+++ /dev/null
-diff --git a/block/Kconfig b/block/Kconfig
-index 377f6dd..27eaed9 100644
---- a/block/Kconfig
-+++ b/block/Kconfig
-@@ -11,4 +11,15 @@ 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 RELAYFS_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
-diff --git a/block/Makefile b/block/Makefile
-index 7e4f93e..c05de0e 100644
---- a/block/Makefile
-+++ b/block/Makefile
-@@ -8,3 +8,5 @@ obj-$(CONFIG_IOSCHED_NOOP) += noop-iosch
- obj-$(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
-diff --git a/block/blktrace.c b/block/blktrace.c
-new file mode 100644
-index 0000000..e786a1d
---- /dev/null
-+++ b/block/blktrace.c
-@@ -0,0 +1,512 @@
-+/*
-+ * Copyright (C) 2006 Jens Axboe <axboe@suse.de>
-+ *
-+ * This program is free software; you can redistribute it and/or modify
-+ * it under the terms of the GNU General Public License version 2 as
-+ * published by the Free Software Foundation.
-+ *
-+ * This program is distributed in the hope that it will be useful,
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-+ * GNU General Public License for more details.
-+ *
-+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ *
-+ */
-+#include <linux/config.h>
-+#include <linux/kernel.h>
-+#include <linux/blkdev.h>
-+#include <linux/blktrace_api.h>
-+#include <linux/percpu.h>
-+#include <linux/init.h>
-+#include <linux/mutex.h>
-+#include <asm/uaccess.h>
-+
-+static DEFINE_PER_CPU(unsigned long long, blk_trace_cpu_offset) = { 0, };
-+static unsigned int blktrace_seq __read_mostly = 1;
-+
-+/*
-+ * Send out a notify for this process, if we haven't done so since a trace
-+ * started
-+ */
-+static void trace_note_tsk(struct blk_trace *bt, struct task_struct *tsk)
-+{
-+ struct blk_io_trace *t;
-+
-+ t = relay_reserve(bt->rchan, sizeof(*t) + sizeof(tsk->comm));
-+ if (t) {
-+ t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION;
-+ t->device = bt->dev;
-+ t->action = BLK_TC_ACT(BLK_TC_NOTIFY);
-+ t->pid = tsk->pid;
-+ t->cpu = smp_processor_id();
-+ t->pdu_len = sizeof(tsk->comm);
-+ memcpy((void *) t + sizeof(*t), tsk->comm, t->pdu_len);
-+ tsk->btrace_seq = blktrace_seq;
-+ }
-+}
-+
-+static int act_log_check(struct blk_trace *bt, u32 what, sector_t sector,
-+ pid_t pid)
-+{
-+ if (((bt->act_mask << BLK_TC_SHIFT) & what) == 0)
-+ return 1;
-+ if (sector < bt->start_lba || sector > bt->end_lba)
-+ return 1;
-+ if (bt->pid && pid != bt->pid)
-+ return 1;
-+
-+ return 0;
-+}
-+
-+/*
-+ * Data direction bit lookup
-+ */
-+static u32 ddir_act[2] __read_mostly = { BLK_TC_ACT(BLK_TC_READ), BLK_TC_ACT(BLK_TC_WRITE) };
-+
-+/*
-+ * Bio action bits of interest
-+ */
-+static u32 bio_act[3] __read_mostly = { 0, BLK_TC_ACT(BLK_TC_BARRIER), BLK_TC_ACT(BLK_TC_SYNC) };
-+
-+/*
-+ * More could be added as needed, taking care to increment the decrementer
-+ * to get correct indexing
-+ */
-+#define trace_barrier_bit(rw) \
-+ (((rw) & (1 << BIO_RW_BARRIER)) >> (BIO_RW_BARRIER - 0))
-+#define trace_sync_bit(rw) \
-+ (((rw) & (1 << BIO_RW_SYNC)) >> (BIO_RW_SYNC - 1))
-+
-+/*
-+ * The worker for the various blk_add_trace*() types. Fills out a
-+ * blk_io_trace structure and places it in a per-cpu subbuffer.
-+ */
-+void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes,
-+ int rw, u32 what, int error, int pdu_len, void *pdu_data)
-+{
-+ struct task_struct *tsk = current;
-+ struct blk_io_trace *t;
-+ unsigned long flags;
-+ unsigned long *sequence;
-+ pid_t pid;
-+ int cpu;
-+
-+ if (unlikely(bt->trace_state != Blktrace_running))
-+ return;
-+
-+ what |= ddir_act[rw & WRITE];
-+ what |= bio_act[trace_barrier_bit(rw)];
-+ what |= bio_act[trace_sync_bit(rw)];
-+
-+ pid = tsk->pid;
-+ if (unlikely(act_log_check(bt, what, sector, pid)))
-+ return;
-+
-+ /*
-+ * A word about the locking here - we disable interrupts to reserve
-+ * some space in the relayfs per-cpu buffer, to prevent an irq
-+ * from coming in and stepping on our toes. Once reserved, it's
-+ * enough to get preemption disabled to prevent read of this data
-+ * before we are through filling it. get_cpu()/put_cpu() does this
-+ * for us
-+ */
-+ local_irq_save(flags);
-+
-+ if (unlikely(tsk->btrace_seq != blktrace_seq))
-+ trace_note_tsk(bt, tsk);
-+
-+ t = relay_reserve(bt->rchan, sizeof(*t) + pdu_len);
-+ if (t) {
-+ cpu = smp_processor_id();
-+ sequence = per_cpu_ptr(bt->sequence, cpu);
-+
-+ t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION;
-+ t->sequence = ++(*sequence);
-+ t->time = sched_clock() - per_cpu(blk_trace_cpu_offset, cpu);
-+ t->sector = sector;
-+ t->bytes = bytes;
-+ t->action = what;
-+ t->pid = pid;
-+ t->device = bt->dev;
-+ t->cpu = cpu;
-+ t->error = error;
-+ t->pdu_len = pdu_len;
-+
-+ if (pdu_len)
-+ memcpy((void *) t + sizeof(*t), pdu_data, pdu_len);
-+ }
-+
-+ local_irq_restore(flags);
-+}
-+
-+EXPORT_SYMBOL_GPL(__blk_add_trace);
-+
-+static struct dentry *blk_tree_root;
-+static struct mutex blk_tree_mutex;
-+
-+static inline void blk_remove_root(void)
-+{
-+ if (relayfs_remove_dir(blk_tree_root) != -ENOTEMPTY)
-+ blk_tree_root = NULL;
-+}
-+
-+static void blk_remove_tree(struct dentry *dir)
-+{
-+ mutex_lock(&blk_tree_mutex);
-+ relayfs_remove_dir(dir);
-+ blk_remove_root();
-+ mutex_unlock(&blk_tree_mutex);
-+}
-+
-+static struct dentry *blk_create_tree(const char *blk_name)
-+{
-+ struct dentry *dir = NULL;
-+
-+ mutex_lock(&blk_tree_mutex);
-+
-+ if (!blk_tree_root) {
-+ blk_tree_root = relayfs_create_dir("block", NULL);
-+ if (!blk_tree_root)
-+ goto err;
-+ }
-+
-+ dir = relayfs_create_dir(blk_name, blk_tree_root);
-+ if (!dir)
-+ blk_remove_root();
-+
-+err:
-+ mutex_unlock(&blk_tree_mutex);
-+ return dir;
-+}
-+
-+static void blk_trace_cleanup(struct blk_trace *bt)
-+{
-+ relay_close(bt->rchan);
-+ relayfs_remove_file(bt->dropped_file);
-+ blk_remove_tree(bt->dir);
-+ free_percpu(bt->sequence);
-+ kfree(bt);
-+}
-+
-+static int blk_trace_remove(request_queue_t *q)
-+{
-+ struct blk_trace *bt;
-+
-+ bt = xchg(&q->blk_trace, NULL);
-+ if (!bt)
-+ return -EINVAL;
-+
-+ if (bt->trace_state == Blktrace_setup ||
-+ bt->trace_state == Blktrace_stopped)
-+ blk_trace_cleanup(bt);
-+
-+ return 0;
-+}
-+
-+static int blk_dropped_open(struct inode *inode, struct file *filp)
-+{
-+ filp->private_data = inode->u.generic_ip;
-+
-+ return 0;
-+}
-+
-+static ssize_t blk_dropped_read(struct file *filp, char __user *buffer,
-+ size_t count, loff_t *ppos)
-+{
-+ struct blk_trace *bt = filp->private_data;
-+ char buf[16];
-+
-+ snprintf(buf, sizeof(buf), "%u\n", atomic_read(&bt->dropped));
-+
-+ return simple_read_from_buffer(buffer, count, ppos, buf, strlen(buf));
-+}
-+
-+static struct file_operations blk_dropped_fops = {
-+ .owner = THIS_MODULE,
-+ .open = blk_dropped_open,
-+ .read = blk_dropped_read,
-+};
-+
-+/*
-+ * Keep track of how many times we encountered a full subbuffer, to aid
-+ * the user space app in telling how many lost events there were.
-+ */
-+static int blk_subbuf_start_callback(struct rchan_buf *buf, void *subbuf,
-+ void *prev_subbuf, size_t prev_padding)
-+{
-+ struct blk_trace *bt;
-+
-+ if (!relay_buf_full(buf))
-+ return 1;
-+
-+ bt = buf->chan->private_data;
-+ atomic_inc(&bt->dropped);
-+ return 0;
-+}
-+
-+static struct rchan_callbacks blk_relay_callbacks = {
-+ .subbuf_start = blk_subbuf_start_callback,
-+};
-+
-+/*
-+ * Setup everything required to start tracing
-+ */
-+static int blk_trace_setup(request_queue_t *q, struct block_device *bdev,
-+ char __user *arg)
-+{
-+ struct blk_user_trace_setup buts;
-+ struct blk_trace *old_bt, *bt = NULL;
-+ struct dentry *dir = NULL;
-+ char b[BDEVNAME_SIZE];
-+ int ret, i;
-+
-+ if (copy_from_user(&buts, arg, sizeof(buts)))
-+ return -EFAULT;
-+
-+ if (!buts.buf_size || !buts.buf_nr)
-+ return -EINVAL;
-+
-+ strcpy(buts.name, bdevname(bdev, b));
-+
-+ /*
-+ * some device names have larger paths - convert the slashes
-+ * to underscores for this to work as expected
-+ */
-+ for (i = 0; i < strlen(buts.name); i++)
-+ if (buts.name[i] == '/')
-+ buts.name[i] = '_';
-+
-+ if (copy_to_user(arg, &buts, sizeof(buts)))
-+ return -EFAULT;
-+
-+ ret = -ENOMEM;
-+ bt = kzalloc(sizeof(*bt), GFP_KERNEL);
-+ if (!bt)
-+ goto err;
-+
-+ bt->sequence = alloc_percpu(unsigned long);
-+ if (!bt->sequence)
-+ goto err;
-+
-+ ret = -ENOENT;
-+ dir = blk_create_tree(buts.name);
-+ if (!dir)
-+ goto err;
-+
-+ bt->dir = dir;
-+ bt->dev = bdev->bd_dev;
-+ atomic_set(&bt->dropped, 0);
-+
-+ ret = -EIO;
-+ bt->dropped_file = relayfs_create_file("dropped", dir, 0, &blk_dropped_fops, bt);
-+ if (!bt->dropped_file)
-+ goto err;
-+
-+ bt->rchan = relay_open("trace", dir, buts.buf_size, buts.buf_nr, &blk_relay_callbacks);
-+ if (!bt->rchan)
-+ goto err;
-+ bt->rchan->private_data = bt;
-+
-+ bt->act_mask = buts.act_mask;
-+ if (!bt->act_mask)
-+ bt->act_mask = (u16) -1;
-+
-+ bt->start_lba = buts.start_lba;
-+ bt->end_lba = buts.end_lba;
-+ if (!bt->end_lba)
-+ bt->end_lba = -1ULL;
-+
-+ bt->pid = buts.pid;
-+ bt->trace_state = Blktrace_setup;
-+
-+ ret = -EBUSY;
-+ old_bt = xchg(&q->blk_trace, bt);
-+ if (old_bt) {
-+ xchg(&q->blk_trace, old_bt);
-+ goto err;
-+ }
-+
-+ return 0;
-+err:
-+ if (dir)
-+ blk_remove_tree(dir);
-+ if (bt) {
-+ if (bt->dropped_file)
-+ relayfs_remove_file(bt->dropped_file);
-+ if (bt->sequence)
-+ free_percpu(bt->sequence);
-+ if (bt->rchan)
-+ relay_close(bt->rchan);
-+ kfree(bt);
-+ }
-+ return ret;
-+}
-+
-+static int blk_trace_startstop(request_queue_t *q, int start)
-+{
-+ struct blk_trace *bt;
-+ int ret;
-+
-+ if ((bt = q->blk_trace) == NULL)
-+ return -EINVAL;
-+
-+ /*
-+ * For starting a trace, we can transition from a setup or stopped
-+ * trace. For stopping a trace, the state must be running
-+ */
-+ ret = -EINVAL;
-+ if (start) {
-+ if (bt->trace_state == Blktrace_setup ||
-+ bt->trace_state == Blktrace_stopped) {
-+ blktrace_seq++;
-+ smp_mb();
-+ bt->trace_state = Blktrace_running;
-+ ret = 0;
-+ }
-+ } else {
-+ if (bt->trace_state == Blktrace_running) {
-+ bt->trace_state = Blktrace_stopped;
-+ ret = 0;
-+ }
-+ }
-+
-+ return ret;
-+}
-+
-+/**
-+ * blk_trace_ioctl: - handle the ioctls associated with tracing
-+ * @bdev: the block device
-+ * @cmd: the ioctl cmd
-+ * @arg: the argument data, if any
-+ *
-+ **/
-+int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg)
-+{
-+ request_queue_t *q;
-+ int ret, start = 0;
-+
-+ q = bdev_get_queue(bdev);
-+ if (!q)
-+ return -ENXIO;
-+
-+ down(&bdev->bd_sem);
-+
-+ switch (cmd) {
-+ case BLKTRACESETUP:
-+ ret = blk_trace_setup(q, bdev, arg);
-+ break;
-+ case BLKTRACESTART:
-+ start = 1;
-+ case BLKTRACESTOP:
-+ ret = blk_trace_startstop(q, start);
-+ break;
-+ case BLKTRACETEARDOWN:
-+ ret = blk_trace_remove(q);
-+ break;
-+ default:
-+ ret = -ENOTTY;
-+ break;
-+ }
-+
-+ up(&bdev->bd_sem);
-+ return ret;
-+}
-+
-+/**
-+ * blk_trace_shutdown: - stop and cleanup trace structures
-+ * @q: the request queue associated with the device
-+ *
-+ **/
-+void blk_trace_shutdown(request_queue_t *q)
-+{
-+ blk_trace_startstop(q, 0);
-+ blk_trace_remove(q);
-+}
-+
-+/*
-+ * Average offset over two calls to sched_clock() with a gettimeofday()
-+ * in the middle
-+ */
-+static void blk_check_time(unsigned long long *t)
-+{
-+ unsigned long long a, b;
-+ struct timeval tv;
-+
-+ a = sched_clock();
-+ do_gettimeofday(&tv);
-+ b = sched_clock();
-+
-+ *t = tv.tv_sec * 1000000000 + tv.tv_usec * 1000;
-+ *t -= (a + b) / 2;
-+}
-+
-+static void blk_trace_check_cpu_time(void *data)
-+{
-+ unsigned long long *t;
-+ int cpu = get_cpu();
-+
-+ t = &per_cpu(blk_trace_cpu_offset, cpu);
-+
-+ /*
-+ * Just call it twice, hopefully the second call will be cache hot
-+ * and a little more precise
-+ */
-+ blk_check_time(t);
-+ blk_check_time(t);
-+
-+ put_cpu();
-+}
-+
-+/*
-+ * Call blk_trace_check_cpu_time() on each CPU to calibrate our inter-CPU
-+ * timings
-+ */
-+static void blk_trace_calibrate_offsets(void)
-+{
-+ unsigned long flags;
-+
-+ smp_call_function(blk_trace_check_cpu_time, NULL, 1, 1);
-+ local_irq_save(flags);
-+ blk_trace_check_cpu_time(NULL);
-+ local_irq_restore(flags);
-+}
-+
-+static void blk_trace_set_ht_offsets(void)
-+{
-+#if defined(CONFIG_SCHED_SMT)
-+ int cpu, i;
-+
-+ /*
-+ * now make sure HT siblings have the same time offset
-+ */
-+ preempt_disable();
-+ for_each_online_cpu(cpu) {
-+ unsigned long long *cpu_off, *sibling_off;
-+
-+ for_each_cpu_mask(i, cpu_sibling_map[cpu]) {
-+ if (i == cpu)
-+ continue;
-+
-+ cpu_off = &per_cpu(blk_trace_cpu_offset, cpu);
-+ sibling_off = &per_cpu(blk_trace_cpu_offset, i);
-+ *sibling_off = *cpu_off;
-+ }
-+ }
-+ preempt_enable();
-+#endif
-+}
-+
-+static __init int blk_trace_init(void)
-+{
-+ mutex_init(&blk_tree_mutex);
-+ blk_trace_calibrate_offsets();
-+ blk_trace_set_ht_offsets();
-+
-+ return 0;
-+}
-+
-+module_init(blk_trace_init);
-+
-diff --git a/block/elevator.c b/block/elevator.c
-index 24b702d..0c9fafe 100644
---- a/block/elevator.c
-+++ b/block/elevator.c
-@@ -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>
-
-@@ -315,6 +316,8 @@ void elv_insert(request_queue_t *q, stru
- struct list_head *pos;
- unsigned ordseq;
-
-+ blk_add_trace_rq(q, rq, BLK_TA_INSERT);
-+
- rq->q = q;
-
- switch (where) {
-@@ -481,6 +484,7 @@ struct request *elv_next_request(request
- * 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) {
-diff --git a/block/ioctl.c b/block/ioctl.c
-index e110949..7acb56c 100644
---- a/block/ioctl.c
-+++ b/block/ioctl.c
-@@ -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 fi
- 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;
- }
-diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
-index 03d9c82..41387f5 100644
---- a/block/ll_rw_blk.c
-+++ b/block/ll_rw_blk.c
-@@ -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
-@@ -1551,8 +1552,10 @@ 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);
-@@ -1616,14 +1619,21 @@ static void blk_backing_dev_unplug(struc
- /*
- * 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);
- }
-
-@@ -1631,6 +1641,9 @@ static void blk_unplug_timeout(unsigned
- {
- 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 @@ void blk_cleanup_queue(request_queue_t *
- if (q->queue_tags)
- __blk_queue_free_tags(q);
-
-+ if (q->blk_trace)
-+ blk_trace_shutdown(q);
-+
- kmem_cache_free(requestq_cachep, q);
- }
-
-@@ -2104,6 +2120,8 @@ rq_starved:
-
- rq_init(q, rq);
- rq->rl = rl;
-+
-+ blk_add_trace_generic(q, bio, rw, BLK_TA_GETRQ);
- out:
- return rq;
- }
-@@ -2132,6 +2150,8 @@ static struct request *get_request_wait(
- 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();
-@@ -2185,6 +2205,8 @@ 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);
-
-@@ -2819,6 +2841,8 @@ static int __make_request(request_queue_
- 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;
-@@ -2834,6 +2858,8 @@ static int __make_request(request_queue_
- 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;
-
-@@ -2951,6 +2977,7 @@ void generic_make_request(struct bio *bi
- 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. */
-@@ -2977,6 +3004,8 @@ void generic_make_request(struct bio *bi
- * 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];
-
-@@ -3009,6 +3038,15 @@ end_io:
- */
- 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);
- }
-@@ -3128,6 +3166,8 @@ static int __end_that_request_first(stru
- 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
- */
-diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
-index 0d65394..234048e 100644
---- a/drivers/block/cciss.c
-+++ b/drivers/block/cciss.c
-@@ -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( ctl
-
- 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);
- }
-
-diff --git a/drivers/md/dm.c b/drivers/md/dm.c
-index e9adeb9..c8f3aa2 100644
---- a/drivers/md/dm.c
-+++ b/drivers/md/dm.c
-@@ -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
- /* 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 *
- struct target_io *tio)
- {
- int r;
-+ sector_t sector;
-
- /*
- * Sanity checks.
