summaryrefslogtreecommitdiffstats
path: root/fs/xfs/linux-2.6
diff options
context:
space:
mode:
Diffstat (limited to 'fs/xfs/linux-2.6')
-rw-r--r--fs/xfs/linux-2.6/sv.h22
-rw-r--r--fs/xfs/linux-2.6/xfs_aops.c66
-rw-r--r--fs/xfs/linux-2.6/xfs_aops.h3
-rw-r--r--fs/xfs/linux-2.6/xfs_buf.c87
-rw-r--r--fs/xfs/linux-2.6/xfs_buf.h30
-rw-r--r--fs/xfs/linux-2.6/xfs_cred.h8
-rw-r--r--fs/xfs/linux-2.6/xfs_export.c1
-rw-r--r--fs/xfs/linux-2.6/xfs_file.c189
-rw-r--r--fs/xfs/linux-2.6/xfs_fs_subr.c23
-rw-r--r--fs/xfs/linux-2.6/xfs_globals.c8
-rw-r--r--fs/xfs/linux-2.6/xfs_globals.h1
-rw-r--r--fs/xfs/linux-2.6/xfs_ioctl.c223
-rw-r--r--fs/xfs/linux-2.6/xfs_ioctl.h82
-rw-r--r--fs/xfs/linux-2.6/xfs_ioctl32.c849
-rw-r--r--fs/xfs/linux-2.6/xfs_ioctl32.h214
-rw-r--r--fs/xfs/linux-2.6/xfs_iops.c122
-rw-r--r--fs/xfs/linux-2.6/xfs_iops.h1
-rw-r--r--fs/xfs/linux-2.6/xfs_linux.h13
-rw-r--r--fs/xfs/linux-2.6/xfs_lrw.c50
-rw-r--r--fs/xfs/linux-2.6/xfs_stats.c6
-rw-r--r--fs/xfs/linux-2.6/xfs_stats.h65
-rw-r--r--fs/xfs/linux-2.6/xfs_super.c884
-rw-r--r--fs/xfs/linux-2.6/xfs_super.h15
-rw-r--r--fs/xfs/linux-2.6/xfs_sync.c762
-rw-r--r--fs/xfs/linux-2.6/xfs_sync.h55
-rw-r--r--fs/xfs/linux-2.6/xfs_sysctl.c11
-rw-r--r--fs/xfs/linux-2.6/xfs_sysctl.h3
-rw-r--r--fs/xfs/linux-2.6/xfs_vfs.h77
-rw-r--r--fs/xfs/linux-2.6/xfs_vnode.c145
-rw-r--r--fs/xfs/linux-2.6/xfs_vnode.h72
30 files changed, 2466 insertions, 1621 deletions
diff --git a/fs/xfs/linux-2.6/sv.h b/fs/xfs/linux-2.6/sv.h
index 351a8f454bd1..4dfc7c370819 100644
--- a/fs/xfs/linux-2.6/sv.h
+++ b/fs/xfs/linux-2.6/sv.h
@@ -32,23 +32,15 @@ typedef struct sv_s {
wait_queue_head_t waiters;
} sv_t;
-#define SV_FIFO 0x0 /* sv_t is FIFO type */
-#define SV_LIFO 0x2 /* sv_t is LIFO type */
-#define SV_PRIO 0x4 /* sv_t is PRIO type */
-#define SV_KEYED 0x6 /* sv_t is KEYED type */
-#define SV_DEFAULT SV_FIFO
-
-
-static inline void _sv_wait(sv_t *sv, spinlock_t *lock, int state,
- unsigned long timeout)
+static inline void _sv_wait(sv_t *sv, spinlock_t *lock)
{
DECLARE_WAITQUEUE(wait, current);
add_wait_queue_exclusive(&sv->waiters, &wait);
- __set_current_state(state);
+ __set_current_state(TASK_UNINTERRUPTIBLE);
spin_unlock(lock);
- schedule_timeout(timeout);
+ schedule();
remove_wait_queue(&sv->waiters, &wait);
}
@@ -58,13 +50,7 @@ static inline void _sv_wait(sv_t *sv, spinlock_t *lock, int state,
#define sv_destroy(sv) \
/*NOTHING*/
#define sv_wait(sv, pri, lock, s) \
- _sv_wait(sv, lock, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT)
-#define sv_wait_sig(sv, pri, lock, s) \
- _sv_wait(sv, lock, TASK_INTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT)
-#define sv_timedwait(sv, pri, lock, s, svf, ts, rts) \
- _sv_wait(sv, lock, TASK_UNINTERRUPTIBLE, timespec_to_jiffies(ts))
-#define sv_timedwait_sig(sv, pri, lock, s, svf, ts, rts) \
- _sv_wait(sv, lock, TASK_INTERRUPTIBLE, timespec_to_jiffies(ts))
+ _sv_wait(sv, lock)
#define sv_signal(sv) \
wake_up(&(sv)->waiters)
#define sv_broadcast(sv) \
diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index a44d68eb50b5..de3a198f771e 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -42,6 +42,40 @@
#include <linux/pagevec.h>
#include <linux/writeback.h>
+
+/*
+ * Prime number of hash buckets since address is used as the key.
+ */
+#define NVSYNC 37
+#define to_ioend_wq(v) (&xfs_ioend_wq[((unsigned long)v) % NVSYNC])
+static wait_queue_head_t xfs_ioend_wq[NVSYNC];
+
+void __init
+xfs_ioend_init(void)
+{
+ int i;
+
+ for (i = 0; i < NVSYNC; i++)
+ init_waitqueue_head(&xfs_ioend_wq[i]);
+}
+
+void
+xfs_ioend_wait(
+ xfs_inode_t *ip)
+{
+ wait_queue_head_t *wq = to_ioend_wq(ip);
+
+ wait_event(*wq, (atomic_read(&ip->i_iocount) == 0));
+}
+
+STATIC void
+xfs_ioend_wake(
+ xfs_inode_t *ip)
+{
+ if (atomic_dec_and_test(&ip->i_iocount))
+ wake_up(to_ioend_wq(ip));
+}
+
STATIC void
xfs_count_page_state(
struct page *page,
@@ -146,16 +180,25 @@ xfs_destroy_ioend(
xfs_ioend_t *ioend)
{
struct buffer_head *bh, *next;
+ struct xfs_inode *ip = XFS_I(ioend->io_inode);
for (bh = ioend->io_buffer_head; bh; bh = next) {
next = bh->b_private;
bh->b_end_io(bh, !ioend->io_error);
}
- if (unlikely(ioend->io_error)) {
- vn_ioerror(XFS_I(ioend->io_inode), ioend->io_error,
- __FILE__,__LINE__);
+
+ /*
+ * Volume managers supporting multiple paths can send back ENODEV
+ * when the final path disappears. In this case continuing to fill
+ * the page cache with dirty data which cannot be written out is
+ * evil, so prevent that.
+ */
+ if (unlikely(ioend->io_error == -ENODEV)) {
+ xfs_do_force_shutdown(ip->i_mount, SHUTDOWN_DEVICE_REQ,
+ __FILE__, __LINE__);
}
- vn_iowake(XFS_I(ioend->io_inode));
+
+ xfs_ioend_wake(ip);
mempool_free(ioend, xfs_ioend_pool);
}
@@ -191,7 +234,7 @@ xfs_setfilesize(
ip->i_d.di_size = isize;
ip->i_update_core = 1;
ip->i_update_size = 1;
- mark_inode_dirty_sync(ioend->io_inode);
+ xfs_mark_inode_dirty_sync(ip);
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
@@ -317,14 +360,9 @@ xfs_map_blocks(
xfs_iomap_t *mapp,
int flags)
{
- xfs_inode_t *ip = XFS_I(inode);
- int error, nmaps = 1;
-
- error = xfs_iomap(ip, offset, count,
- flags, mapp, &nmaps);
- if (!error && (flags & (BMAPI_WRITE|BMAPI_ALLOCATE)))
- xfs_iflags_set(ip, XFS_IMODIFIED);
- return -error;
+ int nmaps = 1;
+
+ return -xfs_iomap(XFS_I(inode), offset, count, flags, mapp, &nmaps);
}
STATIC_INLINE int
@@ -512,7 +550,7 @@ xfs_cancel_ioend(
unlock_buffer(bh);
} while ((bh = next_bh) != NULL);
- vn_iowake(XFS_I(ioend->io_inode));
+ xfs_ioend_wake(XFS_I(ioend->io_inode));
mempool_free(ioend, xfs_ioend_pool);
} while ((ioend = next) != NULL);
}
diff --git a/fs/xfs/linux-2.6/xfs_aops.h b/fs/xfs/linux-2.6/xfs_aops.h
index 3ba0631a3818..7b26f5ff9692 100644
--- a/fs/xfs/linux-2.6/xfs_aops.h
+++ b/fs/xfs/linux-2.6/xfs_aops.h
@@ -43,4 +43,7 @@ typedef struct xfs_ioend {
extern const struct address_space_operations xfs_address_space_operations;
extern int xfs_get_blocks(struct inode *, sector_t, struct buffer_head *, int);
+extern void xfs_ioend_init(void);
+extern void xfs_ioend_wait(struct xfs_inode *);
+
#endif /* __XFS_AOPS_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index 36d5fcd3f593..cb329edc925b 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -630,6 +630,29 @@ xfs_buf_get_flags(
return NULL;
}
+STATIC int
+_xfs_buf_read(
+ xfs_buf_t *bp,
+ xfs_buf_flags_t flags)
+{
+ int status;
+
+ XB_TRACE(bp, "_xfs_buf_read", (unsigned long)flags);
+
+ ASSERT(!(flags & (XBF_DELWRI|XBF_WRITE)));
+ ASSERT(bp->b_bn != XFS_BUF_DADDR_NULL);
+
+ bp->b_flags &= ~(XBF_WRITE | XBF_ASYNC | XBF_DELWRI | \
+ XBF_READ_AHEAD | _XBF_RUN_QUEUES);
+ bp->b_flags |= flags & (XBF_READ | XBF_ASYNC | \
+ XBF_READ_AHEAD | _XBF_RUN_QUEUES);
+
+ status = xfs_buf_iorequest(bp);
+ if (!status && !(flags & XBF_ASYNC))
+ status = xfs_buf_iowait(bp);
+ return status;
+}
+
xfs_buf_t *
xfs_buf_read_flags(
xfs_buftarg_t *target,
@@ -646,7 +669,7 @@ xfs_buf_read_flags(
if (!XFS_BUF_ISDONE(bp)) {
XB_TRACE(bp, "read", (unsigned long)flags);
XFS_STATS_INC(xb_get_read);
- xfs_buf_iostart(bp, flags);
+ _xfs_buf_read(bp, flags);
} else if (flags & XBF_ASYNC) {
XB_TRACE(bp, "read_async", (unsigned long)flags);
/*
@@ -1048,50 +1071,39 @@ xfs_buf_ioerror(
XB_TRACE(bp, "ioerror", (unsigned long)error);
}
-/*
- * Initiate I/O on a buffer, based on the flags supplied.
- * The b_iodone routine in the buffer supplied will only be called
- * when all of the subsidiary I/O requests, if any, have been completed.
- */
int
-xfs_buf_iostart(
- xfs_buf_t *bp,
- xfs_buf_flags_t flags)
+xfs_bawrite(
+ void *mp,
+ struct xfs_buf *bp)
{
- int status = 0;
+ XB_TRACE(bp, "bawrite", 0);
- XB_TRACE(bp, "iostart", (unsigned long)flags);
+ ASSERT(bp->b_bn != XFS_BUF_DADDR_NULL);
- if (flags & XBF_DELWRI) {
- bp->b_flags &= ~(XBF_READ | XBF_WRITE | XBF_ASYNC);
- bp->b_flags |= flags & (XBF_DELWRI | XBF_ASYNC);
- xfs_buf_delwri_queue(bp, 1);
- return 0;
- }
+ xfs_buf_delwri_dequeue(bp);
- bp->b_flags &= ~(XBF_READ | XBF_WRITE | XBF_ASYNC | XBF_DELWRI | \
- XBF_READ_AHEAD | _XBF_RUN_QUEUES);
- bp->b_flags |= flags & (XBF_READ | XBF_WRITE | XBF_ASYNC | \
- XBF_READ_AHEAD | _XBF_RUN_QUEUES);
+ bp->b_flags &= ~(XBF_READ | XBF_DELWRI | XBF_READ_AHEAD);
+ bp->b_flags |= (XBF_WRITE | XBF_ASYNC | _XBF_RUN_QUEUES);
+
+ bp->b_mount = mp;
+ bp->b_strat = xfs_bdstrat_cb;
+ return xfs_bdstrat_cb(bp);
+}
- BUG_ON(bp->b_bn == XFS_BUF_DADDR_NULL);
+void
+xfs_bdwrite(
+ void *mp,
+ struct xfs_buf *bp)
+{
+ XB_TRACE(bp, "bdwrite", 0);
- /* For writes allow an alternate strategy routine to precede
- * the actual I/O request (which may not be issued at all in
- * a shutdown situation, for example).
- */
- status = (flags & XBF_WRITE) ?
- xfs_buf_iostrategy(bp) : xfs_buf_iorequest(bp);
+ bp->b_strat = xfs_bdstrat_cb;
+ bp->b_mount = mp;
- /* Wait for I/O if we are not an async request.
- * Note: async I/O request completion will release the buffer,
- * and that can already be done by this point. So using the
- * buffer pointer from here on, after async I/O, is invalid.
- */
- if (!status && !(flags & XBF_ASYNC))
- status = xfs_buf_iowait(bp);
+ bp->b_flags &= ~XBF_READ;
+ bp->b_flags |= (XBF_DELWRI | XBF_ASYNC);
- return status;
+ xfs_buf_delwri_queue(bp, 1);
}
STATIC_INLINE void
@@ -1114,8 +1126,7 @@ xfs_buf_bio_end_io(
unsigned int blocksize = bp->b_target->bt_bsize;
struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1;
- if (!test_bit(BIO_UPTODATE, &bio->bi_flags))
- bp->b_error = EIO;
+ xfs_buf_ioerror(bp, -error);
do {
struct page *page = bvec->bv_page;
diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h
index 456519a088c7..288ae7c4c800 100644
--- a/fs/xfs/linux-2.6/xfs_buf.h
+++ b/fs/xfs/linux-2.6/xfs_buf.h
@@ -168,7 +168,7 @@ typedef struct xfs_buf {
struct completion b_iowait; /* queue for I/O waiters */
void *b_fspriv;
void *b_fspriv2;
- void *b_fspriv3;
+ struct xfs_mount *b_mount;
unsigned short b_error; /* error code on I/O */
unsigned int b_page_count; /* size of page array */
unsigned int b_offset; /* page offset in first page */
@@ -214,9 +214,10 @@ extern void xfs_buf_lock(xfs_buf_t *);
extern void xfs_buf_unlock(xfs_buf_t *);
/* Buffer Read and Write Routines */
+extern int xfs_bawrite(void *mp, xfs_buf_t *bp);
+extern void xfs_bdwrite(void *mp, xfs_buf_t *bp);
extern void xfs_buf_ioend(xfs_buf_t *, int);
extern void xfs_buf_ioerror(xfs_buf_t *, int);
-extern int xfs_buf_iostart(xfs_buf_t *, xfs_buf_flags_t);
extern int xfs_buf_iorequest(xfs_buf_t *);
extern int xfs_buf_iowait(xfs_buf_t *);
extern void xfs_buf_iomove(xfs_buf_t *, size_t, size_t, xfs_caddr_t,
@@ -311,10 +312,6 @@ extern void xfs_buf_trace(xfs_buf_t *, char *, void *, void *);
#define XFS_BUF_UNORDERED(bp) ((bp)->b_flags &= ~XBF_ORDERED)
#define XFS_BUF_ISORDERED(bp) ((bp)->b_flags & XBF_ORDERED)
-#define XFS_BUF_SHUT(bp) do { } while (0)
-#define XFS_BUF_UNSHUT(bp) do { } while (0)
-#define XFS_BUF_ISSHUT(bp) (0)
-
#define XFS_BUF_HOLD(bp) xfs_buf_hold(bp)
#define XFS_BUF_READ(bp) ((bp)->b_flags |= XBF_READ)
#define XFS_BUF_UNREAD(bp) ((bp)->b_flags &= ~XBF_READ)
@@ -334,8 +331,6 @@ extern void xfs_buf_trace(xfs_buf_t *, char *, void *, void *);
#define XFS_BUF_SET_FSPRIVATE(bp, val) ((bp)->b_fspriv = (void*)(val))
#define XFS_BUF_FSPRIVATE2(bp, type) ((type)(bp)->b_fspriv2)
#define XFS_BUF_SET_FSPRIVATE2(bp, val) ((bp)->b_fspriv2 = (void*)(val))
-#define XFS_BUF_FSPRIVATE3(bp, type) ((type)(bp)->b_fspriv3)
-#define XFS_BUF_SET_FSPRIVATE3(bp, val) ((bp)->b_fspriv3 = (void*)(val))
#define XFS_BUF_SET_START(bp) do { } while (0)
#define XFS_BUF_SET_BRELSE_FUNC(bp, func) ((bp)->b_relse = (func))
@@ -366,14 +361,6 @@ extern void xfs_buf_trace(xfs_buf_t *, char *, void *, void *);
#define XFS_BUF_TARGET(bp) ((bp)->b_target)
#define XFS_BUFTARG_NAME(target) xfs_buf_target_name(target)
-static inline int xfs_bawrite(void *mp, xfs_buf_t *bp)
-{
- bp->b_fspriv3 = mp;
- bp->b_strat = xfs_bdstrat_cb;
- xfs_buf_delwri_dequeue(bp);
- return xfs_buf_iostart(bp, XBF_WRITE | XBF_ASYNC | _XBF_RUN_QUEUES);
-}
-
static inline void xfs_buf_relse(xfs_buf_t *bp)
{
if (!bp->b_relse)
@@ -414,17 +401,6 @@ static inline int XFS_bwrite(xfs_buf_t *bp)
return error;
}
-/*
- * No error can be returned from xfs_buf_iostart for delwri
- * buffers as they are queued and no I/O is issued.
- */
-static inline void xfs_bdwrite(void *mp, xfs_buf_t *bp)
-{
- bp->b_strat = xfs_bdstrat_cb;
- bp->b_fspriv3 = mp;
- (void)xfs_buf_iostart(bp, XBF_DELWRI | XBF_ASYNC);
-}
-
#define XFS_bdstrat(bp) xfs_buf_iorequest(bp)
#define xfs_iowait(bp) xfs_buf_iowait(bp)
diff --git a/fs/xfs/linux-2.6/xfs_cred.h b/fs/xfs/linux-2.6/xfs_cred.h
index 8c022cd0ad67..55bddf3b6091 100644
--- a/fs/xfs/linux-2.6/xfs_cred.h
+++ b/fs/xfs/linux-2.6/xfs_cred.h
@@ -25,12 +25,4 @@
*/
typedef const struct cred cred_t;
-extern cred_t *sys_cred;
-
-/* this is a hack.. (assumes sys_cred is the only cred_t in the system) */
-static inline int capable_cred(cred_t *cr, int cid)
-{
- return (cr == sys_cred) ? 1 : capable(cid);
-}
-
#endif /* __XFS_CRED_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c
index 7f7abec25e14..595751f78350 100644
--- a/fs/xfs/linux-2.6/xfs_export.c
+++ b/fs/xfs/linux-2.6/xfs_export.c
@@ -29,7 +29,6 @@
#include "xfs_vnodeops.h"
#include "xfs_bmap_btree.h"
#include "xfs_inode.h"
-#include "xfs_vfsops.h"
/*
* Note that we only accept fileids which are long enough rather than allow
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index 3fee790f138b..e14c4e3aea0c 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -36,89 +36,54 @@
#include "xfs_inode.h"
#include "xfs_error.h"
#include "xfs_rw.h"
-#include "xfs_ioctl32.h"
#include "xfs_vnodeops.h"
+#include "xfs_da_btree.h"
+#include "xfs_ioctl.h"
#include <linux/dcache.h>
#include <linux/smp_lock.h>
static struct vm_operations_struct xfs_file_vm_ops;
-STATIC_INLINE ssize_t
-__xfs_file_read(
+STATIC ssize_t
+xfs_file_aio_read(
struct kiocb *iocb,
const struct iovec *iov,
unsigned long nr_segs,
- int ioflags,
loff_t pos)
{
struct file *file = iocb->ki_filp;
+ int ioflags = IO_ISAIO;
BUG_ON(iocb->ki_pos != pos);
if (unlikely(file->f_flags & O_DIRECT))
ioflags |= IO_ISDIRECT;
+ if (file->f_mode & FMODE_NOCMTIME)
+ ioflags |= IO_INVIS;
return xfs_read(XFS_I(file->f_path.dentry->d_inode), iocb, iov,
nr_segs, &iocb->ki_pos, ioflags);
}
STATIC ssize_t
-xfs_file_aio_read(
- struct kiocb *iocb,
- const struct iovec *iov,
- unsigned long nr_segs,
- loff_t pos)
-{
- return __xfs_file_read(iocb, iov, nr_segs, IO_ISAIO, pos);
-}
-
-STATIC ssize_t
-xfs_file_aio_read_invis(
- struct kiocb *iocb,
- const struct iovec *iov,
- unsigned long nr_segs,
- loff_t pos)
-{
- return __xfs_file_read(iocb, iov, nr_segs, IO_ISAIO|IO_INVIS, pos);
-}
-
-STATIC_INLINE ssize_t
-__xfs_file_write(
+xfs_file_aio_write(
struct kiocb *iocb,
const struct iovec *iov,
unsigned long nr_segs,
- int ioflags,
loff_t pos)
{
- struct file *file = iocb->ki_filp;
+ struct file *file = iocb->ki_filp;
+ int ioflags = IO_ISAIO;
BUG_ON(iocb->ki_pos != pos);
if (unlikely(file->f_flags & O_DIRECT))
ioflags |= IO_ISDIRECT;
+ if (file->f_mode & FMODE_NOCMTIME)
+ ioflags |= IO_INVIS;
return xfs_write(XFS_I(file->f_mapping->host), iocb, iov, nr_segs,
&iocb->ki_pos, ioflags);
}
STATIC ssize_t
-xfs_file_aio_write(
- struct kiocb *iocb,
- const struct iovec *iov,
- unsigned long nr_segs,
- loff_t pos)
-{
- return __xfs_file_write(iocb, iov, nr_segs, IO_ISAIO, pos);
-}
-
-STATIC ssize_t
-xfs_file_aio_write_invis(
- struct kiocb *iocb,
- const struct iovec *iov,
- unsigned long nr_segs,
- loff_t pos)
-{
- return __xfs_file_write(iocb, iov, nr_segs, IO_ISAIO|IO_INVIS, pos);
-}
-
-STATIC ssize_t
xfs_file_splice_read(
struct file *infilp,
loff_t *ppos,
@@ -126,20 +91,13 @@ xfs_file_splice_read(
size_t len,
unsigned int flags)
{
- return xfs_splice_read(XFS_I(infilp->f_path.dentry->d_inode),
- infilp, ppos, pipe, len, flags, 0);
-}
+ int ioflags = 0;
+
+ if (infilp->f_mode & FMODE_NOCMTIME)
+ ioflags |= IO_INVIS;
-STATIC ssize_t
-xfs_file_splice_read_invis(
- struct file *infilp,
- loff_t *ppos,
- struct pipe_inode_info *pipe,
- size_t len,
- unsigned int flags)
-{
return xfs_splice_read(XFS_I(infilp->f_path.dentry->d_inode),
- infilp, ppos, pipe, len, flags, IO_INVIS);
+ infilp, ppos, pipe, len, flags, ioflags);
}
STATIC ssize_t
@@ -150,30 +108,49 @@ xfs_file_splice_write(
size_t len,
unsigned int flags)
{
- return xfs_splice_write(XFS_I(outfilp->f_path.dentry->d_inode),
- pipe, outfilp, ppos, len, flags, 0);
-}
+ int ioflags = 0;
+
+ if (outfilp->f_mode & FMODE_NOCMTIME)
+ ioflags |= IO_INVIS;
-STATIC ssize_t
-xfs_file_splice_write_invis(
- struct pipe_inode_info *pipe,
- struct file *outfilp,
- loff_t *ppos,
- size_t len,
- unsigned int flags)
-{
return xfs_splice_write(XFS_I(outfilp->f_path.dentry->d_inode),
- pipe, outfilp, ppos, len, flags, IO_INVIS);
+ pipe, outfilp, ppos, len, flags, ioflags);
}
STATIC int
xfs_file_open(
struct inode *inode,
- struct file *filp)
+ struct file *file)
{
- if (!(filp->f_flags & O_LARGEFILE) && i_size_read(inode) > MAX_NON_LFS)
+ if (!(file->f_flags & O_LARGEFILE) && i_size_read(inode) > MAX_NON_LFS)
return -EFBIG;
- return -xfs_open(XFS_I(inode));
+ if (XFS_FORCED_SHUTDOWN(XFS_M(inode->i_sb)))
+ return -EIO;
+ return 0;
+}
+
+STATIC int
+xfs_dir_open(
+ struct inode *inode,
+ struct file *file)
+{
+ struct xfs_inode *ip = XFS_I(inode);
+ int mode;
+ int error;
+
+ error = xfs_file_open(inode, file);
+ if (error)
+ return error;
+
+ /*
+ * If there are any blocks, read-ahead block 0 as we're almost
+ * certain to have the next operation be a read there.
+ */
+ mode = xfs_ilock_map_shared(ip);
+ if (ip->i_d.di_nextents > 0)
+ xfs_da_reada_buf(NULL, ip, 0, XFS_DATA_FORK);
+ xfs_iunlock(ip, mode);
+ return 0;
}
STATIC int
@@ -227,7 +204,7 @@ xfs_file_readdir(
* point we can change the ->readdir prototype to include the
* buffer size.
*/
- bufsize = (size_t)min_t(loff_t, PAGE_SIZE, inode->i_size);
+ bufsize = (size_t)min_t(loff_t, PAGE_SIZE, ip->i_d.di_size);
error = xfs_readdir(ip, dirent, bufsize,
(xfs_off_t *)&filp->f_pos, filldir);
@@ -248,48 +225,6 @@ xfs_file_mmap(
return 0;
}
-STATIC long
-xfs_file_ioctl(
- struct file *filp,
- unsigned int cmd,
- unsigned long p)
-{
- int error;
- struct inode *inode = filp->f_path.dentry->d_inode;
-
- error = xfs_ioctl(XFS_I(inode), filp, 0, cmd, (void __user *)p);
- xfs_iflags_set(XFS_I(inode), XFS_IMODIFIED);
-
- /* NOTE: some of the ioctl's return positive #'s as a
- * byte count indicating success, such as
- * readlink_by_handle. So we don't "sign flip"
- * like most other routines. This means true
- * errors need to be returned as a negative value.
- */
- return error;
-}
-
-STATIC long
-xfs_file_ioctl_invis(
- struct file *filp,
- unsigned int cmd,
- unsigned long p)
-{
- int error;
- struct inode *inode = filp->f_path.dentry->d_inode;
-
- error = xfs_ioctl(XFS_I(inode), filp, IO_INVIS, cmd, (void __user *)p);
- xfs_iflags_set(XFS_I(inode), XFS_IMODIFIED);
-
- /* NOTE: some of the ioctl's return positive #'s as a
- * byte count indicating success, such as
- * readlink_by_handle. So we don't "sign flip"
- * like most other routines. This means true
- * errors need to be returned as a negative value.
