The patch titled
Subject: ocfs2: fix ip_unaligned_aio deadlock with dio work queue
has been added to the -mm tree. Its filename is
ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Ryan Ding <ryan.ding@xxxxxxxxxx>
Subject: ocfs2: fix ip_unaligned_aio deadlock with dio work queue
In the current implementation of unaligned aio+dio, lock order behave as
follow:
in user process context:
-> call io_submit()
-> get i_mutex
<== window1
-> get ip_unaligned_aio
-> submit direct io to block device
-> release i_mutex
-> io_submit() return
in dio work queue context(the work queue is created in __blockdev_direct_IO):
-> release ip_unaligned_aio
<== window2
-> get i_mutex
-> clear unwritten flag & change i_size
-> release i_mutex
There is a limitation to the thread number of dio work queue. 256 at
default. If all 256 thread are in the above 'window2' stage, and there is
a user process in the 'window1' stage, the system will became deadlock.
Since the user process hold i_mutex to wait ip_unaligned_aio lock, while
there is a direct bio hold ip_unaligned_aio mutex who is waiting for a dio
work queue thread to be schedule. But all the dio work queue thread is
waiting for i_mutex lock in 'window2'.
This case only happened in a test which send a large number(more than 256)
of aio at one io_submit() call.
My design is to remove ip_unaligned_aio lock. Change it to a sync io
instead. Just like ip_unaligned_aio lock, serialize the unaligned aio
dio.
Signed-off-by: Ryan Ding <ryan.ding@xxxxxxxxxx>
Cc: Junxiao Bi <junxiao.bi@xxxxxxxxxx>
Cc: Joseph Qi <joseph.qi@xxxxxxxxxx>
Cc: Mark Fasheh <mfasheh@xxxxxxx>
Cc: Joel Becker <jlbec@xxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
fs/ocfs2/aops.c | 6 ------
fs/ocfs2/aops.h | 7 -------
fs/ocfs2/file.c | 27 +++++++++------------------
fs/ocfs2/inode.h | 3 ---
fs/ocfs2/super.c | 1 -
5 files changed, 9 insertions(+), 35 deletions(-)
diff -puN fs/ocfs2/aops.c~ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue fs/ocfs2/aops.c
--- a/fs/ocfs2/aops.c~ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue
+++ a/fs/ocfs2/aops.c
@@ -2388,12 +2388,6 @@ static void ocfs2_dio_end_io(struct kioc
/* this io's submitter should not have unlocked this before we could */
BUG_ON(!ocfs2_iocb_is_rw_locked(iocb));
- if (ocfs2_iocb_is_unaligned_aio(iocb)) {
- ocfs2_iocb_clear_unaligned_aio(iocb);
-
- mutex_unlock(&OCFS2_I(inode)->ip_unaligned_aio);
- }
-
if (private)
ocfs2_dio_end_io_write(inode, private, offset, bytes);
diff -puN fs/ocfs2/aops.h~ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue fs/ocfs2/aops.h
--- a/fs/ocfs2/aops.h~ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue
+++ a/fs/ocfs2/aops.h
@@ -93,11 +93,4 @@ enum ocfs2_iocb_lock_bits {
#define ocfs2_iocb_rw_locked_level(iocb) \
test_bit(OCFS2_IOCB_RW_LOCK_LEVEL, (unsigned long *)&iocb->private)
-#define ocfs2_iocb_set_unaligned_aio(iocb) \
- set_bit(OCFS2_IOCB_UNALIGNED_IO, (unsigned long *)&iocb->private)
-#define ocfs2_iocb_clear_unaligned_aio(iocb) \
- clear_bit(OCFS2_IOCB_UNALIGNED_IO, (unsigned long *)&iocb->private)
-#define ocfs2_iocb_is_unaligned_aio(iocb) \
- test_bit(OCFS2_IOCB_UNALIGNED_IO, (unsigned long *)&iocb->private)
-
#endif /* OCFS2_FILE_H */
diff -puN fs/ocfs2/file.c~ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue fs/ocfs2/file.c
--- a/fs/ocfs2/file.c~ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue
+++ a/fs/ocfs2/file.c
@@ -2170,7 +2170,7 @@ static ssize_t ocfs2_file_write_iter(str
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
int full_coherency = !(osb->s_mount_opt &
OCFS2_MOUNT_COHERENCY_BUFFERED);
- int unaligned_dio = 0;
+ void *saved_ki_complete = NULL;
int append_write = ((iocb->ki_pos + count) >=
i_size_read(inode) ? 1 : 0);
@@ -2233,17 +2233,12 @@ static ssize_t ocfs2_file_write_iter(str
goto out;
}
- if (direct_io && !is_sync_kiocb(iocb))
- unaligned_dio = ocfs2_is_io_unaligned(inode, count, iocb->ki_pos);
-
- if (unaligned_dio) {
+ if (direct_io && !is_sync_kiocb(iocb) &&
+ ocfs2_is_io_unaligned(inode, count, iocb->ki_pos)) {
/*
- * Wait on previous unaligned aio to complete before
- * proceeding.
