aboutsummaryrefslogtreecommitdiff
path: root/fs
diff options
context:
space:
mode:
Diffstat (limited to 'fs')
-rw-r--r--fs/9p/vfs_dir.c93
-rw-r--r--fs/9p/vfs_inode.c5
-rw-r--r--fs/Kconfig4
-rw-r--r--fs/bio.c28
-rw-r--r--fs/block_dev.c2
-rw-r--r--fs/btrfs/acl.c6
-rw-r--r--fs/btrfs/async-thread.c81
-rw-r--r--fs/btrfs/async-thread.h10
-rw-r--r--fs/btrfs/btrfs_inode.h18
-rw-r--r--fs/btrfs/ctree.h19
-rw-r--r--fs/btrfs/disk-io.c50
-rw-r--r--fs/btrfs/extent-tree.c365
-rw-r--r--fs/btrfs/extent_io.c42
-rw-r--r--fs/btrfs/extent_io.h18
-rw-r--r--fs/btrfs/extent_map.c2
-rw-r--r--fs/btrfs/file.c44
-rw-r--r--fs/btrfs/free-space-cache.c2
-rw-r--r--fs/btrfs/inode.c239
-rw-r--r--fs/btrfs/ioctl.c7
-rw-r--r--fs/btrfs/ordered-data.c6
-rw-r--r--fs/btrfs/relocation.c4
-rw-r--r--fs/btrfs/root-tree.c2
-rw-r--r--fs/btrfs/super.c9
-rw-r--r--fs/btrfs/transaction.c64
-rw-r--r--fs/btrfs/transaction.h5
-rw-r--r--fs/btrfs/tree-log.c56
-rw-r--r--fs/btrfs/tree-log.h3
-rw-r--r--fs/btrfs/xattr.c2
-rw-r--r--fs/cifs/cifsproto.h1
-rw-r--r--fs/cifs/connect.c11
-rw-r--r--fs/cifs/inode.c7
-rw-r--r--fs/cifs/misc.c14
-rw-r--r--fs/cifs/readdir.c7
-rw-r--r--fs/compat.c2
-rw-r--r--fs/compat_ioctl.c4
-rw-r--r--fs/dlm/lowcomms.c36
-rw-r--r--fs/ecryptfs/Kconfig3
-rw-r--r--fs/ecryptfs/main.c7
-rw-r--r--fs/exec.c8
-rw-r--r--fs/ext3/fsync.c36
-rw-r--r--fs/ext3/inode.c36
-rw-r--r--fs/ext3/super.c15
-rw-r--r--fs/ext4/ext4.h2
-rw-r--r--fs/ext4/extents.c36
-rw-r--r--fs/ext4/inode.c24
-rw-r--r--fs/ext4/namei.c16
-rw-r--r--fs/ext4/super.c20
-rw-r--r--fs/fcntl.c4
-rw-r--r--fs/file.c1
-rw-r--r--fs/fuse/dir.c4
-rw-r--r--fs/fuse/file.c5
-rw-r--r--fs/hfs/btree.c5
-rw-r--r--fs/hfsplus/wrapper.c4
-rw-r--r--fs/ioctl.c2
-rw-r--r--fs/jbd/journal.c3
-rw-r--r--fs/jbd2/journal.c2
-rw-r--r--fs/nfs/client.c2
-rw-r--r--fs/nfs/dir.c2
-rw-r--r--fs/nfs/direct.c1
-rw-r--r--fs/nfs/nfs4namespace.c12
-rw-r--r--fs/nfs/nfs4proc.c17
-rw-r--r--fs/nfs/nfs4renewd.c6
-rw-r--r--fs/nfs/nfs4xdr.c1
-rw-r--r--fs/nfs/super.c37
-rw-r--r--fs/nfsd/nfs3xdr.c2
-rw-r--r--fs/nilfs2/btnode.c4
-rw-r--r--fs/nilfs2/cpfile.c2
-rw-r--r--fs/nilfs2/inode.c1
-rw-r--r--fs/nilfs2/ioctl.c39
-rw-r--r--fs/nilfs2/segment.c17
-rw-r--r--fs/notify/dnotify/dnotify.c3
-rw-r--r--fs/notify/inode_mark.c6
-rw-r--r--fs/notify/notification.c2
-rw-r--r--fs/partitions/check.c12
-rw-r--r--fs/pipe.c41
-rw-r--r--fs/proc/array.c2
-rw-r--r--fs/proc/base.c3
-rw-r--r--fs/proc/kcore.c1
-rw-r--r--fs/proc/meminfo.c2
-rw-r--r--fs/proc/page.c5
-rw-r--r--fs/romfs/storage.c4
-rw-r--r--fs/sysfs/dir.c7
-rw-r--r--fs/sysfs/file.c14
-rw-r--r--fs/xfs/linux-2.6/xfs_aops.c38
-rw-r--r--fs/xfs/linux-2.6/xfs_file.c9
-rw-r--r--fs/xfs/linux-2.6/xfs_iops.c41
-rw-r--r--fs/xfs/linux-2.6/xfs_lrw.c2
-rw-r--r--fs/xfs/linux-2.6/xfs_quotaops.c2
-rw-r--r--fs/xfs/linux-2.6/xfs_super.c59
-rw-r--r--fs/xfs/linux-2.6/xfs_sync.c36
-rw-r--r--fs/xfs/quota/xfs_qm_syscalls.c1
-rw-r--r--fs/xfs/xfs_dfrag.c8
-rw-r--r--fs/xfs/xfs_dir2_leaf.c4
-rw-r--r--fs/xfs/xfs_ialloc.c1
-rw-r--r--fs/xfs/xfs_inode.c4
-rw-r--r--fs/xfs/xfs_inode.h2
-rw-r--r--fs/xfs/xfs_inode_item.c18
-rw-r--r--fs/xfs/xfs_itable.c21
-rw-r--r--fs/xfs/xfs_log_recover.c4
-rw-r--r--fs/xfs/xfs_trans_ail.c23
-rw-r--r--fs/xfs/xfs_vnodeops.c6
101 files changed, 1414 insertions, 611 deletions
diff --git a/fs/9p/vfs_dir.c b/fs/9p/vfs_dir.c
index 873cd31baa4..15cce53bf61 100644
--- a/fs/9p/vfs_dir.c
+++ b/fs/9p/vfs_dir.c
@@ -40,6 +40,24 @@
#include "fid.h"
/**
+ * struct p9_rdir - readdir accounting
+ * @mutex: mutex protecting readdir
+ * @head: start offset of current dirread buffer
+ * @tail: end offset of current dirread buffer
+ * @buf: dirread buffer
+ *
+ * private structure for keeping track of readdir
+ * allocated on demand
+ */
+
+struct p9_rdir {
+ struct mutex mutex;
+ int head;
+ int tail;
+ uint8_t *buf;
+};
+
+/**
* dt_type - return file type
* @mistat: mistat structure
*
@@ -70,56 +88,79 @@ static int v9fs_dir_readdir(struct file *filp, void *dirent, filldir_t filldir)
{
int over;
struct p9_wstat st;
- int err;
+ int err = 0;
struct p9_fid *fid;
int buflen;
- char *statbuf;
- int n, i = 0;
+ int reclen = 0;
+ struct p9_rdir *rdir;
P9_DPRINTK(P9_DEBUG_VFS, "name %s\n", filp->f_path.dentry->d_name.name);
fid = filp->private_data;
buflen = fid->clnt->msize - P9_IOHDRSZ;
- statbuf = kmalloc(buflen, GFP_KERNEL);
- if (!statbuf)
- return -ENOMEM;
-
- while (1) {
- err = v9fs_file_readn(filp, statbuf, NULL, buflen,
- fid->rdir_fpos);
- if (err <= 0)
- break;
-
- n = err;
- while (i < n) {
- err = p9stat_read(statbuf + i, buflen-i, &st,
- fid->clnt->dotu);
+
+ /* allocate rdir on demand */
+ if (!fid->rdir) {
+ rdir = kmalloc(sizeof(struct p9_rdir) + buflen, GFP_KERNEL);
+
+ if (rdir == NULL) {
+ err = -ENOMEM;
+ goto exit;
+ }
+ spin_lock(&filp->f_dentry->d_lock);
+ if (!fid->rdir) {
+ rdir->buf = (uint8_t *)rdir + sizeof(struct p9_rdir);
+ mutex_init(&rdir->mutex);
+ rdir->head = rdir->tail = 0;
+ fid->rdir = (void *) rdir;
+ rdir = NULL;
+ }
+ spin_unlock(&filp->f_dentry->d_lock);
+ kfree(rdir);
+ }
+ rdir = (struct p9_rdir *) fid->rdir;
+
+ err = mutex_lock_interruptible(&rdir->mutex);
+ while (err == 0) {
+ if (rdir->tail == rdir->head) {
+ err = v9fs_file_readn(filp, rdir->buf, NULL,
+ buflen, filp->f_pos);
+ if (err <= 0)
+ goto unlock_and_exit;
+
+ rdir->head = 0;
+ rdir->tail = err;
+ }
+
+ while (rdir->head < rdir->tail) {
+ err = p9stat_read(rdir->buf + rdir->head,
+ buflen - rdir->head, &st,
+ fid->clnt->dotu);
if (err) {
P9_DPRINTK(P9_DEBUG_VFS, "returned %d\n", err);
err = -EIO;
p9stat_free(&st);
- goto free_and_exit;
+ goto unlock_and_exit;
}
-
- i += st.size+2;
- fid->rdir_fpos += st.size+2;
+ reclen = st.size+2;
over = filldir(dirent, st.name, strlen(st.name),
filp->f_pos, v9fs_qid2ino(&st.qid), dt_type(&st));
- filp->f_pos += st.size+2;
-
p9stat_free(&st);
if (over) {
err = 0;
- goto free_and_exit;
+ goto unlock_and_exit;
}
+ rdir->head += reclen;
+ filp->f_pos += reclen;
}
}
-free_and_exit:
- kfree(statbuf);
+unlock_and_exit:
+ mutex_unlock(&rdir->mutex);
+exit:
return err;
}
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index 5947628aefe..18f74ec4dce 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -994,8 +994,7 @@ static int v9fs_readlink(struct dentry *dentry, char *buffer, int buflen)
P9_DPRINTK(P9_DEBUG_VFS,
"%s -> %s (%s)\n", dentry->d_name.name, st->extension, buffer);
- retval = buflen;
-
+ retval = strnlen(buffer, buflen);
done:
kfree(st);
return retval;
@@ -1062,7 +1061,7 @@ static void *v9fs_vfs_follow_link(struct dentry *dentry, struct nameidata *nd)
__putname(link);
link = ERR_PTR(len);
} else
- link[len] = 0;
+ link[min(len, PATH_MAX-1)] = 0;
}
nd_set_link(nd, link);
diff --git a/fs/Kconfig b/fs/Kconfig
index d4bf8caad8d..64d44efad7a 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -135,8 +135,8 @@ config TMPFS_POSIX_ACL
config HUGETLBFS
bool "HugeTLB file system support"
- depends on X86 || IA64 || PPC64 || SPARC64 || (SUPERH && MMU) || \
- (S390 && 64BIT) || SYS_SUPPORTS_HUGETLBFS || BROKEN
+ depends on X86 || IA64 || SPARC64 || (S390 && 64BIT) || \
+ SYS_SUPPORTS_HUGETLBFS || BROKEN
help
hugetlbfs is a filesystem backing for HugeTLB pages, based on
ramfs. For architectures that support it, say Y here and read
diff --git a/fs/bio.c b/fs/bio.c
index 402cb84a92a..12da5db8682 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -325,8 +325,16 @@ static void bio_fs_destructor(struct bio *bio)
* @gfp_mask: allocation mask to use
* @nr_iovecs: number of iovecs
*
- * Allocate a new bio with @nr_iovecs bvecs. If @gfp_mask
- * contains __GFP_WAIT, the allocation is guaranteed to succeed.
+ * bio_alloc will allocate a bio and associated bio_vec array that can hold
+ * at least @nr_iovecs entries. Allocations will be done from the
+ * fs_bio_set. Also see @bio_alloc_bioset and @bio_kmalloc.
+ *
+ * If %__GFP_WAIT is set, then bio_alloc will always be able to allocate
+ * a bio. This is due to the mempool guarantees. To make this work, callers
+ * must never allocate more than 1 bio at a time from this pool. Callers
+ * that need to allocate more than 1 bio must always submit the previously
+ * allocated bio for IO before attempting to allocate a new one. Failure to
+ * do so can cause livelocks under memory pressure.
*
* RETURNS:
* Pointer to new bio on success, NULL on failure.
@@ -350,21 +358,13 @@ static void bio_kmalloc_destructor(struct bio *bio)
}
/**
- * bio_alloc - allocate a bio for I/O
+ * bio_kmalloc - allocate a bio for I/O using kmalloc()
* @gfp_mask: the GFP_ mask given to the slab allocator
* @nr_iovecs: number of iovecs to pre-allocate
*
* Description:
- * bio_alloc will allocate a bio and associated bio_vec array that can hold
- * at least @nr_iovecs entries. Allocations will be done from the
- * fs_bio_set. Also see @bio_alloc_bioset.
- *
- * If %__GFP_WAIT is set, then bio_alloc will always be able to allocate
- * a bio. This is due to the mempool guarantees. To make this work, callers
- * must never allocate more than 1 bio at a time from this pool. Callers
- * that need to allocate more than 1 bio must always submit the previously
- * allocated bio for IO before attempting to allocate a new one. Failure to
- * do so can cause livelocks under memory pressure.
+ * Allocate a new bio with @nr_iovecs bvecs. If @gfp_mask contains
+ * %__GFP_WAIT, the allocation is guaranteed to succeed.
*
**/
struct bio *bio_kmalloc(gfp_t gfp_mask, int nr_iovecs)
@@ -407,7 +407,7 @@ EXPORT_SYMBOL(zero_fill_bio);
*
* Description:
* Put a reference to a &struct bio, either one you have gotten with
- * bio_alloc or bio_get. The last put of a bio will free it.
+ * bio_alloc, bio_get or bio_clone. The last put of a bio will free it.
**/
void bio_put(struct bio *bio)
{
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 9cf4b926f8e..8bed0557d88 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1248,8 +1248,8 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
bd_set_size(bdev, (loff_t)bdev->bd_part->nr_sects << 9);
}
} else {
- put_disk(disk);
module_put(disk->fops->owner);
+ put_disk(disk);
disk = NULL;
if (bdev->bd_contains == bdev) {
if (bdev->bd_disk->fops->open) {
diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c
index 69b355ae7f4..36160424427 100644
--- a/fs/btrfs/acl.c
+++ b/fs/btrfs/acl.c
@@ -27,7 +27,7 @@
#include "btrfs_inode.h"
#include "xattr.h"
-#ifdef CONFIG_BTRFS_POSIX_ACL
+#ifdef CONFIG_BTRFS_FS_POSIX_ACL
static struct posix_acl *btrfs_get_acl(struct inode *inode, int type)
{
@@ -313,7 +313,7 @@ struct xattr_handler btrfs_xattr_acl_access_handler = {
.set = btrfs_xattr_acl_access_set,
};
-#else /* CONFIG_BTRFS_POSIX_ACL */
+#else /* CONFIG_BTRFS_FS_POSIX_ACL */
int btrfs_acl_chmod(struct inode *inode)
{
@@ -325,4 +325,4 @@ int btrfs_init_acl(struct inode *inode, struct inode *dir)
return 0;
}
-#endif /* CONFIG_BTRFS_POSIX_ACL */
+#endif /* CONFIG_BTRFS_FS_POSIX_ACL */
diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index 282ca085c2f..c0861e781cd 100644
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -64,6 +64,51 @@ struct btrfs_worker_thread {
};
/*
+ * btrfs_start_workers uses kthread_run, which can block waiting for memory
+ * for a very long time. It will actually throttle on page writeback,
+ * and so it may not make progress until after our btrfs worker threads
+ * process all of the pending work structs in their queue
+ *
+ * This means we can't use btrfs_start_workers from inside a btrfs worker
+ * thread that is used as part of cleaning dirty memory, which pretty much
+ * involves all of the worker threads.
+ *
+ * Instead we have a helper queue who never has more than one thread
+ * where we scheduler thread start operations. This worker_start struct
+ * is used to contain the work and hold a pointer to the queue that needs
+ * another worker.
+ */
+struct worker_start {
+ struct btrfs_work work;
+ struct btrfs_workers *queue;
+};
+
+static void start_new_worker_func(struct btrfs_work *work)
+{
+ struct worker_start *start;
+ start = container_of(work, struct worker_start, work);
+ btrfs_start_workers(start->queue, 1);
+ kfree(start);
+}
+
+static int start_new_worker(struct btrfs_workers *queue)
+{
+ struct worker_start *start;
+ int ret;
+
+ start = kzalloc(sizeof(*start), GFP_NOFS);
+ if (!start)
+ return -ENOMEM;
+
+ start->work.func = start_new_worker_func;
+ start->queue = queue;
+ ret = btrfs_queue_worker(queue->atomic_worker_start, &start->work);
+ if (ret)
+ kfree(start);
+ return ret;
+}
+
+/*
* helper function to move a thread onto the idle list after it
* has finished some requests.
*/
@@ -118,11 +163,13 @@ static void check_pending_worker_creates(struct btrfs_worker_thread *worker)
goto out;
workers->atomic_start_pending = 0;
- if (workers->num_workers >= workers->max_workers)
+ if (workers->num_workers + workers->num_workers_starting >=
+ workers->max_workers)
goto out;
+ workers->num_workers_starting += 1;
spin_unlock_irqrestore(&workers->lock, flags);
- btrfs_start_workers(workers, 1);
+ start_new_worker(workers);
return;
out:
@@ -390,9 +437,11 @@ int btrfs_stop_workers(struct btrfs_workers *workers)
/*
* simple init on struct btrfs_workers
*/
-void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max)
+void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max,
+ struct btrfs_workers *async_helper)
{
workers->num_workers = 0;
+ workers->num_workers_starting = 0;
INIT_LIST_HEAD(&workers->worker_list);
INIT_LIST_HEAD(&workers->idle_list);
INIT_LIST_HEAD(&workers->order_list);
@@ -404,14 +453,15 @@ void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max)
workers->name = name;
workers->ordered = 0;
workers->atomic_start_pending = 0;
- workers->atomic_worker_start = 0;
+ workers->atomic_worker_start = async_helper;
}
/*
* starts new worker threads. This does not enforce the max worker
* count in case you need to temporarily go past it.
*/
-int btrfs_start_workers(struct btrfs_workers *workers, int num_workers)
+static int __btrfs_start_workers(struct btrfs_workers *workers,
+ int num_workers)
{
struct btrfs_worker_thread *worker;
int ret = 0;
@@ -444,6 +494,8 @@ int btrfs_start_workers(struct btrfs_workers *workers, int num_workers)
list_add_tail(&worker->worker_list, &workers->idle_list);
worker->idle = 1;
workers->num_workers++;
+ workers->num_workers_starting--;
+ WARN_ON(workers->num_workers_starting < 0);
spin_unlock_irq(&workers->lock);
}
return 0;
@@ -452,6 +504,14 @@ fail:
return ret;
}
+int btrfs_start_workers(struct btrfs_workers *workers, int num_workers)
+{
+ spin_lock_irq(&workers->lock);
+ workers->num_workers_starting += num_workers;
+ spin_unlock_irq(&workers->lock);
+ return __btrfs_start_workers(workers, num_workers);
+}
+
/*
* run through the list and find a worker thread that doesn't have a lot
* to do right now. This can return null if we aren't yet at the thread
@@ -461,7 +521,10 @@ static struct btrfs_worker_thread *next_worker(struct btrfs_workers *workers)
{
struct btrfs_worker_thread *worker;
struct list_head *next;
- int enforce_min = workers->num_workers < workers->max_workers;
+ int enforce_min;
+
+ enforce_min = (workers->num_workers + workers->num_workers_starting) <
+ workers->max_workers;
/*
* if we find an idle thread, don't move it to the end of the
@@ -509,15 +572,17 @@ again:
worker = next_worker(workers);
if (!worker) {
- if (workers->num_workers >= workers->max_workers) {
+ if (workers->num_workers + workers->num_workers_starting >=
+ workers->max_workers) {
goto fallback;
} else if (workers->atomic_worker_start) {
workers->atomic_start_pending = 1;
goto fallback;
} else {
+ workers->num_workers_starting++;
spin_unlock_irqrestore(&workers->lock, flags);
/* we're below the limit, start another worker */
- btrfs_start_workers(workers, 1);
+ __btrfs_start_workers(workers, 1);
goto again;
}
}
diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h
index fc089b95ec1..5077746cf85 100644
--- a/fs/btrfs/async-thread.h
+++ b/fs/btrfs/async-thread.h
@@ -64,6 +64,8 @@ struct btrfs_workers {
/* current number of running workers */
int num_workers;
+ int num_workers_starting;
+
/* max number of workers allowed. changed by btrfs_start_workers */
int max_workers;
@@ -78,9 +80,10 @@ struct btrfs_workers {
/*
* are we allowed to sleep while starting workers or are we required
- * to start them at a later time?
+ * to start them at a later time? If we can't sleep, this indicates
+ * which queue we need to use to schedule thread creation.
*/
- int atomic_worker_start;
+ struct btrfs_workers *atomic_worker_start;
/* list with all the work threads. The workers on the idle thread
* may be actively servicing jobs, but they haven't yet hit the
@@ -109,7 +112,8 @@ struct btrfs_workers {
int btrfs_queue_worker(struct btrfs_workers *workers, struct btrfs_work *work);
int btrfs_start_workers(struct btrfs_workers *workers, int num_workers);
int btrfs_stop_workers(struct btrfs_workers *workers);
-void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max);
+void btrfs_init_workers(struct btrfs_workers *workers, char *name, int max,
+ struct btrfs_workers *async_starter);
int btrfs_requeue_work(struct btrfs_work *work);
void btrfs_set_work_high_prio(struct btrfs_work *work);
#endif
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index a54d354cefc..f6783a42f01 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -86,6 +86,12 @@ struct btrfs_inode {
* transid of the trans_handle that last modified this inode
*/
u64 last_trans;
+
+ /*
+ * log transid when this inode was last modified
+ */
+ u64 last_sub_trans;
+
/*
* transid that last logged this inode
*/
@@ -128,12 +134,14 @@ struct btrfs_inode {
u64 last_unlink_trans;
/*
- * These two counters are for delalloc metadata reservations. We keep
- * track of how many extents we've accounted for vs how many extents we
- * have.
+ * Counters to keep track of the number of extent item's we may use due
+ * to delalloc and such. outstanding_extents is the number of extent
+ * items we think we'll end up using, and reserved_extents is the number
+ * of extent items we've reserved metadata for.
