Merge branch 'xfs-misc-fixes-for-3.20-4' into for-next
diff --git a/Documentation/filesystems/xfs.txt b/Documentation/filesystems/xfs.txt
index 2e56845..0bfafe1 100644
--- a/Documentation/filesystems/xfs.txt
+++ b/Documentation/filesystems/xfs.txt
@@ -349,16 +349,13 @@
Deprecated Sysctls
==================
- fs.xfs.xfsbufd_centisecs (Min: 50 Default: 100 Max: 3000)
- Dirty metadata is now tracked by the log subsystem and
- flushing is driven by log space and idling demands. The
- xfsbufd no longer exists, so this syctl does nothing.
+None at present.
- Due for removal in 3.14.
- fs.xfs.age_buffer_centisecs (Min: 100 Default: 1500 Max: 720000)
- Dirty metadata is now tracked by the log subsystem and
- flushing is driven by log space and idling demands. The
- xfsbufd no longer exists, so this syctl does nothing.
+Removed Sysctls
+===============
- Due for removal in 3.14.
+ Name Removed
+ ---- -------
+ fs.xfs.xfsbufd_centisec v3.20
+ fs.xfs.age_buffer_centisecs v3.20
diff --git a/fs/xfs/kmem.c b/fs/xfs/kmem.c
index 53e95b2..a7a3a63 100644
--- a/fs/xfs/kmem.c
+++ b/fs/xfs/kmem.c
@@ -91,16 +91,6 @@
return ptr;
}
-void
-kmem_free(const void *ptr)
-{
- if (!is_vmalloc_addr(ptr)) {
- kfree(ptr);
- } else {
- vfree(ptr);
- }
-}
-
void *
kmem_realloc(const void *ptr, size_t newsize, size_t oldsize,
xfs_km_flags_t flags)
diff --git a/fs/xfs/kmem.h b/fs/xfs/kmem.h
index 64db0e5..cc6b768 100644
--- a/fs/xfs/kmem.h
+++ b/fs/xfs/kmem.h
@@ -63,7 +63,10 @@
extern void *kmem_alloc(size_t, xfs_km_flags_t);
extern void *kmem_zalloc_large(size_t size, xfs_km_flags_t);
extern void *kmem_realloc(const void *, size_t, size_t, xfs_km_flags_t);
-extern void kmem_free(const void *);
+static inline void kmem_free(const void *ptr)
+{
+ kvfree(ptr);
+}
extern void *kmem_zalloc_greedy(size_t *, size_t, size_t);
diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 63a5bb9..61ec015 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -973,7 +973,11 @@
*firstblock = args.fsbno;
bp = xfs_btree_get_bufl(args.mp, tp, args.fsbno, 0);
- /* initialise the block and copy the data */
+ /*
+ * Initialise the block and copy the data
+ *
+ * Note: init_fn must set the buffer log item type correctly!
+ */
init_fn(tp, bp, ip, ifp);
/* account for the change in fork size and log everything */
diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h
index 44db6db..b9d8a49 100644
--- a/fs/xfs/libxfs/xfs_bmap.h
+++ b/fs/xfs/libxfs/xfs_bmap.h
@@ -28,6 +28,37 @@
extern kmem_zone_t *xfs_bmap_free_item_zone;
/*
+ * Argument structure for xfs_bmap_alloc.
+ */
+struct xfs_bmalloca {
+ xfs_fsblock_t *firstblock; /* i/o first block allocated */
+ struct xfs_bmap_free *flist; /* bmap freelist */
+ struct xfs_trans *tp; /* transaction pointer */
+ struct xfs_inode *ip; /* incore inode pointer */
+ struct xfs_bmbt_irec prev; /* extent before the new one */
+ struct xfs_bmbt_irec got; /* extent after, or delayed */
+
+ xfs_fileoff_t offset; /* offset in file filling in */
+ xfs_extlen_t length; /* i/o length asked/allocated */
+ xfs_fsblock_t blkno; /* starting block of new extent */
+
+ struct xfs_btree_cur *cur; /* btree cursor */
+ xfs_extnum_t idx; /* current extent index */
+ int nallocs;/* number of extents alloc'd */
+ int logflags;/* flags for transaction logging */
+
+ xfs_extlen_t total; /* total blocks needed for xaction */
+ xfs_extlen_t minlen; /* minimum allocation size (blocks) */
+ xfs_extlen_t minleft; /* amount must be left after alloc */
+ bool eof; /* set if allocating past last extent */
+ bool wasdel; /* replacing a delayed allocation */
+ bool userdata;/* set if is user data */
+ bool aeof; /* allocated space at eof */
+ bool conv; /* overwriting unwritten extents */
+ int flags;
+};
+
+/*
* List of extents to be free "later".
* The list is kept sorted on xbf_startblock.
