* [RFC PATCH 1/2] block: add lld busy state exporting interface
2008-09-25 15:53 [RFC PATCH 0/2] Export lld busy state for request stacking drivers Kiyoshi Ueda
@ 2008-09-25 15:54 ` Kiyoshi Ueda
2008-09-25 15:55 ` [RFC PATCH 2/2] scsi: export busy state via q->lld_busy_fn() Kiyoshi Ueda
1 sibling, 0 replies; 3+ messages in thread
From: Kiyoshi Ueda @ 2008-09-25 15:54 UTC (permalink / raw)
To: akpm, jens.axboe, James.Bottomley
Cc: linux-kernel, linux-scsi, dm-devel, j-nomura, k-ueda
This patch adds an new interface, blk_lld_busy(), to check lld's
busy state from the block layer.
blk_lld_busy() calls down into low-level drivers for the checking
if the drivers set q->lld_busy_fn() using blk_queue_lld_busy().
This resolves a performance problem on request stacking devices below.
Some drivers like scsi mid layer stop dispatching request when
they detect busy state on its low-level device like host/target/device.
It allows other requests to stay in the I/O scheduler's queue
for a chance of merging.
Request stacking drivers like request-based dm should follow
the same logic.
However, there is no generic interface for the stacked device
to check if the underlying device(s) are busy.
If the request stacking driver dispatches and submits requests to
the busy underlying device, the requests will stay in
the underlying device's queue without a chance of merging.
This causes performance problem on burst I/O load.
With this patch, busy state of the underlying device is exported
via q->lld_busy_fn(). So the request stacking driver can check it
and stop dispatching requests if busy.
The underlying device driver must return the busy state appropriately:
1: when the device driver can't process requests immediately.
0: when the device driver can process requests immediately,
including abnormal situations where the device driver needs
to kill all requests.
Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Jens Axboe <jens.axboe@oracle.com>
---
block/blk-core.c | 25 +++++++++++++++++++++++++
block/blk-settings.c | 6 ++++++
include/linux/blkdev.h | 4 ++++
3 files changed, 35 insertions(+)
Index: linux-2.6-block/block/blk-settings.c
===================================================================
--- linux-2.6-block.orig/block/blk-settings.c
+++ linux-2.6-block/block/blk-settings.c
@@ -89,6 +89,12 @@ void blk_queue_rq_timed_out(struct reque
}
EXPORT_SYMBOL_GPL(blk_queue_rq_timed_out);
+void blk_queue_lld_busy(struct request_queue *q, lld_busy_fn *fn)
+{
+ q->lld_busy_fn = fn;
+}
+EXPORT_SYMBOL_GPL(blk_queue_lld_busy);
+
/**
* blk_queue_make_request - define an alternate make_request function for a device
* @q: the request queue for the device to be affected
Index: linux-2.6-block/include/linux/blkdev.h
===================================================================
--- linux-2.6-block.orig/include/linux/blkdev.h
+++ linux-2.6-block/include/linux/blkdev.h
@@ -269,6 +269,7 @@ typedef int (merge_bvec_fn) (struct requ
typedef void (prepare_flush_fn) (struct request_queue *, struct request *);
typedef void (softirq_done_fn)(struct request *);
typedef int (dma_drain_needed_fn)(struct request *);
+typedef int (lld_busy_fn) (struct request_queue *q);
enum blk_eh_timer_return {
BLK_EH_NOT_HANDLED,
@@ -325,6 +326,7 @@ struct request_queue
softirq_done_fn *softirq_done_fn;
rq_timed_out_fn *rq_timed_out_fn;
dma_drain_needed_fn *dma_drain_needed;
+ lld_busy_fn *lld_busy_fn;
/*
* Dispatch queue sorting
@@ -699,6 +701,7 @@ extern struct request *blk_get_request(s
extern void blk_insert_request(struct request_queue *, struct request *, int, void *);
extern void blk_requeue_request(struct request_queue *, struct request *);
extern int blk_rq_check_limits(struct request_queue *q, struct request *rq);
+extern int blk_lld_busy(struct request_queue *q);
extern int blk_insert_cloned_request(struct request_queue *q,
struct request *rq);
extern void blk_plug_device(struct request_queue *);
@@ -835,6 +838,7 @@ extern void blk_queue_update_dma_pad(str
extern int blk_queue_dma_drain(struct request_queue *q,
dma_drain_needed_fn *dma_drain_needed,
void *buf, unsigned int size);
+extern void blk_queue_lld_busy(struct request_queue *q, lld_busy_fn *fn);
extern void blk_queue_segment_boundary(struct request_queue *, unsigned long);
extern void blk_queue_prep_rq(struct request_queue *, prep_rq_fn *pfn);
extern void blk_queue_merge_bvec(struct request_queue *, merge_bvec_fn *);
Index: linux-2.6-block/block/blk-core.c
===================================================================
--- linux-2.6-block.orig/block/blk-core.c
+++ linux-2.6-block/block/blk-core.c
@@ -2098,6 +2098,31 @@ void blk_rq_bio_prep(struct request_queu
rq->rq_disk = bio->bi_bdev->bd_disk;
}
+/**
+ * blk_lld_busy - Check if underlying low-level drivers of a device are busy
+ * @q : the queue of the device being checked
+ *
+ * Description:
+ * Check if underlying low-level drivers of a device are busy.
