* [PATCH 1/9] iomap: cleanup up iomap_dio_bio_end_io()
2023-07-21 16:16 [PATCHSET v5 0/9] Improve async iomap DIO performance Jens Axboe
@ 2023-07-21 16:16 ` Jens Axboe
2023-07-21 16:16 ` [PATCH 2/9] iomap: add IOMAP_DIO_INLINE_COMP Jens Axboe
` (7 subsequent siblings)
8 siblings, 0 replies; 20+ messages in thread
From: Jens Axboe @ 2023-07-21 16:16 UTC (permalink / raw)
To: io-uring, linux-xfs; +Cc: hch, andres, david, djwong, Jens Axboe
Make the logic a bit easier to follow:
1) Add a release_bio out path, as everybody needs to touch that, and
have our bio ref check jump there if it's non-zero.
2) Add a kiocb local variable.
3) Add comments for each of the three conditions (sync, inline, or
async workqueue punt).
No functional changes in this patch.
Reviewed-by: Darrick J. Wong <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
---
fs/iomap/direct-io.c | 46 +++++++++++++++++++++++++++++---------------
1 file changed, 31 insertions(+), 15 deletions(-)
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index ea3b868c8355..0ce60e80c901 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -152,27 +152,43 @@ void iomap_dio_bio_end_io(struct bio *bio)
{
struct iomap_dio *dio = bio->bi_private;
bool should_dirty = (dio->flags & IOMAP_DIO_DIRTY);
+ struct kiocb *iocb = dio->iocb;
if (bio->bi_status)
iomap_dio_set_error(dio, blk_status_to_errno(bio->bi_status));
+ if (!atomic_dec_and_test(&dio->ref))
+ goto release_bio;
- if (atomic_dec_and_test(&dio->ref)) {
- if (dio->wait_for_completion) {
- struct task_struct *waiter = dio->submit.waiter;
- WRITE_ONCE(dio->submit.waiter, NULL);
- blk_wake_io_task(waiter);
- } else if (dio->flags & IOMAP_DIO_WRITE) {
- struct inode *inode = file_inode(dio->iocb->ki_filp);
-
- WRITE_ONCE(dio->iocb->private, NULL);
- INIT_WORK(&dio->aio.work, iomap_dio_complete_work);
- queue_work(inode->i_sb->s_dio_done_wq, &dio->aio.work);
- } else {
- WRITE_ONCE(dio->iocb->private, NULL);
- iomap_dio_complete_work(&dio->aio.work);
- }
+ /*
+ * Synchronous dio, task itself will handle any completion work
+ * that needs after IO. All we need to do is wake the task.
+ */
+ if (dio->wait_for_completion) {
+ struct task_struct *waiter = dio->submit.waiter;
+
+ WRITE_ONCE(dio->submit.waiter, NULL);
+ blk_wake_io_task(waiter);
+ goto release_bio;
+ }
+
+ /* Read completion can always complete inline. */
+ if (!(dio->flags & IOMAP_DIO_WRITE)) {
+ WRITE_ONCE(iocb->private, NULL);
+ iomap_dio_complete_work(&dio->aio.work);
+ goto release_bio;
}
+ /*
+ * Async DIO completion that requires filesystem level completion work
+ * gets punted to a work queue to complete as the operation may require
+ * more IO to be issued to finalise filesystem metadata changes or
+ * guarantee data integrity.
+ */
+ WRITE_ONCE(iocb->private, NULL);
+ INIT_WORK(&dio->aio.work, iomap_dio_complete_work);
+ queue_work(file_inode(iocb->ki_filp)->i_sb->s_dio_done_wq,
+ &dio->aio.work);
+release_bio:
if (should_dirty) {
bio_check_pages_dirty(bio);
} else {
--
2.40.1
^ permalink raw reply related [flat|nested] 20+ messages in thread* [PATCH 2/9] iomap: add IOMAP_DIO_INLINE_COMP
2023-07-21 16:16 [PATCHSET v5 0/9] Improve async iomap DIO performance Jens Axboe
2023-07-21 16:16 ` [PATCH 1/9] iomap: cleanup up iomap_dio_bio_end_io() Jens Axboe
@ 2023-07-21 16:16 ` Jens Axboe
2023-07-21 16:16 ` [PATCH 3/9] iomap: treat a write through cache the same as FUA Jens Axboe
` (6 subsequent siblings)
8 siblings, 0 replies; 20+ messages in thread
From: Jens Axboe @ 2023-07-21 16:16 UTC (permalink / raw)
To: io-uring, linux-xfs; +Cc: hch, andres, david, djwong, Jens Axboe
Rather than gate whether or not we need to punt a dio completion to a
workqueue on whether the IO is a write or not, add an explicit flag for
it. For now we treat them the same, reads always set the flags and async
writes do not.
No functional changes in this patch.
Reviewed-by: Darrick J. Wong <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
---
fs/iomap/direct-io.c | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index 0ce60e80c901..c654612b24e5 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -20,6 +20,7 @@
* Private flags for iomap_dio, must not overlap with the public ones in
* iomap.h:
*/
+#define IOMAP_DIO_INLINE_COMP (1 << 27)
#define IOMAP_DIO_WRITE_FUA (1 << 28)
#define IOMAP_DIO_NEED_SYNC (1 << 29)
#define IOMAP_DIO_WRITE (1 << 30)
@@ -171,8 +172,10 @@ void iomap_dio_bio_end_io(struct bio *bio)
goto release_bio;
}
- /* Read completion can always complete inline. */
- if (!(dio->flags & IOMAP_DIO_WRITE)) {
+ /*
+ * Flagged with IOMAP_DIO_INLINE_COMP, we can complete it inline
+ */
+ if (dio->flags & IOMAP_DIO_INLINE_COMP) {
WRITE_ONCE(iocb->private, NULL);
iomap_dio_complete_work(&dio->aio.work);
goto release_bio;
@@ -527,6 +530,9 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
iomi.flags |= IOMAP_NOWAIT;
if (iov_iter_rw(iter) == READ) {
+ /* reads can always complete inline */
+ dio->flags |= IOMAP_DIO_INLINE_COMP;
+
if (iomi.pos >= dio->i_size)
goto out_free_dio;
--
2.40.1
^ permalink raw reply related [flat|nested] 20+ messages in thread* [PATCH 3/9] iomap: treat a write through cache the same as FUA
2023-07-21 16:16 [PATCHSET v5 0/9] Improve async iomap DIO performance Jens Axboe
2023-07-21 16:16 ` [PATCH 1/9] iomap: cleanup up iomap_dio_bio_end_io() Jens Axboe
2023-07-21 16:16 ` [PATCH 2/9] iomap: add IOMAP_DIO_INLINE_COMP Jens Axboe
@ 2023-07-21 16:16 ` Jens Axboe
2023-07-21 16:25 ` Darrick J. Wong
2023-07-21 16:16 ` [PATCH 4/9] iomap: completed polled IO inline Jens Axboe
` (5 subsequent siblings)
8 siblings, 1 reply; 20+ messages in thread
From: Jens Axboe @ 2023-07-21 16:16 UTC (permalink / raw)
To: io-uring, linux-xfs; +Cc: hch, andres, david, djwong, Jens Axboe
Whether we have a write back cache and are using FUA or don't have
a write back cache at all is the same situation. Treat them the same.
This makes the IOMAP_DIO_WRITE_FUA name a bit misleading, as we have
two cases that provide stable writes:
1) Volatile write cache with FUA writes
2) Normal write without a volatile write cache
Rename that flag to IOMAP_DIO_STABLE_WRITE to make that clearer, and
update some of the FUA comments as well.