*/
- int delalloc_reserved_extents;
- int delalloc_extents;
+ spinlock_t accounting_lock;
+ int reserved_extents;
+ int outstanding_extents;
/*
* ordered_data_close is set by truncate when a file that used
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index dd8ced9814c..444b3e9b92a 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -691,14 +691,17 @@ struct btrfs_space_info {
struct list_head list;
+ /* for controlling how we free up space for allocations */
+ wait_queue_head_t allocate_wait;
+ wait_queue_head_t flush_wait;
+ int allocating_chunk;
+ int flushing;
+
/* for block groups in our same type */
struct list_head block_groups;
spinlock_t lock;
struct rw_semaphore groups_sem;
atomic_t caching_threads;
-
- int allocating_chunk;
- wait_queue_head_t wait;
};
/*
@@ -907,6 +910,7 @@ struct btrfs_fs_info {
* A third pool does submit_bio to avoid deadlocking with the other
* two
*/
+ struct btrfs_workers generic_worker;
struct btrfs_workers workers;
struct btrfs_workers delalloc_workers;
struct btrfs_workers endio_workers;
@@ -914,6 +918,7 @@ struct btrfs_fs_info {
struct btrfs_workers endio_meta_write_workers;
struct btrfs_workers endio_write_workers;
struct btrfs_workers submit_workers;
+ struct btrfs_workers enospc_workers;
/*
* fixup workers take dirty pages that didn't properly go through
* the cow mechanism and make them safe to write. It happens
@@ -1004,7 +1009,10 @@ struct btrfs_root {
atomic_t log_writers;
atomic_t log_commit[2];
unsigned long log_transid;
+ unsigned long last_log_commit;
unsigned long log_batch;
+ pid_t log_start_pid;
+ bool log_multiple_pids;
u64 objectid;
u64 last_trans;
@@ -1145,6 +1153,7 @@ struct btrfs_root {
#define BTRFS_MOUNT_FLUSHONCOMMIT (1 << 7)
#define BTRFS_MOUNT_SSD_SPREAD (1 << 8)
#define BTRFS_MOUNT_NOSSD (1 << 9)
+#define BTRFS_MOUNT_DISCARD (1 << 10)
#define btrfs_clear_opt(o, opt) ((o) &= ~BTRFS_MOUNT_##opt)
#define btrfs_set_opt(o, opt) ((o) |= BTRFS_MOUNT_##opt)
@@ -2323,7 +2332,7 @@ int btrfs_orphan_del(struct btrfs_trans_handle *trans, struct inode *inode);
void btrfs_orphan_cleanup(struct btrfs_root *root);
int btrfs_cont_expand(struct inode *inode, loff_t size);
int btrfs_invalidate_inodes(struct btrfs_root *root);
-extern struct dentry_operations btrfs_dentry_operations;
+extern const struct dentry_operations btrfs_dentry_operations;
/* ioctl.c */
long btrfs_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
@@ -2366,7 +2375,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options);
int btrfs_sync_fs(struct super_block *sb, int wait);
/* acl.c */
-#ifdef CONFIG_BTRFS_POSIX_ACL
+#ifdef CONFIG_BTRFS_FS_POSIX_ACL
int btrfs_check_acl(struct inode *inode, int mask);
#else
#define btrfs_check_acl NULL
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index af0435f79fa..02b6afbd745 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -917,6 +917,7 @@ static int __setup_root(u32 nodesize, u32 leafsize, u32 sectorsize,
atomic_set(&root->log_writers, 0);
root->log_batch = 0;
root->log_transid = 0;
+ root->last_log_commit = 0;
extent_io_tree_init(&root->dirty_log_pages,
fs_info->btree_inode->i_mapping, GFP_NOFS);
@@ -1087,6 +1088,7 @@ int btrfs_add_log_tree(struct btrfs_trans_handle *trans,
WARN_ON(root->log_root);
root->log_root = log_root;
root->log_transid = 0;
+ root->last_log_commit = 0;
return 0;
}
@@ -1746,21 +1748,25 @@ struct btrfs_root *open_ctree(struct super_block *sb,
err = -EINVAL;
goto fail_iput;
}
-printk("thread pool is %d\n", fs_info->thread_pool_size);
- /*
- * we need to start all the end_io workers up front because the
- * queue work function gets called at interrupt time, and so it
- * cannot dynamically grow.
- */
+
+ btrfs_init_workers(&fs_info->generic_worker,
+ "genwork", 1, NULL);
+
btrfs_init_workers(&fs_info->workers, "worker",
- fs_info->thread_pool_size);
+ fs_info->thread_pool_size,
+ &fs_info->generic_worker);
btrfs_init_workers(&fs_info->delalloc_workers, "delalloc",
- fs_info->thread_pool_size);
+ fs_info->thread_pool_size,
+ &fs_info->generic_worker);
btrfs_init_workers(&fs_info->submit_workers, "submit",
min_t(u64, fs_devices->num_devices,
- fs_info->thread_pool_size));
+ fs_info->thread_pool_size),
+ &fs_info->generic_worker);
+ btrfs_init_workers(&fs_info->enospc_workers, "enospc",
+ fs_info->thread_pool_size,
+ &fs_info->generic_worker);
/* a higher idle thresh on the submit workers makes it much more
* likely that bios will be send down in a sane order to the
@@ -1774,15 +1780,20 @@ printk("thread pool is %d\n", fs_info->thread_pool_size);
fs_info->delalloc_workers.idle_thresh = 2;
fs_info->delalloc_workers.ordered = 1;
- btrfs_init_workers(&fs_info->fixup_workers, "fixup", 1);
+ btrfs_init_workers(&fs_info->fixup_workers, "fixup", 1,
+ &fs_info->generic_worker);
btrfs_init_workers(&fs_info->endio_workers, "endio",
- fs_info->thread_pool_size);
+ fs_info->thread_pool_size,
+ &fs_info->generic_worker);
btrfs_init_workers(&fs_info->endio_meta_workers, "endio-meta",
- fs_info->thread_pool_size);
+ fs_info->thread_pool_size,
+ &fs_info->generic_worker);
btrfs_init_workers(&fs_info->endio_meta_write_workers,
- "endio-meta-write", fs_info->thread_pool_size);
+ "endio-meta-write", fs_info->thread_pool_size,
+ &fs_info->generic_worker);
btrfs_init_workers(&fs_info->endio_write_workers, "endio-write",
- fs_info->thread_pool_size);
+ fs_info->thread_pool_size,
+ &fs_info->generic_worker);
/*
* endios are largely parallel and should have a very
@@ -1794,12 +1805,8 @@ printk("thread pool is %d\n", fs_info->thread_pool_size);
fs_info->endio_write_workers.idle_thresh = 2;
fs_info->endio_meta_write_workers.idle_thresh = 2;
- fs_info->endio_workers.atomic_worker_start = 1;
- fs_info->endio_meta_workers.atomic_worker_start = 1;
- fs_info->endio_write_workers.atomic_worker_start = 1;
- fs_info->endio_meta_write_workers.atomic_worker_start = 1;
-
btrfs_start_workers(&fs_info->workers, 1);
+ btrfs_start_workers(&fs_info->generic_worker, 1);
btrfs_start_workers(&fs_info->submit_workers, 1);
btrfs_start_workers(&fs_info->delalloc_workers, 1);
btrfs_start_workers(&fs_info->fixup_workers, 1);
@@ -1807,6 +1814,7 @@ printk("thread pool is %d\n", fs_info->thread_pool_size);
btrfs_start_workers(&fs_info->endio_meta_workers, 1);
btrfs_start_workers(&fs_info->endio_meta_write_workers, 1);
btrfs_start_workers(&fs_info->endio_write_workers, 1);
+ btrfs_start_workers(&fs_info->enospc_workers, 1);
fs_info->bdi.ra_pages *= btrfs_super_num_devices(disk_super);
fs_info->bdi.ra_pages = max(fs_info->bdi.ra_pages,
@@ -2012,6 +2020,7 @@ fail_chunk_root:
free_extent_buffer(chunk_root->node);
free_extent_buffer(chunk_root->commit_root);
fail_sb_buffer:
+ btrfs_stop_workers(&fs_info->generic_worker);
btrfs_stop_workers(&fs_info->fixup_workers);
btrfs_stop_workers(&fs_info->delalloc_workers);
btrfs_stop_workers(&fs_info->workers);
@@ -2020,6 +2029,7 @@ fail_sb_buffer:
btrfs_stop_workers(&fs_info->endio_meta_write_workers);
btrfs_stop_workers(&fs_info->endio_write_workers);
btrfs_stop_workers(&fs_info->submit_workers);
+ btrfs_stop_workers(&fs_info->enospc_workers);
fail_iput:
invalidate_inode_pages2(fs_info->btree_inode->i_mapping);
iput(fs_info->btree_inode);
@@ -2437,6 +2447,7 @@ int close_ctree(struct btrfs_root *root)
iput(fs_info->btree_inode);
+ btrfs_stop_workers(&fs_info->generic_worker);
btrfs_stop_workers(&fs_info->fixup_workers);
btrfs_stop_workers(&fs_info->delalloc_workers);
btrfs_stop_workers(&fs_info->workers);
@@ -2445,6 +2456,7 @@ int close_ctree(struct btrfs_root *root)
btrfs_stop_workers(&fs_info->endio_meta_write_workers);
btrfs_stop_workers(&fs_info->endio_write_workers);
btrfs_stop_workers(&fs_info->submit_workers);
+ btrfs_stop_workers(&fs_info->enospc_workers);
btrfs_close_devices(fs_info->fs_devices);
btrfs_mapping_tree_free(&fs_info->mapping_tree);
diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 359a754c782..94627c4cc19 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -1568,23 +1568,23 @@ static int remove_extent_backref(struct btrfs_trans_handle *trans,
return ret;
}
-#ifdef BIO_RW_DISCARD
static void btrfs_issue_discard(struct block_device *bdev,
u64 start, u64 len)
{
blkdev_issue_discard(bdev, start >> 9, len >> 9, GFP_KERNEL,
DISCARD_FL_BARRIER);
}
-#endif
static int btrfs_discard_extent(struct btrfs_root *root, u64 bytenr,
u64 num_bytes)
{
-#ifdef BIO_RW_DISCARD
int ret;
u64 map_length = num_bytes;
struct btrfs_multi_bio *multi = NULL;
+ if (!btrfs_test_opt(root, DISCARD))
+ return 0;
+
/* Tell the block device(s) that the sectors can be discarded */
ret = btrfs_map_block(&root->fs_info->mapping_tree, READ,
bytenr, &map_length, &multi, 0);
@@ -1604,9 +1604,6 @@ static int btrfs_discard_extent(struct btrfs_root *root, u64 bytenr,
}
return ret;
-#else
- return 0;
-#endif
}
int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans,
@@ -2824,14 +2821,17 @@ int btrfs_unreserve_metadata_for_delalloc(struct btrfs_root *root,
num_items);
spin_lock(&meta_sinfo->lock);
- if (BTRFS_I(inode)->delalloc_reserved_extents <=
- BTRFS_I(inode)->delalloc_extents) {
+ spin_lock(&BTRFS_I(inode)->accounting_lock);
+ if (BTRFS_I(inode)->reserved_extents <=
+ BTRFS_I(inode)->outstanding_extents) {
+ spin_unlock(&BTRFS_I(inode)->accounting_lock);
spin_unlock(&meta_sinfo->lock);
return 0;
}
+ spin_unlock(&BTRFS_I(inode)->accounting_lock);
- BTRFS_I(inode)->delalloc_reserved_extents--;
- BUG_ON(BTRFS_I(inode)->delalloc_reserved_extents < 0);
+ BTRFS_I(inode)->reserved_extents--;
+ BUG_ON(BTRFS_I(inode)->reserved_extents < 0);
if (meta_sinfo->bytes_delalloc < num_bytes) {
bug = true;
@@ -2864,6 +2864,107 @@ static void check_force_delalloc(struct btrfs_space_info *meta_sinfo)
meta_sinfo->force_delalloc = 0;
}
+struct async_flush {
+ struct btrfs_root *root;
+ struct btrfs_space_info *info;
+ struct btrfs_work work;
+};
+
+static noinline void flush_delalloc_async(struct btrfs_work *work)
+{
+ struct async_flush *async;
+ struct btrfs_root *root;
+ struct btrfs_space_info *info;
+
+ async = container_of(work, struct async_flush, work);
+ root = async->root;
+ info = async->info;
+
+ btrfs_start_delalloc_inodes(root);
+ wake_up(&info->flush_wait);
+ btrfs_wait_ordered_extents(root, 0);
+
+ spin_lock(&info->lock);
+ info->flushing = 0;
+ spin_unlock(&info->lock);
+ wake_up(&info->flush_wait);
+
+ kfree(async);
+}
+
+static void wait_on_flush(struct btrfs_space_info *info)
+{
+ DEFINE_WAIT(wait);
+ u64 used;
+
+ while (1) {
+ prepare_to_wait(&info->flush_wait, &wait,
+ TASK_UNINTERRUPTIBLE);
+ spin_lock(&info->lock);
+ if (!info->flushing) {
+ spin_unlock(&info->lock);
+ break;
+ }
+
+ used = info->bytes_used + info->bytes_reserved +
+ info->bytes_pinned + info->bytes_readonly +
+ info->bytes_super + info->bytes_root +
+ info->bytes_may_use + info->bytes_delalloc;
+ if (used < info->total_bytes) {
+ spin_unlock(&info->lock);
+ break;
+ }
+ spin_unlock(&info->lock);
+ schedule();
+ }
+ finish_wait(&info->flush_wait, &wait);
+}
+
+static void flush_delalloc(struct btrfs_root *root,
+ struct btrfs_space_info *info)
+{
+ struct async_flush *async;
+ bool wait = false;
+
+ spin_lock(&info->lock);
+
+ if (!info->flushing) {
+ info->flushing = 1;
+ init_waitqueue_head(&info->flush_wait);
+ } else {
+ wait = true;
+ }
+
+ spin_unlock(&info->lock);
+
+ if (wait) {
+ wait_on_flush(info);
+ return;
+ }
+
+ async = kzalloc(sizeof(*async), GFP_NOFS);
+ if (!async)
+ goto flush;
+
+ async->root = root;
+ async->info = info;
+ async->work.func = flush_delalloc_async;
+
+ btrfs_queue_worker(&root->fs_info->enospc_workers,
+ &async->work);
+ wait_on_flush(info);
+ return;
+
+flush:
+ btrfs_start_delalloc_inodes(root);
+ btrfs_wait_ordered_extents(root, 0);
+
+ spin_lock(&info->lock);
+ info->flushing = 0;
+ spin_unlock(&info->lock);
+ wake_up(&info->flush_wait);
+}
+
static int maybe_allocate_chunk(struct btrfs_root *root,
struct btrfs_space_info *info)
{
@@ -2876,10 +2977,10 @@ static int maybe_allocate_chunk(struct btrfs_root *root,
free_space = btrfs_super_total_bytes(disk_super);
/*
- * we allow the metadata to grow to a max of either 5gb or 5% of the
+ * we allow the metadata to grow to a max of either 10gb or 5% of the
* space in the volume.
*/
- min_metadata = min((u64)5 * 1024 * 1024 * 1024,
+ min_metadata = min((u64)10 * 1024 * 1024 * 1024,
div64_u64(free_space * 5, 100));
if (info->total_bytes >= min_metadata) {
spin_unlock(&info->lock);
@@ -2894,7 +2995,7 @@ static int maybe_allocate_chunk(struct btrfs_root *root,
if (!info->allocating_chunk) {
info->force_alloc = 1;
info->allocating_chunk = 1;
- init_waitqueue_head(&info->wait);
+ init_waitqueue_head(&info->allocate_wait);
} else {
wait = true;
}
@@ -2902,7 +3003,7 @@ static int maybe_allocate_chunk(struct btrfs_root *root,
spin_unlock(&info->lock);
if (wait) {
- wait_event(info->wait,
+ wait_event(info->allocate_wait,
!info->allocating_chunk);
return 1;
}
@@ -2923,7 +3024,7 @@ out:
spin_lock(&info->lock);
info->allocating_chunk = 0;
spin_unlock(&info->lock);
- wake_up(&info->wait);
+ wake_up(&info->allocate_wait);
if (ret)
return 0;
@@ -2981,21 +3082,20 @@ again:
filemap_flush(inode->i_mapping);
goto again;
} else if (flushed == 3) {
- btrfs_start_delalloc_inodes(root);
- btrfs_wait_ordered_extents(root, 0);
+ flush_delalloc(root, meta_sinfo);
goto again;
}
spin_lock(&meta_sinfo->lock);
meta_sinfo->bytes_delalloc -= num_bytes;
spin_unlock(&meta_sinfo->lock);
printk(KERN_ERR "enospc, has %d, reserved %d\n",
- BTRFS_I(inode)->delalloc_extents,
- BTRFS_I(inode)->delalloc_reserved_extents);
+ BTRFS_I(inode)->outstanding_extents,
+ BTRFS_I(inode)->reserved_extents);
dump_space_info(meta_sinfo, 0, 0);
return -ENOSPC;
}
- BTRFS_I(inode)->delalloc_reserved_extents++;
+ BTRFS_I(inode)->reserved_extents++;
check_force_delalloc(meta_sinfo);
spin_unlock(&meta_sinfo->lock);
@@ -3094,8 +3194,7 @@ again:
}
if (retries == 2) {
- btrfs_start_delalloc_inodes(root);
- btrfs_wait_ordered_extents(root, 0);
+ flush_delalloc(root, meta_sinfo);
goto again;
}
spin_lock(&meta_sinfo->lock);
@@ -3588,6 +3687,14 @@ static int pin_down_bytes(struct btrfs_trans_handle *trans,
if (is_data)
goto pinit;
+ /*
+ * discard is sloooow, and so triggering discards on
+ * individual btree blocks isn't a good plan. Just
+ * pin everything in discard mode.
+ */
+ if (btrfs_test_opt(root, DISCARD))
+ goto pinit;
+
buf = btrfs_find_tree_block(root, bytenr, num_bytes);
if (!buf)
goto pinit;
@@ -3995,7 +4102,7 @@ wait_block_group_cache_done(struct btrfs_block_group_cache *cache)
}
enum btrfs_loop_type {
- LOOP_CACHED_ONLY = 0,
+ LOOP_FIND_IDEAL = 0,
LOOP_CACHING_NOWAIT = 1,
LOOP_CACHING_WAIT = 2,
LOOP_ALLOC_CHUNK = 3,
@@ -4024,11 +4131,15 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans,
struct btrfs_block_group_cache *block_group = NULL;
int empty_cluster = 2 * 1024 * 1024;
int allowed_chunk_alloc = 0;
+ int done_chunk_alloc = 0;
struct btrfs_space_info *space_info;
int last_ptr_loop = 0;
int loop = 0;
bool found_uncached_bg = false;
bool failed_cluster_refill = false;
+ bool failed_alloc = false;
+ u64 ideal_cache_percent = 0;
+ u64 ideal_cache_offset = 0;
WARN_ON(num_bytes < root->sectorsize);
btrfs_set_key_type(ins, BTRFS_EXTENT_ITEM_KEY);
@@ -4064,14 +4175,19 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans,
empty_cluster = 0;
if (search_start == hint_byte) {
+ideal_cache:
block_group = btrfs_lookup_block_group(root->fs_info,
search_start);
/*
* we don't want to use the block group if it doesn't match our
* allocation bits, or if its not cached.
+ *
+ * However if we are re-searching with an ideal block group
+ * picked out then we don't care that the block group is cached.
*/
if (block_group && block_group_bits(block_group, data) &&
- block_group_cache_done(block_group)) {
+ (block_group->cached != BTRFS_CACHE_NO ||
+ search_start == ideal_cache_offset)) {
down_read(&space_info->groups_sem);
if (list_empty(&block_group->list) ||
block_group->ro) {
@@ -4083,13 +4199,13 @@ static noinline int find_free_extent(struct btrfs_trans_handle *trans,
*/
btrfs_put_block_group(block_group);
up_read(&space_info->groups_sem);
- } else
+ } else {
goto have_block_group;
+ }
} else if (block_group) {
btrfs_put_block_group(block_group);
}
}
-
search:
down_read(&space_info->groups_sem);
list_for_each_entry(block_group, &space_info->block_groups, list) {
@@ -4101,28 +4217,45 @@ search:
have_block_group:
if (unlikely(block_group->cached == BTRFS_CACHE_NO)) {
+ u64 free_percent;
+
+ free_percent = btrfs_block_group_used(&block_group->item);
+ free_percent *= 100;
+ free_percent = div64_u64(free_percent,
+ block_group->key.offset);
+ free_percent = 100 - free_percent;
+ if (free_percent > ideal_cache_percent &&
+ likely(!block_group->ro)) {
+ ideal_cache_offset = block_group->key.objectid;
+ ideal_cache_percent = free_percent;
+ }
+
/*
- * we want to start caching kthreads, but not too many
- * right off the bat so we don't overwhelm the system,
- * so only start them if there are less than 2 and we're
- * in the initial allocation phase.
+ * We only want to start kthread caching if we are at
+ * the point where we will wait for caching to make
+ * progress, or if our ideal search is over and we've
+ * found somebody to start caching.
*/
if (loop > LOOP_CACHING_NOWAIT ||
- atomic_read(&space_info->caching_threads) < 2) {
+ (loop > LOOP_FIND_IDEAL &&
+ atomic_read(&space_info->caching_threads) < 2)) {
ret = cache_block_group(block_group);
BUG_ON(ret);
}
- }
-
- cached = block_group_cache_done(block_group);
- if (unlikely(!cached)) {
found_uncached_bg = true;
- /* if we only want cached bgs, loop */
- if (loop == LOOP_CACHED_ONLY)
+ /*
+ * If loop is set for cached only, try the next block
+ * group.
+ */
+ if (loop == LOOP_FIND_IDEAL)
goto loop;
}
+ cached = block_group_cache_done(block_group);
+ if (unlikely(!cached))
+ found_uncached_bg = true;
+
if (unlikely(block_group->ro))
goto loop;
@@ -4233,14 +4366,23 @@ refill_cluster:
offset = btrfs_find_space_for_alloc(block_group, search_start,
num_bytes, empty_size);
- if (!offset && (cached || (!cached &&
- loop == LOOP_CACHING_NOWAIT))) {
- goto loop;
- } else if (!offset && (!cached &&
- loop > LOOP_CACHING_NOWAIT)) {
+ /*
+ * If we didn't find a chunk, and we haven't failed on this
+ * block group before, and this block group is in the middle of
+ * caching and we are ok with waiting, then go ahead and wait
+ * for progress to be made, and set failed_alloc to true.
+ *
+ * If failed_alloc is true then we've already waited on this
+ * block group once and should move on to the next block group.
+ */
+ if (!offset && !failed_alloc && !cached &&
+ loop > LOOP_CACHING_NOWAIT) {
wait_block_group_cache_progress(block_group,
- num_bytes + empty_size);
+ num_bytes + empty_size);
+ failed_alloc = true;
goto have_block_group;
+ } else if (!offset) {
+ goto loop;
}
checks:
search_start = stripe_align(root, offset);
@@ -4288,13 +4430,16 @@ checks:
break;
loop:
failed_cluster_refill = false;
+ failed_alloc = false;
btrfs_put_block_group(block_group);
}
up_read(&space_info->groups_sem);
- /* LOOP_CACHED_ONLY, only search fully cached block groups
- * LOOP_CACHING_NOWAIT, search partially cached block groups, but
- * dont wait foR them to finish caching
+ /* LOOP_FIND_IDEAL, only search caching/cached bg's, and don't wait for
+ * for them to make caching progress. Also
+ * determine the best possible bg to cache
+ * LOOP_CACHING_NOWAIT, search partially cached block groups, kicking
+ * caching kthreads as we move along
* LOOP_CACHING_WAIT, search everything, and wait if our bg is caching
* LOOP_ALLOC_CHUNK, force a chunk allocation and try again
* LOOP_NO_EMPTY_SIZE, set empty_size and empty_cluster to 0 and try
@@ -4303,12 +4448,47 @@ loop:
if (!ins->objectid && loop < LOOP_NO_EMPTY_SIZE &&
(found_uncached_bg || empty_size || empty_cluster ||
allowed_chunk_alloc)) {
- if (found_uncached_bg) {
+ if (loop == LOOP_FIND_IDEAL && found_uncached_bg) {
found_uncached_bg = false;
- if (loop < LOOP_CACHING_WAIT) {
- loop++;
+ loop++;
+ if (!ideal_cache_percent &&
+ atomic_read(&space_info->caching_threads))
goto search;
- }
+
+ /*
+ * 1 of the following 2 things have happened so far
+ *
+ * 1) We found an ideal block group for caching that
+ * is mostly full and will cache quickly, so we might
+ * as well wait for it.
+ *
+ * 2) We searched for cached only and we didn't find
+ * anything, and we didn't start any caching kthreads
+ * either, so chances are we will loop through and
+ * start a couple caching kthreads, and then come back
+ * around and just wait for them. This will be slower
+ * because we will have 2 caching kthreads reading at
+ * the same time when we could have just started one
+ * and waited for it to get far enough to give us an
+ * allocation, so go ahead and go to the wait caching
+ * loop.
+ */
+ loop = LOOP_CACHING_WAIT;
+ search_start = ideal_cache_offset;
+ ideal_cache_percent = 0;
+ goto ideal_cache;
+ } else if (loop == LOOP_FIND_IDEAL) {
+ /*
+ * Didn't find a uncached bg, wait on anything we find
+ * next.
+ */
+ loop = LOOP_CACHING_WAIT;
+ goto search;
+ }
+
+ if (loop < LOOP_CACHING_WAIT) {
+ loop++;
+ goto search;
}
if (loop == LOOP_ALLOC_CHUNK) {
@@ -4320,7 +4500,8 @@ loop:
ret = do_chunk_alloc(trans, root, num_bytes +
2 * 1024 * 1024, data, 1);
allowed_chunk_alloc = 0;
- } else {
+ done_chunk_alloc = 1;
+ } else if (!done_chunk_alloc) {
space_info->force_alloc = 1;
}
@@ -4799,6 +4980,7 @@ static noinline void reada_walk_down(struct btrfs_trans_handle *trans,
u64 bytenr;
u64 generation;
u64 refs;
+ u64 flags;
u64 last = 0;
u32 nritems;
u32 blocksize;
@@ -4836,15 +5018,19 @@ static noinline void reada_walk_down(struct btrfs_trans_handle *trans,
generation <= root->root_key.offset)
continue;
+ /* We don't lock the tree block, it's OK to be racy here */
+ ret = btrfs_lookup_extent_info(trans, root, bytenr, blocksize,
+ &refs, &flags);
+ BUG_ON(ret);
+ BUG_ON(refs == 0);
+
if (wc->stage == DROP_REFERENCE) {
- ret = btrfs_lookup_extent_info(trans, root,
- bytenr, blocksize,
- &refs, NULL);
- BUG_ON(ret);
- BUG_ON(refs == 0);
if (refs == 1)
goto reada;
+ if (wc->level == 1 &&
+ (flags & BTRFS_BLOCK_FLAG_FULL_BACKREF))
+ continue;
if (!wc->update_ref ||
generation <= root->root_key.offset)
continue;
@@ -4853,6 +5039,10 @@ static noinline void reada_walk_down(struct btrfs_trans_handle *trans,
&wc->update_progress);
if (ret < 0)
continue;
+ } else {
+ if (wc->level == 1 &&
+ (flags & BTRFS_BLOCK_FLAG_FULL_BACKREF))
+ continue;
}
reada:
ret = readahead_tree_block(root, bytenr, blocksize,
@@ -4876,7 +5066,7 @@ reada:
static noinline int walk_down_proc(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
- struct walk_control *wc)
+ struct walk_control *wc, int lookup_info)
{
int level = wc->level;
struct extent_buffer *eb = path->nodes[level];
@@ -4891,8 +5081,9 @@ static noinline int walk_down_proc(struct btrfs_trans_handle *trans,
* when reference count of tree block is 1, it won't increase
* again. once full backref flag is set, we never clear it.
*/
- if ((wc->stage == DROP_REFERENCE && wc->refs[level] != 1) ||
- (wc->stage == UPDATE_BACKREF && !(wc->flags[level] & flag))) {
+ if (lookup_info &&
+ ((wc->stage == DROP_REFERENCE && wc->refs[level] != 1) ||
+ (wc->stage == UPDATE_BACKREF && !(wc->flags[level] & flag)))) {
BUG_ON(!path->locks[level]);
ret = btrfs_lookup_extent_info(trans, root,
eb->start, eb->len,
@@ -4953,7 +5144,7 @@ static noinline int walk_down_proc(struct btrfs_trans_handle *trans,
static noinline int do_walk_down(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
- struct walk_control *wc)
+ struct walk_control *wc, int *lookup_info)
{
u64 bytenr;
u64 generation;
@@ -4973,8 +5164,10 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
* for the subtree
*/
if (wc->stage == UPDATE_BACKREF &&
- generation <= root->root_key.offset)
+ generation <= root->root_key.offset) {
+ *lookup_info = 1;
return 1;
+ }
bytenr = btrfs_node_blockptr(path->nodes[level], path->slots[level]);
blocksize = btrfs_level_size(root, level - 1);
@@ -4987,14 +5180,19 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
btrfs_tree_lock(next);
btrfs_set_lock_blocking(next);
- if (wc->stage == DROP_REFERENCE) {
- ret = btrfs_lookup_extent_info(trans, root, bytenr, blocksize,
- &wc->refs[level - 1],
- &wc->flags[level - 1]);
- BUG_ON(ret);
- BUG_ON(wc->refs[level - 1] == 0);
+ ret = btrfs_lookup_extent_info(trans, root, bytenr, blocksize,
+ &wc->refs[level - 1],
+ &wc->flags[level - 1]);
+ BUG_ON(ret);
+ BUG_ON(wc->refs[level - 1] == 0);
+ *lookup_info = 0;
+ if (wc->stage == DROP_REFERENCE) {
if (wc->refs[level - 1] > 1) {
+ if (level == 1 &&
+ (wc->flags[0] & BTRFS_BLOCK_FLAG_FULL_BACKREF))
+ goto skip;
+
if (!wc->update_ref ||
generation <= root->root_key.offset)
goto skip;
@@ -5008,12 +5206,17 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
wc->stage = UPDATE_BACKREF;
wc->shared_level = level - 1;
}
+ } else {
+ if (level == 1 &&
+ (wc->flags[0] & BTRFS_BLOCK_FLAG_FULL_BACKREF))
+ goto skip;
}
if (!btrfs_buffer_uptodate(next, generation)) {
btrfs_tree_unlock(next);
free_extent_buffer(next);
next = NULL;
+ *lookup_info = 1;
}
if (!next) {
@@ -5036,21 +5239,22 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans,
skip:
wc->refs[level - 1] = 0;
wc->flags[level - 1] = 0;
+ if (wc->stage == DROP_REFERENCE) {
+ if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF) {
+ parent = path->nodes[level]->start;
+ } else {
+ BUG_ON(root->root_key.objectid !=
+ btrfs_header_owner(path->nodes[level]));
+ parent = 0;
+ }
- if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF) {
- parent = path->nodes[level]->start;
- } else {
- BUG_ON(root->root_key.objectid !=
- btrfs_header_owner(path->nodes[level]));
- parent = 0;
+ ret = btrfs_free_extent(trans, root, bytenr, blocksize, parent,
+ root->root_key.objectid, level - 1, 0);
+ BUG_ON(ret);
}
-
- ret = btrfs_free_extent(trans, root, bytenr, blocksize, parent,
- root->root_key.objectid, level - 1, 0);
- BUG_ON(ret);
-
btrfs_tree_unlock(next);
free_extent_buffer(next);
+ *lookup_info = 1;
return 1;
}
@@ -5164,6 +5368,7 @@ static noinline int walk_down_tree(struct btrfs_trans_handle *trans,
struct walk_control *wc)
{
int level = wc->level;
+ int lookup_info = 1;
int ret;
while (level >= 0) {
@@ -5171,14 +5376,14 @@ static noinline int walk_down_tree(struct btrfs_trans_handle *trans,
btrfs_header_nritems(path->nodes[level]))
break;
- ret = walk_down_proc(trans, root, path, wc);
+ ret = walk_down_proc(trans, root, path, wc, lookup_info);
if (ret > 0)
break;
if (level == 0)
break;
- ret = do_walk_down(trans, root, path, wc);
+ ret = do_walk_down(trans, root, path, wc, &lookup_info);
if (ret > 0) {
path->slots[level]++;
continue;
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index de1793ba004..96577e8bf9f 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -460,7 +460,8 @@ static int clear_state_bit(struct extent_io_tree *tree,
struct extent_state *state, int bits, int wake,
int delete)
{
- int ret = state->state & bits;
+ int bits_to_clear = bits & ~EXTENT_DO_ACCOUNTING;
+ int ret = state->state & bits_to_clear;
if ((bits & EXTENT_DIRTY) && (state->state & EXTENT_DIRTY)) {
u64 range = state->end - state->start + 1;
@@ -468,7 +469,7 @@ static int clear_state_bit(struct extent_io_tree *tree,
tree->dirty_bytes -= range;
}
clear_state_cb(tree, state, bits);
- state->state &= ~bits;
+ state->state &= ~bits_to_clear;
if (wake)
wake_up(&state->wq);
if (delete || state->state == 0) {
@@ -956,7 +957,8 @@ int clear_extent_dirty(struct extent_io_tree *tree, u64 start, u64 end,
gfp_t mask)
{
return clear_extent_bit(tree, start, end,
- EXTENT_DIRTY | EXTENT_DELALLOC, 0, 0,
+ EXTENT_DIRTY | EXTENT_DELALLOC |
+ EXTENT_DO_ACCOUNTING, 0, 0,
NULL, mask);
}
@@ -1401,12 +1403,7 @@ out_failed:
int extent_clear_unlock_delalloc(struct inode *inode,
struct extent_io_tree *tree,
u64 start, u64 end, struct page *locked_page,
- int unlock_pages,
- int clear_unlock,
- int clear_delalloc, int clear_dirty,
- int set_writeback,
- int end_writeback,
- int set_private2)
+ unsigned long op)
{
int ret;
struct page *pages[16];
@@ -1416,17 +1413,21 @@ int extent_clear_unlock_delalloc(struct inode *inode,
int i;
int clear_bits = 0;
- if (clear_unlock)
+ if (op & EXTENT_CLEAR_UNLOCK)
clear_bits |= EXTENT_LOCKED;
- if (clear_dirty)
+ if (op & EXTENT_CLEAR_DIRTY)
clear_bits |= EXTENT_DIRTY;
- if (clear_delalloc)
+ if (op & EXTENT_CLEAR_DELALLOC)
clear_bits |= EXTENT_DELALLOC;
+ if (op & EXTENT_CLEAR_ACCOUNTING)
+ clear_bits |= EXTENT_DO_ACCOUNTING;
+
clear_extent_bit(tree, start, end, clear_bits, 1, 0, NULL, GFP_NOFS);
- if (!(unlock_pages || clear_dirty || set_writeback || end_writeback ||
- set_private2))
+ if (!(op & (EXTENT_CLEAR_UNLOCK_PAGE | EXTENT_CLEAR_DIRTY |
+ EXTENT_SET_WRITEBACK | EXTENT_END_WRITEBACK |
+ EXTENT_SET_PRIVATE2)))
return 0;
while (nr_pages > 0) {
@@ -1435,20 +1436,20 @@ int extent_clear_unlock_delalloc(struct inode *inode,
nr_pages, ARRAY_SIZE(pages)), pages);
for (i = 0; i < ret; i++) {
- if (set_private2)
+ if (op & EXTENT_SET_PRIVATE2)
SetPagePrivate2(pages[i]);
if (pages[i] == locked_page) {
page_cache_release(pages[i]);
continue;
}
- if (clear_dirty)
+ if (op & EXTENT_CLEAR_DIRTY)
clear_page_dirty_for_io(pages[i]);
- if (set_writeback)
+ if (op & EXTENT_SET_WRITEBACK)
set_page_writeback(pages[i]);
- if (end_writeback)
+ if (op & EXTENT_END_WRITEBACK)
end_page_writeback(pages[i]);
- if (unlock_pages)
+ if (op & EXTENT_CLEAR_UNLOCK_PAGE)
unlock_page(pages[i]);
page_cache_release(pages[i]);
}
@@ -2714,7 +2715,8 @@ int extent_invalidatepage(struct extent_io_tree *tree,
lock_extent(tree, start, end, GFP_NOFS);
wait_on_page_writeback(page);
clear_extent_bit(tree, start, end,
- EXTENT_LOCKED | EXTENT_DIRTY | EXTENT_DELALLOC,
+ EXTENT_LOCKED | EXTENT_DIRTY | EXTENT_DELALLOC |
+ EXTENT_DO_ACCOUNTING,
1, 1, NULL, GFP_NOFS);
return 0;
}
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 4794ec891fe..36de250a7b2 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -15,6 +15,7 @@
#define EXTENT_BUFFER_FILLED (1 << 8)
#define EXTENT_BOUNDARY (1 << 9)
#define EXTENT_NODATASUM (1 << 10)
+#define EXTENT_DO_ACCOUNTING (1 << 11)
#define EXTENT_IOBITS (EXTENT_LOCKED | EXTENT_WRITEBACK)
/* flags for bio submission */
@@ -25,6 +26,16 @@
#define EXTENT_BUFFER_BLOCKING 1
#define EXTENT_BUFFER_DIRTY 2
+/* these are flags for extent_clear_unlock_delalloc */
+#define EXTENT_CLEAR_UNLOCK_PAGE 0x1
+#define EXTENT_CLEAR_UNLOCK 0x2
+#define EXTENT_CLEAR_DELALLOC 0x4
+#define EXTENT_CLEAR_DIRTY 0x8
+#define EXTENT_SET_WRITEBACK 0x10
+#define EXTENT_END_WRITEBACK 0x20
+#define EXTENT_SET_PRIVATE2 0x40
+#define EXTENT_CLEAR_ACCOUNTING 0x80
+
/*
* page->private values. Every page that is controlled by the extent
* map has page->private set to one.
@@ -288,10 +299,5 @@ int extent_range_uptodate(struct extent_io_tree *tree,
int extent_clear_unlock_delalloc(struct inode *inode,
struct extent_io_tree *tree,
u64 start, u64 end, struct page *locked_page,
- int unlock_page,
- int clear_unlock,
- int clear_delalloc, int clear_dirty,
- int set_writeback,
- int end_writeback,
- int set_private2);
+ unsigned long op);
#endif
diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index 2c726b7b9fa..ccbdcb54ec5 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -208,7 +208,7 @@ int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len)
write_lock(&tree->lock);
em = lookup_extent_mapping(tree, start, len);
- WARN_ON(em->start != start || !em);
+ WARN_ON(!em || em->start != start);
if (!em)
goto out;
diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index f19e1259a97..06550affbd2 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -878,7 +878,8 @@ again:
btrfs_put_ordered_extent(ordered);
clear_extent_bits(&BTRFS_I(inode)->io_tree, start_pos,
- last_pos - 1, EXTENT_DIRTY | EXTENT_DELALLOC,
+ last_pos - 1, EXTENT_DIRTY | EXTENT_DELALLOC |
+ EXTENT_DO_ACCOUNTING,
GFP_NOFS);
unlock_extent(&BTRFS_I(inode)->io_tree,
start_pos, last_pos - 1, GFP_NOFS);
@@ -1085,8 +1086,10 @@ out_nolock:
btrfs_end_transaction(trans, root);
else
btrfs_commit_transaction(trans, root);
- } else {
+ } else if (ret != BTRFS_NO_LOG_SYNC) {
btrfs_commit_transaction(trans, root);
+ } else {
+ btrfs_end_transaction(trans, root);
}
}
if (file->f_flags & O_DIRECT) {
@@ -1136,6 +1139,13 @@ int btrfs_sync_file(struct file *file, struct dentry *dentry, int datasync)
int ret = 0;
struct btrfs_trans_handle *trans;
+
+ /* we wait first, since the writeback may change the inode */
+ root->log_batch++;
+ /* the VFS called filemap_fdatawrite for us */
+ btrfs_wait_ordered_range(inode, 0, (u64)-1);
+ root->log_batch++;
+
/*
* check the transaction that last modified this inode
* and see if its already been committed
@@ -1143,6 +1153,11 @@ int btrfs_sync_file(struct file *file, struct dentry *dentry, int datasync)
if (!BTRFS_I(inode)->last_trans)
goto out;
+ /*
+ * if the last transaction that changed this file was before
+ * the current transaction, we can bail out now without any
+ * syncing
+ */
mutex_lock(&root->fs_info->trans_mutex);
if (BTRFS_I(inode)->last_trans <=
root->fs_info->last_trans_committed) {
@@ -1152,13 +1167,6 @@ int btrfs_sync_file(struct file *file, struct dentry *dentry, int datasync)
}
mutex_unlock(&root->fs_info->trans_mutex);
- root->log_batch++;
- filemap_fdatawrite(inode->i_mapping);
- btrfs_wait_ordered_range(inode, 0, (u64)-1);
- root->log_batch++;
-
- if (datasync && !(inode->i_state & I_DIRTY_PAGES))
- goto out;
/*
* ok we haven't committed the transaction yet, lets do a commit
*/
@@ -1187,14 +1195,18 @@ int btrfs_sync_file(struct file *file, struct dentry *dentry, int datasync)
*/
mutex_unlock(&dentry->d_inode->i_mutex);
- if (ret > 0) {
- ret = btrfs_commit_transaction(trans, root);
- } else {
- ret = btrfs_sync_log(trans, root);
- if (ret == 0)
- ret = btrfs_end_transaction(trans, root);
- else
+ if (ret != BTRFS_NO_LOG_SYNC) {
+ if (ret > 0) {
ret = btrfs_commit_transaction(trans, root);
+ } else {
+ ret = btrfs_sync_log(trans, root);
+ if (ret == 0)
+ ret = btrfs_end_transaction(trans, root);
+ else
+ ret = btrfs_commit_transaction(trans, root);
+ }
+ } else {
+ ret = btrfs_end_transaction(trans, root);
}
mutex_lock(&dentry->d_inode->i_mutex);
out:
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 5c2caad7621..cb2849f0325 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -1296,7 +1296,7 @@ again:
window_start = entry->offset;
window_free = entry->bytes;
last = entry;
- max_extent = 0;
+ max_extent = entry->bytes;
} else {
last = next;
window_free += next->bytes;
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 112e5aa8589..b3ad168a0bf 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -424,9 +424,12 @@ again:
* and free up our temp pages.
*/
extent_clear_unlock_delalloc(inode,
- &BTRFS_I(inode)->io_tree,
- start, end, NULL, 1, 0,
- 0, 1, 1, 1, 0);
+ &BTRFS_I(inode)->io_tree,
+ start, end, NULL,
+ EXTENT_CLEAR_UNLOCK_PAGE | EXTENT_CLEAR_DIRTY |
+ EXTENT_CLEAR_DELALLOC |
+ EXTENT_CLEAR_ACCOUNTING |
+ EXTENT_SET_WRITEBACK | EXTENT_END_WRITEBACK);
ret = 0;
goto free_pages_out;
}
@@ -535,7 +538,7 @@ static noinline int submit_compressed_extents(struct inode *inode,
struct btrfs_root *root = BTRFS_I(inode)->root;
struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
struct extent_io_tree *io_tree;
- int ret;
+ int ret = 0;
if (list_empty(&async_cow->extents))
return 0;
@@ -549,6 +552,7 @@ static noinline int submit_compressed_extents(struct inode *inode,
io_tree = &BTRFS_I(inode)->io_tree;
+retry:
/* did the compression code fall back to uncompressed IO? */
if (!async_extent->pages) {
int page_started = 0;
@@ -559,11 +563,11 @@ static noinline int submit_compressed_extents(struct inode *inode,
async_extent->ram_size - 1, GFP_NOFS);
/* allocate blocks */
- cow_file_range(inode, async_cow->locked_page,
- async_extent->start,
- async_extent->start +
- async_extent->ram_size - 1,
- &page_started, &nr_written, 0);
+ ret = cow_file_range(inode, async_cow->locked_page,
+ async_extent->start,
+ async_extent->start +
+ async_extent->ram_size - 1,
+ &page_started, &nr_written, 0);
/*
* if page_started, cow_file_range inserted an
@@ -571,7 +575,7 @@ static noinline int submit_compressed_extents(struct inode *inode,
* and IO for us. Otherwise, we need to submit
* all those pages down to the drive.
*/
- if (!page_started)
+ if (!page_started && !ret)
extent_write_locked_range(io_tree,
inode, async_extent->start,
async_extent->start +
@@ -599,7 +603,21 @@ static noinline int submit_compressed_extents(struct inode *inode,
async_extent->compressed_size,
0, alloc_hint,
(u64)-1, &ins, 1);
- BUG_ON(ret);
+ if (ret) {
+ int i;
+ for (i = 0; i < async_extent->nr_pages; i++) {
+ WARN_ON(async_extent->pages[i]->mapping);
+ page_cache_release(async_extent->pages[i]);
+ }
+ kfree(async_extent->pages);
+ async_extent->nr_pages = 0;
+ async_extent->pages = NULL;
+ unlock_extent(io_tree, async_extent->start,
+ async_extent->start +
+ async_extent->ram_size - 1, GFP_NOFS);
+ goto retry;
+ }
+
em = alloc_extent_map(GFP_NOFS);
em->start = async_extent->start;
em->len = async_extent->ram_size;
@@ -637,11 +655,14 @@ static noinline int submit_compressed_extents(struct inode *inode,
* clear dirty, set writeback and unlock the pages.
*/
extent_clear_unlock_delalloc(inode,
- &BTRFS_I(inode)->io_tree,
- async_extent->start,
- async_extent->start +
- async_extent->ram_size - 1,
- NULL, 1, 1, 0, 1, 1, 0, 0);
+ &BTRFS_I(inode)->io_tree,
+ async_extent->start,
+ async_extent->start +
+ async_extent->ram_size - 1,
+ NULL, EXTENT_CLEAR_UNLOCK_PAGE |
+ EXTENT_CLEAR_UNLOCK |
+ EXTENT_CLEAR_DELALLOC |
+ EXTENT_CLEAR_DIRTY | EXTENT_SET_WRITEBACK);
ret = btrfs_submit_compressed_write(inode,
async_extent->start,
@@ -712,9 +733,15 @@ static noinline int cow_file_range(struct inode *inode,
start, end, 0, NULL);
if (ret == 0) {
extent_clear_unlock_delalloc(inode,
- &BTRFS_I(inode)->io_tree,
- start, end, NULL, 1, 1,
- 1, 1, 1, 1, 0);
+ &BTRFS_I(inode)->io_tree,
+ start, end, NULL,
+ EXTENT_CLEAR_UNLOCK_PAGE |
+ EXTENT_CLEAR_UNLOCK |
+ EXTENT_CLEAR_DELALLOC |
+ EXTENT_CLEAR_ACCOUNTING |
+ EXTENT_CLEAR_DIRTY |
+ EXTENT_SET_WRITEBACK |
+ EXTENT_END_WRITEBACK);
*nr_written = *nr_written +
(end - start + PAGE_CACHE_SIZE) / PAGE_CACHE_SIZE;
*page_started = 1;
@@ -731,13 +758,29 @@ static noinline int cow_file_range(struct inode *inode,
em = search_extent_mapping(&BTRFS_I(inode)->extent_tree,
start, num_bytes);
if (em) {
- alloc_hint = em->block_start;
- free_extent_map(em);
+ /*
+ * if block start isn't an actual block number then find the
+ * first block in this inode and use that as a hint. If that
+ * block is also bogus then just don't worry about it.
+ */
+ if (em->block_start >= EXTENT_MAP_LAST_BYTE) {
+ free_extent_map(em);
+ em = search_extent_mapping(em_tree, 0, 0);
+ if (em && em->block_start < EXTENT_MAP_LAST_BYTE)
+ alloc_hint = em->block_start;
+ if (em)
+ free_extent_map(em);
+ } else {
+ alloc_hint = em->block_start;
+ free_extent_map(em);
+ }
}
read_unlock(&BTRFS_I(inode)->extent_tree.lock);
btrfs_drop_extent_cache(inode, start, start + num_bytes - 1, 0);
while (disk_num_bytes > 0) {
+ unsigned long op;
+
cur_alloc_size = min(disk_num_bytes, root->fs_info->max_extent);
ret = btrfs_reserve_extent(trans, root, cur_alloc_size,
root->sectorsize, 0, alloc_hint,
@@ -789,10 +832,13 @@ static noinline int cow_file_range(struct inode *inode,
* Do set the Private2 bit so we know this page was properly
* setup for writepage
*/
+ op = unlock ? EXTENT_CLEAR_UNLOCK_PAGE : 0;
+ op |= EXTENT_CLEAR_UNLOCK | EXTENT_CLEAR_DELALLOC |
+ EXTENT_SET_PRIVATE2;
+
extent_clear_unlock_delalloc(inode, &BTRFS_I(inode)->io_tree,
start, start + ram_size - 1,
- locked_page, unlock, 1,
- 1, 0, 0, 0, 1);
+ locked_page, op);
disk_num_bytes -= cur_alloc_size;
num_bytes -= cur_alloc_size;
alloc_hint = ins.objectid + ins.offset;
@@ -864,8 +910,8 @@ static int cow_file_range_async(struct inode *inode, struct page *locked_page,
u64 cur_end;
int limit = 10 * 1024 * 1042;
- clear_extent_bit(&BTRFS_I(inode)->io_tree, start, end, EXTENT_LOCKED |
- EXTENT_DELALLOC, 1, 0, NULL, GFP_NOFS);
+ clear_extent_bit(&BTRFS_I(inode)->io_tree, start, end, EXTENT_LOCKED,
+ 1, 0, NULL, GFP_NOFS);
while (start < end) {
async_cow = kmalloc(sizeof(*async_cow), GFP_NOFS);
async_cow->inode = inode;
@@ -1006,6 +1052,7 @@ next_slot:
if (found_key.offset > cur_offset) {
extent_end = found_key.offset;
+ extent_type = 0;
goto out_check;
}
@@ -1112,8 +1159,10 @@ out_check:
BUG_ON(ret);
extent_clear_unlock_delalloc(inode, &BTRFS_I(inode)->io_tree,
- cur_offset, cur_offset + num_bytes - 1,
- locked_page, 1, 1, 1, 0, 0, 0, 1);
+ cur_offset, cur_offset + num_bytes - 1,
+ locked_page, EXTENT_CLEAR_UNLOCK_PAGE |
+ EXTENT_CLEAR_UNLOCK | EXTENT_CLEAR_DELALLOC |
+ EXTENT_SET_PRIVATE2);
cur_offset = extent_end;
if (cur_offset > end)
break;
@@ -1178,15 +1227,17 @@ static int btrfs_split_extent_hook(struct inode *inode,
root->fs_info->max_extent);
/*
- * if we break a large extent up then leave delalloc_extents be,
- * since we've already accounted for the large extent.
+ * if we break a large extent up then leave oustanding_extents
+ * be, since we've already accounted for the large extent.
