Commit 2056a782 authored by Jens Axboe's avatar Jens Axboe

[PATCH] Block queue IO tracing support (blktrace) as of 2006-03-23

Signed-off-by: default avatarJens Axboe <axboe@suse.de>
parent 6dac40a7
......@@ -11,4 +11,16 @@ config LBD
your machine, or if you want to have a raid or loopback device
bigger than 2TB. Otherwise say N.
config BLK_DEV_IO_TRACE
bool "Support for tracing block io actions"
select RELAY
select DEBUG_FS
help
Say Y here, if you want to be able to trace the block layer actions
on a given queue. Tracing allows you to see any traffic happening
on a block device queue. For more information (and the user space
support tools needed), fetch the blktrace app from:
git://brick.kernel.dk/data/git/blktrace.git
source block/Kconfig.iosched
......@@ -8,3 +8,5 @@ obj-$(CONFIG_IOSCHED_NOOP) += noop-iosched.o
obj-$(CONFIG_IOSCHED_AS) += as-iosched.o
obj-$(CONFIG_IOSCHED_DEADLINE) += deadline-iosched.o
obj-$(CONFIG_IOSCHED_CFQ) += cfq-iosched.o
obj-$(CONFIG_BLK_DEV_IO_TRACE) += blktrace.o
This diff is collapsed.
......@@ -33,6 +33,7 @@
#include <linux/init.h>
#include <linux/compiler.h>
#include <linux/delay.h>
#include <linux/blktrace_api.h>
#include <asm/uaccess.h>
......@@ -333,6 +334,8 @@ void elv_insert(request_queue_t *q, struct request *rq, int where)
struct list_head *pos;
unsigned ordseq;
blk_add_trace_rq(q, rq, BLK_TA_INSERT);
rq->q = q;
switch (where) {
......@@ -499,6 +502,7 @@ struct request *elv_next_request(request_queue_t *q)
* not be passed by new incoming requests
*/
rq->flags |= REQ_STARTED;
blk_add_trace_rq(q, rq, BLK_TA_ISSUE);
}
if (!q->boundary_rq || q->boundary_rq == rq) {
......
......@@ -5,6 +5,7 @@
#include <linux/backing-dev.h>
#include <linux/buffer_head.h>
#include <linux/smp_lock.h>
#include <linux/blktrace_api.h>
#include <asm/uaccess.h>
static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user *arg)
......@@ -189,6 +190,11 @@ static int blkdev_locked_ioctl(struct file *file, struct block_device *bdev,
return put_ulong(arg, bdev->bd_inode->i_size >> 9);
case BLKGETSIZE64:
return put_u64(arg, bdev->bd_inode->i_size);
case BLKTRACESTART:
case BLKTRACESTOP:
case BLKTRACESETUP:
case BLKTRACETEARDOWN:
return blk_trace_ioctl(bdev, cmd, (char __user *) arg);
}
return -ENOIOCTLCMD;
}
......