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
---
fs/iomap/direct-io.c | 29 +++++++++++++++++------------
1 file changed, 17 insertions(+), 12 deletions(-)
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index c654612b24e5..17b695b0e9d6 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -21,7 +21,7 @@
* iomap.h:
*/
#define IOMAP_DIO_INLINE_COMP (1 << 27)
-#define IOMAP_DIO_WRITE_FUA (1 << 28)
+#define IOMAP_DIO_WRITE_THROUGH (1 << 28)
#define IOMAP_DIO_NEED_SYNC (1 << 29)
#define IOMAP_DIO_WRITE (1 << 30)
#define IOMAP_DIO_DIRTY (1 << 31)
@@ -222,7 +222,7 @@ static void iomap_dio_zero(const struct iomap_iter *iter, struct iomap_dio *dio,
/*
* Figure out the bio's operation flags from the dio request, the
* mapping, and whether or not we want FUA. Note that we can end up
- * clearing the WRITE_FUA flag in the dio request.
+ * clearing the WRITE_THROUGH flag in the dio request.
*/
static inline blk_opf_t iomap_dio_bio_opflags(struct iomap_dio *dio,
const struct iomap *iomap, bool use_fua)
@@ -236,7 +236,7 @@ static inline blk_opf_t iomap_dio_bio_opflags(struct iomap_dio *dio,
if (use_fua)
opflags |= REQ_FUA;
else
- dio->flags &= ~IOMAP_DIO_WRITE_FUA;
+ dio->flags &= ~IOMAP_DIO_WRITE_THROUGH;
return opflags;
}
@@ -276,11 +276,13 @@ static loff_t iomap_dio_bio_iter(const struct iomap_iter *iter,
* Use a FUA write if we need datasync semantics, this is a pure
* data IO that doesn't require any metadata updates (including
* after IO completion such as unwritten extent conversion) and
- * the underlying device supports FUA. This allows us to avoid
- * cache flushes on IO completion.
+ * the underlying device either supports FUA or doesn't have
+ * a volatile write cache. This allows us to avoid cache flushes
+ * on IO completion.
*/
if (!(iomap->flags & (IOMAP_F_SHARED|IOMAP_F_DIRTY)) &&
- (dio->flags & IOMAP_DIO_WRITE_FUA) && bdev_fua(iomap->bdev))
+ (dio->flags & IOMAP_DIO_WRITE_THROUGH) &&
+ (bdev_fua(iomap->bdev) || !bdev_write_cache(iomap->bdev)))
use_fua = true;
}
@@ -560,12 +562,15 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
/*
* For datasync only writes, we optimistically try
- * using FUA for this IO. Any non-FUA write that
- * occurs will clear this flag, hence we know before
- * completion whether a cache flush is necessary.
+ * using WRITE_THROUGH for this IO. Stable writes are
+ * either FUA with a write cache, or a normal write to
+ * a device without a volatile write cache. For the
+ * former, Any non-FUA write that occurs will clear this
+ * flag, hence we know before completion whether a cache
+ * flush is necessary.
*/
if (!(iocb->ki_flags & IOCB_SYNC))
- dio->flags |= IOMAP_DIO_WRITE_FUA;
+ dio->flags |= IOMAP_DIO_WRITE_THROUGH;
}
/*
@@ -627,10 +632,10 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
iomap_dio_set_error(dio, ret);
/*
- * If all the writes we issued were FUA, we don't need to flush the
+ * If all the writes we issued were stable, we don't need to flush the
* cache on IO completion. Clear the sync flag for this case.
*/
- if (dio->flags & IOMAP_DIO_WRITE_FUA)
+ if (dio->flags & IOMAP_DIO_WRITE_THROUGH)
dio->flags &= ~IOMAP_DIO_NEED_SYNC;
WRITE_ONCE(iocb->private, dio->submit.poll_bio);
--
2.40.1
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH 3/9] iomap: treat a write through cache the same as FUA
2023-07-21 16:16 ` [PATCH 3/9] iomap: treat a write through cache the same as FUA Jens Axboe
@ 2023-07-21 16:25 ` Darrick J. Wong
2023-07-21 16:27 ` Jens Axboe
0 siblings, 1 reply; 20+ messages in thread
From: Darrick J. Wong @ 2023-07-21 16:25 UTC (permalink / raw)
To: Jens Axboe; +Cc: io-uring, linux-xfs, hch, andres, david
On Fri, Jul 21, 2023 at 10:16:44AM -0600, Jens Axboe wrote:
> Whether we have a write back cache and are using FUA or don't have
> a write back cache at all is the same situation. Treat them the same.
>
> This makes the IOMAP_DIO_WRITE_FUA name a bit misleading, as we have
> two cases that provide stable writes:
>
> 1) Volatile write cache with FUA writes
> 2) Normal write without a volatile write cache
>
> Rename that flag to IOMAP_DIO_STABLE_WRITE to make that clearer, and
> update some of the FUA comments as well.
>
> Reviewed-by: Christoph Hellwig <[email protected]>
> Signed-off-by: Jens Axboe <[email protected]>
> ---
> fs/iomap/direct-io.c | 29 +++++++++++++++++------------
> 1 file changed, 17 insertions(+), 12 deletions(-)
>
> diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
> index c654612b24e5..17b695b0e9d6 100644
> --- a/fs/iomap/direct-io.c
> +++ b/fs/iomap/direct-io.c
> @@ -21,7 +21,7 @@
> * iomap.h:
> */
> #define IOMAP_DIO_INLINE_COMP (1 << 27)
> -#define IOMAP_DIO_WRITE_FUA (1 << 28)
> +#define IOMAP_DIO_WRITE_THROUGH (1 << 28)
> #define IOMAP_DIO_NEED_SYNC (1 << 29)
> #define IOMAP_DIO_WRITE (1 << 30)
> #define IOMAP_DIO_DIRTY (1 << 31)
> @@ -222,7 +222,7 @@ static void iomap_dio_zero(const struct iomap_iter *iter, struct iomap_dio *dio,
> /*
> * Figure out the bio's operation flags from the dio request, the
> * mapping, and whether or not we want FUA. Note that we can end up
> - * clearing the WRITE_FUA flag in the dio request.
> + * clearing the WRITE_THROUGH flag in the dio request.
> */
> static inline blk_opf_t iomap_dio_bio_opflags(struct iomap_dio *dio,
> const struct iomap *iomap, bool use_fua)
> @@ -236,7 +236,7 @@ static inline blk_opf_t iomap_dio_bio_opflags(struct iomap_dio *dio,
> if (use_fua)
> opflags |= REQ_FUA;
> else
> - dio->flags &= ~IOMAP_DIO_WRITE_FUA;
> + dio->flags &= ~IOMAP_DIO_WRITE_THROUGH;
>
> return opflags;
> }
> @@ -276,11 +276,13 @@ static loff_t iomap_dio_bio_iter(const struct iomap_iter *iter,
> * Use a FUA write if we need datasync semantics, this is a pure
> * data IO that doesn't require any metadata updates (including
> * after IO completion such as unwritten extent conversion) and
> - * the underlying device supports FUA. This allows us to avoid
> - * cache flushes on IO completion.
> + * the underlying device either supports FUA or doesn't have
> + * a volatile write cache. This allows us to avoid cache flushes
> + * on IO completion.
> */
> if (!(iomap->flags & (IOMAP_F_SHARED|IOMAP_F_DIRTY)) &&
> - (dio->flags & IOMAP_DIO_WRITE_FUA) && bdev_fua(iomap->bdev))
> + (dio->flags & IOMAP_DIO_WRITE_THROUGH) &&
> + (bdev_fua(iomap->bdev) || !bdev_write_cache(iomap->bdev)))
> use_fua = true;
> }
>
> @@ -560,12 +562,15 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
>
> /*
> * For datasync only writes, we optimistically try
> - * using FUA for this IO. Any non-FUA write that
> - * occurs will clear this flag, hence we know before
> - * completion whether a cache flush is necessary.