*/
if (div64_u64(new_size + root->fs_info->max_extent - 1,
root->fs_info->max_extent) < num_extents)
return 0;
}
- BTRFS_I(inode)->delalloc_extents++;
+ spin_lock(&BTRFS_I(inode)->accounting_lock);
+ BTRFS_I(inode)->outstanding_extents++;
+ spin_unlock(&BTRFS_I(inode)->accounting_lock);
return 0;
}
@@ -1217,7 +1268,9 @@ static int btrfs_merge_extent_hook(struct inode *inode,
/* we're not bigger than the max, unreserve the space and go */
if (new_size <= root->fs_info->max_extent) {
- BTRFS_I(inode)->delalloc_extents--;
+ spin_lock(&BTRFS_I(inode)->accounting_lock);
+ BTRFS_I(inode)->outstanding_extents--;
+ spin_unlock(&BTRFS_I(inode)->accounting_lock);
return 0;
}
@@ -1231,7 +1284,9 @@ static int btrfs_merge_extent_hook(struct inode *inode,
root->fs_info->max_extent) > num_extents)
return 0;
- BTRFS_I(inode)->delalloc_extents--;
+ spin_lock(&BTRFS_I(inode)->accounting_lock);
+ BTRFS_I(inode)->outstanding_extents--;
+ spin_unlock(&BTRFS_I(inode)->accounting_lock);
return 0;
}
@@ -1253,7 +1308,9 @@ static int btrfs_set_bit_hook(struct inode *inode, u64 start, u64 end,
if (!(old & EXTENT_DELALLOC) && (bits & EXTENT_DELALLOC)) {
struct btrfs_root *root = BTRFS_I(inode)->root;
- BTRFS_I(inode)->delalloc_extents++;
+ spin_lock(&BTRFS_I(inode)->accounting_lock);
+ BTRFS_I(inode)->outstanding_extents++;
+ spin_unlock(&BTRFS_I(inode)->accounting_lock);
btrfs_delalloc_reserve_space(root, inode, end - start + 1);
spin_lock(&root->fs_info->delalloc_lock);
BTRFS_I(inode)->delalloc_bytes += end - start + 1;
@@ -1281,8 +1338,12 @@ static int btrfs_clear_bit_hook(struct inode *inode,
if ((state->state & EXTENT_DELALLOC) && (bits & EXTENT_DELALLOC)) {
struct btrfs_root *root = BTRFS_I(inode)->root;
- BTRFS_I(inode)->delalloc_extents--;
- btrfs_unreserve_metadata_for_delalloc(root, inode, 1);
+ if (bits & EXTENT_DO_ACCOUNTING) {
+ spin_lock(&BTRFS_I(inode)->accounting_lock);
+ BTRFS_I(inode)->outstanding_extents--;
+ spin_unlock(&BTRFS_I(inode)->accounting_lock);
+ btrfs_unreserve_metadata_for_delalloc(root, inode, 1);
+ }
spin_lock(&root->fs_info->delalloc_lock);
if (state->end - state->start + 1 >
@@ -2442,7 +2503,19 @@ static int btrfs_unlink(struct inode *dir, struct dentry *dentry)
root = BTRFS_I(dir)->root;
+ /*
+ * 5 items for unlink inode
+ * 1 for orphan
+ */
+ ret = btrfs_reserve_metadata_space(root, 6);
+ if (ret)
+ return ret;
+
trans = btrfs_start_transaction(root, 1);
+ if (IS_ERR(trans)) {
+ btrfs_unreserve_metadata_space(root, 6);
+ return PTR_ERR(trans);
+ }
btrfs_set_trans_block_group(trans, dir);
@@ -2457,6 +2530,7 @@ static int btrfs_unlink(struct inode *dir, struct dentry *dentry)
nr = trans->blocks_used;
btrfs_end_transaction_throttle(trans, root);
+ btrfs_unreserve_metadata_space(root, 6);
btrfs_btree_balance_dirty(root, nr);
return ret;
}
@@ -2537,7 +2611,16 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
inode->i_ino == BTRFS_FIRST_FREE_OBJECTID)
return -ENOTEMPTY;
+ ret = btrfs_reserve_metadata_space(root, 5);
+ if (ret)
+ return ret;
+
trans = btrfs_start_transaction(root, 1);
+ if (IS_ERR(trans)) {
+ btrfs_unreserve_metadata_space(root, 5);
+ return PTR_ERR(trans);
+ }
+
btrfs_set_trans_block_group(trans, dir);
if (unlikely(inode->i_ino == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) {
@@ -2560,6 +2643,7 @@ static int btrfs_rmdir(struct inode *dir, struct dentry *dentry)
out:
nr = trans->blocks_used;
ret = btrfs_end_transaction_throttle(trans, root);
+ btrfs_unreserve_metadata_space(root, 5);
btrfs_btree_balance_dirty(root, nr);
if (ret && !err)
@@ -3000,12 +3084,22 @@ static int btrfs_truncate_page(struct address_space *mapping, loff_t from)
if ((offset & (blocksize - 1)) == 0)
goto out;
+ ret = btrfs_check_data_free_space(root, inode, PAGE_CACHE_SIZE);
+ if (ret)
+ goto out;
+
+ ret = btrfs_reserve_metadata_for_delalloc(root, inode, 1);
+ if (ret)
+ goto out;
ret = -ENOMEM;
again:
page = grab_cache_page(mapping, index);
- if (!page)
+ if (!page) {
+ btrfs_free_reserved_data_space(root, inode, PAGE_CACHE_SIZE);
+ btrfs_unreserve_metadata_for_delalloc(root, inode, 1);
goto out;
+ }
page_start = page_offset(page);
page_end = page_start + PAGE_CACHE_SIZE - 1;
@@ -3038,6 +3132,10 @@ again:
goto again;
}
+ clear_extent_bits(&BTRFS_I(inode)->io_tree, page_start, page_end,
+ EXTENT_DIRTY | EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING,
+ GFP_NOFS);
+
ret = btrfs_set_extent_delalloc(inode, page_start, page_end);
if (ret) {
unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
@@ -3056,6 +3154,9 @@ again:
unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
out_unlock:
+ if (ret)
+ btrfs_free_reserved_data_space(root, inode, PAGE_CACHE_SIZE);
+ btrfs_unreserve_metadata_for_delalloc(root, inode, 1);
unlock_page(page);
page_cache_release(page);
out:
@@ -3079,7 +3180,9 @@ int btrfs_cont_expand(struct inode *inode, loff_t size)
if (size <= hole_start)
return 0;
- btrfs_truncate_page(inode->i_mapping, inode->i_size);
+ err = btrfs_truncate_page(inode->i_mapping, inode->i_size);
+ if (err)
+ return err;
while (1) {
struct btrfs_ordered_extent *ordered;
@@ -3448,6 +3551,7 @@ static noinline void init_btrfs_i(struct inode *inode)
bi->generation = 0;
bi->sequence = 0;
bi->last_trans = 0;
+ bi->last_sub_trans = 0;
bi->logged_trans = 0;
bi->delalloc_bytes = 0;
bi->reserved_bytes = 0;
@@ -3598,12 +3702,14 @@ static int btrfs_dentry_delete(struct dentry *dentry)
{
struct btrfs_root *root;
- if (!dentry->d_inode)
- return 0;
+ if (!dentry->d_inode && !IS_ROOT(dentry))
+ dentry = dentry->d_parent;
- root = BTRFS_I(dentry->d_inode)->root;
- if (btrfs_root_refs(&root->root_item) == 0)
- return 1;
+ if (dentry->d_inode) {
+ root = BTRFS_I(dentry->d_inode)->root;
+ if (btrfs_root_refs(&root->root_item) == 0)
+ return 1;
+ }
return 0;
}
@@ -4808,7 +4914,8 @@ static void btrfs_invalidatepage(struct page *page, unsigned long offset)
*/
clear_extent_bit(tree, page_start, page_end,
EXTENT_DIRTY | EXTENT_DELALLOC |
- EXTENT_LOCKED, 1, 0, NULL, GFP_NOFS);
+ EXTENT_LOCKED | EXTENT_DO_ACCOUNTING, 1, 0,
+ NULL, GFP_NOFS);
/*
* whoever cleared the private bit is responsible
* for the finish_ordered_io
@@ -4821,8 +4928,8 @@ static void btrfs_invalidatepage(struct page *page, unsigned long offset)
lock_extent(tree, page_start, page_end, GFP_NOFS);
}
clear_extent_bit(tree, page_start, page_end,
- EXTENT_LOCKED | EXTENT_DIRTY | EXTENT_DELALLOC,
- 1, 1, NULL, GFP_NOFS);
+ EXTENT_LOCKED | EXTENT_DIRTY | EXTENT_DELALLOC |
+ EXTENT_DO_ACCOUNTING, 1, 1, NULL, GFP_NOFS);
__btrfs_releasepage(page, GFP_NOFS);
ClearPageChecked(page);
@@ -4917,7 +5024,8 @@ again:
* prepare_pages in the normal write path.
*/
clear_extent_bits(&BTRFS_I(inode)->io_tree, page_start, page_end,
- EXTENT_DIRTY | EXTENT_DELALLOC, GFP_NOFS);
+ EXTENT_DIRTY | EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING,
+ GFP_NOFS);
ret = btrfs_set_extent_delalloc(inode, page_start, page_end);
if (ret) {
@@ -4944,7 +5052,9 @@ again:
set_page_dirty(page);
SetPageUptodate(page);
- BTRFS_I(inode)->last_trans = root->fs_info->generation + 1;
+ BTRFS_I(inode)->last_trans = root->fs_info->generation;
+ BTRFS_I(inode)->last_sub_trans = BTRFS_I(inode)->root->log_transid;
+
unlock_extent(io_tree, page_start, page_end, GFP_NOFS);
out_unlock:
@@ -4969,7 +5079,9 @@ static void btrfs_truncate(struct inode *inode)
if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
return;
- btrfs_truncate_page(inode->i_mapping, inode->i_size);
+ ret = btrfs_truncate_page(inode->i_mapping, inode->i_size);
+ if (ret)
+ return;
btrfs_wait_ordered_range(inode, inode->i_size & (~mask), (u64)-1);
trans = btrfs_start_transaction(root, 1);
@@ -5064,9 +5176,12 @@ struct inode *btrfs_alloc_inode(struct super_block *sb)
if (!ei)
return NULL;
ei->last_trans = 0;
+ ei->last_sub_trans = 0;
ei->logged_trans = 0;
- ei->delalloc_extents = 0;
- ei->delalloc_reserved_extents = 0;
+ ei->outstanding_extents = 0;
+ ei->reserved_extents = 0;
+ ei->root = NULL;
+ spin_lock_init(&ei->accounting_lock);
btrfs_ordered_inode_tree_init(&ei->ordered_tree);
INIT_LIST_HEAD(&ei->i_orphan);
INIT_LIST_HEAD(&ei->ordered_operations);
@@ -5082,6 +5197,14 @@ void btrfs_destroy_inode(struct inode *inode)
WARN_ON(inode->i_data.nrpages);
/*
+ * This can happen where we create an inode, but somebody else also
+ * created the same inode and we need to destroy the one we already
+ * created.
+ */
+ if (!root)
+ goto free;
+
+ /*
* Make sure we're properly removed from the ordered operation
* lists.
*/
@@ -5116,6 +5239,7 @@ void btrfs_destroy_inode(struct inode *inode)
}
inode_tree_del(inode);
btrfs_drop_extent_cache(inode, 0, (u64)-1, 0);
+free:
kmem_cache_free(btrfs_inode_cachep, BTRFS_I(inode));
}
@@ -5221,11 +5345,14 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry,
return -ENOTEMPTY;
/*
- * 2 items for dir items
- * 1 item for orphan entry
- * 1 item for ref
+ * We want to reserve the absolute worst case amount of items. So if
+ * both inodes are subvols and we need to unlink them then that would
+ * require 4 item modifications, but if they are both normal inodes it
+ * would require 5 item modifications, so we'll assume their normal
+ * inodes. So 5 * 2 is 10, plus 1 for the new link, so 11 total items
+ * should cover the worst case number of items we'll modify.
*/
- ret = btrfs_reserve_metadata_space(root, 4);
+ ret = btrfs_reserve_metadata_space(root, 11);
if (ret)
return ret;
@@ -5341,7 +5468,7 @@ out_fail:
if (old_inode->i_ino == BTRFS_FIRST_FREE_OBJECTID)
up_read(&root->fs_info->subvol_sem);
- btrfs_unreserve_metadata_space(root, 4);
+ btrfs_unreserve_metadata_space(root, 11);
return ret;
}
@@ -5805,6 +5932,6 @@ static const struct inode_operations btrfs_symlink_inode_operations = {
.removexattr = btrfs_removexattr,
};
-struct dentry_operations btrfs_dentry_operations = {
+const struct dentry_operations btrfs_dentry_operations = {
.d_delete = btrfs_dentry_delete,
};
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 9a780c8d0ac..cdbb054102b 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -830,6 +830,7 @@ out_up_write:
out_unlock:
mutex_unlock(&inode->i_mutex);
if (!err) {
+ shrink_dcache_sb(root->fs_info->sb);
btrfs_invalidate_inodes(dest);
d_delete(dentry);
}
@@ -1122,8 +1123,10 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
datao += off - key.offset;
datal -= off - key.offset;
}
- if (key.offset + datao + datal > off + len)
- datal = off + len - key.offset - datao;
+
+ if (key.offset + datal > off + len)
+ datal = off + len - key.offset;
+
/* disko == 0 means it's a hole */
if (!disko)
datao = 0;
diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c
index 897fba835f8..5799bc46a30 100644
--- a/fs/btrfs/ordered-data.c
+++ b/fs/btrfs/ordered-data.c
@@ -306,6 +306,12 @@ int btrfs_remove_ordered_extent(struct inode *inode,
tree->last = NULL;
set_bit(BTRFS_ORDERED_COMPLETE, &entry->flags);
+ spin_lock(&BTRFS_I(inode)->accounting_lock);
+ BTRFS_I(inode)->outstanding_extents--;
+ spin_unlock(&BTRFS_I(inode)->accounting_lock);
+ btrfs_unreserve_metadata_for_delalloc(BTRFS_I(inode)->root,
+ inode, 1);
+
spin_lock(&BTRFS_I(inode)->root->fs_info->ordered_extent_lock);
list_del_init(&entry->root_extent_list);
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 361ad323faa..cfcc93c93a7 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -3518,7 +3518,7 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start)
BUG_ON(!rc->block_group);
btrfs_init_workers(&rc->workers, "relocate",
- fs_info->thread_pool_size);
+ fs_info->thread_pool_size, NULL);
rc->extent_root = extent_root;
btrfs_prepare_block_group_relocation(extent_root, rc->block_group);
@@ -3701,7 +3701,7 @@ int btrfs_recover_relocation(struct btrfs_root *root)
mapping_tree_init(&rc->reloc_root_tree);
INIT_LIST_HEAD(&rc->reloc_roots);
btrfs_init_workers(&rc->workers, "relocate",
- root->fs_info->thread_pool_size);
+ root->fs_info->thread_pool_size, NULL);
rc->extent_root = root->fs_info->extent_root;
set_reloc_control(rc);
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index 9351428f30e..67fa2d29d66 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -159,7 +159,6 @@ int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root
write_extent_buffer(l, item, ptr, sizeof(*item));
btrfs_mark_buffer_dirty(path->nodes[0]);
out:
- btrfs_release_path(root, path);
btrfs_free_path(path);
return ret;
}
@@ -332,7 +331,6 @@ int btrfs_del_root(struct btrfs_trans_handle *trans, struct btrfs_root *root,
BUG_ON(refs != 0);
ret = btrfs_del_item(trans, root, path);
out:
- btrfs_release_path(root, path);
btrfs_free_path(path);
return ret;
}
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 9de9b223641..752a5463bf5 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -66,7 +66,8 @@ enum {
Opt_degraded, Opt_subvol, Opt_device, Opt_nodatasum, Opt_nodatacow,
Opt_max_extent, Opt_max_inline, Opt_alloc_start, Opt_nobarrier,
Opt_ssd, Opt_nossd, Opt_ssd_spread, Opt_thread_pool, Opt_noacl,
- Opt_compress, Opt_notreelog, Opt_ratio, Opt_flushoncommit, Opt_err,
+ Opt_compress, Opt_notreelog, Opt_ratio, Opt_flushoncommit,
+ Opt_discard, Opt_err,
};
static match_table_t tokens = {
@@ -88,6 +89,7 @@ static match_table_t tokens = {
{Opt_notreelog, "notreelog"},
{Opt_flushoncommit, "flushoncommit"},
{Opt_ratio, "metadata_ratio=%d"},
+ {Opt_discard, "discard"},
{Opt_err, NULL},
};
@@ -257,6 +259,9 @@ int btrfs_parse_options(struct btrfs_root *root, char *options)
info->metadata_ratio);
}
break;
+ case Opt_discard:
+ btrfs_set_opt(info->mount_opt, DISCARD);
+ break;
default:
break;
}
@@ -344,7 +349,7 @@ static int btrfs_fill_super(struct super_block *sb,
sb->s_export_op = &btrfs_export_ops;
sb->s_xattr = btrfs_xattr_handlers;
sb->s_time_gran = 1;
-#ifdef CONFIG_BTRFS_POSIX_ACL
+#ifdef CONFIG_BTRFS_FS_POSIX_ACL
sb->s_flags |= MS_POSIXACL;
#endif
diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c
index 0b8f36d4400..c207e8c32c9 100644
--- a/fs/btrfs/transaction.c
+++ b/fs/btrfs/transaction.c
@@ -163,8 +163,14 @@ static void wait_current_trans(struct btrfs_root *root)
}
}
+enum btrfs_trans_type {
+ TRANS_START,
+ TRANS_JOIN,
+ TRANS_USERSPACE,
+};
+
static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root,
- int num_blocks, int wait)
+ int num_blocks, int type)
{
struct btrfs_trans_handle *h =
kmem_cache_alloc(btrfs_trans_handle_cachep, GFP_NOFS);
@@ -172,7 +178,8 @@ static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root,
mutex_lock(&root->fs_info->trans_mutex);
if (!root->fs_info->log_root_recovering &&
- ((wait == 1 && !root->fs_info->open_ioctl_trans) || wait == 2))
+ ((type == TRANS_START && !root->fs_info->open_ioctl_trans) ||
+ type == TRANS_USERSPACE))
wait_current_trans(root);
ret = join_transaction(root);
BUG_ON(ret);
@@ -186,7 +193,7 @@ static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root,
h->alloc_exclude_start = 0;
h->delayed_ref_updates = 0;
- if (!current->journal_info)
+ if (!current->journal_info && type != TRANS_USERSPACE)
current->journal_info = h;
root->fs_info->running_transaction->use_count++;
@@ -198,18 +205,18 @@ static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root,
struct btrfs_trans_handle *btrfs_start_transaction(struct btrfs_root *root,
int num_blocks)
{
- return start_transaction(root, num_blocks, 1);
+ return start_transaction(root, num_blocks, TRANS_START);
}
struct btrfs_trans_handle *btrfs_join_transaction(struct btrfs_root *root,
int num_blocks)
{
- return start_transaction(root, num_blocks, 0);
+ return start_transaction(root, num_blocks, TRANS_JOIN);
}
struct btrfs_trans_handle *btrfs_start_ioctl_transaction(struct btrfs_root *r,
int num_blocks)
{
- return start_transaction(r, num_blocks, 2);
+ return start_transaction(r, num_blocks, TRANS_USERSPACE);
}
/* wait for a transaction commit to be fully complete */
@@ -344,10 +351,10 @@ int btrfs_end_transaction_throttle(struct btrfs_trans_handle *trans,
/*
* when btree blocks are allocated, they have some corresponding bits set for
* them in one of two extent_io trees. This is used to make sure all of
- * those extents are on disk for transaction or log commit
+ * those extents are sent to disk but does not wait on them
*/
-int btrfs_write_and_wait_marked_extents(struct btrfs_root *root,
- struct extent_io_tree *dirty_pages)
+int btrfs_write_marked_extents(struct btrfs_root *root,
+ struct extent_io_tree *dirty_pages)
{
int ret;
int err = 0;
@@ -394,6 +401,29 @@ int btrfs_write_and_wait_marked_extents(struct btrfs_root *root,
page_cache_release(page);
}
}
+ if (err)
+ werr = err;
+ return werr;
+}
+
+/*
+ * when btree blocks are allocated, they have some corresponding bits set for
+ * them in one of two extent_io trees. This is used to make sure all of
+ * those extents are on disk for transaction or log commit. We wait
+ * on all the pages and clear them from the dirty pages state tree
+ */
+int btrfs_wait_marked_extents(struct btrfs_root *root,
+ struct extent_io_tree *dirty_pages)
+{
+ int ret;
+ int err = 0;
+ int werr = 0;
+ struct page *page;
+ struct inode *btree_inode = root->fs_info->btree_inode;
+ u64 start = 0;
+ u64 end;
+ unsigned long index;
+
while (1) {
ret = find_first_extent_bit(dirty_pages, 0, &start, &end,
EXTENT_DIRTY);
@@ -424,6 +454,22 @@ int btrfs_write_and_wait_marked_extents(struct btrfs_root *root,
return werr;
}
+/*
+ * when btree blocks are allocated, they have some corresponding bits set for
+ * them in one of two extent_io trees. This is used to make sure all of
+ * those extents are on disk for transaction or log commit
+ */
+int btrfs_write_and_wait_marked_extents(struct btrfs_root *root,
+ struct extent_io_tree *dirty_pages)
+{
+ int ret;
+ int ret2;
+
+ ret = btrfs_write_marked_extents(root, dirty_pages);
+ ret2 = btrfs_wait_marked_extents(root, dirty_pages);
+ return ret || ret2;
+}
+
int btrfs_write_and_wait_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root)
{
diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h
index 663c6740491..d4e3e7a6938 100644
--- a/fs/btrfs/transaction.h
+++ b/fs/btrfs/transaction.h
@@ -79,6 +79,7 @@ static inline void btrfs_set_inode_last_trans(struct btrfs_trans_handle *trans,
struct inode *inode)
{
BTRFS_I(inode)->last_trans = trans->transaction->transid;
+ BTRFS_I(inode)->last_sub_trans = BTRFS_I(inode)->root->log_transid;
}
int btrfs_end_transaction(struct btrfs_trans_handle *trans,
@@ -107,5 +108,9 @@ int btrfs_record_root_in_trans(struct btrfs_trans_handle *trans,
struct btrfs_root *root);
int btrfs_write_and_wait_marked_extents(struct btrfs_root *root,
struct extent_io_tree *dirty_pages);
+int btrfs_write_marked_extents(struct btrfs_root *root,
+ struct extent_io_tree *dirty_pages);
+int btrfs_wait_marked_extents(struct btrfs_root *root,
+ struct extent_io_tree *dirty_pages);
int btrfs_transaction_in_commit(struct btrfs_fs_info *info);
#endif
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index 7827841b55c..741666a7676 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -137,11 +137,20 @@ static int start_log_trans(struct btrfs_trans_handle *trans,
mutex_lock(&root->log_mutex);
if (root->log_root) {
+ if (!root->log_start_pid) {
+ root->log_start_pid = current->pid;
+ root->log_multiple_pids = false;
+ } else if (root->log_start_pid != current->pid) {
+ root->log_multiple_pids = true;
+ }
+
root->log_batch++;
atomic_inc(&root->log_writers);
mutex_unlock(&root->log_mutex);
return 0;
}
+ root->log_multiple_pids = false;
+ root->log_start_pid = current->pid;
mutex_lock(&root->fs_info->tree_log_mutex);
if (!root->fs_info->log_root_tree) {
ret = btrfs_init_log_root_tree(trans, root->fs_info);
@@ -1971,6 +1980,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
int ret;
struct btrfs_root *log = root->log_root;
struct btrfs_root *log_root_tree = root->fs_info->log_root_tree;
+ u64 log_transid = 0;
mutex_lock(&root->log_mutex);
index1 = root->log_transid % 2;
@@ -1987,10 +1997,11 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
while (1) {
unsigned long batch = root->log_batch;
- mutex_unlock(&root->log_mutex);
- schedule_timeout_uninterruptible(1);
- mutex_lock(&root->log_mutex);
-
+ if (root->log_multiple_pids) {
+ mutex_unlock(&root->log_mutex);
+ schedule_timeout_uninterruptible(1);
+ mutex_lock(&root->log_mutex);
+ }
wait_for_writer(trans, root);
if (batch == root->log_batch)
break;
@@ -2003,14 +2014,19 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
goto out;
}
- ret = btrfs_write_and_wait_marked_extents(log, &log->dirty_log_pages);
+ /* we start IO on all the marked extents here, but we don't actually
+ * wait for them until later.
