* [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible
@ 2021-07-29 3:42 Ming Lei
2021-07-29 7:17 ` Oleksandr Natalenko
2021-08-05 3:35 ` Ming Lei
0 siblings, 2 replies; 11+ messages in thread
From: Ming Lei @ 2021-07-29 3:42 UTC (permalink / raw
To: Jens Axboe; +Cc: linux-block, Christoph Hellwig, Ming Lei, Oleksandr Natalenko
When merging one bio to request, if they are discard IO and the queue
supports multi-range discard, we need to return ELEVATOR_DISCARD_MERGE
because both block core and related drivers(nvme, virtio-blk) doesn't
handle mixed discard io merge(traditional IO merge together with
discard merge) well.
Fix the issue by returning ELEVATOR_DISCARD_MERGE in this situation,
so both blk-mq and drivers just need to handle multi-range discard.
Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
---
block/bfq-iosched.c | 3 +++
block/blk-merge.c | 16 ----------------
block/elevator.c | 3 +++
block/mq-deadline-main.c | 2 ++
include/linux/blkdev.h | 16 ++++++++++++++++
5 files changed, 24 insertions(+), 16 deletions(-)
diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index 727955918563..673a634eadd9 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -2361,6 +2361,9 @@ static int bfq_request_merge(struct request_queue *q, struct request **req,
__rq = bfq_find_rq_fmerge(bfqd, bio, q);
if (__rq && elv_bio_merge_ok(__rq, bio)) {
*req = __rq;
+
+ if (blk_discard_mergable(__rq))
+ return ELEVATOR_DISCARD_MERGE;
return ELEVATOR_FRONT_MERGE;
}
diff --git a/block/blk-merge.c b/block/blk-merge.c
index a11b3b53717e..f8707ff7e2fc 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -705,22 +705,6 @@ static void blk_account_io_merge_request(struct request *req)
}
}
-/*
- * Two cases of handling DISCARD merge:
- * If max_discard_segments > 1, the driver takes every bio
- * as a range and send them to controller together. The ranges
- * needn't to be contiguous.
- * Otherwise, the bios/requests will be handled as same as
- * others which should be contiguous.
- */
-static inline bool blk_discard_mergable(struct request *req)
-{
- if (req_op(req) == REQ_OP_DISCARD &&
- queue_max_discard_segments(req->q) > 1)
- return true;
- return false;
-}
-
static enum elv_merge blk_try_req_merge(struct request *req,
struct request *next)
{
diff --git a/block/elevator.c b/block/elevator.c
index 52ada14cfe45..a5fe2615ec0f 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -336,6 +336,9 @@ enum elv_merge elv_merge(struct request_queue *q, struct request **req,
__rq = elv_rqhash_find(q, bio->bi_iter.bi_sector);
if (__rq && elv_bio_merge_ok(__rq, bio)) {
*req = __rq;
+
+ if (blk_discard_mergable(__rq))
+ return ELEVATOR_DISCARD_MERGE;
return ELEVATOR_BACK_MERGE;
}
diff --git a/block/mq-deadline-main.c b/block/mq-deadline-main.c
index 6f612e6dc82b..294be0c0db65 100644
--- a/block/mq-deadline-main.c
+++ b/block/mq-deadline-main.c
@@ -677,6 +677,8 @@ static int dd_request_merge(struct request_queue *q, struct request **rq,
if (elv_bio_merge_ok(__rq, bio)) {
*rq = __rq;
+ if (blk_discard_mergable(__rq))
+ return ELEVATOR_DISCARD_MERGE;
return ELEVATOR_FRONT_MERGE;
}
}
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 3177181c4326..87f00292fd7a 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1521,6 +1521,22 @@ static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector
return offset << SECTOR_SHIFT;
}
+/*
+ * Two cases of handling DISCARD merge:
+ * If max_discard_segments > 1, the driver takes every bio
+ * as a range and send them to controller together. The ranges
+ * needn't to be contiguous.
+ * Otherwise, the bios/requests will be handled as same as
+ * others which should be contiguous.