> + * using WRITE_THROUGH for this IO. Stable writes are
"...using WRITE_THROUGH for this IO. This flag requires either FUA
writes through the device's write cache, or a normal write..."
> + * either FUA with a write cache, or a normal write to
> + * a device without a volatile write cache. For the
> + * former, Any non-FUA write that occurs will clear this
> + * flag, hence we know before completion whether a cache
> + * flush is necessary.
> */
> if (!(iocb->ki_flags & IOCB_SYNC))
> - dio->flags |= IOMAP_DIO_WRITE_FUA;
> + dio->flags |= IOMAP_DIO_WRITE_THROUGH;
> }
>
> /*
> @@ -627,10 +632,10 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
> iomap_dio_set_error(dio, ret);
>
> /*
> - * If all the writes we issued were FUA, we don't need to flush the
> + * If all the writes we issued were stable, we don't need to flush the
"If all the writes we issued were already written through to the media,
we don't need to flush..."
With those fixes,
Reviewed-by: Darrick J. Wong <[email protected]>
--D
> * cache on IO completion. Clear the sync flag for this case.
> */
> - if (dio->flags & IOMAP_DIO_WRITE_FUA)
> + if (dio->flags & IOMAP_DIO_WRITE_THROUGH)
> dio->flags &= ~IOMAP_DIO_NEED_SYNC;
>
> WRITE_ONCE(iocb->private, dio->submit.poll_bio);
> --
> 2.40.1
>
^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 3/9] iomap: treat a write through cache the same as FUA
2023-07-21 16:25 ` Darrick J. Wong
@ 2023-07-21 16:27 ` Jens Axboe
2023-07-21 16:47 ` Darrick J. Wong
0 siblings, 1 reply; 20+ messages in thread
From: Jens Axboe @ 2023-07-21 16:27 UTC (permalink / raw)
To: Darrick J. Wong; +Cc: io-uring, linux-xfs, hch, andres, david
On 7/21/23 10:25?AM, Darrick J. Wong wrote:
>> @@ -560,12 +562,15 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
>>
>> /*
>> * For datasync only writes, we optimistically try
>> - * using FUA for this IO. Any non-FUA write that
>> - * occurs will clear this flag, hence we know before
>> - * completion whether a cache flush is necessary.
>> + * using WRITE_THROUGH for this IO. Stable writes are
>
> "...using WRITE_THROUGH for this IO. This flag requires either FUA
> writes through the device's write cache, or a normal write..."
>
>> @@ -627,10 +632,10 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
>> iomap_dio_set_error(dio, ret);
>>
>> /*
>> - * If all the writes we issued were FUA, we don't need to flush the
>> + * If all the writes we issued were stable, we don't need to flush the
>
> "If all the writes we issued were already written through to the media,
> we don't need to flush..."
>
> With those fixes,
> Reviewed-by: Darrick J. Wong <[email protected]>
If you're queueing up this series, could you just make those two edits
while applying? I don't want to spam resend with just a comment change,
at least if I can avoid it...
--
Jens Axboe
^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 3/9] iomap: treat a write through cache the same as FUA
2023-07-21 16:27 ` Jens Axboe
@ 2023-07-21 16:47 ` Darrick J. Wong
2023-07-21 16:52 ` Jens Axboe
0 siblings, 1 reply; 20+ messages in thread
From: Darrick J. Wong @ 2023-07-21 16:47 UTC (permalink / raw)
To: Jens Axboe; +Cc: io-uring, linux-xfs, hch, andres, david
On Fri, Jul 21, 2023 at 10:27:16AM -0600, Jens Axboe wrote:
> On 7/21/23 10:25?AM, Darrick J. Wong wrote:
> >> @@ -560,12 +562,15 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
> >>
> >> /*
> >> * For datasync only writes, we optimistically try
> >> - * using FUA for this IO. Any non-FUA write that
> >> - * occurs will clear this flag, hence we know before
> >> - * completion whether a cache flush is necessary.
> >> + * using WRITE_THROUGH for this IO. Stable writes are
> >
> > "...using WRITE_THROUGH for this IO. This flag requires either FUA
> > writes through the device's write cache, or a normal write..."
> >
> >> @@ -627,10 +632,10 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
> >> iomap_dio_set_error(dio, ret);
> >>
> >> /*
> >> - * If all the writes we issued were FUA, we don't need to flush the
> >> + * If all the writes we issued were stable, we don't need to flush the
> >
> > "If all the writes we issued were already written through to the media,
> > we don't need to flush..."
> >
> > With those fixes,
> > Reviewed-by: Darrick J. Wong <[email protected]>
>
> If you're queueing up this series, could you just make those two edits
> while applying? I don't want to spam resend with just a comment change,
> at least if I can avoid it...
How about pushing the updated branch, tagging it with the cover letter
as the message, and sending me a pull request? Linus has been very
receptive to preserving cover letters this way.
--D
> --
> Jens Axboe
>
^ permalink raw reply [flat|nested] 20+ messages in thread
* Re: [PATCH 3/9] iomap: treat a write through cache the same as FUA
2023-07-21 16:47 ` Darrick J. Wong
@ 2023-07-21 16:52 ` Jens Axboe
0 siblings, 0 replies; 20+ messages in thread
From: Jens Axboe @ 2023-07-21 16:52 UTC (permalink / raw)
To: Darrick J. Wong; +Cc: io-uring, linux-xfs, hch, andres, david
On 7/21/23 10:47 AM, Darrick J. Wong wrote:
> On Fri, Jul 21, 2023 at 10:27:16AM -0600, Jens Axboe wrote:
>> On 7/21/23 10:25?AM, Darrick J. Wong wrote:
>>>> @@ -560,12 +562,15 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
>>>>
>>>> /*
>>>> * For datasync only writes, we optimistically try
>>>> - * using FUA for this IO. Any non-FUA write that
>>>> - * occurs will clear this flag, hence we know before
>>>> - * completion whether a cache flush is necessary.
>>>> + * using WRITE_THROUGH for this IO. Stable writes are
>>>
>>> "...using WRITE_THROUGH for this IO. This flag requires either FUA
>>> writes through the device's write cache, or a normal write..."
>>>
>>>> @@ -627,10 +632,10 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
>>>> iomap_dio_set_error(dio, ret);
>>>>
>>>> /*
>>>> - * If all the writes we issued were FUA, we don't need to flush the
>>>> + * If all the writes we issued were stable, we don't need to flush the
>>>
>>> "If all the writes we issued were already written through to the media,
>>> we don't need to flush..."
>>>
>>> With those fixes,
>>> Reviewed-by: Darrick J. Wong <[email protected]>
>>
>> If you're queueing up this series, could you just make those two edits
>> while applying? I don't want to spam resend with just a comment change,
>> at least if I can avoid it...
>
> How about pushing the updated branch, tagging it with the cover letter
> as the message, and sending me a pull request? Linus has been very
> receptive to preserving cover letters this way.
OK, will do.
--
Jens Axboe
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH 4/9] iomap: completed polled IO inline
2023-07-21 16:16 [PATCHSET v5 0/9] Improve async iomap DIO performance Jens Axboe
` (2 preceding siblings ...)