-@@ -407,10 +411,17 @@ static void __map_bio(struct dm_target *
- * 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 */
-diff --git a/fs/bio.c b/fs/bio.c
-index 1f3bb50..0dd0d81 100644
---- a/fs/bio.c
-+++ b/fs/bio.c
-@@ -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 *b
- 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);
-diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
-index 057e602..893d600 100644
---- a/fs/compat_ioctl.c
-+++ b/fs/compat_ioctl.c
-@@ -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>
-diff --git a/fs/relayfs/buffers.c b/fs/relayfs/buffers.c
-index 1018781..88ed3be 100644
---- a/fs/relayfs/buffers.c
-+++ b/fs/relayfs/buffers.c
-@@ -90,15 +90,16 @@ int relay_mmap_buf(struct rchan_buf *buf
- * @buf: the buffer struct
- * @size: total size of the buffer
- *
-- * Returns a pointer to the resulting buffer, NULL if unsuccessful
-+ * Returns a pointer to the resulting buffer, NULL if unsuccessful. The
-+ * passed in size will get page aligned, if it isn't already.
- */
--static void *relay_alloc_buf(struct rchan_buf *buf, unsigned long size)
-+static void *relay_alloc_buf(struct rchan_buf *buf, unsigned long *size)
- {
- void *mem;
- unsigned int i, j, n_pages;
-
-- size = PAGE_ALIGN(size);
-- n_pages = size >> PAGE_SHIFT;
-+ *size = PAGE_ALIGN(*size);
-+ n_pages = *size >> PAGE_SHIFT;
-
- buf->page_array = kcalloc(n_pages, sizeof(struct page *), GFP_KERNEL);
- if (!buf->page_array)
-@@ -113,7 +114,7 @@ static void *relay_alloc_buf(struct rcha
- if (!mem)
- goto depopulate;
-
-- memset(mem, 0, size);
-+ memset(mem, 0, *size);
- buf->page_count = n_pages;
- return mem;
-
-@@ -141,7 +142,7 @@ struct rchan_buf *relay_create_buf(struc
- if (!buf->padding)
- goto free_buf;
-
-- buf->start = relay_alloc_buf(buf, chan->alloc_size);
-+ buf->start = relay_alloc_buf(buf, &chan->alloc_size);
- if (!buf->start)
- goto free_buf;
-
-diff --git a/fs/relayfs/inode.c b/fs/relayfs/inode.c
-index 3835230..4e498c6 100644
---- a/fs/relayfs/inode.c
-+++ b/fs/relayfs/inode.c
-@@ -293,7 +293,8 @@ static unsigned int relay_file_poll(stru
- if (filp->f_mode & FMODE_READ) {
- poll_wait(filp, &buf->read_wait, wait);
- if (!relay_buf_empty(buf))
-- mask |= POLLIN | POLLRDNORM;
-+ mask |= POLLIN | POLLRDNORM | buf->poll_flags;
-+ buf->poll_flags = 0;
- }
-
- return mask;
-@@ -347,37 +348,33 @@ static void relay_file_read_consume(stru
- */
- static int relay_file_read_avail(struct rchan_buf *buf, size_t read_pos)
- {
-- size_t bytes_produced, bytes_consumed, write_offset;
- size_t subbuf_size = buf->chan->subbuf_size;
- size_t n_subbufs = buf->chan->n_subbufs;
-- size_t produced = buf->subbufs_produced % n_subbufs;
-- size_t consumed = buf->subbufs_consumed % n_subbufs;
-+ size_t produced = buf->subbufs_produced;
-+ size_t consumed = buf->subbufs_consumed;
-
-- write_offset = buf->offset > subbuf_size ? subbuf_size : buf->offset;
-+ relay_file_read_consume(buf, read_pos, 0);
-
-- if (consumed > produced) {
-- if ((produced > n_subbufs) &&
-- (produced + n_subbufs - consumed <= n_subbufs))
-- produced += n_subbufs;
-- } else if (consumed == produced) {
-- if (buf->offset > subbuf_size) {
-- produced += n_subbufs;
-- if (buf->subbufs_produced == buf->subbufs_consumed)
-- consumed += n_subbufs;
-- }
-+ if (unlikely(buf->offset > subbuf_size)) {
-+ if (produced == consumed)
-+ return 0;
-+ return 1;
- }
-
-- if (buf->offset > subbuf_size)
-- bytes_produced = (produced - 1) * subbuf_size + write_offset;
-- else
-- bytes_produced = produced * subbuf_size + write_offset;
-- bytes_consumed = consumed * subbuf_size + buf->bytes_consumed;
--
-- if (bytes_produced == bytes_consumed)
-+ if (unlikely(produced - consumed >= n_subbufs)) {
-+ consumed = (produced / n_subbufs) * n_subbufs;
-+ buf->subbufs_consumed = consumed;
-+ }
-+
-+ produced = (produced % n_subbufs) * subbuf_size + buf->offset;
-+ consumed = (consumed % n_subbufs) * subbuf_size + buf->bytes_consumed;
-+
-+ if (consumed > produced)
-+ produced += n_subbufs * subbuf_size;
-+
-+ if (consumed == produced)
- return 0;
-
-- relay_file_read_consume(buf, read_pos, 0);
--
- return 1;
- }
-
-@@ -498,6 +495,91 @@ out:
- return ret;
- }
-
-+static ssize_t relay_file_sendsubbuf(struct file *filp, loff_t *ppos,
-+ size_t count, read_actor_t actor,
-+ void *target)
-+{
-+ struct rchan_buf *buf = filp->private_data;
-+ read_descriptor_t desc;
-+ size_t read_start, avail;
-+ unsigned long pidx, poff;
-+ unsigned int subbuf_pages;
-+ ssize_t ret = 0;
-+
-+ if (!relay_file_read_avail(buf, *ppos))
-+ return 0;
-+
-+ read_start = relay_file_read_start_pos(*ppos, buf);
-+ avail = relay_file_read_subbuf_avail(read_start, buf);
-+ if (!avail)
-+ return 0;
-+
-+ count = min(count, avail);
-+
-+ desc.written = 0;
-+ desc.count = count;
-+ desc.arg.data = target;
-+ desc.error = 0;
-+
-+ subbuf_pages = buf->chan->alloc_size >> PAGE_SHIFT;
-+ pidx = (read_start / PAGE_SIZE) % subbuf_pages;
-+ poff = read_start & ~PAGE_MASK;
-+ while (count) {
-+ struct page *p = buf->page_array[pidx];
-+ unsigned int len;
-+
-+ len = PAGE_SIZE - poff;
-+ if (len > count)
-+ len = count;
-+
-+ len = actor(&desc, p, poff, len);
-+
-+ if (desc.error) {
-+ if (!ret)
-+ ret = desc.error;
-+ break;
-+ }
-+
-+ count -= len;
-+ ret += len;
-+ poff = 0;
-+ pidx = (pidx + 1) % subbuf_pages;
-+ }
-+
-+ if (ret > 0) {
-+ relay_file_read_consume(buf, read_start, ret);
-+ *ppos = relay_file_read_end_pos(buf, read_start, ret);
-+ }
-+
-+ return ret;
-+}
-+
-+static ssize_t relay_file_sendfile(struct file *filp, loff_t *ppos,
-+ size_t count, read_actor_t actor,
-+ void *target)
-+{
-+ ssize_t sent = 0, ret = 0;
-+
-+ if (!count)
-+ return 0;
-+
-+ mutex_lock(&filp->f_dentry->d_inode->i_mutex);
-+
-+ do {
-+ ret = relay_file_sendsubbuf(filp, ppos, count, actor, target);
-+ if (ret < 0) {
-+ if (!sent)
-+ sent = ret;
-+ break;
-+ }
-+ count -= ret;
-+ sent += ret;
-+ } while (count && ret);
-+
-+ mutex_unlock(&filp->f_dentry->d_inode->i_mutex);
-+ return sent;
-+}
-+
- struct file_operations relay_file_operations = {
- .open = relay_file_open,
- .poll = relay_file_poll,
-@@ -505,6 +587,7 @@ struct file_operations relay_file_operat
- .read = relay_file_read,
- .llseek = no_llseek,
- .release = relay_file_release,
-+ .sendfile = relay_file_sendfile,
- };
-
- static struct super_operations relayfs_ops = {
-diff --git a/fs/relayfs/relay.c b/fs/relayfs/relay.c
-index abf3cea..7969dcc 100644
---- a/fs/relayfs/relay.c
-+++ b/fs/relayfs/relay.c
-@@ -353,6 +353,10 @@ size_t relay_switch_subbuf(struct rchan_
- old_subbuf = buf->subbufs_produced % buf->chan->n_subbufs;
- buf->padding[old_subbuf] = buf->prev_padding;
- buf->subbufs_produced++;
-+ buf->dentry->d_inode->i_size += buf->chan->subbuf_size -
-+ buf->padding[old_subbuf];
-+ buf->poll_flags = POLLMSG;
-+ smp_mb();
- if (waitqueue_active(&buf->read_wait)) {
- PREPARE_WORK(&buf->wake_readers, wakeup_readers, buf);
- schedule_delayed_work(&buf->wake_readers, 1);
-diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
-index 860e7a4..266ce9d 100644
---- a/include/linux/blkdev.h
-+++ b/include/linux/blkdev.h
-@@ -22,6 +22,7 @@ typedef struct request_queue request_que
- 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
- */
-diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h
-new file mode 100644
-index 0000000..86479af
---- /dev/null
-+++ b/include/linux/blktrace_api.h
-@@ -0,0 +1,277 @@
-+#ifndef BLKTRACE_H
-+#define BLKTRACE_H
-+
-+#include <linux/config.h>
-+#include <linux/blkdev.h>
-+#include <linux/relayfs_fs.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
-diff --git a/include/linux/compat_ioctl.h b/include/linux/compat_ioctl.h
-index 8fad50f..a3ed64b 100644
---- a/include/linux/compat_ioctl.h
-+++ b/include/linux/compat_ioctl.h
-@@ -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 */
-diff --git a/include/linux/fs.h b/include/linux/fs.h
-index e059da9..c7a63cd 100644
---- a/include/linux/fs.h
-+++ b/include/linux/fs.h
-@@ -196,6 +196,10 @@ 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 */
-diff --git a/include/linux/relayfs_fs.h b/include/linux/relayfs_fs.h
-index 7342e66..10aa201 100644
---- a/include/linux/relayfs_fs.h
-+++ b/include/linux/relayfs_fs.h
-@@ -46,6 +46,7 @@ struct rchan_buf
- size_t prev_padding; /* temporary variable */
- size_t bytes_consumed; /* bytes consumed in cur read subbuf */
- unsigned int cpu; /* this buf's cpu */
-+ unsigned int poll_flags;
- } ____cacheline_aligned;
-
- /*
-diff --git a/include/linux/sched.h b/include/linux/sched.h
-index b6f51e3..d9d465b 100644
---- a/include/linux/sched.h
-+++ b/include/linux/sched.h
-@@ -705,6 +705,7 @@ struct task_struct {
- prio_array_t *array;
-
- unsigned short ioprio;
-+ unsigned int btrace_seq;
-
- unsigned long sleep_avg;
- unsigned long long timestamp, last_ran;
-diff --git a/kernel/fork.c b/kernel/fork.c
-index fbea12d..f564e9b 100644
---- a/kernel/fork.c
-+++ b/kernel/fork.c
-@@ -179,6 +179,7 @@ static struct task_struct *dup_task_stru
- /* 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;
- }
-
-diff --git a/mm/highmem.c b/mm/highmem.c
-index ce2e7e8..d0ea1ee 100644
---- a/mm/highmem.c
-+++ b/mm/highmem.c
-@@ -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
- pool = isa_page_pool;
- }
-
-+ blk_add_trace_bio(q, *bio_orig, BLK_TA_BOUNCE);
-+
- /*
- * slow path
- */
--- /dev/null
+diff --git a/block/Kconfig b/block/Kconfig
+index 377f6dd..9678364 100644
+--- a/block/Kconfig
++++ b/block/Kconfig
+@@ -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
+diff --git a/block/Makefile b/block/Makefile
+index 7e4f93e..c05de0e 100644
+--- a/block/Makefile
++++ b/block/Makefile
+@@ -8,3 +8,5 @@ obj-$(CONFIG_IOSCHED_NOOP) += noop-iosch
+ 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
+diff --git a/block/blktrace.c b/block/blktrace.c
+new file mode 100644
+index 0000000..7c9ac33
+--- /dev/null
++++ b/block/blktrace.c
+@@ -0,0 +1,533 @@
++/*
++ * Copyright (C) 2006 Jens Axboe <axboe@suse.de>
++ *
++ * This program is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License version 2 as
++ * published by the Free Software Foundation.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program; if not, write to the Free Software
++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ *
++ */
++#include <linux/config.h>
++#include <linux/kernel.h>
++#include <linux/blkdev.h>
++#include <linux/blktrace_api.h>
++#include <linux/percpu.h>
++#include <linux/init.h>
++#include <linux/mutex.h>
++#include <linux/debugfs.h>
++#include <asm/uaccess.h>
++
++static DEFINE_PER_CPU(unsigned long long, blk_trace_cpu_offset) = { 0, };
++static unsigned int blktrace_seq __read_mostly = 1;
++
++/*
++ * Send out a notify for this process, if we haven't done so since a trace
++ * started
++ */
++static void trace_note_tsk(struct blk_trace *bt, struct task_struct *tsk)
++{
++ struct blk_io_trace *t;
++
++ t = relay_reserve(bt->rchan, sizeof(*t) + sizeof(tsk->comm));
++ if (t) {
++ t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION;
++ t->device = bt->dev;
++ t->action = BLK_TC_ACT(BLK_TC_NOTIFY);
++ t->pid = tsk->pid;
++ t->cpu = smp_processor_id();
++ t->pdu_len = sizeof(tsk->comm);
++ memcpy((void *) t + sizeof(*t), tsk->comm, t->pdu_len);
++ tsk->btrace_seq = blktrace_seq;
++ }
++}
++
++static int act_log_check(struct blk_trace *bt, u32 what, sector_t sector,
++ pid_t pid)
++{
++ if (((bt->act_mask << BLK_TC_SHIFT) & what) == 0)
++ return 1;
++ if (sector < bt->start_lba || sector > bt->end_lba)
++ return 1;
++ if (bt->pid && pid != bt->pid)
++ return 1;
++
++ return 0;
++}
++
++/*
++ * Data direction bit lookup
++ */
++static u32 ddir_act[2] __read_mostly = { BLK_TC_ACT(BLK_TC_READ), BLK_TC_ACT(BLK_TC_WRITE) };
++
++/*
++ * Bio action bits of interest
++ */
++static u32 bio_act[3] __read_mostly = { 0, BLK_TC_ACT(BLK_TC_BARRIER), BLK_TC_ACT(BLK_TC_SYNC) };
++
++/*
++ * More could be added as needed, taking care to increment the decrementer
++ * to get correct indexing
++ */
++#define trace_barrier_bit(rw) \
++ (((rw) & (1 << BIO_RW_BARRIER)) >> (BIO_RW_BARRIER - 0))
++#define trace_sync_bit(rw) \
++ (((rw) & (1 << BIO_RW_SYNC)) >> (BIO_RW_SYNC - 1))
++
++/*
++ * The worker for the various blk_add_trace*() types. Fills out a
++ * blk_io_trace structure and places it in a per-cpu subbuffer.