*/
@@ -149,6 +180,8 @@
void xfs_bmap_add_free(xfs_fsblock_t bno, xfs_filblks_t len,
struct xfs_bmap_free *flist, struct xfs_mount *mp);
void xfs_bmap_cancel(struct xfs_bmap_free *flist);
+int xfs_bmap_finish(struct xfs_trans **tp, struct xfs_bmap_free *flist,
+ int *committed);
void xfs_bmap_compute_maxlevels(struct xfs_mount *mp, int whichfork);
int xfs_bmap_first_unused(struct xfs_trans *tp, struct xfs_inode *ip,
xfs_extlen_t len, xfs_fileoff_t *unused, int whichfork);
diff --git a/fs/xfs/libxfs/xfs_format.h b/fs/xfs/libxfs/xfs_format.h
index 749c861..8eb7189 100644
--- a/fs/xfs/libxfs/xfs_format.h
+++ b/fs/xfs/libxfs/xfs_format.h
@@ -307,8 +307,8 @@
#define XFS_SB_ICOUNT XFS_SB_MVAL(ICOUNT)
#define XFS_SB_IFREE XFS_SB_MVAL(IFREE)
#define XFS_SB_FDBLOCKS XFS_SB_MVAL(FDBLOCKS)
-#define XFS_SB_FEATURES2 XFS_SB_MVAL(FEATURES2)
-#define XFS_SB_BAD_FEATURES2 XFS_SB_MVAL(BAD_FEATURES2)
+#define XFS_SB_FEATURES2 (XFS_SB_MVAL(FEATURES2) | \
+ XFS_SB_MVAL(BAD_FEATURES2))
#define XFS_SB_FEATURES_COMPAT XFS_SB_MVAL(FEATURES_COMPAT)
#define XFS_SB_FEATURES_RO_COMPAT XFS_SB_MVAL(FEATURES_RO_COMPAT)
#define XFS_SB_FEATURES_INCOMPAT XFS_SB_MVAL(FEATURES_INCOMPAT)
@@ -322,9 +322,9 @@
XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | XFS_SB_GQUOTINO | \
XFS_SB_QFLAGS | XFS_SB_SHARED_VN | XFS_SB_UNIT | XFS_SB_WIDTH | \
XFS_SB_ICOUNT | XFS_SB_IFREE | XFS_SB_FDBLOCKS | XFS_SB_FEATURES2 | \
- XFS_SB_BAD_FEATURES2 | XFS_SB_FEATURES_COMPAT | \
- XFS_SB_FEATURES_RO_COMPAT | XFS_SB_FEATURES_INCOMPAT | \
- XFS_SB_FEATURES_LOG_INCOMPAT | XFS_SB_PQUOTINO)
+ XFS_SB_FEATURES_COMPAT | XFS_SB_FEATURES_RO_COMPAT | \
+ XFS_SB_FEATURES_INCOMPAT | XFS_SB_FEATURES_LOG_INCOMPAT | \
+ XFS_SB_PQUOTINO)
/*
diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h
similarity index 100%
rename from fs/xfs/xfs_fs.h
rename to fs/xfs/libxfs/xfs_fs.h
diff --git a/fs/xfs/libxfs/xfs_symlink_remote.c b/fs/xfs/libxfs/xfs_symlink_remote.c
index c80c523..e7e26bd 100644
--- a/fs/xfs/libxfs/xfs_symlink_remote.c
+++ b/fs/xfs/libxfs/xfs_symlink_remote.c
@@ -178,6 +178,8 @@
struct xfs_mount *mp = ip->i_mount;
char *buf;
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_SYMLINK_BUF);
+
if (!xfs_sb_version_hascrc(&mp->m_sb)) {
bp->b_ops = NULL;
memcpy(bp->b_addr, ifp->if_u1.if_data, ifp->if_bytes);
diff --git a/fs/xfs/xfs_types.h b/fs/xfs/libxfs/xfs_types.h
similarity index 100%
rename from fs/xfs/xfs_types.h
rename to fs/xfs/libxfs/xfs_types.h
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index 18e2f3b..3a9b7a1 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -135,30 +135,22 @@
*/
STATIC int
xfs_setfilesize(
- struct xfs_ioend *ioend)
+ struct xfs_inode *ip,
+ struct xfs_trans *tp,
+ xfs_off_t offset,
+ size_t size)
{
- struct xfs_inode *ip = XFS_I(ioend->io_inode);
- struct xfs_trans *tp = ioend->io_append_trans;
xfs_fsize_t isize;
- /*
- * The transaction may have been allocated in the I/O submission thread,
- * thus we need to mark ourselves as beeing in a transaction manually.
- * Similarly for freeze protection.
- */
- current_set_flags_nested(&tp->t_pflags, PF_FSTRANS);
- rwsem_acquire_read(&VFS_I(ip)->i_sb->s_writers.lock_map[SB_FREEZE_FS-1],
- 0, 1, _THIS_IP_);
-
xfs_ilock(ip, XFS_ILOCK_EXCL);
- isize = xfs_new_eof(ip, ioend->io_offset + ioend->io_size);
+ isize = xfs_new_eof(ip, offset + size);
if (!isize) {
xfs_iunlock(ip, XFS_ILOCK_EXCL);
xfs_trans_cancel(tp, 0);
return 0;
}
- trace_xfs_setfilesize(ip, ioend->io_offset, ioend->io_size);
+ trace_xfs_setfilesize(ip, offset, size);
ip->i_d.di_size = isize;
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
@@ -167,6 +159,25 @@
return xfs_trans_commit(tp, 0);
}
+STATIC int
+xfs_setfilesize_ioend(
+ struct xfs_ioend *ioend)
+{
+ struct xfs_inode *ip = XFS_I(ioend->io_inode);
+ struct xfs_trans *tp = ioend->io_append_trans;
+
+ /*
+ * The transaction may have been allocated in the I/O submission thread,
+ * thus we need to mark ourselves as being in a transaction manually.
+ * Similarly for freeze protection.
+ */
+ current_set_flags_nested(&tp->t_pflags, PF_FSTRANS);
+ rwsem_acquire_read(&VFS_I(ip)->i_sb->s_writers.lock_map[SB_FREEZE_FS-1],
+ 0, 1, _THIS_IP_);
+
+ return xfs_setfilesize(ip, tp, ioend->io_offset, ioend->io_size);
+}
+
/*
* Schedule IO completion handling on the final put of an ioend.