2023-07-21 16:16 ` [PATCH 3/9] iomap: treat a write through cache the same as FUA Jens Axboe
@ 2023-07-21 16:16 ` Jens Axboe
2023-07-21 16:16 ` [PATCH 5/9] iomap: only set iocb->private for polled bio Jens Axboe
` (4 subsequent siblings)
8 siblings, 0 replies; 20+ messages in thread
From: Jens Axboe @ 2023-07-21 16:16 UTC (permalink / raw)
To: io-uring, linux-xfs; +Cc: hch, andres, david, djwong, Jens Axboe
Polled IO is only allowed for conditions where task completion is safe
anyway, so we can always complete it inline. This cannot easily be
checked with a submission side flag, as the block layer may clear the
polled flag and turn it into a regular IO instead. Hence we need to
check this at completion time. If REQ_POLLED is still set, then we know
that this IO was successfully polled, and is completing in task context.
Reviewed-by: Darrick J. Wong <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
---
fs/iomap/direct-io.c | 14 ++++++++++++--
1 file changed, 12 insertions(+), 2 deletions(-)
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index 17b695b0e9d6..2b453e12dc16 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -173,9 +173,19 @@ void iomap_dio_bio_end_io(struct bio *bio)
}
/*
- * Flagged with IOMAP_DIO_INLINE_COMP, we can complete it inline
+ * Flagged with IOMAP_DIO_INLINE_COMP, we can complete it inline.
+ * Ditto for polled requests - if the flag is still at completion
+ * time, then we know the request was actually polled and completion
+ * is called from the task itself. This is why we need to check it
+ * here rather than flag it at issue time.
*/
- if (dio->flags & IOMAP_DIO_INLINE_COMP) {
+ if ((dio->flags & IOMAP_DIO_INLINE_COMP) || (bio->bi_opf & REQ_POLLED)) {
+ /*
+ * For polled IO, we need to clear ->private as it points to
+ * the bio being polled for. The completion side uses it to
+ * know if a given request has been found yet or not. For
+ * non-polled IO, ->private isn't applicable.
+ */
WRITE_ONCE(iocb->private, NULL);
iomap_dio_complete_work(&dio->aio.work);
goto release_bio;
--
2.40.1
^ permalink raw reply related [flat|nested] 20+ messages in thread* [PATCH 5/9] iomap: only set iocb->private for polled bio
2023-07-21 16:16 [PATCHSET v5 0/9] Improve async iomap DIO performance Jens Axboe
` (3 preceding siblings ...)
2023-07-21 16:16 ` [PATCH 4/9] iomap: completed polled IO inline Jens Axboe
@ 2023-07-21 16:16 ` Jens Axboe
2023-07-21 16:16 ` [PATCH 6/9] fs: add IOCB flags related to passing back dio completions Jens Axboe
` (3 subsequent siblings)
8 siblings, 0 replies; 20+ messages in thread
From: Jens Axboe @ 2023-07-21 16:16 UTC (permalink / raw)
To: io-uring, linux-xfs; +Cc: hch, andres, david, djwong, Jens Axboe
iocb->private is only used for polled IO, where the completer will
find the bio to poll through that field.
Assign it when we're submitting a polled bio, and get rid of the
dio->poll_bio indirection.
Reviewed-by: Darrick J. Wong <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
---
fs/iomap/direct-io.c | 13 +++++--------
1 file changed, 5 insertions(+), 8 deletions(-)
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index 2b453e12dc16..6ffa1b1ebe90 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -42,7 +42,6 @@ struct iomap_dio {
struct {
struct iov_iter *iter;
struct task_struct *waiter;
- struct bio *poll_bio;
} submit;
/* used for aio completion: */
@@ -64,12 +63,14 @@ static struct bio *iomap_dio_alloc_bio(const struct iomap_iter *iter,
static void iomap_dio_submit_bio(const struct iomap_iter *iter,
struct iomap_dio *dio, struct bio *bio, loff_t pos)
{
+ struct kiocb *iocb = dio->iocb;
+
atomic_inc(&dio->ref);
/* Sync dio can't be polled reliably */
- if ((dio->iocb->ki_flags & IOCB_HIPRI) && !is_sync_kiocb(dio->iocb)) {
- bio_set_polled(bio, dio->iocb);
- dio->submit.poll_bio = bio;
+ if ((iocb->ki_flags & IOCB_HIPRI) && !is_sync_kiocb(iocb)) {
+ bio_set_polled(bio, iocb);
+ WRITE_ONCE(iocb->private, bio);
}
if (dio->dops && dio->dops->submit_io)
@@ -197,7 +198,6 @@ void iomap_dio_bio_end_io(struct bio *bio)
* more IO to be issued to finalise filesystem metadata changes or
* guarantee data integrity.
*/
- WRITE_ONCE(iocb->private, NULL);
INIT_WORK(&dio->aio.work, iomap_dio_complete_work);
queue_work(file_inode(iocb->ki_filp)->i_sb->s_dio_done_wq,
&dio->aio.work);
@@ -536,7 +536,6 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
dio->submit.iter = iter;
dio->submit.waiter = current;
- dio->submit.poll_bio = NULL;
if (iocb->ki_flags & IOCB_NOWAIT)
iomi.flags |= IOMAP_NOWAIT;
@@ -648,8 +647,6 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
if (dio->flags & IOMAP_DIO_WRITE_THROUGH)
dio->flags &= ~IOMAP_DIO_NEED_SYNC;
- WRITE_ONCE(iocb->private, dio->submit.poll_bio);
-
/*
* We are about to drop our additional submission reference, which
* might be the last reference to the dio. There are three different
--
2.40.1
^ permalink raw reply related [flat|nested] 20+ messages in thread* [PATCH 6/9] fs: add IOCB flags related to passing back dio completions
2023-07-21 16:16 [PATCHSET v5 0/9] Improve async iomap DIO performance Jens Axboe
` (4 preceding siblings ...)
2023-07-21 16:16 ` [PATCH 5/9] iomap: only set iocb->private for polled bio Jens Axboe
@ 2023-07-21 16:16 ` Jens Axboe
2023-07-21 16:28 ` Darrick J. Wong
2023-07-21 16:16 ` [PATCH 7/9] io_uring/rw: add write support for IOCB_DIO_CALLER_COMP Jens Axboe
` (2 subsequent siblings)
8 siblings, 1 reply; 20+ messages in thread
From: Jens Axboe @ 2023-07-21 16:16 UTC (permalink / raw)
To: io-uring, linux-xfs; +Cc: hch, andres, david, djwong, Jens Axboe
Async dio completions generally happen from hard/soft IRQ context, which
means that users like iomap may need to defer some of the completion
handling to a workqueue. This is less efficient than having the original
issuer handle it, like we do for sync IO, and it adds latency to the
completions.
Add IOCB_DIO_CALLER_COMP, which the issuer can set if it is able to
safely punt these completions to a safe context. If the dio handler is
aware of this flag, assign a callback handler in kiocb->dio_complete and
associated data io kiocb->private. The issuer will then call this
handler with that data from task context.
No functional changes in this patch.
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
---
include/linux/fs.h | 35 +++++++++++++++++++++++++++++++++--
1 file changed, 33 insertions(+), 2 deletions(-)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 6867512907d6..60e2b4ecfc4d 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -338,6 +338,20 @@ enum rw_hint {
#define IOCB_NOIO (1 << 20)
/* can use bio alloc cache */
#define IOCB_ALLOC_CACHE (1 << 21)
+/*
+ * IOCB_DIO_CALLER_COMP can be set by the iocb owner, to indicate that the
+ * iocb completion can be passed back to the owner for execution from a safe
+ * context rather than needing to be punted through a workqueue.If this If this
+ * flag is set, the bio completion handling may set iocb->dio_complete to a
+ * handler function and iocb->private to context information for that handler.
+ * The issuer should call the handler with that context information from task
+ * context to complete the processing of the iocb. Note that while this
+ * provides a task context for the dio_complete() callback, it should only be
+ * used on the completion side for non-IO generating completions. It's fine to
+ * call blocking functions from this callback, but they should not wait for
+ * unrelated IO (like cache flushing, new IO generation, etc).