++ */
++void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes,
++ int rw, u32 what, int error, int pdu_len, void *pdu_data)
++{
++ struct task_struct *tsk = current;
++ struct blk_io_trace *t;
++ unsigned long flags;
++ unsigned long *sequence;
++ pid_t pid;
++ int cpu;
++
++ if (unlikely(bt->trace_state != Blktrace_running))
++ return;
++
++ what |= ddir_act[rw & WRITE];
++ what |= bio_act[trace_barrier_bit(rw)];
++ what |= bio_act[trace_sync_bit(rw)];
++
++ pid = tsk->pid;
++ if (unlikely(act_log_check(bt, what, sector, pid)))
++ return;
++
++ /*
++ * A word about the locking here - we disable interrupts to reserve
++ * some space in the relay per-cpu buffer, to prevent an irq
++ * from coming in and stepping on our toes. Once reserved, it's
++ * enough to get preemption disabled to prevent read of this data
++ * before we are through filling it. get_cpu()/put_cpu() does this
++ * for us
++ */
++ local_irq_save(flags);
++
++ if (unlikely(tsk->btrace_seq != blktrace_seq))
++ trace_note_tsk(bt, tsk);
++
++ t = relay_reserve(bt->rchan, sizeof(*t) + pdu_len);
++ if (t) {
++ cpu = smp_processor_id();
++ sequence = per_cpu_ptr(bt->sequence, cpu);
++
++ t->magic = BLK_IO_TRACE_MAGIC | BLK_IO_TRACE_VERSION;
++ t->sequence = ++(*sequence);
++ t->time = sched_clock() - per_cpu(blk_trace_cpu_offset, cpu);
++ t->sector = sector;
++ t->bytes = bytes;
++ t->action = what;
++ t->pid = pid;
++ t->device = bt->dev;
++ t->cpu = cpu;
++ t->error = error;
++ t->pdu_len = pdu_len;
++
++ if (pdu_len)
++ memcpy((void *) t + sizeof(*t), pdu_data, pdu_len);
++ }
++
++ local_irq_restore(flags);
++}
++
++EXPORT_SYMBOL_GPL(__blk_add_trace);
++
++static struct dentry *blk_tree_root;
++static struct mutex blk_tree_mutex;
++
++static inline void blk_remove_root(void)
++{
++ if (blk_tree_root) {
++ debugfs_remove(blk_tree_root);
++ blk_tree_root = NULL;
++ }
++}
++
++static void blk_remove_tree(struct dentry *dir)
++{
++ mutex_lock(&blk_tree_mutex);
++ debugfs_remove(dir);
++ blk_remove_root();
++ mutex_unlock(&blk_tree_mutex);
++}
++
++static struct dentry *blk_create_tree(const char *blk_name)
++{
++ struct dentry *dir = NULL;
++
++ mutex_lock(&blk_tree_mutex);
++
++ if (!blk_tree_root) {
++ blk_tree_root = debugfs_create_dir("block", NULL);
++ if (!blk_tree_root)
++ goto err;
++ }
++
++ dir = debugfs_create_dir(blk_name, blk_tree_root);
++ if (!dir)
++ blk_remove_root();
++
++err:
++ mutex_unlock(&blk_tree_mutex);
++ return dir;
++}
++
++static void blk_trace_cleanup(struct blk_trace *bt)
++{
++ relay_close(bt->rchan);
++ debugfs_remove(bt->dropped_file);
++ blk_remove_tree(bt->dir);
++ free_percpu(bt->sequence);
++ kfree(bt);
++}
++
++static int blk_trace_remove(request_queue_t *q)
++{
++ struct blk_trace *bt;
++
++ bt = xchg(&q->blk_trace, NULL);
++ if (!bt)
++ return -EINVAL;
++
++ if (bt->trace_state == Blktrace_setup ||
++ bt->trace_state == Blktrace_stopped)
++ blk_trace_cleanup(bt);
++
++ return 0;
++}
++
++static int blk_dropped_open(struct inode *inode, struct file *filp)
++{
++ filp->private_data = inode->u.generic_ip;
++
++ return 0;
++}
++
++static ssize_t blk_dropped_read(struct file *filp, char __user *buffer,
++ size_t count, loff_t *ppos)
++{
++ struct blk_trace *bt = filp->private_data;
++ char buf[16];
++
++ snprintf(buf, sizeof(buf), "%u\n", atomic_read(&bt->dropped));
++
++ return simple_read_from_buffer(buffer, count, ppos, buf, strlen(buf));
++}
++
++static struct file_operations blk_dropped_fops = {
++ .owner = THIS_MODULE,
++ .open = blk_dropped_open,
++ .read = blk_dropped_read,
++};
++
++/*
++ * Keep track of how many times we encountered a full subbuffer, to aid
++ * the user space app in telling how many lost events there were.
++ */
++static int blk_subbuf_start_callback(struct rchan_buf *buf, void *subbuf,
++ void *prev_subbuf, size_t prev_padding)
++{
++ struct blk_trace *bt;
++
++ if (!relay_buf_full(buf))
++ return 1;
++
++ bt = buf->chan->private_data;
++ atomic_inc(&bt->dropped);
++ return 0;
++}
++
++static int blk_remove_buf_file_callback(struct dentry *dentry)
++{
++ debugfs_remove(dentry);
++ return 0;
++}
++
++static struct dentry *blk_create_buf_file_callback(const char *filename,
++ struct dentry *parent,
++ int mode,
++ struct rchan_buf *buf,
++ int *is_global)
++{
++ return debugfs_create_file(filename, mode, parent, buf,
++ &relay_file_operations);
++}
++
++static struct rchan_callbacks blk_relay_callbacks = {
++ .subbuf_start = blk_subbuf_start_callback,
++ .create_buf_file = blk_create_buf_file_callback,
++ .remove_buf_file = blk_remove_buf_file_callback,
++};
++
++/*
++ * Setup everything required to start tracing
++ */
++static int blk_trace_setup(request_queue_t *q, struct block_device *bdev,
++ char __user *arg)
++{
++ struct blk_user_trace_setup buts;
++ struct blk_trace *old_bt, *bt = NULL;
++ struct dentry *dir = NULL;
++ char b[BDEVNAME_SIZE];
++ int ret, i;
++
++ if (copy_from_user(&buts, arg, sizeof(buts)))
++ return -EFAULT;
++
++ if (!buts.buf_size || !buts.buf_nr)
++ return -EINVAL;
++
++ strcpy(buts.name, bdevname(bdev, b));
++
++ /*
++ * some device names have larger paths - convert the slashes
++ * to underscores for this to work as expected
++ */
++ for (i = 0; i < strlen(buts.name); i++)
++ if (buts.name[i] == '/')
++ buts.name[i] = '_';
++
++ if (copy_to_user(arg, &buts, sizeof(buts)))
++ return -EFAULT;
++
++ ret = -ENOMEM;
++ bt = kzalloc(sizeof(*bt), GFP_KERNEL);
++ if (!bt)
++ goto err;
++
++ bt->sequence = alloc_percpu(unsigned long);
++ if (!bt->sequence)
++ goto err;
++
++ ret = -ENOENT;
++ dir = blk_create_tree(buts.name);
++ if (!dir)
++ goto err;
++
++ bt->dir = dir;
++ bt->dev = bdev->bd_dev;
++ atomic_set(&bt->dropped, 0);
++
++ ret = -EIO;
++ bt->dropped_file = debugfs_create_file("dropped", 0444, dir, bt, &blk_dropped_fops);
++ if (!bt->dropped_file)
++ goto err;
++
++ bt->rchan = relay_open("trace", dir, buts.buf_size, buts.buf_nr, &blk_relay_callbacks);
++ if (!bt->rchan)
++ goto err;
++ bt->rchan->private_data = bt;
++
++ bt->act_mask = buts.act_mask;
++ if (!bt->act_mask)
++ bt->act_mask = (u16) -1;
++
++ bt->start_lba = buts.start_lba;
++ bt->end_lba = buts.end_lba;
++ if (!bt->end_lba)
++ bt->end_lba = -1ULL;
++
++ bt->pid = buts.pid;
++ bt->trace_state = Blktrace_setup;
++
++ ret = -EBUSY;
++ old_bt = xchg(&q->blk_trace, bt);
++ if (old_bt) {
++ xchg(&q->blk_trace, old_bt);
++ goto err;
++ }
++
++ return 0;
++err:
++ if (dir)
++ blk_remove_tree(dir);
++ if (bt) {
++ if (bt->dropped_file)
++ debugfs_remove(bt->dropped_file);
++ if (bt->sequence)
++ free_percpu(bt->sequence);
++ if (bt->rchan)
++ relay_close(bt->rchan);
++ kfree(bt);
++ }
++ return ret;
++}
++
++static int blk_trace_startstop(request_queue_t *q, int start)
++{
++ struct blk_trace *bt;
++ int ret;
++
++ if ((bt = q->blk_trace) == NULL)
++ return -EINVAL;
++
++ /*
++ * For starting a trace, we can transition from a setup or stopped
++ * trace. For stopping a trace, the state must be running
++ */
++ ret = -EINVAL;
++ if (start) {
++ if (bt->trace_state == Blktrace_setup ||
++ bt->trace_state == Blktrace_stopped) {
++ blktrace_seq++;
++ smp_mb();
++ bt->trace_state = Blktrace_running;
++ ret = 0;
++ }
++ } else {
++ if (bt->trace_state == Blktrace_running) {
++ bt->trace_state = Blktrace_stopped;
++ ret = 0;
++ }
++ }
++
++ return ret;
++}
++
++/**
++ * blk_trace_ioctl: - handle the ioctls associated with tracing
++ * @bdev: the block device
++ * @cmd: the ioctl cmd
++ * @arg: the argument data, if any
++ *
++ **/
++int blk_trace_ioctl(struct block_device *bdev, unsigned cmd, char __user *arg)
++{
++ request_queue_t *q;
++ int ret, start = 0;
++
++ q = bdev_get_queue(bdev);
++ if (!q)
++ return -ENXIO;
++
++ down(&bdev->bd_sem);
++
++ switch (cmd) {
++ case BLKTRACESETUP:
++ ret = blk_trace_setup(q, bdev, arg);
++ break;
++ case BLKTRACESTART:
++ start = 1;
++ case BLKTRACESTOP:
++ ret = blk_trace_startstop(q, start);
++ break;
++ case BLKTRACETEARDOWN:
++ ret = blk_trace_remove(q);
++ break;
++ default:
++ ret = -ENOTTY;
++ break;
++ }
++
++ up(&bdev->bd_sem);
++ return ret;
++}
++
++/**
++ * blk_trace_shutdown: - stop and cleanup trace structures
++ * @q: the request queue associated with the device
++ *
++ **/
++void blk_trace_shutdown(request_queue_t *q)
++{
++ blk_trace_startstop(q, 0);
++ blk_trace_remove(q);
++}
++
++/*
++ * Average offset over two calls to sched_clock() with a gettimeofday()
++ * in the middle
++ */
++static void blk_check_time(unsigned long long *t)
++{
++ unsigned long long a, b;
++ struct timeval tv;
++
++ a = sched_clock();
++ do_gettimeofday(&tv);
++ b = sched_clock();
++
++ *t = tv.tv_sec * 1000000000 + tv.tv_usec * 1000;
++ *t -= (a + b) / 2;
++}
++
++static void blk_trace_check_cpu_time(void *data)
++{
++ unsigned long long *t;
++ int cpu = get_cpu();
++
++ t = &per_cpu(blk_trace_cpu_offset, cpu);
++
++ /*
++ * Just call it twice, hopefully the second call will be cache hot
++ * and a little more precise
++ */
++ blk_check_time(t);
++ blk_check_time(t);
++
++ put_cpu();
++}
++
++/*
++ * Call blk_trace_check_cpu_time() on each CPU to calibrate our inter-CPU
++ * timings
++ */
++static void blk_trace_calibrate_offsets(void)
++{
++ unsigned long flags;
++
++ smp_call_function(blk_trace_check_cpu_time, NULL, 1, 1);
++ local_irq_save(flags);
++ blk_trace_check_cpu_time(NULL);
++ local_irq_restore(flags);
++}
++
++static void blk_trace_set_ht_offsets(void)
++{
++#if defined(CONFIG_SCHED_SMT)
++ int cpu, i;
++
++ /*
++ * now make sure HT siblings have the same time offset
++ */
++ preempt_disable();
++ for_each_online_cpu(cpu) {
++ unsigned long long *cpu_off, *sibling_off;
++
++ for_each_cpu_mask(i, cpu_sibling_map[cpu]) {
++ if (i == cpu)
++ continue;
++
++ cpu_off = &per_cpu(blk_trace_cpu_offset, cpu);
++ sibling_off = &per_cpu(blk_trace_cpu_offset, i);
++ *sibling_off = *cpu_off;
++ }
++ }
++ preempt_enable();
++#endif
++}
++
++static __init int blk_trace_init(void)
++{
++ mutex_init(&blk_tree_mutex);
++ blk_trace_calibrate_offsets();
++ blk_trace_set_ht_offsets();
++
++ return 0;
++}
++
++module_init(blk_trace_init);
++
+diff --git a/block/elevator.c b/block/elevator.c
+index 24b702d..0c9fafe 100644
+--- a/block/elevator.c
++++ b/block/elevator.c
+@@ -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>
+
+@@ -315,6 +316,8 @@ void elv_insert(request_queue_t *q, stru
+ struct list_head *pos;
+ unsigned ordseq;
+
++ blk_add_trace_rq(q, rq, BLK_TA_INSERT);
++
+ rq->q = q;
+
+ switch (where) {
+@@ -481,6 +484,7 @@ struct request *elv_next_request(request
+ * 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) {
+diff --git a/block/ioctl.c b/block/ioctl.c
+index e110949..7acb56c 100644
+--- a/block/ioctl.c
++++ b/block/ioctl.c
+@@ -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 fi
+ 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;
+ }
+diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c
+index 03d9c82..41387f5 100644
+--- a/block/ll_rw_blk.c
++++ b/block/ll_rw_blk.c
+@@ -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
+@@ -1551,8 +1552,10 @@ 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);
+@@ -1616,14 +1619,21 @@ static void blk_backing_dev_unplug(struc
+ /*
+ * 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);
+ }
+
+@@ -1631,6 +1641,9 @@ static void blk_unplug_timeout(unsigned
+ {
+ 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 @@ void blk_cleanup_queue(request_queue_t *
+ if (q->queue_tags)
+ __blk_queue_free_tags(q);
+
++ if (q->blk_trace)
++ blk_trace_shutdown(q);
++
+ kmem_cache_free(requestq_cachep, q);
+ }
+
+@@ -2104,6 +2120,8 @@ rq_starved:
+
+ rq_init(q, rq);
+ rq->rl = rl;
++
++ blk_add_trace_generic(q, bio, rw, BLK_TA_GETRQ);
+ out:
+ return rq;
+ }
+@@ -2132,6 +2150,8 @@ static struct request *get_request_wait(
+ 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();
+@@ -2185,6 +2205,8 @@ 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);
+
+@@ -2819,6 +2841,8 @@ static int __make_request(request_queue_
+ 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;
+@@ -2834,6 +2858,8 @@ static int __make_request(request_queue_
+ 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;
+
+@@ -2951,6 +2977,7 @@ void generic_make_request(struct bio *bi
+ 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. */
+@@ -2977,6 +3004,8 @@ void generic_make_request(struct bio *bi
+ * 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];
+
+@@ -3009,6 +3038,15 @@ end_io:
+ */
+ 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);
+ }
+@@ -3128,6 +3166,8 @@ static int __end_that_request_first(stru
+ 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
+ */
+diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
+index 0d65394..234048e 100644
+--- a/drivers/block/cciss.c
++++ b/drivers/block/cciss.c
+@@ -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( ctl
+
+ 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);
+ }
+
+diff --git a/drivers/md/dm.c b/drivers/md/dm.c
+index e9adeb9..c8f3aa2 100644
+--- a/drivers/md/dm.c
++++ b/drivers/md/dm.c
+@@ -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
+ /* 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 *
+ struct target_io *tio)
+ {
+ int r;
++ sector_t sector;
+
+ /*
+ * Sanity checks.
+@@ -407,10 +411,17 @@ static void __map_bio(struct dm_target *
+ * 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 */
+diff --git a/fs/Kconfig b/fs/Kconfig
+index e9749b0..c8d0a20 100644
+--- a/fs/Kconfig
++++ b/fs/Kconfig
+@@ -859,18 +859,6 @@ config RAMFS
+ To compile this as a module, choose M here: the module will be called
+ ramfs.
+
+-config RELAYFS_FS
+- tristate "Relayfs file system support"
+- ---help---
+- Relayfs is a high-speed data relay filesystem designed to provide
+- an efficient mechanism for tools and facilities to relay large
+- amounts of data from kernel space to user space.
+-
+- To compile this code as a module, choose M here: the module will be
+- called relayfs.
+-
+- If unsure, say N.
+-
+ config CONFIGFS_FS
+ tristate "Userspace-driven configuration filesystem (EXPERIMENTAL)"
+ depends on EXPERIMENTAL
+diff --git a/fs/Makefile b/fs/Makefile
+index 1db7113..080b386 100644
+--- a/fs/Makefile
++++ b/fs/Makefile
+@@ -91,7 +91,6 @@ obj-$(CONFIG_AUTOFS4_FS) += autofs4/
+ obj-$(CONFIG_ADFS_FS) += adfs/
+ obj-$(CONFIG_FUSE_FS) += fuse/
+ obj-$(CONFIG_UDF_FS) += udf/
+-obj-$(CONFIG_RELAYFS_FS) += relayfs/
+ obj-$(CONFIG_SUN_OPENPROMFS) += openpromfs/
+ obj-$(CONFIG_JFS_FS) += jfs/
+ obj-$(CONFIG_XFS_FS) += xfs/
+diff --git a/fs/bio.c b/fs/bio.c
+index 1f3bb50..0dd0d81 100644
+--- a/fs/bio.c
++++ b/fs/bio.c
+@@ -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 *b
+ 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);
+diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
+index 057e602..893d600 100644
+--- a/fs/compat_ioctl.c
++++ b/fs/compat_ioctl.c
+@@ -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>
+diff --git a/fs/relayfs/Makefile b/fs/relayfs/Makefile
+deleted file mode 100644
+index e76e182..0000000
+--- a/fs/relayfs/Makefile
++++ /dev/null
+@@ -1,4 +0,0 @@
+-obj-$(CONFIG_RELAYFS_FS) += relayfs.o
+-
+-relayfs-y := relay.o inode.o buffers.o
+-
+diff --git a/fs/relayfs/buffers.c b/fs/relayfs/buffers.c
+deleted file mode 100644
+index 1018781..0000000
+--- a/fs/relayfs/buffers.c
++++ /dev/null
+@@ -1,190 +0,0 @@
+-/*
+- * RelayFS buffer management code.
+- *
+- * Copyright (C) 2002-2005 - Tom Zanussi (zanussi@us.ibm.com), IBM Corp
+- * Copyright (C) 1999-2005 - Karim Yaghmour (karim@opersys.com)
+- *
+- * This file is released under the GPL.