......@@ -28,6 +28,7 @@
#include <linux/writeback.h>
#include <linux/interrupt.h>
#include <linux/cpu.h>
#include <linux/blktrace_api.h>
/*
* for max sense size
......@@ -1556,8 +1557,10 @@ void blk_plug_device(request_queue_t *q)
if (test_bit(QUEUE_FLAG_STOPPED, &q->queue_flags))
return;
if (!test_and_set_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags))
if (!test_and_set_bit(QUEUE_FLAG_PLUGGED, &q->queue_flags)) {
mod_timer(&q->unplug_timer, jiffies + q->unplug_delay);
blk_add_trace_generic(q, NULL, 0, BLK_TA_PLUG);
}
}
EXPORT_SYMBOL(blk_plug_device);
......@@ -1621,14 +1624,21 @@ static void blk_backing_dev_unplug(struct backing_dev_info *bdi,
/*
* devices don't necessarily have an ->unplug_fn defined
*/
if (q->unplug_fn)
if (q->unplug_fn) {
blk_add_trace_pdu_int(q, BLK_TA_UNPLUG_IO, NULL,
q->rq.count[READ] + q->rq.count[WRITE]);
q->unplug_fn(q);
}
}
static void blk_unplug_work(void *data)
{
request_queue_t *q = data;
blk_add_trace_pdu_int(q, BLK_TA_UNPLUG_IO, NULL,
q->rq.count[READ] + q->rq.count[WRITE]);
q->unplug_fn(q);
}
......@@ -1636,6 +1646,9 @@ static void blk_unplug_timeout(unsigned long data)
{
request_queue_t *q = (request_queue_t *)data;
blk_add_trace_pdu_int(q, BLK_TA_UNPLUG_TIMER, NULL,
q->rq.count[READ] + q->rq.count[WRITE]);
kblockd_schedule_work(&q->unplug_work);
}
......@@ -1753,6 +1766,9 @@ static void blk_release_queue(struct kobject *kobj)
if (q->queue_tags)
__blk_queue_free_tags(q);
if (q->blk_trace)
blk_trace_shutdown(q);
kmem_cache_free(requestq_cachep, q);
}
......@@ -2129,6 +2145,8 @@ static struct request *get_request(request_queue_t *q, int rw, struct bio *bio,
rq_init(q, rq);
rq->rl = rl;
blk_add_trace_generic(q, bio, rw, BLK_TA_GETRQ);
out:
return rq;
}
......@@ -2157,6 +2175,8 @@ static struct request *get_request_wait(request_queue_t *q, int rw,
if (!rq) {
struct io_context *ioc;
blk_add_trace_generic(q, bio, rw, BLK_TA_SLEEPRQ);
__generic_unplug_device(q);
spin_unlock_irq(q->queue_lock);
io_schedule();
......@@ -2210,6 +2230,8 @@ EXPORT_SYMBOL(blk_get_request);
*/
void blk_requeue_request(request_queue_t *q, struct request *rq)
{
blk_add_trace_rq(q, rq, BLK_TA_REQUEUE);
if (blk_rq_tagged(rq))
blk_queue_end_tag(q, rq);
......@@ -2844,6 +2866,8 @@ static int __make_request(request_queue_t *q, struct bio *bio)
if (!q->back_merge_fn(q, req, bio))
break;
blk_add_trace_bio(q, bio, BLK_TA_BACKMERGE);
req->biotail->bi_next = bio;
req->biotail = bio;
req->nr_sectors = req->hard_nr_sectors += nr_sectors;
......@@ -2859,6 +2883,8 @@ static int __make_request(request_queue_t *q, struct bio *bio)
if (!q->front_merge_fn(q, req, bio))
break;
blk_add_trace_bio(q, bio, BLK_TA_FRONTMERGE);
bio->bi_next = req->bio;
req->bio = bio;
......@@ -2976,6 +3002,7 @@ void generic_make_request(struct bio *bio)
request_queue_t *q;
sector_t maxsector;
int ret, nr_sectors = bio_sectors(bio);
dev_t old_dev;
might_sleep();
/* Test device or partition size, when known. */
......@@ -3002,6 +3029,8 @@ void generic_make_request(struct bio *bio)
* NOTE: we don't repeat the blk_size check for each new device.
* Stacking drivers are expected to know what they are doing.
*/
maxsector = -1;
old_dev = 0;
do {
char b[BDEVNAME_SIZE];
......@@ -3034,6 +3063,15 @@ void generic_make_request(struct bio *bio)
*/
blk_partition_remap(bio);
if (maxsector != -1)
blk_add_trace_remap(q, bio, old_dev, bio->bi_sector,
maxsector);
blk_add_trace_bio(q, bio, BLK_TA_QUEUE);
maxsector = bio->bi_sector;
old_dev = bio->bi_bdev->bd_dev;
ret = q->make_request_fn(q, bio);
} while (ret);
}
......@@ -3153,6 +3191,8 @@ static int __end_that_request_first(struct request *req, int uptodate,
int total_bytes, bio_nbytes, error, next_idx = 0;
struct bio *bio;
blk_add_trace_rq(req->q, req, BLK_TA_COMPLETE);
/*
* extend uptodate bool to allow < 0 value to be direct io error
*/
......
