summaryrefslogtreecommitdiff
path: root/fs/xfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/xfs')
-rw-r--r--fs/xfs/Makefile4
-rw-r--r--fs/xfs/linux-2.6/kmem.c57
-rw-r--r--fs/xfs/linux-2.6/kmem.h21
-rw-r--r--fs/xfs/linux-2.6/xfs_acl.c16
-rw-r--r--fs/xfs/linux-2.6/xfs_aops.c477
-rw-r--r--fs/xfs/linux-2.6/xfs_buf.c424
-rw-r--r--fs/xfs/linux-2.6/xfs_buf.h54
-rw-r--r--fs/xfs/linux-2.6/xfs_export.c31
-rw-r--r--fs/xfs/linux-2.6/xfs_file.c856
-rw-r--r--fs/xfs/linux-2.6/xfs_fs_subr.c2
-rw-r--r--fs/xfs/linux-2.6/xfs_ioctl.c33
-rw-r--r--fs/xfs/linux-2.6/xfs_ioctl.h12
-rw-r--r--fs/xfs/linux-2.6/xfs_ioctl32.c24
-rw-r--r--fs/xfs/linux-2.6/xfs_iops.c36
-rw-r--r--fs/xfs/linux-2.6/xfs_linux.h1
-rw-r--r--fs/xfs/linux-2.6/xfs_lrw.c852
-rw-r--r--fs/xfs/linux-2.6/xfs_lrw.h32
-rw-r--r--fs/xfs/linux-2.6/xfs_quotaops.c29
-rw-r--r--fs/xfs/linux-2.6/xfs_super.c204
-rw-r--r--fs/xfs/linux-2.6/xfs_super.h2
-rw-r--r--fs/xfs/linux-2.6/xfs_sync.c416
-rw-r--r--fs/xfs/linux-2.6/xfs_sync.h6
-rw-r--r--fs/xfs/linux-2.6/xfs_trace.c19
-rw-r--r--fs/xfs/linux-2.6/xfs_trace.h635
-rw-r--r--fs/xfs/linux-2.6/xfs_xattr.c35
-rw-r--r--fs/xfs/quota/xfs_dquot.c242
-rw-r--r--fs/xfs/quota/xfs_dquot.h35
-rw-r--r--fs/xfs/quota/xfs_dquot_item.c127
-rw-r--r--fs/xfs/quota/xfs_dquot_item.h4
-rw-r--r--fs/xfs/quota/xfs_qm.c672
-rw-r--r--fs/xfs/quota/xfs_qm.h23
-rw-r--r--fs/xfs/quota/xfs_qm_bhv.c2
-rw-r--r--fs/xfs/quota/xfs_qm_stats.c2
-rw-r--r--fs/xfs/quota/xfs_qm_syscalls.c196
-rw-r--r--fs/xfs/quota/xfs_quota_priv.h102
-rw-r--r--fs/xfs/quota/xfs_trans_dquot.c78
-rw-r--r--fs/xfs/xfs_acl.h8
-rw-r--r--fs/xfs/xfs_ag.h40
-rw-r--r--fs/xfs/xfs_alloc.c421
-rw-r--r--fs/xfs/xfs_alloc.h7
-rw-r--r--fs/xfs/xfs_alloc_btree.c11
-rw-r--r--fs/xfs/xfs_attr.c52
-rw-r--r--fs/xfs/xfs_attr.h3
-rw-r--r--fs/xfs/xfs_attr_leaf.c30
-rw-r--r--fs/xfs/xfs_attr_sf.h2
-rw-r--r--fs/xfs/xfs_bmap.c223
-rw-r--r--fs/xfs/xfs_bmap_btree.c2
-rw-r--r--fs/xfs/xfs_bmap_btree.h1
-rw-r--r--fs/xfs/xfs_btree.c4
-rw-r--r--fs/xfs/xfs_buf_item.c293
-rw-r--r--fs/xfs/xfs_buf_item.h20
-rw-r--r--fs/xfs/xfs_da_btree.c4
-rw-r--r--fs/xfs/xfs_da_btree.h5
-rw-r--r--fs/xfs/xfs_dfrag.c70
-rw-r--r--fs/xfs/xfs_dfrag.h3
-rw-r--r--fs/xfs/xfs_dir2.c8
-rw-r--r--fs/xfs/xfs_dir2.h4
-rw-r--r--fs/xfs/xfs_dir2_block.c9
-rw-r--r--fs/xfs/xfs_dir2_leaf.c2
-rw-r--r--fs/xfs/xfs_dir2_node.c2
-rw-r--r--fs/xfs/xfs_dir2_node.h2
-rw-r--r--fs/xfs/xfs_dir2_sf.c2
-rw-r--r--fs/xfs/xfs_error.c32
-rw-r--r--fs/xfs/xfs_error.h9
-rw-r--r--fs/xfs/xfs_extfree_item.c22
-rw-r--r--fs/xfs/xfs_filestream.c42
-rw-r--r--fs/xfs/xfs_filestream.h28
-rw-r--r--fs/xfs/xfs_fs.h3
-rw-r--r--fs/xfs/xfs_fsops.c42
-rw-r--r--fs/xfs/xfs_ialloc.c204
-rw-r--r--fs/xfs/xfs_iget.c68
-rw-r--r--fs/xfs/xfs_inode.c335
-rw-r--r--fs/xfs/xfs_inode.h20
-rw-r--r--fs/xfs/xfs_inode_item.c166
-rw-r--r--fs/xfs/xfs_inode_item.h6
-rw-r--r--fs/xfs/xfs_iomap.c123
-rw-r--r--fs/xfs/xfs_iomap.h47
-rw-r--r--fs/xfs/xfs_itable.c297
-rw-r--r--fs/xfs/xfs_itable.h17
-rw-r--r--fs/xfs/xfs_log.c1303
-rw-r--r--fs/xfs/xfs_log.h60
-rw-r--r--fs/xfs/xfs_log_cil.c725
-rw-r--r--fs/xfs/xfs_log_priv.h135
-rw-r--r--fs/xfs/xfs_log_recover.c578
-rw-r--r--fs/xfs/xfs_log_recover.h25
-rw-r--r--fs/xfs/xfs_mount.c303
-rw-r--r--fs/xfs/xfs_mount.h33
-rw-r--r--fs/xfs/xfs_mru_cache.c2
-rw-r--r--fs/xfs/xfs_mru_cache.h1
-rw-r--r--fs/xfs/xfs_quota.h12
-rw-r--r--fs/xfs/xfs_rtalloc.c8
-rw-r--r--fs/xfs/xfs_rtalloc.h11
-rw-r--r--fs/xfs/xfs_rw.c155
-rw-r--r--fs/xfs/xfs_rw.h4
-rw-r--r--fs/xfs/xfs_trans.c1253
-rw-r--r--fs/xfs/xfs_trans.h474
-rw-r--r--fs/xfs/xfs_trans_ail.c34
-rw-r--r--fs/xfs/xfs_trans_buf.c474
-rw-r--r--fs/xfs/xfs_trans_inode.c2
-rw-r--r--fs/xfs/xfs_trans_item.c114
-rw-r--r--fs/xfs/xfs_trans_priv.h15
-rw-r--r--fs/xfs/xfs_types.h6
-rw-r--r--fs/xfs/xfs_vnodeops.c124
-rw-r--r--fs/xfs/xfs_vnodeops.h25
104 files changed, 7474 insertions, 6852 deletions
diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index 56641fe52a23..c8fb13f83b3f 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -16,7 +16,7 @@
# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
#
-EXTRA_CFLAGS += -I$(src) -I$(src)/linux-2.6 -funsigned-char
+EXTRA_CFLAGS += -I$(src) -I$(src)/linux-2.6
XFS_LINUX := linux-2.6
@@ -77,6 +77,7 @@ xfs-y += xfs_alloc.o \
xfs_itable.o \
xfs_dfrag.o \
xfs_log.o \
+ xfs_log_cil.o \
xfs_log_recover.o \
xfs_mount.o \
xfs_mru_cache.o \
@@ -105,7 +106,6 @@ xfs-y += $(addprefix $(XFS_LINUX)/, \
xfs_globals.o \
xfs_ioctl.o \
xfs_iops.o \
- xfs_lrw.o \
xfs_super.o \
xfs_sync.o \
xfs_xattr.o)
diff --git a/fs/xfs/linux-2.6/kmem.c b/fs/xfs/linux-2.6/kmem.c
index 2d3f90afe5f1..666c9db48eb6 100644
--- a/fs/xfs/linux-2.6/kmem.c
+++ b/fs/xfs/linux-2.6/kmem.c
@@ -16,16 +16,33 @@
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include <linux/mm.h>
-#include <linux/vmalloc.h>
#include <linux/highmem.h>
+#include <linux/slab.h>
#include <linux/swap.h>
#include <linux/blkdev.h>
#include <linux/backing-dev.h>
#include "time.h"
#include "kmem.h"
-#define MAX_VMALLOCS 6
-#define MAX_SLAB_SIZE 0x20000
+/*
+ * Greedy allocation. May fail and may return vmalloced memory.
+ *
+ * Must be freed using kmem_free_large.
+ */
+void *
+kmem_zalloc_greedy(size_t *size, size_t minsize, size_t maxsize)
+{
+ void *ptr;
+ size_t kmsize = maxsize;
+
+ while (!(ptr = kmem_zalloc_large(kmsize))) {
+ if ((kmsize >>= 1) <= minsize)
+ kmsize = minsize;
+ }
+ if (ptr)
+ *size = kmsize;
+ return ptr;
+}
void *
kmem_alloc(size_t size, unsigned int __nocast flags)
@@ -34,19 +51,8 @@ kmem_alloc(size_t size, unsigned int __nocast flags)
gfp_t lflags = kmem_flags_convert(flags);
void *ptr;
-#ifdef DEBUG
- if (unlikely(!(flags & KM_LARGE) && (size > PAGE_SIZE))) {
- printk(KERN_WARNING "Large %s attempt, size=%ld\n",
- __func__, (long)size);
- dump_stack();
- }
-#endif
-
do {
- if (size < MAX_SLAB_SIZE || retries > MAX_VMALLOCS)
- ptr = kmalloc(size, lflags);
- else
- ptr = __vmalloc(size, lflags, PAGE_KERNEL);
+ ptr = kmalloc(size, lflags);
if (ptr || (flags & (KM_MAYFAIL|KM_NOSLEEP)))
return ptr;
if (!(++retries % 100))
@@ -68,27 +74,6 @@ kmem_zalloc(size_t size, unsigned int __nocast flags)
return ptr;
}
-void *
-kmem_zalloc_greedy(size_t *size, size_t minsize, size_t maxsize,
- unsigned int __nocast flags)
-{
- void *ptr;
- size_t kmsize = maxsize;
- unsigned int kmflags = (flags & ~KM_SLEEP) | KM_NOSLEEP;
-
- while (!(ptr = kmem_zalloc(kmsize, kmflags))) {
- if ((kmsize <= minsize) && (flags & KM_NOSLEEP))
- break;
- if ((kmsize >>= 1) <= minsize) {
- kmsize = minsize;
- kmflags = flags;
- }
- }
- if (ptr)
- *size = kmsize;
- return ptr;
-}
-
void
kmem_free(const void *ptr)
{
diff --git a/fs/xfs/linux-2.6/kmem.h b/fs/xfs/linux-2.6/kmem.h
index 179cbd630f69..f7c8f7a9ea6d 100644
--- a/fs/xfs/linux-2.6/kmem.h
+++ b/fs/xfs/linux-2.6/kmem.h
@@ -21,6 +21,7 @@
#include <linux/slab.h>
#include <linux/sched.h>
#include <linux/mm.h>
+#include <linux/vmalloc.h>
/*
* General memory allocation interfaces
@@ -30,7 +31,6 @@
#define KM_NOSLEEP 0x0002u
#define KM_NOFS 0x0004u
#define KM_MAYFAIL 0x0008u
-#define KM_LARGE 0x0010u
/*
* We use a special process flag to avoid recursive callbacks into
@@ -42,7 +42,7 @@ kmem_flags_convert(unsigned int __nocast flags)
{
gfp_t lflags;
- BUG_ON(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL|KM_LARGE));
+ BUG_ON(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL));
if (flags & KM_NOSLEEP) {
lflags = GFP_ATOMIC | __GFP_NOWARN;
@@ -56,10 +56,25 @@ kmem_flags_convert(unsigned int __nocast flags)
extern void *kmem_alloc(size_t, unsigned int __nocast);
extern void *kmem_zalloc(size_t, unsigned int __nocast);
-extern void *kmem_zalloc_greedy(size_t *, size_t, size_t, unsigned int __nocast);
extern void *kmem_realloc(const void *, size_t, size_t, unsigned int __nocast);
extern void kmem_free(const void *);
+static inline void *kmem_zalloc_large(size_t size)
+{
+ void *ptr;
+
+ ptr = vmalloc(size);
+ if (ptr)
+ memset(ptr, 0, size);
+ return ptr;
+}
+static inline void kmem_free_large(void *ptr)
+{
+ vfree(ptr);
+}
+
+extern void *kmem_zalloc_greedy(size_t *, size_t, size_t);
+
/*
* Zone interfaces
*/
diff --git a/fs/xfs/linux-2.6/xfs_acl.c b/fs/xfs/linux-2.6/xfs_acl.c
index 883ca5ab8af5..9f769b5b38fc 100644
--- a/fs/xfs/linux-2.6/xfs_acl.c
+++ b/fs/xfs/linux-2.6/xfs_acl.c
@@ -22,6 +22,7 @@
#include "xfs_inode.h"
#include "xfs_vnodeops.h"
#include "xfs_trace.h"
+#include <linux/slab.h>
#include <linux/xattr.h>
#include <linux/posix_acl_xattr.h>
@@ -106,7 +107,7 @@ xfs_get_acl(struct inode *inode, int type)
struct posix_acl *acl;
struct xfs_acl *xfs_acl;
int len = sizeof(struct xfs_acl);
- char *ea_name;
+ unsigned char *ea_name;
int error;
acl = get_cached_acl(inode, type);
@@ -133,7 +134,8 @@ xfs_get_acl(struct inode *inode, int type)
if (!xfs_acl)
return ERR_PTR(-ENOMEM);
- error = -xfs_attr_get(ip, ea_name, (char *)xfs_acl, &len, ATTR_ROOT);
+ error = -xfs_attr_get(ip, ea_name, (unsigned char *)xfs_acl,
+ &len, ATTR_ROOT);
if (error) {
/*
* If the attribute doesn't exist make sure we have a negative
@@ -162,7 +164,7 @@ STATIC int
xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
{
struct xfs_inode *ip = XFS_I(inode);
- char *ea_name;
+ unsigned char *ea_name;
int error;
if (S_ISLNK(inode->i_mode))
@@ -194,7 +196,7 @@ xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl)
(sizeof(struct xfs_acl_entry) *
(XFS_ACL_MAX_ENTRIES - acl->a_count));
- error = -xfs_attr_set(ip, ea_name, (char *)xfs_acl,
+ error = -xfs_attr_set(ip, ea_name, (unsigned char *)xfs_acl,
len, ATTR_ROOT);
kfree(xfs_acl);
@@ -262,7 +264,7 @@ xfs_set_mode(struct inode *inode, mode_t mode)
}
static int
-xfs_acl_exists(struct inode *inode, char *name)
+xfs_acl_exists(struct inode *inode, unsigned char *name)
{
int len = sizeof(struct xfs_acl);
@@ -438,14 +440,14 @@ xfs_xattr_acl_set(struct dentry *dentry, const char *name,
return error;
}
-struct xattr_handler xfs_xattr_acl_access_handler = {
+const struct xattr_handler xfs_xattr_acl_access_handler = {
.prefix = POSIX_ACL_XATTR_ACCESS,
.flags = ACL_TYPE_ACCESS,
.get = xfs_xattr_acl_get,
.set = xfs_xattr_acl_set,
};
-struct xattr_handler xfs_xattr_acl_default_handler = {
+const struct xattr_handler xfs_xattr_acl_default_handler = {
.prefix = POSIX_ACL_XATTR_DEFAULT,
.flags = ACL_TYPE_DEFAULT,
.get = xfs_xattr_acl_get,
diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index 66abe36c1213..34640d6dbdcb 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -39,10 +39,21 @@
#include "xfs_iomap.h"
#include "xfs_vnodeops.h"
#include "xfs_trace.h"
+#include "xfs_bmap.h"
+#include <linux/gfp.h>
#include <linux/mpage.h>
#include <linux/pagevec.h>
#include <linux/writeback.h>
+/*
+ * Types of I/O for bmap clustering and I/O completion tracking.
+ */
+enum {
+ IO_READ, /* mapping for a read */
+ IO_DELAY, /* mapping covers delalloc region */
+ IO_UNWRITTEN, /* mapping covers allocated but uninitialized data */
+ IO_NEW /* just allocated */
+};
/*
* Prime number of hash buckets since address is used as the key.
@@ -101,8 +112,9 @@ xfs_count_page_state(
STATIC struct block_device *
xfs_find_bdev_for_inode(
- struct xfs_inode *ip)
+ struct inode *inode)
{
+ struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
if (XFS_IS_REALTIME_INODE(ip))
@@ -163,14 +175,17 @@ xfs_ioend_new_eof(
}
/*
- * Update on-disk file size now that data has been written to disk.
- * The current in-memory file size is i_size. If a write is beyond
- * eof i_new_size will be the intended file size until i_size is
- * updated. If this write does not extend all the way to the valid
- * file size then restrict this update to the end of the write.
+ * Update on-disk file size now that data has been written to disk. The
+ * current in-memory file size is i_size. If a write is beyond eof i_new_size
+ * will be the intended file size until i_size is updated. If this write does
+ * not extend all the way to the valid file size then restrict this update to
+ * the end of the write.
+ *
+ * This function does not block as blocking on the inode lock in IO completion
+ * can lead to IO completion order dependency deadlocks.. If it can't get the
+ * inode ilock it will return EAGAIN. Callers must handle this.
*/
-
-STATIC void
+STATIC int
xfs_setfilesize(
xfs_ioend_t *ioend)
{
@@ -178,19 +193,43 @@ xfs_setfilesize(
xfs_fsize_t isize;
ASSERT((ip->i_d.di_mode & S_IFMT) == S_IFREG);
- ASSERT(ioend->io_type != IOMAP_READ);
+ ASSERT(ioend->io_type != IO_READ);
if (unlikely(ioend->io_error))
- return;
+ return 0;
+
+ if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL))
+ return EAGAIN;
- xfs_ilock(ip, XFS_ILOCK_EXCL);
isize = xfs_ioend_new_eof(ioend);
if (isize) {
ip->i_d.di_size = isize;
- xfs_mark_inode_dirty_sync(ip);
+ xfs_mark_inode_dirty(ip);
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ return 0;
+}
+
+/*
+ * Schedule IO completion handling on a xfsdatad if this was
+ * the final hold on this ioend. If we are asked to wait,
+ * flush the workqueue.
+ */
+STATIC void
+xfs_finish_ioend(
+ xfs_ioend_t *ioend,
+ int wait)
+{
+ if (atomic_dec_and_test(&ioend->io_remaining)) {
+ struct workqueue_struct *wq;
+
+ wq = (ioend->io_type == IO_UNWRITTEN) ?
+ xfsconvertd_workqueue : xfsdatad_workqueue;
+ queue_work(wq, &ioend->io_work);
+ if (wait)
+ flush_workqueue(wq);
+ }
}
/*
@@ -198,19 +237,18 @@ xfs_setfilesize(
*/
STATIC void
xfs_end_io(
- struct work_struct *work)
+ struct work_struct *work)
{
- xfs_ioend_t *ioend =
- container_of(work, xfs_ioend_t, io_work);
- struct xfs_inode *ip = XFS_I(ioend->io_inode);
+ xfs_ioend_t *ioend = container_of(work, xfs_ioend_t, io_work);
+ struct xfs_inode *ip = XFS_I(ioend->io_inode);
+ int error = 0;
/*
* For unwritten extents we need to issue transactions to convert a
* range to normal written extens after the data I/O has finished.
*/
- if (ioend->io_type == IOMAP_UNWRITTEN &&
+ if (ioend->io_type == IO_UNWRITTEN &&
likely(!ioend->io_error && !XFS_FORCED_SHUTDOWN(ip->i_mount))) {
- int error;
error = xfs_iomap_write_unwritten(ip, ioend->io_offset,
ioend->io_size);
@@ -222,30 +260,23 @@ xfs_end_io(
* We might have to update the on-disk file size after extending
* writes.
*/
- if (ioend->io_type != IOMAP_READ)
- xfs_setfilesize(ioend);
- xfs_destroy_ioend(ioend);
-}
-
-/*
- * Schedule IO completion handling on a xfsdatad if this was
- * the final hold on this ioend. If we are asked to wait,
- * flush the workqueue.
- */
-STATIC void
-xfs_finish_ioend(
- xfs_ioend_t *ioend,
- int wait)
-{
- if (atomic_dec_and_test(&ioend->io_remaining)) {
- struct workqueue_struct *wq;
-
- wq = (ioend->io_type == IOMAP_UNWRITTEN) ?
- xfsconvertd_workqueue : xfsdatad_workqueue;
- queue_work(wq, &ioend->io_work);
- if (wait)
- flush_workqueue(wq);
+ if (ioend->io_type != IO_READ) {
+ error = xfs_setfilesize(ioend);
+ ASSERT(!error || error == EAGAIN);
}
+
+ /*
+ * If we didn't complete processing of the ioend, requeue it to the
+ * tail of the workqueue for another attempt later. Otherwise destroy
+ * it.
+ */
+ if (error == EAGAIN) {
+ atomic_inc(&ioend->io_remaining);
+ xfs_finish_ioend(ioend, 0);
+ /* ensure we don't spin on blocked ioends */
+ delay(1);
+ } else
+ xfs_destroy_ioend(ioend);
}
/*
@@ -288,21 +319,25 @@ xfs_map_blocks(
struct inode *inode,
loff_t offset,
ssize_t count,
- xfs_iomap_t *mapp,
+ struct xfs_bmbt_irec *imap,
int flags)
{
int nmaps = 1;
+ int new = 0;
- return -xfs_iomap(XFS_I(inode), offset, count, flags, mapp, &nmaps);
+ return -xfs_iomap(XFS_I(inode), offset, count, flags, imap, &nmaps, &new);
}
STATIC int
-xfs_iomap_valid(
- xfs_iomap_t *iomapp,
- loff_t offset)
+xfs_imap_valid(
+ struct inode *inode,
+ struct xfs_bmbt_irec *imap,
+ xfs_off_t offset)
{
- return offset >= iomapp->iomap_offset &&
- offset < iomapp->iomap_offset + iomapp->iomap_bsize;
+ offset >>= inode->i_blkbits;
+
+ return offset >= imap->br_startoff &&
+ offset < imap->br_startoff + imap->br_blockcount;
}
/*
@@ -341,7 +376,7 @@ xfs_submit_ioend_bio(
* but don't update the inode size until I/O completion.
*/
if (xfs_ioend_new_eof(ioend))
- xfs_mark_inode_dirty_sync(XFS_I(ioend->io_inode));
+ xfs_mark_inode_dirty(XFS_I(ioend->io_inode));
submit_bio(wbc->sync_mode == WB_SYNC_ALL ?
WRITE_SYNC_PLUG : WRITE, bio);
@@ -533,19 +568,23 @@ xfs_add_to_ioend(
STATIC void
xfs_map_buffer(
+ struct inode *inode,
struct buffer_head *bh,
- xfs_iomap_t *mp,
- xfs_off_t offset,
- uint block_bits)
+ struct xfs_bmbt_irec *imap,
+ xfs_off_t offset)
{
sector_t bn;
+ struct xfs_mount *m = XFS_I(inode)->i_mount;
+ xfs_off_t iomap_offset = XFS_FSB_TO_B(m, imap->br_startoff);
+ xfs_daddr_t iomap_bn = xfs_fsb_to_db(XFS_I(inode), imap->br_startblock);
- ASSERT(mp->iomap_bn != IOMAP_DADDR_NULL);
+ ASSERT(imap->br_startblock != HOLESTARTBLOCK);
+ ASSERT(imap->br_startblock != DELAYSTARTBLOCK);
- bn = (mp->iomap_bn >> (block_bits - BBSHIFT)) +
- ((offset - mp->iomap_offset) >> block_bits);
+ bn = (iomap_bn >> (inode->i_blkbits - BBSHIFT)) +
+ ((offset - iomap_offset) >> inode->i_blkbits);
- ASSERT(bn || (mp->iomap_flags & IOMAP_REALTIME));
+ ASSERT(bn || XFS_IS_REALTIME_INODE(XFS_I(inode)));
bh->b_blocknr = bn;
set_buffer_mapped(bh);
@@ -553,17 +592,17 @@ xfs_map_buffer(
STATIC void
xfs_map_at_offset(
+ struct inode *inode,
struct buffer_head *bh,
- loff_t offset,
- int block_bits,
- xfs_iomap_t *iomapp)
+ struct xfs_bmbt_irec *imap,
+ xfs_off_t offset)
{
- ASSERT(!(iomapp->iomap_flags & IOMAP_HOLE));
- ASSERT(!(iomapp->iomap_flags & IOMAP_DELAY));
+ ASSERT(imap->br_startblock != HOLESTARTBLOCK);
+ ASSERT(imap->br_startblock != DELAYSTARTBLOCK);
lock_buffer(bh);
- xfs_map_buffer(bh, iomapp, offset, block_bits);
- bh->b_bdev = iomapp->iomap_target->bt_bdev;
+ xfs_map_buffer(inode, bh, imap, offset);
+ bh->b_bdev = xfs_find_bdev_for_inode(inode);
set_buffer_mapped(bh);
clear_buffer_delay(bh);
clear_buffer_unwritten(bh);
@@ -692,11 +731,11 @@ xfs_is_delayed_page(
bh = head = page_buffers(page);
do {
if (buffer_unwritten(bh))
- acceptable = (type == IOMAP_UNWRITTEN);
+ acceptable = (type == IO_UNWRITTEN);
else if (buffer_delay(bh))
- acceptable = (type == IOMAP_DELAY);
+ acceptable = (type == IO_DELAY);
else if (buffer_dirty(bh) && buffer_mapped(bh))
- acceptable = (type == IOMAP_NEW);
+ acceptable = (type == IO_NEW);
else
break;
} while ((bh = bh->b_this_page) != head);
@@ -719,7 +758,7 @@ xfs_convert_page(
struct inode *inode,
struct page *page,
loff_t tindex,
- xfs_iomap_t *mp,
+ struct xfs_bmbt_irec *imap,
xfs_ioend_t **ioendp,
struct writeback_control *wbc,
int startio,
@@ -729,7 +768,6 @@ xfs_convert_page(
xfs_off_t end_offset;
unsigned long p_offset;
unsigned int type;
- int bbits = inode->i_blkbits;
int len, page_dirty;
int count = 0, done = 0, uptodate = 1;
xfs_off_t offset = page_offset(page);
@@ -781,19 +819,19 @@ xfs_convert_page(
if (buffer_unwritten(bh) || buffer_delay(bh)) {
if (buffer_unwritten(bh))
- type = IOMAP_UNWRITTEN;
+ type = IO_UNWRITTEN;
else
- type = IOMAP_DELAY;
+ type = IO_DELAY;
- if (!xfs_iomap_valid(mp, offset)) {
+ if (!xfs_imap_valid(inode, imap, offset)) {
done = 1;
continue;
}
- ASSERT(!(mp->iomap_flags & IOMAP_HOLE));
- ASSERT(!(mp->iomap_flags & IOMAP_DELAY));
+ ASSERT(imap->br_startblock != HOLESTARTBLOCK);
+ ASSERT(imap->br_startblock != DELAYSTARTBLOCK);
- xfs_map_at_offset(bh, offset, bbits, mp);
+ xfs_map_at_offset(inode, bh, imap, offset);
if (startio) {
xfs_add_to_ioend(inode, bh, offset,
type, ioendp, done);
@@ -805,7 +843,7 @@ xfs_convert_page(
page_dirty--;
count++;
} else {
- type = IOMAP_NEW;
+ type = IO_NEW;
if (buffer_mapped(bh) && all_bh && startio) {
lock_buffer(bh);
xfs_add_to_ioend(inode, bh, offset,
@@ -845,7 +883,7 @@ STATIC void
xfs_cluster_write(
struct inode *inode,
pgoff_t tindex,
- xfs_iomap_t *iomapp,
+ struct xfs_bmbt_irec *imap,
xfs_ioend_t **ioendp,
struct writeback_control *wbc,
int startio,
@@ -864,7 +902,7 @@ xfs_cluster_write(
for (i = 0; i < pagevec_count(&pvec); i++) {
done = xfs_convert_page(inode, pvec.pages[i], tindex++,
- iomapp, ioendp, wbc, startio, all_bh);
+ imap, ioendp, wbc, startio, all_bh);
if (done)
break;
}
@@ -874,6 +912,125 @@ xfs_cluster_write(
}
}
+STATIC void
+xfs_vm_invalidatepage(
+ struct page *page,
+ unsigned long offset)
+{
+ trace_xfs_invalidatepage(page->mapping->host, page, offset);
+ block_invalidatepage(page, offset);
+}
+
+/*
+ * If the page has delalloc buffers on it, we need to punch them out before we
+ * invalidate the page. If we don't, we leave a stale delalloc mapping on the
+ * inode that can trip a BUG() in xfs_get_blocks() later on if a direct IO read
+ * is done on that same region - the delalloc extent is returned when none is
+ * supposed to be there.
+ *
+ * We prevent this by truncating away the delalloc regions on the page before
+ * invalidating it. Because they are delalloc, we can do this without needing a
+ * transaction. Indeed - if we get ENOSPC errors, we have to be able to do this
+ * truncation without a transaction as there is no space left for block
+ * reservation (typically why we see a ENOSPC in writeback).
+ *
+ * This is not a performance critical path, so for now just do the punching a
+ * buffer head at a time.
+ */
+STATIC void
+xfs_aops_discard_page(
+ struct page *page)
+{
+ struct inode *inode = page->mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ struct buffer_head *bh, *head;
+ loff_t offset = page_offset(page);
+ ssize_t len = 1 << inode->i_blkbits;
+
+ if (!xfs_is_delayed_page(page, IO_DELAY))
+ goto out_invalidate;
+
+ if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+ goto out_invalidate;
+
+ xfs_fs_cmn_err(CE_ALERT, ip->i_mount,
+ "page discard on page %p, inode 0x%llx, offset %llu.",
+ page, ip->i_ino, offset);
+
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ bh = head = page_buffers(page);
+ do {
+ int done;
+ xfs_fileoff_t offset_fsb;
+ xfs_bmbt_irec_t imap;
+ int nimaps = 1;
+ int error;
+ xfs_fsblock_t firstblock;
+ xfs_bmap_free_t flist;
+
+ if (!buffer_delay(bh))
+ goto next_buffer;
+
+ offset_fsb = XFS_B_TO_FSBT(ip->i_mount, offset);
+
+ /*
+ * Map the range first and check that it is a delalloc extent
+ * before trying to unmap the range. Otherwise we will be
+ * trying to remove a real extent (which requires a
+ * transaction) or a hole, which is probably a bad idea...
+ */
+ error = xfs_bmapi(NULL, ip, offset_fsb, 1,
+ XFS_BMAPI_ENTIRE, NULL, 0, &imap,
+ &nimaps, NULL, NULL);
+
+ if (error) {
+ /* something screwed, just bail */
+ if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) {
+ xfs_fs_cmn_err(CE_ALERT, ip->i_mount,
+ "page discard failed delalloc mapping lookup.");
+ }
+ break;
+ }
+ if (!nimaps) {
+ /* nothing there */
+ goto next_buffer;
+ }
+ if (imap.br_startblock != DELAYSTARTBLOCK) {
+ /* been converted, ignore */
+ goto next_buffer;
+ }
+ WARN_ON(imap.br_blockcount == 0);
+
+ /*
+ * Note: while we initialise the firstblock/flist pair, they
+ * should never be used because blocks should never be
+ * allocated or freed for a delalloc extent and hence we need
+ * don't cancel or finish them after the xfs_bunmapi() call.
+ */
+ xfs_bmap_init(&flist, &firstblock);
+ error = xfs_bunmapi(NULL, ip, offset_fsb, 1, 0, 1, &firstblock,
+ &flist, NULL, &done);
+
+ ASSERT(!flist.xbf_count && !flist.xbf_first);
+ if (error) {
+ /* something screwed, just bail */
+ if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) {
+ xfs_fs_cmn_err(CE_ALERT, ip->i_mount,
+ "page discard unable to remove delalloc mapping.");
+ }
+ break;
+ }
+next_buffer:
+ offset += len;
+
+ } while ((bh = bh->b_this_page) != head);
+
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+out_invalidate:
+ xfs_vm_invalidatepage(page, 0);
+ return;
+}
+
/*
* Calling this without startio set means we are being asked to make a dirty
* page ready for freeing it's buffers. When called with startio set then
@@ -902,15 +1059,15 @@ xfs_page_state_convert(
int unmapped) /* also implies page uptodate */
{
struct buffer_head *bh, *head;
- xfs_iomap_t iomap;
+ struct xfs_bmbt_irec imap;
xfs_ioend_t *ioend = NULL, *iohead = NULL;
loff_t offset;
unsigned long p_offset = 0;
unsigned int type;
__uint64_t end_offset;
- pgoff_t end_index, last_index, tlast;
+ pgoff_t end_index, last_index;
ssize_t size, len;
- int flags, err, iomap_valid = 0, uptodate = 1;
+ int flags, err, imap_valid = 0, uptodate = 1;
int page_dirty, count = 0;
int trylock = 0;
int all_bh = unmapped;
@@ -957,7 +1114,7 @@ xfs_page_state_convert(
bh = head = page_buffers(page);
offset = page_offset(page);
flags = BMAPI_READ;
- type = IOMAP_NEW;
+ type = IO_NEW;
/* TODO: cleanup count and page_dirty */
@@ -971,12 +1128,12 @@ xfs_page_state_convert(
* the iomap is actually still valid, but the ioend
* isn't. shouldn't happen too often.
*/
- iomap_valid = 0;
+ imap_valid = 0;
continue;
}
- if (iomap_valid)
- iomap_valid = xfs_iomap_valid(&iomap, offset);
+ if (imap_valid)
+ imap_valid = xfs_imap_valid(inode, &imap, offset);
/*
* First case, map an unwritten extent and prepare for
@@ -997,20 +1154,20 @@ xfs_page_state_convert(
* Make sure we don't use a read-only iomap
*/
if (flags == BMAPI_READ)
- iomap_valid = 0;
+ imap_valid = 0;
if (buffer_unwritten(bh)) {
- type = IOMAP_UNWRITTEN;
+ type = IO_UNWRITTEN;
flags = BMAPI_WRITE | BMAPI_IGNSTATE;
} else if (buffer_delay(bh)) {
- type = IOMAP_DELAY;
+ type = IO_DELAY;
flags = BMAPI_ALLOCATE | trylock;
} else {
- type = IOMAP_NEW;
+ type = IO_NEW;
flags = BMAPI_WRITE | BMAPI_MMAP;
}
- if (!iomap_valid) {
+ if (!imap_valid) {
/*
* if we didn't have a valid mapping then we
* need to ensure that we put the new mapping
@@ -1020,7 +1177,7 @@ xfs_page_state_convert(
* for unwritten extent conversion.
*/
new_ioend = 1;
- if (type == IOMAP_NEW) {
+ if (type == IO_NEW) {
size = xfs_probe_cluster(inode,
page, bh, head, 0);
} else {
@@ -1028,14 +1185,14 @@ xfs_page_state_convert(
}
err = xfs_map_blocks(inode, offset, size,
- &iomap, flags);
+ &imap, flags);
if (err)
goto error;
- iomap_valid = xfs_iomap_valid(&iomap, offset);
+ imap_valid = xfs_imap_valid(inode, &imap,
+ offset);
}
- if (iomap_valid) {
- xfs_map_at_offset(bh, offset,
- inode->i_blkbits, &iomap);
+ if (imap_valid) {
+ xfs_map_at_offset(inode, bh, &imap, offset);
if (startio) {
xfs_add_to_ioend(inode, bh, offset,
type, &ioend,
@@ -1054,40 +1211,41 @@ xfs_page_state_convert(
* That means it must already have extents allocated
* underneath it. Map the extent by reading it.
*/
- if (!iomap_valid || flags != BMAPI_READ) {
+ if (!imap_valid || flags != BMAPI_READ) {
flags = BMAPI_READ;
size = xfs_probe_cluster(inode, page, bh,
head, 1);
err = xfs_map_blocks(inode, offset, size,
- &iomap, flags);
+ &imap, flags);
if (err)
goto error;
- iomap_valid = xfs_iomap_valid(&iomap, offset);
+ imap_valid = xfs_imap_valid(inode, &imap,
+ offset);
}
/*
- * We set the type to IOMAP_NEW in case we are doing a
+ * We set the type to IO_NEW in case we are doing a
* small write at EOF that is extending the file but
* without needing an allocation. We need to update the
* file size on I/O completion in this case so it is
* the same case as having just allocated a new extent
* that we are writing into for the first time.
*/
- type = IOMAP_NEW;
+ type = IO_NEW;
if (trylock_buffer(bh)) {
ASSERT(buffer_mapped(bh));
- if (iomap_valid)
+ if (imap_valid)
all_bh = 1;
xfs_add_to_ioend(inode, bh, offset, type,
- &ioend, !iomap_valid);
+ &ioend, !imap_valid);
page_dirty--;
count++;
} else {
- iomap_valid = 0;
+ imap_valid = 0;
}
} else if ((buffer_uptodate(bh) || PageUptodate(page)) &&
(unmapped || startio)) {
- iomap_valid = 0;
+ imap_valid = 0;
}
if (!iohead)
@@ -1101,12 +1259,23 @@ xfs_page_state_convert(
if (startio)
xfs_start_page_writeback(page, 1, count);
- if (ioend && iomap_valid) {
- offset = (iomap.iomap_offset + iomap.iomap_bsize - 1) >>
- PAGE_CACHE_SHIFT;
- tlast = min_t(pgoff_t, offset, last_index);
- xfs_cluster_write(inode, page->index + 1, &iomap, &ioend,
- wbc, startio, all_bh, tlast);
+ if (ioend && imap_valid) {
+ xfs_off_t end_index;
+
+ end_index = imap.br_startoff + imap.br_blockcount;
+
+ /* to bytes */
+ end_index <<= inode->i_blkbits;
+
+ /* to pages */
+ end_index = (end_index - 1) >> PAGE_CACHE_SHIFT;
+
+ /* check against file size */
+ if (end_index > last_index)
+ end_index = last_index;
+
+ xfs_cluster_write(inode, page->index + 1, &imap, &ioend,
+ wbc, startio, all_bh, end_index);
}
if (iohead)
@@ -1125,7 +1294,7 @@ error:
*/
if (err != -EAGAIN) {
if (!unmapped)
- block_invalidatepage(page, 0);
+ xfs_aops_discard_page(page);
ClearPageUptodate(page);
}
return err;
@@ -1164,6 +1333,21 @@ xfs_vm_writepage(
trace_xfs_writepage(inode, page, 0);
/*
+ * Refuse to write the page out if we are called from reclaim context.
+ *
+ * This is primarily to avoid stack overflows when called from deep
+ * used stacks in random callers for direct reclaim, but disabling
+ * reclaim for kswap is a nice side-effect as kswapd causes rather
+ * suboptimal I/O patters, too.
+ *
+ * This should really be done by the core VM, but until that happens
+ * filesystems like XFS, btrfs and ext4 have to take care of this
+ * by themselves.
+ */
+ if (current->flags & PF_MEMALLOC)
+ goto out_fail;
+
+ /*
* We need a transaction if:
* 1. There are delalloc buffers on the page
* 2. The page is uptodate and we have unmapped buffers
@@ -1197,14 +1381,6 @@ xfs_vm_writepage(
if (!page_has_buffers(page))
create_empty_buffers(page, 1 << inode->i_blkbits, 0);
-
- /*
- * VM calculation for nr_to_write seems off. Bump it way
- * up, this gets simple streaming writes zippy again.
- * To be reviewed again after Jens' writeback changes.
- */
- wbc->nr_to_write *= 4;
-
/*
* Convert delayed allocate, unwritten or unmapped space
* to real space and flush out to disk.
@@ -1308,10 +1484,11 @@ __xfs_get_blocks(
int direct,
bmapi_flags_t flags)
{
- xfs_iomap_t iomap;
+ struct xfs_bmbt_irec imap;
xfs_off_t offset;
ssize_t size;
- int niomap = 1;
+ int nimap = 1;
+ int new = 0;
int error;
offset = (xfs_off_t)iblock << inode->i_blkbits;
@@ -1322,22 +1499,21 @@ __xfs_get_blocks(
return 0;
error = xfs_iomap(XFS_I(inode), offset, size,
- create ? flags : BMAPI_READ, &iomap, &niomap);
+ create ? flags : BMAPI_READ, &imap, &nimap, &new);
if (error)
return -error;
- if (niomap == 0)
+ if (nimap == 0)
return 0;
- if (iomap.iomap_bn != IOMAP_DADDR_NULL) {
+ if (imap.br_startblock != HOLESTARTBLOCK &&
+ imap.br_startblock != DELAYSTARTBLOCK) {
/*
* For unwritten extents do not report a disk address on
* the read case (treat as if we're reading into a hole).
*/
- if (create || !(iomap.iomap_flags & IOMAP_UNWRITTEN)) {
- xfs_map_buffer(bh_result, &iomap, offset,
- inode->i_blkbits);
- }
- if (create && (iomap.iomap_flags & IOMAP_UNWRITTEN)) {
+ if (create || !ISUNWRITTEN(&imap))
+ xfs_map_buffer(inode, bh_result, &imap, offset);
+ if (create && ISUNWRITTEN(&imap)) {
if (direct)
bh_result->b_private = inode;
set_buffer_unwritten(bh_result);
@@ -1348,7 +1524,7 @@ __xfs_get_blocks(
* If this is a realtime file, data may be on a different device.
* to that pointed to from the buffer_head b_bdev currently.
*/
- bh_result->b_bdev = iomap.iomap_target->bt_bdev;
+ bh_result->b_bdev = xfs_find_bdev_for_inode(inode);
/*
* If we previously allocated a block out beyond eof and we are now
@@ -1362,10 +1538,10 @@ __xfs_get_blocks(
if (create &&
((!buffer_mapped(bh_result) && !buffer_uptodate(bh_result)) ||
(offset >= i_size_read(inode)) ||
- (iomap.iomap_flags & (IOMAP_NEW|IOMAP_UNWRITTEN))))
+ (new || ISUNWRITTEN(&imap))))
set_buffer_new(bh_result);
- if (iomap.iomap_flags & IOMAP_DELAY) {
+ if (imap.br_startblock == DELAYSTARTBLOCK) {
BUG_ON(direct);
if (create) {
set_buffer_uptodate(bh_result);
@@ -1374,11 +1550,23 @@ __xfs_get_blocks(
}
}
+ /*
+ * If this is O_DIRECT or the mpage code calling tell them how large
+ * the mapping is, so that we can avoid repeated get_blocks calls.
+ */
if (direct || size > (1 << inode->i_blkbits)) {
- ASSERT(iomap.iomap_bsize - iomap.iomap_delta > 0);
- offset = min_t(xfs_off_t,
- iomap.iomap_bsize - iomap.iomap_delta, size);
- bh_result->b_size = (ssize_t)min_t(xfs_off_t, LONG_MAX, offset);
+ xfs_off_t mapping_size;
+
+ mapping_size = imap.br_startoff + imap.br_blockcount - iblock;
+ mapping_size <<= inode->i_blkbits;
+
+ ASSERT(mapping_size > 0);
+ if (mapping_size > size)
+ mapping_size = size;
+ if (mapping_size > LONG_MAX)
+ mapping_size = LONG_MAX;
+
+ bh_result->b_size = mapping_size;
}
return 0;
@@ -1436,7 +1624,7 @@ xfs_end_io_direct(
*/
ioend->io_offset = offset;
ioend->io_size = size;
- if (ioend->io_type == IOMAP_READ) {
+ if (ioend->io_type == IO_READ) {
xfs_finish_ioend(ioend, 0);
} else if (private && size > 0) {
xfs_finish_ioend(ioend, is_sync_kiocb(iocb));
@@ -1447,7 +1635,7 @@ xfs_end_io_direct(
* didn't map an unwritten extent so switch it's completion
* handler.
*/
- ioend->io_type = IOMAP_NEW;
+ ioend->io_type = IO_NEW;
xfs_finish_ioend(ioend, 0);
}
@@ -1472,10 +1660,10 @@ xfs_vm_direct_IO(
struct block_device *bdev;
ssize_t ret;
- bdev = xfs_find_bdev_for_inode(XFS_I(inode));
+ bdev = xfs_find_bdev_for_inode(inode);
iocb->private = xfs_alloc_ioend(inode, rw == WRITE ?
- IOMAP_UNWRITTEN : IOMAP_READ);
+ IO_UNWRITTEN : IO_READ);
ret = blockdev_direct_IO_no_locking(rw, iocb, inode, bdev, iov,
offset, nr_segs,
@@ -1535,15 +1723,6 @@ xfs_vm_readpages(
return mpage_readpages(mapping, pages, nr_pages, xfs_get_blocks);
}
-STATIC void
-xfs_vm_invalidatepage(
- struct page *page,
- unsigned long offset)
-{
- trace_xfs_invalidatepage(page->mapping->host, page, offset);
- block_invalidatepage(page, offset);
-}
-
const struct address_space_operations xfs_address_space_operations = {
.readpage = xfs_vm_readpage,
.readpages = xfs_vm_readpages,
diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index 77b8be81c769..2ee3f7a60163 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -18,7 +18,7 @@
#include "xfs.h"
#include <linux/stddef.h>
#include <linux/errno.h>
-#include <linux/slab.h>
+#include <linux/gfp.h>
#include <linux/pagemap.h>
#include <linux/init.h>
#include <linux/vmalloc.h>
@@ -33,9 +33,11 @@
#include <linux/migrate.h>
#include <linux/backing-dev.h>
#include <linux/freezer.h>
+#include <linux/list_sort.h>
#include "xfs_sb.h"
#include "xfs_inum.h"
+#include "xfs_log.h"
#include "xfs_ag.h"
#include "xfs_dmapi.h"
#include "xfs_mount.h"
@@ -43,7 +45,7 @@
static kmem_zone_t *xfs_buf_zone;
STATIC int xfsbufd(void *);
-STATIC int xfsbufd_wakeup(int, gfp_t);
+STATIC int xfsbufd_wakeup(struct shrinker *, int, gfp_t);
STATIC void xfs_buf_delwri_queue(xfs_buf_t *, int);
static struct shrinker xfs_buf_shake = {
.shrink = xfsbufd_wakeup,
@@ -76,6 +78,27 @@ struct workqueue_struct *xfsconvertd_workqueue;
#define xfs_buf_deallocate(bp) \
kmem_zone_free(xfs_buf_zone, (bp));
+static inline int
+xfs_buf_is_vmapped(
+ struct xfs_buf *bp)
+{
+ /*
+ * Return true if the buffer is vmapped.
+ *
+ * The XBF_MAPPED flag is set if the buffer should be mapped, but the
+ * code is clever enough to know it doesn't have to map a single page,
+ * so the check has to be both for XBF_MAPPED and bp->b_page_count > 1.
+ */
+ return (bp->b_flags & XBF_MAPPED) && bp->b_page_count > 1;
+}
+
+static inline int
+xfs_buf_vmap_len(
+ struct xfs_buf *bp)
+{
+ return (bp->b_page_count * PAGE_SIZE) - bp->b_offset;
+}
+
/*
* Page Region interfaces.
*
@@ -146,75 +169,6 @@ test_page_region(
}
/*
- * Mapping of multi-page buffers into contiguous virtual space
- */
-
-typedef struct a_list {
- void *vm_addr;
- struct a_list *next;
-} a_list_t;
-
-static a_list_t *as_free_head;
-static int as_list_len;
-static DEFINE_SPINLOCK(as_lock);
-
-/*
- * Try to batch vunmaps because they are costly.
- */
-STATIC void
-free_address(
- void *addr)
-{
- a_list_t *aentry;
-
-#ifdef CONFIG_XEN
- /*
- * Xen needs to be able to make sure it can get an exclusive
- * RO mapping of pages it wants to turn into a pagetable. If
- * a newly allocated page is also still being vmap()ed by xfs,
- * it will cause pagetable construction to fail. This is a
- * quick workaround to always eagerly unmap pages so that Xen
- * is happy.
- */
- vunmap(addr);
- return;
-#endif
-
- aentry = kmalloc(sizeof(a_list_t), GFP_NOWAIT);
- if (likely(aentry)) {
- spin_lock(&as_lock);
- aentry->next = as_free_head;
- aentry->vm_addr = addr;
- as_free_head = aentry;
- as_list_len++;
- spin_unlock(&as_lock);
- } else {
- vunmap(addr);
- }
-}
-
-STATIC void
-purge_addresses(void)
-{
- a_list_t *aentry, *old;
-
- if (as_free_head == NULL)
- return;
-
- spin_lock(&as_lock);
- aentry = as_free_head;
- as_free_head = NULL;
- as_list_len = 0;
- spin_unlock(&as_lock);
-
- while ((old = aentry) != NULL) {
- vunmap(aentry->vm_addr);
- aentry = aentry->next;
- kfree(old);
- }
-}
-
-/*
* Internal xfs_buf_t object manipulation
*/
@@ -314,8 +268,9 @@ xfs_buf_free(
if (bp->b_flags & (_XBF_PAGE_CACHE|_XBF_PAGES)) {
uint i;
- if ((bp->b_flags & XBF_MAPPED) && (bp->b_page_count > 1))
- free_address(bp->b_addr - bp->b_offset);
+ if (xfs_buf_is_vmapped(bp))
+ vm_unmap_ram(bp->b_addr - bp->b_offset,
+ bp->b_page_count);
for (i = 0; i < bp->b_page_count; i++) {
struct page *page = bp->b_pages[i];
@@ -385,7 +340,7 @@ _xfs_buf_lookup_pages(
__func__, gfp_mask);
XFS_STATS_INC(xb_page_retries);
- xfsbufd_wakeup(0, gfp_mask);
+ xfsbufd_wakeup(NULL, 0, gfp_mask);
congestion_wait(BLK_RW_ASYNC, HZ/50);
goto retry;
}
@@ -435,10 +390,8 @@ _xfs_buf_map_pages(
bp->b_addr = page_address(bp->b_pages[0]) + bp->b_offset;
bp->b_flags |= XBF_MAPPED;
} else if (flags & XBF_MAPPED) {
- if (as_list_len > 64)
- purge_addresses();
- bp->b_addr = vmap(bp->b_pages, bp->b_page_count,
- VM_MAP, PAGE_KERNEL);
+ bp->b_addr = vm_map_ram(bp->b_pages, bp->b_page_count,
+ -1, PAGE_KERNEL);
if (unlikely(bp->b_addr == NULL))
return -ENOMEM;
bp->b_addr += bp->b_offset;
@@ -898,6 +851,12 @@ xfs_buf_lock_value(
* Note that this in no way locks the underlying pages, so it is only
* useful for synchronizing concurrent use of buffer objects, not for
* synchronizing independent access to the underlying pages.
+ *
+ * If we come across a stale, pinned, locked buffer, we know that we
+ * are being asked to lock a buffer that has been reallocated. Because
+ * it is pinned, we know that the log has not been pushed to disk and
+ * hence it will still be locked. Rather than sleeping until someone
+ * else pushes the log, push it ourselves before trying to get the lock.
*/
void
xfs_buf_lock(
@@ -905,6 +864,8 @@ xfs_buf_lock(
{
trace_xfs_buf_lock(bp, _RET_IP_);
+ if (atomic_read(&bp->b_pin_count) && (bp->b_flags & XBF_STALE))
+ xfs_log_force(bp->b_mount, 0);
if (atomic_read(&bp->b_io_remaining))
blk_run_address_space(bp->b_target->bt_mapping);
down(&bp->b_sema);
@@ -1051,22 +1012,25 @@ xfs_buf_ioerror(
}
int
-xfs_bawrite(
- void *mp,
+xfs_bwrite(
+ struct xfs_mount *mp,
struct xfs_buf *bp)
{
- trace_xfs_buf_bawrite(bp, _RET_IP_);
+ int error;
- ASSERT(bp->b_bn != XFS_BUF_DADDR_NULL);
+ bp->b_strat = xfs_bdstrat_cb;
+ bp->b_mount = mp;
+ bp->b_flags |= XBF_WRITE;
+ bp->b_flags &= ~(XBF_ASYNC | XBF_READ);
xfs_buf_delwri_dequeue(bp);
+ xfs_buf_iostrategy(bp);
- 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);
+ error = xfs_buf_iowait(bp);
+ if (error)
+ xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
+ xfs_buf_relse(bp);
+ return error;
}
void
@@ -1085,6 +1049,126 @@ xfs_bdwrite(
xfs_buf_delwri_queue(bp, 1);
}
+/*
+ * Called when we want to stop a buffer from getting written or read.
+ * We attach the EIO error, muck with its flags, and call biodone
+ * so that the proper iodone callbacks get called.
+ */
+STATIC int
+xfs_bioerror(
+ xfs_buf_t *bp)
+{
+#ifdef XFSERRORDEBUG
+ ASSERT(XFS_BUF_ISREAD(bp) || bp->b_iodone);
+#endif
+
+ /*
+ * No need to wait until the buffer is unpinned, we aren't flushing it.
+ */
+ XFS_BUF_ERROR(bp, EIO);
+
+ /*
+ * We're calling biodone, so delete XBF_DONE flag.
+ */
+ XFS_BUF_UNREAD(bp);
+ XFS_BUF_UNDELAYWRITE(bp);
+ XFS_BUF_UNDONE(bp);
+ XFS_BUF_STALE(bp);
+
+ XFS_BUF_CLR_BDSTRAT_FUNC(bp);
+ xfs_biodone(bp);
+
+ return EIO;
+}
+
+/*
+ * Same as xfs_bioerror, except that we are releasing the buffer
+ * here ourselves, and avoiding the biodone call.
+ * This is meant for userdata errors; metadata bufs come with
+ * iodone functions attached, so that we can track down errors.
+ */
+STATIC int
+xfs_bioerror_relse(
+ struct xfs_buf *bp)
+{
+ int64_t fl = XFS_BUF_BFLAGS(bp);
+ /*
+ * No need to wait until the buffer is unpinned.
+ * We aren't flushing it.
+ *
+ * chunkhold expects B_DONE to be set, whether
+ * we actually finish the I/O or not. We don't want to
+ * change that interface.
+ */
+ XFS_BUF_UNREAD(bp);
+ XFS_BUF_UNDELAYWRITE(bp);
+ XFS_BUF_DONE(bp);
+ XFS_BUF_STALE(bp);
+ XFS_BUF_CLR_IODONE_FUNC(bp);
+ XFS_BUF_CLR_BDSTRAT_FUNC(bp);
+ if (!(fl & XBF_ASYNC)) {
+ /*
+ * Mark b_error and B_ERROR _both_.
+ * Lot's of chunkcache code assumes that.
+ * There's no reason to mark error for
+ * ASYNC buffers.
+ */
+ XFS_BUF_ERROR(bp, EIO);
+ XFS_BUF_FINISH_IOWAIT(bp);
+ } else {
+ xfs_buf_relse(bp);
+ }
+
+ return EIO;
+}
+
+
+/*
+ * All xfs metadata buffers except log state machine buffers
+ * get this attached as their b_bdstrat callback function.
+ * This is so that we can catch a buffer
+ * after prematurely unpinning it to forcibly shutdown the filesystem.
+ */
+int
+xfs_bdstrat_cb(
+ struct xfs_buf *bp)
+{
+ if (XFS_FORCED_SHUTDOWN(bp->b_mount)) {
+ trace_xfs_bdstrat_shut(bp, _RET_IP_);
+ /*
+ * Metadata write that didn't get logged but
+ * written delayed anyway. These aren't associated
+ * with a transaction, and can be ignored.
+ */
+ if (!bp->b_iodone && !XFS_BUF_ISREAD(bp))
+ return xfs_bioerror_relse(bp);
+ else
+ return xfs_bioerror(bp);
+ }
+
+ xfs_buf_iorequest(bp);
+ return 0;
+}
+
+/*
+ * Wrapper around bdstrat so that we can stop data from going to disk in case
+ * we are shutting down the filesystem. Typically user data goes thru this
+ * path; one of the exceptions is the superblock.
+ */
+void
+xfsbdstrat(
+ struct xfs_mount *mp,
+ struct xfs_buf *bp)
+{
+ if (XFS_FORCED_SHUTDOWN(mp)) {
+ trace_xfs_bdstrat_shut(bp, _RET_IP_);
+ xfs_bioerror_relse(bp);
+ return;
+ }
+
+ xfs_buf_iorequest(bp);
+}
+
STATIC void
_xfs_buf_ioend(
xfs_buf_t *bp,
@@ -1107,6 +1191,9 @@ xfs_buf_bio_end_io(
xfs_buf_ioerror(bp, -error);
+ if (!error && xfs_buf_is_vmapped(bp) && (bp->b_flags & XBF_READ))
+ invalidate_kernel_vmap_range(bp->b_addr, xfs_buf_vmap_len(bp));
+
do {
struct page *page = bvec->bv_page;
@@ -1216,6 +1303,10 @@ next_chunk:
submit_io:
if (likely(bio->bi_size)) {
+ if (xfs_buf_is_vmapped(bp)) {
+ flush_kernel_vmap_range(bp->b_addr,
+ xfs_buf_vmap_len(bp));
+ }
submit_bio(rw, bio);
if (size)
goto next_chunk;
@@ -1296,7 +1387,7 @@ xfs_buf_iomove(
xfs_buf_t *bp, /* buffer to process */
size_t boff, /* starting buffer offset */
size_t bsize, /* length to copy */
- caddr_t data, /* data address */
+ void *data, /* data address */
xfs_buf_rw_t mode) /* read/write/zero flag */
{
size_t bend, cpoff, csize;
@@ -1378,8 +1469,8 @@ xfs_alloc_bufhash(
btp->bt_hashshift = external ? 3 : 8; /* 8 or 256 buckets */
btp->bt_hashmask = (1 << btp->bt_hashshift) - 1;
- btp->bt_hash = kmem_zalloc((1 << btp->bt_hashshift) *
- sizeof(xfs_bufhash_t), KM_SLEEP | KM_LARGE);
+ btp->bt_hash = kmem_zalloc_large((1 << btp->bt_hashshift) *
+ sizeof(xfs_bufhash_t));
for (i = 0; i < (1 << btp->bt_hashshift); i++) {
spin_lock_init(&btp->bt_hash[i].bh_lock);
INIT_LIST_HEAD(&btp->bt_hash[i].bh_list);
@@ -1390,7 +1481,7 @@ STATIC void
xfs_free_bufhash(
xfs_buftarg_t *btp)
{
- kmem_free(btp->bt_hash);
+ kmem_free_large(btp->bt_hash);
btp->bt_hash = NULL;
}
@@ -1527,7 +1618,8 @@ xfs_mapping_buftarg(
STATIC int
xfs_alloc_delwrite_queue(
- xfs_buftarg_t *btp)
+ xfs_buftarg_t *btp,
+ const char *fsname)
{
int error = 0;
@@ -1535,7 +1627,7 @@ xfs_alloc_delwrite_queue(
INIT_LIST_HEAD(&btp->bt_delwrite_queue);
spin_lock_init(&btp->bt_delwrite_lock);
btp->bt_flags = 0;
- btp->bt_task = kthread_run(xfsbufd, btp, "xfsbufd");
+ btp->bt_task = kthread_run(xfsbufd, btp, "xfsbufd/%s", fsname);
if (IS_ERR(btp->bt_task)) {
error = PTR_ERR(btp->bt_task);
goto out_error;
@@ -1548,7 +1640,8 @@ out_error:
xfs_buftarg_t *
xfs_alloc_buftarg(
struct block_device *bdev,
- int external)
+ int external,
+ const char *fsname)
{
xfs_buftarg_t *btp;
@@ -1560,7 +1653,7 @@ xfs_alloc_buftarg(
goto error;
if (xfs_mapping_buftarg(btp, bdev))
goto error;
- if (xfs_alloc_delwrite_queue(btp))
+ if (xfs_alloc_delwrite_queue(btp, fsname))
goto error;
xfs_alloc_bufhash(btp, external);
return btp;
@@ -1595,6 +1688,11 @@ xfs_buf_delwri_queue(
list_del(&bp->b_list);
}
+ if (list_empty(dwq)) {
+ /* start xfsbufd as it is about to have something to do */
+ wake_up_process(bp->b_target->bt_task);
+ }
+
bp->b_flags |= _XBF_DELWRI_Q;
list_add_tail(&bp->b_list, dwq);
bp->b_queuetime = jiffies;
@@ -1626,6 +1724,35 @@ xfs_buf_delwri_dequeue(
trace_xfs_buf_delwri_dequeue(bp, _RET_IP_);
}
+/*
+ * If a delwri buffer needs to be pushed before it has aged out, then promote
+ * it to the head of the delwri queue so that it will be flushed on the next
+ * xfsbufd run. We do this by resetting the queuetime of the buffer to be older
+ * than the age currently needed to flush the buffer. Hence the next time the
+ * xfsbufd sees it is guaranteed to be considered old enough to flush.
+ */
+void
+xfs_buf_delwri_promote(
+ struct xfs_buf *bp)
+{
+ struct xfs_buftarg *btp = bp->b_target;
+ long age = xfs_buf_age_centisecs * msecs_to_jiffies(10) + 1;
+
+ ASSERT(bp->b_flags & XBF_DELWRI);
+ ASSERT(bp->b_flags & _XBF_DELWRI_Q);
+
+ /*
+ * Check the buffer age before locking the delayed write queue as we
+ * don't need to promote buffers that are already past the flush age.
+ */
+ if (bp->b_queuetime < jiffies - age)
+ return;
+ bp->b_queuetime = jiffies - age;
+ spin_lock(&btp->bt_delwrite_lock);
+ list_move(&bp->b_list, &btp->bt_delwrite_queue);
+ spin_unlock(&btp->bt_delwrite_lock);
+}
+
STATIC void
xfs_buf_runall_queues(
struct workqueue_struct *queue)
@@ -1635,6 +1762,7 @@ xfs_buf_runall_queues(
STATIC int
xfsbufd_wakeup(
+ struct shrinker *shrink,
int priority,
gfp_t mask)
{
@@ -1644,6 +1772,8 @@ xfsbufd_wakeup(
list_for_each_entry(btp, &xfs_buftarg_list, bt_list) {
if (test_bit(XBT_FORCE_SLEEP, &btp->bt_flags))
continue;
+ if (list_empty(&btp->bt_delwrite_queue))
+ continue;
set_bit(XBT_FORCE_FLUSH, &btp->bt_flags);
wake_up_process(btp->bt_task);
}
@@ -1694,20 +1824,53 @@ xfs_buf_delwri_split(
}
+/*
+ * Compare function is more complex than it needs to be because
+ * the return value is only 32 bits and we are doing comparisons
+ * on 64 bit values
+ */
+static int
+xfs_buf_cmp(
+ void *priv,
+ struct list_head *a,
+ struct list_head *b)
+{
+ struct xfs_buf *ap = container_of(a, struct xfs_buf, b_list);
+ struct xfs_buf *bp = container_of(b, struct xfs_buf, b_list);
+ xfs_daddr_t diff;
+
+ diff = ap->b_bn - bp->b_bn;
+ if (diff < 0)
+ return -1;
+ if (diff > 0)
+ return 1;
+ return 0;
+}
+
+void
+xfs_buf_delwri_sort(
+ xfs_buftarg_t *target,
+ struct list_head *list)
+{
+ list_sort(NULL, list, xfs_buf_cmp);
+}
+
STATIC int
xfsbufd(
void *data)
{
- struct list_head tmp;
- xfs_buftarg_t *target = (xfs_buftarg_t *)data;
- int count;
- xfs_buf_t *bp;
+ xfs_buftarg_t *target = (xfs_buftarg_t *)data;
current->flags |= PF_MEMALLOC;
set_freezable();
do {
+ long age = xfs_buf_age_centisecs * msecs_to_jiffies(10);
+ long tout = xfs_buf_timer_centisecs * msecs_to_jiffies(10);
+ int count = 0;
+ struct list_head tmp;
+
if (unlikely(freezing(current))) {
set_bit(XBT_FORCE_SLEEP, &target->bt_flags);
refrigerator();
@@ -1715,24 +1878,20 @@ xfsbufd(
clear_bit(XBT_FORCE_SLEEP, &target->bt_flags);
}
- schedule_timeout_interruptible(
- xfs_buf_timer_centisecs * msecs_to_jiffies(10));
+ /* sleep for a long time if there is nothing to do. */
+ if (list_empty(&target->bt_delwrite_queue))
+ tout = MAX_SCHEDULE_TIMEOUT;
+ schedule_timeout_interruptible(tout);
- xfs_buf_delwri_split(target, &tmp,
- xfs_buf_age_centisecs * msecs_to_jiffies(10));
-
- count = 0;
+ xfs_buf_delwri_split(target, &tmp, age);
+ list_sort(NULL, &tmp, xfs_buf_cmp);
while (!list_empty(&tmp)) {
- bp = list_entry(tmp.next, xfs_buf_t, b_list);
- ASSERT(target == bp->b_target);
-
+ struct xfs_buf *bp;
+ bp = list_first_entry(&tmp, struct xfs_buf, b_list);
list_del_init(&bp->b_list);
xfs_buf_iostrategy(bp);
count++;
}
-
- if (as_list_len > 0)
- purge_addresses();
if (count)
blk_run_address_space(target->bt_mapping);
@@ -1751,42 +1910,45 @@ xfs_flush_buftarg(
xfs_buftarg_t *target,
int wait)
{
- struct list_head tmp;
- xfs_buf_t *bp, *n;
+ xfs_buf_t *bp;
int pincount = 0;
+ LIST_HEAD(tmp_list);
+ LIST_HEAD(wait_list);
xfs_buf_runall_queues(xfsconvertd_workqueue);
xfs_buf_runall_queues(xfsdatad_workqueue);
xfs_buf_runall_queues(xfslogd_workqueue);
set_bit(XBT_FORCE_FLUSH, &target->bt_flags);
- pincount = xfs_buf_delwri_split(target, &tmp, 0);
+ pincount = xfs_buf_delwri_split(target, &tmp_list, 0);
/*
- * Dropped the delayed write list lock, now walk the temporary list
+ * Dropped the delayed write list lock, now walk the temporary list.
+ * All I/O is issued async and then if we need to wait for completion
+ * we do that after issuing all the IO.
*/
- list_for_each_entry_safe(bp, n, &tmp, b_list) {
+ list_sort(NULL, &tmp_list, xfs_buf_cmp);
+ while (!list_empty(&tmp_list)) {
+ bp = list_first_entry(&tmp_list, struct xfs_buf, b_list);
ASSERT(target == bp->b_target);
- if (wait)
+ list_del_init(&bp->b_list);
+ if (wait) {
bp->b_flags &= ~XBF_ASYNC;
- else
- list_del_init(&bp->b_list);
-
+ list_add(&bp->b_list, &wait_list);
+ }
xfs_buf_iostrategy(bp);
}
- if (wait)
+ if (wait) {
+ /* Expedite and wait for IO to complete. */
blk_run_address_space(target->bt_mapping);
+ while (!list_empty(&wait_list)) {
+ bp = list_first_entry(&wait_list, struct xfs_buf, b_list);
- /*
- * Remaining list items must be flushed before returning
- */
- while (!list_empty(&tmp)) {
- bp = list_entry(tmp.next, xfs_buf_t, b_list);
-
- list_del_init(&bp->b_list);
- xfs_iowait(bp);
- xfs_buf_relse(bp);
+ list_del_init(&bp->b_list);
+ xfs_iowait(bp);
+ xfs_buf_relse(bp);
+ }
}
return pincount;
diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h
index a34c7b54822d..5fbecefa5dfd 100644
--- a/fs/xfs/linux-2.6/xfs_buf.h
+++ b/fs/xfs/linux-2.6/xfs_buf.h
@@ -232,13 +232,17 @@ 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 int xfs_bwrite(struct xfs_mount *mp, struct xfs_buf *bp);
extern void xfs_bdwrite(void *mp, xfs_buf_t *bp);
+
+extern void xfsbdstrat(struct xfs_mount *, struct xfs_buf *);
+extern int xfs_bdstrat_cb(struct xfs_buf *);
+
extern void xfs_buf_ioend(xfs_buf_t *, int);
extern void xfs_buf_ioerror(xfs_buf_t *, int);
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,
+extern void xfs_buf_iomove(xfs_buf_t *, size_t, size_t, void *,
xfs_buf_rw_t);
static inline int xfs_buf_iostrategy(xfs_buf_t *bp)
@@ -261,6 +265,7 @@ extern int xfs_buf_ispin(xfs_buf_t *);
/* Delayed Write Buffer Routines */
extern void xfs_buf_delwri_dequeue(xfs_buf_t *);
+extern void xfs_buf_delwri_promote(xfs_buf_t *);
/* Buffer Daemon Setup Routines */
extern int xfs_buf_init(void);
@@ -270,33 +275,19 @@ extern void xfs_buf_terminate(void);
({ char __b[BDEVNAME_SIZE]; bdevname((target)->bt_bdev, __b); __b; })
-#define XFS_B_ASYNC XBF_ASYNC
-#define XFS_B_DELWRI XBF_DELWRI
-#define XFS_B_READ XBF_READ
-#define XFS_B_WRITE XBF_WRITE
-#define XFS_B_STALE XBF_STALE
-
-#define XFS_BUF_TRYLOCK XBF_TRYLOCK
-#define XFS_INCORE_TRYLOCK XBF_TRYLOCK
-#define XFS_BUF_LOCK XBF_LOCK
-#define XFS_BUF_MAPPED XBF_MAPPED
-
-#define BUF_BUSY XBF_DONT_BLOCK
-
#define XFS_BUF_BFLAGS(bp) ((bp)->b_flags)
#define XFS_BUF_ZEROFLAGS(bp) ((bp)->b_flags &= \
~(XBF_READ|XBF_WRITE|XBF_ASYNC|XBF_DELWRI|XBF_ORDERED))
-#define XFS_BUF_STALE(bp) ((bp)->b_flags |= XFS_B_STALE)
-#define XFS_BUF_UNSTALE(bp) ((bp)->b_flags &= ~XFS_B_STALE)
-#define XFS_BUF_ISSTALE(bp) ((bp)->b_flags & XFS_B_STALE)
+#define XFS_BUF_STALE(bp) ((bp)->b_flags |= XBF_STALE)
+#define XFS_BUF_UNSTALE(bp) ((bp)->b_flags &= ~XBF_STALE)
+#define XFS_BUF_ISSTALE(bp) ((bp)->b_flags & XBF_STALE)
#define XFS_BUF_SUPER_STALE(bp) do { \
XFS_BUF_STALE(bp); \
xfs_buf_delwri_dequeue(bp); \
XFS_BUF_DONE(bp); \
} while (0)
-#define XFS_BUF_MANAGE XBF_FS_MANAGED
#define XFS_BUF_UNMANAGE(bp) ((bp)->b_flags &= ~XBF_FS_MANAGED)
#define XFS_BUF_DELAYWRITE(bp) ((bp)->b_flags |= XBF_DELWRI)
@@ -385,31 +376,11 @@ static inline void xfs_buf_relse(xfs_buf_t *bp)
#define xfs_biomove(bp, off, len, data, rw) \
xfs_buf_iomove((bp), (off), (len), (data), \
- ((rw) == XFS_B_WRITE) ? XBRW_WRITE : XBRW_READ)
+ ((rw) == XBF_WRITE) ? XBRW_WRITE : XBRW_READ)
#define xfs_biozero(bp, off, len) \
xfs_buf_iomove((bp), (off), (len), NULL, XBRW_ZERO)
-
-static inline int XFS_bwrite(xfs_buf_t *bp)
-{
- int iowait = (bp->b_flags & XBF_ASYNC) == 0;
- int error = 0;
-
- if (!iowait)
- bp->b_flags |= _XBF_RUN_QUEUES;
-
- xfs_buf_delwri_dequeue(bp);
- xfs_buf_iostrategy(bp);
- if (iowait) {
- error = xfs_buf_iowait(bp);
- xfs_buf_relse(bp);
- }
- return error;
-}
-
-#define XFS_bdstrat(bp) xfs_buf_iorequest(bp)
-
#define xfs_iowait(bp) xfs_buf_iowait(bp)
#define xfs_baread(target, rablkno, ralen) \
@@ -419,11 +390,12 @@ static inline int XFS_bwrite(xfs_buf_t *bp)
/*
* Handling of buftargs.
*/
-extern xfs_buftarg_t *xfs_alloc_buftarg(struct block_device *, int);
+extern xfs_buftarg_t *xfs_alloc_buftarg(struct block_device *, int, const char *);
extern void xfs_free_buftarg(struct xfs_mount *, struct xfs_buftarg *);
extern void xfs_wait_buftarg(xfs_buftarg_t *);
extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int, unsigned int);
extern int xfs_flush_buftarg(xfs_buftarg_t *, int);
+
#ifdef CONFIG_KDB_MODULES
extern struct list_head *xfs_get_buftarg_list(void);
#endif
diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c
index 87b8cbd23d4b..e7839ee49e43 100644
--- a/fs/xfs/linux-2.6/xfs_export.c
+++ b/fs/xfs/linux-2.6/xfs_export.c
@@ -29,6 +29,7 @@
#include "xfs_vnodeops.h"
#include "xfs_bmap_btree.h"
#include "xfs_inode.h"
+#include "xfs_inode_item.h"
/*
* Note that we only accept fileids which are long enough rather than allow
@@ -127,13 +128,12 @@ xfs_nfs_get_inode(
return ERR_PTR(-ESTALE);
/*
- * The XFS_IGET_BULKSTAT means that an invalid inode number is just
- * fine and not an indication of a corrupted filesystem. Because
- * clients can send any kind of invalid file handle, e.g. after
- * a restore on the server we have to deal with this case gracefully.
+ * The XFS_IGET_UNTRUSTED means that an invalid inode number is just
+ * fine and not an indication of a corrupted filesystem as clients can
+ * send invalid file handles and we have to handle it gracefully..
*/
- error = xfs_iget(mp, NULL, ino, XFS_IGET_BULKSTAT,
- XFS_ILOCK_SHARED, &ip, 0);
+ error = xfs_iget(mp, NULL, ino, XFS_IGET_UNTRUSTED,
+ XFS_ILOCK_SHARED, &ip);
if (error) {
/*
* EINVAL means the inode cluster doesn't exist anymore.
@@ -215,9 +215,28 @@ xfs_fs_get_parent(
return d_obtain_alias(VFS_I(cip));
}
+STATIC int
+xfs_fs_nfs_commit_metadata(
+ struct inode *inode)
+{
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
+ int error = 0;
+
+ xfs_ilock(ip, XFS_ILOCK_SHARED);
+ if (xfs_ipincount(ip)) {
+ error = _xfs_log_force_lsn(mp, ip->i_itemp->ili_last_lsn,
+ XFS_LOG_SYNC, NULL);
+ }
+ xfs_iunlock(ip, XFS_ILOCK_SHARED);
+
+ return error;
+}
+
const struct export_operations xfs_export_operations = {
.encode_fh = xfs_fs_encode_fh,
.fh_to_dentry = xfs_fs_fh_to_dentry,
.fh_to_parent = xfs_fs_fh_to_parent,
.get_parent = xfs_fs_get_parent,
+ .commit_metadata = xfs_fs_nfs_commit_metadata,
};
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index e4caeb28ce2e..257a56b127cf 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -16,6 +16,7 @@
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "xfs.h"
+#include "xfs_fs.h"
#include "xfs_bit.h"
#include "xfs_log.h"
#include "xfs_inum.h"
@@ -34,52 +35,281 @@
#include "xfs_dir2_sf.h"
#include "xfs_dinode.h"
#include "xfs_inode.h"
+#include "xfs_inode_item.h"
+#include "xfs_bmap.h"
#include "xfs_error.h"
#include "xfs_rw.h"
#include "xfs_vnodeops.h"
#include "xfs_da_btree.h"
#include "xfs_ioctl.h"
+#include "xfs_trace.h"
#include <linux/dcache.h>
static const struct vm_operations_struct xfs_file_vm_ops;
-STATIC ssize_t
-xfs_file_aio_read(
- struct kiocb *iocb,
- const struct iovec *iov,
- unsigned long nr_segs,
- loff_t pos)
+/*
+ * xfs_iozero
+ *
+ * xfs_iozero clears the specified range of buffer supplied,
+ * and marks all the affected blocks as valid and modified. If
+ * an affected block is not allocated, it will be allocated. If
+ * an affected block is not completely overwritten, and is not
+ * valid before the operation, it will be read from disk before
+ * being partially zeroed.
+ */
+STATIC int
+xfs_iozero(
+ struct xfs_inode *ip, /* inode */
+ loff_t pos, /* offset in file */
+ size_t count) /* size of data to zero */
{
- struct file *file = iocb->ki_filp;
- int ioflags = 0;
+ struct page *page;
+ struct address_space *mapping;
+ int status;
- 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);
+ mapping = VFS_I(ip)->i_mapping;
+ do {
+ unsigned offset, bytes;
+ void *fsdata;
+
+ offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */
+ bytes = PAGE_CACHE_SIZE - offset;
+ if (bytes > count)
+ bytes = count;
+
+ status = pagecache_write_begin(NULL, mapping, pos, bytes,
+ AOP_FLAG_UNINTERRUPTIBLE,
+ &page, &fsdata);
+ if (status)
+ break;
+
+ zero_user(page, offset, bytes);
+
+ status = pagecache_write_end(NULL, mapping, pos, bytes, bytes,
+ page, fsdata);
+ WARN_ON(status <= 0); /* can't return less than zero! */
+ pos += bytes;
+ count -= bytes;
+ status = 0;
+ } while (count);
+
+ return (-status);
+}
+
+STATIC int
+xfs_file_fsync(
+ struct file *file,
+ int datasync)
+{
+ struct inode *inode = file->f_mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_trans *tp;
+ int error = 0;
+ int log_flushed = 0;
+
+ xfs_itrace_entry(ip);
+
+ if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+ return -XFS_ERROR(EIO);
+
+ xfs_iflags_clear(ip, XFS_ITRUNCATED);
+
+ xfs_ioend_wait(ip);
+
+ /*
+ * We always need to make sure that the required inode state is safe on
+ * disk. The inode might be clean but we still might need to force the
+ * log because of committed transactions that haven't hit the disk yet.
+ * Likewise, there could be unflushed non-transactional changes to the
+ * inode core that have to go to disk and this requires us to issue
+ * a synchronous transaction to capture these changes correctly.
+ *
+ * This code relies on the assumption that if the i_update_core field
+ * of the inode is clear and the inode is unpinned then it is clean
+ * and no action is required.
+ */
+ xfs_ilock(ip, XFS_ILOCK_SHARED);
+
+ /*
+ * First check if the VFS inode is marked dirty. All the dirtying
+ * of non-transactional updates no goes through mark_inode_dirty*,
+ * which allows us to distinguish beteeen pure timestamp updates
+ * and i_size updates which need to be caught for fdatasync.
+ * After that also theck for the dirty state in the XFS inode, which
+ * might gets cleared when the inode gets written out via the AIL
+ * or xfs_iflush_cluster.
+ */
+ if (((inode->i_state & I_DIRTY_DATASYNC) ||
+ ((inode->i_state & I_DIRTY_SYNC) && !datasync)) &&
+ ip->i_update_core) {
+ /*
+ * Kick off a transaction to log the inode core to get the
+ * updates. The sync transaction will also force the log.
+ */
+ xfs_iunlock(ip, XFS_ILOCK_SHARED);
+ tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_FSYNC_TS);
+ error = xfs_trans_reserve(tp, 0,
+ XFS_FSYNC_TS_LOG_RES(ip->i_mount), 0, 0, 0);
+ if (error) {
+ xfs_trans_cancel(tp, 0);
+ return -error;
+ }
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+
+ /*
+ * Note - it's possible that we might have pushed ourselves out
+ * of the way during trans_reserve which would flush the inode.
+ * But there's no guarantee that the inode buffer has actually
+ * gone out yet (it's delwri). Plus the buffer could be pinned
+ * anyway if it's part of an inode in another recent
+ * transaction. So we play it safe and fire off the
+ * transaction anyway.
+ */
+ xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+ xfs_trans_ihold(tp, ip);
+ xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+ xfs_trans_set_sync(tp);
+ error = _xfs_trans_commit(tp, 0, &log_flushed);
+
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ } else {
+ /*
+ * Timestamps/size haven't changed since last inode flush or
+ * inode transaction commit. That means either nothing got
+ * written or a transaction committed which caught the updates.
+ * If the latter happened and the transaction hasn't hit the
+ * disk yet, the inode will be still be pinned. If it is,
+ * force the log.
+ */
+ if (xfs_ipincount(ip)) {
+ error = _xfs_log_force_lsn(ip->i_mount,
+ ip->i_itemp->ili_last_lsn,
+ XFS_LOG_SYNC, &log_flushed);
+ }
+ xfs_iunlock(ip, XFS_ILOCK_SHARED);
+ }
+
+ if (ip->i_mount->m_flags & XFS_MOUNT_BARRIER) {
+ /*
+ * If the log write didn't issue an ordered tag we need
+ * to flush the disk cache for the data device now.
+ */
+ if (!log_flushed)
+ xfs_blkdev_issue_flush(ip->i_mount->m_ddev_targp);
+
+ /*
+ * If this inode is on the RT dev we need to flush that
+ * cache as well.
+ */
+ if (XFS_IS_REALTIME_INODE(ip))
+ xfs_blkdev_issue_flush(ip->i_mount->m_rtdev_targp);
+ }
+
+ return -error;
}
STATIC ssize_t
-xfs_file_aio_write(
+xfs_file_aio_read(
struct kiocb *iocb,
- const struct iovec *iov,
+ const struct iovec *iovp,
unsigned long nr_segs,
loff_t pos)
{
struct file *file = iocb->ki_filp;
+ struct inode *inode = file->f_mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
+ size_t size = 0;
+ ssize_t ret = 0;
int ioflags = 0;
+ xfs_fsize_t n;
+ unsigned long seg;
+
+ XFS_STATS_INC(xs_read_calls);
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);
+
+ /* START copy & waste from filemap.c */
+ for (seg = 0; seg < nr_segs; seg++) {
+ const struct iovec *iv = &iovp[seg];
+
+ /*
+ * If any segment has a negative length, or the cumulative
+ * length ever wraps negative then return -EINVAL.
+ */
+ size += iv->iov_len;
+ if (unlikely((ssize_t)(size|iv->iov_len) < 0))
+ return XFS_ERROR(-EINVAL);
+ }
+ /* END copy & waste from filemap.c */
+
+ if (unlikely(ioflags & IO_ISDIRECT)) {
+ xfs_buftarg_t *target =
+ XFS_IS_REALTIME_INODE(ip) ?
+ mp->m_rtdev_targp : mp->m_ddev_targp;
+ if ((iocb->ki_pos & target->bt_smask) ||
+ (size & target->bt_smask)) {
+ if (iocb->ki_pos == ip->i_size)
+ return 0;
+ return -XFS_ERROR(EINVAL);
+ }
+ }
+
+ n = XFS_MAXIOFFSET(mp) - iocb->ki_pos;
+ if (n <= 0 || size == 0)
+ return 0;
+
+ if (n < size)
+ size = n;
+
+ if (XFS_FORCED_SHUTDOWN(mp))
+ return -EIO;
+
+ if (unlikely(ioflags & IO_ISDIRECT))
+ mutex_lock(&inode->i_mutex);
+ xfs_ilock(ip, XFS_IOLOCK_SHARED);
+
+ if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) {
+ int dmflags = FILP_DELAY_FLAG(file) | DM_SEM_FLAG_RD(ioflags);
+ int iolock = XFS_IOLOCK_SHARED;
+
+ ret = -XFS_SEND_DATA(mp, DM_EVENT_READ, ip, iocb->ki_pos, size,
+ dmflags, &iolock);
+ if (ret) {
+ xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ if (unlikely(ioflags & IO_ISDIRECT))
+ mutex_unlock(&inode->i_mutex);
+ return ret;
+ }
+ }
+
+ if (unlikely(ioflags & IO_ISDIRECT)) {
+ if (inode->i_mapping->nrpages) {
+ ret = -xfs_flushinval_pages(ip,
+ (iocb->ki_pos & PAGE_CACHE_MASK),
+ -1, FI_REMAPF_LOCKED);
+ }
+ mutex_unlock(&inode->i_mutex);
+ if (ret) {
+ xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ return ret;
+ }
+ }
+
+ trace_xfs_file_read(ip, size, iocb->ki_pos, ioflags);
+
+ ret = generic_file_aio_read(iocb, iovp, nr_segs, iocb->ki_pos);
+ if (ret > 0)
+ XFS_STATS_ADD(xs_read_bytes, ret);
+
+ xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ return ret;
}
STATIC ssize_t
@@ -87,16 +317,44 @@ xfs_file_splice_read(
struct file *infilp,
loff_t *ppos,
struct pipe_inode_info *pipe,
- size_t len,
+ size_t count,
unsigned int flags)
{
+ struct xfs_inode *ip = XFS_I(infilp->f_mapping->host);
+ struct xfs_mount *mp = ip->i_mount;
int ioflags = 0;
+ ssize_t ret;
+
+ XFS_STATS_INC(xs_read_calls);
if (infilp->f_mode & FMODE_NOCMTIME)
ioflags |= IO_INVIS;
- return xfs_splice_read(XFS_I(infilp->f_path.dentry->d_inode),
- infilp, ppos, pipe, len, flags, ioflags);
+ if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+ return -EIO;
+
+ xfs_ilock(ip, XFS_IOLOCK_SHARED);
+
+ if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) {
+ int iolock = XFS_IOLOCK_SHARED;
+ int error;
+
+ error = XFS_SEND_DATA(mp, DM_EVENT_READ, ip, *ppos, count,
+ FILP_DELAY_FLAG(infilp), &iolock);
+ if (error) {
+ xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ return -error;
+ }
+ }
+
+ trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
+
+ ret = generic_file_splice_read(infilp, ppos, pipe, count, flags);
+ if (ret > 0)
+ XFS_STATS_ADD(xs_read_bytes, ret);
+
+ xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ return ret;
}
STATIC ssize_t
@@ -104,16 +362,538 @@ xfs_file_splice_write(
struct pipe_inode_info *pipe,
struct file *outfilp,
loff_t *ppos,
- size_t len,
+ size_t count,
unsigned int flags)
{
+ struct inode *inode = outfilp->f_mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
+ xfs_fsize_t isize, new_size;
int ioflags = 0;
+ ssize_t ret;
+
+ XFS_STATS_INC(xs_write_calls);
if (outfilp->f_mode & FMODE_NOCMTIME)
ioflags |= IO_INVIS;
- return xfs_splice_write(XFS_I(outfilp->f_path.dentry->d_inode),
- pipe, outfilp, ppos, len, flags, ioflags);
+ if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+ return -EIO;
+
+ xfs_ilock(ip, XFS_IOLOCK_EXCL);
+
+ if (DM_EVENT_ENABLED(ip, DM_EVENT_WRITE) && !(ioflags & IO_INVIS)) {
+ int iolock = XFS_IOLOCK_EXCL;
+ int error;
+
+ error = XFS_SEND_DATA(mp, DM_EVENT_WRITE, ip, *ppos, count,
+ FILP_DELAY_FLAG(outfilp), &iolock);
+ if (error) {
+ xfs_iunlock(ip, XFS_IOLOCK_EXCL);
+ return -error;
+ }
+ }
+
+ new_size = *ppos + count;
+
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ if (new_size > ip->i_size)
+ ip->i_new_size = new_size;
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+
+ trace_xfs_file_splice_write(ip, count, *ppos, ioflags);
+
+ ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags);
+ if (ret > 0)
+ XFS_STATS_ADD(xs_write_bytes, ret);
+
+ isize = i_size_read(inode);
+ if (unlikely(ret < 0 && ret != -EFAULT && *ppos > isize))
+ *ppos = isize;
+
+ if (*ppos > ip->i_size) {
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ if (*ppos > ip->i_size)
+ ip->i_size = *ppos;
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ }
+
+ if (ip->i_new_size) {
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ ip->i_new_size = 0;
+ if (ip->i_d.di_size > ip->i_size)
+ ip->i_d.di_size = ip->i_size;
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ }
+ xfs_iunlock(ip, XFS_IOLOCK_EXCL);
+ return ret;
+}
+
+/*
+ * This routine is called to handle zeroing any space in the last
+ * block of the file that is beyond the EOF. We do this since the
+ * size is being increased without writing anything to that block
+ * and we don't want anyone to read the garbage on the disk.
+ */
+STATIC int /* error (positive) */
+xfs_zero_last_block(
+ xfs_inode_t *ip,
+ xfs_fsize_t offset,
+ xfs_fsize_t isize)
+{
+ xfs_fileoff_t last_fsb;
+ xfs_mount_t *mp = ip->i_mount;
+ int nimaps;
+ int zero_offset;
+ int zero_len;
+ int error = 0;
+ xfs_bmbt_irec_t imap;
+
+ ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
+
+ zero_offset = XFS_B_FSB_OFFSET(mp, isize);
+ if (zero_offset == 0) {
+ /*
+ * There are no extra bytes in the last block on disk to
+ * zero, so return.
+ */
+ return 0;
+ }
+
+ last_fsb = XFS_B_TO_FSBT(mp, isize);
+ nimaps = 1;
+ error = xfs_bmapi(NULL, ip, last_fsb, 1, 0, NULL, 0, &imap,
+ &nimaps, NULL, NULL);
+ if (error) {
+ return error;
+ }
+ ASSERT(nimaps > 0);
+ /*
+ * If the block underlying isize is just a hole, then there
+ * is nothing to zero.
+ */
+ if (imap.br_startblock == HOLESTARTBLOCK) {
+ return 0;
+ }
+ /*
+ * Zero the part of the last block beyond the EOF, and write it
+ * out sync. We need to drop the ilock while we do this so we
+ * don't deadlock when the buffer cache calls back to us.
+ */
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+
+ zero_len = mp->m_sb.sb_blocksize - zero_offset;
+ if (isize + zero_len > offset)
+ zero_len = offset - isize;
+ error = xfs_iozero(ip, isize, zero_len);
+
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ ASSERT(error >= 0);
+ return error;
+}
+
+/*
+ * Zero any on disk space between the current EOF and the new,
+ * larger EOF. This handles the normal case of zeroing the remainder
+ * of the last block in the file and the unusual case of zeroing blocks
+ * out beyond the size of the file. This second case only happens
+ * with fixed size extents and when the system crashes before the inode
+ * size was updated but after blocks were allocated. If fill is set,
+ * then any holes in the range are filled and zeroed. If not, the holes
+ * are left alone as holes.
+ */
+
+int /* error (positive) */
+xfs_zero_eof(
+ xfs_inode_t *ip,
+ xfs_off_t offset, /* starting I/O offset */
+ xfs_fsize_t isize) /* current inode size */
+{
+ xfs_mount_t *mp = ip->i_mount;
+ xfs_fileoff_t start_zero_fsb;
+ xfs_fileoff_t end_zero_fsb;
+ xfs_fileoff_t zero_count_fsb;
+ xfs_fileoff_t last_fsb;
+ xfs_fileoff_t zero_off;
+ xfs_fsize_t zero_len;
+ int nimaps;
+ int error = 0;
+ xfs_bmbt_irec_t imap;
+
+ ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
+ ASSERT(offset > isize);
+
+ /*
+ * First handle zeroing the block on which isize resides.
+ * We only zero a part of that block so it is handled specially.
+ */
+ error = xfs_zero_last_block(ip, offset, isize);
+ if (error) {
+ ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
+ return error;
+ }
+
+ /*
+ * Calculate the range between the new size and the old
+ * where blocks needing to be zeroed may exist. To get the
+ * block where the last byte in the file currently resides,
+ * we need to subtract one from the size and truncate back
+ * to a block boundary. We subtract 1 in case the size is
+ * exactly on a block boundary.
+ */
+ last_fsb = isize ? XFS_B_TO_FSBT(mp, isize - 1) : (xfs_fileoff_t)-1;
+ start_zero_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize);
+ end_zero_fsb = XFS_B_TO_FSBT(mp, offset - 1);
+ ASSERT((xfs_sfiloff_t)last_fsb < (xfs_sfiloff_t)start_zero_fsb);
+ if (last_fsb == end_zero_fsb) {
+ /*
+ * The size was only incremented on its last block.
+ * We took care of that above, so just return.
+ */
+ return 0;
+ }
+
+ ASSERT(start_zero_fsb <= end_zero_fsb);
+ while (start_zero_fsb <= end_zero_fsb) {
+ nimaps = 1;
+ zero_count_fsb = end_zero_fsb - start_zero_fsb + 1;
+ error = xfs_bmapi(NULL, ip, start_zero_fsb, zero_count_fsb,
+ 0, NULL, 0, &imap, &nimaps, NULL, NULL);
+ if (error) {
+ ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
+ return error;
+ }
+ ASSERT(nimaps > 0);
+
+ if (imap.br_state == XFS_EXT_UNWRITTEN ||
+ imap.br_startblock == HOLESTARTBLOCK) {
+ /*
+ * This loop handles initializing pages that were
+ * partially initialized by the code below this
+ * loop. It basically zeroes the part of the page
+ * that sits on a hole and sets the page as P_HOLE
+ * and calls remapf if it is a mapped file.
+ */
+ start_zero_fsb = imap.br_startoff + imap.br_blockcount;
+ ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
+ continue;
+ }
+
+ /*
+ * There are blocks we need to zero.
+ * Drop the inode lock while we're doing the I/O.
+ * We'll still have the iolock to protect us.
+ */
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+
+ zero_off = XFS_FSB_TO_B(mp, start_zero_fsb);
+ zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount);
+
+ if ((zero_off + zero_len) > offset)
+ zero_len = offset - zero_off;
+
+ error = xfs_iozero(ip, zero_off, zero_len);
+ if (error) {
+ goto out_lock;
+ }
+
+ start_zero_fsb = imap.br_startoff + imap.br_blockcount;
+ ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
+
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ }
+
+ return 0;
+
+out_lock:
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ ASSERT(error >= 0);
+ return error;
+}
+
+STATIC ssize_t
+xfs_file_aio_write(
+ struct kiocb *iocb,
+ const struct iovec *iovp,
+ unsigned long nr_segs,
+ loff_t pos)
+{
+ struct file *file = iocb->ki_filp;
+ struct address_space *mapping = file->f_mapping;
+ struct inode *inode = mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
+ ssize_t ret = 0, error = 0;
+ int ioflags = 0;
+ xfs_fsize_t isize, new_size;
+ int iolock;
+ int eventsent = 0;
+ size_t ocount = 0, count;
+ int need_i_mutex;
+
+ XFS_STATS_INC(xs_write_calls);
+
+ 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;
+
+ error = generic_segment_checks(iovp, &nr_segs, &ocount, VERIFY_READ);
+ if (error)
+ return error;
+
+ count = ocount;
+ if (count == 0)
+ return 0;
+
+ xfs_wait_for_freeze(mp, SB_FREEZE_WRITE);
+
+ if (XFS_FORCED_SHUTDOWN(mp))
+ return -EIO;
+
+relock:
+ if (ioflags & IO_ISDIRECT) {
+ iolock = XFS_IOLOCK_SHARED;
+ need_i_mutex = 0;
+ } else {
+ iolock = XFS_IOLOCK_EXCL;
+ need_i_mutex = 1;
+ mutex_lock(&inode->i_mutex);
+ }
+
+ xfs_ilock(ip, XFS_ILOCK_EXCL|iolock);
+
+start:
+ error = -generic_write_checks(file, &pos, &count,
+ S_ISBLK(inode->i_mode));
+ if (error) {
+ xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock);
+ goto out_unlock_mutex;
+ }
+
+ if ((DM_EVENT_ENABLED(ip, DM_EVENT_WRITE) &&
+ !(ioflags & IO_INVIS) && !eventsent)) {
+ int dmflags = FILP_DELAY_FLAG(file);
+
+ if (need_i_mutex)
+ dmflags |= DM_FLAGS_IMUX;
+
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ error = XFS_SEND_DATA(ip->i_mount, DM_EVENT_WRITE, ip,
+ pos, count, dmflags, &iolock);
+ if (error) {
+ goto out_unlock_internal;
+ }
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ eventsent = 1;
+
+ /*
+ * The iolock was dropped and reacquired in XFS_SEND_DATA
+ * so we have to recheck the size when appending.
+ * We will only "goto start;" once, since having sent the
+ * event prevents another call to XFS_SEND_DATA, which is
+ * what allows the size to change in the first place.
+ */
+ if ((file->f_flags & O_APPEND) && pos != ip->i_size)
+ goto start;
+ }
+
+ if (ioflags & IO_ISDIRECT) {
+ xfs_buftarg_t *target =
+ XFS_IS_REALTIME_INODE(ip) ?
+ mp->m_rtdev_targp : mp->m_ddev_targp;
+
+ if ((pos & target->bt_smask) || (count & target->bt_smask)) {
+ xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock);
+ return XFS_ERROR(-EINVAL);
+ }
+
+ if (!need_i_mutex && (mapping->nrpages || pos > ip->i_size)) {
+ xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock);
+ iolock = XFS_IOLOCK_EXCL;
+ need_i_mutex = 1;
+ mutex_lock(&inode->i_mutex);
+ xfs_ilock(ip, XFS_ILOCK_EXCL|iolock);
+ goto start;
+ }
+ }
+
+ new_size = pos + count;
+ if (new_size > ip->i_size)
+ ip->i_new_size = new_size;
+
+ if (likely(!(ioflags & IO_INVIS)))
+ file_update_time(file);
+
+ /*
+ * If the offset is beyond the size of the file, we have a couple
+ * of things to do. First, if there is already space allocated
+ * we need to either create holes or zero the disk or ...
+ *
+ * If there is a page where the previous size lands, we need
+ * to zero it out up to the new size.
+ */
+
+ if (pos > ip->i_size) {
+ error = xfs_zero_eof(ip, pos, ip->i_size);
+ if (error) {
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ goto out_unlock_internal;
+ }
+ }
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+
+ /*
+ * If we're writing the file then make sure to clear the
+ * setuid and setgid bits if the process is not being run
+ * by root. This keeps people from modifying setuid and
+ * setgid binaries.
+ */
+ error = -file_remove_suid(file);
+ if (unlikely(error))
+ goto out_unlock_internal;
+
+ /* We can write back this queue in page reclaim */
+ current->backing_dev_info = mapping->backing_dev_info;
+
+ if ((ioflags & IO_ISDIRECT)) {
+ if (mapping->nrpages) {
+ WARN_ON(need_i_mutex == 0);
+ error = xfs_flushinval_pages(ip,
+ (pos & PAGE_CACHE_MASK),
+ -1, FI_REMAPF_LOCKED);
+ if (error)
+ goto out_unlock_internal;
+ }
+
+ if (need_i_mutex) {
+ /* demote the lock now the cached pages are gone */
+ xfs_ilock_demote(ip, XFS_IOLOCK_EXCL);
+ mutex_unlock(&inode->i_mutex);
+
+ iolock = XFS_IOLOCK_SHARED;
+ need_i_mutex = 0;
+ }
+
+ trace_xfs_file_direct_write(ip, count, iocb->ki_pos, ioflags);
+ ret = generic_file_direct_write(iocb, iovp,
+ &nr_segs, pos, &iocb->ki_pos, count, ocount);
+
+ /*
+ * direct-io write to a hole: fall through to buffered I/O
+ * for completing the rest of the request.
+ */
+ if (ret >= 0 && ret != count) {
+ XFS_STATS_ADD(xs_write_bytes, ret);
+
+ pos += ret;
+ count -= ret;
+
+ ioflags &= ~IO_ISDIRECT;
+ xfs_iunlock(ip, iolock);
+ goto relock;
+ }
+ } else {
+ int enospc = 0;
+ ssize_t ret2 = 0;
+
+write_retry:
+ trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, ioflags);
+ ret2 = generic_file_buffered_write(iocb, iovp, nr_segs,
+ pos, &iocb->ki_pos, count, ret);
+ /*
+ * if we just got an ENOSPC, flush the inode now we
+ * aren't holding any page locks and retry *once*
+ */
+ if (ret2 == -ENOSPC && !enospc) {
+ error = xfs_flush_pages(ip, 0, -1, 0, FI_NONE);
+ if (error)
+ goto out_unlock_internal;
+ enospc = 1;
+ goto write_retry;
+ }
+ ret = ret2;
+ }
+
+ current->backing_dev_info = NULL;
+
+ isize = i_size_read(inode);
+ if (unlikely(ret < 0 && ret != -EFAULT && iocb->ki_pos > isize))
+ iocb->ki_pos = isize;
+
+ if (iocb->ki_pos > ip->i_size) {
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ if (iocb->ki_pos > ip->i_size)
+ ip->i_size = iocb->ki_pos;
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ }
+
+ if (ret == -ENOSPC &&
+ DM_EVENT_ENABLED(ip, DM_EVENT_NOSPACE) && !(ioflags & IO_INVIS)) {
+ xfs_iunlock(ip, iolock);
+ if (need_i_mutex)
+ mutex_unlock(&inode->i_mutex);
+ error = XFS_SEND_NAMESP(ip->i_mount, DM_EVENT_NOSPACE, ip,
+ DM_RIGHT_NULL, ip, DM_RIGHT_NULL, NULL, NULL,
+ 0, 0, 0); /* Delay flag intentionally unused */
+ if (need_i_mutex)
+ mutex_lock(&inode->i_mutex);
+ xfs_ilock(ip, iolock);
+ if (error)
+ goto out_unlock_internal;
+ goto start;
+ }
+
+ error = -ret;
+ if (ret <= 0)
+ goto out_unlock_internal;
+
+ XFS_STATS_ADD(xs_write_bytes, ret);
+
+ /* Handle various SYNC-type writes */
+ if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
+ loff_t end = pos + ret - 1;
+ int error2;
+
+ xfs_iunlock(ip, iolock);
+ if (need_i_mutex)
+ mutex_unlock(&inode->i_mutex);
+
+ error2 = filemap_write_and_wait_range(mapping, pos, end);
+ if (!error)
+ error = error2;
+ if (need_i_mutex)
+ mutex_lock(&inode->i_mutex);
+ xfs_ilock(ip, iolock);
+
+ error2 = -xfs_file_fsync(file,
+ (file->f_flags & __O_SYNC) ? 0 : 1);
+ if (!error)
+ error = error2;
+ }
+
+ out_unlock_internal:
+ if (ip->i_new_size) {
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ ip->i_new_size = 0;
+ /*
+ * If this was a direct or synchronous I/O that failed (such
+ * as ENOSPC) then part of the I/O may have been written to
+ * disk before the error occured. In this case the on-disk
+ * file size may have been adjusted beyond the in-memory file
+ * size and now needs to be truncated back.
+ */
+ if (ip->i_d.di_size > ip->i_size)
+ ip->i_d.di_size = ip->i_size;
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ }
+ xfs_iunlock(ip, iolock);
+ out_unlock_mutex:
+ if (need_i_mutex)
+ mutex_unlock(&inode->i_mutex);
+ return -error;
}
STATIC int
@@ -160,28 +940,6 @@ xfs_file_release(
return -xfs_release(XFS_I(inode));
}
-/*
- * We ignore the datasync flag here because a datasync is effectively
- * identical to an fsync. That is, datasync implies that we need to write
- * only the metadata needed to be able to access the data that is written
- * if we crash after the call completes. Hence if we are writing beyond
- * EOF we have to log the inode size change as well, which makes it a
- * full fsync. If we don't write beyond EOF, the inode core will be
- * clean in memory and so we don't need to log the inode, just like
- * fsync.
- */
-STATIC int
-xfs_file_fsync(
- struct file *file,
- struct dentry *dentry,
- int datasync)
-{
- struct xfs_inode *ip = XFS_I(dentry->d_inode);
-
- xfs_iflags_clear(ip, XFS_ITRUNCATED);
- return -xfs_fsync(ip);
-}
-
STATIC int
xfs_file_readdir(
struct file *filp,
@@ -203,9 +961,9 @@ xfs_file_readdir(
*
* Try to give it an estimate that's good enough, maybe at some
* point we can change the ->readdir prototype to include the
- * buffer size.
+ * buffer size. For now we use the current glibc buffer size.