+- */
+-
+-#include <linux/module.h>
+-#include <linux/vmalloc.h>
+-#include <linux/mm.h>
+-#include <linux/relayfs_fs.h>
+-#include "relay.h"
+-#include "buffers.h"
+-
+-/*
+- * close() vm_op implementation for relayfs file mapping.
+- */
+-static void relay_file_mmap_close(struct vm_area_struct *vma)
+-{
+- struct rchan_buf *buf = vma->vm_private_data;
+- buf->chan->cb->buf_unmapped(buf, vma->vm_file);
+-}
+-
+-/*
+- * nopage() vm_op implementation for relayfs file mapping.
+- */
+-static struct page *relay_buf_nopage(struct vm_area_struct *vma,
+- unsigned long address,
+- int *type)
+-{
+- struct page *page;
+- struct rchan_buf *buf = vma->vm_private_data;
+- unsigned long offset = address - vma->vm_start;
+-
+- if (address > vma->vm_end)
+- return NOPAGE_SIGBUS; /* Disallow mremap */
+- if (!buf)
+- return NOPAGE_OOM;
+-
+- page = vmalloc_to_page(buf->start + offset);
+- if (!page)
+- return NOPAGE_OOM;
+- get_page(page);
+-
+- if (type)
+- *type = VM_FAULT_MINOR;
+-
+- return page;
+-}
+-
+-/*
+- * vm_ops for relay file mappings.
+- */
+-static struct vm_operations_struct relay_file_mmap_ops = {
+- .nopage = relay_buf_nopage,
+- .close = relay_file_mmap_close,
+-};
+-
+-/**
+- * relay_mmap_buf: - mmap channel buffer to process address space
+- * @buf: relay channel buffer
+- * @vma: vm_area_struct describing memory to be mapped
+- *
+- * Returns 0 if ok, negative on error
+- *
+- * Caller should already have grabbed mmap_sem.
+- */
+-int relay_mmap_buf(struct rchan_buf *buf, struct vm_area_struct *vma)
+-{
+- unsigned long length = vma->vm_end - vma->vm_start;
+- struct file *filp = vma->vm_file;
+-
+- if (!buf)
+- return -EBADF;
+-
+- if (length != (unsigned long)buf->chan->alloc_size)
+- return -EINVAL;
+-
+- vma->vm_ops = &relay_file_mmap_ops;
+- vma->vm_private_data = buf;
+- buf->chan->cb->buf_mapped(buf, filp);
+-
+- return 0;
+-}
+-
+-/**
+- * relay_alloc_buf - allocate a channel buffer
+- * @buf: the buffer struct
+- * @size: total size of the buffer
+- *
+- * Returns a pointer to the resulting buffer, NULL if unsuccessful
+- */
+-static void *relay_alloc_buf(struct rchan_buf *buf, unsigned long size)
+-{
+- void *mem;
+- unsigned int i, j, n_pages;
+-
+- size = PAGE_ALIGN(size);
+- n_pages = size >> PAGE_SHIFT;
+-
+- buf->page_array = kcalloc(n_pages, sizeof(struct page *), GFP_KERNEL);
+- if (!buf->page_array)
+- return NULL;
+-
+- for (i = 0; i < n_pages; i++) {
+- buf->page_array[i] = alloc_page(GFP_KERNEL);
+- if (unlikely(!buf->page_array[i]))
+- goto depopulate;
+- }
+- mem = vmap(buf->page_array, n_pages, VM_MAP, PAGE_KERNEL);
+- if (!mem)
+- goto depopulate;
+-
+- memset(mem, 0, size);
+- buf->page_count = n_pages;
+- return mem;
+-
+-depopulate:
+- for (j = 0; j < i; j++)
+- __free_page(buf->page_array[j]);
+- kfree(buf->page_array);
+- return NULL;
+-}
+-
+-/**
+- * relay_create_buf - allocate and initialize a channel buffer
+- * @alloc_size: size of the buffer to allocate
+- * @n_subbufs: number of sub-buffers in the channel
+- *
+- * Returns channel buffer if successful, NULL otherwise
+- */
+-struct rchan_buf *relay_create_buf(struct rchan *chan)
+-{
+- struct rchan_buf *buf = kcalloc(1, sizeof(struct rchan_buf), GFP_KERNEL);
+- if (!buf)
+- return NULL;
+-
+- buf->padding = kmalloc(chan->n_subbufs * sizeof(size_t *), GFP_KERNEL);
+- if (!buf->padding)
+- goto free_buf;
+-
+- buf->start = relay_alloc_buf(buf, chan->alloc_size);
+- if (!buf->start)
+- goto free_buf;
+-
+- buf->chan = chan;
+- kref_get(&buf->chan->kref);
+- return buf;
+-
+-free_buf:
+- kfree(buf->padding);
+- kfree(buf);
+- return NULL;
+-}
+-
+-/**
+- * relay_destroy_buf - destroy an rchan_buf struct and associated buffer
+- * @buf: the buffer struct
+- */
+-void relay_destroy_buf(struct rchan_buf *buf)
+-{
+- struct rchan *chan = buf->chan;
+- unsigned int i;
+-
+- if (likely(buf->start)) {
+- vunmap(buf->start);
+- for (i = 0; i < buf->page_count; i++)
+- __free_page(buf->page_array[i]);
+- kfree(buf->page_array);
+- }
+- kfree(buf->padding);
+- kfree(buf);
+- kref_put(&chan->kref, relay_destroy_channel);
+-}
+-
+-/**
+- * relay_remove_buf - remove a channel buffer
+- *
+- * Removes the file from the relayfs fileystem, which also frees the
+- * rchan_buf_struct and the channel buffer. Should only be called from
+- * kref_put().
+- */
+-void relay_remove_buf(struct kref *kref)
+-{
+- struct rchan_buf *buf = container_of(kref, struct rchan_buf, kref);
+- buf->chan->cb->remove_buf_file(buf->dentry);
+- relay_destroy_buf(buf);
+-}
+diff --git a/fs/relayfs/buffers.h b/fs/relayfs/buffers.h
+deleted file mode 100644
+index 37a1249..0000000
+--- a/fs/relayfs/buffers.h
++++ /dev/null
+@@ -1,12 +0,0 @@
+-#ifndef _BUFFERS_H
+-#define _BUFFERS_H
+-
+-/* This inspired by rtai/shmem */
+-#define FIX_SIZE(x) (((x) - 1) & PAGE_MASK) + PAGE_SIZE
+-
+-extern int relay_mmap_buf(struct rchan_buf *buf, struct vm_area_struct *vma);
+-extern struct rchan_buf *relay_create_buf(struct rchan *chan);
+-extern void relay_destroy_buf(struct rchan_buf *buf);
+-extern void relay_remove_buf(struct kref *kref);
+-
+-#endif/* _BUFFERS_H */
+diff --git a/fs/relayfs/inode.c b/fs/relayfs/inode.c
+deleted file mode 100644
+index 3835230..0000000
+--- a/fs/relayfs/inode.c
++++ /dev/null
+@@ -1,581 +0,0 @@
+-/*
+- * VFS-related code for RelayFS, a high-speed data relay filesystem.
+- *
+- * Copyright (C) 2003-2005 - Tom Zanussi <zanussi@us.ibm.com>, IBM Corp
+- * Copyright (C) 2003-2005 - Karim Yaghmour <karim@opersys.com>
+- *
+- * Based on ramfs, Copyright (C) 2002 - Linus Torvalds
+- *
+- * This file is released under the GPL.
+- */
+-
+-#include <linux/module.h>
+-#include <linux/fs.h>
+-#include <linux/mount.h>
+-#include <linux/pagemap.h>
+-#include <linux/init.h>
+-#include <linux/string.h>
+-#include <linux/backing-dev.h>
+-#include <linux/namei.h>
+-#include <linux/poll.h>
+-#include <linux/relayfs_fs.h>
+-#include "relay.h"
+-#include "buffers.h"
+-
+-#define RELAYFS_MAGIC 0xF0B4A981
+-
+-static struct vfsmount * relayfs_mount;
+-static int relayfs_mount_count;
+-
+-static struct backing_dev_info relayfs_backing_dev_info = {
+- .ra_pages = 0, /* No readahead */
+- .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK,
+-};
+-
+-static struct inode *relayfs_get_inode(struct super_block *sb,
+- int mode,
+- struct file_operations *fops,
+- void *data)
+-{
+- struct inode *inode;
+-
+- inode = new_inode(sb);
+- if (!inode)
+- return NULL;
+-
+- inode->i_mode = mode;
+- inode->i_uid = 0;
+- inode->i_gid = 0;
+- inode->i_blksize = PAGE_CACHE_SIZE;
+- inode->i_blocks = 0;
+- inode->i_mapping->backing_dev_info = &relayfs_backing_dev_info;
+- inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
+- switch (mode & S_IFMT) {
+- case S_IFREG:
+- inode->i_fop = fops;
+- if (data)
+- inode->u.generic_ip = data;
+- break;
+- case S_IFDIR:
+- inode->i_op = &simple_dir_inode_operations;
+- inode->i_fop = &simple_dir_operations;
+-
+- /* directory inodes start off with i_nlink == 2 (for "." entry) */
+- inode->i_nlink++;
+- break;
+- default:
+- break;
+- }
+-
+- return inode;
+-}
+-
+-/**
+- * relayfs_create_entry - create a relayfs directory or file
+- * @name: the name of the file to create
+- * @parent: parent directory
+- * @mode: mode
+- * @fops: file operations to use for the file
+- * @data: user-associated data for this file
+- *
+- * Returns the new dentry, NULL on failure
+- *
+- * Creates a file or directory with the specifed permissions.
+- */
+-static struct dentry *relayfs_create_entry(const char *name,
+- struct dentry *parent,
+- int mode,
+- struct file_operations *fops,
+- void *data)
+-{
+- struct dentry *d;
+- struct inode *inode;
+- int error = 0;
+-
+- BUG_ON(!name || !(S_ISREG(mode) || S_ISDIR(mode)));
+-
+- error = simple_pin_fs("relayfs", &relayfs_mount, &relayfs_mount_count);
+- if (error) {
+- printk(KERN_ERR "Couldn't mount relayfs: errcode %d\n", error);
+- return NULL;
+- }
+-
+- if (!parent && relayfs_mount && relayfs_mount->mnt_sb)
+- parent = relayfs_mount->mnt_sb->s_root;
+-
+- if (!parent) {
+- simple_release_fs(&relayfs_mount, &relayfs_mount_count);
+- return NULL;
+- }
+-
+- parent = dget(parent);
+- mutex_lock(&parent->d_inode->i_mutex);
+- d = lookup_one_len(name, parent, strlen(name));
+- if (IS_ERR(d)) {
+- d = NULL;
+- goto release_mount;
+- }
+-
+- if (d->d_inode) {
+- d = NULL;
+- goto release_mount;
+- }
+-
+- inode = relayfs_get_inode(parent->d_inode->i_sb, mode, fops, data);
+- if (!inode) {
+- d = NULL;
+- goto release_mount;
+- }
+-
+- d_instantiate(d, inode);
+- dget(d); /* Extra count - pin the dentry in core */
+-
+- if (S_ISDIR(mode))
+- parent->d_inode->i_nlink++;
+-
+- goto exit;
+-
+-release_mount:
+- simple_release_fs(&relayfs_mount, &relayfs_mount_count);
+-
+-exit:
+- mutex_unlock(&parent->d_inode->i_mutex);
+- dput(parent);
+- return d;
+-}
+-
+-/**
+- * relayfs_create_file - create a file in the relay filesystem
+- * @name: the name of the file to create
+- * @parent: parent directory
+- * @mode: mode, if not specied the default perms are used
+- * @fops: file operations to use for the file
+- * @data: user-associated data for this file
+- *
+- * Returns file dentry if successful, NULL otherwise.
+- *
+- * The file will be created user r on behalf of current user.
+- */
+-struct dentry *relayfs_create_file(const char *name,
+- struct dentry *parent,
+- int mode,
+- struct file_operations *fops,
+- void *data)
+-{
+- BUG_ON(!fops);
+-
+- if (!mode)
+- mode = S_IRUSR;
+- mode = (mode & S_IALLUGO) | S_IFREG;
+-
+- return relayfs_create_entry(name, parent, mode, fops, data);
+-}
+-
+-/**
+- * relayfs_create_dir - create a directory in the relay filesystem
+- * @name: the name of the directory to create
+- * @parent: parent directory, NULL if parent should be fs root
+- *
+- * Returns directory dentry if successful, NULL otherwise.
+- *
+- * The directory will be created world rwx on behalf of current user.
+- */
+-struct dentry *relayfs_create_dir(const char *name, struct dentry *parent)
+-{
+- int mode = S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO;
+- return relayfs_create_entry(name, parent, mode, NULL, NULL);
+-}
+-
+-/**
+- * relayfs_remove - remove a file or directory in the relay filesystem
+- * @dentry: file or directory dentry
+- *
+- * Returns 0 if successful, negative otherwise.
+- */
+-int relayfs_remove(struct dentry *dentry)
+-{
+- struct dentry *parent;
+- int error = 0;
+-
+- if (!dentry)
+- return -EINVAL;
+- parent = dentry->d_parent;
+- if (!parent)
+- return -EINVAL;
+-
+- parent = dget(parent);
+- mutex_lock(&parent->d_inode->i_mutex);
+- if (dentry->d_inode) {
+- if (S_ISDIR(dentry->d_inode->i_mode))
+- error = simple_rmdir(parent->d_inode, dentry);
+- else
+- error = simple_unlink(parent->d_inode, dentry);
+- if (!error)
+- d_delete(dentry);
+- }
+- if (!error)
+- dput(dentry);
+- mutex_unlock(&parent->d_inode->i_mutex);
+- dput(parent);
+-
+- if (!error)
+- simple_release_fs(&relayfs_mount, &relayfs_mount_count);
+-
+- return error;
+-}
+-
+-/**
+- * relayfs_remove_file - remove a file from relay filesystem
+- * @dentry: directory dentry
+- *
+- * Returns 0 if successful, negative otherwise.
+- */
+-int relayfs_remove_file(struct dentry *dentry)
+-{
+- return relayfs_remove(dentry);
+-}
+-
+-/**
+- * relayfs_remove_dir - remove a directory in the relay filesystem
+- * @dentry: directory dentry
+- *
+- * Returns 0 if successful, negative otherwise.
+- */
+-int relayfs_remove_dir(struct dentry *dentry)
+-{
+- return relayfs_remove(dentry);
+-}
+-
+-/**
+- * relay_file_open - open file op for relay files
+- * @inode: the inode
+- * @filp: the file
+- *
+- * Increments the channel buffer refcount.
+- */
+-static int relay_file_open(struct inode *inode, struct file *filp)
+-{
+- struct rchan_buf *buf = inode->u.generic_ip;
+- kref_get(&buf->kref);
+- filp->private_data = buf;
+-
+- return 0;
+-}
+-
+-/**
+- * relay_file_mmap - mmap file op for relay files
+- * @filp: the file
+- * @vma: the vma describing what to map
+- *
+- * Calls upon relay_mmap_buf to map the file into user space.
+- */
+-static int relay_file_mmap(struct file *filp, struct vm_area_struct *vma)
+-{
+- struct rchan_buf *buf = filp->private_data;
+- return relay_mmap_buf(buf, vma);
+-}
+-
+-/**
+- * relay_file_poll - poll file op for relay files
+- * @filp: the file
+- * @wait: poll table
+- *
+- * Poll implemention.
+- */
+-static unsigned int relay_file_poll(struct file *filp, poll_table *wait)
+-{
+- unsigned int mask = 0;
+- struct rchan_buf *buf = filp->private_data;
+-
+- if (buf->finalized)
+- return POLLERR;
+-
+- if (filp->f_mode & FMODE_READ) {
+- poll_wait(filp, &buf->read_wait, wait);
+- if (!relay_buf_empty(buf))
+- mask |= POLLIN | POLLRDNORM;
+- }
+-
+- return mask;
+-}
+-
+-/**
+- * relay_file_release - release file op for relay files
+- * @inode: the inode
+- * @filp: the file
+- *
+- * Decrements the channel refcount, as the filesystem is
+- * no longer using it.
+- */
+-static int relay_file_release(struct inode *inode, struct file *filp)
+-{
+- struct rchan_buf *buf = filp->private_data;
+- kref_put(&buf->kref, relay_remove_buf);
+-
+- return 0;
+-}
+-
+-/**
+- * relay_file_read_consume - update the consumed count for the buffer
+- */
+-static void relay_file_read_consume(struct rchan_buf *buf,
+- size_t read_pos,
+- size_t bytes_consumed)
+-{
+- size_t subbuf_size = buf->chan->subbuf_size;
+- size_t n_subbufs = buf->chan->n_subbufs;
+- size_t read_subbuf;
+-
+- if (buf->bytes_consumed + bytes_consumed > subbuf_size) {
+- relay_subbufs_consumed(buf->chan, buf->cpu, 1);
+- buf->bytes_consumed = 0;
+- }
+-
+- buf->bytes_consumed += bytes_consumed;
+- read_subbuf = read_pos / buf->chan->subbuf_size;
+- if (buf->bytes_consumed + buf->padding[read_subbuf] == subbuf_size) {
+- if ((read_subbuf == buf->subbufs_produced % n_subbufs) &&
+- (buf->offset == subbuf_size))
+- return;
+- relay_subbufs_consumed(buf->chan, buf->cpu, 1);
+- buf->bytes_consumed = 0;
+- }
+-}
+-
+-/**
+- * relay_file_read_avail - boolean, are there unconsumed bytes available?
