[mirror_ubuntu-artful-kernel.git] / include / linux / blktrace_api.h

#ifndef BLKTRACE_H
#define BLKTRACE_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,	/* sync IO */
	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_AHEAD	= 1 << 11,	/* readahead */
	BLK_TC_META	= 1 << 12,	/* metadata */

	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 */
};

/*
 * Notify events.
 */
enum blktrace_notify {
	__BLK_TN_PROCESS = 0,		/* establish pid/name mapping */
	__BLK_TN_TIMESTAMP,		/* include system clock */
};


/*
 * 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_TN_PROCESS		(__BLK_TN_PROCESS | BLK_TC_ACT(BLK_TC_NOTIFY))
#define BLK_TN_TIMESTAMP	(__BLK_TN_TIMESTAMP | BLK_TC_ACT(BLK_TC_NOTIFY))

#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 {
	__be32 device;
	u32 __pad;
	__be64 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->cmd_flags & 0x03;

	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;
	__be64 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
Commit	Line	Data
2056a782 JA	1	#ifndef BLKTRACE_H
	2	#define BLKTRACE_H
	3
2056a782 JA	4	#include <linux/blkdev.h>
	5	#include <linux/relay.h>
	6
	7	/*
	8	* Trace categories
	9	*/
	10	enum blktrace_cat {
	11	BLK_TC_READ = 1 << 0, /* reads */
	12	BLK_TC_WRITE = 1 << 1, /* writes */
	13	BLK_TC_BARRIER = 1 << 2, /* barrier */
fc0a75ce	14	BLK_TC_SYNC = 1 << 3, /* sync IO */
2056a782 JA	15	BLK_TC_QUEUE = 1 << 4, /* queueing/merging */
	16	BLK_TC_REQUEUE = 1 << 5, /* requeueing */
	17	BLK_TC_ISSUE = 1 << 6, /* issue */
	18	BLK_TC_COMPLETE = 1 << 7, /* completions */
	19	BLK_TC_FS = 1 << 8, /* fs requests */
	20	BLK_TC_PC = 1 << 9, /* pc requests */
	21	BLK_TC_NOTIFY = 1 << 10, /* special message */
40359ccb	22	BLK_TC_AHEAD = 1 << 11, /* readahead */
7457e6e2	23	BLK_TC_META = 1 << 12, /* metadata */
2056a782 JA	24
	25	BLK_TC_END = 1 << 15, /* only 16-bits, reminder */
	26	};
	27
	28	#define BLK_TC_SHIFT (16)
	29	#define BLK_TC_ACT(act) ((act) << BLK_TC_SHIFT)
	30
	31	/*
	32	* Basic trace actions
	33	*/
	34	enum blktrace_act {
	35	__BLK_TA_QUEUE = 1, /* queued */
	36	__BLK_TA_BACKMERGE, /* back merged to existing rq */
	37	__BLK_TA_FRONTMERGE, /* front merge to existing rq */
	38	__BLK_TA_GETRQ, /* allocated new request */
	39	__BLK_TA_SLEEPRQ, /* sleeping on rq allocation */
	40	__BLK_TA_REQUEUE, /* request requeued */
	41	__BLK_TA_ISSUE, /* sent to driver */
	42	__BLK_TA_COMPLETE, /* completed by driver */
	43	__BLK_TA_PLUG, /* queue was plugged */
	44	__BLK_TA_UNPLUG_IO, /* queue was unplugged by io */
	45	__BLK_TA_UNPLUG_TIMER, /* queue was unplugged by timer */
	46	__BLK_TA_INSERT, /* insert request */
	47	__BLK_TA_SPLIT, /* bio was split */
	48	__BLK_TA_BOUNCE, /* bio was bounced */
	49	__BLK_TA_REMAP, /* bio was remapped */
	50	};
	51
be1c6341 OK	52	/*
	53	* Notify events.
	54	*/
	55	enum blktrace_notify {
	56	__BLK_TN_PROCESS = 0, /* establish pid/name mapping */
	57	__BLK_TN_TIMESTAMP, /* include system clock */
	58	};
	59
	60
2056a782 JA	61	/*
	62	* Trace actions in full. Additionally, read or write is masked
	63	*/
	64	#define BLK_TA_QUEUE (__BLK_TA_QUEUE \| BLK_TC_ACT(BLK_TC_QUEUE))
	65	#define BLK_TA_BACKMERGE (__BLK_TA_BACKMERGE \| BLK_TC_ACT(BLK_TC_QUEUE))
	66	#define BLK_TA_FRONTMERGE (__BLK_TA_FRONTMERGE \| BLK_TC_ACT(BLK_TC_QUEUE))
	67	#define BLK_TA_GETRQ (__BLK_TA_GETRQ \| BLK_TC_ACT(BLK_TC_QUEUE))
	68	#define BLK_TA_SLEEPRQ (__BLK_TA_SLEEPRQ \| BLK_TC_ACT(BLK_TC_QUEUE))
	69	#define BLK_TA_REQUEUE (__BLK_TA_REQUEUE \| BLK_TC_ACT(BLK_TC_REQUEUE))
	70	#define BLK_TA_ISSUE (__BLK_TA_ISSUE \| BLK_TC_ACT(BLK_TC_ISSUE))
	71	#define BLK_TA_COMPLETE (__BLK_TA_COMPLETE\| BLK_TC_ACT(BLK_TC_COMPLETE))
	72	#define BLK_TA_PLUG (__BLK_TA_PLUG \| BLK_TC_ACT(BLK_TC_QUEUE))
	73	#define BLK_TA_UNPLUG_IO (__BLK_TA_UNPLUG_IO \| BLK_TC_ACT(BLK_TC_QUEUE))
	74	#define BLK_TA_UNPLUG_TIMER (__BLK_TA_UNPLUG_TIMER \| BLK_TC_ACT(BLK_TC_QUEUE))
	75	#define BLK_TA_INSERT (__BLK_TA_INSERT \| BLK_TC_ACT(BLK_TC_QUEUE))
	76	#define BLK_TA_SPLIT (__BLK_TA_SPLIT)
	77	#define BLK_TA_BOUNCE (__BLK_TA_BOUNCE)
	78	#define BLK_TA_REMAP (__BLK_TA_REMAP \| BLK_TC_ACT(BLK_TC_QUEUE))
	79
be1c6341 OK	80	#define BLK_TN_PROCESS (__BLK_TN_PROCESS \| BLK_TC_ACT(BLK_TC_NOTIFY))
	81	#define BLK_TN_TIMESTAMP (__BLK_TN_TIMESTAMP \| BLK_TC_ACT(BLK_TC_NOTIFY))
	82
2056a782 JA	83	#define BLK_IO_TRACE_MAGIC 0x65617400
	84	#define BLK_IO_TRACE_VERSION 0x07
	85
	86	/*
	87	* The trace itself
	88	*/
	89	struct blk_io_trace {
	90	u32 magic; /* MAGIC << 8 \| version */
	91	u32 sequence; /* event number */
	92	u64 time; /* in microseconds */
	93	u64 sector; /* disk offset */
	94	u32 bytes; /* transfer length */
	95	u32 action; /* what happened */
	96	u32 pid; /* who did it */
	97	u32 device; /* device number */
	98	u32 cpu; /* on what cpu did it happen */
	99	u16 error; /* completion error */
	100	u16 pdu_len; /* length of data after this trace */
	101	};
	102
	103	/*
	104	* The remap event
	105	*/
	106	struct blk_io_trace_remap {
fda151d9	107	__be32 device;
2056a782	108	u32 __pad;
fda151d9	109	__be64 sector;
2056a782 JA	110	};
	111
	112	enum {
	113	Blktrace_setup = 1,
	114	Blktrace_running,
	115	Blktrace_stopped,
	116	};
	117
	118	struct blk_trace {
	119	int trace_state;
	120	struct rchan *rchan;
	121	unsigned long *sequence;
	122	u16 act_mask;
	123	u64 start_lba;
	124	u64 end_lba;
	125	u32 pid;
	126	u32 dev;
	127	struct dentry *dir;
	128	struct dentry *dropped_file;
	129	atomic_t dropped;
	130	};
	131
	132	/*
	133	* User setup structure passed with BLKTRACESTART
	134	*/
	135	struct blk_user_trace_setup {
	136	char name[BDEVNAME_SIZE]; /* output */
	137	u16 act_mask; /* input */
	138	u32 buf_size; /* input */
	139	u32 buf_nr; /* input */
	140	u64 start_lba;
	141	u64 end_lba;
	142	u32 pid;
	143	};
	144
	145	#if defined(CONFIG_BLK_DEV_IO_TRACE)
	146	extern int blk_trace_ioctl(struct block_device , unsigned, char __user );
	147	extern void blk_trace_shutdown(request_queue_t *);
	148	extern void __blk_add_trace(struct blk_trace , sector_t, int, int, u32, int, int, void );
	149
	150	/**
	151	* blk_add_trace_rq - Add a trace for a request oriented action
	152	* @q: queue the io is for
	153	* @rq: the source request
	154	* @what: the action
	155	*
	156	* Description:
	157	* Records an action against a request. Will log the bio offset + size.
	158	*
	159	**/
	160	static inline void blk_add_trace_rq(struct request_queue q, struct request rq,
	161	u32 what)
	162	{
	163	struct blk_trace *bt = q->blk_trace;
4aff5e23	164	int rw = rq->cmd_flags & 0x03;
2056a782 JA	165
	166	if (likely(!bt))
	167	return;
	168
	169	if (blk_pc_request(rq)) {
	170	what \|= BLK_TC_ACT(BLK_TC_PC);
	171	__blk_add_trace(bt, 0, rq->data_len, rw, what, rq->errors, sizeof(rq->cmd), rq->cmd);
	172	} else {
	173	what \|= BLK_TC_ACT(BLK_TC_FS);
	174	__blk_add_trace(bt, rq->hard_sector, rq->hard_nr_sectors << 9, rw, what, rq->errors, 0, NULL);
	175	}
	176	}
	177
	178	/**
	179	* blk_add_trace_bio - Add a trace for a bio oriented action
	180	* @q: queue the io is for
	181	* @bio: the source bio
	182	* @what: the action
	183	*
	184	* Description:
	185	* Records an action against a bio. Will log the bio offset + size.
	186	*
	187	**/
	188	static inline void blk_add_trace_bio(struct request_queue q, struct bio bio,
	189	u32 what)
	190	{
	191	struct blk_trace *bt = q->blk_trace;
	192
	193	if (likely(!bt))
	194	return;
	195
	196	__blk_add_trace(bt, bio->bi_sector, bio->bi_size, bio->bi_rw, what, !bio_flagged(bio, BIO_UPTODATE), 0, NULL);
	197	}
	198
	199	/**
	200	* blk_add_trace_generic - Add a trace for a generic action
	201	* @q: queue the io is for
	202	* @bio: the source bio
	203	* @rw: the data direction
	204	* @what: the action
	205	*
	206	* Description:
	207	* Records a simple trace
	208	*
	209	**/
	210	static inline void blk_add_trace_generic(struct request_queue *q,
	211	struct bio *bio, int rw, u32 what)
	212	{
	213	struct blk_trace *bt = q->blk_trace;
	214
	215	if (likely(!bt))
	216	return;
	217
	218	if (bio)
	219	blk_add_trace_bio(q, bio, what);
	220	else
	221	__blk_add_trace(bt, 0, 0, rw, what, 0, 0, NULL);
	222	}
	223
	224	/**
	225	* blk_add_trace_pdu_int - Add a trace for a bio with an integer payload
	226	* @q: queue the io is for
	227	* @what: the action
	228	* @bio: the source bio
229	* @pdu: the integer payload
230	*
231	* Description:
232	* Adds a trace with some integer payload. This might be an unplug
233	* option given as the action, with the depth at unplug time given
234	* as the payload
235	*
236	**/
237	static inline void blk_add_trace_pdu_int(struct request_queue *q, u32 what,
238	struct bio *bio, unsigned int pdu)
239	{
240	struct blk_trace *bt = q->blk_trace;
fda151d9	241	__be64 rpdu = cpu_to_be64(pdu);
2056a782 JA	242
	243	if (likely(!bt))
	244	return;
	245
	246	if (bio)
	247	__blk_add_trace(bt, bio->bi_sector, bio->bi_size, bio->bi_rw, what, !bio_flagged(bio, BIO_UPTODATE), sizeof(rpdu), &rpdu);
	248	else
	249	__blk_add_trace(bt, 0, 0, 0, what, 0, sizeof(rpdu), &rpdu);
	250	}
	251
	252	/**
	253	* blk_add_trace_remap - Add a trace for a remap operation
	254	* @q: queue the io is for
	255	* @bio: the source bio
	256	* @dev: target device
	257	* @from: source sector
	258	* @to: target sector
	259	*
	260	* Description:
	261	* Device mapper or raid target sometimes need to split a bio because
	262	* it spans a stripe (or similar). Add a trace for that action.
	263	*
	264	**/
	265	static inline void blk_add_trace_remap(struct request_queue q, struct bio bio,
	266	dev_t dev, sector_t from, sector_t to)
	267	{
	268	struct blk_trace *bt = q->blk_trace;
	269	struct blk_io_trace_remap r;
	270
	271	if (likely(!bt))
	272	return;
	273
	274	r.device = cpu_to_be32(dev);
	275	r.sector = cpu_to_be64(to);
	276
	277	__blk_add_trace(bt, from, bio->bi_size, bio->bi_rw, BLK_TA_REMAP, !bio_flagged(bio, BIO_UPTODATE), sizeof(r), &r);
	278	}
	279
	280	#else /* !CONFIG_BLK_DEV_IO_TRACE */
	281	#define blk_trace_ioctl(bdev, cmd, arg) (-ENOTTY)
	282	#define blk_trace_shutdown(q) do { } while (0)
	283	#define blk_add_trace_rq(q, rq, what) do { } while (0)
	284	#define blk_add_trace_bio(q, rq, what) do { } while (0)
	285	#define blk_add_trace_generic(q, rq, rw, what) do { } while (0)
	286	#define blk_add_trace_pdu_int(q, what, bio, pdu) do { } while (0)
	287	#define blk_add_trace_remap(q, bio, dev, f, t) do {} while (0)
	288	#endif /* CONFIG_BLK_DEV_IO_TRACE */
	289
	290	#endif