Dave confirmed Jeff's suspicion in [1]:
> That wants to be rw_op = do_aio_write in the else clause. How did you
> test this patch set?
Yep, that's certainly a bug. I concentrated on testing on loop mounts,
running some fio and ltp testcases, but I'm at a loss to explain how
things seem to work as well as they do. Let's see how where fixing this
leads.
This is a fixup to commit daf0f6142dd5 ("fs: use read_iter and write_iter
rather than aio_read and aio_write") from Dave Kleiklamp's GIT tree in [2].
[1] http://www.spinics.net/lists/linux-fsdevel/msg61587.html
[2] https://github.com/kleikamp/linux-shaggy/commits/loop_2013_01_04
Cc: Dave Kleikamp <dave.kleikamp@xxxxxxxxxx>
Cc: Jeff Moyer <jmoyer@xxxxxxxxxx>
Cc: Alexander Viro <viro@xxxxxxxxxxxxxxxxxx>
Cc: linux-fsdevel@xxxxxxxxxxxxxxx
Signed-off-by: Sedat Dilek <sedat.dilek@xxxxxxxxx>
---
fs/aio.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/fs/aio.c b/fs/aio.c
index 08f550a..9635e7a 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -1320,7 +1320,7 @@ static ssize_t aio_rw_vect_retry(struct kiocb *iocb)
rw_op = do_aio_read;
opcode = IOCB_CMD_PREADV;
} else {
- rw_op = do_aio_read;
+ rw_op = do_aio_write;
opcode = IOCB_CMD_PWRITEV;
}
--
1.8.1.1
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