+ */
+#define IOCB_DIO_CALLER_COMP (1 << 22)
/* for use in trace events */
#define TRACE_IOCB_STRINGS \
@@ -351,7 +365,8 @@ enum rw_hint {
{ IOCB_WRITE, "WRITE" }, \
{ IOCB_WAITQ, "WAITQ" }, \
{ IOCB_NOIO, "NOIO" }, \
- { IOCB_ALLOC_CACHE, "ALLOC_CACHE" }
+ { IOCB_ALLOC_CACHE, "ALLOC_CACHE" }, \
+ { IOCB_DIO_CALLER_COMP, "CALLER_COMP" }
struct kiocb {
struct file *ki_filp;
@@ -360,7 +375,23 @@ struct kiocb {
void *private;
int ki_flags;
u16 ki_ioprio; /* See linux/ioprio.h */
- struct wait_page_queue *ki_waitq; /* for async buffered IO */
+ union {
+ /*
+ * Only used for async buffered reads, where it denotes the
+ * page waitqueue associated with completing the read. Valid
+ * IFF IOCB_WAITQ is set.
+ */
+ struct wait_page_queue *ki_waitq;
+ /*
+ * Can be used for O_DIRECT IO, where the completion handling
+ * is punted back to the issuer of the IO. May only be set
+ * if IOCB_DIO_CALLER_COMP is set by the issuer, and the issuer
+ * must then check for presence of this handler when ki_complete
+ * is invoked. The data passed in to this handler must be
+ * assigned to ->private when dio_complete is assigned.
+ */
+ ssize_t (*dio_complete)(void *data);
+ };
};
static inline bool is_sync_kiocb(struct kiocb *kiocb)
--
2.40.1
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH 6/9] fs: add IOCB flags related to passing back dio completions
2023-07-21 16:16 ` [PATCH 6/9] fs: add IOCB flags related to passing back dio completions Jens Axboe
@ 2023-07-21 16:28 ` Darrick J. Wong
2023-07-21 16:30 ` Jens Axboe
0 siblings, 1 reply; 20+ messages in thread
From: Darrick J. Wong @ 2023-07-21 16:28 UTC (permalink / raw)
To: Jens Axboe; +Cc: io-uring, linux-xfs, hch, andres, david
On Fri, Jul 21, 2023 at 10:16:47AM -0600, Jens Axboe wrote:
> Async dio completions generally happen from hard/soft IRQ context, which
> means that users like iomap may need to defer some of the completion
> handling to a workqueue. This is less efficient than having the original
> issuer handle it, like we do for sync IO, and it adds latency to the
> completions.
>
> Add IOCB_DIO_CALLER_COMP, which the issuer can set if it is able to
> safely punt these completions to a safe context. If the dio handler is
> aware of this flag, assign a callback handler in kiocb->dio_complete and
> associated data io kiocb->private. The issuer will then call this
> handler with that data from task context.
>
> No functional changes in this patch.
>
> Reviewed-by: Christoph Hellwig <[email protected]>
> Signed-off-by: Jens Axboe <[email protected]>
> ---
> include/linux/fs.h | 35 +++++++++++++++++++++++++++++++++--
> 1 file changed, 33 insertions(+), 2 deletions(-)
>
> diff --git a/include/linux/fs.h b/include/linux/fs.h
> index 6867512907d6..60e2b4ecfc4d 100644
> --- a/include/linux/fs.h
> +++ b/include/linux/fs.h
> @@ -338,6 +338,20 @@ enum rw_hint {
> #define IOCB_NOIO (1 << 20)
> /* can use bio alloc cache */
> #define IOCB_ALLOC_CACHE (1 << 21)
> +/*
> + * IOCB_DIO_CALLER_COMP can be set by the iocb owner, to indicate that the
> + * iocb completion can be passed back to the owner for execution from a safe
> + * context rather than needing to be punted through a workqueue.If this If this
"...through a workqueue. If this flag is set..."
Need a space after the period, and delete one of the "If this".
With that fixed,
Reviewed-by: Darrick J. Wong <[email protected]>
--D
> + * flag is set, the bio completion handling may set iocb->dio_complete to a
> + * handler function and iocb->private to context information for that handler.
> + * The issuer should call the handler with that context information from task
> + * context to complete the processing of the iocb. Note that while this
> + * provides a task context for the dio_complete() callback, it should only be
> + * used on the completion side for non-IO generating completions. It's fine to
> + * call blocking functions from this callback, but they should not wait for
> + * unrelated IO (like cache flushing, new IO generation, etc).
> + */
> +#define IOCB_DIO_CALLER_COMP (1 << 22)
>
> /* for use in trace events */
> #define TRACE_IOCB_STRINGS \
> @@ -351,7 +365,8 @@ enum rw_hint {
> { IOCB_WRITE, "WRITE" }, \
> { IOCB_WAITQ, "WAITQ" }, \
> { IOCB_NOIO, "NOIO" }, \
> - { IOCB_ALLOC_CACHE, "ALLOC_CACHE" }
> + { IOCB_ALLOC_CACHE, "ALLOC_CACHE" }, \
> + { IOCB_DIO_CALLER_COMP, "CALLER_COMP" }
>
> struct kiocb {
> struct file *ki_filp;
> @@ -360,7 +375,23 @@ struct kiocb {
> void *private;
> int ki_flags;
> u16 ki_ioprio; /* See linux/ioprio.h */
> - struct wait_page_queue *ki_waitq; /* for async buffered IO */
> + union {
> + /*
> + * Only used for async buffered reads, where it denotes the
> + * page waitqueue associated with completing the read. Valid
> + * IFF IOCB_WAITQ is set.
> + */
> + struct wait_page_queue *ki_waitq;
> + /*
> + * Can be used for O_DIRECT IO, where the completion handling
> + * is punted back to the issuer of the IO. May only be set
> + * if IOCB_DIO_CALLER_COMP is set by the issuer, and the issuer
> + * must then check for presence of this handler when ki_complete
> + * is invoked. The data passed in to this handler must be
> + * assigned to ->private when dio_complete is assigned.
> + */
> + ssize_t (*dio_complete)(void *data);
> + };
> };
>
> static inline bool is_sync_kiocb(struct kiocb *kiocb)
> --
> 2.40.1
>
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH 6/9] fs: add IOCB flags related to passing back dio completions
2023-07-21 16:28 ` Darrick J. Wong
@ 2023-07-21 16:30 ` Jens Axboe
2023-07-21 16:43 ` Jens Axboe
0 siblings, 1 reply; 20+ messages in thread
From: Jens Axboe @ 2023-07-21 16:30 UTC (permalink / raw)
To: Darrick J. Wong; +Cc: io-uring, linux-xfs, hch, andres, david
On 7/21/23 10:28?AM, Darrick J. Wong wrote:
>> diff --git a/include/linux/fs.h b/include/linux/fs.h
>> index 6867512907d6..60e2b4ecfc4d 100644
>> --- a/include/linux/fs.h
>> +++ b/include/linux/fs.h
>> @@ -338,6 +338,20 @@ enum rw_hint {
>> #define IOCB_NOIO (1 << 20)
>> /* can use bio alloc cache */
>> #define IOCB_ALLOC_CACHE (1 << 21)
>> +/*
>> + * IOCB_DIO_CALLER_COMP can be set by the iocb owner, to indicate that the
>> + * iocb completion can be passed back to the owner for execution from a safe
>> + * context rather than needing to be punted through a workqueue.If this If this
>
> "...through a workqueue. If this flag is set..."
>
> Need a space after the period, and delete one of the "If this".
>
> With that fixed,
> Reviewed-by: Darrick J. Wong <[email protected]>
Thanks - same ask on the edit. Or let me know if:
a) you're fine with staging this in a separate branch for 6.6, or
b) you want a v5a/v6 edition posted
Either way is no trouble for me, just wary of spamming...
--
Jens Axboe
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH 6/9] fs: add IOCB flags related to passing back dio completions
2023-07-21 16:30 ` Jens Axboe
@ 2023-07-21 16:43 ` Jens Axboe
0 siblings, 0 replies; 20+ messages in thread
From: Jens Axboe @ 2023-07-21 16:43 UTC (permalink / raw)
To: Darrick J. Wong; +Cc: io-uring, linux-xfs, hch, andres, david
On 7/21/23 10:30?AM, Jens Axboe wrote:
> On 7/21/23 10:28?AM, Darrick J. Wong wrote:
>>> diff --git a/include/linux/fs.h b/include/linux/fs.h
>>> index 6867512907d6..60e2b4ecfc4d 100644
>>> --- a/include/linux/fs.h
>>> +++ b/include/linux/fs.h
>>> @@ -338,6 +338,20 @@ enum rw_hint {
>>> #define IOCB_NOIO (1 << 20)
>>> /* can use bio alloc cache */
>>> #define IOCB_ALLOC_CACHE (1 << 21)
>>> +/*
>>> + * IOCB_DIO_CALLER_COMP can be set by the iocb owner, to indicate that the
>>> + * iocb completion can be passed back to the owner for execution from a safe
>>> + * context rather than needing to be punted through a workqueue.If this If this
>>
>> "...through a workqueue. If this flag is set..."
>>
>> Need a space after the period, and delete one of the "If this".
>>
>> With that fixed,
>> Reviewed-by: Darrick J. Wong <[email protected]>
>
> Thanks - same ask on the edit. Or let me know if:
>
> a) you're fine with staging this in a separate branch for 6.6, or
> b) you want a v5a/v6 edition posted
>
> Either way is no trouble for me, just wary of spamming...
FWIW, here's the updated branch:
https://git.kernel.dk/cgit/linux/log/?h=xfs-async-dio.5
--
Jens Axboe
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH 7/9] io_uring/rw: add write support for IOCB_DIO_CALLER_COMP
2023-07-21 16:16 [PATCHSET v5 0/9] Improve async iomap DIO performance Jens Axboe
` (5 preceding siblings ...)
2023-07-21 16:16 ` [PATCH 6/9] fs: add IOCB flags related to passing back dio completions Jens Axboe
@ 2023-07-21 16:16 ` Jens Axboe
2023-07-21 16:29 ` Darrick J. Wong
2023-07-21 16:16 ` [PATCH 8/9] iomap: support IOCB_DIO_CALLER_COMP Jens Axboe
2023-07-21 16:16 ` [PATCH 9/9] iomap: use an unsigned type for IOMAP_DIO_* defines Jens Axboe
8 siblings, 1 reply; 20+ messages in thread
From: Jens Axboe @ 2023-07-21 16:16 UTC (permalink / raw)
To: io-uring, linux-xfs; +Cc: hch, andres, david, djwong, Jens Axboe
If the filesystem dio handler understands IOCB_DIO_CALLER_COMP, we'll
get a kiocb->ki_complete() callback with kiocb->dio_complete set. In
that case, rather than complete the IO directly through task_work, queue
up an intermediate task_work handler that first processes this callback
and then immediately completes the request.
For XFS, this avoids a punt through a workqueue, which is a lot less
efficient and adds latency to lower queue depth (or sync) O_DIRECT
writes.
Only do this for non-polled IO, as polled IO doesn't need this kind
of deferral as it always completes within the task itself. This then
avoids a check for deferral in the polled IO completion handler.
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
---
io_uring/rw.c | 26 +++++++++++++++++++++++---
1 file changed, 23 insertions(+), 3 deletions(-)
diff --git a/io_uring/rw.c b/io_uring/rw.c
index 1bce2208b65c..f19f65b3f0ee 100644
--- a/io_uring/rw.c
+++ b/io_uring/rw.c
@@ -105,6 +105,7 @@ int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe)
} else {
rw->kiocb.ki_ioprio = get_current_ioprio();
}
+ rw->kiocb.dio_complete = NULL;
rw->addr = READ_ONCE(sqe->addr);
rw->len = READ_ONCE(sqe->len);
@@ -285,6 +286,14 @@ static inline int io_fixup_rw_res(struct io_kiocb *req, long res)
void io_req_rw_complete(struct io_kiocb *req, struct io_tw_state *ts)
{
+ struct io_rw *rw = io_kiocb_to_cmd(req, struct io_rw);
+
+ if (rw->kiocb.dio_complete) {
+ long res = rw->kiocb.dio_complete(rw->kiocb.private);
+
+ io_req_set_res(req, io_fixup_rw_res(req, res), 0);
+ }
+
io_req_io_end(req);
if (req->flags & (REQ_F_BUFFER_SELECTED|REQ_F_BUFFER_RING)) {
@@ -300,9 +309,11 @@ static void io_complete_rw(struct kiocb *kiocb, long res)
struct io_rw *rw = container_of(kiocb, struct io_rw, kiocb);
struct io_kiocb *req = cmd_to_io_kiocb(rw);
- if (__io_complete_rw_common(req, res))
- return;
- io_req_set_res(req, io_fixup_rw_res(req, res), 0);
+ if (!rw->kiocb.dio_complete) {
+ if (__io_complete_rw_common(req, res))
+ return;
+ io_req_set_res(req, io_fixup_rw_res(req, res), 0);
+ }
req->io_task_work.func = io_req_rw_complete;
__io_req_task_work_add(req, IOU_F_TWQ_LAZY_WAKE);
}
@@ -916,6 +927,15 @@ int io_write(struct io_kiocb *req, unsigned int issue_flags)
}
kiocb->ki_flags |= IOCB_WRITE;
+ /*
+ * For non-polled IO, set IOCB_DIO_CALLER_COMP, stating that our handler
+ * groks deferring the completion to task context. This isn't
+ * necessary and useful for polled IO as that can always complete
+ * directly.
+ */
+ if (!(kiocb->ki_flags & IOCB_HIPRI))
+ kiocb->ki_flags |= IOCB_DIO_CALLER_COMP;
+
if (likely(req->file->f_op->write_iter))
ret2 = call_write_iter(req->file, kiocb, &s->iter);
else if (req->file->f_op->write)
--
2.40.1
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH 7/9] io_uring/rw: add write support for IOCB_DIO_CALLER_COMP
2023-07-21 16:16 ` [PATCH 7/9] io_uring/rw: add write support for IOCB_DIO_CALLER_COMP Jens Axboe
@ 2023-07-21 16:29 ` Darrick J. Wong
0 siblings, 0 replies; 20+ messages in thread
From: Darrick J. Wong @ 2023-07-21 16:29 UTC (permalink / raw)
To: Jens Axboe; +Cc: io-uring, linux-xfs, hch, andres, david
On Fri, Jul 21, 2023 at 10:16:48AM -0600, Jens Axboe wrote:
> If the filesystem dio handler understands IOCB_DIO_CALLER_COMP, we'll
> get a kiocb->ki_complete() callback with kiocb->dio_complete set. In
> that case, rather than complete the IO directly through task_work, queue
> up an intermediate task_work handler that first processes this callback
> and then immediately completes the request.
>
> For XFS, this avoids a punt through a workqueue, which is a lot less
> efficient and adds latency to lower queue depth (or sync) O_DIRECT
> writes.
>
> Only do this for non-polled IO, as polled IO doesn't need this kind
> of deferral as it always completes within the task itself. This then
> avoids a check for deferral in the polled IO completion handler.
>
> Reviewed-by: Christoph Hellwig <[email protected]>
> Signed-off-by: Jens Axboe <[email protected]>
Looks good!
Reviewed-by: Darrick J. Wong <[email protected]>
--D
> ---
> io_uring/rw.c | 26 +++++++++++++++++++++++---
> 1 file changed, 23 insertions(+), 3 deletions(-)
>
> diff --git a/io_uring/rw.c b/io_uring/rw.c
> index 1bce2208b65c..f19f65b3f0ee 100644
> --- a/io_uring/rw.c
> +++ b/io_uring/rw.c
> @@ -105,6 +105,7 @@ int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe)
> } else {
> rw->kiocb.ki_ioprio = get_current_ioprio();
> }
> + rw->kiocb.dio_complete = NULL;
>
> rw->addr = READ_ONCE(sqe->addr);
> rw->len = READ_ONCE(sqe->len);
> @@ -285,6 +286,14 @@ static inline int io_fixup_rw_res(struct io_kiocb *req, long res)
>
> void io_req_rw_complete(struct io_kiocb *req, struct io_tw_state *ts)
> {
> + struct io_rw *rw = io_kiocb_to_cmd(req, struct io_rw);
> +
> + if (rw->kiocb.dio_complete) {
> + long res = rw->kiocb.dio_complete(rw->kiocb.private);
> +
> + io_req_set_res(req, io_fixup_rw_res(req, res), 0);
> + }
> +
> io_req_io_end(req);
>
> if (req->flags & (REQ_F_BUFFER_SELECTED|REQ_F_BUFFER_RING)) {
> @@ -300,9 +309,11 @@ static void io_complete_rw(struct kiocb *kiocb, long res)
> struct io_rw *rw = container_of(kiocb, struct io_rw, kiocb);
> struct io_kiocb *req = cmd_to_io_kiocb(rw);
>
> - if (__io_complete_rw_common(req, res))
> - return;
> - io_req_set_res(req, io_fixup_rw_res(req, res), 0);
> + if (!rw->kiocb.dio_complete) {
> + if (__io_complete_rw_common(req, res))
> + return;
> + io_req_set_res(req, io_fixup_rw_res(req, res), 0);
> + }
> req->io_task_work.func = io_req_rw_complete;
> __io_req_task_work_add(req, IOU_F_TWQ_LAZY_WAKE);
> }
> @@ -916,6 +927,15 @@ int io_write(struct io_kiocb *req, unsigned int issue_flags)
> }
> kiocb->ki_flags |= IOCB_WRITE;
>
> + /*
> + * For non-polled IO, set IOCB_DIO_CALLER_COMP, stating that our handler
> + * groks deferring the completion to task context. This isn't
> + * necessary and useful for polled IO as that can always complete
> + * directly.
> + */
> + if (!(kiocb->ki_flags & IOCB_HIPRI))
> + kiocb->ki_flags |= IOCB_DIO_CALLER_COMP;
> +
> if (likely(req->file->f_op->write_iter))
> ret2 = call_write_iter(req->file, kiocb, &s->iter);
> else if (req->file->f_op->write)
> --
> 2.40.1
>
^ permalink raw reply [flat|nested] 20+ messages in thread
* [PATCH 8/9] iomap: support IOCB_DIO_CALLER_COMP
2023-07-21 16:16 [PATCHSET v5 0/9] Improve async iomap DIO performance Jens Axboe
` (6 preceding siblings ...)
2023-07-21 16:16 ` [PATCH 7/9] io_uring/rw: add write support for IOCB_DIO_CALLER_COMP Jens Axboe
@ 2023-07-21 16:16 ` Jens Axboe
2023-07-21 16:16 ` [PATCH 9/9] iomap: use an unsigned type for IOMAP_DIO_* defines Jens Axboe
8 siblings, 0 replies; 20+ messages in thread
From: Jens Axboe @ 2023-07-21 16:16 UTC (permalink / raw)
To: io-uring, linux-xfs; +Cc: hch, andres, david, djwong, Jens Axboe
If IOCB_DIO_CALLER_COMP is set, utilize that to set kiocb->dio_complete
handler and data for that callback. Rather than punt the completion to a
workqueue, we pass back the handler and data to the issuer and will get
a callback from a safe task context.
Using the following fio job to randomly dio write 4k blocks at
queue depths of 1..16:
fio --name=dio-write --filename=/data1/file --time_based=1 \
--runtime=10 --bs=4096 --rw=randwrite --norandommap --buffered=0 \
--cpus_allowed=4 --ioengine=io_uring --iodepth=$depth
shows the following results before and after this patch:
Stock Patched Diff
=======================================
QD1 155K 162K + 4.5%
QD2 290K 313K + 7.9%
QD4 533K 597K +12.0%
QD8 604K 827K +36.9%
QD16 615K 845K +37.4%
which shows nice wins all around. If we factored in per-IOP efficiency,
the wins look even nicer. This becomes apparent as queue depth rises,
as the offloaded workqueue completions runs out of steam.
Reviewed-by: Darrick J. Wong <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
---
fs/iomap/direct-io.c | 55 +++++++++++++++++++++++++++++++++++++++++++-
1 file changed, 54 insertions(+), 1 deletion(-)
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index 6ffa1b1ebe90..ae9046d16d71 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -20,6 +20,7 @@
* Private flags for iomap_dio, must not overlap with the public ones in
* iomap.h:
*/
+#define IOMAP_DIO_CALLER_COMP (1 << 26)
#define IOMAP_DIO_INLINE_COMP (1 << 27)
#define IOMAP_DIO_WRITE_THROUGH (1 << 28)
#define IOMAP_DIO_NEED_SYNC (1 << 29)
@@ -132,6 +133,11 @@ ssize_t iomap_dio_complete(struct iomap_dio *dio)
}
EXPORT_SYMBOL_GPL(iomap_dio_complete);
+static ssize_t iomap_dio_deferred_complete(void *data)
+{
+ return iomap_dio_complete(data);
+}
+
static void iomap_dio_complete_work(struct work_struct *work)
{
struct iomap_dio *dio = container_of(work, struct iomap_dio, aio.work);
@@ -192,6 +198,31 @@ void iomap_dio_bio_end_io(struct bio *bio)
goto release_bio;
}
+ /*
+ * If this dio is flagged with IOMAP_DIO_CALLER_COMP, then schedule
+ * our completion that way to avoid an async punt to a workqueue.
+ */
+ if (dio->flags & IOMAP_DIO_CALLER_COMP) {
+ /* only polled IO cares about private cleared */
+ iocb->private = dio;
+ iocb->dio_complete = iomap_dio_deferred_complete;
+
+ /*
+ * Invoke ->ki_complete() directly. We've assigned our
+ * dio_complete callback handler, and since the issuer set
+ * IOCB_DIO_CALLER_COMP, we know their ki_complete handler will
+ * notice ->dio_complete being set and will defer calling that
+ * handler until it can be done from a safe task context.
+ *
+ * Note that the 'res' being passed in here is not important
+ * for this case. The actual completion value of the request
+ * will be gotten from dio_complete when that is run by the
+ * issuer.
+ */
+ iocb->ki_complete(iocb, 0);
+ goto release_bio;
+ }
+
/*
* Async DIO completion that requires filesystem level completion work
* gets punted to a work queue to complete as the operation may require
@@ -288,12 +319,17 @@ static loff_t iomap_dio_bio_iter(const struct iomap_iter *iter,
* after IO completion such as unwritten extent conversion) and
* the underlying device either supports FUA or doesn't have
* a volatile write cache. This allows us to avoid cache flushes
- * on IO completion.