+ */
+static inline bool blk_discard_mergable(struct request *req)
+{
+ if (req_op(req) == REQ_OP_DISCARD &&
+ queue_max_discard_segments(req->q) > 1)
+ return true;
+ return false;
+}
+
static inline int bdev_discard_alignment(struct block_device *bdev)
{
struct request_queue *q = bdev_get_queue(bdev);
--
2.31.1
^ permalink raw reply related [flat|nested] 11+ messages in thread
* Re: [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible
2021-07-29 3:42 [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible Ming Lei
@ 2021-07-29 7:17 ` Oleksandr Natalenko
2021-07-29 9:18 ` Ming Lei
2021-08-05 3:35 ` Ming Lei
1 sibling, 1 reply; 11+ messages in thread
From: Oleksandr Natalenko @ 2021-07-29 7:17 UTC (permalink / raw
To: Jens Axboe, Ming Lei; +Cc: linux-block, Christoph Hellwig, Ming Lei
On čtvrtek 29. července 2021 5:42:26 CEST Ming Lei wrote:
> When merging one bio to request, if they are discard IO and the queue
> supports multi-range discard, we need to return ELEVATOR_DISCARD_MERGE
> because both block core and related drivers(nvme, virtio-blk) doesn't
> handle mixed discard io merge(traditional IO merge together with
> discard merge) well.
>
> Fix the issue by returning ELEVATOR_DISCARD_MERGE in this situation,
> so both blk-mq and drivers just need to handle multi-range discard.
>
> Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
> Signed-off-by: Ming Lei <ming.lei@redhat.com>
> ---
> block/bfq-iosched.c | 3 +++
> block/blk-merge.c | 16 ----------------
> block/elevator.c | 3 +++
> block/mq-deadline-main.c | 2 ++
> include/linux/blkdev.h | 16 ++++++++++++++++
> 5 files changed, 24 insertions(+), 16 deletions(-)
>
> diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
> index 727955918563..673a634eadd9 100644
> --- a/block/bfq-iosched.c
> +++ b/block/bfq-iosched.c
> @@ -2361,6 +2361,9 @@ static int bfq_request_merge(struct request_queue *q,
> struct request **req, __rq = bfq_find_rq_fmerge(bfqd, bio, q);
> if (__rq && elv_bio_merge_ok(__rq, bio)) {
> *req = __rq;
> +
> + if (blk_discard_mergable(__rq))
> + return ELEVATOR_DISCARD_MERGE;
> return ELEVATOR_FRONT_MERGE;
> }
>
> diff --git a/block/blk-merge.c b/block/blk-merge.c
> index a11b3b53717e..f8707ff7e2fc 100644
> --- a/block/blk-merge.c
> +++ b/block/blk-merge.c
> @@ -705,22 +705,6 @@ static void blk_account_io_merge_request(struct request
> *req) }
> }
>
> -/*
> - * Two cases of handling DISCARD merge:
> - * If max_discard_segments > 1, the driver takes every bio
> - * as a range and send them to controller together. The ranges
> - * needn't to be contiguous.
> - * Otherwise, the bios/requests will be handled as same as
> - * others which should be contiguous.
> - */
> -static inline bool blk_discard_mergable(struct request *req)
> -{
> - if (req_op(req) == REQ_OP_DISCARD &&
> - queue_max_discard_segments(req->q) > 1)
> - return true;
> - return false;
> -}
> -
> static enum elv_merge blk_try_req_merge(struct request *req,
> struct request *next)
> {
> diff --git a/block/elevator.c b/block/elevator.c
> index 52ada14cfe45..a5fe2615ec0f 100644
> --- a/block/elevator.c
> +++ b/block/elevator.c
> @@ -336,6 +336,9 @@ enum elv_merge elv_merge(struct request_queue *q, struct
> request **req, __rq = elv_rqhash_find(q, bio->bi_iter.bi_sector);
> if (__rq && elv_bio_merge_ok(__rq, bio)) {
> *req = __rq;
> +
> + if (blk_discard_mergable(__rq))
> + return ELEVATOR_DISCARD_MERGE;
> return ELEVATOR_BACK_MERGE;
> }
>
> diff --git a/block/mq-deadline-main.c b/block/mq-deadline-main.c
> index 6f612e6dc82b..294be0c0db65 100644
> --- a/block/mq-deadline-main.c
> +++ b/block/mq-deadline-main.c
> @@ -677,6 +677,8 @@ static int dd_request_merge(struct request_queue *q,
> struct request **rq,
>
> if (elv_bio_merge_ok(__rq, bio)) {
> *rq = __rq;
> + if (blk_discard_mergable(__rq))
> + return ELEVATOR_DISCARD_MERGE;
> return ELEVATOR_FRONT_MERGE;
> }
> }
> diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
> index 3177181c4326..87f00292fd7a 100644
> --- a/include/linux/blkdev.h
> +++ b/include/linux/blkdev.h
> @@ -1521,6 +1521,22 @@ static inline int
> queue_limit_discard_alignment(struct queue_limits *lim, sector return
> offset << SECTOR_SHIFT;
> }
>
> +/*
> + * Two cases of handling DISCARD merge:
> + * If max_discard_segments > 1, the driver takes every bio
> + * as a range and send them to controller together. The ranges
> + * needn't to be contiguous.
