct dio *dio, str
struct buffer_head *map_bh)
{
int ret = 0;
+ int boundary = sdio->boundary; /* dio_send_cur_page may clear it */

if (dio->rw & WRITE) {
/*
@@ -818,10 +819,10 @@ submit_page_section(struct dio *dio, str
sdio->cur_page_fs_offset = sdio->block_in_file << sdio->blkbits;
out:
/*
- * If sdio->boundary then we want to schedule the IO now to
+ * If boundary then we want to schedule the IO now to
* avoid metadata seeks.
*/
- if (sdio->boundary) {
+ if (boundary) {
ret = dio_send_cur_page(dio, sdio, map_bh);
if (sdio->bio)
dio_bio_submit(dio, sdio);