......@@ -38,6 +38,7 @@
#include <linux/hdreg.h>
#include <linux/spinlock.h>
#include <linux/compat.h>
#include <linux/blktrace_api.h>
#include <asm/uaccess.h>
#include <asm/io.h>
......@@ -2331,6 +2332,7 @@ static inline void complete_command( ctlr_info_t *h, CommandList_struct *cmd,
cmd->rq->completion_data = cmd;
cmd->rq->errors = status;
blk_add_trace_rq(cmd->rq->q, cmd->rq, BLK_TA_COMPLETE);
blk_complete_request(cmd->rq);
}
......
......@@ -17,6 +17,7 @@
#include <linux/mempool.h>
#include <linux/slab.h>
#include <linux/idr.h>
#include <linux/blktrace_api.h>
static const char *_name = DM_NAME;
......@@ -334,6 +335,8 @@ static void dec_pending(struct dm_io *io, int error)
/* nudge anyone waiting on suspend queue */
wake_up(&io->md->wait);
blk_add_trace_bio(io->md->queue, io->bio, BLK_TA_COMPLETE);
bio_endio(io->bio, io->bio->bi_size, io->error);
free_io(io->md, io);
}
......@@ -392,6 +395,7 @@ static void __map_bio(struct dm_target *ti, struct bio *clone,
struct target_io *tio)
{
int r;
sector_t sector;
/*
* Sanity checks.
......@@ -407,10 +411,17 @@ static void __map_bio(struct dm_target *ti, struct bio *clone,
* this io.
*/
atomic_inc(&tio->io->io_count);
sector = clone->bi_sector;
r = ti->type->map(ti, clone, &tio->info);
if (r > 0)
if (r > 0) {
/* the bio has been remapped so dispatch it */
blk_add_trace_remap(bdev_get_queue(clone->bi_bdev), clone,
tio->io->bio->bi_bdev->bd_dev, sector,
clone->bi_sector);
generic_make_request(clone);
}
else if (r < 0) {
/* error the io and bail out */
......
......@@ -25,6 +25,7 @@
#include <linux/module.h>
#include <linux/mempool.h>
#include <linux/workqueue.h>
#include <linux/blktrace_api.h>
#include <scsi/sg.h> /* for struct sg_iovec */
#define BIO_POOL_SIZE 256
......@@ -1095,6 +1096,9 @@ struct bio_pair *bio_split(struct bio *bi, mempool_t *pool, int first_sectors)
if (!bp)
return bp;
blk_add_trace_pdu_int(bdev_get_queue(bi->bi_bdev), BLK_TA_SPLIT, bi,
bi->bi_sector + first_sectors);
BUG_ON(bi->bi_vcnt != 1);
BUG_ON(bi->bi_idx != 0);
atomic_set(&bp->cnt, 3);
......
......@@ -72,6 +72,7 @@
#include <linux/i2c-dev.h>
#include <linux/wireless.h>
#include <linux/atalk.h>
#include <linux/blktrace_api.h>
#include <net/sock.h> /* siocdevprivate_ioctl */
#include <net/bluetooth/bluetooth.h>
......
......@@ -22,6 +22,7 @@ typedef struct request_queue request_queue_t;
struct elevator_queue;
typedef struct elevator_queue elevator_t;
struct request_pm_state;
struct blk_trace;
#define BLKDEV_MIN_RQ 4
#define BLKDEV_MAX_RQ 128 /* Default maximum */
......@@ -416,6 +417,8 @@ struct request_queue
unsigned int sg_reserved_size;
int node;
struct blk_trace *blk_trace;
/*
* reserved for flush operations
*/
......