+ * Make it a sync io if it's an unaligned aio.
*/
- mutex_lock(&OCFS2_I(inode)->ip_unaligned_aio);
- /* Mark the iocb as needing an unlock in ocfs2_dio_end_io */
- ocfs2_iocb_set_unaligned_aio(iocb);
+ saved_ki_complete = xchg(&iocb->ki_complete, NULL);
}
/* communicate with ocfs2_dio_end_io */
@@ -2264,11 +2259,10 @@ static ssize_t ocfs2_file_write_iter(str
*/
if ((written == -EIOCBQUEUED) || (!ocfs2_iocb_is_rw_locked(iocb))) {
rw_level = -1;
- unaligned_dio = 0;
}
if (unlikely(written <= 0))
- goto no_sync;
+ goto out;
if (((file->f_flags & O_DSYNC) && !direct_io) ||
IS_SYNC(inode)) {
@@ -2290,13 +2284,10 @@ static ssize_t ocfs2_file_write_iter(str
iocb->ki_pos - 1);
}
-no_sync:
- if (unaligned_dio && ocfs2_iocb_is_unaligned_aio(iocb)) {
- ocfs2_iocb_clear_unaligned_aio(iocb);
- mutex_unlock(&OCFS2_I(inode)->ip_unaligned_aio);
- }
-
out:
+ if (saved_ki_complete)
+ xchg(&iocb->ki_complete, saved_ki_complete);
+
if (rw_level != -1)
ocfs2_rw_unlock(inode, rw_level);
diff -puN fs/ocfs2/inode.h~ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue fs/ocfs2/inode.h
--- a/fs/ocfs2/inode.h~ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue
+++ a/fs/ocfs2/inode.h
@@ -43,9 +43,6 @@ struct ocfs2_inode_info
/* protects extended attribute changes on this inode */
struct rw_semaphore ip_xattr_sem;
- /* Number of outstanding AIO's which are not page aligned */
- struct mutex ip_unaligned_aio;
-
/* These fields are protected by ip_lock */
spinlock_t ip_lock;
u32 ip_open_count;
diff -puN fs/ocfs2/super.c~ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue fs/ocfs2/super.c
--- a/fs/ocfs2/super.c~ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue
+++ a/fs/ocfs2/super.c
@@ -1742,7 +1742,6 @@ static void ocfs2_inode_init_once(void *
INIT_LIST_HEAD(&oi->ip_io_markers);
INIT_LIST_HEAD(&oi->ip_unwritten_list);
oi->ip_dir_start_lookup = 0;
- mutex_init(&oi->ip_unaligned_aio);
init_rwsem(&oi->ip_alloc_sem);
init_rwsem(&oi->ip_xattr_sem);
mutex_init(&oi->ip_io_mutex);
_
Patches currently in -mm which might be from ryan.ding@xxxxxxxxxx are
ocfs2-add-ocfs2_write_type_t-type-to-identify-the-caller-of-write.patch
ocfs2-use-c_new-to-indicate-newly-allocated-extents.patch
ocfs2-test-target-page-before-change-it.patch
ocfs2-do-not-change-i_size-in-write_end-for-direct-io.patch
ocfs2-return-the-physical-address-in-ocfs2_write_cluster.patch
ocfs2-record-unwritten-extents-when-populate-write-desc.patch
ocfs2-fix-sparse-file-data-ordering-issue-in-direct-io.patch
ocfs2-code-clean-up-for-direct-io.patch
ocfs2-fix-ip_unaligned_aio-deadlock-with-dio-work-queue.patch
ocfs2-take-ip_alloc_sem-in-ocfs2_dio_get_block-ocfs2_dio_end_io_write.patch
ocfs2-fix-disk-file-size-and-memory-file-size-mismatch.patch
ocfs2-fix-a-deadlock-issue-in-ocfs2_dio_end_io_write.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