+ * If the drivers want to export their busy state, they must set own
+ * exporting function using blk_queue_lld_busy() first.
+ *
+ * Basically, this function is used only by request stacking drivers
+ * to stop dispatching requests to underlying devices when underlying
+ * devices are busy. This behavior helps more I/O merging on the queue
+ * of the request stacking driver and prevents I/O throughput regression
+ * on burst I/O load.
+ *
+ * Return:
+ * 0 - Not busy (The request stacking driver should dispatch request)
+ * 1 - Busy (The request stacking driver should stop dispatching request)
+ */
+int blk_lld_busy(struct request_queue *q)
+{
+ return q->lld_busy_fn ? q->lld_busy_fn(q) : 0;
+}
+EXPORT_SYMBOL_GPL(blk_lld_busy);
+
int kblockd_schedule_work(struct request_queue *q, struct work_struct *work)
{
return queue_work(kblockd_workqueue, work);
^ permalink raw reply [flat|nested] 3+ messages in thread* [RFC PATCH 2/2] scsi: export busy state via q->lld_busy_fn()
2008-09-25 15:53 [RFC PATCH 0/2] Export lld busy state for request stacking drivers Kiyoshi Ueda
2008-09-25 15:54 ` [RFC PATCH 1/2] block: add lld busy state exporting interface Kiyoshi Ueda
@ 2008-09-25 15:55 ` Kiyoshi Ueda
1 sibling, 0 replies; 3+ messages in thread
From: Kiyoshi Ueda @ 2008-09-25 15:55 UTC (permalink / raw)
To: akpm, jens.axboe, James.Bottomley
Cc: linux-kernel, linux-scsi, dm-devel, j-nomura, k-ueda
This patch implements q->lld_busy_fn() for scsi mid layer to export
its busy state.
Please note that it checks the cached information (sdev->lld_busy)
for efficiency, instead of checking the actual state of
sdev/starget/shost everytime.
So the care must be taken not to leave sdev->lld_busy = 1 for
the following cases:
- when there is no request in the sdev queue
- when scsi can't dispatch I/Os anymore and needs to kill I/Os
Otherwise, request stacking drivers may not submit any request,
and then, nobody sets back lld_busy = 0 and that causes deadlock.
OTOH, it has no major problem in setting sdev->lld_busy = 0 even when
the starget/shost is actually busy, because newly submitted request
will soon turn it to 1 in scsi_request_fn().
Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
---
drivers/scsi/scsi.c | 4 ++--
drivers/scsi/scsi_lib.c | 20 +++++++++++++++++++-
include/scsi/scsi_device.h | 13 +++++++++++++
3 files changed, 34 insertions(+), 3 deletions(-)
Index: linux-2.6-block/drivers/scsi/scsi_lib.c
===================================================================
--- linux-2.6-block.orig/drivers/scsi/scsi_lib.c
+++ linux-2.6-block/drivers/scsi/scsi_lib.c
@@ -470,6 +470,8 @@ void scsi_device_unbusy(struct scsi_devi
spin_unlock(shost->host_lock);
spin_lock(sdev->request_queue->queue_lock);
sdev->device_busy--;
+ if (sdev->device_busy < sdev->queue_depth && !sdev->device_blocked)
+ sdev->lld_busy = 0;
spin_unlock_irqrestore(sdev->request_queue->queue_lock, flags);
}
@@ -1461,6 +1463,13 @@ static void scsi_softirq_done(struct req
}
}
+static int scsi_lld_busy(struct request_queue *q)
+{
+ struct scsi_device *sdev = q->queuedata;
+
+ return sdev ? sdev->lld_busy : 0;
+}
+
/*
* Function: scsi_request_fn()
*
@@ -1503,9 +1512,14 @@ static void scsi_request_fn(struct reque
* accept it.
*/
req = elv_next_request(q);
- if (!req || !scsi_dev_queue_ready(q, sdev))
+ if (!req)
break;
+ if (!scsi_dev_queue_ready(q, sdev)) {
+ sdev->lld_busy = 1;
+ break;
+ }
+
if (unlikely(!scsi_device_online(sdev))) {
sdev_printk(KERN_ERR, sdev,
"rejecting I/O to offline device\n");
@@ -1576,6 +1590,8 @@ static void scsi_request_fn(struct reque
rtn = scsi_dispatch_cmd(cmd);
spin_lock_irq(q->queue_lock);
if(rtn) {
+ sdev->lld_busy = 1;
+
/* we're refusing the command; because of
* the way locks get dropped, we need to
* check here if plugging is required */
@@ -1600,6 +1616,7 @@ static void scsi_request_fn(struct reque
* later time.
*/
spin_lock_irq(q->queue_lock);
+ sdev->lld_busy = 1;
blk_requeue_request(q, req);
sdev->device_busy--;
if(sdev->device_busy == 0)
@@ -1683,6 +1700,7 @@ struct request_queue *scsi_alloc_queue(s
blk_queue_prep_rq(q, scsi_prep_fn);
blk_queue_softirq_done(q, scsi_softirq_done);
blk_queue_rq_timed_out(q, scsi_times_out);
+ blk_queue_lld_busy(q, scsi_lld_busy);
return q;
}
Index: linux-2.6-block/drivers/scsi/scsi.c
===================================================================
--- linux-2.6-block.orig/drivers/scsi/scsi.c
+++ linux-2.6-block/drivers/scsi/scsi.c
@@ -803,8 +803,6 @@ void scsi_finish_command(struct scsi_cmn
struct scsi_driver *drv;
unsigned int good_bytes;
- scsi_device_unbusy(sdev);
-
/*
* Clear the flags which say that the device/host is no longer
* capable of accepting new commands. These are set in scsi_queue.c
@@ -816,6 +814,8 @@ void scsi_finish_command(struct scsi_cmn
shost->host_blocked = 0;
sdev->device_blocked = 0;
+ scsi_device_unbusy(sdev);
+
/*
* If we have valid sense information, then some kind of recovery
* must have taken place. Make a note of this.
Index: linux-2.6-block/include/scsi/scsi_device.h
===================================================================
--- linux-2.6-block.orig/include/scsi/scsi_device.h
+++ linux-2.6-block/include/scsi/scsi_device.h
@@ -143,6 +143,19 @@ struct scsi_device {
unsigned retry_hwerror:1; /* Retry HARDWARE_ERROR */
unsigned last_sector_bug:1; /* do not use multisector accesses on
SD_LAST_BUGGY_SECTORS */
+ unsigned lld_busy:1; /* Exporting busy state for stacking
+ * drivers.
+ *
+ * 1 - There is one or more requests
+ * in the device's queue,
+ * and the device (or target/host)
+ * can not dispatch any request
+ * immediately.
+ * 0 - The device can dispatch
+ * (or kill) requests immediately,
+ * or there is no request in
+ * the device's queue even though
+ * target/host may busy. */
DECLARE_BITMAP(supported_events, SDEV_EVT_MAXBITS); /* supported events */
struct list_head event_list; /* asserted events */
^ permalink raw reply [flat|nested] 3+ messages in thread