- */
- return error;
-}
-
/*
* mmap()d file has taken write protection fault and is being made
* writable. We can set the page state up correctly for a writable
@@ -325,26 +260,8 @@ const struct file_operations xfs_file_operations = {
#endif
};
-const struct file_operations xfs_invis_file_operations = {
- .llseek = generic_file_llseek,
- .read = do_sync_read,
- .write = do_sync_write,
- .aio_read = xfs_file_aio_read_invis,
- .aio_write = xfs_file_aio_write_invis,
- .splice_read = xfs_file_splice_read_invis,
- .splice_write = xfs_file_splice_write_invis,
- .unlocked_ioctl = xfs_file_ioctl_invis,
-#ifdef CONFIG_COMPAT
- .compat_ioctl = xfs_file_compat_invis_ioctl,
-#endif
- .mmap = xfs_file_mmap,
- .open = xfs_file_open,
- .release = xfs_file_release,
- .fsync = xfs_file_fsync,
-};
-
-
const struct file_operations xfs_dir_file_operations = {
+ .open = xfs_dir_open,
.read = generic_read_dir,
.readdir = xfs_file_readdir,
.llseek = generic_file_llseek,
diff --git a/fs/xfs/linux-2.6/xfs_fs_subr.c b/fs/xfs/linux-2.6/xfs_fs_subr.c
index 36caa6d957df..5aeb77776961 100644
--- a/fs/xfs/linux-2.6/xfs_fs_subr.c
+++ b/fs/xfs/linux-2.6/xfs_fs_subr.c
@@ -24,6 +24,10 @@ int fs_noerr(void) { return 0; }
int fs_nosys(void) { return ENOSYS; }
void fs_noval(void) { return; }
+/*
+ * note: all filemap functions return negative error codes. These
+ * need to be inverted before returning to the xfs core functions.
+ */
void
xfs_tosspages(
xfs_inode_t *ip,
@@ -53,7 +57,7 @@ xfs_flushinval_pages(
if (!ret)
truncate_inode_pages(mapping, first);
}
- return ret;
+ return -ret;
}
int
@@ -72,10 +76,23 @@ xfs_flush_pages(
xfs_iflags_clear(ip, XFS_ITRUNCATED);
ret = filemap_fdatawrite(mapping);
if (flags & XFS_B_ASYNC)
- return ret;
+ return -ret;
ret2 = filemap_fdatawait(mapping);
if (!ret)
ret = ret2;
}
- return ret;
+ return -ret;
+}
+
+int
+xfs_wait_on_pages(
+ xfs_inode_t *ip,
+ xfs_off_t first,
+ xfs_off_t last)
+{
+ struct address_space *mapping = VFS_I(ip)->i_mapping;
+
+ if (mapping_tagged(mapping, PAGECACHE_TAG_WRITEBACK))
+ return -filemap_fdatawait(mapping);
+ return 0;
}
diff --git a/fs/xfs/linux-2.6/xfs_globals.c b/fs/xfs/linux-2.6/xfs_globals.c
index ef90e64641e6..2ae8b1ccb02e 100644
--- a/fs/xfs/linux-2.6/xfs_globals.c
+++ b/fs/xfs/linux-2.6/xfs_globals.c
@@ -26,7 +26,6 @@
*/
xfs_param_t xfs_params = {
/* MIN DFLT MAX */
- .restrict_chown = { 0, 1, 1 },
.sgid_inherit = { 0, 0, 1 },
.symlink_mode = { 0, 0, 1 },
.panic_mask = { 0, 0, 255 },
@@ -43,10 +42,3 @@ xfs_param_t xfs_params = {
.inherit_nodfrg = { 0, 1, 1 },
.fstrm_timer = { 1, 30*100, 3600*100},
};
-
-/*
- * Global system credential structure.
- */
-static cred_t sys_cred_val;
-cred_t *sys_cred = &sys_cred_val;
-
diff --git a/fs/xfs/linux-2.6/xfs_globals.h b/fs/xfs/linux-2.6/xfs_globals.h
index 6eda8a3eb6f1..69f71caf061c 100644
--- a/fs/xfs/linux-2.6/xfs_globals.h
+++ b/fs/xfs/linux-2.6/xfs_globals.h
@@ -19,6 +19,5 @@
#define __XFS_GLOBALS_H__
extern uint64_t xfs_panic_mask; /* set to cause more panics */
-extern cred_t *sys_cred;
#endif /* __XFS_GLOBALS_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c
index 281cbd5a25cf..67205f6198ba 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl.c
@@ -68,26 +68,22 @@
* XFS_IOC_PATH_TO_HANDLE
* returns full handle for a path
*/
-STATIC int
+int
xfs_find_handle(
unsigned int cmd,
- void __user *arg)
+ xfs_fsop_handlereq_t *hreq)
{
int hsize;
xfs_handle_t handle;
- xfs_fsop_handlereq_t hreq;
struct inode *inode;
- if (copy_from_user(&hreq, arg, sizeof(hreq)))
- return -XFS_ERROR(EFAULT);
-
memset((char *)&handle, 0, sizeof(handle));
switch (cmd) {
case XFS_IOC_PATH_TO_FSHANDLE:
case XFS_IOC_PATH_TO_HANDLE: {
struct path path;
- int error = user_lpath((const char __user *)hreq.path, &path);
+ int error = user_lpath((const char __user *)hreq->path, &path);
if (error)
return error;
@@ -101,7 +97,7 @@ xfs_find_handle(
case XFS_IOC_FD_TO_HANDLE: {
struct file *file;
- file = fget(hreq.fd);
+ file = fget(hreq->fd);
if (!file)
return -EBADF;
@@ -158,8 +154,8 @@ xfs_find_handle(
}
/* now copy our handle into the user buffer & write out the size */
- if (copy_to_user(hreq.ohandle, &handle, hsize) ||
- copy_to_user(hreq.ohandlen, &hsize, sizeof(__s32))) {
+ if (copy_to_user(hreq->ohandle, &handle, hsize) ||
+ copy_to_user(hreq->ohandlen, &hsize, sizeof(__s32))) {
iput(inode);
return -XFS_ERROR(EFAULT);
}
@@ -249,10 +245,10 @@ xfs_vget_fsop_handlereq(
return 0;
}
-STATIC int
+int
xfs_open_by_handle(
xfs_mount_t *mp,
- void __user *arg,
+ xfs_fsop_handlereq_t *hreq,
struct file *parfilp,
struct inode *parinode)
{
@@ -263,14 +259,11 @@ xfs_open_by_handle(
struct file *filp;
struct inode *inode;
struct dentry *dentry;
- xfs_fsop_handlereq_t hreq;
if (!capable(CAP_SYS_ADMIN))
return -XFS_ERROR(EPERM);
- if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t)))
- return -XFS_ERROR(EFAULT);
- error = xfs_vget_fsop_handlereq(mp, parinode, &hreq, &inode);
+ error = xfs_vget_fsop_handlereq(mp, parinode, hreq, &inode);
if (error)
return -error;
@@ -281,10 +274,10 @@ xfs_open_by_handle(
}
#if BITS_PER_LONG != 32
- hreq.oflags |= O_LARGEFILE;
+ hreq->oflags |= O_LARGEFILE;
#endif
/* Put open permission in namei format. */
- permflag = hreq.oflags;
+ permflag = hreq->oflags;
if ((permflag+1) & O_ACCMODE)
permflag++;
if (permflag & O_TRUNC)
@@ -322,15 +315,16 @@ xfs_open_by_handle(
mntget(parfilp->f_path.mnt);
/* Create file pointer. */
- filp = dentry_open(dentry, parfilp->f_path.mnt, hreq.oflags, cred);
+ filp = dentry_open(dentry, parfilp->f_path.mnt, hreq->oflags, cred);
if (IS_ERR(filp)) {
put_unused_fd(new_fd);
return -XFS_ERROR(-PTR_ERR(filp));
}
+
if (inode->i_mode & S_IFREG) {
/* invisible operation should not change atime */
filp->f_flags |= O_NOATIME;
- filp->f_op = &xfs_invis_file_operations;
+ filp->f_mode |= FMODE_NOCMTIME;
}
fd_install(new_fd, filp);
@@ -363,24 +357,21 @@ do_readlink(
}
-STATIC int
+int
xfs_readlink_by_handle(
xfs_mount_t *mp,
- void __user *arg,
+ xfs_fsop_handlereq_t *hreq,
struct inode *parinode)
{
struct inode *inode;
- xfs_fsop_handlereq_t hreq;
__u32 olen;
void *link;
int error;
if (!capable(CAP_SYS_ADMIN))
return -XFS_ERROR(EPERM);
- if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t)))
- return -XFS_ERROR(EFAULT);
- error = xfs_vget_fsop_handlereq(mp, parinode, &hreq, &inode);
+ error = xfs_vget_fsop_handlereq(mp, parinode, hreq, &inode);
if (error)
return -error;
@@ -390,7 +381,7 @@ xfs_readlink_by_handle(
goto out_iput;
}
- if (copy_from_user(&olen, hreq.ohandlen, sizeof(__u32))) {
+ if (copy_from_user(&olen, hreq->ohandlen, sizeof(__u32))) {
error = -XFS_ERROR(EFAULT);
goto out_iput;
}
@@ -402,7 +393,7 @@ xfs_readlink_by_handle(
error = -xfs_readlink(XFS_I(inode), link);
if (error)
goto out_kfree;
- error = do_readlink(hreq.ohandle, olen, link);
+ error = do_readlink(hreq->ohandle, olen, link);
if (error)
goto out_kfree;
@@ -501,7 +492,7 @@ xfs_attrlist_by_handle(
return -error;
}
-STATIC int
+int
xfs_attrmulti_attr_get(
struct inode *inode,
char *name,
@@ -530,7 +521,7 @@ xfs_attrmulti_attr_get(
return error;
}
-STATIC int
+int
xfs_attrmulti_attr_set(
struct inode *inode,
char *name,
@@ -560,7 +551,7 @@ xfs_attrmulti_attr_set(
return error;
}
-STATIC int
+int
xfs_attrmulti_attr_remove(
struct inode *inode,
char *name,
@@ -662,19 +653,26 @@ xfs_attrmulti_by_handle(
return -error;
}
-STATIC int
+int
xfs_ioc_space(
struct xfs_inode *ip,
struct inode *inode,
struct file *filp,
int ioflags,
unsigned int cmd,
- void __user *arg)
+ xfs_flock64_t *bf)
{
- xfs_flock64_t bf;
int attr_flags = 0;
int error;
+ /*
+ * Only allow the sys admin to reserve space unless
+ * unwritten extents are enabled.
+ */
+ if (!xfs_sb_version_hasextflgbit(&ip->i_mount->m_sb) &&
+ !capable(CAP_SYS_ADMIN))
+ return -XFS_ERROR(EPERM);
+
if (inode->i_flags & (S_IMMUTABLE|S_APPEND))
return -XFS_ERROR(EPERM);
@@ -684,16 +682,12 @@ xfs_ioc_space(
if (!S_ISREG(inode->i_mode))
return -XFS_ERROR(EINVAL);
- if (copy_from_user(&bf, arg, sizeof(bf)))
- return -XFS_ERROR(EFAULT);
-
if (filp->f_flags & (O_NDELAY|O_NONBLOCK))
attr_flags |= XFS_ATTR_NONBLOCK;
if (ioflags & IO_INVIS)
attr_flags |= XFS_ATTR_DMI;
- error = xfs_change_file_space(ip, cmd, &bf, filp->f_pos,
- NULL, attr_flags);
+ error = xfs_change_file_space(ip, cmd, bf, filp->f_pos, attr_flags);
return -error;
}
@@ -1105,10 +1099,6 @@ xfs_ioctl_setattr(
/*
* Change file ownership. Must be the owner or privileged.
- * If the system was configured with the "restricted_chown"
- * option, the owner is not permitted to give away the file,
- * and can change the group id only to a group of which he
- * or she is a member.
*/
if (mask & FSX_PROJID) {
/*
@@ -1137,7 +1127,7 @@ xfs_ioctl_setattr(
* the superblock version number since projids didn't
* exist before DINODE_VERSION_2 and SB_VERSION_NLINK.
*/
- if (ip->i_d.di_version == XFS_DINODE_VERSION_1)
+ if (ip->i_d.di_version == 1)
xfs_bump_ino_vers2(tp, ip);
}
@@ -1256,43 +1246,67 @@ xfs_ioc_setxflags(
}
STATIC int
+xfs_getbmap_format(void **ap, struct getbmapx *bmv, int *full)
+{
+ struct getbmap __user *base = *ap;
+
+ /* copy only getbmap portion (not getbmapx) */
+ if (copy_to_user(base, bmv, sizeof(struct getbmap)))
+ return XFS_ERROR(EFAULT);
+
+ *ap += sizeof(struct getbmap);
+ return 0;
+}
+
+STATIC int
xfs_ioc_getbmap(
struct xfs_inode *ip,
int ioflags,
unsigned int cmd,
void __user *arg)
{
- struct getbmap bm;
- int iflags;
+ struct getbmapx bmx;
int error;
- if (copy_from_user(&bm, arg, sizeof(bm)))
+ if (copy_from_user(&bmx, arg, sizeof(struct getbmapx)))
return -XFS_ERROR(EFAULT);
- if (bm.bmv_count < 2)
+ if (bmx.bmv_count < 2)
return -XFS_ERROR(EINVAL);
- iflags = (cmd == XFS_IOC_GETBMAPA ? BMV_IF_ATTRFORK : 0);
+ bmx.bmv_iflags = (cmd == XFS_IOC_GETBMAPA ? BMV_IF_ATTRFORK : 0);
if (ioflags & IO_INVIS)
- iflags |= BMV_IF_NO_DMAPI_READ;
+ bmx.bmv_iflags |= BMV_IF_NO_DMAPI_READ;
- error = xfs_getbmap(ip, &bm, (struct getbmap __user *)arg+1, iflags);
+ error = xfs_getbmap(ip, &bmx, xfs_getbmap_format,
+ (struct getbmap *)arg+1);
if (error)
return -error;
- if (copy_to_user(arg, &bm, sizeof(bm)))
+ /* copy back header - only size of getbmap */
+ if (copy_to_user(arg, &bmx, sizeof(struct getbmap)))
return -XFS_ERROR(EFAULT);
return 0;
}
STATIC int
+xfs_getbmapx_format(void **ap, struct getbmapx *bmv, int *full)
+{
+ struct getbmapx __user *base = *ap;
+
+ if (copy_to_user(base, bmv, sizeof(struct getbmapx)))
+ return XFS_ERROR(EFAULT);
+
+ *ap += sizeof(struct getbmapx);
+ return 0;
+}
+
+STATIC int
xfs_ioc_getbmapx(
struct xfs_inode *ip,
void __user *arg)
{
struct getbmapx bmx;
- struct getbmap bm;
- int iflags;
int error;
if (copy_from_user(&bmx, arg, sizeof(bmx)))
@@ -1301,46 +1315,46 @@ xfs_ioc_getbmapx(
if (bmx.bmv_count < 2)
return -XFS_ERROR(EINVAL);
- /*
- * Map input getbmapx structure to a getbmap
- * structure for xfs_getbmap.
- */
- GETBMAP_CONVERT(bmx, bm);
-
- iflags = bmx.bmv_iflags;
-
- if (iflags & (~BMV_IF_VALID))
+ if (bmx.bmv_iflags & (~BMV_IF_VALID))
return -XFS_ERROR(EINVAL);
- iflags |= BMV_IF_EXTENDED;
-
- error = xfs_getbmap(ip, &bm, (struct getbmapx __user *)arg+1, iflags);
+ error = xfs_getbmap(ip, &bmx, xfs_getbmapx_format,
+ (struct getbmapx *)arg+1);
if (error)
return -error;
- GETBMAP_CONVERT(bm, bmx);
-
- if (copy_to_user(arg, &bmx, sizeof(bmx)))
+ /* copy back header */
+ if (copy_to_user(arg, &bmx, sizeof(struct getbmapx)))
return -XFS_ERROR(EFAULT);
return 0;
}
-int
-xfs_ioctl(
- xfs_inode_t *ip,
+/*
+ * Note: some of the ioctl's return positive numbers as a
+ * byte count indicating success, such as readlink_by_handle.
+ * So we don't "sign flip" like most other routines. This means
+ * true errors need to be returned as a negative value.
+ */
+long
+xfs_file_ioctl(
struct file *filp,
- int ioflags,
unsigned int cmd,
- void __user *arg)
+ unsigned long p)
{
struct inode *inode = filp->f_path.dentry->d_inode;
- xfs_mount_t *mp = ip->i_mount;
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
+ void __user *arg = (void __user *)p;
+ int ioflags = 0;
int error;
- xfs_itrace_entry(XFS_I(inode));
- switch (cmd) {
+ if (filp->f_mode & FMODE_NOCMTIME)
+ ioflags |= IO_INVIS;
+ xfs_itrace_entry(ip);
+
+ switch (cmd) {
case XFS_IOC_ALLOCSP:
case XFS_IOC_FREESP:
case XFS_IOC_RESVSP:
@@ -1348,17 +1362,13 @@ xfs_ioctl(
case XFS_IOC_ALLOCSP64:
case XFS_IOC_FREESP64:
case XFS_IOC_RESVSP64:
- case XFS_IOC_UNRESVSP64:
- /*
- * Only allow the sys admin to reserve space unless
- * unwritten extents are enabled.
- */
- if (!xfs_sb_version_hasextflgbit(&mp->m_sb) &&
- !capable(CAP_SYS_ADMIN))
- return -EPERM;
-
- return xfs_ioc_space(ip, inode, filp, ioflags, cmd, arg);
+ case XFS_IOC_UNRESVSP64: {
+ xfs_flock64_t bf;
+ if (copy_from_user(&bf, arg, sizeof(bf)))
+ return -XFS_ERROR(EFAULT);
+ return xfs_ioc_space(ip, inode, filp, ioflags, cmd, &bf);
+ }
case XFS_IOC_DIOINFO: {
struct dioattr da;
xfs_buftarg_t *target =
@@ -1418,18 +1428,30 @@ xfs_ioctl(
case XFS_IOC_FD_TO_HANDLE:
case XFS_IOC_PATH_TO_HANDLE:
- case XFS_IOC_PATH_TO_FSHANDLE:
- return xfs_find_handle(cmd, arg);
+ case XFS_IOC_PATH_TO_FSHANDLE: {
+ xfs_fsop_handlereq_t hreq;
- case XFS_IOC_OPEN_BY_HANDLE:
- return xfs_open_by_handle(mp, arg, filp, inode);
+ if (copy_from_user(&hreq, arg, sizeof(hreq)))
+ return -XFS_ERROR(EFAULT);
+ return xfs_find_handle(cmd, &hreq);
+ }
+ case XFS_IOC_OPEN_BY_HANDLE: {
+ xfs_fsop_handlereq_t hreq;
+ if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t)))
+ return -XFS_ERROR(EFAULT);
+ return xfs_open_by_handle(mp, &hreq, filp, inode);
+ }
case XFS_IOC_FSSETDM_BY_HANDLE:
return xfs_fssetdm_by_handle(mp, arg, inode);
- case XFS_IOC_READLINK_BY_HANDLE:
- return xfs_readlink_by_handle(mp, arg, inode);
+ case XFS_IOC_READLINK_BY_HANDLE: {
+ xfs_fsop_handlereq_t hreq;
+ if (copy_from_user(&hreq, arg, sizeof(xfs_fsop_handlereq_t)))
+ return -XFS_ERROR(EFAULT);
+ return xfs_readlink_by_handle(mp, &hreq, inode);
+ }
case XFS_IOC_ATTRLIST_BY_HANDLE:
return xfs_attrlist_by_handle(mp, arg, inode);
@@ -1437,7 +1459,11 @@ xfs_ioctl(
return xfs_attrmulti_by_handle(mp, arg, filp, inode);
case XFS_IOC_SWAPEXT: {
- error = xfs_swapext((struct xfs_swapext __user *)arg);
+ struct xfs_swapext sxp;
+
+ if (copy_from_user(&sxp, arg, sizeof(xfs_swapext_t)))
+ return -XFS_ERROR(EFAULT);
+ error = xfs_swapext(&sxp);
return -error;
}
@@ -1493,9 +1519,6 @@ xfs_ioctl(
case XFS_IOC_FSGROWFSDATA: {
xfs_growfs_data_t in;
- if (!capable(CAP_SYS_ADMIN))
- return -EPERM;
-
if (copy_from_user(&in, arg, sizeof(in)))
return -XFS_ERROR(EFAULT);
@@ -1506,9 +1529,6 @@ xfs_ioctl(
case XFS_IOC_FSGROWFSLOG: {
xfs_growfs_log_t in;
- if (!capable(CAP_SYS_ADMIN))
- return -EPERM;
-
if (copy_from_user(&in, arg, sizeof(in)))
return -XFS_ERROR(EFAULT);
@@ -1519,9 +1539,6 @@ xfs_ioctl(
case XFS_IOC_FSGROWFSRT: {
xfs_growfs_rt_t in;
- if (!capable(CAP_SYS_ADMIN))
- return -EPERM;
-
if (copy_from_user(&in, arg, sizeof(in)))
return -XFS_ERROR(EFAULT);
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.h b/fs/xfs/linux-2.6/xfs_ioctl.h
new file mode 100644
index 000000000000..8c16bf2d7e03
--- /dev/null
+++ b/fs/xfs/linux-2.6/xfs_ioctl.h
@@ -0,0 +1,82 @@
+/*
+ * Copyright (c) 2008 Silicon Graphics, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+#ifndef __XFS_IOCTL_H__
+#define __XFS_IOCTL_H__
+
+extern int
+xfs_ioc_space(
+ struct xfs_inode *ip,
+ struct inode *inode,
+ struct file *filp,
+ int ioflags,
+ unsigned int cmd,
+ xfs_flock64_t *bf);
+
+extern int
+xfs_find_handle(
+ unsigned int cmd,
+ xfs_fsop_handlereq_t *hreq);
+
+extern int
+xfs_open_by_handle(
+ xfs_mount_t *mp,
+ xfs_fsop_handlereq_t *hreq,
+ struct file *parfilp,
+ struct inode *parinode);
+
+extern int
+xfs_readlink_by_handle(
+ xfs_mount_t *mp,
+ xfs_fsop_handlereq_t *hreq,
+ struct inode *parinode);
+
+extern int
+xfs_attrmulti_attr_get(
+ struct inode *inode,
+ char *name,
+ char __user *ubuf,
+ __uint32_t *len,
+ __uint32_t flags);
+
+extern int
+ xfs_attrmulti_attr_set(
+ struct inode *inode,
+ char *name,
+ const char __user *ubuf,
+ __uint32_t len,
+ __uint32_t flags);
+
+extern int
+xfs_attrmulti_attr_remove(
+ struct inode *inode,
+ char *name,
+ __uint32_t flags);
+
+extern long
+xfs_file_ioctl(
+ struct file *filp,
+ unsigned int cmd,
+ unsigned long p);
+
+extern long
+xfs_file_compat_ioctl(
+ struct file *file,
+ unsigned int cmd,
+ unsigned long arg);
+
+#endif
diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.c b/fs/xfs/linux-2.6/xfs_ioctl32.c
index a4b254eb43b2..0504cece9f66 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl32.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl32.c
@@ -16,11 +16,7 @@
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include <linux/compat.h>
-#include <linux/init.h>
#include <linux/ioctl.h>
-#include <linux/syscalls.h>
-#include <linux/types.h>
-#include <linux/fs.h>
#include <asm/uaccess.h>
#include "xfs.h"
#include "xfs_fs.h"
@@ -36,7 +32,6 @@
#include "xfs_bmap_btree.h"
#include "xfs_attr_sf.h"
#include "xfs_dir2_sf.h"
-#include "xfs_vfs.h"
#include "xfs_vnode.h"
#include "xfs_dinode.h"
#include "xfs_inode.h"
@@ -44,221 +39,219 @@
#include "xfs_error.h"
#include "xfs_dfrag.h"
#include "xfs_vnodeops.h"
+#include "xfs_fsops.h"
+#include "xfs_alloc.h"
+#include "xfs_rtalloc.h"
+#include "xfs_attr.h"
+#include "xfs_ioctl.h"
#include "xfs_ioctl32.h"
#define _NATIVE_IOC(cmd, type) \
_IOC(_IOC_DIR(cmd), _IOC_TYPE(cmd), _IOC_NR(cmd), sizeof(type))
-#if defined(CONFIG_IA64) || defined(CONFIG_X86_64)
-#define BROKEN_X86_ALIGNMENT
-#define _PACKED __attribute__((packed))
-/* on ia32 l_start is on a 32-bit boundary */
-typedef struct xfs_flock64_32 {
- __s16 l_type;
- __s16 l_whence;
- __s64 l_start __attribute__((packed));
- /* len == 0 means until end of file */
- __s64 l_len __attribute__((packed));
- __s32 l_sysid;
- __u32 l_pid;
- __s32 l_pad[4]; /* reserve area */
-} xfs_flock64_32_t;
-
-#define XFS_IOC_ALLOCSP_32 _IOW ('X', 10, struct xfs_flock64_32)
-#define XFS_IOC_FREESP_32 _IOW ('X', 11, struct xfs_flock64_32)
-#define XFS_IOC_ALLOCSP64_32 _IOW ('X', 36, struct xfs_flock64_32)
-#define XFS_IOC_FREESP64_32 _IOW ('X', 37, struct xfs_flock64_32)
-#define XFS_IOC_RESVSP_32 _IOW ('X', 40, struct xfs_flock64_32)
-#define XFS_IOC_UNRESVSP_32 _IOW ('X', 41, struct xfs_flock64_32)
-#define XFS_IOC_RESVSP64_32 _IOW ('X', 42, struct xfs_flock64_32)
-#define XFS_IOC_UNRESVSP64_32 _IOW ('X', 43, struct xfs_flock64_32)
-
-/* just account for different alignment */
-STATIC unsigned long
-xfs_ioctl32_flock(
- unsigned long arg)
+#ifdef BROKEN_X86_ALIGNMENT
+STATIC int
+xfs_compat_flock64_copyin(
+ xfs_flock64_t *bf,
+ compat_xfs_flock64_t __user *arg32)
{
- xfs_flock64_32_t __user *p32 = (void __user *)arg;
- xfs_flock64_t __user *p = compat_alloc_user_space(sizeof(*p));
-
- if (copy_in_user(&p->l_type, &p32->l_type, sizeof(s16)) ||
- copy_in_user(&p->l_whence, &p32->l_whence, sizeof(s16)) ||
- copy_in_user(&p->l_start, &p32->l_start, sizeof(s64)) ||
- copy_in_user(&p->l_len, &p32->l_len, sizeof(s64)) ||
- copy_in_user(&p->l_sysid, &p32->l_sysid, sizeof(s32)) ||
- copy_in_user(&p->l_pid, &p32->l_pid, sizeof(u32)) ||
- copy_in_user(&p->l_pad, &p32->l_pad, 4*sizeof(u32)))
- return -EFAULT;
-
- return (unsigned long)p;
+ if (get_user(bf->l_type, &arg32->l_type) ||
+ get_user(bf->l_whence, &arg32->l_whence) ||
+ get_user(bf->l_start, &arg32->l_start) ||
+ get_user(bf->l_len, &arg32->l_len) ||
+ get_user(bf->l_sysid, &arg32->l_sysid) ||
+ get_user(bf->l_pid, &arg32->l_pid) ||
+ copy_from_user(bf->l_pad, &arg32->l_pad, 4*sizeof(u32)))
+ return -XFS_ERROR(EFAULT);
+ return 0;
}
-typedef struct compat_xfs_fsop_geom_v1 {
- __u32 blocksize; /* filesystem (data) block size */
- __u32 rtextsize; /* realtime extent size */
- __u32 agblocks; /* fsblocks in an AG */
- __u32 agcount; /* number of allocation groups */
- __u32 logblocks; /* fsblocks in the log */
- __u32 sectsize; /* (data) sector size, bytes */
- __u32 inodesize; /* inode size in bytes */
- __u32 imaxpct; /* max allowed inode space(%) */
- __u64 datablocks; /* fsblocks in data subvolume */
- __u64 rtblocks; /* fsblocks in realtime subvol */
- __u64 rtextents; /* rt extents in realtime subvol*/
- __u64 logstart; /* starting fsblock of the log */
- unsigned char uuid[16]; /* unique id of the filesystem */
- __u32 sunit; /* stripe unit, fsblocks */
- __u32 swidth; /* stripe width, fsblocks */
- __s32 version; /* structure version */
- __u32 flags; /* superblock version flags */
- __u32 logsectsize; /* log sector size, bytes */
- __u32 rtsectsize; /* realtime sector size, bytes */
- __u32 dirblocksize; /* directory block size, bytes */
-} __attribute__((packed)) compat_xfs_fsop_geom_v1_t;
-
-#define XFS_IOC_FSGEOMETRY_V1_32 \
- _IOR ('X', 100, struct compat_xfs_fsop_geom_v1)
-
-STATIC unsigned long xfs_ioctl32_geom_v1(unsigned long arg)
+STATIC int
+xfs_compat_ioc_fsgeometry_v1(
+ struct xfs_mount *mp,
+ compat_xfs_fsop_geom_v1_t __user *arg32)
{
- compat_xfs_fsop_geom_v1_t __user *p32 = (void __user *)arg;
- xfs_fsop_geom_v1_t __user *p = compat_alloc_user_space(sizeof(*p));
+ xfs_fsop_geom_t fsgeo;
+ int error;
- if (copy_in_user(p, p32, sizeof(*p32)))
- return -EFAULT;
- return (unsigned long)p;
+ error = xfs_fs_geometry(mp, &fsgeo, 3);
+ if (error)
+ return -error;
+ /* The 32-bit variant simply has some padding at the end */
+ if (copy_to_user(arg32, &fsgeo, sizeof(struct compat_xfs_fsop_geom_v1)))
+ return -XFS_ERROR(EFAULT);
+ return 0;
}
-typedef struct compat_xfs_inogrp {
- __u64 xi_startino; /* starting inode number */
- __s32 xi_alloccount; /* # bits set in allocmask */
- __u64 xi_allocmask; /* mask of allocated inodes */
-} __attribute__((packed)) compat_xfs_inogrp_t;
-
-STATIC int xfs_inumbers_fmt_compat(
- void __user *ubuffer,
- const xfs_inogrp_t *buffer,
- long count,
- long *written)
+STATIC int
+xfs_compat_growfs_data_copyin(
+ struct xfs_growfs_data *in,
+ compat_xfs_growfs_data_t __user *arg32)
{
- compat_xfs_inogrp_t __user *p32 = ubuffer;
- long i;
+ if (get_user(in->newblocks, &arg32->newblocks) ||
+ get_user(in->imaxpct, &arg32->imaxpct))
+ return -XFS_ERROR(EFAULT);
+ return 0;
+}
+
+STATIC int
+xfs_compat_growfs_rt_copyin(
+ struct xfs_growfs_rt *in,
+ compat_xfs_growfs_rt_t __user *arg32)
+{
+ if (get_user(in->newblocks, &arg32->newblocks) ||
+ get_user(in->extsize, &arg32->extsize))
+ return -XFS_ERROR(EFAULT);
+ return 0;
+}
+
+STATIC int
+xfs_inumbers_fmt_compat(
+ void __user *ubuffer,
+ const xfs_inogrp_t *buffer,
+ long count,
+ long *written)
+{
+ compat_xfs_inogrp_t __user *p32 = ubuffer;
+ long i;
for (i = 0; i < count; i++) {
if (put_user(buffer[i].xi_startino, &p32[i].xi_startino) ||
put_user(buffer[i].xi_alloccount, &p32[i].xi_alloccount) ||
put_user(buffer[i].xi_allocmask, &p32[i].xi_allocmask))
- return -EFAULT;
+ return -XFS_ERROR(EFAULT);
}
*written = count * sizeof(*p32);
return 0;
}
#else
-
#define xfs_inumbers_fmt_compat xfs_inumbers_fmt
-#define _PACKED
+#endif /* BROKEN_X86_ALIGNMENT */
-#endif
+STATIC int
+xfs_ioctl32_bstime_copyin(
+ xfs_bstime_t *bstime,
+ compat_xfs_bstime_t __user *bstime32)
+{
+ compat_time_t sec32; /* tv_sec differs on 64 vs. 32 */
-/* XFS_IOC_FSBULKSTAT and friends */
+ if (get_user(sec32, &bstime32->tv_sec) ||
+ get_user(bstime->tv_nsec, &bstime32->tv_nsec))
+ return -XFS_ERROR(EFAULT);
+ bstime->tv_sec = sec32;
+ return 0;
+}
+
+/* xfs_bstat_t has differing alignment on intel, & bstime_t sizes everywhere */
+STATIC int
+xfs_ioctl32_bstat_copyin(
+ xfs_bstat_t *bstat,
+ compat_xfs_bstat_t __user *bstat32)
+{
+ if (get_user(bstat->bs_ino, &bstat32->bs_ino) ||
+ get_user(bstat->bs_mode, &bstat32->bs_mode) ||
+ get_user(bstat->bs_nlink, &bstat32->bs_nlink) ||
+ get_user(bstat->bs_uid, &bstat32->bs_uid) ||
+ get_user(bstat->bs_gid, &bstat32->bs_gid) ||
+ get_user(bstat->bs_rdev, &bstat32->bs_rdev) ||
+ get_user(bstat->bs_blksize, &bstat32->bs_blksize) ||
+ get_user(bstat->bs_size, &bstat32->bs_size) ||
+ xfs_ioctl32_bstime_copyin(&bstat->bs_atime, &bstat32->bs_atime) ||
+ xfs_ioctl32_bstime_copyin(&bstat->bs_mtime, &bstat32->bs_mtime) ||
+ xfs_ioctl32_bstime_copyin(&bstat->bs_ctime, &bstat32->bs_ctime) ||
+ get_user(bstat->bs_blocks, &bstat32->bs_size) ||
+ get_user(bstat->bs_xflags, &bstat32->bs_size) ||
+ get_user(bstat->bs_extsize, &bstat32->bs_extsize) ||
+ get_user(bstat->bs_extents, &bstat32->bs_extents) ||
+ get_user(bstat->bs_gen, &bstat32->bs_gen) ||
+ get_user(bstat->bs_projid, &bstat32->bs_projid) ||
+ get_user(bstat->bs_dmevmask, &bstat32->bs_dmevmask) ||
+ get_user(bstat->bs_dmstate, &bstat32->bs_dmstate) ||
+ get_user(bstat->bs_aextents, &bstat32->bs_aextents))
+ return -XFS_ERROR(EFAULT);
+ return 0;
+}
-typedef struct compat_xfs_bstime {
- __s32 tv_sec; /* seconds */
- __s32 tv_nsec; /* and nanoseconds */
-} compat_xfs_bstime_t;
+/* XFS_IOC_FSBULKSTAT and friends */
-STATIC int xfs_bstime_store_compat(
- compat_xfs_bstime_t __user *p32,
- const xfs_bstime_t *p)
+STATIC int
+xfs_bstime_store_compat(
+ compat_xfs_bstime_t __user *p32,
+ const xfs_bstime_t *p)
{
- __s32 sec32;
+ __s32 sec32;
sec32 = p->tv_sec;
if (put_user(sec32, &p32->tv_sec) ||
put_user(p->tv_nsec, &p32->tv_nsec))
- return -EFAULT;
+ return -XFS_ERROR(EFAULT);
return 0;
}
-typedef struct compat_xfs_bstat {
- __u64 bs_ino; /* inode number */
- __u16 bs_mode; /* type and mode */
- __u16 bs_nlink; /* number of links */
- __u32 bs_uid; /* user id */
- __u32 bs_gid; /* group id */
- __u32 bs_rdev; /* device value */
- __s32 bs_blksize; /* block size */
- __s64 bs_size; /* file size */
- compat_xfs_bstime_t bs_atime; /* access time */
- compat_xfs_bstime_t bs_mtime; /* modify time */
- compat_xfs_bstime_t bs_ctime; /* inode change time */
- int64_t bs_blocks; /* number of blocks */
- __u32 bs_xflags; /* extended flags */
- __s32 bs_extsize; /* extent size */
- __s32 bs_extents; /* number of extents */
- __u32 bs_gen; /* generation count */
- __u16 bs_projid; /* project id */
- unsigned char bs_pad[14]; /* pad space, unused */
- __u32 bs_dmevmask; /* DMIG event mask */
- __u16 bs_dmstate; /* DMIG state info */
- __u16 bs_aextents; /* attribute number of extents */
-} _PACKED compat_xfs_bstat_t;
-
-STATIC int xfs_bulkstat_one_fmt_compat(
+/* Return 0 on success or positive error (to xfs_bulkstat()) */
+STATIC int
+xfs_bulkstat_one_fmt_compat(
void __user *ubuffer,
+ int ubsize,
+ int *ubused,
const xfs_bstat_t *buffer)
{
- compat_xfs_bstat_t __user *p32 = ubuffer;
-
- if (put_user(buffer->bs_ino, &p32->bs_ino) ||
- put_user(buffer->bs_mode, &p32->bs_mode) ||
- put_user(buffer->bs_nlink, &p32->bs_nlink) ||
- put_user(buffer->bs_uid, &p32->bs_uid) ||
- put_user(buffer->bs_gid, &p32->bs_gid) ||
- put_user(buffer->bs_rdev, &p32->bs_rdev) ||
- put_user(buffer->bs_blksize, &p32->bs_blksize) ||
- put_user(buffer->bs_size, &p32->bs_size) ||
+ compat_xfs_bstat_t __user *p32 = ubuffer;
+
+ if (ubsize < sizeof(*p32))
+ return XFS_ERROR(ENOMEM);
+
+ if (put_user(buffer->bs_ino, &p32->bs_ino) ||
+ put_user(buffer->bs_mode, &p32->bs_mode) ||
+ put_user(buffer->bs_nlink, &p32->bs_nlink) ||
+ put_user(buffer->bs_uid, &p32->bs_uid) ||
+ put_user(buffer->bs_gid, &p32->bs_gid) ||
+ put_user(buffer->bs_rdev, &p32->bs_rdev) ||
+ put_user(buffer->bs_blksize, &p32->bs_blksize) ||
+ put_user(buffer->bs_size, &p32->bs_size) ||
xfs_bstime_store_compat(&p32->bs_atime, &buffer->bs_atime) ||
xfs_bstime_store_compat(&p32->bs_mtime, &buffer->bs_mtime) ||
xfs_bstime_store_compat(&p32->bs_ctime, &buffer->bs_ctime) ||
- put_user(buffer->bs_blocks, &p32->bs_blocks) ||
- put_user(buffer->bs_xflags, &p32->bs_xflags) ||
- put_user(buffer->bs_extsize, &p32->bs_extsize) ||
- put_user(buffer->bs_extents, &p32->bs_extents) ||
- put_user(buffer->bs_gen, &p32->bs_gen) ||
- put_user(buffer->bs_projid, &p32->bs_projid) ||
- put_user(buffer->bs_dmevmask, &p32->bs_dmevmask) ||
- put_user(buffer->bs_dmstate, &p32->bs_dmstate) ||
+ put_user(buffer->bs_blocks, &p32->bs_blocks) ||
+ put_user(buffer->bs_xflags, &p32->bs_xflags) ||
+ put_user(buffer->bs_extsize, &p32->bs_extsize) ||
+ put_user(buffer->bs_extents, &p32->bs_extents) ||
+ put_user(buffer->bs_gen, &p32->bs_gen) ||
+ put_user(buffer->bs_projid, &p32->bs_projid) ||
+ put_user(buffer->bs_dmevmask, &p32->bs_dmevmask) ||
+ put_user(buffer->bs_dmstate, &p32->bs_dmstate) ||
put_user(buffer->bs_aextents, &p32->bs_aextents))
- return -EFAULT;
- return sizeof(*p32);
+ return XFS_ERROR(EFAULT);
+ if (ubused)
+ *ubused = sizeof(*p32);
+ return 0;
}
-
-
-typedef struct compat_xfs_fsop_bulkreq {
- compat_uptr_t lastip; /* last inode # pointer */
- __s32 icount; /* count of entries in buffer */
- compat_uptr_t ubuffer; /* user buffer for inode desc. */
- compat_uptr_t ocount; /* output count pointer */
-} compat_xfs_fsop_bulkreq_t;
-
-#define XFS_IOC_FSBULKSTAT_32 \
- _IOWR('X', 101, struct compat_xfs_fsop_bulkreq)
-#define XFS_IOC_FSBULKSTAT_SINGLE_32 \
- _IOWR('X', 102, struct compat_xfs_fsop_bulkreq)
-#define XFS_IOC_FSINUMBERS_32 \
- _IOWR('X', 103, struct compat_xfs_fsop_bulkreq)
+STATIC int
+xfs_bulkstat_one_compat(
+ xfs_mount_t *mp, /* mount point for filesystem */
+ xfs_ino_t ino, /* inode number to get data for */
+ void __user *buffer, /* buffer to place output in */
+ int ubsize, /* size of buffer */
+ void *private_data, /* my private data */
+ xfs_daddr_t bno, /* starting bno of inode cluster */
+ int *ubused, /* bytes used by me */
+ void *dibuff, /* on-disk inode buffer */
+ int *stat) /* BULKSTAT_RV_... */
+{
+ return xfs_bulkstat_one_int(mp, ino, buffer, ubsize,
+ xfs_bulkstat_one_fmt_compat, bno,
+ ubused, dibuff, stat);
+}
/* copied from xfs_ioctl.c */
STATIC int
-xfs_ioc_bulkstat_compat(
- xfs_mount_t *mp,
- unsigned int cmd,
- void __user *arg)
+xfs_compat_ioc_bulkstat(
+ xfs_mount_t *mp,
+ unsigned int cmd,
+ compat_xfs_fsop_bulkreq_t __user *p32)
{
- compat_xfs_fsop_bulkreq_t __user *p32 = (void __user *)arg;
u32 addr;
xfs_fsop_bulkreq_t bulkreq;
int count; /* # of records returned */
@@ -270,20 +263,20 @@ xfs_ioc_bulkstat_compat(
/* should be called again (unused here, but used in dmapi) */
if (!capable(CAP_SYS_ADMIN))
- return -EPERM;
+ return -XFS_ERROR(EPERM);
if (XFS_FORCED_SHUTDOWN(mp))
return -XFS_ERROR(EIO);
if (get_user(addr, &p32->lastip))
- return -EFAULT;
+ return -XFS_ERROR(EFAULT);
bulkreq.lastip = compat_ptr(addr);
if (get_user(bulkreq.icount, &p32->icount) ||
get_user(addr, &p32->ubuffer))
- return -EFAULT;
+ return -XFS_ERROR(EFAULT);
bulkreq.ubuffer = compat_ptr(addr);
if (get_user(addr, &p32->ocount))
- return -EFAULT;
+ return -XFS_ERROR(EFAULT);
bulkreq.ocount = compat_ptr(addr);
if (copy_from_user(&inlast, bulkreq.lastip, sizeof(__s64)))
@@ -295,17 +288,22 @@ xfs_ioc_bulkstat_compat(
if (bulkreq.ubuffer == NULL)
return -XFS_ERROR(EINVAL);
- if (cmd == XFS_IOC_FSINUMBERS)
+ if (cmd == XFS_IOC_FSINUMBERS_32) {
error = xfs_inumbers(mp, &inlast, &count,
bulkreq.ubuffer, xfs_inumbers_fmt_compat);
- else {
- /* declare a var to get a warning in case the type changes */
- bulkstat_one_fmt_pf formatter = xfs_bulkstat_one_fmt_compat;
+ } else if (cmd == XFS_IOC_FSBULKSTAT_SINGLE_32) {
+ int res;
+
+ error = xfs_bulkstat_one_compat(mp, inlast, bulkreq.ubuffer,
+ sizeof(compat_xfs_bstat_t),
+ NULL, 0, NULL, NULL, &res);
+ } else if (cmd == XFS_IOC_FSBULKSTAT_32) {
error = xfs_bulkstat(mp, &inlast, &count,
- xfs_bulkstat_one, formatter,
+ xfs_bulkstat_one_compat, NULL,
sizeof(compat_xfs_bstat_t), bulkreq.ubuffer,
BULKSTAT_FG_QUICK, &done);
- }
+ } else
+ error = XFS_ERROR(EINVAL);
if (error)
return -error;
@@ -321,63 +319,306 @@ xfs_ioc_bulkstat_compat(
return 0;
}
+STATIC int
+xfs_compat_handlereq_copyin(
+ xfs_fsop_handlereq_t *hreq,
+ compat_xfs_fsop_handlereq_t __user *arg32)
+{
+ compat_xfs_fsop_handlereq_t hreq32;
+
+ if (copy_from_user(&hreq32, arg32, sizeof(compat_xfs_fsop_handlereq_t)))
+ return -XFS_ERROR(EFAULT);
+
+ hreq->fd = hreq32.fd;
+ hreq->path = compat_ptr(hreq32.path);
+ hreq->oflags = hreq32.oflags;
+ hreq->ihandle = compat_ptr(hreq32.ihandle);
+ hreq->ihandlen = hreq32.ihandlen;
+ hreq->ohandle = compat_ptr(hreq32.ohandle);
+ hreq->ohandlen = compat_ptr(hreq32.ohandlen);
+ return 0;
+}
-typedef struct compat_xfs_fsop_handlereq {
- __u32 fd; /* fd for FD_TO_HANDLE */
- compat_uptr_t path; /* user pathname */
- __u32 oflags; /* open flags */
- compat_uptr_t ihandle; /* user supplied handle */
- __u32 ihandlen; /* user supplied length */
- compat_uptr_t ohandle; /* user buffer for handle */
- compat_uptr_t ohandlen; /* user buffer length */
-} compat_xfs_fsop_handlereq_t;
-
-#define XFS_IOC_PATH_TO_FSHANDLE_32 \
- _IOWR('X', 104, struct compat_xfs_fsop_handlereq)
-#define XFS_IOC_PATH_TO_HANDLE_32 \
- _IOWR('X', 105, struct compat_xfs_fsop_handlereq)
-#define XFS_IOC_FD_TO_HANDLE_32 \
- _IOWR('X', 106, struct compat_xfs_fsop_handlereq)
-#define XFS_IOC_OPEN_BY_HANDLE_32 \
- _IOWR('X', 107, struct compat_xfs_fsop_handlereq)
-#define XFS_IOC_READLINK_BY_HANDLE_32 \
- _IOWR('X', 108, struct compat_xfs_fsop_handlereq)
-
-STATIC unsigned long xfs_ioctl32_fshandle(unsigned long arg)
+/*
+ * Convert userspace handle data into inode.
+ *
+ * We use the fact that all the fsop_handlereq ioctl calls have a data
+ * structure argument whose first component is always a xfs_fsop_handlereq_t,
+ * so we can pass that sub structure into this handy, shared routine.
+ *
+ * If no error, caller must always iput the returned inode.
+ */
+STATIC int
+xfs_vget_fsop_handlereq_compat(
+ xfs_mount_t *mp,
+ struct inode *parinode, /* parent inode pointer */
+ compat_xfs_fsop_handlereq_t *hreq,
+ struct inode **inode)
{
- compat_xfs_fsop_handlereq_t __user *p32 = (void __user *)arg;
- xfs_fsop_handlereq_t __user *p = compat_alloc_user_space(sizeof(*p));
- u32 addr;
-
- if (copy_in_user(&p->fd, &p32->fd, sizeof(__u32)) ||
- get_user(addr, &p32->path) ||
- put_user(compat_ptr(addr), &p->path) ||
- copy_in_user(&p->oflags, &p32->oflags, sizeof(__u32)) ||
- get_user(addr, &p32->ihandle) ||
- put_user(compat_ptr(addr), &p->ihandle) ||
- copy_in_user(&p->ihandlen, &p32->ihandlen, sizeof(__u32)) ||
- get_user(addr, &p32->ohandle) ||
- put_user(compat_ptr(addr), &p->ohandle) ||
- get_user(addr, &p32->ohandlen) ||
- put_user(compat_ptr(addr), &p->ohandlen))
- return -EFAULT;
-
- return (unsigned long)p;
+ void __user *hanp;
+ size_t hlen;
+ xfs_fid_t *xfid;
+ xfs_handle_t *handlep;
+ xfs_handle_t handle;
+ xfs_inode_t *ip;
+ xfs_ino_t ino;
+ __u32 igen;
+ int error;
+
+ /*
+ * Only allow handle opens under a directory.
+ */
+ if (!S_ISDIR(parinode->i_mode))
+ return XFS_ERROR(ENOTDIR);
+
+ hanp = compat_ptr(hreq->ihandle);
+ hlen = hreq->ihandlen;
+ handlep = &handle;
+
+ if (hlen < sizeof(handlep->ha_fsid) || hlen > sizeof(*handlep))
+ return XFS_ERROR(EINVAL);
+ if (copy_from_user(handlep, hanp, hlen))
+ return XFS_ERROR(EFAULT);
+ if (hlen < sizeof(*handlep))
+ memset(((char *)handlep) + hlen, 0, sizeof(*handlep) - hlen);
+ if (hlen > sizeof(handlep->ha_fsid)) {
+ if (handlep->ha_fid.fid_len !=
+ (hlen - sizeof(handlep->ha_fsid) -
+ sizeof(handlep->ha_fid.fid_len)) ||
+ handlep->ha_fid.fid_pad)
+ return XFS_ERROR(EINVAL);
+ }
+
+ /*
+ * Crack the handle, obtain the inode # & generation #
+ */
+ xfid = (struct xfs_fid *)&handlep->ha_fid;
+ if (xfid->fid_len == sizeof(*xfid) - sizeof(xfid->fid_len)) {
+ ino = xfid->fid_ino;
+ igen = xfid->fid_gen;
+ } else {
+ return XFS_ERROR(EINVAL);
+ }
+
+ /*
+ * Get the XFS inode, building a Linux inode to go with it.
+ */
+ error = xfs_iget(mp, NULL, ino, 0, XFS_ILOCK_SHARED, &ip, 0);
+ if (error)
+ return error;
+ if (ip == NULL)
+ return XFS_ERROR(EIO);
+ if (ip->i_d.di_gen != igen) {
+ xfs_iput_new(ip, XFS_ILOCK_SHARED);
+ return XFS_ERROR(ENOENT);
+ }
+
+ xfs_iunlock(ip, XFS_ILOCK_SHARED);
+
+ *inode = VFS_I(ip);
+ return 0;
}
+STATIC int
+xfs_compat_attrlist_by_handle(
+ xfs_mount_t *mp,
+ void __user *arg,
+ struct inode *parinode)
+{
+ int error;
+ attrlist_cursor_kern_t *cursor;
+ compat_xfs_fsop_attrlist_handlereq_t al_hreq;
+ struct inode *inode;
+ char *kbuf;
+
+ if (!capable(CAP_SYS_ADMIN))
+ return -XFS_ERROR(EPERM);
+ if (copy_from_user(&al_hreq, arg,
+ sizeof(compat_xfs_fsop_attrlist_handlereq_t)))
+ return -XFS_ERROR(EFAULT);
+ if (al_hreq.buflen > XATTR_LIST_MAX)
+ return -XFS_ERROR(EINVAL);
+
+ /*
+ * Reject flags, only allow namespaces.