+ */
+ ret = btrfs_write_marked_extents(log, &log->dirty_log_pages);
BUG_ON(ret);
btrfs_set_root_node(&log->root_item, log->node);
root->log_batch = 0;
+ log_transid = root->log_transid;
root->log_transid++;
log->log_transid = root->log_transid;
+ root->log_start_pid = 0;
smp_mb();
/*
* log tree has been flushed to disk, new modifications of
@@ -2036,6 +2052,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
index2 = log_root_tree->log_transid % 2;
if (atomic_read(&log_root_tree->log_commit[index2])) {
+ btrfs_wait_marked_extents(log, &log->dirty_log_pages);
wait_log_commit(trans, log_root_tree,
log_root_tree->log_transid);
mutex_unlock(&log_root_tree->log_mutex);
@@ -2055,6 +2072,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
* check the full commit flag again
*/
if (root->fs_info->last_trans_log_full_commit == trans->transid) {
+ btrfs_wait_marked_extents(log, &log->dirty_log_pages);
mutex_unlock(&log_root_tree->log_mutex);
ret = -EAGAIN;
goto out_wake_log_root;
@@ -2063,6 +2081,7 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
ret = btrfs_write_and_wait_marked_extents(log_root_tree,
&log_root_tree->dirty_log_pages);
BUG_ON(ret);
+ btrfs_wait_marked_extents(log, &log->dirty_log_pages);
btrfs_set_super_log_root(&root->fs_info->super_for_commit,
log_root_tree->node->start);
@@ -2082,9 +2101,14 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans,
* the running transaction open, so a full commit can't hop
* in and cause problems either.
*/
- write_ctree_super(trans, root->fs_info->tree_root, 2);
+ write_ctree_super(trans, root->fs_info->tree_root, 1);
ret = 0;
+ mutex_lock(&root->log_mutex);
+ if (root->last_log_commit < log_transid)
+ root->last_log_commit = log_transid;
+ mutex_unlock(&root->log_mutex);
+
out_wake_log_root:
atomic_set(&log_root_tree->log_commit[index2], 0);
smp_mb();
@@ -2852,6 +2876,21 @@ out:
return ret;
}
+static int inode_in_log(struct btrfs_trans_handle *trans,
+ struct inode *inode)
+{
+ struct btrfs_root *root = BTRFS_I(inode)->root;
+ int ret = 0;
+
+ mutex_lock(&root->log_mutex);
+ if (BTRFS_I(inode)->logged_trans == trans->transid &&
+ BTRFS_I(inode)->last_sub_trans <= root->last_log_commit)
+ ret = 1;
+ mutex_unlock(&root->log_mutex);
+ return ret;
+}
+
+
/*
* helper function around btrfs_log_inode to make sure newly created
* parent directories also end up in the log. A minimal inode and backref
@@ -2891,6 +2930,11 @@ int btrfs_log_inode_parent(struct btrfs_trans_handle *trans,
if (ret)
goto end_no_trans;
+ if (inode_in_log(trans, inode)) {
+ ret = BTRFS_NO_LOG_SYNC;
+ goto end_no_trans;
+ }
+
start_log_trans(trans, root);
ret = btrfs_log_inode(trans, root, inode, inode_only);
diff --git a/fs/btrfs/tree-log.h b/fs/btrfs/tree-log.h
index d09c7609e16..0776eacb508 100644
--- a/fs/btrfs/tree-log.h
+++ b/fs/btrfs/tree-log.h
@@ -19,6 +19,9 @@
#ifndef __TREE_LOG_
#define __TREE_LOG_
+/* return value for btrfs_log_dentry_safe that means we don't need to log it at all */
+#define BTRFS_NO_LOG_SYNC 256
+
int btrfs_sync_log(struct btrfs_trans_handle *trans,
struct btrfs_root *root);
int btrfs_free_log(struct btrfs_trans_handle *trans, struct btrfs_root *root);
diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c
index b0fc93f95fd..b6dd5967c48 100644
--- a/fs/btrfs/xattr.c
+++ b/fs/btrfs/xattr.c
@@ -260,7 +260,7 @@ err:
* attributes are handled directly.
*/
struct xattr_handler *btrfs_xattr_handlers[] = {
-#ifdef CONFIG_BTRFS_POSIX_ACL
+#ifdef CONFIG_BTRFS_FS_POSIX_ACL
&btrfs_xattr_acl_access_handler,
&btrfs_xattr_acl_default_handler,
#endif
diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h
index 6928c24d1d4..5646727e33f 100644
--- a/fs/cifs/cifsproto.h
+++ b/fs/cifs/cifsproto.h
@@ -388,4 +388,5 @@ extern int CIFSSMBSetPosixACL(const int xid, struct cifsTconInfo *tcon,
const struct nls_table *nls_codepage, int remap_special_chars);
extern int CIFSGetExtAttr(const int xid, struct cifsTconInfo *tcon,
const int netfid, __u64 *pExtAttrBits, __u64 *pMask);
+extern void cifs_autodisable_serverino(struct cifs_sb_info *cifs_sb);
#endif /* _CIFSPROTO_H */
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 43003e0bef1..63ea83ff687 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -1577,7 +1577,8 @@ cifs_get_tcp_session(struct smb_vol *volume_info)
out_err:
if (tcp_ses) {
- kfree(tcp_ses->hostname);
+ if (!IS_ERR(tcp_ses->hostname))
+ kfree(tcp_ses->hostname);
if (tcp_ses->ssocket)
sock_release(tcp_ses->ssocket);
kfree(tcp_ses);
@@ -2219,16 +2220,8 @@ is_path_accessible(int xid, struct cifsTconInfo *tcon,
struct cifs_sb_info *cifs_sb, const char *full_path)
{
int rc;
- __u64 inode_num;
FILE_ALL_INFO *pfile_info;
- rc = CIFSGetSrvInodeNumber(xid, tcon, full_path, &inode_num,
- cifs_sb->local_nls,
- cifs_sb->mnt_cifs_flags &
- CIFS_MOUNT_MAP_SPECIAL_CHR);
- if (rc != -EOPNOTSUPP)
- return rc;
-
pfile_info = kmalloc(sizeof(FILE_ALL_INFO), GFP_KERNEL);
if (pfile_info == NULL)
return -ENOMEM;
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index 5e2492535da..cababd8a52d 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -512,13 +512,10 @@ int cifs_get_inode_info(struct inode **pinode,
cifs_sb->local_nls,
cifs_sb->mnt_cifs_flags &
CIFS_MOUNT_MAP_SPECIAL_CHR);
- if (rc1) {
+ if (rc1 || !fattr.cf_uniqueid) {
cFYI(1, ("GetSrvInodeNum rc %d", rc1));
fattr.cf_uniqueid = iunique(sb, ROOT_I);
- /* disable serverino if call not supported */
- if (rc1 == -EINVAL)
- cifs_sb->mnt_cifs_flags &=
- ~CIFS_MOUNT_SERVER_INUM;
+ cifs_autodisable_serverino(cifs_sb);
}
} else {
fattr.cf_uniqueid = iunique(sb, ROOT_I);
diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c
index 0241b25ac33..d27d4ec6579 100644
--- a/fs/cifs/misc.c
+++ b/fs/cifs/misc.c
@@ -715,3 +715,17 @@ cifsConvertToUCS(__le16 *target, const char *source, int maxlen,
ctoUCS_out:
return i;
}
+
+void
+cifs_autodisable_serverino(struct cifs_sb_info *cifs_sb)
+{
+ if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) {
+ cifs_sb->mnt_cifs_flags &= ~CIFS_MOUNT_SERVER_INUM;
+ cERROR(1, ("Autodisabling the use of server inode numbers on "
+ "%s. This server doesn't seem to support them "
+ "properly. Hardlinks will not be recognized on this "
+ "mount. Consider mounting with the \"noserverino\" "
+ "option to silence this message.",
+ cifs_sb->tcon->treeName));
+ }
+}
diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
index 1f098ca7163..f84062f9a98 100644
--- a/fs/cifs/readdir.c
+++ b/fs/cifs/readdir.c
@@ -727,11 +727,12 @@ static int cifs_filldir(char *pfindEntry, struct file *file, filldir_t filldir,
cifs_dir_info_to_fattr(&fattr, (FILE_DIRECTORY_INFO *)
pfindEntry, cifs_sb);
- /* FIXME: make _to_fattr functions fill this out */
- if (pCifsF->srch_inf.info_level == SMB_FIND_FILE_ID_FULL_DIR_INFO)
+ if (inum && (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM)) {
fattr.cf_uniqueid = inum;
- else
+ } else {
fattr.cf_uniqueid = iunique(sb, ROOT_I);
+ cifs_autodisable_serverino(cifs_sb);
+ }
ino = cifs_uniqueid_to_ino_t(fattr.cf_uniqueid);
tmp_dentry = cifs_readdir_lookup(file->f_dentry, &qstring, &fattr);
diff --git a/fs/compat.c b/fs/compat.c
index d576b552e8e..6c19040ffee 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -1532,6 +1532,8 @@ int compat_do_execve(char * filename,
if (retval < 0)
goto out;
+ current->stack_start = current->mm->start_stack;
+
/* execve succeeded */
current->fs->in_exec = 0;
current->in_execve = 0;
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index f91fd51b32e..d84e7058c29 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -1800,7 +1800,7 @@ struct space_resv_32 {
/* just account for different alignment */
static int compat_ioctl_preallocate(struct file *file, unsigned long arg)
{
- struct space_resv_32 __user *p32 = (void __user *)arg;
+ struct space_resv_32 __user *p32 = compat_ptr(arg);
struct space_resv __user *p = compat_alloc_user_space(sizeof(*p));
if (copy_in_user(&p->l_type, &p32->l_type, sizeof(s16)) ||
@@ -2802,7 +2802,7 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd,
#else
case FS_IOC_RESVSP:
case FS_IOC_RESVSP64:
- error = ioctl_preallocate(filp, (void __user *)arg);
+ error = ioctl_preallocate(filp, compat_ptr(arg));
goto out_fput;
#endif
diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c
index 240cef14fe5..70736eb4b51 100644
--- a/fs/dlm/lowcomms.c
+++ b/fs/dlm/lowcomms.c
@@ -316,6 +316,10 @@ int dlm_lowcomms_connect_node(int nodeid)
{
struct connection *con;
+ /* with sctp there's no connecting without sending */
+ if (dlm_config.ci_protocol != 0)
+ return 0;
+
if (nodeid == dlm_our_nodeid())
return 0;
@@ -455,9 +459,9 @@ static void process_sctp_notification(struct connection *con,
int prim_len, ret;
int addr_len;
struct connection *new_con;
- struct file *file;
sctp_peeloff_arg_t parg;
int parglen = sizeof(parg);
+ int err;
/*
* We get this before any data for an association.
@@ -512,19 +516,22 @@ static void process_sctp_notification(struct connection *con,
ret = kernel_getsockopt(con->sock, IPPROTO_SCTP,
SCTP_SOCKOPT_PEELOFF,
(void *)&parg, &parglen);
- if (ret) {
+ if (ret < 0) {
log_print("Can't peel off a socket for "
- "connection %d to node %d: err=%d\n",
+ "connection %d to node %d: err=%d",
parg.associd, nodeid, ret);
+ return;
+ }
+ new_con->sock = sockfd_lookup(parg.sd, &err);
+ if (!new_con->sock) {
+ log_print("sockfd_lookup error %d", err);
+ return;
}
- file = fget(parg.sd);
- new_con->sock = SOCKET_I(file->f_dentry->d_inode);
add_sock(new_con->sock, new_con);
- fput(file);
- put_unused_fd(parg.sd);
+ sockfd_put(new_con->sock);
- log_print("got new/restarted association %d nodeid %d",
- (int)sn->sn_assoc_change.sac_assoc_id, nodeid);
+ log_print("connecting to %d sctp association %d",
+ nodeid, (int)sn->sn_assoc_change.sac_assoc_id);
/* Send any pending writes */
clear_bit(CF_CONNECT_PENDING, &new_con->flags);
@@ -837,8 +844,6 @@ static void sctp_init_assoc(struct connection *con)
if (con->retries++ > MAX_CONNECT_RETRIES)
return;
- log_print("Initiating association with node %d", con->nodeid);
-
if (nodeid_to_addr(con->nodeid, (struct sockaddr *)&rem_addr)) {
log_print("no address for nodeid %d", con->nodeid);
return;
@@ -855,11 +860,14 @@ static void sctp_init_assoc(struct connection *con)
outmessage.msg_flags = MSG_EOR;
spin_lock(&con->writequeue_lock);
- e = list_entry(con->writequeue.next, struct writequeue_entry,
- list);
- BUG_ON((struct list_head *) e == &con->writequeue);
+ if (list_empty(&con->writequeue)) {
+ spin_unlock(&con->writequeue_lock);
+ log_print("writequeue empty for nodeid %d", con->nodeid);
+ return;
+ }
+ e = list_first_entry(&con->writequeue, struct writequeue_entry, list);
len = e->len;
offset = e->offset;
spin_unlock(&con->writequeue_lock);
diff --git a/fs/ecryptfs/Kconfig b/fs/ecryptfs/Kconfig
index 8aadb99b763..1cd6d9d3e29 100644
--- a/fs/ecryptfs/Kconfig
+++ b/fs/ecryptfs/Kconfig
@@ -1,8 +1,9 @@
config ECRYPT_FS
tristate "eCrypt filesystem layer support (EXPERIMENTAL)"
- depends on EXPERIMENTAL && KEYS && NET
+ depends on EXPERIMENTAL && KEYS && CRYPTO
select CRYPTO_ECB
select CRYPTO_CBC
+ select CRYPTO_MD5
help
Encrypted filesystem that operates on the VFS layer. See
<file:Documentation/filesystems/ecryptfs.txt> to learn more about
diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index 101fe4c7b1e..c6ac85d6c70 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -35,6 +35,7 @@
#include <linux/key.h>
#include <linux/parser.h>
#include <linux/fs_stack.h>
+#include <linux/ima.h>
#include "ecryptfs_kernel.h"
/**
@@ -118,6 +119,7 @@ int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry)
const struct cred *cred = current_cred();
struct ecryptfs_inode_info *inode_info =
ecryptfs_inode_to_private(ecryptfs_dentry->d_inode);
+ int opened_lower_file = 0;
int rc = 0;
mutex_lock(&inode_info->lower_file_mutex);
@@ -134,9 +136,12 @@ int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry)
"for lower_dentry [0x%p] and lower_mnt [0x%p]; "
"rc = [%d]\n", lower_dentry, lower_mnt, rc);
inode_info->lower_file = NULL;
- }
+ } else
+ opened_lower_file = 1;
}
mutex_unlock(&inode_info->lower_file_mutex);
+ if (opened_lower_file)
+ ima_counts_get(inode_info->lower_file);
return rc;
}
diff --git a/fs/exec.c b/fs/exec.c
index d49be6bc179..ba112bd4a33 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -624,10 +624,8 @@ int setup_arg_pages(struct linux_binprm *bprm,
/* Move stack pages down in memory. */
if (stack_shift) {
ret = shift_arg_pages(vma, stack_shift);
- if (ret) {
- up_write(&mm->mmap_sem);
- return ret;
- }
+ if (ret)
+ goto out_unlock;
}
#ifdef CONFIG_STACK_GROWSUP
@@ -641,7 +639,7 @@ int setup_arg_pages(struct linux_binprm *bprm,
out_unlock:
up_write(&mm->mmap_sem);
- return 0;
+ return ret;
}
EXPORT_SYMBOL(setup_arg_pages);
diff --git a/fs/ext3/fsync.c b/fs/ext3/fsync.c
index 451d166bbe9..8209f266e9a 100644
--- a/fs/ext3/fsync.c
+++ b/fs/ext3/fsync.c
@@ -46,19 +46,21 @@
int ext3_sync_file(struct file * file, struct dentry *dentry, int datasync)
{
struct inode *inode = dentry->d_inode;
+ struct ext3_inode_info *ei = EXT3_I(inode);
+ journal_t *journal = EXT3_SB(inode->i_sb)->s_journal;
int ret = 0;
+ tid_t commit_tid;
+
+ if (inode->i_sb->s_flags & MS_RDONLY)
+ return 0;
J_ASSERT(ext3_journal_current_handle() == NULL);
/*
- * data=writeback:
+ * data=writeback,ordered:
* The caller's filemap_fdatawrite()/wait will sync the data.
- * sync_inode() will sync the metadata
- *
- * data=ordered:
- * The caller's filemap_fdatawrite() will write the data and
- * sync_inode() will write the inode if it is dirty. Then the caller's
- * filemap_fdatawait() will wait on the pages.
+ * Metadata is in the journal, we wait for a proper transaction
+ * to commit here.
*
* data=journal:
* filemap_fdatawrite won't do anything (the buffers are clean).
@@ -73,22 +75,16 @@ int ext3_sync_file(struct file * file, struct dentry *dentry, int datasync)
goto out;
}
- if (datasync && !(inode->i_state & I_DIRTY_DATASYNC))
- goto flush;
+ if (datasync)
+ commit_tid = atomic_read(&ei->i_datasync_tid);
+ else
+ commit_tid = atomic_read(&ei->i_sync_tid);
- /*
- * The VFS has written the file data. If the inode is unaltered
- * then we need not start a commit.
- */
- if (inode->i_state & (I_DIRTY_SYNC|I_DIRTY_DATASYNC)) {
- struct writeback_control wbc = {
- .sync_mode = WB_SYNC_ALL,
- .nr_to_write = 0, /* sys_fsync did this */
- };
- ret = sync_inode(inode, &wbc);
+ if (log_start_commit(journal, commit_tid)) {
+ log_wait_commit(journal, commit_tid);
goto out;
}
-flush:
+
/*
* In case we didn't commit a transaction, we have to flush
* disk caches manually so that data really is on persistent
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index acf1b142332..354ed3b47b3 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -699,8 +699,9 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode,
int err = 0;
struct ext3_block_alloc_info *block_i;
ext3_fsblk_t current_block;
+ struct ext3_inode_info *ei = EXT3_I(inode);
- block_i = EXT3_I(inode)->i_block_alloc_info;
+ block_i = ei->i_block_alloc_info;
/*
* If we're splicing into a [td]indirect block (as opposed to the
* inode) then we need to get write access to the [td]indirect block
@@ -741,6 +742,8 @@ static int ext3_splice_branch(handle_t *handle, struct inode *inode,
inode->i_ctime = CURRENT_TIME_SEC;
ext3_mark_inode_dirty(handle, inode);
+ /* ext3_mark_inode_dirty already updated i_sync_tid */
+ atomic_set(&ei->i_datasync_tid, handle->h_transaction->t_tid);
/* had we spliced it onto indirect block? */
if (where->bh) {
@@ -1735,6 +1738,7 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb,
ssize_t ret;
int orphan = 0;
size_t count = iov_length(iov, nr_segs);
+ int retries = 0;
if (rw == WRITE) {
loff_t final_size = offset + count;
@@ -1757,9 +1761,12 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb,
}
}
+retry:
ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov,
offset, nr_segs,
ext3_get_block, NULL);
+ if (ret == -ENOSPC && ext3_should_retry_alloc(inode->i_sb, &retries))
+ goto retry;
if (orphan) {
int err;
@@ -2750,6 +2757,8 @@ struct inode *ext3_iget(struct super_block *sb, unsigned long ino)
struct ext3_inode_info *ei;
struct buffer_head *bh;
struct inode *inode;
+ journal_t *journal = EXT3_SB(sb)->s_journal;
+ transaction_t *transaction;
long ret;
int block;
@@ -2827,6 +2836,30 @@ struct inode *ext3_iget(struct super_block *sb, unsigned long ino)
ei->i_data[block] = raw_inode->i_block[block];
INIT_LIST_HEAD(&ei->i_orphan);
+ /*
+ * Set transaction id's of transactions that have to be committed
+ * to finish f[data]sync. We set them to currently running transaction
+ * as we cannot be sure that the inode or some of its metadata isn't
+ * part of the transaction - the inode could have been reclaimed and
+ * now it is reread from disk.
+ */
+ if (journal) {
+ tid_t tid;
+
+ spin_lock(&journal->j_state_lock);
+ if (journal->j_running_transaction)
+ transaction = journal->j_running_transaction;
+ else
+ transaction = journal->j_committing_transaction;
+ if (transaction)
+ tid = transaction->t_tid;
+ else
+ tid = journal->j_commit_sequence;
+ spin_unlock(&journal->j_state_lock);
+ atomic_set(&ei->i_sync_tid, tid);
+ atomic_set(&ei->i_datasync_tid, tid);
+ }
+
if (inode->i_ino >= EXT3_FIRST_INO(inode->i_sb) + 1 &&
EXT3_INODE_SIZE(inode->i_sb) > EXT3_GOOD_OLD_INODE_SIZE) {
/*
@@ -3011,6 +3044,7 @@ again:
err = rc;
ei->i_state &= ~EXT3_STATE_NEW;
+ atomic_set(&ei->i_sync_tid, handle->h_transaction->t_tid);
out_brelse:
brelse (bh);
ext3_std_error(inode->i_sb, err);
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 72743d36050..427496c4767 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -466,6 +466,8 @@ static struct inode *ext3_alloc_inode(struct super_block *sb)
return NULL;
ei->i_block_alloc_info = NULL;
ei->vfs_inode.i_version = 1;
+ atomic_set(&ei->i_datasync_tid, 0);
+ atomic_set(&ei->i_sync_tid, 0);
return &ei->vfs_inode;
}
@@ -2321,7 +2323,18 @@ static int ext3_commit_super(struct super_block *sb,
if (!sbh)
return error;
- es->s_wtime = cpu_to_le32(get_seconds());
+ /*
+ * If the file system is mounted read-only, don't update the
+ * superblock write time. This avoids updating the superblock
+ * write time when we are mounting the root file system
+ * read/only but we need to replay the journal; at that point,
+ * for people who are east of GMT and who make their clock
+ * tick in localtime for Windows bug-for-bug compatibility,
+ * the clock is set in the future, and this will cause e2fsck
+ * to complain and force a full file system check.
+ */
+ if (!(sb->s_flags & MS_RDONLY))
+ es->s_wtime = cpu_to_le32(get_seconds());
es->s_free_blocks_count = cpu_to_le32(ext3_count_free_blocks(sb));
es->s_free_inodes_count = cpu_to_le32(ext3_count_free_inodes(sb));
BUFFER_TRACE(sbh, "marking dirty");
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 984ca0cb38c..8825515eedd 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -322,6 +322,7 @@ static inline __u32 ext4_mask_flags(umode_t mode, __u32 flags)
#define EXT4_STATE_NO_EXPAND 0x00000008 /* No space for expansion */
#define EXT4_STATE_DA_ALLOC_CLOSE 0x00000010 /* Alloc DA blks on close */
#define EXT4_STATE_EXT_MIGRATE 0x00000020 /* Inode is migrating */
+#define EXT4_STATE_DIO_UNWRITTEN 0x00000040 /* need convert on dio done*/
/* Used to pass group descriptor data when online resize is done */
struct ext4_new_group_input {
@@ -743,6 +744,7 @@ struct ext4_inode_info {
#define EXT4_MOUNT_QUOTA 0x80000 /* Some quota option set */
#define EXT4_MOUNT_USRQUOTA 0x100000 /* "old" user quota */
#define EXT4_MOUNT_GRPQUOTA 0x200000 /* "old" group quota */
+#define EXT4_MOUNT_JOURNAL_CHECKSUM 0x800000 /* Journal checksums */
#define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT 0x1000000 /* Journal Async Commit */
#define EXT4_MOUNT_I_VERSION 0x2000000 /* i_version support */
#define EXT4_MOUNT_DELALLOC 0x8000000 /* Delalloc support */
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 10539e36428..715264b4bae 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -2807,6 +2807,8 @@ fix_extent_len:
* into three uninitialized extent(at most). After IO complete, the part
* being filled will be convert to initialized by the end_io callback function
* via ext4_convert_unwritten_extents().
+ *
+ * Returns the size of uninitialized extent to be written on success.
*/
static int ext4_split_unwritten_extents(handle_t *handle,
struct inode *inode,
@@ -2824,7 +2826,6 @@ static int ext4_split_unwritten_extents(handle_t *handle,
unsigned int allocated, ee_len, depth;
ext4_fsblk_t newblock;
int err = 0;
- int ret = 0;
ext_debug("ext4_split_unwritten_extents: inode %lu,"
"iblock %llu, max_blocks %u\n", inode->i_ino,
@@ -2842,12 +2843,12 @@ static int ext4_split_unwritten_extents(handle_t *handle,
ext4_ext_store_pblock(&orig_ex, ext_pblock(ex));
/*
- * if the entire unintialized extent length less than
- * the size of extent to write, there is no need to split
- * uninitialized extent
+ * If the uninitialized extent begins at the same logical
+ * block where the write begins, and the write completely
+ * covers the extent, then we don't need to split it.