*
@@ -182,8 +193,7 @@
if (ioend->io_type == XFS_IO_UNWRITTEN)
queue_work(mp->m_unwritten_workqueue, &ioend->io_work);
- else if (ioend->io_append_trans ||
- (ioend->io_isdirect && xfs_ioend_is_append(ioend)))
+ else if (ioend->io_append_trans)
queue_work(mp->m_data_workqueue, &ioend->io_work);
else
xfs_destroy_ioend(ioend);
@@ -215,22 +225,8 @@
if (ioend->io_type == XFS_IO_UNWRITTEN) {
error = xfs_iomap_write_unwritten(ip, ioend->io_offset,
ioend->io_size);
- } else if (ioend->io_isdirect && xfs_ioend_is_append(ioend)) {
- /*
- * For direct I/O we do not know if we need to allocate blocks
- * or not so we can't preallocate an append transaction as that
- * results in nested reservations and log space deadlocks. Hence
- * allocate the transaction here. While this is sub-optimal and
- * can block IO completion for some time, we're stuck with doing
- * it this way until we can pass the ioend to the direct IO
- * allocation callbacks and avoid nesting that way.
- */
- error = xfs_setfilesize_trans_alloc(ioend);
- if (error)
- goto done;
- error = xfs_setfilesize(ioend);
} else if (ioend->io_append_trans) {
- error = xfs_setfilesize(ioend);
+ error = xfs_setfilesize_ioend(ioend);
} else {
ASSERT(!xfs_ioend_is_append(ioend));
}
@@ -242,17 +238,6 @@
}
/*
- * Call IO completion handling in caller context on the final put of an ioend.
- */
-STATIC void
-xfs_finish_ioend_sync(
- struct xfs_ioend *ioend)
-{
- if (atomic_dec_and_test(&ioend->io_remaining))
- xfs_end_io(&ioend->io_work);
-}
-
-/*
* Allocate and initialise an IO completion structure.
* We need to track unwritten extent write completion here initially.
* We'll need to extend this for updating the ondisk inode size later
@@ -273,7 +258,6 @@
* all the I/O from calling the completion routine too early.
*/
atomic_set(&ioend->io_remaining, 1);
- ioend->io_isdirect = 0;
ioend->io_error = 0;
ioend->io_list = NULL;
ioend->io_type = type;
@@ -1459,11 +1443,7 @@
*
* If the private argument is non-NULL __xfs_get_blocks signals us that we
* need to issue a transaction to convert the range from unwritten to written
- * extents. In case this is regular synchronous I/O we just call xfs_end_io
- * to do this and we are done. But in case this was a successful AIO
- * request this handler is called from interrupt context, from which we
- * can't start transactions. In that case offload the I/O completion to
- * the workqueues we also use for buffered I/O completion.
+ * extents.
*/
STATIC void
xfs_end_io_direct_write(
@@ -1472,7 +1452,12 @@
ssize_t size,
void *private)
{
- struct xfs_ioend *ioend = iocb->private;
+ struct inode *inode = file_inode(iocb->ki_filp);
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
+
+ if (XFS_FORCED_SHUTDOWN(mp))
+ return;
/*
* While the generic direct I/O code updates the inode size, it does
@@ -1480,22 +1465,33 @@
* end_io handler thinks the on-disk size is outside the in-core
* size. To prevent this just update it a little bit earlier here.
*/
- if (offset + size > i_size_read(ioend->io_inode))
- i_size_write(ioend->io_inode, offset + size);
+ if (offset + size > i_size_read(inode))
+ i_size_write(inode, offset + size);
/*
- * blockdev_direct_IO can return an error even after the I/O
- * completion handler was called. Thus we need to protect
- * against double-freeing.
+ * For direct I/O we do not know if we need to allocate blocks or not,
+ * so we can't preallocate an append transaction, as that results in
+ * nested reservations and log space deadlocks. Hence allocate the
+ * transaction here. While this is sub-optimal and can block IO
+ * completion for some time, we're stuck with doing it this way until
+ * we can pass the ioend to the direct IO allocation callbacks and
+ * avoid nesting that way.
*/
- iocb->private = NULL;
+ if (private && size > 0) {
+ xfs_iomap_write_unwritten(ip, offset, size);
+ } else if (offset + size > ip->i_d.di_size) {
+ struct xfs_trans *tp;
+ int error;
- ioend->io_offset = offset;
- ioend->io_size = size;
- if (private && size > 0)
- ioend->io_type = XFS_IO_UNWRITTEN;
+ tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS);
+ error = xfs_trans_reserve(tp, &M_RES(mp)->tr_fsyncts, 0, 0);
+ if (error) {
+ xfs_trans_cancel(tp, 0);
+ return;
+ }
- xfs_finish_ioend_sync(ioend);
+ xfs_setfilesize(ip, tp, offset, size);
+ }
}
STATIC ssize_t
@@ -1507,39 +1503,16 @@
{
struct inode *inode = iocb->ki_filp->f_mapping->host;
struct block_device *bdev = xfs_find_bdev_for_inode(inode);
- struct xfs_ioend *ioend = NULL;
- ssize_t ret;
if (rw & WRITE) {
- size_t size = iov_iter_count(iter);
-
- /*
- * We cannot preallocate a size update transaction here as we
- * don't know whether allocation is necessary or not. Hence we
- * can only tell IO completion that one is necessary if we are
- * not doing unwritten extent conversion.
- */
- iocb->private = ioend = xfs_alloc_ioend(inode, XFS_IO_DIRECT);
- if (offset + size > XFS_I(inode)->i_d.di_size)
- ioend->io_isdirect = 1;
-
- ret = __blockdev_direct_IO(rw, iocb, inode, bdev, iter,
+ return __blockdev_direct_IO(rw, iocb, inode, bdev, iter,
offset, xfs_get_blocks_direct,
xfs_end_io_direct_write, NULL,
DIO_ASYNC_EXTEND);
- if (ret != -EIOCBQUEUED && iocb->private)
- goto out_destroy_ioend;
- } else {
- ret = __blockdev_direct_IO(rw, iocb, inode, bdev, iter,
- offset, xfs_get_blocks_direct,
- NULL, NULL, 0);
}
-
- return ret;
-
-out_destroy_ioend:
- xfs_destroy_ioend(ioend);
- return ret;
+ return __blockdev_direct_IO(rw, iocb, inode, bdev, iter,
+ offset, xfs_get_blocks_direct,
+ NULL, NULL, 0);
}
/*
diff --git a/fs/xfs/xfs_aops.h b/fs/xfs/xfs_aops.h
index f94dd45..ac644e0 100644
--- a/fs/xfs/xfs_aops.h
+++ b/fs/xfs/xfs_aops.h
@@ -24,14 +24,12 @@
* Types of I/O for bmap clustering and I/O completion tracking.
*/
enum {
- XFS_IO_DIRECT = 0, /* special case for direct I/O ioends */
XFS_IO_DELALLOC, /* covers delalloc region */
XFS_IO_UNWRITTEN, /* covers allocated but uninitialized data */
XFS_IO_OVERWRITE, /* covers already allocated extent */
};
#define XFS_IO_TYPES \
- { 0, "" }, \
{ XFS_IO_DELALLOC, "delalloc" }, \
{ XFS_IO_UNWRITTEN, "unwritten" }, \
{ XFS_IO_OVERWRITE, "overwrite" }
@@ -45,7 +43,6 @@
unsigned int io_type; /* delalloc / unwritten */
int io_error; /* I/O error code */
atomic_t io_remaining; /* hold count */
- unsigned int io_isdirect : 1;/* direct I/O */
struct inode *io_inode; /* file being written to */
struct buffer_head *io_buffer_head;/* buffer linked list head */
struct buffer_head *io_buffer_tail;/* buffer linked list tail */
diff --git a/fs/xfs/xfs_bmap_util.h b/fs/xfs/xfs_bmap_util.h
index 2fdb72d..736429a 100644
--- a/fs/xfs/xfs_bmap_util.h
+++ b/fs/xfs/xfs_bmap_util.h
@@ -26,43 +26,8 @@
struct xfs_inode;
struct xfs_mount;
struct xfs_trans;
+struct xfs_bmalloca;
-/*
- * Argument structure for xfs_bmap_alloc.
- */
-struct xfs_bmalloca {
- xfs_fsblock_t *firstblock; /* i/o first block allocated */
- struct xfs_bmap_free *flist; /* bmap freelist */
- struct xfs_trans *tp; /* transaction pointer */
- struct xfs_inode *ip; /* incore inode pointer */
- struct xfs_bmbt_irec prev; /* extent before the new one */
- struct xfs_bmbt_irec got; /* extent after, or delayed */
-
- xfs_fileoff_t offset; /* offset in file filling in */
- xfs_extlen_t length; /* i/o length asked/allocated */
- xfs_fsblock_t blkno; /* starting block of new extent */
-
- struct xfs_btree_cur *cur; /* btree cursor */
- xfs_extnum_t idx; /* current extent index */
- int nallocs;/* number of extents alloc'd */
- int logflags;/* flags for transaction logging */
-
- xfs_extlen_t total; /* total blocks needed for xaction */
- xfs_extlen_t minlen; /* minimum allocation size (blocks) */
- xfs_extlen_t minleft; /* amount must be left after alloc */
- bool eof; /* set if allocating past last extent */
- bool wasdel; /* replacing a delayed allocation */
- bool userdata;/* set if is user data */
- bool aeof; /* allocated space at eof */
- bool conv; /* overwriting unwritten extents */
- int flags;
- struct completion *done;
- struct work_struct work;
- int result;
-};
-
-int xfs_bmap_finish(struct xfs_trans **tp, struct xfs_bmap_free *flist,
- int *committed);
int xfs_bmap_rtalloc(struct xfs_bmalloca *ap);
int xfs_bmap_eof(struct xfs_inode *ip, xfs_fileoff_t endoff,
int whichfork, int *eof);
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 3f9bd58..507d96a 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -319,6 +319,10 @@
ASSERT(atomic_read(&bip->bli_refcount) > 0);
ASSERT((bip->bli_flags & XFS_BLI_LOGGED) ||
(bip->bli_flags & XFS_BLI_STALE));
+ ASSERT((bip->bli_flags & XFS_BLI_STALE) ||
+ (xfs_blft_from_flags(&bip->__bli_format) > XFS_BLFT_UNKNOWN_BUF
+ && xfs_blft_from_flags(&bip->__bli_format) < XFS_BLFT_MAX_BUF));
+
/*
* If it is an inode buffer, transfer the in-memory state to the
@@ -535,7 +539,7 @@
if ((bp->b_flags & XBF_WRITE_FAIL) &&
___ratelimit(&xfs_buf_write_fail_rl_state, "XFS:")) {
xfs_warn(bp->b_target->bt_mount,
-"Detected failing async write on buffer block 0x%llx. Retrying async write.\n",
+"Detected failing async write on buffer block 0x%llx. Retrying async write.",
(long long)bp->b_bn);
}
diff --git a/fs/xfs/xfs_dquot.h b/fs/xfs/xfs_dquot.h
index c24c67e..2f536f3 100644
--- a/fs/xfs/xfs_dquot.h
+++ b/fs/xfs/xfs_dquot.h
@@ -86,7 +86,7 @@
wait_for_completion(&dqp->q_flush);
}
-static inline int xfs_dqflock_nowait(xfs_dquot_t *dqp)
+static inline bool xfs_dqflock_nowait(xfs_dquot_t *dqp)
{
return try_wait_for_completion(&dqp->q_flush);
}
diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c
index 13e974e..712d312 100644
--- a/fs/xfs/xfs_file.c
+++ b/fs/xfs/xfs_file.c
@@ -127,6 +127,42 @@
return (-status);
}
+int
+xfs_update_prealloc_flags(
+ struct xfs_inode *ip,
+ enum xfs_prealloc_flags flags)
+{
+ struct xfs_trans *tp;
+ int error;
+
+ tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_WRITEID);
+ error = xfs_trans_reserve(tp, &M_RES(ip->i_mount)->tr_writeid, 0, 0);
+ if (error) {
+ xfs_trans_cancel(tp, 0);
+ return error;
+ }
+
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+
+ if (!(flags & XFS_PREALLOC_INVISIBLE)) {
+ ip->i_d.di_mode &= ~S_ISUID;
+ if (ip->i_d.di_mode & S_IXGRP)
+ ip->i_d.di_mode &= ~S_ISGID;
+ xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
+ }
+
+ if (flags & XFS_PREALLOC_SET)
+ ip->i_d.di_flags |= XFS_DIFLAG_PREALLOC;
+ if (flags & XFS_PREALLOC_CLEAR)
+ ip->i_d.di_flags &= ~XFS_DIFLAG_PREALLOC;
+
+ xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+ if (flags & XFS_PREALLOC_SYNC)
+ xfs_trans_set_sync(tp);
+ return xfs_trans_commit(tp, 0);
+}
+
/*
* Fsync operations on directories are much simpler than on regular files,
* as there is no file data to flush, and thus also no need for explicit
@@ -784,8 +820,8 @@
{
struct inode *inode = file_inode(file);
struct xfs_inode *ip = XFS_I(inode);
- struct xfs_trans *tp;
long error;
+ enum xfs_prealloc_flags flags = 0;
loff_t new_size = 0;
if (!S_ISREG(inode->i_mode))
@@ -822,6 +858,8 @@
if (error)
goto out_unlock;
} else {
+ flags |= XFS_PREALLOC_SET;
+
if (!(mode & FALLOC_FL_KEEP_SIZE) &&
offset + len > i_size_read(inode)) {
new_size = offset + len;
@@ -839,28 +877,10 @@
goto out_unlock;
}
- tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_WRITEID);
- error = xfs_trans_reserve(tp, &M_RES(ip->i_mount)->tr_writeid, 0, 0);
- if (error) {
- xfs_trans_cancel(tp, 0);
- goto out_unlock;
- }
-
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
- ip->i_d.di_mode &= ~S_ISUID;
- if (ip->i_d.di_mode & S_IXGRP)
- ip->i_d.di_mode &= ~S_ISGID;
-
- if (!(mode & (FALLOC_FL_PUNCH_HOLE | FALLOC_FL_COLLAPSE_RANGE)))
- ip->i_d.di_flags |= XFS_DIFLAG_PREALLOC;
-
- xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
- xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-
if (file->f_flags & O_DSYNC)
- xfs_trans_set_sync(tp);
- error = xfs_trans_commit(tp, 0);
+ flags |= XFS_PREALLOC_SYNC;
+
+ error = xfs_update_prealloc_flags(ip, flags);
if (error)
goto out_unlock;
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 41f804e..daafa1f 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -1995,6 +1995,7 @@
agi->agi_unlinked[bucket_index] = cpu_to_be32(agino);
offset = offsetof(xfs_agi_t, agi_unlinked) +
(sizeof(xfs_agino_t) * bucket_index);
+ xfs_trans_buf_set_type(tp, agibp, XFS_BLFT_AGI_BUF);
xfs_trans_log_buf(tp, agibp, offset,
(offset + sizeof(xfs_agino_t) - 1));
return 0;
@@ -2086,6 +2087,7 @@
agi->agi_unlinked[bucket_index] = cpu_to_be32(next_agino);
offset = offsetof(xfs_agi_t, agi_unlinked) +
(sizeof(xfs_agino_t) * bucket_index);
+ xfs_trans_buf_set_type(tp, agibp, XFS_BLFT_AGI_BUF);
xfs_trans_log_buf(tp, agibp, offset,
(offset + sizeof(xfs_agino_t) - 1));
} else {
@@ -2656,6 +2658,124 @@
}
/*
+ * xfs_cross_rename()
+ *
+ * responsible for handling RENAME_EXCHANGE flag in renameat2() sytemcall
+ */
+STATIC int
+xfs_cross_rename(
+ struct xfs_trans *tp,
+ struct xfs_inode *dp1,
+ struct xfs_name *name1,
+ struct xfs_inode *ip1,
+ struct xfs_inode *dp2,
+ struct xfs_name *name2,
+ struct xfs_inode *ip2,
+ struct xfs_bmap_free *free_list,
+ xfs_fsblock_t *first_block,
+ int spaceres)
+{
+ int error = 0;
+ int ip1_flags = 0;
+ int ip2_flags = 0;
+ int dp2_flags = 0;
+
+ /* Swap inode number for dirent in first parent */
+ error = xfs_dir_replace(tp, dp1, name1,
+ ip2->i_ino,
+ first_block, free_list, spaceres);
+ if (error)
+ goto out;
+
+ /* Swap inode number for dirent in second parent */
+ error = xfs_dir_replace(tp, dp2, name2,
+ ip1->i_ino,
+ first_block, free_list, spaceres);
+ if (error)
+ goto out;
+
+ /*
+ * If we're renaming one or more directories across different parents,
+ * update the respective ".." entries (and link counts) to match the new
+ * parents.
+ */
+ if (dp1 != dp2) {
+ dp2_flags = XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG;
+
+ if (S_ISDIR(ip2->i_d.di_mode)) {
+ error = xfs_dir_replace(tp, ip2, &xfs_name_dotdot,
+ dp1->i_ino, first_block,
+ free_list, spaceres);
+ if (error)
+ goto out;
+
+ /* transfer ip2 ".." reference to dp1 */
+ if (!S_ISDIR(ip1->i_d.di_mode)) {
+ error = xfs_droplink(tp, dp2);
+ if (error)
+ goto out;
+ error = xfs_bumplink(tp, dp1);
+ if (error)
+ goto out;
+ }
+
+ /*
+ * Although ip1 isn't changed here, userspace needs
+ * to be warned about the change, so that applications
+ * relying on it (like backup ones), will properly
+ * notify the change
+ */
+ ip1_flags |= XFS_ICHGTIME_CHG;
+ ip2_flags |= XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG;
+ }
+
+ if (S_ISDIR(ip1->i_d.di_mode)) {
+ error = xfs_dir_replace(tp, ip1, &xfs_name_dotdot,
+ dp2->i_ino, first_block,
+ free_list, spaceres);
+ if (error)
+ goto out;
+
+ /* transfer ip1 ".." reference to dp2 */
+ if (!S_ISDIR(ip2->i_d.di_mode)) {
+ error = xfs_droplink(tp, dp1);
+ if (error)
+ goto out;
+ error = xfs_bumplink(tp, dp2);
+ if (error)
+ goto out;
+ }
+
+ /*
+ * Although ip2 isn't changed here, userspace needs
+ * to be warned about the change, so that applications
+ * relying on it (like backup ones), will properly
+ * notify the change
+ */
+ ip1_flags |= XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG;
+ ip2_flags |= XFS_ICHGTIME_CHG;
+ }
+ }
+
+ if (ip1_flags) {
+ xfs_trans_ichgtime(tp, ip1, ip1_flags);
+ xfs_trans_log_inode(tp, ip1, XFS_ILOG_CORE);
+ }
+ if (ip2_flags) {
+ xfs_trans_ichgtime(tp, ip2, ip2_flags);
+ xfs_trans_log_inode(tp, ip2, XFS_ILOG_CORE);
+ }
+ if (dp2_flags) {
+ xfs_trans_ichgtime(tp, dp2, dp2_flags);
+ xfs_trans_log_inode(tp, dp2, XFS_ILOG_CORE);
+ }
+ xfs_trans_ichgtime(tp, dp1, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
+ xfs_trans_log_inode(tp, dp1, XFS_ILOG_CORE);
+out:
+ return error;
+}
+
+/*
* xfs_rename
*/
int
@@ -2665,7 +2785,8 @@
xfs_inode_t *src_ip,
xfs_inode_t *target_dp,
struct xfs_name *target_name,
- xfs_inode_t *target_ip)
+ xfs_inode_t *target_ip,
+ unsigned int flags)
{
xfs_trans_t *tp = NULL;
xfs_mount_t *mp = src_dp->i_mount;
@@ -2743,6 +2864,18 @@
}
/*
+ * Handle RENAME_EXCHANGE flags
+ */
+ if (flags & RENAME_EXCHANGE) {
+ error = xfs_cross_rename(tp, src_dp, src_name, src_ip,
+ target_dp, target_name, target_ip,
+ &free_list, &first_block, spaceres);
+ if (error)
+ goto abort_return;
+ goto finish_rename;
+ }
+
+ /*
* Set up the target.
*/
if (target_ip == NULL) {
@@ -2881,6 +3014,7 @@
if (new_parent)
xfs_trans_log_inode(tp, target_dp, XFS_ILOG_CORE);
+finish_rename:
/*
* If this is a synchronous mount, make sure that the
* rename transaction goes to disk before returning to
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 4ed2ba9..86cd6b3 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -338,7 +338,7 @@
int xfs_rename(struct xfs_inode *src_dp, struct xfs_name *src_name,
struct xfs_inode *src_ip, struct xfs_inode *target_dp,
struct xfs_name *target_name,
- struct xfs_inode *target_ip);
+ struct xfs_inode *target_ip, unsigned int flags);
void xfs_ilock(xfs_inode_t *, uint);
int xfs_ilock_nowait(xfs_inode_t *, uint);
@@ -377,6 +377,15 @@
int xfs_bumplink(struct xfs_trans *, struct xfs_inode *);
/* from xfs_file.c */
+enum xfs_prealloc_flags {
+ XFS_PREALLOC_SET = (1 << 1),
+ XFS_PREALLOC_CLEAR = (1 << 2),
+ XFS_PREALLOC_SYNC = (1 << 3),
+ XFS_PREALLOC_INVISIBLE = (1 << 4),
+};
+
+int xfs_update_prealloc_flags(struct xfs_inode *,
+ enum xfs_prealloc_flags);
int xfs_zero_eof(struct xfs_inode *, xfs_off_t, xfs_fsize_t);
int xfs_iozero(struct xfs_inode *, loff_t, size_t);
diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c
index f1dc90b..f7afb86 100644
--- a/fs/xfs/xfs_ioctl.c
+++ b/fs/xfs/xfs_ioctl.c
@@ -606,11 +606,8 @@
unsigned int cmd,
xfs_flock64_t *bf)
{
- struct xfs_mount *mp = ip->i_mount;
- struct xfs_trans *tp;
struct iattr iattr;
- bool setprealloc = false;
- bool clrprealloc = false;
+ enum xfs_prealloc_flags flags = 0;
int error;
/*
@@ -630,6 +627,11 @@
if (!S_ISREG(inode->i_mode))
return -EINVAL;
+ if (filp->f_flags & O_DSYNC)
+ flags |= XFS_PREALLOC_SYNC;
+ if (ioflags & XFS_IO_INVIS)
+ flags |= XFS_PREALLOC_INVISIBLE;
+
error = mnt_want_write_file(filp);
if (error)
return error;
@@ -673,25 +675,23 @@
}
if (bf->l_start < 0 ||
- bf->l_start > mp->m_super->s_maxbytes ||
+ bf->l_start > inode->i_sb->s_maxbytes ||
bf->l_start + bf->l_len < 0 ||
- bf->l_start + bf->l_len >= mp->m_super->s_maxbytes) {
+ bf->l_start + bf->l_len >= inode->i_sb->s_maxbytes) {
error = -EINVAL;
goto out_unlock;
}
switch (cmd) {
case XFS_IOC_ZERO_RANGE:
+ flags |= XFS_PREALLOC_SET;
error = xfs_zero_file_space(ip, bf->l_start, bf->l_len);
- if (!error)
- setprealloc = true;
break;
case XFS_IOC_RESVSP:
case XFS_IOC_RESVSP64:
+ flags |= XFS_PREALLOC_SET;
error = xfs_alloc_file_space(ip, bf->l_start, bf->l_len,
XFS_BMAPI_PREALLOC);
- if (!error)
- setprealloc = true;
break;
case XFS_IOC_UNRESVSP:
case XFS_IOC_UNRESVSP64:
@@ -701,6 +701,7 @@
case XFS_IOC_ALLOCSP64:
case XFS_IOC_FREESP:
case XFS_IOC_FREESP64:
+ flags |= XFS_PREALLOC_CLEAR;
if (bf->l_start > XFS_ISIZE(ip)) {
error = xfs_alloc_file_space(ip, XFS_ISIZE(ip),
bf->l_start - XFS_ISIZE(ip), 0);
@@ -712,8 +713,6 @@
iattr.ia_size = bf->l_start;
error = xfs_setattr_size(ip, &iattr);
- if (!error)
- clrprealloc = true;
break;
default:
ASSERT(0);
@@ -723,32 +722,7 @@
if (error)
goto out_unlock;
- tp = xfs_trans_alloc(mp, XFS_TRANS_WRITEID);
- error = xfs_trans_reserve(tp, &M_RES(mp)->tr_writeid, 0, 0);
- if (error) {
- xfs_trans_cancel(tp, 0);
- goto out_unlock;
- }
-
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
-
- if (!(ioflags & XFS_IO_INVIS)) {
- ip->i_d.di_mode &= ~S_ISUID;
- if (ip->i_d.di_mode & S_IXGRP)
- ip->i_d.di_mode &= ~S_ISGID;
- xfs_trans_ichgtime(tp, ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
- }
-
- if (setprealloc)
- ip->i_d.di_flags |= XFS_DIFLAG_PREALLOC;
- else if (clrprealloc)
- ip->i_d.di_flags &= ~XFS_DIFLAG_PREALLOC;
-
- xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
- if (filp->f_flags & O_DSYNC)
- xfs_trans_set_sync(tp);
- error = xfs_trans_commit(tp, 0);
+ error = xfs_update_prealloc_flags(ip, flags);
out_unlock:
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c
index ec67728..bfc7c7c 100644
--- a/fs/xfs/xfs_ioctl32.c
+++ b/fs/xfs/xfs_ioctl32.c
@@ -423,7 +423,7 @@
ops = memdup_user(compat_ptr(am_hreq.ops), size);
if (IS_ERR(ops)) {
- error = -PTR_ERR(ops);
+ error = PTR_ERR(ops);
goto out_dput;
}
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index c980e2a..ccb1dd0 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -802,7 +802,7 @@
xfs_iomap_write_unwritten(
xfs_inode_t *ip,
xfs_off_t offset,
- size_t count)
+ xfs_off_t count)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb;
diff --git a/fs/xfs/xfs_iomap.h b/fs/xfs/xfs_iomap.h
index 411fbb8..8688e66 100644
--- a/fs/xfs/xfs_iomap.h
+++ b/fs/xfs/xfs_iomap.h
@@ -27,6 +27,6 @@
struct xfs_bmbt_irec *);
int xfs_iomap_write_allocate(struct xfs_inode *, xfs_off_t,
struct xfs_bmbt_irec *);
-int xfs_iomap_write_unwritten(struct xfs_inode *, xfs_off_t, size_t);
+int xfs_iomap_write_unwritten(struct xfs_inode *, xfs_off_t, xfs_off_t);
#endif /* __XFS_IOMAP_H__*/
diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c
index c50311c..ce80eeb 100644
--- a/fs/xfs/xfs_iops.c
+++ b/fs/xfs/xfs_iops.c
@@ -380,18 +380,27 @@
struct inode *odir,
struct dentry *odentry,
struct inode *ndir,
- struct dentry *ndentry)
+ struct dentry *ndentry,
+ unsigned int flags)
{
struct inode *new_inode = ndentry->d_inode;
+ int omode = 0;
struct xfs_name oname;
struct xfs_name nname;
- xfs_dentry_to_name(&oname, odentry, 0);
+ if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE))
+ return -EINVAL;
+
+ /* if we are exchanging files, we need to set i_mode of both files */
+ if (flags & RENAME_EXCHANGE)
+ omode = ndentry->d_inode->i_mode;
+
+ xfs_dentry_to_name(&oname, odentry, omode);
xfs_dentry_to_name(&nname, ndentry, odentry->d_inode->i_mode);
return xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode),
- XFS_I(ndir), &nname, new_inode ?