#ifndef BLKTRACE_H
#define BLKTRACE_H
#include <linux/config.h>
#include <linux/blkdev.h>
#include <linux/relay.h>
/*
* Trace categories
*/
enum blktrace_cat {
BLK_TC_READ = 1 << 0, /* reads */
BLK_TC_WRITE = 1 << 1, /* writes */
BLK_TC_BARRIER = 1 << 2, /* barrier */
BLK_TC_SYNC = 1 << 3, /* barrier */
BLK_TC_QUEUE = 1 << 4, /* queueing/merging */
BLK_TC_REQUEUE = 1 << 5, /* requeueing */
BLK_TC_ISSUE = 1 << 6, /* issue */
BLK_TC_COMPLETE = 1 << 7, /* completions */
BLK_TC_FS = 1 << 8, /* fs requests */
BLK_TC_PC = 1 << 9, /* pc requests */
BLK_TC_NOTIFY = 1 << 10, /* special message */
BLK_TC_END = 1 << 15, /* only 16-bits, reminder */
};
#define BLK_TC_SHIFT (16)
#define BLK_TC_ACT(act) ((act) << BLK_TC_SHIFT)
/*
* Basic trace actions
*/
enum blktrace_act {
__BLK_TA_QUEUE = 1, /* queued */
__BLK_TA_BACKMERGE, /* back merged to existing rq */
__BLK_TA_FRONTMERGE, /* front merge to existing rq */
__BLK_TA_GETRQ, /* allocated new request */
__BLK_TA_SLEEPRQ, /* sleeping on rq allocation */
__BLK_TA_REQUEUE, /* request requeued */
__BLK_TA_ISSUE, /* sent to driver */
__BLK_TA_COMPLETE, /* completed by driver */
__BLK_TA_PLUG, /* queue was plugged */
__BLK_TA_UNPLUG_IO, /* queue was unplugged by io */
__BLK_TA_UNPLUG_TIMER, /* queue was unplugged by timer */
__BLK_TA_INSERT, /* insert request */
__BLK_TA_SPLIT, /* bio was split */
__BLK_TA_BOUNCE, /* bio was bounced */
__BLK_TA_REMAP, /* bio was remapped */
};
/*
* Trace actions in full. Additionally, read or write is masked
*/
#define BLK_TA_QUEUE (__BLK_TA_QUEUE | BLK_TC_ACT(BLK_TC_QUEUE))
#define BLK_TA_BACKMERGE (__BLK_TA_BACKMERGE | BLK_TC_ACT(BLK_TC_QUEUE))
#define BLK_TA_FRONTMERGE (__BLK_TA_FRONTMERGE | BLK_TC_ACT(BLK_TC_QUEUE))
#define BLK_TA_GETRQ (__BLK_TA_GETRQ | BLK_TC_ACT(BLK_TC_QUEUE))
#define BLK_TA_SLEEPRQ (__BLK_TA_SLEEPRQ | BLK_TC_ACT(BLK_TC_QUEUE))
#define BLK_TA_REQUEUE (__BLK_TA_REQUEUE | BLK_TC_ACT(BLK_TC_REQUEUE))
#define BLK_TA_ISSUE (__BLK_TA_ISSUE | BLK_TC_ACT(BLK_TC_ISSUE))
#define BLK_TA_COMPLETE (__BLK_TA_COMPLETE| BLK_TC_ACT(BLK_TC_COMPLETE))
#define BLK_TA_PLUG (__BLK_TA_PLUG | BLK_TC_ACT(BLK_TC_QUEUE))
#define BLK_TA_UNPLUG_IO (__BLK_TA_UNPLUG_IO | BLK_TC_ACT(BLK_TC_QUEUE))
#define BLK_TA_UNPLUG_TIMER (__BLK_TA_UNPLUG_TIMER | BLK_TC_ACT(BLK_TC_QUEUE))
#define BLK_TA_INSERT (__BLK_TA_INSERT | BLK_TC_ACT(BLK_TC_QUEUE))
#define BLK_TA_SPLIT (__BLK_TA_SPLIT)
#define BLK_TA_BOUNCE (__BLK_TA_BOUNCE)
#define BLK_TA_REMAP (__BLK_TA_REMAP | BLK_TC_ACT(BLK_TC_QUEUE))
#define BLK_IO_TRACE_MAGIC 0x65617400
#define BLK_IO_TRACE_VERSION 0x07
/*
* The trace itself
*/
struct blk_io_trace {
u32 magic; /* MAGIC << 8 | version */
u32 sequence; /* event number */
u64 time; /* in microseconds */
u64 sector; /* disk offset */
u32 bytes; /* transfer length */
u32 action; /* what happened */
u32 pid; /* who did it */
u32 device; /* device number */
u32 cpu; /* on what cpu did it happen */
u16 error; /* completion error */
u16 pdu_len; /* length of data after this trace */
};
/*
* The remap event
*/
struct blk_io_trace_remap {
u32 device;
u32 __pad;
u64 sector;
};
enum {
Blktrace_setup = 1,
Blktrace_running,
Blktrace_stopped,
};
struct blk_trace {
int trace_state;
struct rchan *rchan;
unsigned long *sequence;
u16 act_mask;
u64 start_lba;
u64 end_lba;
u32 pid;
u32 dev;