*/
- bufsize = (size_t)min_t(loff_t, PAGE_SIZE, ip->i_d.di_size);
+ bufsize = (size_t)min_t(loff_t, 32768, ip->i_d.di_size);
error = xfs_readdir(ip, dirent, bufsize,
(xfs_off_t *)&filp->f_pos, filldir);
diff --git a/fs/xfs/linux-2.6/xfs_fs_subr.c b/fs/xfs/linux-2.6/xfs_fs_subr.c
index 7501b85fd860..b6918d76bc7b 100644
--- a/fs/xfs/linux-2.6/xfs_fs_subr.c
+++ b/fs/xfs/linux-2.6/xfs_fs_subr.c
@@ -79,7 +79,7 @@ xfs_flush_pages(
xfs_iflags_clear(ip, XFS_ITRUNCATED);
ret = -filemap_fdatawrite(mapping);
}
- if (flags & XFS_B_ASYNC)
+ if (flags & XBF_ASYNC)
return ret;
ret2 = xfs_wait_on_pages(ip, first, last);
if (!ret)
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c
index a034cf624437..e59a81062830 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl.c
@@ -58,6 +58,7 @@
#include <linux/mount.h>
#include <linux/namei.h>
#include <linux/pagemap.h>
+#include <linux/slab.h>
#include <linux/exportfs.h>
/*
@@ -447,12 +448,12 @@ xfs_attrlist_by_handle(
int
xfs_attrmulti_attr_get(
struct inode *inode,
- char *name,
- char __user *ubuf,
+ unsigned char *name,
+ unsigned char __user *ubuf,
__uint32_t *len,
__uint32_t flags)
{
- char *kbuf;
+ unsigned char *kbuf;
int error = EFAULT;
if (*len > XATTR_SIZE_MAX)
@@ -476,12 +477,12 @@ xfs_attrmulti_attr_get(
int
xfs_attrmulti_attr_set(
struct inode *inode,
- char *name,
- const char __user *ubuf,
+ unsigned char *name,
+ const unsigned char __user *ubuf,
__uint32_t len,
__uint32_t flags)
{
- char *kbuf;
+ unsigned char *kbuf;
int error = EFAULT;
if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
@@ -501,7 +502,7 @@ xfs_attrmulti_attr_set(
int
xfs_attrmulti_attr_remove(
struct inode *inode,
- char *name,
+ unsigned char *name,
__uint32_t flags)
{
if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
@@ -519,13 +520,17 @@ xfs_attrmulti_by_handle(
xfs_fsop_attrmulti_handlereq_t am_hreq;
struct dentry *dentry;
unsigned int i, size;
- char *attr_name;
+ unsigned char *attr_name;
if (!capable(CAP_SYS_ADMIN))
return -XFS_ERROR(EPERM);
if (copy_from_user(&am_hreq, arg, sizeof(xfs_fsop_attrmulti_handlereq_t)))
return -XFS_ERROR(EFAULT);
+ /* overflow check */
+ if (am_hreq.opcount >= INT_MAX / sizeof(xfs_attr_multiop_t))
+ return -E2BIG;
+
dentry = xfs_handlereq_to_dentry(parfilp, &am_hreq.hreq);
if (IS_ERR(dentry))
return PTR_ERR(dentry);
@@ -547,7 +552,7 @@ xfs_attrmulti_by_handle(
error = 0;
for (i = 0; i < am_hreq.opcount; i++) {
- ops[i].am_error = strncpy_from_user(attr_name,
+ ops[i].am_error = strncpy_from_user((char *)attr_name,
ops[i].am_attrname, MAXNAMELEN);
if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN)
error = -ERANGE;
@@ -674,10 +679,9 @@ xfs_ioc_bulkstat(
error = xfs_bulkstat_single(mp, &inlast,
bulkreq.ubuffer, &done);
else /* XFS_IOC_FSBULKSTAT */
- error = xfs_bulkstat(mp, &inlast, &count,
- (bulkstat_one_pf)xfs_bulkstat_one, NULL,
- sizeof(xfs_bstat_t), bulkreq.ubuffer,
- BULKSTAT_FG_QUICK, &done);
+ error = xfs_bulkstat(mp, &inlast, &count, xfs_bulkstat_one,
+ sizeof(xfs_bstat_t), bulkreq.ubuffer,
+ &done);
if (error)
return -error;
@@ -1431,6 +1435,9 @@ xfs_file_ioctl(
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
+ if (mp->m_flags & XFS_MOUNT_RDONLY)
+ return -XFS_ERROR(EROFS);
+
if (copy_from_user(&inout, arg, sizeof(inout)))
return -XFS_ERROR(EFAULT);
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.h b/fs/xfs/linux-2.6/xfs_ioctl.h
index 7bd7c6afc1eb..d56173b34a2a 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.h
+++ b/fs/xfs/linux-2.6/xfs_ioctl.h
@@ -45,23 +45,23 @@ xfs_readlink_by_handle(
extern int
xfs_attrmulti_attr_get(
struct inode *inode,
- char *name,
- char __user *ubuf,
+ unsigned char *name,
+ unsigned char __user *ubuf,
__uint32_t *len,
__uint32_t flags);
extern int
- xfs_attrmulti_attr_set(
+xfs_attrmulti_attr_set(
struct inode *inode,
- char *name,
- const char __user *ubuf,
+ unsigned char *name,
+ const unsigned char __user *ubuf,
__uint32_t len,
__uint32_t flags);
extern int
xfs_attrmulti_attr_remove(
struct inode *inode,
- char *name,
+ unsigned char *name,
__uint32_t flags);
extern struct dentry *
diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.c b/fs/xfs/linux-2.6/xfs_ioctl32.c
index be1527b1670c..52ed49e6465c 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl32.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl32.c
@@ -18,6 +18,7 @@
#include <linux/compat.h>
#include <linux/ioctl.h>
#include <linux/mount.h>
+#include <linux/slab.h>
#include <asm/uaccess.h>
#include "xfs.h"
#include "xfs_fs.h"
@@ -236,15 +237,12 @@ xfs_bulkstat_one_compat(
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);
+ xfs_bulkstat_one_fmt_compat,
+ ubused, stat);
}
/* copied from xfs_ioctl.c */
@@ -297,13 +295,11 @@ xfs_compat_ioc_bulkstat(
int res;
error = xfs_bulkstat_one_compat(mp, inlast, bulkreq.ubuffer,
- sizeof(compat_xfs_bstat_t),
- NULL, 0, NULL, NULL, &res);
+ sizeof(compat_xfs_bstat_t), 0, &res);
} else if (cmd == XFS_IOC_FSBULKSTAT_32) {
error = xfs_bulkstat(mp, &inlast, &count,
- xfs_bulkstat_one_compat, NULL,
- sizeof(compat_xfs_bstat_t), bulkreq.ubuffer,
- BULKSTAT_FG_QUICK, &done);
+ xfs_bulkstat_one_compat, sizeof(compat_xfs_bstat_t),
+ bulkreq.ubuffer, &done);
} else
error = XFS_ERROR(EINVAL);
if (error)
@@ -411,7 +407,7 @@ xfs_compat_attrmulti_by_handle(
compat_xfs_fsop_attrmulti_handlereq_t am_hreq;
struct dentry *dentry;
unsigned int i, size;
- char *attr_name;
+ unsigned char *attr_name;
if (!capable(CAP_SYS_ADMIN))
return -XFS_ERROR(EPERM);
@@ -419,6 +415,10 @@ xfs_compat_attrmulti_by_handle(
sizeof(compat_xfs_fsop_attrmulti_handlereq_t)))
return -XFS_ERROR(EFAULT);
+ /* overflow check */
+ if (am_hreq.opcount >= INT_MAX / sizeof(compat_xfs_attr_multiop_t))
+ return -E2BIG;
+
dentry = xfs_compat_handlereq_to_dentry(parfilp, &am_hreq.hreq);
if (IS_ERR(dentry))
return PTR_ERR(dentry);
@@ -440,7 +440,7 @@ xfs_compat_attrmulti_by_handle(
error = 0;
for (i = 0; i < am_hreq.opcount; i++) {
- ops[i].am_error = strncpy_from_user(attr_name,
+ ops[i].am_error = strncpy_from_user((char *)attr_name,
compat_ptr(ops[i].am_attrname),
MAXNAMELEN);
if (ops[i].am_error == 0 || ops[i].am_error == MAXNAMELEN)
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index 225946012d0b..44f0b2de153e 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -56,6 +56,7 @@
#include <linux/security.h>
#include <linux/falloc.h>
#include <linux/fiemap.h>
+#include <linux/slab.h>
/*
* Bring the timestamps in the XFS inode uptodate.
@@ -91,6 +92,16 @@ xfs_mark_inode_dirty_sync(
mark_inode_dirty_sync(inode);
}
+void
+xfs_mark_inode_dirty(
+ xfs_inode_t *ip)
+{
+ struct inode *inode = VFS_I(ip);
+
+ if (!(inode->i_state & (I_WILL_FREE|I_FREEING|I_CLEAR)))
+ mark_inode_dirty(inode);
+}
+
/*
* Change the requested timestamp in the given inode.
* We don't lock across timestamp updates, and we don't log them but
@@ -140,10 +151,10 @@ xfs_init_security(
struct xfs_inode *ip = XFS_I(inode);
size_t length;
void *value;
- char *name;
+ unsigned char *name;
int error;
- error = security_inode_init_security(inode, dir, &name,
+ error = security_inode_init_security(inode, dir, (char **)&name,
&value, &length);
if (error) {
if (error == -EOPNOTSUPP)
@@ -574,11 +585,20 @@ xfs_vn_fallocate(
bf.l_len = len;
xfs_ilock(ip, XFS_IOLOCK_EXCL);
+
+ /* check the new inode size is valid before allocating */
+ if (!(mode & FALLOC_FL_KEEP_SIZE) &&
+ offset + len > i_size_read(inode)) {
+ new_size = offset + len;
+ error = inode_newsize_ok(inode, new_size);
+ if (error)
+ goto out_unlock;
+ }
+
error = -xfs_change_file_space(ip, XFS_IOC_RESVSP, &bf,
0, XFS_ATTR_NOLOCK);
- if (!error && !(mode & FALLOC_FL_KEEP_SIZE) &&
- offset + len > i_size_read(inode))
- new_size = offset + len;
+ if (error)
+ goto out_unlock;
/* Change file size if needed */
if (new_size) {
@@ -589,6 +609,7 @@ xfs_vn_fallocate(
error = -xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK);
}
+out_unlock:
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
out_error:
return error;
@@ -662,7 +683,10 @@ xfs_vn_fiemap(
bm.bmv_length = BTOBB(length);
/* We add one because in getbmap world count includes the header */
- bm.bmv_count = fieinfo->fi_extents_max + 1;
+ bm.bmv_count = !fieinfo->fi_extents_max ? MAXEXTNUM :
+ fieinfo->fi_extents_max + 1;
+ bm.bmv_count = min_t(__s32, bm.bmv_count,
+ (PAGE_SIZE * 16 / sizeof(struct getbmapx)));
bm.bmv_iflags = BMV_IF_PREALLOC;
if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR)
bm.bmv_iflags |= BMV_IF_ATTRFORK;
diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h
index 5af0c81ca1ae..facfb323a706 100644
--- a/fs/xfs/linux-2.6/xfs_linux.h
+++ b/fs/xfs/linux-2.6/xfs_linux.h
@@ -88,7 +88,6 @@
#include <xfs_super.h>
#include <xfs_globals.h>
#include <xfs_fs_subr.h>
-#include <xfs_lrw.h>
#include <xfs_buf.h>
/*
diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c
deleted file mode 100644
index 0d32457abef1..000000000000
--- a/fs/xfs/linux-2.6/xfs_lrw.c
+++ /dev/null
@@ -1,852 +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_fs.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_alloc.h"
-#include "xfs_dmapi.h"
-#include "xfs_quota.h"
-#include "xfs_mount.h"
-#include "xfs_bmap_btree.h"
-#include "xfs_alloc_btree.h"
-#include "xfs_ialloc_btree.h"
-#include "xfs_dir2_sf.h"
-#include "xfs_attr_sf.h"
-#include "xfs_dinode.h"
-#include "xfs_inode.h"
-#include "xfs_bmap.h"
-#include "xfs_btree.h"
-#include "xfs_ialloc.h"
-#include "xfs_rtalloc.h"
-#include "xfs_error.h"
-#include "xfs_itable.h"
-#include "xfs_rw.h"
-#include "xfs_attr.h"
-#include "xfs_inode_item.h"
-#include "xfs_buf_item.h"
-#include "xfs_utils.h"
-#include "xfs_iomap.h"
-#include "xfs_vnodeops.h"
-#include "xfs_trace.h"
-
-#include <linux/capability.h>
-#include <linux/writeback.h>
-
-
-/*
- * xfs_iozero
- *
- * xfs_iozero clears the specified range of buffer supplied,
- * and marks all the affected blocks as valid and modified. If
- * an affected block is not allocated, it will be allocated. If
- * an affected block is not completely overwritten, and is not
- * valid before the operation, it will be read from disk before
- * being partially zeroed.
- */
-STATIC int
-xfs_iozero(
- struct xfs_inode *ip, /* inode */
- loff_t pos, /* offset in file */
- size_t count) /* size of data to zero */
-{
- struct page *page;
- struct address_space *mapping;
- int status;
-
- mapping = VFS_I(ip)->i_mapping;
- do {
- unsigned offset, bytes;
- void *fsdata;
-
- offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */
- bytes = PAGE_CACHE_SIZE - offset;
- if (bytes > count)
- bytes = count;
-
- status = pagecache_write_begin(NULL, mapping, pos, bytes,
- AOP_FLAG_UNINTERRUPTIBLE,
- &page, &fsdata);
- if (status)
- break;
-
- zero_user(page, offset, bytes);
-
- status = pagecache_write_end(NULL, mapping, pos, bytes, bytes,
- page, fsdata);
- WARN_ON(status <= 0); /* can't return less than zero! */
- pos += bytes;
- count -= bytes;
- status = 0;
- } while (count);
-
- return (-status);
-}
-
-ssize_t /* bytes read, or (-) error */
-xfs_read(
- xfs_inode_t *ip,
- struct kiocb *iocb,
- const struct iovec *iovp,
- unsigned int segs,
- loff_t *offset,
- int ioflags)
-{
- struct file *file = iocb->ki_filp;
- struct inode *inode = file->f_mapping->host;
- xfs_mount_t *mp = ip->i_mount;
- size_t size = 0;
- ssize_t ret = 0;
- xfs_fsize_t n;
- unsigned long seg;
-
-
- XFS_STATS_INC(xs_read_calls);
-
- /* START copy & waste from filemap.c */
- for (seg = 0; seg < segs; seg++) {
- const struct iovec *iv = &iovp[seg];
-
- /*
- * If any segment has a negative length, or the cumulative
- * length ever wraps negative then return -EINVAL.
- */
- size += iv->iov_len;
- if (unlikely((ssize_t)(size|iv->iov_len) < 0))
- return XFS_ERROR(-EINVAL);
- }
- /* END copy & waste from filemap.c */
-
- if (unlikely(ioflags & IO_ISDIRECT)) {
- xfs_buftarg_t *target =
- XFS_IS_REALTIME_INODE(ip) ?
- mp->m_rtdev_targp : mp->m_ddev_targp;
- if ((*offset & target->bt_smask) ||
- (size & target->bt_smask)) {
- if (*offset == ip->i_size) {
- return (0);
- }
- return -XFS_ERROR(EINVAL);
- }
- }
-
- n = XFS_MAXIOFFSET(mp) - *offset;
- if ((n <= 0) || (size == 0))
- return 0;
-
- if (n < size)
- size = n;
-
- if (XFS_FORCED_SHUTDOWN(mp))
- return -EIO;
-
- if (unlikely(ioflags & IO_ISDIRECT))
- mutex_lock(&inode->i_mutex);
- xfs_ilock(ip, XFS_IOLOCK_SHARED);
-
- if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) {
- int dmflags = FILP_DELAY_FLAG(file) | DM_SEM_FLAG_RD(ioflags);
- int iolock = XFS_IOLOCK_SHARED;
-
- ret = -XFS_SEND_DATA(mp, DM_EVENT_READ, ip, *offset, size,
- dmflags, &iolock);
- if (ret) {
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- if (unlikely(ioflags & IO_ISDIRECT))
- mutex_unlock(&inode->i_mutex);
- return ret;
- }
- }
-
- if (unlikely(ioflags & IO_ISDIRECT)) {
- if (inode->i_mapping->nrpages)
- ret = -xfs_flushinval_pages(ip, (*offset & PAGE_CACHE_MASK),
- -1, FI_REMAPF_LOCKED);
- mutex_unlock(&inode->i_mutex);
- if (ret) {
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- return ret;
- }
- }
-
- trace_xfs_file_read(ip, size, *offset, ioflags);
-
- iocb->ki_pos = *offset;
- ret = generic_file_aio_read(iocb, iovp, segs, *offset);
- if (ret > 0)
- XFS_STATS_ADD(xs_read_bytes, ret);
-
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- return ret;
-}
-
-ssize_t
-xfs_splice_read(
- xfs_inode_t *ip,
- struct file *infilp,
- loff_t *ppos,
- struct pipe_inode_info *pipe,
- size_t count,
- int flags,
- int ioflags)
-{
- xfs_mount_t *mp = ip->i_mount;
- ssize_t ret;
-
- XFS_STATS_INC(xs_read_calls);
- if (XFS_FORCED_SHUTDOWN(ip->i_mount))
- return -EIO;
-
- xfs_ilock(ip, XFS_IOLOCK_SHARED);
-
- if (DM_EVENT_ENABLED(ip, DM_EVENT_READ) && !(ioflags & IO_INVIS)) {
- int iolock = XFS_IOLOCK_SHARED;
- int error;
-
- error = XFS_SEND_DATA(mp, DM_EVENT_READ, ip, *ppos, count,
- FILP_DELAY_FLAG(infilp), &iolock);
- if (error) {
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- return -error;
- }
- }
-
- trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
-
- ret = generic_file_splice_read(infilp, ppos, pipe, count, flags);
- if (ret > 0)
- XFS_STATS_ADD(xs_read_bytes, ret);
-
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- return ret;
-}
-
-ssize_t
-xfs_splice_write(
- xfs_inode_t *ip,
- struct pipe_inode_info *pipe,
- struct file *outfilp,
- loff_t *ppos,
- size_t count,
- int flags,
- int ioflags)
-{
- xfs_mount_t *mp = ip->i_mount;
- ssize_t ret;
- struct inode *inode = outfilp->f_mapping->host;
- xfs_fsize_t isize, new_size;
-
- XFS_STATS_INC(xs_write_calls);
- if (XFS_FORCED_SHUTDOWN(ip->i_mount))
- return -EIO;
-
- xfs_ilock(ip, XFS_IOLOCK_EXCL);
-
- if (DM_EVENT_ENABLED(ip, DM_EVENT_WRITE) && !(ioflags & IO_INVIS)) {
- int iolock = XFS_IOLOCK_EXCL;
- int error;
-
- error = XFS_SEND_DATA(mp, DM_EVENT_WRITE, ip, *ppos, count,
- FILP_DELAY_FLAG(outfilp), &iolock);
- if (error) {
- xfs_iunlock(ip, XFS_IOLOCK_EXCL);
- return -error;
- }
- }
-
- new_size = *ppos + count;
-
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- if (new_size > ip->i_size)
- ip->i_new_size = new_size;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
-
- trace_xfs_file_splice_write(ip, count, *ppos, ioflags);
-
- ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags);
- if (ret > 0)
- XFS_STATS_ADD(xs_write_bytes, ret);
-
- isize = i_size_read(inode);
- if (unlikely(ret < 0 && ret != -EFAULT && *ppos > isize))
- *ppos = isize;
-
- if (*ppos > ip->i_size) {
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- if (*ppos > ip->i_size)
- ip->i_size = *ppos;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
-
- if (ip->i_new_size) {
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- ip->i_new_size = 0;
- if (ip->i_d.di_size > ip->i_size)
- ip->i_d.di_size = ip->i_size;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
- xfs_iunlock(ip, XFS_IOLOCK_EXCL);
- return ret;
-}
-
-/*
- * This routine is called to handle zeroing any space in the last
- * block of the file that is beyond the EOF. We do this since the
- * size is being increased without writing anything to that block
- * and we don't want anyone to read the garbage on the disk.
- */
-STATIC int /* error (positive) */
-xfs_zero_last_block(
- xfs_inode_t *ip,
- xfs_fsize_t offset,
- xfs_fsize_t isize)
-{
- xfs_fileoff_t last_fsb;
- xfs_mount_t *mp = ip->i_mount;
- int nimaps;
- int zero_offset;
- int zero_len;
- int error = 0;
- xfs_bmbt_irec_t imap;
-
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-
- zero_offset = XFS_B_FSB_OFFSET(mp, isize);
- if (zero_offset == 0) {
- /*
- * There are no extra bytes in the last block on disk to
- * zero, so return.
- */
- return 0;
- }
-
- last_fsb = XFS_B_TO_FSBT(mp, isize);
- nimaps = 1;
- error = xfs_bmapi(NULL, ip, last_fsb, 1, 0, NULL, 0, &imap,
- &nimaps, NULL, NULL);
- if (error) {
- return error;
- }
- ASSERT(nimaps > 0);
- /*
- * If the block underlying isize is just a hole, then there
- * is nothing to zero.
- */
- if (imap.br_startblock == HOLESTARTBLOCK) {
- return 0;
- }
- /*
- * Zero the part of the last block beyond the EOF, and write it
- * out sync. We need to drop the ilock while we do this so we
- * don't deadlock when the buffer cache calls back to us.
- */
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
-
- zero_len = mp->m_sb.sb_blocksize - zero_offset;
- if (isize + zero_len > offset)
- zero_len = offset - isize;
- error = xfs_iozero(ip, isize, zero_len);
-
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- ASSERT(error >= 0);
- return error;
-}
-
-/*
- * Zero any on disk space between the current EOF and the new,
- * larger EOF. This handles the normal case of zeroing the remainder
- * of the last block in the file and the unusual case of zeroing blocks
- * out beyond the size of the file. This second case only happens
- * with fixed size extents and when the system crashes before the inode
- * size was updated but after blocks were allocated. If fill is set,
- * then any holes in the range are filled and zeroed. If not, the holes
- * are left alone as holes.
- */
-
-int /* error (positive) */
-xfs_zero_eof(
- xfs_inode_t *ip,
- xfs_off_t offset, /* starting I/O offset */
- xfs_fsize_t isize) /* current inode size */
-{
- xfs_mount_t *mp = ip->i_mount;
- xfs_fileoff_t start_zero_fsb;
- xfs_fileoff_t end_zero_fsb;
- xfs_fileoff_t zero_count_fsb;
- xfs_fileoff_t last_fsb;
- xfs_fileoff_t zero_off;
- xfs_fsize_t zero_len;
- int nimaps;
- int error = 0;
- xfs_bmbt_irec_t imap;
-
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
- ASSERT(offset > isize);
-
- /*
- * First handle zeroing the block on which isize resides.
- * We only zero a part of that block so it is handled specially.
- */
- error = xfs_zero_last_block(ip, offset, isize);
- if (error) {
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
- return error;
- }
-
- /*
- * Calculate the range between the new size and the old
- * where blocks needing to be zeroed may exist. To get the
- * block where the last byte in the file currently resides,
- * we need to subtract one from the size and truncate back
- * to a block boundary. We subtract 1 in case the size is
- * exactly on a block boundary.
- */
- last_fsb = isize ? XFS_B_TO_FSBT(mp, isize - 1) : (xfs_fileoff_t)-1;
- start_zero_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)isize);
- end_zero_fsb = XFS_B_TO_FSBT(mp, offset - 1);
- ASSERT((xfs_sfiloff_t)last_fsb < (xfs_sfiloff_t)start_zero_fsb);
- if (last_fsb == end_zero_fsb) {
- /*
- * The size was only incremented on its last block.
- * We took care of that above, so just return.
- */
- return 0;
- }
-
- ASSERT(start_zero_fsb <= end_zero_fsb);
- while (start_zero_fsb <= end_zero_fsb) {
- nimaps = 1;
- zero_count_fsb = end_zero_fsb - start_zero_fsb + 1;
- error = xfs_bmapi(NULL, ip, start_zero_fsb, zero_count_fsb,
- 0, NULL, 0, &imap, &nimaps, NULL, NULL);
- if (error) {
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
- return error;
- }
- ASSERT(nimaps > 0);
-
- if (imap.br_state == XFS_EXT_UNWRITTEN ||
- imap.br_startblock == HOLESTARTBLOCK) {
- /*
- * This loop handles initializing pages that were
- * partially initialized by the code below this
- * loop. It basically zeroes the part of the page
- * that sits on a hole and sets the page as P_HOLE
- * and calls remapf if it is a mapped file.
- */
- start_zero_fsb = imap.br_startoff + imap.br_blockcount;
- ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
- continue;
- }
-
- /*
- * There are blocks we need to zero.
- * Drop the inode lock while we're doing the I/O.
- * We'll still have the iolock to protect us.
- */
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
-
- zero_off = XFS_FSB_TO_B(mp, start_zero_fsb);
- zero_len = XFS_FSB_TO_B(mp, imap.br_blockcount);
-
- if ((zero_off + zero_len) > offset)
- zero_len = offset - zero_off;
-
- error = xfs_iozero(ip, zero_off, zero_len);
- if (error) {
- goto out_lock;
- }
-
- start_zero_fsb = imap.br_startoff + imap.br_blockcount;
- ASSERT(start_zero_fsb <= (end_zero_fsb + 1));
-
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- }
-
- return 0;
-
-out_lock:
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- ASSERT(error >= 0);
- return error;
-}
-
-ssize_t /* bytes written, or (-) error */
-xfs_write(
- struct xfs_inode *xip,
- struct kiocb *iocb,
- const struct iovec *iovp,
- unsigned int nsegs,
- loff_t *offset,
- int ioflags)
-{
- struct file *file = iocb->ki_filp;
- struct address_space *mapping = file->f_mapping;
- struct inode *inode = mapping->host;
- unsigned long segs = nsegs;
- xfs_mount_t *mp;
- ssize_t ret = 0, error = 0;
- xfs_fsize_t isize, new_size;
- int iolock;
- int eventsent = 0;
- size_t ocount = 0, count;
- loff_t pos;
- int need_i_mutex;
-
- XFS_STATS_INC(xs_write_calls);
-
- error = generic_segment_checks(iovp, &segs, &ocount, VERIFY_READ);
- if (error)
- return error;
-
- count = ocount;
- pos = *offset;
-
- if (count == 0)
- return 0;
-
- mp = xip->i_mount;
-
- xfs_wait_for_freeze(mp, SB_FREEZE_WRITE);
-
- if (XFS_FORCED_SHUTDOWN(mp))
- return -EIO;
-
-relock:
- if (ioflags & IO_ISDIRECT) {
- iolock = XFS_IOLOCK_SHARED;
- need_i_mutex = 0;
- } else {
- iolock = XFS_IOLOCK_EXCL;
- need_i_mutex = 1;
- mutex_lock(&inode->i_mutex);
- }
-
- xfs_ilock(xip, XFS_ILOCK_EXCL|iolock);
-
-start:
- error = -generic_write_checks(file, &pos, &count,
- S_ISBLK(inode->i_mode));
- if (error) {
- xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
- goto out_unlock_mutex;
- }
-
- if ((DM_EVENT_ENABLED(xip, DM_EVENT_WRITE) &&
- !(ioflags & IO_INVIS) && !eventsent)) {
- int dmflags = FILP_DELAY_FLAG(file);
-
- if (need_i_mutex)
- dmflags |= DM_FLAGS_IMUX;
-
- xfs_iunlock(xip, XFS_ILOCK_EXCL);
- error = XFS_SEND_DATA(xip->i_mount, DM_EVENT_WRITE, xip,
- pos, count, dmflags, &iolock);
- if (error) {
- goto out_unlock_internal;
- }
- xfs_ilock(xip, XFS_ILOCK_EXCL);
- eventsent = 1;
-
- /*
- * The iolock was dropped and reacquired in XFS_SEND_DATA
- * so we have to recheck the size when appending.
- * We will only "goto start;" once, since having sent the
- * event prevents another call to XFS_SEND_DATA, which is
- * what allows the size to change in the first place.
- */
- if ((file->f_flags & O_APPEND) && pos != xip->i_size)
- goto start;
- }
-
- if (ioflags & IO_ISDIRECT) {
- xfs_buftarg_t *target =
- XFS_IS_REALTIME_INODE(xip) ?
- mp->m_rtdev_targp : mp->m_ddev_targp;
-
- if ((pos & target->bt_smask) || (count & target->bt_smask)) {
- xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
- return XFS_ERROR(-EINVAL);
- }
-
- if (!need_i_mutex && (mapping->nrpages || pos > xip->i_size)) {
- xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
- iolock = XFS_IOLOCK_EXCL;
- need_i_mutex = 1;
- mutex_lock(&inode->i_mutex);
- xfs_ilock(xip, XFS_ILOCK_EXCL|iolock);
- goto start;
- }
- }
-
- new_size = pos + count;
- if (new_size > xip->i_size)
- xip->i_new_size = new_size;
-
- if (likely(!(ioflags & IO_INVIS)))
- file_update_time(file);
-
- /*
- * If the offset is beyond the size of the file, we have a couple
- * of things to do. First, if there is already space allocated
- * we need to either create holes or zero the disk or ...
- *
- * If there is a page where the previous size lands, we need
- * to zero it out up to the new size.
- */
-
- if (pos > xip->i_size) {
- error = xfs_zero_eof(xip, pos, xip->i_size);
- if (error) {
- xfs_iunlock(xip, XFS_ILOCK_EXCL);
- goto out_unlock_internal;
- }
- }
- xfs_iunlock(xip, XFS_ILOCK_EXCL);
-
- /*
- * If we're writing the file then make sure to clear the
- * setuid and setgid bits if the process is not being run
- * by root. This keeps people from modifying setuid and
- * setgid binaries.
- */
-
- if (((xip->i_d.di_mode & S_ISUID) ||
- ((xip->i_d.di_mode & (S_ISGID | S_IXGRP)) ==
- (S_ISGID | S_IXGRP))) &&
- !capable(CAP_FSETID)) {
- error = xfs_write_clear_setuid(xip);
- if (likely(!error))
- error = -file_remove_suid(file);
- if (unlikely(error)) {
- goto out_unlock_internal;
- }
- }
-
- /* We can write back this queue in page reclaim */
- current->backing_dev_info = mapping->backing_dev_info;
-
- if ((ioflags & IO_ISDIRECT)) {
- if (mapping->nrpages) {
- WARN_ON(need_i_mutex == 0);
- error = xfs_flushinval_pages(xip,
- (pos & PAGE_CACHE_MASK),
- -1, FI_REMAPF_LOCKED);
- if (error)
- goto out_unlock_internal;
- }
-
- if (need_i_mutex) {
- /* demote the lock now the cached pages are gone */
- xfs_ilock_demote(xip, XFS_IOLOCK_EXCL);
- mutex_unlock(&inode->i_mutex);
-
- iolock = XFS_IOLOCK_SHARED;
- need_i_mutex = 0;
- }
-
- trace_xfs_file_direct_write(xip, count, *offset, ioflags);
- ret = generic_file_direct_write(iocb, iovp,
- &segs, pos, offset, count, ocount);
-
- /*
- * direct-io write to a hole: fall through to buffered I/O
- * for completing the rest of the request.
- */
- if (ret >= 0 && ret != count) {
- XFS_STATS_ADD(xs_write_bytes, ret);
-
- pos += ret;
- count -= ret;
-
- ioflags &= ~IO_ISDIRECT;
- xfs_iunlock(xip, iolock);
- goto relock;
- }
- } else {
- int enospc = 0;
- ssize_t ret2 = 0;
-
-write_retry:
- trace_xfs_file_buffered_write(xip, count, *offset, ioflags);
- ret2 = generic_file_buffered_write(iocb, iovp, segs,
- pos, offset, count, ret);
- /*
- * if we just got an ENOSPC, flush the inode now we
- * aren't holding any page locks and retry *once*
- */
- if (ret2 == -ENOSPC && !enospc) {
- error = xfs_flush_pages(xip, 0, -1, 0, FI_NONE);
- if (error)
- goto out_unlock_internal;
- enospc = 1;
- goto write_retry;
- }
- ret = ret2;
- }
-
- current->backing_dev_info = NULL;
-
- 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);
- if (need_i_mutex)
- mutex_unlock(&inode->i_mutex);
- error = XFS_SEND_NAMESP(xip->i_mount, DM_EVENT_NOSPACE, xip,
- DM_RIGHT_NULL, xip, DM_RIGHT_NULL, NULL, NULL,
- 0, 0, 0); /* Delay flag intentionally unused */
- if (need_i_mutex)
- mutex_lock(&inode->i_mutex);
- xfs_ilock(xip, iolock);
- if (error)
- goto out_unlock_internal;
- goto start;
- }
-
- error = -ret;
- if (ret <= 0)
- goto out_unlock_internal;
-
- XFS_STATS_ADD(xs_write_bytes, ret);
-
- /* Handle various SYNC-type writes */
- if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
- loff_t end = pos + ret - 1;
- int error2;
-
- xfs_iunlock(xip, iolock);
- if (need_i_mutex)
- mutex_unlock(&inode->i_mutex);
-
- error2 = filemap_write_and_wait_range(mapping, pos, end);
- if (!error)
- error = error2;
- if (need_i_mutex)
- mutex_lock(&inode->i_mutex);
- xfs_ilock(xip, iolock);
-
- error2 = xfs_fsync(xip);
- if (!error)
- error = error2;
- }
-
- out_unlock_internal:
- if (xip->i_new_size) {
- xfs_ilock(xip, XFS_ILOCK_EXCL);
- xip->i_new_size = 0;
- /*
- * If this was a direct or synchronous I/O that failed (such
- * as ENOSPC) then part of the I/O may have been written to
- * disk before the error occured. In this case the on-disk
- * file size may have been adjusted beyond the in-memory file
- * size and now needs to be truncated back.
- */
- if (xip->i_d.di_size > xip->i_size)
- xip->i_d.di_size = xip->i_size;
- xfs_iunlock(xip, XFS_ILOCK_EXCL);
- }
- xfs_iunlock(xip, iolock);
- out_unlock_mutex:
- if (need_i_mutex)
- mutex_unlock(&inode->i_mutex);
- return -error;
-}
-
-/*
- * All xfs metadata buffers except log state machine buffers
- * get this attached as their b_bdstrat callback function.
- * This is so that we can catch a buffer
- * after prematurely unpinning it to forcibly shutdown the filesystem.
- */
-int
-xfs_bdstrat_cb(struct xfs_buf *bp)
-{
- if (XFS_FORCED_SHUTDOWN(bp->b_mount)) {
- trace_xfs_bdstrat_shut(bp, _RET_IP_);
- /*
- * Metadata write that didn't get logged but
- * written delayed anyway. These aren't associated
- * with a transaction, and can be ignored.
- */
- if (XFS_BUF_IODONE_FUNC(bp) == NULL &&
- (XFS_BUF_ISREAD(bp)) == 0)
- return (xfs_bioerror_relse(bp));
- else
- return (xfs_bioerror(bp));
- }
-
- xfs_buf_iorequest(bp);
- return 0;
-}
-
-/*
- * Wrapper around bdstrat so that we can stop data from going to disk in case
- * we are shutting down the filesystem. Typically user data goes thru this
- * path; one of the exceptions is the superblock.
- */
-void
-xfsbdstrat(
- struct xfs_mount *mp,
- struct xfs_buf *bp)
-{
- ASSERT(mp);
- if (!XFS_FORCED_SHUTDOWN(mp)) {
- xfs_buf_iorequest(bp);
- return;
- }
-
- trace_xfs_bdstrat_shut(bp, _RET_IP_);
- xfs_bioerror_relse(bp);
-}
-
-/*
- * If the underlying (data/log/rt) device is readonly, there are some
- * operations that cannot proceed.
- */
-int
-xfs_dev_is_read_only(
- xfs_mount_t *mp,
- char *message)
-{
- if (xfs_readonly_buftarg(mp->m_ddev_targp) ||
- xfs_readonly_buftarg(mp->m_logdev_targp) ||
- (mp->m_rtdev_targp && xfs_readonly_buftarg(mp->m_rtdev_targp))) {
- cmn_err(CE_NOTE,
- "XFS: %s required on read-only device.", message);
- cmn_err(CE_NOTE,
- "XFS: write access unavailable, cannot proceed.");
- return EROFS;
- }
- return 0;
-}
diff --git a/fs/xfs/linux-2.6/xfs_lrw.h b/fs/xfs/linux-2.6/xfs_lrw.h
deleted file mode 100644
index d1f7789c7ffb..000000000000
--- a/fs/xfs/linux-2.6/xfs_lrw.h
+++ /dev/null
@@ -1,32 +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
- */
-#ifndef __XFS_LRW_H__
-#define __XFS_LRW_H__
-
-struct xfs_mount;
-struct xfs_inode;
-struct xfs_buf;
-
-/* errors from xfsbdstrat() must be extracted from the buffer */
-extern void xfsbdstrat(struct xfs_mount *, struct xfs_buf *);
-extern int xfs_bdstrat_cb(struct xfs_buf *);
-extern int xfs_dev_is_read_only(struct xfs_mount *, char *);
-
-extern int xfs_zero_eof(struct xfs_inode *, xfs_off_t, xfs_fsize_t);
-
-#endif /* __XFS_LRW_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_quotaops.c b/fs/xfs/linux-2.6/xfs_quotaops.c
index 3d4a0c84d634..067cafbfc635 100644
--- a/fs/xfs/linux-2.6/xfs_quotaops.c
+++ b/fs/xfs/linux-2.6/xfs_quotaops.c
@@ -19,10 +19,10 @@
#include "xfs_dmapi.h"
#include "xfs_sb.h"
#include "xfs_inum.h"
+#include "xfs_log.h"
#include "xfs_ag.h"
#include "xfs_mount.h"
#include "xfs_quota.h"
-#include "xfs_log.h"
#include "xfs_trans.h"
#include "xfs_bmap_btree.h"
#include "xfs_inode.h"
@@ -44,20 +44,6 @@ xfs_quota_type(int type)
}
STATIC int
-xfs_fs_quota_sync(
- struct super_block *sb,
- int type)
-{
- struct xfs_mount *mp = XFS_M(sb);
-
- if (sb->s_flags & MS_RDONLY)
- return -EROFS;
- if (!XFS_IS_QUOTA_RUNNING(mp))
- return -ENOSYS;
- return -xfs_sync_data(mp, 0);
-}
-
-STATIC int
xfs_fs_get_xstate(
struct super_block *sb,
struct fs_quota_stat *fqs)
@@ -82,8 +68,6 @@ xfs_fs_set_xstate(
return -EROFS;
if (op != Q_XQUOTARM && !XFS_IS_QUOTA_RUNNING(mp))
return -ENOSYS;
- if (!capable(CAP_SYS_ADMIN))
- return -EPERM;
if (uflags & XFS_QUOTA_UDQ_ACCT)
flags |= XFS_UQUOTA_ACCT;
@@ -113,7 +97,7 @@ xfs_fs_set_xstate(
}
STATIC int
-xfs_fs_get_xquota(
+xfs_fs_get_dqblk(
struct super_block *sb,
int type,
qid_t id,
@@ -130,7 +114,7 @@ xfs_fs_get_xquota(
}
STATIC int
-xfs_fs_set_xquota(
+xfs_fs_set_dqblk(
struct super_block *sb,
int type,
qid_t id,
@@ -144,16 +128,13 @@ xfs_fs_set_xquota(
return -ENOSYS;
if (!XFS_IS_QUOTA_ON(mp))
return -ESRCH;
- if (!capable(CAP_SYS_ADMIN))
- return -EPERM;
return -xfs_qm_scall_setqlim(mp, id, xfs_quota_type(type), fdq);
}
const struct quotactl_ops xfs_quotactl_operations = {
- .quota_sync = xfs_fs_quota_sync,
.get_xstate = xfs_fs_get_xstate,
.set_xstate = xfs_fs_set_xstate,
- .get_xquota = xfs_fs_get_xquota,
- .set_xquota = xfs_fs_set_xquota,
+ .get_dqblk = xfs_fs_get_dqblk,
+ .set_dqblk = xfs_fs_set_dqblk,
};
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 77414db10dc2..80938c736c27 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -61,6 +61,7 @@
#include <linux/namei.h>
#include <linux/init.h>
+#include <linux/slab.h>
#include <linux/mount.h>
#include <linux/mempool.h>
#include <linux/writeback.h>
@@ -118,6 +119,8 @@ mempool_t *xfs_ioend_pool;
#define MNTOPT_DMAPI "dmapi" /* DMI enabled (DMAPI / XDSM) */
#define MNTOPT_XDSM "xdsm" /* DMI enabled (DMAPI / XDSM) */
#define MNTOPT_DMI "dmi" /* DMI enabled (DMAPI / XDSM) */
+#define MNTOPT_DELAYLOG "delaylog" /* Delayed loging enabled */
+#define MNTOPT_NODELAYLOG "nodelaylog" /* Delayed loging disabled */
/*
* Table driven mount option parser.
@@ -373,6 +376,13 @@ xfs_parseargs(
mp->m_flags |= XFS_MOUNT_DMAPI;
} else if (!strcmp(this_char, MNTOPT_DMI)) {
mp->m_flags |= XFS_MOUNT_DMAPI;
+ } else if (!strcmp(this_char, MNTOPT_DELAYLOG)) {
+ mp->m_flags |= XFS_MOUNT_DELAYLOG;
+ cmn_err(CE_WARN,
+ "Enabling EXPERIMENTAL delayed logging feature "
+ "- use at your own risk.\n");
+ } else if (!strcmp(this_char, MNTOPT_NODELAYLOG)) {
+ mp->m_flags &= ~XFS_MOUNT_DELAYLOG;
} else if (!strcmp(this_char, "ihashsize")) {
cmn_err(CE_WARN,
"XFS: ihashsize no longer used, option is deprecated.");
@@ -534,6 +544,7 @@ xfs_showargs(
{ XFS_MOUNT_FILESTREAMS, "," MNTOPT_FILESTREAM },
{ XFS_MOUNT_DMAPI, "," MNTOPT_DMAPI },
{ XFS_MOUNT_GRPID, "," MNTOPT_GRPID },
+ { XFS_MOUNT_DELAYLOG, "," MNTOPT_DELAYLOG },
{ 0, NULL }
};
static struct proc_xfs_info xfs_info_unset[] = {
@@ -724,7 +735,8 @@ void
xfs_blkdev_issue_flush(
xfs_buftarg_t *buftarg)
{
- blkdev_issue_flush(buftarg->bt_bdev, NULL);
+ blkdev_issue_flush(buftarg->bt_bdev, GFP_KERNEL, NULL,
+ BLKDEV_IFL_WAIT);
}
STATIC void
@@ -788,18 +800,18 @@ xfs_open_devices(
* Setup xfs_mount buffer target pointers
*/
error = ENOMEM;
- mp->m_ddev_targp = xfs_alloc_buftarg(ddev, 0);
+ mp->m_ddev_targp = xfs_alloc_buftarg(ddev, 0, mp->m_fsname);
if (!mp->m_ddev_targp)
goto out_close_rtdev;
if (rtdev) {
- mp->m_rtdev_targp = xfs_alloc_buftarg(rtdev, 1);
+ mp->m_rtdev_targp = xfs_alloc_buftarg(rtdev, 1, mp->m_fsname);
if (!mp->m_rtdev_targp)
goto out_free_ddev_targ;
}
if (logdev && logdev != ddev) {
- mp->m_logdev_targp = xfs_alloc_buftarg(logdev, 1);
+ mp->m_logdev_targp = xfs_alloc_buftarg(logdev, 1, mp->m_fsname);
if (!mp->m_logdev_targp)
goto out_free_rtdev_targ;
} else {
@@ -877,12 +889,11 @@ xfsaild(
{
struct xfs_ail *ailp = data;
xfs_lsn_t last_pushed_lsn = 0;
- long tout = 0;
+ long tout = 0; /* milliseconds */
while (!kthread_should_stop()) {
- if (tout)
- schedule_timeout_interruptible(msecs_to_jiffies(tout));
- tout = 1000;
+ schedule_timeout_interruptible(tout ?
+ msecs_to_jiffies(tout) : MAX_SCHEDULE_TIMEOUT);
/* swsusp */
try_to_freeze();
@@ -902,7 +913,8 @@ xfsaild_start(
struct xfs_ail *ailp)
{
ailp->xa_target = 0;
- ailp->xa_task = kthread_run(xfsaild, ailp, "xfsaild");
+ ailp->xa_task = kthread_run(xfsaild, ailp, "xfsaild/%s",
+ ailp->xa_mount->m_fsname);
if (IS_ERR(ailp->xa_task))
return -PTR_ERR(ailp->xa_task);
return 0;
@@ -1022,59 +1034,109 @@ xfs_fs_dirty_inode(
XFS_I(inode)->i_update_core = 1;
}
-/*
- * Attempt to flush the inode, this will actually fail
- * if the inode is pinned, but we dirty the inode again
- * at the point when it is unpinned after a log write,
- * since this is when the inode itself becomes flushable.
- */
+STATIC int
+xfs_log_inode(
+ struct xfs_inode *ip)
+{
+ struct xfs_mount *mp = ip->i_mount;
+ struct xfs_trans *tp;
+ int error;
+
+ xfs_iunlock(ip, XFS_ILOCK_SHARED);
+ tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS);
+ error = xfs_trans_reserve(tp, 0, XFS_FSYNC_TS_LOG_RES(mp), 0, 0, 0);
+
+ if (error) {
+ xfs_trans_cancel(tp, 0);
+ /* we need to return with the lock hold shared */
+ xfs_ilock(ip, XFS_ILOCK_SHARED);
+ return error;
+ }
+
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+
+ /*
+ * Note - it's possible that we might have pushed ourselves out of the
+ * way during trans_reserve which would flush the inode. But there's
+ * no guarantee that the inode buffer has actually gone out yet (it's
+ * delwri). Plus the buffer could be pinned anyway if it's part of
+ * an inode in another recent transaction. So we play it safe and
+ * fire off the transaction anyway.
+ */
+ xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
+ xfs_trans_ihold(tp, ip);
+ xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+ xfs_trans_set_sync(tp);
+ error = xfs_trans_commit(tp, 0);
+ xfs_ilock_demote(ip, XFS_ILOCK_EXCL);
+
+ return error;
+}
+
STATIC int
xfs_fs_write_inode(
struct inode *inode,
- int sync)
+ struct writeback_control *wbc)
{
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
- int error = 0;
+ int error = EAGAIN;
xfs_itrace_entry(ip);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
- if (sync) {
- error = xfs_wait_on_pages(ip, 0, -1);
- if (error)
+ if (wbc->sync_mode == WB_SYNC_ALL) {
+ /*
+ * Make sure the inode has hit stable storage. By using the
+ * log and the fsync transactions we reduce the IOs we have
+ * to do here from two (log and inode) to just the log.
+ *
+ * Note: We still need to do a delwri write of the inode after
+ * this to flush it to the backing buffer so that bulkstat
+ * works properly if this is the first time the inode has been
+ * written. Because we hold the ilock atomically over the
+ * transaction commit and the inode flush we are guaranteed
+ * that the inode is not pinned when it returns. If the flush
+ * lock is already held, then the inode has already been
+ * flushed once and we don't need to flush it again. Hence
+ * the code will only flush the inode if it isn't already
+ * being flushed.
+ */
+ xfs_ioend_wait(ip);
+ xfs_ilock(ip, XFS_ILOCK_SHARED);
+ if (ip->i_update_core) {
+ error = xfs_log_inode(ip);
+ if (error)
+ goto out_unlock;
+ }
+ } else {
+ /*
+ * We make this non-blocking if the inode is contended, return
+ * EAGAIN to indicate to the caller that they did not succeed.
+ * This prevents the flush path from blocking on inodes inside
+ * another operation right now, they get caught later by xfs_sync.
+ */
+ if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED))
goto out;
}
- /*
- * Bypass inodes which have already been cleaned by
- * the inode flush clustering code inside xfs_iflush
- */
- if (xfs_inode_clean(ip))
- goto out;
+ if (xfs_ipincount(ip) || !xfs_iflock_nowait(ip))
+ goto out_unlock;
/*
- * We make this non-blocking if the inode is contended, return
- * EAGAIN to indicate to the caller that they did not succeed.
- * This prevents the flush path from blocking on inodes inside
- * another operation right now, they get caught later by xfs_sync.
+ * Now we have the flush lock and the inode is not pinned, we can check
+ * if the inode is really clean as we know that there are no pending
+ * transaction completions, it is not waiting on the delayed write
+ * queue and there is no IO in progress.
*/
- if (sync) {
- xfs_ilock(ip, XFS_ILOCK_SHARED);
- xfs_iflock(ip);
-
- error = xfs_iflush(ip, XFS_IFLUSH_SYNC);
- } else {
- error = EAGAIN;
- if (!xfs_ilock_nowait(ip, XFS_ILOCK_SHARED))
- goto out;
- if (xfs_ipincount(ip) || !xfs_iflock_nowait(ip))
- goto out_unlock;
-
- error = xfs_iflush(ip, XFS_IFLUSH_ASYNC_NOBLOCK);
+ if (xfs_inode_clean(ip)) {
+ xfs_ifunlock(ip);
+ error = 0;
+ goto out_unlock;
}
+ error = xfs_iflush(ip, 0);
out_unlock:
xfs_iunlock(ip, XFS_ILOCK_SHARED);
@@ -1160,6 +1222,7 @@ xfs_fs_put_super(
xfs_unmountfs(mp);
xfs_freesb(mp);
+ xfs_inode_shrinker_unregister(mp);
xfs_icsb_destroy_counters(mp);
xfs_close_devices(mp);
xfs_dmops_put(mp);
@@ -1257,6 +1320,29 @@ xfs_fs_statfs(
return 0;
}
+STATIC void
+xfs_save_resvblks(struct xfs_mount *mp)
+{
+ __uint64_t resblks = 0;
+
+ mp->m_resblks_save = mp->m_resblks;
+ xfs_reserve_blocks(mp, &resblks, NULL);
+}
+
+STATIC void
+xfs_restore_resvblks(struct xfs_mount *mp)
+{
+ __uint64_t resblks;
+
+ if (mp->m_resblks_save) {
+ resblks = mp->m_resblks_save;
+ mp->m_resblks_save = 0;
+ } else
+ resblks = xfs_default_resblks(mp);
+
+ xfs_reserve_blocks(mp, &resblks, NULL);
+}
+
STATIC int
xfs_fs_remount(
struct super_block *sb,
@@ -1336,11 +1422,27 @@ xfs_fs_remount(
}
mp->m_update_flags = 0;
}
+
+ /*
+ * Fill out the reserve pool if it is empty. Use the stashed
+ * value if it is non-zero, otherwise go with the default.
+ */
+ xfs_restore_resvblks(mp);
}
/* rw -> ro */
if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (*flags & MS_RDONLY)) {
+ /*
+ * After we have synced the data but before we sync the
+ * metadata, we need to free up the reserve block pool so that
+ * the used block count in the superblock on disk is correct at
+ * the end of the remount. Stash the current reserve pool size
+ * so that if we get remounted rw, we can return it to the same
+ * size.
+ */
+
xfs_quiesce_data(mp);
+ xfs_save_resvblks(mp);
xfs_quiesce_attr(mp);
mp->m_flags |= XFS_MOUNT_RDONLY;
}
@@ -1359,11 +1461,22 @@ xfs_fs_freeze(
{
struct xfs_mount *mp = XFS_M(sb);
+ xfs_save_resvblks(mp);
xfs_quiesce_attr(mp);
return -xfs_fs_log_dummy(mp);
}
STATIC int
+xfs_fs_unfreeze(
+ struct super_block *sb)
+{
+ struct xfs_mount *mp = XFS_M(sb);
+
+ xfs_restore_resvblks(mp);
+ return 0;
+}
+
+STATIC int
xfs_fs_show_options(
struct seq_file *m,
struct vfsmount *mnt)
@@ -1523,6 +1636,8 @@ xfs_fs_fill_super(
if (error)
goto fail_vnrele;
+ xfs_inode_shrinker_register(mp);
+
kfree(mtpt);
return 0;
@@ -1585,6 +1700,7 @@ static const struct super_operations xfs_super_operations = {
.put_super = xfs_fs_put_super,
.sync_fs = xfs_fs_sync_fs,
.freeze_fs = xfs_fs_freeze,
+ .unfreeze_fs = xfs_fs_unfreeze,
.statfs = xfs_fs_statfs,
.remount_fs = xfs_fs_remount,
.show_options = xfs_fs_show_options,
@@ -1649,7 +1765,7 @@ xfs_init_zones(void)
* but it is much faster.
*/
xfs_buf_item_zone = kmem_zone_init((sizeof(xfs_buf_log_item_t) +
- (((XFS_MAX_BLOCKSIZE / XFS_BLI_CHUNK) /
+ (((XFS_MAX_BLOCKSIZE / XFS_BLF_CHUNK) /
NBWORD) * sizeof(int))), "xfs_buf_item");
if (!xfs_buf_item_zone)
goto out_destroy_trans_zone;
diff --git a/fs/xfs/linux-2.6/xfs_super.h b/fs/xfs/linux-2.6/xfs_super.h
index 233d4b9881b1..519618e9279e 100644
--- a/fs/xfs/linux-2.6/xfs_super.h
+++ b/fs/xfs/linux-2.6/xfs_super.h
@@ -85,7 +85,7 @@ extern __uint64_t xfs_max_file_offset(unsigned int);
extern void xfs_blkdev_issue_flush(struct xfs_buftarg *);
extern const struct export_operations xfs_export_operations;
-extern struct xattr_handler *xfs_xattr_handlers[];
+extern const struct xattr_handler *xfs_xattr_handlers[];
extern const struct quotactl_ops xfs_quotactl_operations;
#define XFS_M(sb) ((struct xfs_mount *)((sb)->s_fs_info))
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index 1f5e4bb5e970..a51a07c3a70c 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -90,14 +90,14 @@ xfs_inode_ag_lookup(
STATIC int
xfs_inode_ag_walk(
struct xfs_mount *mp,
- xfs_agnumber_t ag,
+ struct xfs_perag *pag,
int (*execute)(struct xfs_inode *ip,
struct xfs_perag *pag, int flags),
int flags,
int tag,
- int exclusive)
+ int exclusive,
+ int *nr_to_scan)
{
- struct xfs_perag *pag = &mp->m_perag[ag];
uint32_t first_index;
int last_error = 0;
int skipped;
@@ -135,17 +135,50 @@ restart:
if (error == EFSCORRUPTED)
break;
- } while (1);
+ } while ((*nr_to_scan)--);
if (skipped) {
delay(1);
goto restart;
}
-
- xfs_put_perag(mp, pag);
return last_error;
}
+/*
+ * Select the next per-ag structure to iterate during the walk. The reclaim
+ * walk is optimised only to walk AGs with reclaimable inodes in them.
+ */
+static struct xfs_perag *
+xfs_inode_ag_iter_next_pag(
+ struct xfs_mount *mp,
+ xfs_agnumber_t *first,
+ int tag)
+{
+ struct xfs_perag *pag = NULL;
+
+ if (tag == XFS_ICI_RECLAIM_TAG) {
+ int found;
+ int ref;
+
+ spin_lock(&mp->m_perag_lock);
+ found = radix_tree_gang_lookup_tag(&mp->m_perag_tree,
+ (void **)&pag, *first, 1, tag);
+ if (found <= 0) {
+ spin_unlock(&mp->m_perag_lock);
+ return NULL;
+ }
+ *first = pag->pag_agno + 1;
+ /* open coded pag reference increment */
+ ref = atomic_inc_return(&pag->pag_ref);
+ spin_unlock(&mp->m_perag_lock);
+ trace_xfs_perag_get_reclaim(mp, pag->pag_agno, ref, _RET_IP_);
+ } else {
+ pag = xfs_perag_get(mp, *first);
+ (*first)++;
+ }
+ return pag;
+}
+
int
xfs_inode_ag_iterator(
struct xfs_mount *mp,
@@ -153,23 +186,31 @@ xfs_inode_ag_iterator(
struct xfs_perag *pag, int flags),
int flags,
int tag,
- int exclusive)
+ int exclusive,
+ int *nr_to_scan)
{
+ struct xfs_perag *pag;
int error = 0;
int last_error = 0;
xfs_agnumber_t ag;
-
- for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) {
- if (!mp->m_perag[ag].pag_ici_init)
- continue;
- error = xfs_inode_ag_walk(mp, ag, execute, flags, tag,
- exclusive);
+ int nr;
+
+ nr = nr_to_scan ? *nr_to_scan : INT_MAX;
+ ag = 0;
+ while ((pag = xfs_inode_ag_iter_next_pag(mp, &ag, tag))) {
+ error = xfs_inode_ag_walk(mp, pag, execute, flags, tag,
+ exclusive, &nr);
+ xfs_perag_put(pag);
if (error) {
last_error = error;
if (error == EFSCORRUPTED)
break;
}
+ if (nr <= 0)
+ break;
}
+ if (nr_to_scan)
+ *nr_to_scan = nr;
return XFS_ERROR(last_error);
}
@@ -231,7 +272,7 @@ xfs_sync_inode_data(
}
error = xfs_flush_pages(ip, 0, -1, (flags & SYNC_WAIT) ?
- 0 : XFS_B_ASYNC, FI_NONE);
+ 0 : XBF_ASYNC, FI_NONE);
xfs_iunlock(ip, XFS_IOLOCK_SHARED);
out_wait:
@@ -267,8 +308,7 @@ xfs_sync_inode_attr(
goto out_unlock;
}
- error = xfs_iflush(ip, (flags & SYNC_WAIT) ?
- XFS_IFLUSH_SYNC : XFS_IFLUSH_DELWRI);
+ error = xfs_iflush(ip, flags);
out_unlock:
xfs_iunlock(ip, XFS_ILOCK_SHARED);
@@ -289,14 +329,11 @@ xfs_sync_data(
ASSERT((flags & ~(SYNC_TRYLOCK|SYNC_WAIT)) == 0);
error = xfs_inode_ag_iterator(mp, xfs_sync_inode_data, flags,
- XFS_ICI_NO_TAG, 0);
+ XFS_ICI_NO_TAG, 0, NULL);
if (error)
return XFS_ERROR(error);
- xfs_log_force(mp, 0,
- (flags & SYNC_WAIT) ?
- XFS_LOG_FORCE | XFS_LOG_SYNC :
- XFS_LOG_FORCE);
+ xfs_log_force(mp, (flags & SYNC_WAIT) ? XFS_LOG_SYNC : 0);
return 0;
}
@@ -311,7 +348,7 @@ xfs_sync_attr(
ASSERT((flags & ~SYNC_WAIT) == 0);
return xfs_inode_ag_iterator(mp, xfs_sync_inode_attr, flags,
- XFS_ICI_NO_TAG, 0);
+ XFS_ICI_NO_TAG, 0, NULL);
}
STATIC int
@@ -322,10 +359,6 @@ xfs_commit_dummy_trans(
struct xfs_inode *ip = mp->m_rootip;
struct xfs_trans *tp;
int error;
- int log_flags = XFS_LOG_FORCE;
-
- if (flags & SYNC_WAIT)
- log_flags |= XFS_LOG_SYNC;
/*
* Put a dummy transaction in the log to tell recovery
@@ -347,74 +380,29 @@ xfs_commit_dummy_trans(
xfs_iunlock(ip, XFS_ILOCK_EXCL);
/* the log force ensures this transaction is pushed to disk */
- xfs_log_force(mp, 0, log_flags);
+ xfs_log_force(mp, (flags & SYNC_WAIT) ? XFS_LOG_SYNC : 0);
return error;
}
-int
+STATIC int
xfs_sync_fsdata(
- struct xfs_mount *mp,
- int flags)
+ struct xfs_mount *mp)
{
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 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 (flags & SYNC_TRYLOCK) {
- 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);
+ bp = xfs_getsb(mp, 0);
+ if (XFS_BUF_ISPINNED(bp))
+ xfs_log_force(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);
-
- error = xfs_bwrite(mp, bp);
- if (error)
- return error;
-
- /*
- * If this is a data integrity sync make sure all pending buffers
- * are flushed out for the log coverage check below.
- */
- if (flags & SYNC_WAIT)
- xfs_flush_buftarg(mp->m_ddev_targp, 1);
-
- if (xfs_log_need_covered(mp))
- error = xfs_commit_dummy_trans(mp, flags);
- return error;
-
- out_brelse:
- xfs_buf_relse(bp);
- out:
- return error;
+ return xfs_bwrite(mp, bp);
}
/*
@@ -438,7 +426,7 @@ int
xfs_quiesce_data(
struct xfs_mount *mp)
{
- int error;
+ int error, error2 = 0;
/* push non-blocking */
xfs_sync_data(mp, 0);
@@ -448,17 +436,21 @@ xfs_quiesce_data(
xfs_sync_data(mp, SYNC_WAIT);
xfs_qm_sync(mp, SYNC_WAIT);
- /* drop inode references pinned by filestreams */
- xfs_filestream_flush(mp);
-
/* write superblock and hoover up shutdown errors */
- error = xfs_sync_fsdata(mp, SYNC_WAIT);
+ error = xfs_sync_fsdata(mp);
+
+ /* make sure all delwri buffers are written out */
+ xfs_flush_buftarg(mp->m_ddev_targp, 1);
+
+ /* mark the log as covered if needed */
+ if (xfs_log_need_covered(mp))
+ error2 = xfs_commit_dummy_trans(mp, SYNC_WAIT);
/* flush data-only devices */
if (mp->m_rtdev_targp)
XFS_bflush(mp->m_rtdev_targp);
- return error;
+ return error ? error : error2;
}
STATIC void
@@ -467,16 +459,18 @@ xfs_quiesce_fs(
{
int count = 0, pincount;
+ xfs_reclaim_inodes(mp, 0);
xfs_flush_buftarg(mp->m_ddev_targp, 0);
- xfs_reclaim_inodes(mp, 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.
+ * logged before we can write the unmount record. We also so sync
+ * reclaim of inodes to catch any that the above delwri flush skipped.
*/
do {
+ xfs_reclaim_inodes(mp, SYNC_WAIT);
xfs_sync_attr(mp, SYNC_WAIT);
pincount = xfs_flush_buftarg(mp->m_ddev_targp, 1);
if (!pincount) {
@@ -575,13 +569,13 @@ xfs_flush_inodes(
igrab(inode);
xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_inodes_work, &completion);
wait_for_completion(&completion);
- xfs_log_force(ip->i_mount, (xfs_lsn_t)0, XFS_LOG_FORCE|XFS_LOG_SYNC);
+ xfs_log_force(ip->i_mount, 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.
+ * Every sync period we need to unpin all items, reclaim inodes and sync
+ * disk quotas. We might need to cover the log to indicate that the
+ * filesystem is idle.
*/
STATIC void
xfs_sync_worker(
@@ -591,11 +585,12 @@ xfs_sync_worker(
int error;
if (!(mp->m_flags & XFS_MOUNT_RDONLY)) {
- xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE);
- xfs_reclaim_inodes(mp, XFS_IFLUSH_DELWRI_ELSE_ASYNC);
+ xfs_log_force(mp, 0);
+ xfs_reclaim_inodes(mp, 0);
/* dgc: errors ignored here */
error = xfs_qm_sync(mp, SYNC_TRYLOCK);
- error = xfs_sync_fsdata(mp, SYNC_TRYLOCK);
+ if (xfs_log_need_covered(mp))
+ error = xfs_commit_dummy_trans(mp, 0);
}
mp->m_sync_seq++;
wake_up(&mp->m_wait_single_sync_task);
@@ -613,7 +608,8 @@ xfssyncd(
set_freezable();
timeleft = xfs_syncd_centisecs * msecs_to_jiffies(10);
for (;;) {
- timeleft = schedule_timeout_interruptible(timeleft);
+ if (list_empty(&mp->m_sync_list))
+ timeleft = schedule_timeout_interruptible(timeleft);
/* swsusp */
try_to_freeze();
if (kthread_should_stop() && list_empty(&mp->m_sync_list))
@@ -633,8 +629,7 @@ xfssyncd(
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);
+ list_splice_init(&mp->m_sync_list, &tmp);
spin_unlock(&mp->m_sync_lock);
list_for_each_entry_safe(work, n, &tmp, w_list) {
@@ -658,7 +653,7 @@ xfs_syncd_init(
mp->m_sync_work.w_syncer = xfs_sync_worker;
mp->m_sync_work.w_mount = mp;
mp->m_sync_work.w_completion = NULL;
- mp->m_sync_task = kthread_run(xfssyncd, mp, "xfssyncd");
+ mp->m_sync_task = kthread_run(xfssyncd, mp, "xfssyncd/%s", mp->m_fsname);
if (IS_ERR(mp->m_sync_task))
return -PTR_ERR(mp->m_sync_task);
return 0;
@@ -679,6 +674,18 @@ __xfs_inode_set_reclaim_tag(
radix_tree_tag_set(&pag->pag_ici_root,
XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino),
XFS_ICI_RECLAIM_TAG);
+
+ if (!pag->pag_ici_reclaimable) {
+ /* propagate the reclaim tag up into the perag radix tree */
+ spin_lock(&ip->i_mount->m_perag_lock);
+ radix_tree_tag_set(&ip->i_mount->m_perag_tree,
+ XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino),
+ XFS_ICI_RECLAIM_TAG);
+ spin_unlock(&ip->i_mount->m_perag_lock);
+ trace_xfs_perag_set_reclaim(ip->i_mount, pag->pag_agno,
+ -1, _RET_IP_);
+ }
+ pag->pag_ici_reclaimable++;
}
/*
@@ -690,16 +697,17 @@ 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);
+ struct xfs_mount *mp = ip->i_mount;
+ struct xfs_perag *pag;
- read_lock(&pag->pag_ici_lock);
+ pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino));
+ write_lock(&pag->pag_ici_lock);
spin_lock(&ip->i_flags_lock);
__xfs_inode_set_reclaim_tag(pag, ip);
__xfs_iflags_set(ip, XFS_IRECLAIMABLE);
spin_unlock(&ip->i_flags_lock);
- read_unlock(&pag->pag_ici_lock);
- xfs_put_perag(mp, pag);
+ write_unlock(&pag->pag_ici_lock);
+ xfs_perag_put(pag);
}
void
@@ -710,14 +718,77 @@ __xfs_inode_clear_reclaim_tag(
{
radix_tree_tag_clear(&pag->pag_ici_root,
XFS_INO_TO_AGINO(mp, ip->i_ino), XFS_ICI_RECLAIM_TAG);
+ pag->pag_ici_reclaimable--;
+ if (!pag->pag_ici_reclaimable) {
+ /* clear the reclaim tag from the perag radix tree */
+ spin_lock(&ip->i_mount->m_perag_lock);
+ radix_tree_tag_clear(&ip->i_mount->m_perag_tree,
+ XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino),
+ XFS_ICI_RECLAIM_TAG);
+ spin_unlock(&ip->i_mount->m_perag_lock);
+ trace_xfs_perag_clear_reclaim(ip->i_mount, pag->pag_agno,
+ -1, _RET_IP_);
+ }
}
+/*
+ * Inodes in different states need to be treated differently, and the return
+ * value of xfs_iflush is not sufficient to get this right. The following table
+ * lists the inode states and the reclaim actions necessary for non-blocking
+ * reclaim:
+ *
+ *
+ * inode state iflush ret required action
+ * --------------- ---------- ---------------
+ * bad - reclaim
+ * shutdown EIO unpin and reclaim
+ * clean, unpinned 0 reclaim
+ * stale, unpinned 0 reclaim
+ * clean, pinned(*) 0 requeue
+ * stale, pinned EAGAIN requeue
+ * dirty, delwri ok 0 requeue
+ * dirty, delwri blocked EAGAIN requeue
+ * dirty, sync flush 0 reclaim
+ *
+ * (*) dgc: I don't think the clean, pinned state is possible but it gets
+ * handled anyway given the order of checks implemented.
+ *
+ * As can be seen from the table, the return value of xfs_iflush() is not
+ * sufficient to correctly decide the reclaim action here. The checks in
+ * xfs_iflush() might look like duplicates, but they are not.
+ *
+ * Also, because we get the flush lock first, we know that any inode that has
+ * been flushed delwri has had the flush completed by the time we check that
+ * the inode is clean. The clean inode check needs to be done before flushing
+ * the inode delwri otherwise we would loop forever requeuing clean inodes as
+ * we cannot tell apart a successful delwri flush and a clean inode from the
+ * return value of xfs_iflush().
+ *
+ * Note that because the inode is flushed delayed write by background
+ * writeback, the flush lock may already be held here and waiting on it can
+ * result in very long latencies. Hence for sync reclaims, where we wait on the
+ * flush lock, the caller should push out delayed write inodes first before
+ * trying to reclaim them to minimise the amount of time spent waiting. For
+ * background relaim, we just requeue the inode for the next pass.
+ *
+ * Hence the order of actions after gaining the locks should be:
+ * bad => reclaim
+ * shutdown => unpin and reclaim
+ * pinned, delwri => requeue
+ * pinned, sync => unpin
+ * stale => reclaim
+ * clean => reclaim
+ * dirty, delwri => flush and requeue
+ * dirty, sync => flush, wait and reclaim
+ */
STATIC int
xfs_reclaim_inode(
struct xfs_inode *ip,
struct xfs_perag *pag,
int sync_mode)
{
+ int error = 0;
+
/*
* The radix tree lock here protects a thread in xfs_iget from racing
* with us starting reclaim on the inode. Once we have the
@@ -735,33 +806,70 @@ xfs_reclaim_inode(
spin_unlock(&ip->i_flags_lock);
write_unlock(&pag->pag_ici_lock);
- /*
- * 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.
- */
xfs_ilock(ip, XFS_ILOCK_EXCL);
- xfs_iflock(ip);
+ if (!xfs_iflock_nowait(ip)) {
+ if (!(sync_mode & SYNC_WAIT))
+ goto out;
+ xfs_iflock(ip);
+ }
+
+ if (is_bad_inode(VFS_I(ip)))
+ goto reclaim;
+ if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
+ xfs_iunpin_wait(ip);
+ goto reclaim;
+ }
+ if (xfs_ipincount(ip)) {
+ if (!(sync_mode & SYNC_WAIT)) {
+ xfs_ifunlock(ip);
+ goto out;
+ }
+ xfs_iunpin_wait(ip);
+ }
+ if (xfs_iflags_test(ip, XFS_ISTALE))
+ goto reclaim;
+ if (xfs_inode_clean(ip))
+ goto reclaim;
+
+ /* Now we have an inode that needs flushing */
+ error = xfs_iflush(ip, sync_mode);
+ if (sync_mode & SYNC_WAIT) {
+ xfs_iflock(ip);
+ goto reclaim;
+ }
/*
- * In the case of a forced shutdown we rely on xfs_iflush() to
- * wait for the inode to be unpinned before returning an error.
+ * When we have to flush an inode but don't have SYNC_WAIT set, we
+ * flush the inode out using a delwri buffer and wait for the next
+ * call into reclaim to find it in a clean state instead of waiting for
+ * it now. We also don't return errors here - if the error is transient
+ * then the next reclaim pass will flush the inode, and if the error
+ * is permanent then the next sync reclaim will reclaim the inode and
+ * pass on the 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);
+ if (error && error != EAGAIN && !XFS_FORCED_SHUTDOWN(ip->i_mount)) {
+ xfs_fs_cmn_err(CE_WARN, ip->i_mount,
+ "inode 0x%llx background reclaim flush failed with %d",
+ (long long)ip->i_ino, error);
}
+out:
+ xfs_iflags_clear(ip, XFS_IRECLAIM);
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ /*
+ * We could return EAGAIN here to make reclaim rescan the inode tree in
+ * a short while. However, this just burns CPU time scanning the tree
+ * waiting for IO to complete and xfssyncd never goes back to the idle
+ * state. Instead, return 0 to let the next scheduled background reclaim
+ * attempt to reclaim the inode again.
+ */
+ return 0;
+reclaim:
+ xfs_ifunlock(ip);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
xfs_ireclaim(ip);
- return 0;
+ return error;
+
}
int
@@ -770,5 +878,57 @@ xfs_reclaim_inodes(
int mode)
{
return xfs_inode_ag_iterator(mp, xfs_reclaim_inode, mode,
- XFS_ICI_RECLAIM_TAG, 1);
+ XFS_ICI_RECLAIM_TAG, 1, NULL);
+}
+
+/*
+ * Shrinker infrastructure.
+ */
+static int
+xfs_reclaim_inode_shrink(
+ struct shrinker *shrink,
+ int nr_to_scan,
+ gfp_t gfp_mask)
+{
+ struct xfs_mount *mp;
+ struct xfs_perag *pag;
+ xfs_agnumber_t ag;
+ int reclaimable;
+
+ mp = container_of(shrink, struct xfs_mount, m_inode_shrink);
+ if (nr_to_scan) {
+ if (!(gfp_mask & __GFP_FS))
+ return -1;
+
+ xfs_inode_ag_iterator(mp, xfs_reclaim_inode, 0,
+ XFS_ICI_RECLAIM_TAG, 1, &nr_to_scan);
+ /* if we don't exhaust the scan, don't bother coming back */
+ if (nr_to_scan > 0)
+ return -1;
+ }
+
+ reclaimable = 0;
+ ag = 0;
+ while ((pag = xfs_inode_ag_iter_next_pag(mp, &ag,
+ XFS_ICI_RECLAIM_TAG))) {
+ reclaimable += pag->pag_ici_reclaimable;
+ xfs_perag_put(pag);
+ }
+ return reclaimable;
+}
+
+void
+xfs_inode_shrinker_register(
+ struct xfs_mount *mp)
+{
+ mp->m_inode_shrink.shrink = xfs_reclaim_inode_shrink;
+ mp->m_inode_shrink.seeks = DEFAULT_SEEKS;
+ register_shrinker(&mp->m_inode_shrink);
+}
+
+void
+xfs_inode_shrinker_unregister(
+ struct xfs_mount *mp)
+{
+ unregister_shrinker(&mp->m_inode_shrink);
}
diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h
index ea932b43335d..e28139aaa4aa 100644
--- a/fs/xfs/linux-2.6/xfs_sync.h
+++ b/fs/xfs/linux-2.6/xfs_sync.h
@@ -37,7 +37,6 @@ void xfs_syncd_stop(struct xfs_mount *mp);
int xfs_sync_attr(struct xfs_mount *mp, int flags);
int xfs_sync_data(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);
@@ -54,6 +53,9 @@ void __xfs_inode_clear_reclaim_tag(struct xfs_mount *mp, struct xfs_perag *pag,
int xfs_sync_inode_valid(struct xfs_inode *ip, struct xfs_perag *pag);
int xfs_inode_ag_iterator(struct xfs_mount *mp,
int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag, int flags),
- int flags, int tag, int write_lock);
+ int flags, int tag, int write_lock, int *nr_to_scan);
+
+void xfs_inode_shrinker_register(struct xfs_mount *mp);
+void xfs_inode_shrinker_unregister(struct xfs_mount *mp);
#endif
diff --git a/fs/xfs/linux-2.6/xfs_trace.c b/fs/xfs/linux-2.6/xfs_trace.c
index 856eb3c8d605..d12be8470cba 100644
--- a/fs/xfs/linux-2.6/xfs_trace.c
+++ b/fs/xfs/linux-2.6/xfs_trace.c
@@ -41,7 +41,6 @@
#include "xfs_alloc.h"
#include "xfs_bmap.h"
#include "xfs_attr.h"
-#include "xfs_attr_sf.h"
#include "xfs_attr_leaf.h"
#include "xfs_log_priv.h"
#include "xfs_buf_item.h"
@@ -50,22 +49,8 @@
#include "xfs_aops.h"
#include "quota/xfs_dquot_item.h"
#include "quota/xfs_dquot.h"
-
-/*
- * Format fsblock number into a static buffer & return it.
- */
-STATIC char *xfs_fmtfsblock(xfs_fsblock_t bno)
-{
- static char rval[50];
-
- if (bno == NULLFSBLOCK)
- sprintf(rval, "NULLFSBLOCK");
- else if (isnullstartblock(bno))
- sprintf(rval, "NULLSTARTBLOCK(%lld)", startblockval(bno));
- else
- sprintf(rval, "%lld", (xfs_dfsbno_t)bno);
- return rval;
-}
+#include "xfs_log_recover.h"
+#include "xfs_inode_item.h"
/*
* We include this last to have the helpers above available for the trace
diff --git a/fs/xfs/linux-2.6/xfs_trace.h b/fs/xfs/linux-2.6/xfs_trace.h
index c22a608321a3..302820690904 100644
--- a/fs/xfs/linux-2.6/xfs_trace.h
+++ b/fs/xfs/linux-2.6/xfs_trace.h
@@ -32,6 +32,10 @@ struct xfs_da_node_entry;
struct xfs_dquot;
struct xlog_ticket;
struct log;
+struct xlog_recover;
+struct xlog_recover_item;
+struct xfs_buf_log_format;
+struct xfs_inode_log_format;
DECLARE_EVENT_CLASS(xfs_attr_list_class,
TP_PROTO(struct xfs_attr_list_context *ctx),
@@ -91,6 +95,40 @@ DEFINE_ATTR_LIST_EVENT(xfs_attr_list_add);
DEFINE_ATTR_LIST_EVENT(xfs_attr_list_wrong_blk);
DEFINE_ATTR_LIST_EVENT(xfs_attr_list_notfound);
+DECLARE_EVENT_CLASS(xfs_perag_class,
+ TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, int refcount,
+ unsigned long caller_ip),
+ TP_ARGS(mp, agno, refcount, caller_ip),
+ TP_STRUCT__entry(
+ __field(dev_t, dev)
+ __field(xfs_agnumber_t, agno)
+ __field(int, refcount)
+ __field(unsigned long, caller_ip)
+ ),
+ TP_fast_assign(
+ __entry->dev = mp->m_super->s_dev;
+ __entry->agno = agno;
+ __entry->refcount = refcount;
+ __entry->caller_ip = caller_ip;
+ ),
+ TP_printk("dev %d:%d agno %u refcount %d caller %pf",
+ MAJOR(__entry->dev), MINOR(__entry->dev),
+ __entry->agno,
+ __entry->refcount,
+ (char *)__entry->caller_ip)
+);
+
+#define DEFINE_PERAG_REF_EVENT(name) \
+DEFINE_EVENT(xfs_perag_class, name, \
+ TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, int refcount, \
+ unsigned long caller_ip), \
+ TP_ARGS(mp, agno, refcount, caller_ip))
+DEFINE_PERAG_REF_EVENT(xfs_perag_get);
+DEFINE_PERAG_REF_EVENT(xfs_perag_get_reclaim);
+DEFINE_PERAG_REF_EVENT(xfs_perag_put);
+DEFINE_PERAG_REF_EVENT(xfs_perag_set_reclaim);
+DEFINE_PERAG_REF_EVENT(xfs_perag_clear_reclaim);
+
TRACE_EVENT(xfs_attr_list_node_descend,
TP_PROTO(struct xfs_attr_list_context *ctx,
struct xfs_da_node_entry *btree),
@@ -170,13 +208,13 @@ TRACE_EVENT(xfs_iext_insert,
__entry->caller_ip = caller_ip;
),
TP_printk("dev %d:%d ino 0x%llx state %s idx %ld "
- "offset %lld block %s count %lld flag %d caller %pf",
+ "offset %lld block %lld count %lld flag %d caller %pf",
MAJOR(__entry->dev), MINOR(__entry->dev),
__entry->ino,
__print_flags(__entry->bmap_state, "|", XFS_BMAP_EXT_FLAGS),
(long)__entry->idx,
__entry->startoff,
- xfs_fmtfsblock(__entry->startblock),
+ (__int64_t)__entry->startblock,
__entry->blockcount,
__entry->state,
(char *)__entry->caller_ip)
@@ -214,13 +252,13 @@ DECLARE_EVENT_CLASS(xfs_bmap_class,
__entry->caller_ip = caller_ip;
),
TP_printk("dev %d:%d ino 0x%llx state %s idx %ld "
- "offset %lld block %s count %lld flag %d caller %pf",
+ "offset %lld block %lld count %lld flag %d caller %pf",
MAJOR(__entry->dev), MINOR(__entry->dev),
__entry->ino,
__print_flags(__entry->bmap_state, "|", XFS_BMAP_EXT_FLAGS),
(long)__entry->idx,
__entry->startoff,
- xfs_fmtfsblock(__entry->startblock),
+ (__int64_t)__entry->startblock,
__entry->blockcount,
__entry->state,
(char *)__entry->caller_ip)
@@ -456,6 +494,7 @@ DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unlock);
DEFINE_BUF_ITEM_EVENT(xfs_buf_item_unlock_stale);
DEFINE_BUF_ITEM_EVENT(xfs_buf_item_committed);
DEFINE_BUF_ITEM_EVENT(xfs_buf_item_push);
+DEFINE_BUF_ITEM_EVENT(xfs_buf_item_pushbuf);
DEFINE_BUF_ITEM_EVENT(xfs_trans_get_buf);
DEFINE_BUF_ITEM_EVENT(xfs_trans_get_buf_recur);
DEFINE_BUF_ITEM_EVENT(xfs_trans_getsb);
@@ -534,18 +573,21 @@ DECLARE_EVENT_CLASS(xfs_inode_class,
__field(dev_t, dev)
__field(xfs_ino_t, ino)
__field(int, count)
+ __field(int, pincount)
__field(unsigned long, caller_ip)
),
TP_fast_assign(
__entry->dev = VFS_I(ip)->i_sb->s_dev;
__entry->ino = ip->i_ino;
__entry->count = atomic_read(&VFS_I(ip)->i_count);
+ __entry->pincount = atomic_read(&ip->i_pincount);
__entry->caller_ip = caller_ip;
),
- TP_printk("dev %d:%d ino 0x%llx count %d caller %pf",
+ TP_printk("dev %d:%d ino 0x%llx count %d pincount %d caller %pf",
MAJOR(__entry->dev), MINOR(__entry->dev),
__entry->ino,
__entry->count,
+ __entry->pincount,
(char *)__entry->caller_ip)
)
@@ -555,6 +597,10 @@ DEFINE_EVENT(xfs_inode_class, name, \
TP_ARGS(ip, caller_ip))
DEFINE_INODE_EVENT(xfs_ihold);
DEFINE_INODE_EVENT(xfs_irele);
+DEFINE_INODE_EVENT(xfs_inode_pin);
+DEFINE_INODE_EVENT(xfs_inode_unpin);
+DEFINE_INODE_EVENT(xfs_inode_unpin_nowait);
+
/* the old xfs_itrace_entry tracer - to be replaced by s.th. in the VFS */
DEFINE_INODE_EVENT(xfs_inode);
#define xfs_itrace_entry(ip) \
@@ -565,7 +611,7 @@ DECLARE_EVENT_CLASS(xfs_dquot_class,
TP_ARGS(dqp),
TP_STRUCT__entry(
__field(dev_t, dev)
- __field(__be32, id)
+ __field(u32, id)
__field(unsigned, flags)
__field(unsigned, nrefs)
__field(unsigned long long, res_bcount)
@@ -578,7 +624,7 @@ DECLARE_EVENT_CLASS(xfs_dquot_class,
), \
TP_fast_assign(
__entry->dev = dqp->q_mount->m_super->s_dev;
- __entry->id = dqp->q_core.d_id;
+ __entry->id = be32_to_cpu(dqp->q_core.d_id);
__entry->flags = dqp->dq_flags;
__entry->nrefs = dqp->q_nrefs;
__entry->res_bcount = dqp->q_res_bcount;
@@ -594,10 +640,10 @@ DECLARE_EVENT_CLASS(xfs_dquot_class,
be64_to_cpu(dqp->q_core.d_ino_softlimit);
),
TP_printk("dev %d:%d id 0x%x flags %s nrefs %u res_bc 0x%llx "
- "bcnt 0x%llx [hard 0x%llx | soft 0x%llx] "
- "icnt 0x%llx [hard 0x%llx | soft 0x%llx]",
+ "bcnt 0x%llx bhardlimit 0x%llx bsoftlimit 0x%llx "
+ "icnt 0x%llx ihardlimit 0x%llx isoftlimit 0x%llx]",
MAJOR(__entry->dev), MINOR(__entry->dev),
- be32_to_cpu(__entry->id),
+ __entry->id,
__print_flags(__entry->flags, "|", XFS_DQ_FLAGS),
__entry->nrefs,
__entry->res_bcount,
@@ -614,8 +660,6 @@ DEFINE_EVENT(xfs_dquot_class, name, \
TP_PROTO(struct xfs_dquot *dqp), \
TP_ARGS(dqp))
DEFINE_DQUOT_EVENT(xfs_dqadjust);
-DEFINE_DQUOT_EVENT(xfs_dqshake_dirty);
-DEFINE_DQUOT_EVENT(xfs_dqshake_unlink);
DEFINE_DQUOT_EVENT(xfs_dqreclaim_want);
DEFINE_DQUOT_EVENT(xfs_dqreclaim_dirty);
DEFINE_DQUOT_EVENT(xfs_dqreclaim_unlink);
@@ -630,7 +674,6 @@ DEFINE_DQUOT_EVENT(xfs_dqread_fail);
DEFINE_DQUOT_EVENT(xfs_dqlookup_found);
DEFINE_DQUOT_EVENT(xfs_dqlookup_want);
DEFINE_DQUOT_EVENT(xfs_dqlookup_freelist);
-DEFINE_DQUOT_EVENT(xfs_dqlookup_move);
DEFINE_DQUOT_EVENT(xfs_dqlookup_done);
DEFINE_DQUOT_EVENT(xfs_dqget_hit);
DEFINE_DQUOT_EVENT(xfs_dqget_miss);
@@ -739,165 +782,181 @@ DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_enter);
DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_exit);
DEFINE_LOGGRANT_EVENT(xfs_log_ungrant_sub);
-#define DEFINE_RW_EVENT(name) \
-TRACE_EVENT(name, \
- TP_PROTO(struct xfs_inode *ip, size_t count, loff_t offset, int flags), \
- TP_ARGS(ip, count, offset, flags), \
- TP_STRUCT__entry( \
- __field(dev_t, dev) \
- __field(xfs_ino_t, ino) \
- __field(xfs_fsize_t, size) \
- __field(xfs_fsize_t, new_size) \
- __field(loff_t, offset) \
- __field(size_t, count) \
- __field(int, flags) \
- ), \
- TP_fast_assign( \
- __entry->dev = VFS_I(ip)->i_sb->s_dev; \
- __entry->ino = ip->i_ino; \
- __entry->size = ip->i_d.di_size; \
- __entry->new_size = ip->i_new_size; \
- __entry->offset = offset; \
- __entry->count = count; \
- __entry->flags = flags; \
- ), \
- TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx " \
- "offset 0x%llx count 0x%zx ioflags %s", \
- MAJOR(__entry->dev), MINOR(__entry->dev), \
- __entry->ino, \
- __entry->size, \
- __entry->new_size, \
- __entry->offset, \
- __entry->count, \
- __print_flags(__entry->flags, "|", XFS_IO_FLAGS)) \
+DECLARE_EVENT_CLASS(xfs_file_class,
+ TP_PROTO(struct xfs_inode *ip, size_t count, loff_t offset, int flags),
+ TP_ARGS(ip, count, offset, flags),
+ TP_STRUCT__entry(
+ __field(dev_t, dev)
+ __field(xfs_ino_t, ino)
+ __field(xfs_fsize_t, size)
+ __field(xfs_fsize_t, new_size)
+ __field(loff_t, offset)
+ __field(size_t, count)
+ __field(int, flags)
+ ),
+ TP_fast_assign(
+ __entry->dev = VFS_I(ip)->i_sb->s_dev;
+ __entry->ino = ip->i_ino;
+ __entry->size = ip->i_d.di_size;
+ __entry->new_size = ip->i_new_size;
+ __entry->offset = offset;
+ __entry->count = count;
+ __entry->flags = flags;
+ ),
+ TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx "
+ "offset 0x%llx count 0x%zx ioflags %s",
+ MAJOR(__entry->dev), MINOR(__entry->dev),
+ __entry->ino,
+ __entry->size,
+ __entry->new_size,
+ __entry->offset,
+ __entry->count,
+ __print_flags(__entry->flags, "|", XFS_IO_FLAGS))
)
+
+#define DEFINE_RW_EVENT(name) \
+DEFINE_EVENT(xfs_file_class, name, \
+ TP_PROTO(struct xfs_inode *ip, size_t count, loff_t offset, int flags), \
+ TP_ARGS(ip, count, offset, flags))
DEFINE_RW_EVENT(xfs_file_read);
DEFINE_RW_EVENT(xfs_file_buffered_write);
DEFINE_RW_EVENT(xfs_file_direct_write);
DEFINE_RW_EVENT(xfs_file_splice_read);
DEFINE_RW_EVENT(xfs_file_splice_write);
+DECLARE_EVENT_CLASS(xfs_page_class,
+ TP_PROTO(struct inode *inode, struct page *page, unsigned long off),
+ TP_ARGS(inode, page, off),
+ TP_STRUCT__entry(
+ __field(dev_t, dev)
+ __field(xfs_ino_t, ino)
+ __field(pgoff_t, pgoff)
+ __field(loff_t, size)
+ __field(unsigned long, offset)
+ __field(int, delalloc)
+ __field(int, unmapped)
+ __field(int, unwritten)
+ ),
+ TP_fast_assign(
+ int delalloc = -1, unmapped = -1, unwritten = -1;
-#define DEFINE_PAGE_EVENT(name) \
-TRACE_EVENT(name, \
- TP_PROTO(struct inode *inode, struct page *page, unsigned long off), \
- TP_ARGS(inode, page, off), \
- TP_STRUCT__entry( \
- __field(dev_t, dev) \
- __field(xfs_ino_t, ino) \
- __field(pgoff_t, pgoff) \
- __field(loff_t, size) \
- __field(unsigned long, offset) \
- __field(int, delalloc) \
- __field(int, unmapped) \
- __field(int, unwritten) \
- ), \
- TP_fast_assign( \
- int delalloc = -1, unmapped = -1, unwritten = -1; \
- \
- if (page_has_buffers(page)) \
- xfs_count_page_state(page, &delalloc, \
- &unmapped, &unwritten); \
- __entry->dev = inode->i_sb->s_dev; \
- __entry->ino = XFS_I(inode)->i_ino; \
- __entry->pgoff = page_offset(page); \
- __entry->size = i_size_read(inode); \
- __entry->offset = off; \
- __entry->delalloc = delalloc; \
- __entry->unmapped = unmapped; \
- __entry->unwritten = unwritten; \
- ), \
- TP_printk("dev %d:%d ino 0x%llx pgoff 0x%lx size 0x%llx offset %lx " \
- "delalloc %d unmapped %d unwritten %d", \
- MAJOR(__entry->dev), MINOR(__entry->dev), \
- __entry->ino, \
- __entry->pgoff, \
- __entry->size, \
- __entry->offset, \
- __entry->delalloc, \
- __entry->unmapped, \
- __entry->unwritten) \
+ if (page_has_buffers(page))
+ xfs_count_page_state(page, &delalloc,
+ &unmapped, &unwritten);
+ __entry->dev = inode->i_sb->s_dev;
+ __entry->ino = XFS_I(inode)->i_ino;
+ __entry->pgoff = page_offset(page);
+ __entry->size = i_size_read(inode);
+ __entry->offset = off;
+ __entry->delalloc = delalloc;
+ __entry->unmapped = unmapped;
+ __entry->unwritten = unwritten;
+ ),
+ TP_printk("dev %d:%d ino 0x%llx pgoff 0x%lx size 0x%llx offset %lx "
+ "delalloc %d unmapped %d unwritten %d",
+ MAJOR(__entry->dev), MINOR(__entry->dev),
+ __entry->ino,
+ __entry->pgoff,
+ __entry->size,
+ __entry->offset,
+ __entry->delalloc,
+ __entry->unmapped,
+ __entry->unwritten)
)
+
+#define DEFINE_PAGE_EVENT(name) \
+DEFINE_EVENT(xfs_page_class, name, \
+ TP_PROTO(struct inode *inode, struct page *page, unsigned long off), \
+ TP_ARGS(inode, page, off))
DEFINE_PAGE_EVENT(xfs_writepage);
DEFINE_PAGE_EVENT(xfs_releasepage);
DEFINE_PAGE_EVENT(xfs_invalidatepage);
-#define DEFINE_IOMAP_EVENT(name) \
-TRACE_EVENT(name, \
- TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count, \
- int flags, struct xfs_bmbt_irec *irec), \
- TP_ARGS(ip, offset, count, flags, irec), \
- TP_STRUCT__entry( \
- __field(dev_t, dev) \
- __field(xfs_ino_t, ino) \
- __field(loff_t, size) \
- __field(loff_t, new_size) \
- __field(loff_t, offset) \
- __field(size_t, count) \
- __field(int, flags) \
- __field(xfs_fileoff_t, startoff) \
- __field(xfs_fsblock_t, startblock) \
- __field(xfs_filblks_t, blockcount) \
- ), \
- TP_fast_assign( \
- __entry->dev = VFS_I(ip)->i_sb->s_dev; \
- __entry->ino = ip->i_ino; \
- __entry->size = ip->i_d.di_size; \
- __entry->new_size = ip->i_new_size; \
- __entry->offset = offset; \
- __entry->count = count; \
- __entry->flags = flags; \
- __entry->startoff = irec ? irec->br_startoff : 0; \
- __entry->startblock = irec ? irec->br_startblock : 0; \
- __entry->blockcount = irec ? irec->br_blockcount : 0; \
- ), \
- TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx " \
- "offset 0x%llx count %zd flags %s " \
- "startoff 0x%llx startblock %s blockcount 0x%llx", \
- MAJOR(__entry->dev), MINOR(__entry->dev), \
- __entry->ino, \
- __entry->size, \
- __entry->new_size, \
- __entry->offset, \
- __entry->count, \
- __print_flags(__entry->flags, "|", BMAPI_FLAGS), \
- __entry->startoff, \
- xfs_fmtfsblock(__entry->startblock), \
- __entry->blockcount) \
+DECLARE_EVENT_CLASS(xfs_iomap_class,
+ TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count,
+ int flags, struct xfs_bmbt_irec *irec),
+ TP_ARGS(ip, offset, count, flags, irec),
+ TP_STRUCT__entry(
+ __field(dev_t, dev)
+ __field(xfs_ino_t, ino)
+ __field(loff_t, size)
+ __field(loff_t, new_size)
+ __field(loff_t, offset)
+ __field(size_t, count)
+ __field(int, flags)
+ __field(xfs_fileoff_t, startoff)
+ __field(xfs_fsblock_t, startblock)
+ __field(xfs_filblks_t, blockcount)
+ ),
+ TP_fast_assign(
+ __entry->dev = VFS_I(ip)->i_sb->s_dev;
+ __entry->ino = ip->i_ino;
+ __entry->size = ip->i_d.di_size;
+ __entry->new_size = ip->i_new_size;
+ __entry->offset = offset;
+ __entry->count = count;
+ __entry->flags = flags;
+ __entry->startoff = irec ? irec->br_startoff : 0;
+ __entry->startblock = irec ? irec->br_startblock : 0;
+ __entry->blockcount = irec ? irec->br_blockcount : 0;
+ ),
+ TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx "
+ "offset 0x%llx count %zd flags %s "
+ "startoff 0x%llx startblock %lld blockcount 0x%llx",
+ MAJOR(__entry->dev), MINOR(__entry->dev),
+ __entry->ino,
+ __entry->size,
+ __entry->new_size,
+ __entry->offset,
+ __entry->count,
+ __print_flags(__entry->flags, "|", BMAPI_FLAGS),
+ __entry->startoff,
+ (__int64_t)__entry->startblock,
+ __entry->blockcount)
)
+
+#define DEFINE_IOMAP_EVENT(name) \
+DEFINE_EVENT(xfs_iomap_class, name, \
+ TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count, \
+ int flags, struct xfs_bmbt_irec *irec), \
+ TP_ARGS(ip, offset, count, flags, irec))
DEFINE_IOMAP_EVENT(xfs_iomap_enter);
DEFINE_IOMAP_EVENT(xfs_iomap_found);
DEFINE_IOMAP_EVENT(xfs_iomap_alloc);
-#define DEFINE_SIMPLE_IO_EVENT(name) \
-TRACE_EVENT(name, \
- TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count), \
- TP_ARGS(ip, offset, count), \
- TP_STRUCT__entry( \
- __field(dev_t, dev) \
- __field(xfs_ino_t, ino) \
- __field(loff_t, size) \
- __field(loff_t, new_size) \
- __field(loff_t, offset) \
- __field(size_t, count) \
- ), \
- TP_fast_assign( \
- __entry->dev = VFS_I(ip)->i_sb->s_dev; \
- __entry->ino = ip->i_ino; \
- __entry->size = ip->i_d.di_size; \
- __entry->new_size = ip->i_new_size; \
- __entry->offset = offset; \
- __entry->count = count; \
- ), \
- TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx " \
- "offset 0x%llx count %zd", \
- MAJOR(__entry->dev), MINOR(__entry->dev), \
- __entry->ino, \
- __entry->size, \
- __entry->new_size, \
- __entry->offset, \
- __entry->count) \
+DECLARE_EVENT_CLASS(xfs_simple_io_class,
+ TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count),
+ TP_ARGS(ip, offset, count),
+ TP_STRUCT__entry(
+ __field(dev_t, dev)
+ __field(xfs_ino_t, ino)
+ __field(loff_t, size)
+ __field(loff_t, new_size)
+ __field(loff_t, offset)
+ __field(size_t, count)
+ ),
+ TP_fast_assign(
+ __entry->dev = VFS_I(ip)->i_sb->s_dev;
+ __entry->ino = ip->i_ino;
+ __entry->size = ip->i_d.di_size;
+ __entry->new_size = ip->i_new_size;
+ __entry->offset = offset;
+ __entry->count = count;
+ ),
+ TP_printk("dev %d:%d ino 0x%llx size 0x%llx new_size 0x%llx "
+ "offset 0x%llx count %zd",
+ MAJOR(__entry->dev), MINOR(__entry->dev),
+ __entry->ino,
+ __entry->size,
+ __entry->new_size,
+ __entry->offset,
+ __entry->count)
);
+
+#define DEFINE_SIMPLE_IO_EVENT(name) \
+DEFINE_EVENT(xfs_simple_io_class, name, \
+ TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count), \
+ TP_ARGS(ip, offset, count))
DEFINE_SIMPLE_IO_EVENT(xfs_delalloc_enospc);
DEFINE_SIMPLE_IO_EVENT(xfs_unwritten_convert);
@@ -1023,83 +1082,112 @@ TRACE_EVENT(xfs_bunmap,
);
+#define XFS_BUSY_SYNC \
+ { 0, "async" }, \
+ { 1, "sync" }
+
TRACE_EVENT(xfs_alloc_busy,
- TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno,
- xfs_extlen_t len, int slot),
- TP_ARGS(mp, agno, agbno, len, slot),
+ TP_PROTO(struct xfs_trans *trans, xfs_agnumber_t agno,
+ xfs_agblock_t agbno, xfs_extlen_t len, int sync),
+ TP_ARGS(trans, agno, agbno, len, sync),
TP_STRUCT__entry(
__field(dev_t, dev)
+ __field(struct xfs_trans *, tp)
+ __field(int, tid)
__field(xfs_agnumber_t, agno)
__field(xfs_agblock_t, agbno)
__field(xfs_extlen_t, len)
- __field(int, slot)
+ __field(int, sync)
),
TP_fast_assign(
- __entry->dev = mp->m_super->s_dev;
+ __entry->dev = trans->t_mountp->m_super->s_dev;
+ __entry->tp = trans;
+ __entry->tid = trans->t_ticket->t_tid;
__entry->agno = agno;
__entry->agbno = agbno;
__entry->len = len;
- __entry->slot = slot;
+ __entry->sync = sync;
),
- TP_printk("dev %d:%d agno %u agbno %u len %u slot %d",
+ TP_printk("dev %d:%d trans 0x%p tid 0x%x agno %u agbno %u len %u %s",
MAJOR(__entry->dev), MINOR(__entry->dev),
+ __entry->tp,
+ __entry->tid,
__entry->agno,
__entry->agbno,
__entry->len,
- __entry->slot)
+ __print_symbolic(__entry->sync, XFS_BUSY_SYNC))
);
-#define XFS_BUSY_STATES \
- { 0, "found" }, \
- { 1, "missing" }
-
TRACE_EVENT(xfs_alloc_unbusy,
TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno,
- int slot, int found),
- TP_ARGS(mp, agno, slot, found),
+ xfs_agblock_t agbno, xfs_extlen_t len),
+ TP_ARGS(mp, agno, agbno, len),
TP_STRUCT__entry(
__field(dev_t, dev)
__field(xfs_agnumber_t, agno)
- __field(int, slot)
- __field(int, found)
+ __field(xfs_agblock_t, agbno)
+ __field(xfs_extlen_t, len)
),
TP_fast_assign(
__entry->dev = mp->m_super->s_dev;
__entry->agno = agno;
- __entry->slot = slot;
- __entry->found = found;
+ __entry->agbno = agbno;
+ __entry->len = len;
),
- TP_printk("dev %d:%d agno %u slot %d %s",
+ TP_printk("dev %d:%d agno %u agbno %u len %u",
MAJOR(__entry->dev), MINOR(__entry->dev),
__entry->agno,
- __entry->slot,
- __print_symbolic(__entry->found, XFS_BUSY_STATES))
+ __entry->agbno,
+ __entry->len)
);
+#define XFS_BUSY_STATES \
+ { 0, "missing" }, \
+ { 1, "found" }
+
TRACE_EVENT(xfs_alloc_busysearch,
- TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, xfs_agblock_t agbno,
- xfs_extlen_t len, xfs_lsn_t lsn),
- TP_ARGS(mp, agno, agbno, len, lsn),
+ TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno,
+ xfs_agblock_t agbno, xfs_extlen_t len, int found),
+ TP_ARGS(mp, agno, agbno, len, found),
TP_STRUCT__entry(
__field(dev_t, dev)
__field(xfs_agnumber_t, agno)
__field(xfs_agblock_t, agbno)
__field(xfs_extlen_t, len)
- __field(xfs_lsn_t, lsn)
+ __field(int, found)
),
TP_fast_assign(
__entry->dev = mp->m_super->s_dev;
__entry->agno = agno;
__entry->agbno = agbno;
__entry->len = len;
- __entry->lsn = lsn;
+ __entry->found = found;
),
- TP_printk("dev %d:%d agno %u agbno %u len %u force lsn 0x%llx",
+ TP_printk("dev %d:%d agno %u agbno %u len %u %s",
MAJOR(__entry->dev), MINOR(__entry->dev),
__entry->agno,
__entry->agbno,
__entry->len,
+ __print_symbolic(__entry->found, XFS_BUSY_STATES))
+);
+
+TRACE_EVENT(xfs_trans_commit_lsn,
+ TP_PROTO(struct xfs_trans *trans),
+ TP_ARGS(trans),
+ TP_STRUCT__entry(
+ __field(dev_t, dev)
+ __field(struct xfs_trans *, tp)
+ __field(xfs_lsn_t, lsn)
+ ),
+ TP_fast_assign(
+ __entry->dev = trans->t_mountp->m_super->s_dev;
+ __entry->tp = trans;
+ __entry->lsn = trans->t_commit_lsn;
+ ),
+ TP_printk("dev %d:%d trans 0x%p commit_lsn 0x%llx",
+ MAJOR(__entry->dev), MINOR(__entry->dev),
+ __entry->tp,
__entry->lsn)
);
@@ -1414,6 +1502,193 @@ TRACE_EVENT(xfs_dir2_leafn_moveents,
__entry->count)
);
+#define XFS_SWAPEXT_INODES \
+ { 0, "target" }, \
+ { 1, "temp" }
+
+#define XFS_INODE_FORMAT_STR \
+ { 0, "invalid" }, \
+ { 1, "local" }, \
+ { 2, "extent" }, \
+ { 3, "btree" }
+
+DECLARE_EVENT_CLASS(xfs_swap_extent_class,
+ TP_PROTO(struct xfs_inode *ip, int which),
+ TP_ARGS(ip, which),
+ TP_STRUCT__entry(
+ __field(dev_t, dev)
+ __field(int, which)
+ __field(xfs_ino_t, ino)
+ __field(int, format)
+ __field(int, nex)
+ __field(int, max_nex)
+ __field(int, broot_size)
+ __field(int, fork_off)
+ ),
+ TP_fast_assign(
+ __entry->dev = VFS_I(ip)->i_sb->s_dev;
+ __entry->which = which;
+ __entry->ino = ip->i_ino;
+ __entry->format = ip->i_d.di_format;
+ __entry->nex = ip->i_d.di_nextents;
+ __entry->max_nex = ip->i_df.if_ext_max;
+ __entry->broot_size = ip->i_df.if_broot_bytes;
+ __entry->fork_off = XFS_IFORK_BOFF(ip);
+ ),
+ TP_printk("dev %d:%d ino 0x%llx (%s), %s format, num_extents %d, "
+ "Max in-fork extents %d, broot size %d, fork offset %d",
+ MAJOR(__entry->dev), MINOR(__entry->dev),
+ __entry->ino,
+ __print_symbolic(__entry->which, XFS_SWAPEXT_INODES),
+ __print_symbolic(__entry->format, XFS_INODE_FORMAT_STR),
+ __entry->nex,
+ __entry->max_nex,
+ __entry->broot_size,
+ __entry->fork_off)
+)
+
+#define DEFINE_SWAPEXT_EVENT(name) \
+DEFINE_EVENT(xfs_swap_extent_class, name, \
+ TP_PROTO(struct xfs_inode *ip, int which), \
+ TP_ARGS(ip, which))
+
+DEFINE_SWAPEXT_EVENT(xfs_swap_extent_before);
+DEFINE_SWAPEXT_EVENT(xfs_swap_extent_after);
+
+DECLARE_EVENT_CLASS(xfs_log_recover_item_class,
+ TP_PROTO(struct log *log, struct xlog_recover *trans,
+ struct xlog_recover_item *item, int pass),
+ TP_ARGS(log, trans, item, pass),
+ TP_STRUCT__entry(
+ __field(dev_t, dev)
+ __field(unsigned long, item)
+ __field(xlog_tid_t, tid)
+ __field(int, type)
+ __field(int, pass)
+ __field(int, count)
+ __field(int, total)
+ ),
+ TP_fast_assign(
+ __entry->dev = log->l_mp->m_super->s_dev;
+ __entry->item = (unsigned long)item;
+ __entry->tid = trans->r_log_tid;
+ __entry->type = ITEM_TYPE(item);
+ __entry->pass = pass;
+ __entry->count = item->ri_cnt;
+ __entry->total = item->ri_total;
+ ),
+ TP_printk("dev %d:%d trans 0x%x, pass %d, item 0x%p, item type %s "
+ "item region count/total %d/%d",
+ MAJOR(__entry->dev), MINOR(__entry->dev),
+ __entry->tid,
+ __entry->pass,
+ (void *)__entry->item,
+ __print_symbolic(__entry->type, XFS_LI_TYPE_DESC),
+ __entry->count,
+ __entry->total)
+)
+
+#define DEFINE_LOG_RECOVER_ITEM(name) \
+DEFINE_EVENT(xfs_log_recover_item_class, name, \
+ TP_PROTO(struct log *log, struct xlog_recover *trans, \
+ struct xlog_recover_item *item, int pass), \
+ TP_ARGS(log, trans, item, pass))
+
+DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_add);
+DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_add_cont);
+DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_reorder_head);
+DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_reorder_tail);
+DEFINE_LOG_RECOVER_ITEM(xfs_log_recover_item_recover);
+
+DECLARE_EVENT_CLASS(xfs_log_recover_buf_item_class,
+ TP_PROTO(struct log *log, struct xfs_buf_log_format *buf_f),
+ TP_ARGS(log, buf_f),
+ TP_STRUCT__entry(
+ __field(dev_t, dev)
+ __field(__int64_t, blkno)
+ __field(unsigned short, len)
+ __field(unsigned short, flags)
+ __field(unsigned short, size)
+ __field(unsigned int, map_size)
+ ),
+ TP_fast_assign(
+ __entry->dev = log->l_mp->m_super->s_dev;
+ __entry->blkno = buf_f->blf_blkno;
+ __entry->len = buf_f->blf_len;
+ __entry->flags = buf_f->blf_flags;
+ __entry->size = buf_f->blf_size;
+ __entry->map_size = buf_f->blf_map_size;
+ ),
+ TP_printk("dev %d:%d blkno 0x%llx, len %u, flags 0x%x, size %d, "
+ "map_size %d",
+ MAJOR(__entry->dev), MINOR(__entry->dev),
+ __entry->blkno,
+ __entry->len,
+ __entry->flags,
+ __entry->size,
+ __entry->map_size)
+)
+
+#define DEFINE_LOG_RECOVER_BUF_ITEM(name) \
+DEFINE_EVENT(xfs_log_recover_buf_item_class, name, \
+ TP_PROTO(struct log *log, struct xfs_buf_log_format *buf_f), \
+ TP_ARGS(log, buf_f))
+
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_not_cancel);
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_cancel);
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_cancel_add);
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_cancel_ref_inc);
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_recover);
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_inode_buf);
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_reg_buf);
+DEFINE_LOG_RECOVER_BUF_ITEM(xfs_log_recover_buf_dquot_buf);
+
+DECLARE_EVENT_CLASS(xfs_log_recover_ino_item_class,
+ TP_PROTO(struct log *log, struct xfs_inode_log_format *in_f),
+ TP_ARGS(log, in_f),
+ TP_STRUCT__entry(
+ __field(dev_t, dev)
+ __field(xfs_ino_t, ino)
+ __field(unsigned short, size)
+ __field(int, fields)
+ __field(unsigned short, asize)
+ __field(unsigned short, dsize)
+ __field(__int64_t, blkno)
+ __field(int, len)
+ __field(int, boffset)
+ ),
+ TP_fast_assign(
+ __entry->dev = log->l_mp->m_super->s_dev;
+ __entry->ino = in_f->ilf_ino;
+ __entry->size = in_f->ilf_size;
+ __entry->fields = in_f->ilf_fields;
+ __entry->asize = in_f->ilf_asize;
+ __entry->dsize = in_f->ilf_dsize;
+ __entry->blkno = in_f->ilf_blkno;
+ __entry->len = in_f->ilf_len;
+ __entry->boffset = in_f->ilf_boffset;
+ ),
+ TP_printk("dev %d:%d ino 0x%llx, size %u, fields 0x%x, asize %d, "
+ "dsize %d, blkno 0x%llx, len %d, boffset %d",
+ MAJOR(__entry->dev), MINOR(__entry->dev),
+ __entry->ino,
+ __entry->size,
+ __entry->fields,
+ __entry->asize,
+ __entry->dsize,
+ __entry->blkno,
+ __entry->len,
+ __entry->boffset)
+)
+#define DEFINE_LOG_RECOVER_INO_ITEM(name) \
+DEFINE_EVENT(xfs_log_recover_ino_item_class, name, \
+ TP_PROTO(struct log *log, struct xfs_inode_log_format *in_f), \
+ TP_ARGS(log, in_f))
+
+DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_recover);
+DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_cancel);
+DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_skip);
+
#endif /* _TRACE_XFS_H */
#undef TRACE_INCLUDE_PATH
diff --git a/fs/xfs/linux-2.6/xfs_xattr.c b/fs/xfs/linux-2.6/xfs_xattr.c
index 0b1878857fc3..87d3e03878c8 100644
--- a/fs/xfs/linux-2.6/xfs_xattr.c
+++ b/fs/xfs/linux-2.6/xfs_xattr.c
@@ -45,7 +45,7 @@ xfs_xattr_get(struct dentry *dentry, const char *name,
value = NULL;
}
- error = -xfs_attr_get(ip, name, value, &asize, xflags);
+ error = -xfs_attr_get(ip, (unsigned char *)name, value, &asize, xflags);
if (error)
return error;
return asize;
@@ -67,32 +67,33 @@ xfs_xattr_set(struct dentry *dentry, const char *name, const void *value,
xflags |= ATTR_REPLACE;
if (!value)
- return -xfs_attr_remove(ip, name, xflags);
- return -xfs_attr_set(ip, name, (void *)value, size, xflags);
+ return -xfs_attr_remove(ip, (unsigned char *)name, xflags);
+ return -xfs_attr_set(ip, (unsigned char *)name,
+ (void *)value, size, xflags);
}
-static struct xattr_handler xfs_xattr_user_handler = {
+static const struct xattr_handler xfs_xattr_user_handler = {
.prefix = XATTR_USER_PREFIX,
.flags = 0, /* no flags implies user namespace */
.get = xfs_xattr_get,
.set = xfs_xattr_set,
};
-static struct xattr_handler xfs_xattr_trusted_handler = {
+static const struct xattr_handler xfs_xattr_trusted_handler = {
.prefix = XATTR_TRUSTED_PREFIX,
.flags = ATTR_ROOT,
.get = xfs_xattr_get,
.set = xfs_xattr_set,
};
-static struct xattr_handler xfs_xattr_security_handler = {
+static const struct xattr_handler xfs_xattr_security_handler = {
.prefix = XATTR_SECURITY_PREFIX,
.flags = ATTR_SECURE,
.get = xfs_xattr_get,
.set = xfs_xattr_set,
};
-struct xattr_handler *xfs_xattr_handlers[] = {
+const struct xattr_handler *xfs_xattr_handlers[] = {
&xfs_xattr_user_handler,
&xfs_xattr_trusted_handler,
&xfs_xattr_security_handler,
@@ -124,8 +125,13 @@ static const char *xfs_xattr_prefix(int flags)
}
static int
-xfs_xattr_put_listent(struct xfs_attr_list_context *context, int flags,
- char *name, int namelen, int valuelen, char *value)
+xfs_xattr_put_listent(
+ struct xfs_attr_list_context *context,
+ int flags,
+ unsigned char *name,
+ int namelen,
+ int valuelen,
+ unsigned char *value)
{
unsigned int prefix_len = xfs_xattr_prefix_len(flags);
char *offset;
@@ -148,7 +154,7 @@ xfs_xattr_put_listent(struct xfs_attr_list_context *context, int flags,
offset = (char *)context->alist + context->count;
strncpy(offset, xfs_xattr_prefix(flags), prefix_len);
offset += prefix_len;
- strncpy(offset, name, namelen); /* real name */
+ strncpy(offset, (char *)name, namelen); /* real name */
offset += namelen;
*offset = '\0';
context->count += prefix_len + namelen + 1;
@@ -156,8 +162,13 @@ xfs_xattr_put_listent(struct xfs_attr_list_context *context, int flags,
}
static int
-xfs_xattr_put_listent_sizes(struct xfs_attr_list_context *context, int flags,
- char *name, int namelen, int valuelen, char *value)
+xfs_xattr_put_listent_sizes(
+ struct xfs_attr_list_context *context,
+ int flags,
+ unsigned char *name,
+ int namelen,
+ int valuelen,
+ unsigned char *value)
{
context->count += xfs_xattr_prefix_len(flags) + namelen + 1;
return 0;
diff --git a/fs/xfs/quota/xfs_dquot.c b/fs/xfs/quota/xfs_dquot.c
index d7c7eea09fc2..585e7633dfc7 100644
--- a/fs/xfs/quota/xfs_dquot.c
+++ b/fs/xfs/quota/xfs_dquot.c
@@ -101,7 +101,7 @@ xfs_qm_dqinit(
* No need to re-initialize these if this is a reclaimed dquot.
*/
if (brandnewdquot) {
- dqp->dq_flnext = dqp->dq_flprev = dqp;
+ INIT_LIST_HEAD(&dqp->q_freelist);
mutex_init(&dqp->q_qlock);
init_waitqueue_head(&dqp->q_pinwait);
@@ -119,20 +119,20 @@ xfs_qm_dqinit(
* Only the q_core portion was zeroed in dqreclaim_one().
* So, we need to reset others.
*/
- dqp->q_nrefs = 0;
- dqp->q_blkno = 0;
- dqp->MPL_NEXT = dqp->HL_NEXT = NULL;
- dqp->HL_PREVP = dqp->MPL_PREVP = NULL;
- dqp->q_bufoffset = 0;
- dqp->q_fileoffset = 0;
- dqp->q_transp = NULL;
- dqp->q_gdquot = NULL;
- dqp->q_res_bcount = 0;
- dqp->q_res_icount = 0;
- dqp->q_res_rtbcount = 0;
- atomic_set(&dqp->q_pincount, 0);
- dqp->q_hash = NULL;
- ASSERT(dqp->dq_flnext == dqp->dq_flprev);
+ dqp->q_nrefs = 0;
+ dqp->q_blkno = 0;
+ INIT_LIST_HEAD(&dqp->q_mplist);
+ INIT_LIST_HEAD(&dqp->q_hashlist);
+ dqp->q_bufoffset = 0;
+ dqp->q_fileoffset = 0;
+ dqp->q_transp = NULL;
+ dqp->q_gdquot = NULL;
+ dqp->q_res_bcount = 0;
+ dqp->q_res_icount = 0;
+ dqp->q_res_rtbcount = 0;
+ atomic_set(&dqp->q_pincount, 0);
+ dqp->q_hash = NULL;
+ ASSERT(list_empty(&dqp->q_freelist));
trace_xfs_dqreuse(dqp);
}
@@ -158,7 +158,7 @@ void
xfs_qm_dqdestroy(
xfs_dquot_t *dqp)
{
- ASSERT(! XFS_DQ_IS_ON_FREELIST(dqp));
+ ASSERT(list_empty(&dqp->q_freelist));
mutex_destroy(&dqp->q_qlock);
sv_destroy(&dqp->q_pinwait);
@@ -252,7 +252,7 @@ xfs_qm_adjust_dqtimers(
(be64_to_cpu(d->d_bcount) >=
be64_to_cpu(d->d_blk_hardlimit)))) {
d->d_btimer = cpu_to_be32(get_seconds() +
- XFS_QI_BTIMELIMIT(mp));
+ mp->m_quotainfo->qi_btimelimit);
} else {
d->d_bwarns = 0;
}
@@ -275,7 +275,7 @@ xfs_qm_adjust_dqtimers(
(be64_to_cpu(d->d_icount) >=
be64_to_cpu(d->d_ino_hardlimit)))) {
d->d_itimer = cpu_to_be32(get_seconds() +
- XFS_QI_ITIMELIMIT(mp));
+ mp->m_quotainfo->qi_itimelimit);
} else {
d->d_iwarns = 0;
}
@@ -298,7 +298,7 @@ xfs_qm_adjust_dqtimers(
(be64_to_cpu(d->d_rtbcount) >=
be64_to_cpu(d->d_rtb_hardlimit)))) {
d->d_rtbtimer = cpu_to_be32(get_seconds() +
- XFS_QI_RTBTIMELIMIT(mp));
+ mp->m_quotainfo->qi_rtbtimelimit);
} else {
d->d_rtbwarns = 0;
}
@@ -325,6 +325,7 @@ xfs_qm_init_dquot_blk(
uint type,
xfs_buf_t *bp)
{
+ struct xfs_quotainfo *q = mp->m_quotainfo;
xfs_dqblk_t *d;
int curid, i;
@@ -337,16 +338,16 @@ xfs_qm_init_dquot_blk(
/*
* ID of the first dquot in the block - id's are zero based.
*/
- curid = id - (id % XFS_QM_DQPERBLK(mp));
+ curid = id - (id % q->qi_dqperchunk);
ASSERT(curid >= 0);
- memset(d, 0, BBTOB(XFS_QI_DQCHUNKLEN(mp)));
- for (i = 0; i < XFS_QM_DQPERBLK(mp); i++, d++, curid++)
+ memset(d, 0, BBTOB(q->qi_dqchunklen));
+ for (i = 0; i < q->qi_dqperchunk; i++, d++, curid++)
xfs_qm_dqinit_core(curid, type, d);
xfs_trans_dquot_buf(tp, bp,
- (type & XFS_DQ_USER ? XFS_BLI_UDQUOT_BUF :
- ((type & XFS_DQ_PROJ) ? XFS_BLI_PDQUOT_BUF :
- XFS_BLI_GDQUOT_BUF)));
- xfs_trans_log_buf(tp, bp, 0, BBTOB(XFS_QI_DQCHUNKLEN(mp)) - 1);
+ (type & XFS_DQ_USER ? XFS_BLF_UDQUOT_BUF :
+ ((type & XFS_DQ_PROJ) ? XFS_BLF_PDQUOT_BUF :
+ XFS_BLF_GDQUOT_BUF)));
+ xfs_trans_log_buf(tp, bp, 0, BBTOB(q->qi_dqchunklen) - 1);
}
@@ -419,7 +420,7 @@ xfs_qm_dqalloc(
/* now we can just get the buffer (there's nothing to read yet) */
bp = xfs_trans_get_buf(tp, mp->m_ddev_targp,
dqp->q_blkno,
- XFS_QI_DQCHUNKLEN(mp),
+ mp->m_quotainfo->qi_dqchunklen,
0);
if (!bp || (error = XFS_BUF_GETERROR(bp)))
goto error1;
@@ -500,7 +501,8 @@ xfs_qm_dqtobp(
*/
if (dqp->q_blkno == (xfs_daddr_t) 0) {
/* We use the id as an index */
- dqp->q_fileoffset = (xfs_fileoff_t)id / XFS_QM_DQPERBLK(mp);
+ dqp->q_fileoffset = (xfs_fileoff_t)id /
+ mp->m_quotainfo->qi_dqperchunk;
nmaps = 1;
quotip = XFS_DQ_TO_QIP(dqp);
xfs_ilock(quotip, XFS_ILOCK_SHARED);
@@ -529,7 +531,7 @@ xfs_qm_dqtobp(
/*
* offset of dquot in the (fixed sized) dquot chunk.
*/
- dqp->q_bufoffset = (id % XFS_QM_DQPERBLK(mp)) *
+ dqp->q_bufoffset = (id % mp->m_quotainfo->qi_dqperchunk) *
sizeof(xfs_dqblk_t);
if (map.br_startblock == HOLESTARTBLOCK) {
/*
@@ -559,15 +561,13 @@ xfs_qm_dqtobp(
* Read in the buffer, unless we've just done the allocation
* (in which case we already have the buf).
*/
- if (! newdquot) {
+ if (!newdquot) {
trace_xfs_dqtobp_read(dqp);
- if ((error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
- dqp->q_blkno,
- XFS_QI_DQCHUNKLEN(mp),
- 0, &bp))) {
- return (error);
- }
+ error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
+ dqp->q_blkno,
+ mp->m_quotainfo->qi_dqchunklen,
+ 0, &bp);
if (error || !bp)
return XFS_ERROR(error);
}
@@ -689,14 +689,14 @@ xfs_qm_idtodq(
tp = NULL;
if (flags & XFS_QMOPT_DQALLOC) {
tp = xfs_trans_alloc(mp, XFS_TRANS_QM_DQALLOC);
- if ((error = xfs_trans_reserve(tp,
- XFS_QM_DQALLOC_SPACE_RES(mp),
- XFS_WRITE_LOG_RES(mp) +
- BBTOB(XFS_QI_DQCHUNKLEN(mp)) - 1 +
- 128,
- 0,
- XFS_TRANS_PERM_LOG_RES,
- XFS_WRITE_LOG_COUNT))) {
+ error = xfs_trans_reserve(tp, XFS_QM_DQALLOC_SPACE_RES(mp),
+ XFS_WRITE_LOG_RES(mp) +
+ BBTOB(mp->m_quotainfo->qi_dqchunklen) - 1 +
+ 128,
+ 0,
+ XFS_TRANS_PERM_LOG_RES,
+ XFS_WRITE_LOG_COUNT);
+ if (error) {
cancelflags = 0;
goto error0;
}
@@ -751,7 +751,6 @@ xfs_qm_dqlookup(
{
xfs_dquot_t *dqp;
uint flist_locked;
- xfs_dquot_t *d;
ASSERT(mutex_is_locked(&qh->qh_lock));
@@ -760,7 +759,7 @@ xfs_qm_dqlookup(
/*
* Traverse the hashchain looking for a match
*/
- for (dqp = qh->qh_next; dqp != NULL; dqp = dqp->HL_NEXT) {
+ list_for_each_entry(dqp, &qh->qh_list, q_hashlist) {
/*
* We already have the hashlock. We don't need the
* dqlock to look at the id field of the dquot, since the
@@ -772,12 +771,12 @@ xfs_qm_dqlookup(
/*
* All in core dquots must be on the dqlist of mp
*/
- ASSERT(dqp->MPL_PREVP != NULL);
+ ASSERT(!list_empty(&dqp->q_mplist));
xfs_dqlock(dqp);
if (dqp->q_nrefs == 0) {
- ASSERT (XFS_DQ_IS_ON_FREELIST(dqp));
- if (! xfs_qm_freelist_lock_nowait(xfs_Gqm)) {
+ ASSERT(!list_empty(&dqp->q_freelist));
+ if (!mutex_trylock(&xfs_Gqm->qm_dqfrlist_lock)) {
trace_xfs_dqlookup_want(dqp);
/*
@@ -787,7 +786,7 @@ xfs_qm_dqlookup(
*/
dqp->dq_flags |= XFS_DQ_WANT;
xfs_dqunlock(dqp);
- xfs_qm_freelist_lock(xfs_Gqm);
+ mutex_lock(&xfs_Gqm->qm_dqfrlist_lock);
xfs_dqlock(dqp);
dqp->dq_flags &= ~(XFS_DQ_WANT);
}
@@ -802,46 +801,28 @@ xfs_qm_dqlookup(
if (flist_locked) {
if (dqp->q_nrefs != 0) {
- xfs_qm_freelist_unlock(xfs_Gqm);
+ mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
flist_locked = B_FALSE;
} else {
- /*
- * take it off the freelist
- */
+ /* take it off the freelist */
trace_xfs_dqlookup_freelist(dqp);
- XQM_FREELIST_REMOVE(dqp);
- /* xfs_qm_freelist_print(&(xfs_Gqm->
- qm_dqfreelist),
- "after removal"); */
+ list_del_init(&dqp->q_freelist);
+ xfs_Gqm->qm_dqfrlist_cnt--;
}
}
- /*
- * grab a reference
- */
XFS_DQHOLD(dqp);
if (flist_locked)
- xfs_qm_freelist_unlock(xfs_Gqm);
+ mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
/*
* move the dquot to the front of the hashchain
*/
ASSERT(mutex_is_locked(&qh->qh_lock));
- if (dqp->HL_PREVP != &qh->qh_next) {
- trace_xfs_dqlookup_move(dqp);
- if ((d = dqp->HL_NEXT))
- d->HL_PREVP = dqp->HL_PREVP;
- *(dqp->HL_PREVP) = d;
- d = qh->qh_next;
- d->HL_PREVP = &dqp->HL_NEXT;
- dqp->HL_NEXT = d;
- dqp->HL_PREVP = &qh->qh_next;
- qh->qh_next = dqp;
- }
+ list_move(&dqp->q_hashlist, &qh->qh_list);
trace_xfs_dqlookup_done(dqp);
*O_dqpp = dqp;
- ASSERT(mutex_is_locked(&qh->qh_lock));
- return (0);
+ return 0;
}
}
@@ -975,16 +956,17 @@ xfs_qm_dqget(
*/
if (ip) {
xfs_ilock(ip, XFS_ILOCK_EXCL);
- if (! XFS_IS_DQTYPE_ON(mp, type)) {
- /* inode stays locked on return */
- xfs_qm_dqdestroy(dqp);
- return XFS_ERROR(ESRCH);
- }
+
/*
* A dquot could be attached to this inode by now, since
* we had dropped the ilock.
*/
if (type == XFS_DQ_USER) {
+ if (!XFS_IS_UQUOTA_ON(mp)) {
+ /* inode stays locked on return */
+ xfs_qm_dqdestroy(dqp);
+ return XFS_ERROR(ESRCH);
+ }
if (ip->i_udquot) {
xfs_qm_dqdestroy(dqp);
dqp = ip->i_udquot;
@@ -992,6 +974,11 @@ xfs_qm_dqget(
goto dqret;
}
} else {
+ if (!XFS_IS_OQUOTA_ON(mp)) {
+ /* inode stays locked on return */
+ xfs_qm_dqdestroy(dqp);
+ return XFS_ERROR(ESRCH);
+ }
if (ip->i_gdquot) {
xfs_qm_dqdestroy(dqp);
dqp = ip->i_gdquot;
@@ -1033,13 +1020,14 @@ xfs_qm_dqget(
*/
ASSERT(mutex_is_locked(&h->qh_lock));
dqp->q_hash = h;
- XQM_HASHLIST_INSERT(h, dqp);
+ list_add(&dqp->q_hashlist, &h->qh_list);
+ h->qh_version++;
/*
* Attach this dquot to this filesystem's list of all dquots,
* kept inside the mount structure in m_quotainfo field
*/
- xfs_qm_mplist_lock(mp);
+ mutex_lock(&mp->m_quotainfo->qi_dqlist_lock);
/*
* We return a locked dquot to the caller, with a reference taken
@@ -1047,9 +1035,9 @@ xfs_qm_dqget(
xfs_dqlock(dqp);
dqp->q_nrefs = 1;
- XQM_MPLIST_INSERT(&(XFS_QI_MPL_LIST(mp)), dqp);
-
- xfs_qm_mplist_unlock(mp);
+ list_add(&dqp->q_mplist, &mp->m_quotainfo->qi_dqlist);
+ mp->m_quotainfo->qi_dquots++;
+ mutex_unlock(&mp->m_quotainfo->qi_dqlist_lock);
mutex_unlock(&h->qh_lock);
dqret:
ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL));
@@ -1086,10 +1074,10 @@ xfs_qm_dqput(
* drop the dqlock and acquire the freelist and dqlock
* in the right order; but try to get it out-of-order first
*/
- if (! xfs_qm_freelist_lock_nowait(xfs_Gqm)) {
+ if (!mutex_trylock(&xfs_Gqm->qm_dqfrlist_lock)) {
trace_xfs_dqput_wait(dqp);
xfs_dqunlock(dqp);
- xfs_qm_freelist_lock(xfs_Gqm);
+ mutex_lock(&xfs_Gqm->qm_dqfrlist_lock);
xfs_dqlock(dqp);
}
@@ -1100,10 +1088,8 @@ xfs_qm_dqput(
if (--dqp->q_nrefs == 0) {
trace_xfs_dqput_free(dqp);
- /*
- * insert at end of the freelist.
- */
- XQM_FREELIST_INSERT(&(xfs_Gqm->qm_dqfreelist), dqp);
+ list_add_tail(&dqp->q_freelist, &xfs_Gqm->qm_dqfrlist);
+ xfs_Gqm->qm_dqfrlist_cnt++;
/*
* If we just added a udquot to the freelist, then
@@ -1118,10 +1104,6 @@ xfs_qm_dqput(
xfs_dqlock(gdqp);
dqp->q_gdquot = NULL;
}
-
- /* xfs_qm_freelist_print(&(xfs_Gqm->qm_dqfreelist),
- "@@@@@++ Free list (after append) @@@@@+");
- */
}
xfs_dqunlock(dqp);
@@ -1133,7 +1115,7 @@ xfs_qm_dqput(
break;
dqp = gdqp;
}
- xfs_qm_freelist_unlock(xfs_Gqm);
+ mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
}
/*
@@ -1187,7 +1169,7 @@ xfs_qm_dqflush(
* block, nada.
*/
if (!XFS_DQ_IS_DIRTY(dqp) ||
- (!(flags & XFS_QMOPT_SYNC) && atomic_read(&dqp->q_pincount) > 0)) {
+ (!(flags & SYNC_WAIT) && atomic_read(&dqp->q_pincount) > 0)) {
xfs_dqfunlock(dqp);
return 0;
}
@@ -1248,23 +1230,20 @@ xfs_qm_dqflush(
*/
if (XFS_BUF_ISPINNED(bp)) {
trace_xfs_dqflush_force(dqp);
- xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE);
+ xfs_log_force(mp, 0);
}
- if (flags & XFS_QMOPT_DELWRI) {
- xfs_bdwrite(mp, bp);
- } else if (flags & XFS_QMOPT_ASYNC) {
- error = xfs_bawrite(mp, bp);
- } else {
+ if (flags & SYNC_WAIT)
error = xfs_bwrite(mp, bp);
- }
+ else
+ xfs_bdwrite(mp, bp);
trace_xfs_dqflush_done(dqp);
/*
* dqp is still locked, but caller is free to unlock it now.
*/
- return (error);
+ return error;
}
@@ -1389,10 +1368,10 @@ int
xfs_qm_dqpurge(
xfs_dquot_t *dqp)
{
- xfs_dqhash_t *thishash;
+ xfs_dqhash_t *qh = dqp->q_hash;
xfs_mount_t *mp = dqp->q_mount;
- ASSERT(XFS_QM_IS_MPLIST_LOCKED(mp));
+ ASSERT(mutex_is_locked(&mp->m_quotainfo->qi_dqlist_lock));
ASSERT(mutex_is_locked(&dqp->q_hash->qh_lock));
xfs_dqlock(dqp);
@@ -1410,7 +1389,7 @@ xfs_qm_dqpurge(
return (1);
}
- ASSERT(XFS_DQ_IS_ON_FREELIST(dqp));
+ ASSERT(!list_empty(&dqp->q_freelist));
/*
* If we're turning off quotas, we have to make sure that, for
@@ -1445,7 +1424,7 @@ xfs_qm_dqpurge(
* We don't care about getting disk errors here. We need
* to purge this dquot anyway, so we go ahead regardless.
*/
- error = xfs_qm_dqflush(dqp, XFS_QMOPT_SYNC);
+ error = xfs_qm_dqflush(dqp, SYNC_WAIT);
if (error)
xfs_fs_cmn_err(CE_WARN, mp,
"xfs_qm_dqpurge: dquot %p flush failed", dqp);
@@ -1455,14 +1434,16 @@ xfs_qm_dqpurge(
ASSERT(XFS_FORCED_SHUTDOWN(mp) ||
!(dqp->q_logitem.qli_item.li_flags & XFS_LI_IN_AIL));
- thishash = dqp->q_hash;
- XQM_HASHLIST_REMOVE(thishash, dqp);
- XQM_MPLIST_REMOVE(&(XFS_QI_MPL_LIST(mp)), dqp);
+ list_del_init(&dqp->q_hashlist);
+ qh->qh_version++;
+ list_del_init(&dqp->q_mplist);
+ mp->m_quotainfo->qi_dqreclaims++;
+ mp->m_quotainfo->qi_dquots--;
/*
* XXX Move this to the front of the freelist, if we can get the
* freelist lock.
*/
- ASSERT(XFS_DQ_IS_ON_FREELIST(dqp));
+ ASSERT(!list_empty(&dqp->q_freelist));
dqp->q_mount = NULL;
dqp->q_hash = NULL;
@@ -1470,7 +1451,7 @@ xfs_qm_dqpurge(
memset(&dqp->q_core, 0, sizeof(dqp->q_core));
xfs_dqfunlock(dqp);
xfs_dqunlock(dqp);
- mutex_unlock(&thishash->qh_lock);
+ mutex_unlock(&qh->qh_lock);
return (0);
}
@@ -1520,6 +1501,7 @@ void
xfs_qm_dqflock_pushbuf_wait(
xfs_dquot_t *dqp)
{
+ xfs_mount_t *mp = dqp->q_mount;
xfs_buf_t *bp;
/*
@@ -1528,26 +1510,18 @@ xfs_qm_dqflock_pushbuf_wait(
* out immediately. We'll be able to acquire
* the flush lock when the I/O completes.
*/
- bp = xfs_incore(dqp->q_mount->m_ddev_targp, dqp->q_blkno,
- XFS_QI_DQCHUNKLEN(dqp->q_mount),
- XFS_INCORE_TRYLOCK);
- if (bp != NULL) {
- if (XFS_BUF_ISDELAYWRITE(bp)) {
- int error;
- if (XFS_BUF_ISPINNED(bp)) {
- xfs_log_force(dqp->q_mount,
- (xfs_lsn_t)0,
- XFS_LOG_FORCE);
- }
- error = xfs_bawrite(dqp->q_mount, bp);
- if (error)
- xfs_fs_cmn_err(CE_WARN, dqp->q_mount,
- "xfs_qm_dqflock_pushbuf_wait: "
- "pushbuf error %d on dqp %p, bp %p",
- error, dqp, bp);
- } else {
- xfs_buf_relse(bp);
- }
+ bp = xfs_incore(mp->m_ddev_targp, dqp->q_blkno,
+ mp->m_quotainfo->qi_dqchunklen, XBF_TRYLOCK);
+ if (!bp)
+ goto out_lock;
+
+ if (XFS_BUF_ISDELAYWRITE(bp)) {
+ if (XFS_BUF_ISPINNED(bp))
+ xfs_log_force(mp, 0);
+ xfs_buf_delwri_promote(bp);
+ wake_up_process(bp->b_target->bt_task);
}
+ xfs_buf_relse(bp);
+out_lock:
xfs_dqflock(dqp);
}
diff --git a/fs/xfs/quota/xfs_dquot.h b/fs/xfs/quota/xfs_dquot.h
index a0f7da586d1b..5da3a23b820d 100644
--- a/fs/xfs/quota/xfs_dquot.h
+++ b/fs/xfs/quota/xfs_dquot.h
@@ -33,40 +33,23 @@
* The hash chain headers (hash buckets)
*/
typedef struct xfs_dqhash {
- struct xfs_dquot *qh_next;
+ struct list_head qh_list;
struct mutex qh_lock;
uint qh_version; /* ever increasing version */
uint qh_nelems; /* number of dquots on the list */
} xfs_dqhash_t;
-typedef struct xfs_dqlink {
- struct xfs_dquot *ql_next; /* forward link */
- struct xfs_dquot **ql_prevp; /* pointer to prev ql_next */
-} xfs_dqlink_t;
-
struct xfs_mount;
struct xfs_trans;
/*
- * This is the marker which is designed to occupy the first few
- * bytes of the xfs_dquot_t structure. Even inside this, the freelist pointers
- * must come first.
- * This serves as the marker ("sentinel") when we have to restart list
- * iterations because of locking considerations.
- */
-typedef struct xfs_dqmarker {
- struct xfs_dquot*dqm_flnext; /* link to freelist: must be first */
- struct xfs_dquot*dqm_flprev;
- xfs_dqlink_t dqm_mplist; /* link to mount's list of dquots */
- xfs_dqlink_t dqm_hashlist; /* link to the hash chain */
- uint dqm_flags; /* various flags (XFS_DQ_*) */
-} xfs_dqmarker_t;
-
-/*
* The incore dquot structure
*/
typedef struct xfs_dquot {
- xfs_dqmarker_t q_lists; /* list ptrs, q_flags (marker) */
+ uint dq_flags; /* various flags (XFS_DQ_*) */
+ struct list_head q_freelist; /* global free list of dquots */
+ struct list_head q_mplist; /* mount's list of dquots */
+ struct list_head q_hashlist; /* gloabl hash list of dquots */
xfs_dqhash_t *q_hash; /* the hashchain header */
struct xfs_mount*q_mount; /* filesystem this relates to */
struct xfs_trans*q_transp; /* trans this belongs to currently */
@@ -87,13 +70,6 @@ typedef struct xfs_dquot {
wait_queue_head_t q_pinwait; /* dquot pinning wait queue */
} xfs_dquot_t;
-
-#define dq_flnext q_lists.dqm_flnext
-#define dq_flprev q_lists.dqm_flprev
-#define dq_mplist q_lists.dqm_mplist
-#define dq_hashlist q_lists.dqm_hashlist
-#define dq_flags q_lists.dqm_flags
-
/*
* Lock hierarchy for q_qlock:
* XFS_QLOCK_NORMAL is the implicit default,
@@ -127,7 +103,6 @@ static inline void xfs_dqfunlock(xfs_dquot_t *dqp)
}
#define XFS_DQ_IS_LOCKED(dqp) (mutex_is_locked(&((dqp)->q_qlock)))
-#define XFS_DQ_IS_ON_FREELIST(dqp) ((dqp)->dq_flnext != (dqp))
#define XFS_DQ_IS_DIRTY(dqp) ((dqp)->dq_flags & XFS_DQ_DIRTY)
#define XFS_QM_ISUDQ(dqp) ((dqp)->dq_flags & XFS_DQ_USER)
#define XFS_QM_ISPDQ(dqp) ((dqp)->dq_flags & XFS_DQ_PROJ)
diff --git a/fs/xfs/quota/xfs_dquot_item.c b/fs/xfs/quota/xfs_dquot_item.c
index d0d4a9a0bbd7..8d89a24ae324 100644
--- a/fs/xfs/quota/xfs_dquot_item.c
+++ b/fs/xfs/quota/xfs_dquot_item.c
@@ -74,11 +74,11 @@ xfs_qm_dquot_logitem_format(
logvec->i_addr = (xfs_caddr_t)&logitem->qli_format;
logvec->i_len = sizeof(xfs_dq_logformat_t);
- XLOG_VEC_SET_TYPE(logvec, XLOG_REG_TYPE_QFORMAT);
+ logvec->i_type = XLOG_REG_TYPE_QFORMAT;
logvec++;
logvec->i_addr = (xfs_caddr_t)&logitem->qli_dquot->q_core;
logvec->i_len = sizeof(xfs_disk_dquot_t);
- XLOG_VEC_SET_TYPE(logvec, XLOG_REG_TYPE_DQUOT);
+ logvec->i_type = XLOG_REG_TYPE_DQUOT;
ASSERT(2 == logitem->qli_item.li_desc->lid_size);
logitem->qli_format.qlf_size = 2;
@@ -107,8 +107,7 @@ xfs_qm_dquot_logitem_pin(
/* ARGSUSED */
STATIC void
xfs_qm_dquot_logitem_unpin(
- xfs_dq_logitem_t *logitem,
- int stale)
+ xfs_dq_logitem_t *logitem)
{
xfs_dquot_t *dqp = logitem->qli_dquot;
@@ -123,7 +122,7 @@ xfs_qm_dquot_logitem_unpin_remove(
xfs_dq_logitem_t *logitem,
xfs_trans_t *tp)
{
- xfs_qm_dquot_logitem_unpin(logitem, 0);
+ xfs_qm_dquot_logitem_unpin(logitem);
}
/*
@@ -153,7 +152,7 @@ xfs_qm_dquot_logitem_push(
* lock without sleeping, then there must not have been
* anyone in the process of flushing the dquot.
*/
- error = xfs_qm_dqflush(dqp, XFS_QMOPT_DELWRI);
+ error = xfs_qm_dqflush(dqp, 0);
if (error)
xfs_fs_cmn_err(CE_WARN, dqp->q_mount,
"xfs_qm_dquot_logitem_push: push error %d on dqp %p",
@@ -190,7 +189,7 @@ xfs_qm_dqunpin_wait(
/*
* Give the log a push so we don't wait here too long.
*/
- xfs_log_force(dqp->q_mount, (xfs_lsn_t)0, XFS_LOG_FORCE);
+ xfs_log_force(dqp->q_mount, 0);
wait_event(dqp->q_pinwait, (atomic_read(&dqp->q_pincount) == 0));
}
@@ -212,68 +211,31 @@ xfs_qm_dquot_logitem_pushbuf(
xfs_dquot_t *dqp;
xfs_mount_t *mp;
xfs_buf_t *bp;
- uint dopush;
dqp = qip->qli_dquot;
ASSERT(XFS_DQ_IS_LOCKED(dqp));
/*
- * The qli_pushbuf_flag keeps others from
- * trying to duplicate our effort.
- */
- ASSERT(qip->qli_pushbuf_flag != 0);
- ASSERT(qip->qli_push_owner == current_pid());
-
- /*
* If flushlock isn't locked anymore, chances are that the
* inode flush completed and the inode was taken off the AIL.
* So, just get out.
*/
if (completion_done(&dqp->q_flush) ||
((qip->qli_item.li_flags & XFS_LI_IN_AIL) == 0)) {
- qip->qli_pushbuf_flag = 0;
xfs_dqunlock(dqp);
return;
}
mp = dqp->q_mount;
bp = xfs_incore(mp->m_ddev_targp, qip->qli_format.qlf_blkno,
- XFS_QI_DQCHUNKLEN(mp),
- XFS_INCORE_TRYLOCK);
- if (bp != NULL) {
- if (XFS_BUF_ISDELAYWRITE(bp)) {
- dopush = ((qip->qli_item.li_flags & XFS_LI_IN_AIL) &&
- !completion_done(&dqp->q_flush));
- qip->qli_pushbuf_flag = 0;
- xfs_dqunlock(dqp);
-
- if (XFS_BUF_ISPINNED(bp)) {
- xfs_log_force(mp, (xfs_lsn_t)0,
- XFS_LOG_FORCE);
- }
- if (dopush) {
- int error;
-#ifdef XFSRACEDEBUG
- delay_for_intr();
- delay(300);
-#endif
- error = xfs_bawrite(mp, bp);
- if (error)
- xfs_fs_cmn_err(CE_WARN, mp,
- "xfs_qm_dquot_logitem_pushbuf: pushbuf error %d on qip %p, bp %p",
- error, qip, bp);
- } else {
- xfs_buf_relse(bp);
- }
- } else {
- qip->qli_pushbuf_flag = 0;
- xfs_dqunlock(dqp);
- xfs_buf_relse(bp);
- }
+ mp->m_quotainfo->qi_dqchunklen, XBF_TRYLOCK);
+ xfs_dqunlock(dqp);
+ if (!bp)
return;
- }
+ if (XFS_BUF_ISDELAYWRITE(bp))
+ xfs_buf_delwri_promote(bp);
+ xfs_buf_relse(bp);
+ return;
- qip->qli_pushbuf_flag = 0;
- xfs_dqunlock(dqp);
}
/*
@@ -291,50 +253,24 @@ xfs_qm_dquot_logitem_trylock(
xfs_dq_logitem_t *qip)
{
xfs_dquot_t *dqp;
- uint retval;
dqp = qip->qli_dquot;
if (atomic_read(&dqp->q_pincount) > 0)
- return (XFS_ITEM_PINNED);
+ return XFS_ITEM_PINNED;
if (! xfs_qm_dqlock_nowait(dqp))
- return (XFS_ITEM_LOCKED);
+ return XFS_ITEM_LOCKED;
- retval = XFS_ITEM_SUCCESS;
if (!xfs_dqflock_nowait(dqp)) {
/*
- * The dquot is already being flushed. It may have been
- * flushed delayed write, however, and we don't want to
- * get stuck waiting for that to complete. So, we want to check
- * to see if we can lock the dquot's buffer without sleeping.
- * If we can and it is marked for delayed write, then we
- * hold it and send it out from the push routine. We don't
- * want to do that now since we might sleep in the device
- * strategy routine. We also don't want to grab the buffer lock
- * here because we'd like not to call into the buffer cache
- * while holding the AIL lock.
- * Make sure to only return PUSHBUF if we set pushbuf_flag
- * ourselves. If someone else is doing it then we don't
- * want to go to the push routine and duplicate their efforts.
+ * dquot has already been flushed to the backing buffer,
+ * leave it locked, pushbuf routine will unlock it.
*/
- if (qip->qli_pushbuf_flag == 0) {
- qip->qli_pushbuf_flag = 1;
- ASSERT(qip->qli_format.qlf_blkno == dqp->q_blkno);
-#ifdef DEBUG
- qip->qli_push_owner = current_pid();
-#endif
- /*
- * The dquot is left locked.
- */
- retval = XFS_ITEM_PUSHBUF;
- } else {
- retval = XFS_ITEM_FLUSHING;
- xfs_dqunlock_nonotify(dqp);
- }
+ return XFS_ITEM_PUSHBUF;
}
ASSERT(qip->qli_item.li_flags & XFS_LI_IN_AIL);
- return (retval);
+ return XFS_ITEM_SUCCESS;
}
@@ -392,8 +328,7 @@ static struct xfs_item_ops xfs_dquot_item_ops = {
.iop_format = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
xfs_qm_dquot_logitem_format,
.iop_pin = (void(*)(xfs_log_item_t*))xfs_qm_dquot_logitem_pin,
- .iop_unpin = (void(*)(xfs_log_item_t*, int))
- xfs_qm_dquot_logitem_unpin,
+ .iop_unpin = (void(*)(xfs_log_item_t*))xfs_qm_dquot_logitem_unpin,
.iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t*))
xfs_qm_dquot_logitem_unpin_remove,
.iop_trylock = (uint(*)(xfs_log_item_t*))
@@ -420,9 +355,8 @@ xfs_qm_dquot_logitem_init(
xfs_dq_logitem_t *lp;
lp = &dqp->q_logitem;
- lp->qli_item.li_type = XFS_LI_DQUOT;
- lp->qli_item.li_ops = &xfs_dquot_item_ops;
- lp->qli_item.li_mountp = dqp->q_mount;
+ xfs_log_item_init(dqp->q_mount, &lp->qli_item, XFS_LI_DQUOT,
+ &xfs_dquot_item_ops);
lp->qli_dquot = dqp;
lp->qli_format.qlf_type = XFS_LI_DQUOT;
lp->qli_format.qlf_id = be32_to_cpu(dqp->q_core.d_id);
@@ -467,7 +401,7 @@ xfs_qm_qoff_logitem_format(xfs_qoff_logitem_t *qf,
log_vector->i_addr = (xfs_caddr_t)&(qf->qql_format);
log_vector->i_len = sizeof(xfs_qoff_logitem_t);
- XLOG_VEC_SET_TYPE(log_vector, XLOG_REG_TYPE_QUOTAOFF);
+ log_vector->i_type = XLOG_REG_TYPE_QUOTAOFF;
qf->qql_format.qf_size = 1;
}
@@ -489,7 +423,7 @@ xfs_qm_qoff_logitem_pin(xfs_qoff_logitem_t *qf)
*/
/*ARGSUSED*/
STATIC void
-xfs_qm_qoff_logitem_unpin(xfs_qoff_logitem_t *qf, int stale)
+xfs_qm_qoff_logitem_unpin(xfs_qoff_logitem_t *qf)
{
return;
}
@@ -600,8 +534,7 @@ static struct xfs_item_ops xfs_qm_qoffend_logitem_ops = {
.iop_format = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
xfs_qm_qoff_logitem_format,
.iop_pin = (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_pin,
- .iop_unpin = (void(*)(xfs_log_item_t* ,int))
- xfs_qm_qoff_logitem_unpin,
+ .iop_unpin = (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_unpin,
.iop_unpin_remove = (void(*)(xfs_log_item_t*,xfs_trans_t*))
xfs_qm_qoff_logitem_unpin_remove,
.iop_trylock = (uint(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_trylock,
@@ -622,8 +555,7 @@ static struct xfs_item_ops xfs_qm_qoff_logitem_ops = {
.iop_format = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
xfs_qm_qoff_logitem_format,
.iop_pin = (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_pin,
- .iop_unpin = (void(*)(xfs_log_item_t*, int))
- xfs_qm_qoff_logitem_unpin,
+ .iop_unpin = (void(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_unpin,
.iop_unpin_remove = (void(*)(xfs_log_item_t*,xfs_trans_t*))
xfs_qm_qoff_logitem_unpin_remove,
.iop_trylock = (uint(*)(xfs_log_item_t*))xfs_qm_qoff_logitem_trylock,
@@ -649,11 +581,8 @@ xfs_qm_qoff_logitem_init(
qf = (xfs_qoff_logitem_t*) kmem_zalloc(sizeof(xfs_qoff_logitem_t), KM_SLEEP);
- qf->qql_item.li_type = XFS_LI_QUOTAOFF;
- if (start)
- qf->qql_item.li_ops = &xfs_qm_qoffend_logitem_ops;
- else
- qf->qql_item.li_ops = &xfs_qm_qoff_logitem_ops;
+ xfs_log_item_init(mp, &qf->qql_item, XFS_LI_QUOTAOFF, start ?
+ &xfs_qm_qoffend_logitem_ops : &xfs_qm_qoff_logitem_ops);
qf->qql_item.li_mountp = mp;
qf->qql_format.qf_type = XFS_LI_QUOTAOFF;
qf->qql_format.qf_flags = flags;
diff --git a/fs/xfs/quota/xfs_dquot_item.h b/fs/xfs/quota/xfs_dquot_item.h
index 5a632531f843..5acae2ada70b 100644
--- a/fs/xfs/quota/xfs_dquot_item.h
+++ b/fs/xfs/quota/xfs_dquot_item.h
@@ -27,10 +27,6 @@ typedef struct xfs_dq_logitem {
xfs_log_item_t qli_item; /* common portion */
struct xfs_dquot *qli_dquot; /* dquot ptr */
xfs_lsn_t qli_flush_lsn; /* lsn at last flush */
- unsigned short qli_pushbuf_flag; /* 1 bit used in push_ail */
-#ifdef DEBUG
- uint64_t qli_push_owner;
-#endif
xfs_dq_logformat_t qli_format; /* logged structure */
} xfs_dq_logitem_t;
diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c
index 9e627a8b5b0e..67c018392d62 100644
--- a/fs/xfs/quota/xfs_qm.c
+++ b/fs/xfs/quota/xfs_qm.c
@@ -67,12 +67,9 @@ static cred_t xfs_zerocr;
STATIC void xfs_qm_list_init(xfs_dqlist_t *, char *, int);
STATIC void xfs_qm_list_destroy(xfs_dqlist_t *);
-STATIC void xfs_qm_freelist_init(xfs_frlist_t *);
-STATIC void xfs_qm_freelist_destroy(xfs_frlist_t *);
-
STATIC int xfs_qm_init_quotainos(xfs_mount_t *);
STATIC int xfs_qm_init_quotainfo(xfs_mount_t *);
-STATIC int xfs_qm_shake(int, gfp_t);
+STATIC int xfs_qm_shake(struct shrinker *, int, gfp_t);
static struct shrinker xfs_qm_shaker = {
.shrink = xfs_qm_shake,
@@ -84,21 +81,25 @@ extern struct mutex qcheck_lock;
#endif
#ifdef QUOTADEBUG
-#define XQM_LIST_PRINT(l, NXT, title) \
-{ \
- xfs_dquot_t *dqp; int i = 0; \
- cmn_err(CE_DEBUG, "%s (#%d)", title, (int) (l)->qh_nelems); \
- for (dqp = (l)->qh_next; dqp != NULL; dqp = dqp->NXT) { \
- cmn_err(CE_DEBUG, " %d. \"%d (%s)\" " \
- "bcnt = %d, icnt = %d, refs = %d", \
- ++i, (int) be32_to_cpu(dqp->q_core.d_id), \
- DQFLAGTO_TYPESTR(dqp), \
- (int) be64_to_cpu(dqp->q_core.d_bcount), \
- (int) be64_to_cpu(dqp->q_core.d_icount), \
- (int) dqp->q_nrefs); } \
+static void
+xfs_qm_dquot_list_print(
+ struct xfs_mount *mp)
+{
+ xfs_dquot_t *dqp;
+ int i = 0;
+
+ list_for_each_entry(dqp, &mp->m_quotainfo->qi_dqlist_lock, qi_mplist) {
+ cmn_err(CE_DEBUG, " %d. \"%d (%s)\" "
+ "bcnt = %lld, icnt = %lld, refs = %d",
+ i++, be32_to_cpu(dqp->q_core.d_id),
+ DQFLAGTO_TYPESTR(dqp),
+ (long long)be64_to_cpu(dqp->q_core.d_bcount),
+ (long long)be64_to_cpu(dqp->q_core.d_icount),
+ dqp->q_nrefs);
+ }
}
#else
-#define XQM_LIST_PRINT(l, NXT, title) do { } while (0)
+static void xfs_qm_dquot_list_print(struct xfs_mount *mp) { }
#endif
/*
@@ -118,9 +119,14 @@ xfs_Gqm_init(void)
*/
udqhash = kmem_zalloc_greedy(&hsize,
XFS_QM_HASHSIZE_LOW * sizeof(xfs_dqhash_t),
- XFS_QM_HASHSIZE_HIGH * sizeof(xfs_dqhash_t),
- KM_SLEEP | KM_MAYFAIL | KM_LARGE);
- gdqhash = kmem_zalloc(hsize, KM_SLEEP | KM_LARGE);
+ XFS_QM_HASHSIZE_HIGH * sizeof(xfs_dqhash_t));
+ if (!udqhash)
+ goto out;
+
+ gdqhash = kmem_zalloc_large(hsize);
+ if (!gdqhash)
+ goto out_free_udqhash;
+
hsize /= sizeof(xfs_dqhash_t);
ndquot = hsize << 8;
@@ -139,7 +145,9 @@ xfs_Gqm_init(void)
/*
* Freelist of all dquots of all file systems
*/
- xfs_qm_freelist_init(&(xqm->qm_dqfreelist));
+ INIT_LIST_HEAD(&xqm->qm_dqfrlist);
+ xqm->qm_dqfrlist_cnt = 0;
+ mutex_init(&xqm->qm_dqfrlist_lock);
/*
* dquot zone. we register our own low-memory callback.
@@ -170,6 +178,11 @@ xfs_Gqm_init(void)
mutex_init(&qcheck_lock);
#endif
return xqm;
+
+ out_free_udqhash:
+ kmem_free_large(udqhash);
+ out:
+ return NULL;
}
/*
@@ -179,6 +192,7 @@ STATIC void
xfs_qm_destroy(
struct xfs_qm *xqm)
{
+ struct xfs_dquot *dqp, *n;
int hsize, i;
ASSERT(xqm != NULL);
@@ -189,12 +203,26 @@ xfs_qm_destroy(
xfs_qm_list_destroy(&(xqm->qm_usr_dqhtable[i]));
xfs_qm_list_destroy(&(xqm->qm_grp_dqhtable[i]));
}
- kmem_free(xqm->qm_usr_dqhtable);
- kmem_free(xqm->qm_grp_dqhtable);
+ kmem_free_large(xqm->qm_usr_dqhtable);
+ kmem_free_large(xqm->qm_grp_dqhtable);
xqm->qm_usr_dqhtable = NULL;
xqm->qm_grp_dqhtable = NULL;
xqm->qm_dqhashmask = 0;
- xfs_qm_freelist_destroy(&(xqm->qm_dqfreelist));
+
+ /* frlist cleanup */
+ mutex_lock(&xqm->qm_dqfrlist_lock);
+ list_for_each_entry_safe(dqp, n, &xqm->qm_dqfrlist, q_freelist) {
+ xfs_dqlock(dqp);
+#ifdef QUOTADEBUG
+ cmn_err(CE_DEBUG, "FREELIST destroy 0x%p", dqp);
+#endif
+ list_del_init(&dqp->q_freelist);
+ xfs_Gqm->qm_dqfrlist_cnt--;
+ xfs_dqunlock(dqp);
+ xfs_qm_dqdestroy(dqp);
+ }
+ mutex_unlock(&xqm->qm_dqfrlist_lock);
+ mutex_destroy(&xqm->qm_dqfrlist_lock);
#ifdef DEBUG
mutex_destroy(&qcheck_lock);
#endif
@@ -219,8 +247,14 @@ xfs_qm_hold_quotafs_ref(
*/
mutex_lock(&xfs_Gqm_lock);
- if (xfs_Gqm == NULL)
+ if (!xfs_Gqm) {
xfs_Gqm = xfs_Gqm_init();
+ if (!xfs_Gqm) {
+ mutex_unlock(&xfs_Gqm_lock);
+ return ENOMEM;
+ }
+ }
+
/*
* We can keep a list of all filesystems with quotas mounted for
* debugging and statistical purposes, but ...
@@ -242,7 +276,7 @@ STATIC void
xfs_qm_rele_quotafs_ref(
struct xfs_mount *mp)
{
- xfs_dquot_t *dqp, *nextdqp;
+ xfs_dquot_t *dqp, *n;
ASSERT(xfs_Gqm);
ASSERT(xfs_Gqm->qm_nrefs > 0);
@@ -250,26 +284,24 @@ xfs_qm_rele_quotafs_ref(
/*
* Go thru the freelist and destroy all inactive dquots.
*/
- xfs_qm_freelist_lock(xfs_Gqm);
+ mutex_lock(&xfs_Gqm->qm_dqfrlist_lock);
- for (dqp = xfs_Gqm->qm_dqfreelist.qh_next;
- dqp != (xfs_dquot_t *)&(xfs_Gqm->qm_dqfreelist); ) {
+ list_for_each_entry_safe(dqp, n, &xfs_Gqm->qm_dqfrlist, q_freelist) {
xfs_dqlock(dqp);
- nextdqp = dqp->dq_flnext;
if (dqp->dq_flags & XFS_DQ_INACTIVE) {
ASSERT(dqp->q_mount == NULL);
ASSERT(! XFS_DQ_IS_DIRTY(dqp));
- ASSERT(dqp->HL_PREVP == NULL);
- ASSERT(dqp->MPL_PREVP == NULL);
- XQM_FREELIST_REMOVE(dqp);
+ ASSERT(list_empty(&dqp->q_hashlist));
+ ASSERT(list_empty(&dqp->q_mplist));
+ list_del_init(&dqp->q_freelist);
+ xfs_Gqm->qm_dqfrlist_cnt--;
xfs_dqunlock(dqp);
xfs_qm_dqdestroy(dqp);
} else {
xfs_dqunlock(dqp);
}
- dqp = nextdqp;
}
- xfs_qm_freelist_unlock(xfs_Gqm);
+ mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
/*
* Destroy the entire XQM. If somebody mounts with quotaon, this'll
@@ -291,7 +323,7 @@ xfs_qm_unmount(
struct xfs_mount *mp)
{
if (mp->m_quotainfo) {
- xfs_qm_dqpurge_all(mp, XFS_QMOPT_QUOTALL | XFS_QMOPT_UMOUNTING);
+ xfs_qm_dqpurge_all(mp, XFS_QMOPT_QUOTALL);
xfs_qm_destroy_quotainfo(mp);
}
}
@@ -435,20 +467,21 @@ xfs_qm_unmount_quotas(
*/
STATIC int
xfs_qm_dqflush_all(
- xfs_mount_t *mp,
- int flags)
+ struct xfs_mount *mp,
+ int sync_mode)
{
- int recl;
- xfs_dquot_t *dqp;
- int niters;
- int error;
+ struct xfs_quotainfo *q = mp->m_quotainfo;
+ int recl;
+ struct xfs_dquot *dqp;
+ int niters;
+ int error;
- if (mp->m_quotainfo == NULL)
+ if (!q)
return 0;
niters = 0;
again:
- xfs_qm_mplist_lock(mp);
- FOREACH_DQUOT_IN_MP(dqp, mp) {
+ mutex_lock(&q->qi_dqlist_lock);
+ list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) {
xfs_dqlock(dqp);
if (! XFS_DQ_IS_DIRTY(dqp)) {
xfs_dqunlock(dqp);
@@ -456,7 +489,7 @@ again:
}
/* XXX a sentinel would be better */
- recl = XFS_QI_MPLRECLAIMS(mp);
+ recl = q->qi_dqreclaims;
if (!xfs_dqflock_nowait(dqp)) {
/*
* If we can't grab the flush lock then check
@@ -471,21 +504,21 @@ again:
* Let go of the mplist lock. We don't want to hold it
* across a disk write.
*/
- xfs_qm_mplist_unlock(mp);
- error = xfs_qm_dqflush(dqp, flags);
+ mutex_unlock(&q->qi_dqlist_lock);
+ error = xfs_qm_dqflush(dqp, sync_mode);
xfs_dqunlock(dqp);
if (error)
return error;
- xfs_qm_mplist_lock(mp);
- if (recl != XFS_QI_MPLRECLAIMS(mp)) {
- xfs_qm_mplist_unlock(mp);
+ mutex_lock(&q->qi_dqlist_lock);
+ if (recl != q->qi_dqreclaims) {
+ mutex_unlock(&q->qi_dqlist_lock);
/* XXX restart limit */
goto again;
}
}
- xfs_qm_mplist_unlock(mp);
+ mutex_unlock(&q->qi_dqlist_lock);
/* return ! busy */
return 0;
}
@@ -495,15 +528,15 @@ again:
*/
STATIC void
xfs_qm_detach_gdquots(
- xfs_mount_t *mp)
+ struct xfs_mount *mp)
{
- xfs_dquot_t *dqp, *gdqp;
- int nrecl;
+ struct xfs_quotainfo *q = mp->m_quotainfo;
+ struct xfs_dquot *dqp, *gdqp;
+ int nrecl;
again:
- ASSERT(XFS_QM_IS_MPLIST_LOCKED(mp));
- dqp = XFS_QI_MPLNEXT(mp);
- while (dqp) {
+ ASSERT(mutex_is_locked(&q->qi_dqlist_lock));
+ list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) {
xfs_dqlock(dqp);
if ((gdqp = dqp->q_gdquot)) {
xfs_dqlock(gdqp);
@@ -516,15 +549,14 @@ xfs_qm_detach_gdquots(
* Can't hold the mplist lock across a dqput.
* XXXmust convert to marker based iterations here.
*/
- nrecl = XFS_QI_MPLRECLAIMS(mp);
- xfs_qm_mplist_unlock(mp);
+ nrecl = q->qi_dqreclaims;
+ mutex_unlock(&q->qi_dqlist_lock);
xfs_qm_dqput(gdqp);
- xfs_qm_mplist_lock(mp);
- if (nrecl != XFS_QI_MPLRECLAIMS(mp))
+ mutex_lock(&q->qi_dqlist_lock);
+ if (nrecl != q->qi_dqreclaims)
goto again;
}
- dqp = dqp->MPL_NEXT;
}
}
@@ -536,23 +568,23 @@ xfs_qm_detach_gdquots(
*/
STATIC int
xfs_qm_dqpurge_int(
- xfs_mount_t *mp,
- uint flags) /* QUOTAOFF/UMOUNTING/UQUOTA/PQUOTA/GQUOTA */
+ struct xfs_mount *mp,
+ uint flags)
{
- xfs_dquot_t *dqp;
- uint dqtype;
- int nrecl;
- xfs_dquot_t *nextdqp;
- int nmisses;
+ struct xfs_quotainfo *q = mp->m_quotainfo;
+ struct xfs_dquot *dqp, *n;
+ uint dqtype;
+ int nrecl;
+ int nmisses;
- if (mp->m_quotainfo == NULL)
+ if (!q)
return 0;
dqtype = (flags & XFS_QMOPT_UQUOTA) ? XFS_DQ_USER : 0;
dqtype |= (flags & XFS_QMOPT_PQUOTA) ? XFS_DQ_PROJ : 0;
dqtype |= (flags & XFS_QMOPT_GQUOTA) ? XFS_DQ_GROUP : 0;
- xfs_qm_mplist_lock(mp);
+ mutex_lock(&q->qi_dqlist_lock);
/*
* In the first pass through all incore dquots of this filesystem,
@@ -564,28 +596,25 @@ xfs_qm_dqpurge_int(
again:
nmisses = 0;
- ASSERT(XFS_QM_IS_MPLIST_LOCKED(mp));
+ ASSERT(mutex_is_locked(&q->qi_dqlist_lock));
/*
* Try to get rid of all of the unwanted dquots. The idea is to
* get them off mplist and hashlist, but leave them on freelist.
*/
- dqp = XFS_QI_MPLNEXT(mp);
- while (dqp) {
+ list_for_each_entry_safe(dqp, n, &q->qi_dqlist, q_mplist) {
/*
* It's OK to look at the type without taking dqlock here.
* We're holding the mplist lock here, and that's needed for
* a dqreclaim.
*/
- if ((dqp->dq_flags & dqtype) == 0) {
- dqp = dqp->MPL_NEXT;
+ if ((dqp->dq_flags & dqtype) == 0)
continue;
- }
if (!mutex_trylock(&dqp->q_hash->qh_lock)) {
- nrecl = XFS_QI_MPLRECLAIMS(mp);
- xfs_qm_mplist_unlock(mp);
+ nrecl = q->qi_dqreclaims;
+ mutex_unlock(&q->qi_dqlist_lock);
mutex_lock(&dqp->q_hash->qh_lock);
- xfs_qm_mplist_lock(mp);
+ mutex_lock(&q->qi_dqlist_lock);
/*
* XXXTheoretically, we can get into a very long
@@ -593,7 +622,7 @@ xfs_qm_dqpurge_int(
* No one can be adding dquots to the mplist at
* this point, but somebody might be taking things off.
*/
- if (nrecl != XFS_QI_MPLRECLAIMS(mp)) {
+ if (nrecl != q->qi_dqreclaims) {
mutex_unlock(&dqp->q_hash->qh_lock);
goto again;
}
@@ -603,11 +632,9 @@ xfs_qm_dqpurge_int(
* Take the dquot off the mplist and hashlist. It may remain on
* freelist in INACTIVE state.
*/
- nextdqp = dqp->MPL_NEXT;
nmisses += xfs_qm_dqpurge(dqp);
- dqp = nextdqp;
}
- xfs_qm_mplist_unlock(mp);
+ mutex_unlock(&q->qi_dqlist_lock);
return nmisses;
}
@@ -907,33 +934,33 @@ xfs_qm_dqdetach(
int
xfs_qm_sync(
- xfs_mount_t *mp,
- int flags)
+ struct xfs_mount *mp,
+ int flags)
{
- int recl, restarts;
- xfs_dquot_t *dqp;
- uint flush_flags;
- int error;
+ struct xfs_quotainfo *q = mp->m_quotainfo;
+ int recl, restarts;
+ struct xfs_dquot *dqp;
+ int error;
if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp))
return 0;
- flush_flags = (flags & SYNC_WAIT) ? XFS_QMOPT_SYNC : XFS_QMOPT_DELWRI;
restarts = 0;
again:
- xfs_qm_mplist_lock(mp);
+ mutex_lock(&q->qi_dqlist_lock);
/*
* dqpurge_all() also takes the mplist lock and iterate thru all dquots
* in quotaoff. However, if the QUOTA_ACTIVE bits are not cleared
* when we have the mplist lock, we know that dquots will be consistent
* as long as we have it locked.
*/
- if (! XFS_IS_QUOTA_ON(mp)) {
- xfs_qm_mplist_unlock(mp);
+ if (!XFS_IS_QUOTA_ON(mp)) {
+ mutex_unlock(&q->qi_dqlist_lock);
return 0;
}
- FOREACH_DQUOT_IN_MP(dqp, mp) {
+ ASSERT(mutex_is_locked(&q->qi_dqlist_lock));
+ list_for_each_entry(dqp, &q->qi_dqlist, q_mplist) {
/*
* If this is vfs_sync calling, then skip the dquots that
* don't 'seem' to be dirty. ie. don't acquire dqlock.
@@ -957,7 +984,7 @@ xfs_qm_sync(
}
/* XXX a sentinel would be better */
- recl = XFS_QI_MPLRECLAIMS(mp);
+ recl = q->qi_dqreclaims;
if (!xfs_dqflock_nowait(dqp)) {
if (flags & SYNC_TRYLOCK) {
xfs_dqunlock(dqp);
@@ -977,25 +1004,25 @@ xfs_qm_sync(
* Let go of the mplist lock. We don't want to hold it
* across a disk write
*/
- xfs_qm_mplist_unlock(mp);
- error = xfs_qm_dqflush(dqp, flush_flags);
+ mutex_unlock(&q->qi_dqlist_lock);
+ error = xfs_qm_dqflush(dqp, flags);
xfs_dqunlock(dqp);
if (error && XFS_FORCED_SHUTDOWN(mp))
return 0; /* Need to prevent umount failure */
else if (error)
return error;
- xfs_qm_mplist_lock(mp);
- if (recl != XFS_QI_MPLRECLAIMS(mp)) {
+ mutex_lock(&q->qi_dqlist_lock);
+ if (recl != q->qi_dqreclaims) {
if (++restarts >= XFS_QM_SYNC_MAX_RESTARTS)
break;
- xfs_qm_mplist_unlock(mp);
+ mutex_unlock(&q->qi_dqlist_lock);
goto again;
}
}
- xfs_qm_mplist_unlock(mp);
+ mutex_unlock(&q->qi_dqlist_lock);
return 0;
}
@@ -1040,8 +1067,9 @@ xfs_qm_init_quotainfo(
return error;
}
- xfs_qm_list_init(&qinf->qi_dqlist, "mpdqlist", 0);
- lockdep_set_class(&qinf->qi_dqlist.qh_lock, &xfs_quota_mplist_class);
+ INIT_LIST_HEAD(&qinf->qi_dqlist);
+ mutex_init(&qinf->qi_dqlist_lock);
+ lockdep_set_class(&qinf->qi_dqlist_lock, &xfs_quota_mplist_class);
qinf->qi_dqreclaims = 0;
@@ -1138,7 +1166,8 @@ xfs_qm_destroy_quotainfo(
*/
xfs_qm_rele_quotafs_ref(mp);
- xfs_qm_list_destroy(&qi->qi_dqlist);
+ ASSERT(list_empty(&qi->qi_dqlist));
+ mutex_destroy(&qi->qi_dqlist_lock);
if (qi->qi_uquotaip) {
IRELE(qi->qi_uquotaip);
@@ -1165,7 +1194,7 @@ xfs_qm_list_init(
int n)
{
mutex_init(&list->qh_lock);
- list->qh_next = NULL;
+ INIT_LIST_HEAD(&list->qh_list);
list->qh_version = 0;
list->qh_nelems = 0;
}
@@ -1304,9 +1333,6 @@ xfs_qm_qino_alloc(
*/
spin_lock(&mp->m_sb_lock);
if (flags & XFS_QMOPT_SBVERSION) {
-#if defined(DEBUG) && defined(XFS_LOUD_RECOVERY)
- unsigned oldv = mp->m_sb.sb_versionnum;
-#endif
ASSERT(!xfs_sb_version_hasquota(&mp->m_sb));
ASSERT((sbfields & (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO |
XFS_SB_GQUOTINO | XFS_SB_QFLAGS)) ==
@@ -1319,11 +1345,6 @@ xfs_qm_qino_alloc(
/* qflags will get updated _after_ quotacheck */
mp->m_sb.sb_qflags = 0;
-#if defined(DEBUG) && defined(XFS_LOUD_RECOVERY)
- cmn_err(CE_NOTE,
- "Old superblock version %x, converting to %x.",
- oldv, mp->m_sb.sb_versionnum);
-#endif
}
if (flags & XFS_QMOPT_UQUOTA)
mp->m_sb.sb_uquotino = (*ip)->i_ino;
@@ -1359,10 +1380,10 @@ xfs_qm_reset_dqcounts(
#ifdef DEBUG
j = XFS_FSB_TO_B(mp, XFS_DQUOT_CLUSTER_SIZE_FSB);
do_div(j, sizeof(xfs_dqblk_t));
- ASSERT(XFS_QM_DQPERBLK(mp) == j);
+ ASSERT(mp->m_quotainfo->qi_dqperchunk == j);
#endif
ddq = (xfs_disk_dquot_t *)XFS_BUF_PTR(bp);
- for (j = 0; j < XFS_QM_DQPERBLK(mp); j++) {
+ for (j = 0; j < mp->m_quotainfo->qi_dqperchunk; j++) {
/*
* Do a sanity check, and if needed, repair the dqblk. Don't
* output any warnings because it's perfectly possible to
@@ -1417,7 +1438,7 @@ xfs_qm_dqiter_bufs(
while (blkcnt--) {
error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp,
XFS_FSB_TO_DADDR(mp, bno),
- (int)XFS_QI_DQCHUNKLEN(mp), 0, &bp);
+ mp->m_quotainfo->qi_dqchunklen, 0, &bp);
if (error)
break;
@@ -1427,7 +1448,7 @@ xfs_qm_dqiter_bufs(
* goto the next block.
*/
bno++;
- firstid += XFS_QM_DQPERBLK(mp);
+ firstid += mp->m_quotainfo->qi_dqperchunk;
}
return error;
}
@@ -1493,7 +1514,7 @@ xfs_qm_dqiterate(
continue;
firstid = (xfs_dqid_t) map[i].br_startoff *
- XFS_QM_DQPERBLK(mp);
+ mp->m_quotainfo->qi_dqperchunk;
/*
* Do a read-ahead on the next extent.
*/
@@ -1504,7 +1525,7 @@ xfs_qm_dqiterate(
while (rablkcnt--) {
xfs_baread(mp->m_ddev_targp,
XFS_FSB_TO_DADDR(mp, rablkno),
- (int)XFS_QI_DQCHUNKLEN(mp));
+ mp->m_quotainfo->qi_dqchunklen);
rablkno++;
}
}
@@ -1564,8 +1585,10 @@ xfs_qm_quotacheck_dqadjust(
/*
* Set default limits, adjust timers (since we changed usages)
+ *
+ * There are no timers for the default values set in the root dquot.
*/
- if (! XFS_IS_SUSER_DQUOT(dqp)) {
+ if (dqp->q_core.d_id) {
xfs_qm_adjust_dqlimits(dqp->q_mount, &dqp->q_core);
xfs_qm_adjust_dqtimers(dqp->q_mount, &dqp->q_core);
}
@@ -1609,10 +1632,7 @@ xfs_qm_dqusage_adjust(
xfs_ino_t ino, /* inode number to get data for */
void __user *buffer, /* not used */
int ubsize, /* not used */
- void *private_data, /* not used */
- xfs_daddr_t bno, /* starting block of inode cluster */
int *ubused, /* not used */
- void *dip, /* on-disk inode pointer (not used) */
int *res) /* result code value */
{
xfs_inode_t *ip;
@@ -1637,7 +1657,7 @@ xfs_qm_dqusage_adjust(
* the case in all other instances. It's OK that we do this because
* quotacheck is done only at mount time.
*/
- if ((error = xfs_iget(mp, NULL, ino, 0, XFS_ILOCK_EXCL, &ip, bno))) {
+ if ((error = xfs_iget(mp, NULL, ino, 0, XFS_ILOCK_EXCL, &ip))) {
*res = BULKSTAT_RV_NOTHING;
return error;
}
@@ -1735,14 +1755,14 @@ xfs_qm_quotacheck(
lastino = 0;
flags = 0;
- ASSERT(XFS_QI_UQIP(mp) || XFS_QI_GQIP(mp));
+ ASSERT(mp->m_quotainfo->qi_uquotaip || mp->m_quotainfo->qi_gquotaip);
ASSERT(XFS_IS_QUOTA_RUNNING(mp));
/*
* There should be no cached dquots. The (simplistic) quotacheck
* algorithm doesn't like that.
*/
- ASSERT(XFS_QI_MPLNDQUOTS(mp) == 0);
+ ASSERT(list_empty(&mp->m_quotainfo->qi_dqlist));
cmn_err(CE_NOTE, "XFS quotacheck %s: Please wait.", mp->m_fsname);
@@ -1751,15 +1771,19 @@ xfs_qm_quotacheck(
* their counters to zero. We need a clean slate.
* We don't log our changes till later.
*/
- if ((uip = XFS_QI_UQIP(mp))) {
- if ((error = xfs_qm_dqiterate(mp, uip, XFS_QMOPT_UQUOTA)))
+ uip = mp->m_quotainfo->qi_uquotaip;
+ if (uip) {
+ error = xfs_qm_dqiterate(mp, uip, XFS_QMOPT_UQUOTA);
+ if (error)
goto error_return;
flags |= XFS_UQUOTA_CHKD;
}
- if ((gip = XFS_QI_GQIP(mp))) {
- if ((error = xfs_qm_dqiterate(mp, gip, XFS_IS_GQUOTA_ON(mp) ?
- XFS_QMOPT_GQUOTA : XFS_QMOPT_PQUOTA)))
+ gip = mp->m_quotainfo->qi_gquotaip;
+ if (gip) {
+ error = xfs_qm_dqiterate(mp, gip, XFS_IS_GQUOTA_ON(mp) ?
+ XFS_QMOPT_GQUOTA : XFS_QMOPT_PQUOTA);
+ if (error)
goto error_return;
flags |= XFS_OQUOTA_CHKD;
}
@@ -1769,12 +1793,13 @@ xfs_qm_quotacheck(
* Iterate thru all the inodes in the file system,
* adjusting the corresponding dquot counters in core.
*/
- if ((error = xfs_bulkstat(mp, &lastino, &count,
- xfs_qm_dqusage_adjust, NULL,
- structsz, NULL, BULKSTAT_FG_IGET, &done)))
+ error = xfs_bulkstat(mp, &lastino, &count,
+ xfs_qm_dqusage_adjust,
+ structsz, NULL, &done);
+ if (error)
break;
- } while (! done);
+ } while (!done);
/*
* We've made all the changes that we need to make incore.
@@ -1782,7 +1807,7 @@ xfs_qm_quotacheck(
* successfully.
*/
if (!error)
- error = xfs_qm_dqflush_all(mp, XFS_QMOPT_DELWRI);
+ error = xfs_qm_dqflush_all(mp, 0);
/*
* We can get this error if we couldn't do a dquot allocation inside
@@ -1792,7 +1817,7 @@ xfs_qm_quotacheck(
* at this point (because we intentionally didn't in dqget_noattach).
*/
if (error) {
- xfs_qm_dqpurge_all(mp, XFS_QMOPT_QUOTALL | XFS_QMOPT_QUOTAOFF);
+ xfs_qm_dqpurge_all(mp, XFS_QMOPT_QUOTALL);
goto error_return;
}
@@ -1813,7 +1838,7 @@ xfs_qm_quotacheck(
mp->m_qflags &= ~(XFS_OQUOTA_CHKD | XFS_UQUOTA_CHKD);
mp->m_qflags |= flags;
- XQM_LIST_PRINT(&(XFS_QI_MPL_LIST(mp)), MPL_NEXT, "++++ Mp list +++");
+ xfs_qm_dquot_list_print(mp);
error_return:
if (error) {
@@ -1862,14 +1887,14 @@ xfs_qm_init_quotainos(
mp->m_sb.sb_uquotino != NULLFSINO) {
ASSERT(mp->m_sb.sb_uquotino > 0);
if ((error = xfs_iget(mp, NULL, mp->m_sb.sb_uquotino,
- 0, 0, &uip, 0)))
+ 0, 0, &uip)))
return XFS_ERROR(error);
}
if (XFS_IS_OQUOTA_ON(mp) &&
mp->m_sb.sb_gquotino != NULLFSINO) {
ASSERT(mp->m_sb.sb_gquotino > 0);
if ((error = xfs_iget(mp, NULL, mp->m_sb.sb_gquotino,
- 0, 0, &gip, 0))) {
+ 0, 0, &gip))) {
if (uip)
IRELE(uip);
return XFS_ERROR(error);
@@ -1908,59 +1933,53 @@ xfs_qm_init_quotainos(
}
}
- XFS_QI_UQIP(mp) = uip;
- XFS_QI_GQIP(mp) = gip;
+ mp->m_quotainfo->qi_uquotaip = uip;
+ mp->m_quotainfo->qi_gquotaip = gip;
return 0;
}
+
/*
- * Traverse the freelist of dquots and attempt to reclaim a maximum of
- * 'howmany' dquots. This operation races with dqlookup(), and attempts to
- * favor the lookup function ...
- * XXXsup merge this with qm_reclaim_one().
+ * Just pop the least recently used dquot off the freelist and
+ * recycle it. The returned dquot is locked.
*/
-STATIC int
-xfs_qm_shake_freelist(
- int howmany)
+STATIC xfs_dquot_t *
+xfs_qm_dqreclaim_one(void)
{
- int nreclaimed;
- xfs_dqhash_t *hash;
- xfs_dquot_t *dqp, *nextdqp;
+ xfs_dquot_t *dqpout;
+ xfs_dquot_t *dqp;
int restarts;
- int nflushes;
- if (howmany <= 0)
- return 0;
-
- nreclaimed = 0;
restarts = 0;
- nflushes = 0;
+ dqpout = NULL;
-#ifdef QUOTADEBUG
- cmn_err(CE_DEBUG, "Shake free 0x%x", howmany);
-#endif
- /* lock order is : hashchainlock, freelistlock, mplistlock */
- tryagain:
- xfs_qm_freelist_lock(xfs_Gqm);
+ /* lockorder: hashchainlock, freelistlock, mplistlock, dqlock, dqflock */
+startagain:
+ mutex_lock(&xfs_Gqm->qm_dqfrlist_lock);
- for (dqp = xfs_Gqm->qm_dqfreelist.qh_next;
- ((dqp != (xfs_dquot_t *) &xfs_Gqm->qm_dqfreelist) &&
- nreclaimed < howmany); ) {
+ list_for_each_entry(dqp, &xfs_Gqm->qm_dqfrlist, q_freelist) {
+ struct xfs_mount *mp = dqp->q_mount;
xfs_dqlock(dqp);
/*
* We are racing with dqlookup here. Naturally we don't
- * want to reclaim a dquot that lookup wants.
+ * want to reclaim a dquot that lookup wants. We release the
+ * freelist lock and start over, so that lookup will grab
+ * both the dquot and the freelistlock.
*/
if (dqp->dq_flags & XFS_DQ_WANT) {
+ ASSERT(! (dqp->dq_flags & XFS_DQ_INACTIVE));
+
+ trace_xfs_dqreclaim_want(dqp);
+
xfs_dqunlock(dqp);
- xfs_qm_freelist_unlock(xfs_Gqm);
+ mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
if (++restarts >= XFS_QM_RECLAIM_MAX_RESTARTS)
- return nreclaimed;
+ return NULL;
XQM_STATS_INC(xqmstats.xs_qm_dqwants);
- goto tryagain;
+ goto startagain;
}
/*
@@ -1969,23 +1988,27 @@ xfs_qm_shake_freelist(
* life easier.
*/
if (dqp->dq_flags & XFS_DQ_INACTIVE) {
- ASSERT(dqp->q_mount == NULL);
+ ASSERT(mp == NULL);
ASSERT(! XFS_DQ_IS_DIRTY(dqp));
- ASSERT(dqp->HL_PREVP == NULL);
- ASSERT(dqp->MPL_PREVP == NULL);
+ ASSERT(list_empty(&dqp->q_hashlist));
+ ASSERT(list_empty(&dqp->q_mplist));
+ list_del_init(&dqp->q_freelist);
+ xfs_Gqm->qm_dqfrlist_cnt--;
+ xfs_dqunlock(dqp);
+ dqpout = dqp;
XQM_STATS_INC(xqmstats.xs_qm_dqinact_reclaims);
- nextdqp = dqp->dq_flnext;
- goto off_freelist;
+ break;
}
- ASSERT(dqp->MPL_PREVP);
+ ASSERT(dqp->q_hash);
+ ASSERT(!list_empty(&dqp->q_mplist));
+
/*
* Try to grab the flush lock. If this dquot is in the process of
* getting flushed to disk, we don't want to reclaim it.
*/
if (!xfs_dqflock_nowait(dqp)) {
xfs_dqunlock(dqp);
- dqp = dqp->dq_flnext;
continue;
}
@@ -1998,21 +2021,21 @@ xfs_qm_shake_freelist(
if (XFS_DQ_IS_DIRTY(dqp)) {
int error;
- trace_xfs_dqshake_dirty(dqp);
+ trace_xfs_dqreclaim_dirty(dqp);
/*
* We flush it delayed write, so don't bother
- * releasing the mplock.
+ * releasing the freelist lock.
*/
- error = xfs_qm_dqflush(dqp, XFS_QMOPT_DELWRI);
+ error = xfs_qm_dqflush(dqp, 0);
if (error) {
- xfs_fs_cmn_err(CE_WARN, dqp->q_mount,
- "xfs_qm_dqflush_all: dquot %p flush failed", dqp);
+ xfs_fs_cmn_err(CE_WARN, mp,
+ "xfs_qm_dqreclaim: dquot %p flush failed", dqp);
}
xfs_dqunlock(dqp); /* dqflush unlocks dqflock */
- dqp = dqp->dq_flnext;
continue;
}
+
/*
* We're trying to get the hashlock out of order. This races
* with dqlookup; so, we giveup and goto the next dquot if
@@ -2021,62 +2044,83 @@ xfs_qm_shake_freelist(
* waiting for the freelist lock.
*/
if (!mutex_trylock(&dqp->q_hash->qh_lock)) {
- xfs_dqfunlock(dqp);
- xfs_dqunlock(dqp);
- dqp = dqp->dq_flnext;
- continue;
+ restarts++;
+ goto dqfunlock;
}
+
/*
* This races with dquot allocation code as well as dqflush_all
* and reclaim code. So, if we failed to grab the mplist lock,
* giveup everything and start over.
*/
- hash = dqp->q_hash;
- ASSERT(hash);
- if (! xfs_qm_mplist_nowait(dqp->q_mount)) {
- /* XXX put a sentinel so that we can come back here */
+ if (!mutex_trylock(&mp->m_quotainfo->qi_dqlist_lock)) {
+ restarts++;
+ mutex_unlock(&dqp->q_hash->qh_lock);
xfs_dqfunlock(dqp);
xfs_dqunlock(dqp);
- mutex_unlock(&hash->qh_lock);
- xfs_qm_freelist_unlock(xfs_Gqm);
- if (++restarts >= XFS_QM_RECLAIM_MAX_RESTARTS)
- return nreclaimed;
- goto tryagain;
+ mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
+ if (restarts++ >= XFS_QM_RECLAIM_MAX_RESTARTS)
+ return NULL;
+ goto startagain;
}
- trace_xfs_dqshake_unlink(dqp);
-
-#ifdef QUOTADEBUG
- cmn_err(CE_DEBUG, "Shake 0x%p, ID 0x%x\n",
- dqp, be32_to_cpu(dqp->q_core.d_id));
-#endif
ASSERT(dqp->q_nrefs == 0);
- nextdqp = dqp->dq_flnext;
- XQM_MPLIST_REMOVE(&(XFS_QI_MPL_LIST(dqp->q_mount)), dqp);
- XQM_HASHLIST_REMOVE(hash, dqp);
+ list_del_init(&dqp->q_mplist);
+ mp->m_quotainfo->qi_dquots--;
+ mp->m_quotainfo->qi_dqreclaims++;
+ list_del_init(&dqp->q_hashlist);
+ dqp->q_hash->qh_version++;
+ list_del_init(&dqp->q_freelist);
+ xfs_Gqm->qm_dqfrlist_cnt--;
+ dqpout = dqp;
+ mutex_unlock(&mp->m_quotainfo->qi_dqlist_lock);
+ mutex_unlock(&dqp->q_hash->qh_lock);
+dqfunlock:
xfs_dqfunlock(dqp);
- xfs_qm_mplist_unlock(dqp->q_mount);
- mutex_unlock(&hash->qh_lock);
-
- off_freelist:
- XQM_FREELIST_REMOVE(dqp);
xfs_dqunlock(dqp);
- nreclaimed++;
- XQM_STATS_INC(xqmstats.xs_qm_dqshake_reclaims);
+ if (dqpout)
+ break;
+ if (restarts >= XFS_QM_RECLAIM_MAX_RESTARTS)
+ return NULL;
+ }
+ mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
+ return dqpout;
+}
+
+/*
+ * Traverse the freelist of dquots and attempt to reclaim a maximum of
+ * 'howmany' dquots. This operation races with dqlookup(), and attempts to
+ * favor the lookup function ...
+ */
+STATIC int
+xfs_qm_shake_freelist(
+ int howmany)
+{
+ int nreclaimed = 0;
+ xfs_dquot_t *dqp;
+
+ if (howmany <= 0)
+ return 0;
+
+ while (nreclaimed < howmany) {
+ dqp = xfs_qm_dqreclaim_one();
+ if (!dqp)
+ return nreclaimed;
xfs_qm_dqdestroy(dqp);
- dqp = nextdqp;
+ nreclaimed++;
}
- xfs_qm_freelist_unlock(xfs_Gqm);
return nreclaimed;
}
-
/*
* The kmem_shake interface is invoked when memory is running low.
*/
/* ARGSUSED */
STATIC int
-xfs_qm_shake(int nr_to_scan, gfp_t gfp_mask)
+xfs_qm_shake(
+ struct shrinker *shrink,
+ int nr_to_scan,
+ gfp_t gfp_mask)
{
int ndqused, nfree, n;
@@ -2085,7 +2129,7 @@ xfs_qm_shake(int nr_to_scan, gfp_t gfp_mask)
if (!xfs_Gqm)
return 0;
- nfree = xfs_Gqm->qm_dqfreelist.qh_nelems; /* free dquots */
+ nfree = xfs_Gqm->qm_dqfrlist_cnt; /* free dquots */
/* incore dquots in all f/s's */
ndqused = atomic_read(&xfs_Gqm->qm_totaldquots) - nfree;
@@ -2101,131 +2145,6 @@ xfs_qm_shake(int nr_to_scan, gfp_t gfp_mask)
}
-/*
- * Just pop the least recently used dquot off the freelist and
- * recycle it. The returned dquot is locked.
- */
-STATIC xfs_dquot_t *
-xfs_qm_dqreclaim_one(void)
-{
- xfs_dquot_t *dqpout;
- xfs_dquot_t *dqp;
- int restarts;
- int nflushes;
-
- restarts = 0;
- dqpout = NULL;
- nflushes = 0;
-
- /* lockorder: hashchainlock, freelistlock, mplistlock, dqlock, dqflock */
- startagain:
- xfs_qm_freelist_lock(xfs_Gqm);
-
- FOREACH_DQUOT_IN_FREELIST(dqp, &(xfs_Gqm->qm_dqfreelist)) {
- xfs_dqlock(dqp);
-
- /*
- * We are racing with dqlookup here. Naturally we don't
- * want to reclaim a dquot that lookup wants. We release the
- * freelist lock and start over, so that lookup will grab
- * both the dquot and the freelistlock.
- */
- if (dqp->dq_flags & XFS_DQ_WANT) {
- ASSERT(! (dqp->dq_flags & XFS_DQ_INACTIVE));
-
- trace_xfs_dqreclaim_want(dqp);
-
- xfs_dqunlock(dqp);
- xfs_qm_freelist_unlock(xfs_Gqm);
- if (++restarts >= XFS_QM_RECLAIM_MAX_RESTARTS)
- return NULL;
- XQM_STATS_INC(xqmstats.xs_qm_dqwants);
- goto startagain;
- }
-
- /*
- * If the dquot is inactive, we are assured that it is
- * not on the mplist or the hashlist, and that makes our
- * life easier.
- */
- if (dqp->dq_flags & XFS_DQ_INACTIVE) {
- ASSERT(dqp->q_mount == NULL);
- ASSERT(! XFS_DQ_IS_DIRTY(dqp));
- ASSERT(dqp->HL_PREVP == NULL);
- ASSERT(dqp->MPL_PREVP == NULL);
- XQM_FREELIST_REMOVE(dqp);
- xfs_dqunlock(dqp);
- dqpout = dqp;
- XQM_STATS_INC(xqmstats.xs_qm_dqinact_reclaims);
- break;
- }
-
- ASSERT(dqp->q_hash);
- ASSERT(dqp->MPL_PREVP);
-
- /*
- * Try to grab the flush lock. If this dquot is in the process of
- * getting flushed to disk, we don't want to reclaim it.
- */
- if (!xfs_dqflock_nowait(dqp)) {
- xfs_dqunlock(dqp);
- continue;
- }
-
- /*
- * We have the flush lock so we know that this is not in the
- * process of being flushed. So, if this is dirty, flush it
- * DELWRI so that we don't get a freelist infested with
- * dirty dquots.
- */
- if (XFS_DQ_IS_DIRTY(dqp)) {
- int error;
-
- trace_xfs_dqreclaim_dirty(dqp);
-
- /*
- * We flush it delayed write, so don't bother
- * releasing the freelist lock.
- */
- error = xfs_qm_dqflush(dqp, XFS_QMOPT_DELWRI);
- if (error) {
- xfs_fs_cmn_err(CE_WARN, dqp->q_mount,
- "xfs_qm_dqreclaim: dquot %p flush failed", dqp);
- }
- xfs_dqunlock(dqp); /* dqflush unlocks dqflock */
- continue;
- }
-
- if (! xfs_qm_mplist_nowait(dqp->q_mount)) {
- xfs_dqfunlock(dqp);
- xfs_dqunlock(dqp);
- continue;
- }
-
- if (!mutex_trylock(&dqp->q_hash->qh_lock))
- goto mplistunlock;
-
- trace_xfs_dqreclaim_unlink(dqp);
-
- ASSERT(dqp->q_nrefs == 0);
- XQM_MPLIST_REMOVE(&(XFS_QI_MPL_LIST(dqp->q_mount)), dqp);
- XQM_HASHLIST_REMOVE(dqp->q_hash, dqp);
- XQM_FREELIST_REMOVE(dqp);
- dqpout = dqp;
- mutex_unlock(&dqp->q_hash->qh_lock);
- mplistunlock:
- xfs_qm_mplist_unlock(dqp->q_mount);
- xfs_dqfunlock(dqp);
- xfs_dqunlock(dqp);
- if (dqpout)
- break;
- }
-
- xfs_qm_freelist_unlock(xfs_Gqm);
- return dqpout;
-}
-
-
/*------------------------------------------------------------------*/
/*
@@ -2650,66 +2569,3 @@ xfs_qm_vop_create_dqattach(
}
}
-/* ------------- list stuff -----------------*/
-STATIC void
-xfs_qm_freelist_init(xfs_frlist_t *ql)
-{
- ql->qh_next = ql->qh_prev = (xfs_dquot_t *) ql;
- mutex_init(&ql->qh_lock);
- ql->qh_version = 0;
- ql->qh_nelems = 0;
-}
-
-STATIC void
-xfs_qm_freelist_destroy(xfs_frlist_t *ql)
-{
- xfs_dquot_t *dqp, *nextdqp;
-
- mutex_lock(&ql->qh_lock);
- for (dqp = ql->qh_next;
- dqp != (xfs_dquot_t *)ql; ) {
- xfs_dqlock(dqp);
- nextdqp = dqp->dq_flnext;
-#ifdef QUOTADEBUG
- cmn_err(CE_DEBUG, "FREELIST destroy 0x%p", dqp);
-#endif
- XQM_FREELIST_REMOVE(dqp);
- xfs_dqunlock(dqp);
- xfs_qm_dqdestroy(dqp);
- dqp = nextdqp;
- }
- mutex_unlock(&ql->qh_lock);
- mutex_destroy(&ql->qh_lock);
-
- ASSERT(ql->qh_nelems == 0);
-}
-
-STATIC void
-xfs_qm_freelist_insert(xfs_frlist_t *ql, xfs_dquot_t *dq)
-{
- dq->dq_flnext = ql->qh_next;
- dq->dq_flprev = (xfs_dquot_t *)ql;
- ql->qh_next = dq;
- dq->dq_flnext->dq_flprev = dq;
- xfs_Gqm->qm_dqfreelist.qh_nelems++;
- xfs_Gqm->qm_dqfreelist.qh_version++;
-}
-
-void
-xfs_qm_freelist_unlink(xfs_dquot_t *dq)
-{
- xfs_dquot_t *next = dq->dq_flnext;
- xfs_dquot_t *prev = dq->dq_flprev;
-
- next->dq_flprev = prev;
- prev->dq_flnext = next;
- dq->dq_flnext = dq->dq_flprev = dq;
- xfs_Gqm->qm_dqfreelist.qh_nelems--;
- xfs_Gqm->qm_dqfreelist.qh_version++;
-}
-
-void
-xfs_qm_freelist_append(xfs_frlist_t *ql, xfs_dquot_t *dq)
-{
- xfs_qm_freelist_insert((xfs_frlist_t *)ql->qh_prev, dq);
-}
diff --git a/fs/xfs/quota/xfs_qm.h b/fs/xfs/quota/xfs_qm.h
index 495564b8af38..c9446f1c726d 100644
--- a/fs/xfs/quota/xfs_qm.h
+++ b/fs/xfs/quota/xfs_qm.h
@@ -72,17 +72,6 @@ extern kmem_zone_t *qm_dqtrxzone;
#define XFS_QM_MAX_DQCLUSTER_LOGSZ 3
typedef xfs_dqhash_t xfs_dqlist_t;
-/*
- * The freelist head. The first two fields match the first two in the
- * xfs_dquot_t structure (in xfs_dqmarker_t)
- */
-typedef struct xfs_frlist {
- struct xfs_dquot *qh_next;
- struct xfs_dquot *qh_prev;
- struct mutex qh_lock;
- uint qh_version;
- uint qh_nelems;
-} xfs_frlist_t;
/*
* Quota Manager (global) structure. Lives only in core.
@@ -91,7 +80,9 @@ typedef struct xfs_qm {
xfs_dqlist_t *qm_usr_dqhtable;/* udquot hash table */
xfs_dqlist_t *qm_grp_dqhtable;/* gdquot hash table */
uint qm_dqhashmask; /* # buckets in dq hashtab - 1 */
- xfs_frlist_t qm_dqfreelist; /* freelist of dquots */
+ struct list_head qm_dqfrlist; /* freelist of dquots */
+ struct mutex qm_dqfrlist_lock;
+ int qm_dqfrlist_cnt;
atomic_t qm_totaldquots; /* total incore dquots */
uint qm_nrefs; /* file systems with quota on */
int qm_dqfree_ratio;/* ratio of free to inuse dquots */
@@ -106,7 +97,9 @@ typedef struct xfs_qm {
typedef struct xfs_quotainfo {
xfs_inode_t *qi_uquotaip; /* user quota inode */
xfs_inode_t *qi_gquotaip; /* group quota inode */
- xfs_dqlist_t qi_dqlist; /* all dquots in filesys */
+ struct list_head qi_dqlist; /* all dquots in filesys */
+ struct mutex qi_dqlist_lock;
+ int qi_dquots;
int qi_dqreclaims; /* a change here indicates
a removal in the dqlist */
time_t qi_btimelimit; /* limit for blks timer */
@@ -175,10 +168,6 @@ extern int xfs_qm_scall_getqstat(xfs_mount_t *, fs_quota_stat_t *);
extern int xfs_qm_scall_quotaon(xfs_mount_t *, uint);
extern int xfs_qm_scall_quotaoff(xfs_mount_t *, uint);
-/* list stuff */
-extern void xfs_qm_freelist_append(xfs_frlist_t *, xfs_dquot_t *);
-extern void xfs_qm_freelist_unlink(xfs_dquot_t *);
-
#ifdef DEBUG
extern int xfs_qm_internalqcheck(xfs_mount_t *);
#else
diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c
index a5346630dfae..97b410c12794 100644
--- a/fs/xfs/quota/xfs_qm_bhv.c
+++ b/fs/xfs/quota/xfs_qm_bhv.c
@@ -59,7 +59,7 @@ xfs_fill_statvfs_from_dquot(
be64_to_cpu(dp->d_blk_hardlimit);
if (limit && statp->f_blocks > limit) {
statp->f_blocks = limit;
- statp->f_bfree =
+ statp->f_bfree = statp->f_bavail =
(statp->f_blocks > be64_to_cpu(dp->d_bcount)) ?
(statp->f_blocks - be64_to_cpu(dp->d_bcount)) : 0;
}
diff --git a/fs/xfs/quota/xfs_qm_stats.c b/fs/xfs/quota/xfs_qm_stats.c
index 83e7ea3e25fa..3d1fc79532e2 100644
--- a/fs/xfs/quota/xfs_qm_stats.c
+++ b/fs/xfs/quota/xfs_qm_stats.c
@@ -55,7 +55,7 @@ static int xqm_proc_show(struct seq_file *m, void *v)
ndquot,
xfs_Gqm? atomic_read(&xfs_Gqm->qm_totaldquots) : 0,
xfs_Gqm? xfs_Gqm->qm_dqfree_ratio : 0,
- xfs_Gqm? xfs_Gqm->qm_dqfreelist.qh_nelems : 0);
+ xfs_Gqm? xfs_Gqm->qm_dqfrlist_cnt : 0);
return 0;
}
diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c
index 873e07e29074..b4487764e923 100644
--- a/fs/xfs/quota/xfs_qm_syscalls.c
+++ b/fs/xfs/quota/xfs_qm_syscalls.c
@@ -79,6 +79,7 @@ xfs_qm_scall_quotaoff(
xfs_mount_t *mp,
uint flags)
{
+ struct xfs_quotainfo *q = mp->m_quotainfo;
uint dqtype;
int error;
uint inactivate_flags;
@@ -102,11 +103,8 @@ xfs_qm_scall_quotaoff(
* critical thing.
* If quotaoff, then we must be dealing with the root filesystem.
*/
- ASSERT(mp->m_quotainfo);
- if (mp->m_quotainfo)
- mutex_lock(&(XFS_QI_QOFFLOCK(mp)));
-
- ASSERT(mp->m_quotainfo);
+ ASSERT(q);
+ mutex_lock(&q->qi_quotaofflock);
/*
* If we're just turning off quota enforcement, change mp and go.
@@ -117,7 +115,7 @@ xfs_qm_scall_quotaoff(
spin_lock(&mp->m_sb_lock);
mp->m_sb.sb_qflags = mp->m_qflags;
spin_unlock(&mp->m_sb_lock);
- mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
+ mutex_unlock(&q->qi_quotaofflock);
/* XXX what to do if error ? Revert back to old vals incore ? */
error = xfs_qm_write_sb_changes(mp, XFS_SB_QFLAGS);
@@ -150,10 +148,8 @@ xfs_qm_scall_quotaoff(
* Nothing to do? Don't complain. This happens when we're just
* turning off quota enforcement.
*/
- if ((mp->m_qflags & flags) == 0) {
- mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
- return (0);
- }
+ if ((mp->m_qflags & flags) == 0)
+ goto out_unlock;
/*
* Write the LI_QUOTAOFF log record, and do SB changes atomically,
@@ -162,7 +158,7 @@ xfs_qm_scall_quotaoff(
*/
error = xfs_qm_log_quotaoff(mp, &qoffstart, flags);
if (error)
- goto out_error;
+ goto out_unlock;
/*
* Next we clear the XFS_MOUNT_*DQ_ACTIVE bit(s) in the mount struct
@@ -204,7 +200,7 @@ xfs_qm_scall_quotaoff(
* So, if we couldn't purge all the dquots from the filesystem,
* we can't get rid of the incore data structures.
*/
- while ((nculprits = xfs_qm_dqpurge_all(mp, dqtype|XFS_QMOPT_QUOTAOFF)))
+ while ((nculprits = xfs_qm_dqpurge_all(mp, dqtype)))
delay(10 * nculprits);
/*
@@ -222,7 +218,7 @@ xfs_qm_scall_quotaoff(
if (error) {
/* We're screwed now. Shutdown is the only option. */
xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
- goto out_error;
+ goto out_unlock;
}
/*
@@ -230,27 +226,26 @@ xfs_qm_scall_quotaoff(
*/
if (((flags & XFS_MOUNT_QUOTA_ALL) == XFS_MOUNT_QUOTA_SET1) ||
((flags & XFS_MOUNT_QUOTA_ALL) == XFS_MOUNT_QUOTA_SET2)) {
- mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
+ mutex_unlock(&q->qi_quotaofflock);
xfs_qm_destroy_quotainfo(mp);
return (0);
}
/*
- * Release our quotainode references, and vn_purge them,
- * if we don't need them anymore.
+ * Release our quotainode references if we don't need them anymore.
*/
- if ((dqtype & XFS_QMOPT_UQUOTA) && XFS_QI_UQIP(mp)) {
- IRELE(XFS_QI_UQIP(mp));
- XFS_QI_UQIP(mp) = NULL;
+ if ((dqtype & XFS_QMOPT_UQUOTA) && q->qi_uquotaip) {
+ IRELE(q->qi_uquotaip);
+ q->qi_uquotaip = NULL;
}
- if ((dqtype & (XFS_QMOPT_GQUOTA|XFS_QMOPT_PQUOTA)) && XFS_QI_GQIP(mp)) {
- IRELE(XFS_QI_GQIP(mp));
- XFS_QI_GQIP(mp) = NULL;
+ if ((dqtype & (XFS_QMOPT_GQUOTA|XFS_QMOPT_PQUOTA)) && q->qi_gquotaip) {
+ IRELE(q->qi_gquotaip);
+ q->qi_gquotaip = NULL;
}
-out_error:
- mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
- return (error);
+out_unlock:
+ mutex_unlock(&q->qi_quotaofflock);
+ return error;
}
int
@@ -267,7 +262,7 @@ xfs_qm_scall_trunc_qfiles(
}
if ((flags & XFS_DQ_USER) && mp->m_sb.sb_uquotino != NULLFSINO) {
- error = xfs_iget(mp, NULL, mp->m_sb.sb_uquotino, 0, 0, &qip, 0);
+ error = xfs_iget(mp, NULL, mp->m_sb.sb_uquotino, 0, 0, &qip);
if (!error) {
error = xfs_truncate_file(mp, qip);
IRELE(qip);
@@ -276,7 +271,7 @@ xfs_qm_scall_trunc_qfiles(
if ((flags & (XFS_DQ_GROUP|XFS_DQ_PROJ)) &&
mp->m_sb.sb_gquotino != NULLFSINO) {
- error2 = xfs_iget(mp, NULL, mp->m_sb.sb_gquotino, 0, 0, &qip, 0);
+ error2 = xfs_iget(mp, NULL, mp->m_sb.sb_gquotino, 0, 0, &qip);
if (!error2) {
error2 = xfs_truncate_file(mp, qip);
IRELE(qip);
@@ -379,9 +374,9 @@ xfs_qm_scall_quotaon(
/*
* Switch on quota enforcement in core.
*/
- mutex_lock(&(XFS_QI_QOFFLOCK(mp)));
+ mutex_lock(&mp->m_quotainfo->qi_quotaofflock);
mp->m_qflags |= (flags & XFS_ALL_QUOTA_ENFD);
- mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
+ mutex_unlock(&mp->m_quotainfo->qi_quotaofflock);
return (0);
}
@@ -392,11 +387,12 @@ xfs_qm_scall_quotaon(
*/
int
xfs_qm_scall_getqstat(
- xfs_mount_t *mp,
- fs_quota_stat_t *out)
+ struct xfs_mount *mp,
+ struct fs_quota_stat *out)
{
- xfs_inode_t *uip, *gip;
- boolean_t tempuqip, tempgqip;
+ struct xfs_quotainfo *q = mp->m_quotainfo;
+ struct xfs_inode *uip, *gip;
+ boolean_t tempuqip, tempgqip;
uip = gip = NULL;
tempuqip = tempgqip = B_FALSE;
@@ -415,18 +411,18 @@ xfs_qm_scall_getqstat(
out->qs_uquota.qfs_ino = mp->m_sb.sb_uquotino;
out->qs_gquota.qfs_ino = mp->m_sb.sb_gquotino;
- if (mp->m_quotainfo) {
- uip = mp->m_quotainfo->qi_uquotaip;
- gip = mp->m_quotainfo->qi_gquotaip;
+ if (q) {
+ uip = q->qi_uquotaip;
+ gip = q->qi_gquotaip;
}
if (!uip && mp->m_sb.sb_uquotino != NULLFSINO) {
if (xfs_iget(mp, NULL, mp->m_sb.sb_uquotino,
- 0, 0, &uip, 0) == 0)
+ 0, 0, &uip) == 0)
tempuqip = B_TRUE;
}
if (!gip && mp->m_sb.sb_gquotino != NULLFSINO) {
if (xfs_iget(mp, NULL, mp->m_sb.sb_gquotino,
- 0, 0, &gip, 0) == 0)
+ 0, 0, &gip) == 0)
tempgqip = B_TRUE;
}
if (uip) {
@@ -441,17 +437,20 @@ xfs_qm_scall_getqstat(
if (tempgqip)
IRELE(gip);
}
- if (mp->m_quotainfo) {
- out->qs_incoredqs = XFS_QI_MPLNDQUOTS(mp);
- out->qs_btimelimit = XFS_QI_BTIMELIMIT(mp);
- out->qs_itimelimit = XFS_QI_ITIMELIMIT(mp);
- out->qs_rtbtimelimit = XFS_QI_RTBTIMELIMIT(mp);
- out->qs_bwarnlimit = XFS_QI_BWARNLIMIT(mp);
- out->qs_iwarnlimit = XFS_QI_IWARNLIMIT(mp);
+ if (q) {
+ out->qs_incoredqs = q->qi_dquots;
+ out->qs_btimelimit = q->qi_btimelimit;
+ out->qs_itimelimit = q->qi_itimelimit;
+ out->qs_rtbtimelimit = q->qi_rtbtimelimit;
+ out->qs_bwarnlimit = q->qi_bwarnlimit;
+ out->qs_iwarnlimit = q->qi_iwarnlimit;
}
- return (0);
+ return 0;
}
+#define XFS_DQ_MASK \
+ (FS_DQ_LIMIT_MASK | FS_DQ_TIMER_MASK | FS_DQ_WARNS_MASK)
+
/*
* Adjust quota limits, and start/stop timers accordingly.
*/
@@ -462,15 +461,17 @@ xfs_qm_scall_setqlim(
uint type,
fs_disk_quota_t *newlim)
{
+ struct xfs_quotainfo *q = mp->m_quotainfo;
xfs_disk_dquot_t *ddq;
xfs_dquot_t *dqp;
xfs_trans_t *tp;
int error;
xfs_qcnt_t hard, soft;
- if ((newlim->d_fieldmask &
- (FS_DQ_LIMIT_MASK|FS_DQ_TIMER_MASK|FS_DQ_WARNS_MASK)) == 0)
- return (0);
+ if (newlim->d_fieldmask & ~XFS_DQ_MASK)
+ return EINVAL;
+ if ((newlim->d_fieldmask & XFS_DQ_MASK) == 0)
+ return 0;
tp = xfs_trans_alloc(mp, XFS_TRANS_QM_SETQLIM);
if ((error = xfs_trans_reserve(tp, 0, sizeof(xfs_disk_dquot_t) + 128,
@@ -485,7 +486,7 @@ xfs_qm_scall_setqlim(
* a quotaoff from happening). (XXXThis doesn't currently happen
* because we take the vfslock before calling xfs_qm_sysent).
*/
- mutex_lock(&(XFS_QI_QOFFLOCK(mp)));
+ mutex_lock(&q->qi_quotaofflock);
/*
* Get the dquot (locked), and join it to the transaction.
@@ -493,9 +494,8 @@ xfs_qm_scall_setqlim(
*/
if ((error = xfs_qm_dqget(mp, NULL, id, type, XFS_QMOPT_DQALLOC, &dqp))) {
xfs_trans_cancel(tp, XFS_TRANS_ABORT);
- mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
ASSERT(error != ENOENT);
- return (error);
+ goto out_unlock;
}
xfs_trans_dqjoin(tp, dqp);
ddq = &dqp->q_core;
@@ -513,8 +513,8 @@ xfs_qm_scall_setqlim(
ddq->d_blk_hardlimit = cpu_to_be64(hard);
ddq->d_blk_softlimit = cpu_to_be64(soft);
if (id == 0) {
- mp->m_quotainfo->qi_bhardlimit = hard;
- mp->m_quotainfo->qi_bsoftlimit = soft;
+ q->qi_bhardlimit = hard;
+ q->qi_bsoftlimit = soft;
}
} else {
qdprintk("blkhard %Ld < blksoft %Ld\n", hard, soft);
@@ -529,8 +529,8 @@ xfs_qm_scall_setqlim(
ddq->d_rtb_hardlimit = cpu_to_be64(hard);
ddq->d_rtb_softlimit = cpu_to_be64(soft);
if (id == 0) {
- mp->m_quotainfo->qi_rtbhardlimit = hard;
- mp->m_quotainfo->qi_rtbsoftlimit = soft;
+ q->qi_rtbhardlimit = hard;
+ q->qi_rtbsoftlimit = soft;
}
} else {
qdprintk("rtbhard %Ld < rtbsoft %Ld\n", hard, soft);
@@ -546,8 +546,8 @@ xfs_qm_scall_setqlim(
ddq->d_ino_hardlimit = cpu_to_be64(hard);
ddq->d_ino_softlimit = cpu_to_be64(soft);
if (id == 0) {
- mp->m_quotainfo->qi_ihardlimit = hard;
- mp->m_quotainfo->qi_isoftlimit = soft;
+ q->qi_ihardlimit = hard;
+ q->qi_isoftlimit = soft;
}
} else {
qdprintk("ihard %Ld < isoft %Ld\n", hard, soft);
@@ -572,23 +572,23 @@ xfs_qm_scall_setqlim(
* for warnings.
*/
if (newlim->d_fieldmask & FS_DQ_BTIMER) {
- mp->m_quotainfo->qi_btimelimit = newlim->d_btimer;
+ q->qi_btimelimit = newlim->d_btimer;
ddq->d_btimer = cpu_to_be32(newlim->d_btimer);
}
if (newlim->d_fieldmask & FS_DQ_ITIMER) {
- mp->m_quotainfo->qi_itimelimit = newlim->d_itimer;
+ q->qi_itimelimit = newlim->d_itimer;
ddq->d_itimer = cpu_to_be32(newlim->d_itimer);
}
if (newlim->d_fieldmask & FS_DQ_RTBTIMER) {
- mp->m_quotainfo->qi_rtbtimelimit = newlim->d_rtbtimer;
+ q->qi_rtbtimelimit = newlim->d_rtbtimer;
ddq->d_rtbtimer = cpu_to_be32(newlim->d_rtbtimer);
}
if (newlim->d_fieldmask & FS_DQ_BWARNS)
- mp->m_quotainfo->qi_bwarnlimit = newlim->d_bwarns;
+ q->qi_bwarnlimit = newlim->d_bwarns;
if (newlim->d_fieldmask & FS_DQ_IWARNS)
- mp->m_quotainfo->qi_iwarnlimit = newlim->d_iwarns;
+ q->qi_iwarnlimit = newlim->d_iwarns;
if (newlim->d_fieldmask & FS_DQ_RTBWARNS)
- mp->m_quotainfo->qi_rtbwarnlimit = newlim->d_rtbwarns;
+ q->qi_rtbwarnlimit = newlim->d_rtbwarns;
} else {
/*
* If the user is now over quota, start the timelimit.
@@ -605,8 +605,9 @@ xfs_qm_scall_setqlim(
error = xfs_trans_commit(tp, 0);
xfs_qm_dqprint(dqp);
xfs_qm_dqrele(dqp);
- mutex_unlock(&(XFS_QI_QOFFLOCK(mp)));
+ out_unlock:
+ mutex_unlock(&q->qi_quotaofflock);
return error;
}
@@ -853,7 +854,8 @@ xfs_dqrele_inode(
int error;
/* skip quota inodes */
- if (ip == XFS_QI_UQIP(ip->i_mount) || ip == XFS_QI_GQIP(ip->i_mount)) {
+ if (ip == ip->i_mount->m_quotainfo->qi_uquotaip ||
+ ip == ip->i_mount->m_quotainfo->qi_gquotaip) {
ASSERT(ip->i_udquot == NULL);
ASSERT(ip->i_gdquot == NULL);
read_unlock(&pag->pag_ici_lock);
@@ -891,7 +893,8 @@ xfs_qm_dqrele_all_inodes(
uint flags)
{
ASSERT(mp->m_quotainfo);
- xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags, XFS_ICI_NO_TAG, 0);
+ xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags,
+ XFS_ICI_NO_TAG, 0, NULL);
}
/*------------------------------------------------------------------------*/
@@ -930,7 +933,8 @@ struct mutex qcheck_lock;
}
typedef struct dqtest {
- xfs_dqmarker_t q_lists;
+ uint dq_flags; /* various flags (XFS_DQ_*) */
+ struct list_head q_hashlist;
xfs_dqhash_t *q_hash; /* the hashchain header */
xfs_mount_t *q_mount; /* filesystem this relates to */
xfs_dqid_t d_id; /* user id or group id */
@@ -941,14 +945,9 @@ typedef struct dqtest {
STATIC void
xfs_qm_hashinsert(xfs_dqhash_t *h, xfs_dqtest_t *dqp)
{
- xfs_dquot_t *d;
- if (((d) = (h)->qh_next))
- (d)->HL_PREVP = &((dqp)->HL_NEXT);
- (dqp)->HL_NEXT = d;
- (dqp)->HL_PREVP = &((h)->qh_next);
- (h)->qh_next = (xfs_dquot_t *)dqp;
- (h)->qh_version++;
- (h)->qh_nelems++;
+ list_add(&dqp->q_hashlist, &h->qh_list);
+ h->qh_version++;
+ h->qh_nelems++;
}
STATIC void
xfs_qm_dqtest_print(
@@ -1060,9 +1059,7 @@ xfs_qm_internalqcheck_dqget(
xfs_dqhash_t *h;
h = DQTEST_HASH(mp, id, type);
- for (d = (xfs_dqtest_t *) h->qh_next; d != NULL;
- d = (xfs_dqtest_t *) d->HL_NEXT) {
- /* DQTEST_LIST_PRINT(h, HL_NEXT, "@@@@@ dqtestlist @@@@@"); */
+ list_for_each_entry(d, &h->qh_list, q_hashlist) {
if (d->d_id == id && mp == d->q_mount) {
*O_dq = d;
return (0);
@@ -1073,6 +1070,7 @@ xfs_qm_internalqcheck_dqget(
d->d_id = id;
d->q_mount = mp;
d->q_hash = h;
+ INIT_LIST_HEAD(&d->q_hashlist);
xfs_qm_hashinsert(h, d);
*O_dq = d;
return (0);
@@ -1111,10 +1109,7 @@ xfs_qm_internalqcheck_adjust(
xfs_ino_t ino, /* inode number to get data for */
void __user *buffer, /* not used */
int ubsize, /* not used */
- void *private_data, /* not used */
- xfs_daddr_t bno, /* starting block of inode cluster */
int *ubused, /* not used */
- void *dip, /* not used */
int *res) /* bulkstat result code */
{
xfs_inode_t *ip;
@@ -1136,7 +1131,7 @@ xfs_qm_internalqcheck_adjust(
ipreleased = B_FALSE;
again:
lock_flags = XFS_ILOCK_SHARED;
- if ((error = xfs_iget(mp, NULL, ino, 0, lock_flags, &ip, bno))) {
+ if ((error = xfs_iget(mp, NULL, ino, 0, lock_flags, &ip))) {
*res = BULKSTAT_RV_NOTHING;
return (error);
}
@@ -1179,8 +1174,6 @@ xfs_qm_internalqcheck(
xfs_ino_t lastino;
int done, count;
int i;
- xfs_dqtest_t *d, *e;
- xfs_dqhash_t *h1;
int error;
lastino = 0;
@@ -1192,9 +1185,9 @@ xfs_qm_internalqcheck(
if (! XFS_IS_QUOTA_ON(mp))
return XFS_ERROR(ESRCH);
- xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE | XFS_LOG_SYNC);
+ xfs_log_force(mp, XFS_LOG_SYNC);
XFS_bflush(mp->m_ddev_targp);
- xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE | XFS_LOG_SYNC);
+ xfs_log_force(mp, XFS_LOG_SYNC);
XFS_bflush(mp->m_ddev_targp);
mutex_lock(&qcheck_lock);
@@ -1209,30 +1202,29 @@ xfs_qm_internalqcheck(
* Iterate thru all the inodes in the file system,
* adjusting the corresponding dquot counters
*/
- if ((error = xfs_bulkstat(mp, &lastino, &count,
- xfs_qm_internalqcheck_adjust, NULL,
- 0, NULL, BULKSTAT_FG_IGET, &done))) {
+ error = xfs_bulkstat(mp, &lastino, &count,
+ xfs_qm_internalqcheck_adjust,
+ 0, NULL, &done);
+ if (error) {
+ cmn_err(CE_DEBUG, "Bulkstat returned error 0x%x", error);
break;
}
- } while (! done);
- if (error) {
- cmn_err(CE_DEBUG, "Bulkstat returned error 0x%x", error);
- }
+ } while (!done);
+
cmn_err(CE_DEBUG, "Checking results against system dquots");
for (i = 0; i < qmtest_hashmask; i++) {
- h1 = &qmtest_udqtab[i];
- for (d = (xfs_dqtest_t *) h1->qh_next; d != NULL; ) {
+ xfs_dqtest_t *d, *n;
+ xfs_dqhash_t *h;
+
+ h = &qmtest_udqtab[i];
+ list_for_each_entry_safe(d, n, &h->qh_list, q_hashlist) {
xfs_dqtest_cmp(d);
- e = (xfs_dqtest_t *) d->HL_NEXT;
kmem_free(d);
- d = e;
}
- h1 = &qmtest_gdqtab[i];
- for (d = (xfs_dqtest_t *) h1->qh_next; d != NULL; ) {
+ h = &qmtest_gdqtab[i];
+ list_for_each_entry_safe(d, n, &h->qh_list, q_hashlist) {
xfs_dqtest_cmp(d);
- e = (xfs_dqtest_t *) d->HL_NEXT;
kmem_free(d);
- d = e;
}
}
diff --git a/fs/xfs/quota/xfs_quota_priv.h b/fs/xfs/quota/xfs_quota_priv.h
index 8286b2842b6b..94a3d927d716 100644
--- a/fs/xfs/quota/xfs_quota_priv.h
+++ b/fs/xfs/quota/xfs_quota_priv.h
@@ -24,43 +24,6 @@
*/
#define XFS_DQITER_MAP_SIZE 10
-/* Number of dquots that fit in to a dquot block */
-#define XFS_QM_DQPERBLK(mp) ((mp)->m_quotainfo->qi_dqperchunk)
-
-#define XFS_DQ_IS_ADDEDTO_TRX(t, d) ((d)->q_transp == (t))
-
-#define XFS_QI_MPLRECLAIMS(mp) ((mp)->m_quotainfo->qi_dqreclaims)
-#define XFS_QI_UQIP(mp) ((mp)->m_quotainfo->qi_uquotaip)
-#define XFS_QI_GQIP(mp) ((mp)->m_quotainfo->qi_gquotaip)
-#define XFS_QI_DQCHUNKLEN(mp) ((mp)->m_quotainfo->qi_dqchunklen)
-#define XFS_QI_BTIMELIMIT(mp) ((mp)->m_quotainfo->qi_btimelimit)
-#define XFS_QI_RTBTIMELIMIT(mp) ((mp)->m_quotainfo->qi_rtbtimelimit)
-#define XFS_QI_ITIMELIMIT(mp) ((mp)->m_quotainfo->qi_itimelimit)
-#define XFS_QI_BWARNLIMIT(mp) ((mp)->m_quotainfo->qi_bwarnlimit)
-#define XFS_QI_RTBWARNLIMIT(mp) ((mp)->m_quotainfo->qi_rtbwarnlimit)
-#define XFS_QI_IWARNLIMIT(mp) ((mp)->m_quotainfo->qi_iwarnlimit)
-#define XFS_QI_QOFFLOCK(mp) ((mp)->m_quotainfo->qi_quotaofflock)
-
-#define XFS_QI_MPL_LIST(mp) ((mp)->m_quotainfo->qi_dqlist)
-#define XFS_QI_MPLNEXT(mp) ((mp)->m_quotainfo->qi_dqlist.qh_next)
-#define XFS_QI_MPLNDQUOTS(mp) ((mp)->m_quotainfo->qi_dqlist.qh_nelems)
-
-#define xfs_qm_mplist_lock(mp) \
- mutex_lock(&(XFS_QI_MPL_LIST(mp).qh_lock))
-#define xfs_qm_mplist_nowait(mp) \
- mutex_trylock(&(XFS_QI_MPL_LIST(mp).qh_lock))
-#define xfs_qm_mplist_unlock(mp) \
- mutex_unlock(&(XFS_QI_MPL_LIST(mp).qh_lock))
-#define XFS_QM_IS_MPLIST_LOCKED(mp) \
- mutex_is_locked(&(XFS_QI_MPL_LIST(mp).qh_lock))
-
-#define xfs_qm_freelist_lock(qm) \
- mutex_lock(&((qm)->qm_dqfreelist.qh_lock))
-#define xfs_qm_freelist_lock_nowait(qm) \
- mutex_trylock(&((qm)->qm_dqfreelist.qh_lock))
-#define xfs_qm_freelist_unlock(qm) \
- mutex_unlock(&((qm)->qm_dqfreelist.qh_lock))
-
/*
* Hash into a bucket in the dquot hash table, based on <mp, id>.
*/
@@ -72,9 +35,6 @@
XFS_DQ_HASHVAL(mp, id)) : \
(xfs_Gqm->qm_grp_dqhtable + \
XFS_DQ_HASHVAL(mp, id)))
-#define XFS_IS_DQTYPE_ON(mp, type) (type == XFS_DQ_USER ? \
- XFS_IS_UQUOTA_ON(mp) : \
- XFS_IS_OQUOTA_ON(mp))
#define XFS_IS_DQUOT_UNINITIALIZED(dqp) ( \
!dqp->q_core.d_blk_hardlimit && \
!dqp->q_core.d_blk_softlimit && \
@@ -86,68 +46,6 @@
!dqp->q_core.d_rtbcount && \
!dqp->q_core.d_icount)
-#define HL_PREVP dq_hashlist.ql_prevp
-#define HL_NEXT dq_hashlist.ql_next
-#define MPL_PREVP dq_mplist.ql_prevp
-#define MPL_NEXT dq_mplist.ql_next
-
-
-#define _LIST_REMOVE(h, dqp, PVP, NXT) \
- { \
- xfs_dquot_t *d; \
- if (((d) = (dqp)->NXT)) \
- (d)->PVP = (dqp)->PVP; \
- *((dqp)->PVP) = d; \
- (dqp)->NXT = NULL; \
- (dqp)->PVP = NULL; \
- (h)->qh_version++; \
- (h)->qh_nelems--; \
- }
-
-#define _LIST_INSERT(h, dqp, PVP, NXT) \
- { \
- xfs_dquot_t *d; \
- if (((d) = (h)->qh_next)) \
- (d)->PVP = &((dqp)->NXT); \
- (dqp)->NXT = d; \
- (dqp)->PVP = &((h)->qh_next); \
- (h)->qh_next = dqp; \
- (h)->qh_version++; \
- (h)->qh_nelems++; \
- }
-
-#define FOREACH_DQUOT_IN_MP(dqp, mp) \
- for ((dqp) = XFS_QI_MPLNEXT(mp); (dqp) != NULL; (dqp) = (dqp)->MPL_NEXT)
-
-#define FOREACH_DQUOT_IN_FREELIST(dqp, qlist) \
-for ((dqp) = (qlist)->qh_next; (dqp) != (xfs_dquot_t *)(qlist); \
- (dqp) = (dqp)->dq_flnext)
-
-#define XQM_HASHLIST_INSERT(h, dqp) \
- _LIST_INSERT(h, dqp, HL_PREVP, HL_NEXT)
-
-#define XQM_FREELIST_INSERT(h, dqp) \
- xfs_qm_freelist_append(h, dqp)
-
-#define XQM_MPLIST_INSERT(h, dqp) \
- _LIST_INSERT(h, dqp, MPL_PREVP, MPL_NEXT)
-
-#define XQM_HASHLIST_REMOVE(h, dqp) \
- _LIST_REMOVE(h, dqp, HL_PREVP, HL_NEXT)
-#define XQM_FREELIST_REMOVE(dqp) \
- xfs_qm_freelist_unlink(dqp)
-#define XQM_MPLIST_REMOVE(h, dqp) \
- { _LIST_REMOVE(h, dqp, MPL_PREVP, MPL_NEXT); \
- XFS_QI_MPLRECLAIMS((dqp)->q_mount)++; }
-
-#define XFS_DQ_IS_LOGITEM_INITD(dqp) ((dqp)->q_logitem.qli_dquot == (dqp))
-
-#define XFS_QM_DQP_TO_DQACCT(tp, dqp) (XFS_QM_ISUDQ(dqp) ? \
- (tp)->t_dqinfo->dqa_usrdquots : \
- (tp)->t_dqinfo->dqa_grpdquots)
-#define XFS_IS_SUSER_DQUOT(dqp) \
- (!((dqp)->q_core.d_id))
-
#define DQFLAGTO_TYPESTR(d) (((d)->dq_flags & XFS_DQ_USER) ? "USR" : \
(((d)->dq_flags & XFS_DQ_GROUP) ? "GRP" : \
(((d)->dq_flags & XFS_DQ_PROJ) ? "PRJ":"???")))
diff --git a/fs/xfs/quota/xfs_trans_dquot.c b/fs/xfs/quota/xfs_trans_dquot.c
index 97ac9640be98..061d827da33c 100644
--- a/fs/xfs/quota/xfs_trans_dquot.c
+++ b/fs/xfs/quota/xfs_trans_dquot.c
@@ -59,12 +59,11 @@ xfs_trans_dqjoin(
xfs_trans_t *tp,
xfs_dquot_t *dqp)
{
- xfs_dq_logitem_t *lp;
+ xfs_dq_logitem_t *lp = &dqp->q_logitem;
- ASSERT(! XFS_DQ_IS_ADDEDTO_TRX(tp, dqp));
+ ASSERT(dqp->q_transp != tp);
ASSERT(XFS_DQ_IS_LOCKED(dqp));
- ASSERT(XFS_DQ_IS_LOGITEM_INITD(dqp));
- lp = &dqp->q_logitem;
+ ASSERT(lp->qli_dquot == dqp);
/*
* Get a log_item_desc to point at the new item.
@@ -96,7 +95,7 @@ xfs_trans_log_dquot(
{
xfs_log_item_desc_t *lidp;
- ASSERT(XFS_DQ_IS_ADDEDTO_TRX(tp, dqp));
+ ASSERT(dqp->q_transp == tp);
ASSERT(XFS_DQ_IS_LOCKED(dqp));
lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)(&dqp->q_logitem));
@@ -198,16 +197,16 @@ xfs_trans_get_dqtrx(
int i;
xfs_dqtrx_t *qa;
- for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) {
- qa = XFS_QM_DQP_TO_DQACCT(tp, dqp);
+ qa = XFS_QM_ISUDQ(dqp) ?
+ tp->t_dqinfo->dqa_usrdquots : tp->t_dqinfo->dqa_grpdquots;
+ for (i = 0; i < XFS_QM_TRANS_MAXDQS; i++) {
if (qa[i].qt_dquot == NULL ||
- qa[i].qt_dquot == dqp) {
- return (&qa[i]);
- }
+ qa[i].qt_dquot == dqp)
+ return &qa[i];
}
- return (NULL);
+ return NULL;
}
/*
@@ -381,7 +380,7 @@ xfs_trans_apply_dquot_deltas(
break;
ASSERT(XFS_DQ_IS_LOCKED(dqp));
- ASSERT(XFS_DQ_IS_ADDEDTO_TRX(tp, dqp));
+ ASSERT(dqp->q_transp == tp);
/*
* adjust the actual number of blocks used
@@ -589,12 +588,18 @@ xfs_trans_unreserve_and_mod_dquots(
}
}
-STATIC int
-xfs_quota_error(uint flags)
+STATIC void
+xfs_quota_warn(
+ struct xfs_mount *mp,
+ struct xfs_dquot *dqp,
+ int type)
{
- if (flags & XFS_QMOPT_ENOSPC)
- return ENOSPC;
- return EDQUOT;
+ /* no warnings for project quotas - we just return ENOSPC later */
+ if (dqp->dq_flags & XFS_DQ_PROJ)
+ return;
+ quota_send_warning((dqp->dq_flags & XFS_DQ_USER) ? USRQUOTA : GRPQUOTA,
+ be32_to_cpu(dqp->q_core.d_id), mp->m_super->s_dev,
+ type);
}
/*
@@ -612,7 +617,6 @@ xfs_trans_dqresv(
long ninos,
uint flags)
{
- int error;
xfs_qcnt_t hardlimit;
xfs_qcnt_t softlimit;
time_t timer;
@@ -634,7 +638,7 @@ xfs_trans_dqresv(
softlimit = q->qi_bsoftlimit;
timer = be32_to_cpu(dqp->q_core.d_btimer);
warns = be16_to_cpu(dqp->q_core.d_bwarns);
- warnlimit = XFS_QI_BWARNLIMIT(dqp->q_mount);
+ warnlimit = dqp->q_mount->m_quotainfo->qi_bwarnlimit;
resbcountp = &dqp->q_res_bcount;
} else {
ASSERT(flags & XFS_TRANS_DQ_RES_RTBLKS);
@@ -646,10 +650,9 @@ xfs_trans_dqresv(
softlimit = q->qi_rtbsoftlimit;
timer = be32_to_cpu(dqp->q_core.d_rtbtimer);
warns = be16_to_cpu(dqp->q_core.d_rtbwarns);
- warnlimit = XFS_QI_RTBWARNLIMIT(dqp->q_mount);
+ warnlimit = dqp->q_mount->m_quotainfo->qi_rtbwarnlimit;
resbcountp = &dqp->q_res_rtbcount;
}
- error = 0;
if ((flags & XFS_QMOPT_FORCE_RES) == 0 &&
dqp->q_core.d_id &&
@@ -667,40 +670,46 @@ xfs_trans_dqresv(
* nblks.
*/
if (hardlimit > 0ULL &&
- (hardlimit <= nblks + *resbcountp)) {
- error = xfs_quota_error(flags);
+ hardlimit <= nblks + *resbcountp) {
+ xfs_quota_warn(mp, dqp, QUOTA_NL_BHARDWARN);
goto error_return;
}
-
if (softlimit > 0ULL &&
- (softlimit <= nblks + *resbcountp)) {
+ softlimit <= nblks + *resbcountp) {
if ((timer != 0 && get_seconds() > timer) ||
(warns != 0 && warns >= warnlimit)) {
- error = xfs_quota_error(flags);
+ xfs_quota_warn(mp, dqp,
+ QUOTA_NL_BSOFTLONGWARN);
goto error_return;
}
+
+ xfs_quota_warn(mp, dqp, QUOTA_NL_BSOFTWARN);
}
}
if (ninos > 0) {
count = be64_to_cpu(dqp->q_core.d_icount);
timer = be32_to_cpu(dqp->q_core.d_itimer);
warns = be16_to_cpu(dqp->q_core.d_iwarns);
- warnlimit = XFS_QI_IWARNLIMIT(dqp->q_mount);
+ warnlimit = dqp->q_mount->m_quotainfo->qi_iwarnlimit;
hardlimit = be64_to_cpu(dqp->q_core.d_ino_hardlimit);
if (!hardlimit)
hardlimit = q->qi_ihardlimit;
softlimit = be64_to_cpu(dqp->q_core.d_ino_softlimit);
if (!softlimit)
softlimit = q->qi_isoftlimit;
+
if (hardlimit > 0ULL && count >= hardlimit) {
- error = xfs_quota_error(flags);
+ xfs_quota_warn(mp, dqp, QUOTA_NL_IHARDWARN);
goto error_return;
- } else if (softlimit > 0ULL && count >= softlimit) {
- if ((timer != 0 && get_seconds() > timer) ||
+ }
+ if (softlimit > 0ULL && count >= softlimit) {
+ if ((timer != 0 && get_seconds() > timer) ||
(warns != 0 && warns >= warnlimit)) {
- error = xfs_quota_error(flags);
+ xfs_quota_warn(mp, dqp,
+ QUOTA_NL_ISOFTLONGWARN);
goto error_return;
}
+ xfs_quota_warn(mp, dqp, QUOTA_NL_ISOFTWARN);
}
}
}
@@ -736,9 +745,14 @@ xfs_trans_dqresv(
ASSERT(dqp->q_res_rtbcount >= be64_to_cpu(dqp->q_core.d_rtbcount));
ASSERT(dqp->q_res_icount >= be64_to_cpu(dqp->q_core.d_icount));
+ xfs_dqunlock(dqp);
+ return 0;
+
error_return:
xfs_dqunlock(dqp);
- return error;
+ if (flags & XFS_QMOPT_ENOSPC)
+ return ENOSPC;
+ return EDQUOT;
}
diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h
index 00fd357c3e46..0135e2a669d7 100644
--- a/fs/xfs/xfs_acl.h
+++ b/fs/xfs/xfs_acl.h
@@ -36,8 +36,8 @@ struct xfs_acl {
};
/* On-disk XFS extended attribute names */
-#define SGI_ACL_FILE "SGI_ACL_FILE"
-#define SGI_ACL_DEFAULT "SGI_ACL_DEFAULT"
+#define SGI_ACL_FILE (unsigned char *)"SGI_ACL_FILE"
+#define SGI_ACL_DEFAULT (unsigned char *)"SGI_ACL_DEFAULT"
#define SGI_ACL_FILE_SIZE (sizeof(SGI_ACL_FILE)-1)
#define SGI_ACL_DEFAULT_SIZE (sizeof(SGI_ACL_DEFAULT)-1)
@@ -49,8 +49,8 @@ extern int xfs_acl_chmod(struct inode *inode);
extern int posix_acl_access_exists(struct inode *inode);
extern int posix_acl_default_exists(struct inode *inode);
-extern struct xattr_handler xfs_xattr_acl_access_handler;
-extern struct xattr_handler xfs_xattr_acl_default_handler;
+extern const struct xattr_handler xfs_xattr_acl_access_handler;
+extern const struct xattr_handler xfs_xattr_acl_default_handler;
#else
# define xfs_check_acl NULL
# define xfs_get_acl(inode, type) NULL
diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h
index 6702bd865811..4917d4eed4ed 100644
--- a/fs/xfs/xfs_ag.h
+++ b/fs/xfs/xfs_ag.h
@@ -175,29 +175,31 @@ typedef struct xfs_agfl {
} xfs_agfl_t;
/*
- * Busy block/extent entry. Used in perag to mark blocks that have been freed
- * but whose transactions aren't committed to disk yet.
+ * Busy block/extent entry. Indexed by a rbtree in perag to mark blocks that
+ * have been freed but whose transactions aren't committed to disk yet.
+ *
+ * Note that we use the transaction ID to record the transaction, not the
+ * transaction structure itself. See xfs_alloc_busy_insert() for details.
*/
-typedef struct xfs_perag_busy {
- xfs_agblock_t busy_start;
- xfs_extlen_t busy_length;
- struct xfs_trans *busy_tp; /* transaction that did the free */
-} xfs_perag_busy_t;
+struct xfs_busy_extent {
+ struct rb_node rb_node; /* ag by-bno indexed search tree */
+ struct list_head list; /* transaction busy extent list */
+ xfs_agnumber_t agno;
+ xfs_agblock_t bno;
+ xfs_extlen_t length;
+ xlog_tid_t tid; /* transaction that created this */
+};
/*
* Per-ag incore structure, copies of information in agf and agi,
* to improve the performance of allocation group selection.
- *
- * pick sizes which fit in allocation buckets well
*/
-#if (BITS_PER_LONG == 32)
-#define XFS_PAGB_NUM_SLOTS 84
-#elif (BITS_PER_LONG == 64)
#define XFS_PAGB_NUM_SLOTS 128
-#endif
-typedef struct xfs_perag
-{
+typedef struct xfs_perag {
+ struct xfs_mount *pag_mount; /* owner filesystem */
+ xfs_agnumber_t pag_agno; /* AG this structure belongs to */
+ atomic_t pag_ref; /* perag reference count */
char pagf_init; /* this agf's entry is initialized */
char pagi_init; /* this agi's entry is initialized */
char pagf_metadata; /* the agf is preferred to be metadata */
@@ -210,8 +212,6 @@ typedef struct xfs_perag
__uint32_t pagf_btreeblks; /* # of blocks held in AGF btrees */
xfs_agino_t pagi_freecount; /* number of free inodes */
xfs_agino_t pagi_count; /* number of allocated inodes */
- int pagb_count; /* pagb slots in use */
- xfs_perag_busy_t *pagb_list; /* unstable blocks */
/*
* Inode allocation search lookup optimisation.
@@ -222,14 +222,16 @@ typedef struct xfs_perag
xfs_agino_t pagl_leftrec;
xfs_agino_t pagl_rightrec;
#ifdef __KERNEL__
- spinlock_t pagb_lock; /* lock for pagb_list */
+ spinlock_t pagb_lock; /* lock for pagb_tree */
+ struct rb_root pagb_tree; /* ordered tree of busy extents */
atomic_t pagf_fstrms; /* # of filestreams active in this AG */
- int pag_ici_init; /* incore inode cache initialised */
rwlock_t pag_ici_lock; /* incore inode lock */
struct radix_tree_root pag_ici_root; /* incore inode cache root */
+ int pag_ici_reclaimable; /* reclaimable inodes */
#endif
+ int pagb_count; /* pagb slots in use */
} xfs_perag_t;
/*
diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c
index 275b1f4f9430..a7fbe8a99b12 100644
--- a/fs/xfs/xfs_alloc.c
+++ b/fs/xfs/xfs_alloc.c
@@ -46,11 +46,9 @@
#define XFSA_FIXUP_BNO_OK 1
#define XFSA_FIXUP_CNT_OK 2
-STATIC void
-xfs_alloc_search_busy(xfs_trans_t *tp,
- xfs_agnumber_t agno,
- xfs_agblock_t bno,
- xfs_extlen_t len);
+static int
+xfs_alloc_busy_search(struct xfs_mount *mp, xfs_agnumber_t agno,
+ xfs_agblock_t bno, xfs_extlen_t len);
/*
* Prototypes for per-ag allocation routines
@@ -540,9 +538,16 @@ xfs_alloc_ag_vextent(
be32_to_cpu(agf->agf_length));
xfs_alloc_log_agf(args->tp, args->agbp,
XFS_AGF_FREEBLKS);
- /* search the busylist for these blocks */
- xfs_alloc_search_busy(args->tp, args->agno,
- args->agbno, args->len);
+ /*
+ * Search the busylist for these blocks and mark the
+ * transaction as synchronous if blocks are found. This
+ * avoids the need to block due to a synchronous log
+ * force to ensure correct ordering as the synchronous
+ * transaction will guarantee that for us.
+ */
+ if (xfs_alloc_busy_search(args->mp, args->agno,
+ args->agbno, args->len))
+ xfs_trans_set_sync(args->tp);
}
if (!args->isfl)
xfs_trans_mod_sb(args->tp,
@@ -1662,11 +1667,13 @@ xfs_free_ag_extent(
xfs_agf_t *agf;
xfs_perag_t *pag; /* per allocation group data */
+ pag = xfs_perag_get(mp, agno);
+ pag->pagf_freeblks += len;
+ xfs_perag_put(pag);
+
agf = XFS_BUF_TO_AGF(agbp);
- pag = &mp->m_perag[agno];
be32_add_cpu(&agf->agf_freeblks, len);
xfs_trans_agblocks_delta(tp, len);
- pag->pagf_freeblks += len;
XFS_WANT_CORRUPTED_GOTO(
be32_to_cpu(agf->agf_freeblks) <=
be32_to_cpu(agf->agf_length),
@@ -1691,7 +1698,7 @@ xfs_free_ag_extent(
* when the iclog commits to disk. If a busy block is allocated,
* the iclog is pushed up to the LSN that freed the block.
*/
- xfs_alloc_mark_busy(tp, agno, bno, len);
+ xfs_alloc_busy_insert(tp, agno, bno, len);
return 0;
error0:
@@ -1969,10 +1976,12 @@ xfs_alloc_get_freelist(
xfs_trans_brelse(tp, agflbp);
if (be32_to_cpu(agf->agf_flfirst) == XFS_AGFL_SIZE(mp))
agf->agf_flfirst = 0;
- pag = &mp->m_perag[be32_to_cpu(agf->agf_seqno)];
+
+ pag = xfs_perag_get(mp, be32_to_cpu(agf->agf_seqno));
be32_add_cpu(&agf->agf_flcount, -1);
xfs_trans_agflist_delta(tp, -1);
pag->pagf_flcount--;
+ xfs_perag_put(pag);
logflags = XFS_AGF_FLFIRST | XFS_AGF_FLCOUNT;
if (btreeblk) {
@@ -1985,14 +1994,20 @@ xfs_alloc_get_freelist(
*bnop = bno;
/*
- * As blocks are freed, they are added to the per-ag busy list
- * and remain there until the freeing transaction is committed to
- * disk. Now that we have allocated blocks, this list must be
- * searched to see if a block is being reused. If one is, then
- * the freeing transaction must be pushed to disk NOW by forcing
- * to disk all iclogs up that transaction's LSN.
+ * As blocks are freed, they are added to the per-ag busy list and
+ * remain there until the freeing transaction is committed to disk.
+ * Now that we have allocated blocks, this list must be searched to see
+ * if a block is being reused. If one is, then the freeing transaction
+ * must be pushed to disk before this transaction.
+ *
+ * We do this by setting the current transaction to a sync transaction
+ * which guarantees that the freeing transaction is on disk before this
+ * transaction. This is done instead of a synchronous log force here so
+ * that we don't sit and wait with the AGF locked in the transaction
+ * during the log force.
*/
- xfs_alloc_search_busy(tp, be32_to_cpu(agf->agf_seqno), bno, 1);
+ if (xfs_alloc_busy_search(mp, be32_to_cpu(agf->agf_seqno), bno, 1))
+ xfs_trans_set_sync(tp);
return 0;
}
@@ -2078,7 +2093,8 @@ xfs_alloc_put_freelist(
be32_add_cpu(&agf->agf_fllast, 1);
if (be32_to_cpu(agf->agf_fllast) == XFS_AGFL_SIZE(mp))
agf->agf_fllast = 0;
- pag = &mp->m_perag[be32_to_cpu(agf->agf_seqno)];
+
+ pag = xfs_perag_get(mp, be32_to_cpu(agf->agf_seqno));
be32_add_cpu(&agf->agf_flcount, 1);
xfs_trans_agflist_delta(tp, 1);
pag->pagf_flcount++;
@@ -2089,6 +2105,7 @@ xfs_alloc_put_freelist(
pag->pagf_btreeblks--;
logflags |= XFS_AGF_BTREEBLKS;
}
+ xfs_perag_put(pag);
xfs_alloc_log_agf(tp, agbp, logflags);
@@ -2152,7 +2169,6 @@ xfs_read_agf(
xfs_trans_brelse(tp, *bpp);
return XFS_ERROR(EFSCORRUPTED);
}
-
XFS_BUF_SET_VTYPE_REF(*bpp, B_FS_AGF, XFS_AGF_REF);
return 0;
}
@@ -2175,7 +2191,7 @@ xfs_alloc_read_agf(
ASSERT(agno != NULLAGNUMBER);
error = xfs_read_agf(mp, tp, agno,
- (flags & XFS_ALLOC_FLAG_TRYLOCK) ? XFS_BUF_TRYLOCK : 0,
+ (flags & XFS_ALLOC_FLAG_TRYLOCK) ? XBF_TRYLOCK : 0,
bpp);
if (error)
return error;
@@ -2184,7 +2200,7 @@ xfs_alloc_read_agf(
ASSERT(!XFS_BUF_GETERROR(*bpp));
agf = XFS_BUF_TO_AGF(*bpp);
- pag = &mp->m_perag[agno];
+ pag = xfs_perag_get(mp, agno);
if (!pag->pagf_init) {
pag->pagf_freeblks = be32_to_cpu(agf->agf_freeblks);
pag->pagf_btreeblks = be32_to_cpu(agf->agf_btreeblks);
@@ -2195,8 +2211,8 @@ xfs_alloc_read_agf(
pag->pagf_levels[XFS_BTNUM_CNTi] =
be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNTi]);
spin_lock_init(&pag->pagb_lock);
- pag->pagb_list = kmem_zalloc(XFS_PAGB_NUM_SLOTS *
- sizeof(xfs_perag_busy_t), KM_SLEEP);
+ pag->pagb_count = 0;
+ pag->pagb_tree = RB_ROOT;
pag->pagf_init = 1;
}
#ifdef DEBUG
@@ -2211,6 +2227,7 @@ xfs_alloc_read_agf(
be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNTi]));
}
#endif
+ xfs_perag_put(pag);
return 0;
}
@@ -2270,8 +2287,7 @@ xfs_alloc_vextent(
* These three force us into a single a.g.
*/
args->agno = XFS_FSB_TO_AGNO(mp, args->fsbno);
- down_read(&mp->m_peraglock);
- args->pag = &mp->m_perag[args->agno];
+ args->pag = xfs_perag_get(mp, args->agno);
args->minleft = 0;
error = xfs_alloc_fix_freelist(args, 0);
args->minleft = minleft;
@@ -2280,14 +2296,12 @@ xfs_alloc_vextent(
goto error0;
}
if (!args->agbp) {
- up_read(&mp->m_peraglock);
trace_xfs_alloc_vextent_noagbp(args);
break;
}
args->agbno = XFS_FSB_TO_AGBNO(mp, args->fsbno);
if ((error = xfs_alloc_ag_vextent(args)))
goto error0;
- up_read(&mp->m_peraglock);
break;
case XFS_ALLOCTYPE_START_BNO:
/*
@@ -2339,9 +2353,8 @@ xfs_alloc_vextent(
* Loop over allocation groups twice; first time with
* trylock set, second time without.
*/
- down_read(&mp->m_peraglock);
for (;;) {
- args->pag = &mp->m_perag[args->agno];
+ args->pag = xfs_perag_get(mp, args->agno);
if (no_min) args->minleft = 0;
error = xfs_alloc_fix_freelist(args, flags);
args->minleft = minleft;
@@ -2400,8 +2413,8 @@ xfs_alloc_vextent(
}
}
}
+ xfs_perag_put(args->pag);
}
- up_read(&mp->m_peraglock);
if (bump_rotor || (type == XFS_ALLOCTYPE_ANY_AG)) {
if (args->agno == sagno)
mp->m_agfrotor = (mp->m_agfrotor + 1) %
@@ -2427,9 +2440,10 @@ xfs_alloc_vextent(
args->len);
#endif
}
+ xfs_perag_put(args->pag);
return 0;
error0:
- up_read(&mp->m_peraglock);
+ xfs_perag_put(args->pag);
return error;
}
@@ -2454,8 +2468,7 @@ xfs_free_extent(
args.agno = XFS_FSB_TO_AGNO(args.mp, bno);
ASSERT(args.agno < args.mp->m_sb.sb_agcount);
args.agbno = XFS_FSB_TO_AGBNO(args.mp, bno);
- down_read(&args.mp->m_peraglock);
- args.pag = &args.mp->m_perag[args.agno];
+ args.pag = xfs_perag_get(args.mp, args.agno);
if ((error = xfs_alloc_fix_freelist(&args, XFS_ALLOC_FLAG_FREEING)))
goto error0;
#ifdef DEBUG
@@ -2465,7 +2478,7 @@ xfs_free_extent(
#endif
error = xfs_free_ag_extent(tp, args.agbp, args.agno, args.agbno, len, 0);
error0:
- up_read(&args.mp->m_peraglock);
+ xfs_perag_put(args.pag);
return error;
}
@@ -2477,127 +2490,263 @@ error0:
* list is reused, the transaction that freed it must be forced to disk
* before continuing to use the block.
*
- * xfs_alloc_mark_busy - add to the per-ag busy list
- * xfs_alloc_clear_busy - remove an item from the per-ag busy list
+ * xfs_alloc_busy_insert - add to the per-ag busy list
+ * xfs_alloc_busy_clear - remove an item from the per-ag busy list
+ * xfs_alloc_busy_search - search for a busy extent
+ */
+
+/*
+ * Insert a new extent into the busy tree.
+ *
+ * The busy extent tree is indexed by the start block of the busy extent.
+ * there can be multiple overlapping ranges in the busy extent tree but only
+ * ever one entry at a given start block. The reason for this is that
+ * multi-block extents can be freed, then smaller chunks of that extent
+ * allocated and freed again before the first transaction commit is on disk.
+ * If the exact same start block is freed a second time, we have to wait for
+ * that busy extent to pass out of the tree before the new extent is inserted.
+ * There are two main cases we have to handle here.
+ *
+ * The first case is a transaction that triggers a "free - allocate - free"
+ * cycle. This can occur during btree manipulations as a btree block is freed
+ * to the freelist, then allocated from the free list, then freed again. In
+ * this case, the second extxpnet free is what triggers the duplicate and as
+ * such the transaction IDs should match. Because the extent was allocated in
+ * this transaction, the transaction must be marked as synchronous. This is
+ * true for all cases where the free/alloc/free occurs in the one transaction,
+ * hence the addition of the ASSERT(tp->t_flags & XFS_TRANS_SYNC) to this case.
+ * This serves to catch violations of the second case quite effectively.
+ *
+ * The second case is where the free/alloc/free occur in different
+ * transactions. In this case, the thread freeing the extent the second time
+ * can't mark the extent busy immediately because it is already tracked in a
+ * transaction that may be committing. When the log commit for the existing
+ * busy extent completes, the busy extent will be removed from the tree. If we
+ * allow the second busy insert to continue using that busy extent structure,
+ * it can be freed before this transaction is safely in the log. Hence our
+ * only option in this case is to force the log to remove the existing busy
+ * extent from the list before we insert the new one with the current
+ * transaction ID.
+ *
+ * The problem we are trying to avoid in the free-alloc-free in separate
+ * transactions is most easily described with a timeline:
+ *
+ * Thread 1 Thread 2 Thread 3 xfslogd
+ * xact alloc
+ * free X
+ * mark busy
+ * commit xact
+ * free xact
+ * xact alloc
+ * alloc X
+ * busy search
+ * mark xact sync
+ * commit xact
+ * free xact
+ * force log
+ * checkpoint starts
+ * ....
+ * xact alloc
+ * free X
+ * mark busy
+ * finds match
+ * *** KABOOM! ***
+ * ....
+ * log IO completes
+ * unbusy X
+ * checkpoint completes
+ *
+ * By issuing a log force in thread 3 @ "KABOOM", the thread will block until
+ * the checkpoint completes, and the busy extent it matched will have been
+ * removed from the tree when it is woken. Hence it can then continue safely.
+ *
+ * However, to ensure this matching process is robust, we need to use the
+ * transaction ID for identifying transaction, as delayed logging results in
+ * the busy extent and transaction lifecycles being different. i.e. the busy
+ * extent is active for a lot longer than the transaction. Hence the
+ * transaction structure can be freed and reallocated, then mark the same
+ * extent busy again in the new transaction. In this case the new transaction
+ * will have a different tid but can have the same address, and hence we need
+ * to check against the tid.
+ *
+ * Future: for delayed logging, we could avoid the log force if the extent was
+ * first freed in the current checkpoint sequence. This, however, requires the
+ * ability to pin the current checkpoint in memory until this transaction
+ * commits to ensure that both the original free and the current one combine
+ * logically into the one checkpoint. If the checkpoint sequences are
+ * different, however, we still need to wait on a log force.
*/
void
-xfs_alloc_mark_busy(xfs_trans_t *tp,
- xfs_agnumber_t agno,
- xfs_agblock_t bno,
- xfs_extlen_t len)
+xfs_alloc_busy_insert(
+ struct xfs_trans *tp,
+ xfs_agnumber_t agno,
+ xfs_agblock_t bno,
+ xfs_extlen_t len)
{
- xfs_mount_t *mp;
- xfs_perag_busy_t *bsy;
- int n;
+ struct xfs_busy_extent *new;
+ struct xfs_busy_extent *busyp;
+ struct xfs_perag *pag;
+ struct rb_node **rbp;
+ struct rb_node *parent;
+ int match;
- mp = tp->t_mountp;
- spin_lock(&mp->m_perag[agno].pagb_lock);
-
- /* search pagb_list for an open slot */
- for (bsy = mp->m_perag[agno].pagb_list, n = 0;
- n < XFS_PAGB_NUM_SLOTS;
- bsy++, n++) {
- if (bsy->busy_tp == NULL) {
- break;
- }
- }
- trace_xfs_alloc_busy(mp, agno, bno, len, n);
-
- if (n < XFS_PAGB_NUM_SLOTS) {
- bsy = &mp->m_perag[agno].pagb_list[n];
- mp->m_perag[agno].pagb_count++;
- bsy->busy_start = bno;
- bsy->busy_length = len;
- bsy->busy_tp = tp;
- xfs_trans_add_busy(tp, agno, n);
- } else {
+ new = kmem_zalloc(sizeof(struct xfs_busy_extent), KM_MAYFAIL);
+ if (!new) {
/*
- * The busy list is full! Since it is now not possible to
- * track the free block, make this a synchronous transaction
- * to insure that the block is not reused before this
- * transaction commits.
+ * No Memory! Since it is now not possible to track the free
+ * block, make this a synchronous transaction to insure that
+ * the block is not reused before this transaction commits.
*/
+ trace_xfs_alloc_busy(tp, agno, bno, len, 1);
xfs_trans_set_sync(tp);
+ return;
}
- spin_unlock(&mp->m_perag[agno].pagb_lock);
-}
-
-void
-xfs_alloc_clear_busy(xfs_trans_t *tp,
- xfs_agnumber_t agno,
- int idx)
-{
- xfs_mount_t *mp;
- xfs_perag_busy_t *list;
-
- mp = tp->t_mountp;
-
- spin_lock(&mp->m_perag[agno].pagb_lock);
- list = mp->m_perag[agno].pagb_list;
-
- ASSERT(idx < XFS_PAGB_NUM_SLOTS);
-
- trace_xfs_alloc_unbusy(mp, agno, idx, list[idx].busy_tp == tp);
-
- if (list[idx].busy_tp == tp) {
- list[idx].busy_tp = NULL;
- mp->m_perag[agno].pagb_count--;
+ new->agno = agno;
+ new->bno = bno;
+ new->length = len;
+ new->tid = xfs_log_get_trans_ident(tp);
+
+ INIT_LIST_HEAD(&new->list);
+
+ /* trace before insert to be able to see failed inserts */
+ trace_xfs_alloc_busy(tp, agno, bno, len, 0);
+
+ pag = xfs_perag_get(tp->t_mountp, new->agno);
+restart:
+ spin_lock(&pag->pagb_lock);
+ rbp = &pag->pagb_tree.rb_node;
+ parent = NULL;
+ busyp = NULL;
+ match = 0;
+ while (*rbp && match >= 0) {
+ parent = *rbp;
+ busyp = rb_entry(parent, struct xfs_busy_extent, rb_node);
+
+ if (new->bno < busyp->bno) {
+ /* may overlap, but exact start block is lower */
+ rbp = &(*rbp)->rb_left;
+ if (new->bno + new->length > busyp->bno)
+ match = busyp->tid == new->tid ? 1 : -1;
+ } else if (new->bno > busyp->bno) {
+ /* may overlap, but exact start block is higher */
+ rbp = &(*rbp)->rb_right;
+ if (bno < busyp->bno + busyp->length)
+ match = busyp->tid == new->tid ? 1 : -1;
+ } else {
+ match = busyp->tid == new->tid ? 1 : -1;
+ break;
+ }
+ }
+ if (match < 0) {
+ /* overlap marked busy in different transaction */
+ spin_unlock(&pag->pagb_lock);
+ xfs_log_force(tp->t_mountp, XFS_LOG_SYNC);
+ goto restart;
}
+ if (match > 0) {
+ /*
+ * overlap marked busy in same transaction. Update if exact
+ * start block match, otherwise combine the busy extents into
+ * a single range.
+ */
+ if (busyp->bno == new->bno) {
+ busyp->length = max(busyp->length, new->length);
+ spin_unlock(&pag->pagb_lock);
+ ASSERT(tp->t_flags & XFS_TRANS_SYNC);
+ xfs_perag_put(pag);
+ kmem_free(new);
+ return;
+ }
+ rb_erase(&busyp->rb_node, &pag->pagb_tree);
+ new->length = max(busyp->bno + busyp->length,
+ new->bno + new->length) -
+ min(busyp->bno, new->bno);
+ new->bno = min(busyp->bno, new->bno);
+ } else
+ busyp = NULL;
- spin_unlock(&mp->m_perag[agno].pagb_lock);
-}
+ rb_link_node(&new->rb_node, parent, rbp);
+ rb_insert_color(&new->rb_node, &pag->pagb_tree);
+ list_add(&new->list, &tp->t_busy);
+ spin_unlock(&pag->pagb_lock);
+ xfs_perag_put(pag);
+ kmem_free(busyp);
+}
/*
- * If we find the extent in the busy list, force the log out to get the
- * extent out of the busy list so the caller can use it straight away.
+ * Search for a busy extent within the range of the extent we are about to
+ * allocate. You need to be holding the busy extent tree lock when calling
+ * xfs_alloc_busy_search(). This function returns 0 for no overlapping busy
+ * extent, -1 for an overlapping but not exact busy extent, and 1 for an exact
+ * match. This is done so that a non-zero return indicates an overlap that
+ * will require a synchronous transaction, but it can still be
+ * used to distinguish between a partial or exact match.
*/
-STATIC void
-xfs_alloc_search_busy(xfs_trans_t *tp,
- xfs_agnumber_t agno,
- xfs_agblock_t bno,
- xfs_extlen_t len)
+static int
+xfs_alloc_busy_search(
+ struct xfs_mount *mp,
+ xfs_agnumber_t agno,
+ xfs_agblock_t bno,
+ xfs_extlen_t len)
{
- xfs_mount_t *mp;
- xfs_perag_busy_t *bsy;
- xfs_agblock_t uend, bend;
- xfs_lsn_t lsn = 0;
- int cnt;
-
- mp = tp->t_mountp;
+ struct xfs_perag *pag;
+ struct rb_node *rbp;
+ struct xfs_busy_extent *busyp;
+ int match = 0;
+
+ pag = xfs_perag_get(mp, agno);
+ spin_lock(&pag->pagb_lock);
+
+ rbp = pag->pagb_tree.rb_node;
+
+ /* find closest start bno overlap */
+ while (rbp) {
+ busyp = rb_entry(rbp, struct xfs_busy_extent, rb_node);
+ if (bno < busyp->bno) {
+ /* may overlap, but exact start block is lower */
+ if (bno + len > busyp->bno)
+ match = -1;
+ rbp = rbp->rb_left;
+ } else if (bno > busyp->bno) {
+ /* may overlap, but exact start block is higher */
+ if (bno < busyp->bno + busyp->length)
+ match = -1;
+ rbp = rbp->rb_right;
+ } else {
+ /* bno matches busyp, length determines exact match */
+ match = (busyp->length == len) ? 1 : -1;
+ break;
+ }
+ }
+ spin_unlock(&pag->pagb_lock);
+ trace_xfs_alloc_busysearch(mp, agno, bno, len, !!match);
+ xfs_perag_put(pag);
+ return match;
+}
- spin_lock(&mp->m_perag[agno].pagb_lock);
+void
+xfs_alloc_busy_clear(
+ struct xfs_mount *mp,
+ struct xfs_busy_extent *busyp)
+{
+ struct xfs_perag *pag;
- uend = bno + len - 1;
+ trace_xfs_alloc_unbusy(mp, busyp->agno, busyp->bno,
+ busyp->length);
- /*
- * search pagb_list for this slot, skipping open slots. We have to
- * search the entire array as there may be multiple overlaps and
- * we have to get the most recent LSN for the log force to push out
- * all the transactions that span the range.
- */
- for (cnt = 0; cnt < mp->m_perag[agno].pagb_count; cnt++) {
- bsy = &mp->m_perag[agno].pagb_list[cnt];
- if (!bsy->busy_tp)
- continue;
+ ASSERT(xfs_alloc_busy_search(mp, busyp->agno, busyp->bno,
+ busyp->length) == 1);
- bend = bsy->busy_start + bsy->busy_length - 1;
- if (bno > bend || uend < bsy->busy_start)
- continue;
+ list_del_init(&busyp->list);
- /* (start1,length1) within (start2, length2) */
- if (XFS_LSN_CMP(bsy->busy_tp->t_commit_lsn, lsn) > 0)
- lsn = bsy->busy_tp->t_commit_lsn;
- }
- spin_unlock(&mp->m_perag[agno].pagb_lock);
- trace_xfs_alloc_busysearch(tp->t_mountp, agno, bno, len, lsn);
+ pag = xfs_perag_get(mp, busyp->agno);
+ spin_lock(&pag->pagb_lock);
+ rb_erase(&busyp->rb_node, &pag->pagb_tree);
+ spin_unlock(&pag->pagb_lock);
+ xfs_perag_put(pag);
- /*
- * If a block was found, force the log through the LSN of the
- * transaction that freed the block
- */
- if (lsn)
- xfs_log_force(mp, lsn, XFS_LOG_FORCE|XFS_LOG_SYNC);
+ kmem_free(busyp);
}
diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/xfs_alloc.h
index 599bffa39784..6d05199b667c 100644
--- a/fs/xfs/xfs_alloc.h
+++ b/fs/xfs/xfs_alloc.h
@@ -22,6 +22,7 @@ struct xfs_buf;
struct xfs_mount;
struct xfs_perag;
struct xfs_trans;
+struct xfs_busy_extent;
/*
* Freespace allocation types. Argument to xfs_alloc_[v]extent.
@@ -119,15 +120,13 @@ xfs_alloc_longest_free_extent(struct xfs_mount *mp,
#ifdef __KERNEL__
void
-xfs_alloc_mark_busy(xfs_trans_t *tp,
+xfs_alloc_busy_insert(xfs_trans_t *tp,
xfs_agnumber_t agno,
xfs_agblock_t bno,
xfs_extlen_t len);
void
-xfs_alloc_clear_busy(xfs_trans_t *tp,
- xfs_agnumber_t ag,
- int idx);
+xfs_alloc_busy_clear(struct xfs_mount *mp, struct xfs_busy_extent *busyp);
#endif /* __KERNEL__ */
diff --git a/fs/xfs/xfs_alloc_btree.c b/fs/xfs/xfs_alloc_btree.c
index adbd9141aea1..83f494218759 100644
--- a/fs/xfs/xfs_alloc_btree.c
+++ b/fs/xfs/xfs_alloc_btree.c
@@ -61,12 +61,14 @@ xfs_allocbt_set_root(
struct xfs_agf *agf = XFS_BUF_TO_AGF(agbp);
xfs_agnumber_t seqno = be32_to_cpu(agf->agf_seqno);
int btnum = cur->bc_btnum;
+ struct xfs_perag *pag = xfs_perag_get(cur->bc_mp, seqno);
ASSERT(ptr->s != 0);
agf->agf_roots[btnum] = ptr->s;
be32_add_cpu(&agf->agf_levels[btnum], inc);
- cur->bc_mp->m_perag[seqno].pagf_levels[btnum] += inc;
+ pag->pagf_levels[btnum] += inc;
+ xfs_perag_put(pag);
xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_ROOTS | XFS_AGF_LEVELS);
}
@@ -132,7 +134,7 @@ xfs_allocbt_free_block(
* disk. If a busy block is allocated, the iclog is pushed up to the
* LSN that freed the block.
*/
- xfs_alloc_mark_busy(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1);
+ xfs_alloc_busy_insert(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1);
xfs_trans_agbtree_delta(cur->bc_tp, -1);
return 0;
}
@@ -150,6 +152,7 @@ xfs_allocbt_update_lastrec(
{
struct xfs_agf *agf = XFS_BUF_TO_AGF(cur->bc_private.a.agbp);
xfs_agnumber_t seqno = be32_to_cpu(agf->agf_seqno);
+ struct xfs_perag *pag;
__be32 len;
int numrecs;
@@ -193,7 +196,9 @@ xfs_allocbt_update_lastrec(
}
agf->agf_longest = len;
- cur->bc_mp->m_perag[seqno].pagf_longest = be32_to_cpu(len);
+ pag = xfs_perag_get(cur->bc_mp, seqno);
+ pag->pagf_longest = be32_to_cpu(len);
+ xfs_perag_put(pag);
xfs_alloc_log_agf(cur->bc_tp, cur->bc_private.a.agbp, XFS_AGF_LONGEST);
}
diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c
index e953b6cfb2a8..b9c196a53c42 100644
--- a/fs/xfs/xfs_attr.c
+++ b/fs/xfs/xfs_attr.c
@@ -93,12 +93,12 @@ STATIC int xfs_attr_rmtval_remove(xfs_da_args_t *args);
STATIC int
xfs_attr_name_to_xname(
struct xfs_name *xname,
- const char *aname)
+ const unsigned char *aname)
{
if (!aname)
return EINVAL;
xname->name = aname;
- xname->len = strlen(aname);
+ xname->len = strlen((char *)aname);
if (xname->len >= MAXNAMELEN)
return EFAULT; /* match IRIX behaviour */
@@ -124,7 +124,7 @@ STATIC int
xfs_attr_get_int(
struct xfs_inode *ip,
struct xfs_name *name,
- char *value,
+ unsigned char *value,
int *valuelenp,
int flags)
{
@@ -171,8 +171,8 @@ xfs_attr_get_int(
int
xfs_attr_get(
xfs_inode_t *ip,
- const char *name,
- char *value,
+ const unsigned char *name,
+ unsigned char *value,
int *valuelenp,
int flags)
{
@@ -197,7 +197,7 @@ xfs_attr_get(
/*
* Calculate how many blocks we need for the new attribute,
*/
-int
+STATIC int
xfs_attr_calc_size(
struct xfs_inode *ip,
int namelen,
@@ -235,8 +235,12 @@ xfs_attr_calc_size(
}
STATIC int
-xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name,
- char *value, int valuelen, int flags)
+xfs_attr_set_int(
+ struct xfs_inode *dp,
+ struct xfs_name *name,
+ unsigned char *value,
+ int valuelen,
+ int flags)
{
xfs_da_args_t args;
xfs_fsblock_t firstblock;
@@ -452,8 +456,8 @@ out:
int
xfs_attr_set(
xfs_inode_t *dp,
- const char *name,
- char *value,
+ const unsigned char *name,
+ unsigned char *value,
int valuelen,
int flags)
{
@@ -600,7 +604,7 @@ out:
int
xfs_attr_remove(
xfs_inode_t *dp,
- const char *name,
+ const unsigned char *name,
int flags)
{
int error;
@@ -669,9 +673,13 @@ xfs_attr_list_int(xfs_attr_list_context_t *context)
*/
/*ARGSUSED*/
STATIC int
-xfs_attr_put_listent(xfs_attr_list_context_t *context, int flags,
- char *name, int namelen,
- int valuelen, char *value)
+xfs_attr_put_listent(
+ xfs_attr_list_context_t *context,
+ int flags,
+ unsigned char *name,
+ int namelen,
+ int valuelen,
+ unsigned char *value)
{
struct attrlist *alist = (struct attrlist *)context->alist;
attrlist_ent_t *aep;
@@ -1980,7 +1988,7 @@ xfs_attr_rmtval_get(xfs_da_args_t *args)
xfs_bmbt_irec_t map[ATTR_RMTVALUE_MAPSIZE];
xfs_mount_t *mp;
xfs_daddr_t dblkno;
- xfs_caddr_t dst;
+ void *dst;
xfs_buf_t *bp;
int nmap, error, tmp, valuelen, blkcnt, i;
xfs_dablk_t lblkno;
@@ -2007,15 +2015,14 @@ xfs_attr_rmtval_get(xfs_da_args_t *args)
dblkno = XFS_FSB_TO_DADDR(mp, map[i].br_startblock);
blkcnt = XFS_FSB_TO_BB(mp, map[i].br_blockcount);
error = xfs_read_buf(mp, mp->m_ddev_targp, dblkno,
- blkcnt,
- XFS_BUF_LOCK | XBF_DONT_BLOCK,
+ blkcnt, XBF_LOCK | XBF_DONT_BLOCK,
&bp);
if (error)
return(error);
tmp = (valuelen < XFS_BUF_SIZE(bp))
? valuelen : XFS_BUF_SIZE(bp);
- xfs_biomove(bp, 0, tmp, dst, XFS_B_READ);
+ xfs_biomove(bp, 0, tmp, dst, XBF_READ);
xfs_buf_relse(bp);
dst += tmp;
valuelen -= tmp;
@@ -2039,7 +2046,7 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
xfs_inode_t *dp;
xfs_bmbt_irec_t map;
xfs_daddr_t dblkno;
- xfs_caddr_t src;
+ void *src;
xfs_buf_t *bp;
xfs_dablk_t lblkno;
int blkcnt, valuelen, nmap, error, tmp, committed;
@@ -2141,13 +2148,13 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
blkcnt = XFS_FSB_TO_BB(mp, map.br_blockcount);
bp = xfs_buf_get(mp->m_ddev_targp, dblkno, blkcnt,
- XFS_BUF_LOCK | XBF_DONT_BLOCK);
+ XBF_LOCK | XBF_DONT_BLOCK);
ASSERT(bp);
ASSERT(!XFS_BUF_GETERROR(bp));
tmp = (valuelen < XFS_BUF_SIZE(bp)) ? valuelen :
XFS_BUF_SIZE(bp);
- xfs_biomove(bp, 0, tmp, src, XFS_B_WRITE);
+ xfs_biomove(bp, 0, tmp, src, XBF_WRITE);
if (tmp < XFS_BUF_SIZE(bp))
xfs_biozero(bp, tmp, XFS_BUF_SIZE(bp) - tmp);
if ((error = xfs_bwrite(mp, bp))) {/* GROT: NOTE: synchronous write */
@@ -2208,8 +2215,7 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args)
/*
* If the "remote" value is in the cache, remove it.
*/
- bp = xfs_incore(mp->m_ddev_targp, dblkno, blkcnt,
- XFS_INCORE_TRYLOCK);
+ bp = xfs_incore(mp->m_ddev_targp, dblkno, blkcnt, XBF_TRYLOCK);
if (bp) {
XFS_BUF_STALE(bp);
XFS_BUF_UNDELAYWRITE(bp);
diff --git a/fs/xfs/xfs_attr.h b/fs/xfs/xfs_attr.h
index 59b410ce69a1..e920d68ef509 100644
--- a/fs/xfs/xfs_attr.h
+++ b/fs/xfs/xfs_attr.h
@@ -113,7 +113,7 @@ typedef struct attrlist_cursor_kern {
typedef int (*put_listent_func_t)(struct xfs_attr_list_context *, int,
- char *, int, int, char *);
+ unsigned char *, int, int, unsigned char *);
typedef struct xfs_attr_list_context {
struct xfs_inode *dp; /* inode */
@@ -139,7 +139,6 @@ typedef struct xfs_attr_list_context {
/*
* Overall external interface routines.
*/
-int xfs_attr_calc_size(struct xfs_inode *, int, int, int *);
int xfs_attr_inactive(struct xfs_inode *dp);
int xfs_attr_rmtval_get(struct xfs_da_args *args);
int xfs_attr_list_int(struct xfs_attr_list_context *);
diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c
index baf41b5af756..a90ce74fc256 100644
--- a/fs/xfs/xfs_attr_leaf.c
+++ b/fs/xfs/xfs_attr_leaf.c
@@ -521,11 +521,11 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
sfe = &sf->list[0];
for (i = 0; i < sf->hdr.count; i++) {
- nargs.name = (char *)sfe->nameval;
+ nargs.name = sfe->nameval;
nargs.namelen = sfe->namelen;
- nargs.value = (char *)&sfe->nameval[nargs.namelen];
+ nargs.value = &sfe->nameval[nargs.namelen];
nargs.valuelen = sfe->valuelen;
- nargs.hashval = xfs_da_hashname((char *)sfe->nameval,
+ nargs.hashval = xfs_da_hashname(sfe->nameval,
sfe->namelen);
nargs.flags = XFS_ATTR_NSP_ONDISK_TO_ARGS(sfe->flags);
error = xfs_attr_leaf_lookup_int(bp, &nargs); /* set a->index */
@@ -612,10 +612,10 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
for (i = 0, sfe = &sf->list[0]; i < sf->hdr.count; i++) {
error = context->put_listent(context,
sfe->flags,
- (char *)sfe->nameval,
+ sfe->nameval,
(int)sfe->namelen,
(int)sfe->valuelen,
- (char*)&sfe->nameval[sfe->namelen]);
+ &sfe->nameval[sfe->namelen]);
/*
* Either search callback finished early or
@@ -659,8 +659,8 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
}
sbp->entno = i;
- sbp->hash = xfs_da_hashname((char *)sfe->nameval, sfe->namelen);
- sbp->name = (char *)sfe->nameval;
+ sbp->hash = xfs_da_hashname(sfe->nameval, sfe->namelen);
+ sbp->name = sfe->nameval;
sbp->namelen = sfe->namelen;
/* These are bytes, and both on-disk, don't endian-flip */
sbp->valuelen = sfe->valuelen;
@@ -818,9 +818,9 @@ xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args, int forkoff)
continue;
ASSERT(entry->flags & XFS_ATTR_LOCAL);
name_loc = xfs_attr_leaf_name_local(leaf, i);
- nargs.name = (char *)name_loc->nameval;
+ nargs.name = name_loc->nameval;
nargs.namelen = name_loc->namelen;
- nargs.value = (char *)&name_loc->nameval[nargs.namelen];
+ nargs.value = &name_loc->nameval[nargs.namelen];
nargs.valuelen = be16_to_cpu(name_loc->valuelen);
nargs.hashval = be32_to_cpu(entry->hashval);
nargs.flags = XFS_ATTR_NSP_ONDISK_TO_ARGS(entry->flags);
@@ -2370,10 +2370,10 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context)
retval = context->put_listent(context,
entry->flags,
- (char *)name_loc->nameval,
+ name_loc->nameval,
(int)name_loc->namelen,
be16_to_cpu(name_loc->valuelen),
- (char *)&name_loc->nameval[name_loc->namelen]);
+ &name_loc->nameval[name_loc->namelen]);
if (retval)
return retval;
} else {
@@ -2397,15 +2397,15 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context)
return retval;
retval = context->put_listent(context,
entry->flags,
- (char *)name_rmt->name,
+ name_rmt->name,
(int)name_rmt->namelen,
valuelen,
- (char*)args.value);
+ args.value);
kmem_free(args.value);
} else {
retval = context->put_listent(context,
entry->flags,
- (char *)name_rmt->name,
+ name_rmt->name,
(int)name_rmt->namelen,
valuelen,
NULL);
@@ -2950,7 +2950,7 @@ xfs_attr_leaf_freextent(xfs_trans_t **trans, xfs_inode_t *dp,
map.br_blockcount);
bp = xfs_trans_get_buf(*trans,
dp->i_mount->m_ddev_targp,
- dblkno, dblkcnt, XFS_BUF_LOCK);
+ dblkno, dblkcnt, XBF_LOCK);
xfs_trans_binval(*trans, bp);
/*
* Roll to next transaction.
diff --git a/fs/xfs/xfs_attr_sf.h b/fs/xfs/xfs_attr_sf.h
index 76ab7b0cbb3a..919756e3ba53 100644
--- a/fs/xfs/xfs_attr_sf.h
+++ b/fs/xfs/xfs_attr_sf.h
@@ -52,7 +52,7 @@ typedef struct xfs_attr_sf_sort {
__uint8_t valuelen; /* length of value */
__uint8_t flags; /* flags bits (see xfs_attr_leaf.h) */
xfs_dahash_t hash; /* this entry's hash value */
- char *name; /* name value, pointer into buffer */
+ unsigned char *name; /* name value, pointer into buffer */
} xfs_attr_sf_sort_t;
#define XFS_ATTR_SF_ENTSIZE_BYNAME(nlen,vlen) /* space name/value uses */ \
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c
index 98251cdc52aa..99587ded043f 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/xfs_bmap.c
@@ -2550,22 +2550,134 @@ xfs_bmap_rtalloc(
}
STATIC int
+xfs_bmap_btalloc_nullfb(
+ struct xfs_bmalloca *ap,
+ struct xfs_alloc_arg *args,
+ xfs_extlen_t *blen)
+{
+ struct xfs_mount *mp = ap->ip->i_mount;
+ struct xfs_perag *pag;
+ xfs_agnumber_t ag, startag;
+ int notinit = 0;
+ int error;
+
+ if (ap->userdata && xfs_inode_is_filestream(ap->ip))
+ args->type = XFS_ALLOCTYPE_NEAR_BNO;
+ else
+ args->type = XFS_ALLOCTYPE_START_BNO;
+ args->total = ap->total;
+
+ /*
+ * Search for an allocation group with a single extent large enough
+ * for the request. If one isn't found, then adjust the minimum
+ * allocation size to the largest space found.
+ */
+ startag = ag = XFS_FSB_TO_AGNO(mp, args->fsbno);
+ if (startag == NULLAGNUMBER)
+ startag = ag = 0;
+
+ pag = xfs_perag_get(mp, ag);
+ while (*blen < ap->alen) {
+ if (!pag->pagf_init) {
+ error = xfs_alloc_pagf_init(mp, args->tp, ag,
+ XFS_ALLOC_FLAG_TRYLOCK);
+ if (error) {
+ xfs_perag_put(pag);
+ return error;
+ }
+ }
+
+ /*
+ * See xfs_alloc_fix_freelist...
+ */
+ if (pag->pagf_init) {
+ xfs_extlen_t longest;
+ longest = xfs_alloc_longest_free_extent(mp, pag);
+ if (*blen < longest)
+ *blen = longest;
+ } else
+ notinit = 1;
+
+ if (xfs_inode_is_filestream(ap->ip)) {
+ if (*blen >= ap->alen)
+ break;
+
+ if (ap->userdata) {
+ /*
+ * If startag is an invalid AG, we've
+ * come here once before and
+ * xfs_filestream_new_ag picked the
+ * best currently available.
+ *
+ * Don't continue looping, since we
+ * could loop forever.
+ */
+ if (startag == NULLAGNUMBER)
+ break;
+
+ error = xfs_filestream_new_ag(ap, &ag);
+ xfs_perag_put(pag);
+ if (error)
+ return error;
+
+ /* loop again to set 'blen'*/
+ startag = NULLAGNUMBER;
+ pag = xfs_perag_get(mp, ag);
+ continue;
+ }
+ }
+ if (++ag == mp->m_sb.sb_agcount)
+ ag = 0;
+ if (ag == startag)
+ break;
+ xfs_perag_put(pag);
+ pag = xfs_perag_get(mp, ag);
+ }
+ xfs_perag_put(pag);
+
+ /*
+ * Since the above loop did a BUF_TRYLOCK, it is
+ * possible that there is space for this request.
+ */
+ if (notinit || *blen < ap->minlen)
+ args->minlen = ap->minlen;
+ /*
+ * If the best seen length is less than the request
+ * length, use the best as the minimum.
+ */
+ else if (*blen < ap->alen)
+ args->minlen = *blen;
+ /*
+ * Otherwise we've seen an extent as big as alen,
+ * use that as the minimum.
+ */
+ else
+ args->minlen = ap->alen;
+
+ /*
+ * set the failure fallback case to look in the selected
+ * AG as the stream may have moved.
+ */
+ if (xfs_inode_is_filestream(ap->ip))
+ ap->rval = args->fsbno = XFS_AGB_TO_FSB(mp, ag, 0);
+
+ return 0;
+}
+
+STATIC int
xfs_bmap_btalloc(
xfs_bmalloca_t *ap) /* bmap alloc argument struct */
{
xfs_mount_t *mp; /* mount point structure */
xfs_alloctype_t atype = 0; /* type for allocation routines */
xfs_extlen_t align; /* minimum allocation alignment */
- xfs_agnumber_t ag;
xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */
- xfs_agnumber_t startag;
+ xfs_agnumber_t ag;
xfs_alloc_arg_t args;
xfs_extlen_t blen;
xfs_extlen_t nextminlen = 0;
- xfs_perag_t *pag;
int nullfb; /* true if ap->firstblock isn't set */
int isaligned;
- int notinit;
int tryagain;
int error;
@@ -2612,102 +2724,9 @@ xfs_bmap_btalloc(
args.firstblock = ap->firstblock;
blen = 0;
if (nullfb) {
- if (ap->userdata && xfs_inode_is_filestream(ap->ip))
- args.type = XFS_ALLOCTYPE_NEAR_BNO;
- else
- args.type = XFS_ALLOCTYPE_START_BNO;
- args.total = ap->total;
-
- /*
- * Search for an allocation group with a single extent
- * large enough for the request.
- *
- * If one isn't found, then adjust the minimum allocation
- * size to the largest space found.
- */
- startag = ag = XFS_FSB_TO_AGNO(mp, args.fsbno);
- if (startag == NULLAGNUMBER)
- startag = ag = 0;
- notinit = 0;
- down_read(&mp->m_peraglock);
- while (blen < ap->alen) {
- pag = &mp->m_perag[ag];
- if (!pag->pagf_init &&
- (error = xfs_alloc_pagf_init(mp, args.tp,
- ag, XFS_ALLOC_FLAG_TRYLOCK))) {
- up_read(&mp->m_peraglock);
- return error;
- }
- /*
- * See xfs_alloc_fix_freelist...
- */
- if (pag->pagf_init) {
- xfs_extlen_t longest;
- longest = xfs_alloc_longest_free_extent(mp, pag);
- if (blen < longest)
- blen = longest;
- } else
- notinit = 1;
-
- if (xfs_inode_is_filestream(ap->ip)) {
- if (blen >= ap->alen)
- break;
-
- if (ap->userdata) {
- /*
- * If startag is an invalid AG, we've
- * come here once before and
- * xfs_filestream_new_ag picked the
- * best currently available.
- *
- * Don't continue looping, since we
- * could loop forever.
- */
- if (startag == NULLAGNUMBER)
- break;
-
- error = xfs_filestream_new_ag(ap, &ag);
- if (error) {
- up_read(&mp->m_peraglock);
- return error;
- }
-
- /* loop again to set 'blen'*/
- startag = NULLAGNUMBER;
- continue;
- }
- }
- if (++ag == mp->m_sb.sb_agcount)
- ag = 0;
- if (ag == startag)
- break;
- }
- up_read(&mp->m_peraglock);
- /*
- * Since the above loop did a BUF_TRYLOCK, it is
- * possible that there is space for this request.
- */
- if (notinit || blen < ap->minlen)
- args.minlen = ap->minlen;
- /*
- * If the best seen length is less than the request
- * length, use the best as the minimum.
- */
- else if (blen < ap->alen)
- args.minlen = blen;
- /*
- * Otherwise we've seen an extent as big as alen,
- * use that as the minimum.
- */
- else
- args.minlen = ap->alen;
-
- /*
- * set the failure fallback case to look in the selected
- * AG as the stream may have moved.
- */
- if (xfs_inode_is_filestream(ap->ip))
- ap->rval = args.fsbno = XFS_AGB_TO_FSB(mp, ag, 0);
+ error = xfs_bmap_btalloc_nullfb(ap, &args, &blen);
+ if (error)
+ return error;
} else if (ap->low) {
if (xfs_inode_is_filestream(ap->ip))
args.type = XFS_ALLOCTYPE_FIRST_AG;
@@ -3810,7 +3829,7 @@ xfs_bmap_add_attrfork(
}
if ((error = xfs_bmap_finish(&tp, &flist, &committed)))
goto error2;
- error = xfs_trans_commit(tp, XFS_TRANS_PERM_LOG_RES);
+ error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
ASSERT(ip->i_df.if_ext_max ==
XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t));
return error;
@@ -4470,7 +4489,7 @@ xfs_bmapi(
xfs_fsblock_t abno; /* allocated block number */
xfs_extlen_t alen; /* allocated extent length */
xfs_fileoff_t aoff; /* allocated file offset */
- xfs_bmalloca_t bma; /* args for xfs_bmap_alloc */
+ xfs_bmalloca_t bma = { 0 }; /* args for xfs_bmap_alloc */
xfs_btree_cur_t *cur; /* bmap btree cursor */
xfs_fileoff_t end; /* end of mapped file region */
int eof; /* we've hit the end of extents */
diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c
index 38751d5fac6f..416e47e54b83 100644
--- a/fs/xfs/xfs_bmap_btree.c
+++ b/fs/xfs/xfs_bmap_btree.c
@@ -334,7 +334,7 @@ xfs_bmbt_disk_set_allf(
/*
* Set all the fields in a bmap extent record from the uncompressed form.
*/
-void
+STATIC void
xfs_bmbt_disk_set_all(
xfs_bmbt_rec_t *r,
xfs_bmbt_irec_t *s)
diff --git a/fs/xfs/xfs_bmap_btree.h b/fs/xfs/xfs_bmap_btree.h
index cf07ca7c22e7..0e66c4ea0f85 100644
--- a/fs/xfs/xfs_bmap_btree.h
+++ b/fs/xfs/xfs_bmap_btree.h
@@ -223,7 +223,6 @@ extern void xfs_bmbt_set_startblock(xfs_bmbt_rec_host_t *r, xfs_fsblock_t v);
extern void xfs_bmbt_set_startoff(xfs_bmbt_rec_host_t *r, xfs_fileoff_t v);
extern void xfs_bmbt_set_state(xfs_bmbt_rec_host_t *r, xfs_exntst_t v);
-extern void xfs_bmbt_disk_set_all(xfs_bmbt_rec_t *r, xfs_bmbt_irec_t *s);
extern void xfs_bmbt_disk_set_allf(xfs_bmbt_rec_t *r, xfs_fileoff_t o,
xfs_fsblock_t b, xfs_filblks_t c, xfs_exntst_t v);
diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c
index 36a0992dd669..96be4b0f2496 100644
--- a/fs/xfs/xfs_btree.c
+++ b/fs/xfs/xfs_btree.c
@@ -977,7 +977,7 @@ xfs_btree_get_buf_block(
xfs_daddr_t d;
/* need to sort out how callers deal with failures first */
- ASSERT(!(flags & XFS_BUF_TRYLOCK));
+ ASSERT(!(flags & XBF_TRYLOCK));
d = xfs_btree_ptr_to_daddr(cur, ptr);
*bpp = xfs_trans_get_buf(cur->bc_tp, mp->m_ddev_targp, d,
@@ -1008,7 +1008,7 @@ xfs_btree_read_buf_block(
int error;
/* need to sort out how callers deal with failures first */
- ASSERT(!(flags & XFS_BUF_TRYLOCK));
+ ASSERT(!(flags & XBF_TRYLOCK));
d = xfs_btree_ptr_to_daddr(cur, ptr);
error = xfs_trans_read_buf(mp, cur->bc_tp, mp->m_ddev_targp, d,
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index a30f7e9eb2b9..02a80984aa05 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -64,7 +64,7 @@ xfs_buf_item_log_debug(
nbytes = last - first + 1;
bfset(bip->bli_logged, first, nbytes);
for (x = 0; x < nbytes; x++) {
- chunk_num = byte >> XFS_BLI_SHIFT;
+ chunk_num = byte >> XFS_BLF_SHIFT;
word_num = chunk_num >> BIT_TO_WORD_SHIFT;
bit_num = chunk_num & (NBWORD - 1);
wordp = &(bip->bli_format.blf_data_map[word_num]);
@@ -166,7 +166,7 @@ xfs_buf_item_size(
* cancel flag in it.
*/
trace_xfs_buf_item_size_stale(bip);
- ASSERT(bip->bli_format.blf_flags & XFS_BLI_CANCEL);
+ ASSERT(bip->bli_format.blf_flags & XFS_BLF_CANCEL);
return 1;
}
@@ -197,9 +197,9 @@ xfs_buf_item_size(
} else if (next_bit != last_bit + 1) {
last_bit = next_bit;
nvecs++;
- } else if (xfs_buf_offset(bp, next_bit * XFS_BLI_CHUNK) !=
- (xfs_buf_offset(bp, last_bit * XFS_BLI_CHUNK) +
- XFS_BLI_CHUNK)) {
+ } else if (xfs_buf_offset(bp, next_bit * XFS_BLF_CHUNK) !=
+ (xfs_buf_offset(bp, last_bit * XFS_BLF_CHUNK) +
+ XFS_BLF_CHUNK)) {
last_bit = next_bit;
nvecs++;
} else {
@@ -250,10 +250,24 @@ xfs_buf_item_format(
((bip->bli_format.blf_map_size - 1) * sizeof(uint)));
vecp->i_addr = (xfs_caddr_t)&bip->bli_format;
vecp->i_len = base_size;
- XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_BFORMAT);
+ vecp->i_type = XLOG_REG_TYPE_BFORMAT;
vecp++;
nvecs = 1;
+ /*
+ * If it is an inode buffer, transfer the in-memory state to the
+ * format flags and clear the in-memory state. We do not transfer
+ * this state if the inode buffer allocation has not yet been committed
+ * to the log as setting the XFS_BLI_INODE_BUF flag will prevent
+ * correct replay of the inode allocation.
+ */
+ if (bip->bli_flags & XFS_BLI_INODE_BUF) {
+ if (!((bip->bli_flags & XFS_BLI_INODE_ALLOC_BUF) &&
+ xfs_log_item_in_current_chkpt(&bip->bli_item)))
+ bip->bli_format.blf_flags |= XFS_BLF_INODE_BUF;
+ bip->bli_flags &= ~XFS_BLI_INODE_BUF;
+ }
+
if (bip->bli_flags & XFS_BLI_STALE) {
/*
* The buffer is stale, so all we need to log
@@ -261,7 +275,7 @@ xfs_buf_item_format(
* cancel flag in it.
*/
trace_xfs_buf_item_format_stale(bip);
- ASSERT(bip->bli_format.blf_flags & XFS_BLI_CANCEL);
+ ASSERT(bip->bli_format.blf_flags & XFS_BLF_CANCEL);
bip->bli_format.blf_size = nvecs;
return;
}
@@ -294,29 +308,29 @@ xfs_buf_item_format(
* keep counting and scanning.
*/
if (next_bit == -1) {
- buffer_offset = first_bit * XFS_BLI_CHUNK;
+ buffer_offset = first_bit * XFS_BLF_CHUNK;
vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
- vecp->i_len = nbits * XFS_BLI_CHUNK;
- XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_BCHUNK);
+ vecp->i_len = nbits * XFS_BLF_CHUNK;
+ vecp->i_type = XLOG_REG_TYPE_BCHUNK;
nvecs++;
break;
} else if (next_bit != last_bit + 1) {
- buffer_offset = first_bit * XFS_BLI_CHUNK;
+ buffer_offset = first_bit * XFS_BLF_CHUNK;
vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
- vecp->i_len = nbits * XFS_BLI_CHUNK;
- XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_BCHUNK);
+ vecp->i_len = nbits * XFS_BLF_CHUNK;
+ vecp->i_type = XLOG_REG_TYPE_BCHUNK;
nvecs++;
vecp++;
first_bit = next_bit;
last_bit = next_bit;
nbits = 1;
- } else if (xfs_buf_offset(bp, next_bit << XFS_BLI_SHIFT) !=
- (xfs_buf_offset(bp, last_bit << XFS_BLI_SHIFT) +
- XFS_BLI_CHUNK)) {
- buffer_offset = first_bit * XFS_BLI_CHUNK;
+ } else if (xfs_buf_offset(bp, next_bit << XFS_BLF_SHIFT) !=
+ (xfs_buf_offset(bp, last_bit << XFS_BLF_SHIFT) +
+ XFS_BLF_CHUNK)) {
+ buffer_offset = first_bit * XFS_BLF_CHUNK;
vecp->i_addr = xfs_buf_offset(bp, buffer_offset);
- vecp->i_len = nbits * XFS_BLI_CHUNK;
- XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_BCHUNK);
+ vecp->i_len = nbits * XFS_BLF_CHUNK;
+ vecp->i_type = XLOG_REG_TYPE_BCHUNK;
/* You would think we need to bump the nvecs here too, but we do not
* this number is used by recovery, and it gets confused by the boundary
* split here
@@ -341,10 +355,15 @@ xfs_buf_item_format(
}
/*
- * This is called to pin the buffer associated with the buf log
- * item in memory so it cannot be written out. Simply call bpin()
- * on the buffer to do this.
+ * This is called to pin the buffer associated with the buf log item in memory
+ * so it cannot be written out. Simply call bpin() on the buffer to do this.
+ *
+ * We also always take a reference to the buffer log item here so that the bli
+ * is held while the item is pinned in memory. This means that we can
+ * unconditionally drop the reference count a transaction holds when the
+ * transaction is completed.
*/
+
STATIC void
xfs_buf_item_pin(
xfs_buf_log_item_t *bip)
@@ -356,6 +375,7 @@ xfs_buf_item_pin(
ASSERT(atomic_read(&bip->bli_refcount) > 0);
ASSERT((bip->bli_flags & XFS_BLI_LOGGED) ||
(bip->bli_flags & XFS_BLI_STALE));
+ atomic_inc(&bip->bli_refcount);
trace_xfs_buf_item_pin(bip);
xfs_bpin(bp);
}
@@ -372,12 +392,12 @@ xfs_buf_item_pin(
*/
STATIC void
xfs_buf_item_unpin(
- xfs_buf_log_item_t *bip,
- int stale)
+ xfs_buf_log_item_t *bip)
{
struct xfs_ail *ailp;
xfs_buf_t *bp;
int freed;
+ int stale = bip->bli_flags & XFS_BLI_STALE;
bp = bip->bli_buf;
ASSERT(bp != NULL);
@@ -393,7 +413,7 @@ xfs_buf_item_unpin(
ASSERT(XFS_BUF_VALUSEMA(bp) <= 0);
ASSERT(!(XFS_BUF_ISDELAYWRITE(bp)));
ASSERT(XFS_BUF_ISSTALE(bp));
- ASSERT(bip->bli_format.blf_flags & XFS_BLI_CANCEL);
+ ASSERT(bip->bli_format.blf_flags & XFS_BLF_CANCEL);
trace_xfs_buf_item_unpin_stale(bip);
/*
@@ -428,47 +448,43 @@ xfs_buf_item_unpin_remove(
xfs_buf_log_item_t *bip,
xfs_trans_t *tp)
{
- xfs_buf_t *bp;
- xfs_log_item_desc_t *lidp;
- int stale = 0;
-
- bp = bip->bli_buf;
- /*
- * will xfs_buf_item_unpin() call xfs_buf_item_relse()?
- */
+ /* will xfs_buf_item_unpin() call xfs_buf_item_relse()? */
if ((atomic_read(&bip->bli_refcount) == 1) &&
(bip->bli_flags & XFS_BLI_STALE)) {
+ /*
+ * yes -- We can safely do some work here and then call
+ * buf_item_unpin to do the rest because we are
+ * are holding the buffer locked so no one else will be
+ * able to bump up the refcount. We have to remove the
+ * log item from the transaction as we are about to release
+ * our reference to the buffer. If we don't, the unlock that
+ * occurs later in the xfs_trans_uncommit() will try to
+ * reference the buffer which we no longer have a hold on.
+ */
+ struct xfs_log_item_desc *lidp;
+
ASSERT(XFS_BUF_VALUSEMA(bip->bli_buf) <= 0);
trace_xfs_buf_item_unpin_stale(bip);
- /*
- * yes -- clear the xaction descriptor in-use flag
- * and free the chunk if required. We can safely
- * do some work here and then call buf_item_unpin
- * to do the rest because if the if is true, then
- * we are holding the buffer locked so no one else
- * will be able to bump up the refcount.
- */
- lidp = xfs_trans_find_item(tp, (xfs_log_item_t *) bip);
- stale = lidp->lid_flags & XFS_LID_BUF_STALE;
+ lidp = xfs_trans_find_item(tp, (xfs_log_item_t *)bip);
xfs_trans_free_item(tp, lidp);
+
/*
- * Since the transaction no longer refers to the buffer,
- * the buffer should no longer refer to the transaction.
+ * Since the transaction no longer refers to the buffer, the
+ * buffer should no longer refer to the transaction.
*/
- XFS_BUF_SET_FSPRIVATE2(bp, NULL);
+ XFS_BUF_SET_FSPRIVATE2(bip->bli_buf, NULL);
}
-
- xfs_buf_item_unpin(bip, stale);
-
- return;
+ xfs_buf_item_unpin(bip);
}
/*
* This is called to attempt to lock the buffer associated with this
* buf log item. Don't sleep on the buffer lock. If we can't get
- * the lock right away, return 0. If we can get the lock, pull the
- * buffer from the free list, mark it busy, and return 1.
+ * the lock right away, return 0. If we can get the lock, take a
+ * reference to the buffer. If this is a delayed write buffer that
+ * needs AIL help to be written back, invoke the pushbuf routine
+ * rather than the normal success path.
*/
STATIC uint
xfs_buf_item_trylock(
@@ -477,42 +493,39 @@ xfs_buf_item_trylock(
xfs_buf_t *bp;
bp = bip->bli_buf;
-
- if (XFS_BUF_ISPINNED(bp)) {
+ if (XFS_BUF_ISPINNED(bp))
return XFS_ITEM_PINNED;
- }
-
- if (!XFS_BUF_CPSEMA(bp)) {
+ if (!XFS_BUF_CPSEMA(bp))
return XFS_ITEM_LOCKED;
- }
- /*
- * Remove the buffer from the free list. Only do this
- * if it's on the free list. Private buffers like the
- * superblock buffer are not.
- */
+ /* take a reference to the buffer. */
XFS_BUF_HOLD(bp);
ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
trace_xfs_buf_item_trylock(bip);
+ if (XFS_BUF_ISDELAYWRITE(bp))
+ return XFS_ITEM_PUSHBUF;
return XFS_ITEM_SUCCESS;
}
/*
- * Release the buffer associated with the buf log item.
- * If there is no dirty logged data associated with the
- * buffer recorded in the buf log item, then free the
- * buf log item and remove the reference to it in the
- * buffer.
+ * Release the buffer associated with the buf log item. If there is no dirty
+ * logged data associated with the buffer recorded in the buf log item, then
+ * free the buf log item and remove the reference to it in the buffer.
+ *
+ * This call ignores the recursion count. It is only called when the buffer
+ * should REALLY be unlocked, regardless of the recursion count.
*
- * This call ignores the recursion count. It is only called
- * when the buffer should REALLY be unlocked, regardless
- * of the recursion count.
+ * We unconditionally drop the transaction's reference to the log item. If the
+ * item was logged, then another reference was taken when it was pinned, so we
+ * can safely drop the transaction reference now. This also allows us to avoid
+ * potential races with the unpin code freeing the bli by not referencing the
+ * bli after we've dropped the reference count.
*
- * If the XFS_BLI_HOLD flag is set in the buf log item, then
- * free the log item if necessary but do not unlock the buffer.
- * This is for support of xfs_trans_bhold(). Make sure the
- * XFS_BLI_HOLD field is cleared if we don't free the item.
+ * If the XFS_BLI_HOLD flag is set in the buf log item, then free the log item
+ * if necessary but do not unlock the buffer. This is for support of
+ * xfs_trans_bhold(). Make sure the XFS_BLI_HOLD field is cleared if we don't
+ * free the item.
*/
STATIC void
xfs_buf_item_unlock(
@@ -524,73 +537,54 @@ xfs_buf_item_unlock(
bp = bip->bli_buf;
- /*
- * Clear the buffer's association with this transaction.
- */
+ /* Clear the buffer's association with this transaction. */
XFS_BUF_SET_FSPRIVATE2(bp, NULL);
/*
- * If this is a transaction abort, don't return early.
- * Instead, allow the brelse to happen.
- * Normally it would be done for stale (cancelled) buffers
- * at unpin time, but we'll never go through the pin/unpin
- * cycle if we abort inside commit.
+ * If this is a transaction abort, don't return early. Instead, allow
+ * the brelse to happen. Normally it would be done for stale
+ * (cancelled) buffers at unpin time, but we'll never go through the
+ * pin/unpin cycle if we abort inside commit.
*/
aborted = (bip->bli_item.li_flags & XFS_LI_ABORTED) != 0;
/*
- * If the buf item is marked stale, then don't do anything.
- * We'll unlock the buffer and free the buf item when the
- * buffer is unpinned for the last time.
+ * Before possibly freeing the buf item, determine if we should
+ * release the buffer at the end of this routine.
*/
- if (bip->bli_flags & XFS_BLI_STALE) {
- bip->bli_flags &= ~XFS_BLI_LOGGED;
- trace_xfs_buf_item_unlock_stale(bip);
- ASSERT(bip->bli_format.blf_flags & XFS_BLI_CANCEL);
- if (!aborted)
- return;
- }
+ hold = bip->bli_flags & XFS_BLI_HOLD;
+
+ /* Clear the per transaction state. */
+ bip->bli_flags &= ~(XFS_BLI_LOGGED | XFS_BLI_HOLD);
/*
- * Drop the transaction's reference to the log item if
- * it was not logged as part of the transaction. Otherwise
- * we'll drop the reference in xfs_buf_item_unpin() when
- * the transaction is really through with the buffer.
+ * If the buf item is marked stale, then don't do anything. We'll
+ * unlock the buffer and free the buf item when the buffer is unpinned
+ * for the last time.
*/
- if (!(bip->bli_flags & XFS_BLI_LOGGED)) {
- atomic_dec(&bip->bli_refcount);
- } else {
- /*
- * Clear the logged flag since this is per
- * transaction state.
- */
- bip->bli_flags &= ~XFS_BLI_LOGGED;
+ if (bip->bli_flags & XFS_BLI_STALE) {
+ trace_xfs_buf_item_unlock_stale(bip);
+ ASSERT(bip->bli_format.blf_flags & XFS_BLF_CANCEL);
+ if (!aborted) {
+ atomic_dec(&bip->bli_refcount);
+ return;
+ }
}
- /*
- * Before possibly freeing the buf item, determine if we should
- * release the buffer at the end of this routine.
- */
- hold = bip->bli_flags & XFS_BLI_HOLD;
trace_xfs_buf_item_unlock(bip);
/*
- * If the buf item isn't tracking any data, free it.
- * Otherwise, if XFS_BLI_HOLD is set clear it.
+ * If the buf item isn't tracking any data, free it, otherwise drop the
+ * reference we hold to it.
*/
if (xfs_bitmap_empty(bip->bli_format.blf_data_map,
- bip->bli_format.blf_map_size)) {
+ bip->bli_format.blf_map_size))
xfs_buf_item_relse(bp);
- } else if (hold) {
- bip->bli_flags &= ~XFS_BLI_HOLD;
- }
+ else
+ atomic_dec(&bip->bli_refcount);
- /*
- * Release the buffer if XFS_BLI_HOLD was not set.
- */
- if (!hold) {
+ if (!hold)
xfs_buf_relse(bp);
- }
}
/*
@@ -626,11 +620,9 @@ xfs_buf_item_committed(
}
/*
- * This is called to asynchronously write the buffer associated with this
- * buf log item out to disk. The buffer will already have been locked by
- * a successful call to xfs_buf_item_trylock(). If the buffer still has
- * B_DELWRI set, then get it going out to disk with a call to bawrite().
- * If not, then just release the buffer.
+ * The buffer is locked, but is not a delayed write buffer. This happens
+ * if we race with IO completion and hence we don't want to try to write it
+ * again. Just release the buffer.
*/
STATIC void
xfs_buf_item_push(
@@ -642,17 +634,29 @@ xfs_buf_item_push(
trace_xfs_buf_item_push(bip);
bp = bip->bli_buf;
+ ASSERT(!XFS_BUF_ISDELAYWRITE(bp));
+ xfs_buf_relse(bp);
+}
- if (XFS_BUF_ISDELAYWRITE(bp)) {
- int error;
- error = xfs_bawrite(bip->bli_item.li_mountp, bp);
- if (error)
- xfs_fs_cmn_err(CE_WARN, bip->bli_item.li_mountp,
- "xfs_buf_item_push: pushbuf error %d on bip %p, bp %p",
- error, bip, bp);
- } else {
- xfs_buf_relse(bp);
- }
+/*
+ * The buffer is locked and is a delayed write buffer. Promote the buffer
+ * in the delayed write queue as the caller knows that they must invoke
+ * the xfsbufd to get this buffer written. We have to unlock the buffer
+ * to allow the xfsbufd to write it, too.
+ */
+STATIC void
+xfs_buf_item_pushbuf(
+ xfs_buf_log_item_t *bip)
+{
+ xfs_buf_t *bp;
+
+ ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
+ trace_xfs_buf_item_pushbuf(bip);
+
+ bp = bip->bli_buf;
+ ASSERT(XFS_BUF_ISDELAYWRITE(bp));
+ xfs_buf_delwri_promote(bp);
+ xfs_buf_relse(bp);
}
/* ARGSUSED */
@@ -669,7 +673,7 @@ static struct xfs_item_ops xfs_buf_item_ops = {
.iop_format = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
xfs_buf_item_format,
.iop_pin = (void(*)(xfs_log_item_t*))xfs_buf_item_pin,
- .iop_unpin = (void(*)(xfs_log_item_t*, int))xfs_buf_item_unpin,
+ .iop_unpin = (void(*)(xfs_log_item_t*))xfs_buf_item_unpin,
.iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t *))
xfs_buf_item_unpin_remove,
.iop_trylock = (uint(*)(xfs_log_item_t*))xfs_buf_item_trylock,
@@ -677,7 +681,7 @@ static struct xfs_item_ops xfs_buf_item_ops = {
.iop_committed = (xfs_lsn_t(*)(xfs_log_item_t*, xfs_lsn_t))
xfs_buf_item_committed,
.iop_push = (void(*)(xfs_log_item_t*))xfs_buf_item_push,
- .iop_pushbuf = NULL,
+ .iop_pushbuf = (void(*)(xfs_log_item_t*))xfs_buf_item_pushbuf,
.iop_committing = (void(*)(xfs_log_item_t*, xfs_lsn_t))
xfs_buf_item_committing
};
@@ -717,20 +721,17 @@ xfs_buf_item_init(
}
/*
- * chunks is the number of XFS_BLI_CHUNK size pieces
+ * chunks is the number of XFS_BLF_CHUNK size pieces
* the buffer can be divided into. Make sure not to
* truncate any pieces. map_size is the size of the
* bitmap needed to describe the chunks of the buffer.
*/
- chunks = (int)((XFS_BUF_COUNT(bp) + (XFS_BLI_CHUNK - 1)) >> XFS_BLI_SHIFT);
+ chunks = (int)((XFS_BUF_COUNT(bp) + (XFS_BLF_CHUNK - 1)) >> XFS_BLF_SHIFT);
map_size = (int)((chunks + NBWORD) >> BIT_TO_WORD_SHIFT);
bip = (xfs_buf_log_item_t*)kmem_zone_zalloc(xfs_buf_item_zone,
KM_SLEEP);
- bip->bli_item.li_type = XFS_LI_BUF;
- bip->bli_item.li_ops = &xfs_buf_item_ops;
- bip->bli_item.li_mountp = mp;
- bip->bli_item.li_ailp = mp->m_ail;
+ xfs_log_item_init(mp, &bip->bli_item, XFS_LI_BUF, &xfs_buf_item_ops);
bip->bli_buf = bp;
xfs_buf_hold(bp);
bip->bli_format.blf_type = XFS_LI_BUF;
@@ -793,8 +794,8 @@ xfs_buf_item_log(
/*
* Convert byte offsets to bit numbers.
*/
- first_bit = first >> XFS_BLI_SHIFT;
- last_bit = last >> XFS_BLI_SHIFT;
+ first_bit = first >> XFS_BLF_SHIFT;
+ last_bit = last >> XFS_BLF_SHIFT;
/*
* Calculate the total number of bits to be set.
diff --git a/fs/xfs/xfs_buf_item.h b/fs/xfs/xfs_buf_item.h
index 217f34af00cb..f20bb472d582 100644
--- a/fs/xfs/xfs_buf_item.h
+++ b/fs/xfs/xfs_buf_item.h
@@ -26,7 +26,7 @@ extern kmem_zone_t *xfs_buf_item_zone;
* have been logged.
* For 6.2 and beyond, this is XFS_LI_BUF. We use this to log everything.
*/
-typedef struct xfs_buf_log_format_t {
+typedef struct xfs_buf_log_format {
unsigned short blf_type; /* buf log item type indicator */
unsigned short blf_size; /* size of this item */
ushort blf_flags; /* misc state */
@@ -41,22 +41,22 @@ typedef struct xfs_buf_log_format_t {
* This flag indicates that the buffer contains on disk inodes
* and requires special recovery handling.
*/
-#define XFS_BLI_INODE_BUF 0x1
+#define XFS_BLF_INODE_BUF 0x1
/*
* This flag indicates that the buffer should not be replayed
* during recovery because its blocks are being freed.
*/
-#define XFS_BLI_CANCEL 0x2
+#define XFS_BLF_CANCEL 0x2
/*
* This flag indicates that the buffer contains on disk
* user or group dquots and may require special recovery handling.
*/
-#define XFS_BLI_UDQUOT_BUF 0x4
-#define XFS_BLI_PDQUOT_BUF 0x8
-#define XFS_BLI_GDQUOT_BUF 0x10
+#define XFS_BLF_UDQUOT_BUF 0x4
+#define XFS_BLF_PDQUOT_BUF 0x8
+#define XFS_BLF_GDQUOT_BUF 0x10
-#define XFS_BLI_CHUNK 128
-#define XFS_BLI_SHIFT 7
+#define XFS_BLF_CHUNK 128
+#define XFS_BLF_SHIFT 7
#define BIT_TO_WORD_SHIFT 5
#define NBWORD (NBBY * sizeof(unsigned int))
@@ -69,6 +69,7 @@ typedef struct xfs_buf_log_format_t {
#define XFS_BLI_LOGGED 0x08
#define XFS_BLI_INODE_ALLOC_BUF 0x10
#define XFS_BLI_STALE_INODE 0x20
+#define XFS_BLI_INODE_BUF 0x40
#define XFS_BLI_FLAGS \
{ XFS_BLI_HOLD, "HOLD" }, \
@@ -76,7 +77,8 @@ typedef struct xfs_buf_log_format_t {
{ XFS_BLI_STALE, "STALE" }, \
{ XFS_BLI_LOGGED, "LOGGED" }, \
{ XFS_BLI_INODE_ALLOC_BUF, "INODE_ALLOC" }, \
- { XFS_BLI_STALE_INODE, "STALE_INODE" }
+ { XFS_BLI_STALE_INODE, "STALE_INODE" }, \
+ { XFS_BLI_INODE_BUF, "INODE_BUF" }
#ifdef __KERNEL__
diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c
index c0c8869115b1..0ca556b4bf31 100644
--- a/fs/xfs/xfs_da_btree.c
+++ b/fs/xfs/xfs_da_btree.c
@@ -1534,8 +1534,8 @@ xfs_da_hashname(const __uint8_t *name, int namelen)
enum xfs_dacmp
xfs_da_compname(
struct xfs_da_args *args,
- const char *name,
- int len)
+ const unsigned char *name,
+ int len)
{
return (args->namelen == len && memcmp(args->name, name, len) == 0) ?
XFS_CMP_EXACT : XFS_CMP_DIFFERENT;
diff --git a/fs/xfs/xfs_da_btree.h b/fs/xfs/xfs_da_btree.h
index 30cd08f56a3a..fe9f5a8c1d2a 100644
--- a/fs/xfs/xfs_da_btree.h
+++ b/fs/xfs/xfs_da_btree.h
@@ -209,7 +209,8 @@ typedef struct xfs_da_state {
*/
struct xfs_nameops {
xfs_dahash_t (*hashname)(struct xfs_name *);
- enum xfs_dacmp (*compname)(struct xfs_da_args *, const char *, int);
+ enum xfs_dacmp (*compname)(struct xfs_da_args *,
+ const unsigned char *, int);
};
@@ -260,7 +261,7 @@ int xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno,
uint xfs_da_hashname(const __uint8_t *name_string, int name_length);
enum xfs_dacmp xfs_da_compname(struct xfs_da_args *args,
- const char *name, int len);
+ const unsigned char *name, int len);
xfs_da_state_t *xfs_da_state_alloc(void);
diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c
index 84ca1cf16a1e..7f159d2a429a 100644
--- a/fs/xfs/xfs_dfrag.c
+++ b/fs/xfs/xfs_dfrag.c
@@ -45,15 +45,21 @@
#include "xfs_vnodeops.h"
#include "xfs_trace.h"
+
+static int xfs_swap_extents(
+ xfs_inode_t *ip, /* target inode */
+ xfs_inode_t *tip, /* tmp inode */
+ xfs_swapext_t *sxp);
+
/*
- * Syssgi interface for swapext
+ * ioctl interface for swapext
*/
int
xfs_swapext(
xfs_swapext_t *sxp)
{
xfs_inode_t *ip, *tip;
- struct file *file, *target_file;
+ struct file *file, *tmp_file;
int error = 0;
/* Pull information for the target fd */
@@ -63,51 +69,54 @@ xfs_swapext(
goto out;
}
- if (!(file->f_mode & FMODE_WRITE) || (file->f_flags & O_APPEND)) {
+ if (!(file->f_mode & FMODE_WRITE) ||
+ !(file->f_mode & FMODE_READ) ||
+ (file->f_flags & O_APPEND)) {
error = XFS_ERROR(EBADF);
goto out_put_file;
}
- target_file = fget((int)sxp->sx_fdtmp);
- if (!target_file) {
+ tmp_file = fget((int)sxp->sx_fdtmp);
+ if (!tmp_file) {
error = XFS_ERROR(EINVAL);
goto out_put_file;
}
- if (!(target_file->f_mode & FMODE_WRITE) ||
- (target_file->f_flags & O_APPEND)) {
+ if (!(tmp_file->f_mode & FMODE_WRITE) ||
+ !(tmp_file->f_mode & FMODE_READ) ||
+ (tmp_file->f_flags & O_APPEND)) {
error = XFS_ERROR(EBADF);
- goto out_put_target_file;
+ goto out_put_tmp_file;
}
if (IS_SWAPFILE(file->f_path.dentry->d_inode) ||
- IS_SWAPFILE(target_file->f_path.dentry->d_inode)) {
+ IS_SWAPFILE(tmp_file->f_path.dentry->d_inode)) {
error = XFS_ERROR(EINVAL);
- goto out_put_target_file;
+ goto out_put_tmp_file;
}
ip = XFS_I(file->f_path.dentry->d_inode);
- tip = XFS_I(target_file->f_path.dentry->d_inode);
+ tip = XFS_I(tmp_file->f_path.dentry->d_inode);
if (ip->i_mount != tip->i_mount) {
error = XFS_ERROR(EINVAL);
- goto out_put_target_file;
+ goto out_put_tmp_file;
}
if (ip->i_ino == tip->i_ino) {
error = XFS_ERROR(EINVAL);
- goto out_put_target_file;
+ goto out_put_tmp_file;
}
if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
error = XFS_ERROR(EIO);
- goto out_put_target_file;
+ goto out_put_tmp_file;
}
error = xfs_swap_extents(ip, tip, sxp);
- out_put_target_file:
- fput(target_file);
+ out_put_tmp_file:
+ fput(tmp_file);
out_put_file:
fput(file);
out:
@@ -171,22 +180,32 @@ xfs_swap_extents_check_format(
XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) > tip->i_df.if_ext_max)
return EINVAL;
- /* Check root block of temp in btree form to max in target */
+ /*
+ * If we are in a btree format, check that the temp root block will fit
+ * in the target and that it has enough extents to be in btree format
+ * in the target.
+ *
+ * Note that we have to be careful to allow btree->extent conversions
+ * (a common defrag case) which will occur when the temp inode is in
+ * extent format...
+ */
if (tip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
- XFS_IFORK_BOFF(ip) &&
- tip->i_df.if_broot_bytes > XFS_IFORK_BOFF(ip))
+ ((XFS_IFORK_BOFF(ip) &&
+ tip->i_df.if_broot_bytes > XFS_IFORK_BOFF(ip)) ||
+ XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) <= ip->i_df.if_ext_max))
return EINVAL;
- /* Check root block of target in btree form to max in temp */
+ /* Reciprocal target->temp btree format checks */
if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
- XFS_IFORK_BOFF(tip) &&
- ip->i_df.if_broot_bytes > XFS_IFORK_BOFF(tip))
+ ((XFS_IFORK_BOFF(tip) &&
+ ip->i_df.if_broot_bytes > XFS_IFORK_BOFF(tip)) ||
+ XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) <= tip->i_df.if_ext_max))
return EINVAL;
return 0;
}
-int
+static int
xfs_swap_extents(
xfs_inode_t *ip, /* target inode */
xfs_inode_t *tip, /* tmp inode */
@@ -254,6 +273,9 @@ xfs_swap_extents(
goto out_unlock;
}
+ trace_xfs_swap_extent_before(ip, 0);
+ trace_xfs_swap_extent_before(tip, 1);
+
/* check inode formats now that data is flushed */
error = xfs_swap_extents_check_format(ip, tip);
if (error) {
@@ -421,6 +443,8 @@ xfs_swap_extents(
error = xfs_trans_commit(tp, XFS_TRANS_SWAPEXT);
+ trace_xfs_swap_extent_after(ip, 0);
+ trace_xfs_swap_extent_after(tip, 1);
out:
kmem_free(tempifp);
return error;
diff --git a/fs/xfs/xfs_dfrag.h b/fs/xfs/xfs_dfrag.h
index 4f55a6306558..20bdd935c121 100644
--- a/fs/xfs/xfs_dfrag.h
+++ b/fs/xfs/xfs_dfrag.h
@@ -48,9 +48,6 @@ typedef struct xfs_swapext
*/
int xfs_swapext(struct xfs_swapext *sx);
-int xfs_swap_extents(struct xfs_inode *ip, struct xfs_inode *tip,
- struct xfs_swapext *sxp);
-
#endif /* __KERNEL__ */
#endif /* __XFS_DFRAG_H__ */
diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c
index 93634a7e90e9..42520f041265 100644
--- a/fs/xfs/xfs_dir2.c
+++ b/fs/xfs/xfs_dir2.c
@@ -44,7 +44,7 @@
#include "xfs_vnodeops.h"
#include "xfs_trace.h"
-struct xfs_name xfs_name_dotdot = {"..", 2};
+struct xfs_name xfs_name_dotdot = { (unsigned char *)"..", 2};
/*
* ASCII case-insensitive (ie. A-Z) support for directories that was
@@ -66,8 +66,8 @@ xfs_ascii_ci_hashname(
STATIC enum xfs_dacmp
xfs_ascii_ci_compname(
struct xfs_da_args *args,
- const char *name,
- int len)
+ const unsigned char *name,
+ int len)
{
enum xfs_dacmp result;
int i;
@@ -247,7 +247,7 @@ xfs_dir_createname(
int
xfs_dir_cilookup_result(
struct xfs_da_args *args,
- const char *name,
+ const unsigned char *name,
int len)
{
if (args->cmpresult == XFS_CMP_DIFFERENT)
diff --git a/fs/xfs/xfs_dir2.h b/fs/xfs/xfs_dir2.h
index 1d9ef96f33aa..74a3b1057685 100644
--- a/fs/xfs/xfs_dir2.h
+++ b/fs/xfs/xfs_dir2.h
@@ -100,7 +100,7 @@ extern int xfs_dir2_isleaf(struct xfs_trans *tp, struct xfs_inode *dp,
extern int xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db,
struct xfs_dabuf *bp);
-extern int xfs_dir_cilookup_result(struct xfs_da_args *args, const char *name,
- int len);
+extern int xfs_dir_cilookup_result(struct xfs_da_args *args,
+ const unsigned char *name, int len);
#endif /* __XFS_DIR2_H__ */
diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c
index ddc4ecc7807f..779a267b0a84 100644
--- a/fs/xfs/xfs_dir2_block.c
+++ b/fs/xfs/xfs_dir2_block.c
@@ -57,8 +57,8 @@ static xfs_dahash_t xfs_dir_hash_dot, xfs_dir_hash_dotdot;
void
xfs_dir_startup(void)
{
- xfs_dir_hash_dot = xfs_da_hashname(".", 1);
- xfs_dir_hash_dotdot = xfs_da_hashname("..", 2);
+ xfs_dir_hash_dot = xfs_da_hashname((unsigned char *)".", 1);
+ xfs_dir_hash_dotdot = xfs_da_hashname((unsigned char *)"..", 2);
}
/*
@@ -513,8 +513,9 @@ xfs_dir2_block_getdents(
/*
* If it didn't fit, set the final offset to here & return.
*/
- if (filldir(dirent, dep->name, dep->namelen, cook & 0x7fffffff,
- be64_to_cpu(dep->inumber), DT_UNKNOWN)) {
+ if (filldir(dirent, (char *)dep->name, dep->namelen,
+ cook & 0x7fffffff, be64_to_cpu(dep->inumber),
+ DT_UNKNOWN)) {
*offset = cook & 0x7fffffff;
xfs_da_brelse(NULL, bp);
return 0;
diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c
index 29f484c11b3a..e2d89854ec9e 100644
--- a/fs/xfs/xfs_dir2_leaf.c
+++ b/fs/xfs/xfs_dir2_leaf.c
@@ -1081,7 +1081,7 @@ xfs_dir2_leaf_getdents(
dep = (xfs_dir2_data_entry_t *)ptr;
length = xfs_dir2_data_entsize(dep->namelen);
- if (filldir(dirent, dep->name, dep->namelen,
+ if (filldir(dirent, (char *)dep->name, dep->namelen,
xfs_dir2_byte_to_dataptr(mp, curoff) & 0x7fffffff,
be64_to_cpu(dep->inumber), DT_UNKNOWN))
break;
diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c
index ce6e355199b5..78fc4d9ae756 100644
--- a/fs/xfs/xfs_dir2_node.c
+++ b/fs/xfs/xfs_dir2_node.c
@@ -65,7 +65,7 @@ static int xfs_dir2_node_addname_int(xfs_da_args_t *args,
/*
* Log entries from a freespace block.
*/
-void
+STATIC void
xfs_dir2_free_log_bests(
xfs_trans_t *tp, /* transaction pointer */
xfs_dabuf_t *bp, /* freespace buffer */
diff --git a/fs/xfs/xfs_dir2_node.h b/fs/xfs/xfs_dir2_node.h
index dde72db3d695..82dfe7147195 100644
--- a/fs/xfs/xfs_dir2_node.h
+++ b/fs/xfs/xfs_dir2_node.h
@@ -75,8 +75,6 @@ xfs_dir2_db_to_fdindex(struct xfs_mount *mp, xfs_dir2_db_t db)
return ((db) % XFS_DIR2_MAX_FREE_BESTS(mp));
}
-extern void xfs_dir2_free_log_bests(struct xfs_trans *tp, struct xfs_dabuf *bp,
- int first, int last);
extern int xfs_dir2_leaf_to_node(struct xfs_da_args *args,
struct xfs_dabuf *lbp);
extern xfs_dahash_t xfs_dir2_leafn_lasthash(struct xfs_dabuf *bp, int *count);
diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/xfs_dir2_sf.c
index 9d4f17a69676..c1a5945d463a 100644
--- a/fs/xfs/xfs_dir2_sf.c
+++ b/fs/xfs/xfs_dir2_sf.c
@@ -782,7 +782,7 @@ xfs_dir2_sf_getdents(
}
ino = xfs_dir2_sf_get_inumber(sfp, xfs_dir2_sf_inumberp(sfep));
- if (filldir(dirent, sfep->name, sfep->namelen,
+ if (filldir(dirent, (char *)sfep->name, sfep->namelen,
off & 0x7fffffff, ino, DT_UNKNOWN)) {
*offset = off & 0x7fffffff;
return 0;
diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index 92d5cd5bf4f2..047b8a8e5c29 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -170,7 +170,7 @@ xfs_cmn_err(int panic_tag, int level, xfs_mount_t *mp, char *fmt, ...)
va_list ap;
#ifdef DEBUG
- xfs_panic_mask |= XFS_PTAG_SHUTDOWN_CORRUPT;
+ xfs_panic_mask |= (XFS_PTAG_SHUTDOWN_CORRUPT | XFS_PTAG_LOGRES);
#endif
if (xfs_panic_mask && (xfs_panic_mask & panic_tag)
@@ -186,18 +186,18 @@ xfs_cmn_err(int panic_tag, int level, xfs_mount_t *mp, char *fmt, ...)
void
xfs_error_report(
- char *tag,
- int level,
- xfs_mount_t *mp,
- char *fname,
- int linenum,
- inst_t *ra)
+ const char *tag,
+ int level,
+ struct xfs_mount *mp,
+ const char *filename,
+ int linenum,
+ inst_t *ra)
{
if (level <= xfs_error_level) {
xfs_cmn_err(XFS_PTAG_ERROR_REPORT,
CE_ALERT, mp,
"XFS internal error %s at line %d of file %s. Caller 0x%p\n",
- tag, linenum, fname, ra);
+ tag, linenum, filename, ra);
xfs_stack_trace();
}
@@ -205,15 +205,15 @@ xfs_error_report(
void
xfs_corruption_error(
- char *tag,
- int level,
- xfs_mount_t *mp,
- void *p,
- char *fname,
- int linenum,
- inst_t *ra)
+ const char *tag,
+ int level,
+ struct xfs_mount *mp,
+ void *p,
+ const char *filename,
+ int linenum,
+ inst_t *ra)
{
if (level <= xfs_error_level)
xfs_hex_dump(p, 16);
- xfs_error_report(tag, level, mp, fname, linenum, ra);
+ xfs_error_report(tag, level, mp, filename, linenum, ra);
}
diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h
index 0c93051c4651..c2c1a072bb82 100644
--- a/fs/xfs/xfs_error.h
+++ b/fs/xfs/xfs_error.h
@@ -29,10 +29,11 @@ extern int xfs_error_trap(int);
struct xfs_mount;
-extern void xfs_error_report(char *tag, int level, struct xfs_mount *mp,
- char *fname, int linenum, inst_t *ra);
-extern void xfs_corruption_error(char *tag, int level, struct xfs_mount *mp,
- void *p, char *fname, int linenum, inst_t *ra);
+extern void xfs_error_report(const char *tag, int level, struct xfs_mount *mp,
+ const char *filename, int linenum, inst_t *ra);
+extern void xfs_corruption_error(const char *tag, int level,
+ struct xfs_mount *mp, void *p, const char *filename,
+ int linenum, inst_t *ra);
#define XFS_ERROR_REPORT(e, lvl, mp) \
xfs_error_report(e, lvl, mp, __FILE__, __LINE__, __return_address)
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index 05a4bdd4be39..409fe81585fd 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -82,7 +82,7 @@ xfs_efi_item_format(xfs_efi_log_item_t *efip,
log_vector->i_addr = (xfs_caddr_t)&(efip->efi_format);
log_vector->i_len = size;
- XLOG_VEC_SET_TYPE(log_vector, XLOG_REG_TYPE_EFI_FORMAT);
+ log_vector->i_type = XLOG_REG_TYPE_EFI_FORMAT;
ASSERT(size >= sizeof(xfs_efi_log_format_t));
}
@@ -106,7 +106,7 @@ xfs_efi_item_pin(xfs_efi_log_item_t *efip)
*/
/*ARGSUSED*/
STATIC void
-xfs_efi_item_unpin(xfs_efi_log_item_t *efip, int stale)
+xfs_efi_item_unpin(xfs_efi_log_item_t *efip)
{
struct xfs_ail *ailp = efip->efi_item.li_ailp;
@@ -224,7 +224,7 @@ static struct xfs_item_ops xfs_efi_item_ops = {
.iop_format = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
xfs_efi_item_format,
.iop_pin = (void(*)(xfs_log_item_t*))xfs_efi_item_pin,
- .iop_unpin = (void(*)(xfs_log_item_t*, int))xfs_efi_item_unpin,
+ .iop_unpin = (void(*)(xfs_log_item_t*))xfs_efi_item_unpin,
.iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t *))
xfs_efi_item_unpin_remove,
.iop_trylock = (uint(*)(xfs_log_item_t*))xfs_efi_item_trylock,
@@ -259,10 +259,7 @@ xfs_efi_init(xfs_mount_t *mp,
KM_SLEEP);
}
- efip->efi_item.li_type = XFS_LI_EFI;
- efip->efi_item.li_ops = &xfs_efi_item_ops;
- efip->efi_item.li_mountp = mp;
- efip->efi_item.li_ailp = mp->m_ail;
+ xfs_log_item_init(mp, &efip->efi_item, XFS_LI_EFI, &xfs_efi_item_ops);
efip->efi_format.efi_nextents = nextents;
efip->efi_format.efi_id = (__psint_t)(void*)efip;
@@ -406,7 +403,7 @@ xfs_efd_item_format(xfs_efd_log_item_t *efdp,
log_vector->i_addr = (xfs_caddr_t)&(efdp->efd_format);
log_vector->i_len = size;
- XLOG_VEC_SET_TYPE(log_vector, XLOG_REG_TYPE_EFD_FORMAT);
+ log_vector->i_type = XLOG_REG_TYPE_EFD_FORMAT;
ASSERT(size >= sizeof(xfs_efd_log_format_t));
}
@@ -428,7 +425,7 @@ xfs_efd_item_pin(xfs_efd_log_item_t *efdp)
*/
/*ARGSUSED*/
STATIC void
-xfs_efd_item_unpin(xfs_efd_log_item_t *efdp, int stale)
+xfs_efd_item_unpin(xfs_efd_log_item_t *efdp)
{
return;
}
@@ -518,7 +515,7 @@ static struct xfs_item_ops xfs_efd_item_ops = {
.iop_format = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
xfs_efd_item_format,
.iop_pin = (void(*)(xfs_log_item_t*))xfs_efd_item_pin,
- .iop_unpin = (void(*)(xfs_log_item_t*, int))xfs_efd_item_unpin,
+ .iop_unpin = (void(*)(xfs_log_item_t*))xfs_efd_item_unpin,
.iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t*))
xfs_efd_item_unpin_remove,
.iop_trylock = (uint(*)(xfs_log_item_t*))xfs_efd_item_trylock,
@@ -554,10 +551,7 @@ xfs_efd_init(xfs_mount_t *mp,
KM_SLEEP);
}
- efdp->efd_item.li_type = XFS_LI_EFD;
- efdp->efd_item.li_ops = &xfs_efd_item_ops;
- efdp->efd_item.li_mountp = mp;
- efdp->efd_item.li_ailp = mp->m_ail;
+ xfs_log_item_init(mp, &efdp->efd_item, XFS_LI_EFD, &xfs_efd_item_ops);
efdp->efd_efip = efip;
efdp->efd_format.efd_nextents = nextents;
efdp->efd_format.efd_efi_id = efip->efi_format.efi_id;
diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c
index a631e1451abb..390850ee6603 100644
--- a/fs/xfs/xfs_filestream.c
+++ b/fs/xfs/xfs_filestream.c
@@ -140,6 +140,7 @@ _xfs_filestream_pick_ag(
int flags,
xfs_extlen_t minlen)
{
+ int streams, max_streams;
int err, trylock, nscan;
xfs_extlen_t longest, free, minfree, maxfree = 0;
xfs_agnumber_t ag, max_ag = NULLAGNUMBER;
@@ -155,15 +156,15 @@ _xfs_filestream_pick_ag(
trylock = XFS_ALLOC_FLAG_TRYLOCK;
for (nscan = 0; 1; nscan++) {
-
- TRACE_AG_SCAN(mp, ag, xfs_filestream_peek_ag(mp, ag));
-
- pag = mp->m_perag + ag;
+ pag = xfs_perag_get(mp, ag);
+ TRACE_AG_SCAN(mp, ag, atomic_read(&pag->pagf_fstrms));
if (!pag->pagf_init) {
err = xfs_alloc_pagf_init(mp, NULL, ag, trylock);
- if (err && !trylock)
+ if (err && !trylock) {
+ xfs_perag_put(pag);
return err;
+ }
}
/* Might fail sometimes during the 1st pass with trylock set. */
@@ -173,6 +174,7 @@ _xfs_filestream_pick_ag(
/* Keep track of the AG with the most free blocks. */
if (pag->pagf_freeblks > maxfree) {
maxfree = pag->pagf_freeblks;
+ max_streams = atomic_read(&pag->pagf_fstrms);
max_ag = ag;
}
@@ -195,6 +197,8 @@ _xfs_filestream_pick_ag(
/* Break out, retaining the reference on the AG. */
free = pag->pagf_freeblks;
+ streams = atomic_read(&pag->pagf_fstrms);
+ xfs_perag_put(pag);
*agp = ag;
break;
}
@@ -202,6 +206,7 @@ _xfs_filestream_pick_ag(
/* Drop the reference on this AG, it's not usable. */
xfs_filestream_put_ag(mp, ag);
next_ag:
+ xfs_perag_put(pag);
/* Move to the next AG, wrapping to AG 0 if necessary. */
if (++ag >= mp->m_sb.sb_agcount)
ag = 0;
@@ -229,6 +234,7 @@ next_ag:
if (max_ag != NULLAGNUMBER) {
xfs_filestream_get_ag(mp, max_ag);
TRACE_AG_PICK1(mp, max_ag, maxfree);
+ streams = max_streams;
free = maxfree;
*agp = max_ag;
break;
@@ -240,16 +246,14 @@ next_ag:
return 0;
}
- TRACE_AG_PICK2(mp, startag, *agp, xfs_filestream_peek_ag(mp, *agp),
- free, nscan, flags);
+ TRACE_AG_PICK2(mp, startag, *agp, streams, free, nscan, flags);
return 0;
}
/*
* Set the allocation group number for a file or a directory, updating inode
- * references and per-AG references as appropriate. Must be called with the
- * m_peraglock held in read mode.
+ * references and per-AG references as appropriate.
*/
static int
_xfs_filestream_update_ag(
@@ -451,20 +455,6 @@ xfs_filestream_unmount(
}
/*
- * If the mount point's m_perag array is going to be reallocated, all
- * outstanding cache entries must be flushed to avoid accessing reference count
- * addresses that have been freed. The call to xfs_filestream_flush() must be
- * made inside the block that holds the m_peraglock in write mode to do the
- * reallocation.
- */
-void
-xfs_filestream_flush(
- xfs_mount_t *mp)
-{
- xfs_mru_cache_flush(mp->m_filestream);
-}
-
-/*
* Return the AG of the filestream the file or directory belongs to, or
* NULLAGNUMBER otherwise.
*/
@@ -526,7 +516,6 @@ xfs_filestream_associate(
mp = pip->i_mount;
cache = mp->m_filestream;
- down_read(&mp->m_peraglock);
/*
* We have a problem, Houston.
@@ -543,10 +532,8 @@ xfs_filestream_associate(
*
* So, if we can't get the iolock without sleeping then just give up
*/
- if (!xfs_ilock_nowait(pip, XFS_IOLOCK_EXCL)) {
- up_read(&mp->m_peraglock);
+ if (!xfs_ilock_nowait(pip, XFS_IOLOCK_EXCL))
return 1;
- }
/* If the parent directory is already in the cache, use its AG. */
item = xfs_mru_cache_lookup(cache, pip->i_ino);
@@ -601,7 +588,6 @@ exit_did_pick:
exit:
xfs_iunlock(pip, XFS_IOLOCK_EXCL);
- up_read(&mp->m_peraglock);
return -err;
}
diff --git a/fs/xfs/xfs_filestream.h b/fs/xfs/xfs_filestream.h
index 4aba67c5f64f..260f757bbc5d 100644
--- a/fs/xfs/xfs_filestream.h
+++ b/fs/xfs/xfs_filestream.h
@@ -79,12 +79,21 @@ extern ktrace_t *xfs_filestreams_trace_buf;
* the cache that reference per-ag array elements that have since been
* reallocated.
*/
+/*
+ * xfs_filestream_peek_ag is only used in tracing code
+ */
static inline int
xfs_filestream_peek_ag(
xfs_mount_t *mp,
xfs_agnumber_t agno)
{
- return atomic_read(&mp->m_perag[agno].pagf_fstrms);
+ struct xfs_perag *pag;
+ int ret;
+
+ pag = xfs_perag_get(mp, agno);
+ ret = atomic_read(&pag->pagf_fstrms);
+ xfs_perag_put(pag);
+ return ret;
}
static inline int
@@ -92,7 +101,13 @@ xfs_filestream_get_ag(
xfs_mount_t *mp,
xfs_agnumber_t agno)
{
- return atomic_inc_return(&mp->m_perag[agno].pagf_fstrms);
+ struct xfs_perag *pag;
+ int ret;
+
+ pag = xfs_perag_get(mp, agno);
+ ret = atomic_inc_return(&pag->pagf_fstrms);
+ xfs_perag_put(pag);
+ return ret;
}
static inline int
@@ -100,7 +115,13 @@ xfs_filestream_put_ag(
xfs_mount_t *mp,
xfs_agnumber_t agno)
{
- return atomic_dec_return(&mp->m_perag[agno].pagf_fstrms);
+ struct xfs_perag *pag;
+ int ret;
+
+ pag = xfs_perag_get(mp, agno);
+ ret = atomic_dec_return(&pag->pagf_fstrms);
+ xfs_perag_put(pag);
+ return ret;
}
/* allocation selection flags */
@@ -114,7 +135,6 @@ int xfs_filestream_init(void);
void xfs_filestream_uninit(void);
int xfs_filestream_mount(struct xfs_mount *mp);
void xfs_filestream_unmount(struct xfs_mount *mp);
-void xfs_filestream_flush(struct xfs_mount *mp);
xfs_agnumber_t xfs_filestream_lookup_ag(struct xfs_inode *ip);
int xfs_filestream_associate(struct xfs_inode *dip, struct xfs_inode *ip);
void xfs_filestream_deassociate(struct xfs_inode *ip);
diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h
index f52ac276277e..7cf7220e7d5f 100644
--- a/fs/xfs/xfs_fs.h
+++ b/fs/xfs/xfs_fs.h
@@ -292,7 +292,8 @@ typedef struct xfs_bstat {
__s32 bs_extents; /* number of extents */
__u32 bs_gen; /* generation count */
__u16 bs_projid; /* project id */
- unsigned char bs_pad[14]; /* pad space, unused */
+ __u16 bs_forkoff; /* inode fork offset in bytes */
+ unsigned char bs_pad[12]; /* pad space, unused */
__u32 bs_dmevmask; /* DMIG event mask */
__u16 bs_dmstate; /* DMIG state info */
__u16 bs_aextents; /* attribute number of extents */
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index a13919a6a364..37a6f62c57b6 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -167,27 +167,14 @@ xfs_growfs_data_private(
}
new = nb - mp->m_sb.sb_dblocks;
oagcount = mp->m_sb.sb_agcount;
- if (nagcount > oagcount) {
- void *new_perag, *old_perag;
-
- xfs_filestream_flush(mp);
-
- new_perag = kmem_zalloc(sizeof(xfs_perag_t) * nagcount,
- KM_MAYFAIL);
- if (!new_perag)
- return XFS_ERROR(ENOMEM);
-
- down_write(&mp->m_peraglock);
- memcpy(new_perag, mp->m_perag, sizeof(xfs_perag_t) * oagcount);
- old_perag = mp->m_perag;
- mp->m_perag = new_perag;
-
- mp->m_flags |= XFS_MOUNT_32BITINODES;
- nagimax = xfs_initialize_perag(mp, nagcount);
- up_write(&mp->m_peraglock);
- kmem_free(old_perag);
+ /* allocate the new per-ag structures */
+ if (nagcount > oagcount) {
+ error = xfs_initialize_perag(mp, nagcount, &nagimax);
+ if (error)
+ return error;
}
+
tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFS);
tp->t_flags |= XFS_TRANS_RESERVE;
if ((error = xfs_trans_reserve(tp, XFS_GROWFS_SPACE_RES(mp),
@@ -196,6 +183,11 @@ xfs_growfs_data_private(
return error;
}
+ /*
+ * Write new AG headers to disk. Non-transactional, but written
+ * synchronously so they are completed prior to the growfs transaction
+ * being logged.
+ */
nfree = 0;
for (agno = nagcount - 1; agno >= oagcount; agno--, new -= agsize) {
/*
@@ -359,6 +351,12 @@ xfs_growfs_data_private(
goto error0;
}
}
+
+ /*
+ * Update changed superblock fields transactionally. These are not
+ * seen by the rest of the world until the transaction commit applies
+ * them atomically to the superblock.
+ */
if (nagcount > oagcount)
xfs_trans_mod_sb(tp, XFS_TRANS_SB_AGCOUNT, nagcount - oagcount);
if (nb > mp->m_sb.sb_dblocks)
@@ -369,9 +367,9 @@ xfs_growfs_data_private(
if (dpct)
xfs_trans_mod_sb(tp, XFS_TRANS_SB_IMAXPCT, dpct);
error = xfs_trans_commit(tp, 0);
- if (error) {
+ if (error)
return error;
- }
+
/* New allocation groups fully initialized, so update mount struct */
if (nagimax)
mp->m_maxagi = nagimax;
@@ -381,6 +379,8 @@ xfs_growfs_data_private(
mp->m_maxicount = icount << mp->m_sb.sb_inopblog;
} else
mp->m_maxicount = 0;
+
+ /* update secondary superblocks. */
for (agno = 1; agno < nagcount; agno++) {
error = xfs_read_buf(mp, mp->m_ddev_targp,
XFS_AGB_TO_DADDR(mp, agno, XFS_SB_BLOCK(mp)),
diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c
index cb907ba69c4c..c7142a064c48 100644
--- a/fs/xfs/xfs_ialloc.c
+++ b/fs/xfs/xfs_ialloc.c
@@ -205,7 +205,7 @@ xfs_ialloc_inode_init(
d = XFS_AGB_TO_DADDR(mp, agno, agbno + (j * blks_per_cluster));
fbuf = xfs_trans_get_buf(tp, mp->m_ddev_targp, d,
mp->m_bsize * blks_per_cluster,
- XFS_BUF_LOCK);
+ XBF_LOCK);
ASSERT(fbuf);
ASSERT(!XFS_BUF_GETERROR(fbuf));
@@ -253,6 +253,7 @@ xfs_ialloc_ag_alloc(
xfs_agino_t thisino; /* current inode number, for loop */
int isaligned = 0; /* inode allocation at stripe unit */
/* boundary */
+ struct xfs_perag *pag;
args.tp = tp;
args.mp = tp->t_mountp;
@@ -382,9 +383,9 @@ xfs_ialloc_ag_alloc(
newino = XFS_OFFBNO_TO_AGINO(args.mp, args.agbno, 0);
be32_add_cpu(&agi->agi_count, newlen);
be32_add_cpu(&agi->agi_freecount, newlen);
- down_read(&args.mp->m_peraglock);
- args.mp->m_perag[agno].pagi_freecount += newlen;
- up_read(&args.mp->m_peraglock);
+ pag = xfs_perag_get(args.mp, agno);
+ pag->pagi_freecount += newlen;
+ xfs_perag_put(pag);
agi->agi_newino = cpu_to_be32(newino);
/*
@@ -486,9 +487,8 @@ xfs_ialloc_ag_select(
*/
agno = pagno;
flags = XFS_ALLOC_FLAG_TRYLOCK;
- down_read(&mp->m_peraglock);
for (;;) {
- pag = &mp->m_perag[agno];
+ pag = xfs_perag_get(mp, agno);
if (!pag->pagi_init) {
if (xfs_ialloc_read_agi(mp, tp, agno, &agbp)) {
agbp = NULL;
@@ -527,7 +527,7 @@ xfs_ialloc_ag_select(
agbp = NULL;
goto nextag;
}
- up_read(&mp->m_peraglock);
+ xfs_perag_put(pag);
return agbp;
}
}
@@ -535,22 +535,19 @@ unlock_nextag:
if (agbp)
xfs_trans_brelse(tp, agbp);
nextag:
+ xfs_perag_put(pag);
/*
* No point in iterating over the rest, if we're shutting
* down.
*/
- if (XFS_FORCED_SHUTDOWN(mp)) {
- up_read(&mp->m_peraglock);
+ if (XFS_FORCED_SHUTDOWN(mp))
return NULL;
- }
agno++;
if (agno >= agcount)
agno = 0;
if (agno == pagno) {
- if (flags == 0) {
- up_read(&mp->m_peraglock);
+ if (flags == 0)
return NULL;
- }
flags = 0;
}
}
@@ -672,6 +669,7 @@ xfs_dialloc(
xfs_agnumber_t tagno; /* testing allocation group number */
xfs_btree_cur_t *tcur; /* temp cursor */
xfs_inobt_rec_incore_t trec; /* temp inode allocation record */
+ struct xfs_perag *pag;
if (*IO_agbp == NULL) {
@@ -771,13 +769,13 @@ nextag:
*inop = NULLFSINO;
return noroom ? ENOSPC : 0;
}
- down_read(&mp->m_peraglock);
- if (mp->m_perag[tagno].pagi_inodeok == 0) {
- up_read(&mp->m_peraglock);
+ pag = xfs_perag_get(mp, tagno);
+ if (pag->pagi_inodeok == 0) {
+ xfs_perag_put(pag);
goto nextag;
}
error = xfs_ialloc_read_agi(mp, tp, tagno, &agbp);
- up_read(&mp->m_peraglock);
+ xfs_perag_put(pag);
if (error)
goto nextag;
agi = XFS_BUF_TO_AGI(agbp);
@@ -790,6 +788,7 @@ nextag:
*/
agno = tagno;
*IO_agbp = NULL;
+ pag = xfs_perag_get(mp, agno);
restart_pagno:
cur = xfs_inobt_init_cursor(mp, tp, agbp, be32_to_cpu(agi->agi_seqno));
@@ -808,7 +807,6 @@ nextag:
* If in the same AG as the parent, try to get near the parent.
*/
if (pagno == agno) {
- xfs_perag_t *pag = &mp->m_perag[agno];
int doneleft; /* done, to the left */
int doneright; /* done, to the right */
int searchdistance = 10;
@@ -1006,9 +1004,7 @@ alloc_inode:
goto error0;
be32_add_cpu(&agi->agi_freecount, -1);
xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT);
- down_read(&mp->m_peraglock);
- mp->m_perag[tagno].pagi_freecount--;
- up_read(&mp->m_peraglock);
+ pag->pagi_freecount--;
error = xfs_check_agi_freecount(cur, agi);
if (error)
@@ -1016,12 +1012,14 @@ alloc_inode:
xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, -1);
+ xfs_perag_put(pag);
*inop = ino;
return 0;
error1:
xfs_btree_del_cursor(tcur, XFS_BTREE_ERROR);
error0:
xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
+ xfs_perag_put(pag);
return error;
}
@@ -1052,6 +1050,7 @@ xfs_difree(
xfs_mount_t *mp; /* mount structure for filesystem */
int off; /* offset of inode in inode chunk */
xfs_inobt_rec_incore_t rec; /* btree record */
+ struct xfs_perag *pag;
mp = tp->t_mountp;
@@ -1088,9 +1087,7 @@ xfs_difree(
/*
* Get the allocation group header.
*/
- down_read(&mp->m_peraglock);
error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
- up_read(&mp->m_peraglock);
if (error) {
cmn_err(CE_WARN,
"xfs_difree: xfs_ialloc_read_agi() returned an error %d on %s. Returning error.",
@@ -1157,9 +1154,9 @@ xfs_difree(
be32_add_cpu(&agi->agi_count, -ilen);
be32_add_cpu(&agi->agi_freecount, -(ilen - 1));
xfs_ialloc_log_agi(tp, agbp, XFS_AGI_COUNT | XFS_AGI_FREECOUNT);
- down_read(&mp->m_peraglock);
- mp->m_perag[agno].pagi_freecount -= ilen - 1;
- up_read(&mp->m_peraglock);
+ pag = xfs_perag_get(mp, agno);
+ pag->pagi_freecount -= ilen - 1;
+ xfs_perag_put(pag);
xfs_trans_mod_sb(tp, XFS_TRANS_SB_ICOUNT, -ilen);
xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, -(ilen - 1));
@@ -1188,9 +1185,9 @@ xfs_difree(
*/
be32_add_cpu(&agi->agi_freecount, 1);
xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT);
- down_read(&mp->m_peraglock);
- mp->m_perag[agno].pagi_freecount++;
- up_read(&mp->m_peraglock);
+ pag = xfs_perag_get(mp, agno);
+ pag->pagi_freecount++;
+ xfs_perag_put(pag);
xfs_trans_mod_sb(tp, XFS_TRANS_SB_IFREE, 1);
}
@@ -1206,6 +1203,63 @@ error0:
return error;
}
+STATIC int
+xfs_imap_lookup(
+ struct xfs_mount *mp,
+ struct xfs_trans *tp,
+ xfs_agnumber_t agno,
+ xfs_agino_t agino,
+ xfs_agblock_t agbno,
+ xfs_agblock_t *chunk_agbno,
+ xfs_agblock_t *offset_agbno,
+ int flags)
+{
+ struct xfs_inobt_rec_incore rec;
+ struct xfs_btree_cur *cur;
+ struct xfs_buf *agbp;
+ xfs_agino_t startino;
+ int error;
+ int i;
+
+ error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
+ if (error) {
+ xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
+ "xfs_ialloc_read_agi() returned "
+ "error %d, agno %d",
+ error, agno);
+ return error;
+ }
+
+ /*
+ * derive and lookup the exact inode record for the given agino. If the
+ * record cannot be found, then it's an invalid inode number and we
+ * should abort.
+ */
+ cur = xfs_inobt_init_cursor(mp, tp, agbp, agno);
+ startino = agino & ~(XFS_IALLOC_INODES(mp) - 1);
+ error = xfs_inobt_lookup(cur, startino, XFS_LOOKUP_EQ, &i);
+ if (!error) {
+ if (i)
+ error = xfs_inobt_get_rec(cur, &rec, &i);
+ if (!error && i == 0)
+ error = EINVAL;
+ }
+
+ xfs_trans_brelse(tp, agbp);
+ xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
+ if (error)
+ return error;
+
+ /* for untrusted inodes check it is allocated first */
+ if ((flags & XFS_IGET_UNTRUSTED) &&
+ (rec.ir_free & XFS_INOBT_MASK(agino - rec.ir_startino)))
+ return EINVAL;
+
+ *chunk_agbno = XFS_AGINO_TO_AGBNO(mp, rec.ir_startino);
+ *offset_agbno = agbno - *chunk_agbno;
+ return 0;
+}
+
/*
* Return the location of the inode in imap, for mapping it into a buffer.
*/
@@ -1238,8 +1292,11 @@ xfs_imap(
if (agno >= mp->m_sb.sb_agcount || agbno >= mp->m_sb.sb_agblocks ||
ino != XFS_AGINO_TO_INO(mp, agno, agino)) {
#ifdef DEBUG
- /* no diagnostics for bulkstat, ino comes from userspace */
- if (flags & XFS_IGET_BULKSTAT)
+ /*
+ * Don't output diagnostic information for untrusted inodes
+ * as they can be invalid without implying corruption.
+ */
+ if (flags & XFS_IGET_UNTRUSTED)
return XFS_ERROR(EINVAL);
if (agno >= mp->m_sb.sb_agcount) {
xfs_fs_cmn_err(CE_ALERT, mp,
@@ -1266,6 +1323,23 @@ xfs_imap(
return XFS_ERROR(EINVAL);
}
+ blks_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_blocklog;
+
+ /*
+ * For bulkstat and handle lookups, we have an untrusted inode number
+ * that we have to verify is valid. We cannot do this just by reading
+ * the inode buffer as it may have been unlinked and removed leaving
+ * inodes in stale state on disk. Hence we have to do a btree lookup
+ * in all cases where an untrusted inode number is passed.
+ */
+ if (flags & XFS_IGET_UNTRUSTED) {
+ error = xfs_imap_lookup(mp, tp, agno, agino, agbno,
+ &chunk_agbno, &offset_agbno, flags);
+ if (error)
+ return error;
+ goto out_map;
+ }
+
/*
* If the inode cluster size is the same as the blocksize or
* smaller we get to the buffer by simple arithmetics.
@@ -1280,24 +1354,6 @@ xfs_imap(
return 0;
}
- blks_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_blocklog;
-
- /*
- * If we get a block number passed from bulkstat we can use it to
- * find the buffer easily.
- */
- if (imap->im_blkno) {
- offset = XFS_INO_TO_OFFSET(mp, ino);
- ASSERT(offset < mp->m_sb.sb_inopblock);
-
- cluster_agbno = xfs_daddr_to_agbno(mp, imap->im_blkno);
- offset += (agbno - cluster_agbno) * mp->m_sb.sb_inopblock;
-
- imap->im_len = XFS_FSB_TO_BB(mp, blks_per_cluster);
- imap->im_boffset = (ushort)(offset << mp->m_sb.sb_inodelog);
- return 0;
- }
-
/*
* If the inode chunks are aligned then use simple maths to
* find the location. Otherwise we have to do a btree
@@ -1307,52 +1363,13 @@ xfs_imap(
offset_agbno = agbno & mp->m_inoalign_mask;
chunk_agbno = agbno - offset_agbno;
} else {
- xfs_btree_cur_t *cur; /* inode btree cursor */
- xfs_inobt_rec_incore_t chunk_rec;
- xfs_buf_t *agbp; /* agi buffer */
- int i; /* temp state */
-
- down_read(&mp->m_peraglock);
- error = xfs_ialloc_read_agi(mp, tp, agno, &agbp);
- up_read(&mp->m_peraglock);
- if (error) {
- xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
- "xfs_ialloc_read_agi() returned "
- "error %d, agno %d",
- error, agno);
- return error;
- }
-
- cur = xfs_inobt_init_cursor(mp, tp, agbp, agno);
- error = xfs_inobt_lookup(cur, agino, XFS_LOOKUP_LE, &i);
- if (error) {
- xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
- "xfs_inobt_lookup() failed");
- goto error0;
- }
-
- error = xfs_inobt_get_rec(cur, &chunk_rec, &i);
- if (error) {
- xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
- "xfs_inobt_get_rec() failed");
- goto error0;
- }
- if (i == 0) {
-#ifdef DEBUG
- xfs_fs_cmn_err(CE_ALERT, mp, "xfs_imap: "
- "xfs_inobt_get_rec() failed");
-#endif /* DEBUG */
- error = XFS_ERROR(EINVAL);
- }
- error0:
- xfs_trans_brelse(tp, agbp);
- xfs_btree_del_cursor(cur, XFS_BTREE_NOERROR);
+ error = xfs_imap_lookup(mp, tp, agno, agino, agbno,
+ &chunk_agbno, &offset_agbno, flags);
if (error)
return error;
- chunk_agbno = XFS_AGINO_TO_AGBNO(mp, chunk_rec.ir_startino);
- offset_agbno = agbno - chunk_agbno;
}
+out_map:
ASSERT(agbno >= chunk_agbno);
cluster_agbno = chunk_agbno +
((offset_agbno / blks_per_cluster) * blks_per_cluster);
@@ -1379,7 +1396,6 @@ xfs_imap(
XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks));
return XFS_ERROR(EINVAL);
}
-
return 0;
}
@@ -1523,8 +1539,7 @@ xfs_ialloc_read_agi(
return error;
agi = XFS_BUF_TO_AGI(*bpp);
- pag = &mp->m_perag[agno];
-
+ pag = xfs_perag_get(mp, agno);
if (!pag->pagi_init) {
pag->pagi_freecount = be32_to_cpu(agi->agi_freecount);
pag->pagi_count = be32_to_cpu(agi->agi_count);
@@ -1537,6 +1552,7 @@ xfs_ialloc_read_agi(
*/
ASSERT(pag->pagi_freecount == be32_to_cpu(agi->agi_freecount) ||
XFS_FORCED_SHUTDOWN(mp));
+ xfs_perag_put(pag);
return 0;
}
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index 155e798f30a1..8f8b91be2c99 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -190,13 +190,12 @@ xfs_iget_cache_hit(
trace_xfs_iget_reclaim(ip);
/*
- * We need to set XFS_INEW atomically with clearing the
- * reclaimable tag so that we do have an indicator of the
- * inode still being initialized.
+ * We need to set XFS_IRECLAIM to prevent xfs_reclaim_inode
+ * from stomping over us while we recycle the inode. We can't
+ * clear the radix tree reclaimable tag yet as it requires
+ * pag_ici_lock to be held exclusive.
*/
- ip->i_flags |= XFS_INEW;
- ip->i_flags &= ~XFS_IRECLAIMABLE;
- __xfs_inode_clear_reclaim_tag(mp, pag, ip);
+ ip->i_flags |= XFS_IRECLAIM;
spin_unlock(&ip->i_flags_lock);
read_unlock(&pag->pag_ici_lock);
@@ -216,7 +215,15 @@ xfs_iget_cache_hit(
trace_xfs_iget_reclaim(ip);
goto out_error;
}
+
+ write_lock(&pag->pag_ici_lock);
+ spin_lock(&ip->i_flags_lock);
+ ip->i_flags &= ~(XFS_IRECLAIMABLE | XFS_IRECLAIM);
+ ip->i_flags |= XFS_INEW;
+ __xfs_inode_clear_reclaim_tag(mp, pag, ip);
inode->i_state = I_NEW;
+ spin_unlock(&ip->i_flags_lock);
+ write_unlock(&pag->pag_ici_lock);
} else {
/* If the VFS inode is being torn down, pause and try again. */
if (!igrab(inode)) {
@@ -252,7 +259,6 @@ xfs_iget_cache_miss(
xfs_trans_t *tp,
xfs_ino_t ino,
struct xfs_inode **ipp,
- xfs_daddr_t bno,
int flags,
int lock_flags)
{
@@ -265,7 +271,7 @@ xfs_iget_cache_miss(
if (!ip)
return ENOMEM;
- error = xfs_iread(mp, tp, ip, bno, flags);
+ error = xfs_iread(mp, tp, ip, flags);
if (error)
goto out_destroy;
@@ -351,8 +357,6 @@ out_destroy:
* within the file system for the inode being requested.
* lock_flags -- flags indicating how to lock the inode. See the comment
* for xfs_ilock() for a list of valid values.
- * bno -- the block number starting the buffer containing the inode,
- * if known (as by bulkstat), else 0.
*/
int
xfs_iget(
@@ -361,8 +365,7 @@ xfs_iget(
xfs_ino_t ino,
uint flags,
uint lock_flags,
- xfs_inode_t **ipp,
- xfs_daddr_t bno)
+ xfs_inode_t **ipp)
{
xfs_inode_t *ip;
int error;
@@ -374,10 +377,7 @@ xfs_iget(
return EINVAL;
/* get the perag structure and ensure that it's inode capable */
- pag = xfs_get_perag(mp, ino);
- if (!pag->pagi_inodeok)
- return EINVAL;
- ASSERT(pag->pag_ici_init);
+ pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ino));
agino = XFS_INO_TO_AGINO(mp, ino);
again:
@@ -393,12 +393,12 @@ again:
read_unlock(&pag->pag_ici_lock);
XFS_STATS_INC(xs_ig_missed);
- error = xfs_iget_cache_miss(mp, pag, tp, ino, &ip, bno,
+ error = xfs_iget_cache_miss(mp, pag, tp, ino, &ip,
flags, lock_flags);
if (error)
goto out_error_or_again;
}
- xfs_put_perag(mp, pag);
+ xfs_perag_put(pag);
*ipp = ip;
@@ -417,7 +417,7 @@ out_error_or_again:
delay(1);
goto again;
}
- xfs_put_perag(mp, pag);
+ xfs_perag_put(pag);
return error;
}
@@ -488,12 +488,12 @@ xfs_ireclaim(
* added to the tree assert that it's been there before to catch
* problems with the inode life time early on.
*/
- pag = xfs_get_perag(mp, ip->i_ino);
+ pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino));
write_lock(&pag->pag_ici_lock);
if (!radix_tree_delete(&pag->pag_ici_root, agino))
ASSERT(0);
write_unlock(&pag->pag_ici_lock);
- xfs_put_perag(mp, pag);
+ xfs_perag_put(pag);
/*
* Here we do an (almost) spurious inode lock in order to coordinate
@@ -737,30 +737,24 @@ xfs_ilock_demote(
}
#ifdef DEBUG
-/*
- * Debug-only routine, without additional rw_semaphore APIs, we can
- * now only answer requests regarding whether we hold the lock for write
- * (reader state is outside our visibility, we only track writer state).
- *
- * Note: this means !xfs_isilocked would give false positives, so don't do that.
- */
int
xfs_isilocked(
xfs_inode_t *ip,
uint lock_flags)
{
- if ((lock_flags & (XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)) ==
- XFS_ILOCK_EXCL) {
- if (!ip->i_lock.mr_writer)
- return 0;
+ if (lock_flags & (XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)) {
+ if (!(lock_flags & XFS_ILOCK_SHARED))
+ return !!ip->i_lock.mr_writer;
+ return rwsem_is_locked(&ip->i_lock.mr_lock);
}
- if ((lock_flags & (XFS_IOLOCK_EXCL|XFS_IOLOCK_SHARED)) ==
- XFS_IOLOCK_EXCL) {
- if (!ip->i_iolock.mr_writer)
- return 0;
+ if (lock_flags & (XFS_IOLOCK_EXCL|XFS_IOLOCK_SHARED)) {
+ if (!(lock_flags & XFS_IOLOCK_SHARED))
+ return !!ip->i_iolock.mr_writer;
+ return rwsem_is_locked(&ip->i_iolock.mr_lock);
}
- return 1;
+ ASSERT(0);
+ return 0;
}
#endif
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index ef77fd88c8e3..b76a829d7e20 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -151,7 +151,7 @@ xfs_imap_to_bp(
"an error %d on %s. Returning error.",
error, mp->m_fsname);
} else {
- ASSERT(buf_flags & XFS_BUF_TRYLOCK);
+ ASSERT(buf_flags & XBF_TRYLOCK);
}
return error;
}
@@ -177,7 +177,7 @@ xfs_imap_to_bp(
if (unlikely(XFS_TEST_ERROR(!di_ok, mp,
XFS_ERRTAG_ITOBP_INOTOBP,
XFS_RANDOM_ITOBP_INOTOBP))) {
- if (iget_flags & XFS_IGET_BULKSTAT) {
+ if (iget_flags & XFS_IGET_UNTRUSTED) {
xfs_trans_brelse(tp, bp);
return XFS_ERROR(EINVAL);
}
@@ -239,7 +239,7 @@ xfs_inotobp(
if (error)
return error;
- error = xfs_imap_to_bp(mp, tp, &imap, &bp, XFS_BUF_LOCK, imap_flags);
+ error = xfs_imap_to_bp(mp, tp, &imap, &bp, XBF_LOCK, imap_flags);
if (error)
return error;
@@ -285,7 +285,7 @@ xfs_itobp(
return error;
if (!bp) {
- ASSERT(buf_flags & XFS_BUF_TRYLOCK);
+ ASSERT(buf_flags & XBF_TRYLOCK);
ASSERT(tp == NULL);
*bpp = NULL;
return EAGAIN;
@@ -787,7 +787,6 @@ xfs_iread(
xfs_mount_t *mp,
xfs_trans_t *tp,
xfs_inode_t *ip,
- xfs_daddr_t bno,
uint iget_flags)
{
xfs_buf_t *bp;
@@ -797,17 +796,15 @@ xfs_iread(
/*
* Fill in the location information in the in-core inode.
*/
- ip->i_imap.im_blkno = bno;
error = xfs_imap(mp, tp, ip->i_ino, &ip->i_imap, iget_flags);
if (error)
return error;
- ASSERT(bno == 0 || bno == ip->i_imap.im_blkno);
/*
* Get pointers to the on-disk inode and the buffer containing it.
*/
error = xfs_imap_to_bp(mp, tp, &ip->i_imap, &bp,
- XFS_BUF_LOCK, iget_flags);
+ XBF_LOCK, iget_flags);
if (error)
return error;
dip = (xfs_dinode_t *)xfs_buf_offset(bp, ip->i_imap.im_boffset);
@@ -1751,7 +1748,7 @@ xfs_iunlink(
* Here we put the head pointer into our next pointer,
* and then we fall through to point the head at us.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &ibp, XFS_BUF_LOCK);
+ error = xfs_itobp(mp, tp, ip, &dip, &ibp, XBF_LOCK);
if (error)
return error;
@@ -1833,7 +1830,7 @@ xfs_iunlink_remove(
* of dealing with the buffer when there is no need to
* change it.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &ibp, XFS_BUF_LOCK);
+ error = xfs_itobp(mp, tp, ip, &dip, &ibp, XBF_LOCK);
if (error) {
cmn_err(CE_WARN,
"xfs_iunlink_remove: xfs_itobp() returned an error %d on %s. Returning error.",
@@ -1895,7 +1892,7 @@ xfs_iunlink_remove(
* Now last_ibp points to the buffer previous to us on
* the unlinked list. Pull us from the list.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &ibp, XFS_BUF_LOCK);
+ error = xfs_itobp(mp, tp, ip, &dip, &ibp, XBF_LOCK);
if (error) {
cmn_err(CE_WARN,
"xfs_iunlink_remove: xfs_itobp() returned an error %d on %s. Returning error.",
@@ -1940,14 +1937,15 @@ xfs_ifree_cluster(
int blks_per_cluster;
int nbufs;
int ninodes;
- int i, j, found, pre_flushed;
+ int i, j;
xfs_daddr_t blkno;
xfs_buf_t *bp;
- xfs_inode_t *ip, **ip_found;
+ xfs_inode_t *ip;
xfs_inode_log_item_t *iip;
xfs_log_item_t *lip;
- xfs_perag_t *pag = xfs_get_perag(mp, inum);
+ struct xfs_perag *pag;
+ pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, inum));
if (mp->m_sb.sb_blocksize >= XFS_INODE_CLUSTER_SIZE(mp)) {
blks_per_cluster = 1;
ninodes = mp->m_sb.sb_inopblock;
@@ -1959,114 +1957,97 @@ xfs_ifree_cluster(
nbufs = XFS_IALLOC_BLOCKS(mp) / blks_per_cluster;
}
- ip_found = kmem_alloc(ninodes * sizeof(xfs_inode_t *), KM_NOFS);
-
for (j = 0; j < nbufs; j++, inum += ninodes) {
+ int found = 0;
+
blkno = XFS_AGB_TO_DADDR(mp, XFS_INO_TO_AGNO(mp, inum),
XFS_INO_TO_AGBNO(mp, inum));
+ /*
+ * We obtain and lock the backing buffer first in the process
+ * here, as we have to ensure that any dirty inode that we
+ * can't get the flush lock on is attached to the buffer.
+ * If we scan the in-memory inodes first, then buffer IO can
+ * complete before we get a lock on it, and hence we may fail
+ * to mark all the active inodes on the buffer stale.
+ */
+ bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, blkno,
+ mp->m_bsize * blks_per_cluster,
+ XBF_LOCK);
/*
- * Look for each inode in memory and attempt to lock it,
- * we can be racing with flush and tail pushing here.
- * any inode we get the locks on, add to an array of
- * inode items to process later.
+ * Walk the inodes already attached to the buffer and mark them
+ * stale. These will all have the flush locks held, so an
+ * in-memory inode walk can't lock them.
+ */
+ lip = XFS_BUF_FSPRIVATE(bp, xfs_log_item_t *);
+ while (lip) {
+ if (lip->li_type == XFS_LI_INODE) {
+ iip = (xfs_inode_log_item_t *)lip;
+ ASSERT(iip->ili_logged == 1);
+ lip->li_cb = (void(*)(xfs_buf_t*,xfs_log_item_t*)) xfs_istale_done;
+ xfs_trans_ail_copy_lsn(mp->m_ail,
+ &iip->ili_flush_lsn,
+ &iip->ili_item.li_lsn);
+ xfs_iflags_set(iip->ili_inode, XFS_ISTALE);
+ found++;
+ }
+ lip = lip->li_bio_list;
+ }
+
+ /*
+ * For each inode in memory attempt to add it to the inode
+ * buffer and set it up for being staled on buffer IO
+ * completion. This is safe as we've locked out tail pushing
+ * and flushing by locking the buffer.
*
- * The get the buffer lock, we could beat a flush
- * or tail pushing thread to the lock here, in which
- * case they will go looking for the inode buffer
- * and fail, we need some other form of interlock
- * here.
+ * We have already marked every inode that was part of a
+ * transaction stale above, which means there is no point in
+ * even trying to lock them.
*/
- found = 0;
for (i = 0; i < ninodes; i++) {
read_lock(&pag->pag_ici_lock);
ip = radix_tree_lookup(&pag->pag_ici_root,
XFS_INO_TO_AGINO(mp, (inum + i)));
- /* Inode not in memory or we found it already,
- * nothing to do
- */
+ /* Inode not in memory or stale, nothing to do */
if (!ip || xfs_iflags_test(ip, XFS_ISTALE)) {
read_unlock(&pag->pag_ici_lock);
continue;
}
- if (xfs_inode_clean(ip)) {
- read_unlock(&pag->pag_ici_lock);
- continue;
- }
-
- /* If we can get the locks then add it to the
- * list, otherwise by the time we get the bp lock
- * below it will already be attached to the
- * inode buffer.
- */
-
- /* This inode will already be locked - by us, lets
- * keep it that way.
- */
-
- if (ip == free_ip) {
- if (xfs_iflock_nowait(ip)) {
- xfs_iflags_set(ip, XFS_ISTALE);
- if (xfs_inode_clean(ip)) {
- xfs_ifunlock(ip);
- } else {
- ip_found[found++] = ip;
- }
- }
+ /* don't try to lock/unlock the current inode */
+ if (ip != free_ip &&
+ !xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) {
read_unlock(&pag->pag_ici_lock);
continue;
}
+ read_unlock(&pag->pag_ici_lock);
- if (xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) {
- if (xfs_iflock_nowait(ip)) {
- xfs_iflags_set(ip, XFS_ISTALE);
-
- if (xfs_inode_clean(ip)) {
- xfs_ifunlock(ip);
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- } else {
- ip_found[found++] = ip;
- }
- } else {
+ if (!xfs_iflock_nowait(ip)) {
+ if (ip != free_ip)
xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
+ continue;
}
- read_unlock(&pag->pag_ici_lock);
- }
- bp = xfs_trans_get_buf(tp, mp->m_ddev_targp, blkno,
- mp->m_bsize * blks_per_cluster,
- XFS_BUF_LOCK);
-
- pre_flushed = 0;
- lip = XFS_BUF_FSPRIVATE(bp, xfs_log_item_t *);
- while (lip) {
- if (lip->li_type == XFS_LI_INODE) {
- iip = (xfs_inode_log_item_t *)lip;
- ASSERT(iip->ili_logged == 1);
- lip->li_cb = (void(*)(xfs_buf_t*,xfs_log_item_t*)) xfs_istale_done;
- xfs_trans_ail_copy_lsn(mp->m_ail,
- &iip->ili_flush_lsn,
- &iip->ili_item.li_lsn);
- xfs_iflags_set(iip->ili_inode, XFS_ISTALE);
- pre_flushed++;
+ xfs_iflags_set(ip, XFS_ISTALE);
+ if (xfs_inode_clean(ip)) {
+ ASSERT(ip != free_ip);
+ xfs_ifunlock(ip);
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ continue;
}
- lip = lip->li_bio_list;
- }
- for (i = 0; i < found; i++) {
- ip = ip_found[i];
iip = ip->i_itemp;
-
if (!iip) {
+ /* inode with unlogged changes only */
+ ASSERT(ip != free_ip);
ip->i_update_core = 0;
xfs_ifunlock(ip);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
continue;
}
+ found++;
iip->ili_last_fields = iip->ili_format.ilf_fields;
iip->ili_format.ilf_fields = 0;
@@ -2077,18 +2058,17 @@ xfs_ifree_cluster(
xfs_buf_attach_iodone(bp,
(void(*)(xfs_buf_t*,xfs_log_item_t*))
xfs_istale_done, (xfs_log_item_t *)iip);
- if (ip != free_ip) {
+
+ if (ip != free_ip)
xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
}
- if (found || pre_flushed)
+ if (found)
xfs_trans_stale_inode_buf(tp, bp);
xfs_trans_binval(tp, bp);
}
- kmem_free(ip_found);
- xfs_put_perag(mp, pag);
+ xfs_perag_put(pag);
}
/*
@@ -2150,7 +2130,7 @@ xfs_ifree(
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
- error = xfs_itobp(ip->i_mount, tp, ip, &dip, &ibp, XFS_BUF_LOCK);
+ error = xfs_itobp(ip->i_mount, tp, ip, &dip, &ibp, XBF_LOCK);
if (error)
return error;
@@ -2438,72 +2418,33 @@ xfs_idestroy_fork(
}
/*
- * Increment the pin count of the given buffer.
- * This value is protected by ipinlock spinlock in the mount structure.
+ * This is called to unpin an inode. The caller must have the inode locked
+ * in at least shared mode so that the buffer cannot be subsequently pinned
+ * once someone is waiting for it to be unpinned.
*/
-void
-xfs_ipin(
- xfs_inode_t *ip)
-{
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
-
- atomic_inc(&ip->i_pincount);
-}
-
-/*
- * Decrement the pin count of the given inode, and wake up
- * anyone in xfs_iwait_unpin() if the count goes to 0. The
- * inode must have been previously pinned with a call to xfs_ipin().
- */
-void
-xfs_iunpin(
- xfs_inode_t *ip)
-{
- ASSERT(atomic_read(&ip->i_pincount) > 0);
-
- if (atomic_dec_and_test(&ip->i_pincount))
- wake_up(&ip->i_ipin_wait);
-}
-
-/*
- * This is called to unpin an inode. It can be directed to wait or to return
- * immediately without waiting for the inode to be unpinned. The caller must
- * have the inode locked in at least shared mode so that the buffer cannot be
- * subsequently pinned once someone is waiting for it to be unpinned.
- */
-STATIC void
-__xfs_iunpin_wait(
- xfs_inode_t *ip,
- int wait)
+static void
+xfs_iunpin_nowait(
+ struct xfs_inode *ip)
{
- xfs_inode_log_item_t *iip = ip->i_itemp;
-
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
- if (atomic_read(&ip->i_pincount) == 0)
- return;
+
+ trace_xfs_inode_unpin_nowait(ip, _RET_IP_);
/* Give the log a push to start the unpinning I/O */
- xfs_log_force(ip->i_mount, (iip && iip->ili_last_lsn) ?
- iip->ili_last_lsn : 0, XFS_LOG_FORCE);
- if (wait)
- wait_event(ip->i_ipin_wait, (atomic_read(&ip->i_pincount) == 0));
-}
+ xfs_log_force_lsn(ip->i_mount, ip->i_itemp->ili_last_lsn, 0);
-static inline void
-xfs_iunpin_wait(
- xfs_inode_t *ip)
-{
- __xfs_iunpin_wait(ip, 1);
}
-static inline void
-xfs_iunpin_nowait(
- xfs_inode_t *ip)
+void
+xfs_iunpin_wait(
+ struct xfs_inode *ip)
{
- __xfs_iunpin_wait(ip, 0);
+ if (xfs_ipincount(ip)) {
+ xfs_iunpin_nowait(ip);
+ wait_event(ip->i_ipin_wait, (xfs_ipincount(ip) == 0));
+ }
}
-
/*
* xfs_iextents_copy()
*
@@ -2675,7 +2616,7 @@ xfs_iflush_cluster(
xfs_buf_t *bp)
{
xfs_mount_t *mp = ip->i_mount;
- xfs_perag_t *pag = xfs_get_perag(mp, ip->i_ino);
+ struct xfs_perag *pag;
unsigned long first_index, mask;
unsigned long inodes_per_cluster;
int ilist_size;
@@ -2686,14 +2627,13 @@ xfs_iflush_cluster(
int bufwasdelwri;
int i;
- ASSERT(pag->pagi_inodeok);
- ASSERT(pag->pag_ici_init);
+ pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino));
inodes_per_cluster = XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog;
ilist_size = inodes_per_cluster * sizeof(xfs_inode_t *);
ilist = kmem_alloc(ilist_size, KM_MAYFAIL|KM_NOFS);
if (!ilist)
- return 0;
+ goto out_put;
mask = ~(((XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog)) - 1);
first_index = XFS_INO_TO_AGINO(mp, ip->i_ino) & mask;
@@ -2762,6 +2702,8 @@ xfs_iflush_cluster(
out_free:
read_unlock(&pag->pag_ici_lock);
kmem_free(ilist);
+out_put:
+ xfs_perag_put(pag);
return 0;
@@ -2805,6 +2747,7 @@ cluster_corrupt_out:
*/
xfs_iflush_abort(iq);
kmem_free(ilist);
+ xfs_perag_put(pag);
return XFS_ERROR(EFSCORRUPTED);
}
@@ -2827,8 +2770,6 @@ xfs_iflush(
xfs_dinode_t *dip;
xfs_mount_t *mp;
int error;
- int noblock = (flags == XFS_IFLUSH_ASYNC_NOBLOCK);
- enum { INT_DELWRI = (1 << 0), INT_ASYNC = (1 << 1) };
XFS_STATS_INC(xs_iflush_count);
@@ -2841,15 +2782,6 @@ xfs_iflush(
mp = ip->i_mount;
/*
- * If the inode isn't dirty, then just release the inode flush lock and
- * do nothing.
- */
- if (xfs_inode_clean(ip)) {
- xfs_ifunlock(ip);
- return 0;
- }
-
- /*
* We can't flush the inode until it is unpinned, so wait for it if we
* are allowed to block. We know noone new can pin it, because we are
* holding the inode lock shared and you need to hold it exclusively to
@@ -2860,7 +2792,7 @@ xfs_iflush(
* in the same cluster are dirty, they will probably write the inode
* out for us if they occur after the log force completes.
*/
- if (noblock && xfs_ipincount(ip)) {
+ if (!(flags & SYNC_WAIT) && xfs_ipincount(ip)) {
xfs_iunpin_nowait(ip);
xfs_ifunlock(ip);
return EAGAIN;
@@ -2894,60 +2826,10 @@ xfs_iflush(
}
/*
- * Decide how buffer will be flushed out. This is done before
- * the call to xfs_iflush_int because this field is zeroed by it.
- */
- if (iip != NULL && iip->ili_format.ilf_fields != 0) {
- /*
- * Flush out the inode buffer according to the directions
- * of the caller. In the cases where the caller has given
- * us a choice choose the non-delwri case. This is because
- * the inode is in the AIL and we need to get it out soon.
- */
- switch (flags) {
- case XFS_IFLUSH_SYNC:
- case XFS_IFLUSH_DELWRI_ELSE_SYNC:
- flags = 0;
- break;
- case XFS_IFLUSH_ASYNC_NOBLOCK:
- case XFS_IFLUSH_ASYNC:
- case XFS_IFLUSH_DELWRI_ELSE_ASYNC:
- flags = INT_ASYNC;
- break;
- case XFS_IFLUSH_DELWRI:
- flags = INT_DELWRI;
- break;
- default:
- ASSERT(0);
- flags = 0;
- break;
- }
- } else {
- switch (flags) {
- case XFS_IFLUSH_DELWRI_ELSE_SYNC:
- case XFS_IFLUSH_DELWRI_ELSE_ASYNC:
- case XFS_IFLUSH_DELWRI:
- flags = INT_DELWRI;
- break;
- case XFS_IFLUSH_ASYNC_NOBLOCK:
- case XFS_IFLUSH_ASYNC:
- flags = INT_ASYNC;
- break;
- case XFS_IFLUSH_SYNC:
- flags = 0;
- break;
- default:
- ASSERT(0);
- flags = 0;
- break;
- }
- }
-
- /*
* Get the buffer containing the on-disk inode.
*/
error = xfs_itobp(mp, NULL, ip, &dip, &bp,
- noblock ? XFS_BUF_TRYLOCK : XFS_BUF_LOCK);
+ (flags & SYNC_WAIT) ? XBF_LOCK : XBF_TRYLOCK);
if (error || !bp) {
xfs_ifunlock(ip);
return error;
@@ -2965,7 +2847,7 @@ xfs_iflush(
* get stuck waiting in the write for too long.
*/
if (XFS_BUF_ISPINNED(bp))
- xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE);
+ xfs_log_force(mp, 0);
/*
* inode clustering:
@@ -2975,13 +2857,10 @@ xfs_iflush(
if (error)
goto cluster_corrupt_out;
- if (flags & INT_DELWRI) {
- xfs_bdwrite(mp, bp);
- } else if (flags & INT_ASYNC) {
- error = xfs_bawrite(mp, bp);
- } else {
+ if (flags & SYNC_WAIT)
error = xfs_bwrite(mp, bp);
- }
+ else
+ xfs_bdwrite(mp, bp);
return error;
corrupt_out:
@@ -3016,16 +2895,6 @@ xfs_iflush_int(
iip = ip->i_itemp;
mp = ip->i_mount;
-
- /*
- * If the inode isn't dirty, then just release the inode
- * flush lock and do nothing.
- */
- if (xfs_inode_clean(ip)) {
- xfs_ifunlock(ip);
- return 0;
- }
-
/* set *dip = inode's place in the buffer */
dip = (xfs_dinode_t *)xfs_buf_offset(bp, ip->i_imap.im_boffset);
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index ec1f28c4fc4f..78550df13cd6 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -420,16 +420,6 @@ static inline void xfs_ifunlock(xfs_inode_t *ip)
#define XFS_ILOCK_DEP(flags) (((flags) & XFS_ILOCK_DEP_MASK) >> XFS_ILOCK_SHIFT)
/*
- * Flags for xfs_iflush()
- */
-#define XFS_IFLUSH_DELWRI_ELSE_SYNC 1
-#define XFS_IFLUSH_DELWRI_ELSE_ASYNC 2
-#define XFS_IFLUSH_SYNC 3
-#define XFS_IFLUSH_ASYNC 4
-#define XFS_IFLUSH_DELWRI 5
-#define XFS_IFLUSH_ASYNC_NOBLOCK 6
-
-/*
* Flags for xfs_itruncate_start().
*/
#define XFS_ITRUNC_DEFINITE 0x1
@@ -452,7 +442,7 @@ static inline void xfs_ifunlock(xfs_inode_t *ip)
* xfs_iget.c prototypes.
*/
int xfs_iget(struct xfs_mount *, struct xfs_trans *, xfs_ino_t,
- uint, uint, xfs_inode_t **, xfs_daddr_t);
+ uint, uint, xfs_inode_t **);
void xfs_iput(xfs_inode_t *, uint);
void xfs_iput_new(xfs_inode_t *, uint);
void xfs_ilock(xfs_inode_t *, uint);
@@ -481,14 +471,14 @@ int xfs_itruncate_finish(struct xfs_trans **, xfs_inode_t *,
int xfs_iunlink(struct xfs_trans *, xfs_inode_t *);
void xfs_iext_realloc(xfs_inode_t *, int, int);
-void xfs_ipin(xfs_inode_t *);
-void xfs_iunpin(xfs_inode_t *);
+void xfs_iunpin_wait(xfs_inode_t *);
int xfs_iflush(xfs_inode_t *, uint);
void xfs_ichgtime(xfs_inode_t *, int);
void xfs_lock_inodes(xfs_inode_t **, int, uint);
void xfs_lock_two_inodes(xfs_inode_t *, xfs_inode_t *, uint);
void xfs_synchronize_times(xfs_inode_t *);
+void xfs_mark_inode_dirty(xfs_inode_t *);
void xfs_mark_inode_dirty_sync(xfs_inode_t *);
#define IHOLD(ip) \
@@ -510,7 +500,7 @@ do { \
* Flags for xfs_iget()
*/
#define XFS_IGET_CREATE 0x1
-#define XFS_IGET_BULKSTAT 0x2
+#define XFS_IGET_UNTRUSTED 0x2
int xfs_inotobp(struct xfs_mount *, struct xfs_trans *,
xfs_ino_t, struct xfs_dinode **,
@@ -519,7 +509,7 @@ int xfs_itobp(struct xfs_mount *, struct xfs_trans *,
struct xfs_inode *, struct xfs_dinode **,
struct xfs_buf **, uint);
int xfs_iread(struct xfs_mount *, struct xfs_trans *,
- struct xfs_inode *, xfs_daddr_t, uint);
+ struct xfs_inode *, uint);
void xfs_dinode_to_disk(struct xfs_dinode *,
struct xfs_icdinode *);
void xfs_idestroy_fork(struct xfs_inode *, int);
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index f38855d21ea5..cf8249a60004 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -228,7 +228,7 @@ xfs_inode_item_format(
vecp->i_addr = (xfs_caddr_t)&iip->ili_format;
vecp->i_len = sizeof(xfs_inode_log_format_t);
- XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_IFORMAT);
+ vecp->i_type = XLOG_REG_TYPE_IFORMAT;
vecp++;
nvecs = 1;
@@ -279,7 +279,7 @@ xfs_inode_item_format(
vecp->i_addr = (xfs_caddr_t)&ip->i_d;
vecp->i_len = sizeof(struct xfs_icdinode);
- XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_ICORE);
+ vecp->i_type = XLOG_REG_TYPE_ICORE;
vecp++;
nvecs++;
iip->ili_format.ilf_fields |= XFS_ILOG_CORE;
@@ -336,7 +336,7 @@ xfs_inode_item_format(
vecp->i_addr =
(char *)(ip->i_df.if_u1.if_extents);
vecp->i_len = ip->i_df.if_bytes;
- XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_IEXT);
+ vecp->i_type = XLOG_REG_TYPE_IEXT;
} else
#endif
{
@@ -355,7 +355,7 @@ xfs_inode_item_format(
vecp->i_addr = (xfs_caddr_t)ext_buffer;
vecp->i_len = xfs_iextents_copy(ip, ext_buffer,
XFS_DATA_FORK);
- XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_IEXT);
+ vecp->i_type = XLOG_REG_TYPE_IEXT;
}
ASSERT(vecp->i_len <= ip->i_df.if_bytes);
iip->ili_format.ilf_dsize = vecp->i_len;
@@ -373,7 +373,7 @@ xfs_inode_item_format(
ASSERT(ip->i_df.if_broot != NULL);
vecp->i_addr = (xfs_caddr_t)ip->i_df.if_broot;
vecp->i_len = ip->i_df.if_broot_bytes;
- XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_IBROOT);
+ vecp->i_type = XLOG_REG_TYPE_IBROOT;
vecp++;
nvecs++;
iip->ili_format.ilf_dsize = ip->i_df.if_broot_bytes;
@@ -399,7 +399,7 @@ xfs_inode_item_format(
ASSERT((ip->i_df.if_real_bytes == 0) ||
(ip->i_df.if_real_bytes == data_bytes));
vecp->i_len = (int)data_bytes;
- XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_ILOCAL);
+ vecp->i_type = XLOG_REG_TYPE_ILOCAL;
vecp++;
nvecs++;
iip->ili_format.ilf_dsize = (unsigned)data_bytes;
@@ -477,7 +477,7 @@ xfs_inode_item_format(
vecp->i_len = xfs_iextents_copy(ip, ext_buffer,
XFS_ATTR_FORK);
#endif
- XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_IATTR_EXT);
+ vecp->i_type = XLOG_REG_TYPE_IATTR_EXT;
iip->ili_format.ilf_asize = vecp->i_len;
vecp++;
nvecs++;
@@ -492,7 +492,7 @@ xfs_inode_item_format(
ASSERT(ip->i_afp->if_broot != NULL);
vecp->i_addr = (xfs_caddr_t)ip->i_afp->if_broot;
vecp->i_len = ip->i_afp->if_broot_bytes;
- XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_IATTR_BROOT);
+ vecp->i_type = XLOG_REG_TYPE_IATTR_BROOT;
vecp++;
nvecs++;
iip->ili_format.ilf_asize = ip->i_afp->if_broot_bytes;
@@ -516,7 +516,7 @@ xfs_inode_item_format(
ASSERT((ip->i_afp->if_real_bytes == 0) ||
(ip->i_afp->if_real_bytes == data_bytes));
vecp->i_len = (int)data_bytes;
- XLOG_VEC_SET_TYPE(vecp, XLOG_REG_TYPE_IATTR_LOCAL);
+ vecp->i_type = XLOG_REG_TYPE_IATTR_LOCAL;
vecp++;
nvecs++;
iip->ili_format.ilf_asize = (unsigned)data_bytes;
@@ -535,31 +535,36 @@ xfs_inode_item_format(
/*
* This is called to pin the inode associated with the inode log
- * item in memory so it cannot be written out. Do this by calling
- * xfs_ipin() to bump the pin count in the inode while holding the
- * inode pin lock.
+ * item in memory so it cannot be written out.
*/
STATIC void
xfs_inode_item_pin(
xfs_inode_log_item_t *iip)
{
ASSERT(xfs_isilocked(iip->ili_inode, XFS_ILOCK_EXCL));
- xfs_ipin(iip->ili_inode);
+
+ trace_xfs_inode_pin(iip->ili_inode, _RET_IP_);
+ atomic_inc(&iip->ili_inode->i_pincount);
}
/*
* This is called to unpin the inode associated with the inode log
* item which was previously pinned with a call to xfs_inode_item_pin().
- * Just call xfs_iunpin() on the inode to do this.
+ *
+ * Also wake up anyone in xfs_iunpin_wait() if the count goes to 0.
*/
/* ARGSUSED */
STATIC void
xfs_inode_item_unpin(
- xfs_inode_log_item_t *iip,
- int stale)
+ xfs_inode_log_item_t *iip)
{
- xfs_iunpin(iip->ili_inode);
+ struct xfs_inode *ip = iip->ili_inode;
+
+ trace_xfs_inode_unpin(ip, _RET_IP_);
+ ASSERT(atomic_read(&ip->i_pincount) > 0);
+ if (atomic_dec_and_test(&ip->i_pincount))
+ wake_up(&ip->i_ipin_wait);
}
/* ARGSUSED */
@@ -568,7 +573,7 @@ xfs_inode_item_unpin_remove(
xfs_inode_log_item_t *iip,
xfs_trans_t *tp)
{
- xfs_iunpin(iip->ili_inode);
+ xfs_inode_item_unpin(iip);
}
/*
@@ -602,33 +607,20 @@ xfs_inode_item_trylock(
if (!xfs_iflock_nowait(ip)) {
/*
- * If someone else isn't already trying to push the inode
- * buffer, we get to do it.
+ * inode has already been flushed to the backing buffer,
+ * leave it locked in shared mode, pushbuf routine will
+ * unlock it.
*/
- if (iip->ili_pushbuf_flag == 0) {
- iip->ili_pushbuf_flag = 1;
-#ifdef DEBUG
- iip->ili_push_owner = current_pid();
-#endif
- /*
- * Inode is left locked in shared mode.
- * Pushbuf routine gets to unlock it.
- */
- return XFS_ITEM_PUSHBUF;
- } else {
- /*
- * We hold the AIL lock, so we must specify the
- * NONOTIFY flag so that we won't double trip.
- */
- xfs_iunlock(ip, XFS_ILOCK_SHARED|XFS_IUNLOCK_NONOTIFY);
- return XFS_ITEM_FLUSHING;
- }
- /* NOTREACHED */
+ return XFS_ITEM_PUSHBUF;
}
/* Stale items should force out the iclog */
if (ip->i_flags & XFS_ISTALE) {
xfs_ifunlock(ip);
+ /*
+ * we hold the AIL lock - notify the unlock routine of this
+ * so it doesn't try to get the lock again.
+ */
xfs_iunlock(ip, XFS_ILOCK_SHARED|XFS_IUNLOCK_NONOTIFY);
return XFS_ITEM_PINNED;
}
@@ -746,11 +738,8 @@ xfs_inode_item_committed(
* This gets called by xfs_trans_push_ail(), when IOP_TRYLOCK
* failed to get the inode flush lock but did get the inode locked SHARED.
* Here we're trying to see if the inode buffer is incore, and if so whether it's
- * marked delayed write. If that's the case, we'll initiate a bawrite on that
- * buffer to expedite the process.
- *
- * We aren't holding the AIL lock (or the flush lock) when this gets called,
- * so it is inherently race-y.
+ * marked delayed write. If that's the case, we'll promote it and that will
+ * allow the caller to write the buffer by triggering the xfsbufd to run.
*/
STATIC void
xfs_inode_item_pushbuf(
@@ -759,82 +748,30 @@ xfs_inode_item_pushbuf(
xfs_inode_t *ip;
xfs_mount_t *mp;
xfs_buf_t *bp;
- uint dopush;
ip = iip->ili_inode;
-
ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED));
/*
- * The ili_pushbuf_flag keeps others from
- * trying to duplicate our effort.
- */
- ASSERT(iip->ili_pushbuf_flag != 0);
- ASSERT(iip->ili_push_owner == current_pid());
-
- /*
* If a flush is not in progress anymore, chances are that the
* inode was taken off the AIL. So, just get out.
*/
if (completion_done(&ip->i_flush) ||
((iip->ili_item.li_flags & XFS_LI_IN_AIL) == 0)) {
- iip->ili_pushbuf_flag = 0;
xfs_iunlock(ip, XFS_ILOCK_SHARED);
return;
}
mp = ip->i_mount;
bp = xfs_incore(mp->m_ddev_targp, iip->ili_format.ilf_blkno,
- iip->ili_format.ilf_len, XFS_INCORE_TRYLOCK);
-
- if (bp != NULL) {
- if (XFS_BUF_ISDELAYWRITE(bp)) {
- /*
- * We were racing with iflush because we don't hold
- * the AIL lock or the flush lock. However, at this point,
- * we have the buffer, and we know that it's dirty.
- * So, it's possible that iflush raced with us, and
- * this item is already taken off the AIL.
- * If not, we can flush it async.
- */
- dopush = ((iip->ili_item.li_flags & XFS_LI_IN_AIL) &&
- !completion_done(&ip->i_flush));
- iip->ili_pushbuf_flag = 0;
- xfs_iunlock(ip, XFS_ILOCK_SHARED);
-
- trace_xfs_inode_item_push(bp, _RET_IP_);
+ iip->ili_format.ilf_len, XBF_TRYLOCK);
- if (XFS_BUF_ISPINNED(bp)) {
- xfs_log_force(mp, (xfs_lsn_t)0,
- XFS_LOG_FORCE);
- }
- if (dopush) {
- int error;
- error = xfs_bawrite(mp, bp);
- if (error)
- xfs_fs_cmn_err(CE_WARN, mp,
- "xfs_inode_item_pushbuf: pushbuf error %d on iip %p, bp %p",
- error, iip, bp);
- } else {
- xfs_buf_relse(bp);
- }
- } else {
- iip->ili_pushbuf_flag = 0;
- xfs_iunlock(ip, XFS_ILOCK_SHARED);
- xfs_buf_relse(bp);
- }
- return;
- }
- /*
- * We have to be careful about resetting pushbuf flag too early (above).
- * Even though in theory we can do it as soon as we have the buflock,
- * we don't want others to be doing work needlessly. They'll come to
- * this function thinking that pushing the buffer is their
- * responsibility only to find that the buffer is still locked by
- * another doing the same thing
- */
- iip->ili_pushbuf_flag = 0;
xfs_iunlock(ip, XFS_ILOCK_SHARED);
+ if (!bp)
+ return;
+ if (XFS_BUF_ISDELAYWRITE(bp))
+ xfs_buf_delwri_promote(bp);
+ xfs_buf_relse(bp);
return;
}
@@ -867,10 +804,14 @@ xfs_inode_item_push(
iip->ili_format.ilf_fields != 0);
/*
- * Write out the inode. The completion routine ('iflush_done') will
- * pull it from the AIL, mark it clean, unlock the flush lock.
+ * Push the inode to it's backing buffer. This will not remove the
+ * inode from the AIL - a further push will be required to trigger a
+ * buffer push. However, this allows all the dirty inodes to be pushed
+ * to the buffer before it is pushed to disk. THe buffer IO completion
+ * will pull th einode from the AIL, mark it clean and unlock the flush
+ * lock.
*/
- (void) xfs_iflush(ip, XFS_IFLUSH_ASYNC);
+ (void) xfs_iflush(ip, 0);
xfs_iunlock(ip, XFS_ILOCK_SHARED);
return;
@@ -898,7 +839,7 @@ static struct xfs_item_ops xfs_inode_item_ops = {
.iop_format = (void(*)(xfs_log_item_t*, xfs_log_iovec_t*))
xfs_inode_item_format,
.iop_pin = (void(*)(xfs_log_item_t*))xfs_inode_item_pin,
- .iop_unpin = (void(*)(xfs_log_item_t*, int))xfs_inode_item_unpin,
+ .iop_unpin = (void(*)(xfs_log_item_t*))xfs_inode_item_unpin,
.iop_unpin_remove = (void(*)(xfs_log_item_t*, xfs_trans_t*))
xfs_inode_item_unpin_remove,
.iop_trylock = (uint(*)(xfs_log_item_t*))xfs_inode_item_trylock,
@@ -925,18 +866,9 @@ xfs_inode_item_init(
ASSERT(ip->i_itemp == NULL);
iip = ip->i_itemp = kmem_zone_zalloc(xfs_ili_zone, KM_SLEEP);
- iip->ili_item.li_type = XFS_LI_INODE;
- iip->ili_item.li_ops = &xfs_inode_item_ops;
- iip->ili_item.li_mountp = mp;
- iip->ili_item.li_ailp = mp->m_ail;
iip->ili_inode = ip;
-
- /*
- We have zeroed memory. No need ...
- iip->ili_extents_buf = NULL;
- iip->ili_pushbuf_flag = 0;
- */
-
+ xfs_log_item_init(mp, &iip->ili_item, XFS_LI_INODE,
+ &xfs_inode_item_ops);
iip->ili_format.ilf_type = XFS_LI_INODE;
iip->ili_format.ilf_ino = ip->i_ino;
iip->ili_format.ilf_blkno = ip->i_imap.im_blkno;
diff --git a/fs/xfs/xfs_inode_item.h b/fs/xfs/xfs_inode_item.h
index cc8df1ac7783..9a467958ecdd 100644
--- a/fs/xfs/xfs_inode_item.h
+++ b/fs/xfs/xfs_inode_item.h
@@ -144,12 +144,6 @@ typedef struct xfs_inode_log_item {
data exts */
struct xfs_bmbt_rec *ili_aextents_buf; /* array of logged
attr exts */
- unsigned int ili_pushbuf_flag; /* one bit used in push_ail */
-
-#ifdef DEBUG
- uint64_t ili_push_owner; /* one who sets pushbuf_flag
- above gets to push the buf */
-#endif
#ifdef XFS_TRANS_DEBUG
int ili_root_size;
char *ili_orig_root;
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 0b65039951a0..ef14943829da 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -55,71 +55,33 @@
#define XFS_STRAT_WRITE_IMAPS 2
#define XFS_WRITE_IMAPS XFS_BMAP_MAX_NMAP
-STATIC int
-xfs_imap_to_bmap(
- xfs_inode_t *ip,
- xfs_off_t offset,
- xfs_bmbt_irec_t *imap,
- xfs_iomap_t *iomapp,
- int imaps, /* Number of imap entries */
- int iomaps, /* Number of iomap entries */
- int flags)
-{
- xfs_mount_t *mp = ip->i_mount;
- int pbm;
- xfs_fsblock_t start_block;
-
-
- for (pbm = 0; imaps && pbm < iomaps; imaps--, iomapp++, imap++, pbm++) {
- iomapp->iomap_offset = XFS_FSB_TO_B(mp, imap->br_startoff);
- iomapp->iomap_delta = offset - iomapp->iomap_offset;
- iomapp->iomap_bsize = XFS_FSB_TO_B(mp, imap->br_blockcount);
- iomapp->iomap_flags = flags;
-
- if (XFS_IS_REALTIME_INODE(ip)) {
- iomapp->iomap_flags |= IOMAP_REALTIME;
- iomapp->iomap_target = mp->m_rtdev_targp;
- } else {
- iomapp->iomap_target = mp->m_ddev_targp;
- }
- start_block = imap->br_startblock;
- if (start_block == HOLESTARTBLOCK) {
- iomapp->iomap_bn = IOMAP_DADDR_NULL;
- iomapp->iomap_flags |= IOMAP_HOLE;
- } else if (start_block == DELAYSTARTBLOCK) {
- iomapp->iomap_bn = IOMAP_DADDR_NULL;
- iomapp->iomap_flags |= IOMAP_DELAY;
- } else {
- iomapp->iomap_bn = xfs_fsb_to_db(ip, start_block);
- if (ISUNWRITTEN(imap))
- iomapp->iomap_flags |= IOMAP_UNWRITTEN;
- }
-
- offset += iomapp->iomap_bsize - iomapp->iomap_delta;
- }
- return pbm; /* Return the number filled */
-}
+STATIC int xfs_iomap_write_direct(struct xfs_inode *, xfs_off_t, size_t,
+ int, struct xfs_bmbt_irec *, int *);
+STATIC int xfs_iomap_write_delay(struct xfs_inode *, xfs_off_t, size_t, int,
+ struct xfs_bmbt_irec *, int *);
+STATIC int xfs_iomap_write_allocate(struct xfs_inode *, xfs_off_t, size_t,
+ struct xfs_bmbt_irec *, int *);
int
xfs_iomap(
- xfs_inode_t *ip,
- xfs_off_t offset,
- ssize_t count,
- int flags,
- xfs_iomap_t *iomapp,
- int *niomaps)
+ struct xfs_inode *ip,
+ xfs_off_t offset,
+ ssize_t count,
+ int flags,
+ struct xfs_bmbt_irec *imap,
+ int *nimaps,
+ int *new)
{
- xfs_mount_t *mp = ip->i_mount;
- xfs_fileoff_t offset_fsb, end_fsb;
- int error = 0;
- int lockmode = 0;
- xfs_bmbt_irec_t imap;
- int nimaps = 1;
- int bmapi_flags = 0;
- int iomap_flags = 0;
+ struct xfs_mount *mp = ip->i_mount;
+ xfs_fileoff_t offset_fsb, end_fsb;
+ int error = 0;
+ int lockmode = 0;
+ int bmapi_flags = 0;
ASSERT((ip->i_d.di_mode & S_IFMT) == S_IFREG);
+ *new = 0;
+
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
@@ -160,8 +122,8 @@ xfs_iomap(
error = xfs_bmapi(NULL, ip, offset_fsb,
(xfs_filblks_t)(end_fsb - offset_fsb),
- bmapi_flags, NULL, 0, &imap,
- &nimaps, NULL, NULL);
+ bmapi_flags, NULL, 0, imap,
+ nimaps, NULL, NULL);
if (error)
goto out;
@@ -169,46 +131,41 @@ xfs_iomap(
switch (flags & (BMAPI_WRITE|BMAPI_ALLOCATE)) {
case BMAPI_WRITE:
/* If we found an extent, return it */
- if (nimaps &&
- (imap.br_startblock != HOLESTARTBLOCK) &&
- (imap.br_startblock != DELAYSTARTBLOCK)) {
- trace_xfs_iomap_found(ip, offset, count, flags, &imap);
+ if (*nimaps &&
+ (imap->br_startblock != HOLESTARTBLOCK) &&
+ (imap->br_startblock != DELAYSTARTBLOCK)) {
+ trace_xfs_iomap_found(ip, offset, count, flags, imap);
break;
}
if (flags & (BMAPI_DIRECT|BMAPI_MMAP)) {
error = xfs_iomap_write_direct(ip, offset, count, flags,
- &imap, &nimaps, nimaps);
+ imap, nimaps);
} else {
error = xfs_iomap_write_delay(ip, offset, count, flags,
- &imap, &nimaps);
+ imap, nimaps);
}
if (!error) {
- trace_xfs_iomap_alloc(ip, offset, count, flags, &imap);
+ trace_xfs_iomap_alloc(ip, offset, count, flags, imap);
}
- iomap_flags = IOMAP_NEW;
+ *new = 1;
break;
case BMAPI_ALLOCATE:
/* If we found an extent, return it */
xfs_iunlock(ip, lockmode);
lockmode = 0;
- if (nimaps && !isnullstartblock(imap.br_startblock)) {
- trace_xfs_iomap_found(ip, offset, count, flags, &imap);
+ if (*nimaps && !isnullstartblock(imap->br_startblock)) {
+ trace_xfs_iomap_found(ip, offset, count, flags, imap);
break;
}
error = xfs_iomap_write_allocate(ip, offset, count,
- &imap, &nimaps);
+ imap, nimaps);
break;
}
- if (nimaps) {
- *niomaps = xfs_imap_to_bmap(ip, offset, &imap,
- iomapp, nimaps, *niomaps, iomap_flags);
- } else if (niomaps) {
- *niomaps = 0;
- }
+ ASSERT(*nimaps <= 1);
out:
if (lockmode)
@@ -216,7 +173,6 @@ out:
return XFS_ERROR(error);
}
-
STATIC int
xfs_iomap_eof_align_last_fsb(
xfs_mount_t *mp,
@@ -285,15 +241,14 @@ xfs_cmn_err_fsblock_zero(
return EFSCORRUPTED;
}
-int
+STATIC int
xfs_iomap_write_direct(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
int flags,
xfs_bmbt_irec_t *ret_imap,
- int *nmaps,
- int found)
+ int *nmaps)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb;
@@ -330,7 +285,7 @@ xfs_iomap_write_direct(
if (error)
goto error_out;
} else {
- if (found && (ret_imap->br_startblock == HOLESTARTBLOCK))
+ if (*nmaps && (ret_imap->br_startblock == HOLESTARTBLOCK))
last_fsb = MIN(last_fsb, (xfs_fileoff_t)
ret_imap->br_blockcount +
ret_imap->br_startoff);
@@ -485,7 +440,7 @@ xfs_iomap_eof_want_preallocate(
return 0;
}
-int
+STATIC int
xfs_iomap_write_delay(
xfs_inode_t *ip,
xfs_off_t offset,
@@ -588,7 +543,7 @@ retry:
* We no longer bother to look at the incoming map - all we have to
* guarantee is that whatever we allocate fills the required range.
*/
-int
+STATIC int
xfs_iomap_write_allocate(
xfs_inode_t *ip,
xfs_off_t offset,
diff --git a/fs/xfs/xfs_iomap.h b/fs/xfs/xfs_iomap.h
index 174f29990991..81ac4afd45b3 100644
--- a/fs/xfs/xfs_iomap.h
+++ b/fs/xfs/xfs_iomap.h
@@ -18,19 +18,6 @@
#ifndef __XFS_IOMAP_H__
#define __XFS_IOMAP_H__
-#define IOMAP_DADDR_NULL ((xfs_daddr_t) (-1LL))
-
-
-typedef enum { /* iomap_flags values */
- IOMAP_READ = 0, /* mapping for a read */
- IOMAP_HOLE = 0x02, /* mapping covers a hole */
- IOMAP_DELAY = 0x04, /* mapping covers delalloc region */
- IOMAP_REALTIME = 0x10, /* mapping on the realtime device */
- IOMAP_UNWRITTEN = 0x20, /* mapping covers allocated */
- /* but uninitialized file data */
- IOMAP_NEW = 0x40 /* just allocate */
-} iomap_flags_t;
-
typedef enum {
/* base extent manipulation calls */
BMAPI_READ = (1 << 0), /* read extents */
@@ -52,43 +39,11 @@ typedef enum {
{ BMAPI_MMAP, "MMAP" }, \
{ BMAPI_TRYLOCK, "TRYLOCK" }
-/*
- * xfs_iomap_t: File system I/O map
- *
- * The iomap_bn field is expressed in 512-byte blocks, and is where the
- * mapping starts on disk.
- *
- * The iomap_offset, iomap_bsize and iomap_delta fields are in bytes.
- * iomap_offset is the offset of the mapping in the file itself.
- * iomap_bsize is the size of the mapping, iomap_delta is the
- * desired data's offset into the mapping, given the offset supplied
- * to the file I/O map routine.
- *
- * When a request is made to read beyond the logical end of the object,
- * iomap_size may be set to 0, but iomap_offset and iomap_length should be set
- * to the actual amount of underlying storage that has been allocated, if any.
- */
-
-typedef struct xfs_iomap {
- xfs_daddr_t iomap_bn; /* first 512B blk of mapping */
- xfs_buftarg_t *iomap_target;
- xfs_off_t iomap_offset; /* offset of mapping, bytes */
- xfs_off_t iomap_bsize; /* size of mapping, bytes */
- xfs_off_t iomap_delta; /* offset into mapping, bytes */
- iomap_flags_t iomap_flags;
-} xfs_iomap_t;
-
struct xfs_inode;
struct xfs_bmbt_irec;
extern int xfs_iomap(struct xfs_inode *, xfs_off_t, ssize_t, int,
- struct xfs_iomap *, int *);
-extern int xfs_iomap_write_direct(struct xfs_inode *, xfs_off_t, size_t,
- int, struct xfs_bmbt_irec *, int *, int);
-extern int xfs_iomap_write_delay(struct xfs_inode *, xfs_off_t, size_t, int,
- struct xfs_bmbt_irec *, int *);
-extern int xfs_iomap_write_allocate(struct xfs_inode *, xfs_off_t, size_t,
- struct xfs_bmbt_irec *, int *);
+ struct xfs_bmbt_irec *, int *, int *);
extern int xfs_iomap_write_unwritten(struct xfs_inode *, xfs_off_t, size_t);
#endif /* __XFS_IOMAP_H__*/
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index 62efab2f3839..2b86f8610512 100644
--- a/fs/xfs/xfs_itable.c
+++ b/fs/xfs/xfs_itable.c
@@ -49,24 +49,40 @@ xfs_internal_inum(
(ino == mp->m_sb.sb_uquotino || ino == mp->m_sb.sb_gquotino)));
}
-STATIC int
-xfs_bulkstat_one_iget(
- xfs_mount_t *mp, /* mount point for filesystem */
- xfs_ino_t ino, /* inode number to get data for */
- xfs_daddr_t bno, /* starting bno of inode cluster */
- xfs_bstat_t *buf, /* return buffer */
- int *stat) /* BULKSTAT_RV_... */
+/*
+ * Return stat information for one inode.
+ * Return 0 if ok, else errno.
+ */
+int
+xfs_bulkstat_one_int(
+ struct xfs_mount *mp, /* mount point for filesystem */
+ xfs_ino_t ino, /* inode to get data for */
+ void __user *buffer, /* buffer to place output in */
+ int ubsize, /* size of buffer */
+ bulkstat_one_fmt_pf formatter, /* formatter, copy to user */
+ int *ubused, /* bytes used by me */
+ int *stat) /* BULKSTAT_RV_... */
{
- xfs_icdinode_t *dic; /* dinode core info pointer */
- xfs_inode_t *ip; /* incore inode pointer */
- struct inode *inode;
- int error;
+ struct xfs_icdinode *dic; /* dinode core info pointer */
+ struct xfs_inode *ip; /* incore inode pointer */
+ struct inode *inode;
+ struct xfs_bstat *buf; /* return buffer */
+ int error = 0; /* error value */
+
+ *stat = BULKSTAT_RV_NOTHING;
+
+ if (!buffer || xfs_internal_inum(mp, ino))
+ return XFS_ERROR(EINVAL);
+
+ buf = kmem_alloc(sizeof(*buf), KM_SLEEP | KM_MAYFAIL);
+ if (!buf)
+ return XFS_ERROR(ENOMEM);
error = xfs_iget(mp, NULL, ino,
- XFS_IGET_BULKSTAT, XFS_ILOCK_SHARED, &ip, bno);
+ XFS_IGET_UNTRUSTED, XFS_ILOCK_SHARED, &ip);
if (error) {
*stat = BULKSTAT_RV_NOTHING;
- return error;
+ goto out_free;
}
ASSERT(ip != NULL);
@@ -106,6 +122,7 @@ xfs_bulkstat_one_iget(
buf->bs_dmevmask = dic->di_dmevmask;
buf->bs_dmstate = dic->di_dmstate;
buf->bs_aextents = dic->di_anextents;
+ buf->bs_forkoff = XFS_IFORK_BOFF(ip);
switch (dic->di_format) {
case XFS_DINODE_FMT_DEV:
@@ -126,76 +143,16 @@ xfs_bulkstat_one_iget(
buf->bs_blocks = dic->di_nblocks + ip->i_delayed_blks;
break;
}
-
xfs_iput(ip, XFS_ILOCK_SHARED);
- return error;
-}
-STATIC void
-xfs_bulkstat_one_dinode(
- xfs_mount_t *mp, /* mount point for filesystem */
- xfs_ino_t ino, /* inode number to get data for */
- xfs_dinode_t *dic, /* dinode inode pointer */
- xfs_bstat_t *buf) /* return buffer */
-{
- /*
- * The inode format changed when we moved the link count and
- * made it 32 bits long. If this is an old format inode,
- * convert it in memory to look like a new one. If it gets
- * flushed to disk we will convert back before flushing or
- * logging it. We zero out the new projid field and the old link
- * count field. We'll handle clearing the pad field (the remains
- * of the old uuid field) when we actually convert the inode to
- * the new format. We don't change the version number so that we
- * can distinguish this from a real new format inode.
- */
- if (dic->di_version == 1) {
- buf->bs_nlink = be16_to_cpu(dic->di_onlink);
- buf->bs_projid = 0;
- } else {
- buf->bs_nlink = be32_to_cpu(dic->di_nlink);
- buf->bs_projid = be16_to_cpu(dic->di_projid);
- }
+ error = formatter(buffer, ubsize, ubused, buf);
- buf->bs_ino = ino;
- buf->bs_mode = be16_to_cpu(dic->di_mode);
- buf->bs_uid = be32_to_cpu(dic->di_uid);
- buf->bs_gid = be32_to_cpu(dic->di_gid);
- buf->bs_size = be64_to_cpu(dic->di_size);
- buf->bs_atime.tv_sec = be32_to_cpu(dic->di_atime.t_sec);
- buf->bs_atime.tv_nsec = be32_to_cpu(dic->di_atime.t_nsec);
- buf->bs_mtime.tv_sec = be32_to_cpu(dic->di_mtime.t_sec);
- buf->bs_mtime.tv_nsec = be32_to_cpu(dic->di_mtime.t_nsec);
- buf->bs_ctime.tv_sec = be32_to_cpu(dic->di_ctime.t_sec);
- buf->bs_ctime.tv_nsec = be32_to_cpu(dic->di_ctime.t_nsec);
- buf->bs_xflags = xfs_dic2xflags(dic);
- buf->bs_extsize = be32_to_cpu(dic->di_extsize) << mp->m_sb.sb_blocklog;
- buf->bs_extents = be32_to_cpu(dic->di_nextents);
- buf->bs_gen = be32_to_cpu(dic->di_gen);
- memset(buf->bs_pad, 0, sizeof(buf->bs_pad));
- buf->bs_dmevmask = be32_to_cpu(dic->di_dmevmask);
- buf->bs_dmstate = be16_to_cpu(dic->di_dmstate);
- buf->bs_aextents = be16_to_cpu(dic->di_anextents);
+ if (!error)
+ *stat = BULKSTAT_RV_DIDONE;
- switch (dic->di_format) {
- case XFS_DINODE_FMT_DEV:
- buf->bs_rdev = xfs_dinode_get_rdev(dic);
- buf->bs_blksize = BLKDEV_IOSIZE;
- buf->bs_blocks = 0;
- break;
- case XFS_DINODE_FMT_LOCAL:
- case XFS_DINODE_FMT_UUID:
- buf->bs_rdev = 0;
- buf->bs_blksize = mp->m_sb.sb_blocksize;
- buf->bs_blocks = 0;
- break;
- case XFS_DINODE_FMT_EXTENTS:
- case XFS_DINODE_FMT_BTREE:
- buf->bs_rdev = 0;
- buf->bs_blksize = mp->m_sb.sb_blocksize;
- buf->bs_blocks = be64_to_cpu(dic->di_nblocks);
- break;
- }
+ out_free:
+ kmem_free(buf);
+ return error;
}
/* Return 0 on success or positive error */
@@ -215,118 +172,17 @@ xfs_bulkstat_one_fmt(
return 0;
}
-/*
- * Return stat information for one inode.
- * Return 0 if ok, else errno.
- */
-int /* error status */
-xfs_bulkstat_one_int(
- 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 */
- bulkstat_one_fmt_pf formatter, /* formatter, copy to user */
- 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_... */
-{
- xfs_bstat_t *buf; /* return buffer */
- int error = 0; /* error value */
- xfs_dinode_t *dip; /* dinode inode pointer */
-
- dip = (xfs_dinode_t *)dibuff;
- *stat = BULKSTAT_RV_NOTHING;
-
- if (!buffer || xfs_internal_inum(mp, ino))
- return XFS_ERROR(EINVAL);
-
- buf = kmem_alloc(sizeof(*buf), KM_SLEEP);
-
- if (dip == NULL) {
- /* We're not being passed a pointer to a dinode. This happens
- * if BULKSTAT_FG_IGET is selected. Do the iget.
- */
- error = xfs_bulkstat_one_iget(mp, ino, bno, buf, stat);
- if (error)
- goto out_free;
- } else {
- xfs_bulkstat_one_dinode(mp, ino, dip, buf);
- }
-
- error = formatter(buffer, ubsize, ubused, buf);
- if (error)
- goto out_free;
-
- *stat = BULKSTAT_RV_DIDONE;
-
- out_free:
- kmem_free(buf);
- return error;
-}
-
int
xfs_bulkstat_one(
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, bno,
- ubused, dibuff, stat);
-}
-
-/*
- * Test to see whether we can use the ondisk inode directly, based
- * on the given bulkstat flags, filling in dipp accordingly.
- * Returns zero if the inode is dodgey.
- */
-STATIC int
-xfs_bulkstat_use_dinode(
- xfs_mount_t *mp,
- int flags,
- xfs_buf_t *bp,
- int clustidx,
- xfs_dinode_t **dipp)
-{
- xfs_dinode_t *dip;
- unsigned int aformat;
-
- *dipp = NULL;
- if (!bp || (flags & BULKSTAT_FG_IGET))
- return 1;
- dip = (xfs_dinode_t *)
- xfs_buf_offset(bp, clustidx << mp->m_sb.sb_inodelog);
- /*
- * Check the buffer containing the on-disk inode for di_mode == 0.
- * This is to prevent xfs_bulkstat from picking up just reclaimed
- * inodes that have their in-core state initialized but not flushed
- * to disk yet. This is a temporary hack that would require a proper
- * fix in the future.
- */
- if (be16_to_cpu(dip->di_magic) != XFS_DINODE_MAGIC ||
- !XFS_DINODE_GOOD_VERSION(dip->di_version) ||
- !dip->di_mode)
- return 0;
- if (flags & BULKSTAT_FG_QUICK) {
- *dipp = dip;
- return 1;
- }
- /* BULKSTAT_FG_INLINE: if attr fork is local, or not there, use it */
- aformat = dip->di_aformat;
- if ((XFS_DFORK_Q(dip) == 0) ||
- (aformat == XFS_DINODE_FMT_LOCAL) ||
- (aformat == XFS_DINODE_FMT_EXTENTS && !dip->di_anextents)) {
- *dipp = dip;
- return 1;
- }
- return 1;
+ xfs_bulkstat_one_fmt, ubused, stat);
}
#define XFS_BULKSTAT_UBLEFT(ubleft) ((ubleft) >= statstruct_size)
@@ -340,10 +196,8 @@ xfs_bulkstat(
xfs_ino_t *lastinop, /* last inode returned */
int *ubcountp, /* size of buffer/count returned */
bulkstat_one_pf formatter, /* func that'd fill a single buf */
- void *private_data,/* private data for formatter */
size_t statstruct_size, /* sizeof struct filling */
char __user *ubuffer, /* buffer with inode stats */
- int flags, /* defined in xfs_itable.h */
int *done) /* 1 if there are more stats to get */
{
xfs_agblock_t agbno=0;/* allocation group block number */
@@ -378,14 +232,12 @@ xfs_bulkstat(
int ubelem; /* spaces used in user's buffer */
int ubused; /* bytes used by formatter */
xfs_buf_t *bp; /* ptr to on-disk inode cluster buf */
- xfs_dinode_t *dip; /* ptr into bp for specific inode */
/*
* Get the last inode value, see if there's nothing to do.
*/
ino = (xfs_ino_t)*lastinop;
lastino = ino;
- dip = NULL;
agno = XFS_INO_TO_AGNO(mp, ino);
agino = XFS_INO_TO_AGINO(mp, ino);
if (agno >= mp->m_sb.sb_agcount ||
@@ -408,8 +260,10 @@ xfs_bulkstat(
(XFS_INODE_CLUSTER_SIZE(mp) >> mp->m_sb.sb_inodelog);
nimask = ~(nicluster - 1);
nbcluster = nicluster >> mp->m_sb.sb_inopblog;
- irbuf = kmem_zalloc_greedy(&irbsize, PAGE_SIZE, PAGE_SIZE * 4,
- KM_SLEEP | KM_MAYFAIL | KM_LARGE);
+ irbuf = kmem_zalloc_greedy(&irbsize, PAGE_SIZE, PAGE_SIZE * 4);
+ if (!irbuf)
+ return ENOMEM;
+
nirbuf = irbsize / sizeof(*irbuf);
/*
@@ -420,9 +274,7 @@ xfs_bulkstat(
while (XFS_BULKSTAT_UBLEFT(ubleft) && agno < mp->m_sb.sb_agcount) {
cond_resched();
bp = NULL;
- down_read(&mp->m_peraglock);
error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp);
- up_read(&mp->m_peraglock);
if (error) {
/*
* Skip this allocation group and go to the next one.
@@ -610,37 +462,6 @@ xfs_bulkstat(
irbp->ir_startino) +
((chunkidx & nimask) >>
mp->m_sb.sb_inopblog);
-
- if (flags & (BULKSTAT_FG_QUICK |
- BULKSTAT_FG_INLINE)) {
- int offset;
-
- ino = XFS_AGINO_TO_INO(mp, agno,
- agino);
- bno = XFS_AGB_TO_DADDR(mp, agno,
- agbno);
-
- /*
- * Get the inode cluster buffer
- */
- if (bp)
- xfs_buf_relse(bp);
-
- error = xfs_inotobp(mp, NULL, ino, &dip,
- &bp, &offset,
- XFS_IGET_BULKSTAT);
-
- if (!error)
- clustidx = offset / mp->m_sb.sb_inodesize;
- if (XFS_TEST_ERROR(error != 0,
- mp, XFS_ERRTAG_BULKSTAT_READ_CHUNK,
- XFS_RANDOM_BULKSTAT_READ_CHUNK)) {
- bp = NULL;
- ubleft = 0;
- rval = error;
- break;
- }
- }
}
ino = XFS_AGINO_TO_INO(mp, agno, agino);
bno = XFS_AGB_TO_DADDR(mp, agno, agbno);
@@ -656,35 +477,13 @@ xfs_bulkstat(
* when the chunk is used up.
*/
irbp->ir_freecount++;
- if (!xfs_bulkstat_use_dinode(mp, flags, bp,
- clustidx, &dip)) {
- lastino = ino;
- continue;
- }
- /*
- * If we need to do an iget, cannot hold bp.
- * Drop it, until starting the next cluster.
- */
- if ((flags & BULKSTAT_FG_INLINE) && !dip) {
- if (bp)
- xfs_buf_relse(bp);
- bp = NULL;
- }
/*
* Get the inode and fill in a single buffer.
- * BULKSTAT_FG_QUICK uses dip to fill it in.
- * BULKSTAT_FG_IGET uses igets.
- * BULKSTAT_FG_INLINE uses dip if we have an
- * inline attr fork, else igets.
- * See: xfs_bulkstat_one & xfs_dm_bulkstat_one.
- * This is also used to count inodes/blks, etc
- * in xfs_qm_quotacheck.
*/
ubused = statstruct_size;
- error = formatter(mp, ino, ubufp,
- ubleft, private_data,
- bno, &ubused, dip, &fmterror);
+ error = formatter(mp, ino, ubufp, ubleft,
+ &ubused, &fmterror);
if (fmterror == BULKSTAT_RV_NOTHING) {
if (error && error != ENOENT &&
error != EINVAL) {
@@ -729,7 +528,7 @@ xfs_bulkstat(
/*
* Done, we're either out of filesystem or space to put the data.
*/
- kmem_free(irbuf);
+ kmem_free_large(irbuf);
*ubcountp = ubelem;
/*
* Found some inodes, return them now and return the error next time.
@@ -776,8 +575,7 @@ xfs_bulkstat_single(
*/
ino = (xfs_ino_t)*lastinop;
- error = xfs_bulkstat_one(mp, ino, buffer, sizeof(xfs_bstat_t),
- NULL, 0, NULL, NULL, &res);
+ error = xfs_bulkstat_one(mp, ino, buffer, sizeof(xfs_bstat_t), 0, &res);
if (error) {
/*
* Special case way failed, do it the "long" way
@@ -786,8 +584,7 @@ xfs_bulkstat_single(
(*lastinop)--;
count = 1;
if (xfs_bulkstat(mp, lastinop, &count, xfs_bulkstat_one,
- NULL, sizeof(xfs_bstat_t), buffer,
- BULKSTAT_FG_IGET, done))
+ sizeof(xfs_bstat_t), buffer, done))
return error;
if (count == 0 || (xfs_ino_t)*lastinop != ino)
return error == EFSCORRUPTED ?
@@ -849,9 +646,7 @@ xfs_inumbers(
agbp = NULL;
while (left > 0 && agno < mp->m_sb.sb_agcount) {
if (agbp == NULL) {
- down_read(&mp->m_peraglock);
error = xfs_ialloc_read_agi(mp, NULL, agno, &agbp);
- up_read(&mp->m_peraglock);
if (error) {
/*
* If we can't read the AGI of this ag,
diff --git a/fs/xfs/xfs_itable.h b/fs/xfs/xfs_itable.h
index 20792bf45946..97295d91d170 100644
--- a/fs/xfs/xfs_itable.h
+++ b/fs/xfs/xfs_itable.h
@@ -27,10 +27,7 @@ typedef int (*bulkstat_one_pf)(struct xfs_mount *mp,
xfs_ino_t ino,
void __user *buffer,
int ubsize,
- void *private_data,
- xfs_daddr_t bno,
int *ubused,
- void *dip,
int *stat);
/*
@@ -41,13 +38,6 @@ typedef int (*bulkstat_one_pf)(struct xfs_mount *mp,
#define BULKSTAT_RV_GIVEUP 2
/*
- * Values for bulkstat flag argument.
- */
-#define BULKSTAT_FG_IGET 0x1 /* Go through the buffer cache */
-#define BULKSTAT_FG_QUICK 0x2 /* No iget, walk the dinode cluster */
-#define BULKSTAT_FG_INLINE 0x4 /* No iget if inline attrs */
-
-/*
* Return stat information in bulk (by-inode) for the filesystem.
*/
int /* error status */
@@ -56,10 +46,8 @@ xfs_bulkstat(
xfs_ino_t *lastino, /* last inode returned */
int *count, /* size of buffer/count returned */
bulkstat_one_pf formatter, /* func that'd fill a single buf */
- void *private_data, /* private data for formatter */
size_t statstruct_size,/* sizeof struct that we're filling */
char __user *ubuffer,/* buffer with inode stats */
- int flags, /* flag to control access method */
int *done); /* 1 if there are more stats to get */
int
@@ -82,9 +70,7 @@ xfs_bulkstat_one_int(
void __user *buffer,
int ubsize,
bulkstat_one_fmt_pf formatter,
- xfs_daddr_t bno,
int *ubused,
- void *dibuff,
int *stat);
int
@@ -93,10 +79,7 @@ xfs_bulkstat_one(
xfs_ino_t ino,
void __user *buffer,
int ubsize,
- void *private_data,
- xfs_daddr_t bno,
int *ubused,
- void *dibuff,
int *stat);
typedef int (*inumbers_fmt_pf)(
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 600b5b06aaeb..5215abc8023a 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -44,14 +44,8 @@
kmem_zone_t *xfs_log_ticket_zone;
-#define xlog_write_adv_cnt(ptr, len, off, bytes) \
- { (ptr) += (bytes); \
- (len) -= (bytes); \
- (off) += (bytes);}
-
/* Local miscellaneous function prototypes */
-STATIC int xlog_bdstrat_cb(struct xfs_buf *);
-STATIC int xlog_commit_record(xfs_mount_t *mp, xlog_ticket_t *ticket,
+STATIC int xlog_commit_record(struct log *log, struct xlog_ticket *ticket,
xlog_in_core_t **, xfs_lsn_t *);
STATIC xlog_t * xlog_alloc_log(xfs_mount_t *mp,
xfs_buftarg_t *log_target,
@@ -60,11 +54,6 @@ STATIC xlog_t * xlog_alloc_log(xfs_mount_t *mp,
STATIC int xlog_space_left(xlog_t *log, int cycle, int bytes);
STATIC int xlog_sync(xlog_t *log, xlog_in_core_t *iclog);
STATIC void xlog_dealloc_log(xlog_t *log);
-STATIC int xlog_write(xfs_mount_t *mp, xfs_log_iovec_t region[],
- int nentries, xfs_log_ticket_t tic,
- xfs_lsn_t *start_lsn,
- xlog_in_core_t **commit_iclog,
- uint flags);
/* local state machine functions */
STATIC void xlog_state_done_syncing(xlog_in_core_t *iclog, int);
@@ -80,11 +69,6 @@ STATIC int xlog_state_release_iclog(xlog_t *log,
STATIC void xlog_state_switch_iclogs(xlog_t *log,
xlog_in_core_t *iclog,
int eventual_size);
-STATIC int xlog_state_sync(xlog_t *log,
- xfs_lsn_t lsn,
- uint flags,
- int *log_flushed);
-STATIC int xlog_state_sync_all(xlog_t *log, uint flags, int *log_flushed);
STATIC void xlog_state_want_sync(xlog_t *log, xlog_in_core_t *iclog);
/* local functions to manipulate grant head */
@@ -99,16 +83,8 @@ STATIC int xlog_regrant_write_log_space(xlog_t *log,
STATIC void xlog_ungrant_log_space(xlog_t *log,
xlog_ticket_t *ticket);
-
-/* local ticket functions */
-STATIC xlog_ticket_t *xlog_ticket_alloc(xlog_t *log,
- int unit_bytes,
- int count,
- char clientid,
- uint flags);
-
#if defined(DEBUG)
-STATIC void xlog_verify_dest_ptr(xlog_t *log, __psint_t ptr);
+STATIC void xlog_verify_dest_ptr(xlog_t *log, char *ptr);
STATIC void xlog_verify_grant_head(xlog_t *log, int equals);
STATIC void xlog_verify_iclog(xlog_t *log, xlog_in_core_t *iclog,
int count, boolean_t syncing);
@@ -249,14 +225,14 @@ xlog_tic_add_region(xlog_ticket_t *tic, uint len, uint type)
* out when the next write occurs.
*/
xfs_lsn_t
-xfs_log_done(xfs_mount_t *mp,
- xfs_log_ticket_t xtic,
- void **iclog,
- uint flags)
+xfs_log_done(
+ struct xfs_mount *mp,
+ struct xlog_ticket *ticket,
+ struct xlog_in_core **iclog,
+ uint flags)
{
- xlog_t *log = mp->m_log;
- xlog_ticket_t *ticket = (xfs_log_ticket_t) xtic;
- xfs_lsn_t lsn = 0;
+ struct log *log = mp->m_log;
+ xfs_lsn_t lsn = 0;
if (XLOG_FORCED_SHUTDOWN(log) ||
/*
@@ -264,8 +240,7 @@ xfs_log_done(xfs_mount_t *mp,
* If we get an error, just continue and give back the log ticket.
*/
(((ticket->t_flags & XLOG_TIC_INITED) == 0) &&
- (xlog_commit_record(mp, ticket,
- (xlog_in_core_t **)iclog, &lsn)))) {
+ (xlog_commit_record(log, ticket, iclog, &lsn)))) {
lsn = (xfs_lsn_t) -1;
if (ticket->t_flags & XLOG_TIC_PERM_RESERV) {
flags |= XFS_LOG_REL_PERM_RESERV;
@@ -295,67 +270,8 @@ xfs_log_done(xfs_mount_t *mp,
}
return lsn;
-} /* xfs_log_done */
-
-
-/*
- * Force the in-core log to disk. If flags == XFS_LOG_SYNC,
- * the force is done synchronously.
- *
- * Asynchronous forces are implemented by setting the WANT_SYNC
- * bit in the appropriate in-core log and then returning.
- *
- * Synchronous forces are implemented with a signal variable. All callers
- * to force a given lsn to disk will wait on a the sv attached to the
- * specific in-core log. When given in-core log finally completes its
- * write to disk, that thread will wake up all threads waiting on the
- * sv.
- */
-int
-_xfs_log_force(
- xfs_mount_t *mp,
- xfs_lsn_t lsn,
- uint flags,
- int *log_flushed)
-{
- xlog_t *log = mp->m_log;
- int dummy;
-
- if (!log_flushed)
- log_flushed = &dummy;
-
- ASSERT(flags & XFS_LOG_FORCE);
-
- XFS_STATS_INC(xs_log_force);
-
- if (log->l_flags & XLOG_IO_ERROR)
- return XFS_ERROR(EIO);
- if (lsn == 0)
- return xlog_state_sync_all(log, flags, log_flushed);
- else
- return xlog_state_sync(log, lsn, flags, log_flushed);
-} /* _xfs_log_force */
-
-/*
- * Wrapper for _xfs_log_force(), to be used when caller doesn't care
- * about errors or whether the log was flushed or not. This is the normal
- * interface to use when trying to unpin items or move the log forward.
- */
-void
-xfs_log_force(
- xfs_mount_t *mp,
- xfs_lsn_t lsn,
- uint flags)
-{
- int error;
- error = _xfs_log_force(mp, lsn, flags, NULL);
- if (error) {
- xfs_fs_cmn_err(CE_WARN, mp, "xfs_log_force: "
- "error %d returned.", error);
- }
}
-
/*
* Attaches a new iclog I/O completion callback routine during
* transaction commit. If the log is in error state, a non-zero
@@ -363,11 +279,11 @@ xfs_log_force(
* executing the callback at an appropriate time.
*/
int
-xfs_log_notify(xfs_mount_t *mp, /* mount of partition */
- void *iclog_hndl, /* iclog to hang callback off */
- xfs_log_callback_t *cb)
+xfs_log_notify(
+ struct xfs_mount *mp,
+ struct xlog_in_core *iclog,
+ xfs_log_callback_t *cb)
{
- xlog_in_core_t *iclog = (xlog_in_core_t *)iclog_hndl;
int abortflg;
spin_lock(&iclog->ic_callback_lock);
@@ -381,16 +297,14 @@ xfs_log_notify(xfs_mount_t *mp, /* mount of partition */
}
spin_unlock(&iclog->ic_callback_lock);
return abortflg;
-} /* xfs_log_notify */
+}
int
-xfs_log_release_iclog(xfs_mount_t *mp,
- void *iclog_hndl)
+xfs_log_release_iclog(
+ struct xfs_mount *mp,
+ struct xlog_in_core *iclog)
{
- xlog_t *log = mp->m_log;
- xlog_in_core_t *iclog = (xlog_in_core_t *)iclog_hndl;
-
- if (xlog_state_release_iclog(log, iclog)) {
+ if (xlog_state_release_iclog(mp->m_log, iclog)) {
xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
return EIO;
}
@@ -409,17 +323,18 @@ xfs_log_release_iclog(xfs_mount_t *mp,
* reservation, we prevent over allocation problems.
*/
int
-xfs_log_reserve(xfs_mount_t *mp,
- int unit_bytes,
- int cnt,
- xfs_log_ticket_t *ticket,
- __uint8_t client,
- uint flags,
- uint t_type)
+xfs_log_reserve(
+ struct xfs_mount *mp,
+ int unit_bytes,
+ int cnt,
+ struct xlog_ticket **ticket,
+ __uint8_t client,
+ uint flags,
+ uint t_type)
{
- xlog_t *log = mp->m_log;
- xlog_ticket_t *internal_ticket;
- int retval = 0;
+ struct log *log = mp->m_log;
+ struct xlog_ticket *internal_ticket;
+ int retval = 0;
ASSERT(client == XFS_TRANSACTION || client == XFS_LOG);
ASSERT((flags & XFS_LOG_NOSLEEP) == 0);
@@ -432,7 +347,16 @@ xfs_log_reserve(xfs_mount_t *mp,
if (*ticket != NULL) {
ASSERT(flags & XFS_LOG_PERM_RESERV);
- internal_ticket = (xlog_ticket_t *)*ticket;
+ internal_ticket = *ticket;
+
+ /*
+ * this is a new transaction on the ticket, so we need to
+ * change the transaction ID so that the next transaction has a
+ * different TID in the log. Just add one to the existing tid
+ * so that we can see chains of rolling transactions in the log
+ * easily.
+ */
+ internal_ticket->t_tid++;
trace_xfs_log_reserve(log, internal_ticket);
@@ -441,7 +365,8 @@ xfs_log_reserve(xfs_mount_t *mp,
} else {
/* may sleep if need to allocate more tickets */
internal_ticket = xlog_ticket_alloc(log, unit_bytes, cnt,
- client, flags);
+ client, flags,
+ KM_SLEEP|KM_MAYFAIL);
if (!internal_ticket)
return XFS_ERROR(ENOMEM);
internal_ticket->t_trans_type = t_type;
@@ -526,6 +451,13 @@ xfs_log_mount(
/* Normal transactions can now occur */
mp->m_log->l_flags &= ~XLOG_ACTIVE_RECOVERY;
+ /*
+ * Now the log has been fully initialised and we know were our
+ * space grant counters are, we can initialise the permanent ticket
+ * needed for delayed logging to work.
+ */
+ xlog_cil_init_post_recovery(mp->m_log);
+
return 0;
out_destroy_ail:
@@ -583,18 +515,10 @@ xfs_log_unmount_write(xfs_mount_t *mp)
#ifdef DEBUG
xlog_in_core_t *first_iclog;
#endif
- xfs_log_iovec_t reg[1];
- xfs_log_ticket_t tic = NULL;
+ xlog_ticket_t *tic = NULL;
xfs_lsn_t lsn;
int error;
- /* the data section must be 32 bit size aligned */
- struct {
- __uint16_t magic;
- __uint16_t pad1;
- __uint32_t pad2; /* may as well make it 64 bits */
- } magic = { XLOG_UNMOUNT_TYPE, 0, 0 };
-
/*
* Don't write out unmount record on read-only mounts.
* Or, if we are doing a forced umount (typically because of IO errors).
@@ -602,7 +526,7 @@ xfs_log_unmount_write(xfs_mount_t *mp)
if (mp->m_flags & XFS_MOUNT_RDONLY)
return 0;
- error = _xfs_log_force(mp, 0, XFS_LOG_FORCE|XFS_LOG_SYNC, NULL);
+ error = _xfs_log_force(mp, XFS_LOG_SYNC, NULL);
ASSERT(error || !(XLOG_FORCED_SHUTDOWN(log)));
#ifdef DEBUG
@@ -616,16 +540,30 @@ xfs_log_unmount_write(xfs_mount_t *mp)
} while (iclog != first_iclog);
#endif
if (! (XLOG_FORCED_SHUTDOWN(log))) {
- reg[0].i_addr = (void*)&magic;
- reg[0].i_len = sizeof(magic);
- XLOG_VEC_SET_TYPE(&reg[0], XLOG_REG_TYPE_UNMOUNT);
-
error = xfs_log_reserve(mp, 600, 1, &tic,
XFS_LOG, 0, XLOG_UNMOUNT_REC_TYPE);
if (!error) {
+ /* the data section must be 32 bit size aligned */
+ struct {
+ __uint16_t magic;
+ __uint16_t pad1;
+ __uint32_t pad2; /* may as well make it 64 bits */
+ } magic = {
+ .magic = XLOG_UNMOUNT_TYPE,
+ };
+ struct xfs_log_iovec reg = {
+ .i_addr = (void *)&magic,
+ .i_len = sizeof(magic),
+ .i_type = XLOG_REG_TYPE_UNMOUNT,
+ };
+ struct xfs_log_vec vec = {
+ .lv_niovecs = 1,
+ .lv_iovecp = &reg,
+ };
+
/* remove inited flag */
- ((xlog_ticket_t *)tic)->t_flags = 0;
- error = xlog_write(mp, reg, 1, tic, &lsn,
+ tic->t_flags = 0;
+ error = xlog_write(log, &vec, tic, &lsn,
NULL, XLOG_UNMOUNT_TRANS);
/*
* At this point, we're umounting anyway,
@@ -715,30 +653,54 @@ xfs_log_unmount(xfs_mount_t *mp)
xlog_dealloc_log(mp->m_log);
}
+void
+xfs_log_item_init(
+ struct xfs_mount *mp,
+ struct xfs_log_item *item,
+ int type,
+ struct xfs_item_ops *ops)
+{
+ item->li_mountp = mp;
+ item->li_ailp = mp->m_ail;
+ item->li_type = type;
+ item->li_ops = ops;
+ item->li_lv = NULL;
+
+ INIT_LIST_HEAD(&item->li_ail);
+ INIT_LIST_HEAD(&item->li_cil);
+}
+
/*
* Write region vectors to log. The write happens using the space reservation
* of the ticket (tic). It is not a requirement that all writes for a given
- * transaction occur with one call to xfs_log_write().
+ * transaction occur with one call to xfs_log_write(). However, it is important
+ * to note that the transaction reservation code makes an assumption about the
+ * number of log headers a transaction requires that may be violated if you
+ * don't pass all the transaction vectors in one call....
*/
int
-xfs_log_write(xfs_mount_t * mp,
- xfs_log_iovec_t reg[],
- int nentries,
- xfs_log_ticket_t tic,
- xfs_lsn_t *start_lsn)
+xfs_log_write(
+ struct xfs_mount *mp,
+ struct xfs_log_iovec reg[],
+ int nentries,
+ struct xlog_ticket *tic,
+ xfs_lsn_t *start_lsn)
{
- int error;
- xlog_t *log = mp->m_log;
+ struct log *log = mp->m_log;
+ int error;
+ struct xfs_log_vec vec = {
+ .lv_niovecs = nentries,
+ .lv_iovecp = reg,
+ };
if (XLOG_FORCED_SHUTDOWN(log))
return XFS_ERROR(EIO);
- if ((error = xlog_write(mp, reg, nentries, tic, start_lsn, NULL, 0))) {
+ error = xlog_write(log, &vec, tic, start_lsn, NULL, 0);
+ if (error)
xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
- }
return error;
-} /* xfs_log_write */
-
+}
void
xfs_log_move_tail(xfs_mount_t *mp,
@@ -812,9 +774,16 @@ xfs_log_move_tail(xfs_mount_t *mp,
/*
* Determine if we have a transaction that has gone to disk
- * that needs to be covered. Log activity needs to be idle (no AIL and
- * nothing in the iclogs). And, we need to be in the right state indicating
- * something has gone out.
+ * that needs to be covered. To begin the transition to the idle state
+ * firstly the log needs to be idle (no AIL and nothing in the iclogs).
+ * If we are then in a state where covering is needed, the caller is informed
+ * that dummy transactions are required to move the log into the idle state.
+ *
+ * Because this is called as part of the sync process, we should also indicate
+ * that dummy transactions should be issued in anything but the covered or
+ * idle states. This ensures that the log tail is accurately reflected in
+ * the log at the end of the sync, hence if a crash occurrs avoids replay
+ * of transactions where the metadata is already on disk.
*/
int
xfs_log_need_covered(xfs_mount_t *mp)
@@ -826,17 +795,24 @@ xfs_log_need_covered(xfs_mount_t *mp)
return 0;
spin_lock(&log->l_icloglock);
- if (((log->l_covered_state == XLOG_STATE_COVER_NEED) ||
- (log->l_covered_state == XLOG_STATE_COVER_NEED2))
- && !xfs_trans_ail_tail(log->l_ailp)
- && xlog_iclogs_empty(log)) {
- if (log->l_covered_state == XLOG_STATE_COVER_NEED)
- log->l_covered_state = XLOG_STATE_COVER_DONE;
- else {
- ASSERT(log->l_covered_state == XLOG_STATE_COVER_NEED2);
- log->l_covered_state = XLOG_STATE_COVER_DONE2;
+ switch (log->l_covered_state) {
+ case XLOG_STATE_COVER_DONE:
+ case XLOG_STATE_COVER_DONE2:
+ case XLOG_STATE_COVER_IDLE:
+ break;
+ case XLOG_STATE_COVER_NEED:
+ case XLOG_STATE_COVER_NEED2:
+ if (!xfs_trans_ail_tail(log->l_ailp) &&
+ xlog_iclogs_empty(log)) {
+ if (log->l_covered_state == XLOG_STATE_COVER_NEED)
+ log->l_covered_state = XLOG_STATE_COVER_DONE;
+ else
+ log->l_covered_state = XLOG_STATE_COVER_DONE2;
}
+ /* FALLTHRU */
+ default:
needed = 1;
+ break;
}
spin_unlock(&log->l_icloglock);
return needed;
@@ -988,35 +964,6 @@ xlog_iodone(xfs_buf_t *bp)
} /* xlog_iodone */
/*
- * The bdstrat callback function for log bufs. This gives us a central
- * place to trap bufs in case we get hit by a log I/O error and need to
- * shutdown. Actually, in practice, even when we didn't get a log error,
- * we transition the iclogs to IOERROR state *after* flushing all existing
- * iclogs to disk. This is because we don't want anymore new transactions to be
- * started or completed afterwards.
- */
-STATIC int
-xlog_bdstrat_cb(struct xfs_buf *bp)
-{
- xlog_in_core_t *iclog;
-
- iclog = XFS_BUF_FSPRIVATE(bp, xlog_in_core_t *);
-
- if ((iclog->ic_state & XLOG_STATE_IOERROR) == 0) {
- /* note for irix bstrat will need struct bdevsw passed
- * Fix the following macro if the code ever is merged
- */
- XFS_bdstrat(bp);
- return 0;
- }
-
- XFS_BUF_ERROR(bp, EIO);
- XFS_BUF_STALE(bp);
- xfs_biodone(bp);
- return XFS_ERROR(EIO);
-}
-
-/*
* Return size of each in-core log record buffer.
*
* All machines get 8 x 32kB buffers by default, unless tuned otherwise.
@@ -1102,6 +1049,7 @@ xlog_alloc_log(xfs_mount_t *mp,
int i;
int iclogsize;
int error = ENOMEM;
+ uint log2_size = 0;
log = kmem_zalloc(sizeof(xlog_t), KM_MAYFAIL);
if (!log) {
@@ -1127,29 +1075,30 @@ xlog_alloc_log(xfs_mount_t *mp,
error = EFSCORRUPTED;
if (xfs_sb_version_hassector(&mp->m_sb)) {
- log->l_sectbb_log = mp->m_sb.sb_logsectlog - BBSHIFT;
- if (log->l_sectbb_log < 0 ||
- log->l_sectbb_log > mp->m_sectbb_log) {
- xlog_warn("XFS: Log sector size (0x%x) out of range.",
- log->l_sectbb_log);
+ log2_size = mp->m_sb.sb_logsectlog;
+ if (log2_size < BBSHIFT) {
+ xlog_warn("XFS: Log sector size too small "
+ "(0x%x < 0x%x)", log2_size, BBSHIFT);
goto out_free_log;
}
- /* for larger sector sizes, must have v2 or external log */
- if (log->l_sectbb_log != 0 &&
- (log->l_logBBstart != 0 &&
- !xfs_sb_version_haslogv2(&mp->m_sb))) {
- xlog_warn("XFS: log sector size (0x%x) invalid "
- "for configuration.", log->l_sectbb_log);
+ log2_size -= BBSHIFT;
+ if (log2_size > mp->m_sectbb_log) {
+ xlog_warn("XFS: Log sector size too large "
+ "(0x%x > 0x%x)", log2_size, mp->m_sectbb_log);
goto out_free_log;
}
- if (mp->m_sb.sb_logsectlog < BBSHIFT) {
- xlog_warn("XFS: Log sector log (0x%x) too small.",
- mp->m_sb.sb_logsectlog);
+
+ /* for larger sector sizes, must have v2 or external log */
+ if (log2_size && log->l_logBBstart > 0 &&
+ !xfs_sb_version_haslogv2(&mp->m_sb)) {
+
+ xlog_warn("XFS: log sector size (0x%x) invalid "
+ "for configuration.", log2_size);
goto out_free_log;
}
}
- log->l_sectbb_mask = (1 << log->l_sectbb_log) - 1;
+ log->l_sectBBsize = 1 << log2_size;
xlog_get_iclog_buffer_size(mp, log);
@@ -1158,7 +1107,6 @@ xlog_alloc_log(xfs_mount_t *mp,
if (!bp)
goto out_free_log;
XFS_BUF_SET_IODONE_FUNC(bp, xlog_iodone);
- XFS_BUF_SET_BDSTRAT_FUNC(bp, xlog_bdstrat_cb);
XFS_BUF_SET_FSPRIVATE2(bp, (unsigned long)1);
ASSERT(XFS_BUF_ISBUSY(bp));
ASSERT(XFS_BUF_VALUSEMA(bp) <= 0);
@@ -1196,7 +1144,6 @@ xlog_alloc_log(xfs_mount_t *mp,
if (!XFS_BUF_CPSEMA(bp))
ASSERT(0);
XFS_BUF_SET_IODONE_FUNC(bp, xlog_iodone);
- XFS_BUF_SET_BDSTRAT_FUNC(bp, xlog_bdstrat_cb);
XFS_BUF_SET_FSPRIVATE2(bp, (unsigned long)1);
iclog->ic_bp = bp;
iclog->ic_data = bp->b_addr;
@@ -1231,6 +1178,9 @@ xlog_alloc_log(xfs_mount_t *mp,
*iclogp = log->l_iclog; /* complete ring */
log->l_iclog->ic_prev = prev_iclog; /* re-write 1st prev ptr */
+ error = xlog_cil_init(log);
+ if (error)
+ goto out_free_iclog;
return log;
out_free_iclog:
@@ -1258,26 +1208,31 @@ out:
* ticket. Return the lsn of the commit record.
*/
STATIC int
-xlog_commit_record(xfs_mount_t *mp,
- xlog_ticket_t *ticket,
- xlog_in_core_t **iclog,
- xfs_lsn_t *commitlsnp)
+xlog_commit_record(
+ struct log *log,
+ struct xlog_ticket *ticket,
+ struct xlog_in_core **iclog,
+ xfs_lsn_t *commitlsnp)
{
- int error;
- xfs_log_iovec_t reg[1];
-
- reg[0].i_addr = NULL;
- reg[0].i_len = 0;
- XLOG_VEC_SET_TYPE(&reg[0], XLOG_REG_TYPE_COMMIT);
+ struct xfs_mount *mp = log->l_mp;
+ int error;
+ struct xfs_log_iovec reg = {
+ .i_addr = NULL,
+ .i_len = 0,
+ .i_type = XLOG_REG_TYPE_COMMIT,
+ };
+ struct xfs_log_vec vec = {
+ .lv_niovecs = 1,
+ .lv_iovecp = &reg,
+ };
ASSERT_ALWAYS(iclog);
- if ((error = xlog_write(mp, reg, 1, ticket, commitlsnp,
- iclog, XLOG_COMMIT_TRANS))) {
+ error = xlog_write(log, &vec, ticket, commitlsnp, iclog,
+ XLOG_COMMIT_TRANS);
+ if (error)
xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
- }
return error;
-} /* xlog_commit_record */
-
+}
/*
* Push on the buffer cache code if we ever use more than 75% of the on-disk
@@ -1343,6 +1298,37 @@ xlog_grant_push_ail(xfs_mount_t *mp,
xfs_trans_ail_push(log->l_ailp, threshold_lsn);
} /* xlog_grant_push_ail */
+/*
+ * The bdstrat callback function for log bufs. This gives us a central
+ * place to trap bufs in case we get hit by a log I/O error and need to
+ * shutdown. Actually, in practice, even when we didn't get a log error,
+ * we transition the iclogs to IOERROR state *after* flushing all existing
+ * iclogs to disk. This is because we don't want anymore new transactions to be
+ * started or completed afterwards.
+ */
+STATIC int
+xlog_bdstrat(
+ struct xfs_buf *bp)
+{
+ struct xlog_in_core *iclog;
+
+ iclog = XFS_BUF_FSPRIVATE(bp, xlog_in_core_t *);
+ if (iclog->ic_state & XLOG_STATE_IOERROR) {
+ XFS_BUF_ERROR(bp, EIO);
+ XFS_BUF_STALE(bp);
+ xfs_biodone(bp);
+ /*
+ * It would seem logical to return EIO here, but we rely on
+ * the log state machine to propagate I/O errors instead of
+ * doing it here.
+ */
+ return 0;
+ }
+
+ bp->b_flags |= _XBF_RUN_QUEUES;
+ xfs_buf_iorequest(bp);
+ return 0;
+}
/*
* Flush out the in-core log (iclog) to the on-disk log in an asynchronous
@@ -1462,7 +1448,7 @@ xlog_sync(xlog_t *log,
*/
XFS_BUF_WRITE(bp);
- if ((error = XFS_bwrite(bp))) {
+ if ((error = xlog_bdstrat(bp))) {
xfs_ioerror_alert("xlog_sync", log->l_mp, bp,
XFS_BUF_ADDR(bp));
return error;
@@ -1502,7 +1488,7 @@ xlog_sync(xlog_t *log,
/* account for internal log which doesn't start at block #0 */
XFS_BUF_SET_ADDR(bp, XFS_BUF_ADDR(bp) + log->l_logBBstart);
XFS_BUF_WRITE(bp);
- if ((error = XFS_bwrite(bp))) {
+ if ((error = xlog_bdstrat(bp))) {
xfs_ioerror_alert("xlog_sync (split)", log->l_mp,
bp, XFS_BUF_ADDR(bp));
return error;
@@ -1521,6 +1507,8 @@ xlog_dealloc_log(xlog_t *log)
xlog_in_core_t *iclog, *next_iclog;
int i;
+ xlog_cil_destroy(log);
+
iclog = log->l_iclog;
for (i=0; i<log->l_iclog_bufs; i++) {
sv_destroy(&iclog->ic_force_wait);
@@ -1563,8 +1551,10 @@ xlog_state_finish_copy(xlog_t *log,
* print out info relating to regions written which consume
* the reservation
*/
-STATIC void
-xlog_print_tic_res(xfs_mount_t *mp, xlog_ticket_t *ticket)
+void
+xlog_print_tic_res(
+ struct xfs_mount *mp,
+ struct xlog_ticket *ticket)
{
uint i;
uint ophdr_spc = ticket->t_res_num_ophdrs * (uint)sizeof(xlog_op_header_t);
@@ -1664,6 +1654,196 @@ xlog_print_tic_res(xfs_mount_t *mp, xlog_ticket_t *ticket)
"bad-rtype" : res_type_str[r_type-1]),
ticket->t_res_arr[i].r_len);
}
+
+ xfs_cmn_err(XFS_PTAG_LOGRES, CE_ALERT, mp,
+ "xfs_log_write: reservation ran out. Need to up reservation");
+ xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
+}
+
+/*
+ * Calculate the potential space needed by the log vector. Each region gets
+ * its own xlog_op_header_t and may need to be double word aligned.
+ */
+static int
+xlog_write_calc_vec_length(
+ struct xlog_ticket *ticket,
+ struct xfs_log_vec *log_vector)
+{
+ struct xfs_log_vec *lv;
+ int headers = 0;
+ int len = 0;
+ int i;
+
+ /* acct for start rec of xact */
+ if (ticket->t_flags & XLOG_TIC_INITED)
+ headers++;
+
+ for (lv = log_vector; lv; lv = lv->lv_next) {
+ headers += lv->lv_niovecs;
+
+ for (i = 0; i < lv->lv_niovecs; i++) {
+ struct xfs_log_iovec *vecp = &lv->lv_iovecp[i];
+
+ len += vecp->i_len;
+ xlog_tic_add_region(ticket, vecp->i_len, vecp->i_type);
+ }
+ }
+
+ ticket->t_res_num_ophdrs += headers;
+ len += headers * sizeof(struct xlog_op_header);
+
+ return len;
+}
+
+/*
+ * If first write for transaction, insert start record We can't be trying to
+ * commit if we are inited. We can't have any "partial_copy" if we are inited.
+ */
+static int
+xlog_write_start_rec(
+ struct xlog_op_header *ophdr,
+ struct xlog_ticket *ticket)
+{
+ if (!(ticket->t_flags & XLOG_TIC_INITED))
+ return 0;
+
+ ophdr->oh_tid = cpu_to_be32(ticket->t_tid);
+ ophdr->oh_clientid = ticket->t_clientid;
+ ophdr->oh_len = 0;
+ ophdr->oh_flags = XLOG_START_TRANS;
+ ophdr->oh_res2 = 0;
+
+ ticket->t_flags &= ~XLOG_TIC_INITED;
+
+ return sizeof(struct xlog_op_header);
+}
+
+static xlog_op_header_t *
+xlog_write_setup_ophdr(
+ struct log *log,
+ struct xlog_op_header *ophdr,
+ struct xlog_ticket *ticket,
+ uint flags)
+{
+ ophdr->oh_tid = cpu_to_be32(ticket->t_tid);
+ ophdr->oh_clientid = ticket->t_clientid;
+ ophdr->oh_res2 = 0;
+
+ /* are we copying a commit or unmount record? */
+ ophdr->oh_flags = flags;
+
+ /*
+ * We've seen logs corrupted with bad transaction client ids. This
+ * makes sure that XFS doesn't generate them on. Turn this into an EIO
+ * and shut down the filesystem.
+ */
+ switch (ophdr->oh_clientid) {
+ case XFS_TRANSACTION:
+ case XFS_VOLUME:
+ case XFS_LOG:
+ break;
+ default:
+ xfs_fs_cmn_err(CE_WARN, log->l_mp,
+ "Bad XFS transaction clientid 0x%x in ticket 0x%p",
+ ophdr->oh_clientid, ticket);
+ return NULL;
+ }
+
+ return ophdr;
+}
+
+/*
+ * Set up the parameters of the region copy into the log. This has
+ * to handle region write split across multiple log buffers - this
+ * state is kept external to this function so that this code can
+ * can be written in an obvious, self documenting manner.
+ */
+static int
+xlog_write_setup_copy(
+ struct xlog_ticket *ticket,
+ struct xlog_op_header *ophdr,
+ int space_available,
+ int space_required,
+ int *copy_off,
+ int *copy_len,
+ int *last_was_partial_copy,
+ int *bytes_consumed)
+{
+ int still_to_copy;
+
+ still_to_copy = space_required - *bytes_consumed;
+ *copy_off = *bytes_consumed;
+
+ if (still_to_copy <= space_available) {
+ /* write of region completes here */
+ *copy_len = still_to_copy;
+ ophdr->oh_len = cpu_to_be32(*copy_len);
+ if (*last_was_partial_copy)
+ ophdr->oh_flags |= (XLOG_END_TRANS|XLOG_WAS_CONT_TRANS);
+ *last_was_partial_copy = 0;
+ *bytes_consumed = 0;
+ return 0;
+ }
+
+ /* partial write of region, needs extra log op header reservation */
+ *copy_len = space_available;
+ ophdr->oh_len = cpu_to_be32(*copy_len);
+ ophdr->oh_flags |= XLOG_CONTINUE_TRANS;
+ if (*last_was_partial_copy)
+ ophdr->oh_flags |= XLOG_WAS_CONT_TRANS;
+ *bytes_consumed += *copy_len;
+ (*last_was_partial_copy)++;
+
+ /* account for new log op header */
+ ticket->t_curr_res -= sizeof(struct xlog_op_header);
+ ticket->t_res_num_ophdrs++;
+
+ return sizeof(struct xlog_op_header);
+}
+
+static int
+xlog_write_copy_finish(
+ struct log *log,
+ struct xlog_in_core *iclog,
+ uint flags,
+ int *record_cnt,
+ int *data_cnt,
+ int *partial_copy,
+ int *partial_copy_len,
+ int log_offset,
+ struct xlog_in_core **commit_iclog)
+{
+ if (*partial_copy) {
+ /*
+ * This iclog has already been marked WANT_SYNC by
+ * xlog_state_get_iclog_space.
+ */
+ xlog_state_finish_copy(log, iclog, *record_cnt, *data_cnt);
+ *record_cnt = 0;
+ *data_cnt = 0;
+ return xlog_state_release_iclog(log, iclog);
+ }
+
+ *partial_copy = 0;
+ *partial_copy_len = 0;
+
+ if (iclog->ic_size - log_offset <= sizeof(xlog_op_header_t)) {
+ /* no more space in this iclog - push it. */
+ xlog_state_finish_copy(log, iclog, *record_cnt, *data_cnt);
+ *record_cnt = 0;
+ *data_cnt = 0;
+
+ spin_lock(&log->l_icloglock);
+ xlog_state_want_sync(log, iclog);
+ spin_unlock(&log->l_icloglock);
+
+ if (!commit_iclog)
+ return xlog_state_release_iclog(log, iclog);
+ ASSERT(flags & XLOG_COMMIT_TRANS);
+ *commit_iclog = iclog;
+ }
+
+ return 0;
}
/*
@@ -1706,211 +1886,163 @@ xlog_print_tic_res(xfs_mount_t *mp, xlog_ticket_t *ticket)
* we don't update ic_offset until the end when we know exactly how many
* bytes have been written out.
*/
-STATIC int
-xlog_write(xfs_mount_t * mp,
- xfs_log_iovec_t reg[],
- int nentries,
- xfs_log_ticket_t tic,
- xfs_lsn_t *start_lsn,
- xlog_in_core_t **commit_iclog,
- uint flags)
-{
- xlog_t *log = mp->m_log;
- xlog_ticket_t *ticket = (xlog_ticket_t *)tic;
- xlog_in_core_t *iclog = NULL; /* ptr to current in-core log */
- xlog_op_header_t *logop_head; /* ptr to log operation header */
- __psint_t ptr; /* copy address into data region */
- int len; /* # xlog_write() bytes 2 still copy */
- int index; /* region index currently copying */
- int log_offset; /* offset (from 0) into data region */
- int start_rec_copy; /* # bytes to copy for start record */
- int partial_copy; /* did we split a region? */
- int partial_copy_len;/* # bytes copied if split region */
- int need_copy; /* # bytes need to memcpy this region */
- int copy_len; /* # bytes actually memcpy'ing */
- int copy_off; /* # bytes from entry start */
- int contwr; /* continued write of in-core log? */
- int error;
- int record_cnt = 0, data_cnt = 0;
-
- partial_copy_len = partial_copy = 0;
-
- /* Calculate potential maximum space. Each region gets its own
- * xlog_op_header_t and may need to be double word aligned.
- */
- len = 0;
- if (ticket->t_flags & XLOG_TIC_INITED) { /* acct for start rec of xact */
- len += sizeof(xlog_op_header_t);
- ticket->t_res_num_ophdrs++;
- }
-
- for (index = 0; index < nentries; index++) {
- len += sizeof(xlog_op_header_t); /* each region gets >= 1 */
- ticket->t_res_num_ophdrs++;
- len += reg[index].i_len;
- xlog_tic_add_region(ticket, reg[index].i_len, reg[index].i_type);
- }
- contwr = *start_lsn = 0;
+int
+xlog_write(
+ struct log *log,
+ struct xfs_log_vec *log_vector,
+ struct xlog_ticket *ticket,
+ xfs_lsn_t *start_lsn,
+ struct xlog_in_core **commit_iclog,
+ uint flags)
+{
+ struct xlog_in_core *iclog = NULL;
+ struct xfs_log_iovec *vecp;
+ struct xfs_log_vec *lv;
+ int len;
+ int index;
+ int partial_copy = 0;
+ int partial_copy_len = 0;
+ int contwr = 0;
+ int record_cnt = 0;
+ int data_cnt = 0;
+ int error;
+
+ *start_lsn = 0;
+
+ len = xlog_write_calc_vec_length(ticket, log_vector);
+ if (log->l_cilp) {
+ /*
+ * Region headers and bytes are already accounted for.
+ * We only need to take into account start records and
+ * split regions in this function.
+ */
+ if (ticket->t_flags & XLOG_TIC_INITED)
+ ticket->t_curr_res -= sizeof(xlog_op_header_t);
- if (ticket->t_curr_res < len) {
- xlog_print_tic_res(mp, ticket);
-#ifdef DEBUG
- xlog_panic(
- "xfs_log_write: reservation ran out. Need to up reservation");
-#else
- /* Customer configurable panic */
- xfs_cmn_err(XFS_PTAG_LOGRES, CE_ALERT, mp,
- "xfs_log_write: reservation ran out. Need to up reservation");
- /* If we did not panic, shutdown the filesystem */
- xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
-#endif
- } else
- ticket->t_curr_res -= len;
+ /*
+ * Commit record headers need to be accounted for. These
+ * come in as separate writes so are easy to detect.
+ */
+ if (flags & (XLOG_COMMIT_TRANS | XLOG_UNMOUNT_TRANS))
+ ticket->t_curr_res -= sizeof(xlog_op_header_t);
+ } else
+ ticket->t_curr_res -= len;
+
+ if (ticket->t_curr_res < 0)
+ xlog_print_tic_res(log->l_mp, ticket);
+
+ index = 0;
+ lv = log_vector;
+ vecp = lv->lv_iovecp;
+ while (lv && index < lv->lv_niovecs) {
+ void *ptr;
+ int log_offset;
+
+ error = xlog_state_get_iclog_space(log, len, &iclog, ticket,
+ &contwr, &log_offset);
+ if (error)
+ return error;
- for (index = 0; index < nentries; ) {
- if ((error = xlog_state_get_iclog_space(log, len, &iclog, ticket,
- &contwr, &log_offset)))
- return error;
+ ASSERT(log_offset <= iclog->ic_size - 1);
+ ptr = iclog->ic_datap + log_offset;
- ASSERT(log_offset <= iclog->ic_size - 1);
- ptr = (__psint_t) ((char *)iclog->ic_datap+log_offset);
+ /* start_lsn is the first lsn written to. That's all we need. */
+ if (!*start_lsn)
+ *start_lsn = be64_to_cpu(iclog->ic_header.h_lsn);
- /* start_lsn is the first lsn written to. That's all we need. */
- if (! *start_lsn)
- *start_lsn = be64_to_cpu(iclog->ic_header.h_lsn);
+ /*
+ * This loop writes out as many regions as can fit in the amount
+ * of space which was allocated by xlog_state_get_iclog_space().
+ */
+ while (lv && index < lv->lv_niovecs) {
+ struct xfs_log_iovec *reg = &vecp[index];
+ struct xlog_op_header *ophdr;
+ int start_rec_copy;
+ int copy_len;
+ int copy_off;
+
+ ASSERT(reg->i_len % sizeof(__int32_t) == 0);
+ ASSERT((unsigned long)ptr % sizeof(__int32_t) == 0);
+
+ start_rec_copy = xlog_write_start_rec(ptr, ticket);
+ if (start_rec_copy) {
+ record_cnt++;
+ xlog_write_adv_cnt(&ptr, &len, &log_offset,
+ start_rec_copy);
+ }
- /* This loop writes out as many regions as can fit in the amount
- * of space which was allocated by xlog_state_get_iclog_space().
- */
- while (index < nentries) {
- ASSERT(reg[index].i_len % sizeof(__int32_t) == 0);
- ASSERT((__psint_t)ptr % sizeof(__int32_t) == 0);
- start_rec_copy = 0;
-
- /* If first write for transaction, insert start record.
- * We can't be trying to commit if we are inited. We can't
- * have any "partial_copy" if we are inited.
- */
- if (ticket->t_flags & XLOG_TIC_INITED) {
- logop_head = (xlog_op_header_t *)ptr;
- logop_head->oh_tid = cpu_to_be32(ticket->t_tid);
- logop_head->oh_clientid = ticket->t_clientid;
- logop_head->oh_len = 0;
- logop_head->oh_flags = XLOG_START_TRANS;
- logop_head->oh_res2 = 0;
- ticket->t_flags &= ~XLOG_TIC_INITED; /* clear bit */
- record_cnt++;
-
- start_rec_copy = sizeof(xlog_op_header_t);
- xlog_write_adv_cnt(ptr, len, log_offset, start_rec_copy);
- }
+ ophdr = xlog_write_setup_ophdr(log, ptr, ticket, flags);
+ if (!ophdr)
+ return XFS_ERROR(EIO);
- /* Copy log operation header directly into data section */
- logop_head = (xlog_op_header_t *)ptr;
- logop_head->oh_tid = cpu_to_be32(ticket->t_tid);
- logop_head->oh_clientid = ticket->t_clientid;
- logop_head->oh_res2 = 0;
+ xlog_write_adv_cnt(&ptr, &len, &log_offset,
+ sizeof(struct xlog_op_header));
+
+ len += xlog_write_setup_copy(ticket, ophdr,
+ iclog->ic_size-log_offset,
+ reg->i_len,
+ &copy_off, &copy_len,
+ &partial_copy,
+ &partial_copy_len);
+ xlog_verify_dest_ptr(log, ptr);
+
+ /* copy region */
+ ASSERT(copy_len >= 0);
+ memcpy(ptr, reg->i_addr + copy_off, copy_len);
+ xlog_write_adv_cnt(&ptr, &len, &log_offset, copy_len);
+
+ copy_len += start_rec_copy + sizeof(xlog_op_header_t);
+ record_cnt++;
+ data_cnt += contwr ? copy_len : 0;
+
+ error = xlog_write_copy_finish(log, iclog, flags,
+ &record_cnt, &data_cnt,
+ &partial_copy,
+ &partial_copy_len,
+ log_offset,
+ commit_iclog);
+ if (error)
+ return error;
- /* header copied directly */
- xlog_write_adv_cnt(ptr, len, log_offset, sizeof(xlog_op_header_t));
+ /*
+ * if we had a partial copy, we need to get more iclog
+ * space but we don't want to increment the region
+ * index because there is still more is this region to
+ * write.
+ *
+ * If we completed writing this region, and we flushed
+ * the iclog (indicated by resetting of the record
+ * count), then we also need to get more log space. If
+ * this was the last record, though, we are done and
+ * can just return.
+ */
+ if (partial_copy)
+ break;
- /* are we copying a commit or unmount record? */
- logop_head->oh_flags = flags;
+ if (++index == lv->lv_niovecs) {
+ lv = lv->lv_next;
+ index = 0;
+ if (lv)
+ vecp = lv->lv_iovecp;
+ }
+ if (record_cnt == 0) {
+ if (!lv)
+ return 0;
+ break;
+ }
+ }
+ }
- /*
- * We've seen logs corrupted with bad transaction client
- * ids. This makes sure that XFS doesn't generate them on.
- * Turn this into an EIO and shut down the filesystem.
- */
- switch (logop_head->oh_clientid) {
- case XFS_TRANSACTION:
- case XFS_VOLUME:
- case XFS_LOG:
- break;
- default:
- xfs_fs_cmn_err(CE_WARN, mp,
- "Bad XFS transaction clientid 0x%x in ticket 0x%p",
- logop_head->oh_clientid, tic);
- return XFS_ERROR(EIO);
- }
+ ASSERT(len == 0);
- /* Partial write last time? => (partial_copy != 0)
- * need_copy is the amount we'd like to copy if everything could
- * fit in the current memcpy.
- */
- need_copy = reg[index].i_len - partial_copy_len;
-
- copy_off = partial_copy_len;
- if (need_copy <= iclog->ic_size - log_offset) { /*complete write */
- copy_len = need_copy;
- logop_head->oh_len = cpu_to_be32(copy_len);
- if (partial_copy)
- logop_head->oh_flags|= (XLOG_END_TRANS|XLOG_WAS_CONT_TRANS);
- partial_copy_len = partial_copy = 0;
- } else { /* partial write */
- copy_len = iclog->ic_size - log_offset;
- logop_head->oh_len = cpu_to_be32(copy_len);
- logop_head->oh_flags |= XLOG_CONTINUE_TRANS;
- if (partial_copy)
- logop_head->oh_flags |= XLOG_WAS_CONT_TRANS;
- partial_copy_len += copy_len;
- partial_copy++;
- len += sizeof(xlog_op_header_t); /* from splitting of region */
- /* account for new log op header */
- ticket->t_curr_res -= sizeof(xlog_op_header_t);
- ticket->t_res_num_ophdrs++;
- }
- xlog_verify_dest_ptr(log, ptr);
-
- /* copy region */
- ASSERT(copy_len >= 0);
- memcpy((xfs_caddr_t)ptr, reg[index].i_addr + copy_off, copy_len);
- xlog_write_adv_cnt(ptr, len, log_offset, copy_len);
-
- /* make copy_len total bytes copied, including headers */
- copy_len += start_rec_copy + sizeof(xlog_op_header_t);
- record_cnt++;
- data_cnt += contwr ? copy_len : 0;
- if (partial_copy) { /* copied partial region */
- /* already marked WANT_SYNC by xlog_state_get_iclog_space */
- xlog_state_finish_copy(log, iclog, record_cnt, data_cnt);
- record_cnt = data_cnt = 0;
- if ((error = xlog_state_release_iclog(log, iclog)))
- return error;
- break; /* don't increment index */
- } else { /* copied entire region */
- index++;
- partial_copy_len = partial_copy = 0;
-
- if (iclog->ic_size - log_offset <= sizeof(xlog_op_header_t)) {
- xlog_state_finish_copy(log, iclog, record_cnt, data_cnt);
- record_cnt = data_cnt = 0;
- spin_lock(&log->l_icloglock);
- xlog_state_want_sync(log, iclog);
- spin_unlock(&log->l_icloglock);
- if (commit_iclog) {
- ASSERT(flags & XLOG_COMMIT_TRANS);
- *commit_iclog = iclog;
- } else if ((error = xlog_state_release_iclog(log, iclog)))
- return error;
- if (index == nentries)
- return 0; /* we are done */
- else
- break;
- }
- } /* if (partial_copy) */
- } /* while (index < nentries) */
- } /* for (index = 0; index < nentries; ) */
- ASSERT(len == 0);
+ xlog_state_finish_copy(log, iclog, record_cnt, data_cnt);
+ if (!commit_iclog)
+ return xlog_state_release_iclog(log, iclog);
- xlog_state_finish_copy(log, iclog, record_cnt, data_cnt);
- if (commit_iclog) {
ASSERT(flags & XLOG_COMMIT_TRANS);
*commit_iclog = iclog;
return 0;
- }
- return xlog_state_release_iclog(log, iclog);
-} /* xlog_write */
+}
/*****************************************************************************
@@ -2854,7 +2986,6 @@ xlog_state_switch_iclogs(xlog_t *log,
log->l_iclog = iclog->ic_next;
} /* xlog_state_switch_iclogs */
-
/*
* Write out all data in the in-core log as of this exact moment in time.
*
@@ -2882,11 +3013,19 @@ xlog_state_switch_iclogs(xlog_t *log,
* b) when we return from flushing out this iclog, it is still
* not in the active nor dirty state.
*/
-STATIC int
-xlog_state_sync_all(xlog_t *log, uint flags, int *log_flushed)
+int
+_xfs_log_force(
+ struct xfs_mount *mp,
+ uint flags,
+ int *log_flushed)
{
- xlog_in_core_t *iclog;
- xfs_lsn_t lsn;
+ struct log *log = mp->m_log;
+ struct xlog_in_core *iclog;
+ xfs_lsn_t lsn;
+
+ XFS_STATS_INC(xs_log_force);
+
+ xlog_cil_push(log, 1);
spin_lock(&log->l_icloglock);
@@ -2932,7 +3071,9 @@ xlog_state_sync_all(xlog_t *log, uint flags, int *log_flushed)
if (xlog_state_release_iclog(log, iclog))
return XFS_ERROR(EIO);
- *log_flushed = 1;
+
+ if (log_flushed)
+ *log_flushed = 1;
spin_lock(&log->l_icloglock);
if (be64_to_cpu(iclog->ic_header.h_lsn) == lsn &&
iclog->ic_state != XLOG_STATE_DIRTY)
@@ -2976,19 +3117,37 @@ maybe_sleep:
*/
if (iclog->ic_state & XLOG_STATE_IOERROR)
return XFS_ERROR(EIO);
- *log_flushed = 1;
-
+ if (log_flushed)
+ *log_flushed = 1;
} else {
no_sleep:
spin_unlock(&log->l_icloglock);
}
return 0;
-} /* xlog_state_sync_all */
+}
+/*
+ * Wrapper for _xfs_log_force(), to be used when caller doesn't care
+ * about errors or whether the log was flushed or not. This is the normal
+ * interface to use when trying to unpin items or move the log forward.
+ */
+void
+xfs_log_force(
+ xfs_mount_t *mp,
+ uint flags)
+{
+ int error;
+
+ error = _xfs_log_force(mp, flags, NULL);
+ if (error) {
+ xfs_fs_cmn_err(CE_WARN, mp, "xfs_log_force: "
+ "error %d returned.", error);
+ }
+}
/*
- * Used by code which implements synchronous log forces.
+ * Force the in-core log to disk for a specific LSN.
*
* Find in-core log with lsn.
* If it is in the DIRTY state, just return.
@@ -2996,109 +3155,148 @@ no_sleep:
* state and go to sleep or return.
* If it is in any other state, go to sleep or return.
*
- * If filesystem activity goes to zero, the iclog will get flushed only by
- * bdflush().
+ * Synchronous forces are implemented with a signal variable. All callers
+ * to force a given lsn to disk will wait on a the sv attached to the
+ * specific in-core log. When given in-core log finally completes its
+ * write to disk, that thread will wake up all threads waiting on the
+ * sv.
*/
-STATIC int
-xlog_state_sync(xlog_t *log,
- xfs_lsn_t lsn,
- uint flags,
- int *log_flushed)
+int
+_xfs_log_force_lsn(
+ struct xfs_mount *mp,
+ xfs_lsn_t lsn,
+ uint flags,
+ int *log_flushed)
{
- xlog_in_core_t *iclog;
- int already_slept = 0;
+ struct log *log = mp->m_log;
+ struct xlog_in_core *iclog;
+ int already_slept = 0;
-try_again:
- spin_lock(&log->l_icloglock);
- iclog = log->l_iclog;
+ ASSERT(lsn != 0);
- if (iclog->ic_state & XLOG_STATE_IOERROR) {
- spin_unlock(&log->l_icloglock);
- return XFS_ERROR(EIO);
- }
+ XFS_STATS_INC(xs_log_force);
- do {
- if (be64_to_cpu(iclog->ic_header.h_lsn) != lsn) {
- iclog = iclog->ic_next;
- continue;
+ if (log->l_cilp) {
+ lsn = xlog_cil_push_lsn(log, lsn);
+ if (lsn == NULLCOMMITLSN)
+ return 0;
}
- if (iclog->ic_state == XLOG_STATE_DIRTY) {
+try_again:
+ spin_lock(&log->l_icloglock);
+ iclog = log->l_iclog;
+ if (iclog->ic_state & XLOG_STATE_IOERROR) {
spin_unlock(&log->l_icloglock);
- return 0;
+ return XFS_ERROR(EIO);
}
- if (iclog->ic_state == XLOG_STATE_ACTIVE) {
- /*
- * We sleep here if we haven't already slept (e.g.
- * this is the first time we've looked at the correct
- * iclog buf) and the buffer before us is going to
- * be sync'ed. The reason for this is that if we
- * are doing sync transactions here, by waiting for
- * the previous I/O to complete, we can allow a few
- * more transactions into this iclog before we close
- * it down.
- *
- * Otherwise, we mark the buffer WANT_SYNC, and bump
- * up the refcnt so we can release the log (which drops
- * the ref count). The state switch keeps new transaction
- * commits from using this buffer. When the current commits
- * finish writing into the buffer, the refcount will drop to
- * zero and the buffer will go out then.
- */
- if (!already_slept &&
- (iclog->ic_prev->ic_state & (XLOG_STATE_WANT_SYNC |
- XLOG_STATE_SYNCING))) {
- ASSERT(!(iclog->ic_state & XLOG_STATE_IOERROR));
- XFS_STATS_INC(xs_log_force_sleep);
- sv_wait(&iclog->ic_prev->ic_write_wait, PSWP,
- &log->l_icloglock, s);
- *log_flushed = 1;
- already_slept = 1;
- goto try_again;
- } else {
+ do {
+ if (be64_to_cpu(iclog->ic_header.h_lsn) != lsn) {
+ iclog = iclog->ic_next;
+ continue;
+ }
+
+ if (iclog->ic_state == XLOG_STATE_DIRTY) {
+ spin_unlock(&log->l_icloglock);
+ return 0;
+ }
+
+ if (iclog->ic_state == XLOG_STATE_ACTIVE) {
+ /*
+ * We sleep here if we haven't already slept (e.g.
+ * this is the first time we've looked at the correct
+ * iclog buf) and the buffer before us is going to
+ * be sync'ed. The reason for this is that if we
+ * are doing sync transactions here, by waiting for
+ * the previous I/O to complete, we can allow a few
+ * more transactions into this iclog before we close
+ * it down.
+ *
+ * Otherwise, we mark the buffer WANT_SYNC, and bump
+ * up the refcnt so we can release the log (which
+ * drops the ref count). The state switch keeps new
+ * transaction commits from using this buffer. When
+ * the current commits finish writing into the buffer,
+ * the refcount will drop to zero and the buffer will
+ * go out then.
+ */
+ if (!already_slept &&
+ (iclog->ic_prev->ic_state &
+ (XLOG_STATE_WANT_SYNC | XLOG_STATE_SYNCING))) {
+ ASSERT(!(iclog->ic_state & XLOG_STATE_IOERROR));
+
+ XFS_STATS_INC(xs_log_force_sleep);
+
+ sv_wait(&iclog->ic_prev->ic_write_wait,
+ PSWP, &log->l_icloglock, s);
+ if (log_flushed)
+ *log_flushed = 1;
+ already_slept = 1;
+ goto try_again;
+ }
atomic_inc(&iclog->ic_refcnt);
xlog_state_switch_iclogs(log, iclog, 0);
spin_unlock(&log->l_icloglock);
if (xlog_state_release_iclog(log, iclog))
return XFS_ERROR(EIO);
- *log_flushed = 1;
+ if (log_flushed)
+ *log_flushed = 1;
spin_lock(&log->l_icloglock);
}
- }
- if ((flags & XFS_LOG_SYNC) && /* sleep */
- !(iclog->ic_state & (XLOG_STATE_ACTIVE | XLOG_STATE_DIRTY))) {
+ if ((flags & XFS_LOG_SYNC) && /* sleep */
+ !(iclog->ic_state &
+ (XLOG_STATE_ACTIVE | XLOG_STATE_DIRTY))) {
+ /*
+ * Don't wait on completion if we know that we've
+ * gotten a log write error.
+ */
+ if (iclog->ic_state & XLOG_STATE_IOERROR) {
+ spin_unlock(&log->l_icloglock);
+ return XFS_ERROR(EIO);
+ }
+ XFS_STATS_INC(xs_log_force_sleep);
+ sv_wait(&iclog->ic_force_wait, PSWP, &log->l_icloglock, s);
+ /*
+ * No need to grab the log lock here since we're
+ * only deciding whether or not to return EIO
+ * and the memory read should be atomic.
+ */
+ if (iclog->ic_state & XLOG_STATE_IOERROR)
+ return XFS_ERROR(EIO);
- /*
- * Don't wait on completion if we know that we've
- * gotten a log write error.
- */
- if (iclog->ic_state & XLOG_STATE_IOERROR) {
+ if (log_flushed)
+ *log_flushed = 1;
+ } else { /* just return */
spin_unlock(&log->l_icloglock);
- return XFS_ERROR(EIO);
}
- XFS_STATS_INC(xs_log_force_sleep);
- sv_wait(&iclog->ic_force_wait, PSWP, &log->l_icloglock, s);
- /*
- * No need to grab the log lock here since we're
- * only deciding whether or not to return EIO
- * and the memory read should be atomic.
- */
- if (iclog->ic_state & XLOG_STATE_IOERROR)
- return XFS_ERROR(EIO);
- *log_flushed = 1;
- } else { /* just return */
- spin_unlock(&log->l_icloglock);
- }
- return 0;
- } while (iclog != log->l_iclog);
+ return 0;
+ } while (iclog != log->l_iclog);
- spin_unlock(&log->l_icloglock);
- return 0;
-} /* xlog_state_sync */
+ spin_unlock(&log->l_icloglock);
+ return 0;
+}
+/*
+ * Wrapper for _xfs_log_force_lsn(), to be used when caller doesn't care
+ * about errors or whether the log was flushed or not. This is the normal
+ * interface to use when trying to unpin items or move the log forward.
+ */
+void
+xfs_log_force_lsn(
+ xfs_mount_t *mp,
+ xfs_lsn_t lsn,
+ uint flags)
+{
+ int error;
+
+ error = _xfs_log_force_lsn(mp, lsn, flags, NULL);
+ if (error) {
+ xfs_fs_cmn_err(CE_WARN, mp, "xfs_log_force: "
+ "error %d returned.", error);
+ }
+}
/*
* Called when we want to mark the current iclog as being ready to sync to
@@ -3148,20 +3346,30 @@ xfs_log_ticket_get(
return ticket;
}
+xlog_tid_t
+xfs_log_get_trans_ident(
+ struct xfs_trans *tp)
+{
+ return tp->t_ticket->t_tid;
+}
+
/*
* Allocate and initialise a new log ticket.
*/
-STATIC xlog_ticket_t *
-xlog_ticket_alloc(xlog_t *log,
- int unit_bytes,
- int cnt,
- char client,
- uint xflags)
+xlog_ticket_t *
+xlog_ticket_alloc(
+ struct log *log,
+ int unit_bytes,
+ int cnt,
+ char client,
+ uint xflags,
+ int alloc_flags)
{
- xlog_ticket_t *tic;
+ struct xlog_ticket *tic;
uint num_headers;
+ int iclog_space;
- tic = kmem_zone_zalloc(xfs_log_ticket_zone, KM_SLEEP|KM_MAYFAIL);
+ tic = kmem_zone_zalloc(xfs_log_ticket_zone, alloc_flags);
if (!tic)
return NULL;
@@ -3203,16 +3411,40 @@ xlog_ticket_alloc(xlog_t *log,
/* for start-rec */
unit_bytes += sizeof(xlog_op_header_t);
- /* for LR headers */
- num_headers = ((unit_bytes + log->l_iclog_size-1) >> log->l_iclog_size_log);
+ /*
+ * for LR headers - the space for data in an iclog is the size minus
+ * the space used for the headers. If we use the iclog size, then we
+ * undercalculate the number of headers required.
+ *
+ * Furthermore - the addition of op headers for split-recs might
+ * increase the space required enough to require more log and op
+ * headers, so take that into account too.
+ *
+ * IMPORTANT: This reservation makes the assumption that if this
+ * transaction is the first in an iclog and hence has the LR headers
+ * accounted to it, then the remaining space in the iclog is
+ * exclusively for this transaction. i.e. if the transaction is larger
+ * than the iclog, it will be the only thing in that iclog.
+ * Fundamentally, this means we must pass the entire log vector to
+ * xlog_write to guarantee this.
+ */
+ iclog_space = log->l_iclog_size - log->l_iclog_hsize;
+ num_headers = howmany(unit_bytes, iclog_space);
+
+ /* for split-recs - ophdrs added when data split over LRs */
+ unit_bytes += sizeof(xlog_op_header_t) * num_headers;
+
+ /* add extra header reservations if we overrun */
+ while (!num_headers ||
+ howmany(unit_bytes, iclog_space) > num_headers) {
+ unit_bytes += sizeof(xlog_op_header_t);
+ num_headers++;
+ }
unit_bytes += log->l_iclog_hsize * num_headers;
/* for commit-rec LR header - note: padding will subsume the ophdr */
unit_bytes += log->l_iclog_hsize;
- /* for split-recs - ophdrs added when data split over LRs */
- unit_bytes += sizeof(xlog_op_header_t) * num_headers;
-
/* for roundoff padding for transaction data and one for commit record */
if (xfs_sb_version_haslogv2(&log->l_mp->m_sb) &&
log->l_mp->m_sb.sb_logsunit > 1) {
@@ -3228,13 +3460,13 @@ xlog_ticket_alloc(xlog_t *log,
tic->t_curr_res = unit_bytes;
tic->t_cnt = cnt;
tic->t_ocnt = cnt;
- tic->t_tid = (xlog_tid_t)((__psint_t)tic & 0xffffffff);
+ tic->t_tid = random32();
tic->t_clientid = client;
tic->t_flags = XLOG_TIC_INITED;
tic->t_trans_type = 0;
if (xflags & XFS_LOG_PERM_RESERV)
tic->t_flags |= XLOG_TIC_PERM_RESERV;
- sv_init(&(tic->t_wait), SV_DEFAULT, "logtick");
+ sv_init(&tic->t_wait, SV_DEFAULT, "logtick");
xlog_tic_reset_res(tic);
@@ -3255,20 +3487,22 @@ xlog_ticket_alloc(xlog_t *log,
* part of the log in case we trash the log structure.
*/
void
-xlog_verify_dest_ptr(xlog_t *log,
- __psint_t ptr)
+xlog_verify_dest_ptr(
+ struct log *log,
+ char *ptr)
{
int i;
int good_ptr = 0;
- for (i=0; i < log->l_iclog_bufs; i++) {
- if (ptr >= (__psint_t)log->l_iclog_bak[i] &&
- ptr <= (__psint_t)log->l_iclog_bak[i]+log->l_iclog_size)
+ for (i = 0; i < log->l_iclog_bufs; i++) {
+ if (ptr >= log->l_iclog_bak[i] &&
+ ptr <= log->l_iclog_bak[i] + log->l_iclog_size)
good_ptr++;
}
- if (! good_ptr)
+
+ if (!good_ptr)
xlog_panic("xlog_verify_dest_ptr: invalid ptr");
-} /* xlog_verify_dest_ptr */
+}
STATIC void
xlog_verify_grant_head(xlog_t *log, int equals)
@@ -3454,6 +3688,11 @@ xlog_state_ioerror(
* c. nothing new gets queued up after (a) and (b) are done.
* d. if !logerror, flush the iclogs to disk, then seal them off
* for business.
+ *
+ * Note: for delayed logging the !logerror case needs to flush the regions
+ * held in memory out to the iclogs before flushing them to disk. This needs
+ * to be done before the log is marked as shutdown, otherwise the flush to the
+ * iclogs will fail.
*/
int
xfs_log_force_umount(
@@ -3463,7 +3702,6 @@ xfs_log_force_umount(
xlog_ticket_t *tic;
xlog_t *log;
int retval;
- int dummy;
log = mp->m_log;
@@ -3488,6 +3726,16 @@ xfs_log_force_umount(
return 1;
}
retval = 0;
+
+ /*
+ * Flush the in memory commit item list before marking the log as
+ * being shut down. We need to do it in this order to ensure all the
+ * completed transactions are flushed to disk with the xfs_log_force()
+ * call below.
+ */
+ if (!logerror && (mp->m_flags & XFS_MOUNT_DELAYLOG))
+ xlog_cil_push(log, 1);
+
/*
* We must hold both the GRANT lock and the LOG lock,
* before we mark the filesystem SHUTDOWN and wake
@@ -3537,13 +3785,14 @@ xfs_log_force_umount(
}
spin_unlock(&log->l_grant_lock);
- if (! (log->l_iclog->ic_state & XLOG_STATE_IOERROR)) {
+ if (!(log->l_iclog->ic_state & XLOG_STATE_IOERROR)) {
ASSERT(!logerror);
/*
* Force the incore logs to disk before shutting the
* log down completely.
*/
- xlog_state_sync_all(log, XFS_LOG_FORCE|XFS_LOG_SYNC, &dummy);
+ _xfs_log_force(mp, XFS_LOG_SYNC, NULL);
+
spin_lock(&log->l_icloglock);
retval = xlog_state_ioerror(log);
spin_unlock(&log->l_icloglock);
diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h
index d0c9baa50b1a..04c78e642cc8 100644
--- a/fs/xfs/xfs_log.h
+++ b/fs/xfs/xfs_log.h
@@ -19,7 +19,6 @@
#define __XFS_LOG_H__
/* get lsn fields */
-
#define CYCLE_LSN(lsn) ((uint)((lsn)>>32))
#define BLOCK_LSN(lsn) ((uint)(lsn))
@@ -70,14 +69,8 @@ static inline xfs_lsn_t _lsn_cmp(xfs_lsn_t lsn1, xfs_lsn_t lsn2)
* Flags to xfs_log_force()
*
* XFS_LOG_SYNC: Synchronous force in-core log to disk
- * XFS_LOG_FORCE: Start in-core log write now.
- * XFS_LOG_URGE: Start write within some window of time.
- *
- * Note: Either XFS_LOG_FORCE or XFS_LOG_URGE must be set.
*/
#define XFS_LOG_SYNC 0x1
-#define XFS_LOG_FORCE 0x2
-#define XFS_LOG_URGE 0x4
#endif /* __KERNEL__ */
@@ -110,15 +103,20 @@ static inline xfs_lsn_t _lsn_cmp(xfs_lsn_t lsn1, xfs_lsn_t lsn2)
#define XLOG_REG_TYPE_TRANSHDR 19
#define XLOG_REG_TYPE_MAX 19
-#define XLOG_VEC_SET_TYPE(vecp, t) ((vecp)->i_type = (t))
-
typedef struct xfs_log_iovec {
- xfs_caddr_t i_addr; /* beginning address of region */
+ xfs_caddr_t i_addr; /* beginning address of region */
int i_len; /* length in bytes of region */
uint i_type; /* type of region */
} xfs_log_iovec_t;
-typedef void* xfs_log_ticket_t;
+struct xfs_log_vec {
+ struct xfs_log_vec *lv_next; /* next lv in build list */
+ int lv_niovecs; /* number of iovecs in lv */
+ struct xfs_log_iovec *lv_iovecp; /* iovec array */
+ struct xfs_log_item *lv_item; /* owner */
+ char *lv_buf; /* formatted buffer */
+ int lv_buf_len; /* size of formatted buffer */
+};
/*
* Structure used to pass callback function and the function's argument
@@ -134,18 +132,33 @@ typedef struct xfs_log_callback {
#ifdef __KERNEL__
/* Log manager interfaces */
struct xfs_mount;
+struct xlog_in_core;
struct xlog_ticket;
+struct xfs_log_item;
+struct xfs_item_ops;
+struct xfs_trans;
+
+void xfs_log_item_init(struct xfs_mount *mp,
+ struct xfs_log_item *item,
+ int type,
+ struct xfs_item_ops *ops);
+
xfs_lsn_t xfs_log_done(struct xfs_mount *mp,
- xfs_log_ticket_t ticket,
- void **iclog,
+ struct xlog_ticket *ticket,
+ struct xlog_in_core **iclog,
uint flags);
int _xfs_log_force(struct xfs_mount *mp,
- xfs_lsn_t lsn,
uint flags,
int *log_forced);
void xfs_log_force(struct xfs_mount *mp,
- xfs_lsn_t lsn,
uint flags);
+int _xfs_log_force_lsn(struct xfs_mount *mp,
+ xfs_lsn_t lsn,
+ uint flags,
+ int *log_forced);
+void xfs_log_force_lsn(struct xfs_mount *mp,
+ xfs_lsn_t lsn,
+ uint flags);
int xfs_log_mount(struct xfs_mount *mp,
struct xfs_buftarg *log_target,
xfs_daddr_t start_block,
@@ -154,21 +167,21 @@ int xfs_log_mount_finish(struct xfs_mount *mp);
void xfs_log_move_tail(struct xfs_mount *mp,
xfs_lsn_t tail_lsn);
int xfs_log_notify(struct xfs_mount *mp,
- void *iclog,
+ struct xlog_in_core *iclog,
xfs_log_callback_t *callback_entry);
int xfs_log_release_iclog(struct xfs_mount *mp,
- void *iclog_hndl);
+ struct xlog_in_core *iclog);
int xfs_log_reserve(struct xfs_mount *mp,
int length,
int count,
- xfs_log_ticket_t *ticket,
+ struct xlog_ticket **ticket,
__uint8_t clientid,
uint flags,
uint t_type);
int xfs_log_write(struct xfs_mount *mp,
xfs_log_iovec_t region[],
int nentries,
- xfs_log_ticket_t ticket,
+ struct xlog_ticket *ticket,
xfs_lsn_t *start_lsn);
int xfs_log_unmount_write(struct xfs_mount *mp);
void xfs_log_unmount(struct xfs_mount *mp);
@@ -177,9 +190,16 @@ int xfs_log_need_covered(struct xfs_mount *mp);
void xlog_iodone(struct xfs_buf *);
-struct xlog_ticket * xfs_log_ticket_get(struct xlog_ticket *ticket);
+struct xlog_ticket *xfs_log_ticket_get(struct xlog_ticket *ticket);
void xfs_log_ticket_put(struct xlog_ticket *ticket);
+xlog_tid_t xfs_log_get_trans_ident(struct xfs_trans *tp);
+
+int xfs_log_commit_cil(struct xfs_mount *mp, struct xfs_trans *tp,
+ struct xfs_log_vec *log_vector,
+ xfs_lsn_t *commit_lsn, int flags);
+bool xfs_log_item_in_current_chkpt(struct xfs_log_item *lip);
+
#endif
diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c
new file mode 100644
index 000000000000..bb17cc044bf3
--- /dev/null
+++ b/fs/xfs/xfs_log_cil.c
@@ -0,0 +1,725 @@
+/*
+ * Copyright (c) 2010 Red Hat, 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_trans_priv.h"
+#include "xfs_log_priv.h"
+#include "xfs_sb.h"
+#include "xfs_ag.h"
+#include "xfs_dir2.h"
+#include "xfs_dmapi.h"
+#include "xfs_mount.h"
+#include "xfs_error.h"
+#include "xfs_alloc.h"
+
+/*
+ * Perform initial CIL structure initialisation. If the CIL is not
+ * enabled in this filesystem, ensure the log->l_cilp is null so
+ * we can check this conditional to determine if we are doing delayed
+ * logging or not.
+ */
+int
+xlog_cil_init(
+ struct log *log)
+{
+ struct xfs_cil *cil;
+ struct xfs_cil_ctx *ctx;
+
+ log->l_cilp = NULL;
+ if (!(log->l_mp->m_flags & XFS_MOUNT_DELAYLOG))
+ return 0;
+
+ cil = kmem_zalloc(sizeof(*cil), KM_SLEEP|KM_MAYFAIL);
+ if (!cil)
+ return ENOMEM;
+
+ ctx = kmem_zalloc(sizeof(*ctx), KM_SLEEP|KM_MAYFAIL);
+ if (!ctx) {
+ kmem_free(cil);
+ return ENOMEM;
+ }
+
+ INIT_LIST_HEAD(&cil->xc_cil);
+ INIT_LIST_HEAD(&cil->xc_committing);
+ spin_lock_init(&cil->xc_cil_lock);
+ init_rwsem(&cil->xc_ctx_lock);
+ sv_init(&cil->xc_commit_wait, SV_DEFAULT, "cilwait");
+
+ INIT_LIST_HEAD(&ctx->committing);
+ INIT_LIST_HEAD(&ctx->busy_extents);
+ ctx->sequence = 1;
+ ctx->cil = cil;
+ cil->xc_ctx = ctx;
+
+ cil->xc_log = log;
+ log->l_cilp = cil;
+ return 0;
+}
+
+void
+xlog_cil_destroy(
+ struct log *log)
+{
+ if (!log->l_cilp)
+ return;
+
+ if (log->l_cilp->xc_ctx) {
+ if (log->l_cilp->xc_ctx->ticket)
+ xfs_log_ticket_put(log->l_cilp->xc_ctx->ticket);
+ kmem_free(log->l_cilp->xc_ctx);
+ }
+
+ ASSERT(list_empty(&log->l_cilp->xc_cil));
+ kmem_free(log->l_cilp);
+}
+
+/*
+ * Allocate a new ticket. Failing to get a new ticket makes it really hard to
+ * recover, so we don't allow failure here. Also, we allocate in a context that
+ * we don't want to be issuing transactions from, so we need to tell the
+ * allocation code this as well.
+ *
+ * We don't reserve any space for the ticket - we are going to steal whatever
+ * space we require from transactions as they commit. To ensure we reserve all
+ * the space required, we need to set the current reservation of the ticket to
+ * zero so that we know to steal the initial transaction overhead from the
+ * first transaction commit.
+ */
+static struct xlog_ticket *
+xlog_cil_ticket_alloc(
+ struct log *log)
+{
+ struct xlog_ticket *tic;
+
+ tic = xlog_ticket_alloc(log, 0, 1, XFS_TRANSACTION, 0,
+ KM_SLEEP|KM_NOFS);
+ tic->t_trans_type = XFS_TRANS_CHECKPOINT;
+
+ /*
+ * set the current reservation to zero so we know to steal the basic
+ * transaction overhead reservation from the first transaction commit.
+ */
+ tic->t_curr_res = 0;
+ return tic;
+}
+
+/*
+ * After the first stage of log recovery is done, we know where the head and
+ * tail of the log are. We need this log initialisation done before we can
+ * initialise the first CIL checkpoint context.
+ *
+ * Here we allocate a log ticket to track space usage during a CIL push. This
+ * ticket is passed to xlog_write() directly so that we don't slowly leak log
+ * space by failing to account for space used by log headers and additional
+ * region headers for split regions.
+ */
+void
+xlog_cil_init_post_recovery(
+ struct log *log)
+{
+ if (!log->l_cilp)
+ return;
+
+ log->l_cilp->xc_ctx->ticket = xlog_cil_ticket_alloc(log);
+ log->l_cilp->xc_ctx->sequence = 1;
+ log->l_cilp->xc_ctx->commit_lsn = xlog_assign_lsn(log->l_curr_cycle,
+ log->l_curr_block);
+}
+
+/*
+ * Insert the log item into the CIL and calculate the difference in space
+ * consumed by the item. Add the space to the checkpoint ticket and calculate
+ * if the change requires additional log metadata. If it does, take that space
+ * as well. Remove the amount of space we addded to the checkpoint ticket from
+ * the current transaction ticket so that the accounting works out correctly.
+ *
+ * If this is the first time the item is being placed into the CIL in this
+ * context, pin it so it can't be written to disk until the CIL is flushed to
+ * the iclog and the iclog written to disk.
+ */
+static void
+xlog_cil_insert(
+ struct log *log,
+ struct xlog_ticket *ticket,
+ struct xfs_log_item *item,
+ struct xfs_log_vec *lv)
+{
+ struct xfs_cil *cil = log->l_cilp;
+ struct xfs_log_vec *old = lv->lv_item->li_lv;
+ struct xfs_cil_ctx *ctx = cil->xc_ctx;
+ int len;
+ int diff_iovecs;
+ int iclog_space;
+
+ if (old) {
+ /* existing lv on log item, space used is a delta */
+ ASSERT(!list_empty(&item->li_cil));
+ ASSERT(old->lv_buf && old->lv_buf_len && old->lv_niovecs);
+
+ len = lv->lv_buf_len - old->lv_buf_len;
+ diff_iovecs = lv->lv_niovecs - old->lv_niovecs;
+ kmem_free(old->lv_buf);
+ kmem_free(old);
+ } else {
+ /* new lv, must pin the log item */
+ ASSERT(!lv->lv_item->li_lv);
+ ASSERT(list_empty(&item->li_cil));
+
+ len = lv->lv_buf_len;
+ diff_iovecs = lv->lv_niovecs;
+ IOP_PIN(lv->lv_item);
+
+ }
+ len += diff_iovecs * sizeof(xlog_op_header_t);
+
+ /* attach new log vector to log item */
+ lv->lv_item->li_lv = lv;
+
+ spin_lock(&cil->xc_cil_lock);
+ list_move_tail(&item->li_cil, &cil->xc_cil);
+ ctx->nvecs += diff_iovecs;
+
+ /*
+ * If this is the first time the item is being committed to the CIL,
+ * store the sequence number on the log item so we can tell
+ * in future commits whether this is the first checkpoint the item is
+ * being committed into.
+ */
+ if (!item->li_seq)
+ item->li_seq = ctx->sequence;
+
+ /*
+ * Now transfer enough transaction reservation to the context ticket
+ * for the checkpoint. The context ticket is special - the unit
+ * reservation has to grow as well as the current reservation as we
+ * steal from tickets so we can correctly determine the space used
+ * during the transaction commit.
+ */
+ if (ctx->ticket->t_curr_res == 0) {
+ /* first commit in checkpoint, steal the header reservation */
+ ASSERT(ticket->t_curr_res >= ctx->ticket->t_unit_res + len);
+ ctx->ticket->t_curr_res = ctx->ticket->t_unit_res;
+ ticket->t_curr_res -= ctx->ticket->t_unit_res;
+ }
+
+ /* do we need space for more log record headers? */
+ iclog_space = log->l_iclog_size - log->l_iclog_hsize;
+ if (len > 0 && (ctx->space_used / iclog_space !=
+ (ctx->space_used + len) / iclog_space)) {
+ int hdrs;
+
+ hdrs = (len + iclog_space - 1) / iclog_space;
+ /* need to take into account split region headers, too */
+ hdrs *= log->l_iclog_hsize + sizeof(struct xlog_op_header);
+ ctx->ticket->t_unit_res += hdrs;
+ ctx->ticket->t_curr_res += hdrs;
+ ticket->t_curr_res -= hdrs;
+ ASSERT(ticket->t_curr_res >= len);
+ }
+ ticket->t_curr_res -= len;
+ ctx->space_used += len;
+
+ spin_unlock(&cil->xc_cil_lock);
+}
+
+/*
+ * Format log item into a flat buffers
+ *
+ * For delayed logging, we need to hold a formatted buffer containing all the
+ * changes on the log item. This enables us to relog the item in memory and
+ * write it out asynchronously without needing to relock the object that was
+ * modified at the time it gets written into the iclog.
+ *
+ * This function builds a vector for the changes in each log item in the
+ * transaction. It then works out the length of the buffer needed for each log
+ * item, allocates them and formats the vector for the item into the buffer.
+ * The buffer is then attached to the log item are then inserted into the
+ * Committed Item List for tracking until the next checkpoint is written out.
+ *
+ * We don't set up region headers during this process; we simply copy the
+ * regions into the flat buffer. We can do this because we still have to do a
+ * formatting step to write the regions into the iclog buffer. Writing the
+ * ophdrs during the iclog write means that we can support splitting large
+ * regions across iclog boundares without needing a change in the format of the
+ * item/region encapsulation.
+ *
+ * Hence what we need to do now is change the rewrite the vector array to point
+ * to the copied region inside the buffer we just allocated. This allows us to
+ * format the regions into the iclog as though they are being formatted
+ * directly out of the objects themselves.
+ */
+static void
+xlog_cil_format_items(
+ struct log *log,
+ struct xfs_log_vec *log_vector,
+ struct xlog_ticket *ticket,
+ xfs_lsn_t *start_lsn)
+{
+ struct xfs_log_vec *lv;
+
+ if (start_lsn)
+ *start_lsn = log->l_cilp->xc_ctx->sequence;
+
+ ASSERT(log_vector);
+ for (lv = log_vector; lv; lv = lv->lv_next) {
+ void *ptr;
+ int index;
+ int len = 0;
+
+ /* build the vector array and calculate it's length */
+ IOP_FORMAT(lv->lv_item, lv->lv_iovecp);
+ for (index = 0; index < lv->lv_niovecs; index++)
+ len += lv->lv_iovecp[index].i_len;
+
+ lv->lv_buf_len = len;
+ lv->lv_buf = kmem_zalloc(lv->lv_buf_len, KM_SLEEP|KM_NOFS);
+ ptr = lv->lv_buf;
+
+ for (index = 0; index < lv->lv_niovecs; index++) {
+ struct xfs_log_iovec *vec = &lv->lv_iovecp[index];
+
+ memcpy(ptr, vec->i_addr, vec->i_len);
+ vec->i_addr = ptr;
+ ptr += vec->i_len;
+ }
+ ASSERT(ptr == lv->lv_buf + lv->lv_buf_len);
+
+ xlog_cil_insert(log, ticket, lv->lv_item, lv);
+ }
+}
+
+static void
+xlog_cil_free_logvec(
+ struct xfs_log_vec *log_vector)
+{
+ struct xfs_log_vec *lv;
+
+ for (lv = log_vector; lv; ) {
+ struct xfs_log_vec *next = lv->lv_next;
+ kmem_free(lv->lv_buf);
+ kmem_free(lv);
+ lv = next;
+ }
+}
+
+/*
+ * Commit a transaction with the given vector to the Committed Item List.
+ *
+ * To do this, we need to format the item, pin it in memory if required and
+ * account for the space used by the transaction. Once we have done that we
+ * need to release the unused reservation for the transaction, attach the
+ * transaction to the checkpoint context so we carry the busy extents through
+ * to checkpoint completion, and then unlock all the items in the transaction.
+ *
+ * For more specific information about the order of operations in
+ * xfs_log_commit_cil() please refer to the comments in
+ * xfs_trans_commit_iclog().
+ *
+ * Called with the context lock already held in read mode to lock out
+ * background commit, returns without it held once background commits are
+ * allowed again.
+ */
+int
+xfs_log_commit_cil(
+ struct xfs_mount *mp,
+ struct xfs_trans *tp,
+ struct xfs_log_vec *log_vector,
+ xfs_lsn_t *commit_lsn,
+ int flags)
+{
+ struct log *log = mp->m_log;
+ int log_flags = 0;
+ int push = 0;
+
+ if (flags & XFS_TRANS_RELEASE_LOG_RES)
+ log_flags = XFS_LOG_REL_PERM_RESERV;
+
+ if (XLOG_FORCED_SHUTDOWN(log)) {
+ xlog_cil_free_logvec(log_vector);
+ return XFS_ERROR(EIO);
+ }
+
+ /* lock out background commit */
+ down_read(&log->l_cilp->xc_ctx_lock);
+ xlog_cil_format_items(log, log_vector, tp->t_ticket, commit_lsn);
+
+ /* check we didn't blow the reservation */
+ if (tp->t_ticket->t_curr_res < 0)
+ xlog_print_tic_res(log->l_mp, tp->t_ticket);
+
+ /* attach the transaction to the CIL if it has any busy extents */
+ if (!list_empty(&tp->t_busy)) {
+ spin_lock(&log->l_cilp->xc_cil_lock);
+ list_splice_init(&tp->t_busy,
+ &log->l_cilp->xc_ctx->busy_extents);
+ spin_unlock(&log->l_cilp->xc_cil_lock);
+ }
+
+ tp->t_commit_lsn = *commit_lsn;
+ xfs_log_done(mp, tp->t_ticket, NULL, log_flags);
+ xfs_trans_unreserve_and_mod_sb(tp);
+
+ /* check for background commit before unlock */
+ if (log->l_cilp->xc_ctx->space_used > XLOG_CIL_SPACE_LIMIT(log))
+ push = 1;
+ up_read(&log->l_cilp->xc_ctx_lock);
+
+ /*
+ * We need to push CIL every so often so we don't cache more than we
+ * can fit in the log. The limit really is that a checkpoint can't be
+ * more than half the log (the current checkpoint is not allowed to
+ * overwrite the previous checkpoint), but commit latency and memory
+ * usage limit this to a smaller size in most cases.
+ */
+ if (push)
+ xlog_cil_push(log, 0);
+ return 0;
+}
+
+/*
+ * Mark all items committed and clear busy extents. We free the log vector
+ * chains in a separate pass so that we unpin the log items as quickly as
+ * possible.
+ */
+static void
+xlog_cil_committed(
+ void *args,
+ int abort)
+{
+ struct xfs_cil_ctx *ctx = args;
+ struct xfs_log_vec *lv;
+ int abortflag = abort ? XFS_LI_ABORTED : 0;
+ struct xfs_busy_extent *busyp, *n;
+
+ /* unpin all the log items */
+ for (lv = ctx->lv_chain; lv; lv = lv->lv_next ) {
+ xfs_trans_item_committed(lv->lv_item, ctx->start_lsn,
+ abortflag);
+ }
+
+ list_for_each_entry_safe(busyp, n, &ctx->busy_extents, list)
+ xfs_alloc_busy_clear(ctx->cil->xc_log->l_mp, busyp);
+
+ spin_lock(&ctx->cil->xc_cil_lock);
+ list_del(&ctx->committing);
+ spin_unlock(&ctx->cil->xc_cil_lock);
+
+ xlog_cil_free_logvec(ctx->lv_chain);
+ kmem_free(ctx);
+}
+
+/*
+ * Push the Committed Item List to the log. If the push_now flag is not set,
+ * then it is a background flush and so we can chose to ignore it.
+ */
+int
+xlog_cil_push(
+ struct log *log,
+ int push_now)
+{
+ struct xfs_cil *cil = log->l_cilp;
+ struct xfs_log_vec *lv;
+ struct xfs_cil_ctx *ctx;
+ struct xfs_cil_ctx *new_ctx;
+ struct xlog_in_core *commit_iclog;
+ struct xlog_ticket *tic;
+ int num_lv;
+ int num_iovecs;
+ int len;
+ int error = 0;
+ struct xfs_trans_header thdr;
+ struct xfs_log_iovec lhdr;
+ struct xfs_log_vec lvhdr = { NULL };
+ xfs_lsn_t commit_lsn;
+
+ if (!cil)
+ return 0;
+
+ new_ctx = kmem_zalloc(sizeof(*new_ctx), KM_SLEEP|KM_NOFS);
+ new_ctx->ticket = xlog_cil_ticket_alloc(log);
+
+ /* lock out transaction commit, but don't block on background push */
+ if (!down_write_trylock(&cil->xc_ctx_lock)) {
+ if (!push_now)
+ goto out_free_ticket;
+ down_write(&cil->xc_ctx_lock);
+ }
+ ctx = cil->xc_ctx;
+
+ /* check if we've anything to push */
+ if (list_empty(&cil->xc_cil))
+ goto out_skip;
+
+ /* check for spurious background flush */
+ if (!push_now && cil->xc_ctx->space_used < XLOG_CIL_SPACE_LIMIT(log))
+ goto out_skip;
+
+ /*
+ * pull all the log vectors off the items in the CIL, and
+ * remove the items from the CIL. We don't need the CIL lock
+ * here because it's only needed on the transaction commit
+ * side which is currently locked out by the flush lock.
+ */
+ lv = NULL;
+ num_lv = 0;
+ num_iovecs = 0;
+ len = 0;
+ while (!list_empty(&cil->xc_cil)) {
+ struct xfs_log_item *item;
+ int i;
+
+ item = list_first_entry(&cil->xc_cil,
+ struct xfs_log_item, li_cil);
+ list_del_init(&item->li_cil);
+ if (!ctx->lv_chain)
+ ctx->lv_chain = item->li_lv;
+ else
+ lv->lv_next = item->li_lv;
+ lv = item->li_lv;
+ item->li_lv = NULL;
+
+ num_lv++;
+ num_iovecs += lv->lv_niovecs;
+ for (i = 0; i < lv->lv_niovecs; i++)
+ len += lv->lv_iovecp[i].i_len;
+ }
+
+ /*
+ * initialise the new context and attach it to the CIL. Then attach
+ * the current context to the CIL committing lsit so it can be found
+ * during log forces to extract the commit lsn of the sequence that
+ * needs to be forced.
+ */
+ INIT_LIST_HEAD(&new_ctx->committing);
+ INIT_LIST_HEAD(&new_ctx->busy_extents);
+ new_ctx->sequence = ctx->sequence + 1;
+ new_ctx->cil = cil;
+ cil->xc_ctx = new_ctx;
+
+ /*
+ * The switch is now done, so we can drop the context lock and move out
+ * of a shared context. We can't just go straight to the commit record,
+ * though - we need to synchronise with previous and future commits so
+ * that the commit records are correctly ordered in the log to ensure
+ * that we process items during log IO completion in the correct order.
+ *
+ * For example, if we get an EFI in one checkpoint and the EFD in the
+ * next (e.g. due to log forces), we do not want the checkpoint with
+ * the EFD to be committed before the checkpoint with the EFI. Hence
+ * we must strictly order the commit records of the checkpoints so
+ * that: a) the checkpoint callbacks are attached to the iclogs in the
+ * correct order; and b) the checkpoints are replayed in correct order
+ * in log recovery.
+ *
+ * Hence we need to add this context to the committing context list so
+ * that higher sequences will wait for us to write out a commit record
+ * before they do.
+ */
+ spin_lock(&cil->xc_cil_lock);
+ list_add(&ctx->committing, &cil->xc_committing);
+ spin_unlock(&cil->xc_cil_lock);
+ up_write(&cil->xc_ctx_lock);
+
+ /*
+ * Build a checkpoint transaction header and write it to the log to
+ * begin the transaction. We need to account for the space used by the
+ * transaction header here as it is not accounted for in xlog_write().
+ *
+ * The LSN we need to pass to the log items on transaction commit is
+ * the LSN reported by the first log vector write. If we use the commit
+ * record lsn then we can move the tail beyond the grant write head.
+ */
+ tic = ctx->ticket;
+ thdr.th_magic = XFS_TRANS_HEADER_MAGIC;
+ thdr.th_type = XFS_TRANS_CHECKPOINT;
+ thdr.th_tid = tic->t_tid;
+ thdr.th_num_items = num_iovecs;
+ lhdr.i_addr = (xfs_caddr_t)&thdr;
+ lhdr.i_len = sizeof(xfs_trans_header_t);
+ lhdr.i_type = XLOG_REG_TYPE_TRANSHDR;
+ tic->t_curr_res -= lhdr.i_len + sizeof(xlog_op_header_t);
+
+ lvhdr.lv_niovecs = 1;
+ lvhdr.lv_iovecp = &lhdr;
+ lvhdr.lv_next = ctx->lv_chain;
+
+ error = xlog_write(log, &lvhdr, tic, &ctx->start_lsn, NULL, 0);
+ if (error)
+ goto out_abort;
+
+ /*
+ * now that we've written the checkpoint into the log, strictly
+ * order the commit records so replay will get them in the right order.
+ */
+restart:
+ spin_lock(&cil->xc_cil_lock);
+ list_for_each_entry(new_ctx, &cil->xc_committing, committing) {
+ /*
+ * Higher sequences will wait for this one so skip them.
+ * Don't wait for own own sequence, either.
+ */
+ if (new_ctx->sequence >= ctx->sequence)
+ continue;
+ if (!new_ctx->commit_lsn) {
+ /*
+ * It is still being pushed! Wait for the push to
+ * complete, then start again from the beginning.
+ */
+ sv_wait(&cil->xc_commit_wait, 0, &cil->xc_cil_lock, 0);
+ goto restart;
+ }
+ }
+ spin_unlock(&cil->xc_cil_lock);
+
+ commit_lsn = xfs_log_done(log->l_mp, tic, &commit_iclog, 0);
+ if (error || commit_lsn == -1)
+ goto out_abort;
+
+ /* attach all the transactions w/ busy extents to iclog */
+ ctx->log_cb.cb_func = xlog_cil_committed;
+ ctx->log_cb.cb_arg = ctx;
+ error = xfs_log_notify(log->l_mp, commit_iclog, &ctx->log_cb);
+ if (error)
+ goto out_abort;
+
+ /*
+ * now the checkpoint commit is complete and we've attached the
+ * callbacks to the iclog we can assign the commit LSN to the context
+ * and wake up anyone who is waiting for the commit to complete.
+ */
+ spin_lock(&cil->xc_cil_lock);
+ ctx->commit_lsn = commit_lsn;
+ sv_broadcast(&cil->xc_commit_wait);
+ spin_unlock(&cil->xc_cil_lock);
+
+ /* release the hounds! */
+ return xfs_log_release_iclog(log->l_mp, commit_iclog);
+
+out_skip:
+ up_write(&cil->xc_ctx_lock);
+out_free_ticket:
+ xfs_log_ticket_put(new_ctx->ticket);
+ kmem_free(new_ctx);
+ return 0;
+
+out_abort:
+ xlog_cil_committed(ctx, XFS_LI_ABORTED);
+ return XFS_ERROR(EIO);
+}
+
+/*
+ * Conditionally push the CIL based on the sequence passed in.
+ *
+ * We only need to push if we haven't already pushed the sequence
+ * number given. Hence the only time we will trigger a push here is
+ * if the push sequence is the same as the current context.
+ *
+ * We return the current commit lsn to allow the callers to determine if a
+ * iclog flush is necessary following this call.
+ *
+ * XXX: Initially, just push the CIL unconditionally and return whatever
+ * commit lsn is there. It'll be empty, so this is broken for now.
+ */
+xfs_lsn_t
+xlog_cil_push_lsn(
+ struct log *log,
+ xfs_lsn_t push_seq)
+{
+ struct xfs_cil *cil = log->l_cilp;
+ struct xfs_cil_ctx *ctx;
+ xfs_lsn_t commit_lsn = NULLCOMMITLSN;
+
+restart:
+ down_write(&cil->xc_ctx_lock);
+ ASSERT(push_seq <= cil->xc_ctx->sequence);
+
+ /* check to see if we need to force out the current context */
+ if (push_seq == cil->xc_ctx->sequence) {
+ up_write(&cil->xc_ctx_lock);
+ xlog_cil_push(log, 1);
+ goto restart;
+ }
+
+ /*
+ * See if we can find a previous sequence still committing.
+ * We can drop the flush lock as soon as we have the cil lock
+ * because we are now only comparing contexts protected by
+ * the cil lock.
+ *
+ * We need to wait for all previous sequence commits to complete
+ * before allowing the force of push_seq to go ahead. Hence block
+ * on commits for those as well.
+ */
+ spin_lock(&cil->xc_cil_lock);
+ up_write(&cil->xc_ctx_lock);
+ list_for_each_entry(ctx, &cil->xc_committing, committing) {
+ if (ctx->sequence > push_seq)
+ continue;
+ if (!ctx->commit_lsn) {
+ /*
+ * It is still being pushed! Wait for the push to
+ * complete, then start again from the beginning.
+ */
+ sv_wait(&cil->xc_commit_wait, 0, &cil->xc_cil_lock, 0);
+ goto restart;
+ }
+ if (ctx->sequence != push_seq)
+ continue;
+ /* found it! */
+ commit_lsn = ctx->commit_lsn;
+ }
+ spin_unlock(&cil->xc_cil_lock);
+ return commit_lsn;
+}
+
+/*
+ * Check if the current log item was first committed in this sequence.
+ * We can't rely on just the log item being in the CIL, we have to check
+ * the recorded commit sequence number.
+ *
+ * Note: for this to be used in a non-racy manner, it has to be called with
+ * CIL flushing locked out. As a result, it should only be used during the
+ * transaction commit process when deciding what to format into the item.
+ */
+bool
+xfs_log_item_in_current_chkpt(
+ struct xfs_log_item *lip)
+{
+ struct xfs_cil_ctx *ctx;
+
+ if (!(lip->li_mountp->m_flags & XFS_MOUNT_DELAYLOG))
+ return false;
+ if (list_empty(&lip->li_cil))
+ return false;
+
+ ctx = lip->li_mountp->m_log->l_cilp->xc_ctx;
+
+ /*
+ * li_seq is written on the first commit of a log item to record the
+ * first checkpoint it is written to. Hence if it is different to the
+ * current sequence, we're in a new checkpoint.
+ */
+ if (XFS_LSN_CMP(lip->li_seq, ctx->sequence) != 0)
+ return false;
+ return true;
+}
diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h
index d55662db7077..8c072618965c 100644
--- a/fs/xfs/xfs_log_priv.h
+++ b/fs/xfs/xfs_log_priv.h
@@ -152,8 +152,6 @@ static inline uint xlog_get_client_id(__be32 i)
#define XLOG_RECOVERY_NEEDED 0x4 /* log was recovered */
#define XLOG_IO_ERROR 0x8 /* log hit an I/O error, and being
shutdown */
-typedef __uint32_t xlog_tid_t;
-
#ifdef __KERNEL__
/*
@@ -379,6 +377,99 @@ typedef struct xlog_in_core {
} xlog_in_core_t;
/*
+ * The CIL context is used to aggregate per-transaction details as well be
+ * passed to the iclog for checkpoint post-commit processing. After being
+ * passed to the iclog, another context needs to be allocated for tracking the
+ * next set of transactions to be aggregated into a checkpoint.
+ */
+struct xfs_cil;
+
+struct xfs_cil_ctx {
+ struct xfs_cil *cil;
+ xfs_lsn_t sequence; /* chkpt sequence # */
+ xfs_lsn_t start_lsn; /* first LSN of chkpt commit */
+ xfs_lsn_t commit_lsn; /* chkpt commit record lsn */
+ struct xlog_ticket *ticket; /* chkpt ticket */
+ int nvecs; /* number of regions */
+ int space_used; /* aggregate size of regions */
+ struct list_head busy_extents; /* busy extents in chkpt */
+ struct xfs_log_vec *lv_chain; /* logvecs being pushed */
+ xfs_log_callback_t log_cb; /* completion callback hook. */
+ struct list_head committing; /* ctx committing list */
+};
+
+/*
+ * Committed Item List structure
+ *
+ * This structure is used to track log items that have been committed but not
+ * yet written into the log. It is used only when the delayed logging mount
+ * option is enabled.
+ *
+ * This structure tracks the list of committing checkpoint contexts so
+ * we can avoid the problem of having to hold out new transactions during a
+ * flush until we have a the commit record LSN of the checkpoint. We can
+ * traverse the list of committing contexts in xlog_cil_push_lsn() to find a
+ * sequence match and extract the commit LSN directly from there. If the
+ * checkpoint is still in the process of committing, we can block waiting for
+ * the commit LSN to be determined as well. This should make synchronous
+ * operations almost as efficient as the old logging methods.
+ */
+struct xfs_cil {
+ struct log *xc_log;
+ struct list_head xc_cil;
+ spinlock_t xc_cil_lock;
+ struct xfs_cil_ctx *xc_ctx;
+ struct rw_semaphore xc_ctx_lock;
+ struct list_head xc_committing;
+ sv_t xc_commit_wait;
+};
+
+/*
+ * The amount of log space we should the CIL to aggregate is difficult to size.
+ * Whatever we chose we have to make we can get a reservation for the log space
+ * effectively, that it is large enough to capture sufficient relogging to
+ * reduce log buffer IO significantly, but it is not too large for the log or
+ * induces too much latency when writing out through the iclogs. We track both
+ * space consumed and the number of vectors in the checkpoint context, so we
+ * need to decide which to use for limiting.
+ *
+ * Every log buffer we write out during a push needs a header reserved, which
+ * is at least one sector and more for v2 logs. Hence we need a reservation of
+ * at least 512 bytes per 32k of log space just for the LR headers. That means
+ * 16KB of reservation per megabyte of delayed logging space we will consume,
+ * plus various headers. The number of headers will vary based on the num of
+ * io vectors, so limiting on a specific number of vectors is going to result
+ * in transactions of varying size. IOWs, it is more consistent to track and
+ * limit space consumed in the log rather than by the number of objects being
+ * logged in order to prevent checkpoint ticket overruns.
+ *
+ * Further, use of static reservations through the log grant mechanism is
+ * problematic. It introduces a lot of complexity (e.g. reserve grant vs write
+ * grant) and a significant deadlock potential because regranting write space
+ * can block on log pushes. Hence if we have to regrant log space during a log
+ * push, we can deadlock.
+ *
+ * However, we can avoid this by use of a dynamic "reservation stealing"
+ * technique during transaction commit whereby unused reservation space in the
+ * transaction ticket is transferred to the CIL ctx commit ticket to cover the
+ * space needed by the checkpoint transaction. This means that we never need to
+ * specifically reserve space for the CIL checkpoint transaction, nor do we
+ * need to regrant space once the checkpoint completes. This also means the
+ * checkpoint transaction ticket is specific to the checkpoint context, rather
+ * than the CIL itself.
+ *
+ * With dynamic reservations, we can basically make up arbitrary limits for the
+ * checkpoint size so long as they don't violate any other size rules. Hence
+ * the initial maximum size for the checkpoint transaction will be set to a
+ * quarter of the log or 8MB, which ever is smaller. 8MB is an arbitrary limit
+ * right now based on the latency of writing out a large amount of data through
+ * the circular iclog buffers.
+ */
+
+#define XLOG_CIL_SPACE_LIMIT(log) \
+ (min((log->l_logsize >> 2), (8 * 1024 * 1024)))
+
+/*
* The reservation head lsn is not made up of a cycle number and block number.
* Instead, it uses a cycle number and byte number. Logs don't expect to
* overflow 31 bits worth of byte offset, so using a byte number will mean
@@ -388,6 +479,7 @@ typedef struct log {
/* The following fields don't need locking */
struct xfs_mount *l_mp; /* mount point */
struct xfs_ail *l_ailp; /* AIL log is working with */
+ struct xfs_cil *l_cilp; /* CIL log is working with */
struct xfs_buf *l_xbuf; /* extra buffer for log
* wrapping */
struct xfs_buftarg *l_targ; /* buftarg of log */
@@ -396,9 +488,7 @@ typedef struct log {
struct xfs_buf_cancel **l_buf_cancel_table;
int l_iclog_hsize; /* size of iclog header */
int l_iclog_heads; /* # of iclog header sectors */
- uint l_sectbb_log; /* log2 of sector size in BBs */
- uint l_sectbb_mask; /* sector size (in BBs)
- * alignment mask */
+ uint l_sectBBsize; /* sector size in BBs (2^n) */
int l_iclog_size; /* size of log in bytes */
int l_iclog_size_log; /* log power size of log */
int l_iclog_bufs; /* number of iclog buffers */
@@ -440,19 +530,40 @@ typedef struct log {
#define XLOG_FORCED_SHUTDOWN(log) ((log)->l_flags & XLOG_IO_ERROR)
-
/* common routines */
extern xfs_lsn_t xlog_assign_tail_lsn(struct xfs_mount *mp);
-extern int xlog_find_tail(xlog_t *log,
- xfs_daddr_t *head_blk,
- xfs_daddr_t *tail_blk);
extern int xlog_recover(xlog_t *log);
extern int xlog_recover_finish(xlog_t *log);
extern void xlog_pack_data(xlog_t *log, xlog_in_core_t *iclog, int);
-extern struct xfs_buf *xlog_get_bp(xlog_t *, int);
-extern void xlog_put_bp(struct xfs_buf *);
-extern kmem_zone_t *xfs_log_ticket_zone;
+extern kmem_zone_t *xfs_log_ticket_zone;
+struct xlog_ticket *xlog_ticket_alloc(struct log *log, int unit_bytes,
+ int count, char client, uint xflags,
+ int alloc_flags);
+
+
+static inline void
+xlog_write_adv_cnt(void **ptr, int *len, int *off, size_t bytes)
+{
+ *ptr += bytes;
+ *len -= bytes;
+ *off += bytes;
+}
+
+void xlog_print_tic_res(struct xfs_mount *mp, struct xlog_ticket *ticket);
+int xlog_write(struct log *log, struct xfs_log_vec *log_vector,
+ struct xlog_ticket *tic, xfs_lsn_t *start_lsn,
+ xlog_in_core_t **commit_iclog, uint flags);
+
+/*
+ * Committed Item List interfaces
+ */
+int xlog_cil_init(struct log *log);
+void xlog_cil_init_post_recovery(struct log *log);
+void xlog_cil_destroy(struct log *log);
+
+int xlog_cil_push(struct log *log, int push_now);
+xfs_lsn_t xlog_cil_push_lsn(struct log *log, xfs_lsn_t push_sequence);
/*
* Unmount record type is used as a pseudo transaction type for the ticket.
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 69ac2e5ef20c..9ac5cfab27b9 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -50,51 +50,81 @@
STATIC int xlog_find_zeroed(xlog_t *, xfs_daddr_t *);
STATIC int xlog_clear_stale_blocks(xlog_t *, xfs_lsn_t);
-STATIC void xlog_recover_insert_item_backq(xlog_recover_item_t **q,
- xlog_recover_item_t *item);
#if defined(DEBUG)
STATIC void xlog_recover_check_summary(xlog_t *);
#else
#define xlog_recover_check_summary(log)
#endif
-
/*
* Sector aligned buffer routines for buffer create/read/write/access
*/
-#define XLOG_SECTOR_ROUNDUP_BBCOUNT(log, bbs) \
- ( ((log)->l_sectbb_mask && (bbs & (log)->l_sectbb_mask)) ? \
- ((bbs + (log)->l_sectbb_mask + 1) & ~(log)->l_sectbb_mask) : (bbs) )
-#define XLOG_SECTOR_ROUNDDOWN_BLKNO(log, bno) ((bno) & ~(log)->l_sectbb_mask)
+/*
+ * Verify the given count of basic blocks is valid number of blocks
+ * to specify for an operation involving the given XFS log buffer.
+ * Returns nonzero if the count is valid, 0 otherwise.
+ */
-xfs_buf_t *
+static inline int
+xlog_buf_bbcount_valid(
+ xlog_t *log,
+ int bbcount)
+{
+ return bbcount > 0 && bbcount <= log->l_logBBsize;
+}
+
+/*
+ * Allocate a buffer to hold log data. The buffer needs to be able
+ * to map to a range of nbblks basic blocks at any valid (basic
+ * block) offset within the log.
+ */
+STATIC xfs_buf_t *
xlog_get_bp(
xlog_t *log,
int nbblks)
{
- if (nbblks <= 0 || nbblks > log->l_logBBsize) {
- xlog_warn("XFS: Invalid block length (0x%x) given for buffer", nbblks);
- XFS_ERROR_REPORT("xlog_get_bp(1)",
- XFS_ERRLEVEL_HIGH, log->l_mp);
+ if (!xlog_buf_bbcount_valid(log, nbblks)) {
+ xlog_warn("XFS: Invalid block length (0x%x) given for buffer",
+ nbblks);
+ XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
return NULL;
}
- if (log->l_sectbb_log) {
- if (nbblks > 1)
- nbblks += XLOG_SECTOR_ROUNDUP_BBCOUNT(log, 1);
- nbblks = XLOG_SECTOR_ROUNDUP_BBCOUNT(log, nbblks);
- }
+ /*
+ * We do log I/O in units of log sectors (a power-of-2
+ * multiple of the basic block size), so we round up the
+ * requested size to acommodate the basic blocks required
+ * for complete log sectors.
+ *
+ * In addition, the buffer may be used for a non-sector-
+ * aligned block offset, in which case an I/O of the
+ * requested size could extend beyond the end of the
+ * buffer. If the requested size is only 1 basic block it
+ * will never straddle a sector boundary, so this won't be
+ * an issue. Nor will this be a problem if the log I/O is
+ * done in basic blocks (sector size 1). But otherwise we
+ * extend the buffer by one extra log sector to ensure
+ * there's space to accomodate this possiblility.
+ */
+ if (nbblks > 1 && log->l_sectBBsize > 1)
+ nbblks += log->l_sectBBsize;
+ nbblks = round_up(nbblks, log->l_sectBBsize);
+
return xfs_buf_get_noaddr(BBTOB(nbblks), log->l_mp->m_logdev_targp);
}
-void
+STATIC void
xlog_put_bp(
xfs_buf_t *bp)
{
xfs_buf_free(bp);
}
+/*
+ * Return the address of the start of the given block number's data
+ * in a log buffer. The buffer covers a log sector-aligned region.
+ */
STATIC xfs_caddr_t
xlog_align(
xlog_t *log,
@@ -102,15 +132,10 @@ xlog_align(
int nbblks,
xfs_buf_t *bp)
{
- xfs_caddr_t ptr;
-
- if (!log->l_sectbb_log)
- return XFS_BUF_PTR(bp);
+ xfs_daddr_t offset = blk_no & ((xfs_daddr_t)log->l_sectBBsize - 1);
- ptr = XFS_BUF_PTR(bp) + BBTOB((int)blk_no & log->l_sectbb_mask);
- ASSERT(XFS_BUF_SIZE(bp) >=
- BBTOB(nbblks + (blk_no & log->l_sectbb_mask)));
- return ptr;
+ ASSERT(BBTOB(offset + nbblks) <= XFS_BUF_SIZE(bp));
+ return XFS_BUF_PTR(bp) + BBTOB(offset);
}
@@ -126,21 +151,18 @@ xlog_bread_noalign(
{
int error;
- if (nbblks <= 0 || nbblks > log->l_logBBsize) {
- xlog_warn("XFS: Invalid block length (0x%x) given for buffer", nbblks);
- XFS_ERROR_REPORT("xlog_bread(1)",
- XFS_ERRLEVEL_HIGH, log->l_mp);
+ if (!xlog_buf_bbcount_valid(log, nbblks)) {
+ xlog_warn("XFS: Invalid block length (0x%x) given for buffer",
+ nbblks);
+ XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
return EFSCORRUPTED;
}
- if (log->l_sectbb_log) {
- blk_no = XLOG_SECTOR_ROUNDDOWN_BLKNO(log, blk_no);
- nbblks = XLOG_SECTOR_ROUNDUP_BBCOUNT(log, nbblks);
- }
+ blk_no = round_down(blk_no, log->l_sectBBsize);
+ nbblks = round_up(nbblks, log->l_sectBBsize);
ASSERT(nbblks > 0);
ASSERT(BBTOB(nbblks) <= XFS_BUF_SIZE(bp));
- ASSERT(bp);
XFS_BUF_SET_ADDR(bp, log->l_logBBstart + blk_no);
XFS_BUF_READ(bp);
@@ -188,17 +210,15 @@ xlog_bwrite(
{
int error;
- if (nbblks <= 0 || nbblks > log->l_logBBsize) {
- xlog_warn("XFS: Invalid block length (0x%x) given for buffer", nbblks);
- XFS_ERROR_REPORT("xlog_bwrite(1)",
- XFS_ERRLEVEL_HIGH, log->l_mp);
+ if (!xlog_buf_bbcount_valid(log, nbblks)) {
+ xlog_warn("XFS: Invalid block length (0x%x) given for buffer",
+ nbblks);
+ XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_HIGH, log->l_mp);
return EFSCORRUPTED;
}
- if (log->l_sectbb_log) {
- blk_no = XLOG_SECTOR_ROUNDDOWN_BLKNO(log, blk_no);
- nbblks = XLOG_SECTOR_ROUNDUP_BBCOUNT(log, nbblks);
- }
+ blk_no = round_down(blk_no, log->l_sectBBsize);
+ nbblks = round_up(nbblks, log->l_sectBBsize);
ASSERT(nbblks > 0);
ASSERT(BBTOB(nbblks) <= XFS_BUF_SIZE(bp));
@@ -329,39 +349,38 @@ xlog_find_cycle_start(
{
xfs_caddr_t offset;
xfs_daddr_t mid_blk;
+ xfs_daddr_t end_blk;
uint mid_cycle;
int error;
- mid_blk = BLK_AVG(first_blk, *last_blk);
- while (mid_blk != first_blk && mid_blk != *last_blk) {
+ end_blk = *last_blk;
+ mid_blk = BLK_AVG(first_blk, end_blk);
+ while (mid_blk != first_blk && mid_blk != end_blk) {
error = xlog_bread(log, mid_blk, 1, bp, &offset);
if (error)
return error;
mid_cycle = xlog_get_cycle(offset);
- if (mid_cycle == cycle) {
- *last_blk = mid_blk;
- /* last_half_cycle == mid_cycle */
- } else {
- first_blk = mid_blk;
- /* first_half_cycle == mid_cycle */
- }
- mid_blk = BLK_AVG(first_blk, *last_blk);
+ if (mid_cycle == cycle)
+ end_blk = mid_blk; /* last_half_cycle == mid_cycle */
+ else
+ first_blk = mid_blk; /* first_half_cycle == mid_cycle */
+ mid_blk = BLK_AVG(first_blk, end_blk);
}
- ASSERT((mid_blk == first_blk && mid_blk+1 == *last_blk) ||
- (mid_blk == *last_blk && mid_blk-1 == first_blk));
+ ASSERT((mid_blk == first_blk && mid_blk+1 == end_blk) ||
+ (mid_blk == end_blk && mid_blk-1 == first_blk));
+
+ *last_blk = end_blk;
return 0;
}
/*
- * Check that the range of blocks does not contain the cycle number
- * given. The scan needs to occur from front to back and the ptr into the
- * region must be updated since a later routine will need to perform another
- * test. If the region is completely good, we end up returning the same
- * last block number.
- *
- * Set blkno to -1 if we encounter no errors. This is an invalid block number
- * since we don't ever expect logs to get this large.
+ * Check that a range of blocks does not contain stop_on_cycle_no.
+ * Fill in *new_blk with the block offset where such a block is
+ * found, or with -1 (an invalid block number) if there is no such
+ * block in the range. The scan needs to occur from front to back
+ * and the pointer into the region must be updated since a later
+ * routine will need to perform another test.
*/
STATIC int
xlog_find_verify_cycle(
@@ -378,12 +397,16 @@ xlog_find_verify_cycle(
xfs_caddr_t buf = NULL;
int error = 0;
+ /*
+ * Greedily allocate a buffer big enough to handle the full
+ * range of basic blocks we'll be examining. If that fails,
+ * try a smaller size. We need to be able to read at least
+ * a log sector, or we're out of luck.
+ */
bufblks = 1 << ffs(nbblks);
-
while (!(bp = xlog_get_bp(log, bufblks))) {
- /* can't get enough memory to do everything in one big buffer */
bufblks >>= 1;
- if (bufblks <= log->l_sectbb_log)
+ if (bufblks < log->l_sectBBsize)
return ENOMEM;
}
@@ -631,7 +654,7 @@ xlog_find_head(
* In this case we want to find the first block with cycle
* number matching last_half_cycle. We expect the log to be
* some variation on
- * x + 1 ... | x ...
+ * x + 1 ... | x ... | x
* The first block with cycle number x (last_half_cycle) will
* be where the new head belongs. First we do a binary search
* for the first occurrence of last_half_cycle. The binary
@@ -641,11 +664,13 @@ xlog_find_head(
* the log, then we look for occurrences of last_half_cycle - 1
* at the end of the log. The cases we're looking for look
* like
- * x + 1 ... | x | x + 1 | x ...
- * ^ binary search stopped here
+ * v binary search stopped here
+ * x + 1 ... | x | x + 1 | x ... | x
+ * ^ but we want to locate this spot
* or
- * x + 1 ... | x ... | x - 1 | x
* <---------> less than scan distance
+ * x + 1 ... | x ... | x - 1 | x
+ * ^ we want to locate this spot
*/
stop_on_cycle = last_half_cycle;
if ((error = xlog_find_cycle_start(log, bp, first_blk,
@@ -701,16 +726,16 @@ xlog_find_head(
* certainly not the head of the log. By searching for
* last_half_cycle-1 we accomplish that.
*/
- start_blk = log_bbnum - num_scan_bblks + head_blk;
ASSERT(head_blk <= INT_MAX &&
- (xfs_daddr_t) num_scan_bblks - head_blk >= 0);
+ (xfs_daddr_t) num_scan_bblks >= head_blk);
+ start_blk = log_bbnum - (num_scan_bblks - head_blk);
if ((error = xlog_find_verify_cycle(log, start_blk,
num_scan_bblks - (int)head_blk,
(stop_on_cycle - 1), &new_blk)))
goto bp_err;
if (new_blk != -1) {
head_blk = new_blk;
- goto bad_blk;
+ goto validate_head;
}
/*
@@ -728,7 +753,7 @@ xlog_find_head(
head_blk = new_blk;
}
- bad_blk:
+validate_head:
/*
* Now we need to make sure head_blk is not pointing to a block in
* the middle of a log record.
@@ -750,7 +775,7 @@ xlog_find_head(
if ((error = xlog_find_verify_log_record(log, start_blk,
&head_blk, 0)) == -1) {
/* We hit the beginning of the log during our search */
- start_blk = log_bbnum - num_scan_bblks + head_blk;
+ start_blk = log_bbnum - (num_scan_bblks - head_blk);
new_blk = log_bbnum;
ASSERT(start_blk <= INT_MAX &&
(xfs_daddr_t) log_bbnum-start_blk >= 0);
@@ -805,7 +830,7 @@ xlog_find_head(
* We could speed up search by using current head_blk buffer, but it is not
* available.
*/
-int
+STATIC int
xlog_find_tail(
xlog_t *log,
xfs_daddr_t *head_blk,
@@ -835,12 +860,12 @@ xlog_find_tail(
if (*head_blk == 0) { /* special case */
error = xlog_bread(log, 0, 1, bp, &offset);
if (error)
- goto bread_err;
+ goto done;
if (xlog_get_cycle(offset) == 0) {
*tail_blk = 0;
/* leave all other log inited values alone */
- goto exit;
+ goto done;
}
}
@@ -851,7 +876,7 @@ xlog_find_tail(
for (i = (int)(*head_blk) - 1; i >= 0; i--) {
error = xlog_bread(log, i, 1, bp, &offset);
if (error)
- goto bread_err;
+ goto done;
if (XLOG_HEADER_MAGIC_NUM == be32_to_cpu(*(__be32 *)offset)) {
found = 1;
@@ -868,7 +893,7 @@ xlog_find_tail(
for (i = log->l_logBBsize - 1; i >= (int)(*head_blk); i--) {
error = xlog_bread(log, i, 1, bp, &offset);
if (error)
- goto bread_err;
+ goto done;
if (XLOG_HEADER_MAGIC_NUM ==
be32_to_cpu(*(__be32 *)offset)) {
@@ -943,7 +968,7 @@ xlog_find_tail(
umount_data_blk = (i + hblks) % log->l_logBBsize;
error = xlog_bread(log, umount_data_blk, 1, bp, &offset);
if (error)
- goto bread_err;
+ goto done;
op_head = (xlog_op_header_t *)offset;
if (op_head->oh_flags & XLOG_UNMOUNT_TRANS) {
@@ -989,12 +1014,10 @@ xlog_find_tail(
* But... if the -device- itself is readonly, just skip this.
* We can't recover this device anyway, so it won't matter.
*/
- if (!xfs_readonly_buftarg(log->l_mp->m_logdev_targp)) {
+ if (!xfs_readonly_buftarg(log->l_mp->m_logdev_targp))
error = xlog_clear_stale_blocks(log, tail_lsn);
- }
-bread_err:
-exit:
+done:
xlog_put_bp(bp);
if (error)
@@ -1154,16 +1177,22 @@ xlog_write_log_records(
xfs_caddr_t offset;
xfs_buf_t *bp;
int balign, ealign;
- int sectbb = XLOG_SECTOR_ROUNDUP_BBCOUNT(log, 1);
+ int sectbb = log->l_sectBBsize;
int end_block = start_block + blocks;
int bufblks;
int error = 0;
int i, j = 0;
+ /*
+ * Greedily allocate a buffer big enough to handle the full
+ * range of basic blocks to be written. If that fails, try
+ * a smaller size. We need to be able to write at least a
+ * log sector, or we're out of luck.
+ */
bufblks = 1 << ffs(blocks);
while (!(bp = xlog_get_bp(log, bufblks))) {
bufblks >>= 1;
- if (bufblks <= log->l_sectbb_log)
+ if (bufblks < sectbb)
return ENOMEM;
}
@@ -1171,7 +1200,7 @@ xlog_write_log_records(
* the buffer in the starting sector not covered by the first
* write below.
*/
- balign = XLOG_SECTOR_ROUNDDOWN_BLKNO(log, start_block);
+ balign = round_down(start_block, sectbb);
if (balign != start_block) {
error = xlog_bread_noalign(log, start_block, 1, bp);
if (error)
@@ -1190,7 +1219,7 @@ xlog_write_log_records(
* the buffer in the final sector not covered by the write.
* If this is the same sector as the above read, skip it.
*/
- ealign = XLOG_SECTOR_ROUNDDOWN_BLKNO(log, end_block);
+ ealign = round_down(end_block, sectbb);
if (j == 0 && (start_block + endcount > ealign)) {
offset = XFS_BUF_PTR(bp);
balign = BBTOB(ealign - start_block);
@@ -1367,40 +1396,50 @@ xlog_clear_stale_blocks(
STATIC xlog_recover_t *
xlog_recover_find_tid(
- xlog_recover_t *q,
+ struct hlist_head *head,
xlog_tid_t tid)
{
- xlog_recover_t *p = q;
+ xlog_recover_t *trans;
+ struct hlist_node *n;
- while (p != NULL) {
- if (p->r_log_tid == tid)
- break;
- p = p->r_next;
+ hlist_for_each_entry(trans, n, head, r_list) {
+ if (trans->r_log_tid == tid)
+ return trans;
}
- return p;
+ return NULL;
}
STATIC void
-xlog_recover_put_hashq(
- xlog_recover_t **q,
- xlog_recover_t *trans)
+xlog_recover_new_tid(
+ struct hlist_head *head,
+ xlog_tid_t tid,
+ xfs_lsn_t lsn)
{
- trans->r_next = *q;
- *q = trans;
+ xlog_recover_t *trans;
+
+ trans = kmem_zalloc(sizeof(xlog_recover_t), KM_SLEEP);
+ trans->r_log_tid = tid;
+ trans->r_lsn = lsn;
+ INIT_LIST_HEAD(&trans->r_itemq);
+
+ INIT_HLIST_NODE(&trans->r_list);
+ hlist_add_head(&trans->r_list, head);
}
STATIC void
xlog_recover_add_item(
- xlog_recover_item_t **itemq)
+ struct list_head *head)
{
xlog_recover_item_t *item;
item = kmem_zalloc(sizeof(xlog_recover_item_t), KM_SLEEP);
- xlog_recover_insert_item_backq(itemq, item);
+ INIT_LIST_HEAD(&item->ri_list);
+ list_add_tail(&item->ri_list, head);
}
STATIC int
xlog_recover_add_to_cont_trans(
+ struct log *log,
xlog_recover_t *trans,
xfs_caddr_t dp,
int len)
@@ -1409,8 +1448,7 @@ xlog_recover_add_to_cont_trans(
xfs_caddr_t ptr, old_ptr;
int old_len;
- item = trans->r_itemq;
- if (item == NULL) {
+ if (list_empty(&trans->r_itemq)) {
/* finish copying rest of trans header */
xlog_recover_add_item(&trans->r_itemq);
ptr = (xfs_caddr_t) &trans->r_theader +
@@ -1418,7 +1456,8 @@ xlog_recover_add_to_cont_trans(
memcpy(ptr, dp, len); /* d, s, l */
return 0;
}
- item = item->ri_prev;
+ /* take the tail entry */
+ item = list_entry(trans->r_itemq.prev, xlog_recover_item_t, ri_list);
old_ptr = item->ri_buf[item->ri_cnt-1].i_addr;
old_len = item->ri_buf[item->ri_cnt-1].i_len;
@@ -1427,6 +1466,7 @@ xlog_recover_add_to_cont_trans(
memcpy(&ptr[old_len], dp, len); /* d, s, l */
item->ri_buf[item->ri_cnt-1].i_len += len;
item->ri_buf[item->ri_cnt-1].i_addr = ptr;
+ trace_xfs_log_recover_item_add_cont(log, trans, item, 0);
return 0;
}
@@ -1445,6 +1485,7 @@ xlog_recover_add_to_cont_trans(
*/
STATIC int
xlog_recover_add_to_trans(
+ struct log *log,
xlog_recover_t *trans,
xfs_caddr_t dp,
int len)
@@ -1455,8 +1496,7 @@ xlog_recover_add_to_trans(
if (!len)
return 0;
- item = trans->r_itemq;
- if (item == NULL) {
+ if (list_empty(&trans->r_itemq)) {
/* we need to catch log corruptions here */
if (*(uint *)dp != XFS_TRANS_HEADER_MAGIC) {
xlog_warn("XFS: xlog_recover_add_to_trans: "
@@ -1474,12 +1514,15 @@ xlog_recover_add_to_trans(
memcpy(ptr, dp, len);
in_f = (xfs_inode_log_format_t *)ptr;
- if (item->ri_prev->ri_total != 0 &&
- item->ri_prev->ri_total == item->ri_prev->ri_cnt) {
+ /* take the tail entry */
+ item = list_entry(trans->r_itemq.prev, xlog_recover_item_t, ri_list);
+ if (item->ri_total != 0 &&
+ item->ri_total == item->ri_cnt) {
+ /* tail item is in use, get a new one */
xlog_recover_add_item(&trans->r_itemq);
+ item = list_entry(trans->r_itemq.prev,
+ xlog_recover_item_t, ri_list);
}
- item = trans->r_itemq;
- item = item->ri_prev;
if (item->ri_total == 0) { /* first region to be added */
if (in_f->ilf_size == 0 ||
@@ -1501,99 +1544,37 @@ xlog_recover_add_to_trans(
item->ri_buf[item->ri_cnt].i_addr = ptr;
item->ri_buf[item->ri_cnt].i_len = len;
item->ri_cnt++;
+ trace_xfs_log_recover_item_add(log, trans, item, 0);
return 0;
}
-STATIC void
-xlog_recover_new_tid(
- xlog_recover_t **q,
- xlog_tid_t tid,
- xfs_lsn_t lsn)
-{
- xlog_recover_t *trans;
-
- trans = kmem_zalloc(sizeof(xlog_recover_t), KM_SLEEP);
- trans->r_log_tid = tid;
- trans->r_lsn = lsn;
- xlog_recover_put_hashq(q, trans);
-}
-
-STATIC int
-xlog_recover_unlink_tid(
- xlog_recover_t **q,
- xlog_recover_t *trans)
-{
- xlog_recover_t *tp;
- int found = 0;
-
- ASSERT(trans != NULL);
- if (trans == *q) {
- *q = (*q)->r_next;
- } else {
- tp = *q;
- while (tp) {
- if (tp->r_next == trans) {
- found = 1;
- break;
- }
- tp = tp->r_next;
- }
- if (!found) {
- xlog_warn(
- "XFS: xlog_recover_unlink_tid: trans not found");
- ASSERT(0);
- return XFS_ERROR(EIO);
- }
- tp->r_next = tp->r_next->r_next;
- }
- return 0;
-}
-
-STATIC void
-xlog_recover_insert_item_backq(
- xlog_recover_item_t **q,
- xlog_recover_item_t *item)
-{
- if (*q == NULL) {
- item->ri_prev = item->ri_next = item;
- *q = item;
- } else {
- item->ri_next = *q;
- item->ri_prev = (*q)->ri_prev;
- (*q)->ri_prev = item;
- item->ri_prev->ri_next = item;
- }
-}
-
-STATIC void
-xlog_recover_insert_item_frontq(
- xlog_recover_item_t **q,
- xlog_recover_item_t *item)
-{
- xlog_recover_insert_item_backq(q, item);
- *q = item;
-}
-
+/*
+ * Sort the log items in the transaction. Cancelled buffers need
+ * to be put first so they are processed before any items that might
+ * modify the buffers. If they are cancelled, then the modifications
+ * don't need to be replayed.
+ */
STATIC int
xlog_recover_reorder_trans(
- xlog_recover_t *trans)
+ struct log *log,
+ xlog_recover_t *trans,
+ int pass)
{
- xlog_recover_item_t *first_item, *itemq, *itemq_next;
- xfs_buf_log_format_t *buf_f;
- ushort flags = 0;
+ xlog_recover_item_t *item, *n;
+ LIST_HEAD(sort_list);
+
+ list_splice_init(&trans->r_itemq, &sort_list);
+ list_for_each_entry_safe(item, n, &sort_list, ri_list) {
+ xfs_buf_log_format_t *buf_f;
- first_item = itemq = trans->r_itemq;
- trans->r_itemq = NULL;
- do {
- itemq_next = itemq->ri_next;
- buf_f = (xfs_buf_log_format_t *)itemq->ri_buf[0].i_addr;
+ buf_f = (xfs_buf_log_format_t *)item->ri_buf[0].i_addr;
- switch (ITEM_TYPE(itemq)) {
+ switch (ITEM_TYPE(item)) {
case XFS_LI_BUF:
- flags = buf_f->blf_flags;
- if (!(flags & XFS_BLI_CANCEL)) {
- xlog_recover_insert_item_frontq(&trans->r_itemq,
- itemq);
+ if (!(buf_f->blf_flags & XFS_BLF_CANCEL)) {
+ trace_xfs_log_recover_item_reorder_head(log,
+ trans, item, pass);
+ list_move(&item->ri_list, &trans->r_itemq);
break;
}
case XFS_LI_INODE:
@@ -1601,7 +1582,9 @@ xlog_recover_reorder_trans(
case XFS_LI_QUOTAOFF:
case XFS_LI_EFD:
case XFS_LI_EFI:
- xlog_recover_insert_item_backq(&trans->r_itemq, itemq);
+ trace_xfs_log_recover_item_reorder_tail(log,
+ trans, item, pass);
+ list_move_tail(&item->ri_list, &trans->r_itemq);
break;
default:
xlog_warn(
@@ -1609,8 +1592,8 @@ xlog_recover_reorder_trans(
ASSERT(0);
return XFS_ERROR(EIO);
}
- itemq = itemq_next;
- } while (first_item != itemq);
+ }
+ ASSERT(list_empty(&sort_list));
return 0;
}
@@ -1650,8 +1633,10 @@ xlog_recover_do_buffer_pass1(
/*
* If this isn't a cancel buffer item, then just return.
*/
- if (!(flags & XFS_BLI_CANCEL))
+ if (!(flags & XFS_BLF_CANCEL)) {
+ trace_xfs_log_recover_buf_not_cancel(log, buf_f);
return;
+ }
/*
* Insert an xfs_buf_cancel record into the hash table of
@@ -1685,6 +1670,7 @@ xlog_recover_do_buffer_pass1(
while (nextp != NULL) {
if (nextp->bc_blkno == blkno && nextp->bc_len == len) {
nextp->bc_refcount++;
+ trace_xfs_log_recover_buf_cancel_ref_inc(log, buf_f);
return;
}
prevp = nextp;
@@ -1698,13 +1684,14 @@ xlog_recover_do_buffer_pass1(
bcp->bc_refcount = 1;
bcp->bc_next = NULL;
prevp->bc_next = bcp;
+ trace_xfs_log_recover_buf_cancel_add(log, buf_f);
}
/*
* Check to see whether the buffer being recovered has a corresponding
* entry in the buffer cancel record table. If it does then return 1
* so that it will be cancelled, otherwise return 0. If the buffer is
- * actually a buffer cancel item (XFS_BLI_CANCEL is set), then decrement
+ * actually a buffer cancel item (XFS_BLF_CANCEL is set), then decrement
* the refcount on the entry in the table and remove it from the table
* if this is the last reference.
*
@@ -1729,7 +1716,7 @@ xlog_check_buffer_cancelled(
* There is nothing in the table built in pass one,
* so this buffer must not be cancelled.
*/
- ASSERT(!(flags & XFS_BLI_CANCEL));
+ ASSERT(!(flags & XFS_BLF_CANCEL));
return 0;
}
@@ -1741,7 +1728,7 @@ xlog_check_buffer_cancelled(
* There is no corresponding entry in the table built
* in pass one, so this buffer has not been cancelled.
*/
- ASSERT(!(flags & XFS_BLI_CANCEL));
+ ASSERT(!(flags & XFS_BLF_CANCEL));
return 0;
}
@@ -1760,7 +1747,7 @@ xlog_check_buffer_cancelled(
* one in the table and remove it if this is the
* last reference.
*/
- if (flags & XFS_BLI_CANCEL) {
+ if (flags & XFS_BLF_CANCEL) {
bcp->bc_refcount--;
if (bcp->bc_refcount == 0) {
if (prevp == NULL) {
@@ -1780,7 +1767,7 @@ xlog_check_buffer_cancelled(
* We didn't find a corresponding entry in the table, so
* return 0 so that the buffer is NOT cancelled.
*/
- ASSERT(!(flags & XFS_BLI_CANCEL));
+ ASSERT(!(flags & XFS_BLF_CANCEL));
return 0;
}
@@ -1837,6 +1824,8 @@ xlog_recover_do_inode_buffer(
unsigned int *data_map = NULL;
unsigned int map_size = 0;
+ trace_xfs_log_recover_buf_inode_buf(mp->m_log, buf_f);
+
switch (buf_f->blf_type) {
case XFS_LI_BUF:
data_map = buf_f->blf_data_map;
@@ -1880,8 +1869,8 @@ xlog_recover_do_inode_buffer(
nbits = xfs_contig_bits(data_map, map_size,
bit);
ASSERT(nbits > 0);
- reg_buf_offset = bit << XFS_BLI_SHIFT;
- reg_buf_bytes = nbits << XFS_BLI_SHIFT;
+ reg_buf_offset = bit << XFS_BLF_SHIFT;
+ reg_buf_bytes = nbits << XFS_BLF_SHIFT;
item_index++;
}
@@ -1895,7 +1884,7 @@ xlog_recover_do_inode_buffer(
}
ASSERT(item->ri_buf[item_index].i_addr != NULL);
- ASSERT((item->ri_buf[item_index].i_len % XFS_BLI_CHUNK) == 0);
+ ASSERT((item->ri_buf[item_index].i_len % XFS_BLF_CHUNK) == 0);
ASSERT((reg_buf_offset + reg_buf_bytes) <= XFS_BUF_COUNT(bp));
/*
@@ -1932,6 +1921,7 @@ xlog_recover_do_inode_buffer(
/*ARGSUSED*/
STATIC void
xlog_recover_do_reg_buffer(
+ struct xfs_mount *mp,
xlog_recover_item_t *item,
xfs_buf_t *bp,
xfs_buf_log_format_t *buf_f)
@@ -1943,6 +1933,8 @@ xlog_recover_do_reg_buffer(
unsigned int map_size = 0;
int error;
+ trace_xfs_log_recover_buf_reg_buf(mp->m_log, buf_f);
+
switch (buf_f->blf_type) {
case XFS_LI_BUF:
data_map = buf_f->blf_data_map;
@@ -1958,9 +1950,9 @@ xlog_recover_do_reg_buffer(
nbits = xfs_contig_bits(data_map, map_size, bit);
ASSERT(nbits > 0);
ASSERT(item->ri_buf[i].i_addr != NULL);
- ASSERT(item->ri_buf[i].i_len % XFS_BLI_CHUNK == 0);
+ ASSERT(item->ri_buf[i].i_len % XFS_BLF_CHUNK == 0);
ASSERT(XFS_BUF_COUNT(bp) >=
- ((uint)bit << XFS_BLI_SHIFT)+(nbits<<XFS_BLI_SHIFT));
+ ((uint)bit << XFS_BLF_SHIFT)+(nbits<<XFS_BLF_SHIFT));
/*
* Do a sanity check if this is a dquot buffer. Just checking
@@ -1969,7 +1961,7 @@ xlog_recover_do_reg_buffer(
*/
error = 0;
if (buf_f->blf_flags &
- (XFS_BLI_UDQUOT_BUF|XFS_BLI_PDQUOT_BUF|XFS_BLI_GDQUOT_BUF)) {
+ (XFS_BLF_UDQUOT_BUF|XFS_BLF_PDQUOT_BUF|XFS_BLF_GDQUOT_BUF)) {
if (item->ri_buf[i].i_addr == NULL) {
cmn_err(CE_ALERT,
"XFS: NULL dquot in %s.", __func__);
@@ -1990,9 +1982,9 @@ xlog_recover_do_reg_buffer(
}
memcpy(xfs_buf_offset(bp,
- (uint)bit << XFS_BLI_SHIFT), /* dest */
+ (uint)bit << XFS_BLF_SHIFT), /* dest */
item->ri_buf[i].i_addr, /* source */
- nbits<<XFS_BLI_SHIFT); /* length */
+ nbits<<XFS_BLF_SHIFT); /* length */
next:
i++;
bit += nbits;
@@ -2141,6 +2133,8 @@ xlog_recover_do_dquot_buffer(
{
uint type;
+ trace_xfs_log_recover_buf_dquot_buf(log, buf_f);
+
/*
* Filesystems are required to send in quota flags at mount time.
*/
@@ -2149,11 +2143,11 @@ xlog_recover_do_dquot_buffer(
}
type = 0;
- if (buf_f->blf_flags & XFS_BLI_UDQUOT_BUF)
+ if (buf_f->blf_flags & XFS_BLF_UDQUOT_BUF)
type |= XFS_DQ_USER;
- if (buf_f->blf_flags & XFS_BLI_PDQUOT_BUF)
+ if (buf_f->blf_flags & XFS_BLF_PDQUOT_BUF)
type |= XFS_DQ_PROJ;
- if (buf_f->blf_flags & XFS_BLI_GDQUOT_BUF)
+ if (buf_f->blf_flags & XFS_BLF_GDQUOT_BUF)
type |= XFS_DQ_GROUP;
/*
* This type of quotas was turned off, so ignore this buffer
@@ -2161,7 +2155,7 @@ xlog_recover_do_dquot_buffer(
if (log->l_quotaoffs_flag & type)
return;
- xlog_recover_do_reg_buffer(item, bp, buf_f);
+ xlog_recover_do_reg_buffer(mp, item, bp, buf_f);
}
/*
@@ -2174,7 +2168,7 @@ xlog_recover_do_dquot_buffer(
* here which overlaps that may be stale.
*
* When meta-data buffers are freed at run time we log a buffer item
- * with the XFS_BLI_CANCEL bit set to indicate that previous copies
+ * with the XFS_BLF_CANCEL bit set to indicate that previous copies
* of the buffer in the log should not be replayed at recovery time.
* This is so that if the blocks covered by the buffer are reused for
* file data before we crash we don't end up replaying old, freed
@@ -2208,7 +2202,7 @@ xlog_recover_do_buffer_trans(
if (pass == XLOG_RECOVER_PASS1) {
/*
* In this pass we're only looking for buf items
- * with the XFS_BLI_CANCEL bit set.
+ * with the XFS_BLF_CANCEL bit set.
*/
xlog_recover_do_buffer_pass1(log, buf_f);
return 0;
@@ -2222,9 +2216,11 @@ xlog_recover_do_buffer_trans(
*/
cancel = xlog_recover_do_buffer_pass2(log, buf_f);
if (cancel) {
+ trace_xfs_log_recover_buf_cancel(log, buf_f);
return 0;
}
}
+ trace_xfs_log_recover_buf_recover(log, buf_f);
switch (buf_f->blf_type) {
case XFS_LI_BUF:
blkno = buf_f->blf_blkno;
@@ -2242,9 +2238,9 @@ xlog_recover_do_buffer_trans(
}
mp = log->l_mp;
- buf_flags = XFS_BUF_LOCK;
- if (!(flags & XFS_BLI_INODE_BUF))
- buf_flags |= XFS_BUF_MAPPED;
+ buf_flags = XBF_LOCK;
+ if (!(flags & XFS_BLF_INODE_BUF))
+ buf_flags |= XBF_MAPPED;
bp = xfs_buf_read(mp->m_ddev_targp, blkno, len, buf_flags);
if (XFS_BUF_ISERROR(bp)) {
@@ -2256,13 +2252,13 @@ xlog_recover_do_buffer_trans(
}
error = 0;
- if (flags & XFS_BLI_INODE_BUF) {
+ if (flags & XFS_BLF_INODE_BUF) {
error = xlog_recover_do_inode_buffer(mp, item, bp, buf_f);
} else if (flags &
- (XFS_BLI_UDQUOT_BUF|XFS_BLI_PDQUOT_BUF|XFS_BLI_GDQUOT_BUF)) {
+ (XFS_BLF_UDQUOT_BUF|XFS_BLF_PDQUOT_BUF|XFS_BLF_GDQUOT_BUF)) {
xlog_recover_do_dquot_buffer(mp, log, item, bp, buf_f);
} else {
- xlog_recover_do_reg_buffer(item, bp, buf_f);
+ xlog_recover_do_reg_buffer(mp, item, bp, buf_f);
}
if (error)
return XFS_ERROR(error);
@@ -2342,11 +2338,13 @@ xlog_recover_do_inode_trans(
if (xlog_check_buffer_cancelled(log, in_f->ilf_blkno,
in_f->ilf_len, 0)) {
error = 0;
+ trace_xfs_log_recover_inode_cancel(log, in_f);
goto error;
}
+ trace_xfs_log_recover_inode_recover(log, in_f);
bp = xfs_buf_read(mp->m_ddev_targp, in_f->ilf_blkno, in_f->ilf_len,
- XFS_BUF_LOCK);
+ XBF_LOCK);
if (XFS_BUF_ISERROR(bp)) {
xfs_ioerror_alert("xlog_recover_do..(read#2)", mp,
bp, in_f->ilf_blkno);
@@ -2395,6 +2393,7 @@ xlog_recover_do_inode_trans(
/* do nothing */
} else {
xfs_buf_relse(bp);
+ trace_xfs_log_recover_inode_skip(log, in_f);
error = 0;
goto error;
}
@@ -2814,14 +2813,14 @@ xlog_recover_do_trans(
int pass)
{
int error = 0;
- xlog_recover_item_t *item, *first_item;
+ xlog_recover_item_t *item;
- error = xlog_recover_reorder_trans(trans);
+ error = xlog_recover_reorder_trans(log, trans, pass);
if (error)
return error;
- first_item = item = trans->r_itemq;
- do {
+ list_for_each_entry(item, &trans->r_itemq, ri_list) {
+ trace_xfs_log_recover_item_recover(log, trans, item, pass);
switch (ITEM_TYPE(item)) {
case XFS_LI_BUF:
error = xlog_recover_do_buffer_trans(log, item, pass);
@@ -2854,8 +2853,7 @@ xlog_recover_do_trans(
if (error)
return error;
- item = item->ri_next;
- } while (first_item != item);
+ }
return 0;
}
@@ -2869,21 +2867,18 @@ STATIC void
xlog_recover_free_trans(
xlog_recover_t *trans)
{
- xlog_recover_item_t *first_item, *item, *free_item;
+ xlog_recover_item_t *item, *n;
int i;
- item = first_item = trans->r_itemq;
- do {
- free_item = item;
- item = item->ri_next;
- /* Free the regions in the item. */
- for (i = 0; i < free_item->ri_cnt; i++) {
- kmem_free(free_item->ri_buf[i].i_addr);
- }
+ list_for_each_entry_safe(item, n, &trans->r_itemq, ri_list) {
+ /* Free the regions in the item. */
+ list_del(&item->ri_list);
+ for (i = 0; i < item->ri_cnt; i++)
+ kmem_free(item->ri_buf[i].i_addr);
/* Free the item itself */
- kmem_free(free_item->ri_buf);
- kmem_free(free_item);
- } while (first_item != item);
+ kmem_free(item->ri_buf);
+ kmem_free(item);
+ }
/* Free the transaction recover structure */
kmem_free(trans);
}
@@ -2891,14 +2886,12 @@ xlog_recover_free_trans(
STATIC int
xlog_recover_commit_trans(
xlog_t *log,
- xlog_recover_t **q,
xlog_recover_t *trans,
int pass)
{
int error;
- if ((error = xlog_recover_unlink_tid(q, trans)))
- return error;
+ hlist_del(&trans->r_list);
if ((error = xlog_recover_do_trans(log, trans, pass)))
return error;
xlog_recover_free_trans(trans); /* no error */
@@ -2926,7 +2919,7 @@ xlog_recover_unmount_trans(
STATIC int
xlog_recover_process_data(
xlog_t *log,
- xlog_recover_t *rhash[],
+ struct hlist_head rhash[],
xlog_rec_header_t *rhead,
xfs_caddr_t dp,
int pass)
@@ -2960,7 +2953,7 @@ xlog_recover_process_data(
}
tid = be32_to_cpu(ohead->oh_tid);
hash = XLOG_RHASH(tid);
- trans = xlog_recover_find_tid(rhash[hash], tid);
+ trans = xlog_recover_find_tid(&rhash[hash], tid);
if (trans == NULL) { /* not found; add new tid */
if (ohead->oh_flags & XLOG_START_TRANS)
xlog_recover_new_tid(&rhash[hash], tid,
@@ -2978,14 +2971,15 @@ xlog_recover_process_data(
switch (flags) {
case XLOG_COMMIT_TRANS:
error = xlog_recover_commit_trans(log,
- &rhash[hash], trans, pass);
+ trans, pass);
break;
case XLOG_UNMOUNT_TRANS:
error = xlog_recover_unmount_trans(trans);
break;
case XLOG_WAS_CONT_TRANS:
- error = xlog_recover_add_to_cont_trans(trans,
- dp, be32_to_cpu(ohead->oh_len));
+ error = xlog_recover_add_to_cont_trans(log,
+ trans, dp,
+ be32_to_cpu(ohead->oh_len));
break;
case XLOG_START_TRANS:
xlog_warn(
@@ -2995,7 +2989,7 @@ xlog_recover_process_data(
break;
case 0:
case XLOG_CONTINUE_TRANS:
- error = xlog_recover_add_to_trans(trans,
+ error = xlog_recover_add_to_trans(log, trans,
dp, be32_to_cpu(ohead->oh_len));
break;
default:
@@ -3204,14 +3198,14 @@ xlog_recover_process_one_iunlink(
int error;
ino = XFS_AGINO_TO_INO(mp, agno, agino);
- error = xfs_iget(mp, NULL, ino, 0, 0, &ip, 0);
+ error = xfs_iget(mp, NULL, ino, 0, 0, &ip);
if (error)
goto fail;
/*
* Get the on disk inode to find the next inode in the bucket.
*/
- error = xfs_itobp(mp, NULL, ip, &dip, &ibp, XFS_BUF_LOCK);
+ error = xfs_itobp(mp, NULL, ip, &dip, &ibp, XBF_LOCK);
if (error)
goto fail_iput;
@@ -3396,42 +3390,6 @@ xlog_pack_data(
}
}
-#if defined(DEBUG) && defined(XFS_LOUD_RECOVERY)
-STATIC void
-xlog_unpack_data_checksum(
- xlog_rec_header_t *rhead,
- xfs_caddr_t dp,
- xlog_t *log)
-{
- __be32 *up = (__be32 *)dp;
- uint chksum = 0;
- int i;
-
- /* divide length by 4 to get # words */
- for (i=0; i < be32_to_cpu(rhead->h_len) >> 2; i++) {
- chksum ^= be32_to_cpu(*up);
- up++;
- }
- if (chksum != be32_to_cpu(rhead->h_chksum)) {
- if (rhead->h_chksum ||
- ((log->l_flags & XLOG_CHKSUM_MISMATCH) == 0)) {
- cmn_err(CE_DEBUG,
- "XFS: LogR chksum mismatch: was (0x%x) is (0x%x)\n",
- be32_to_cpu(rhead->h_chksum), chksum);
- cmn_err(CE_DEBUG,
-"XFS: Disregard message if filesystem was created with non-DEBUG kernel");
- if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) {
- cmn_err(CE_DEBUG,
- "XFS: LogR this is a LogV2 filesystem\n");
- }
- log->l_flags |= XLOG_CHKSUM_MISMATCH;
- }
- }
-}
-#else
-#define xlog_unpack_data_checksum(rhead, dp, log)
-#endif
-
STATIC void
xlog_unpack_data(
xlog_rec_header_t *rhead,
@@ -3455,8 +3413,6 @@ xlog_unpack_data(
dp += BBSIZE;
}
}
-
- xlog_unpack_data_checksum(rhead, dp, log);
}
STATIC int
@@ -3517,7 +3473,7 @@ xlog_do_recovery_pass(
int error = 0, h_size;
int bblks, split_bblks;
int hblks, split_hblks, wrapped_hblks;
- xlog_recover_t *rhash[XLOG_RHASH_SIZE];
+ struct hlist_head rhash[XLOG_RHASH_SIZE];
ASSERT(head_blk != tail_blk);
@@ -3555,7 +3511,7 @@ xlog_do_recovery_pass(
hblks = 1;
}
} else {
- ASSERT(log->l_sectbb_log == 0);
+ ASSERT(log->l_sectBBsize == 1);
hblks = 1;
hbp = xlog_get_bp(log, 1);
h_size = XLOG_BIG_RECORD_BSIZE;
@@ -3978,8 +3934,7 @@ xlog_recover_finish(
* case the unlink transactions would have problems
* pushing the EFIs out of the way.
*/
- xfs_log_force(log->l_mp, (xfs_lsn_t)0,
- (XFS_LOG_FORCE | XFS_LOG_SYNC));
+ xfs_log_force(log->l_mp, XFS_LOG_SYNC);
xlog_recover_process_iunlinks(log);
@@ -4012,10 +3967,6 @@ xlog_recover_check_summary(
xfs_agf_t *agfp;
xfs_buf_t *agfbp;
xfs_buf_t *agibp;
- xfs_buf_t *sbbp;
-#ifdef XFS_LOUD_RECOVERY
- xfs_sb_t *sbp;
-#endif
xfs_agnumber_t agno;
__uint64_t freeblks;
__uint64_t itotal;
@@ -4050,30 +4001,5 @@ xlog_recover_check_summary(
xfs_buf_relse(agibp);
}
}
-
- sbbp = xfs_getsb(mp, 0);
-#ifdef XFS_LOUD_RECOVERY
- sbp = &mp->m_sb;
- xfs_sb_from_disk(sbp, XFS_BUF_TO_SBP(sbbp));
- cmn_err(CE_NOTE,
- "xlog_recover_check_summary: sb_icount %Lu itotal %Lu",
- sbp->sb_icount, itotal);
- cmn_err(CE_NOTE,
- "xlog_recover_check_summary: sb_ifree %Lu itotal %Lu",
- sbp->sb_ifree, ifree);
- cmn_err(CE_NOTE,
- "xlog_recover_check_summary: sb_fdblocks %Lu freeblks %Lu",
- sbp->sb_fdblocks, freeblks);
-#if 0
- /*
- * This is turned off until I account for the allocation
- * btree blocks which live in free space.
- */
- ASSERT(sbp->sb_icount == itotal);
- ASSERT(sbp->sb_ifree == ifree);
- ASSERT(sbp->sb_fdblocks == freeblks);
-#endif
-#endif
- xfs_buf_relse(sbbp);
}
#endif /* DEBUG */
diff --git a/fs/xfs/xfs_log_recover.h b/fs/xfs/xfs_log_recover.h
index b22545555301..1c55ccbb379d 100644
--- a/fs/xfs/xfs_log_recover.h
+++ b/fs/xfs/xfs_log_recover.h
@@ -28,29 +28,28 @@
#define XLOG_RHASH(tid) \
((((__uint32_t)tid)>>XLOG_RHASH_SHIFT) & (XLOG_RHASH_SIZE-1))
-#define XLOG_MAX_REGIONS_IN_ITEM (XFS_MAX_BLOCKSIZE / XFS_BLI_CHUNK / 2 + 1)
+#define XLOG_MAX_REGIONS_IN_ITEM (XFS_MAX_BLOCKSIZE / XFS_BLF_CHUNK / 2 + 1)
/*
* item headers are in ri_buf[0]. Additional buffers follow.
*/
typedef struct xlog_recover_item {
- struct xlog_recover_item *ri_next;
- struct xlog_recover_item *ri_prev;
- int ri_type;
- int ri_cnt; /* count of regions found */
- int ri_total; /* total regions */
- xfs_log_iovec_t *ri_buf; /* ptr to regions buffer */
+ struct list_head ri_list;
+ int ri_type;
+ int ri_cnt; /* count of regions found */
+ int ri_total; /* total regions */
+ xfs_log_iovec_t *ri_buf; /* ptr to regions buffer */
} xlog_recover_item_t;
struct xlog_tid;
typedef struct xlog_recover {
- struct xlog_recover *r_next;
- xlog_tid_t r_log_tid; /* log's transaction id */
- xfs_trans_header_t r_theader; /* trans header for partial */
- int r_state; /* not needed */
- xfs_lsn_t r_lsn; /* xact lsn */
- xlog_recover_item_t *r_itemq; /* q for items */
+ struct hlist_node r_list;
+ xlog_tid_t r_log_tid; /* log's transaction id */
+ xfs_trans_header_t r_theader; /* trans header for partial */
+ int r_state; /* not needed */
+ xfs_lsn_t r_lsn; /* xact lsn */
+ struct list_head r_itemq; /* q for items */
} xlog_recover_t;
#define ITEM_TYPE(i) (*(ushort *)(i)->ri_buf[0].i_addr)
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index eb403b40e120..69f62d8b2816 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -201,6 +201,38 @@ xfs_uuid_unmount(
/*
+ * Reference counting access wrappers to the perag structures.
+ */
+struct xfs_perag *
+xfs_perag_get(struct xfs_mount *mp, xfs_agnumber_t agno)
+{
+ struct xfs_perag *pag;
+ int ref = 0;
+
+ spin_lock(&mp->m_perag_lock);
+ pag = radix_tree_lookup(&mp->m_perag_tree, agno);
+ if (pag) {
+ ASSERT(atomic_read(&pag->pag_ref) >= 0);
+ /* catch leaks in the positive direction during testing */
+ ASSERT(atomic_read(&pag->pag_ref) < 1000);
+ ref = atomic_inc_return(&pag->pag_ref);
+ }
+ spin_unlock(&mp->m_perag_lock);
+ trace_xfs_perag_get(mp, agno, ref, _RET_IP_);
+ return pag;
+}
+
+void
+xfs_perag_put(struct xfs_perag *pag)
+{
+ int ref;
+
+ ASSERT(atomic_read(&pag->pag_ref) > 0);
+ ref = atomic_dec_return(&pag->pag_ref);
+ trace_xfs_perag_put(pag->pag_mount, pag->pag_agno, ref, _RET_IP_);
+}
+
+/*
* Free up the resources associated with a mount structure. Assume that
* the structure was initially zeroed, so we can tell which fields got
* initialized.
@@ -209,13 +241,16 @@ STATIC void
xfs_free_perag(
xfs_mount_t *mp)
{
- if (mp->m_perag) {
- int agno;
+ xfs_agnumber_t agno;
+ struct xfs_perag *pag;
- for (agno = 0; agno < mp->m_maxagi; agno++)
- if (mp->m_perag[agno].pagb_list)
- kmem_free(mp->m_perag[agno].pagb_list);
- kmem_free(mp->m_perag);
+ for (agno = 0; agno < mp->m_sb.sb_agcount; agno++) {
+ spin_lock(&mp->m_perag_lock);
+ pag = radix_tree_delete(&mp->m_perag_tree, agno);
+ ASSERT(pag);
+ ASSERT(atomic_read(&pag->pag_ref) == 0);
+ spin_unlock(&mp->m_perag_lock);
+ kmem_free(pag);
}
}
@@ -233,10 +268,10 @@ xfs_sb_validate_fsb_count(
#if XFS_BIG_BLKNOS /* Limited by ULONG_MAX of page cache index */
if (nblocks >> (PAGE_CACHE_SHIFT - sbp->sb_blocklog) > ULONG_MAX)
- return E2BIG;
+ return EFBIG;
#else /* Limited by UINT_MAX of sectors */
if (nblocks << (sbp->sb_blocklog - BBSHIFT) > UINT_MAX)
- return E2BIG;
+ return EFBIG;
#endif
return 0;
}
@@ -358,7 +393,7 @@ xfs_mount_validate_sb(
xfs_sb_validate_fsb_count(sbp, sbp->sb_rblocks)) {
xfs_fs_mount_cmn_err(flags,
"file system too large to be mounted on this system.");
- return XFS_ERROR(E2BIG);
+ return XFS_ERROR(EFBIG);
}
if (unlikely(sbp->sb_inprogress)) {
@@ -378,46 +413,73 @@ xfs_mount_validate_sb(
return 0;
}
-STATIC void
-xfs_initialize_perag_icache(
- xfs_perag_t *pag)
-{
- if (!pag->pag_ici_init) {
- rwlock_init(&pag->pag_ici_lock);
- INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC);
- pag->pag_ici_init = 1;
- }
-}
-
-xfs_agnumber_t
+int
xfs_initialize_perag(
xfs_mount_t *mp,
- xfs_agnumber_t agcount)
+ xfs_agnumber_t agcount,
+ xfs_agnumber_t *maxagi)
{
xfs_agnumber_t index, max_metadata;
+ xfs_agnumber_t first_initialised = 0;
xfs_perag_t *pag;
xfs_agino_t agino;
xfs_ino_t ino;
xfs_sb_t *sbp = &mp->m_sb;
- xfs_ino_t max_inum = XFS_MAXINUMBER_32;
+ int error = -ENOMEM;
- /* Check to see if the filesystem can overflow 32 bit inodes */
+ /*
+ * Walk the current per-ag tree so we don't try to initialise AGs
+ * that already exist (growfs case). Allocate and insert all the
+ * AGs we don't find ready for initialisation.
+ */
+ for (index = 0; index < agcount; index++) {
+ pag = xfs_perag_get(mp, index);
+ if (pag) {
+ xfs_perag_put(pag);
+ continue;
+ }
+ if (!first_initialised)
+ first_initialised = index;
+
+ pag = kmem_zalloc(sizeof(*pag), KM_MAYFAIL);
+ if (!pag)
+ goto out_unwind;
+ pag->pag_agno = index;
+ pag->pag_mount = mp;
+ rwlock_init(&pag->pag_ici_lock);
+ INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC);
+
+ if (radix_tree_preload(GFP_NOFS))
+ goto out_unwind;
+
+ spin_lock(&mp->m_perag_lock);
+ if (radix_tree_insert(&mp->m_perag_tree, index, pag)) {
+ BUG();
+ spin_unlock(&mp->m_perag_lock);
+ radix_tree_preload_end();
+ error = -EEXIST;
+ goto out_unwind;
+ }
+ spin_unlock(&mp->m_perag_lock);
+ radix_tree_preload_end();
+ }
+
+ /*
+ * If we mount with the inode64 option, or no inode overflows
+ * the legacy 32-bit address space clear the inode32 option.
+ */
agino = XFS_OFFBNO_TO_AGINO(mp, sbp->sb_agblocks - 1, 0);
ino = XFS_AGINO_TO_INO(mp, agcount - 1, agino);
- /* Clear the mount flag if no inode can overflow 32 bits
- * on this filesystem, or if specifically requested..
- */
- if ((mp->m_flags & XFS_MOUNT_SMALL_INUMS) && ino > max_inum) {
+ if ((mp->m_flags & XFS_MOUNT_SMALL_INUMS) && ino > XFS_MAXINUMBER_32)
mp->m_flags |= XFS_MOUNT_32BITINODES;
- } else {
+ else
mp->m_flags &= ~XFS_MOUNT_32BITINODES;
- }
- /* If we can overflow then setup the ag headers accordingly */
if (mp->m_flags & XFS_MOUNT_32BITINODES) {
- /* Calculate how much should be reserved for inodes to
- * meet the max inode percentage.
+ /*
+ * Calculate how much should be reserved for inodes to meet
+ * the max inode percentage.
*/
if (mp->m_maxicount) {
__uint64_t icount;
@@ -430,29 +492,39 @@ xfs_initialize_perag(
} else {
max_metadata = agcount;
}
+
for (index = 0; index < agcount; index++) {
ino = XFS_AGINO_TO_INO(mp, index, agino);
- if (ino > max_inum) {
+ if (ino > XFS_MAXINUMBER_32) {
index++;
break;
}
- /* This ag is preferred for inodes */
- pag = &mp->m_perag[index];
+ pag = xfs_perag_get(mp, index);
pag->pagi_inodeok = 1;
if (index < max_metadata)
pag->pagf_metadata = 1;
- xfs_initialize_perag_icache(pag);
+ xfs_perag_put(pag);
}
} else {
- /* Setup default behavior for smaller filesystems */
for (index = 0; index < agcount; index++) {
- pag = &mp->m_perag[index];
+ pag = xfs_perag_get(mp, index);
pag->pagi_inodeok = 1;
- xfs_initialize_perag_icache(pag);
+ xfs_perag_put(pag);
}
}
- return index;
+
+ if (maxagi)
+ *maxagi = index;
+ return 0;
+
+out_unwind:
+ kmem_free(pag);
+ for (; index > first_initialised; index--) {
+ pag = radix_tree_delete(&mp->m_perag_tree, index);
+ kmem_free(pag);
+ }
+ return error;
}
void
@@ -583,7 +655,7 @@ xfs_readsb(xfs_mount_t *mp, int flags)
* access to the superblock.
*/
sector_size = xfs_getsize_buftarg(mp->m_ddev_targp);
- extra_flags = XFS_BUF_LOCK | XFS_BUF_MANAGE | XFS_BUF_MAPPED;
+ extra_flags = XBF_LOCK | XBF_FS_MANAGED | XBF_MAPPED;
bp = xfs_buf_read(mp->m_ddev_targp, XFS_SB_DADDR, BTOBB(sector_size),
extra_flags);
@@ -731,12 +803,13 @@ xfs_initialize_perag_data(xfs_mount_t *mp, xfs_agnumber_t agcount)
error = xfs_ialloc_pagi_init(mp, NULL, index);
if (error)
return error;
- pag = &mp->m_perag[index];
+ pag = xfs_perag_get(mp, index);
ifree += pag->pagi_freecount;
ialloc += pag->pagi_count;
bfree += pag->pagf_freeblks;
bfreelst += pag->pagf_flcount;
btree += pag->pagf_btreeblks;
+ xfs_perag_put(pag);
}
/*
* Overwrite incore superblock counters with just-read data
@@ -926,7 +999,7 @@ xfs_check_sizes(xfs_mount_t *mp)
d = (xfs_daddr_t)XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks);
if (XFS_BB_TO_FSB(mp, d) != mp->m_sb.sb_dblocks) {
cmn_err(CE_WARN, "XFS: size check 1 failed");
- return XFS_ERROR(E2BIG);
+ return XFS_ERROR(EFBIG);
}
error = xfs_read_buf(mp, mp->m_ddev_targp,
d - XFS_FSS_TO_BB(mp, 1),
@@ -936,7 +1009,7 @@ xfs_check_sizes(xfs_mount_t *mp)
} else {
cmn_err(CE_WARN, "XFS: size check 2 failed");
if (error == ENOSPC)
- error = XFS_ERROR(E2BIG);
+ error = XFS_ERROR(EFBIG);
return error;
}
@@ -944,7 +1017,7 @@ xfs_check_sizes(xfs_mount_t *mp)
d = (xfs_daddr_t)XFS_FSB_TO_BB(mp, mp->m_sb.sb_logblocks);
if (XFS_BB_TO_FSB(mp, d) != mp->m_sb.sb_logblocks) {
cmn_err(CE_WARN, "XFS: size check 3 failed");
- return XFS_ERROR(E2BIG);
+ return XFS_ERROR(EFBIG);
}
error = xfs_read_buf(mp, mp->m_logdev_targp,
d - XFS_FSB_TO_BB(mp, 1),
@@ -954,7 +1027,7 @@ xfs_check_sizes(xfs_mount_t *mp)
} else {
cmn_err(CE_WARN, "XFS: size check 3 failed");
if (error == ENOSPC)
- error = XFS_ERROR(E2BIG);
+ error = XFS_ERROR(EFBIG);
return error;
}
}
@@ -1008,6 +1081,24 @@ xfs_mount_reset_sbqflags(
return xfs_trans_commit(tp, 0);
}
+__uint64_t
+xfs_default_resblks(xfs_mount_t *mp)
+{
+ __uint64_t resblks;
+
+ /*
+ * We default to 5% or 8192 fsbs of space reserved, whichever is
+ * smaller. This is intended to cover concurrent allocation
+ * transactions when we initially hit enospc. These each require a 4
+ * block reservation. Hence by default we cover roughly 2000 concurrent
+ * allocation reservations.
+ */
+ resblks = mp->m_sb.sb_dblocks;
+ do_div(resblks, 20);
+ resblks = min_t(__uint64_t, resblks, 8192);
+ return resblks;
+}
+
/*
* This function does the following on an initial mount of a file system:
* - reads the superblock from disk and init the mount struct
@@ -1152,13 +1243,13 @@ xfs_mountfs(
/*
* Allocate and initialize the per-ag data.
*/
- init_rwsem(&mp->m_peraglock);
- mp->m_perag = kmem_zalloc(sbp->sb_agcount * sizeof(xfs_perag_t),
- KM_MAYFAIL);
- if (!mp->m_perag)
+ spin_lock_init(&mp->m_perag_lock);
+ INIT_RADIX_TREE(&mp->m_perag_tree, GFP_ATOMIC);
+ error = xfs_initialize_perag(mp, sbp->sb_agcount, &mp->m_maxagi);
+ if (error) {
+ cmn_err(CE_WARN, "XFS: Failed per-ag init: %d", error);
goto out_remove_uuid;
-
- mp->m_maxagi = xfs_initialize_perag(mp, sbp->sb_agcount);
+ }
if (!sbp->sb_logblocks) {
cmn_err(CE_WARN, "XFS: no log defined");
@@ -1209,7 +1300,7 @@ xfs_mountfs(
* Get and sanity-check the root inode.
* Save the pointer to it in the mount structure.
*/
- error = xfs_iget(mp, NULL, sbp->sb_rootino, 0, XFS_ILOCK_EXCL, &rip, 0);
+ error = xfs_iget(mp, NULL, sbp->sb_rootino, 0, XFS_ILOCK_EXCL, &rip);
if (error) {
cmn_err(CE_WARN, "XFS: failed to read root inode");
goto out_log_dealloc;
@@ -1304,13 +1395,6 @@ xfs_mountfs(
xfs_qm_mount_quotas(mp);
}
-#if defined(DEBUG) && defined(XFS_LOUD_RECOVERY)
- if (XFS_IS_QUOTA_ON(mp))
- xfs_fs_cmn_err(CE_NOTE, mp, "Disk quotas turned on");
- else
- xfs_fs_cmn_err(CE_NOTE, mp, "Disk quotas not turned on");
-#endif
-
/*
* Now we are mounted, reserve a small amount of unused space for
* privileged transactions. This is needed so that transaction
@@ -1319,17 +1403,16 @@ xfs_mountfs(
* attr, unwritten extent conversion at ENOSPC, etc. Data allocations
* are not allowed to use this reserved space.
*
- * We default to 5% or 1024 fsbs of space reserved, whichever is smaller.
* This may drive us straight to ENOSPC on mount, but that implies
* we were already there on the last unmount. Warn if this occurs.
*/
- resblks = mp->m_sb.sb_dblocks;
- do_div(resblks, 20);
- resblks = min_t(__uint64_t, resblks, 1024);
- error = xfs_reserve_blocks(mp, &resblks, NULL);
- if (error)
- cmn_err(CE_WARN, "XFS: Unable to allocate reserve blocks. "
- "Continuing without a reserve pool.");
+ if (!(mp->m_flags & XFS_MOUNT_RDONLY)) {
+ resblks = xfs_default_resblks(mp);
+ error = xfs_reserve_blocks(mp, &resblks, NULL);
+ if (error)
+ cmn_err(CE_WARN, "XFS: Unable to allocate reserve "
+ "blocks. Continuing without a reserve pool.");
+ }
return 0;
@@ -1372,8 +1455,19 @@ xfs_unmountfs(
* push out the iclog we will never get that unlocked. hence we
* need to force the log first.
*/
- xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE | XFS_LOG_SYNC);
- xfs_reclaim_inodes(mp, XFS_IFLUSH_ASYNC);
+ xfs_log_force(mp, XFS_LOG_SYNC);
+
+ /*
+ * Do a delwri reclaim pass first so that as many dirty inodes are
+ * queued up for IO as possible. Then flush the buffers before making
+ * a synchronous path to catch all the remaining inodes are reclaimed.
+ * This makes the reclaim process as quick as possible by avoiding
+ * synchronous writeout and blocking on inodes already in the delwri
+ * state as much as possible.
+ */
+ xfs_reclaim_inodes(mp, 0);
+ XFS_bflush(mp->m_ddev_targp);
+ xfs_reclaim_inodes(mp, SYNC_WAIT);
xfs_qm_unmount(mp);
@@ -1382,7 +1476,7 @@ xfs_unmountfs(
* that nothing is pinned. This is important because bflush()
* will skip pinned buffers.
*/
- xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE | XFS_LOG_SYNC);
+ xfs_log_force(mp, XFS_LOG_SYNC);
xfs_binval(mp->m_ddev_targp);
if (mp->m_rtdev_targp) {
@@ -1548,15 +1642,14 @@ xfs_mod_sb(xfs_trans_t *tp, __int64_t fields)
xfs_sb_to_disk(XFS_BUF_TO_SBP(bp), &mp->m_sb, fields);
/* find modified range */
+ f = (xfs_sb_field_t)xfs_highbit64((__uint64_t)fields);
+ ASSERT((1LL << f) & XFS_SB_MOD_BITS);
+ last = xfs_sb_info[f + 1].offset - 1;
f = (xfs_sb_field_t)xfs_lowbit64((__uint64_t)fields);
ASSERT((1LL << f) & XFS_SB_MOD_BITS);
first = xfs_sb_info[f].offset;
- f = (xfs_sb_field_t)xfs_highbit64((__uint64_t)fields);
- ASSERT((1LL << f) & XFS_SB_MOD_BITS);
- last = xfs_sb_info[f + 1].offset - 1;
-
xfs_trans_log_buf(tp, bp, first, last);
}
@@ -1620,26 +1713,30 @@ xfs_mod_incore_sb_unlocked(
lcounter += rem;
}
} else { /* Taking blocks away */
-
lcounter += delta;
+ if (lcounter >= 0) {
+ mp->m_sb.sb_fdblocks = lcounter +
+ XFS_ALLOC_SET_ASIDE(mp);
+ return 0;
+ }
- /*
- * If were out of blocks, use any available reserved blocks if
- * were allowed to.
- */
+ /*
+ * We are out of blocks, use any available reserved
+ * blocks if were allowed to.
+ */
+ if (!rsvd)
+ return XFS_ERROR(ENOSPC);
- if (lcounter < 0) {
- if (rsvd) {
- lcounter = (long long)mp->m_resblks_avail + delta;
- if (lcounter < 0) {
- return XFS_ERROR(ENOSPC);
- }
- mp->m_resblks_avail = lcounter;
- return 0;
- } else { /* not reserved */
- return XFS_ERROR(ENOSPC);
- }
+ lcounter = (long long)mp->m_resblks_avail + delta;
+ if (lcounter >= 0) {
+ mp->m_resblks_avail = lcounter;
+ return 0;
}
+ printk_once(KERN_WARNING
+ "Filesystem \"%s\": reserve blocks depleted! "
+ "Consider increasing reserve pool size.",
+ mp->m_fsname);
+ return XFS_ERROR(ENOSPC);
}
mp->m_sb.sb_fdblocks = lcounter + XFS_ALLOC_SET_ASIDE(mp);
@@ -1887,7 +1984,7 @@ xfs_getsb(
ASSERT(mp->m_sb_bp != NULL);
bp = mp->m_sb_bp;
- if (flags & XFS_BUF_TRYLOCK) {
+ if (flags & XBF_TRYLOCK) {
if (!XFS_BUF_CPSEMA(bp)) {
return NULL;
}
@@ -1947,6 +2044,26 @@ xfs_mount_log_sb(
return error;
}
+/*
+ * If the underlying (data/log/rt) device is readonly, there are some
+ * operations that cannot proceed.
+ */
+int
+xfs_dev_is_read_only(
+ struct xfs_mount *mp,
+ char *message)
+{
+ if (xfs_readonly_buftarg(mp->m_ddev_targp) ||
+ xfs_readonly_buftarg(mp->m_logdev_targp) ||
+ (mp->m_rtdev_targp && xfs_readonly_buftarg(mp->m_rtdev_targp))) {
+ cmn_err(CE_NOTE,
+ "XFS: %s required on read-only device.", message);
+ cmn_err(CE_NOTE,
+ "XFS: write access unavailable, cannot proceed.");
+ return EROFS;
+ }
+ return 0;
+}
#ifdef HAVE_PERCPU_SB
/*
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index 1df7e4502967..5761087ee8ea 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -78,7 +78,8 @@ typedef int (*xfs_send_destroy_t)(struct xfs_inode *, dm_right_t);
typedef int (*xfs_send_namesp_t)(dm_eventtype_t, struct xfs_mount *,
struct xfs_inode *, dm_right_t,
struct xfs_inode *, dm_right_t,
- const char *, const char *, mode_t, int, int);
+ const unsigned char *, const unsigned char *,
+ mode_t, int, int);
typedef int (*xfs_send_mount_t)(struct xfs_mount *, dm_right_t,
char *, char *);
typedef void (*xfs_send_unmount_t)(struct xfs_mount *, struct xfs_inode *,
@@ -207,8 +208,8 @@ typedef struct xfs_mount {
uint m_ag_maxlevels; /* XFS_AG_MAXLEVELS */
uint m_bm_maxlevels[2]; /* XFS_BM_MAXLEVELS */
uint m_in_maxlevels; /* max inobt btree levels. */
- struct xfs_perag *m_perag; /* per-ag accounting info */
- struct rw_semaphore m_peraglock; /* lock for m_perag (pointer) */
+ struct radix_tree_root m_perag_tree; /* per-ag accounting info */
+ spinlock_t m_perag_lock; /* lock for m_perag_tree */
struct mutex m_growlock; /* growfs mutex */
int m_fixedfsid[2]; /* unchanged for life of FS */
uint m_dmevmask; /* DMI events for this FS */
@@ -224,6 +225,7 @@ typedef struct xfs_mount {
__uint64_t m_maxioffset; /* maximum inode offset */
__uint64_t m_resblks; /* total reserved blocks */
__uint64_t m_resblks_avail;/* available reserved blocks */
+ __uint64_t m_resblks_save; /* reserved blks @ remount,ro */
int m_dalign; /* stripe unit */
int m_swidth; /* stripe width */
int m_sinoalign; /* stripe unit inode alignment */
@@ -243,7 +245,7 @@ typedef struct xfs_mount {
struct xfs_qmops *m_qm_ops; /* vector of XQM ops */
atomic_t m_active_trans; /* number trans frozen */
#ifdef HAVE_PERCPU_SB
- xfs_icsb_cnts_t *m_sb_cnts; /* per-cpu superblock counters */
+ xfs_icsb_cnts_t __percpu *m_sb_cnts; /* per-cpu superblock counters */
unsigned long m_icsb_counters; /* disabled per-cpu counters */
struct notifier_block m_icsb_notifier; /* hotplug cpu notifier */
struct mutex m_icsb_mutex; /* balancer sync lock */
@@ -257,6 +259,7 @@ typedef struct xfs_mount {
wait_queue_head_t m_wait_single_sync_task;
__int64_t m_update_flags; /* sb flags we need to update
on the next remount,rw */
+ struct shrinker m_inode_shrink; /* inode reclaim shrinker */
} xfs_mount_t;
/*
@@ -265,6 +268,7 @@ typedef struct xfs_mount {
#define XFS_MOUNT_WSYNC (1ULL << 0) /* for nfs - all metadata ops
must be synchronous except
for space allocations */
+#define XFS_MOUNT_DELAYLOG (1ULL << 1) /* delayed logging is enabled */
#define XFS_MOUNT_DMAPI (1ULL << 2) /* dmapi is enabled */
#define XFS_MOUNT_WAS_CLEAN (1ULL << 3)
#define XFS_MOUNT_FS_SHUTDOWN (1ULL << 4) /* atomic stop of all filesystem
@@ -384,19 +388,10 @@ xfs_daddr_to_agbno(struct xfs_mount *mp, xfs_daddr_t d)
}
/*
- * perag get/put wrappers for eventual ref counting
+ * perag get/put wrappers for ref counting
*/
-static inline xfs_perag_t *
-xfs_get_perag(struct xfs_mount *mp, xfs_ino_t ino)
-{
- return &mp->m_perag[XFS_INO_TO_AGNO(mp, ino)];
-}
-
-static inline void
-xfs_put_perag(struct xfs_mount *mp, xfs_perag_t *pag)
-{
- /* nothing to see here, move along */
-}
+struct xfs_perag *xfs_perag_get(struct xfs_mount *mp, xfs_agnumber_t agno);
+void xfs_perag_put(struct xfs_perag *pag);
/*
* Per-cpu superblock locking functions
@@ -428,6 +423,7 @@ typedef struct xfs_mod_sb {
} xfs_mod_sb_t;
extern int xfs_log_sbcount(xfs_mount_t *, uint);
+extern __uint64_t xfs_default_resblks(xfs_mount_t *mp);
extern int xfs_mountfs(xfs_mount_t *mp);
extern void xfs_unmountfs(xfs_mount_t *);
@@ -442,6 +438,8 @@ extern void xfs_freesb(xfs_mount_t *);
extern int xfs_fs_writable(xfs_mount_t *);
extern int xfs_sb_validate_fsb_count(struct xfs_sb *, __uint64_t);
+extern int xfs_dev_is_read_only(struct xfs_mount *, char *);
+
extern int xfs_dmops_get(struct xfs_mount *);
extern void xfs_dmops_put(struct xfs_mount *);
@@ -450,7 +448,8 @@ extern struct xfs_dmops xfs_dmcore_xfs;
#endif /* __KERNEL__ */
extern void xfs_mod_sb(struct xfs_trans *, __int64_t);
-extern xfs_agnumber_t xfs_initialize_perag(struct xfs_mount *, xfs_agnumber_t);
+extern int xfs_initialize_perag(struct xfs_mount *, xfs_agnumber_t,
+ xfs_agnumber_t *);
extern void xfs_sb_from_disk(struct xfs_sb *, struct xfs_dsb *);
extern void xfs_sb_to_disk(struct xfs_dsb *, struct xfs_sb *, __int64_t);
diff --git a/fs/xfs/xfs_mru_cache.c b/fs/xfs/xfs_mru_cache.c
index 4b0613d99faa..45ce15dc5b2b 100644
--- a/fs/xfs/xfs_mru_cache.c
+++ b/fs/xfs/xfs_mru_cache.c
@@ -398,7 +398,7 @@ exit:
* guaranteed that all the free functions for all the elements have finished
* executing and the reaper is not running.
*/
-void
+static void
xfs_mru_cache_flush(
xfs_mru_cache_t *mru)
{
diff --git a/fs/xfs/xfs_mru_cache.h b/fs/xfs/xfs_mru_cache.h
index 5d439f34b0c9..36dd3ec8b4eb 100644
--- a/fs/xfs/xfs_mru_cache.h
+++ b/fs/xfs/xfs_mru_cache.h
@@ -42,7 +42,6 @@ void xfs_mru_cache_uninit(void);
int xfs_mru_cache_create(struct xfs_mru_cache **mrup, unsigned int lifetime_ms,
unsigned int grp_count,
xfs_mru_cache_free_func_t free_func);
-void xfs_mru_cache_flush(xfs_mru_cache_t *mru);
void xfs_mru_cache_destroy(struct xfs_mru_cache *mru);
int xfs_mru_cache_insert(struct xfs_mru_cache *mru, unsigned long key,
void *value);
diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h
index 91bfd60f4c74..e0e64b113bd6 100644
--- a/fs/xfs/xfs_quota.h
+++ b/fs/xfs/xfs_quota.h
@@ -201,9 +201,6 @@ typedef struct xfs_qoff_logformat {
#define XFS_QMOPT_FORCE_RES 0x0000010 /* ignore quota limits */
#define XFS_QMOPT_DQSUSER 0x0000020 /* don't cache super users dquot */
#define XFS_QMOPT_SBVERSION 0x0000040 /* change superblock version num */
-#define XFS_QMOPT_QUOTAOFF 0x0000080 /* quotas are being turned off */
-#define XFS_QMOPT_UMOUNTING 0x0000100 /* filesys is being unmounted */
-#define XFS_QMOPT_DOLOG 0x0000200 /* log buf changes (in quotacheck) */
#define XFS_QMOPT_DOWARN 0x0000400 /* increase warning cnt if needed */
#define XFS_QMOPT_DQREPAIR 0x0001000 /* repair dquot if damaged */
#define XFS_QMOPT_GQUOTA 0x0002000 /* group dquot requested */
@@ -223,16 +220,9 @@ typedef struct xfs_qoff_logformat {
#define XFS_QMOPT_RES_INOS 0x0800000
/*
- * flags for dqflush and dqflush_all.
- */
-#define XFS_QMOPT_SYNC 0x1000000
-#define XFS_QMOPT_ASYNC 0x2000000
-#define XFS_QMOPT_DELWRI 0x4000000
-
-/*
* flags for dqalloc.
*/
-#define XFS_QMOPT_INHERIT 0x8000000
+#define XFS_QMOPT_INHERIT 0x1000000
/*
* flags to xfs_trans_mod_dquot.
diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index 6be05f756d59..a2d32ce335aa 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -2247,7 +2247,7 @@ xfs_rtmount_init(
cmn_err(CE_WARN, "XFS: realtime mount -- %llu != %llu",
(unsigned long long) XFS_BB_TO_FSB(mp, d),
(unsigned long long) mp->m_sb.sb_rblocks);
- return XFS_ERROR(E2BIG);
+ return XFS_ERROR(EFBIG);
}
error = xfs_read_buf(mp, mp->m_rtdev_targp,
d - XFS_FSB_TO_BB(mp, 1),
@@ -2256,7 +2256,7 @@ xfs_rtmount_init(
cmn_err(CE_WARN,
"XFS: realtime mount -- xfs_read_buf failed, returned %d", error);
if (error == ENOSPC)
- return XFS_ERROR(E2BIG);
+ return XFS_ERROR(EFBIG);
return error;
}
xfs_buf_relse(bp);
@@ -2277,12 +2277,12 @@ xfs_rtmount_inodes(
sbp = &mp->m_sb;
if (sbp->sb_rbmino == NULLFSINO)
return 0;
- error = xfs_iget(mp, NULL, sbp->sb_rbmino, 0, 0, &mp->m_rbmip, 0);
+ error = xfs_iget(mp, NULL, sbp->sb_rbmino, 0, 0, &mp->m_rbmip);
if (error)
return error;
ASSERT(mp->m_rbmip != NULL);
ASSERT(sbp->sb_rsumino != NULLFSINO);
- error = xfs_iget(mp, NULL, sbp->sb_rsumino, 0, 0, &mp->m_rsumip, 0);
+ error = xfs_iget(mp, NULL, sbp->sb_rsumino, 0, 0, &mp->m_rsumip);
if (error) {
IRELE(mp->m_rbmip);
return error;
diff --git a/fs/xfs/xfs_rtalloc.h b/fs/xfs/xfs_rtalloc.h
index b2d67adb6a08..ff614c29b441 100644
--- a/fs/xfs/xfs_rtalloc.h
+++ b/fs/xfs/xfs_rtalloc.h
@@ -147,7 +147,16 @@ xfs_growfs_rt(
# define xfs_rtfree_extent(t,b,l) (ENOSYS)
# define xfs_rtpick_extent(m,t,l,rb) (ENOSYS)
# define xfs_growfs_rt(mp,in) (ENOSYS)
-# define xfs_rtmount_init(m) (((mp)->m_sb.sb_rblocks == 0)? 0 : (ENOSYS))
+static inline int /* error */
+xfs_rtmount_init(
+ xfs_mount_t *mp) /* file system mount structure */
+{
+ if (mp->m_sb.sb_rblocks == 0)
+ return 0;
+
+ cmn_err(CE_WARN, "XFS: Not built with CONFIG_XFS_RT");
+ return ENOSYS;
+}
# define xfs_rtmount_inodes(m) (((mp)->m_sb.sb_rblocks == 0)? 0 : (ENOSYS))
# define xfs_rtunmount_inodes(m)
#endif /* CONFIG_XFS_RT */
diff --git a/fs/xfs/xfs_rw.c b/fs/xfs/xfs_rw.c
index 5aa07caea5f1..e336742a58a4 100644
--- a/fs/xfs/xfs_rw.c
+++ b/fs/xfs/xfs_rw.c
@@ -47,48 +47,6 @@
#include "xfs_trace.h"
/*
- * This is a subroutine for xfs_write() and other writers (xfs_ioctl)
- * which clears the setuid and setgid bits when a file is written.
- */
-int
-xfs_write_clear_setuid(
- xfs_inode_t *ip)
-{
- xfs_mount_t *mp;
- xfs_trans_t *tp;
- int error;
-
- mp = ip->i_mount;
- tp = xfs_trans_alloc(mp, XFS_TRANS_WRITEID);
- if ((error = xfs_trans_reserve(tp, 0,
- XFS_WRITEID_LOG_RES(mp),
- 0, 0, 0))) {
- 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);
- ip->i_d.di_mode &= ~S_ISUID;
-
- /*
- * Note that we don't have to worry about mandatory
- * file locking being disabled here because we only
- * clear the S_ISGID bit if the Group execute bit is
- * on, but if it was on then mandatory locking wouldn't
- * have been enabled.
- */
- if (ip->i_d.di_mode & S_IXGRP) {
- ip->i_d.di_mode &= ~S_ISGID;
- }
- xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
- xfs_trans_set_sync(tp);
- error = xfs_trans_commit(tp, 0);
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- return 0;
-}
-
-/*
* Force a shutdown of the filesystem instantly while keeping
* the filesystem consistent. We don't do an unmount here; just shutdown
* the shop, make sure that absolutely nothing persistent happens to
@@ -153,88 +111,6 @@ xfs_do_force_shutdown(
}
}
-
-/*
- * Called when we want to stop a buffer from getting written or read.
- * We attach the EIO error, muck with its flags, and call biodone
- * so that the proper iodone callbacks get called.
- */
-int
-xfs_bioerror(
- xfs_buf_t *bp)
-{
-
-#ifdef XFSERRORDEBUG
- ASSERT(XFS_BUF_ISREAD(bp) || bp->b_iodone);
-#endif
-
- /*
- * No need to wait until the buffer is unpinned.
- * We aren't flushing it.
- */
- XFS_BUF_ERROR(bp, EIO);
- /*
- * We're calling biodone, so delete B_DONE flag. Either way
- * we have to call the iodone callback, and calling biodone
- * probably is the best way since it takes care of
- * GRIO as well.
- */
- XFS_BUF_UNREAD(bp);
- XFS_BUF_UNDELAYWRITE(bp);
- XFS_BUF_UNDONE(bp);
- XFS_BUF_STALE(bp);
-
- XFS_BUF_CLR_BDSTRAT_FUNC(bp);
- xfs_biodone(bp);
-
- return (EIO);
-}
-
-/*
- * Same as xfs_bioerror, except that we are releasing the buffer
- * here ourselves, and avoiding the biodone call.
- * This is meant for userdata errors; metadata bufs come with
- * iodone functions attached, so that we can track down errors.
- */
-int
-xfs_bioerror_relse(
- xfs_buf_t *bp)
-{
- int64_t fl;
-
- ASSERT(XFS_BUF_IODONE_FUNC(bp) != xfs_buf_iodone_callbacks);
- ASSERT(XFS_BUF_IODONE_FUNC(bp) != xlog_iodone);
-
- fl = XFS_BUF_BFLAGS(bp);
- /*
- * No need to wait until the buffer is unpinned.
- * We aren't flushing it.
- *
- * chunkhold expects B_DONE to be set, whether
- * we actually finish the I/O or not. We don't want to
- * change that interface.
- */
- XFS_BUF_UNREAD(bp);
- XFS_BUF_UNDELAYWRITE(bp);
- XFS_BUF_DONE(bp);
- XFS_BUF_STALE(bp);
- XFS_BUF_CLR_IODONE_FUNC(bp);
- XFS_BUF_CLR_BDSTRAT_FUNC(bp);
- if (!(fl & XFS_B_ASYNC)) {
- /*
- * Mark b_error and B_ERROR _both_.
- * Lot's of chunkcache code assumes that.
- * There's no reason to mark error for
- * ASYNC buffers.
- */
- XFS_BUF_ERROR(bp, EIO);
- XFS_BUF_FINISH_IOWAIT(bp);
- } else {
- xfs_buf_relse(bp);
- }
- return (EIO);
-}
-
/*
* Prints out an ALERT message about I/O error.
*/
@@ -306,37 +182,6 @@ xfs_read_buf(
}
/*
- * Wrapper around bwrite() so that we can trap
- * write errors, and act accordingly.
- */
-int
-xfs_bwrite(
- struct xfs_mount *mp,
- struct xfs_buf *bp)
-{
- int error;
-
- /*
- * XXXsup how does this work for quotas.
- */
- XFS_BUF_SET_BDSTRAT_FUNC(bp, xfs_bdstrat_cb);
- bp->b_mount = mp;
- XFS_BUF_WRITE(bp);
-
- if ((error = XFS_bwrite(bp))) {
- ASSERT(mp);
- /*
- * Cannot put a buftrace here since if the buffer is not
- * B_HOLD then we will brelse() the buffer before returning
- * from bwrite and we could be tracing a buffer that has
- * been reused.
- */
- xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
- }
- return (error);
-}
-
-/*
* helper function to extract extent size hint from inode
*/
xfs_extlen_t
diff --git a/fs/xfs/xfs_rw.h b/fs/xfs/xfs_rw.h
index 571f2174435c..11c41ec6ed75 100644
--- a/fs/xfs/xfs_rw.h
+++ b/fs/xfs/xfs_rw.h
@@ -39,10 +39,6 @@ xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb)
/*
* Prototypes for functions in xfs_rw.c.
*/
-extern int xfs_write_clear_setuid(struct xfs_inode *ip);
-extern int xfs_bwrite(struct xfs_mount *mp, struct xfs_buf *bp);
-extern int xfs_bioerror(struct xfs_buf *bp);
-extern int xfs_bioerror_relse(struct xfs_buf *bp);
extern int xfs_read_buf(struct xfs_mount *mp, xfs_buftarg_t *btp,
xfs_daddr_t blkno, int len, uint flags,
struct xfs_buf **bpp);
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index 237badcbac3b..28547dfce037 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -44,148 +44,493 @@
#include "xfs_trans_priv.h"
#include "xfs_trans_space.h"
#include "xfs_inode_item.h"
-
-
-STATIC void xfs_trans_apply_sb_deltas(xfs_trans_t *);
-STATIC uint xfs_trans_count_vecs(xfs_trans_t *);
-STATIC void xfs_trans_fill_vecs(xfs_trans_t *, xfs_log_iovec_t *);
-STATIC void xfs_trans_uncommit(xfs_trans_t *, uint);
-STATIC void xfs_trans_committed(xfs_trans_t *, int);
-STATIC void xfs_trans_chunk_committed(xfs_log_item_chunk_t *, xfs_lsn_t, int);
-STATIC void xfs_trans_free(xfs_trans_t *);
+#include "xfs_trace.h"
kmem_zone_t *xfs_trans_zone;
/*
- * Reservation functions here avoid a huge stack in xfs_trans_init
- * due to register overflow from temporaries in the calculations.
+ * Various log reservation values.
+ *
+ * These are based on the size of the file system block because that is what
+ * most transactions manipulate. Each adds in an additional 128 bytes per
+ * item logged to try to account for the overhead of the transaction mechanism.
+ *
+ * Note: Most of the reservations underestimate the number of allocation
+ * groups into which they could free extents in the xfs_bmap_finish() call.
+ * This is because the number in the worst case is quite high and quite
+ * unusual. In order to fix this we need to change xfs_bmap_finish() to free
+ * extents in only a single AG at a time. This will require changes to the
+ * EFI code as well, however, so that the EFI for the extents not freed is
+ * logged again in each transaction. See SGI PV #261917.
+ *
+ * Reservation functions here avoid a huge stack in xfs_trans_init due to
+ * register overflow from temporaries in the calculations.
*/
+
+/*
+ * In a write transaction we can allocate a maximum of 2
+ * extents. This gives:
+ * the inode getting the new extents: inode size
+ * the inode's bmap btree: max depth * block size
+ * the agfs of the ags from which the extents are allocated: 2 * sector
+ * the superblock free block counter: sector size
+ * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size
+ * And the bmap_finish transaction can free bmap blocks in a join:
+ * the agfs of the ags containing the blocks: 2 * sector size
+ * the agfls of the ags containing the blocks: 2 * sector size
+ * the super block free block counter: sector size
+ * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size
+ */
STATIC uint
-xfs_calc_write_reservation(xfs_mount_t *mp)
+xfs_calc_write_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_WRITE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+ return XFS_DQUOT_LOGRES(mp) +
+ MAX((mp->m_sb.sb_inodesize +
+ XFS_FSB_TO_B(mp, XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK)) +
+ 2 * mp->m_sb.sb_sectsize +
+ mp->m_sb.sb_sectsize +
+ XFS_ALLOCFREE_LOG_RES(mp, 2) +
+ 128 * (4 + XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) +
+ XFS_ALLOCFREE_LOG_COUNT(mp, 2))),
+ (2 * mp->m_sb.sb_sectsize +
+ 2 * mp->m_sb.sb_sectsize +
+ mp->m_sb.sb_sectsize +
+ XFS_ALLOCFREE_LOG_RES(mp, 2) +
+ 128 * (5 + XFS_ALLOCFREE_LOG_COUNT(mp, 2))));
}
+/*
+ * In truncating a file we free up to two extents at once. We can modify:
+ * the inode being truncated: inode size
+ * the inode's bmap btree: (max depth + 1) * block size
+ * And the bmap_finish transaction can free the blocks and bmap blocks:
+ * the agf for each of the ags: 4 * sector size
+ * the agfl for each of the ags: 4 * sector size
+ * the super block to reflect the freed blocks: sector size
+ * worst case split in allocation btrees per extent assuming 4 extents:
+ * 4 exts * 2 trees * (2 * max depth - 1) * block size
+ * the inode btree: max depth * blocksize
+ * the allocation btrees: 2 trees * (max depth - 1) * block size
+ */
STATIC uint
-xfs_calc_itruncate_reservation(xfs_mount_t *mp)
+xfs_calc_itruncate_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_ITRUNCATE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+ return XFS_DQUOT_LOGRES(mp) +
+ MAX((mp->m_sb.sb_inodesize +
+ XFS_FSB_TO_B(mp, XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) + 1) +
+ 128 * (2 + XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK))),
+ (4 * mp->m_sb.sb_sectsize +
+ 4 * mp->m_sb.sb_sectsize +
+ mp->m_sb.sb_sectsize +
+ XFS_ALLOCFREE_LOG_RES(mp, 4) +
+ 128 * (9 + XFS_ALLOCFREE_LOG_COUNT(mp, 4)) +
+ 128 * 5 +
+ XFS_ALLOCFREE_LOG_RES(mp, 1) +
+ 128 * (2 + XFS_IALLOC_BLOCKS(mp) + mp->m_in_maxlevels +
+ XFS_ALLOCFREE_LOG_COUNT(mp, 1))));
}
+/*
+ * In renaming a files we can modify:
+ * the four inodes involved: 4 * inode size
+ * the two directory btrees: 2 * (max depth + v2) * dir block size
+ * the two directory bmap btrees: 2 * max depth * block size
+ * And the bmap_finish transaction can free dir and bmap blocks (two sets
+ * of bmap blocks) giving:
+ * the agf for the ags in which the blocks live: 3 * sector size
+ * the agfl for the ags in which the blocks live: 3 * sector size
+ * the superblock for the free block count: sector size
+ * the allocation btrees: 3 exts * 2 trees * (2 * max depth - 1) * block size
+ */
STATIC uint
-xfs_calc_rename_reservation(xfs_mount_t *mp)
+xfs_calc_rename_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_RENAME_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+ return XFS_DQUOT_LOGRES(mp) +
+ MAX((4 * mp->m_sb.sb_inodesize +
+ 2 * XFS_DIROP_LOG_RES(mp) +
+ 128 * (4 + 2 * XFS_DIROP_LOG_COUNT(mp))),
+ (3 * mp->m_sb.sb_sectsize +
+ 3 * mp->m_sb.sb_sectsize +
+ mp->m_sb.sb_sectsize +
+ XFS_ALLOCFREE_LOG_RES(mp, 3) +
+ 128 * (7 + XFS_ALLOCFREE_LOG_COUNT(mp, 3))));
}
+/*
+ * For creating a link to an inode:
+ * the parent directory inode: inode size
+ * the linked inode: inode size
+ * the directory btree could split: (max depth + v2) * dir block size
+ * the directory bmap btree could join or split: (max depth + v2) * blocksize
+ * And the bmap_finish transaction can free some bmap blocks giving:
+ * the agf for the ag in which the blocks live: sector size
+ * the agfl for the ag in which the blocks live: sector size
+ * the superblock for the free block count: sector size
+ * the allocation btrees: 2 trees * (2 * max depth - 1) * block size
+ */
STATIC uint
-xfs_calc_link_reservation(xfs_mount_t *mp)
+xfs_calc_link_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_LINK_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+ return XFS_DQUOT_LOGRES(mp) +
+ MAX((mp->m_sb.sb_inodesize +
+ mp->m_sb.sb_inodesize +
+ XFS_DIROP_LOG_RES(mp) +
+ 128 * (2 + XFS_DIROP_LOG_COUNT(mp))),
+ (mp->m_sb.sb_sectsize +
+ mp->m_sb.sb_sectsize +
+ mp->m_sb.sb_sectsize +
+ XFS_ALLOCFREE_LOG_RES(mp, 1) +
+ 128 * (3 + XFS_ALLOCFREE_LOG_COUNT(mp, 1))));
}
+/*
+ * For removing a directory entry we can modify:
+ * the parent directory inode: inode size
+ * the removed inode: inode size
+ * the directory btree could join: (max depth + v2) * dir block size
+ * the directory bmap btree could join or split: (max depth + v2) * blocksize
+ * And the bmap_finish transaction can free the dir and bmap blocks giving:
+ * the agf for the ag in which the blocks live: 2 * sector size
+ * the agfl for the ag in which the blocks live: 2 * sector size
+ * the superblock for the free block count: sector size
+ * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size
+ */
STATIC uint
-xfs_calc_remove_reservation(xfs_mount_t *mp)
+xfs_calc_remove_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_REMOVE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+ return XFS_DQUOT_LOGRES(mp) +
+ MAX((mp->m_sb.sb_inodesize +
+ mp->m_sb.sb_inodesize +
+ XFS_DIROP_LOG_RES(mp) +
+ 128 * (2 + XFS_DIROP_LOG_COUNT(mp))),
+ (2 * mp->m_sb.sb_sectsize +
+ 2 * mp->m_sb.sb_sectsize +
+ mp->m_sb.sb_sectsize +
+ XFS_ALLOCFREE_LOG_RES(mp, 2) +
+ 128 * (5 + XFS_ALLOCFREE_LOG_COUNT(mp, 2))));
}
+/*
+ * For symlink we can modify:
+ * the parent directory inode: inode size
+ * the new inode: inode size
+ * the inode btree entry: 1 block
+ * the directory btree: (max depth + v2) * dir block size
+ * the directory inode's bmap btree: (max depth + v2) * block size
+ * the blocks for the symlink: 1 kB
+ * Or in the first xact we allocate some inodes giving:
+ * the agi and agf of the ag getting the new inodes: 2 * sectorsize
+ * the inode blocks allocated: XFS_IALLOC_BLOCKS * blocksize
+ * the inode btree: max depth * blocksize
+ * the allocation btrees: 2 trees * (2 * max depth - 1) * block size
+ */
STATIC uint
-xfs_calc_symlink_reservation(xfs_mount_t *mp)
+xfs_calc_symlink_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_SYMLINK_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+ return XFS_DQUOT_LOGRES(mp) +
+ MAX((mp->m_sb.sb_inodesize +
+ mp->m_sb.sb_inodesize +
+ XFS_FSB_TO_B(mp, 1) +
+ XFS_DIROP_LOG_RES(mp) +
+ 1024 +
+ 128 * (4 + XFS_DIROP_LOG_COUNT(mp))),
+ (2 * mp->m_sb.sb_sectsize +
+ XFS_FSB_TO_B(mp, XFS_IALLOC_BLOCKS(mp)) +
+ XFS_FSB_TO_B(mp, mp->m_in_maxlevels) +
+ XFS_ALLOCFREE_LOG_RES(mp, 1) +
+ 128 * (2 + XFS_IALLOC_BLOCKS(mp) + mp->m_in_maxlevels +
+ XFS_ALLOCFREE_LOG_COUNT(mp, 1))));
}
+/*
+ * For create we can modify:
+ * the parent directory inode: inode size
+ * the new inode: inode size
+ * the inode btree entry: block size
+ * the superblock for the nlink flag: sector size
+ * the directory btree: (max depth + v2) * dir block size
+ * the directory inode's bmap btree: (max depth + v2) * block size
+ * Or in the first xact we allocate some inodes giving:
+ * the agi and agf of the ag getting the new inodes: 2 * sectorsize
+ * the superblock for the nlink flag: sector size
+ * the inode blocks allocated: XFS_IALLOC_BLOCKS * blocksize
+ * the inode btree: max depth * blocksize
+ * the allocation btrees: 2 trees * (max depth - 1) * block size
+ */
STATIC uint
-xfs_calc_create_reservation(xfs_mount_t *mp)
+xfs_calc_create_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_CREATE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+ return XFS_DQUOT_LOGRES(mp) +
+ MAX((mp->m_sb.sb_inodesize +
+ mp->m_sb.sb_inodesize +
+ mp->m_sb.sb_sectsize +
+ XFS_FSB_TO_B(mp, 1) +
+ XFS_DIROP_LOG_RES(mp) +
+ 128 * (3 + XFS_DIROP_LOG_COUNT(mp))),
+ (3 * mp->m_sb.sb_sectsize +
+ XFS_FSB_TO_B(mp, XFS_IALLOC_BLOCKS(mp)) +
+ XFS_FSB_TO_B(mp, mp->m_in_maxlevels) +
+ XFS_ALLOCFREE_LOG_RES(mp, 1) +
+ 128 * (2 + XFS_IALLOC_BLOCKS(mp) + mp->m_in_maxlevels +
+ XFS_ALLOCFREE_LOG_COUNT(mp, 1))));
}
+/*
+ * Making a new directory is the same as creating a new file.
+ */
STATIC uint
-xfs_calc_mkdir_reservation(xfs_mount_t *mp)
+xfs_calc_mkdir_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_MKDIR_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+ return xfs_calc_create_reservation(mp);
}
+/*
+ * In freeing an inode we can modify:
+ * the inode being freed: inode size
+ * the super block free inode counter: sector size
+ * the agi hash list and counters: sector size
+ * the inode btree entry: block size
+ * the on disk inode before ours in the agi hash list: inode cluster size
+ * the inode btree: max depth * blocksize
+ * the allocation btrees: 2 trees * (max depth - 1) * block size
+ */
STATIC uint
-xfs_calc_ifree_reservation(xfs_mount_t *mp)
+xfs_calc_ifree_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_IFREE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+ return XFS_DQUOT_LOGRES(mp) +
+ mp->m_sb.sb_inodesize +
+ mp->m_sb.sb_sectsize +
+ mp->m_sb.sb_sectsize +
+ XFS_FSB_TO_B(mp, 1) +
+ MAX((__uint16_t)XFS_FSB_TO_B(mp, 1),
+ XFS_INODE_CLUSTER_SIZE(mp)) +
+ 128 * 5 +
+ XFS_ALLOCFREE_LOG_RES(mp, 1) +
+ 128 * (2 + XFS_IALLOC_BLOCKS(mp) + mp->m_in_maxlevels +
+ XFS_ALLOCFREE_LOG_COUNT(mp, 1));
}
+/*
+ * When only changing the inode we log the inode and possibly the superblock
+ * We also add a bit of slop for the transaction stuff.
+ */
STATIC uint
-xfs_calc_ichange_reservation(xfs_mount_t *mp)
+xfs_calc_ichange_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_ICHANGE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+ return XFS_DQUOT_LOGRES(mp) +
+ mp->m_sb.sb_inodesize +
+ mp->m_sb.sb_sectsize +
+ 512;
+
}
+/*
+ * Growing the data section of the filesystem.
+ * superblock
+ * agi and agf
+ * allocation btrees
+ */
STATIC uint
-xfs_calc_growdata_reservation(xfs_mount_t *mp)
+xfs_calc_growdata_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_GROWDATA_LOG_RES(mp);
+ return mp->m_sb.sb_sectsize * 3 +
+ XFS_ALLOCFREE_LOG_RES(mp, 1) +
+ 128 * (3 + XFS_ALLOCFREE_LOG_COUNT(mp, 1));
}
+/*
+ * Growing the rt section of the filesystem.
+ * In the first set of transactions (ALLOC) we allocate space to the
+ * bitmap or summary files.
+ * superblock: sector size
+ * agf of the ag from which the extent is allocated: sector size
+ * bmap btree for bitmap/summary inode: max depth * blocksize
+ * bitmap/summary inode: inode size
+ * allocation btrees for 1 block alloc: 2 * (2 * maxdepth - 1) * blocksize
+ */
STATIC uint
-xfs_calc_growrtalloc_reservation(xfs_mount_t *mp)
+xfs_calc_growrtalloc_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_GROWRTALLOC_LOG_RES(mp);
+ return 2 * mp->m_sb.sb_sectsize +
+ XFS_FSB_TO_B(mp, XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK)) +
+ mp->m_sb.sb_inodesize +
+ XFS_ALLOCFREE_LOG_RES(mp, 1) +
+ 128 * (3 + XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) +
+ XFS_ALLOCFREE_LOG_COUNT(mp, 1));
}
+/*
+ * Growing the rt section of the filesystem.
+ * In the second set of transactions (ZERO) we zero the new metadata blocks.
+ * one bitmap/summary block: blocksize
+ */
STATIC uint
-xfs_calc_growrtzero_reservation(xfs_mount_t *mp)
+xfs_calc_growrtzero_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_GROWRTZERO_LOG_RES(mp);
+ return mp->m_sb.sb_blocksize + 128;
}
+/*
+ * Growing the rt section of the filesystem.
+ * In the third set of transactions (FREE) we update metadata without
+ * allocating any new blocks.
+ * superblock: sector size
+ * bitmap inode: inode size
+ * summary inode: inode size
+ * one bitmap block: blocksize
+ * summary blocks: new summary size
+ */
STATIC uint
-xfs_calc_growrtfree_reservation(xfs_mount_t *mp)
+xfs_calc_growrtfree_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_GROWRTFREE_LOG_RES(mp);
+ return mp->m_sb.sb_sectsize +
+ 2 * mp->m_sb.sb_inodesize +
+ mp->m_sb.sb_blocksize +
+ mp->m_rsumsize +
+ 128 * 5;
}
+/*
+ * Logging the inode modification timestamp on a synchronous write.
+ * inode
+ */
STATIC uint
-xfs_calc_swrite_reservation(xfs_mount_t *mp)
+xfs_calc_swrite_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_SWRITE_LOG_RES(mp);
+ return mp->m_sb.sb_inodesize + 128;
}
+/*
+ * Logging the inode mode bits when writing a setuid/setgid file
+ * inode
+ */
STATIC uint
xfs_calc_writeid_reservation(xfs_mount_t *mp)
{
- return XFS_CALC_WRITEID_LOG_RES(mp);
+ return mp->m_sb.sb_inodesize + 128;
}
+/*
+ * Converting the inode from non-attributed to attributed.
+ * the inode being converted: inode size
+ * agf block and superblock (for block allocation)
+ * the new block (directory sized)
+ * bmap blocks for the new directory block
+ * allocation btrees
+ */
STATIC uint
-xfs_calc_addafork_reservation(xfs_mount_t *mp)
+xfs_calc_addafork_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_ADDAFORK_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+ return XFS_DQUOT_LOGRES(mp) +
+ mp->m_sb.sb_inodesize +
+ mp->m_sb.sb_sectsize * 2 +
+ mp->m_dirblksize +
+ XFS_FSB_TO_B(mp, XFS_DAENTER_BMAP1B(mp, XFS_DATA_FORK) + 1) +
+ XFS_ALLOCFREE_LOG_RES(mp, 1) +
+ 128 * (4 + XFS_DAENTER_BMAP1B(mp, XFS_DATA_FORK) + 1 +
+ XFS_ALLOCFREE_LOG_COUNT(mp, 1));
}
+/*
+ * Removing the attribute fork of a file
+ * the inode being truncated: inode size
+ * the inode's bmap btree: max depth * block size
+ * And the bmap_finish transaction can free the blocks and bmap blocks:
+ * the agf for each of the ags: 4 * sector size
+ * the agfl for each of the ags: 4 * sector size
+ * the super block to reflect the freed blocks: sector size
+ * worst case split in allocation btrees per extent assuming 4 extents:
+ * 4 exts * 2 trees * (2 * max depth - 1) * block size
+ */
STATIC uint
-xfs_calc_attrinval_reservation(xfs_mount_t *mp)
+xfs_calc_attrinval_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_ATTRINVAL_LOG_RES(mp);
+ return MAX((mp->m_sb.sb_inodesize +
+ XFS_FSB_TO_B(mp, XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK)) +
+ 128 * (1 + XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK))),
+ (4 * mp->m_sb.sb_sectsize +
+ 4 * mp->m_sb.sb_sectsize +
+ mp->m_sb.sb_sectsize +
+ XFS_ALLOCFREE_LOG_RES(mp, 4) +
+ 128 * (9 + XFS_ALLOCFREE_LOG_COUNT(mp, 4))));
}
+/*
+ * Setting an attribute.
+ * the inode getting the attribute
+ * the superblock for allocations
+ * the agfs extents are allocated from
+ * the attribute btree * max depth
+ * the inode allocation btree
+ * Since attribute transaction space is dependent on the size of the attribute,
+ * the calculation is done partially at mount time and partially at runtime.
+ */
STATIC uint
-xfs_calc_attrset_reservation(xfs_mount_t *mp)
+xfs_calc_attrset_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_ATTRSET_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+ return XFS_DQUOT_LOGRES(mp) +
+ mp->m_sb.sb_inodesize +
+ mp->m_sb.sb_sectsize +
+ XFS_FSB_TO_B(mp, XFS_DA_NODE_MAXDEPTH) +
+ 128 * (2 + XFS_DA_NODE_MAXDEPTH);
}
+/*
+ * Removing an attribute.
+ * the inode: inode size
+ * the attribute btree could join: max depth * block size
+ * the inode bmap btree could join or split: max depth * block size
+ * And the bmap_finish transaction can free the attr blocks freed giving:
+ * the agf for the ag in which the blocks live: 2 * sector size
+ * the agfl for the ag in which the blocks live: 2 * sector size
+ * the superblock for the free block count: sector size
+ * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size
+ */
STATIC uint
-xfs_calc_attrrm_reservation(xfs_mount_t *mp)
+xfs_calc_attrrm_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_ATTRRM_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+ return XFS_DQUOT_LOGRES(mp) +
+ MAX((mp->m_sb.sb_inodesize +
+ XFS_FSB_TO_B(mp, XFS_DA_NODE_MAXDEPTH) +
+ XFS_FSB_TO_B(mp, XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK)) +
+ 128 * (1 + XFS_DA_NODE_MAXDEPTH +
+ XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK))),
+ (2 * mp->m_sb.sb_sectsize +
+ 2 * mp->m_sb.sb_sectsize +
+ mp->m_sb.sb_sectsize +
+ XFS_ALLOCFREE_LOG_RES(mp, 2) +
+ 128 * (5 + XFS_ALLOCFREE_LOG_COUNT(mp, 2))));
}
+/*
+ * Clearing a bad agino number in an agi hash bucket.
+ */
STATIC uint
-xfs_calc_clear_agi_bucket_reservation(xfs_mount_t *mp)
+xfs_calc_clear_agi_bucket_reservation(
+ struct xfs_mount *mp)
{
- return XFS_CALC_CLEAR_AGI_BUCKET_LOG_RES(mp);
+ return mp->m_sb.sb_sectsize + 128;
}
/*
@@ -194,11 +539,10 @@ xfs_calc_clear_agi_bucket_reservation(xfs_mount_t *mp)
*/
void
xfs_trans_init(
- xfs_mount_t *mp)
+ struct xfs_mount *mp)
{
- xfs_trans_reservations_t *resp;
+ struct xfs_trans_reservations *resp = &mp->m_reservations;
- resp = &(mp->m_reservations);
resp->tr_write = xfs_calc_write_reservation(mp);
resp->tr_itruncate = xfs_calc_itruncate_reservation(mp);
resp->tr_rename = xfs_calc_rename_reservation(mp);
@@ -254,13 +598,30 @@ _xfs_trans_alloc(
tp->t_type = type;
tp->t_mountp = mp;
tp->t_items_free = XFS_LIC_NUM_SLOTS;
- tp->t_busy_free = XFS_LBC_NUM_SLOTS;
xfs_lic_init(&(tp->t_items));
- XFS_LBC_INIT(&(tp->t_busy));
+ INIT_LIST_HEAD(&tp->t_busy);
return tp;
}
/*
+ * Free the transaction structure. If there is more clean up
+ * to do when the structure is freed, add it here.
