bcache: fix inaccurate io state for detached bcache devices

Commit Message

[e-mail] Jan. 9, 2018, 2:27 a.m. UTC

From: Tang Junhui <[e-mail]>

When we run IO in a detached device,  and run iostat to shows IO status,
normally it will show like bellow (Omitted some fields):
Device: ... avgrq-sz avgqu-sz   await r_await w_await  svctm  %util
sdd        ... 15.89     0.53    1.82    0.20    2.23   1.81  52.30
bcache0    ... 15.89   115.42    0.00    0.00    0.00   2.40  69.60
but after IO stopped, there are still very big avgqu-sz and %util 
values as bellow:
Device: ... avgrq-sz avgqu-sz   await r_await w_await  svctm  %util
bcache0   ...      0   5326.32    0.00    0.00    0.00   0.00 100.10

The reason for this issue is that, only generic_start_io_acct() called
and no generic_end_io_acct() called for detached device in
cached_dev_make_request(). See the code:
//start generic_start_io_acct()
generic_start_io_acct(q, rw, bio_sectors(bio), &d->disk->part0);
if (cached_dev_get(dc)) {
	//will callback generic_end_io_acct()
}
else {
	//will not call generic_end_io_acct()
}

This patch calls generic_end_io_acct() in the end of IO for detached
devices, so we can show IO state correctly.

Signed-off-by: Tang Junhui <[e-mail]>
---
 drivers/md/bcache/request.c | 58 +++++++++++++++++++++++++++++++++++++++------
 1 file changed, 51 insertions(+), 7 deletions(-)

Comments

Coly Li Jan. 9, 2018, 2:35 a.m. UTC | #1

On 09/01/2018 10:27 AM, [e-mail] wrote:
> From: Tang Junhui <[e-mail]>
> 
> When we run IO in a detached device,  and run iostat to shows IO status,
> normally it will show like bellow (Omitted some fields):
> Device: ... avgrq-sz avgqu-sz   await r_await w_await  svctm  %util
> sdd        ... 15.89     0.53    1.82    0.20    2.23   1.81  52.30
> bcache0    ... 15.89   115.42    0.00    0.00    0.00   2.40  69.60
> but after IO stopped, there are still very big avgqu-sz and %util 
> values as bellow:
> Device: ... avgrq-sz avgqu-sz   await r_await w_await  svctm  %util
> bcache0   ...      0   5326.32    0.00    0.00    0.00   0.00 100.10
> 
> The reason for this issue is that, only generic_start_io_acct() called
> and no generic_end_io_acct() called for detached device in
> cached_dev_make_request(). See the code:
> //start generic_start_io_acct()
> generic_start_io_acct(q, rw, bio_sectors(bio), &d->disk->part0);
> if (cached_dev_get(dc)) {
> 	//will callback generic_end_io_acct()
> }
> else {
> 	//will not call generic_end_io_acct()
> }
> 
> This patch calls generic_end_io_acct() in the end of IO for detached
> devices, so we can show IO state correctly.
> 
> Signed-off-by: Tang Junhui <[e-mail]>

It looks good to me. Reviewed-by: Coly Li <[e-mail]>

I also modify the backing device bi_end_io to catch I/O status in my
backing device failure patches, so I am thinking of pick this patch into
my device failure patch set, then Mike does not need to merge the conflict.

Coly Li

[code snipped]

Patch

diff --git a/drivers/md/bcache/request.c b/drivers/md/bcache/request.c
index 3475d66..28c3740 100644
--- a/drivers/md/bcache/request.c
+++ b/drivers/md/bcache/request.c
@@ -945,6 +945,55 @@  static void cached_dev_nodata(struct closure *cl)
 	continue_at(cl, cached_dev_bio_complete, NULL);
 }
 
+struct detached_dev_io_private {
+	struct bcache_device	*d;
+	unsigned long		start_time;
+	bio_end_io_t		*bi_end_io;
+	void			*bi_private;
+};
+
+static void detatched_dev_end_io(struct bio *bio)
+{
+	struct detached_dev_io_private *ddip;
+
+	ddip = bio->bi_private;
+	bio->bi_end_io = ddip->bi_end_io;
+	bio->bi_private = ddip->bi_private;
+
+	generic_end_io_acct(ddip->d->disk->queue,
+			    bio_data_dir(bio),
+			    &ddip->d->disk->part0, ddip->start_time);
+
+	kfree(ddip);
+
+	bio->bi_end_io(bio);
+}
+
+static void detached_dev_do_request(struct bcache_device *d, struct bio *bio)
+{
+	struct detached_dev_io_private *ddip;
+	struct cached_dev *dc = container_of(d, struct cached_dev, disk);
+
+	/*
+	 * no need to call closure_get(&dc->disk.cl),
+	 * because upper layer had already opened bcache device,
+	 * which would call closure_get(&dc->disk.cl)
+	 */
+	ddip = kzalloc(sizeof(struct detached_dev_io_private), GFP_KERNEL);
+	ddip->d = d;
+	ddip->start_time = jiffies;
+	ddip->bi_end_io = bio->bi_end_io;
+	ddip->bi_private = bio->bi_private;
+	bio->bi_end_io = detatched_dev_end_io;
+	bio->bi_private = ddip;
+
+	if ((bio_op(bio) == REQ_OP_DISCARD) &&
+	    !blk_queue_discard(bdev_get_queue(dc->bdev)))
+		bio->bi_end_io(bio);
+	else
+		generic_make_request(bio);
+}
+
 /* Cached devices - read & write stuff */
 
 static blk_qc_t cached_dev_make_request(struct request_queue *q,
@@ -980,13 +1029,8 @@  static blk_qc_t cached_dev_make_request(struct request_queue *q,
 			else
 				cached_dev_read(dc, s);
 		}
-	} else {
-		if ((bio_op(bio) == REQ_OP_DISCARD) &&
-		    !blk_queue_discard(bdev_get_queue(dc->bdev)))
-			bio_endio(bio);
-		else
-			generic_make_request(bio);
-	}
+	} else
+		detached_dev_do_request(d, bio);
 
 	return BLK_QC_T_NONE;
 }