*/
- if (allocated <= max_blocks)
- return ret;
+ if ((iblock == ee_block) && (allocated <= max_blocks))
+ return allocated;
err = ext4_ext_get_access(handle, inode, path + depth);
if (err)
@@ -3048,12 +3049,18 @@ ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode,
ret = ext4_split_unwritten_extents(handle,
inode, path, iblock,
max_blocks, flags);
- /* flag the io_end struct that we need convert when IO done */
+ /*
+ * Flag the inode(non aio case) or end_io struct (aio case)
+ * that this IO needs to convertion to written when IO is
+ * completed
+ */
if (io)
io->flag = DIO_AIO_UNWRITTEN;
+ else
+ EXT4_I(inode)->i_state |= EXT4_STATE_DIO_UNWRITTEN;
goto out;
}
- /* DIO end_io complete, convert the filled extent to written */
+ /* async DIO end_io complete, convert the filled extent to written */
if (flags == EXT4_GET_BLOCKS_DIO_CONVERT_EXT) {
ret = ext4_convert_unwritten_extents_dio(handle, inode,
path);
@@ -3295,10 +3302,16 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode,
* To avoid unecessary convertion for every aio dio rewrite
* to the mid of file, here we flag the IO that is really
* need the convertion.
- *
+ * For non asycn direct IO case, flag the inode state
+ * that we need to perform convertion when IO is done.
*/
- if (io && flags == EXT4_GET_BLOCKS_DIO_CREATE_EXT)
- io->flag = DIO_AIO_UNWRITTEN;
+ if (flags == EXT4_GET_BLOCKS_DIO_CREATE_EXT) {
+ if (io)
+ io->flag = DIO_AIO_UNWRITTEN;
+ else
+ EXT4_I(inode)->i_state |=
+ EXT4_STATE_DIO_UNWRITTEN;;
+ }
}
err = ext4_ext_insert_extent(handle, inode, path, &newex, flags);
if (err) {
@@ -3519,6 +3532,7 @@ retry:
*
* This function is called from the direct IO end io call back
* function, to convert the fallocated extents after IO is completed.
+ * Returns 0 on success.
*/
int ext4_convert_unwritten_extents(struct inode *inode, loff_t offset,
loff_t len)
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 5c5bc5dafff..2c8caa51add 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -193,7 +193,7 @@ static int try_to_extend_transaction(handle_t *handle, struct inode *inode)
* so before we call here everything must be consistently dirtied against
* this transaction.
*/
- int ext4_truncate_restart_trans(handle_t *handle, struct inode *inode,
+int ext4_truncate_restart_trans(handle_t *handle, struct inode *inode,
int nblocks)
{
int ret;
@@ -209,6 +209,7 @@ static int try_to_extend_transaction(handle_t *handle, struct inode *inode)
up_write(&EXT4_I(inode)->i_data_sem);
ret = ext4_journal_restart(handle, blocks_for_truncate(inode));
down_write(&EXT4_I(inode)->i_data_sem);
+ ext4_discard_preallocations(inode);
return ret;
}
@@ -3445,8 +3446,6 @@ out:
return ret;
}
-/* Maximum number of blocks we map for direct IO at once. */
-
static int ext4_get_block_dio_write(struct inode *inode, sector_t iblock,
struct buffer_head *bh_result, int create)
{
@@ -3654,13 +3653,14 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset,
ext4_io_end_t *io_end = iocb->private;
struct workqueue_struct *wq;
+ /* if not async direct IO or dio with 0 bytes write, just return */
+ if (!io_end || !size)
+ return;
+
ext_debug("ext4_end_io_dio(): io_end 0x%p"
"for inode %lu, iocb 0x%p, offset %llu, size %llu\n",
iocb->private, io_end->inode->i_ino, iocb, offset,
size);
- /* if not async direct IO or dio with 0 bytes write, just return */
- if (!io_end || !size)
- return;
/* if not aio dio with unwritten extents, just free io and return */
if (io_end->flag != DIO_AIO_UNWRITTEN){
@@ -3771,13 +3771,19 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb,
if (ret != -EIOCBQUEUED && ret <= 0 && iocb->private) {
ext4_free_io_end(iocb->private);
iocb->private = NULL;
- } else if (ret > 0)
+ } else if (ret > 0 && (EXT4_I(inode)->i_state &
+ EXT4_STATE_DIO_UNWRITTEN)) {
+ int err;
/*
* for non AIO case, since the IO is already
* completed, we could do the convertion right here
*/
- ret = ext4_convert_unwritten_extents(inode,
- offset, ret);
+ err = ext4_convert_unwritten_extents(inode,
+ offset, ret);
+ if (err < 0)
+ ret = err;
+ EXT4_I(inode)->i_state &= ~EXT4_STATE_DIO_UNWRITTEN;
+ }
return ret;
}
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index 7c8fe80bacd..6d2c1b897fc 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -1518,12 +1518,8 @@ static int ext4_add_entry(handle_t *handle, struct dentry *dentry,
return retval;
if (blocks == 1 && !dx_fallback &&
- EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_DIR_INDEX)) {
- retval = make_indexed_dir(handle, dentry, inode, bh);
- if (retval == -ENOSPC)
- brelse(bh);
- return retval;
- }
+ EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_DIR_INDEX))
+ return make_indexed_dir(handle, dentry, inode, bh);
brelse(bh);
}
bh = ext4_append(handle, dir, &block, &retval);
@@ -1532,10 +1528,7 @@ static int ext4_add_entry(handle_t *handle, struct dentry *dentry,
de = (struct ext4_dir_entry_2 *) bh->b_data;
de->inode = 0;
de->rec_len = ext4_rec_len_to_disk(blocksize, blocksize);
- retval = add_dirent_to_buf(handle, dentry, inode, de, bh);
- if (retval == -ENOSPC)
- brelse(bh);
- return retval;
+ return add_dirent_to_buf(handle, dentry, inode, de, bh);
}
/*
@@ -1664,8 +1657,7 @@ static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry,
if (!de)
goto cleanup;
err = add_dirent_to_buf(handle, dentry, inode, de, bh);
- if (err != -ENOSPC)
- bh = NULL;
+ bh = NULL;
goto cleanup;
journal_error:
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 312211ee05a..d4ca92aab51 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1300,9 +1300,11 @@ static int parse_options(char *options, struct super_block *sb,
*journal_devnum = option;
break;
case Opt_journal_checksum:
- break; /* Kept for backwards compatibility */
+ set_opt(sbi->s_mount_opt, JOURNAL_CHECKSUM);
+ break;
case Opt_journal_async_commit:
set_opt(sbi->s_mount_opt, JOURNAL_ASYNC_COMMIT);
+ set_opt(sbi->s_mount_opt, JOURNAL_CHECKSUM);
break;
case Opt_noload:
set_opt(sbi->s_mount_opt, NOLOAD);
@@ -2759,14 +2761,20 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
goto failed_mount4;
}
- jbd2_journal_set_features(sbi->s_journal,
- JBD2_FEATURE_COMPAT_CHECKSUM, 0, 0);
- if (test_opt(sb, JOURNAL_ASYNC_COMMIT))
- jbd2_journal_set_features(sbi->s_journal, 0, 0,
+ if (test_opt(sb, JOURNAL_ASYNC_COMMIT)) {
+ jbd2_journal_set_features(sbi->s_journal,
+ JBD2_FEATURE_COMPAT_CHECKSUM, 0,
JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
- else
+ } else if (test_opt(sb, JOURNAL_CHECKSUM)) {
+ jbd2_journal_set_features(sbi->s_journal,
+ JBD2_FEATURE_COMPAT_CHECKSUM, 0, 0);
jbd2_journal_clear_features(sbi->s_journal, 0, 0,
JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
+ } else {
+ jbd2_journal_clear_features(sbi->s_journal,
+ JBD2_FEATURE_COMPAT_CHECKSUM, 0,
+ JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT);
+ }
/* We have now updated the journal if required, so we can
* validate the data journaling mode. */
diff --git a/fs/fcntl.c b/fs/fcntl.c
index fc089f2f7f5..2cf93ec40a6 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -284,7 +284,7 @@ static int f_setown_ex(struct file *filp, unsigned long arg)
type = PIDTYPE_PID;
break;
- case F_OWNER_GID:
+ case F_OWNER_PGRP:
type = PIDTYPE_PGID;
break;
@@ -321,7 +321,7 @@ static int f_getown_ex(struct file *filp, unsigned long arg)
break;
case PIDTYPE_PGID:
- owner.type = F_OWNER_GID;
+ owner.type = F_OWNER_PGRP;
break;
default:
diff --git a/fs/file.c b/fs/file.c
index f313314f996..87e129030ab 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -10,6 +10,7 @@
#include <linux/fs.h>
#include <linux/mm.h>
#include <linux/time.h>
+#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/vmalloc.h>
#include <linux/file.h>
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 992f6c9410b..8ada78aade5 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -712,8 +712,10 @@ static int fuse_rename(struct inode *olddir, struct dentry *oldent,
fuse_invalidate_attr(newdir);
/* newent will end up negative */
- if (newent->d_inode)
+ if (newent->d_inode) {
+ fuse_invalidate_attr(newent->d_inode);
fuse_invalidate_entry_cache(newent);
+ }
} else if (err == -EINTR) {
/* If request was interrupted, DEITY only knows if the
rename actually took place. If the invalidation
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index a3492f7d207..c18913a777a 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -1063,7 +1063,8 @@ ssize_t fuse_direct_io(struct file *file, const char __user *buf,
break;
}
}
- fuse_put_request(fc, req);
+ if (!IS_ERR(req))
+ fuse_put_request(fc, req);
if (res > 0)
*ppos = pos;
@@ -1599,7 +1600,7 @@ static int fuse_ioctl_copy_user(struct page **pages, struct iovec *iov,
kaddr += copy;
}
- kunmap(map);
+ kunmap(page);
}
return 0;
diff --git a/fs/hfs/btree.c b/fs/hfs/btree.c
index 9b9d6395bad..052f214ea6f 100644
--- a/fs/hfs/btree.c
+++ b/fs/hfs/btree.c
@@ -58,6 +58,11 @@ struct hfs_btree *hfs_btree_open(struct super_block *sb, u32 id, btree_keycmp ke
}
unlock_new_inode(tree->inode);
+ if (!HFS_I(tree->inode)->first_blocks) {
+ printk(KERN_ERR "hfs: invalid btree extent records (0 size).\n");
+ goto free_inode;
+ }
+
mapping = tree->inode->i_mapping;
page = read_mapping_page(mapping, 0, NULL);
if (IS_ERR(page))
diff --git a/fs/hfsplus/wrapper.c b/fs/hfsplus/wrapper.c
index 175d08eacc8..bed78ac8f6d 100644
--- a/fs/hfsplus/wrapper.c
+++ b/fs/hfsplus/wrapper.c
@@ -99,6 +99,10 @@ int hfsplus_read_wrapper(struct super_block *sb)
if (hfsplus_get_last_session(sb, &part_start, &part_size))
return -EINVAL;
+ if ((u64)part_start + part_size > 0x100000000ULL) {
+ pr_err("hfs: volumes larger than 2TB are not supported yet\n");
+ return -EINVAL;
+ }
while (1) {
bh = sb_bread512(sb, part_start + HFSPLUS_VOLHEAD_SECTOR, vhdr);
if (!bh)
diff --git a/fs/ioctl.c b/fs/ioctl.c
index 7b17a14396f..6c751106c2e 100644
--- a/fs/ioctl.c
+++ b/fs/ioctl.c
@@ -254,7 +254,7 @@ int __generic_block_fiemap(struct inode *inode,
u64 len, get_block_t *get_block)
{
struct buffer_head tmp;
- unsigned int start_blk;
+ unsigned long long start_blk;
long long length = 0, map_len = 0;
u64 logical = 0, phys = 0, size = 0;
u32 flags = FIEMAP_EXTENT_MERGED;
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index bd3c073b485..4160afad6d0 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -73,6 +73,7 @@ EXPORT_SYMBOL(journal_errno);
EXPORT_SYMBOL(journal_ack_err);
EXPORT_SYMBOL(journal_clear_err);
EXPORT_SYMBOL(log_wait_commit);
+EXPORT_SYMBOL(log_start_commit);
EXPORT_SYMBOL(journal_start_commit);
EXPORT_SYMBOL(journal_force_commit_nested);
EXPORT_SYMBOL(journal_wipe);
@@ -756,6 +757,7 @@ journal_t * journal_init_dev(struct block_device *bdev,
return journal;
out_err:
+ kfree(journal->j_wbuf);
kfree(journal);
return NULL;
}
@@ -820,6 +822,7 @@ journal_t * journal_init_inode (struct inode *inode)
return journal;
out_err:
+ kfree(journal->j_wbuf);
kfree(journal);
return NULL;
}
diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c
index b0ab5219bec..fed85388ee8 100644
--- a/fs/jbd2/journal.c
+++ b/fs/jbd2/journal.c
@@ -913,6 +913,7 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev,
return journal;
out_err:
+ kfree(journal->j_wbuf);
jbd2_stats_proc_exit(journal);
kfree(journal);
return NULL;
@@ -986,6 +987,7 @@ journal_t * jbd2_journal_init_inode (struct inode *inode)
return journal;
out_err:
+ kfree(journal->j_wbuf);
jbd2_stats_proc_exit(journal);
kfree(journal);
return NULL;
diff --git a/fs/nfs/client.c b/fs/nfs/client.c
index 63976c0ccc2..99ea196f071 100644
--- a/fs/nfs/client.c
+++ b/fs/nfs/client.c
@@ -1180,7 +1180,7 @@ static int nfs4_init_client(struct nfs_client *clp,
1, flags & NFS_MOUNT_NORESVPORT);
if (error < 0)
goto error;
- memcpy(clp->cl_ipaddr, ip_addr, sizeof(clp->cl_ipaddr));
+ strlcpy(clp->cl_ipaddr, ip_addr, sizeof(clp->cl_ipaddr));
error = nfs_idmap_new(clp);
if (error < 0) {
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 32062c33c85..7cb298525ee 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1536,6 +1536,8 @@ nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry)
old_dentry->d_parent->d_name.name, old_dentry->d_name.name,
dentry->d_parent->d_name.name, dentry->d_name.name);
+ nfs_inode_return_delegation(inode);
+
d_drop(dentry);
error = NFS_PROTO(dir)->link(inode, dir, &dentry->d_name);
if (error == 0) {
diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c
index 6c3210099d5..e1d415e9784 100644
--- a/fs/nfs/direct.c
+++ b/fs/nfs/direct.c
@@ -457,6 +457,7 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq)
};
struct rpc_task_setup task_setup_data = {
.rpc_client = NFS_CLIENT(inode),
+ .rpc_message = &msg,
.callback_ops = &nfs_write_direct_ops,
.workqueue = nfsiod_workqueue,
.flags = RPC_TASK_ASYNC,
diff --git a/fs/nfs/nfs4namespace.c b/fs/nfs/nfs4namespace.c
index 2636c26d56f..fa3408f2011 100644
--- a/fs/nfs/nfs4namespace.c
+++ b/fs/nfs/nfs4namespace.c
@@ -121,7 +121,7 @@ static struct vfsmount *try_location(struct nfs_clone_mount *mountdata,
mnt_path = nfs4_pathname_string(&location->rootpath, page2, PAGE_SIZE);
if (IS_ERR(mnt_path))
- return mnt;
+ return ERR_CAST(mnt_path);
mountdata->mnt_path = mnt_path;
maxbuflen = mnt_path - 1 - page2;
@@ -132,15 +132,15 @@ static struct vfsmount *try_location(struct nfs_clone_mount *mountdata,
if (buf->len <= 0 || buf->len >= maxbuflen)
continue;
- mountdata->addr = (struct sockaddr *)&addr;
-
if (memchr(buf->data, IPV6_SCOPE_DELIMITER, buf->len))
continue;
- mountdata->addrlen = nfs_parse_server_name(buf->data,
- buf->len,
- mountdata->addr, mountdata->addrlen);
+
+ mountdata->addrlen = nfs_parse_server_name(buf->data, buf->len,
+ (struct sockaddr *)&addr, sizeof(addr));
if (mountdata->addrlen == 0)
continue;
+
+ mountdata->addr = (struct sockaddr *)&addr;
rpc_set_port(mountdata->addr, NFS_PORT);
memcpy(page2, buf->data, buf->len);
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index ed7c269e251..741a562177f 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -72,12 +72,17 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle,
/* Prevent leaks of NFSv4 errors into userland */
static int nfs4_map_errors(int err)
{
- if (err < -1000) {
+ if (err >= -1000)
+ return err;
+ switch (err) {
+ case -NFS4ERR_RESOURCE:
+ return -EREMOTEIO;
+ default:
dprintk("%s could not handle NFSv4 error %d\n",
__func__, -err);
- return -EIO;
+ break;
}
- return err;
+ return -EIO;
}
/*
@@ -2762,7 +2767,7 @@ static int _nfs4_proc_readdir(struct dentry *dentry, struct rpc_cred *cred,
.pages = &page,
.pgbase = 0,
.count = count,
- .bitmask = NFS_SERVER(dentry->d_inode)->cache_consistency_bitmask,
+ .bitmask = NFS_SERVER(dentry->d_inode)->attr_bitmask,
};
struct nfs4_readdir_res res;
struct rpc_message msg = {
@@ -3060,9 +3065,6 @@ static void nfs4_renew_done(struct rpc_task *task, void *data)
if (time_before(clp->cl_last_renewal,timestamp))
clp->cl_last_renewal = timestamp;
spin_unlock(&clp->cl_lock);
- dprintk("%s calling put_rpccred on rpc_cred %p\n", __func__,
- task->tk_msg.rpc_cred);
- put_rpccred(task->tk_msg.rpc_cred);
}
static const struct rpc_call_ops nfs4_renew_ops = {
@@ -4877,7 +4879,6 @@ void nfs41_sequence_call_done(struct rpc_task *task, void *data)
nfs41_sequence_free_slot(clp, task->tk_msg.rpc_resp);
dprintk("%s rpc_cred %p\n", __func__, task->tk_msg.rpc_cred);
- put_rpccred(task->tk_msg.rpc_cred);
kfree(task->tk_msg.rpc_argp);
kfree(task->tk_msg.rpc_resp);
diff --git a/fs/nfs/nfs4renewd.c b/fs/nfs/nfs4renewd.c
index e27c6cef18f..0156c01c212 100644
--- a/fs/nfs/nfs4renewd.c
+++ b/fs/nfs/nfs4renewd.c
@@ -127,12 +127,6 @@ nfs4_schedule_state_renewal(struct nfs_client *clp)
}
void
-nfs4_renewd_prepare_shutdown(struct nfs_server *server)
-{
- cancel_delayed_work(&server->nfs_client->cl_renewd);
-}
-
-void
nfs4_kill_renewd(struct nfs_client *clp)
{
cancel_delayed_work_sync(&clp->cl_renewd);
diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c
index 83ad47cbdd8..20b4e30e6c8 100644
--- a/fs/nfs/nfs4xdr.c
+++ b/fs/nfs/nfs4xdr.c
@@ -5681,7 +5681,6 @@ static struct {
{ NFS4ERR_SERVERFAULT, -ESERVERFAULT },
{ NFS4ERR_BADTYPE, -EBADTYPE },
{ NFS4ERR_LOCKED, -EAGAIN },
- { NFS4ERR_RESOURCE, -EREMOTEIO },
{ NFS4ERR_SYMLINK, -ELOOP },
{ NFS4ERR_OP_ILLEGAL, -EOPNOTSUPP },
{ NFS4ERR_DEADLOCK, -EDEADLK },
diff --git a/fs/nfs/super.c b/fs/nfs/super.c
index 29786d3b932..90be551b80c 100644
--- a/fs/nfs/super.c
+++ b/fs/nfs/super.c
@@ -728,22 +728,24 @@ static void nfs_umount_begin(struct super_block *sb)
unlock_kernel();
}
-static struct nfs_parsed_mount_data *nfs_alloc_parsed_mount_data(int flags)
+static struct nfs_parsed_mount_data *nfs_alloc_parsed_mount_data(unsigned int version)
{
struct nfs_parsed_mount_data *data;
data = kzalloc(sizeof(*data), GFP_KERNEL);
if (data) {
- data->flags = flags;
data->rsize = NFS_MAX_FILE_IO_SIZE;
data->wsize = NFS_MAX_FILE_IO_SIZE;
data->acregmin = NFS_DEF_ACREGMIN;
data->acregmax = NFS_DEF_ACREGMAX;
data->acdirmin = NFS_DEF_ACDIRMIN;
data->acdirmax = NFS_DEF_ACDIRMAX;
+ data->mount_server.port = NFS_UNSPEC_PORT;
data->nfs_server.port = NFS_UNSPEC_PORT;
+ data->nfs_server.protocol = XPRT_TRANSPORT_TCP;
data->auth_flavors[0] = RPC_AUTH_UNIX;
data->auth_flavor_len = 1;
+ data->version = version;
data->minorversion = 0;
}
return data;
@@ -776,15 +778,13 @@ static int nfs_verify_server_address(struct sockaddr *addr)
* Select between a default port value and a user-specified port value.
* If a zero value is set, then autobind will be used.
*/
-static void nfs_set_default_port(struct sockaddr *sap, const int parsed_port,
+static void nfs_set_port(struct sockaddr *sap, int *port,
const unsigned short default_port)
{
- unsigned short port = default_port;
+ if (*port == NFS_UNSPEC_PORT)
+ *port = default_port;
- if (parsed_port != NFS_UNSPEC_PORT)
- port = parsed_port;
-
- rpc_set_port(sap, port);
+ rpc_set_port(sap, *port);
}
/*
@@ -1253,6 +1253,7 @@ static int nfs_parse_mount_options(char *raw,
default:
dfprintk(MOUNT, "NFS: unrecognized "
"transport protocol\n");
+ kfree(string);
return 0;
}
break;
@@ -1475,7 +1476,7 @@ static int nfs_try_mount(struct nfs_parsed_mount_data *args,
args->mount_server.addrlen = args->nfs_server.addrlen;
}
request.salen = args->mount_server.addrlen;
- nfs_set_default_port(request.sap, args->mount_server.port, 0);
+ nfs_set_port(request.sap, &args->mount_server.port, 0);
/*
* Now ask the mount server to map our export path
@@ -1711,8 +1712,6 @@ static int nfs_validate_mount_data(void *options,
if (!(data->flags & NFS_MOUNT_TCP))
args->nfs_server.protocol = XPRT_TRANSPORT_UDP;
- else
- args->nfs_server.protocol = XPRT_TRANSPORT_TCP;
/* N.B. caller will free nfs_server.hostname in all cases */
args->nfs_server.hostname = kstrdup(data->hostname, GFP_KERNEL);
args->namlen = data->namlen;
@@ -1767,7 +1766,7 @@ static int nfs_validate_mount_data(void *options,
goto out_v4_not_compiled;
#endif
- nfs_set_default_port(sap, args->nfs_server.port, 0);
+ nfs_set_port(sap, &args->nfs_server.port, 0);
nfs_set_mount_transport_protocol(args);
@@ -1848,9 +1847,10 @@ nfs_compare_remount_data(struct nfs_server *nfss,
data->acdirmin != nfss->acdirmin / HZ ||
data->acdirmax != nfss->acdirmax / HZ ||
data->timeo != (10U * nfss->client->cl_timeout->to_initval / HZ) ||
+ data->nfs_server.port != nfss->port ||
data->nfs_server.addrlen != nfss->nfs_client->cl_addrlen ||
- memcmp(&data->nfs_server.address, &nfss->nfs_client->cl_addr,
- data->nfs_server.addrlen) != 0)
+ !rpc_cmp_addr((struct sockaddr *)&data->nfs_server.address,
+ (struct sockaddr *)&nfss->nfs_client->cl_addr))
return -EINVAL;
return 0;
@@ -1893,6 +1893,7 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data)
data->acdirmin = nfss->acdirmin / HZ;
data->acdirmax = nfss->acdirmax / HZ;
data->timeo = 10U * nfss->client->cl_timeout->to_initval / HZ;
+ data->nfs_server.port = nfss->port;
data->nfs_server.addrlen = nfss->nfs_client->cl_addrlen;
memcpy(&data->nfs_server.address, &nfss->nfs_client->cl_addr,
data->nfs_server.addrlen);
@@ -2106,7 +2107,7 @@ static int nfs_get_sb(struct file_system_type *fs_type,
};
int error = -ENOMEM;
- data = nfs_alloc_parsed_mount_data(NFS_MOUNT_VER3 | NFS_MOUNT_TCP);
+ data = nfs_alloc_parsed_mount_data(3);
mntfh = kzalloc(sizeof(*mntfh), GFP_KERNEL);
if (data == NULL || mntfh == NULL)
goto out_free_fh;
@@ -2331,7 +2332,7 @@ static int nfs4_validate_text_mount_data(void *options,
{
struct sockaddr *sap = (struct sockaddr *)&args->nfs_server.address;
- nfs_set_default_port(sap, args->nfs_server.port, NFS_PORT);
+ nfs_set_port(sap, &args->nfs_server.port, NFS_PORT);
nfs_validate_transport_protocol(args);
@@ -2376,7 +2377,6 @@ static int nfs4_validate_mount_data(void *options,
if (data == NULL)
goto out_no_data;
- args->version = 4;
switch (data->version) {
case 1:
if (data->host_addrlen > sizeof(args->nfs_server.address))
@@ -2660,7 +2660,7 @@ static int nfs4_get_sb(struct file_system_type *fs_type,
struct nfs_parsed_mount_data *data;
int error = -ENOMEM;
- data = nfs_alloc_parsed_mount_data(0);
+ data = nfs_alloc_parsed_mount_data(4);
if (data == NULL)
goto out_free_data;
@@ -2690,7 +2690,6 @@ static void nfs4_kill_super(struct super_block *sb)
dprintk("--> %s\n", __func__);
nfs_super_return_all_delegations(sb);
kill_anon_super(sb);
- nfs4_renewd_prepare_shutdown(server);
nfs_fscache_release_super_cookie(sb);
nfs_free_server(server);
dprintk("<-- %s\n", __func__);
diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c
index edf926e1062..d0a2ce1b432 100644
--- a/fs/nfsd/nfs3xdr.c
+++ b/fs/nfsd/nfs3xdr.c
@@ -958,7 +958,7 @@ encode_entry(struct readdir_cd *ccd, const char *name, int namlen,
p1 = encode_entry_baggage(cd, p1, name, namlen, ino);
if (plus)
- p = encode_entryplus_baggage(cd, p1, name, namlen);
+ p1 = encode_entryplus_baggage(cd, p1, name, namlen);
/* determine entry word length and lengths to go in pages */
num_entry_words = p1 - tmp;
diff --git a/fs/nilfs2/btnode.c b/fs/nilfs2/btnode.c
index 5941958f1e4..84c25382f8e 100644
--- a/fs/nilfs2/btnode.c
+++ b/fs/nilfs2/btnode.c
@@ -87,6 +87,7 @@ int nilfs_btnode_submit_block(struct address_space *btnc, __u64 blocknr,
brelse(bh);
BUG();
}
+ memset(bh->b_data, 0, 1 << inode->i_blkbits);
bh->b_bdev = NILFS_I_NILFS(inode)->ns_bdev;
bh->b_blocknr = blocknr;
set_buffer_mapped(bh);
@@ -276,8 +277,7 @@ void nilfs_btnode_commit_change_key(struct address_space *btnc,
"invalid oldkey %lld (newkey=%lld)",
(unsigned long long)oldkey,
(unsigned long long)newkey);
- if (!test_set_buffer_dirty(obh) && TestSetPageDirty(opage))
- BUG();
+ nilfs_btnode_mark_dirty(obh);
spin_lock_irq(&btnc->tree_lock);
radix_tree_delete(&btnc->page_tree, oldkey);
diff --git a/fs/nilfs2/cpfile.c b/fs/nilfs2/cpfile.c
index 1c6cfb59128..3f5d5d06f53 100644
--- a/fs/nilfs2/cpfile.c
+++ b/fs/nilfs2/cpfile.c
@@ -871,7 +871,6 @@ int nilfs_cpfile_change_cpmode(struct inode *cpfile, __u64 cno, int mode)
* exclusive with a new mount job. Though it doesn't cover
* umount, it's enough for the purpose.
*/
- mutex_lock(&nilfs->ns_mount_mutex);
if (nilfs_checkpoint_is_mounted(nilfs, cno, 1)) {
/* Current implementation does not have to protect
plain read-only mounts since they are exclusive
@@ -880,7 +879,6 @@ int nilfs_cpfile_change_cpmode(struct inode *cpfile, __u64 cno, int mode)
ret = -EBUSY;
} else
ret = nilfs_cpfile_clear_snapshot(cpfile, cno);
- mutex_unlock(&nilfs->ns_mount_mutex);
return ret;
case NILFS_SNAPSHOT:
return nilfs_cpfile_set_snapshot(cpfile, cno);
diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c
index 5040220c373..2a0a5a3ac13 100644
--- a/fs/nilfs2/inode.c
+++ b/fs/nilfs2/inode.c
@@ -664,7 +664,6 @@ int nilfs_load_inode_block(struct nilfs_sb_info *sbi, struct inode *inode,
int err;
spin_lock(&sbi->s_inode_lock);
- /* Caller of this function MUST lock s_inode_lock */
if (ii->i_bh == NULL) {
spin_unlock(&sbi->s_inode_lock);
err = nilfs_ifile_get_inode_block(sbi->s_ifile, inode->i_ino,
diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c
index 6572ea4bc4d..f6af76042d8 100644
--- a/fs/nilfs2/ioctl.c
+++ b/fs/nilfs2/ioctl.c
@@ -99,7 +99,8 @@ static int nilfs_ioctl_wrap_copy(struct the_nilfs *nilfs,
static int nilfs_ioctl_change_cpmode(struct inode *inode, struct file *filp,
unsigned int cmd, void __user *argp)
{
- struct inode *cpfile = NILFS_SB(inode->i_sb)->s_nilfs->ns_cpfile;
+ struct the_nilfs *nilfs = NILFS_SB(inode->i_sb)->s_nilfs;
+ struct inode *cpfile = nilfs->ns_cpfile;
struct nilfs_transaction_info ti;
struct nilfs_cpmode cpmode;
int ret;
@@ -109,14 +110,17 @@ static int nilfs_ioctl_change_cpmode(struct inode *inode, struct file *filp,
if (copy_from_user(&cpmode, argp, sizeof(cpmode)))
return -EFAULT;
+ mutex_lock(&nilfs->ns_mount_mutex);
nilfs_transaction_begin(inode->i_sb, &ti, 0);
ret = nilfs_cpfile_change_cpmode(
cpfile, cpmode.cm_cno, cpmode.cm_mode);
if (unlikely(ret < 0)) {
nilfs_transaction_abort(inode->i_sb);
+ mutex_unlock(&nilfs->ns_mount_mutex);
return ret;
}
nilfs_transaction_commit(inode->i_sb); /* never fails */
+ mutex_unlock(&nilfs->ns_mount_mutex);
return ret;
}
@@ -297,7 +301,18 @@ static int nilfs_ioctl_move_inode_block(struct inode *inode,
(unsigned long long)vdesc->vd_vblocknr);
return ret;
}
- bh->b_private = vdesc;
+ if (unlikely(!list_empty(&bh->b_assoc_buffers))) {
+ printk(KERN_CRIT "%s: conflicting %s buffer: ino=%llu, "
+ "cno=%llu, offset=%llu, blocknr=%llu, vblocknr=%llu\n",
+ __func__, vdesc->vd_flags ? "node" : "data",
+ (unsigned long long)vdesc->vd_ino,
+ (unsigned long long)vdesc->vd_cno,
+ (unsigned long long)vdesc->vd_offset,
+ (unsigned long long)vdesc->vd_blocknr,
+ (unsigned long long)vdesc->vd_vblocknr);
+ brelse(bh);
+ return -EEXIST;
+ }
list_add_tail(&bh->b_assoc_buffers, buffers);
return 0;
}
@@ -335,24 +350,10 @@ static int nilfs_ioctl_move_blocks(struct the_nilfs *nilfs,
list_for_each_entry_safe(bh, n, &buffers, b_assoc_buffers) {
ret = nilfs_gccache_wait_and_mark_dirty(bh);
if (unlikely(ret < 0)) {
- if (ret == -EEXIST) {
- vdesc = bh->b_private;
- printk(KERN_CRIT
- "%s: conflicting %s buffer: "
- "ino=%llu, cno=%llu, offset=%llu, "
- "blocknr=%llu, vblocknr=%llu\n",
- __func__,
- vdesc->vd_flags ? "node" : "data",
- (unsigned long long)vdesc->vd_ino,
- (unsigned long long)vdesc->vd_cno,
- (unsigned long long)vdesc->vd_offset,
- (unsigned long long)vdesc->vd_blocknr,
- (unsigned long long)vdesc->vd_vblocknr);
- }
+ WARN_ON(ret == -EEXIST);
goto failed;
}
list_del_init(&bh->b_assoc_buffers);
- bh->b_private = NULL;
brelse(bh);
}
return nmembs;
@@ -360,7 +361,6 @@ static int nilfs_ioctl_move_blocks(struct the_nilfs *nilfs,
failed:
list_for_each_entry_safe(bh, n, &buffers, b_assoc_buffers) {
list_del_init(&bh->b_assoc_buffers);
- bh->b_private = NULL;
brelse(bh);
}
return ret;
@@ -471,7 +471,6 @@ int nilfs_ioctl_prepare_clean_segments(struct the_nilfs *nilfs,
return 0;
failed:
- nilfs_remove_all_gcinode(nilfs);
printk(KERN_ERR "NILFS: GC failed during preparation: %s: err=%d\n",
msg, ret);
return ret;
@@ -560,6 +559,8 @@ static int nilfs_ioctl_clean_segments(struct inode *inode, struct file *filp,
else
ret = nilfs_clean_segments(inode->i_sb, argv, kbufs);
+ if (ret < 0)
+ nilfs_remove_all_gcinode(nilfs);
clear_nilfs_gc_running(nilfs);
out_free:
diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c
index 683df89dbae..6eff66a070d 100644
--- a/fs/nilfs2/segment.c
+++ b/fs/nilfs2/segment.c
@@ -2468,17 +2468,22 @@ static void nilfs_segctor_notify(struct nilfs_sc_info *sci,
/* Clear requests (even when the construction failed) */
spin_lock(&sci->sc_state_lock);
- sci->sc_state &= ~NILFS_SEGCTOR_COMMIT;
-
if (req->mode == SC_LSEG_SR) {
+ sci->sc_state &= ~NILFS_SEGCTOR_COMMIT;
sci->sc_seq_done = req->seq_accepted;
nilfs_segctor_wakeup(sci, req->sc_err ? : req->sb_err);
sci->sc_flush_request = 0;
- } else if (req->mode == SC_FLUSH_FILE)
- sci->sc_flush_request &= ~FLUSH_FILE_BIT;
- else if (req->mode == SC_FLUSH_DAT)
- sci->sc_flush_request &= ~FLUSH_DAT_BIT;
+ } else {
+ if (req->mode == SC_FLUSH_FILE)
+ sci->sc_flush_request &= ~FLUSH_FILE_BIT;
+ else if (req->mode == SC_FLUSH_DAT)
+ sci->sc_flush_request &= ~FLUSH_DAT_BIT;
+ /* re-enable timer if checkpoint creation was not done */
+ if (sci->sc_timer && (sci->sc_state & NILFS_SEGCTOR_COMMIT) &&
+ time_before(jiffies, sci->sc_timer->expires))
+ add_timer(sci->sc_timer);
+ }
spin_unlock(&sci->sc_state_lock);
}
diff --git a/fs/notify/dnotify/dnotify.c b/fs/notify/dnotify/dnotify.c
index 828a889be90..7e54e52964d 100644
--- a/fs/notify/dnotify/dnotify.c
+++ b/fs/notify/dnotify/dnotify.c
@@ -91,6 +91,7 @@ static int dnotify_handle_event(struct fsnotify_group *group,
struct dnotify_struct *dn;
struct dnotify_struct **prev;
struct fown_struct *fown;
+ __u32 test_mask = event->mask & ~FS_EVENT_ON_CHILD;
to_tell = event->to_tell;
@@ -106,7 +107,7 @@ static int dnotify_handle_event(struct fsnotify_group *group,
spin_lock(&entry->lock);
prev = &dnentry->dn;
while ((dn = *prev) != NULL) {
- if ((dn->dn_mask & event->mask) == 0) {
+ if ((dn->dn_mask & test_mask) == 0) {
prev = &dn->dn_next;
continue;
}
diff --git a/fs/notify/inode_mark.c b/fs/notify/inode_mark.c
index c8a07c65482..3165d85aada 100644
--- a/fs/notify/inode_mark.c
+++ b/fs/notify/inode_mark.c
@@ -324,11 +324,11 @@ int fsnotify_add_mark(struct fsnotify_mark_entry *entry,
spin_lock(&group->mark_lock);
spin_lock(&inode->i_lock);
- entry->group = group;
- entry->inode = inode;
-
lentry = fsnotify_find_mark_entry(group, inode);
if (!lentry) {
+ entry->group = group;
+ entry->inode = inode;
+
hlist_add_head(&entry->i_list, &inode->i_fsnotify_mark_entries);
list_add(&entry->g_list, &group->mark_entries);
diff --git a/fs/notify/notification.c b/fs/notify/notification.c
index 3816d5750dd..b8bf53b4c10 100644
--- a/fs/notify/notification.c
+++ b/fs/notify/notification.c
@@ -143,7 +143,7 @@ static bool event_compare(struct fsnotify_event *old, struct fsnotify_event *new
/* remember, after old was put on the wait_q we aren't
* allowed to look at the inode any more, only thing
* left to check was if the file_name is the same */
- if (old->name_len &&
+ if (!old->name_len ||
!strcmp(old->file_name, new->file_name))
return true;
break;
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index f38fee0311a..7b685e10cba 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -248,11 +248,19 @@ ssize_t part_stat_show(struct device *dev,
part_stat_read(p, merges[WRITE]),
(unsigned long long)part_stat_read(p, sectors[WRITE]),
jiffies_to_msecs(part_stat_read(p, ticks[WRITE])),
- p->in_flight,
+ part_in_flight(p),
jiffies_to_msecs(part_stat_read(p, io_ticks)),
jiffies_to_msecs(part_stat_read(p, time_in_queue)));
}
+ssize_t part_inflight_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct hd_struct *p = dev_to_part(dev);
+
+ return sprintf(buf, "%8u %8u\n", p->in_flight[0], p->in_flight[1]);
+}
+
#ifdef CONFIG_FAIL_MAKE_REQUEST
ssize_t part_fail_show(struct device *dev,
struct device_attribute *attr, char *buf)
@@ -281,6 +289,7 @@ static DEVICE_ATTR(start, S_IRUGO, part_start_show, NULL);
static DEVICE_ATTR(size, S_IRUGO, part_size_show, NULL);
static DEVICE_ATTR(alignment_offset, S_IRUGO, part_alignment_offset_show, NULL);
static DEVICE_ATTR(stat, S_IRUGO, part_stat_show, NULL);
+static DEVICE_ATTR(inflight, S_IRUGO, part_inflight_show, NULL);
#ifdef CONFIG_FAIL_MAKE_REQUEST
static struct device_attribute dev_attr_fail =
__ATTR(make-it-fail, S_IRUGO|S_IWUSR, part_fail_show, part_fail_store);
@@ -292,6 +301,7 @@ static struct attribute *part_attrs[] = {
&dev_attr_size.attr,
&dev_attr_alignment_offset.attr,
&dev_attr_stat.attr,
+ &dev_attr_inflight.attr,
#ifdef CONFIG_FAIL_MAKE_REQUEST
&dev_attr_fail.attr,
#endif
diff --git a/fs/pipe.c b/fs/pipe.c
index 52c41511483..ae17d026aaa 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -777,36 +777,55 @@ pipe_rdwr_release(struct inode *inode, struct file *filp)
static int
pipe_read_open(struct inode *inode, struct file *filp)
{
- /* We could have perhaps used atomic_t, but this and friends
- below are the only places. So it doesn't seem worthwhile. */
+ int ret = -ENOENT;
+
mutex_lock(&inode->i_mutex);
- inode->i_pipe->readers++;
+
+ if (inode->i_pipe) {
+ ret = 0;
+ inode->i_pipe->readers++;
+ }
+
mutex_unlock(&inode->i_mutex);
- return 0;
+ return ret;
}
static int
pipe_write_open(struct inode *inode, struct file *filp)
{
+ int ret = -ENOENT;
+
mutex_lock(&inode->i_mutex);
- inode->i_pipe->writers++;
+
+ if (inode->i_pipe) {
+ ret = 0;
+ inode->i_pipe->writers++;
+ }
+
mutex_unlock(&inode->i_mutex);
- return 0;
+ return ret;
}
static int
pipe_rdwr_open(struct inode *inode, struct file *filp)
{
+ int ret = -ENOENT;
+
mutex_lock(&inode->i_mutex);
- if (filp->f_mode & FMODE_READ)
- inode->i_pipe->readers++;
- if (filp->f_mode & FMODE_WRITE)
- inode->i_pipe->writers++;
+
+ if (inode->i_pipe) {
+ ret = 0;
+ if (filp->f_mode & FMODE_READ)
+ inode->i_pipe->readers++;
+ if (filp->f_mode & FMODE_WRITE)
+ inode->i_pipe->writers++;
+ }
+
mutex_unlock(&inode->i_mutex);
- return 0;
+ return ret;
}
/*
diff --git a/fs/proc/array.c b/fs/proc/array.c
index 07f77a7945c..822c2d50651 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -571,7 +571,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns,
rsslim,
mm ? mm->start_code : 0,
mm ? mm->end_code : 0,
- (permitted) ? task->stack_start : 0,
+ (permitted && mm) ? task->stack_start : 0,
esp,
eip,
/* The signal information here is obsolete.
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 837469a9659..af643b5aefe 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2597,8 +2597,7 @@ static void proc_flush_task_mnt(struct vfsmount *mnt, pid_t pid, pid_t tgid)
name.len = snprintf(buf, sizeof(buf), "%d", pid);
dentry = d_hash_and_lookup(mnt->mnt_root, &name);
if (dentry) {
- if (!(current->flags & PF_EXITING))
- shrink_dcache_parent(dentry);
+ shrink_dcache_parent(dentry);
d_drop(dentry);
dput(dentry);
}
diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c
index 56013371f9f..a44a7897fd4 100644
--- a/fs/proc/kcore.c
+++ b/fs/proc/kcore.c
@@ -23,7 +23,6 @@
#include <asm/io.h>
#include <linux/list.h>
#include <linux/ioport.h>
-#include <linux/mm.h>
#include <linux/memory.h>
#include <asm/sections.h>
diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
index c7bff4f603f..a65239cfd97 100644
--- a/fs/proc/meminfo.c
+++ b/fs/proc/meminfo.c
@@ -99,7 +99,7 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
"VmallocUsed: %8lu kB\n"
"VmallocChunk: %8lu kB\n"
#ifdef CONFIG_MEMORY_FAILURE
- "HardwareCorrupted: %8lu kB\n"
+ "HardwareCorrupted: %5lu kB\n"
#endif
,
K(i.totalram),
diff --git a/fs/proc/page.c b/fs/proc/page.c
index 2281c2cbfe2..5033ce0d254 100644
--- a/fs/proc/page.c
+++ b/fs/proc/page.c
@@ -94,6 +94,7 @@ static const struct file_operations proc_kpagecount_operations = {
#define KPF_COMPOUND_TAIL 16
#define KPF_HUGE 17
#define KPF_UNEVICTABLE 18
+#define KPF_HWPOISON 19
#define KPF_NOPAGE 20
#define KPF_KSM 21
@@ -180,6 +181,10 @@ static u64 get_uflags(struct page *page)
u |= kpf_copy_bit(k, KPF_UNEVICTABLE, PG_unevictable);
u |= kpf_copy_bit(k, KPF_MLOCKED, PG_mlocked);
+#ifdef CONFIG_MEMORY_FAILURE
+ u |= kpf_copy_bit(k, KPF_HWPOISON, PG_hwpoison);
+#endif
+
#ifdef CONFIG_IA64_UNCACHED_ALLOCATOR
u |= kpf_copy_bit(k, KPF_UNCACHED, PG_uncached);
#endif
diff --git a/fs/romfs/storage.c b/fs/romfs/storage.c
index b3208adf8e7..71e2b4d50a0 100644
--- a/fs/romfs/storage.c
+++ b/fs/romfs/storage.c
@@ -253,11 +253,11 @@ ssize_t romfs_dev_strnlen(struct super_block *sb,
#ifdef CONFIG_ROMFS_ON_MTD
if (sb->s_mtd)
- return romfs_mtd_strnlen(sb, pos, limit);
+ return romfs_mtd_strnlen(sb, pos, maxlen);
#endif
#ifdef CONFIG_ROMFS_ON_BLOCK
if (sb->s_bdev)
- return romfs_blk_strnlen(sb, pos, limit);
+ return romfs_blk_strnlen(sb, pos, maxlen);
#endif
return -EIO;
}
diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c
index 0050fc40e8c..e0201837d24 100644
--- a/fs/sysfs/dir.c
+++ b/fs/sysfs/dir.c
@@ -21,6 +21,7 @@
#include <linux/completion.h>
#include <linux/mutex.h>
#include <linux/slab.h>
+#include <linux/security.h>
#include "sysfs.h"
DEFINE_MUTEX(sysfs_mutex);
@@ -285,6 +286,9 @@ void release_sysfs_dirent(struct sysfs_dirent * sd)
sysfs_put(sd->s_symlink.target_sd);
if (sysfs_type(sd) & SYSFS_COPY_NAME)
kfree(sd->s_name);
+ if (sd->s_iattr && sd->s_iattr->ia_secdata)
+ security_release_secctx(sd->s_iattr->ia_secdata,
+ sd->s_iattr->ia_secdata_len);
kfree(sd->s_iattr);
sysfs_free_ino(sd->s_ino);
kmem_cache_free(sysfs_dir_cachep, sd);
@@ -894,7 +898,8 @@ int sysfs_move_dir(struct kobject *kobj, struct kobject *new_parent_kobj)
mutex_lock(&sysfs_rename_mutex);
BUG_ON(!sd->s_parent);
- new_parent_sd = new_parent_kobj->sd ? new_parent_kobj->sd : &sysfs_root;
+ new_parent_sd = (new_parent_kobj && new_parent_kobj->sd) ?
+ new_parent_kobj->sd : &sysfs_root;
error = 0;
if (sd->s_parent == new_parent_sd)
diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c
index 561a9c050ce..f5ea4680f15 100644
--- a/fs/sysfs/file.c
+++ b/fs/sysfs/file.c
@@ -268,7 +268,7 @@ static int sysfs_get_open_dirent(struct sysfs_dirent *sd,
struct sysfs_open_dirent *od, *new_od = NULL;
retry:
- spin_lock(&sysfs_open_dirent_lock);
+ spin_lock_irq(&sysfs_open_dirent_lock);
if (!sd->s_attr.open && new_od) {
sd->s_attr.open = new_od;
@@ -281,7 +281,7 @@ static int sysfs_get_open_dirent(struct sysfs_dirent *sd,
list_add_tail(&buffer->list, &od->buffers);
}
- spin_unlock(&sysfs_open_dirent_lock);
+ spin_unlock_irq(&sysfs_open_dirent_lock);
if (od) {
kfree(new_od);
@@ -315,8 +315,9 @@ static void sysfs_put_open_dirent(struct sysfs_dirent *sd,
struct sysfs_buffer *buffer)
{
struct sysfs_open_dirent *od = sd->s_attr.open;
+ unsigned long flags;
- spin_lock(&sysfs_open_dirent_lock);
+ spin_lock_irqsave(&sysfs_open_dirent_lock, flags);
list_del(&buffer->list);
if (atomic_dec_and_test(&od->refcnt))
@@ -324,7 +325,7 @@ static void sysfs_put_open_dirent(struct sysfs_dirent *sd,
else
od = NULL;
- spin_unlock(&sysfs_open_dirent_lock);
+ spin_unlock_irqrestore(&sysfs_open_dirent_lock, flags);
kfree(od);
}
@@ -456,8 +457,9 @@ static unsigned int sysfs_poll(struct file *filp, poll_table *wait)
void sysfs_notify_dirent(struct sysfs_dirent *sd)
{
struct sysfs_open_dirent *od;
+ unsigned long flags;
- spin_lock(&sysfs_open_dirent_lock);
+ spin_lock_irqsave(&sysfs_open_dirent_lock, flags);
od = sd->s_attr.open;
if (od) {
@@ -465,7 +467,7 @@ void sysfs_notify_dirent(struct sysfs_dirent *sd)
wake_up_interruptible(&od->poll);
}
- spin_unlock(&sysfs_open_dirent_lock);
+ spin_unlock_irqrestore(&sysfs_open_dirent_lock, flags);
}
EXPORT_SYMBOL_GPL(sysfs_notify_dirent);
diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index 381854461b2..c2e30eea74d 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -186,19 +186,37 @@ xfs_destroy_ioend(
}
/*
+ * If the end of the current ioend is beyond the current EOF,
+ * return the new EOF value, otherwise zero.
+ */
+STATIC xfs_fsize_t
+xfs_ioend_new_eof(
+ xfs_ioend_t *ioend)
+{
+ xfs_inode_t *ip = XFS_I(ioend->io_inode);
+ xfs_fsize_t isize;
+ xfs_fsize_t bsize;
+
+ bsize = ioend->io_offset + ioend->io_size;
+ isize = MAX(ip->i_size, ip->i_new_size);
+ isize = MIN(isize, bsize);
+ return isize > ip->i_d.di_size ? isize : 0;
+}
+
+/*
* Update on-disk file size now that data has been written to disk.
* The current in-memory file size is i_size. If a write is beyond
* eof i_new_size will be the intended file size until i_size is
* updated. If this write does not extend all the way to the valid
* file size then restrict this update to the end of the write.
*/
+
STATIC void
xfs_setfilesize(
xfs_ioend_t *ioend)
{
xfs_inode_t *ip = XFS_I(ioend->io_inode);
xfs_fsize_t isize;
- xfs_fsize_t bsize;
ASSERT((ip->i_d.di_mode & S_IFMT) == S_IFREG);
ASSERT(ioend->io_type != IOMAP_READ);
@@ -206,16 +224,10 @@ xfs_setfilesize(
if (unlikely(ioend->io_error))
return;
- bsize = ioend->io_offset + ioend->io_size;
-
xfs_ilock(ip, XFS_ILOCK_EXCL);
-
- isize = MAX(ip->i_size, ip->i_new_size);
- isize = MIN(isize, bsize);
-
- if (ip->i_d.di_size < isize) {
+ isize = xfs_ioend_new_eof(ioend);
+ if (isize) {
ip->i_d.di_size = isize;
- ip->i_update_core = 1;
xfs_mark_inode_dirty_sync(ip);
}
@@ -404,10 +416,16 @@ xfs_submit_ioend_bio(
struct bio *bio)
{
atomic_inc(&ioend->io_remaining);
-
bio->bi_private = ioend;
bio->bi_end_io = xfs_end_bio;
+ /*
+ * If the I/O is beyond EOF we mark the inode dirty immediately
+ * but don't update the inode size until I/O completion.
+ */
+ if (xfs_ioend_new_eof(ioend))
+ xfs_mark_inode_dirty_sync(XFS_I(ioend->io_inode));
+
submit_bio(WRITE, bio);
ASSERT(!bio_flagged(bio, BIO_EOPNOTSUPP));
bio_put(bio);
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index 629370974e5..eff61e2732a 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -176,14 +176,7 @@ xfs_file_fsync(
struct dentry *dentry,
int datasync)
{
- struct inode *inode = dentry->d_inode;
- struct xfs_inode *ip = XFS_I(inode);
- int error;
-
- /* capture size updates in I/O completion before writing the inode. */
- error = filemap_fdatawait(inode->i_mapping);
- if (error)
- return error;
+ struct xfs_inode *ip = XFS_I(dentry->d_inode);
xfs_iflags_clear(ip, XFS_ITRUNCATED);
return -xfs_fsync(ip);
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index da0159d99f8..cd42ef78f6b 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -57,19 +57,22 @@
#include <linux/fiemap.h>
/*
- * Bring the atime in the XFS inode uptodate.
- * Used before logging the inode to disk or when the Linux inode goes away.
+ * Bring the timestamps in the XFS inode uptodate.
+ *
+ * Used before writing the inode to disk.
*/
void
-xfs_synchronize_atime(
+xfs_synchronize_times(
xfs_inode_t *ip)
{
struct inode *inode = VFS_I(ip);
- if (!(inode->i_state & I_CLEAR)) {
- ip->i_d.di_atime.t_sec = (__int32_t)inode->i_atime.tv_sec;
- ip->i_d.di_atime.t_nsec = (__int32_t)inode->i_atime.tv_nsec;
- }
+ ip->i_d.di_atime.t_sec = (__int32_t)inode->i_atime.tv_sec;
+ ip->i_d.di_atime.t_nsec = (__int32_t)inode->i_atime.tv_nsec;
+ ip->i_d.di_ctime.t_sec = (__int32_t)inode->i_ctime.tv_sec;
+ ip->i_d.di_ctime.t_nsec = (__int32_t)inode->i_ctime.tv_nsec;
+ ip->i_d.di_mtime.t_sec = (__int32_t)inode->i_mtime.tv_sec;
+ ip->i_d.di_mtime.t_nsec = (__int32_t)inode->i_mtime.tv_nsec;
}
/*
@@ -106,32 +109,20 @@ xfs_ichgtime(
if ((flags & XFS_ICHGTIME_MOD) &&
!timespec_equal(&inode->i_mtime, &tv)) {
inode->i_mtime = tv;
- ip->i_d.di_mtime.t_sec = (__int32_t)tv.tv_sec;
- ip->i_d.di_mtime.t_nsec = (__int32_t)tv.tv_nsec;
sync_it = 1;
}
if ((flags & XFS_ICHGTIME_CHG) &&
!timespec_equal(&inode->i_ctime, &tv)) {
inode->i_ctime = tv;
- ip->i_d.di_ctime.t_sec = (__int32_t)tv.tv_sec;
- ip->i_d.di_ctime.t_nsec = (__int32_t)tv.tv_nsec;
sync_it = 1;
}
/*
- * We update the i_update_core field _after_ changing
- * the timestamps in order to coordinate properly with
- * xfs_iflush() so that we don't lose timestamp updates.
- * This keeps us from having to hold the inode lock
- * while doing this. We use the SYNCHRONIZE macro to
- * ensure that the compiler does not reorder the update
- * of i_update_core above the timestamp updates above.
+ * Update complete - now make sure everyone knows that the inode
+ * is dirty.
*/
- if (sync_it) {
- SYNCHRONIZE();
- ip->i_update_core = 1;
+ if (sync_it)
xfs_mark_inode_dirty_sync(ip);
- }
}
/*
@@ -506,10 +497,8 @@ xfs_vn_getattr(
stat->gid = ip->i_d.di_gid;
stat->ino = ip->i_ino;
stat->atime = inode->i_atime;
- stat->mtime.tv_sec = ip->i_d.di_mtime.t_sec;
- stat->mtime.tv_nsec = ip->i_d.di_mtime.t_nsec;
- stat->ctime.tv_sec = ip->i_d.di_ctime.t_sec;
- stat->ctime.tv_nsec = ip->i_d.di_ctime.t_nsec;
+ stat->mtime = inode->i_mtime;
+ stat->ctime = inode->i_ctime;
stat->blocks =
XFS_FSB_TO_BB(mp, ip->i_d.di_nblocks + ip->i_delayed_blks);
diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c
index 49e4a6aea73..072050f8d34 100644
--- a/fs/xfs/linux-2.6/xfs_lrw.c
+++ b/fs/xfs/linux-2.6/xfs_lrw.c
@@ -667,7 +667,7 @@ start:
xip->i_new_size = new_size;
if (likely(!(ioflags & IO_INVIS)))
- xfs_ichgtime(xip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
+ file_update_time(file);
/*
* If the offset is beyond the size of the file, we have a couple
diff --git a/fs/xfs/linux-2.6/xfs_quotaops.c b/fs/xfs/linux-2.6/xfs_quotaops.c
index 9e41f91aa26..3d4a0c84d63 100644
--- a/fs/xfs/linux-2.6/xfs_quotaops.c
+++ b/fs/xfs/linux-2.6/xfs_quotaops.c
@@ -80,7 +80,7 @@ xfs_fs_set_xstate(
if (sb->s_flags & MS_RDONLY)
return -EROFS;
- if (!XFS_IS_QUOTA_RUNNING(mp))
+ if (op != Q_XQUOTARM && !XFS_IS_QUOTA_RUNNING(mp))
return -ENOSYS;
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index bdd41c8c342..18a4b8e11df 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -977,6 +977,28 @@ xfs_fs_inode_init_once(
}
/*
+ * Dirty the XFS inode when mark_inode_dirty_sync() is called so that
+ * we catch unlogged VFS level updates to the inode. Care must be taken
+ * here - the transaction code calls mark_inode_dirty_sync() to mark the
+ * VFS inode dirty in a transaction and clears the i_update_core field;
+ * it must clear the field after calling mark_inode_dirty_sync() to
+ * correctly indicate that the dirty state has been propagated into the
+ * inode log item.
+ *
+ * We need the barrier() to maintain correct ordering between unlogged
+ * updates and the transaction commit code that clears the i_update_core
+ * field. This requires all updates to be completed before marking the
+ * inode dirty.
+ */
+STATIC void
+xfs_fs_dirty_inode(
+ struct inode *inode)
+{
+ barrier();
+ XFS_I(inode)->i_update_core = 1;
+}
+
+/*
* Attempt to flush the inode, this will actually fail
* if the inode is pinned, but we dirty the inode again
* at the point when it is unpinned after a log write,
@@ -1126,7 +1148,7 @@ xfs_fs_put_super(
}
STATIC int
-xfs_fs_sync_super(
+xfs_fs_sync_fs(
struct super_block *sb,
int wait)
{
@@ -1134,23 +1156,23 @@ xfs_fs_sync_super(
int error;
/*
- * Treat a sync operation like a freeze. This is to work
- * around a race in sync_inodes() which works in two phases
- * - an asynchronous flush, which can write out an inode
- * without waiting for file size updates to complete, and a
- * synchronous flush, which wont do anything because the
- * async flush removed the inode's dirty flag. Also
- * sync_inodes() will not see any files that just have
- * outstanding transactions to be flushed because we don't
- * dirty the Linux inode until after the transaction I/O
- * completes.
+ * Not much we can do for the first async pass. Writing out the
+ * superblock would be counter-productive as we are going to redirty
+ * when writing out other data and metadata (and writing out a single
+ * block is quite fast anyway).
+ *
+ * Try to asynchronously kick off quota syncing at least.
*/
- if (wait || unlikely(sb->s_frozen == SB_FREEZE_WRITE))
- error = xfs_quiesce_data(mp);
- else
- error = xfs_sync_fsdata(mp, 0);
+ if (!wait) {
+ xfs_qm_sync(mp, SYNC_TRYLOCK);
+ return 0;
+ }
+
+ error = xfs_quiesce_data(mp);
+ if (error)
+ return -error;
- if (unlikely(laptop_mode)) {
+ if (laptop_mode) {
int prev_sync_seq = mp->m_sync_seq;
/*
@@ -1169,7 +1191,7 @@ xfs_fs_sync_super(
mp->m_sync_seq != prev_sync_seq);
}
- return -error;
+ return 0;
}
STATIC int
@@ -1539,10 +1561,11 @@ xfs_fs_get_sb(
static const struct super_operations xfs_super_operations = {
.alloc_inode = xfs_fs_alloc_inode,
.destroy_inode = xfs_fs_destroy_inode,
+ .dirty_inode = xfs_fs_dirty_inode,
.write_inode = xfs_fs_write_inode,
.clear_inode = xfs_fs_clear_inode,
.put_super = xfs_fs_put_super,
- .sync_fs = xfs_fs_sync_super,
+ .sync_fs = xfs_fs_sync_fs,
.freeze_fs = xfs_fs_freeze,
.statfs = xfs_fs_statfs,
.remount_fs = xfs_fs_remount,
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index 320be6aea49..961df0a22c7 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -309,11 +309,15 @@ xfs_sync_attr(
STATIC int
xfs_commit_dummy_trans(
struct xfs_mount *mp,
- uint log_flags)
+ uint flags)
{
struct xfs_inode *ip = mp->m_rootip;
struct xfs_trans *tp;
int error;
+ int log_flags = XFS_LOG_FORCE;
+
+ if (flags & SYNC_WAIT)
+ log_flags |= XFS_LOG_SYNC;
/*
* Put a dummy transaction in the log to tell recovery
@@ -331,13 +335,12 @@ xfs_commit_dummy_trans(
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
- /* XXX(hch): ignoring the error here.. */
error = xfs_trans_commit(tp, 0);
-
xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ /* the log force ensures this transaction is pushed to disk */
xfs_log_force(mp, 0, log_flags);
- return 0;
+ return error;
}
int
@@ -385,7 +388,20 @@ xfs_sync_fsdata(
else
XFS_BUF_ASYNC(bp);
- return xfs_bwrite(mp, bp);
+ error = xfs_bwrite(mp, bp);
+ if (error)
+ return error;
+
+ /*
+ * If this is a data integrity sync make sure all pending buffers
+ * are flushed out for the log coverage check below.
+ */
+ if (flags & SYNC_WAIT)
+ xfs_flush_buftarg(mp->m_ddev_targp, 1);
+
+ if (xfs_log_need_covered(mp))
+ error = xfs_commit_dummy_trans(mp, flags);
+ return error;
out_brelse:
xfs_buf_relse(bp);
@@ -419,14 +435,16 @@ xfs_quiesce_data(
/* push non-blocking */
xfs_sync_data(mp, 0);
xfs_qm_sync(mp, SYNC_TRYLOCK);
- xfs_filestream_flush(mp);
- /* push and block */
+ /* push and block till complete */
xfs_sync_data(mp, SYNC_WAIT);
xfs_qm_sync(mp, SYNC_WAIT);
+ /* drop inode references pinned by filestreams */
+ xfs_filestream_flush(mp);
+
/* write superblock and hoover up shutdown errors */
- error = xfs_sync_fsdata(mp, 0);
+ error = xfs_sync_fsdata(mp, SYNC_WAIT);
/* flush data-only devices */
if (mp->m_rtdev_targp)
@@ -570,8 +588,6 @@ xfs_sync_worker(
/* dgc: errors ignored here */
error = xfs_qm_sync(mp, SYNC_TRYLOCK);
error = xfs_sync_fsdata(mp, SYNC_TRYLOCK);
- if (xfs_log_need_covered(mp))
- error = xfs_commit_dummy_trans(mp, XFS_LOG_FORCE);
}
mp->m_sync_seq++;
wake_up(&mp->m_wait_single_sync_task);
diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c
index 4e4276b956e..5d1a3b98a6e 100644
--- a/fs/xfs/quota/xfs_qm_syscalls.c
+++ b/fs/xfs/quota/xfs_qm_syscalls.c
@@ -876,7 +876,6 @@ xfs_dqrele_inode(
ip->i_gdquot = NULL;
}
xfs_iput(ip, XFS_ILOCK_EXCL);
- IRELE(ip);
return 0;
}
diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c
index 7465f9ee125..ab89a7e94a0 100644
--- a/fs/xfs/xfs_dfrag.c
+++ b/fs/xfs/xfs_dfrag.c
@@ -206,10 +206,10 @@ xfs_swap_extents(
* process that the file was not changed out from
* under it.
*/
- if ((sbp->bs_ctime.tv_sec != ip->i_d.di_ctime.t_sec) ||
- (sbp->bs_ctime.tv_nsec != ip->i_d.di_ctime.t_nsec) ||
- (sbp->bs_mtime.tv_sec != ip->i_d.di_mtime.t_sec) ||
- (sbp->bs_mtime.tv_nsec != ip->i_d.di_mtime.t_nsec)) {
+ if ((sbp->bs_ctime.tv_sec != VFS_I(ip)->i_ctime.tv_sec) ||
+ (sbp->bs_ctime.tv_nsec != VFS_I(ip)->i_ctime.tv_nsec) ||
+ (sbp->bs_mtime.tv_sec != VFS_I(ip)->i_mtime.tv_sec) ||
+ (sbp->bs_mtime.tv_nsec != VFS_I(ip)->i_mtime.tv_nsec)) {
error = XFS_ERROR(EBUSY);
goto out_unlock;
}
diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c
index fa913e45944..41ad537c49e 100644
--- a/fs/xfs/xfs_dir2_leaf.c
+++ b/fs/xfs/xfs_dir2_leaf.c
@@ -854,6 +854,7 @@ xfs_dir2_leaf_getdents(
*/
ra_want = howmany(bufsize + mp->m_dirblksize,
mp->m_sb.sb_blocksize) - 1;
+ ASSERT(ra_want >= 0);
/*
* If we don't have as many as we want, and we haven't
@@ -1088,7 +1089,8 @@ xfs_dir2_leaf_getdents(
*/
ptr += length;
curoff += length;
- bufsize -= length;
+ /* bufsize may have just been a guess; don't go negative */
+ bufsize = bufsize > length ? bufsize - length : 0;
}
/*
diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c
index ab64f3efb43..0785797db82 100644
--- a/fs/xfs/xfs_ialloc.c
+++ b/fs/xfs/xfs_ialloc.c
@@ -880,6 +880,7 @@ nextag:
* Not in range - save last search
* location and allocate a new inode
*/
+ xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR);
pag->pagl_leftrec = trec.ir_startino;
pag->pagl_rightrec = rec.ir_startino;
pag->pagl_pagino = pagino;
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index c1dc7ef5a1d..b92a4fa2a0a 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -3068,9 +3068,9 @@ xfs_iflush_int(
SYNCHRONIZE();
/*
- * Make sure to get the latest atime from the Linux inode.
+ * Make sure to get the latest timestamps from the Linux inode.
*/
- xfs_synchronize_atime(ip);
+ xfs_synchronize_times(ip);
if (XFS_TEST_ERROR(be16_to_cpu(dip->di_magic) != XFS_DINODE_MAGIC,
mp, XFS_ERRTAG_IFLUSH_1, XFS_RANDOM_IFLUSH_1)) {
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 0b38b9a869e..41555de1d1d 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -504,7 +504,7 @@ void xfs_ichgtime(xfs_inode_t *, int);
void xfs_lock_inodes(xfs_inode_t **, int, uint);
void xfs_lock_two_inodes(xfs_inode_t *, xfs_inode_t *, uint);
-void xfs_synchronize_atime(xfs_inode_t *);
+void xfs_synchronize_times(xfs_inode_t *);
void xfs_mark_inode_dirty_sync(xfs_inode_t *);
#if defined(XFS_INODE_TRACE)
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index 47d5b663c37..9794b876d6f 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -232,6 +232,15 @@ xfs_inode_item_format(
nvecs = 1;
/*
+ * Make sure the linux inode is dirty. We do this before
+ * clearing i_update_core as the VFS will call back into
+ * XFS here and set i_update_core, so we need to dirty the
+ * inode first so that the ordering of i_update_core and
+ * unlogged modifications still works as described below.
+ */
+ xfs_mark_inode_dirty_sync(ip);
+
+ /*
* Clear i_update_core if the timestamps (or any other
* non-transactional modification) need flushing/logging
* and we're about to log them with the rest of the core.
@@ -263,14 +272,9 @@ xfs_inode_item_format(
}
/*
- * Make sure to get the latest atime from the Linux inode.
+ * Make sure to get the latest timestamps from the Linux inode.
*/
- xfs_synchronize_atime(ip);
-
- /*
- * make sure the linux inode is dirty
- */
- xfs_mark_inode_dirty_sync(ip);
+ xfs_synchronize_times(ip);
vecp->i_addr = (xfs_caddr_t)&ip->i_d;
vecp->i_len = sizeof(struct xfs_icdinode);
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index b68f9107e26..62efab2f383 100644
--- a/fs/xfs/xfs_itable.c
+++ b/fs/xfs/xfs_itable.c
@@ -59,6 +59,7 @@ xfs_bulkstat_one_iget(
{
xfs_icdinode_t *dic; /* dinode core info pointer */
xfs_inode_t *ip; /* incore inode pointer */
+ struct inode *inode;
int error;
error = xfs_iget(mp, NULL, ino,
@@ -72,6 +73,7 @@ xfs_bulkstat_one_iget(
ASSERT(ip->i_imap.im_blkno != 0);
dic = &ip->i_d;
+ inode = VFS_I(ip);
/* xfs_iget returns the following without needing
* further change.
@@ -83,16 +85,19 @@ xfs_bulkstat_one_iget(
buf->bs_uid = dic->di_uid;
buf->bs_gid = dic->di_gid;
buf->bs_size = dic->di_size;
+
/*
- * We are reading the atime from the Linux inode because the
- * dinode might not be uptodate.
+ * We need to read the timestamps from the Linux inode because
+ * the VFS keeps writing directly into the inode structure instead
+ * of telling us about the updates.
*/
- buf->bs_atime.tv_sec = VFS_I(ip)->i_atime.tv_sec;
- buf->bs_atime.tv_nsec = VFS_I(ip)->i_atime.tv_nsec;
- buf->bs_mtime.tv_sec = dic->di_mtime.t_sec;
- buf->bs_mtime.tv_nsec = dic->di_mtime.t_nsec;
- buf->bs_ctime.tv_sec = dic->di_ctime.t_sec;
- buf->bs_ctime.tv_nsec = dic->di_ctime.t_nsec;
+ buf->bs_atime.tv_sec = inode->i_atime.tv_sec;
+ buf->bs_atime.tv_nsec = inode->i_atime.tv_nsec;
+ buf->bs_mtime.tv_sec = inode->i_mtime.tv_sec;
+ buf->bs_mtime.tv_nsec = inode->i_mtime.tv_nsec;
+ buf->bs_ctime.tv_sec = inode->i_ctime.tv_sec;
+ buf->bs_ctime.tv_nsec = inode->i_ctime.tv_nsec;
+
buf->bs_xflags = xfs_ip2xflags(ip);
buf->bs_extsize = dic->di_extsize << mp->m_sb.sb_blocklog;
buf->bs_extents = dic->di_nextents;
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 1099395d7d6..fb17f8226b0 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -1980,7 +1980,7 @@ xlog_recover_do_reg_buffer(
"XFS: NULL dquot in %s.", __func__);
goto next;
}
- if (item->ri_buf[i].i_len < sizeof(xfs_dqblk_t)) {
+ if (item->ri_buf[i].i_len < sizeof(xfs_disk_dquot_t)) {
cmn_err(CE_ALERT,
"XFS: dquot too small (%d) in %s.",
item->ri_buf[i].i_len, __func__);
@@ -2635,7 +2635,7 @@ xlog_recover_do_dquot_trans(
"XFS: NULL dquot in %s.", __func__);
return XFS_ERROR(EIO);
}
- if (item->ri_buf[1].i_len < sizeof(xfs_dqblk_t)) {
+ if (item->ri_buf[1].i_len < sizeof(xfs_disk_dquot_t)) {
cmn_err(CE_ALERT,
"XFS: dquot too small (%d) in %s.",
item->ri_buf[1].i_len, __func__);
diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c
index f31271c30de..2ffc570679b 100644
--- a/fs/xfs/xfs_trans_ail.c
+++ b/fs/xfs/xfs_trans_ail.c
@@ -467,6 +467,7 @@ xfs_trans_ail_update(
{
xfs_log_item_t *dlip = NULL;
xfs_log_item_t *mlip; /* ptr to minimum lip */
+ xfs_lsn_t tail_lsn;
mlip = xfs_ail_min(ailp);
@@ -483,8 +484,16 @@ xfs_trans_ail_update(
if (mlip == dlip) {
mlip = xfs_ail_min(ailp);
+ /*
+ * It is not safe to access mlip after the AIL lock is
+ * dropped, so we must get a copy of li_lsn before we do
+ * so. This is especially important on 32-bit platforms
+ * where accessing and updating 64-bit values like li_lsn
+ * is not atomic.
+ */
+ tail_lsn = mlip->li_lsn;
spin_unlock(&ailp->xa_lock);
- xfs_log_move_tail(ailp->xa_mount, mlip->li_lsn);
+ xfs_log_move_tail(ailp->xa_mount, tail_lsn);
} else {
spin_unlock(&ailp->xa_lock);
}
@@ -514,6 +523,7 @@ xfs_trans_ail_delete(
{
xfs_log_item_t *dlip;
xfs_log_item_t *mlip;
+ xfs_lsn_t tail_lsn;
if (lip->li_flags & XFS_LI_IN_AIL) {
mlip = xfs_ail_min(ailp);
@@ -527,9 +537,16 @@ xfs_trans_ail_delete(
if (mlip == dlip) {
mlip = xfs_ail_min(ailp);
+ /*
+ * It is not safe to access mlip after the AIL lock
+ * is dropped, so we must get a copy of li_lsn
+ * before we do so. This is especially important
+ * on 32-bit platforms where accessing and updating
+ * 64-bit values like li_lsn is not atomic.
+ */
+ tail_lsn = mlip ? mlip->li_lsn : 0;
spin_unlock(&ailp->xa_lock);
- xfs_log_move_tail(ailp->xa_mount,
- (mlip ? mlip->li_lsn : 0));
+ xfs_log_move_tail(ailp->xa_mount, tail_lsn);
} else {
spin_unlock(&ailp->xa_lock);
}
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index a434f287962..b572f7e840e 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -2476,12 +2476,6 @@ xfs_reclaim(
ASSERT(XFS_FORCED_SHUTDOWN(ip->i_mount) || ip->i_delayed_blks == 0);
/*
- * Make sure the atime in the XFS inode is correct before freeing the
- * Linux inode.
- */
- xfs_synchronize_atime(ip);
-
- /*
* If we have nothing to flush with this inode then complete the
* teardown now, otherwise break the link between the xfs inode and the
* linux inode and clean up the xfs inode later. This avoids flushing