+ */
+STATIC void
+xfs_trans_free(
+ struct xfs_trans *tp)
+{
+ struct xfs_busy_extent *busyp, *n;
+
+ list_for_each_entry_safe(busyp, n, &tp->t_busy, list)
+ xfs_alloc_busy_clear(tp->t_mountp, busyp);
+
+ atomic_dec(&tp->t_mountp->m_active_trans);
+ xfs_trans_free_dqinfo(tp);
+ kmem_zone_free(xfs_trans_zone, tp);
+}
+
+/*
* This is called to create a new transaction which will share the
* permanent log reservation of the given transaction. The remaining
* unused block and rt extent reservations are also inherited. This
@@ -283,9 +644,8 @@ xfs_trans_dup(
ntp->t_type = tp->t_type;
ntp->t_mountp = tp->t_mountp;
ntp->t_items_free = XFS_LIC_NUM_SLOTS;
- ntp->t_busy_free = XFS_LBC_NUM_SLOTS;
xfs_lic_init(&(ntp->t_items));
- XFS_LBC_INIT(&(ntp->t_busy));
+ INIT_LIST_HEAD(&ntp->t_busy);
ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
ASSERT(tp->t_ticket != NULL);
@@ -421,7 +781,6 @@ undo_blocks:
return error;
}
-
/*
* Record the indicated change to the given field for application
* to the file system's superblock when the transaction commits.
@@ -650,7 +1009,7 @@ xfs_trans_apply_sb_deltas(
* XFS_TRANS_SB_DIRTY will not be set when the transaction is updated but we
* still need to update the incore superblock with the changes.
*/
-STATIC void
+void
xfs_trans_unreserve_and_mod_sb(
xfs_trans_t *tp)
{
@@ -764,94 +1123,256 @@ xfs_trans_unreserve_and_mod_sb(
}
}
+/*
+ * Total up the number of log iovecs needed to commit this
+ * transaction. The transaction itself needs one for the
+ * transaction header. Ask each dirty item in turn how many
+ * it needs to get the total.
+ */
+static uint
+xfs_trans_count_vecs(
+ struct xfs_trans *tp)
+{
+ int nvecs;
+ xfs_log_item_desc_t *lidp;
+
+ nvecs = 1;
+ lidp = xfs_trans_first_item(tp);
+ ASSERT(lidp != NULL);
+
+ /* In the non-debug case we need to start bailing out if we
+ * didn't find a log_item here, return zero and let trans_commit
+ * deal with it.
+ */
+ if (lidp == NULL)
+ return 0;
+
+ while (lidp != NULL) {
+ /*
+ * Skip items which aren't dirty in this transaction.
+ */
+ if (!(lidp->lid_flags & XFS_LID_DIRTY)) {
+ lidp = xfs_trans_next_item(tp, lidp);
+ continue;
+ }
+ lidp->lid_size = IOP_SIZE(lidp->lid_item);
+ nvecs += lidp->lid_size;
+ lidp = xfs_trans_next_item(tp, lidp);
+ }
+
+ return nvecs;
+}
/*
- * xfs_trans_commit
+ * Fill in the vector with pointers to data to be logged
+ * by this transaction. The transaction header takes
+ * the first vector, and then each dirty item takes the
+ * number of vectors it indicated it needed in xfs_trans_count_vecs().
*
- * Commit the given transaction to the log a/synchronously.
+ * As each item fills in the entries it needs, also pin the item
+ * so that it cannot be flushed out until the log write completes.
+ */
+static void
+xfs_trans_fill_vecs(
+ struct xfs_trans *tp,
+ struct xfs_log_iovec *log_vector)
+{
+ xfs_log_item_desc_t *lidp;
+ struct xfs_log_iovec *vecp;
+ uint nitems;
+
+ /*
+ * Skip over the entry for the transaction header, we'll
+ * fill that in at the end.
+ */
+ vecp = log_vector + 1;
+
+ nitems = 0;
+ lidp = xfs_trans_first_item(tp);
+ ASSERT(lidp);
+ while (lidp) {
+ /* Skip items which aren't dirty in this transaction. */
+ if (!(lidp->lid_flags & XFS_LID_DIRTY)) {
+ lidp = xfs_trans_next_item(tp, lidp);
+ continue;
+ }
+
+ /*
+ * The item may be marked dirty but not log anything. This can
+ * be used to get called when a transaction is committed.
+ */
+ if (lidp->lid_size)
+ nitems++;
+ IOP_FORMAT(lidp->lid_item, vecp);
+ vecp += lidp->lid_size;
+ IOP_PIN(lidp->lid_item);
+ lidp = xfs_trans_next_item(tp, lidp);
+ }
+
+ /*
+ * Now that we've counted the number of items in this transaction, fill
+ * in the transaction header. Note that the transaction header does not
+ * have a log item.
+ */
+ tp->t_header.th_magic = XFS_TRANS_HEADER_MAGIC;
+ tp->t_header.th_type = tp->t_type;
+ tp->t_header.th_num_items = nitems;
+ log_vector->i_addr = (xfs_caddr_t)&tp->t_header;
+ log_vector->i_len = sizeof(xfs_trans_header_t);
+ log_vector->i_type = XLOG_REG_TYPE_TRANSHDR;
+}
+
+/*
+ * The committed item processing consists of calling the committed routine of
+ * each logged item, updating the item's position in the AIL if necessary, and
+ * unpinning each item. If the committed routine returns -1, then do nothing
+ * further with the item because it may have been freed.
*
- * XFS disk error handling mechanism is not based on a typical
- * transaction abort mechanism. Logically after the filesystem
- * gets marked 'SHUTDOWN', we can't let any new transactions
- * be durable - ie. committed to disk - because some metadata might
- * be inconsistent. In such cases, this returns an error, and the
- * caller may assume that all locked objects joined to the transaction
- * have already been unlocked as if the commit had succeeded.
- * Do not reference the transaction structure after this call.
+ * Since items are unlocked when they are copied to the incore log, it is
+ * possible for two transactions to be completing and manipulating the same
+ * item simultaneously. The AIL lock will protect the lsn field of each item.
+ * The value of this field can never go backwards.
+ *
+ * We unpin the items after repositioning them in the AIL, because otherwise
+ * they could be immediately flushed and we'd have to race with the flusher
+ * trying to pull the item from the AIL as we add it.
*/
- /*ARGSUSED*/
-int
-_xfs_trans_commit(
- xfs_trans_t *tp,
- uint flags,
- int *log_flushed)
+void
+xfs_trans_item_committed(
+ struct xfs_log_item *lip,
+ xfs_lsn_t commit_lsn,
+ int aborted)
{
- xfs_log_iovec_t *log_vector;
- int nvec;
- xfs_mount_t *mp;
- xfs_lsn_t commit_lsn;
- /* REFERENCED */
- int error;
- int log_flags;
- int sync;
-#define XFS_TRANS_LOGVEC_COUNT 16
- xfs_log_iovec_t log_vector_fast[XFS_TRANS_LOGVEC_COUNT];
- void *commit_iclog;
- int shutdown;
+ xfs_lsn_t item_lsn;
+ struct xfs_ail *ailp;
+
+ if (aborted)
+ lip->li_flags |= XFS_LI_ABORTED;
+ item_lsn = IOP_COMMITTED(lip, commit_lsn);
- commit_lsn = -1;
+ /* If the committed routine returns -1, item has been freed. */
+ if (XFS_LSN_CMP(item_lsn, (xfs_lsn_t)-1) == 0)
+ return;
/*
- * Determine whether this commit is releasing a permanent
- * log reservation or not.
+ * If the returned lsn is greater than what it contained before, update
+ * the location of the item in the AIL. If it is not, then do nothing.
+ * Items can never move backwards in the AIL.
+ *
+ * While the new lsn should usually be greater, it is possible that a
+ * later transaction completing simultaneously with an earlier one
+ * using the same item could complete first with a higher lsn. This
+ * would cause the earlier transaction to fail the test below.
*/
- if (flags & XFS_TRANS_RELEASE_LOG_RES) {
- ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
- log_flags = XFS_LOG_REL_PERM_RESERV;
+ ailp = lip->li_ailp;
+ spin_lock(&ailp->xa_lock);
+ if (XFS_LSN_CMP(item_lsn, lip->li_lsn) > 0) {
+ /*
+ * This will set the item's lsn to item_lsn and update the
+ * position of the item in the AIL.
+ *
+ * xfs_trans_ail_update() drops the AIL lock.
+ */
+ xfs_trans_ail_update(ailp, lip, item_lsn);
} else {
- log_flags = 0;
+ spin_unlock(&ailp->xa_lock);
}
- mp = tp->t_mountp;
/*
- * If there is nothing to be logged by the transaction,
- * then unlock all of the items associated with the
- * transaction and free the transaction structure.
- * Also make sure to return any reserved blocks to
- * the free pool.
+ * Now that we've repositioned the item in the AIL, unpin it so it can
+ * be flushed. Pass information about buffer stale state down from the
+ * log item flags, if anyone else stales the buffer we do not want to
+ * pay any attention to it.
*/
-shut_us_down:
- shutdown = XFS_FORCED_SHUTDOWN(mp) ? EIO : 0;
- if (!(tp->t_flags & XFS_TRANS_DIRTY) || shutdown) {
- xfs_trans_unreserve_and_mod_sb(tp);
+ IOP_UNPIN(lip);
+}
+
+/*
+ * This is typically called by the LM when a transaction has been fully
+ * committed to disk. It needs to unpin the items which have
+ * been logged by the transaction and update their positions
+ * in the AIL if necessary.
+ *
+ * This also gets called when the transactions didn't get written out
+ * because of an I/O error. Abortflag & XFS_LI_ABORTED is set then.
+ */
+STATIC void
+xfs_trans_committed(
+ struct xfs_trans *tp,
+ int abortflag)
+{
+ xfs_log_item_desc_t *lidp;
+ xfs_log_item_chunk_t *licp;
+ xfs_log_item_chunk_t *next_licp;
+
+ /* Call the transaction's completion callback if there is one. */
+ if (tp->t_callback != NULL)
+ tp->t_callback(tp, tp->t_callarg);
+
+ for (lidp = xfs_trans_first_item(tp);
+ lidp != NULL;
+ lidp = xfs_trans_next_item(tp, lidp)) {
+ xfs_trans_item_committed(lidp->lid_item, tp->t_lsn, abortflag);
+ }
+
+ /* free the item chunks, ignoring the embedded chunk */
+ for (licp = tp->t_items.lic_next; licp != NULL; licp = next_licp) {
+ next_licp = licp->lic_next;
+ kmem_free(licp);
+ }
+
+ xfs_trans_free(tp);
+}
+
+/*
+ * Called from the trans_commit code when we notice that
+ * the filesystem is in the middle of a forced shutdown.
+ */
+STATIC void
+xfs_trans_uncommit(
+ struct xfs_trans *tp,
+ uint flags)
+{
+ xfs_log_item_desc_t *lidp;
+
+ for (lidp = xfs_trans_first_item(tp);
+ lidp != NULL;
+ lidp = xfs_trans_next_item(tp, lidp)) {
/*
- * It is indeed possible for the transaction to be
- * not dirty but the dqinfo portion to be. All that
- * means is that we have some (non-persistent) quota
- * reservations that need to be unreserved.
+ * Unpin all but those that aren't dirty.
*/
- xfs_trans_unreserve_and_mod_dquots(tp);
- if (tp->t_ticket) {
- commit_lsn = xfs_log_done(mp, tp->t_ticket,
- NULL, log_flags);
- if (commit_lsn == -1 && !shutdown)
- shutdown = XFS_ERROR(EIO);
- }
- current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
- xfs_trans_free_items(tp, shutdown? XFS_TRANS_ABORT : 0);
- xfs_trans_free_busy(tp);
- xfs_trans_free(tp);
- XFS_STATS_INC(xs_trans_empty);
- return (shutdown);
+ if (lidp->lid_flags & XFS_LID_DIRTY)
+ IOP_UNPIN_REMOVE(lidp->lid_item, tp);
}
- ASSERT(tp->t_ticket != NULL);
- /*
- * If we need to update the superblock, then do it now.
- */
- if (tp->t_flags & XFS_TRANS_SB_DIRTY)
- xfs_trans_apply_sb_deltas(tp);
- xfs_trans_apply_dquot_deltas(tp);
+ xfs_trans_unreserve_and_mod_sb(tp);
+ xfs_trans_unreserve_and_mod_dquots(tp);
+
+ xfs_trans_free_items(tp, NULLCOMMITLSN, flags);
+ xfs_trans_free(tp);
+}
+
+/*
+ * Format the transaction direct to the iclog. This isolates the physical
+ * transaction commit operation from the logical operation and hence allows
+ * other methods to be introduced without affecting the existing commit path.
+ */
+static int
+xfs_trans_commit_iclog(
+ struct xfs_mount *mp,
+ struct xfs_trans *tp,
+ xfs_lsn_t *commit_lsn,
+ int flags)
+{
+ int shutdown;
+ int error;
+ int log_flags = 0;
+ struct xlog_in_core *commit_iclog;
+#define XFS_TRANS_LOGVEC_COUNT 16
+ struct xfs_log_iovec log_vector_fast[XFS_TRANS_LOGVEC_COUNT];
+ struct xfs_log_iovec *log_vector;
+ uint nvec;
+
/*
* Ask each log item how many log_vector entries it will
@@ -861,8 +1382,7 @@ shut_us_down:
*/
nvec = xfs_trans_count_vecs(tp);
if (nvec == 0) {
- xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
- goto shut_us_down;
+ return ENOMEM; /* triggers a shutdown! */
} else if (nvec <= XFS_TRANS_LOGVEC_COUNT) {
log_vector = log_vector_fast;
} else {
@@ -877,6 +1397,9 @@ shut_us_down:
*/
xfs_trans_fill_vecs(tp, log_vector);
+ if (flags & XFS_TRANS_RELEASE_LOG_RES)
+ log_flags = XFS_LOG_REL_PERM_RESERV;
+
error = xfs_log_write(mp, log_vector, nvec, tp->t_ticket, &(tp->t_lsn));
/*
@@ -884,18 +1407,19 @@ shut_us_down:
* at any time after this call. However, all the items associated
* with the transaction are still locked and pinned in memory.
*/
- commit_lsn = xfs_log_done(mp, tp->t_ticket, &commit_iclog, log_flags);
+ *commit_lsn = xfs_log_done(mp, tp->t_ticket, &commit_iclog, log_flags);
- tp->t_commit_lsn = commit_lsn;
- if (nvec > XFS_TRANS_LOGVEC_COUNT) {
+ tp->t_commit_lsn = *commit_lsn;
+ trace_xfs_trans_commit_lsn(tp);
+
+ if (nvec > XFS_TRANS_LOGVEC_COUNT)
kmem_free(log_vector);
- }
/*
* If we got a log write error. Unpin the logitems that we
* had pinned, clean up, free trans structure, and return error.
*/
- if (error || commit_lsn == -1) {
+ if (error || *commit_lsn == -1) {
current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
xfs_trans_uncommit(tp, flags|XFS_TRANS_ABORT);
return XFS_ERROR(EIO);
@@ -909,8 +1433,6 @@ shut_us_down:
*/
xfs_trans_unreserve_and_mod_sb(tp);
- sync = tp->t_flags & XFS_TRANS_SYNC;
-
/*
* Tell the LM to call the transaction completion routine
* when the log write with LSN commit_lsn completes (e.g.
@@ -953,7 +1475,7 @@ shut_us_down:
* the commit lsn of this transaction for dependency tracking
* purposes.
*/
- xfs_trans_unlock_items(tp, commit_lsn);
+ xfs_trans_unlock_items(tp, *commit_lsn);
/*
* If we detected a log error earlier, finish committing
@@ -973,157 +1495,204 @@ shut_us_down:
* and the items are released we can finally allow the iclog to
* go to disk.
*/
- error = xfs_log_release_iclog(mp, commit_iclog);
-
- /*
- * If the transaction needs to be synchronous, then force the
- * log out now and wait for it.
- */
- if (sync) {
- if (!error) {
- error = _xfs_log_force(mp, commit_lsn,
- XFS_LOG_FORCE | XFS_LOG_SYNC,
- log_flushed);
- }
- XFS_STATS_INC(xs_trans_sync);
- } else {
- XFS_STATS_INC(xs_trans_async);
- }
-
- return (error);
+ return xfs_log_release_iclog(mp, commit_iclog);
}
-
/*
- * Total up the number of log iovecs needed to commit this
- * transaction. The transaction itself needs one for the
- * transaction header. Ask each dirty item in turn how many
- * it needs to get the total.
+ * Walk the log items and allocate log vector structures for
+ * each item large enough to fit all the vectors they require.
+ * Note that this format differs from the old log vector format in
+ * that there is no transaction header in these log vectors.
*/
-STATIC uint
-xfs_trans_count_vecs(
+STATIC struct xfs_log_vec *
+xfs_trans_alloc_log_vecs(
xfs_trans_t *tp)
{
- int nvecs;
xfs_log_item_desc_t *lidp;
+ struct xfs_log_vec *lv = NULL;
+ struct xfs_log_vec *ret_lv = NULL;
- nvecs = 1;
lidp = xfs_trans_first_item(tp);
- ASSERT(lidp != NULL);
- /* In the non-debug case we need to start bailing out if we
- * didn't find a log_item here, return zero and let trans_commit
- * deal with it.
- */
- if (lidp == NULL)
- return 0;
+ /* Bail out if we didn't find a log item. */
+ if (!lidp) {
+ ASSERT(0);
+ return NULL;
+ }
while (lidp != NULL) {
- /*
- * Skip items which aren't dirty in this transaction.
- */
+ struct xfs_log_vec *new_lv;
+
+ /* Skip items which aren't dirty in this transaction. */
if (!(lidp->lid_flags & XFS_LID_DIRTY)) {
lidp = xfs_trans_next_item(tp, lidp);
continue;
}
+
+ /* Skip items that do not have any vectors for writing */
lidp->lid_size = IOP_SIZE(lidp->lid_item);
- nvecs += lidp->lid_size;
+ if (!lidp->lid_size) {
+ lidp = xfs_trans_next_item(tp, lidp);
+ continue;
+ }
+
+ new_lv = kmem_zalloc(sizeof(*new_lv) +
+ lidp->lid_size * sizeof(struct xfs_log_iovec),
+ KM_SLEEP);
+
+ /* The allocated iovec region lies beyond the log vector. */
+ new_lv->lv_iovecp = (struct xfs_log_iovec *)&new_lv[1];
+ new_lv->lv_niovecs = lidp->lid_size;
+ new_lv->lv_item = lidp->lid_item;
+ if (!ret_lv)
+ ret_lv = new_lv;
+ else
+ lv->lv_next = new_lv;
+ lv = new_lv;
lidp = xfs_trans_next_item(tp, lidp);
}
- return nvecs;
+ return ret_lv;
}
-/*
- * Called from the trans_commit code when we notice that
- * the filesystem is in the middle of a forced shutdown.
- */
-STATIC void
-xfs_trans_uncommit(
- xfs_trans_t *tp,
- uint flags)
+static int
+xfs_trans_commit_cil(
+ struct xfs_mount *mp,
+ struct xfs_trans *tp,
+ xfs_lsn_t *commit_lsn,
+ int flags)
{
- xfs_log_item_desc_t *lidp;
+ struct xfs_log_vec *log_vector;
+ int error;
- for (lidp = xfs_trans_first_item(tp);
- lidp != NULL;
- lidp = xfs_trans_next_item(tp, lidp)) {
- /*
- * Unpin all but those that aren't dirty.
- */
- if (lidp->lid_flags & XFS_LID_DIRTY)
- IOP_UNPIN_REMOVE(lidp->lid_item, tp);
- }
+ /*
+ * Get each log item to allocate a vector structure for
+ * the log item to to pass to the log write code. The
+ * CIL commit code will format the vector and save it away.
+ */
+ log_vector = xfs_trans_alloc_log_vecs(tp);
+ if (!log_vector)
+ return ENOMEM;
- xfs_trans_unreserve_and_mod_sb(tp);
- xfs_trans_unreserve_and_mod_dquots(tp);
+ error = xfs_log_commit_cil(mp, tp, log_vector, commit_lsn, flags);
+ if (error)
+ return error;
+
+ current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
- xfs_trans_free_items(tp, flags);
- xfs_trans_free_busy(tp);
+ /* xfs_trans_free_items() unlocks them first */
+ xfs_trans_free_items(tp, *commit_lsn, 0);
xfs_trans_free(tp);
+ return 0;
}
/*
- * Fill in the vector with pointers to data to be logged
- * by this transaction. The transaction header takes
- * the first vector, and then each dirty item takes the
- * number of vectors it indicated it needed in xfs_trans_count_vecs().
+ * xfs_trans_commit
*
- * As each item fills in the entries it needs, also pin the item
- * so that it cannot be flushed out until the log write completes.
+ * Commit the given transaction to the log a/synchronously.
+ *
+ * XFS disk error handling mechanism is not based on a typical
+ * transaction abort mechanism. Logically after the filesystem
+ * gets marked 'SHUTDOWN', we can't let any new transactions
+ * be durable - ie. committed to disk - because some metadata might
+ * be inconsistent. In such cases, this returns an error, and the
+ * caller may assume that all locked objects joined to the transaction
+ * have already been unlocked as if the commit had succeeded.
+ * Do not reference the transaction structure after this call.
*/
-STATIC void
-xfs_trans_fill_vecs(
- xfs_trans_t *tp,
- xfs_log_iovec_t *log_vector)
+int
+_xfs_trans_commit(
+ struct xfs_trans *tp,
+ uint flags,
+ int *log_flushed)
{
- xfs_log_item_desc_t *lidp;
- xfs_log_iovec_t *vecp;
- uint nitems;
+ struct xfs_mount *mp = tp->t_mountp;
+ xfs_lsn_t commit_lsn = -1;
+ int error = 0;
+ int log_flags = 0;
+ int sync = tp->t_flags & XFS_TRANS_SYNC;
/*
- * Skip over the entry for the transaction header, we'll
- * fill that in at the end.
+ * Determine whether this commit is releasing a permanent
+ * log reservation or not.
*/
- vecp = log_vector + 1; /* pointer arithmetic */
+ if (flags & XFS_TRANS_RELEASE_LOG_RES) {
+ ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
+ log_flags = XFS_LOG_REL_PERM_RESERV;
+ }
- nitems = 0;
- lidp = xfs_trans_first_item(tp);
- ASSERT(lidp != NULL);
- while (lidp != NULL) {
- /*
- * Skip items which aren't dirty in this transaction.
- */
- if (!(lidp->lid_flags & XFS_LID_DIRTY)) {
- lidp = xfs_trans_next_item(tp, lidp);
- continue;
- }
- /*
- * The item may be marked dirty but not log anything.
- * This can be used to get called when a transaction
- * is committed.
- */
- if (lidp->lid_size) {
- nitems++;
+ /*
+ * If there is nothing to be logged by the transaction,
+ * then unlock all of the items associated with the
+ * transaction and free the transaction structure.
+ * Also make sure to return any reserved blocks to
+ * the free pool.
+ */
+ if (!(tp->t_flags & XFS_TRANS_DIRTY))
+ goto out_unreserve;
+
+ if (XFS_FORCED_SHUTDOWN(mp)) {
+ error = XFS_ERROR(EIO);
+ goto out_unreserve;
+ }
+
+ ASSERT(tp->t_ticket != NULL);
+
+ /*
+ * If we need to update the superblock, then do it now.
+ */
+ if (tp->t_flags & XFS_TRANS_SB_DIRTY)
+ xfs_trans_apply_sb_deltas(tp);
+ xfs_trans_apply_dquot_deltas(tp);
+
+ if (mp->m_flags & XFS_MOUNT_DELAYLOG)
+ error = xfs_trans_commit_cil(mp, tp, &commit_lsn, flags);
+ else
+ error = xfs_trans_commit_iclog(mp, tp, &commit_lsn, flags);
+
+ if (error == ENOMEM) {
+ xfs_force_shutdown(mp, SHUTDOWN_LOG_IO_ERROR);
+ error = XFS_ERROR(EIO);
+ goto out_unreserve;
+ }
+
+ /*
+ * If the transaction needs to be synchronous, then force the
+ * log out now and wait for it.
+ */
+ if (sync) {
+ if (!error) {
+ error = _xfs_log_force_lsn(mp, commit_lsn,
+ XFS_LOG_SYNC, log_flushed);
}
- IOP_FORMAT(lidp->lid_item, vecp);
- vecp += lidp->lid_size; /* pointer arithmetic */
- IOP_PIN(lidp->lid_item);
- lidp = xfs_trans_next_item(tp, lidp);
+ XFS_STATS_INC(xs_trans_sync);
+ } else {
+ XFS_STATS_INC(xs_trans_async);
}
+ return error;
+
+out_unreserve:
+ xfs_trans_unreserve_and_mod_sb(tp);
+
/*
- * Now that we've counted the number of items in this
- * transaction, fill in the transaction header.
+ * It is indeed possible for the transaction to be not dirty but
+ * the dqinfo portion to be. All that means is that we have some
+ * (non-persistent) quota reservations that need to be unreserved.
*/
- tp->t_header.th_magic = XFS_TRANS_HEADER_MAGIC;
- tp->t_header.th_type = tp->t_type;
- tp->t_header.th_num_items = nitems;
- log_vector->i_addr = (xfs_caddr_t)&tp->t_header;
- log_vector->i_len = sizeof(xfs_trans_header_t);
- XLOG_VEC_SET_TYPE(log_vector, XLOG_REG_TYPE_TRANSHDR);
-}
+ xfs_trans_unreserve_and_mod_dquots(tp);
+ if (tp->t_ticket) {
+ commit_lsn = xfs_log_done(mp, tp->t_ticket, NULL, log_flags);
+ if (commit_lsn == -1 && !error)
+ error = XFS_ERROR(EIO);
+ }
+ current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
+ xfs_trans_free_items(tp, NULLCOMMITLSN, error ? XFS_TRANS_ABORT : 0);
+ xfs_trans_free(tp);
+ XFS_STATS_INC(xs_trans_empty);
+ return error;
+}
/*
* Unlock all of the transaction's items and free the transaction.
@@ -1196,25 +1765,10 @@ xfs_trans_cancel(
/* mark this thread as no longer being in a transaction */
current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
- xfs_trans_free_items(tp, flags);
- xfs_trans_free_busy(tp);
+ xfs_trans_free_items(tp, NULLCOMMITLSN, flags);
xfs_trans_free(tp);
}
-
-/*
- * Free the transaction structure. If there is more clean up
- * to do when the structure is freed, add it here.
- */
-STATIC void
-xfs_trans_free(
- xfs_trans_t *tp)
-{
- atomic_dec(&tp->t_mountp->m_active_trans);
- xfs_trans_free_dqinfo(tp);
- kmem_zone_free(xfs_trans_zone, tp);
-}
-
/*
* Roll from one trans in the sequence of PERMANENT transactions to
* the next: permanent transactions are only flushed out when
@@ -1284,174 +1838,3 @@ xfs_trans_roll(
xfs_trans_ihold(trans, dp);
return 0;
}
-
-/*
- * THIS SHOULD BE REWRITTEN TO USE xfs_trans_next_item().
- *
- * This is typically called by the LM when a transaction has been fully
- * committed to disk. It needs to unpin the items which have
- * been logged by the transaction and update their positions
- * in the AIL if necessary.
- * This also gets called when the transactions didn't get written out
- * because of an I/O error. Abortflag & XFS_LI_ABORTED is set then.
- *
- * Call xfs_trans_chunk_committed() to process the items in
- * each chunk.
- */
-STATIC void
-xfs_trans_committed(
- xfs_trans_t *tp,
- int abortflag)
-{
- xfs_log_item_chunk_t *licp;
- xfs_log_item_chunk_t *next_licp;
- xfs_log_busy_chunk_t *lbcp;
- xfs_log_busy_slot_t *lbsp;
- int i;
-
- /*
- * Call the transaction's completion callback if there
- * is one.
- */
- if (tp->t_callback != NULL) {
- tp->t_callback(tp, tp->t_callarg);
- }
-
- /*
- * Special case the chunk embedded in the transaction.
- */
- licp = &(tp->t_items);
- if (!(xfs_lic_are_all_free(licp))) {
- xfs_trans_chunk_committed(licp, tp->t_lsn, abortflag);
- }
-
- /*
- * Process the items in each chunk in turn.
- */
- licp = licp->lic_next;
- while (licp != NULL) {
- ASSERT(!xfs_lic_are_all_free(licp));
- xfs_trans_chunk_committed(licp, tp->t_lsn, abortflag);
- next_licp = licp->lic_next;
- kmem_free(licp);
- licp = next_licp;
- }
-
- /*
- * Clear all the per-AG busy list items listed in this transaction
- */
- lbcp = &tp->t_busy;
- while (lbcp != NULL) {
- for (i = 0, lbsp = lbcp->lbc_busy; i < lbcp->lbc_unused; i++, lbsp++) {
- if (!XFS_LBC_ISFREE(lbcp, i)) {
- xfs_alloc_clear_busy(tp, lbsp->lbc_ag,
- lbsp->lbc_idx);
- }
- }
- lbcp = lbcp->lbc_next;
- }
- xfs_trans_free_busy(tp);
-
- /*
- * That's it for the transaction structure. Free it.
- */
- xfs_trans_free(tp);
-}
-
-/*
- * This is called to perform the commit processing for each
- * item described by the given chunk.
- *
- * The commit processing consists of unlocking items which were
- * held locked with the SYNC_UNLOCK attribute, calling the committed
- * routine of each logged item, updating the item's position in the AIL
- * if necessary, and unpinning each item. If the committed routine
- * returns -1, then do nothing further with the item because it
- * may have been freed.
- *
- * Since items are unlocked when they are copied to the incore
- * log, it is possible for two transactions to be completing
- * and manipulating the same item simultaneously. The AIL lock
- * will protect the lsn field of each item. The value of this
- * field can never go backwards.
- *
- * We unpin the items after repositioning them in the AIL, because
- * otherwise they could be immediately flushed and we'd have to race
- * with the flusher trying to pull the item from the AIL as we add it.
- */
-STATIC void
-xfs_trans_chunk_committed(
- xfs_log_item_chunk_t *licp,
- xfs_lsn_t lsn,
- int aborted)
-{
- xfs_log_item_desc_t *lidp;
- xfs_log_item_t *lip;
- xfs_lsn_t item_lsn;
- int i;
-
- lidp = licp->lic_descs;
- for (i = 0; i < licp->lic_unused; i++, lidp++) {
- struct xfs_ail *ailp;
-
- if (xfs_lic_isfree(licp, i)) {
- continue;
- }
-
- lip = lidp->lid_item;
- if (aborted)
- lip->li_flags |= XFS_LI_ABORTED;
-
- /*
- * Send in the ABORTED flag to the COMMITTED routine
- * so that it knows whether the transaction was aborted
- * or not.
- */
- item_lsn = IOP_COMMITTED(lip, lsn);
-
- /*
- * If the committed routine returns -1, make
- * no more references to the item.
- */
- if (XFS_LSN_CMP(item_lsn, (xfs_lsn_t)-1) == 0) {
- continue;
- }
-
- /*
- * If the returned lsn is greater than what it
- * contained before, update the location of the
- * item in the AIL. If it is not, then do nothing.
- * Items can never move backwards in the AIL.
- *
- * While the new lsn should usually be greater, it
- * is possible that a later transaction completing
- * simultaneously with an earlier one using the
- * same item could complete first with a higher lsn.
- * This would cause the earlier transaction to fail
- * the test below.
- */
- ailp = lip->li_ailp;
- spin_lock(&ailp->xa_lock);
- if (XFS_LSN_CMP(item_lsn, lip->li_lsn) > 0) {
- /*
- * This will set the item's lsn to item_lsn
- * and update the position of the item in
- * the AIL.
- *
- * xfs_trans_ail_update() drops the AIL lock.
- */
- xfs_trans_ail_update(ailp, lip, item_lsn);
- } else {
- spin_unlock(&ailp->xa_lock);
- }
-
- /*
- * Now that we've repositioned the item in the AIL,
- * unpin it so it can be flushed. Pass information
- * about buffer stale state down from the log item
- * flags, if anyone else stales the buffer we do not
- * want to pay any attention to it.
- */
- IOP_UNPIN(lip, lidp->lid_flags & XFS_LID_BUF_STALE);
- }
-}
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index ca64f33c63a3..e639e8e9a2a9 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -49,6 +49,15 @@ typedef struct xfs_trans_header {
#define XFS_LI_DQUOT 0x123d
#define XFS_LI_QUOTAOFF 0x123e
+#define XFS_LI_TYPE_DESC \
+ { XFS_LI_EFI, "XFS_LI_EFI" }, \
+ { XFS_LI_EFD, "XFS_LI_EFD" }, \
+ { XFS_LI_IUNLINK, "XFS_LI_IUNLINK" }, \
+ { XFS_LI_INODE, "XFS_LI_INODE" }, \
+ { XFS_LI_BUF, "XFS_LI_BUF" }, \
+ { XFS_LI_DQUOT, "XFS_LI_DQUOT" }, \
+ { XFS_LI_QUOTAOFF, "XFS_LI_QUOTAOFF" }
+
/*
* Transaction types. Used to distinguish types of buffers.
*/
@@ -97,7 +106,8 @@ typedef struct xfs_trans_header {
#define XFS_TRANS_GROWFSRT_FREE 39
#define XFS_TRANS_SWAPEXT 40
#define XFS_TRANS_SB_COUNT 41
-#define XFS_TRANS_TYPE_MAX 41
+#define XFS_TRANS_CHECKPOINT 42
+#define XFS_TRANS_TYPE_MAX 42
/* new transaction types need to be reflected in xfs_logprint(8) */
#define XFS_TRANS_TYPES \
@@ -139,6 +149,7 @@ typedef struct xfs_trans_header {
{ XFS_TRANS_GROWFSRT_FREE, "GROWFSRT_FREE" }, \
{ XFS_TRANS_SWAPEXT, "SWAPEXT" }, \
{ XFS_TRANS_SB_COUNT, "SB_COUNT" }, \
+ { XFS_TRANS_CHECKPOINT, "CHECKPOINT" }, \
{ XFS_TRANS_DUMMY1, "DUMMY1" }, \
{ XFS_TRANS_DUMMY2, "DUMMY2" }, \
{ XLOG_UNMOUNT_REC_TYPE, "UNMOUNT" }
@@ -159,7 +170,6 @@ typedef struct xfs_log_item_desc {
#define XFS_LID_DIRTY 0x1
#define XFS_LID_PINNED 0x2
-#define XFS_LID_BUF_STALE 0x8
/*
* This structure is used to maintain a chunk list of log_item_desc
@@ -290,24 +300,6 @@ xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
/*
- * Various log reservation values.
- * These are based on the size of the file system block
- * because that is what most transactions manipulate.
- * Each adds in an additional 128 bytes per item logged to
- * try to account for the overhead of the transaction mechanism.
- *
- * Note:
- * Most of the reservations underestimate the number of allocation
- * groups into which they could free extents in the xfs_bmap_finish()
- * call. This is because the number in the worst case is quite high
- * and quite unusual. In order to fix this we need to change
- * xfs_bmap_finish() to free extents in only a single AG at a time.
- * This will require changes to the EFI code as well, however, so that
- * the EFI for the extents not freed is logged again in each transaction.
- * See bug 261917.
- */
-
-/*
* Per-extent log reservation for the allocation btree changes
* involved in freeing or allocating an extent.
* 2 trees * (2 blocks/level * max depth - 1) * block size
@@ -331,429 +323,36 @@ xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
(XFS_DAENTER_BLOCKS(mp, XFS_DATA_FORK) + \
XFS_DAENTER_BMAPS(mp, XFS_DATA_FORK) + 1)
-/*
- * In a write transaction we can allocate a maximum of 2
- * extents. This gives:
- * the inode getting the new extents: inode size
- * the inode's bmap btree: max depth * block size
- * the agfs of the ags from which the extents are allocated: 2 * sector
- * the superblock free block counter: sector size
- * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size
- * And the bmap_finish transaction can free bmap blocks in a join:
- * the agfs of the ags containing the blocks: 2 * sector size
- * the agfls of the ags containing the blocks: 2 * sector size
- * the super block free block counter: sector size
- * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size
- */
-#define XFS_CALC_WRITE_LOG_RES(mp) \
- (MAX( \
- ((mp)->m_sb.sb_inodesize + \
- XFS_FSB_TO_B((mp), XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK)) + \
- (2 * (mp)->m_sb.sb_sectsize) + \
- (mp)->m_sb.sb_sectsize + \
- XFS_ALLOCFREE_LOG_RES(mp, 2) + \
- (128 * (4 + XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) + XFS_ALLOCFREE_LOG_COUNT(mp, 2)))),\
- ((2 * (mp)->m_sb.sb_sectsize) + \
- (2 * (mp)->m_sb.sb_sectsize) + \
- (mp)->m_sb.sb_sectsize + \
- XFS_ALLOCFREE_LOG_RES(mp, 2) + \
- (128 * (5 + XFS_ALLOCFREE_LOG_COUNT(mp, 2))))))
#define XFS_WRITE_LOG_RES(mp) ((mp)->m_reservations.tr_write)
-
-/*
- * In truncating a file we free up to two extents at once. We can modify:
- * the inode being truncated: inode size
- * the inode's bmap btree: (max depth + 1) * block size
- * And the bmap_finish transaction can free the blocks and bmap blocks:
- * the agf for each of the ags: 4 * sector size
- * the agfl for each of the ags: 4 * sector size
- * the super block to reflect the freed blocks: sector size
- * worst case split in allocation btrees per extent assuming 4 extents:
- * 4 exts * 2 trees * (2 * max depth - 1) * block size
- * the inode btree: max depth * blocksize
- * the allocation btrees: 2 trees * (max depth - 1) * block size
- */
-#define XFS_CALC_ITRUNCATE_LOG_RES(mp) \
- (MAX( \
- ((mp)->m_sb.sb_inodesize + \
- XFS_FSB_TO_B((mp), XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) + 1) + \
- (128 * (2 + XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK)))), \
- ((4 * (mp)->m_sb.sb_sectsize) + \
- (4 * (mp)->m_sb.sb_sectsize) + \
- (mp)->m_sb.sb_sectsize + \
- XFS_ALLOCFREE_LOG_RES(mp, 4) + \
- (128 * (9 + XFS_ALLOCFREE_LOG_COUNT(mp, 4))) + \
- (128 * 5) + \
- XFS_ALLOCFREE_LOG_RES(mp, 1) + \
- (128 * (2 + XFS_IALLOC_BLOCKS(mp) + (mp)->m_in_maxlevels + \
- XFS_ALLOCFREE_LOG_COUNT(mp, 1))))))
-
#define XFS_ITRUNCATE_LOG_RES(mp) ((mp)->m_reservations.tr_itruncate)
-
-/*
- * In renaming a files we can modify:
- * the four inodes involved: 4 * inode size
- * the two directory btrees: 2 * (max depth + v2) * dir block size
- * the two directory bmap btrees: 2 * max depth * block size
- * And the bmap_finish transaction can free dir and bmap blocks (two sets
- * of bmap blocks) giving:
- * the agf for the ags in which the blocks live: 3 * sector size
- * the agfl for the ags in which the blocks live: 3 * sector size
- * the superblock for the free block count: sector size
- * the allocation btrees: 3 exts * 2 trees * (2 * max depth - 1) * block size
- */
-#define XFS_CALC_RENAME_LOG_RES(mp) \
- (MAX( \
- ((4 * (mp)->m_sb.sb_inodesize) + \
- (2 * XFS_DIROP_LOG_RES(mp)) + \
- (128 * (4 + 2 * XFS_DIROP_LOG_COUNT(mp)))), \
- ((3 * (mp)->m_sb.sb_sectsize) + \
- (3 * (mp)->m_sb.sb_sectsize) + \
- (mp)->m_sb.sb_sectsize + \
- XFS_ALLOCFREE_LOG_RES(mp, 3) + \
- (128 * (7 + XFS_ALLOCFREE_LOG_COUNT(mp, 3))))))
-
#define XFS_RENAME_LOG_RES(mp) ((mp)->m_reservations.tr_rename)
-
-/*
- * For creating a link to an inode:
- * the parent directory inode: inode size
- * the linked inode: inode size
- * the directory btree could split: (max depth + v2) * dir block size
- * the directory bmap btree could join or split: (max depth + v2) * blocksize
- * And the bmap_finish transaction can free some bmap blocks giving:
- * the agf for the ag in which the blocks live: sector size
- * the agfl for the ag in which the blocks live: sector size
- * the superblock for the free block count: sector size
- * the allocation btrees: 2 trees * (2 * max depth - 1) * block size
- */
-#define XFS_CALC_LINK_LOG_RES(mp) \
- (MAX( \
- ((mp)->m_sb.sb_inodesize + \
- (mp)->m_sb.sb_inodesize + \
- XFS_DIROP_LOG_RES(mp) + \
- (128 * (2 + XFS_DIROP_LOG_COUNT(mp)))), \
- ((mp)->m_sb.sb_sectsize + \
- (mp)->m_sb.sb_sectsize + \
- (mp)->m_sb.sb_sectsize + \
- XFS_ALLOCFREE_LOG_RES(mp, 1) + \
- (128 * (3 + XFS_ALLOCFREE_LOG_COUNT(mp, 1))))))
-
#define XFS_LINK_LOG_RES(mp) ((mp)->m_reservations.tr_link)
-
-/*
- * For removing a directory entry we can modify:
- * the parent directory inode: inode size
- * the removed inode: inode size
- * the directory btree could join: (max depth + v2) * dir block size
- * the directory bmap btree could join or split: (max depth + v2) * blocksize
- * And the bmap_finish transaction can free the dir and bmap blocks giving:
- * the agf for the ag in which the blocks live: 2 * sector size
- * the agfl for the ag in which the blocks live: 2 * sector size
- * the superblock for the free block count: sector size
- * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size
- */
-#define XFS_CALC_REMOVE_LOG_RES(mp) \
- (MAX( \
- ((mp)->m_sb.sb_inodesize + \
- (mp)->m_sb.sb_inodesize + \
- XFS_DIROP_LOG_RES(mp) + \
- (128 * (2 + XFS_DIROP_LOG_COUNT(mp)))), \
- ((2 * (mp)->m_sb.sb_sectsize) + \
- (2 * (mp)->m_sb.sb_sectsize) + \
- (mp)->m_sb.sb_sectsize + \
- XFS_ALLOCFREE_LOG_RES(mp, 2) + \
- (128 * (5 + XFS_ALLOCFREE_LOG_COUNT(mp, 2))))))
-
#define XFS_REMOVE_LOG_RES(mp) ((mp)->m_reservations.tr_remove)
-
-/*
- * For symlink we can modify:
- * the parent directory inode: inode size
- * the new inode: inode size
- * the inode btree entry: 1 block
- * the directory btree: (max depth + v2) * dir block size
- * the directory inode's bmap btree: (max depth + v2) * block size
- * the blocks for the symlink: 1 kB
- * Or in the first xact we allocate some inodes giving:
- * the agi and agf of the ag getting the new inodes: 2 * sectorsize
- * the inode blocks allocated: XFS_IALLOC_BLOCKS * blocksize
- * the inode btree: max depth * blocksize
- * the allocation btrees: 2 trees * (2 * max depth - 1) * block size
- */
-#define XFS_CALC_SYMLINK_LOG_RES(mp) \
- (MAX( \
- ((mp)->m_sb.sb_inodesize + \
- (mp)->m_sb.sb_inodesize + \
- XFS_FSB_TO_B(mp, 1) + \
- XFS_DIROP_LOG_RES(mp) + \
- 1024 + \
- (128 * (4 + XFS_DIROP_LOG_COUNT(mp)))), \
- (2 * (mp)->m_sb.sb_sectsize + \
- XFS_FSB_TO_B((mp), XFS_IALLOC_BLOCKS((mp))) + \
- XFS_FSB_TO_B((mp), (mp)->m_in_maxlevels) + \
- XFS_ALLOCFREE_LOG_RES(mp, 1) + \
- (128 * (2 + XFS_IALLOC_BLOCKS(mp) + (mp)->m_in_maxlevels + \
- XFS_ALLOCFREE_LOG_COUNT(mp, 1))))))
-
#define XFS_SYMLINK_LOG_RES(mp) ((mp)->m_reservations.tr_symlink)
-
-/*
- * For create we can modify:
- * the parent directory inode: inode size
- * the new inode: inode size
- * the inode btree entry: block size
- * the superblock for the nlink flag: sector size
- * the directory btree: (max depth + v2) * dir block size
- * the directory inode's bmap btree: (max depth + v2) * block size
- * Or in the first xact we allocate some inodes giving:
- * the agi and agf of the ag getting the new inodes: 2 * sectorsize
- * the superblock for the nlink flag: sector size
- * the inode blocks allocated: XFS_IALLOC_BLOCKS * blocksize
- * the inode btree: max depth * blocksize
- * the allocation btrees: 2 trees * (max depth - 1) * block size
- */
-#define XFS_CALC_CREATE_LOG_RES(mp) \
- (MAX( \
- ((mp)->m_sb.sb_inodesize + \
- (mp)->m_sb.sb_inodesize + \
- (mp)->m_sb.sb_sectsize + \
- XFS_FSB_TO_B(mp, 1) + \
- XFS_DIROP_LOG_RES(mp) + \
- (128 * (3 + XFS_DIROP_LOG_COUNT(mp)))), \
- (3 * (mp)->m_sb.sb_sectsize + \
- XFS_FSB_TO_B((mp), XFS_IALLOC_BLOCKS((mp))) + \
- XFS_FSB_TO_B((mp), (mp)->m_in_maxlevels) + \
- XFS_ALLOCFREE_LOG_RES(mp, 1) + \
- (128 * (2 + XFS_IALLOC_BLOCKS(mp) + (mp)->m_in_maxlevels + \
- XFS_ALLOCFREE_LOG_COUNT(mp, 1))))))
-
#define XFS_CREATE_LOG_RES(mp) ((mp)->m_reservations.tr_create)
-
-/*
- * Making a new directory is the same as creating a new file.
- */
-#define XFS_CALC_MKDIR_LOG_RES(mp) XFS_CALC_CREATE_LOG_RES(mp)
-
#define XFS_MKDIR_LOG_RES(mp) ((mp)->m_reservations.tr_mkdir)
-
-/*
- * In freeing an inode we can modify:
- * the inode being freed: inode size
- * the super block free inode counter: sector size
- * the agi hash list and counters: sector size
- * the inode btree entry: block size
- * the on disk inode before ours in the agi hash list: inode cluster size
- * the inode btree: max depth * blocksize
- * the allocation btrees: 2 trees * (max depth - 1) * block size
- */
-#define XFS_CALC_IFREE_LOG_RES(mp) \
- ((mp)->m_sb.sb_inodesize + \
- (mp)->m_sb.sb_sectsize + \
- (mp)->m_sb.sb_sectsize + \
- XFS_FSB_TO_B((mp), 1) + \
- MAX((__uint16_t)XFS_FSB_TO_B((mp), 1), XFS_INODE_CLUSTER_SIZE(mp)) + \
- (128 * 5) + \
- XFS_ALLOCFREE_LOG_RES(mp, 1) + \
- (128 * (2 + XFS_IALLOC_BLOCKS(mp) + (mp)->m_in_maxlevels + \
- XFS_ALLOCFREE_LOG_COUNT(mp, 1))))
-
-
#define XFS_IFREE_LOG_RES(mp) ((mp)->m_reservations.tr_ifree)
-
-/*
- * When only changing the inode we log the inode and possibly the superblock
- * We also add a bit of slop for the transaction stuff.
- */
-#define XFS_CALC_ICHANGE_LOG_RES(mp) ((mp)->m_sb.sb_inodesize + \
- (mp)->m_sb.sb_sectsize + 512)
-
#define XFS_ICHANGE_LOG_RES(mp) ((mp)->m_reservations.tr_ichange)
-
-/*
- * Growing the data section of the filesystem.
- * superblock
- * agi and agf
- * allocation btrees
- */
-#define XFS_CALC_GROWDATA_LOG_RES(mp) \
- ((mp)->m_sb.sb_sectsize * 3 + \
- XFS_ALLOCFREE_LOG_RES(mp, 1) + \
- (128 * (3 + XFS_ALLOCFREE_LOG_COUNT(mp, 1))))
-
#define XFS_GROWDATA_LOG_RES(mp) ((mp)->m_reservations.tr_growdata)
-
-/*
- * Growing the rt section of the filesystem.
- * In the first set of transactions (ALLOC) we allocate space to the
- * bitmap or summary files.
- * superblock: sector size
- * agf of the ag from which the extent is allocated: sector size
- * bmap btree for bitmap/summary inode: max depth * blocksize
- * bitmap/summary inode: inode size
- * allocation btrees for 1 block alloc: 2 * (2 * maxdepth - 1) * blocksize
- */
-#define XFS_CALC_GROWRTALLOC_LOG_RES(mp) \
- (2 * (mp)->m_sb.sb_sectsize + \
- XFS_FSB_TO_B((mp), XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK)) + \
- (mp)->m_sb.sb_inodesize + \
- XFS_ALLOCFREE_LOG_RES(mp, 1) + \
- (128 * \
- (3 + XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK) + \
- XFS_ALLOCFREE_LOG_COUNT(mp, 1))))
-
#define XFS_GROWRTALLOC_LOG_RES(mp) ((mp)->m_reservations.tr_growrtalloc)
-
-/*
- * Growing the rt section of the filesystem.
- * In the second set of transactions (ZERO) we zero the new metadata blocks.
- * one bitmap/summary block: blocksize
- */
-#define XFS_CALC_GROWRTZERO_LOG_RES(mp) \
- ((mp)->m_sb.sb_blocksize + 128)
-
#define XFS_GROWRTZERO_LOG_RES(mp) ((mp)->m_reservations.tr_growrtzero)
-
-/*
- * Growing the rt section of the filesystem.
- * In the third set of transactions (FREE) we update metadata without
- * allocating any new blocks.
- * superblock: sector size
- * bitmap inode: inode size
- * summary inode: inode size
- * one bitmap block: blocksize
- * summary blocks: new summary size
- */
-#define XFS_CALC_GROWRTFREE_LOG_RES(mp) \
- ((mp)->m_sb.sb_sectsize + \
- 2 * (mp)->m_sb.sb_inodesize + \
- (mp)->m_sb.sb_blocksize + \
- (mp)->m_rsumsize + \
- (128 * 5))
-
#define XFS_GROWRTFREE_LOG_RES(mp) ((mp)->m_reservations.tr_growrtfree)
-
-/*
- * Logging the inode modification timestamp on a synchronous write.
- * inode
- */
-#define XFS_CALC_SWRITE_LOG_RES(mp) \
- ((mp)->m_sb.sb_inodesize + 128)
-
#define XFS_SWRITE_LOG_RES(mp) ((mp)->m_reservations.tr_swrite)
-
/*
* Logging the inode timestamps on an fsync -- same as SWRITE
* as long as SWRITE logs the entire inode core
*/
#define XFS_FSYNC_TS_LOG_RES(mp) ((mp)->m_reservations.tr_swrite)
-
-/*
- * Logging the inode mode bits when writing a setuid/setgid file
- * inode
- */
-#define XFS_CALC_WRITEID_LOG_RES(mp) \
- ((mp)->m_sb.sb_inodesize + 128)
-
#define XFS_WRITEID_LOG_RES(mp) ((mp)->m_reservations.tr_swrite)
-
-/*
- * Converting the inode from non-attributed to attributed.
- * the inode being converted: inode size
- * agf block and superblock (for block allocation)
- * the new block (directory sized)
- * bmap blocks for the new directory block
- * allocation btrees
- */
-#define XFS_CALC_ADDAFORK_LOG_RES(mp) \
- ((mp)->m_sb.sb_inodesize + \
- (mp)->m_sb.sb_sectsize * 2 + \
- (mp)->m_dirblksize + \
- XFS_FSB_TO_B(mp, (XFS_DAENTER_BMAP1B(mp, XFS_DATA_FORK) + 1)) + \
- XFS_ALLOCFREE_LOG_RES(mp, 1) + \
- (128 * (4 + (XFS_DAENTER_BMAP1B(mp, XFS_DATA_FORK) + 1) + \
- XFS_ALLOCFREE_LOG_COUNT(mp, 1))))
-
#define XFS_ADDAFORK_LOG_RES(mp) ((mp)->m_reservations.tr_addafork)
-
-/*
- * Removing the attribute fork of a file
- * the inode being truncated: inode size
- * the inode's bmap btree: max depth * block size
- * And the bmap_finish transaction can free the blocks and bmap blocks:
- * the agf for each of the ags: 4 * sector size
- * the agfl for each of the ags: 4 * sector size
- * the super block to reflect the freed blocks: sector size
- * worst case split in allocation btrees per extent assuming 4 extents:
- * 4 exts * 2 trees * (2 * max depth - 1) * block size
- */
-#define XFS_CALC_ATTRINVAL_LOG_RES(mp) \
- (MAX( \
- ((mp)->m_sb.sb_inodesize + \
- XFS_FSB_TO_B((mp), XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK)) + \
- (128 * (1 + XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK)))), \
- ((4 * (mp)->m_sb.sb_sectsize) + \
- (4 * (mp)->m_sb.sb_sectsize) + \
- (mp)->m_sb.sb_sectsize + \
- XFS_ALLOCFREE_LOG_RES(mp, 4) + \
- (128 * (9 + XFS_ALLOCFREE_LOG_COUNT(mp, 4))))))
-
#define XFS_ATTRINVAL_LOG_RES(mp) ((mp)->m_reservations.tr_attrinval)
-
-/*
- * Setting an attribute.
- * the inode getting the attribute
- * the superblock for allocations
- * the agfs extents are allocated from
- * the attribute btree * max depth
- * the inode allocation btree
- * Since attribute transaction space is dependent on the size of the attribute,
- * the calculation is done partially at mount time and partially at runtime.
- */
-#define XFS_CALC_ATTRSET_LOG_RES(mp) \
- ((mp)->m_sb.sb_inodesize + \
- (mp)->m_sb.sb_sectsize + \
- XFS_FSB_TO_B((mp), XFS_DA_NODE_MAXDEPTH) + \
- (128 * (2 + XFS_DA_NODE_MAXDEPTH)))
-
#define XFS_ATTRSET_LOG_RES(mp, ext) \
((mp)->m_reservations.tr_attrset + \
(ext * (mp)->m_sb.sb_sectsize) + \
(ext * XFS_FSB_TO_B((mp), XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK))) + \
(128 * (ext + (ext * XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK)))))
-
-/*
- * Removing an attribute.
- * the inode: inode size
- * the attribute btree could join: max depth * block size
- * the inode bmap btree could join or split: max depth * block size
- * And the bmap_finish transaction can free the attr blocks freed giving:
- * the agf for the ag in which the blocks live: 2 * sector size
- * the agfl for the ag in which the blocks live: 2 * sector size
- * the superblock for the free block count: sector size
- * the allocation btrees: 2 exts * 2 trees * (2 * max depth - 1) * block size
- */
-#define XFS_CALC_ATTRRM_LOG_RES(mp) \
- (MAX( \
- ((mp)->m_sb.sb_inodesize + \
- XFS_FSB_TO_B((mp), XFS_DA_NODE_MAXDEPTH) + \
- XFS_FSB_TO_B((mp), XFS_BM_MAXLEVELS(mp, XFS_ATTR_FORK)) + \
- (128 * (1 + XFS_DA_NODE_MAXDEPTH + XFS_BM_MAXLEVELS(mp, XFS_DATA_FORK)))), \
- ((2 * (mp)->m_sb.sb_sectsize) + \
- (2 * (mp)->m_sb.sb_sectsize) + \
- (mp)->m_sb.sb_sectsize + \
- XFS_ALLOCFREE_LOG_RES(mp, 2) + \
- (128 * (5 + XFS_ALLOCFREE_LOG_COUNT(mp, 2))))))
-
#define XFS_ATTRRM_LOG_RES(mp) ((mp)->m_reservations.tr_attrrm)
-
-/*
- * Clearing a bad agino number in an agi hash bucket.
- */
-#define XFS_CALC_CLEAR_AGI_BUCKET_LOG_RES(mp) \
- ((mp)->m_sb.sb_sectsize + 128)
-
#define XFS_CLEAR_AGI_BUCKET_LOG_RES(mp) ((mp)->m_reservations.tr_clearagi)
@@ -805,6 +404,7 @@ struct xfs_log_item_desc;
struct xfs_mount;
struct xfs_trans;
struct xfs_dquot_acct;
+struct xfs_busy_extent;
typedef struct xfs_log_item {
struct list_head li_ail; /* AIL pointers */
@@ -820,6 +420,11 @@ typedef struct xfs_log_item {
/* buffer item iodone */
/* callback func */
struct xfs_item_ops *li_ops; /* function list */
+
+ /* delayed logging */
+ struct list_head li_cil; /* CIL pointers */
+ struct xfs_log_vec *li_lv; /* active log vector */
+ xfs_lsn_t li_seq; /* CIL commit seq */
} xfs_log_item_t;
#define XFS_LI_IN_AIL 0x1
@@ -833,7 +438,7 @@ typedef struct xfs_item_ops {
uint (*iop_size)(xfs_log_item_t *);
void (*iop_format)(xfs_log_item_t *, struct xfs_log_iovec *);
void (*iop_pin)(xfs_log_item_t *);
- void (*iop_unpin)(xfs_log_item_t *, int);
+ void (*iop_unpin)(xfs_log_item_t *);
void (*iop_unpin_remove)(xfs_log_item_t *, struct xfs_trans *);
uint (*iop_trylock)(xfs_log_item_t *);
void (*iop_unlock)(xfs_log_item_t *);
@@ -846,7 +451,7 @@ typedef struct xfs_item_ops {
#define IOP_SIZE(ip) (*(ip)->li_ops->iop_size)(ip)
#define IOP_FORMAT(ip,vp) (*(ip)->li_ops->iop_format)(ip, vp)
#define IOP_PIN(ip) (*(ip)->li_ops->iop_pin)(ip)
-#define IOP_UNPIN(ip, flags) (*(ip)->li_ops->iop_unpin)(ip, flags)
+#define IOP_UNPIN(ip) (*(ip)->li_ops->iop_unpin)(ip)
#define IOP_UNPIN_REMOVE(ip,tp) (*(ip)->li_ops->iop_unpin_remove)(ip, tp)
#define IOP_TRYLOCK(ip) (*(ip)->li_ops->iop_trylock)(ip)
#define IOP_UNLOCK(ip) (*(ip)->li_ops->iop_unlock)(ip)
@@ -861,36 +466,7 @@ typedef struct xfs_item_ops {
#define XFS_ITEM_SUCCESS 0
#define XFS_ITEM_PINNED 1
#define XFS_ITEM_LOCKED 2
-#define XFS_ITEM_FLUSHING 3
-#define XFS_ITEM_PUSHBUF 4
-
-/*
- * This structure is used to maintain a list of block ranges that have been
- * freed in the transaction. The ranges are listed in the perag[] busy list
- * between when they're freed and the transaction is committed to disk.
- */
-
-typedef struct xfs_log_busy_slot {
- xfs_agnumber_t lbc_ag;
- ushort lbc_idx; /* index in perag.busy[] */
-} xfs_log_busy_slot_t;
-
-#define XFS_LBC_NUM_SLOTS 31
-typedef struct xfs_log_busy_chunk {
- struct xfs_log_busy_chunk *lbc_next;
- uint lbc_free; /* free slots bitmask */
- ushort lbc_unused; /* first unused */
- xfs_log_busy_slot_t lbc_busy[XFS_LBC_NUM_SLOTS];
-} xfs_log_busy_chunk_t;
-
-#define XFS_LBC_MAX_SLOT (XFS_LBC_NUM_SLOTS - 1)
-#define XFS_LBC_FREEMASK ((1U << XFS_LBC_NUM_SLOTS) - 1)
-
-#define XFS_LBC_INIT(cp) ((cp)->lbc_free = XFS_LBC_FREEMASK)
-#define XFS_LBC_CLAIM(cp, slot) ((cp)->lbc_free &= ~(1 << (slot)))
-#define XFS_LBC_SLOT(cp, slot) (&((cp)->lbc_busy[(slot)]))
-#define XFS_LBC_VACANCY(cp) (((cp)->lbc_free) & XFS_LBC_FREEMASK)
-#define XFS_LBC_ISFREE(cp, slot) ((cp)->lbc_free & (1 << (slot)))
+#define XFS_ITEM_PUSHBUF 3
/*
* This is the type of function which can be given to xfs_trans_callback()
@@ -911,7 +487,7 @@ typedef struct xfs_trans {
unsigned int t_blk_res_used; /* # of resvd blocks used */
unsigned int t_rtx_res; /* # of rt extents resvd */
unsigned int t_rtx_res_used; /* # of resvd rt extents used */
- xfs_log_ticket_t t_ticket; /* log mgr ticket */
+ struct xlog_ticket *t_ticket; /* log mgr ticket */
xfs_lsn_t t_lsn; /* log seq num of start of
* transaction. */
xfs_lsn_t t_commit_lsn; /* log seq num of end of
@@ -943,8 +519,7 @@ typedef struct xfs_trans {
unsigned int t_items_free; /* log item descs free */
xfs_log_item_chunk_t t_items; /* first log item desc chunk */
xfs_trans_header_t t_header; /* header for in-log trans */
- unsigned int t_busy_free; /* busy descs free */
- xfs_log_busy_chunk_t t_busy; /* busy/async free blocks */
+ struct list_head t_busy; /* list of busy extents */
unsigned long t_pflags; /* saved process flags state */
} xfs_trans_t;
@@ -1018,9 +593,6 @@ int _xfs_trans_commit(xfs_trans_t *,
void xfs_trans_cancel(xfs_trans_t *, int);
int xfs_trans_ail_init(struct xfs_mount *);
void xfs_trans_ail_destroy(struct xfs_mount *);
-xfs_log_busy_slot_t *xfs_trans_add_busy(xfs_trans_t *tp,
- xfs_agnumber_t ag,
- xfs_extlen_t idx);
extern kmem_zone_t *xfs_trans_zone;
diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c
index 2ffc570679be..e799824f7245 100644
--- a/fs/xfs/xfs_trans_ail.c
+++ b/fs/xfs/xfs_trans_ail.c
@@ -237,14 +237,15 @@ out:
}
/*
- * Function that does the work of pushing on the AIL
+ * xfsaild_push does the work of pushing on the AIL. Returning a timeout of
+ * zero indicates that the caller should sleep until woken.
*/
long
xfsaild_push(
struct xfs_ail *ailp,
xfs_lsn_t *last_lsn)
{
- long tout = 1000; /* milliseconds */
+ long tout = 0;
xfs_lsn_t last_pushed_lsn = *last_lsn;
xfs_lsn_t target = ailp->xa_target;
xfs_lsn_t lsn;
@@ -252,6 +253,7 @@ xfsaild_push(
int flush_log, count, stuck;
xfs_mount_t *mp = ailp->xa_mount;
struct xfs_ail_cursor *cur = &ailp->xa_cursors;
+ int push_xfsbufd = 0;
spin_lock(&ailp->xa_lock);
xfs_trans_ail_cursor_init(ailp, cur);
@@ -262,7 +264,7 @@ xfsaild_push(
*/
xfs_trans_ail_cursor_done(ailp, cur);
spin_unlock(&ailp->xa_lock);
- last_pushed_lsn = 0;
+ *last_lsn = 0;
return tout;
}
@@ -279,7 +281,6 @@ xfsaild_push(
* prevents use from spinning when we can't do anything or there is
* lots of contention on the AIL lists.
*/
- tout = 10;
lsn = lip->li_lsn;
flush_log = stuck = count = 0;
while ((XFS_LSN_CMP(lip->li_lsn, target) < 0)) {
@@ -308,6 +309,7 @@ xfsaild_push(
XFS_STATS_INC(xs_push_ail_pushbuf);
IOP_PUSHBUF(lip);
last_pushed_lsn = lsn;
+ push_xfsbufd = 1;
break;
case XFS_ITEM_PINNED:
@@ -322,12 +324,6 @@ xfsaild_push(
stuck++;
break;
- case XFS_ITEM_FLUSHING:
- XFS_STATS_INC(xs_push_ail_flushing);
- last_pushed_lsn = lsn;
- stuck++;
- break;
-
default:
ASSERT(0);
break;
@@ -371,19 +367,24 @@ xfsaild_push(
* move forward in the AIL.
*/
XFS_STATS_INC(xs_push_ail_flush);
- xfs_log_force(mp, (xfs_lsn_t)0, XFS_LOG_FORCE);
+ xfs_log_force(mp, 0);
+ }
+
+ if (push_xfsbufd) {
+ /* we've got delayed write buffers to flush */
+ wake_up_process(mp->m_ddev_targp->bt_task);
}
if (!count) {
/* We're past our target or empty, so idle */
- tout = 1000;
+ last_pushed_lsn = 0;
} else if (XFS_LSN_CMP(lsn, target) >= 0) {
/*
* We reached the target so wait a bit longer for I/O to
* complete and remove pushed items from the AIL before we
* start the next scan from the start of the AIL.
*/
- tout += 20;
+ tout = 50;
last_pushed_lsn = 0;
} else if ((stuck * 100) / count > 90) {
/*
@@ -395,11 +396,14 @@ xfsaild_push(
* Backoff a bit more to allow some I/O to complete before
* continuing from where we were.
*/
- tout += 10;
+ tout = 20;
+ } else {
+ /* more to do, but wait a short while before continuing */
+ tout = 10;
}
*last_lsn = last_pushed_lsn;
return tout;
-} /* xfsaild_push */
+}
/*
diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c
index 49130628d5ef..63d81a22f4fd 100644
--- a/fs/xfs/xfs_trans_buf.c
+++ b/fs/xfs/xfs_trans_buf.c
@@ -40,12 +40,111 @@
#include "xfs_rw.h"
#include "xfs_trace.h"
+/*
+ * Check to see if a buffer matching the given parameters is already
+ * a part of the given transaction.
+ */
+STATIC struct xfs_buf *
+xfs_trans_buf_item_match(
+ struct xfs_trans *tp,
+ struct xfs_buftarg *target,
+ xfs_daddr_t blkno,
+ int len)
+{
+ xfs_log_item_chunk_t *licp;
+ xfs_log_item_desc_t *lidp;
+ xfs_buf_log_item_t *blip;
+ int i;
+
+ len = BBTOB(len);
+ for (licp = &tp->t_items; licp != NULL; licp = licp->lic_next) {
+ if (xfs_lic_are_all_free(licp)) {
+ ASSERT(licp == &tp->t_items);
+ ASSERT(licp->lic_next == NULL);
+ return NULL;
+ }
+
+ for (i = 0; i < licp->lic_unused; i++) {
+ /*
+ * Skip unoccupied slots.
+ */
+ if (xfs_lic_isfree(licp, i))
+ continue;
+
+ lidp = xfs_lic_slot(licp, i);
+ blip = (xfs_buf_log_item_t *)lidp->lid_item;
+ if (blip->bli_item.li_type != XFS_LI_BUF)
+ continue;
-STATIC xfs_buf_t *xfs_trans_buf_item_match(xfs_trans_t *, xfs_buftarg_t *,
- xfs_daddr_t, int);
-STATIC xfs_buf_t *xfs_trans_buf_item_match_all(xfs_trans_t *, xfs_buftarg_t *,
- xfs_daddr_t, int);
+ if (XFS_BUF_TARGET(blip->bli_buf) == target &&
+ XFS_BUF_ADDR(blip->bli_buf) == blkno &&
+ XFS_BUF_COUNT(blip->bli_buf) == len)
+ return blip->bli_buf;
+ }
+ }
+ return NULL;
+}
+
+/*
+ * Add the locked buffer to the transaction.
+ *
+ * The buffer must be locked, and it cannot be associated with any
+ * transaction.
+ *
+ * If the buffer does not yet have a buf log item associated with it,
+ * then allocate one for it. Then add the buf item to the transaction.
+ */
+STATIC void
+_xfs_trans_bjoin(
+ struct xfs_trans *tp,
+ struct xfs_buf *bp,
+ int reset_recur)
+{
+ struct xfs_buf_log_item *bip;
+
+ ASSERT(XFS_BUF_ISBUSY(bp));
+ ASSERT(XFS_BUF_FSPRIVATE2(bp, void *) == NULL);
+
+ /*
+ * The xfs_buf_log_item pointer is stored in b_fsprivate. If
+ * it doesn't have one yet, then allocate one and initialize it.
+ * The checks to see if one is there are in xfs_buf_item_init().
+ */
+ xfs_buf_item_init(bp, tp->t_mountp);
+ bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
+ ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
+ ASSERT(!(bip->bli_format.blf_flags & XFS_BLF_CANCEL));
+ ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED));
+ if (reset_recur)
+ bip->bli_recur = 0;
+
+ /*
+ * Take a reference for this transaction on the buf item.
+ */
+ atomic_inc(&bip->bli_refcount);
+
+ /*
+ * Get a log_item_desc to point at the new item.
+ */
+ (void) xfs_trans_add_item(tp, (xfs_log_item_t *)bip);
+
+ /*
+ * Initialize b_fsprivate2 so we can find it with incore_match()
+ * in xfs_trans_get_buf() and friends above.
+ */
+ XFS_BUF_SET_FSPRIVATE2(bp, tp);
+
+}
+
+void
+xfs_trans_bjoin(
+ struct xfs_trans *tp,
+ struct xfs_buf *bp)
+{
+ _xfs_trans_bjoin(tp, bp, 0);
+ trace_xfs_trans_bjoin(bp->b_fspriv);
+}
/*
* Get and lock the buffer for the caller if it is not already
@@ -53,14 +152,6 @@ STATIC xfs_buf_t *xfs_trans_buf_item_match_all(xfs_trans_t *, xfs_buftarg_t *,
* within the transaction, just increment its lock recursion count
* and return a pointer to it.
*
- * Use the fast path function xfs_trans_buf_item_match() or the buffer
- * cache routine incore_match() to find the buffer
- * if it is already owned by this transaction.
- *
- * If we don't already own the buffer, use get_buf() to get it.
- * If it doesn't yet have an associated xfs_buf_log_item structure,
- * then allocate one and add the item to this transaction.
- *
* If the transaction pointer is NULL, make this just a normal
* get_buf() call.
*/
@@ -75,13 +166,14 @@ xfs_trans_get_buf(xfs_trans_t *tp,
xfs_buf_log_item_t *bip;
if (flags == 0)
- flags = XFS_BUF_LOCK | XFS_BUF_MAPPED;
+ flags = XBF_LOCK | XBF_MAPPED;
/*
* Default to a normal get_buf() call if the tp is NULL.
*/
if (tp == NULL)
- return xfs_buf_get(target_dev, blkno, len, flags | BUF_BUSY);
+ return xfs_buf_get(target_dev, blkno, len,
+ flags | XBF_DONT_BLOCK);
/*
* If we find the buffer in the cache with this transaction
@@ -89,11 +181,7 @@ xfs_trans_get_buf(xfs_trans_t *tp,
* have it locked. In this case we just increment the lock
* recursion count and return the buffer to the caller.
*/
- if (tp->t_items.lic_next == NULL) {
- bp = xfs_trans_buf_item_match(tp, target_dev, blkno, len);
- } else {
- bp = xfs_trans_buf_item_match_all(tp, target_dev, blkno, len);
- }
+ bp = xfs_trans_buf_item_match(tp, target_dev, blkno, len);
if (bp != NULL) {
ASSERT(XFS_BUF_VALUSEMA(bp) <= 0);
if (XFS_FORCED_SHUTDOWN(tp->t_mountp))
@@ -117,54 +205,22 @@ xfs_trans_get_buf(xfs_trans_t *tp,
}
/*
- * We always specify the BUF_BUSY flag within a transaction so
- * that get_buf does not try to push out a delayed write buffer
+ * We always specify the XBF_DONT_BLOCK flag within a transaction
+ * so that get_buf does not try to push out a delayed write buffer
* which might cause another transaction to take place (if the
* buffer was delayed alloc). Such recursive transactions can
* easily deadlock with our current transaction as well as cause
* us to run out of stack space.
*/
- bp = xfs_buf_get(target_dev, blkno, len, flags | BUF_BUSY);
+ bp = xfs_buf_get(target_dev, blkno, len, flags | XBF_DONT_BLOCK);
if (bp == NULL) {
return NULL;
}
ASSERT(!XFS_BUF_GETERROR(bp));
- /*
- * The xfs_buf_log_item pointer is stored in b_fsprivate. If
- * it doesn't have one yet, then allocate one and initialize it.
- * The checks to see if one is there are in xfs_buf_item_init().
- */
- xfs_buf_item_init(bp, tp->t_mountp);
-
- /*
- * Set the recursion count for the buffer within this transaction
- * to 0.
- */
- bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t*);
- ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
- ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL));
- ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED));
- bip->bli_recur = 0;
-
- /*
- * Take a reference for this transaction on the buf item.
- */
- atomic_inc(&bip->bli_refcount);
-
- /*
- * Get a log_item_desc to point at the new item.
- */
- (void) xfs_trans_add_item(tp, (xfs_log_item_t*)bip);
-
- /*
- * Initialize b_fsprivate2 so we can find it with incore_match()
- * above.
- */
- XFS_BUF_SET_FSPRIVATE2(bp, tp);
-
- trace_xfs_trans_get_buf(bip);
+ _xfs_trans_bjoin(tp, bp, 1);
+ trace_xfs_trans_get_buf(bp->b_fspriv);
return (bp);
}
@@ -209,44 +265,11 @@ xfs_trans_getsb(xfs_trans_t *tp,
}
bp = xfs_getsb(mp, flags);
- if (bp == NULL) {
+ if (bp == NULL)
return NULL;
- }
-
- /*
- * The xfs_buf_log_item pointer is stored in b_fsprivate. If
- * it doesn't have one yet, then allocate one and initialize it.
- * The checks to see if one is there are in xfs_buf_item_init().
- */
- xfs_buf_item_init(bp, mp);
-
- /*
- * Set the recursion count for the buffer within this transaction
- * to 0.
- */
- bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t*);
- ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
- ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL));
- ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED));
- bip->bli_recur = 0;
-
- /*
- * Take a reference for this transaction on the buf item.
- */
- atomic_inc(&bip->bli_refcount);
- /*
- * Get a log_item_desc to point at the new item.
- */
- (void) xfs_trans_add_item(tp, (xfs_log_item_t*)bip);
-
- /*
- * Initialize b_fsprivate2 so we can find it with incore_match()
- * above.
- */
- XFS_BUF_SET_FSPRIVATE2(bp, tp);
-
- trace_xfs_trans_getsb(bip);
+ _xfs_trans_bjoin(tp, bp, 1);
+ trace_xfs_trans_getsb(bp->b_fspriv);
return (bp);
}
@@ -264,14 +287,6 @@ int xfs_error_mod = 33;
* within the transaction and already read in, just increment its
* lock recursion count and return a pointer to it.
*
- * Use the fast path function xfs_trans_buf_item_match() or the buffer
- * cache routine incore_match() to find the buffer
- * if it is already owned by this transaction.
- *
- * If we don't already own the buffer, use read_buf() to get it.
- * If it doesn't yet have an associated xfs_buf_log_item structure,
- * then allocate one and add the item to this transaction.
- *
* If the transaction pointer is NULL, make this just a normal
* read_buf() call.
*/
@@ -290,15 +305,15 @@ xfs_trans_read_buf(
int error;
if (flags == 0)
- flags = XFS_BUF_LOCK | XFS_BUF_MAPPED;
+ flags = XBF_LOCK | XBF_MAPPED;
/*
* Default to a normal get_buf() call if the tp is NULL.
*/
if (tp == NULL) {
- bp = xfs_buf_read(target, blkno, len, flags | BUF_BUSY);
+ bp = xfs_buf_read(target, blkno, len, flags | XBF_DONT_BLOCK);
if (!bp)
- return (flags & XFS_BUF_TRYLOCK) ?
+ return (flags & XBF_TRYLOCK) ?
EAGAIN : XFS_ERROR(ENOMEM);
if (XFS_BUF_GETERROR(bp) != 0) {
@@ -333,11 +348,7 @@ xfs_trans_read_buf(
* If the buffer is not yet read in, then we read it in, increment
* the lock recursion count, and return it to the caller.
*/
- if (tp->t_items.lic_next == NULL) {
- bp = xfs_trans_buf_item_match(tp, target, blkno, len);
- } else {
- bp = xfs_trans_buf_item_match_all(tp, target, blkno, len);
- }
+ bp = xfs_trans_buf_item_match(tp, target, blkno, len);
if (bp != NULL) {
ASSERT(XFS_BUF_VALUSEMA(bp) <= 0);
ASSERT(XFS_BUF_FSPRIVATE2(bp, xfs_trans_t *) == tp);
@@ -385,14 +396,14 @@ xfs_trans_read_buf(
}
/*
- * We always specify the BUF_BUSY flag within a transaction so
- * that get_buf does not try to push out a delayed write buffer
+ * We always specify the XBF_DONT_BLOCK flag within a transaction
+ * so that get_buf does not try to push out a delayed write buffer
* which might cause another transaction to take place (if the
* buffer was delayed alloc). Such recursive transactions can
* easily deadlock with our current transaction as well as cause
* us to run out of stack space.
*/
- bp = xfs_buf_read(target, blkno, len, flags | BUF_BUSY);
+ bp = xfs_buf_read(target, blkno, len, flags | XBF_DONT_BLOCK);
if (bp == NULL) {
*bpp = NULL;
return 0;
@@ -424,40 +435,9 @@ xfs_trans_read_buf(
if (XFS_FORCED_SHUTDOWN(mp))
goto shutdown_abort;
- /*
- * The xfs_buf_log_item pointer is stored in b_fsprivate. If
- * it doesn't have one yet, then allocate one and initialize it.
- * The checks to see if one is there are in xfs_buf_item_init().
- */
- xfs_buf_item_init(bp, tp->t_mountp);
-
- /*
- * Set the recursion count for the buffer within this transaction
- * to 0.
- */
- bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t*);
- ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
- ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL));
- ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED));
- bip->bli_recur = 0;
-
- /*
- * Take a reference for this transaction on the buf item.
- */
- atomic_inc(&bip->bli_refcount);
+ _xfs_trans_bjoin(tp, bp, 1);
+ trace_xfs_trans_read_buf(bp->b_fspriv);
- /*
- * Get a log_item_desc to point at the new item.
- */
- (void) xfs_trans_add_item(tp, (xfs_log_item_t*)bip);
-
- /*
- * Initialize b_fsprivate2 so we can find it with incore_match()
- * above.
- */
- XFS_BUF_SET_FSPRIVATE2(bp, tp);
-
- trace_xfs_trans_read_buf(bip);
*bpp = bp;
return 0;
@@ -472,8 +452,8 @@ shutdown_abort:
if (XFS_BUF_ISSTALE(bp) && XFS_BUF_ISDELAYWRITE(bp))
cmn_err(CE_NOTE, "about to pop assert, bp == 0x%p", bp);
#endif
- ASSERT((XFS_BUF_BFLAGS(bp) & (XFS_B_STALE|XFS_B_DELWRI)) !=
- (XFS_B_STALE|XFS_B_DELWRI));
+ ASSERT((XFS_BUF_BFLAGS(bp) & (XBF_STALE|XBF_DELWRI)) !=
+ (XBF_STALE|XBF_DELWRI));
trace_xfs_trans_read_buf_shut(bp, _RET_IP_);
xfs_buf_relse(bp);
@@ -531,7 +511,7 @@ xfs_trans_brelse(xfs_trans_t *tp,
bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
ASSERT(bip->bli_item.li_type == XFS_LI_BUF);
ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
- ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL));
+ ASSERT(!(bip->bli_format.blf_flags & XFS_BLF_CANCEL));
ASSERT(atomic_read(&bip->bli_refcount) > 0);
/*
@@ -622,53 +602,6 @@ xfs_trans_brelse(xfs_trans_t *tp,
}
/*
- * Add the locked buffer to the transaction.
- * The buffer must be locked, and it cannot be associated with any
- * transaction.
- *
- * If the buffer does not yet have a buf log item associated with it,
- * then allocate one for it. Then add the buf item to the transaction.
- */
-void
-xfs_trans_bjoin(xfs_trans_t *tp,
- xfs_buf_t *bp)
-{
- xfs_buf_log_item_t *bip;
-
- ASSERT(XFS_BUF_ISBUSY(bp));
- ASSERT(XFS_BUF_FSPRIVATE2(bp, void *) == NULL);
-
- /*
- * The xfs_buf_log_item pointer is stored in b_fsprivate. If
- * it doesn't have one yet, then allocate one and initialize it.
- * The checks to see if one is there are in xfs_buf_item_init().
- */
- xfs_buf_item_init(bp, tp->t_mountp);
- bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
- ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
- ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL));
- ASSERT(!(bip->bli_flags & XFS_BLI_LOGGED));
-
- /*
- * Take a reference for this transaction on the buf item.
- */
- atomic_inc(&bip->bli_refcount);
-
- /*
- * Get a log_item_desc to point at the new item.
- */
- (void) xfs_trans_add_item(tp, (xfs_log_item_t *)bip);
-
- /*
- * Initialize b_fsprivate2 so we can find it with incore_match()
- * in xfs_trans_get_buf() and friends above.
- */
- XFS_BUF_SET_FSPRIVATE2(bp, tp);
-
- trace_xfs_trans_bjoin(bip);
-}
-
-/*
* Mark the buffer as not needing to be unlocked when the buf item's
* IOP_UNLOCK() routine is called. The buffer must already be locked
* and associated with the given transaction.
@@ -686,7 +619,7 @@ xfs_trans_bhold(xfs_trans_t *tp,
bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
- ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL));
+ ASSERT(!(bip->bli_format.blf_flags & XFS_BLF_CANCEL));
ASSERT(atomic_read(&bip->bli_refcount) > 0);
bip->bli_flags |= XFS_BLI_HOLD;
trace_xfs_trans_bhold(bip);
@@ -708,7 +641,7 @@ xfs_trans_bhold_release(xfs_trans_t *tp,
bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
ASSERT(!(bip->bli_flags & XFS_BLI_STALE));
- ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_CANCEL));
+ ASSERT(!(bip->bli_format.blf_flags & XFS_BLF_CANCEL));
ASSERT(atomic_read(&bip->bli_refcount) > 0);
ASSERT(bip->bli_flags & XFS_BLI_HOLD);
bip->bli_flags &= ~XFS_BLI_HOLD;
@@ -771,7 +704,7 @@ xfs_trans_log_buf(xfs_trans_t *tp,
bip->bli_flags &= ~XFS_BLI_STALE;
ASSERT(XFS_BUF_ISSTALE(bp));
XFS_BUF_UNSTALE(bp);
- bip->bli_format.blf_flags &= ~XFS_BLI_CANCEL;
+ bip->bli_format.blf_flags &= ~XFS_BLF_CANCEL;
}
lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)bip);
@@ -779,7 +712,6 @@ xfs_trans_log_buf(xfs_trans_t *tp,
tp->t_flags |= XFS_TRANS_DIRTY;
lidp->lid_flags |= XFS_LID_DIRTY;
- lidp->lid_flags &= ~XFS_LID_BUF_STALE;
bip->bli_flags |= XFS_BLI_LOGGED;
xfs_buf_item_log(bip, first, last);
}
@@ -830,8 +762,8 @@ xfs_trans_binval(
ASSERT(!(XFS_BUF_ISDELAYWRITE(bp)));
ASSERT(XFS_BUF_ISSTALE(bp));
ASSERT(!(bip->bli_flags & (XFS_BLI_LOGGED | XFS_BLI_DIRTY)));
- ASSERT(!(bip->bli_format.blf_flags & XFS_BLI_INODE_BUF));
- ASSERT(bip->bli_format.blf_flags & XFS_BLI_CANCEL);
+ ASSERT(!(bip->bli_format.blf_flags & XFS_BLF_INODE_BUF));
+ ASSERT(bip->bli_format.blf_flags & XFS_BLF_CANCEL);
ASSERT(lidp->lid_flags & XFS_LID_DIRTY);
ASSERT(tp->t_flags & XFS_TRANS_DIRTY);
return;
@@ -842,7 +774,7 @@ xfs_trans_binval(
* in the buf log item. The STALE flag will be used in
* xfs_buf_item_unpin() to determine if it should clean up
* when the last reference to the buf item is given up.
- * We set the XFS_BLI_CANCEL flag in the buf log format structure
+ * We set the XFS_BLF_CANCEL flag in the buf log format structure
* and log the buf item. This will be used at recovery time
* to determine that copies of the buffer in the log before
* this should not be replayed.
@@ -860,26 +792,26 @@ xfs_trans_binval(
XFS_BUF_UNDELAYWRITE(bp);
XFS_BUF_STALE(bp);
bip->bli_flags |= XFS_BLI_STALE;
- bip->bli_flags &= ~(XFS_BLI_LOGGED | XFS_BLI_DIRTY);
- bip->bli_format.blf_flags &= ~XFS_BLI_INODE_BUF;
- bip->bli_format.blf_flags |= XFS_BLI_CANCEL;
+ bip->bli_flags &= ~(XFS_BLI_INODE_BUF | XFS_BLI_LOGGED | XFS_BLI_DIRTY);
+ bip->bli_format.blf_flags &= ~XFS_BLF_INODE_BUF;
+ bip->bli_format.blf_flags |= XFS_BLF_CANCEL;
memset((char *)(bip->bli_format.blf_data_map), 0,
(bip->bli_format.blf_map_size * sizeof(uint)));
- lidp->lid_flags |= XFS_LID_DIRTY|XFS_LID_BUF_STALE;
+ lidp->lid_flags |= XFS_LID_DIRTY;
tp->t_flags |= XFS_TRANS_DIRTY;
}
/*
- * This call is used to indicate that the buffer contains on-disk
- * inodes which must be handled specially during recovery. They
- * require special handling because only the di_next_unlinked from
- * the inodes in the buffer should be recovered. The rest of the
- * data in the buffer is logged via the inodes themselves.
+ * This call is used to indicate that the buffer contains on-disk inodes which
+ * must be handled specially during recovery. They require special handling
+ * because only the di_next_unlinked from the inodes in the buffer should be
+ * recovered. The rest of the data in the buffer is logged via the inodes
+ * themselves.
*
- * All we do is set the XFS_BLI_INODE_BUF flag in the buffer's log
- * format structure so that we'll know what to do at recovery time.
+ * All we do is set the XFS_BLI_INODE_BUF flag in the items flags so it can be
+ * transferred to the buffer's log format structure so that we'll know what to
+ * do at recovery time.
*/
-/* ARGSUSED */
void
xfs_trans_inode_buf(
xfs_trans_t *tp,
@@ -894,7 +826,7 @@ xfs_trans_inode_buf(
bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
ASSERT(atomic_read(&bip->bli_refcount) > 0);
- bip->bli_format.blf_flags |= XFS_BLI_INODE_BUF;
+ bip->bli_flags |= XFS_BLI_INODE_BUF;
}
/*
@@ -976,120 +908,12 @@ xfs_trans_dquot_buf(
ASSERT(XFS_BUF_ISBUSY(bp));
ASSERT(XFS_BUF_FSPRIVATE2(bp, xfs_trans_t *) == tp);
ASSERT(XFS_BUF_FSPRIVATE(bp, void *) != NULL);
- ASSERT(type == XFS_BLI_UDQUOT_BUF ||
- type == XFS_BLI_PDQUOT_BUF ||
- type == XFS_BLI_GDQUOT_BUF);
+ ASSERT(type == XFS_BLF_UDQUOT_BUF ||
+ type == XFS_BLF_PDQUOT_BUF ||
+ type == XFS_BLF_GDQUOT_BUF);
bip = XFS_BUF_FSPRIVATE(bp, xfs_buf_log_item_t *);
ASSERT(atomic_read(&bip->bli_refcount) > 0);
bip->bli_format.blf_flags |= type;
}
-
-/*
- * Check to see if a buffer matching the given parameters is already
- * a part of the given transaction. Only check the first, embedded
- * chunk, since we don't want to spend all day scanning large transactions.
- */
-STATIC xfs_buf_t *
-xfs_trans_buf_item_match(
- xfs_trans_t *tp,
- xfs_buftarg_t *target,
- xfs_daddr_t blkno,
- int len)
-{
- xfs_log_item_chunk_t *licp;
- xfs_log_item_desc_t *lidp;
- xfs_buf_log_item_t *blip;
- xfs_buf_t *bp;
- int i;
-
- bp = NULL;
- len = BBTOB(len);
- licp = &tp->t_items;
- if (!xfs_lic_are_all_free(licp)) {
- for (i = 0; i < licp->lic_unused; i++) {
- /*
- * Skip unoccupied slots.
- */
- if (xfs_lic_isfree(licp, i)) {
- continue;
- }
-
- lidp = xfs_lic_slot(licp, i);
- blip = (xfs_buf_log_item_t *)lidp->lid_item;
- if (blip->bli_item.li_type != XFS_LI_BUF) {
- continue;
- }
-
- bp = blip->bli_buf;
- if ((XFS_BUF_TARGET(bp) == target) &&
- (XFS_BUF_ADDR(bp) == blkno) &&
- (XFS_BUF_COUNT(bp) == len)) {
- /*
- * We found it. Break out and
- * return the pointer to the buffer.
- */
- break;
- } else {
- bp = NULL;
- }
- }
- }
- return bp;
-}
-
-/*
- * Check to see if a buffer matching the given parameters is already
- * a part of the given transaction. Check all the chunks, we
- * want to be thorough.
- */
-STATIC xfs_buf_t *
-xfs_trans_buf_item_match_all(
- xfs_trans_t *tp,
- xfs_buftarg_t *target,
- xfs_daddr_t blkno,
- int len)
-{
- xfs_log_item_chunk_t *licp;
- xfs_log_item_desc_t *lidp;
- xfs_buf_log_item_t *blip;
- xfs_buf_t *bp;
- int i;
-
- bp = NULL;
- len = BBTOB(len);
- for (licp = &tp->t_items; licp != NULL; licp = licp->lic_next) {
- if (xfs_lic_are_all_free(licp)) {
- ASSERT(licp == &tp->t_items);
- ASSERT(licp->lic_next == NULL);
- return NULL;
- }
- for (i = 0; i < licp->lic_unused; i++) {
- /*
- * Skip unoccupied slots.
- */
- if (xfs_lic_isfree(licp, i)) {
- continue;
- }
-
- lidp = xfs_lic_slot(licp, i);
- blip = (xfs_buf_log_item_t *)lidp->lid_item;
- if (blip->bli_item.li_type != XFS_LI_BUF) {
- continue;
- }
-
- bp = blip->bli_buf;
- if ((XFS_BUF_TARGET(bp) == target) &&
- (XFS_BUF_ADDR(bp) == blkno) &&
- (XFS_BUF_COUNT(bp) == len)) {
- /*
- * We found it. Break out and
- * return the pointer to the buffer.
- */
- return bp;
- }
- }
- }
- return NULL;
-}
diff --git a/fs/xfs/xfs_trans_inode.c b/fs/xfs/xfs_trans_inode.c
index 785ff101da0a..2559dfec946b 100644
--- a/fs/xfs/xfs_trans_inode.c
+++ b/fs/xfs/xfs_trans_inode.c
@@ -62,7 +62,7 @@ xfs_trans_iget(
{
int error;
- error = xfs_iget(mp, tp, ino, flags, lock_flags, ipp, 0);
+ error = xfs_iget(mp, tp, ino, flags, lock_flags, ipp);
if (!error && tp)
xfs_trans_ijoin(tp, *ipp, lock_flags);
return error;
diff --git a/fs/xfs/xfs_trans_item.c b/fs/xfs/xfs_trans_item.c
index eb3fc57f9eef..f11d37d06dcc 100644
--- a/fs/xfs/xfs_trans_item.c
+++ b/fs/xfs/xfs_trans_item.c
@@ -299,6 +299,7 @@ xfs_trans_next_item(xfs_trans_t *tp, xfs_log_item_desc_t *lidp)
void
xfs_trans_free_items(
xfs_trans_t *tp,
+ xfs_lsn_t commit_lsn,
int flags)
{
xfs_log_item_chunk_t *licp;
@@ -311,7 +312,7 @@ xfs_trans_free_items(
* Special case the embedded chunk so we don't free it below.
*/
if (!xfs_lic_are_all_free(licp)) {
- (void) xfs_trans_unlock_chunk(licp, 1, abort, NULLCOMMITLSN);
+ (void) xfs_trans_unlock_chunk(licp, 1, abort, commit_lsn);
xfs_lic_all_free(licp);
licp->lic_unused = 0;
}
@@ -322,7 +323,7 @@ xfs_trans_free_items(
*/
while (licp != NULL) {
ASSERT(!xfs_lic_are_all_free(licp));
- (void) xfs_trans_unlock_chunk(licp, 1, abort, NULLCOMMITLSN);
+ (void) xfs_trans_unlock_chunk(licp, 1, abort, commit_lsn);
next_licp = licp->lic_next;
kmem_free(licp);
licp = next_licp;
@@ -438,112 +439,3 @@ xfs_trans_unlock_chunk(
return freed;
}
-
-
-/*
- * This is called to add the given busy item to the transaction's
- * list of busy items. It must find a free busy item descriptor
- * or allocate a new one and add the item to that descriptor.
- * The function returns a pointer to busy descriptor used to point
- * to the new busy entry. The log busy entry will now point to its new
- * descriptor with its ???? field.
- */
-xfs_log_busy_slot_t *
-xfs_trans_add_busy(xfs_trans_t *tp, xfs_agnumber_t ag, xfs_extlen_t idx)
-{
- xfs_log_busy_chunk_t *lbcp;
- xfs_log_busy_slot_t *lbsp;
- int i=0;
-
- /*
- * If there are no free descriptors, allocate a new chunk
- * of them and put it at the front of the chunk list.
- */
- if (tp->t_busy_free == 0) {
- lbcp = (xfs_log_busy_chunk_t*)
- kmem_alloc(sizeof(xfs_log_busy_chunk_t), KM_SLEEP);
- ASSERT(lbcp != NULL);
- /*
- * Initialize the chunk, and then
- * claim the first slot in the newly allocated chunk.
- */
- XFS_LBC_INIT(lbcp);
- XFS_LBC_CLAIM(lbcp, 0);
- lbcp->lbc_unused = 1;
- lbsp = XFS_LBC_SLOT(lbcp, 0);
-
- /*
- * Link in the new chunk and update the free count.
- */
- lbcp->lbc_next = tp->t_busy.lbc_next;
- tp->t_busy.lbc_next = lbcp;
- tp->t_busy_free = XFS_LIC_NUM_SLOTS - 1;
-
- /*
- * Initialize the descriptor and the generic portion
- * of the log item.
- *
- * Point the new slot at this item and return it.
- * Also point the log item at its currently active
- * descriptor and set the item's mount pointer.
- */
- lbsp->lbc_ag = ag;
- lbsp->lbc_idx = idx;
- return lbsp;
- }
-
- /*
- * Find the free descriptor. It is somewhere in the chunklist
- * of descriptors.
- */
- lbcp = &tp->t_busy;
- while (lbcp != NULL) {
- if (XFS_LBC_VACANCY(lbcp)) {
- if (lbcp->lbc_unused <= XFS_LBC_MAX_SLOT) {
- i = lbcp->lbc_unused;
- break;
- } else {
- /* out-of-order vacancy */
- cmn_err(CE_DEBUG, "OOO vacancy lbcp 0x%p\n", lbcp);
- ASSERT(0);
- }
- }
- lbcp = lbcp->lbc_next;
- }
- ASSERT(lbcp != NULL);
- /*
- * If we find a free descriptor, claim it,
- * initialize it, and return it.
- */
- XFS_LBC_CLAIM(lbcp, i);
- if (lbcp->lbc_unused <= i) {
- lbcp->lbc_unused = i + 1;
- }
- lbsp = XFS_LBC_SLOT(lbcp, i);
- tp->t_busy_free--;
- lbsp->lbc_ag = ag;
- lbsp->lbc_idx = idx;
- return lbsp;
-}
-
-
-/*
- * xfs_trans_free_busy
- * Free all of the busy lists from a transaction
- */
-void
-xfs_trans_free_busy(xfs_trans_t *tp)
-{
- xfs_log_busy_chunk_t *lbcp;
- xfs_log_busy_chunk_t *lbcq;
-
- lbcp = tp->t_busy.lbc_next;
- while (lbcp != NULL) {
- lbcq = lbcp->lbc_next;
- kmem_free(lbcp);
- lbcp = lbcq;
- }
-
- XFS_LBC_INIT(&tp->t_busy);
- tp->t_busy.lbc_unused = 0;
-}
diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h
index 73e2ad397432..c6e4f2c8de6e 100644
--- a/fs/xfs/xfs_trans_priv.h
+++ b/fs/xfs/xfs_trans_priv.h
@@ -35,13 +35,14 @@ struct xfs_log_item_desc *xfs_trans_find_item(struct xfs_trans *,
struct xfs_log_item_desc *xfs_trans_first_item(struct xfs_trans *);
struct xfs_log_item_desc *xfs_trans_next_item(struct xfs_trans *,
struct xfs_log_item_desc *);
-void xfs_trans_free_items(struct xfs_trans *, int);
-void xfs_trans_unlock_items(struct xfs_trans *,
- xfs_lsn_t);
-void xfs_trans_free_busy(xfs_trans_t *tp);
-xfs_log_busy_slot_t *xfs_trans_add_busy(xfs_trans_t *tp,
- xfs_agnumber_t ag,
- xfs_extlen_t idx);
+
+void xfs_trans_unlock_items(struct xfs_trans *tp, xfs_lsn_t commit_lsn);
+void xfs_trans_free_items(struct xfs_trans *tp, xfs_lsn_t commit_lsn,
+ int flags);
+
+void xfs_trans_item_committed(struct xfs_log_item *lip,
+ xfs_lsn_t commit_lsn, int aborted);
+void xfs_trans_unreserve_and_mod_sb(struct xfs_trans *tp);
/*
* AIL traversal cursor.
diff --git a/fs/xfs/xfs_types.h b/fs/xfs/xfs_types.h
index d725428c9df6..320775295e32 100644
--- a/fs/xfs/xfs_types.h
+++ b/fs/xfs/xfs_types.h
@@ -75,6 +75,8 @@ typedef __uint32_t xfs_dahash_t; /* dir/attr hash value */
typedef __uint16_t xfs_prid_t; /* prid_t truncated to 16bits in XFS */
+typedef __uint32_t xlog_tid_t; /* transaction ID type */
+
/*
* These types are 64 bits on disk but are either 32 or 64 bits in memory.
* Disk based types:
@@ -151,8 +153,8 @@ typedef enum {
} xfs_btnum_t;
struct xfs_name {
- const char *name;
- int len;
+ const unsigned char *name;
+ int len;
};
#endif /* __XFS_TYPES_H__ */
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index 6f268756bf36..c1646838898f 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -256,7 +256,7 @@ xfs_setattr(
iattr->ia_size > ip->i_d.di_size) {
code = xfs_flush_pages(ip,
ip->i_d.di_size, iattr->ia_size,
- XFS_B_ASYNC, FI_NONE);
+ XBF_ASYNC, FI_NONE);
}
/* wait for all I/O to complete */
@@ -267,7 +267,7 @@ xfs_setattr(
if (code) {
ASSERT(tp == NULL);
lock_flags &= ~XFS_ILOCK_EXCL;
- ASSERT(lock_flags == XFS_IOLOCK_EXCL);
+ ASSERT(lock_flags == XFS_IOLOCK_EXCL || !need_iolock);
goto error_return;
}
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_SIZE);
@@ -584,116 +584,6 @@ xfs_readlink(
}
/*
- * xfs_fsync
- *
- * This is called to sync the inode and its data out to disk. We need to hold
- * the I/O lock while flushing the data, and the inode lock while flushing the
- * inode. The inode lock CANNOT be held while flushing the data, so acquire
- * after we're done with that.
- */
-int
-xfs_fsync(
- xfs_inode_t *ip)
-{
- xfs_trans_t *tp;
- int error = 0;
- int log_flushed = 0, changed = 1;
-
- xfs_itrace_entry(ip);
-
- if (XFS_FORCED_SHUTDOWN(ip->i_mount))
- return XFS_ERROR(EIO);
-
- /*
- * We always need to make sure that the required inode state is safe on
- * disk. The inode might be clean but we still might need to force the
- * log because of committed transactions that haven't hit the disk yet.
- * Likewise, there could be unflushed non-transactional changes to the
- * inode core that have to go to disk and this requires us to issue
- * a synchronous transaction to capture these changes correctly.
- *
- * This code relies on the assumption that if the update_* fields
- * of the inode are clear and the inode is unpinned then it is clean
- * and no action is required.
- */
- xfs_ilock(ip, XFS_ILOCK_SHARED);
-
- if (!ip->i_update_core) {
- /*
- * Timestamps/size haven't changed since last inode flush or
- * inode transaction commit. That means either nothing got
- * written or a transaction committed which caught the updates.
- * If the latter happened and the transaction hasn't hit the
- * disk yet, the inode will be still be pinned. If it is,
- * force the log.
- */
-
- xfs_iunlock(ip, XFS_ILOCK_SHARED);
-
- if (xfs_ipincount(ip)) {
- error = _xfs_log_force(ip->i_mount, (xfs_lsn_t)0,
- XFS_LOG_FORCE | XFS_LOG_SYNC,
- &log_flushed);
- } else {
- /*
- * If the inode is not pinned and nothing has changed
- * we don't need to flush the cache.
- */
- changed = 0;
- }
- } else {
- /*
- * Kick off a transaction to log the inode core to get the
- * updates. The sync transaction will also force the log.
- */
- xfs_iunlock(ip, XFS_ILOCK_SHARED);
- tp = xfs_trans_alloc(ip->i_mount, XFS_TRANS_FSYNC_TS);
- error = xfs_trans_reserve(tp, 0,
- XFS_FSYNC_TS_LOG_RES(ip->i_mount), 0, 0, 0);
- if (error) {
- xfs_trans_cancel(tp, 0);
- return error;
- }
- xfs_ilock(ip, XFS_ILOCK_EXCL);
-
- /*
- * Note - it's possible that we might have pushed ourselves out
- * of the way during trans_reserve which would flush the inode.
- * But there's no guarantee that the inode buffer has actually
- * gone out yet (it's delwri). Plus the buffer could be pinned
- * anyway if it's part of an inode in another recent
- * transaction. So we play it safe and fire off the
- * transaction anyway.
- */
- xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
- xfs_trans_ihold(tp, ip);
- xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
- xfs_trans_set_sync(tp);
- error = _xfs_trans_commit(tp, 0, &log_flushed);
-
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
-
- if ((ip->i_mount->m_flags & XFS_MOUNT_BARRIER) && changed) {
- /*
- * If the log write didn't issue an ordered tag we need
- * to flush the disk cache for the data device now.
- */
- if (!log_flushed)
- xfs_blkdev_issue_flush(ip->i_mount->m_ddev_targp);
-
- /*
- * If this inode is on the RT dev we need to flush that
- * cache as well.
- */
- if (XFS_IS_REALTIME_INODE(ip))
- xfs_blkdev_issue_flush(ip->i_mount->m_rtdev_targp);
- }
-
- return error;
-}
-
-/*
* Flags for xfs_free_eofblocks
*/
#define XFS_FREE_EOF_TRYLOCK (1<<0)
@@ -1096,7 +986,7 @@ xfs_release(
*/
truncated = xfs_iflags_test_and_clear(ip, XFS_ITRUNCATED);
if (truncated && VN_DIRTY(VFS_I(ip)) && ip->i_delayed_blks > 0)
- xfs_flush_pages(ip, 0, -1, XFS_B_ASYNC, FI_NONE);
+ xfs_flush_pages(ip, 0, -1, XBF_ASYNC, FI_NONE);
}
if (ip->i_d.di_nlink != 0) {
@@ -1379,7 +1269,7 @@ xfs_lookup(
if (error)
goto out;
- error = xfs_iget(dp->i_mount, NULL, inum, 0, 0, ipp, 0);
+ error = xfs_iget(dp->i_mount, NULL, inum, 0, 0, ipp);
if (error)
goto out_free_name;
@@ -2199,7 +2089,8 @@ xfs_symlink(
if (DM_EVENT_ENABLED(dp, DM_EVENT_SYMLINK)) {
error = XFS_SEND_NAMESP(mp, DM_EVENT_SYMLINK, dp,
DM_RIGHT_NULL, NULL, DM_RIGHT_NULL,
- link_name->name, target_path, 0, 0, 0);
+ link_name->name,
+ (unsigned char *)target_path, 0, 0, 0);
if (error)
return error;
}
@@ -2395,7 +2286,8 @@ std_return:
dp, DM_RIGHT_NULL,
error ? NULL : ip,
DM_RIGHT_NULL, link_name->name,
- target_path, 0, error, 0);
+ (unsigned char *)target_path,
+ 0, error, 0);
}
if (!error)
diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h
index 167a467403a5..d8dfa8d0dadd 100644
--- a/fs/xfs/xfs_vnodeops.h
+++ b/fs/xfs/xfs_vnodeops.h
@@ -21,7 +21,6 @@ int xfs_setattr(struct xfs_inode *ip, struct iattr *vap, int flags);
#define XFS_ATTR_NOACL 0x08 /* Don't call xfs_acl_chmod */
int xfs_readlink(struct xfs_inode *ip, char *link);
-int xfs_fsync(struct xfs_inode *ip);
int xfs_release(struct xfs_inode *ip);
int xfs_inactive(struct xfs_inode *ip);
int xfs_lookup(struct xfs_inode *dp, struct xfs_name *name,
@@ -43,25 +42,13 @@ int xfs_change_file_space(struct xfs_inode *ip, int cmd,
int xfs_rename(struct xfs_inode *src_dp, struct xfs_name *src_name,
struct xfs_inode *src_ip, struct xfs_inode *target_dp,
struct xfs_name *target_name, struct xfs_inode *target_ip);
-int xfs_attr_get(struct xfs_inode *ip, const char *name, char *value,
- int *valuelenp, int flags);
-int xfs_attr_set(struct xfs_inode *dp, const char *name, char *value,
- int valuelen, int flags);
-int xfs_attr_remove(struct xfs_inode *dp, const char *name, int flags);
+int xfs_attr_get(struct xfs_inode *ip, const unsigned char *name,
+ unsigned char *value, int *valuelenp, int flags);
+int xfs_attr_set(struct xfs_inode *dp, const unsigned char *name,
+ unsigned char *value, int valuelen, int flags);
+int xfs_attr_remove(struct xfs_inode *dp, const unsigned char *name, int flags);
int xfs_attr_list(struct xfs_inode *dp, char *buffer, int bufsize,
int flags, struct attrlist_cursor_kern *cursor);
-ssize_t xfs_read(struct xfs_inode *ip, struct kiocb *iocb,
- const struct iovec *iovp, unsigned int segs,
- loff_t *offset, int ioflags);
-ssize_t xfs_splice_read(struct xfs_inode *ip, struct file *infilp,
- loff_t *ppos, struct pipe_inode_info *pipe, size_t count,
- int flags, int ioflags);
-ssize_t xfs_splice_write(struct xfs_inode *ip,
- struct pipe_inode_info *pipe, struct file *outfilp,
- loff_t *ppos, size_t count, int flags, int ioflags);
-ssize_t xfs_write(struct xfs_inode *xip, struct kiocb *iocb,
- const struct iovec *iovp, unsigned int nsegs,
- loff_t *offset, int ioflags);
int xfs_bmap(struct xfs_inode *ip, xfs_off_t offset, ssize_t count,
int flags, struct xfs_iomap *iomapp, int *niomaps);
void xfs_tosspages(struct xfs_inode *inode, xfs_off_t first,
@@ -72,4 +59,6 @@ int xfs_flush_pages(struct xfs_inode *ip, xfs_off_t first,
xfs_off_t last, uint64_t flags, int fiopt);
int xfs_wait_on_pages(struct xfs_inode *ip, xfs_off_t first, xfs_off_t last);
+int xfs_zero_eof(struct xfs_inode *, xfs_off_t, xfs_fsize_t);
+
#endif /* _XFS_VNODEOPS_H */