struct dentry *dir;
struct dentry *dropped_file;
atomic_t dropped;
};
/*
* User setup structure passed with BLKTRACESTART
*/
struct blk_user_trace_setup {
char name[BDEVNAME_SIZE]; /* output */
u16 act_mask; /* input */
u32 buf_size; /* input */
u32 buf_nr; /* input */
u64 start_lba;
u64 end_lba;
u32 pid;
};
#if defined(CONFIG_BLK_DEV_IO_TRACE)
extern int blk_trace_ioctl(struct block_device *, unsigned, char __user *);
extern void blk_trace_shutdown(request_queue_t *);
extern void __blk_add_trace(struct blk_trace *, sector_t, int, int, u32, int, int, void *);
/**
* blk_add_trace_rq - Add a trace for a request oriented action
* @q: queue the io is for
* @rq: the source request
* @what: the action
*
* Description:
* Records an action against a request. Will log the bio offset + size.
*
**/
static inline void blk_add_trace_rq(struct request_queue *q, struct request *rq,
u32 what)
{
struct blk_trace *bt = q->blk_trace;
int rw = rq->flags & 0x07;
if (likely(!bt))
return;
if (blk_pc_request(rq)) {
what |= BLK_TC_ACT(BLK_TC_PC);
__blk_add_trace(bt, 0, rq->data_len, rw, what, rq->errors, sizeof(rq->cmd), rq->cmd);
} else {
what |= BLK_TC_ACT(BLK_TC_FS);
__blk_add_trace(bt, rq->hard_sector, rq->hard_nr_sectors << 9, rw, what, rq->errors, 0, NULL);
}
}
/**
* blk_add_trace_bio - Add a trace for a bio oriented action
* @q: queue the io is for
* @bio: the source bio
* @what: the action
*
* Description:
* Records an action against a bio. Will log the bio offset + size.
*
**/
static inline void blk_add_trace_bio(struct request_queue *q, struct bio *bio,
u32 what)
{
struct blk_trace *bt = q->blk_trace;
if (likely(!bt))
return;
__blk_add_trace(bt, bio->bi_sector, bio->bi_size, bio->bi_rw, what, !bio_flagged(bio, BIO_UPTODATE), 0, NULL);
}
/**
* blk_add_trace_generic - Add a trace for a generic action
* @q: queue the io is for
* @bio: the source bio
* @rw: the data direction
* @what: the action
*
* Description:
* Records a simple trace
*
**/
static inline void blk_add_trace_generic(struct request_queue *q,
struct bio *bio, int rw, u32 what)
{
struct blk_trace *bt = q->blk_trace;
if (likely(!bt))
return;
if (bio)
blk_add_trace_bio(q, bio, what);
else
__blk_add_trace(bt, 0, 0, rw, what, 0, 0, NULL);
}
/**
* blk_add_trace_pdu_int - Add a trace for a bio with an integer payload
* @q: queue the io is for
* @what: the action
* @bio: the source bio
* @pdu: the integer payload
*
* Description:
* Adds a trace with some integer payload. This might be an unplug
* option given as the action, with the depth at unplug time given
* as the payload
*
**/
static inline void blk_add_trace_pdu_int(struct request_queue *q, u32 what,
struct bio *bio, unsigned int pdu)
{
struct blk_trace *bt = q->blk_trace;
u64 rpdu = cpu_to_be64(pdu);
if (likely(!bt))
return;
if (bio)
__blk_add_trace(bt, bio->bi_sector, bio->bi_size, bio->bi_rw, what, !bio_flagged(bio, BIO_UPTODATE), sizeof(rpdu), &rpdu);
else
__blk_add_trace(bt, 0, 0, 0, what, 0, sizeof(rpdu), &rpdu);
}
/**
* blk_add_trace_remap - Add a trace for a remap operation
* @q: queue the io is for
* @bio: the source bio
* @dev: target device
* @from: source sector
* @to: target sector
*
* Description:
* Device mapper or raid target sometimes need to split a bio because
* it spans a stripe (or similar). Add a trace for that action.