+- */
+-static int relay_file_read_avail(struct rchan_buf *buf, size_t read_pos)
+-{
+- size_t bytes_produced, bytes_consumed, write_offset;
+- size_t subbuf_size = buf->chan->subbuf_size;
+- size_t n_subbufs = buf->chan->n_subbufs;
+- size_t produced = buf->subbufs_produced % n_subbufs;
+- size_t consumed = buf->subbufs_consumed % n_subbufs;
+-
+- write_offset = buf->offset > subbuf_size ? subbuf_size : buf->offset;
+-
+- if (consumed > produced) {
+- if ((produced > n_subbufs) &&
+- (produced + n_subbufs - consumed <= n_subbufs))
+- produced += n_subbufs;
+- } else if (consumed == produced) {
+- if (buf->offset > subbuf_size) {
+- produced += n_subbufs;
+- if (buf->subbufs_produced == buf->subbufs_consumed)
+- consumed += n_subbufs;
+- }
+- }
+-
+- if (buf->offset > subbuf_size)
+- bytes_produced = (produced - 1) * subbuf_size + write_offset;
+- else
+- bytes_produced = produced * subbuf_size + write_offset;
+- bytes_consumed = consumed * subbuf_size + buf->bytes_consumed;
+-
+- if (bytes_produced == bytes_consumed)
+- return 0;
+-
+- relay_file_read_consume(buf, read_pos, 0);
+-
+- return 1;
+-}
+-
+-/**
+- * relay_file_read_subbuf_avail - return bytes available in sub-buffer
+- */
+-static size_t relay_file_read_subbuf_avail(size_t read_pos,
+- struct rchan_buf *buf)
+-{
+- size_t padding, avail = 0;
+- size_t read_subbuf, read_offset, write_subbuf, write_offset;
+- size_t subbuf_size = buf->chan->subbuf_size;
+-
+- write_subbuf = (buf->data - buf->start) / subbuf_size;
+- write_offset = buf->offset > subbuf_size ? subbuf_size : buf->offset;
+- read_subbuf = read_pos / subbuf_size;
+- read_offset = read_pos % subbuf_size;
+- padding = buf->padding[read_subbuf];
+-
+- if (read_subbuf == write_subbuf) {
+- if (read_offset + padding < write_offset)
+- avail = write_offset - (read_offset + padding);
+- } else
+- avail = (subbuf_size - padding) - read_offset;
+-
+- return avail;
+-}
+-
+-/**
+- * relay_file_read_start_pos - find the first available byte to read
+- *
+- * If the read_pos is in the middle of padding, return the
+- * position of the first actually available byte, otherwise
+- * return the original value.
+- */
+-static size_t relay_file_read_start_pos(size_t read_pos,
+- struct rchan_buf *buf)
+-{
+- size_t read_subbuf, padding, padding_start, padding_end;
+- size_t subbuf_size = buf->chan->subbuf_size;
+- size_t n_subbufs = buf->chan->n_subbufs;
+-
+- read_subbuf = read_pos / subbuf_size;
+- padding = buf->padding[read_subbuf];
+- padding_start = (read_subbuf + 1) * subbuf_size - padding;
+- padding_end = (read_subbuf + 1) * subbuf_size;
+- if (read_pos >= padding_start && read_pos < padding_end) {
+- read_subbuf = (read_subbuf + 1) % n_subbufs;
+- read_pos = read_subbuf * subbuf_size;
+- }
+-
+- return read_pos;
+-}
+-
+-/**
+- * relay_file_read_end_pos - return the new read position
+- */
+-static size_t relay_file_read_end_pos(struct rchan_buf *buf,
+- size_t read_pos,
+- size_t count)
+-{
+- size_t read_subbuf, padding, end_pos;
+- size_t subbuf_size = buf->chan->subbuf_size;
+- size_t n_subbufs = buf->chan->n_subbufs;
+-
+- read_subbuf = read_pos / subbuf_size;
+- padding = buf->padding[read_subbuf];
+- if (read_pos % subbuf_size + count + padding == subbuf_size)
+- end_pos = (read_subbuf + 1) * subbuf_size;
+- else
+- end_pos = read_pos + count;
+- if (end_pos >= subbuf_size * n_subbufs)
+- end_pos = 0;
+-
+- return end_pos;
+-}
+-
+-/**
+- * relay_file_read - read file op for relay files
+- * @filp: the file
+- * @buffer: the userspace buffer
+- * @count: number of bytes to read
+- * @ppos: position to read from
+- *
+- * Reads count bytes or the number of bytes available in the
+- * current sub-buffer being read, whichever is smaller.
+- */
+-static ssize_t relay_file_read(struct file *filp,
+- char __user *buffer,
+- size_t count,
+- loff_t *ppos)
+-{
+- struct rchan_buf *buf = filp->private_data;
+- struct inode *inode = filp->f_dentry->d_inode;
+- size_t read_start, avail;
+- ssize_t ret = 0;
+- void *from;
+-
+- mutex_lock(&inode->i_mutex);
+- if(!relay_file_read_avail(buf, *ppos))
+- goto out;
+-
+- read_start = relay_file_read_start_pos(*ppos, buf);
+- avail = relay_file_read_subbuf_avail(read_start, buf);
+- if (!avail)
+- goto out;
+-
+- from = buf->start + read_start;
+- ret = count = min(count, avail);
+- if (copy_to_user(buffer, from, count)) {
+- ret = -EFAULT;
+- goto out;
+- }
+- relay_file_read_consume(buf, read_start, count);
+- *ppos = relay_file_read_end_pos(buf, read_start, count);
+-out:
+- mutex_unlock(&inode->i_mutex);
+- return ret;
+-}
+-
+-struct file_operations relay_file_operations = {
+- .open = relay_file_open,
+- .poll = relay_file_poll,
+- .mmap = relay_file_mmap,
+- .read = relay_file_read,
+- .llseek = no_llseek,
+- .release = relay_file_release,
+-};
+-
+-static struct super_operations relayfs_ops = {
+- .statfs = simple_statfs,
+- .drop_inode = generic_delete_inode,
+-};
+-
+-static int relayfs_fill_super(struct super_block * sb, void * data, int silent)
+-{
+- struct inode *inode;
+- struct dentry *root;
+- int mode = S_IFDIR | S_IRWXU | S_IRUGO | S_IXUGO;
+-
+- sb->s_blocksize = PAGE_CACHE_SIZE;
+- sb->s_blocksize_bits = PAGE_CACHE_SHIFT;
+- sb->s_magic = RELAYFS_MAGIC;
+- sb->s_op = &relayfs_ops;
+- inode = relayfs_get_inode(sb, mode, NULL, NULL);
+-
+- if (!inode)
+- return -ENOMEM;
+-
+- root = d_alloc_root(inode);
+- if (!root) {
+- iput(inode);
+- return -ENOMEM;
+- }
+- sb->s_root = root;
+-
+- return 0;
+-}
+-
+-static struct super_block * relayfs_get_sb(struct file_system_type *fs_type,
+- int flags, const char *dev_name,
+- void *data)
+-{
+- return get_sb_single(fs_type, flags, data, relayfs_fill_super);
+-}
+-
+-static struct file_system_type relayfs_fs_type = {
+- .owner = THIS_MODULE,
+- .name = "relayfs",
+- .get_sb = relayfs_get_sb,
+- .kill_sb = kill_litter_super,
+-};
+-
+-static int __init init_relayfs_fs(void)
+-{
+- return register_filesystem(&relayfs_fs_type);
+-}
+-
+-static void __exit exit_relayfs_fs(void)
+-{
+-
+-
+-
+-
+-
+- unregister_filesystem(&relayfs_fs_type);
+-}
+-
+-module_init(init_relayfs_fs)
+-module_exit(exit_relayfs_fs)
+-
+-EXPORT_SYMBOL_GPL(relay_file_operations);
+-EXPORT_SYMBOL_GPL(relayfs_create_dir);
+-EXPORT_SYMBOL_GPL(relayfs_remove_dir);
+-EXPORT_SYMBOL_GPL(relayfs_create_file);
+-EXPORT_SYMBOL_GPL(relayfs_remove_file);
+-
+-MODULE_AUTHOR("Tom Zanussi <zanussi@us.ibm.com> and Karim Yaghmour <karim@opersys.com>");
+-MODULE_DESCRIPTION("Relay Filesystem");
+-MODULE_LICENSE("GPL");
+-
+diff --git a/fs/relayfs/relay.c b/fs/relayfs/relay.c
+deleted file mode 100644
+index abf3cea..0000000
+--- a/fs/relayfs/relay.c
++++ /dev/null
+@@ -1,482 +0,0 @@
+-/*
+- * Public API and common code for RelayFS.
+- *
+- * See Documentation/filesystems/relayfs.txt for an overview of relayfs.
+- *
+- * Copyright (C) 2002-2005 - Tom Zanussi (zanussi@us.ibm.com), IBM Corp
+- * Copyright (C) 1999-2005 - Karim Yaghmour (karim@opersys.com)
+- *
+- * This file is released under the GPL.
+- */
+-
+-#include <linux/errno.h>
+-#include <linux/stddef.h>
+-#include <linux/slab.h>
+-#include <linux/module.h>
+-#include <linux/string.h>
+-#include <linux/relayfs_fs.h>
+-#include "relay.h"
+-#include "buffers.h"
+-
+-/**
+- * relay_buf_empty - boolean, is the channel buffer empty?
+- * @buf: channel buffer
+- *
+- * Returns 1 if the buffer is empty, 0 otherwise.
+- */
+-int relay_buf_empty(struct rchan_buf *buf)
+-{
+- return (buf->subbufs_produced - buf->subbufs_consumed) ? 0 : 1;
+-}
+-
+-/**
+- * relay_buf_full - boolean, is the channel buffer full?
+- * @buf: channel buffer
+- *
+- * Returns 1 if the buffer is full, 0 otherwise.
+- */
+-int relay_buf_full(struct rchan_buf *buf)
+-{
+- size_t ready = buf->subbufs_produced - buf->subbufs_consumed;
+- return (ready >= buf->chan->n_subbufs) ? 1 : 0;
+-}
+-
+-/*
+- * High-level relayfs kernel API and associated functions.
+- */
+-
+-/*
+- * rchan_callback implementations defining default channel behavior. Used
+- * in place of corresponding NULL values in client callback struct.
+- */
+-
+-/*
+- * subbuf_start() default callback. Does nothing.
+- */
+-static int subbuf_start_default_callback (struct rchan_buf *buf,
+- void *subbuf,
+- void *prev_subbuf,
+- size_t prev_padding)
+-{
+- if (relay_buf_full(buf))
+- return 0;
+-
+- return 1;
+-}
+-
+-/*
+- * buf_mapped() default callback. Does nothing.
+- */
+-static void buf_mapped_default_callback(struct rchan_buf *buf,
+- struct file *filp)
+-{
+-}
+-
+-/*
+- * buf_unmapped() default callback. Does nothing.
+- */
+-static void buf_unmapped_default_callback(struct rchan_buf *buf,
+- struct file *filp)
+-{
+-}
+-
+-/*
+- * create_buf_file_create() default callback. Creates file to represent buf.
+- */
+-static struct dentry *create_buf_file_default_callback(const char *filename,
+- struct dentry *parent,
+- int mode,
+- struct rchan_buf *buf,
+- int *is_global)
+-{
+- return relayfs_create_file(filename, parent, mode,
+- &relay_file_operations, buf);
+-}
+-
+-/*
+- * remove_buf_file() default callback. Removes file representing relay buffer.
+- */
+-static int remove_buf_file_default_callback(struct dentry *dentry)
+-{
+- return relayfs_remove(dentry);
+-}
+-
+-/* relay channel default callbacks */
+-static struct rchan_callbacks default_channel_callbacks = {
+- .subbuf_start = subbuf_start_default_callback,
+- .buf_mapped = buf_mapped_default_callback,
+- .buf_unmapped = buf_unmapped_default_callback,
+- .create_buf_file = create_buf_file_default_callback,
+- .remove_buf_file = remove_buf_file_default_callback,
+-};
+-
+-/**
+- * wakeup_readers - wake up readers waiting on a channel
+- * @private: the channel buffer
+- *
+- * This is the work function used to defer reader waking. The
+- * reason waking is deferred is that calling directly from write
+- * causes problems if you're writing from say the scheduler.
+- */
+-static void wakeup_readers(void *private)
+-{
+- struct rchan_buf *buf = private;
+- wake_up_interruptible(&buf->read_wait);
+-}
+-
+-/**
+- * __relay_reset - reset a channel buffer
+- * @buf: the channel buffer
+- * @init: 1 if this is a first-time initialization
+- *
+- * See relay_reset for description of effect.
+- */
+-static inline void __relay_reset(struct rchan_buf *buf, unsigned int init)
+-{
+- size_t i;
+-
+- if (init) {
+- init_waitqueue_head(&buf->read_wait);
+- kref_init(&buf->kref);
+- INIT_WORK(&buf->wake_readers, NULL, NULL);
+- } else {
+- cancel_delayed_work(&buf->wake_readers);
+- flush_scheduled_work();
+- }
+-
+- buf->subbufs_produced = 0;
+- buf->subbufs_consumed = 0;
+- buf->bytes_consumed = 0;
+- buf->finalized = 0;
+- buf->data = buf->start;
+- buf->offset = 0;
+-
+- for (i = 0; i < buf->chan->n_subbufs; i++)
+- buf->padding[i] = 0;
+-
+- buf->chan->cb->subbuf_start(buf, buf->data, NULL, 0);
+-}
+-
+-/**
+- * relay_reset - reset the channel
+- * @chan: the channel
+- *
+- * This has the effect of erasing all data from all channel buffers
+- * and restarting the channel in its initial state. The buffers
+- * are not freed, so any mappings are still in effect.
+- *
+- * NOTE: Care should be taken that the channel isn't actually
+- * being used by anything when this call is made.
+- */
+-void relay_reset(struct rchan *chan)
+-{
+- unsigned int i;
+- struct rchan_buf *prev = NULL;
+-
+- if (!chan)
+- return;
+-
+- for (i = 0; i < NR_CPUS; i++) {
+- if (!chan->buf[i] || chan->buf[i] == prev)
+- break;
+- __relay_reset(chan->buf[i], 0);
+- prev = chan->buf[i];
+- }
+-}
+-
+-/**
+- * relay_open_buf - create a new channel buffer in relayfs
+- *
+- * Internal - used by relay_open().
+- */
+-static struct rchan_buf *relay_open_buf(struct rchan *chan,
+- const char *filename,
+- struct dentry *parent,
+- int *is_global)
+-{
+- struct rchan_buf *buf;
+- struct dentry *dentry;
+-
+- if (*is_global)
+- return chan->buf[0];
+-
+- buf = relay_create_buf(chan);
+- if (!buf)
+- return NULL;
+-
+- /* Create file in fs */
+- dentry = chan->cb->create_buf_file(filename, parent, S_IRUSR,
+- buf, is_global);
+- if (!dentry) {
+- relay_destroy_buf(buf);
+- return NULL;
+- }
+-
+- buf->dentry = dentry;
+- __relay_reset(buf, 1);
+-
+- return buf;
+-}
+-
+-/**
+- * relay_close_buf - close a channel buffer
+- * @buf: channel buffer
+- *
+- * Marks the buffer finalized and restores the default callbacks.
+- * The channel buffer and channel buffer data structure are then freed
+- * automatically when the last reference is given up.
+- */
+-static inline void relay_close_buf(struct rchan_buf *buf)
+-{
+- buf->finalized = 1;
+- buf->chan->cb = &default_channel_callbacks;
+- cancel_delayed_work(&buf->wake_readers);
+- flush_scheduled_work();
+- kref_put(&buf->kref, relay_remove_buf);
+-}
+-
+-static inline void setup_callbacks(struct rchan *chan,
+- struct rchan_callbacks *cb)
+-{
+- if (!cb) {
+- chan->cb = &default_channel_callbacks;
+- return;
+- }
+-
+- if (!cb->subbuf_start)
+- cb->subbuf_start = subbuf_start_default_callback;
+- if (!cb->buf_mapped)
+- cb->buf_mapped = buf_mapped_default_callback;
+- if (!cb->buf_unmapped)
+- cb->buf_unmapped = buf_unmapped_default_callback;
+- if (!cb->create_buf_file)
+- cb->create_buf_file = create_buf_file_default_callback;
+- if (!cb->remove_buf_file)
+- cb->remove_buf_file = remove_buf_file_default_callback;
+- chan->cb = cb;
+-}
+-
+-/**
+- * relay_open - create a new relayfs channel
+- * @base_filename: base name of files to create
+- * @parent: dentry of parent directory, NULL for root directory
+- * @subbuf_size: size of sub-buffers
+- * @n_subbufs: number of sub-buffers
+- * @cb: client callback functions
+- *
+- * Returns channel pointer if successful, NULL otherwise.
+- *
+- * Creates a channel buffer for each cpu using the sizes and
+- * attributes specified. The created channel buffer files
+- * will be named base_filename0...base_filenameN-1. File
+- * permissions will be S_IRUSR.
+- */
+-struct rchan *relay_open(const char *base_filename,
+- struct dentry *parent,
+- size_t subbuf_size,
+- size_t n_subbufs,
+- struct rchan_callbacks *cb)
+-{
+- unsigned int i;
+- struct rchan *chan;
+- char *tmpname;
+- int is_global = 0;
+-
+- if (!base_filename)
+- return NULL;
+-
+- if (!(subbuf_size && n_subbufs))
+- return NULL;
+-
+- chan = kcalloc(1, sizeof(struct rchan), GFP_KERNEL);
+- if (!chan)
+- return NULL;
+-
+- chan->version = RELAYFS_CHANNEL_VERSION;
+- chan->n_subbufs = n_subbufs;
+- chan->subbuf_size = subbuf_size;
+- chan->alloc_size = FIX_SIZE(subbuf_size * n_subbufs);
+- setup_callbacks(chan, cb);
+- kref_init(&chan->kref);
+-
+- tmpname = kmalloc(NAME_MAX + 1, GFP_KERNEL);
+- if (!tmpname)
+- goto free_chan;
+-
+- for_each_online_cpu(i) {
+- sprintf(tmpname, "%s%d", base_filename, i);
+- chan->buf[i] = relay_open_buf(chan, tmpname, parent,
+- &is_global);
+- chan->buf[i]->cpu = i;
+- if (!chan->buf[i])
+- goto free_bufs;
+- }
+-
+- kfree(tmpname);
+- return chan;
+-
+-free_bufs:
+- for (i = 0; i < NR_CPUS; i++) {
+- if (!chan->buf[i])
+- break;
+- relay_close_buf(chan->buf[i]);
+- if (is_global)
+- break;
+- }
+- kfree(tmpname);
+-
+-free_chan:
+- kref_put(&chan->kref, relay_destroy_channel);
+- return NULL;
+-}
+-
+-/**
+- * relay_switch_subbuf - switch to a new sub-buffer
+- * @buf: channel buffer
+- * @length: size of current event
+- *
+- * Returns either the length passed in or 0 if full.
+-
+- * Performs sub-buffer-switch tasks such as invoking callbacks,
+- * updating padding counts, waking up readers, etc.
+- */
+-size_t relay_switch_subbuf(struct rchan_buf *buf, size_t length)
+-{
+- void *old, *new;
+- size_t old_subbuf, new_subbuf;
+-
+- if (unlikely(length > buf->chan->subbuf_size))
+- goto toobig;
+-
+- if (buf->offset != buf->chan->subbuf_size + 1) {
+- buf->prev_padding = buf->chan->subbuf_size - buf->offset;
+- old_subbuf = buf->subbufs_produced % buf->chan->n_subbufs;
+- buf->padding[old_subbuf] = buf->prev_padding;
+- buf->subbufs_produced++;
+- if (waitqueue_active(&buf->read_wait)) {
+- PREPARE_WORK(&buf->wake_readers, wakeup_readers, buf);
+- schedule_delayed_work(&buf->wake_readers, 1);
+- }
+- }
+-
+- old = buf->data;
+- new_subbuf = buf->subbufs_produced % buf->chan->n_subbufs;
+- new = buf->start + new_subbuf * buf->chan->subbuf_size;
+- buf->offset = 0;
+- if (!buf->chan->cb->subbuf_start(buf, new, old, buf->prev_padding)) {
+- buf->offset = buf->chan->subbuf_size + 1;
+- return 0;
+- }
+- buf->data = new;
+- buf->padding[new_subbuf] = 0;
+-
+- if (unlikely(length + buf->offset > buf->chan->subbuf_size))
+- goto toobig;
+-
+- return length;
+-
+-toobig:
+- buf->chan->last_toobig = length;
+- return 0;
+-}
+-
+-/**
+- * relay_subbufs_consumed - update the buffer's sub-buffers-consumed count
+- * @chan: the channel
+- * @cpu: the cpu associated with the channel buffer to update
+- * @subbufs_consumed: number of sub-buffers to add to current buf's count
+- *
+- * Adds to the channel buffer's consumed sub-buffer count.
+- * subbufs_consumed should be the number of sub-buffers newly consumed,
+- * not the total consumed.
+- *
+- * NOTE: kernel clients don't need to call this function if the channel
+- * mode is 'overwrite'.
+- */
+-void relay_subbufs_consumed(struct rchan *chan,
+- unsigned int cpu,
+- size_t subbufs_consumed)
+-{
+- struct rchan_buf *buf;
+-
+- if (!chan)
+- return;
+-
+- if (cpu >= NR_CPUS || !chan->buf[cpu])
+- return;
+-
+- buf = chan->buf[cpu];
+- buf->subbufs_consumed += subbufs_consumed;
+- if (buf->subbufs_consumed > buf->subbufs_produced)
+- buf->subbufs_consumed = buf->subbufs_produced;
+-}
+-
+-/**
+- * relay_destroy_channel - free the channel struct
+- *
+- * Should only be called from kref_put().
+- */
+-void relay_destroy_channel(struct kref *kref)
+-{
+- struct rchan *chan = container_of(kref, struct rchan, kref);
+- kfree(chan);
+-}
+-
+-/**
+- * relay_close - close the channel
+- * @chan: the channel
+- *
+- * Closes all channel buffers and frees the channel.
+- */
+-void relay_close(struct rchan *chan)
+-{
+- unsigned int i;
+- struct rchan_buf *prev = NULL;
+-
+- if (!chan)
+- return;
+-
+- for (i = 0; i < NR_CPUS; i++) {
+- if (!chan->buf[i] || chan->buf[i] == prev)
+- break;
+- relay_close_buf(chan->buf[i]);
+- prev = chan->buf[i];
+- }
+-
+- if (chan->last_toobig)
+- printk(KERN_WARNING "relayfs: one or more items not logged "
+- "[item size (%Zd) > sub-buffer size (%Zd)]\n",
+- chan->last_toobig, chan->subbuf_size);
+-
+- kref_put(&chan->kref, relay_destroy_channel);
+-}
+-
+-/**
+- * relay_flush - close the channel
+- * @chan: the channel
+- *
+- * Flushes all channel buffers i.e. forces buffer switch.
+- */
+-void relay_flush(struct rchan *chan)
+-{
+- unsigned int i;
+- struct rchan_buf *prev = NULL;
+-
+- if (!chan)
+- return;
+-
+- for (i = 0; i < NR_CPUS; i++) {
+- if (!chan->buf[i] || chan->buf[i] == prev)
+- break;
+- relay_switch_subbuf(chan->buf[i], 0);
+- prev = chan->buf[i];
+- }
+-}
+-
+-EXPORT_SYMBOL_GPL(relay_open);
+-EXPORT_SYMBOL_GPL(relay_close);
+-EXPORT_SYMBOL_GPL(relay_flush);
+-EXPORT_SYMBOL_GPL(relay_reset);
+-EXPORT_SYMBOL_GPL(relay_subbufs_consumed);
+-EXPORT_SYMBOL_GPL(relay_switch_subbuf);
+-EXPORT_SYMBOL_GPL(relay_buf_full);
+diff --git a/fs/relayfs/relay.h b/fs/relayfs/relay.h
+deleted file mode 100644
+index 0993d3e..0000000
+--- a/fs/relayfs/relay.h
++++ /dev/null
+@@ -1,8 +0,0 @@
+-#ifndef _RELAY_H
+-#define _RELAY_H
+-
+-extern int relayfs_remove(struct dentry *dentry);
+-extern int relay_buf_empty(struct rchan_buf *buf);
+-extern void relay_destroy_channel(struct kref *kref);
+-
+-#endif /* _RELAY_H */
+diff --git a/fs/sysfs/Makefile b/fs/sysfs/Makefile
+index 7a1ceb9..7bd4d8f 100644
+--- a/fs/sysfs/Makefile
++++ b/fs/sysfs/Makefile
+@@ -4,3 +4,4 @@
+
+ obj-y := inode.o file.o dir.o symlink.o mount.o bin.o \
+ group.o
++obj-$(CONFIG_RELAY) += relay.o
+diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c
+index 49bd219..52a759b 100644
+--- a/fs/sysfs/dir.c
++++ b/fs/sysfs/dir.c
+@@ -87,6 +87,20 @@ static int init_file(struct inode * inod
+ return 0;
+ }
+
++#ifdef CONFIG_RELAY
++static int init_relay_file(struct inode * inode)
++{
++ extern struct file_operations relay_file_operations;
++ inode->i_fop = &relay_file_operations;
++ return 0;
++}
++#else
++static int init_relay_file(struct inode * inode)
++{
++ return -EINVAL;
++}
++#endif
++
+ static int init_symlink(struct inode * inode)
+ {
+ inode->i_op = &sysfs_symlink_inode_operations;
+@@ -165,6 +179,7 @@ int sysfs_create_dir(struct kobject * ko
+ static int sysfs_attach_attr(struct sysfs_dirent * sd, struct dentry * dentry)
+ {
+ struct attribute * attr = NULL;
++ struct relay_attribute * rel_attr = NULL;
+ struct bin_attribute * bin_attr = NULL;
+ int (* init) (struct inode *) = NULL;
+ int error = 0;
+@@ -172,6 +187,10 @@ static int sysfs_attach_attr(struct sysf
+ if (sd->s_type & SYSFS_KOBJ_BIN_ATTR) {
+ bin_attr = sd->s_element;
+ attr = &bin_attr->attr;
++ } else if (sd->s_type & SYSFS_KOBJ_RELAY_ATTR) {
++ rel_attr = sd->s_element;
++ attr = &rel_attr->attr;
++ init = init_relay_file;
+ } else {
+ attr = sd->s_element;
+ init = init_file;
+@@ -189,6 +208,13 @@ static int sysfs_attach_attr(struct sysf
+ dentry->d_inode->i_size = bin_attr->size;
+ dentry->d_inode->i_fop = &bin_fops;
+ }
++
++ if (rel_attr) {
++ unsigned int size = rel_attr->subbuf_size * rel_attr->n_subbufs;
++ dentry->d_inode->i_size = ((size - 1) & PAGE_MASK) + PAGE_SIZE;
++ dentry->d_inode->u.generic_ip = rel_attr->rchan_buf;
++ }
++
+ dentry->d_op = &sysfs_dentry_ops;
+ d_rehash(dentry);
+
+diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c
+index 689f7bc..d41cb6d 100644
+--- a/fs/sysfs/inode.c
++++ b/fs/sysfs/inode.c
+@@ -174,6 +174,7 @@ const unsigned char * sysfs_get_name(str
+ {
+ struct attribute * attr;
+ struct bin_attribute * bin_attr;
++ struct relay_attribute * rel_attr;
+ struct sysfs_symlink * sl;
+
+ if (!sd || !sd->s_element)
+@@ -192,6 +193,10 @@ const unsigned char * sysfs_get_name(str
+ bin_attr = sd->s_element;
+ return bin_attr->attr.name;
+
++ case SYSFS_KOBJ_RELAY_ATTR:
++ rel_attr = sd->s_element;
++ return rel_attr->attr.name;
++
+ case SYSFS_KOBJ_LINK:
+ sl = sd->s_element;
+ return sl->link_name;
+diff --git a/fs/sysfs/relay.c b/fs/sysfs/relay.c
+new file mode 100644
+index 0000000..f40de18
+--- /dev/null
++++ b/fs/sysfs/relay.c
+@@ -0,0 +1,70 @@
++/*
++ * relay.c - relay channel buffer support for sysfs
++ *
++ * Copyright (C) 2006 Paul Mundt
++ *
++ * This file is subject to the terms and conditions of the GNU General Public
++ * License. See the file "COPYING" in the main directory of this archive
++ * for more details.
++ */
++#include <linux/relay.h>
++#include <linux/namei.h>
++#include <linux/module.h>
++#include "sysfs.h"
++
++static struct dentry *
++sysfs_create_buf_file(const char *filename, struct dentry *dentry, int mode,
++ struct rchan_buf *buf, int *is_global)
++{
++ struct relay_attribute *attr = (struct relay_attribute *)dentry;
++ struct dentry *parent = attr->kobj->dentry;
++ int ret;
++
++ attr->attr.mode = mode;
++ attr->rchan_buf = buf;
++
++ ret = sysfs_add_file(parent, &attr->attr, SYSFS_KOBJ_RELAY_ATTR);
++ if (unlikely(ret != 0))
++ return NULL;
++
++ dentry = lookup_one_len(filename, parent, strlen(filename));
++ if (IS_ERR(dentry))
++ sysfs_hash_and_remove(parent, filename);
++
++ return dentry;
++}
++
++static int sysfs_remove_buf_file(struct dentry *dentry)
++{
++ sysfs_hash_and_remove(dentry, dentry->d_name.name);
++ return 0;
++}
++
++struct rchan_callbacks sysfs_rchan_callbacks = {
++ .create_buf_file = sysfs_create_buf_file,
++ .remove_buf_file = sysfs_remove_buf_file,
++};
++
++int sysfs_create_relay_file(struct kobject *kobj, struct relay_attribute *attr)
++{
++ BUG_ON(!kobj || !kobj->dentry || !attr);
++
++ attr->kobj = kobj;
++
++ attr->rchan = relay_open(attr->attr.name, (void *)attr,
++ attr->subbuf_size, attr->n_subbufs,
++ &sysfs_rchan_callbacks);
++ if (IS_ERR(attr->rchan))
++ return PTR_ERR(attr->rchan);
++
++ return 0;
++}
++EXPORT_SYMBOL_GPL(sysfs_create_relay_file);
++
++void sysfs_remove_relay_file(struct kobject *kobj, struct relay_attribute *attr)
++{
++ BUG_ON(!attr);
++
++ relay_close(attr->rchan);
++}
++EXPORT_SYMBOL_GPL(sysfs_remove_relay_file);
+diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
+index 860e7a4..266ce9d 100644
+--- a/include/linux/blkdev.h
++++ b/include/linux/blkdev.h
+@@ -22,6 +22,7 @@ typedef struct request_queue request_que
+ 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
+ */
+diff --git a/include/linux/blktrace_api.h b/include/linux/blktrace_api.h
+new file mode 100644
+index 0000000..b34d3e7
+--- /dev/null
++++ b/include/linux/blktrace_api.h
+@@ -0,0 +1,277 @@
++#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
+diff --git a/include/linux/compat_ioctl.h b/include/linux/compat_ioctl.h
+index 8fad50f..a3ed64b 100644
+--- a/include/linux/compat_ioctl.h
++++ b/include/linux/compat_ioctl.h
+@@ -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 */
+diff --git a/include/linux/fs.h b/include/linux/fs.h
+index e059da9..c7a63cd 100644
+--- a/include/linux/fs.h
++++ b/include/linux/fs.h
+@@ -196,6 +196,10 @@ 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 */
+diff --git a/include/linux/relayfs_fs.h b/include/linux/relay.h
+similarity index 89%
+rename from include/linux/relayfs_fs.h
+rename to include/linux/relay.h
+index 7342e66..0b456ec 100644
+--- a/include/linux/relayfs_fs.h
++++ b/include/linux/relay.h
+@@ -1,14 +1,14 @@
+ /*
+- * linux/include/linux/relayfs_fs.h
++ * linux/include/linux/relay.h
+ *
+ * Copyright (C) 2002, 2003 - Tom Zanussi (zanussi@us.ibm.com), IBM Corp
+ * Copyright (C) 1999, 2000, 2001, 2002 - Karim Yaghmour (karim@opersys.com)
+ *
+- * RelayFS definitions and declarations
++ * CONFIG_RELAY definitions and declarations
+ */
+
+-#ifndef _LINUX_RELAYFS_FS_H
+-#define _LINUX_RELAYFS_FS_H
++#ifndef _LINUX_RELAY_H
++#define _LINUX_RELAY_H
+
+ #include <linux/config.h>
+ #include <linux/types.h>
+@@ -19,6 +19,9 @@
+ #include <linux/poll.h>
+ #include <linux/kref.h>
+
++/* Needs a _much_ better name... */
++#define FIX_SIZE(x) ((((x) - 1) & PAGE_MASK) + PAGE_SIZE)
++
+ /*
+ * Tracks changes to rchan/rchan_buf structs
+ */
+@@ -46,6 +49,7 @@ struct rchan_buf
+ size_t prev_padding; /* temporary variable */
+ size_t bytes_consumed; /* bytes consumed in cur read subbuf */
+ unsigned int cpu; /* this buf's cpu */
++ unsigned int poll_flags;
+ } ____cacheline_aligned;
+
+ /*
+@@ -92,26 +96,26 @@ struct rchan_callbacks
+ size_t prev_padding);
+
+ /*
+- * buf_mapped - relayfs buffer mmap notification
++ * buf_mapped - relay buffer mmap notification
+ * @buf: the channel buffer
+- * @filp: relayfs file pointer
++ * @filp: relay file pointer
+ *
+- * Called when a relayfs file is successfully mmapped
++ * Called when a relay file is successfully mmapped
+ */
+ void (*buf_mapped)(struct rchan_buf *buf,
+ struct file *filp);
+
+ /*
+- * buf_unmapped - relayfs buffer unmap notification
++ * buf_unmapped - relay buffer unmap notification
+ * @buf: the channel buffer
+- * @filp: relayfs file pointer
++ * @filp: relay file pointer
+ *
+- * Called when a relayfs file is successfully unmapped
++ * Called when a relay file is successfully unmapped
+ */
+ void (*buf_unmapped)(struct rchan_buf *buf,
+ struct file *filp);
+ /*
+- * create_buf_file - create file to represent a relayfs channel buffer
++ * create_buf_file - create file to represent a relay channel buffer
+ * @filename: the name of the file to create
+ * @parent: the parent of the file to create
+ * @mode: the mode of the file to create
+@@ -121,7 +125,7 @@ struct rchan_callbacks
+ * Called during relay_open(), once for each per-cpu buffer,
+ * to allow the client to create a file to be used to
+ * represent the corresponding channel buffer. If the file is
+- * created outside of relayfs, the parent must also exist in
++ * created outside of relay, the parent must also exist in
+ * that filesystem.
+ *
+ * The callback should return the dentry of the file created
+@@ -140,7 +144,7 @@ struct rchan_callbacks
+ int *is_global);
+
+ /*
+- * remove_buf_file - remove file representing a relayfs channel buffer
++ * remove_buf_file - remove file representing a relay channel buffer
+ * @dentry: the dentry of the file to remove
+ *
+ * Called during relay_close(), once for each per-cpu buffer,
+@@ -153,7 +157,7 @@ struct rchan_callbacks
+ };
+
+ /*
+- * relayfs kernel API, fs/relayfs/relay.c
++ * CONFIG_RELAY kernel API, kernel/relay.c
+ */
+
+ struct rchan *relay_open(const char *base_filename,
+@@ -171,15 +175,6 @@ extern int relay_buf_full(struct rchan_b
+
+ extern size_t relay_switch_subbuf(struct rchan_buf *buf,
+ size_t length);
+-extern struct dentry *relayfs_create_dir(const char *name,
+- struct dentry *parent);
+-extern int relayfs_remove_dir(struct dentry *dentry);
+-extern struct dentry *relayfs_create_file(const char *name,
+- struct dentry *parent,
+- int mode,
+- struct file_operations *fops,
+- void *data);
+-extern int relayfs_remove_file(struct dentry *dentry);
+
+ /**
+ * relay_write - write data into the channel
+@@ -279,9 +274,9 @@ static inline void subbuf_start_reserve(
+ }
+
+ /*
+- * exported relay file operations, fs/relayfs/inode.c
++ * exported relay file operations, kernel/relay.c
+ */
+ extern struct file_operations relay_file_operations;
+
+-#endif /* _LINUX_RELAYFS_FS_H */
++#endif /* _LINUX_RELAY_H */
+
+diff --git a/include/linux/sched.h b/include/linux/sched.h
+index b6f51e3..d9d465b 100644
+--- a/include/linux/sched.h
++++ b/include/linux/sched.h
+@@ -705,6 +705,7 @@ struct task_struct {
+ prio_array_t *array;
+
+ unsigned short ioprio;
++ unsigned int btrace_seq;
+
+ unsigned long sleep_avg;
+ unsigned long long timestamp, last_ran;
+diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h
+index 392da5a..36a078e 100644
+--- a/include/linux/sysfs.h
++++ b/include/linux/sysfs.h
+@@ -44,6 +44,16 @@ struct attribute_group {
+ .show = _name##_show, \
+ }
+
++#define __ATTR_RELAY(_name,_buffer_size,_nr_buffers) { \
++ .attr = { \
++ .owner = THIS_MODULE, \
++ .name = __stringify(_name), \
++ .mode = 0400, \
++ }, \
++ .subbuf_size = _buffer_size, \
++ .n_subbufs = _nr_buffers, \
++}
++
+ #define __ATTR_NULL { .attr = { .name = NULL } }
+
+ #define attr_name(_attr) (_attr).attr.name
+@@ -60,6 +70,15 @@ struct bin_attribute {
+ struct vm_area_struct *vma);
+ };
+
++struct relay_attribute {
++ struct attribute attr;
++ struct rchan *rchan;
++ struct rchan_buf *rchan_buf;
++ struct kobject *kobj;
++ size_t subbuf_size;
++ size_t n_subbufs;
++};
++
+ struct sysfs_ops {
+ ssize_t (*show)(struct kobject *, struct attribute *,char *);
+ ssize_t (*store)(struct kobject *,struct attribute *,const char *, size_t);
+@@ -81,7 +100,23 @@ struct sysfs_dirent {
+ #define SYSFS_KOBJ_ATTR 0x0004
+ #define SYSFS_KOBJ_BIN_ATTR 0x0008
+ #define SYSFS_KOBJ_LINK 0x0020
+-#define SYSFS_NOT_PINNED (SYSFS_KOBJ_ATTR | SYSFS_KOBJ_BIN_ATTR | SYSFS_KOBJ_LINK)
++#define SYSFS_KOBJ_RELAY_ATTR 0x0040
++#define SYSFS_NOT_PINNED (SYSFS_KOBJ_ATTR | SYSFS_KOBJ_BIN_ATTR | \
++ SYSFS_KOBJ_LINK | SYSFS_KOBJ_RELAY_ATTR)
++
++#if defined(CONFIG_RELAY) && defined(CONFIG_SYSFS)
++int sysfs_create_relay_file(struct kobject *, struct relay_attribute *);
++void sysfs_remove_relay_file(struct kobject *, struct relay_attribute *);
++#else
++static inline int sysfs_create_relay_file(struct kobject *kobj, struct relay_attribute *attr)
++{
++ return 0;
++}
++
++static inline void sysfs_remove_relay_file(struct kobject *kobj, struct relay_attribute *attr)
++{
++}
++#endif /* CONFIG_RELAY && CONFIG_SYSFS */
+
+ #ifdef CONFIG_SYSFS
+
+diff --git a/init/Kconfig b/init/Kconfig
+index 38416a1..fb74fdd 100644
+--- a/init/Kconfig
++++ b/init/Kconfig
+@@ -214,6 +214,17 @@ config CPUSETS
+
+ Say N if unsure.
+
++config RELAY
++ bool "Kernel->user space relay support (formerly relayfs)"
++ help
++ This option enables support for relay interface support in
++ certain file systems (such as sysfs, and trivially debugfs).
++ It is designed to provide an efficient mechanism for tools and
++ facilities to relay large amounts of data from kernel space to
++ user space.
++
++ If unsure, say N.
++
+ source "usr/Kconfig"
+
+ config UID16
+diff --git a/kernel/Makefile b/kernel/Makefile
+index 4ae0fbd..aebd7a7 100644
+--- a/kernel/Makefile
++++ b/kernel/Makefile
+@@ -34,6 +34,7 @@ obj-$(CONFIG_DETECT_SOFTLOCKUP) += softl
+ obj-$(CONFIG_GENERIC_HARDIRQS) += irq/
+ obj-$(CONFIG_SECCOMP) += seccomp.o
+ obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o
++obj-$(CONFIG_RELAY) += relay.o
+
+ ifneq ($(CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER),y)
+ # According to Alan Modra <alan@linuxcare.com.au>, the -fno-omit-frame-pointer is
+diff --git a/kernel/fork.c b/kernel/fork.c
+index fbea12d..f564e9b 100644
+--- a/kernel/fork.c
++++ b/kernel/fork.c
+@@ -179,6 +179,7 @@ static struct task_struct *dup_task_stru
+ /* 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;
+ }
+
+diff --git a/kernel/relay.c b/kernel/relay.c
+new file mode 100644
+index 0000000..7fbdf78
+--- /dev/null
++++ b/kernel/relay.c
+@@ -0,0 +1,1006 @@
++/*
++ * Public API and common code for kernel->userspace relay file support.
++ *
++ * See Documentation/filesystems/relayfs.txt for an overview of relayfs.
++ *
++ * Copyright (C) 2002-2005 - Tom Zanussi (zanussi@us.ibm.com), IBM Corp
++ * Copyright (C) 1999-2005 - Karim Yaghmour (karim@opersys.com)
++ *
++ * Moved to kernel/relay.c by Paul Mundt, 2006.
++ *
++ * This file is released under the GPL.
++ */
++#include <linux/errno.h>
++#include <linux/stddef.h>
++#include <linux/slab.h>
++#include <linux/module.h>
++#include <linux/string.h>
++#include <linux/relay.h>
++#include <linux/vmalloc.h>
++#include <linux/mm.h>
++
++/*
++ * close() vm_op implementation for relay file mapping.
++ */
++static void relay_file_mmap_close(struct vm_area_struct *vma)
++{
++ struct rchan_buf *buf = vma->vm_private_data;
++ buf->chan->cb->buf_unmapped(buf, vma->vm_file);
++}
++
++/*
++ * nopage() vm_op implementation for relay file mapping.
++ */
++static struct page *relay_buf_nopage(struct vm_area_struct *vma,
++ unsigned long address,
++ int *type)
++{
++ struct page *page;
++ struct rchan_buf *buf = vma->vm_private_data;
++ unsigned long offset = address - vma->vm_start;
++
++ if (address > vma->vm_end)
++ return NOPAGE_SIGBUS; /* Disallow mremap */
++ if (!buf)
++ return NOPAGE_OOM;
++
++ page = vmalloc_to_page(buf->start + offset);
++ if (!page)
++ return NOPAGE_OOM;
++ get_page(page);
++
++ if (type)
++ *type = VM_FAULT_MINOR;
++
++ return page;
++}
++
++/*
++ * vm_ops for relay file mappings.
++ */
++static struct vm_operations_struct relay_file_mmap_ops = {
++ .nopage = relay_buf_nopage,
++ .close = relay_file_mmap_close,
++};
++
++/**
++ * relay_mmap_buf: - mmap channel buffer to process address space
++ * @buf: relay channel buffer
++ * @vma: vm_area_struct describing memory to be mapped
++ *
++ * Returns 0 if ok, negative on error
++ *
++ * Caller should already have grabbed mmap_sem.
++ */
++int relay_mmap_buf(struct rchan_buf *buf, struct vm_area_struct *vma)
++{
++ unsigned long length = vma->vm_end - vma->vm_start;
++ struct file *filp = vma->vm_file;
++
++ if (!buf)
++ return -EBADF;
++
++ if (length != (unsigned long)buf->chan->alloc_size)
++ return -EINVAL;
++
++ vma->vm_ops = &relay_file_mmap_ops;
++ vma->vm_private_data = buf;
++ buf->chan->cb->buf_mapped(buf, filp);
++
++ return 0;
++}
++
++/**
++ * relay_alloc_buf - allocate a channel buffer
++ * @buf: the buffer struct
++ * @size: total size of the buffer
++ *
++ * Returns a pointer to the resulting buffer, NULL if unsuccessful. The
++ * passed in size will get page aligned, if it isn't already.
++ */
++static void *relay_alloc_buf(struct rchan_buf *buf, size_t *size)
++{
++ void *mem;
++ unsigned int i, j, n_pages;
++
++ *size = PAGE_ALIGN(*size);
++ n_pages = *size >> PAGE_SHIFT;
++
++ buf->page_array = kcalloc(n_pages, sizeof(struct page *), GFP_KERNEL);
++ if (!buf->page_array)
++ return NULL;
++
++ for (i = 0; i < n_pages; i++) {
++ buf->page_array[i] = alloc_page(GFP_KERNEL);
++ if (unlikely(!buf->page_array[i]))
++ goto depopulate;
++ }
++ mem = vmap(buf->page_array, n_pages, VM_MAP, PAGE_KERNEL);
++ if (!mem)
++ goto depopulate;
++
++ memset(mem, 0, *size);
++ buf->page_count = n_pages;
++ return mem;
++
++depopulate:
++ for (j = 0; j < i; j++)
++ __free_page(buf->page_array[j]);
++ kfree(buf->page_array);
++ return NULL;
++}
++
++/**
++ * relay_create_buf - allocate and initialize a channel buffer
++ * @alloc_size: size of the buffer to allocate
++ * @n_subbufs: number of sub-buffers in the channel
++ *
++ * Returns channel buffer if successful, NULL otherwise
++ */
++struct rchan_buf *relay_create_buf(struct rchan *chan)
++{
++ struct rchan_buf *buf = kcalloc(1, sizeof(struct rchan_buf), GFP_KERNEL);
++ if (!buf)
++ return NULL;
++
++ buf->padding = kmalloc(chan->n_subbufs * sizeof(size_t *), GFP_KERNEL);
++ if (!buf->padding)
++ goto free_buf;
++
++ buf->start = relay_alloc_buf(buf, &chan->alloc_size);
++ if (!buf->start)
++ goto free_buf;
++
++ buf->chan = chan;
++ kref_get(&buf->chan->kref);
++ return buf;
++
++free_buf:
++ kfree(buf->padding);
++ kfree(buf);
++ return NULL;
++}
++
++/**
++ * relay_destroy_channel - free the channel struct
++ *
++ * Should only be called from kref_put().
++ */
++void relay_destroy_channel(struct kref *kref)
++{
++ struct rchan *chan = container_of(kref, struct rchan, kref);
++ kfree(chan);
++}
++
++/**
++ * relay_destroy_buf - destroy an rchan_buf struct and associated buffer
++ * @buf: the buffer struct
++ */
++void relay_destroy_buf(struct rchan_buf *buf)
++{
++ struct rchan *chan = buf->chan;
++ unsigned int i;
++
++ if (likely(buf->start)) {
++ vunmap(buf->start);
++ for (i = 0; i < buf->page_count; i++)
++ __free_page(buf->page_array[i]);
++ kfree(buf->page_array);
++ }
++ kfree(buf->padding);
++ kfree(buf);
++ kref_put(&chan->kref, relay_destroy_channel);
++}
++
++/**
++ * relay_remove_buf - remove a channel buffer
++ *
++ * Removes the file from the fileystem, which also frees the
++ * rchan_buf_struct and the channel buffer. Should only be called from
++ * kref_put().
++ */
++void relay_remove_buf(struct kref *kref)
++{
++ struct rchan_buf *buf = container_of(kref, struct rchan_buf, kref);
++ buf->chan->cb->remove_buf_file(buf->dentry);
++ relay_destroy_buf(buf);
++}
++
++/**
++ * relay_buf_empty - boolean, is the channel buffer empty?
++ * @buf: channel buffer
++ *
++ * Returns 1 if the buffer is empty, 0 otherwise.
++ */
++int relay_buf_empty(struct rchan_buf *buf)
++{
++ return (buf->subbufs_produced - buf->subbufs_consumed) ? 0 : 1;
++}
++EXPORT_SYMBOL_GPL(relay_buf_empty);
++
++/**
++ * relay_buf_full - boolean, is the channel buffer full?
++ * @buf: channel buffer
++ *
++ * Returns 1 if the buffer is full, 0 otherwise.
++ */
++int relay_buf_full(struct rchan_buf *buf)
++{
++ size_t ready = buf->subbufs_produced - buf->subbufs_consumed;
++ return (ready >= buf->chan->n_subbufs) ? 1 : 0;
++}
++EXPORT_SYMBOL_GPL(relay_buf_full);
++
++/*
++ * High-level relay kernel API and associated functions.
++ */
++
++/*
++ * rchan_callback implementations defining default channel behavior. Used
++ * in place of corresponding NULL values in client callback struct.
++ */
++
++/*
++ * subbuf_start() default callback. Does nothing.
++ */
++static int subbuf_start_default_callback (struct rchan_buf *buf,
++ void *subbuf,
++ void *prev_subbuf,
++ size_t prev_padding)
++{
++ if (relay_buf_full(buf))
++ return 0;
++
++ return 1;
++}
++
++/*
++ * buf_mapped() default callback. Does nothing.
++ */
++static void buf_mapped_default_callback(struct rchan_buf *buf,
++ struct file *filp)
++{
++}
++
++/*
++ * buf_unmapped() default callback. Does nothing.
++ */
++static void buf_unmapped_default_callback(struct rchan_buf *buf,
++ struct file *filp)
++{
++}
++
++/*
++ * create_buf_file_create() default callback. Does nothing.
++ */
++static struct dentry *create_buf_file_default_callback(const char *filename,
++ struct dentry *parent,
++ int mode,
++ struct rchan_buf *buf,
++ int *is_global)
++{
++ return NULL;
++}
++
++/*
++ * remove_buf_file() default callback. Does nothing.
++ */
++static int remove_buf_file_default_callback(struct dentry *dentry)
++{
++ return -EINVAL;
++}
++
++/* relay channel default callbacks */
++static struct rchan_callbacks default_channel_callbacks = {
++ .subbuf_start = subbuf_start_default_callback,
++ .buf_mapped = buf_mapped_default_callback,
++ .buf_unmapped = buf_unmapped_default_callback,
++ .create_buf_file = create_buf_file_default_callback,
++ .remove_buf_file = remove_buf_file_default_callback,
++};
++
++/**
++ * wakeup_readers - wake up readers waiting on a channel
++ * @private: the channel buffer
++ *
++ * This is the work function used to defer reader waking. The
++ * reason waking is deferred is that calling directly from write
++ * causes problems if you're writing from say the scheduler.
++ */
++static void wakeup_readers(void *private)
++{
++ struct rchan_buf *buf = private;
++ wake_up_interruptible(&buf->read_wait);
++}
++
++/**
++ * __relay_reset - reset a channel buffer
++ * @buf: the channel buffer
++ * @init: 1 if this is a first-time initialization
++ *
++ * See relay_reset for description of effect.
++ */
++static inline void __relay_reset(struct rchan_buf *buf, unsigned int init)
++{
++ size_t i;
++
++ if (init) {
++ init_waitqueue_head(&buf->read_wait);
++ kref_init(&buf->kref);
++ INIT_WORK(&buf->wake_readers, NULL, NULL);
++ } else {
++ cancel_delayed_work(&buf->wake_readers);
++ flush_scheduled_work();
++ }
++
++ buf->subbufs_produced = 0;
++ buf->subbufs_consumed = 0;
++ buf->bytes_consumed = 0;
++ buf->finalized = 0;
++ buf->data = buf->start;
++ buf->offset = 0;
++
++ for (i = 0; i < buf->chan->n_subbufs; i++)
++ buf->padding[i] = 0;
++
++ buf->chan->cb->subbuf_start(buf, buf->data, NULL, 0);
++}
++
++/**
++ * relay_reset - reset the channel
++ * @chan: the channel
++ *
++ * This has the effect of erasing all data from all channel buffers
++ * and restarting the channel in its initial state. The buffers
++ * are not freed, so any mappings are still in effect.
++ *
++ * NOTE: Care should be taken that the channel isn't actually
++ * being used by anything when this call is made.
++ */
++void relay_reset(struct rchan *chan)
++{
++ unsigned int i;
++ struct rchan_buf *prev = NULL;
++
++ if (!chan)
++ return;
++
++ for (i = 0; i < NR_CPUS; i++) {
++ if (!chan->buf[i] || chan->buf[i] == prev)
++ break;
++ __relay_reset(chan->buf[i], 0);
++ prev = chan->buf[i];
++ }
++}
++EXPORT_SYMBOL_GPL(relay_reset);
++
++/**
++ * relay_open_buf - create a new relay channel buffer
++ *
++ * Internal - used by relay_open().
++ */
++static struct rchan_buf *relay_open_buf(struct rchan *chan,
++ const char *filename,
++ struct dentry *parent,
++ int *is_global)
++{
++ struct rchan_buf *buf;
++ struct dentry *dentry;
++
++ if (*is_global)
++ return chan->buf[0];
++
++ buf = relay_create_buf(chan);
++ if (!buf)
++ return NULL;
++
++ /* Create file in fs */
++ dentry = chan->cb->create_buf_file(filename, parent, S_IRUSR,
++ buf, is_global);
++ if (!dentry) {
++ relay_destroy_buf(buf);
++ return NULL;
++ }
++
++ buf->dentry = dentry;
++ __relay_reset(buf, 1);
++
++ return buf;
++}
++
++/**
++ * relay_close_buf - close a channel buffer
++ * @buf: channel buffer
++ *
++ * Marks the buffer finalized and restores the default callbacks.
++ * The channel buffer and channel buffer data structure are then freed
++ * automatically when the last reference is given up.
++ */
++static inline void relay_close_buf(struct rchan_buf *buf)
++{
++ buf->finalized = 1;
++ cancel_delayed_work(&buf->wake_readers);
++ flush_scheduled_work();
++ kref_put(&buf->kref, relay_remove_buf);
++}
++
++static inline void setup_callbacks(struct rchan *chan,
++ struct rchan_callbacks *cb)
++{
++ if (!cb) {
++ chan->cb = &default_channel_callbacks;
++ return;
++ }
++
++ if (!cb->subbuf_start)
++ cb->subbuf_start = subbuf_start_default_callback;
++ if (!cb->buf_mapped)
++ cb->buf_mapped = buf_mapped_default_callback;
++ if (!cb->buf_unmapped)
++ cb->buf_unmapped = buf_unmapped_default_callback;
++ if (!cb->create_buf_file)
++ cb->create_buf_file = create_buf_file_default_callback;
++ if (!cb->remove_buf_file)
++ cb->remove_buf_file = remove_buf_file_default_callback;
++ chan->cb = cb;
++}
++
++/**
++ * relay_open - create a new relay channel
++ * @base_filename: base name of files to create
++ * @parent: dentry of parent directory, NULL for root directory
++ * @subbuf_size: size of sub-buffers
++ * @n_subbufs: number of sub-buffers
++ * @cb: client callback functions
++ *
++ * Returns channel pointer if successful, NULL otherwise.
++ *
++ * Creates a channel buffer for each cpu using the sizes and
++ * attributes specified. The created channel buffer files
++ * will be named base_filename0...base_filenameN-1. File
++ * permissions will be S_IRUSR.
++ */
++struct rchan *relay_open(const char *base_filename,
++ struct dentry *parent,
++ size_t subbuf_size,
++ size_t n_subbufs,
++ struct rchan_callbacks *cb)
++{
++ unsigned int i;
++ struct rchan *chan;
++ char *tmpname;
++ int is_global = 0;
++
++ if (!base_filename)
++ return NULL;
++
++ if (!(subbuf_size && n_subbufs))
++ return NULL;
++
++ chan = kcalloc(1, sizeof(struct rchan), GFP_KERNEL);
++ if (!chan)
++ return NULL;
++
++ chan->version = RELAYFS_CHANNEL_VERSION;
++ chan->n_subbufs = n_subbufs;
++ chan->subbuf_size = subbuf_size;
++ chan->alloc_size = FIX_SIZE(subbuf_size * n_subbufs);
++ setup_callbacks(chan, cb);
++ kref_init(&chan->kref);
++
++ tmpname = kmalloc(NAME_MAX + 1, GFP_KERNEL);
++ if (!tmpname)
++ goto free_chan;
++
++ for_each_online_cpu(i) {
++ sprintf(tmpname, "%s%d", base_filename, i);
++ chan->buf[i] = relay_open_buf(chan, tmpname, parent,
++ &is_global);
++ chan->buf[i]->cpu = i;
++ if (!chan->buf[i])
++ goto free_bufs;
++ }
++
++ kfree(tmpname);
++ return chan;
++
++free_bufs:
++ for (i = 0; i < NR_CPUS; i++) {
++ if (!chan->buf[i])
++ break;
++ relay_close_buf(chan->buf[i]);
++ if (is_global)
++ break;
++ }
++ kfree(tmpname);
++
++free_chan:
++ kref_put(&chan->kref, relay_destroy_channel);
++ return NULL;
++}
++EXPORT_SYMBOL_GPL(relay_open);
++
++/**
++ * relay_switch_subbuf - switch to a new sub-buffer
++ * @buf: channel buffer
++ * @length: size of current event
++ *
++ * Returns either the length passed in or 0 if full.
++ *
++ * Performs sub-buffer-switch tasks such as invoking callbacks,
++ * updating padding counts, waking up readers, etc.
++ */
++size_t relay_switch_subbuf(struct rchan_buf *buf, size_t length)
++{
++ void *old, *new;
++ size_t old_subbuf, new_subbuf;
++
++ if (unlikely(length > buf->chan->subbuf_size))
++ goto toobig;
++
++ if (buf->offset != buf->chan->subbuf_size + 1) {
++ buf->prev_padding = buf->chan->subbuf_size - buf->offset;
++ old_subbuf = buf->subbufs_produced % buf->chan->n_subbufs;
++ buf->padding[old_subbuf] = buf->prev_padding;
++ buf->subbufs_produced++;
++ buf->dentry->d_inode->i_size += buf->chan->subbuf_size -
++ buf->padding[old_subbuf];
++ buf->poll_flags = POLLMSG;
++ smp_mb();
++ if (waitqueue_active(&buf->read_wait)) {
++ PREPARE_WORK(&buf->wake_readers, wakeup_readers, buf);
++ schedule_delayed_work(&buf->wake_readers, 1);
++ }
++ }
++
++ old = buf->data;
++ new_subbuf = buf->subbufs_produced % buf->chan->n_subbufs;
++ new = buf->start + new_subbuf * buf->chan->subbuf_size;
++ buf->offset = 0;
++ if (!buf->chan->cb->subbuf_start(buf, new, old, buf->prev_padding)) {
++ buf->offset = buf->chan->subbuf_size + 1;
++ return 0;
++ }
++ buf->data = new;
++ buf->padding[new_subbuf] = 0;
++
++ if (unlikely(length + buf->offset > buf->chan->subbuf_size))
++ goto toobig;
++
++ return length;
++
++toobig:
++ buf->chan->last_toobig = length;
++ return 0;
++}
++EXPORT_SYMBOL_GPL(relay_switch_subbuf);
++
++/**
++ * relay_subbufs_consumed - update the buffer's sub-buffers-consumed count
++ * @chan: the channel
++ * @cpu: the cpu associated with the channel buffer to update
++ * @subbufs_consumed: number of sub-buffers to add to current buf's count
++ *
++ * Adds to the channel buffer's consumed sub-buffer count.
++ * subbufs_consumed should be the number of sub-buffers newly consumed,
++ * not the total consumed.
++ *
++ * NOTE: kernel clients don't need to call this function if the channel
++ * mode is 'overwrite'.
++ */
++void relay_subbufs_consumed(struct rchan *chan,
++ unsigned int cpu,
++ size_t subbufs_consumed)
++{
++ struct rchan_buf *buf;
++
++ if (!chan)
++ return;
++
++ if (cpu >= NR_CPUS || !chan->buf[cpu])
++ return;
++
++ buf = chan->buf[cpu];
++ buf->subbufs_consumed += subbufs_consumed;
++ if (buf->subbufs_consumed > buf->subbufs_produced)
++ buf->subbufs_consumed = buf->subbufs_produced;
++}
++EXPORT_SYMBOL_GPL(relay_subbufs_consumed);
++
++/**
++ * relay_close - close the channel
++ * @chan: the channel
++ *
++ * Closes all channel buffers and frees the channel.
++ */
++void relay_close(struct rchan *chan)
++{
++ unsigned int i;
++ struct rchan_buf *prev = NULL;
++
++ if (!chan)
++ return;
++
++ for (i = 0; i < NR_CPUS; i++) {
++ if (!chan->buf[i] || chan->buf[i] == prev)
++ break;
++ relay_close_buf(chan->buf[i]);
++ prev = chan->buf[i];
++ }
++
++ if (chan->last_toobig)
++ printk(KERN_WARNING "relay: one or more items not logged "
++ "[item size (%Zd) > sub-buffer size (%Zd)]\n",
++ chan->last_toobig, chan->subbuf_size);
++
++ kref_put(&chan->kref, relay_destroy_channel);
++}
++EXPORT_SYMBOL_GPL(relay_close);
++
++/**
++ * relay_flush - close the channel
++ * @chan: the channel
++ *
++ * Flushes all channel buffers i.e. forces buffer switch.
++ */
++void relay_flush(struct rchan *chan)
++{
++ unsigned int i;
++ struct rchan_buf *prev = NULL;
++
++ if (!chan)
++ return;
++
++ for (i = 0; i < NR_CPUS; i++) {
++ if (!chan->buf[i] || chan->buf[i] == prev)
++ break;
++ relay_switch_subbuf(chan->buf[i], 0);
++ prev = chan->buf[i];
++ }
++}
++EXPORT_SYMBOL_GPL(relay_flush);
++
++/**
++ * relay_file_open - open file op for relay files
++ * @inode: the inode
++ * @filp: the file
++ *
++ * Increments the channel buffer refcount.
++ */
++static int relay_file_open(struct inode *inode, struct file *filp)
++{
++ struct rchan_buf *buf = inode->u.generic_ip;
++ kref_get(&buf->kref);
++ filp->private_data = buf;
++
++ return 0;
++}
++
++/**
++ * relay_file_mmap - mmap file op for relay files
++ * @filp: the file
++ * @vma: the vma describing what to map
++ *
++ * Calls upon relay_mmap_buf to map the file into user space.
++ */
++static int relay_file_mmap(struct file *filp, struct vm_area_struct *vma)
++{
++ struct rchan_buf *buf = filp->private_data;
++ return relay_mmap_buf(buf, vma);
++}
++
++/**
++ * relay_file_poll - poll file op for relay files
++ * @filp: the file
++ * @wait: poll table
++ *
++ * Poll implemention.
++ */
++static unsigned int relay_file_poll(struct file *filp, poll_table *wait)
++{
++ unsigned int mask = 0;
++ struct rchan_buf *buf = filp->private_data;
++
++ if (buf->finalized)
++ return POLLERR;
++
++ if (filp->f_mode & FMODE_READ) {
++ poll_wait(filp, &buf->read_wait, wait);
++ if (!relay_buf_empty(buf))
++ mask |= POLLIN | POLLRDNORM | buf->poll_flags;
++ buf->poll_flags = 0;
++ }
++
++ return mask;
++}
++
++/**
++ * relay_file_release - release file op for relay files
++ * @inode: the inode
++ * @filp: the file
++ *
++ * Decrements the channel refcount, as the filesystem is
++ * no longer using it.
++ */
++static int relay_file_release(struct inode *inode, struct file *filp)
++{
++ struct rchan_buf *buf = filp->private_data;
++ kref_put(&buf->kref, relay_remove_buf);
++
++ return 0;
++}
++
++/**
++ * relay_file_read_consume - update the consumed count for the buffer
++ */
++static void relay_file_read_consume(struct rchan_buf *buf,
++ size_t read_pos,
++ size_t bytes_consumed)
++{
++ size_t subbuf_size = buf->chan->subbuf_size;
++ size_t n_subbufs = buf->chan->n_subbufs;
++ size_t read_subbuf;
++
++ if (buf->bytes_consumed + bytes_consumed > subbuf_size) {
++ relay_subbufs_consumed(buf->chan, buf->cpu, 1);
++ buf->bytes_consumed = 0;
++ }
++
++ buf->bytes_consumed += bytes_consumed;
++ read_subbuf = read_pos / buf->chan->subbuf_size;
++ if (buf->bytes_consumed + buf->padding[read_subbuf] == subbuf_size) {
++ if ((read_subbuf == buf->subbufs_produced % n_subbufs) &&
++ (buf->offset == subbuf_size))
++ return;
++ relay_subbufs_consumed(buf->chan, buf->cpu, 1);
++ buf->bytes_consumed = 0;
++ }
++}
++
++/**
++ * relay_file_read_avail - boolean, are there unconsumed bytes available?
++ */
++static int relay_file_read_avail(struct rchan_buf *buf, size_t read_pos)
++{
++ size_t subbuf_size = buf->chan->subbuf_size;
++ size_t n_subbufs = buf->chan->n_subbufs;
++ size_t produced = buf->subbufs_produced;
++ size_t consumed = buf->subbufs_consumed;
++
++ relay_file_read_consume(buf, read_pos, 0);
++
++ if (unlikely(buf->offset > subbuf_size)) {
++ if (produced == consumed)
++ return 0;
++ return 1;
++ }
++
++ if (unlikely(produced - consumed >= n_subbufs)) {
++ consumed = (produced / n_subbufs) * n_subbufs;
++ buf->subbufs_consumed = consumed;
++ }
++
++ produced = (produced % n_subbufs) * subbuf_size + buf->offset;
++ consumed = (consumed % n_subbufs) * subbuf_size + buf->bytes_consumed;
++
++ if (consumed > produced)
++ produced += n_subbufs * subbuf_size;
++
++ if (consumed == produced)
++ return 0;
++
++ return 1;
++}
++
++/**
++ * relay_file_read_subbuf_avail - return bytes available in sub-buffer
++ */
++static size_t relay_file_read_subbuf_avail(size_t read_pos,
++ struct rchan_buf *buf)
++{
++ size_t padding, avail = 0;
++ size_t read_subbuf, read_offset, write_subbuf, write_offset;
++ size_t subbuf_size = buf->chan->subbuf_size;
++
++ write_subbuf = (buf->data - buf->start) / subbuf_size;
++ write_offset = buf->offset > subbuf_size ? subbuf_size : buf->offset;
++ read_subbuf = read_pos / subbuf_size;
++ read_offset = read_pos % subbuf_size;
++ padding = buf->padding[read_subbuf];
++
++ if (read_subbuf == write_subbuf) {
++ if (read_offset + padding < write_offset)
++ avail = write_offset - (read_offset + padding);
++ } else
++ avail = (subbuf_size - padding) - read_offset;
++
++ return avail;
++}
++
++/**
++ * relay_file_read_start_pos - find the first available byte to read
++ *
++ * If the read_pos is in the middle of padding, return the
++ * position of the first actually available byte, otherwise
++ * return the original value.
++ */
++static size_t relay_file_read_start_pos(size_t read_pos,
++ struct rchan_buf *buf)
++{
++ size_t read_subbuf, padding, padding_start, padding_end;
++ size_t subbuf_size = buf->chan->subbuf_size;
++ size_t n_subbufs = buf->chan->n_subbufs;
++
++ read_subbuf = read_pos / subbuf_size;
++ padding = buf->padding[read_subbuf];
++ padding_start = (read_subbuf + 1) * subbuf_size - padding;
++ padding_end = (read_subbuf + 1) * subbuf_size;
++ if (read_pos >= padding_start && read_pos < padding_end) {
++ read_subbuf = (read_subbuf + 1) % n_subbufs;
++ read_pos = read_subbuf * subbuf_size;
++ }
++
++ return read_pos;
++}
++
++/**
++ * relay_file_read_end_pos - return the new read position
++ */
++static size_t relay_file_read_end_pos(struct rchan_buf *buf,
++ size_t read_pos,
++ size_t count)
++{
++ size_t read_subbuf, padding, end_pos;
++ size_t subbuf_size = buf->chan->subbuf_size;
++ size_t n_subbufs = buf->chan->n_subbufs;
++
++ read_subbuf = read_pos / subbuf_size;
++ padding = buf->padding[read_subbuf];
++ if (read_pos % subbuf_size + count + padding == subbuf_size)
++ end_pos = (read_subbuf + 1) * subbuf_size;
++ else
++ end_pos = read_pos + count;
++ if (end_pos >= subbuf_size * n_subbufs)
++ end_pos = 0;
++
++ return end_pos;
++}
++
++/**
++ * relay_file_read - read file op for relay files
++ * @filp: the file
++ * @buffer: the userspace buffer
++ * @count: number of bytes to read
++ * @ppos: position to read from
++ *
++ * Reads count bytes or the number of bytes available in the
++ * current sub-buffer being read, whichever is smaller.
++ */
++static ssize_t relay_file_read(struct file *filp,
++ char __user *buffer,
++ size_t count,
++ loff_t *ppos)
++{
++ struct rchan_buf *buf = filp->private_data;
++ struct inode *inode = filp->f_dentry->d_inode;
++ size_t read_start, avail;
++ ssize_t ret = 0;
++ void *from;
++
++ mutex_lock(&inode->i_mutex);
++ if(!relay_file_read_avail(buf, *ppos))
++ goto out;
++
++ read_start = relay_file_read_start_pos(*ppos, buf);
++ avail = relay_file_read_subbuf_avail(read_start, buf);
++ if (!avail)
++ goto out;
++
++ from = buf->start + read_start;
++ ret = count = min(count, avail);
++ if (copy_to_user(buffer, from, count)) {
++ ret = -EFAULT;
++ goto out;
++ }
++ relay_file_read_consume(buf, read_start, count);
++ *ppos = relay_file_read_end_pos(buf, read_start, count);
++out:
++ mutex_unlock(&inode->i_mutex);
++ return ret;
++}
++
++static ssize_t relay_file_sendsubbuf(struct file *filp, loff_t *ppos,
++ size_t count, read_actor_t actor,
++ void *target)
++{
++ struct rchan_buf *buf = filp->private_data;
++ read_descriptor_t desc;
++ size_t read_start, avail;
++ unsigned long pidx, poff;
++ unsigned int subbuf_pages;
++ ssize_t ret = 0;
++
++ if (!relay_file_read_avail(buf, *ppos))
++ return 0;
++
++ read_start = relay_file_read_start_pos(*ppos, buf);
++ avail = relay_file_read_subbuf_avail(read_start, buf);
++ if (!avail)
++ return 0;
++
++ count = min(count, avail);
++
++ desc.written = 0;
++ desc.count = count;
++ desc.arg.data = target;
++ desc.error = 0;
++
++ subbuf_pages = buf->chan->alloc_size >> PAGE_SHIFT;
++ pidx = (read_start / PAGE_SIZE) % subbuf_pages;
++ poff = read_start & ~PAGE_MASK;
++ while (count) {
++ struct page *p = buf->page_array[pidx];
++ unsigned int len;
++
++ len = PAGE_SIZE - poff;
++ if (len > count)
++ len = count;
++
++ len = actor(&desc, p, poff, len);
++
++ if (desc.error) {
++ if (!ret)
++ ret = desc.error;
++ break;
++ }
++
++ count -= len;
++ ret += len;
++ poff = 0;
++ pidx = (pidx + 1) % subbuf_pages;
++ }
++
++ if (ret > 0) {
++ relay_file_read_consume(buf, read_start, ret);
++ *ppos = relay_file_read_end_pos(buf, read_start, ret);
++ }
++
++ return ret;
++}
++
++static ssize_t relay_file_sendfile(struct file *filp, loff_t *ppos,
++ size_t count, read_actor_t actor,
++ void *target)
++{
++ ssize_t sent = 0, ret = 0;
++
++ if (!count)
++ return 0;
++
++ mutex_lock(&filp->f_dentry->d_inode->i_mutex);
++
++ do {
++ ret = relay_file_sendsubbuf(filp, ppos, count, actor, target);
++ if (ret < 0) {
++ if (!sent)
++ sent = ret;
++ break;
++ }
++ count -= ret;
++ sent += ret;
++ } while (count && ret);
++
++ mutex_unlock(&filp->f_dentry->d_inode->i_mutex);
++ return sent;
++}
++
++struct file_operations relay_file_operations = {
++ .open = relay_file_open,
++ .poll = relay_file_poll,
++ .mmap = relay_file_mmap,
++ .read = relay_file_read,
++ .llseek = no_llseek,
++ .release = relay_file_release,
++ .sendfile = relay_file_sendfile,
++};
++EXPORT_SYMBOL_GPL(relay_file_operations);
+diff --git a/mm/highmem.c b/mm/highmem.c
+index ce2e7e8..d0ea1ee 100644
+--- a/mm/highmem.c
++++ b/mm/highmem.c
+@@ -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
+ pool = isa_page_pool;
+ }
+
++ blk_add_trace_bio(q, *bio_orig, BLK_TA_BOUNCE);
++
+ /*
+ * slow path
+ */
projects / blktrace.git / commitdiff