> + * Otherwise, the bios/requests will be handled as same as
> + * others which should be contiguous.
> + */
> +static inline bool blk_discard_mergable(struct request *req)
> +{
> + if (req_op(req) == REQ_OP_DISCARD &&
> + queue_max_discard_segments(req->q) > 1)
> + return true;
> + return false;
> +}
> +
> static inline int bdev_discard_alignment(struct block_device *bdev)
> {
> struct request_queue *q = bdev_get_queue(bdev);
Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Also,
Fixes: ?
and possibly:
CC: stable@ # v5.x?
Thanks.
--
Oleksandr Natalenko (post-factum)
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible
2021-07-29 7:17 ` Oleksandr Natalenko
@ 2021-07-29 9:18 ` Ming Lei
0 siblings, 0 replies; 11+ messages in thread
From: Ming Lei @ 2021-07-29 9:18 UTC (permalink / raw
To: Oleksandr Natalenko; +Cc: Jens Axboe, linux-block, Christoph Hellwig
On Thu, Jul 29, 2021 at 09:17:28AM +0200, Oleksandr Natalenko wrote:
> On čtvrtek 29. července 2021 5:42:26 CEST Ming Lei wrote:
> > When merging one bio to request, if they are discard IO and the queue
> > supports multi-range discard, we need to return ELEVATOR_DISCARD_MERGE
> > because both block core and related drivers(nvme, virtio-blk) doesn't
> > handle mixed discard io merge(traditional IO merge together with
> > discard merge) well.
> >
> > Fix the issue by returning ELEVATOR_DISCARD_MERGE in this situation,
> > so both blk-mq and drivers just need to handle multi-range discard.
> >
> > Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
> > Signed-off-by: Ming Lei <ming.lei@redhat.com>
> > ---
> > block/bfq-iosched.c | 3 +++
> > block/blk-merge.c | 16 ----------------
> > block/elevator.c | 3 +++
> > block/mq-deadline-main.c | 2 ++
> > include/linux/blkdev.h | 16 ++++++++++++++++
> > 5 files changed, 24 insertions(+), 16 deletions(-)
> >
> > diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
> > index 727955918563..673a634eadd9 100644
> > --- a/block/bfq-iosched.c
> > +++ b/block/bfq-iosched.c
> > @@ -2361,6 +2361,9 @@ static int bfq_request_merge(struct request_queue *q,
> > struct request **req, __rq = bfq_find_rq_fmerge(bfqd, bio, q);
> > if (__rq && elv_bio_merge_ok(__rq, bio)) {
> > *req = __rq;
> > +
> > + if (blk_discard_mergable(__rq))
> > + return ELEVATOR_DISCARD_MERGE;
> > return ELEVATOR_FRONT_MERGE;
> > }
> >
> > diff --git a/block/blk-merge.c b/block/blk-merge.c
> > index a11b3b53717e..f8707ff7e2fc 100644
> > --- a/block/blk-merge.c
> > +++ b/block/blk-merge.c
> > @@ -705,22 +705,6 @@ static void blk_account_io_merge_request(struct request
> > *req) }
> > }
> >
> > -/*
> > - * Two cases of handling DISCARD merge:
> > - * If max_discard_segments > 1, the driver takes every bio
> > - * as a range and send them to controller together. The ranges
> > - * needn't to be contiguous.
> > - * Otherwise, the bios/requests will be handled as same as
> > - * others which should be contiguous.
> > - */
> > -static inline bool blk_discard_mergable(struct request *req)
> > -{
> > - if (req_op(req) == REQ_OP_DISCARD &&
> > - queue_max_discard_segments(req->q) > 1)
> > - return true;
> > - return false;
> > -}
> > -
> > static enum elv_merge blk_try_req_merge(struct request *req,
> > struct request *next)
> > {
> > diff --git a/block/elevator.c b/block/elevator.c
> > index 52ada14cfe45..a5fe2615ec0f 100644
> > --- a/block/elevator.c
> > +++ b/block/elevator.c
> > @@ -336,6 +336,9 @@ enum elv_merge elv_merge(struct request_queue *q, struct
> > request **req, __rq = elv_rqhash_find(q, bio->bi_iter.bi_sector);
> > if (__rq && elv_bio_merge_ok(__rq, bio)) {
> > *req = __rq;
> > +
> > + if (blk_discard_mergable(__rq))
> > + return ELEVATOR_DISCARD_MERGE;
> > return ELEVATOR_BACK_MERGE;
> > }
> >
> > diff --git a/block/mq-deadline-main.c b/block/mq-deadline-main.c
> > index 6f612e6dc82b..294be0c0db65 100644
> > --- a/block/mq-deadline-main.c
> > +++ b/block/mq-deadline-main.c
> > @@ -677,6 +677,8 @@ static int dd_request_merge(struct request_queue *q,
> > struct request **rq,
> >
> > if (elv_bio_merge_ok(__rq, bio)) {
> > *rq = __rq;
> > + if (blk_discard_mergable(__rq))
> > + return ELEVATOR_DISCARD_MERGE;
> > return ELEVATOR_FRONT_MERGE;
> > }
> > }
> > diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
> > index 3177181c4326..87f00292fd7a 100644
> > --- a/include/linux/blkdev.h
> > +++ b/include/linux/blkdev.h
> > @@ -1521,6 +1521,22 @@ static inline int
> > queue_limit_discard_alignment(struct queue_limits *lim, sector return
> > offset << SECTOR_SHIFT;
> > }
> >
> > +/*
> > + * Two cases of handling DISCARD merge:
> > + * If max_discard_segments > 1, the driver takes every bio
> > + * as a range and send them to controller together. The ranges
> > + * needn't to be contiguous.
> > + * Otherwise, the bios/requests will be handled as same as
> > + * others which should be contiguous.
> > + */
> > +static inline bool blk_discard_mergable(struct request *req)
> > +{
> > + if (req_op(req) == REQ_OP_DISCARD &&
> > + queue_max_discard_segments(req->q) > 1)
> > + return true;
> > + return false;
> > +}
> > +
> > static inline int bdev_discard_alignment(struct block_device *bdev)
> > {
> > struct request_queue *q = bdev_get_queue(bdev);
>
> Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
>
> Also,
>
> Fixes: ?
Fixes: 2705dfb20947 ("block: fix discard request merge")
In theory, the fixes should be 1e739730c5b9 ("block: optionally merge
discontiguous discard bios into a single request"). The reason why it
isn't be triggered before 2705dfb20947 is that the nr_segment of each
bio is exactly 1 for multi-range device.
>
> and possibly:
>
> CC: stable@ # v5.x?
CC: stable@ # v5.13
Thanks,
Ming
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible
2021-07-29 3:42 [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible Ming Lei
2021-07-29 7:17 ` Oleksandr Natalenko
@ 2021-08-05 3:35 ` Ming Lei
2021-08-09 20:30 ` Oleksandr Natalenko
2021-08-09 20:36 ` Jens Axboe
1 sibling, 2 replies; 11+ messages in thread
From: Ming Lei @ 2021-08-05 3:35 UTC (permalink / raw
To: Jens Axboe; +Cc: linux-block, Christoph Hellwig, Oleksandr Natalenko
On Thu, Jul 29, 2021 at 11:42:26AM +0800, Ming Lei wrote:
> When merging one bio to request, if they are discard IO and the queue
> supports multi-range discard, we need to return ELEVATOR_DISCARD_MERGE
> because both block core and related drivers(nvme, virtio-blk) doesn't
> handle mixed discard io merge(traditional IO merge together with
> discard merge) well.
>
> Fix the issue by returning ELEVATOR_DISCARD_MERGE in this situation,
> so both blk-mq and drivers just need to handle multi-range discard.
>
> Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
> Signed-off-by: Ming Lei <ming.lei@redhat.com>
Hello Jens and Guys,
Ping...
Thanks,
Ming
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible
2021-08-05 3:35 ` Ming Lei
@ 2021-08-09 20:30 ` Oleksandr Natalenko
2021-08-09 20:36 ` Jens Axboe
1 sibling, 0 replies; 11+ messages in thread
From: Oleksandr Natalenko @ 2021-08-09 20:30 UTC (permalink / raw
To: Jens Axboe, Ming Lei; +Cc: linux-block, Christoph Hellwig
Hello.
On čtvrtek 5. srpna 2021 5:35:00 CEST Ming Lei wrote:
> On Thu, Jul 29, 2021 at 11:42:26AM +0800, Ming Lei wrote:
> > When merging one bio to request, if they are discard IO and the queue
> > supports multi-range discard, we need to return ELEVATOR_DISCARD_MERGE
> > because both block core and related drivers(nvme, virtio-blk) doesn't
> > handle mixed discard io merge(traditional IO merge together with
> > discard merge) well.
> >
> > Fix the issue by returning ELEVATOR_DISCARD_MERGE in this situation,
> > so both blk-mq and drivers just need to handle multi-range discard.
> >
> > Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
> > Signed-off-by: Ming Lei <ming.lei@redhat.com>
>
> Hello Jens and Guys,
>
> Ping...
Ping v2 here… I'm still running a couple of affected systems with this patch,
and the issue doesn't pop up any more.
Thanks.
--
Oleksandr Natalenko (post-factum)
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible
2021-08-05 3:35 ` Ming Lei
2021-08-09 20:30 ` Oleksandr Natalenko
@ 2021-08-09 20:36 ` Jens Axboe
2021-08-10 2:47 ` Ming Lei
1 sibling, 1 reply; 11+ messages in thread
From: Jens Axboe @ 2021-08-09 20:36 UTC (permalink / raw
To: Ming Lei; +Cc: linux-block, Christoph Hellwig, Oleksandr Natalenko
On 8/4/21 9:35 PM, Ming Lei wrote:
> On Thu, Jul 29, 2021 at 11:42:26AM +0800, Ming Lei wrote:
>> When merging one bio to request, if they are discard IO and the queue
>> supports multi-range discard, we need to return ELEVATOR_DISCARD_MERGE
>> because both block core and related drivers(nvme, virtio-blk) doesn't
>> handle mixed discard io merge(traditional IO merge together with
>> discard merge) well.
>>
>> Fix the issue by returning ELEVATOR_DISCARD_MERGE in this situation,
>> so both blk-mq and drivers just need to handle multi-range discard.
>>
>> Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
>> Signed-off-by: Ming Lei <ming.lei@redhat.com>
>
> Hello Jens and Guys,
>
> Ping...
Since this isn't a new regression this release and since this kind
of change always makes me a bit nervous, any objections to queueing
it up for 5.15 with the stable/fixes tags?
--
Jens Axboe
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible
2021-08-09 20:36 ` Jens Axboe
@ 2021-08-10 2:47 ` Ming Lei
2021-08-10 2:55 ` Jens Axboe
0 siblings, 1 reply; 11+ messages in thread
From: Ming Lei @ 2021-08-10 2:47 UTC (permalink / raw
To: Jens Axboe; +Cc: linux-block, Christoph Hellwig, Oleksandr Natalenko
On Mon, Aug 09, 2021 at 02:36:25PM -0600, Jens Axboe wrote:
> On 8/4/21 9:35 PM, Ming Lei wrote:
> > On Thu, Jul 29, 2021 at 11:42:26AM +0800, Ming Lei wrote:
> >> When merging one bio to request, if they are discard IO and the queue
> >> supports multi-range discard, we need to return ELEVATOR_DISCARD_MERGE
> >> because both block core and related drivers(nvme, virtio-blk) doesn't
> >> handle mixed discard io merge(traditional IO merge together with
> >> discard merge) well.
> >>
> >> Fix the issue by returning ELEVATOR_DISCARD_MERGE in this situation,
> >> so both blk-mq and drivers just need to handle multi-range discard.
> >>
> >> Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
> >> Signed-off-by: Ming Lei <ming.lei@redhat.com>
> >
> > Hello Jens and Guys,
> >
> > Ping...
>
> Since this isn't a new regression this release and since this kind
> of change always makes me a bit nervous, any objections to queueing
> it up for 5.15 with the stable/fixes tags?
Fine, will post a new version with fixes tag.
--
Ming
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible
2021-08-10 2:47 ` Ming Lei
@ 2021-08-10 2:55 ` Jens Axboe
2021-08-10 3:28 ` Ming Lei
0 siblings, 1 reply; 11+ messages in thread
From: Jens Axboe @ 2021-08-10 2:55 UTC (permalink / raw
To: Ming Lei; +Cc: linux-block, Christoph Hellwig, Oleksandr Natalenko
On 8/9/21 8:47 PM, Ming Lei wrote:
> On Mon, Aug 09, 2021 at 02:36:25PM -0600, Jens Axboe wrote:
>> On 8/4/21 9:35 PM, Ming Lei wrote:
>>> On Thu, Jul 29, 2021 at 11:42:26AM +0800, Ming Lei wrote:
>>>> When merging one bio to request, if they are discard IO and the queue
>>>> supports multi-range discard, we need to return ELEVATOR_DISCARD_MERGE
>>>> because both block core and related drivers(nvme, virtio-blk) doesn't
>>>> handle mixed discard io merge(traditional IO merge together with
>>>> discard merge) well.
>>>>
>>>> Fix the issue by returning ELEVATOR_DISCARD_MERGE in this situation,
>>>> so both blk-mq and drivers just need to handle multi-range discard.
>>>>
>>>> Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
>>>> Signed-off-by: Ming Lei <ming.lei@redhat.com>
>>>
>>> Hello Jens and Guys,
>>>
>>> Ping...
>>
>> Since this isn't a new regression this release and since this kind
>> of change always makes me a bit nervous, any objections to queueing
>> it up for 5.15 with the stable/fixes tags?
>
> Fine, will post a new version with fixes tag.
Sorry if I wasn't clear, I mean are you fine with queueing this for
5.15? I already did add the fixes tag.
--
Jens Axboe
^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible
2021-08-10 2:55 ` Jens Axboe
@ 2021-08-10 3:28 ` Ming Lei
0 siblings, 0 replies; 11+ messages in thread
From: Ming Lei @ 2021-08-10 3:28 UTC (permalink / raw
To: Jens Axboe; +Cc: linux-block, Christoph Hellwig, Oleksandr Natalenko
On Mon, Aug 09, 2021 at 08:55:16PM -0600, Jens Axboe wrote:
> On 8/9/21 8:47 PM, Ming Lei wrote:
> > On Mon, Aug 09, 2021 at 02:36:25PM -0600, Jens Axboe wrote:
> >> On 8/4/21 9:35 PM, Ming Lei wrote:
> >>> On Thu, Jul 29, 2021 at 11:42:26AM +0800, Ming Lei wrote:
> >>>> When merging one bio to request, if they are discard IO and the queue
> >>>> supports multi-range discard, we need to return ELEVATOR_DISCARD_MERGE
> >>>> because both block core and related drivers(nvme, virtio-blk) doesn't
> >>>> handle mixed discard io merge(traditional IO merge together with
> >>>> discard merge) well.
> >>>>
> >>>> Fix the issue by returning ELEVATOR_DISCARD_MERGE in this situation,
> >>>> so both blk-mq and drivers just need to handle multi-range discard.
> >>>>
> >>>> Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
> >>>> Signed-off-by: Ming Lei <ming.lei@redhat.com>
> >>>
> >>> Hello Jens and Guys,
> >>>
> >>> Ping...
> >>
> >> Since this isn't a new regression this release and since this kind
> >> of change always makes me a bit nervous, any objections to queueing
> >> it up for 5.15 with the stable/fixes tags?
> >
> > Fine, will post a new version with fixes tag.
>
> Sorry if I wasn't clear, I mean are you fine with queueing this for
> 5.15? I already did add the fixes tag.
Sorry for misunderstanding your point, yeah, I am fine with queuing in
5.15.
Thanks,
Ming
^ permalink raw reply [flat|nested] 11+ messages in thread
* [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible
@ 2022-05-21 18:56 Gwendal Grignou
2022-05-23 15:13 ` Greg KH
0 siblings, 1 reply; 11+ messages in thread
From: Gwendal Grignou @ 2022-05-21 18:56 UTC (permalink / raw
To: paolo.valente, axboe, gregkh, xieyongji, ming.lei
Cc: linux-block, Oleksandr Natalenko, stable, Gwendal Grignou
From: Ming Lei <ming.lei@redhat.com>
When merging one bio to request, if they are discard IO and the queue
supports multi-range discard, we need to return ELEVATOR_DISCARD_MERGE
because both block core and related drivers(nvme, virtio-blk) doesn't
handle mixed discard io merge(traditional IO merge together with
discard merge) well.
Fix the issue by returning ELEVATOR_DISCARD_MERGE in this situation,
so both blk-mq and drivers just need to handle multi-range discard.
Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Signed-off-by: Ming Lei <ming.lei@redhat.com>
Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
Fixes: 2705dfb20947 ("block: fix discard request merge")
Link: https://lore.kernel.org/r/20210729034226.1591070-1-ming.lei@redhat.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
commit 866663b7b52d2 upstream.
Similar to commit 87aa69aa10b42 ("block: return ELEVATOR_DISCARD_MERGE if possible")
in 5.10 kernel.
Conflicts:
block/blk-merge.c: function at a different place.
block/mq-deadline-main.c: not in 5.4, use mq-deadline.c instead.
Cc: <stable@vger.kernel.org> # 5.4.y
Signed-off-by: Gwendal Grignou <gwendal@chromium.org>
---
block/bfq-iosched.c | 3 +++
block/blk-merge.c | 15 ---------------
block/elevator.c | 3 +++
block/mq-deadline.c | 2 ++
include/linux/blkdev.h | 16 ++++++++++++++++
5 files changed, 24 insertions(+), 15 deletions(-)
diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c
index 1d443d17cf7c5..d46806182b051 100644
--- a/block/bfq-iosched.c
+++ b/block/bfq-iosched.c
@@ -2251,6 +2251,9 @@ static int bfq_request_merge(struct request_queue *q, struct request **req,
__rq = bfq_find_rq_fmerge(bfqd, bio, q);
if (__rq && elv_bio_merge_ok(__rq, bio)) {
*req = __rq;
+
+ if (blk_discard_mergable(__rq))
+ return ELEVATOR_DISCARD_MERGE;
return ELEVATOR_FRONT_MERGE;
}
diff --git a/block/blk-merge.c b/block/blk-merge.c
index a62692d135660..5219064cd72bb 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -721,21 +721,6 @@ static void blk_account_io_merge(struct request *req)
part_stat_unlock();
}
}
-/*
- * Two cases of handling DISCARD merge:
- * If max_discard_segments > 1, the driver takes every bio
- * as a range and send them to controller together. The ranges
- * needn't to be contiguous.
- * Otherwise, the bios/requests will be handled as same as
- * others which should be contiguous.
- */
-static inline bool blk_discard_mergable(struct request *req)
-{
- if (req_op(req) == REQ_OP_DISCARD &&
- queue_max_discard_segments(req->q) > 1)
- return true;
- return false;
-}
static enum elv_merge blk_try_req_merge(struct request *req,
struct request *next)
diff --git a/block/elevator.c b/block/elevator.c
index 78805c74ea8a4..3ba826230c578 100644
--- a/block/elevator.c
+++ b/block/elevator.c
@@ -337,6 +337,9 @@ enum elv_merge elv_merge(struct request_queue *q, struct request **req,
__rq = elv_rqhash_find(q, bio->bi_iter.bi_sector);
if (__rq && elv_bio_merge_ok(__rq, bio)) {
*req = __rq;
+
+ if (blk_discard_mergable(__rq))
+ return ELEVATOR_DISCARD_MERGE;
return ELEVATOR_BACK_MERGE;
}
diff --git a/block/mq-deadline.c b/block/mq-deadline.c
index 19c6922e85f1b..6d6dda5cfffa3 100644
--- a/block/mq-deadline.c
+++ b/block/mq-deadline.c
@@ -452,6 +452,8 @@ static int dd_request_merge(struct request_queue *q, struct request **rq,
if (elv_bio_merge_ok(__rq, bio)) {
*rq = __rq;
+ if (blk_discard_mergable(__rq))
+ return ELEVATOR_DISCARD_MERGE;
return ELEVATOR_FRONT_MERGE;
}
}
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 8cc766743270f..308c2d8cdca19 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1409,6 +1409,22 @@ static inline int queue_limit_discard_alignment(struct queue_limits *lim, sector
return offset << SECTOR_SHIFT;
}
+/*
+ * Two cases of handling DISCARD merge:
+ * If max_discard_segments > 1, the driver takes every bio
+ * as a range and send them to controller together. The ranges
+ * needn't to be contiguous.
+ * Otherwise, the bios/requests will be handled as same as
+ * others which should be contiguous.
+ */
+static inline bool blk_discard_mergable(struct request *req)
+{
+ if (req_op(req) == REQ_OP_DISCARD &&
+ queue_max_discard_segments(req->q) > 1)
+ return true;
+ return false;
+}
+
static inline int bdev_discard_alignment(struct block_device *bdev)
{
struct request_queue *q = bdev_get_queue(bdev);
--
2.36.1.124.g0e6072fb45-goog
^ permalink raw reply related [flat|nested] 11+ messages in thread
* Re: [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible
2022-05-21 18:56 Gwendal Grignou
@ 2022-05-23 15:13 ` Greg KH
0 siblings, 0 replies; 11+ messages in thread
From: Greg KH @ 2022-05-23 15:13 UTC (permalink / raw
To: Gwendal Grignou
Cc: paolo.valente, axboe, xieyongji, ming.lei, linux-block,
Oleksandr Natalenko, stable
On Sat, May 21, 2022 at 11:56:26AM -0700, Gwendal Grignou wrote:
> From: Ming Lei <ming.lei@redhat.com>
>
> When merging one bio to request, if they are discard IO and the queue
> supports multi-range discard, we need to return ELEVATOR_DISCARD_MERGE
> because both block core and related drivers(nvme, virtio-blk) doesn't
> handle mixed discard io merge(traditional IO merge together with
> discard merge) well.
>
> Fix the issue by returning ELEVATOR_DISCARD_MERGE in this situation,
> so both blk-mq and drivers just need to handle multi-range discard.
>
> Reported-by: Oleksandr Natalenko <oleksandr@natalenko.name>
> Signed-off-by: Ming Lei <ming.lei@redhat.com>
> Tested-by: Oleksandr Natalenko <oleksandr@natalenko.name>
> Fixes: 2705dfb20947 ("block: fix discard request merge")
> Link: https://lore.kernel.org/r/20210729034226.1591070-1-ming.lei@redhat.com
> Signed-off-by: Jens Axboe <axboe@kernel.dk>
>
> commit 866663b7b52d2 upstream.
>
> Similar to commit 87aa69aa10b42 ("block: return ELEVATOR_DISCARD_MERGE if possible")
> in 5.10 kernel.
>
> Conflicts:
> block/blk-merge.c: function at a different place.
> block/mq-deadline-main.c: not in 5.4, use mq-deadline.c instead.
>
> Cc: <stable@vger.kernel.org> # 5.4.y
> Signed-off-by: Gwendal Grignou <gwendal@chromium.org>
Now queued up, thanks.
greg k-h
^ permalink raw reply [flat|nested] 11+ messages in thread
end of thread, other threads:[~2022-05-23 15:13 UTC | newest]
Thread overview: 11+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2021-07-29 3:42 [PATCH] block: return ELEVATOR_DISCARD_MERGE if possible Ming Lei
2021-07-29 7:17 ` Oleksandr Natalenko
2021-07-29 9:18 ` Ming Lei
2021-08-05 3:35 ` Ming Lei
2021-08-09 20:30 ` Oleksandr Natalenko
2021-08-09 20:36 ` Jens Axboe
2021-08-10 2:47 ` Ming Lei
2021-08-10 2:55 ` Jens Axboe
2021-08-10 3:28 ` Ming Lei
-- strict thread matches above, loose matches on Subject: below --
2022-05-21 18:56 Gwendal Grignou
2022-05-23 15:13 ` Greg KH
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.