+ */
+ if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE))
+ return -XFS_ERROR(EINVAL);
+
+ error = xfs_vget_fsop_handlereq_compat(mp, parinode, &al_hreq.hreq,
+ &inode);
+ if (error)
+ goto out;
+
+ kbuf = kmalloc(al_hreq.buflen, GFP_KERNEL);
+ if (!kbuf)
+ goto out_vn_rele;
+
+ cursor = (attrlist_cursor_kern_t *)&al_hreq.pos;
+ error = xfs_attr_list(XFS_I(inode), kbuf, al_hreq.buflen,
+ al_hreq.flags, cursor);
+ if (error)
+ goto out_kfree;
+
+ if (copy_to_user(compat_ptr(al_hreq.buffer), kbuf, al_hreq.buflen))
+ error = -EFAULT;
+
+ out_kfree:
+ kfree(kbuf);
+ out_vn_rele:
+ iput(inode);
+ out:
+ return -error;
+}
-STATIC long
-xfs_compat_ioctl(
- int mode,
- struct file *file,
- unsigned cmd,
- unsigned long arg)
+STATIC int
+xfs_compat_attrmulti_by_handle(
+ xfs_mount_t *mp,
+ void __user *arg,
+ struct inode *parinode)
+{
+ int error;
+ compat_xfs_attr_multiop_t *ops;
+ compat_xfs_fsop_attrmulti_handlereq_t am_hreq;
+ struct inode *inode;
+ unsigned int i, size;
+ char *attr_name;
+
+ if (!capable(CAP_SYS_ADMIN))
+ return -XFS_ERROR(EPERM);
+ if (copy_from_user(&am_hreq, arg,
+ sizeof(compat_xfs_fsop_attrmulti_handlereq_t)))
+ return -XFS_ERROR(EFAULT);
+
+ error = xfs_vget_fsop_handlereq_compat(mp, parinode, &am_hreq.hreq,
+ &inode);
+ if (error)
+ goto out;
+
+ error = E2BIG;
+ size = am_hreq.opcount * sizeof(compat_xfs_attr_multiop_t);
+ if (!size || size > 16 * PAGE_SIZE)
+ goto out_vn_rele;
+
+ error = ENOMEM;
+ ops = kmalloc(size, GFP_KERNEL);
+ if (!ops)
+ goto out_vn_rele;
+
+ error = EFAULT;
+ if (copy_from_user(ops, compat_ptr(am_hreq.ops), size))
+ goto out_kfree_ops;
+
+ attr_name = kmalloc(MAXNAMELEN, GFP_KERNEL);
+ if (!attr_name)
+ goto out_kfree_ops;
+
+
+ error = 0;
+ for (i = 0; i < am_hreq.opcount; i++) {
+ ops[i].am_error = strncpy_from_user(attr_name,
+ compat_ptr(ops[i].am_attrname),
+ MAXNAMELEN);
+ if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN)
+ error = -ERANGE;
+ if (ops[i].am_error < 0)
+ break;
+
+ switch (ops[i].am_opcode) {
+ case ATTR_OP_GET:
+ ops[i].am_error = xfs_attrmulti_attr_get(inode,
+ attr_name,
+ compat_ptr(ops[i].am_attrvalue),
+ &ops[i].am_length, ops[i].am_flags);
+ break;
+ case ATTR_OP_SET:
+ ops[i].am_error = xfs_attrmulti_attr_set(inode,
+ attr_name,
+ compat_ptr(ops[i].am_attrvalue),
+ ops[i].am_length, ops[i].am_flags);
+ break;
+ case ATTR_OP_REMOVE:
+ ops[i].am_error = xfs_attrmulti_attr_remove(inode,
+ attr_name, ops[i].am_flags);
+ break;
+ default:
+ ops[i].am_error = EINVAL;
+ }
+ }
+
+ if (copy_to_user(compat_ptr(am_hreq.ops), ops, size))
+ error = XFS_ERROR(EFAULT);
+
+ kfree(attr_name);
+ out_kfree_ops:
+ kfree(ops);
+ out_vn_rele:
+ iput(inode);
+ out:
+ return -error;
+}
+
+STATIC int
+xfs_compat_fssetdm_by_handle(
+ xfs_mount_t *mp,
+ void __user *arg,
+ struct inode *parinode)
+{
+ int error;
+ struct fsdmidata fsd;
+ compat_xfs_fsop_setdm_handlereq_t dmhreq;
+ struct inode *inode;
+
+ if (!capable(CAP_MKNOD))
+ return -XFS_ERROR(EPERM);
+ if (copy_from_user(&dmhreq, arg,
+ sizeof(compat_xfs_fsop_setdm_handlereq_t)))
+ return -XFS_ERROR(EFAULT);
+
+ error = xfs_vget_fsop_handlereq_compat(mp, parinode, &dmhreq.hreq,
+ &inode);
+ if (error)
+ return -error;
+
+ if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) {
+ error = -XFS_ERROR(EPERM);
+ goto out;
+ }
+
+ if (copy_from_user(&fsd, compat_ptr(dmhreq.data), sizeof(fsd))) {
+ error = -XFS_ERROR(EFAULT);
+ goto out;
+ }
+
+ error = -xfs_set_dmattrs(XFS_I(inode), fsd.fsd_dmevmask,
+ fsd.fsd_dmstate);
+
+out:
+ iput(inode);
+ return error;
+}
+
+long
+xfs_file_compat_ioctl(
+ struct file *filp,
+ unsigned cmd,
+ unsigned long p)
{
- struct inode *inode = file->f_path.dentry->d_inode;
- int error;
+ struct inode *inode = filp->f_path.dentry->d_inode;
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
+ void __user *arg = (void __user *)p;
+ int ioflags = 0;
+ int error;
+
+ if (filp->f_mode & FMODE_NOCMTIME)
+ ioflags |= IO_INVIS;
+
+ xfs_itrace_entry(ip);
switch (cmd) {
+ /* No size or alignment issues on any arch */
case XFS_IOC_DIOINFO:
case XFS_IOC_FSGEOMETRY:
case XFS_IOC_FSGETXATTR:
@@ -387,48 +628,18 @@ xfs_compat_ioctl(
case XFS_IOC_GETBMAP:
case XFS_IOC_GETBMAPA:
case XFS_IOC_GETBMAPX:
-/* not handled
- case XFS_IOC_FSSETDM_BY_HANDLE:
- case XFS_IOC_ATTRLIST_BY_HANDLE:
- case XFS_IOC_ATTRMULTI_BY_HANDLE:
-*/
case XFS_IOC_FSCOUNTS:
case XFS_IOC_SET_RESBLKS:
case XFS_IOC_GET_RESBLKS:
- case XFS_IOC_FSGROWFSDATA:
case XFS_IOC_FSGROWFSLOG:
- case XFS_IOC_FSGROWFSRT:
case XFS_IOC_FREEZE:
case XFS_IOC_THAW:
case XFS_IOC_GOINGDOWN:
case XFS_IOC_ERROR_INJECTION:
case XFS_IOC_ERROR_CLEARALL:
- break;
-
- case XFS_IOC32_GETXFLAGS:
- case XFS_IOC32_SETXFLAGS:
- case XFS_IOC32_GETVERSION:
- cmd = _NATIVE_IOC(cmd, long);
- break;
-#ifdef BROKEN_X86_ALIGNMENT
- /* xfs_flock_t has wrong u32 vs u64 alignment */
- case XFS_IOC_ALLOCSP_32:
- case XFS_IOC_FREESP_32:
- case XFS_IOC_ALLOCSP64_32:
- case XFS_IOC_FREESP64_32:
- case XFS_IOC_RESVSP_32:
- case XFS_IOC_UNRESVSP_32:
- case XFS_IOC_RESVSP64_32:
- case XFS_IOC_UNRESVSP64_32:
- arg = xfs_ioctl32_flock(arg);
- cmd = _NATIVE_IOC(cmd, struct xfs_flock64);
- break;
- case XFS_IOC_FSGEOMETRY_V1_32:
- arg = xfs_ioctl32_geom_v1(arg);
- cmd = _NATIVE_IOC(cmd, struct xfs_fsop_geom_v1);
- break;
-
-#else /* These are handled fine if no alignment issues */
+ return xfs_file_ioctl(filp, cmd, p);
+#ifndef BROKEN_X86_ALIGNMENT
+ /* These are handled fine if no alignment issues */
case XFS_IOC_ALLOCSP:
case XFS_IOC_FREESP:
case XFS_IOC_RESVSP:
@@ -438,51 +649,97 @@ xfs_compat_ioctl(
case XFS_IOC_RESVSP64:
case XFS_IOC_UNRESVSP64:
case XFS_IOC_FSGEOMETRY_V1:
- break;
+ case XFS_IOC_FSGROWFSDATA:
+ case XFS_IOC_FSGROWFSRT:
+ return xfs_file_ioctl(filp, cmd, p);
+#else
+ case XFS_IOC_ALLOCSP_32:
+ case XFS_IOC_FREESP_32:
+ case XFS_IOC_ALLOCSP64_32:
+ case XFS_IOC_FREESP64_32:
+ case XFS_IOC_RESVSP_32:
+ case XFS_IOC_UNRESVSP_32:
+ case XFS_IOC_RESVSP64_32:
+ case XFS_IOC_UNRESVSP64_32: {
+ struct xfs_flock64 bf;
- /* xfs_bstat_t still has wrong u32 vs u64 alignment */
- case XFS_IOC_SWAPEXT:
- break;
+ if (xfs_compat_flock64_copyin(&bf, arg))
+ return -XFS_ERROR(EFAULT);
+ cmd = _NATIVE_IOC(cmd, struct xfs_flock64);
+ return xfs_ioc_space(ip, inode, filp, ioflags, cmd, &bf);
+ }
+ case XFS_IOC_FSGEOMETRY_V1_32:
+ return xfs_compat_ioc_fsgeometry_v1(mp, arg);
+ case XFS_IOC_FSGROWFSDATA_32: {
+ struct xfs_growfs_data in;
+
+ if (xfs_compat_growfs_data_copyin(&in, arg))
+ return -XFS_ERROR(EFAULT);
+ error = xfs_growfs_data(mp, &in);
+ return -error;
+ }
+ case XFS_IOC_FSGROWFSRT_32: {
+ struct xfs_growfs_rt in;
+ if (xfs_compat_growfs_rt_copyin(&in, arg))
+ return -XFS_ERROR(EFAULT);
+ error = xfs_growfs_rt(mp, &in);
+ return -error;
+ }
#endif
+ /* long changes size, but xfs only copiese out 32 bits */
+ case XFS_IOC_GETXFLAGS_32:
+ case XFS_IOC_SETXFLAGS_32:
+ case XFS_IOC_GETVERSION_32:
+ cmd = _NATIVE_IOC(cmd, long);
+ return xfs_file_ioctl(filp, cmd, p);
+ case XFS_IOC_SWAPEXT: {
+ struct xfs_swapext sxp;
+ struct compat_xfs_swapext __user *sxu = arg;
+
+ /* Bulk copy in up to the sx_stat field, then copy bstat */
+ if (copy_from_user(&sxp, sxu,
+ offsetof(struct xfs_swapext, sx_stat)) ||
+ xfs_ioctl32_bstat_copyin(&sxp.sx_stat, &sxu->sx_stat))
+ return -XFS_ERROR(EFAULT);
+ error = xfs_swapext(&sxp);
+ return -error;
+ }
case XFS_IOC_FSBULKSTAT_32:
case XFS_IOC_FSBULKSTAT_SINGLE_32:
case XFS_IOC_FSINUMBERS_32:
- cmd = _NATIVE_IOC(cmd, struct xfs_fsop_bulkreq);
- return xfs_ioc_bulkstat_compat(XFS_I(inode)->i_mount,
- cmd, (void __user*)arg);
+ return xfs_compat_ioc_bulkstat(mp, cmd, arg);
case XFS_IOC_FD_TO_HANDLE_32:
case XFS_IOC_PATH_TO_HANDLE_32:
- case XFS_IOC_PATH_TO_FSHANDLE_32:
- case XFS_IOC_OPEN_BY_HANDLE_32:
- case XFS_IOC_READLINK_BY_HANDLE_32:
- arg = xfs_ioctl32_fshandle(arg);
+ case XFS_IOC_PATH_TO_FSHANDLE_32: {
+ struct xfs_fsop_handlereq hreq;
+
+ if (xfs_compat_handlereq_copyin(&hreq, arg))
+ return -XFS_ERROR(EFAULT);
cmd = _NATIVE_IOC(cmd, struct xfs_fsop_handlereq);
- break;
- default:
- return -ENOIOCTLCMD;
+ return xfs_find_handle(cmd, &hreq);
}
+ case XFS_IOC_OPEN_BY_HANDLE_32: {
+ struct xfs_fsop_handlereq hreq;
- error = xfs_ioctl(XFS_I(inode), file, mode, cmd, (void __user *)arg);
- xfs_iflags_set(XFS_I(inode), XFS_IMODIFIED);
-
- return error;
-}
-
-long
-xfs_file_compat_ioctl(
- struct file *file,
- unsigned cmd,
- unsigned long arg)
-{
- return xfs_compat_ioctl(0, file, cmd, arg);
-}
+ if (xfs_compat_handlereq_copyin(&hreq, arg))
+ return -XFS_ERROR(EFAULT);
+ return xfs_open_by_handle(mp, &hreq, filp, inode);
+ }
+ case XFS_IOC_READLINK_BY_HANDLE_32: {
+ struct xfs_fsop_handlereq hreq;
-long
-xfs_file_compat_invis_ioctl(
- struct file *file,
- unsigned cmd,
- unsigned long arg)
-{
- return xfs_compat_ioctl(IO_INVIS, file, cmd, arg);
+ if (xfs_compat_handlereq_copyin(&hreq, arg))
+ return -XFS_ERROR(EFAULT);
+ return xfs_readlink_by_handle(mp, &hreq, inode);
+ }
+ case XFS_IOC_ATTRLIST_BY_HANDLE_32:
+ return xfs_compat_attrlist_by_handle(mp, arg, inode);
+ case XFS_IOC_ATTRMULTI_BY_HANDLE_32:
+ return xfs_compat_attrmulti_by_handle(mp, arg, inode);
+ case XFS_IOC_FSSETDM_BY_HANDLE_32:
+ return xfs_compat_fssetdm_by_handle(mp, arg, inode);
+ default:
+ return -XFS_ERROR(ENOIOCTLCMD);
+ }
}
diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.h b/fs/xfs/linux-2.6/xfs_ioctl32.h
index 02de6e62ee37..1024c4f8ba0d 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl32.h
+++ b/fs/xfs/linux-2.6/xfs_ioctl32.h
@@ -18,7 +18,217 @@
#ifndef __XFS_IOCTL32_H__
#define __XFS_IOCTL32_H__
-extern long xfs_file_compat_ioctl(struct file *, unsigned, unsigned long);
-extern long xfs_file_compat_invis_ioctl(struct file *, unsigned, unsigned long);
+#include <linux/compat.h>
+
+/*
+ * on 32-bit arches, ioctl argument structures may have different sizes
+ * and/or alignment. We define compat structures which match the
+ * 32-bit sizes/alignments here, and their associated ioctl numbers.
+ *
+ * xfs_ioctl32.c contains routines to copy these structures in and out.
+ */
+
+/* stock kernel-level ioctls we support */
+#define XFS_IOC_GETXFLAGS_32 FS_IOC32_GETFLAGS
+#define XFS_IOC_SETXFLAGS_32 FS_IOC32_SETFLAGS
+#define XFS_IOC_GETVERSION_32 FS_IOC32_GETVERSION
+
+/*
+ * On intel, even if sizes match, alignment and/or padding may differ.
+ */
+#if defined(CONFIG_IA64) || defined(CONFIG_X86_64)
+#define BROKEN_X86_ALIGNMENT
+#define __compat_packed __attribute__((packed))
+#else
+#define __compat_packed
+#endif
+
+typedef struct compat_xfs_bstime {
+ compat_time_t tv_sec; /* seconds */
+ __s32 tv_nsec; /* and nanoseconds */
+} compat_xfs_bstime_t;
+
+typedef struct compat_xfs_bstat {
+ __u64 bs_ino; /* inode number */
+ __u16 bs_mode; /* type and mode */
+ __u16 bs_nlink; /* number of links */
+ __u32 bs_uid; /* user id */
+ __u32 bs_gid; /* group id */
+ __u32 bs_rdev; /* device value */
+ __s32 bs_blksize; /* block size */
+ __s64 bs_size; /* file size */
+ compat_xfs_bstime_t bs_atime; /* access time */
+ compat_xfs_bstime_t bs_mtime; /* modify time */
+ compat_xfs_bstime_t bs_ctime; /* inode change time */
+ int64_t bs_blocks; /* number of blocks */
+ __u32 bs_xflags; /* extended flags */
+ __s32 bs_extsize; /* extent size */
+ __s32 bs_extents; /* number of extents */
+ __u32 bs_gen; /* generation count */
+ __u16 bs_projid; /* project id */
+ unsigned char bs_pad[14]; /* pad space, unused */
+ __u32 bs_dmevmask; /* DMIG event mask */
+ __u16 bs_dmstate; /* DMIG state info */
+ __u16 bs_aextents; /* attribute number of extents */
+} __compat_packed compat_xfs_bstat_t;
+
+typedef struct compat_xfs_fsop_bulkreq {
+ compat_uptr_t lastip; /* last inode # pointer */
+ __s32 icount; /* count of entries in buffer */
+ compat_uptr_t ubuffer; /* user buffer for inode desc. */
+ compat_uptr_t ocount; /* output count pointer */
+} compat_xfs_fsop_bulkreq_t;
+
+#define XFS_IOC_FSBULKSTAT_32 \
+ _IOWR('X', 101, struct compat_xfs_fsop_bulkreq)
+#define XFS_IOC_FSBULKSTAT_SINGLE_32 \
+ _IOWR('X', 102, struct compat_xfs_fsop_bulkreq)
+#define XFS_IOC_FSINUMBERS_32 \
+ _IOWR('X', 103, struct compat_xfs_fsop_bulkreq)
+
+typedef struct compat_xfs_fsop_handlereq {
+ __u32 fd; /* fd for FD_TO_HANDLE */
+ compat_uptr_t path; /* user pathname */
+ __u32 oflags; /* open flags */
+ compat_uptr_t ihandle; /* user supplied handle */
+ __u32 ihandlen; /* user supplied length */
+ compat_uptr_t ohandle; /* user buffer for handle */
+ compat_uptr_t ohandlen; /* user buffer length */
+} compat_xfs_fsop_handlereq_t;
+
+#define XFS_IOC_PATH_TO_FSHANDLE_32 \
+ _IOWR('X', 104, struct compat_xfs_fsop_handlereq)
+#define XFS_IOC_PATH_TO_HANDLE_32 \
+ _IOWR('X', 105, struct compat_xfs_fsop_handlereq)
+#define XFS_IOC_FD_TO_HANDLE_32 \
+ _IOWR('X', 106, struct compat_xfs_fsop_handlereq)
+#define XFS_IOC_OPEN_BY_HANDLE_32 \
+ _IOWR('X', 107, struct compat_xfs_fsop_handlereq)
+#define XFS_IOC_READLINK_BY_HANDLE_32 \
+ _IOWR('X', 108, struct compat_xfs_fsop_handlereq)
+
+/* The bstat field in the swapext struct needs translation */
+typedef struct compat_xfs_swapext {
+ __int64_t sx_version; /* version */
+ __int64_t sx_fdtarget; /* fd of target file */
+ __int64_t sx_fdtmp; /* fd of tmp file */
+ xfs_off_t sx_offset; /* offset into file */
+ xfs_off_t sx_length; /* leng from offset */
+ char sx_pad[16]; /* pad space, unused */
+ compat_xfs_bstat_t sx_stat; /* stat of target b4 copy */
+} __compat_packed compat_xfs_swapext_t;
+
+#define XFS_IOC_SWAPEXT_32 _IOWR('X', 109, struct compat_xfs_swapext)
+
+typedef struct compat_xfs_fsop_attrlist_handlereq {
+ struct compat_xfs_fsop_handlereq hreq; /* handle interface structure */
+ struct xfs_attrlist_cursor pos; /* opaque cookie, list offset */
+ __u32 flags; /* which namespace to use */
+ __u32 buflen; /* length of buffer supplied */
+ compat_uptr_t buffer; /* returned names */
+} __compat_packed compat_xfs_fsop_attrlist_handlereq_t;
+
+/* Note: actually this is read/write */
+#define XFS_IOC_ATTRLIST_BY_HANDLE_32 \
+ _IOW('X', 122, struct compat_xfs_fsop_attrlist_handlereq)
+
+/* am_opcodes defined in xfs_fs.h */
+typedef struct compat_xfs_attr_multiop {
+ __u32 am_opcode;
+ __s32 am_error;
+ compat_uptr_t am_attrname;
+ compat_uptr_t am_attrvalue;
+ __u32 am_length;
+ __u32 am_flags;
+} compat_xfs_attr_multiop_t;
+
+typedef struct compat_xfs_fsop_attrmulti_handlereq {
+ struct compat_xfs_fsop_handlereq hreq; /* handle interface structure */
+ __u32 opcount;/* count of following multiop */
+ /* ptr to compat_xfs_attr_multiop */
+ compat_uptr_t ops; /* attr_multi data */
+} compat_xfs_fsop_attrmulti_handlereq_t;
+
+#define XFS_IOC_ATTRMULTI_BY_HANDLE_32 \
+ _IOW('X', 123, struct compat_xfs_fsop_attrmulti_handlereq)
+
+typedef struct compat_xfs_fsop_setdm_handlereq {
+ struct compat_xfs_fsop_handlereq hreq; /* handle information */
+ /* ptr to struct fsdmidata */
+ compat_uptr_t data; /* DMAPI data */
+} compat_xfs_fsop_setdm_handlereq_t;
+
+#define XFS_IOC_FSSETDM_BY_HANDLE_32 \
+ _IOW('X', 121, struct compat_xfs_fsop_setdm_handlereq)
+
+#ifdef BROKEN_X86_ALIGNMENT
+/* on ia32 l_start is on a 32-bit boundary */
+typedef struct compat_xfs_flock64 {
+ __s16 l_type;
+ __s16 l_whence;
+ __s64 l_start __attribute__((packed));
+ /* len == 0 means until end of file */
+ __s64 l_len __attribute__((packed));
+ __s32 l_sysid;
+ __u32 l_pid;
+ __s32 l_pad[4]; /* reserve area */
+} compat_xfs_flock64_t;
+
+#define XFS_IOC_ALLOCSP_32 _IOW('X', 10, struct compat_xfs_flock64)
+#define XFS_IOC_FREESP_32 _IOW('X', 11, struct compat_xfs_flock64)
+#define XFS_IOC_ALLOCSP64_32 _IOW('X', 36, struct compat_xfs_flock64)
+#define XFS_IOC_FREESP64_32 _IOW('X', 37, struct compat_xfs_flock64)
+#define XFS_IOC_RESVSP_32 _IOW('X', 40, struct compat_xfs_flock64)
+#define XFS_IOC_UNRESVSP_32 _IOW('X', 41, struct compat_xfs_flock64)
+#define XFS_IOC_RESVSP64_32 _IOW('X', 42, struct compat_xfs_flock64)
+#define XFS_IOC_UNRESVSP64_32 _IOW('X', 43, struct compat_xfs_flock64)
+
+typedef struct compat_xfs_fsop_geom_v1 {
+ __u32 blocksize; /* filesystem (data) block size */
+ __u32 rtextsize; /* realtime extent size */
+ __u32 agblocks; /* fsblocks in an AG */
+ __u32 agcount; /* number of allocation groups */
+ __u32 logblocks; /* fsblocks in the log */
+ __u32 sectsize; /* (data) sector size, bytes */
+ __u32 inodesize; /* inode size in bytes */
+ __u32 imaxpct; /* max allowed inode space(%) */
+ __u64 datablocks; /* fsblocks in data subvolume */
+ __u64 rtblocks; /* fsblocks in realtime subvol */
+ __u64 rtextents; /* rt extents in realtime subvol*/
+ __u64 logstart; /* starting fsblock of the log */
+ unsigned char uuid[16]; /* unique id of the filesystem */
+ __u32 sunit; /* stripe unit, fsblocks */
+ __u32 swidth; /* stripe width, fsblocks */
+ __s32 version; /* structure version */
+ __u32 flags; /* superblock version flags */
+ __u32 logsectsize; /* log sector size, bytes */
+ __u32 rtsectsize; /* realtime sector size, bytes */
+ __u32 dirblocksize; /* directory block size, bytes */
+} __attribute__((packed)) compat_xfs_fsop_geom_v1_t;
+
+#define XFS_IOC_FSGEOMETRY_V1_32 \
+ _IOR('X', 100, struct compat_xfs_fsop_geom_v1)
+
+typedef struct compat_xfs_inogrp {
+ __u64 xi_startino; /* starting inode number */
+ __s32 xi_alloccount; /* # bits set in allocmask */
+ __u64 xi_allocmask; /* mask of allocated inodes */
+} __attribute__((packed)) compat_xfs_inogrp_t;
+
+/* These growfs input structures have padding on the end, so must translate */
+typedef struct compat_xfs_growfs_data {
+ __u64 newblocks; /* new data subvol size, fsblocks */
+ __u32 imaxpct; /* new inode space percentage limit */
+} __attribute__((packed)) compat_xfs_growfs_data_t;
+
+typedef struct compat_xfs_growfs_rt {
+ __u64 newblocks; /* new realtime size, fsblocks */
+ __u32 extsize; /* new realtime extent size, fsblocks */
+} __attribute__((packed)) compat_xfs_growfs_rt_t;
+
+#define XFS_IOC_FSGROWFSDATA_32 _IOW('X', 110, struct compat_xfs_growfs_data)
+#define XFS_IOC_FSGROWFSRT_32 _IOW('X', 112, struct compat_xfs_growfs_rt)
+
+#endif /* BROKEN_X86_ALIGNMENT */
#endif /* __XFS_IOCTL32_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index 095d271f3434..7aa53fefc67f 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -53,6 +53,7 @@
#include <linux/namei.h>
#include <linux/security.h>
#include <linux/falloc.h>
+#include <linux/fiemap.h>
/*
* Bring the atime in the XFS inode uptodate.
@@ -64,14 +65,14 @@ xfs_synchronize_atime(
{
struct inode *inode = VFS_I(ip);
- if (inode) {
+ 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;
}
}
/*
- * If the linux inode exists, mark it dirty.
+ * If the linux inode is valid, mark it dirty.
* Used when commiting a dirty inode into a transaction so that
* the inode will get written back by the linux code
*/
@@ -81,7 +82,7 @@ xfs_mark_inode_dirty_sync(
{
struct inode *inode = VFS_I(ip);
- if (inode)
+ if (!(inode->i_state & (I_WILL_FREE|I_FREEING|I_CLEAR)))
mark_inode_dirty_sync(inode);
}
@@ -128,7 +129,7 @@ xfs_ichgtime(
if (sync_it) {
SYNCHRONIZE();
ip->i_update_core = 1;
- mark_inode_dirty_sync(inode);
+ xfs_mark_inode_dirty_sync(ip);
}
}
@@ -158,8 +159,6 @@ xfs_init_security(
}
error = xfs_attr_set(ip, name, value, length, ATTR_SECURE);
- if (!error)
- xfs_iflags_set(ip, XFS_IMODIFIED);
kfree(name);
kfree(value);
@@ -260,7 +259,6 @@ xfs_vn_mknod(
error = _ACL_INHERIT(inode, mode, default_acl);
if (unlikely(error))
goto out_cleanup_inode;
- xfs_iflags_set(ip, XFS_IMODIFIED);
_ACL_FREE(default_acl);
}
@@ -366,21 +364,17 @@ xfs_vn_link(
struct inode *dir,
struct dentry *dentry)
{
- struct inode *inode; /* inode of guy being linked to */
+ struct inode *inode = old_dentry->d_inode;
struct xfs_name name;
int error;
- inode = old_dentry->d_inode;
xfs_dentry_to_name(&name, dentry);
- igrab(inode);
error = xfs_link(XFS_I(dir), XFS_I(inode), &name);
- if (unlikely(error)) {
- iput(inode);
+ if (unlikely(error))
return -error;
- }
- xfs_iflags_set(XFS_I(dir), XFS_IMODIFIED);
+ atomic_inc(&inode->i_count);
d_instantiate(dentry, inode);
return 0;
}
@@ -601,7 +595,7 @@ xfs_vn_setattr(
struct dentry *dentry,
struct iattr *iattr)
{
- return -xfs_setattr(XFS_I(dentry->d_inode), iattr, 0, NULL);
+ return -xfs_setattr(XFS_I(dentry->d_inode), iattr, 0);
}
/*
@@ -642,7 +636,7 @@ xfs_vn_fallocate(
xfs_ilock(ip, XFS_IOLOCK_EXCL);
error = xfs_change_file_space(ip, XFS_IOC_RESVSP, &bf,
- 0, NULL, XFS_ATTR_NOLOCK);
+ 0, XFS_ATTR_NOLOCK);
if (!error && !(mode & FALLOC_FL_KEEP_SIZE) &&
offset + len > i_size_read(inode))
new_size = offset + len;
@@ -653,7 +647,7 @@ xfs_vn_fallocate(
iattr.ia_valid = ATTR_SIZE;
iattr.ia_size = new_size;
- error = xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK, NULL);
+ error = xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK);
}
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
@@ -661,6 +655,88 @@ out_error:
return error;
}
+#define XFS_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR)
+
+/*
+ * Call fiemap helper to fill in user data.
+ * Returns positive errors to xfs_getbmap.