- XFS_I(new_inode) : NULL);
+ XFS_I(ndir), &nname,
+ new_inode ? XFS_I(new_inode) : NULL, flags);
}
/*
@@ -1144,7 +1153,7 @@
*/
.rmdir = xfs_vn_unlink,
.mknod = xfs_vn_mknod,
- .rename = xfs_vn_rename,
+ .rename2 = xfs_vn_rename,
.get_acl = xfs_get_acl,
.set_acl = xfs_set_acl,
.getattr = xfs_vn_getattr,
@@ -1172,7 +1181,7 @@
*/
.rmdir = xfs_vn_unlink,
.mknod = xfs_vn_mknod,
- .rename = xfs_vn_rename,
+ .rename2 = xfs_vn_rename,
.get_acl = xfs_get_acl,
.set_acl = xfs_set_acl,
.getattr = xfs_vn_getattr,
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 2b8dcf2..bcc7cfa 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -1407,6 +1407,8 @@
ASSERT(xfs_buf_islocked(bp));
xfs_buf_unlock(bp);
+ /* use high priority wq for log I/O completion */
+ bp->b_ioend_wq = mp->m_log_workqueue;
bp->b_iodone = xlog_iodone;
log->l_xbuf = bp;
@@ -1439,6 +1441,8 @@
ASSERT(xfs_buf_islocked(bp));
xfs_buf_unlock(bp);
+ /* use high priority wq for log I/O completion */
+ bp->b_ioend_wq = mp->m_log_workqueue;
bp->b_iodone = xlog_iodone;
iclog->ic_bp = bp;
iclog->ic_data = bp->b_addr;
@@ -1818,8 +1822,6 @@
XFS_BUF_ZEROFLAGS(bp);
XFS_BUF_ASYNC(bp);
bp->b_flags |= XBF_SYNCIO;
- /* use high priority completion wq */
- bp->b_ioend_wq = log->l_mp->m_log_workqueue;
if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) {
bp->b_flags |= XBF_FUA;
@@ -1868,8 +1870,6 @@
bp->b_flags |= XBF_SYNCIO;
if (log->l_mp->m_flags & XFS_MOUNT_BARRIER)
bp->b_flags |= XBF_FUA;
- /* use high priority completion wq */
- bp->b_ioend_wq = log->l_mp->m_log_workqueue;
ASSERT(XFS_BUF_ADDR(bp) <= log->l_logBBsize-1);
ASSERT(XFS_BUF_ADDR(bp) + BTOBB(count) <= log->l_logBBsize);
@@ -2039,7 +2039,7 @@
" total reg = %u bytes (o/flow = %u bytes)\n"
" ophdrs = %u (ophdr space = %u bytes)\n"
" ophdr + reg = %u bytes\n"
- " num regions = %u\n",
+ " num regions = %u",
((ticket->t_trans_type <= 0 ||
ticket->t_trans_type > XFS_TRANS_TYPE_MAX) ?
"bad-trans-type" : trans_type_str[ticket->t_trans_type-1]),
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index 26afa43..f2449fd 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -685,7 +685,7 @@
mp);
if (IS_ERR(*bdevp)) {
error = PTR_ERR(*bdevp);
- xfs_warn(mp, "Invalid device [%s], error=%d\n", name, error);
+ xfs_warn(mp, "Invalid device [%s], error=%d", name, error);
}
return error;
diff --git a/fs/xfs/xfs_sysctl.c b/fs/xfs/xfs_sysctl.c
index 1743b9f..a0c8067 100644
--- a/fs/xfs/xfs_sysctl.c
+++ b/fs/xfs/xfs_sysctl.c
@@ -149,24 +149,6 @@
.extra2 = &xfs_params.inherit_noatim.max
},
{
- .procname = "xfsbufd_centisecs",
- .data = &xfs_params.xfs_buf_timer.val,
- .maxlen = sizeof(int),
- .mode = 0644,
- .proc_handler = proc_dointvec_minmax,
- .extra1 = &xfs_params.xfs_buf_timer.min,
- .extra2 = &xfs_params.xfs_buf_timer.max
- },
- {
- .procname = "age_buffer_centisecs",
- .data = &xfs_params.xfs_buf_age.val,
- .maxlen = sizeof(int),
- .mode = 0644,
- .proc_handler = proc_dointvec_minmax,
- .extra1 = &xfs_params.xfs_buf_age.min,
- .extra2 = &xfs_params.xfs_buf_age.max
- },
- {
.procname = "inherit_nosymlinks",
.data = &xfs_params.inherit_nosym.val,
.maxlen = sizeof(int),
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index fa3135b..eb90cd5 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -472,6 +472,7 @@
whole = 1;
}
+ xfs_trans_buf_set_type(tp, bp, XFS_BLFT_SB_BUF);
if (whole)
/*
* Log the whole thing, the fields are noncontiguous.