+ * on IO completion. If we can't use writethrough and need to
+ * sync, disable in-task completions as dio completion will
+ * need to call generic_write_sync() which will do a blocking
+ * fsync / cache flush call.
*/
if (!(iomap->flags & (IOMAP_F_SHARED|IOMAP_F_DIRTY)) &&
(dio->flags & IOMAP_DIO_WRITE_THROUGH) &&
(bdev_fua(iomap->bdev) || !bdev_write_cache(iomap->bdev)))
use_fua = true;
+ else if (dio->flags & IOMAP_DIO_NEED_SYNC)
+ dio->flags &= ~IOMAP_DIO_CALLER_COMP;
}
/*
@@ -319,6 +355,14 @@ static loff_t iomap_dio_bio_iter(const struct iomap_iter *iter,
pad = pos & (fs_block_size - 1);
if (pad)
iomap_dio_zero(iter, dio, pos - pad, pad);
+
+ /*
+ * If need_zeroout is set, then this is a new or unwritten
+ * extent, or dirty file metadata have not been persisted to
+ * disk. These need extra handling at completion time, so
+ * disable in-task deferred completion for those.
+ */
+ dio->flags &= ~IOMAP_DIO_CALLER_COMP;
}
/*
@@ -557,6 +601,15 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter,
iomi.flags |= IOMAP_WRITE;
dio->flags |= IOMAP_DIO_WRITE;
+ /*
+ * Flag as supporting deferred completions, if the issuer
+ * groks it. This can avoid a workqueue punt for writes.
+ * We may later clear this flag if we need to do other IO
+ * as part of this IO completion.
+ */
+ if (iocb->ki_flags & IOCB_DIO_CALLER_COMP)
+ dio->flags |= IOMAP_DIO_CALLER_COMP;
+
if (dio_flags & IOMAP_DIO_OVERWRITE_ONLY) {
ret = -EAGAIN;
if (iomi.pos >= dio->i_size ||
--
2.40.1
^ permalink raw reply related [flat|nested] 20+ messages in thread* [PATCH 9/9] iomap: use an unsigned type for IOMAP_DIO_* defines
2023-07-21 16:16 [PATCHSET v5 0/9] Improve async iomap DIO performance Jens Axboe
` (7 preceding siblings ...)
2023-07-21 16:16 ` [PATCH 8/9] iomap: support IOCB_DIO_CALLER_COMP Jens Axboe
@ 2023-07-21 16:16 ` Jens Axboe
2023-07-21 16:29 ` Darrick J. Wong
2023-07-24 16:36 ` Christoph Hellwig
8 siblings, 2 replies; 20+ messages in thread
From: Jens Axboe @ 2023-07-21 16:16 UTC (permalink / raw)
To: io-uring, linux-xfs; +Cc: hch, andres, david, djwong, Jens Axboe
IOMAP_DIO_DIRTY shifts by 31 bits, which makes UBSAN unhappy. Clean up
all the defines by making the shifted value an unsigned value.
Reported-by: Darrick J. Wong <[email protected]>
Signed-off-by: Jens Axboe <[email protected]>
---
fs/iomap/direct-io.c | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
index ae9046d16d71..dc9fe2ac9136 100644
--- a/fs/iomap/direct-io.c
+++ b/fs/iomap/direct-io.c
@@ -20,12 +20,12 @@
* Private flags for iomap_dio, must not overlap with the public ones in
* iomap.h:
*/
-#define IOMAP_DIO_CALLER_COMP (1 << 26)
-#define IOMAP_DIO_INLINE_COMP (1 << 27)
-#define IOMAP_DIO_WRITE_THROUGH (1 << 28)
-#define IOMAP_DIO_NEED_SYNC (1 << 29)
-#define IOMAP_DIO_WRITE (1 << 30)
-#define IOMAP_DIO_DIRTY (1 << 31)
+#define IOMAP_DIO_CALLER_COMP (1U << 26)
+#define IOMAP_DIO_INLINE_COMP (1U << 27)
+#define IOMAP_DIO_WRITE_THROUGH (1U << 28)
+#define IOMAP_DIO_NEED_SYNC (1U << 29)
+#define IOMAP_DIO_WRITE (1U << 30)
+#define IOMAP_DIO_DIRTY (1U << 31)
struct iomap_dio {
struct kiocb *iocb;
--
2.40.1
^ permalink raw reply related [flat|nested] 20+ messages in thread* Re: [PATCH 9/9] iomap: use an unsigned type for IOMAP_DIO_* defines
2023-07-21 16:16 ` [PATCH 9/9] iomap: use an unsigned type for IOMAP_DIO_* defines Jens Axboe
@ 2023-07-21 16:29 ` Darrick J. Wong
2023-07-24 16:36 ` Christoph Hellwig
1 sibling, 0 replies; 20+ messages in thread
From: Darrick J. Wong @ 2023-07-21 16:29 UTC (permalink / raw)
To: Jens Axboe; +Cc: io-uring, linux-xfs, hch, andres, david
On Fri, Jul 21, 2023 at 10:16:50AM -0600, Jens Axboe wrote:
> IOMAP_DIO_DIRTY shifts by 31 bits, which makes UBSAN unhappy. Clean up
> all the defines by making the shifted value an unsigned value.
>
> Reported-by: Darrick J. Wong <[email protected]>
> Signed-off-by: Jens Axboe <[email protected]>
Thanks!
Reviewed-by: Darrick J. Wong <[email protected]>
--D
> ---
> fs/iomap/direct-io.c | 12 ++++++------
> 1 file changed, 6 insertions(+), 6 deletions(-)
>
> diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c
> index ae9046d16d71..dc9fe2ac9136 100644
> --- a/fs/iomap/direct-io.c
> +++ b/fs/iomap/direct-io.c
> @@ -20,12 +20,12 @@
> * Private flags for iomap_dio, must not overlap with the public ones in
> * iomap.h:
> */
> -#define IOMAP_DIO_CALLER_COMP (1 << 26)
> -#define IOMAP_DIO_INLINE_COMP (1 << 27)
> -#define IOMAP_DIO_WRITE_THROUGH (1 << 28)
> -#define IOMAP_DIO_NEED_SYNC (1 << 29)
> -#define IOMAP_DIO_WRITE (1 << 30)
> -#define IOMAP_DIO_DIRTY (1 << 31)
> +#define IOMAP_DIO_CALLER_COMP (1U << 26)
> +#define IOMAP_DIO_INLINE_COMP (1U << 27)
> +#define IOMAP_DIO_WRITE_THROUGH (1U << 28)
> +#define IOMAP_DIO_NEED_SYNC (1U << 29)
> +#define IOMAP_DIO_WRITE (1U << 30)
> +#define IOMAP_DIO_DIRTY (1U << 31)
>
> struct iomap_dio {
> struct kiocb *iocb;
> --
> 2.40.1
>
^ permalink raw reply [flat|nested] 20+ messages in thread* Re: [PATCH 9/9] iomap: use an unsigned type for IOMAP_DIO_* defines
2023-07-21 16:16 ` [PATCH 9/9] iomap: use an unsigned type for IOMAP_DIO_* defines Jens Axboe
2023-07-21 16:29 ` Darrick J. Wong
@ 2023-07-24 16:36 ` Christoph Hellwig
1 sibling, 0 replies; 20+ messages in thread
From: Christoph Hellwig @ 2023-07-24 16:36 UTC (permalink / raw)
To: Jens Axboe; +Cc: io-uring, linux-xfs, hch, andres, david, djwong
Looks good:
Reviewed-by: Christoph Hellwig <[email protected]>
^ permalink raw reply [flat|nested] 20+ messages in thread