+ */
+STATIC int
+xfs_fiemap_format(
+ void **arg,
+ struct getbmapx *bmv,
+ int *full)
+{
+ int error;
+ struct fiemap_extent_info *fieinfo = *arg;
+ u32 fiemap_flags = 0;
+ u64 logical, physical, length;
+
+ /* Do nothing for a hole */
+ if (bmv->bmv_block == -1LL)
+ return 0;
+
+ logical = BBTOB(bmv->bmv_offset);
+ physical = BBTOB(bmv->bmv_block);
+ length = BBTOB(bmv->bmv_length);
+
+ if (bmv->bmv_oflags & BMV_OF_PREALLOC)
+ fiemap_flags |= FIEMAP_EXTENT_UNWRITTEN;
+ else if (bmv->bmv_oflags & BMV_OF_DELALLOC) {
+ fiemap_flags |= FIEMAP_EXTENT_DELALLOC;
+ physical = 0; /* no block yet */
+ }
+ if (bmv->bmv_oflags & BMV_OF_LAST)
+ fiemap_flags |= FIEMAP_EXTENT_LAST;
+
+ error = fiemap_fill_next_extent(fieinfo, logical, physical,
+ length, fiemap_flags);
+ if (error > 0) {
+ error = 0;
+ *full = 1; /* user array now full */
+ }
+
+ return -error;
+}
+
+STATIC int
+xfs_vn_fiemap(
+ struct inode *inode,
+ struct fiemap_extent_info *fieinfo,
+ u64 start,
+ u64 length)
+{
+ xfs_inode_t *ip = XFS_I(inode);
+ struct getbmapx bm;
+ int error;
+
+ error = fiemap_check_flags(fieinfo, XFS_FIEMAP_FLAGS);
+ if (error)
+ return error;
+
+ /* Set up bmap header for xfs internal routine */
+ bm.bmv_offset = BTOBB(start);
+ /* Special case for whole file */
+ if (length == FIEMAP_MAX_OFFSET)
+ bm.bmv_length = -1LL;
+ else
+ bm.bmv_length = BTOBB(length);
+
+ /* our formatter will tell xfs_getbmap when to stop. */
+ bm.bmv_count = MAXEXTNUM;
+ bm.bmv_iflags = BMV_IF_PREALLOC;
+ if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR)
+ bm.bmv_iflags |= BMV_IF_ATTRFORK;
+ if (!(fieinfo->fi_flags & FIEMAP_FLAG_SYNC))
+ bm.bmv_iflags |= BMV_IF_DELALLOC;
+
+ error = xfs_getbmap(ip, &bm, xfs_fiemap_format, fieinfo);
+ if (error)
+ return -error;
+
+ return 0;
+}
+
static const struct inode_operations xfs_inode_operations = {
.permission = xfs_vn_permission,
.truncate = xfs_vn_truncate,
@@ -671,6 +747,7 @@ static const struct inode_operations xfs_inode_operations = {
.removexattr = generic_removexattr,
.listxattr = xfs_vn_listxattr,
.fallocate = xfs_vn_fallocate,
+ .fiemap = xfs_vn_fiemap,
};
static const struct inode_operations xfs_dir_inode_operations = {
@@ -766,12 +843,20 @@ xfs_diflags_to_iflags(
* When reading existing inodes from disk this is called directly
* from xfs_iget, when creating a new inode it is called from
* xfs_ialloc after setting up the inode.
+ *
+ * We are always called with an uninitialised linux inode here.
+ * We need to initialise the necessary fields and take a reference
+ * on it.
*/
void
xfs_setup_inode(
struct xfs_inode *ip)
{
- struct inode *inode = ip->i_vnode;
+ struct inode *inode = &ip->i_vnode;
+
+ inode->i_ino = ip->i_ino;
+ inode->i_state = I_NEW|I_LOCK;
+ inode_add_to_lists(ip->i_mount->m_super, inode);
inode->i_mode = ip->i_d.di_mode;
inode->i_nlink = ip->i_d.di_nlink;
@@ -799,7 +884,6 @@ xfs_setup_inode(
inode->i_ctime.tv_sec = ip->i_d.di_ctime.t_sec;
inode->i_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec;
xfs_diflags_to_iflags(inode, ip);
- xfs_iflags_clear(ip, XFS_IMODIFIED);
switch (inode->i_mode & S_IFMT) {
case S_IFREG:
diff --git a/fs/xfs/linux-2.6/xfs_iops.h b/fs/xfs/linux-2.6/xfs_iops.h
index 8b1a1e31dc21..ef41c92ce66e 100644
--- a/fs/xfs/linux-2.6/xfs_iops.h
+++ b/fs/xfs/linux-2.6/xfs_iops.h
@@ -22,7 +22,6 @@ struct xfs_inode;
extern const struct file_operations xfs_file_operations;
extern const struct file_operations xfs_dir_file_operations;
-extern const struct file_operations xfs_invis_file_operations;
extern ssize_t xfs_vn_listxattr(struct dentry *, char *data, size_t size);
diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h
index cc0f7b3a9795..507492d6dccd 100644
--- a/fs/xfs/linux-2.6/xfs_linux.h
+++ b/fs/xfs/linux-2.6/xfs_linux.h
@@ -21,18 +21,12 @@
#include <linux/types.h>
/*
- * Some types are conditional depending on the target system.
* XFS_BIG_BLKNOS needs block layer disk addresses to be 64 bits.
- * XFS_BIG_INUMS needs the VFS inode number to be 64 bits, as well
- * as requiring XFS_BIG_BLKNOS to be set.
+ * XFS_BIG_INUMS requires XFS_BIG_BLKNOS to be set.
*/
#if defined(CONFIG_LBD) || (BITS_PER_LONG == 64)
# define XFS_BIG_BLKNOS 1
-# if BITS_PER_LONG == 64
-# define XFS_BIG_INUMS 1
-# else
-# define XFS_BIG_INUMS 0
-# endif
+# define XFS_BIG_INUMS 1
#else
# define XFS_BIG_BLKNOS 0
# define XFS_BIG_INUMS 0
@@ -77,6 +71,7 @@
#include <linux/spinlock.h>
#include <linux/random.h>
#include <linux/ctype.h>
+#include <linux/writeback.h>
#include <asm/page.h>
#include <asm/div64.h>
@@ -85,7 +80,6 @@
#include <asm/byteorder.h>
#include <asm/unaligned.h>
-#include <xfs_vfs.h>
#include <xfs_cred.h>
#include <xfs_vnode.h>
#include <xfs_stats.h>
@@ -107,7 +101,6 @@
#undef HAVE_PERCPU_SB /* per cpu superblock counters are a 2.6 feature */
#endif
-#define restricted_chown xfs_params.restrict_chown.val
#define irix_sgid_inherit xfs_params.sgid_inherit.val
#define irix_symlink_mode xfs_params.symlink_mode.val
#define xfs_panic_mask xfs_params.panic_mask.val
diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c
index 1957e5357d04..7e90daa0d1d1 100644
--- a/fs/xfs/linux-2.6/xfs_lrw.c
+++ b/fs/xfs/linux-2.6/xfs_lrw.c
@@ -51,7 +51,6 @@
#include "xfs_vnodeops.h"
#include <linux/capability.h>
-#include <linux/mount.h>
#include <linux/writeback.h>
@@ -243,7 +242,7 @@ xfs_read(
if (unlikely(ioflags & IO_ISDIRECT)) {
if (inode->i_mapping->nrpages)
- ret = xfs_flushinval_pages(ip, (*offset & PAGE_CACHE_MASK),
+ ret = -xfs_flushinval_pages(ip, (*offset & PAGE_CACHE_MASK),
-1, FI_REMAPF_LOCKED);
mutex_unlock(&inode->i_mutex);
if (ret) {
@@ -668,15 +667,8 @@ start:
if (new_size > xip->i_size)
xip->i_new_size = new_size;
- /*
- * We're not supposed to change timestamps in readonly-mounted
- * filesystems. Throw it away if anyone asks us.
- */
- if (likely(!(ioflags & IO_INVIS) &&
- !mnt_want_write(file->f_path.mnt))) {
+ if (likely(!(ioflags & IO_INVIS)))
xfs_ichgtime(xip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
- mnt_drop_write(file->f_path.mnt);
- }
/*
* If the offset is beyond the size of the file, we have a couple
@@ -715,7 +707,6 @@ start:
}
}
-retry:
/* We can write back this queue in page reclaim */
current->backing_dev_info = mapping->backing_dev_info;
@@ -771,6 +762,17 @@ retry:
if (ret == -EIOCBQUEUED && !(ioflags & IO_ISAIO))
ret = wait_on_sync_kiocb(iocb);
+ isize = i_size_read(inode);
+ if (unlikely(ret < 0 && ret != -EFAULT && *offset > isize))
+ *offset = isize;
+
+ if (*offset > xip->i_size) {
+ xfs_ilock(xip, XFS_ILOCK_EXCL);
+ if (*offset > xip->i_size)
+ xip->i_size = *offset;
+ xfs_iunlock(xip, XFS_ILOCK_EXCL);
+ }
+
if (ret == -ENOSPC &&
DM_EVENT_ENABLED(xip, DM_EVENT_NOSPACE) && !(ioflags & IO_INVIS)) {
xfs_iunlock(xip, iolock);
@@ -784,20 +786,7 @@ retry:
xfs_ilock(xip, iolock);
if (error)
goto out_unlock_internal;
- pos = xip->i_size;
- ret = 0;
- goto retry;
- }
-
- isize = i_size_read(inode);
- if (unlikely(ret < 0 && ret != -EFAULT && *offset > isize))
- *offset = isize;
-
- if (*offset > xip->i_size) {
- xfs_ilock(xip, XFS_ILOCK_EXCL);
- if (*offset > xip->i_size)
- xip->i_size = *offset;
- xfs_iunlock(xip, XFS_ILOCK_EXCL);
+ goto start;
}
error = -ret;
@@ -855,13 +844,7 @@ retry:
int
xfs_bdstrat_cb(struct xfs_buf *bp)
{
- xfs_mount_t *mp;
-
- mp = XFS_BUF_FSPRIVATE3(bp, xfs_mount_t *);
- if (!XFS_FORCED_SHUTDOWN(mp)) {
- xfs_buf_iorequest(bp);
- return 0;
- } else {
+ if (XFS_FORCED_SHUTDOWN(bp->b_mount)) {
xfs_buftrace("XFS__BDSTRAT IOERROR", bp);
/*
* Metadata write that didn't get logged but
@@ -874,6 +857,9 @@ xfs_bdstrat_cb(struct xfs_buf *bp)
else
return (xfs_bioerror(bp));
}
+
+ xfs_buf_iorequest(bp);
+ return 0;
}
/*
diff --git a/fs/xfs/linux-2.6/xfs_stats.c b/fs/xfs/linux-2.6/xfs_stats.c
index 3d5b67c075c7..c3526d445f6a 100644
--- a/fs/xfs/linux-2.6/xfs_stats.c
+++ b/fs/xfs/linux-2.6/xfs_stats.c
@@ -53,11 +53,15 @@ xfs_read_xfsstats(
{ "icluster", XFSSTAT_END_INODE_CLUSTER },
{ "vnodes", XFSSTAT_END_VNODE_OPS },
{ "buf", XFSSTAT_END_BUF },
+ { "abtb2", XFSSTAT_END_ABTB_V2 },
+ { "abtc2", XFSSTAT_END_ABTC_V2 },
+ { "bmbt2", XFSSTAT_END_BMBT_V2 },
+ { "ibt2", XFSSTAT_END_IBT_V2 },
};
/* Loop over all stats groups */
for (i=j=len = 0; i < ARRAY_SIZE(xstats); i++) {
- len += sprintf(buffer + len, xstats[i].desc);
+ len += sprintf(buffer + len, "%s", xstats[i].desc);
/* inner loop does each group */
while (j < xstats[i].endpoint) {
val = 0;
diff --git a/fs/xfs/linux-2.6/xfs_stats.h b/fs/xfs/linux-2.6/xfs_stats.h
index e83820febc9f..736854b1ca1a 100644
--- a/fs/xfs/linux-2.6/xfs_stats.h
+++ b/fs/xfs/linux-2.6/xfs_stats.h
@@ -118,6 +118,71 @@ struct xfsstats {
__uint32_t xb_page_retries;
__uint32_t xb_page_found;
__uint32_t xb_get_read;
+/* Version 2 btree counters */
+#define XFSSTAT_END_ABTB_V2 (XFSSTAT_END_BUF+15)
+ __uint32_t xs_abtb_2_lookup;
+ __uint32_t xs_abtb_2_compare;
+ __uint32_t xs_abtb_2_insrec;
+ __uint32_t xs_abtb_2_delrec;
+ __uint32_t xs_abtb_2_newroot;
+ __uint32_t xs_abtb_2_killroot;
+ __uint32_t xs_abtb_2_increment;
+ __uint32_t xs_abtb_2_decrement;
+ __uint32_t xs_abtb_2_lshift;
+ __uint32_t xs_abtb_2_rshift;
+ __uint32_t xs_abtb_2_split;
+ __uint32_t xs_abtb_2_join;
+ __uint32_t xs_abtb_2_alloc;
+ __uint32_t xs_abtb_2_free;
+ __uint32_t xs_abtb_2_moves;
+#define XFSSTAT_END_ABTC_V2 (XFSSTAT_END_ABTB_V2+15)
+ __uint32_t xs_abtc_2_lookup;
+ __uint32_t xs_abtc_2_compare;
+ __uint32_t xs_abtc_2_insrec;
+ __uint32_t xs_abtc_2_delrec;
+ __uint32_t xs_abtc_2_newroot;
+ __uint32_t xs_abtc_2_killroot;
+ __uint32_t xs_abtc_2_increment;
+ __uint32_t xs_abtc_2_decrement;
+ __uint32_t xs_abtc_2_lshift;
+ __uint32_t xs_abtc_2_rshift;
+ __uint32_t xs_abtc_2_split;
+ __uint32_t xs_abtc_2_join;
+ __uint32_t xs_abtc_2_alloc;
+ __uint32_t xs_abtc_2_free;
+ __uint32_t xs_abtc_2_moves;
+#define XFSSTAT_END_BMBT_V2 (XFSSTAT_END_ABTC_V2+15)
+ __uint32_t xs_bmbt_2_lookup;
+ __uint32_t xs_bmbt_2_compare;
+ __uint32_t xs_bmbt_2_insrec;
+ __uint32_t xs_bmbt_2_delrec;
+ __uint32_t xs_bmbt_2_newroot;
+ __uint32_t xs_bmbt_2_killroot;
+ __uint32_t xs_bmbt_2_increment;
+ __uint32_t xs_bmbt_2_decrement;
+ __uint32_t xs_bmbt_2_lshift;
+ __uint32_t xs_bmbt_2_rshift;
+ __uint32_t xs_bmbt_2_split;
+ __uint32_t xs_bmbt_2_join;
+ __uint32_t xs_bmbt_2_alloc;
+ __uint32_t xs_bmbt_2_free;
+ __uint32_t xs_bmbt_2_moves;
+#define XFSSTAT_END_IBT_V2 (XFSSTAT_END_BMBT_V2+15)
+ __uint32_t xs_ibt_2_lookup;
+ __uint32_t xs_ibt_2_compare;
+ __uint32_t xs_ibt_2_insrec;
+ __uint32_t xs_ibt_2_delrec;
+ __uint32_t xs_ibt_2_newroot;
+ __uint32_t xs_ibt_2_killroot;
+ __uint32_t xs_ibt_2_increment;
+ __uint32_t xs_ibt_2_decrement;
+ __uint32_t xs_ibt_2_lshift;
+ __uint32_t xs_ibt_2_rshift;
+ __uint32_t xs_ibt_2_split;
+ __uint32_t xs_ibt_2_join;
+ __uint32_t xs_ibt_2_alloc;
+ __uint32_t xs_ibt_2_free;
+ __uint32_t xs_ibt_2_moves;
/* Extra precision counters */
__uint64_t xs_xstrat_bytes;
__uint64_t xs_write_bytes;
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 37ebe36056eb..36f6cc703ef2 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -18,7 +18,6 @@
#include "xfs.h"
#include "xfs_bit.h"
#include "xfs_log.h"
-#include "xfs_clnt.h"
#include "xfs_inum.h"
#include "xfs_trans.h"
#include "xfs_sb.h"
@@ -36,6 +35,7 @@
#include "xfs_dinode.h"
#include "xfs_inode.h"
#include "xfs_btree.h"
+#include "xfs_btree_trace.h"
#include "xfs_ialloc.h"
#include "xfs_bmap.h"
#include "xfs_rtalloc.h"
@@ -48,7 +48,6 @@
#include "xfs_buf_item.h"
#include "xfs_utils.h"
#include "xfs_vnodeops.h"
-#include "xfs_vfsops.h"
#include "xfs_version.h"
#include "xfs_log_priv.h"
#include "xfs_trans_priv.h"
@@ -58,6 +57,7 @@
#include "xfs_extfree_item.h"
#include "xfs_mru_cache.h"
#include "xfs_inode_item.h"
+#include "xfs_sync.h"
#include <linux/namei.h>
#include <linux/init.h>
@@ -70,36 +70,9 @@
static struct quotactl_ops xfs_quotactl_operations;
static struct super_operations xfs_super_operations;
-static kmem_zone_t *xfs_vnode_zone;
static kmem_zone_t *xfs_ioend_zone;
mempool_t *xfs_ioend_pool;
-STATIC struct xfs_mount_args *
-xfs_args_allocate(
- struct super_block *sb,
- int silent)
-{
- struct xfs_mount_args *args;
-
- args = kzalloc(sizeof(struct xfs_mount_args), GFP_KERNEL);
- if (!args)
- return NULL;
-
- args->logbufs = args->logbufsize = -1;
- strncpy(args->fsname, sb->s_id, MAXNAMELEN);
-
- /* Copy the already-parsed mount(2) flags we're interested in */
- if (sb->s_flags & MS_DIRSYNC)
- args->flags |= XFSMNT_DIRSYNC;
- if (sb->s_flags & MS_SYNCHRONOUS)
- args->flags |= XFSMNT_WSYNC;
- if (silent)
- args->flags |= XFSMNT_QUIET;
- args->flags |= XFSMNT_32BITINODES;
-
- return args;
-}
-
#define MNTOPT_LOGBUFS "logbufs" /* number of XFS log buffers */
#define MNTOPT_LOGBSIZE "logbsize" /* size of XFS log buffers */
#define MNTOPT_LOGDEV "logdev" /* log device */
@@ -188,26 +161,54 @@ suffix_strtoul(char *s, char **endp, unsigned int base)
return simple_strtoul((const char *)s, endp, base) << shift_left_factor;
}
+/*
+ * This function fills in xfs_mount_t fields based on mount args.
+ * Note: the superblock has _not_ yet been read in.
+ *
+ * Note that this function leaks the various device name allocations on
+ * failure. The caller takes care of them.
+ */
STATIC int
xfs_parseargs(
struct xfs_mount *mp,
char *options,
- struct xfs_mount_args *args,
- int update)
+ char **mtpt)
{
+ struct super_block *sb = mp->m_super;
char *this_char, *value, *eov;
- int dsunit, dswidth, vol_dsunit, vol_dswidth;
- int iosize;
+ int dsunit = 0;
+ int dswidth = 0;
+ int iosize = 0;
int dmapi_implies_ikeep = 1;
+ uchar_t iosizelog = 0;
+
+ /*
+ * Copy binary VFS mount flags we are interested in.
+ */
+ if (sb->s_flags & MS_RDONLY)
+ mp->m_flags |= XFS_MOUNT_RDONLY;
+ if (sb->s_flags & MS_DIRSYNC)
+ mp->m_flags |= XFS_MOUNT_DIRSYNC;
+ if (sb->s_flags & MS_SYNCHRONOUS)
+ mp->m_flags |= XFS_MOUNT_WSYNC;
+
+ /*
+ * Set some default flags that could be cleared by the mount option
+ * parsing.
+ */
+ mp->m_flags |= XFS_MOUNT_BARRIER;
+ mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
+ mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
- args->flags |= XFSMNT_BARRIER;
- args->flags2 |= XFSMNT2_COMPAT_IOSIZE;
+ /*
+ * These can be overridden by the mount option parsing.
+ */
+ mp->m_logbufs = -1;
+ mp->m_logbsize = -1;
if (!options)
goto done;
- iosize = dsunit = dswidth = vol_dsunit = vol_dswidth = 0;
-
while ((this_char = strsep(&options, ",")) != NULL) {
if (!*this_char)
continue;
@@ -221,7 +222,7 @@ xfs_parseargs(
this_char);
return EINVAL;
}
- args->logbufs = simple_strtoul(value, &eov, 10);
+ mp->m_logbufs = simple_strtoul(value, &eov, 10);
} else if (!strcmp(this_char, MNTOPT_LOGBSIZE)) {
if (!value || !*value) {
cmn_err(CE_WARN,
@@ -229,7 +230,7 @@ xfs_parseargs(
this_char);
return EINVAL;
}
- args->logbufsize = suffix_strtoul(value, &eov, 10);
+ mp->m_logbsize = suffix_strtoul(value, &eov, 10);
} else if (!strcmp(this_char, MNTOPT_LOGDEV)) {
if (!value || !*value) {
cmn_err(CE_WARN,
@@ -237,7 +238,9 @@ xfs_parseargs(
this_char);
return EINVAL;
}
- strncpy(args->logname, value, MAXNAMELEN);
+ mp->m_logname = kstrndup(value, MAXNAMELEN, GFP_KERNEL);
+ if (!mp->m_logname)
+ return ENOMEM;
} else if (!strcmp(this_char, MNTOPT_MTPT)) {
if (!value || !*value) {
cmn_err(CE_WARN,
@@ -245,7 +248,9 @@ xfs_parseargs(
this_char);
return EINVAL;
}
- strncpy(args->mtpt, value, MAXNAMELEN);
+ *mtpt = kstrndup(value, MAXNAMELEN, GFP_KERNEL);
+ if (!*mtpt)
+ return ENOMEM;
} else if (!strcmp(this_char, MNTOPT_RTDEV)) {
if (!value || !*value) {
cmn_err(CE_WARN,
@@ -253,7 +258,9 @@ xfs_parseargs(
this_char);
return EINVAL;
}
- strncpy(args->rtname, value, MAXNAMELEN);
+ mp->m_rtname = kstrndup(value, MAXNAMELEN, GFP_KERNEL);
+ if (!mp->m_rtname)
+ return ENOMEM;
} else if (!strcmp(this_char, MNTOPT_BIOSIZE)) {
if (!value || !*value) {
cmn_err(CE_WARN,
@@ -262,8 +269,7 @@ xfs_parseargs(
return EINVAL;
}
iosize = simple_strtoul(value, &eov, 10);
- args->flags |= XFSMNT_IOSIZE;
- args->iosizelog = (uint8_t) iosize;
+ iosizelog = ffs(iosize) - 1;
} else if (!strcmp(this_char, MNTOPT_ALLOCSIZE)) {
if (!value || !*value) {
cmn_err(CE_WARN,
@@ -272,8 +278,7 @@ xfs_parseargs(
return EINVAL;
}
iosize = suffix_strtoul(value, &eov, 10);
- args->flags |= XFSMNT_IOSIZE;
- args->iosizelog = ffs(iosize) - 1;
+ iosizelog = ffs(iosize) - 1;
} else if (!strcmp(this_char, MNTOPT_GRPID) ||
!strcmp(this_char, MNTOPT_BSDGROUPS)) {
mp->m_flags |= XFS_MOUNT_GRPID;
@@ -281,23 +286,25 @@ xfs_parseargs(
!strcmp(this_char, MNTOPT_SYSVGROUPS)) {
mp->m_flags &= ~XFS_MOUNT_GRPID;
} else if (!strcmp(this_char, MNTOPT_WSYNC)) {
- args->flags |= XFSMNT_WSYNC;
+ mp->m_flags |= XFS_MOUNT_WSYNC;
} else if (!strcmp(this_char, MNTOPT_OSYNCISOSYNC)) {
- args->flags |= XFSMNT_OSYNCISOSYNC;
+ mp->m_flags |= XFS_MOUNT_OSYNCISOSYNC;
} else if (!strcmp(this_char, MNTOPT_NORECOVERY)) {
- args->flags |= XFSMNT_NORECOVERY;
+ mp->m_flags |= XFS_MOUNT_NORECOVERY;
} else if (!strcmp(this_char, MNTOPT_INO64)) {
- args->flags |= XFSMNT_INO64;
-#if !XFS_BIG_INUMS
+#if XFS_BIG_INUMS
+ mp->m_flags |= XFS_MOUNT_INO64;
+ mp->m_inoadd = XFS_INO64_OFFSET;
+#else
cmn_err(CE_WARN,
"XFS: %s option not allowed on this system",
this_char);
return EINVAL;
#endif
} else if (!strcmp(this_char, MNTOPT_NOALIGN)) {
- args->flags |= XFSMNT_NOALIGN;
+ mp->m_flags |= XFS_MOUNT_NOALIGN;
} else if (!strcmp(this_char, MNTOPT_SWALLOC)) {
- args->flags |= XFSMNT_SWALLOC;
+ mp->m_flags |= XFS_MOUNT_SWALLOC;
} else if (!strcmp(this_char, MNTOPT_SUNIT)) {
if (!value || !*value) {
cmn_err(CE_WARN,
@@ -315,7 +322,7 @@ xfs_parseargs(
}
dswidth = simple_strtoul(value, &eov, 10);
} else if (!strcmp(this_char, MNTOPT_64BITINODE)) {
- args->flags &= ~XFSMNT_32BITINODES;
+ mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS;
#if !XFS_BIG_INUMS
cmn_err(CE_WARN,
"XFS: %s option not allowed on this system",
@@ -323,56 +330,61 @@ xfs_parseargs(
return EINVAL;
#endif
} else if (!strcmp(this_char, MNTOPT_NOUUID)) {
- args->flags |= XFSMNT_NOUUID;
+ mp->m_flags |= XFS_MOUNT_NOUUID;
} else if (!strcmp(this_char, MNTOPT_BARRIER)) {
- args->flags |= XFSMNT_BARRIER;
+ mp->m_flags |= XFS_MOUNT_BARRIER;
} else if (!strcmp(this_char, MNTOPT_NOBARRIER)) {
- args->flags &= ~XFSMNT_BARRIER;
+ mp->m_flags &= ~XFS_MOUNT_BARRIER;
} else if (!strcmp(this_char, MNTOPT_IKEEP)) {
- args->flags |= XFSMNT_IKEEP;
+ mp->m_flags |= XFS_MOUNT_IKEEP;
} else if (!strcmp(this_char, MNTOPT_NOIKEEP)) {
dmapi_implies_ikeep = 0;
- args->flags &= ~XFSMNT_IKEEP;
+ mp->m_flags &= ~XFS_MOUNT_IKEEP;
} else if (!strcmp(this_char, MNTOPT_LARGEIO)) {
- args->flags2 &= ~XFSMNT2_COMPAT_IOSIZE;
+ mp->m_flags &= ~XFS_MOUNT_COMPAT_IOSIZE;
} else if (!strcmp(this_char, MNTOPT_NOLARGEIO)) {
- args->flags2 |= XFSMNT2_COMPAT_IOSIZE;
+ mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
} else if (!strcmp(this_char, MNTOPT_ATTR2)) {
- args->flags |= XFSMNT_ATTR2;
+ mp->m_flags |= XFS_MOUNT_ATTR2;
} else if (!strcmp(this_char, MNTOPT_NOATTR2)) {
- args->flags &= ~XFSMNT_ATTR2;
- args->flags |= XFSMNT_NOATTR2;
+ mp->m_flags &= ~XFS_MOUNT_ATTR2;
+ mp->m_flags |= XFS_MOUNT_NOATTR2;
} else if (!strcmp(this_char, MNTOPT_FILESTREAM)) {
- args->flags2 |= XFSMNT2_FILESTREAMS;
+ mp->m_flags |= XFS_MOUNT_FILESTREAMS;
} else if (!strcmp(this_char, MNTOPT_NOQUOTA)) {
- args->flags &= ~(XFSMNT_UQUOTAENF|XFSMNT_UQUOTA);
- args->flags &= ~(XFSMNT_GQUOTAENF|XFSMNT_GQUOTA);
+ mp->m_qflags &= ~(XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE |
+ XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE |
+ XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE |
+ XFS_UQUOTA_ENFD | XFS_OQUOTA_ENFD);
} else if (!strcmp(this_char, MNTOPT_QUOTA) ||
!strcmp(this_char, MNTOPT_UQUOTA) ||
!strcmp(this_char, MNTOPT_USRQUOTA)) {
- args->flags |= XFSMNT_UQUOTA | XFSMNT_UQUOTAENF;
+ mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE |
+ XFS_UQUOTA_ENFD);
} else if (!strcmp(this_char, MNTOPT_QUOTANOENF) ||
!strcmp(this_char, MNTOPT_UQUOTANOENF)) {
- args->flags |= XFSMNT_UQUOTA;
- args->flags &= ~XFSMNT_UQUOTAENF;
+ mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE);
+ mp->m_qflags &= ~XFS_UQUOTA_ENFD;
} else if (!strcmp(this_char, MNTOPT_PQUOTA) ||
!strcmp(this_char, MNTOPT_PRJQUOTA)) {
- args->flags |= XFSMNT_PQUOTA | XFSMNT_PQUOTAENF;
+ mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE |
+ XFS_OQUOTA_ENFD);
} else if (!strcmp(this_char, MNTOPT_PQUOTANOENF)) {
- args->flags |= XFSMNT_PQUOTA;
- args->flags &= ~XFSMNT_PQUOTAENF;
+ mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE);
+ mp->m_qflags &= ~XFS_OQUOTA_ENFD;
} else if (!strcmp(this_char, MNTOPT_GQUOTA) ||
!strcmp(this_char, MNTOPT_GRPQUOTA)) {
- args->flags |= XFSMNT_GQUOTA | XFSMNT_GQUOTAENF;
+ mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE |
+ XFS_OQUOTA_ENFD);
} else if (!strcmp(this_char, MNTOPT_GQUOTANOENF)) {
- args->flags |= XFSMNT_GQUOTA;
- args->flags &= ~XFSMNT_GQUOTAENF;
+ mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE);
+ mp->m_qflags &= ~XFS_OQUOTA_ENFD;
} else if (!strcmp(this_char, MNTOPT_DMAPI)) {
- args->flags |= XFSMNT_DMAPI;
+ mp->m_flags |= XFS_MOUNT_DMAPI;
} else if (!strcmp(this_char, MNTOPT_XDSM)) {
- args->flags |= XFSMNT_DMAPI;
+ mp->m_flags |= XFS_MOUNT_DMAPI;
} else if (!strcmp(this_char, MNTOPT_DMI)) {
- args->flags |= XFSMNT_DMAPI;
+ mp->m_flags |= XFS_MOUNT_DMAPI;
} else if (!strcmp(this_char, "ihashsize")) {
cmn_err(CE_WARN,
"XFS: ihashsize no longer used, option is deprecated.");
@@ -390,27 +402,29 @@ xfs_parseargs(
}
}
- if (args->flags & XFSMNT_NORECOVERY) {
- if ((mp->m_flags & XFS_MOUNT_RDONLY) == 0) {
- cmn_err(CE_WARN,
- "XFS: no-recovery mounts must be read-only.");
- return EINVAL;
- }
+ /*
+ * no recovery flag requires a read-only mount
+ */
+ if ((mp->m_flags & XFS_MOUNT_NORECOVERY) &&
+ !(mp->m_flags & XFS_MOUNT_RDONLY)) {
+ cmn_err(CE_WARN, "XFS: no-recovery mounts must be read-only.");
+ return EINVAL;
}
- if ((args->flags & XFSMNT_NOALIGN) && (dsunit || dswidth)) {
+ if ((mp->m_flags & XFS_MOUNT_NOALIGN) && (dsunit || dswidth)) {
cmn_err(CE_WARN,
"XFS: sunit and swidth options incompatible with the noalign option");
return EINVAL;
}
- if ((args->flags & XFSMNT_GQUOTA) && (args->flags & XFSMNT_PQUOTA)) {
+ if ((mp->m_qflags & (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE)) &&
+ (mp->m_qflags & (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE))) {
cmn_err(CE_WARN,
"XFS: cannot mount with both project and group quota");
return EINVAL;
}
- if ((args->flags & XFSMNT_DMAPI) && *args->mtpt == '\0') {
+ if ((mp->m_flags & XFS_MOUNT_DMAPI) && (!*mtpt || *mtpt[0] == '\0')) {
printk("XFS: %s option needs the mount point option as well\n",
MNTOPT_DMAPI);
return EINVAL;
@@ -438,27 +452,66 @@ xfs_parseargs(
* Note that if "ikeep" or "noikeep" mount options are
* supplied, then they are honored.
*/
- if ((args->flags & XFSMNT_DMAPI) && dmapi_implies_ikeep)
- args->flags |= XFSMNT_IKEEP;
+ if ((mp->m_flags & XFS_MOUNT_DMAPI) && dmapi_implies_ikeep)
+ mp->m_flags |= XFS_MOUNT_IKEEP;
- if ((args->flags & XFSMNT_NOALIGN) != XFSMNT_NOALIGN) {
+done:
+ if (!(mp->m_flags & XFS_MOUNT_NOALIGN)) {
+ /*
+ * At this point the superblock has not been read
+ * in, therefore we do not know the block size.
+ * Before the mount call ends we will convert
+ * these to FSBs.
+ */
if (dsunit) {
- args->sunit = dsunit;
- args->flags |= XFSMNT_RETERR;
- } else {
- args->sunit = vol_dsunit;
+ mp->m_dalign = dsunit;
+ mp->m_flags |= XFS_MOUNT_RETERR;
}
- dswidth ? (args->swidth = dswidth) :
- (args->swidth = vol_dswidth);
- } else {
- args->sunit = args->swidth = 0;
+
+ if (dswidth)
+ mp->m_swidth = dswidth;
+ }
+
+ if (mp->m_logbufs != -1 &&
+ mp->m_logbufs != 0 &&
+ (mp->m_logbufs < XLOG_MIN_ICLOGS ||
+ mp->m_logbufs > XLOG_MAX_ICLOGS)) {
+ cmn_err(CE_WARN,
+ "XFS: invalid logbufs value: %d [not %d-%d]",
+ mp->m_logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS);
+ return XFS_ERROR(EINVAL);
+ }
+ if (mp->m_logbsize != -1 &&
+ mp->m_logbsize != 0 &&
+ (mp->m_logbsize < XLOG_MIN_RECORD_BSIZE ||
+ mp->m_logbsize > XLOG_MAX_RECORD_BSIZE ||
+ !is_power_of_2(mp->m_logbsize))) {
+ cmn_err(CE_WARN,
+ "XFS: invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]",
+ mp->m_logbsize);
+ return XFS_ERROR(EINVAL);
+ }
+
+ mp->m_fsname = kstrndup(sb->s_id, MAXNAMELEN, GFP_KERNEL);
+ if (!mp->m_fsname)
+ return ENOMEM;
+ mp->m_fsname_len = strlen(mp->m_fsname) + 1;
+
+ if (iosizelog) {
+ if (iosizelog > XFS_MAX_IO_LOG ||
+ iosizelog < XFS_MIN_IO_LOG) {
+ cmn_err(CE_WARN,
+ "XFS: invalid log iosize: %d [not %d-%d]",
+ iosizelog, XFS_MIN_IO_LOG,
+ XFS_MAX_IO_LOG);
+ return XFS_ERROR(EINVAL);
+ }
+
+ mp->m_flags |= XFS_MOUNT_DFLT_IOSIZE;
+ mp->m_readio_log = iosizelog;
+ mp->m_writeio_log = iosizelog;
}
-done:
- if (args->flags & XFSMNT_32BITINODES)
- mp->m_flags |= XFS_MOUNT_SMALL_INUMS;
- if (args->flags2)
- args->flags |= XFSMNT_FLAGS2;
return 0;
}
@@ -704,8 +757,7 @@ xfs_close_devices(
*/
STATIC int
xfs_open_devices(
- struct xfs_mount *mp,
- struct xfs_mount_args *args)
+ struct xfs_mount *mp)
{
struct block_device *ddev = mp->m_super->s_bdev;
struct block_device *logdev = NULL, *rtdev = NULL;
@@ -714,14 +766,14 @@ xfs_open_devices(
/*
* Open real time and log devices - order is important.
*/
- if (args->logname[0]) {
- error = xfs_blkdev_get(mp, args->logname, &logdev);
+ if (mp->m_logname) {
+ error = xfs_blkdev_get(mp, mp->m_logname, &logdev);
if (error)
goto out;
}
- if (args->rtname[0]) {
- error = xfs_blkdev_get(mp, args->rtname, &rtdev);
+ if (mp->m_rtname) {
+ error = xfs_blkdev_get(mp, mp->m_rtname, &rtdev);
if (error)
goto out_close_logdev;
@@ -813,18 +865,18 @@ xfs_setup_devices(
*/
void
xfsaild_wakeup(
- xfs_mount_t *mp,
+ struct xfs_ail *ailp,
xfs_lsn_t threshold_lsn)
{
- mp->m_ail.xa_target = threshold_lsn;
- wake_up_process(mp->m_ail.xa_task);
+ ailp->xa_target = threshold_lsn;
+ wake_up_process(ailp->xa_task);
}
int
xfsaild(
void *data)
{
- xfs_mount_t *mp = (xfs_mount_t *)data;
+ struct xfs_ail *ailp = data;
xfs_lsn_t last_pushed_lsn = 0;
long tout = 0;
@@ -836,11 +888,11 @@ xfsaild(
/* swsusp */
try_to_freeze();
- ASSERT(mp->m_log);
- if (XFS_FORCED_SHUTDOWN(mp))
+ ASSERT(ailp->xa_mount->m_log);
+ if (XFS_FORCED_SHUTDOWN(ailp->xa_mount))
continue;
- tout = xfsaild_push(mp, &last_pushed_lsn);
+ tout = xfsaild_push(ailp, &last_pushed_lsn);
}
return 0;
@@ -848,43 +900,82 @@ xfsaild(
int
xfsaild_start(
- xfs_mount_t *mp)
+ struct xfs_ail *ailp)
{
- mp->m_ail.xa_target = 0;
- mp->m_ail.xa_task = kthread_run(xfsaild, mp, "xfsaild");
- if (IS_ERR(mp->m_ail.xa_task))
- return -PTR_ERR(mp->m_ail.xa_task);
+ ailp->xa_target = 0;
+ ailp->xa_task = kthread_run(xfsaild, ailp, "xfsaild");
+ if (IS_ERR(ailp->xa_task))
+ return -PTR_ERR(ailp->xa_task);
return 0;
}
void
xfsaild_stop(
- xfs_mount_t *mp)
+ struct xfs_ail *ailp)
{
- kthread_stop(mp->m_ail.xa_task);
+ kthread_stop(ailp->xa_task);
}
-
+/* Catch misguided souls that try to use this interface on XFS */
STATIC struct inode *
xfs_fs_alloc_inode(
struct super_block *sb)
{
- return kmem_zone_alloc(xfs_vnode_zone, KM_SLEEP);
+ BUG();
+ return NULL;
}
+/*
+ * Now that the generic code is guaranteed not to be accessing
+ * the linux inode, we can reclaim the inode.
+ */
STATIC void
xfs_fs_destroy_inode(
- struct inode *inode)
+ struct inode *inode)
{
- kmem_zone_free(xfs_vnode_zone, inode);
+ xfs_inode_t *ip = XFS_I(inode);
+
+ XFS_STATS_INC(vn_reclaim);
+ if (xfs_reclaim(ip))
+ panic("%s: cannot reclaim 0x%p\n", __func__, inode);
}
+/*
+ * Slab object creation initialisation for the XFS inode.
+ * This covers only the idempotent fields in the XFS inode;
+ * all other fields need to be initialised on allocation
+ * from the slab. This avoids the need to repeatedly intialise
+ * fields in the xfs inode that left in the initialise state
+ * when freeing the inode.
+ */
STATIC void
xfs_fs_inode_init_once(
- void *vnode)
+ void *inode)
{
- inode_init_once((struct inode *)vnode);
+ struct xfs_inode *ip = inode;
+
+ memset(ip, 0, sizeof(struct xfs_inode));
+
+ /* vfs inode */
+ inode_init_once(VFS_I(ip));
+
+ /* xfs inode */
+ atomic_set(&ip->i_iocount, 0);
+ atomic_set(&ip->i_pincount, 0);
+ spin_lock_init(&ip->i_flags_lock);
+ init_waitqueue_head(&ip->i_ipin_wait);
+ /*
+ * Because we want to use a counting completion, complete
+ * the flush completion once to allow a single access to
+ * the flush completion without blocking.
+ */
+ init_completion(&ip->i_flush);
+ complete(&ip->i_flush);
+
+ mrlock_init(&ip->i_lock, MRLOCK_ALLOW_EQUAL_PRI|MRLOCK_BARRIER,
+ "xfsino", ip->i_ino);
+ mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino);
}
/*
@@ -898,21 +989,26 @@ xfs_fs_write_inode(
struct inode *inode,
int sync)
{
+ struct xfs_inode *ip = XFS_I(inode);
int error = 0;
int flags = 0;
- xfs_itrace_entry(XFS_I(inode));
+ xfs_itrace_entry(ip);
if (sync) {
- filemap_fdatawait(inode->i_mapping);
+ error = xfs_wait_on_pages(ip, 0, -1);
+ if (error)
+ goto out_error;
flags |= FLUSH_SYNC;
}
- error = xfs_inode_flush(XFS_I(inode), flags);
+ error = xfs_inode_flush(ip, flags);
+
+out_error:
/*
* if we failed to write out the inode then mark
* it dirty again so we'll try again later.
*/
if (error)
- mark_inode_dirty_sync(inode);
+ xfs_mark_inode_dirty_sync(ip);
return -error;
}
@@ -923,164 +1019,12 @@ xfs_fs_clear_inode(
{
xfs_inode_t *ip = XFS_I(inode);
- /*
- * ip can be null when xfs_iget_core calls xfs_idestroy if we
- * find an inode with di_mode == 0 but without IGET_CREATE set.
- */
- if (ip) {
- xfs_itrace_entry(ip);
- XFS_STATS_INC(vn_rele);
- XFS_STATS_INC(vn_remove);
- XFS_STATS_INC(vn_reclaim);
- XFS_STATS_DEC(vn_active);
-
- xfs_inactive(ip);
- xfs_iflags_clear(ip, XFS_IMODIFIED);
- if (xfs_reclaim(ip))
- panic("%s: cannot reclaim 0x%p\n", __func__, inode);
- }
-
- ASSERT(XFS_I(inode) == NULL);
-}
+ xfs_itrace_entry(ip);
+ XFS_STATS_INC(vn_rele);
+ XFS_STATS_INC(vn_remove);
+ XFS_STATS_DEC(vn_active);
-/*
- * Enqueue a work item to be picked up by the vfs xfssyncd thread.
- * Doing this has two advantages:
- * - It saves on stack space, which is tight in certain situations
- * - It can be used (with care) as a mechanism to avoid deadlocks.
- * Flushing while allocating in a full filesystem requires both.
- */
-STATIC void
-xfs_syncd_queue_work(
- struct xfs_mount *mp,
- void *data,
- void (*syncer)(struct xfs_mount *, void *))
-{
- struct bhv_vfs_sync_work *work;
-
- work = kmem_alloc(sizeof(struct bhv_vfs_sync_work), KM_SLEEP);
- INIT_LIST_HEAD(&work->w_list);
- work->w_syncer = syncer;
- work->w_data = data;
- work->w_mount = mp;
- spin_lock(&mp->m_sync_lock);
- list_add_tail(&work->w_list, &mp->m_sync_list);
- spin_unlock(&mp->m_sync_lock);
- wake_up_process(mp->m_sync_task);
-}
-
-/*
- * Flush delayed allocate data, attempting to free up reserved space
- * from existing allocations. At this point a new allocation attempt
- * has failed with ENOSPC and we are in the process of scratching our
- * heads, looking about for more room...
- */
-STATIC void
-xfs_flush_inode_work(
- struct xfs_mount *mp,
- void *arg)
-{
- struct inode *inode = arg;
- filemap_flush(inode->i_mapping);
- iput(inode);
-}
-
-void
-xfs_flush_inode(
- xfs_inode_t *ip)
-{
- struct inode *inode = VFS_I(ip);
-
- igrab(inode);
- xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_inode_work);
- delay(msecs_to_jiffies(500));
-}
-
-/*
- * This is the "bigger hammer" version of xfs_flush_inode_work...
- * (IOW, "If at first you don't succeed, use a Bigger Hammer").
- */
-STATIC void
-xfs_flush_device_work(
- struct xfs_mount *mp,
- void *arg)
-{
- struct inode *inode = arg;
- sync_blockdev(mp->m_super->s_bdev);
- iput(inode);
-}
-
-void
-xfs_flush_device(
- xfs_inode_t *ip)
-{
- struct inode *inode = VFS_I(ip);
-
- igrab(inode);
- xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_device_work);
- delay(msecs_to_jiffies(500));
- xfs_log_force(ip->i_mount, (xfs_lsn_t)0, XFS_LOG_FORCE|XFS_LOG_SYNC);
-}
-
-STATIC void
-xfs_sync_worker(
- struct xfs_mount *mp,
- void *unused)
-{
- int error;
-
- if (!(mp->m_flags & XFS_MOUNT_RDONLY))
- error = xfs_sync(mp, SYNC_FSDATA | SYNC_BDFLUSH | SYNC_ATTR);
- mp->m_sync_seq++;
- wake_up(&mp->m_wait_single_sync_task);
-}
-
-STATIC int
-xfssyncd(
- void *arg)
-{
- struct xfs_mount *mp = arg;
- long timeleft;
- bhv_vfs_sync_work_t *work, *n;
- LIST_HEAD (tmp);
-
- set_freezable();
- timeleft = xfs_syncd_centisecs * msecs_to_jiffies(10);
- for (;;) {
- timeleft = schedule_timeout_interruptible(timeleft);
- /* swsusp */
- try_to_freeze();
- if (kthread_should_stop() && list_empty(&mp->m_sync_list))
- break;
-
- spin_lock(&mp->m_sync_lock);
- /*
- * We can get woken by laptop mode, to do a sync -
- * that's the (only!) case where the list would be
- * empty with time remaining.
- */
- if (!timeleft || list_empty(&mp->m_sync_list)) {
- if (!timeleft)
- timeleft = xfs_syncd_centisecs *
- msecs_to_jiffies(10);
- INIT_LIST_HEAD(&mp->m_sync_work.w_list);
- list_add_tail(&mp->m_sync_work.w_list,
- &mp->m_sync_list);
- }
- list_for_each_entry_safe(work, n, &mp->m_sync_list, w_list)
- list_move(&work->w_list, &tmp);
- spin_unlock(&mp->m_sync_lock);
-
- list_for_each_entry_safe(work, n, &tmp, w_list) {
- (*work->w_syncer)(mp, work->w_data);
- list_del(&work->w_list);
- if (work == &mp->m_sync_work)
- continue;
- kmem_free(work);
- }
- }
-
- return 0;
+ xfs_inactive(ip);
}
STATIC void
@@ -1099,11 +1043,9 @@ xfs_fs_put_super(
struct xfs_mount *mp = XFS_M(sb);
struct xfs_inode *rip = mp->m_rootip;
int unmount_event_flags = 0;
- int error;
- kthread_stop(mp->m_sync_task);
-
- xfs_sync(mp, SYNC_ATTR | SYNC_DELWRI);
+ xfs_syncd_stop(mp);
+ xfs_sync_inodes(mp, SYNC_ATTR|SYNC_DELWRI);
#ifdef HAVE_DMAPI
if (mp->m_flags & XFS_MOUNT_DMAPI) {
@@ -1128,18 +1070,6 @@ xfs_fs_put_super(
xfs_filestream_unmount(mp);
XFS_bflush(mp->m_ddev_targp);
- error = xfs_unmount_flush(mp, 0);
- WARN_ON(error);
-
- /*
- * If we're forcing a shutdown, typically because of a media error,
- * we want to make sure we invalidate dirty pages that belong to
- * referenced vnodes as well.
- */
- if (XFS_FORCED_SHUTDOWN(mp)) {
- error = xfs_sync(mp, SYNC_WAIT | SYNC_CLOSE);
- ASSERT(error != EFSCORRUPTED);
- }
if (mp->m_flags & XFS_MOUNT_DMAPI) {
XFS_SEND_UNMOUNT(mp, rip, DM_RIGHT_NULL, 0, 0,
@@ -1161,7 +1091,7 @@ xfs_fs_write_super(
struct super_block *sb)
{
if (!(sb->s_flags & MS_RDONLY))
- xfs_sync(XFS_M(sb), SYNC_FSDATA);
+ xfs_sync_fsdata(XFS_M(sb), 0);
sb->s_dirt = 0;
}
@@ -1172,7 +1102,6 @@ xfs_fs_sync_super(
{
struct xfs_mount *mp = XFS_M(sb);
int error;
- int flags;
/*
* Treat a sync operation like a freeze. This is to work
@@ -1186,20 +1115,10 @@ xfs_fs_sync_super(
* dirty the Linux inode until after the transaction I/O
* completes.
*/
- if (wait || unlikely(sb->s_frozen == SB_FREEZE_WRITE)) {
- /*
- * First stage of freeze - no more writers will make progress
- * now we are here, so we flush delwri and delalloc buffers
- * here, then wait for all I/O to complete. Data is frozen at
- * that point. Metadata is not frozen, transactions can still
- * occur here so don't bother flushing the buftarg (i.e
- * SYNC_QUIESCE) because it'll just get dirty again.
- */
- flags = SYNC_DATA_QUIESCE;
- } else
- flags = SYNC_FSDATA;
-
- error = xfs_sync(mp, flags);
+ if (wait || unlikely(sb->s_frozen == SB_FREEZE_WRITE))
+ error = xfs_quiesce_data(mp);
+ else
+ error = xfs_sync_fsdata(mp, 0);
sb->s_dirt = 0;
if (unlikely(laptop_mode)) {
@@ -1337,9 +1256,8 @@ xfs_fs_remount(
/* rw -> ro */
if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (*flags & MS_RDONLY)) {
- xfs_filestream_flush(mp);
- xfs_sync(mp, SYNC_DATA_QUIESCE);
- xfs_attr_quiesce(mp);
+ xfs_quiesce_data(mp);
+ xfs_quiesce_attr(mp);
mp->m_flags |= XFS_MOUNT_RDONLY;
}
@@ -1348,7 +1266,7 @@ xfs_fs_remount(
/*
* Second stage of a freeze. The data is already frozen so we only
- * need to take care of themetadata. Once that's done write a dummy
+ * need to take care of the metadata. Once that's done write a dummy
* record to dirty the log in case of a crash while frozen.
*/
STATIC void
@@ -1357,7 +1275,7 @@ xfs_fs_lockfs(
{
struct xfs_mount *mp = XFS_M(sb);
- xfs_attr_quiesce(mp);
+ xfs_quiesce_attr(mp);
xfs_fs_log_dummy(mp);
}
@@ -1422,175 +1340,28 @@ xfs_fs_setxquota(
/*
* This function fills in xfs_mount_t fields based on mount args.
- * Note: the superblock has _not_ yet been read in.
- */
-STATIC int
-xfs_start_flags(
- struct xfs_mount_args *ap,
- struct xfs_mount *mp)
-{
- int error;
-
- /* Values are in BBs */
- if ((ap->flags & XFSMNT_NOALIGN) != XFSMNT_NOALIGN) {
- /*
- * At this point the superblock has not been read
- * in, therefore we do not know the block size.
- * Before the mount call ends we will convert
- * these to FSBs.
- */
- mp->m_dalign = ap->sunit;
- mp->m_swidth = ap->swidth;
- }
-
- if (ap->logbufs != -1 &&
- ap->logbufs != 0 &&
- (ap->logbufs < XLOG_MIN_ICLOGS ||
- ap->logbufs > XLOG_MAX_ICLOGS)) {
- cmn_err(CE_WARN,
- "XFS: invalid logbufs value: %d [not %d-%d]",
- ap->logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS);
- return XFS_ERROR(EINVAL);
- }
- mp->m_logbufs = ap->logbufs;
- if (ap->logbufsize != -1 &&
- ap->logbufsize != 0 &&
- (ap->logbufsize < XLOG_MIN_RECORD_BSIZE ||
- ap->logbufsize > XLOG_MAX_RECORD_BSIZE ||
- !is_power_of_2(ap->logbufsize))) {
- cmn_err(CE_WARN,
- "XFS: invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]",
- ap->logbufsize);
- return XFS_ERROR(EINVAL);
- }
-
- error = ENOMEM;
-
- mp->m_logbsize = ap->logbufsize;
- mp->m_fsname_len = strlen(ap->fsname) + 1;
-
- mp->m_fsname = kstrdup(ap->fsname, GFP_KERNEL);
- if (!mp->m_fsname)
- goto out;
-
- if (ap->rtname[0]) {
- mp->m_rtname = kstrdup(ap->rtname, GFP_KERNEL);
- if (!mp->m_rtname)
- goto out_free_fsname;
-
- }
-
- if (ap->logname[0]) {
- mp->m_logname = kstrdup(ap->logname, GFP_KERNEL);
- if (!mp->m_logname)
- goto out_free_rtname;
- }
-
- if (ap->flags & XFSMNT_WSYNC)
- mp->m_flags |= XFS_MOUNT_WSYNC;
-#if XFS_BIG_INUMS
- if (ap->flags & XFSMNT_INO64) {
- mp->m_flags |= XFS_MOUNT_INO64;
- mp->m_inoadd = XFS_INO64_OFFSET;
- }
-#endif
- if (ap->flags & XFSMNT_RETERR)
- mp->m_flags |= XFS_MOUNT_RETERR;
- if (ap->flags & XFSMNT_NOALIGN)
- mp->m_flags |= XFS_MOUNT_NOALIGN;
- if (ap->flags & XFSMNT_SWALLOC)
- mp->m_flags |= XFS_MOUNT_SWALLOC;
- if (ap->flags & XFSMNT_OSYNCISOSYNC)
- mp->m_flags |= XFS_MOUNT_OSYNCISOSYNC;
- if (ap->flags & XFSMNT_32BITINODES)
- mp->m_flags |= XFS_MOUNT_32BITINODES;
-
- if (ap->flags & XFSMNT_IOSIZE) {
- if (ap->iosizelog > XFS_MAX_IO_LOG ||
- ap->iosizelog < XFS_MIN_IO_LOG) {
- cmn_err(CE_WARN,
- "XFS: invalid log iosize: %d [not %d-%d]",
- ap->iosizelog, XFS_MIN_IO_LOG,
- XFS_MAX_IO_LOG);
- return XFS_ERROR(EINVAL);
- }
-
- mp->m_flags |= XFS_MOUNT_DFLT_IOSIZE;
- mp->m_readio_log = mp->m_writeio_log = ap->iosizelog;
- }
-
- if (ap->flags & XFSMNT_IKEEP)
- mp->m_flags |= XFS_MOUNT_IKEEP;
- if (ap->flags & XFSMNT_DIRSYNC)
- mp->m_flags |= XFS_MOUNT_DIRSYNC;
- if (ap->flags & XFSMNT_ATTR2)
- mp->m_flags |= XFS_MOUNT_ATTR2;
- if (ap->flags & XFSMNT_NOATTR2)
- mp->m_flags |= XFS_MOUNT_NOATTR2;
-
- if (ap->flags2 & XFSMNT2_COMPAT_IOSIZE)
- mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
-
- /*
- * no recovery flag requires a read-only mount
- */
- if (ap->flags & XFSMNT_NORECOVERY) {
- if (!(mp->m_flags & XFS_MOUNT_RDONLY)) {
- cmn_err(CE_WARN,
- "XFS: tried to mount a FS read-write without recovery!");
- return XFS_ERROR(EINVAL);
- }
- mp->m_flags |= XFS_MOUNT_NORECOVERY;
- }
-
- if (ap->flags & XFSMNT_NOUUID)
- mp->m_flags |= XFS_MOUNT_NOUUID;
- if (ap->flags & XFSMNT_BARRIER)
- mp->m_flags |= XFS_MOUNT_BARRIER;
- else
- mp->m_flags &= ~XFS_MOUNT_BARRIER;
-
- if (ap->flags2 & XFSMNT2_FILESTREAMS)
- mp->m_flags |= XFS_MOUNT_FILESTREAMS;
-
- if (ap->flags & XFSMNT_DMAPI)
- mp->m_flags |= XFS_MOUNT_DMAPI;
- return 0;
-
-
- out_free_rtname:
- kfree(mp->m_rtname);
- out_free_fsname:
- kfree(mp->m_fsname);
- out:
- return error;
-}
-
-/*
- * This function fills in xfs_mount_t fields based on mount args.
* Note: the superblock _has_ now been read in.
*/
STATIC int
xfs_finish_flags(
- struct xfs_mount_args *ap,
struct xfs_mount *mp)
{
int ronly = (mp->m_flags & XFS_MOUNT_RDONLY);
/* Fail a mount where the logbuf is smaller then the log stripe */
if (xfs_sb_version_haslogv2(&mp->m_sb)) {
- if ((ap->logbufsize <= 0) &&
- (mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE)) {
+ if (mp->m_logbsize <= 0 &&
+ mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE) {
mp->m_logbsize = mp->m_sb.sb_logsunit;
- } else if (ap->logbufsize > 0 &&
- ap->logbufsize < mp->m_sb.sb_logsunit) {
+ } else if (mp->m_logbsize > 0 &&
+ mp->m_logbsize < mp->m_sb.sb_logsunit) {
cmn_err(CE_WARN,
"XFS: logbuf size must be greater than or equal to log stripe size");
return XFS_ERROR(EINVAL);
}
} else {
/* Fail a mount if the logbuf is larger than 32K */
- if (ap->logbufsize > XLOG_BIG_RECORD_BSIZE) {
+ if (mp->m_logbsize > XLOG_BIG_RECORD_BSIZE) {
cmn_err(CE_WARN,
"XFS: logbuf size for version 1 logs must be 16K or 32K");
return XFS_ERROR(EINVAL);
@@ -1602,7 +1373,7 @@ xfs_finish_flags(
* told by noattr2 to turn it off
*/
if (xfs_sb_version_hasattr2(&mp->m_sb) &&
- !(ap->flags & XFSMNT_NOATTR2))
+ !(mp->m_flags & XFS_MOUNT_NOATTR2))
mp->m_flags |= XFS_MOUNT_ATTR2;
/*
@@ -1614,48 +1385,6 @@ xfs_finish_flags(
return XFS_ERROR(EROFS);
}
- /*
- * check for shared mount.
- */
- if (ap->flags & XFSMNT_SHARED) {
- if (!xfs_sb_version_hasshared(&mp->m_sb))
- return XFS_ERROR(EINVAL);
-
- /*
- * For IRIX 6.5, shared mounts must have the shared
- * version bit set, have the persistent readonly
- * field set, must be version 0 and can only be mounted
- * read-only.
- */
- if (!ronly || !(mp->m_sb.sb_flags & XFS_SBF_READONLY) ||
- (mp->m_sb.sb_shared_vn != 0))
- return XFS_ERROR(EINVAL);
-
- mp->m_flags |= XFS_MOUNT_SHARED;
-
- /*
- * Shared XFS V0 can't deal with DMI. Return EINVAL.
- */
- if (mp->m_sb.sb_shared_vn == 0 && (ap->flags & XFSMNT_DMAPI))
- return XFS_ERROR(EINVAL);
- }
-
- if (ap->flags & XFSMNT_UQUOTA) {
- mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE);
- if (ap->flags & XFSMNT_UQUOTAENF)
- mp->m_qflags |= XFS_UQUOTA_ENFD;
- }
-
- if (ap->flags & XFSMNT_GQUOTA) {
- mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE);
- if (ap->flags & XFSMNT_GQUOTAENF)
- mp->m_qflags |= XFS_OQUOTA_ENFD;
- } else if (ap->flags & XFSMNT_PQUOTA) {
- mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE);
- if (ap->flags & XFSMNT_PQUOTAENF)
- mp->m_qflags |= XFS_OQUOTA_ENFD;
- }
-
return 0;
}
@@ -1667,19 +1396,14 @@ xfs_fs_fill_super(
{
struct inode *root;
struct xfs_mount *mp = NULL;
- struct xfs_mount_args *args;
int flags = 0, error = ENOMEM;
-
- args = xfs_args_allocate(sb, silent);
- if (!args)
- return -ENOMEM;
+ char *mtpt = NULL;
mp = kzalloc(sizeof(struct xfs_mount), GFP_KERNEL);
if (!mp)
- goto out_free_args;
+ goto out;
spin_lock_init(&mp->m_sb_lock);
- mutex_init(&mp->m_ilock);
mutex_init(&mp->m_growlock);
atomic_set(&mp->m_active_trans, 0);
INIT_LIST_HEAD(&mp->m_sync_list);
@@ -1689,12 +1413,9 @@ xfs_fs_fill_super(
mp->m_super = sb;
sb->s_fs_info = mp;
- if (sb->s_flags & MS_RDONLY)
- mp->m_flags |= XFS_MOUNT_RDONLY;
-
- error = xfs_parseargs(mp, (char *)data, args, 0);
+ error = xfs_parseargs(mp, (char *)data, &mtpt);
if (error)
- goto out_free_mp;
+ goto out_free_fsname;
sb_min_blocksize(sb, BBSIZE);
sb->s_xattr = xfs_xattr_handlers;
@@ -1702,33 +1423,28 @@ xfs_fs_fill_super(
sb->s_qcop = &xfs_quotactl_operations;
sb->s_op = &xfs_super_operations;
- error = xfs_dmops_get(mp, args);
+ error = xfs_dmops_get(mp);
if (error)
- goto out_free_mp;
- error = xfs_qmops_get(mp, args);
+ goto out_free_fsname;
+ error = xfs_qmops_get(mp);
if (error)
goto out_put_dmops;
- if (args->flags & XFSMNT_QUIET)
+ if (silent)
flags |= XFS_MFSI_QUIET;
- error = xfs_open_devices(mp, args);
+ error = xfs_open_devices(mp);
if (error)
goto out_put_qmops;
if (xfs_icsb_init_counters(mp))
mp->m_flags |= XFS_MOUNT_NO_PERCPU_SB;
- /*
- * Setup flags based on mount(2) options and then the superblock
- */
- error = xfs_start_flags(args, mp);
- if (error)
- goto out_free_fsname;
error = xfs_readsb(mp, flags);
if (error)
- goto out_free_fsname;
- error = xfs_finish_flags(args, mp);
+ goto out_destroy_counters;
+
+ error = xfs_finish_flags(mp);
if (error)
goto out_free_sb;
@@ -1747,7 +1463,7 @@ xfs_fs_fill_super(
if (error)
goto out_filestream_unmount;
- XFS_SEND_MOUNT(mp, DM_RIGHT_NULL, args->mtpt, args->fsname);
+ XFS_SEND_MOUNT(mp, DM_RIGHT_NULL, mtpt, mp->m_fsname);
sb->s_dirt = 1;
sb->s_magic = XFS_SB_MAGIC;
@@ -1772,35 +1488,31 @@ xfs_fs_fill_super(
goto fail_vnrele;
}
- mp->m_sync_work.w_syncer = xfs_sync_worker;
- mp->m_sync_work.w_mount = mp;
- mp->m_sync_task = kthread_run(xfssyncd, mp, "xfssyncd");
- if (IS_ERR(mp->m_sync_task)) {
- error = -PTR_ERR(mp->m_sync_task);
+ error = xfs_syncd_init(mp);
+ if (error)
goto fail_vnrele;
- }
- xfs_itrace_exit(XFS_I(sb->s_root->d_inode));
+ kfree(mtpt);
- kfree(args);
+ xfs_itrace_exit(XFS_I(sb->s_root->d_inode));
return 0;
out_filestream_unmount:
xfs_filestream_unmount(mp);
out_free_sb:
xfs_freesb(mp);
- out_free_fsname:
- xfs_free_fsname(mp);
+ out_destroy_counters:
xfs_icsb_destroy_counters(mp);
xfs_close_devices(mp);
out_put_qmops:
xfs_qmops_put(mp);
out_put_dmops:
xfs_dmops_put(mp);
- out_free_mp:
+ out_free_fsname:
+ xfs_free_fsname(mp);
+ kfree(mtpt);
kfree(mp);
- out_free_args:
- kfree(args);
+ out:
return -error;
fail_vnrele:
@@ -1820,8 +1532,6 @@ xfs_fs_fill_super(
xfs_filestream_unmount(mp);
XFS_bflush(mp->m_ddev_targp);
- error = xfs_unmount_flush(mp, 0);
- WARN_ON(error);
xfs_unmountfs(mp);
goto out_free_sb;
@@ -1882,10 +1592,19 @@ xfs_alloc_trace_bufs(void)
if (!xfs_bmap_trace_buf)
goto out_free_alloc_trace;
#endif
-#ifdef XFS_BMBT_TRACE
+#ifdef XFS_BTREE_TRACE
+ xfs_allocbt_trace_buf = ktrace_alloc(XFS_ALLOCBT_TRACE_SIZE,
+ KM_MAYFAIL);
+ if (!xfs_allocbt_trace_buf)
+ goto out_free_bmap_trace;
+
+ xfs_inobt_trace_buf = ktrace_alloc(XFS_INOBT_TRACE_SIZE, KM_MAYFAIL);
+ if (!xfs_inobt_trace_buf)
+ goto out_free_allocbt_trace;
+
xfs_bmbt_trace_buf = ktrace_alloc(XFS_BMBT_TRACE_SIZE, KM_MAYFAIL);
if (!xfs_bmbt_trace_buf)
- goto out_free_bmap_trace;
+ goto out_free_inobt_trace;
#endif
#ifdef XFS_ATTR_TRACE
xfs_attr_trace_buf = ktrace_alloc(XFS_ATTR_TRACE_SIZE, KM_MAYFAIL);
@@ -1907,8 +1626,12 @@ xfs_alloc_trace_bufs(void)
ktrace_free(xfs_attr_trace_buf);
out_free_bmbt_trace:
#endif
-#ifdef XFS_BMBT_TRACE
+#ifdef XFS_BTREE_TRACE
ktrace_free(xfs_bmbt_trace_buf);
+ out_free_inobt_trace:
+ ktrace_free(xfs_inobt_trace_buf);
+ out_free_allocbt_trace:
+ ktrace_free(xfs_allocbt_trace_buf);
out_free_bmap_trace:
#endif
#ifdef XFS_BMAP_TRACE
@@ -1931,8 +1654,10 @@ xfs_free_trace_bufs(void)
#ifdef XFS_ATTR_TRACE
ktrace_free(xfs_attr_trace_buf);
#endif
-#ifdef XFS_BMBT_TRACE
+#ifdef XFS_BTREE_TRACE
ktrace_free(xfs_bmbt_trace_buf);
+ ktrace_free(xfs_inobt_trace_buf);
+ ktrace_free(xfs_allocbt_trace_buf);
#endif
#ifdef XFS_BMAP_TRACE
ktrace_free(xfs_bmap_trace_buf);
@@ -1945,16 +1670,10 @@ xfs_free_trace_bufs(void)
STATIC int __init
xfs_init_zones(void)
{
- xfs_vnode_zone = kmem_zone_init_flags(sizeof(struct inode), "xfs_vnode",
- KM_ZONE_HWALIGN | KM_ZONE_RECLAIM |
- KM_ZONE_SPREAD,
- xfs_fs_inode_init_once);
- if (!xfs_vnode_zone)
- goto out;
xfs_ioend_zone = kmem_zone_init(sizeof(xfs_ioend_t), "xfs_ioend");
if (!xfs_ioend_zone)
- goto out_destroy_vnode_zone;
+ goto out;
xfs_ioend_pool = mempool_create_slab_pool(4 * MAX_BUF_PER_PAGE,
xfs_ioend_zone);
@@ -1970,6 +1689,7 @@ xfs_init_zones(void)
"xfs_bmap_free_item");
if (!xfs_bmap_free_item_zone)
goto out_destroy_log_ticket_zone;
+
xfs_btree_cur_zone = kmem_zone_init(sizeof(xfs_btree_cur_t),
"xfs_btree_cur");
if (!xfs_btree_cur_zone)
@@ -2017,8 +1737,8 @@ xfs_init_zones(void)
xfs_inode_zone =
kmem_zone_init_flags(sizeof(xfs_inode_t), "xfs_inode",
- KM_ZONE_HWALIGN | KM_ZONE_RECLAIM |
- KM_ZONE_SPREAD, NULL);
+ KM_ZONE_HWALIGN | KM_ZONE_RECLAIM | KM_ZONE_SPREAD,
+ xfs_fs_inode_init_once);
if (!xfs_inode_zone)
goto out_destroy_efi_zone;
@@ -2066,8 +1786,6 @@ xfs_init_zones(void)
mempool_destroy(xfs_ioend_pool);
out_destroy_ioend_zone:
kmem_zone_destroy(xfs_ioend_zone);
- out_destroy_vnode_zone:
- kmem_zone_destroy(xfs_vnode_zone);
out:
return -ENOMEM;
}
@@ -2092,7 +1810,6 @@ xfs_destroy_zones(void)
kmem_zone_destroy(xfs_log_ticket_zone);
mempool_destroy(xfs_ioend_pool);
kmem_zone_destroy(xfs_ioend_zone);
- kmem_zone_destroy(xfs_vnode_zone);
}
@@ -2100,13 +1817,12 @@ STATIC int __init
init_xfs_fs(void)
{
int error;
- static char message[] __initdata = KERN_INFO \
- XFS_VERSION_STRING " with " XFS_BUILD_OPTIONS " enabled\n";
- printk(message);
+ printk(KERN_INFO XFS_VERSION_STRING " with "
+ XFS_BUILD_OPTIONS " enabled\n");
ktrace_init(64);
- vn_init();
+ xfs_ioend_init();
xfs_dir_startup();
error = xfs_init_zones();
diff --git a/fs/xfs/linux-2.6/xfs_super.h b/fs/xfs/linux-2.6/xfs_super.h
index fe2ef4e6a0f9..d5d776d4cd67 100644
--- a/fs/xfs/linux-2.6/xfs_super.h
+++ b/fs/xfs/linux-2.6/xfs_super.h
@@ -20,24 +20,12 @@
#include <linux/exportfs.h>
-#ifdef CONFIG_XFS_DMAPI
-# define vfs_insertdmapi(vfs) vfs_insertops(vfsp, &xfs_dmops)
-# define vfs_initdmapi() dmapi_init()
-# define vfs_exitdmapi() dmapi_uninit()
-#else
-# define vfs_insertdmapi(vfs) do { } while (0)
-# define vfs_initdmapi() do { } while (0)
-# define vfs_exitdmapi() do { } while (0)
-#endif
-
#ifdef CONFIG_XFS_QUOTA
-# define vfs_insertquota(vfs) vfs_insertops(vfsp, &xfs_qmops)
extern void xfs_qm_init(void);
extern void xfs_qm_exit(void);
# define vfs_initquota() xfs_qm_init()
# define vfs_exitquota() xfs_qm_exit()
#else
-# define vfs_insertquota(vfs) do { } while (0)
# define vfs_initquota() do { } while (0)
# define vfs_exitquota() do { } while (0)
#endif
@@ -101,9 +89,6 @@ struct block_device;
extern __uint64_t xfs_max_file_offset(unsigned int);
-extern void xfs_flush_inode(struct xfs_inode *);
-extern void xfs_flush_device(struct xfs_inode *);
-
extern void xfs_blkdev_issue_flush(struct xfs_buftarg *);
extern const struct export_operations xfs_export_operations;
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
new file mode 100644
index 000000000000..2ed035354c26
--- /dev/null
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -0,0 +1,762 @@
+/*
+ * Copyright (c) 2000-2005 Silicon Graphics, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+#include "xfs.h"
+#include "xfs_fs.h"
+#include "xfs_types.h"
+#include "xfs_bit.h"
+#include "xfs_log.h"
+#include "xfs_inum.h"
+#include "xfs_trans.h"
+#include "xfs_sb.h"
+#include "xfs_ag.h"
+#include "xfs_dir2.h"
+#include "xfs_dmapi.h"
+#include "xfs_mount.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_alloc_btree.h"
+#include "xfs_ialloc_btree.h"
+#include "xfs_btree.h"
+#include "xfs_dir2_sf.h"
+#include "xfs_attr_sf.h"
+#include "xfs_inode.h"
+#include "xfs_dinode.h"
+#include "xfs_error.h"
+#include "xfs_mru_cache.h"
+#include "xfs_filestream.h"
+#include "xfs_vnodeops.h"
+#include "xfs_utils.h"
+#include "xfs_buf_item.h"
+#include "xfs_inode_item.h"
+#include "xfs_rw.h"
+
+#include <linux/kthread.h>
+#include <linux/freezer.h>
+
+/*
+ * Sync all the inodes in the given AG according to the
+ * direction given by the flags.
+ */
+STATIC int
+xfs_sync_inodes_ag(
+ xfs_mount_t *mp,
+ int ag,
+ int flags)
+{
+ xfs_perag_t *pag = &mp->m_perag[ag];
+ int nr_found;
+ uint32_t first_index = 0;
+ int error = 0;
+ int last_error = 0;
+ int fflag = XFS_B_ASYNC;
+
+ if (flags & SYNC_DELWRI)
+ fflag = XFS_B_DELWRI;
+ if (flags & SYNC_WAIT)
+ fflag = 0; /* synchronous overrides all */
+
+ do {
+ struct inode *inode;
+ xfs_inode_t *ip = NULL;
+ int lock_flags = XFS_ILOCK_SHARED;
+
+ /*
+ * use a gang lookup to find the next inode in the tree
+ * as the tree is sparse and a gang lookup walks to find
+ * the number of objects requested.
+ */
+ read_lock(&pag->pag_ici_lock);
+ nr_found = radix_tree_gang_lookup(&pag->pag_ici_root,
+ (void**)&ip, first_index, 1);
+
+ if (!nr_found) {
+ read_unlock(&pag->pag_ici_lock);
+ break;
+ }
+
+ /*
+ * Update the index for the next lookup. Catch overflows
+ * into the next AG range which can occur if we have inodes
+ * in the last block of the AG and we are currently
+ * pointing to the last inode.
+ */
+ first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1);
+ if (first_index < XFS_INO_TO_AGINO(mp, ip->i_ino)) {
+ read_unlock(&pag->pag_ici_lock);
+ break;
+ }
+
+ /* nothing to sync during shutdown */
+ if (XFS_FORCED_SHUTDOWN(mp)) {
+ read_unlock(&pag->pag_ici_lock);
+ return 0;
+ }
+
+ /*
+ * If we can't get a reference on the inode, it must be
+ * in reclaim. Leave it for the reclaim code to flush.
+ */
+ inode = VFS_I(ip);
+ if (!igrab(inode)) {
+ read_unlock(&pag->pag_ici_lock);
+ continue;
+ }
+ read_unlock(&pag->pag_ici_lock);
+
+ /* avoid new or bad inodes */
+ if (is_bad_inode(inode) ||
+ xfs_iflags_test(ip, XFS_INEW)) {
+ IRELE(ip);
+ continue;
+ }
+
+ /*
+ * If we have to flush data or wait for I/O completion
+ * we need to hold the iolock.
+ */
+ if ((flags & SYNC_DELWRI) && VN_DIRTY(inode)) {
+ xfs_ilock(ip, XFS_IOLOCK_SHARED);
+ lock_flags |= XFS_IOLOCK_SHARED;
+ error = xfs_flush_pages(ip, 0, -1, fflag, FI_NONE);
+ if (flags & SYNC_IOWAIT)
+ xfs_ioend_wait(ip);
+ }
+ xfs_ilock(ip, XFS_ILOCK_SHARED);
+
+ if ((flags & SYNC_ATTR) && !xfs_inode_clean(ip)) {
+ if (flags & SYNC_WAIT) {
+ xfs_iflock(ip);
+ if (!xfs_inode_clean(ip))
+ error = xfs_iflush(ip, XFS_IFLUSH_SYNC);
+ else
+ xfs_ifunlock(ip);
+ } else if (xfs_iflock_nowait(ip)) {
+ if (!xfs_inode_clean(ip))
+ error = xfs_iflush(ip, XFS_IFLUSH_DELWRI);
+ else
+ xfs_ifunlock(ip);
+ }
+ }
+ xfs_iput(ip, lock_flags);
+
+ if (error)
+ last_error = error;
+ /*
+ * bail out if the filesystem is corrupted.
+ */
+ if (error == EFSCORRUPTED)
+ return XFS_ERROR(error);
+
+ } while (nr_found);
+
+ return last_error;
+}
+
+int
+xfs_sync_inodes(
+ xfs_mount_t *mp,
+ int flags)
+{
+ int error;
+ int last_error;
+ int i;
+ int lflags = XFS_LOG_FORCE;
+
+ if (mp->m_flags & XFS_MOUNT_RDONLY)
+ return 0;
+ error = 0;
+ last_error = 0;
+
+ if (flags & SYNC_WAIT)
+ lflags |= XFS_LOG_SYNC;
+
+ for (i = 0; i < mp->m_sb.sb_agcount; i++) {
+ if (!mp->m_perag[i].pag_ici_init)
+ continue;
+ error = xfs_sync_inodes_ag(mp, i, flags);
+ if (error)
+ last_error = error;
+ if (error == EFSCORRUPTED)
+ break;
+ }
+ if (flags & SYNC_DELWRI)
+ xfs_log_force(mp, 0, lflags);
+
+ return XFS_ERROR(last_error);
+}
+
+STATIC int
+xfs_commit_dummy_trans(
+ struct xfs_mount *mp,
+ uint log_flags)
+{
+ struct xfs_inode *ip = mp->m_rootip;
+ struct xfs_trans *tp;
+ int error;
+
+ /*
+ * Put a dummy transaction in the log to tell recovery
+ * that all others are OK.
+ */
+ tp = xfs_trans_alloc(mp, XFS_TRANS_DUMMY1);
+ error = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 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);
+ 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);
+
+ xfs_log_force(mp, 0, log_flags);
+ return 0;
+}
+
+int
+xfs_sync_fsdata(
+ struct xfs_mount *mp,
+ int flags)
+{
+ struct xfs_buf *bp;
+ struct xfs_buf_log_item *bip;
+ int error = 0;
+
+ /*
+ * If this is xfssyncd() then only sync the superblock if we can
+ * lock it without sleeping and it is not pinned.
+ */
+ if (flags & SYNC_BDFLUSH) {
+ ASSERT(!(flags & SYNC_WAIT));
+
+ bp = xfs_getsb(mp, XFS_BUF_TRYLOCK);
+ if (!bp)
+ goto out;
+
+ bip = XFS_BUF_FSPRIVATE(bp, struct xfs_buf_log_item *);
+ if (!bip || !xfs_buf_item_dirty(bip) || XFS_BUF_ISPINNED(bp))
+ goto out_brelse;
+ } else {
+ bp = xfs_getsb(mp, 0);
+
+ /*
+ * If the buffer is pinned then push on the log so we won't
+ * get stuck waiting in the write for someone, maybe
+ * ourselves, to flush the log.
+ *
+ * Even though we just pushed the log above, we did not have
+ * the superblock buffer locked at that point so it can
+ * become pinned in between there and here.
+ */
+ if (XFS_BUF_ISPINNED(bp))
+ xfs_log_force(mp, 0, XFS_LOG_FORCE);
+ }
+
+
+ if (flags & SYNC_WAIT)
+ XFS_BUF_UNASYNC(bp);
+ else
+ XFS_BUF_ASYNC(bp);
+
+ return xfs_bwrite(mp, bp);
+
+ out_brelse:
+ xfs_buf_relse(bp);
+ out:
+ return error;
+}
+
+/*
+ * When remounting a filesystem read-only or freezing the filesystem, we have
+ * two phases to execute. This first phase is syncing the data before we
+ * quiesce the filesystem, and the second is flushing all the inodes out after
+ * we've waited for all the transactions created by the first phase to
+ * complete. The second phase ensures that the inodes are written to their
+ * location on disk rather than just existing in transactions in the log. This
+ * means after a quiesce there is no log replay required to write the inodes to
+ * disk (this is the main difference between a sync and a quiesce).
+ */
+/*
+ * First stage of freeze - no writers will make progress now we are here,
+ * so we flush delwri and delalloc buffers here, then wait for all I/O to
+ * complete. Data is frozen at that point. Metadata is not frozen,
+ * transactions can still occur here so don't bother flushing the buftarg
+ * because it'll just get dirty again.
+ */
+int
+xfs_quiesce_data(
+ struct xfs_mount *mp)
+{
+ int error;
+
+ /* push non-blocking */
+ xfs_sync_inodes(mp, SYNC_DELWRI|SYNC_BDFLUSH);
+ XFS_QM_DQSYNC(mp, SYNC_BDFLUSH);
+ xfs_filestream_flush(mp);
+
+ /* push and block */
+ xfs_sync_inodes(mp, SYNC_DELWRI|SYNC_WAIT|SYNC_IOWAIT);
+ XFS_QM_DQSYNC(mp, SYNC_WAIT);
+
+ /* write superblock and hoover up shutdown errors */
+ error = xfs_sync_fsdata(mp, 0);
+
+ /* flush data-only devices */
+ if (mp->m_rtdev_targp)
+ XFS_bflush(mp->m_rtdev_targp);
+
+ return error;
+}
+
+STATIC void
+xfs_quiesce_fs(
+ struct xfs_mount *mp)
+{
+ int count = 0, pincount;
+
+ xfs_flush_buftarg(mp->m_ddev_targp, 0);
+ xfs_reclaim_inodes(mp, 0, XFS_IFLUSH_DELWRI_ELSE_ASYNC);
+
+ /*
+ * This loop must run at least twice. The first instance of the loop
+ * will flush most meta data but that will generate more meta data
+ * (typically directory updates). Which then must be flushed and
+ * logged before we can write the unmount record.
+ */
+ do {
+ xfs_sync_inodes(mp, SYNC_ATTR|SYNC_WAIT);
+ pincount = xfs_flush_buftarg(mp->m_ddev_targp, 1);
+ if (!pincount) {
+ delay(50);
+ count++;
+ }
+ } while (count < 2);
+}
+
+/*
+ * Second stage of a quiesce. The data is already synced, now we have to take
+ * care of the metadata. New transactions are already blocked, so we need to
+ * wait for any remaining transactions to drain out before proceding.
+ */
+void
+xfs_quiesce_attr(
+ struct xfs_mount *mp)
+{
+ int error = 0;
+
+ /* wait for all modifications to complete */
+ while (atomic_read(&mp->m_active_trans) > 0)
+ delay(100);
+
+ /* flush inodes and push all remaining buffers out to disk */
+ xfs_quiesce_fs(mp);
+
+ ASSERT_ALWAYS(atomic_read(&mp->m_active_trans) == 0);
+
+ /* Push the superblock and write an unmount record */
+ error = xfs_log_sbcount(mp, 1);
+ if (error)
+ xfs_fs_cmn_err(CE_WARN, mp,
+ "xfs_attr_quiesce: failed to log sb changes. "
+ "Frozen image may not be consistent.");
+ xfs_log_unmount_write(mp);
+ xfs_unmountfs_writesb(mp);
+}
+
+/*
+ * Enqueue a work item to be picked up by the vfs xfssyncd thread.
+ * Doing this has two advantages:
+ * - It saves on stack space, which is tight in certain situations
+ * - It can be used (with care) as a mechanism to avoid deadlocks.
+ * Flushing while allocating in a full filesystem requires both.
+ */
+STATIC void
+xfs_syncd_queue_work(
+ struct xfs_mount *mp,
+ void *data,
+ void (*syncer)(struct xfs_mount *, void *))
+{
+ struct bhv_vfs_sync_work *work;
+
+ work = kmem_alloc(sizeof(struct bhv_vfs_sync_work), KM_SLEEP);
+ INIT_LIST_HEAD(&work->w_list);
+ work->w_syncer = syncer;
+ work->w_data = data;
+ work->w_mount = mp;
+ spin_lock(&mp->m_sync_lock);
+ list_add_tail(&work->w_list, &mp->m_sync_list);
+ spin_unlock(&mp->m_sync_lock);
+ wake_up_process(mp->m_sync_task);
+}
+
+/*
+ * Flush delayed allocate data, attempting to free up reserved space
+ * from existing allocations. At this point a new allocation attempt
+ * has failed with ENOSPC and we are in the process of scratching our
+ * heads, looking about for more room...
+ */
+STATIC void
+xfs_flush_inode_work(
+ struct xfs_mount *mp,
+ void *arg)
+{
+ struct inode *inode = arg;
+ filemap_flush(inode->i_mapping);
+ iput(inode);
+}
+
+void
+xfs_flush_inode(
+ xfs_inode_t *ip)
+{
+ struct inode *inode = VFS_I(ip);
+
+ igrab(inode);
+ xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_inode_work);
+ delay(msecs_to_jiffies(500));
+}
+
+/*
+ * This is the "bigger hammer" version of xfs_flush_inode_work...
+ * (IOW, "If at first you don't succeed, use a Bigger Hammer").
+ */
+STATIC void
+xfs_flush_device_work(
+ struct xfs_mount *mp,
+ void *arg)
+{
+ struct inode *inode = arg;
+ sync_blockdev(mp->m_super->s_bdev);
+ iput(inode);
+}
+
+void
+xfs_flush_device(
+ xfs_inode_t *ip)
+{
+ struct inode *inode = VFS_I(ip);
+
+ igrab(inode);
+ xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_device_work);
+ delay(msecs_to_jiffies(500));
+ xfs_log_force(ip->i_mount, (xfs_lsn_t)0, XFS_LOG_FORCE|XFS_LOG_SYNC);
+}
+
+/*
+ * Every sync period we need to unpin all items, reclaim inodes, sync
+ * quota and write out the superblock. We might need to cover the log
+ * to indicate it is idle.
+ */
+STATIC void
+xfs_sync_worker(
+ struct xfs_mount *mp,
+ void *unused)
+{
+ int error;
+
+ if (!(mp->m_flags & XFS_MOUNT_RDONLY)) {
+ xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE);
+ xfs_reclaim_inodes(mp, 0, XFS_IFLUSH_DELWRI_ELSE_ASYNC);
+ /* dgc: errors ignored here */
+ error = XFS_QM_DQSYNC(mp, SYNC_BDFLUSH);
+ error = xfs_sync_fsdata(mp, SYNC_BDFLUSH);
+ 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);
+}
+
+STATIC int
+xfssyncd(
+ void *arg)
+{
+ struct xfs_mount *mp = arg;
+ long timeleft;
+ bhv_vfs_sync_work_t *work, *n;
+ LIST_HEAD (tmp);
+
+ set_freezable();
+ timeleft = xfs_syncd_centisecs * msecs_to_jiffies(10);
+ for (;;) {
+ timeleft = schedule_timeout_interruptible(timeleft);
+ /* swsusp */
+ try_to_freeze();
+ if (kthread_should_stop() && list_empty(&mp->m_sync_list))
+ break;
+
+ spin_lock(&mp->m_sync_lock);
+ /*
+ * We can get woken by laptop mode, to do a sync -
+ * that's the (only!) case where the list would be
+ * empty with time remaining.
+ */
+ if (!timeleft || list_empty(&mp->m_sync_list)) {
+ if (!timeleft)
+ timeleft = xfs_syncd_centisecs *
+ msecs_to_jiffies(10);
+ INIT_LIST_HEAD(&mp->m_sync_work.w_list);
+ list_add_tail(&mp->m_sync_work.w_list,
+ &mp->m_sync_list);
+ }
+ list_for_each_entry_safe(work, n, &mp->m_sync_list, w_list)
+ list_move(&work->w_list, &tmp);
+ spin_unlock(&mp->m_sync_lock);
+
+ list_for_each_entry_safe(work, n, &tmp, w_list) {
+ (*work->w_syncer)(mp, work->w_data);
+ list_del(&work->w_list);
+ if (work == &mp->m_sync_work)
+ continue;
+ kmem_free(work);
+ }
+ }
+
+ return 0;
+}
+
+int
+xfs_syncd_init(
+ struct xfs_mount *mp)
+{
+ mp->m_sync_work.w_syncer = xfs_sync_worker;
+ mp->m_sync_work.w_mount = mp;
+ mp->m_sync_task = kthread_run(xfssyncd, mp, "xfssyncd");
+ if (IS_ERR(mp->m_sync_task))
+ return -PTR_ERR(mp->m_sync_task);
+ return 0;
+}
+
+void
+xfs_syncd_stop(
+ struct xfs_mount *mp)
+{
+ kthread_stop(mp->m_sync_task);
+}
+
+int
+xfs_reclaim_inode(
+ xfs_inode_t *ip,
+ int locked,
+ int sync_mode)
+{
+ xfs_perag_t *pag = xfs_get_perag(ip->i_mount, ip->i_ino);
+
+ /* The hash lock here protects a thread in xfs_iget_core from
+ * racing with us on linking the inode back with a vnode.
+ * Once we have the XFS_IRECLAIM flag set it will not touch
+ * us.
+ */
+ write_lock(&pag->pag_ici_lock);
+ spin_lock(&ip->i_flags_lock);
+ if (__xfs_iflags_test(ip, XFS_IRECLAIM) ||
+ !__xfs_iflags_test(ip, XFS_IRECLAIMABLE)) {
+ spin_unlock(&ip->i_flags_lock);
+ write_unlock(&pag->pag_ici_lock);
+ if (locked) {
+ xfs_ifunlock(ip);
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ }
+ return 1;
+ }
+ __xfs_iflags_set(ip, XFS_IRECLAIM);
+ spin_unlock(&ip->i_flags_lock);
+ write_unlock(&pag->pag_ici_lock);
+ xfs_put_perag(ip->i_mount, pag);
+
+ /*
+ * If the inode is still dirty, then flush it out. If the inode
+ * is not in the AIL, then it will be OK to flush it delwri as
+ * long as xfs_iflush() does not keep any references to the inode.
+ * We leave that decision up to xfs_iflush() since it has the
+ * knowledge of whether it's OK to simply do a delwri flush of
+ * the inode or whether we need to wait until the inode is
+ * pulled from the AIL.
+ * We get the flush lock regardless, though, just to make sure
+ * we don't free it while it is being flushed.
+ */
+ if (!locked) {
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ xfs_iflock(ip);
+ }
+
+ /*
+ * In the case of a forced shutdown we rely on xfs_iflush() to
+ * wait for the inode to be unpinned before returning an error.
+ */
+ if (!is_bad_inode(VFS_I(ip)) && xfs_iflush(ip, sync_mode) == 0) {
+ /* synchronize with xfs_iflush_done */
+ xfs_iflock(ip);
+ xfs_ifunlock(ip);
+ }
+
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ xfs_ireclaim(ip);
+ return 0;
+}
+
+/*
+ * We set the inode flag atomically with the radix tree tag.
+ * Once we get tag lookups on the radix tree, this inode flag
+ * can go away.
+ */
+void
+xfs_inode_set_reclaim_tag(
+ xfs_inode_t *ip)
+{
+ xfs_mount_t *mp = ip->i_mount;
+ xfs_perag_t *pag = xfs_get_perag(mp, ip->i_ino);
+
+ read_lock(&pag->pag_ici_lock);
+ spin_lock(&ip->i_flags_lock);
+ radix_tree_tag_set(&pag->pag_ici_root,
+ XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG);
+ __xfs_iflags_set(ip, XFS_IRECLAIMABLE);
+ spin_unlock(&ip->i_flags_lock);
+ read_unlock(&pag->pag_ici_lock);
+ xfs_put_perag(mp, pag);
+}
+
+void
+__xfs_inode_clear_reclaim_tag(
+ xfs_mount_t *mp,
+ xfs_perag_t *pag,
+ xfs_inode_t *ip)
+{
+ radix_tree_tag_clear(&pag->pag_ici_root,
+ XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG);
+}
+
+void
+xfs_inode_clear_reclaim_tag(
+ xfs_inode_t *ip)
+{
+ xfs_mount_t *mp = ip->i_mount;
+ xfs_perag_t *pag = xfs_get_perag(mp, ip->i_ino);
+
+ read_lock(&pag->pag_ici_lock);
+ spin_lock(&ip->i_flags_lock);
+ __xfs_inode_clear_reclaim_tag(mp, pag, ip);
+ spin_unlock(&ip->i_flags_lock);
+ read_unlock(&pag->pag_ici_lock);
+ xfs_put_perag(mp, pag);
+}
+
+
+STATIC void
+xfs_reclaim_inodes_ag(
+ xfs_mount_t *mp,
+ int ag,
+ int noblock,
+ int mode)
+{
+ xfs_inode_t *ip = NULL;
+ xfs_perag_t *pag = &mp->m_perag[ag];
+ int nr_found;
+ uint32_t first_index;
+ int skipped;
+
+restart:
+ first_index = 0;
+ skipped = 0;
+ do {
+ /*
+ * use a gang lookup to find the next inode in the tree
+ * as the tree is sparse and a gang lookup walks to find
+ * the number of objects requested.
+ */
+ read_lock(&pag->pag_ici_lock);
+ nr_found = radix_tree_gang_lookup_tag(&pag->pag_ici_root,
+ (void**)&ip, first_index, 1,
+ XFS_ICI_RECLAIM_TAG);
+
+ if (!nr_found) {
+ read_unlock(&pag->pag_ici_lock);
+ break;
+ }
+
+ /*
+ * Update the index for the next lookup. Catch overflows
+ * into the next AG range which can occur if we have inodes
+ * in the last block of the AG and we are currently
+ * pointing to the last inode.
+ */
+ first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1);
+ if (first_index < XFS_INO_TO_AGINO(mp, ip->i_ino)) {
+ read_unlock(&pag->pag_ici_lock);
+ break;
+ }
+
+ /* ignore if already under reclaim */
+ if (xfs_iflags_test(ip, XFS_IRECLAIM)) {
+ read_unlock(&pag->pag_ici_lock);
+ continue;
+ }
+
+ if (noblock) {
+ if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) {
+ read_unlock(&pag->pag_ici_lock);
+ continue;
+ }
+ if (xfs_ipincount(ip) ||
+ !xfs_iflock_nowait(ip)) {
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ read_unlock(&pag->pag_ici_lock);
+ continue;
+ }
+ }
+ read_unlock(&pag->pag_ici_lock);
+
+ /*
+ * hmmm - this is an inode already in reclaim. Do
+ * we even bother catching it here?
+ */
+ if (xfs_reclaim_inode(ip, noblock, mode))
+ skipped++;
+ } while (nr_found);
+
+ if (skipped) {
+ delay(1);
+ goto restart;
+ }
+ return;
+
+}
+
+int
+xfs_reclaim_inodes(
+ xfs_mount_t *mp,
+ int noblock,
+ int mode)
+{
+ int i;
+
+ for (i = 0; i < mp->m_sb.sb_agcount; i++) {
+ if (!mp->m_perag[i].pag_ici_init)
+ continue;
+ xfs_reclaim_inodes_ag(mp, i, noblock, mode);
+ }
+ return 0;
+}
+
+
diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h
new file mode 100644
index 000000000000..5f6de1efe1f6
--- /dev/null
+++ b/fs/xfs/linux-2.6/xfs_sync.h
@@ -0,0 +1,55 @@
+/*
+ * Copyright (c) 2000-2006 Silicon Graphics, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+#ifndef XFS_SYNC_H
+#define XFS_SYNC_H 1
+
+struct xfs_mount;
+
+typedef struct bhv_vfs_sync_work {
+ struct list_head w_list;
+ struct xfs_mount *w_mount;
+ void *w_data; /* syncer routine argument */
+ void (*w_syncer)(struct xfs_mount *, void *);
+} bhv_vfs_sync_work_t;
+
+#define SYNC_ATTR 0x0001 /* sync attributes */
+#define SYNC_DELWRI 0x0002 /* look at delayed writes */
+#define SYNC_WAIT 0x0004 /* wait for i/o to complete */
+#define SYNC_BDFLUSH 0x0008 /* BDFLUSH is calling -- don't block */
+#define SYNC_IOWAIT 0x0010 /* wait for all I/O to complete */
+
+int xfs_syncd_init(struct xfs_mount *mp);
+void xfs_syncd_stop(struct xfs_mount *mp);
+
+int xfs_sync_inodes(struct xfs_mount *mp, int flags);
+int xfs_sync_fsdata(struct xfs_mount *mp, int flags);
+
+int xfs_quiesce_data(struct xfs_mount *mp);
+void xfs_quiesce_attr(struct xfs_mount *mp);
+
+void xfs_flush_inode(struct xfs_inode *ip);
+void xfs_flush_device(struct xfs_inode *ip);
+
+int xfs_reclaim_inode(struct xfs_inode *ip, int locked, int sync_mode);
+int xfs_reclaim_inodes(struct xfs_mount *mp, int noblock, int mode);
+
+void xfs_inode_set_reclaim_tag(struct xfs_inode *ip);
+void xfs_inode_clear_reclaim_tag(struct xfs_inode *ip);
+void __xfs_inode_clear_reclaim_tag(struct xfs_mount *mp, struct xfs_perag *pag,
+ struct xfs_inode *ip);
+#endif
diff --git a/fs/xfs/linux-2.6/xfs_sysctl.c b/fs/xfs/linux-2.6/xfs_sysctl.c
index 7dacb5bbde3f..916c0ffb6083 100644
--- a/fs/xfs/linux-2.6/xfs_sysctl.c
+++ b/fs/xfs/linux-2.6/xfs_sysctl.c
@@ -56,17 +56,6 @@ xfs_stats_clear_proc_handler(
static ctl_table xfs_table[] = {
{
- .ctl_name = XFS_RESTRICT_CHOWN,
- .procname = "restrict_chown",
- .data = &xfs_params.restrict_chown.val,
- .maxlen = sizeof(int),
- .mode = 0644,
- .proc_handler = &proc_dointvec_minmax,
- .strategy = &sysctl_intvec,
- .extra1 = &xfs_params.restrict_chown.min,
- .extra2 = &xfs_params.restrict_chown.max
- },
- {
.ctl_name = XFS_SGID_INHERIT,
.procname = "irix_sgid_inherit",
.data = &xfs_params.sgid_inherit.val,
diff --git a/fs/xfs/linux-2.6/xfs_sysctl.h b/fs/xfs/linux-2.6/xfs_sysctl.h
index 4aadb8056c37..b9937d450f8e 100644
--- a/fs/xfs/linux-2.6/xfs_sysctl.h
+++ b/fs/xfs/linux-2.6/xfs_sysctl.h
@@ -31,7 +31,6 @@ typedef struct xfs_sysctl_val {
} xfs_sysctl_val_t;
typedef struct xfs_param {
- xfs_sysctl_val_t restrict_chown;/* Root/non-root can give away files.*/
xfs_sysctl_val_t sgid_inherit; /* Inherit S_ISGID if process' GID is
* not a member of parent dir GID. */
xfs_sysctl_val_t symlink_mode; /* Link creat mode affected by umask */
@@ -68,7 +67,7 @@ typedef struct xfs_param {
enum {
/* XFS_REFCACHE_SIZE = 1 */
/* XFS_REFCACHE_PURGE = 2 */
- XFS_RESTRICT_CHOWN = 3,
+ /* XFS_RESTRICT_CHOWN = 3 */
XFS_SGID_INHERIT = 4,
XFS_SYMLINK_MODE = 5,
XFS_PANIC_MASK = 6,
diff --git a/fs/xfs/linux-2.6/xfs_vfs.h b/fs/xfs/linux-2.6/xfs_vfs.h
deleted file mode 100644
index 7e60c7776b1c..000000000000
--- a/fs/xfs/linux-2.6/xfs_vfs.h
+++ /dev/null
@@ -1,77 +0,0 @@
-/*
- * Copyright (c) 2000-2006 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write the Free Software Foundation,
- * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
- */
-#ifndef __XFS_VFS_H__
-#define __XFS_VFS_H__
-
-#include <linux/vfs.h>
-#include "xfs_fs.h"
-
-struct inode;
-
-struct fid;
-struct cred;
-struct seq_file;
-struct super_block;
-struct xfs_inode;
-struct xfs_mount;
-struct xfs_mount_args;
-
-typedef struct kstatfs bhv_statvfs_t;
-
-typedef struct bhv_vfs_sync_work {
- struct list_head w_list;
- struct xfs_mount *w_mount;
- void *w_data; /* syncer routine argument */
- void (*w_syncer)(struct xfs_mount *, void *);
-} bhv_vfs_sync_work_t;
-
-#define SYNC_ATTR 0x0001 /* sync attributes */
-#define SYNC_CLOSE 0x0002 /* close file system down */
-#define SYNC_DELWRI 0x0004 /* look at delayed writes */
-#define SYNC_WAIT 0x0008 /* wait for i/o to complete */
-#define SYNC_BDFLUSH 0x0010 /* BDFLUSH is calling -- don't block */
-#define SYNC_FSDATA 0x0020 /* flush fs data (e.g. superblocks) */
-#define SYNC_REFCACHE 0x0040 /* prune some of the nfs ref cache */
-#define SYNC_REMOUNT 0x0080 /* remount readonly, no dummy LRs */
-#define SYNC_IOWAIT 0x0100 /* wait for all I/O to complete */
-
-/*
- * When remounting a filesystem read-only or freezing the filesystem,
- * we have two phases to execute. This first phase is syncing the data
- * before we quiesce the fielsystem, and the second is flushing all the
- * inodes out after we've waited for all the transactions created by
- * the first phase to complete. The second phase uses SYNC_INODE_QUIESCE
- * to ensure that the inodes are written to their location on disk
- * rather than just existing in transactions in the log. This means
- * after a quiesce there is no log replay required to write the inodes
- * to disk (this is the main difference between a sync and a quiesce).
- */
-#define SYNC_DATA_QUIESCE (SYNC_DELWRI|SYNC_FSDATA|SYNC_WAIT|SYNC_IOWAIT)
-#define SYNC_INODE_QUIESCE (SYNC_REMOUNT|SYNC_ATTR|SYNC_WAIT)
-
-#define SHUTDOWN_META_IO_ERROR 0x0001 /* write attempt to metadata failed */
-#define SHUTDOWN_LOG_IO_ERROR 0x0002 /* write attempt to the log failed */
-#define SHUTDOWN_FORCE_UMOUNT 0x0004 /* shutdown from a forced unmount */
-#define SHUTDOWN_CORRUPT_INCORE 0x0008 /* corrupt in-memory data structures */
-#define SHUTDOWN_REMOTE_REQ 0x0010 /* shutdown came from remote cell */
-#define SHUTDOWN_DEVICE_REQ 0x0020 /* failed all paths to the device */
-
-#define xfs_test_for_freeze(mp) ((mp)->m_super->s_frozen)
-#define xfs_wait_for_freeze(mp,l) vfs_check_frozen((mp)->m_super, (l))
-
-#endif /* __XFS_VFS_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_vnode.c b/fs/xfs/linux-2.6/xfs_vnode.c
deleted file mode 100644
index b52528bbbfff..000000000000
--- a/fs/xfs/linux-2.6/xfs_vnode.c
+++ /dev/null
@@ -1,145 +0,0 @@
-/*
- * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc.
- * All Rights Reserved.
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public License as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write the Free Software Foundation,
- * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
- */
-#include "xfs.h"
-#include "xfs_vnodeops.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_inode.h"
-
-/*
- * And this gunk is needed for xfs_mount.h"
- */
-#include "xfs_log.h"
-#include "xfs_trans.h"
-#include "xfs_sb.h"
-#include "xfs_dmapi.h"
-#include "xfs_inum.h"
-#include "xfs_ag.h"
-#include "xfs_mount.h"
-
-
-/*
- * Dedicated vnode inactive/reclaim sync wait queues.
- * Prime number of hash buckets since address is used as the key.
- */
-#define NVSYNC 37
-#define vptosync(v) (&vsync[((unsigned long)v) % NVSYNC])
-static wait_queue_head_t vsync[NVSYNC];
-
-void __init
-vn_init(void)
-{
- int i;
-
- for (i = 0; i < NVSYNC; i++)
- init_waitqueue_head(&vsync[i]);
-}
-
-void
-vn_iowait(
- xfs_inode_t *ip)
-{
- wait_queue_head_t *wq = vptosync(ip);
-
- wait_event(*wq, (atomic_read(&ip->i_iocount) == 0));
-}
-
-void
-vn_iowake(
- xfs_inode_t *ip)
-{
- if (atomic_dec_and_test(&ip->i_iocount))
- wake_up(vptosync(ip));
-}
-
-/*
- * Volume managers supporting multiple paths can send back ENODEV when the
- * final path disappears. In this case continuing to fill the page cache
- * with dirty data which cannot be written out is evil, so prevent that.
- */
-void
-vn_ioerror(
- xfs_inode_t *ip,
- int error,
- char *f,
- int l)
-{
- if (unlikely(error == -ENODEV))
- xfs_do_force_shutdown(ip->i_mount, SHUTDOWN_DEVICE_REQ, f, l);
-}
-
-#ifdef XFS_INODE_TRACE
-
-/*
- * Reference count of Linux inode if present, -1 if the xfs_inode
- * has no associated Linux inode.
- */
-static inline int xfs_icount(struct xfs_inode *ip)
-{
- struct inode *vp = VFS_I(ip);
-
- if (vp)
- return vn_count(vp);
- return -1;
-}
-
-#define KTRACE_ENTER(ip, vk, s, line, ra) \
- ktrace_enter( (ip)->i_trace, \
-/* 0 */ (void *)(__psint_t)(vk), \
-/* 1 */ (void *)(s), \
-/* 2 */ (void *)(__psint_t) line, \
-/* 3 */ (void *)(__psint_t)xfs_icount(ip), \
-/* 4 */ (void *)(ra), \
-/* 5 */ NULL, \
-/* 6 */ (void *)(__psint_t)current_cpu(), \
-/* 7 */ (void *)(__psint_t)current_pid(), \
-/* 8 */ (void *)__return_address, \
-/* 9 */ NULL, NULL, NULL, NULL, NULL, NULL, NULL)
-
-/*
- * Vnode tracing code.
- */
-void
-_xfs_itrace_entry(xfs_inode_t *ip, const char *func, inst_t *ra)
-{
- KTRACE_ENTER(ip, INODE_KTRACE_ENTRY, func, 0, ra);
-}
-
-void
-_xfs_itrace_exit(xfs_inode_t *ip, const char *func, inst_t *ra)
-{
- KTRACE_ENTER(ip, INODE_KTRACE_EXIT, func, 0, ra);
-}
-
-void
-xfs_itrace_hold(xfs_inode_t *ip, char *file, int line, inst_t *ra)
-{
- KTRACE_ENTER(ip, INODE_KTRACE_HOLD, file, line, ra);
-}
-
-void
-_xfs_itrace_ref(xfs_inode_t *ip, char *file, int line, inst_t *ra)
-{
- KTRACE_ENTER(ip, INODE_KTRACE_REF, file, line, ra);
-}
-
-void
-xfs_itrace_rele(xfs_inode_t *ip, char *file, int line, inst_t *ra)
-{
- KTRACE_ENTER(ip, INODE_KTRACE_RELE, file, line, ra);
-}
-#endif /* XFS_INODE_TRACE */
diff --git a/fs/xfs/linux-2.6/xfs_vnode.h b/fs/xfs/linux-2.6/xfs_vnode.h
index 683ce16210ff..f65983a230d3 100644
--- a/fs/xfs/linux-2.6/xfs_vnode.h
+++ b/fs/xfs/linux-2.6/xfs_vnode.h
@@ -18,7 +18,10 @@
#ifndef __XFS_VNODE_H__
#define __XFS_VNODE_H__
+#include "xfs_fs.h"
+
struct file;
+struct xfs_inode;
struct xfs_iomap;
struct attrlist_cursor_kern;
@@ -51,40 +54,6 @@ struct attrlist_cursor_kern;
Prevent VM access to the pages until
the operation completes. */
-
-extern void vn_init(void);
-
-/*
- * Yeah, these don't take vnode anymore at all, all this should be
- * cleaned up at some point.
- */
-extern void vn_iowait(struct xfs_inode *ip);
-extern void vn_iowake(struct xfs_inode *ip);
-extern void vn_ioerror(struct xfs_inode *ip, int error, char *f, int l);
-
-static inline int vn_count(struct inode *vp)
-{
- return atomic_read(&vp->i_count);
-}
-
-#define IHOLD(ip) \
-do { \
- ASSERT(atomic_read(&VFS_I(ip)->i_count) > 0) ; \
- atomic_inc(&(VFS_I(ip)->i_count)); \
- xfs_itrace_hold((ip), __FILE__, __LINE__, (inst_t *)__return_address); \
-} while (0)
-
-#define IRELE(ip) \
-do { \
- xfs_itrace_rele((ip), __FILE__, __LINE__, (inst_t *)__return_address); \
- iput(VFS_I(ip)); \
-} while (0)
-
-static inline struct inode *vn_grab(struct inode *vp)
-{
- return igrab(vp);
-}
-
/*
* Dealing with bad inodes
*/
@@ -121,39 +90,4 @@ static inline void vn_atime_to_time_t(struct inode *vp, time_t *tt)
PAGECACHE_TAG_DIRTY)
-/*
- * Tracking vnode activity.
- */
-#if defined(XFS_INODE_TRACE)
-
-#define INODE_TRACE_SIZE 16 /* number of trace entries */
-#define INODE_KTRACE_ENTRY 1
-#define INODE_KTRACE_EXIT 2
-#define INODE_KTRACE_HOLD 3
-#define INODE_KTRACE_REF 4
-#define INODE_KTRACE_RELE 5
-
-extern void _xfs_itrace_entry(struct xfs_inode *, const char *, inst_t *);
-extern void _xfs_itrace_exit(struct xfs_inode *, const char *, inst_t *);
-extern void xfs_itrace_hold(struct xfs_inode *, char *, int, inst_t *);
-extern void _xfs_itrace_ref(struct xfs_inode *, char *, int, inst_t *);
-extern void xfs_itrace_rele(struct xfs_inode *, char *, int, inst_t *);
-#define xfs_itrace_entry(ip) \
- _xfs_itrace_entry(ip, __func__, (inst_t *)__return_address)
-#define xfs_itrace_exit(ip) \
- _xfs_itrace_exit(ip, __func__, (inst_t *)__return_address)
-#define xfs_itrace_exit_tag(ip, tag) \
- _xfs_itrace_exit(ip, tag, (inst_t *)__return_address)
-#define xfs_itrace_ref(ip) \
- _xfs_itrace_ref(ip, __FILE__, __LINE__, (inst_t *)__return_address)
-
-#else
-#define xfs_itrace_entry(a)
-#define xfs_itrace_exit(a)
-#define xfs_itrace_exit_tag(a, b)
-#define xfs_itrace_hold(a, b, c, d)
-#define xfs_itrace_ref(a)
-#define xfs_itrace_rele(a, b, c, d)
-#endif
-
#endif /* __XFS_VNODE_H__ */