*
**/
static inline void blk_add_trace_remap(struct request_queue *q, struct bio *bio,
dev_t dev, sector_t from, sector_t to)
{
struct blk_trace *bt = q->blk_trace;
struct blk_io_trace_remap r;
if (likely(!bt))
return;
r.device = cpu_to_be32(dev);
r.sector = cpu_to_be64(to);
__blk_add_trace(bt, from, bio->bi_size, bio->bi_rw, BLK_TA_REMAP, !bio_flagged(bio, BIO_UPTODATE), sizeof(r), &r);
}
#else /* !CONFIG_BLK_DEV_IO_TRACE */
#define blk_trace_ioctl(bdev, cmd, arg) (-ENOTTY)
#define blk_trace_shutdown(q) do { } while (0)
#define blk_add_trace_rq(q, rq, what) do { } while (0)
#define blk_add_trace_bio(q, rq, what) do { } while (0)
#define blk_add_trace_generic(q, rq, rw, what) do { } while (0)
#define blk_add_trace_pdu_int(q, what, bio, pdu) do { } while (0)
#define blk_add_trace_remap(q, bio, dev, f, t) do {} while (0)
#endif /* CONFIG_BLK_DEV_IO_TRACE */
#endif
......@@ -97,6 +97,10 @@ COMPATIBLE_IOCTL(BLKRRPART)
COMPATIBLE_IOCTL(BLKFLSBUF)
COMPATIBLE_IOCTL(BLKSECTSET)
COMPATIBLE_IOCTL(BLKSSZGET)
COMPATIBLE_IOCTL(BLKTRACESTART)
COMPATIBLE_IOCTL(BLKTRACESTOP)
COMPATIBLE_IOCTL(BLKTRACESETUP)
COMPATIBLE_IOCTL(BLKTRACETEARDOWN)
ULONG_IOCTL(BLKRASET)
ULONG_IOCTL(BLKFRASET)
/* RAID */
......
......@@ -197,6 +197,10 @@ extern int dir_notify_enable;
#define BLKBSZGET _IOR(0x12,112,size_t)
#define BLKBSZSET _IOW(0x12,113,size_t)
#define BLKGETSIZE64 _IOR(0x12,114,size_t) /* return device size in bytes (u64 *arg) */
#define BLKTRACESETUP _IOWR(0x12,115,struct blk_user_trace_setup)
#define BLKTRACESTART _IO(0x12,116)
#define BLKTRACESTOP _IO(0x12,117)
#define BLKTRACETEARDOWN _IO(0x12,118)
#define BMAP_IOCTL 1 /* obsolete - kept for compatibility */
#define FIBMAP _IO(0x00,1) /* bmap access */
......
......@@ -706,6 +706,7 @@ struct task_struct {
prio_array_t *array;
unsigned short ioprio;
unsigned int btrace_seq;
unsigned long sleep_avg;
unsigned long long timestamp, last_ran;
......
......@@ -181,6 +181,7 @@ static struct task_struct *dup_task_struct(struct task_struct *orig)
/* One for us, one for whoever does the "release_task()" (usually parent) */
atomic_set(&tsk->usage,2);
atomic_set(&tsk->fs_excl, 0);
tsk->btrace_seq = 0;
return tsk;
}
......
......@@ -26,6 +26,7 @@
#include <linux/init.h>
#include <linux/hash.h>
#include <linux/highmem.h>
#include <linux/blktrace_api.h>
#include <asm/tlbflush.h>
static mempool_t *page_pool, *isa_page_pool;
......@@ -483,6 +484,8 @@ void blk_queue_bounce(request_queue_t *q, struct bio **bio_orig)
pool = isa_page_pool;
}
blk_add_trace_bio(q, *bio_orig, BLK_TA_BOUNCE);
/*
* slow path
*/
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment