summaryrefslogtreecommitdiffstats
path: root/fs/xfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/xfs')
-rw-r--r--fs/xfs/Makefile3
-rw-r--r--fs/xfs/linux-2.6/kmem.c6
-rw-r--r--fs/xfs/linux-2.6/kmem.h6
-rw-r--r--fs/xfs/linux-2.6/sema.h52
-rw-r--r--fs/xfs/linux-2.6/xfs_aops.c18
-rw-r--r--fs/xfs/linux-2.6/xfs_buf.c33
-rw-r--r--fs/xfs/linux-2.6/xfs_buf.h14
-rw-r--r--fs/xfs/linux-2.6/xfs_export.c12
-rw-r--r--fs/xfs/linux-2.6/xfs_file.c1
-rw-r--r--fs/xfs/linux-2.6/xfs_fs_subr.c6
-rw-r--r--fs/xfs/linux-2.6/xfs_ioctl.c392
-rw-r--r--fs/xfs/linux-2.6/xfs_iops.c534
-rw-r--r--fs/xfs/linux-2.6/xfs_iops.h15
-rw-r--r--fs/xfs/linux-2.6/xfs_linux.h14
-rw-r--r--fs/xfs/linux-2.6/xfs_lrw.c8
-rw-r--r--fs/xfs/linux-2.6/xfs_stats.c15
-rw-r--r--fs/xfs/linux-2.6/xfs_stats.h12
-rw-r--r--fs/xfs/linux-2.6/xfs_super.c1108
-rw-r--r--fs/xfs/linux-2.6/xfs_super.h7
-rw-r--r--fs/xfs/linux-2.6/xfs_sysctl.c8
-rw-r--r--fs/xfs/linux-2.6/xfs_sysctl.h4
-rw-r--r--fs/xfs/linux-2.6/xfs_vnode.c72
-rw-r--r--fs/xfs/linux-2.6/xfs_vnode.h154
-rw-r--r--fs/xfs/linux-2.6/xfs_xattr.c330
-rw-r--r--fs/xfs/quota/xfs_dquot.c41
-rw-r--r--fs/xfs/quota/xfs_dquot.h31
-rw-r--r--fs/xfs/quota/xfs_dquot_item.c12
-rw-r--r--fs/xfs/quota/xfs_qm.c38
-rw-r--r--fs/xfs/quota/xfs_qm.h2
-rw-r--r--fs/xfs/quota/xfs_qm_bhv.c7
-rw-r--r--fs/xfs/quota/xfs_qm_syscalls.c16
-rw-r--r--fs/xfs/quota/xfs_quota_priv.h3
-rw-r--r--fs/xfs/support/ktrace.c4
-rw-r--r--fs/xfs/support/uuid.c8
-rw-r--r--fs/xfs/support/uuid.h1
-rw-r--r--fs/xfs/xfs_acl.c69
-rw-r--r--fs/xfs/xfs_acl.h18
-rw-r--r--fs/xfs/xfs_arch.h68
-rw-r--r--fs/xfs/xfs_attr.c718
-rw-r--r--fs/xfs/xfs_attr.h91
-rw-r--r--fs/xfs/xfs_attr_leaf.c174
-rw-r--r--fs/xfs/xfs_attr_leaf.h31
-rw-r--r--fs/xfs/xfs_attr_sf.h10
-rw-r--r--fs/xfs/xfs_bit.c103
-rw-r--r--fs/xfs/xfs_bit.h34
-rw-r--r--fs/xfs/xfs_bmap.c152
-rw-r--r--fs/xfs/xfs_bmap.h13
-rw-r--r--fs/xfs/xfs_bmap_btree.c76
-rw-r--r--fs/xfs/xfs_btree.c105
-rw-r--r--fs/xfs/xfs_btree.h8
-rw-r--r--fs/xfs/xfs_buf_item.c48
-rw-r--r--fs/xfs/xfs_clnt.h1
-rw-r--r--fs/xfs/xfs_da_btree.c48
-rw-r--r--fs/xfs/xfs_da_btree.h36
-rw-r--r--fs/xfs/xfs_dfrag.c44
-rw-r--r--fs/xfs/xfs_dir2.c125
-rw-r--r--fs/xfs/xfs_dir2.h6
-rw-r--r--fs/xfs/xfs_dir2_block.c56
-rw-r--r--fs/xfs/xfs_dir2_data.c5
-rw-r--r--fs/xfs/xfs_dir2_leaf.c93
-rw-r--r--fs/xfs/xfs_dir2_node.c402
-rw-r--r--fs/xfs/xfs_dir2_sf.c83
-rw-r--r--fs/xfs/xfs_dir2_sf.h6
-rw-r--r--fs/xfs/xfs_dir2_trace.c20
-rw-r--r--fs/xfs/xfs_dmapi.h3
-rw-r--r--fs/xfs/xfs_error.c18
-rw-r--r--fs/xfs/xfs_error.h13
-rw-r--r--fs/xfs/xfs_extfree_item.c6
-rw-r--r--fs/xfs/xfs_filestream.c6
-rw-r--r--fs/xfs/xfs_fs.h4
-rw-r--r--fs/xfs/xfs_fsops.c4
-rw-r--r--fs/xfs/xfs_ialloc_btree.c30
-rw-r--r--fs/xfs/xfs_iget.c48
-rw-r--r--fs/xfs/xfs_inode.c235
-rw-r--r--fs/xfs/xfs_inode.h49
-rw-r--r--fs/xfs/xfs_inode_item.c18
-rw-r--r--fs/xfs/xfs_iomap.c10
-rw-r--r--fs/xfs/xfs_itable.c10
-rw-r--r--fs/xfs/xfs_log.c215
-rw-r--r--fs/xfs/xfs_log.h2
-rw-r--r--fs/xfs/xfs_log_priv.h21
-rw-r--r--fs/xfs/xfs_log_recover.c28
-rw-r--r--fs/xfs/xfs_mount.c188
-rw-r--r--fs/xfs/xfs_mount.h32
-rw-r--r--fs/xfs/xfs_mru_cache.c21
-rw-r--r--fs/xfs/xfs_rename.c22
-rw-r--r--fs/xfs/xfs_rtalloc.c21
-rw-r--r--fs/xfs/xfs_rw.c2
-rw-r--r--fs/xfs/xfs_sb.h17
-rw-r--r--fs/xfs/xfs_trans.c79
-rw-r--r--fs/xfs/xfs_trans.h12
-rw-r--r--fs/xfs/xfs_trans_buf.c12
-rw-r--r--fs/xfs/xfs_trans_inode.c2
-rw-r--r--fs/xfs/xfs_trans_item.c74
-rw-r--r--fs/xfs/xfs_utils.c4
-rw-r--r--fs/xfs/xfs_utils.h3
-rw-r--r--fs/xfs/xfs_vfsops.c623
-rw-r--r--fs/xfs/xfs_vfsops.h5
-rw-r--r--fs/xfs/xfs_vnodeops.c934
-rw-r--r--fs/xfs/xfs_vnodeops.h12
100 files changed, 3912 insertions, 4501 deletions
diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index 36ec614e699..737c9a42536 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -106,7 +106,8 @@ xfs-y += $(addprefix $(XFS_LINUX)/, \
xfs_iops.o \
xfs_lrw.o \
xfs_super.o \
- xfs_vnode.o)
+ xfs_vnode.o \
+ xfs_xattr.o)
# Objects in support/
xfs-y += $(addprefix support/, \
diff --git a/fs/xfs/linux-2.6/kmem.c b/fs/xfs/linux-2.6/kmem.c
index 9b1bb17a050..1cd3b55ee3d 100644
--- a/fs/xfs/linux-2.6/kmem.c
+++ b/fs/xfs/linux-2.6/kmem.c
@@ -90,7 +90,7 @@ kmem_zalloc_greedy(size_t *size, size_t minsize, size_t maxsize,
}
void
-kmem_free(void *ptr, size_t size)
+kmem_free(const void *ptr)
{
if (!is_vmalloc_addr(ptr)) {
kfree(ptr);
@@ -100,7 +100,7 @@ kmem_free(void *ptr, size_t size)
}
void *
-kmem_realloc(void *ptr, size_t newsize, size_t oldsize,
+kmem_realloc(const void *ptr, size_t newsize, size_t oldsize,
unsigned int __nocast flags)
{
void *new;
@@ -110,7 +110,7 @@ kmem_realloc(void *ptr, size_t newsize, size_t oldsize,
if (new)
memcpy(new, ptr,
((oldsize < newsize) ? oldsize : newsize));
- kmem_free(ptr, oldsize);
+ kmem_free(ptr);
}
return new;
}
diff --git a/fs/xfs/linux-2.6/kmem.h b/fs/xfs/linux-2.6/kmem.h
index 5e956490297..af6843c7ee4 100644
--- a/fs/xfs/linux-2.6/kmem.h
+++ b/fs/xfs/linux-2.6/kmem.h
@@ -57,8 +57,8 @@ 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(void *, size_t, size_t, unsigned int __nocast);
-extern void kmem_free(void *, size_t);
+extern void *kmem_realloc(const void *, size_t, size_t, unsigned int __nocast);
+extern void kmem_free(const void *);
/*
* Zone interfaces
@@ -79,7 +79,7 @@ kmem_zone_init(int size, char *zone_name)
static inline kmem_zone_t *
kmem_zone_init_flags(int size, char *zone_name, unsigned long flags,
- void (*construct)(kmem_zone_t *, void *))
+ void (*construct)(void *))
{
return kmem_cache_create(zone_name, size, 0, flags, construct);
}
diff --git a/fs/xfs/linux-2.6/sema.h b/fs/xfs/linux-2.6/sema.h
deleted file mode 100644
index 3abe7e9ceb3..00000000000
--- a/fs/xfs/linux-2.6/sema.h
+++ /dev/null
@@ -1,52 +0,0 @@
-/*
- * Copyright (c) 2000-2002,2005 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_SUPPORT_SEMA_H__
-#define __XFS_SUPPORT_SEMA_H__
-
-#include <linux/time.h>
-#include <linux/wait.h>
-#include <linux/semaphore.h>
-#include <asm/atomic.h>
-
-/*
- * sema_t structure just maps to struct semaphore in Linux kernel.
- */
-
-typedef struct semaphore sema_t;
-
-#define initnsema(sp, val, name) sema_init(sp, val)
-#define psema(sp, b) down(sp)
-#define vsema(sp) up(sp)
-#define freesema(sema) do { } while (0)
-
-static inline int issemalocked(sema_t *sp)
-{
- return down_trylock(sp) || (up(sp), 0);
-}
-
-/*
- * Map cpsema (try to get the sema) to down_trylock. We need to switch
- * the return values since cpsema returns 1 (acquired) 0 (failed) and
- * down_trylock returns the reverse 0 (acquired) 1 (failed).
- */
-static inline int cpsema(sema_t *sp)
-{
- return down_trylock(sp) ? 0 : 1;
-}
-
-#endif /* __XFS_SUPPORT_SEMA_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index a55c3b26d84..a44d68eb50b 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -73,7 +73,6 @@ xfs_page_trace(
unsigned long pgoff)
{
xfs_inode_t *ip;
- bhv_vnode_t *vp = vn_from_inode(inode);
loff_t isize = i_size_read(inode);
loff_t offset = page_offset(page);
int delalloc = -1, unmapped = -1, unwritten = -1;
@@ -81,7 +80,7 @@ xfs_page_trace(
if (page_has_buffers(page))
xfs_count_page_state(page, &delalloc, &unmapped, &unwritten);
- ip = xfs_vtoi(vp);
+ ip = XFS_I(inode);
if (!ip->i_rwtrace)
return;
@@ -409,7 +408,6 @@ xfs_start_buffer_writeback(
STATIC void
xfs_start_page_writeback(
struct page *page,
- struct writeback_control *wbc,
int clear_dirty,
int buffers)
{
@@ -676,7 +674,7 @@ xfs_probe_cluster(
} else
pg_offset = PAGE_CACHE_SIZE;
- if (page->index == tindex && !TestSetPageLocked(page)) {
+ if (page->index == tindex && trylock_page(page)) {
pg_len = xfs_probe_page(page, pg_offset, mapped);
unlock_page(page);
}
@@ -760,7 +758,7 @@ xfs_convert_page(
if (page->index != tindex)
goto fail;
- if (TestSetPageLocked(page))
+ if (!trylock_page(page))
goto fail;
if (PageWriteback(page))
goto fail_unlock_page;
@@ -858,7 +856,7 @@ xfs_convert_page(
done = 1;
}
}
- xfs_start_page_writeback(page, wbc, !page_dirty, count);
+ xfs_start_page_writeback(page, !page_dirty, count);
}
return done;
@@ -1105,7 +1103,7 @@ xfs_page_state_convert(
* that we are writing into for the first time.
*/
type = IOMAP_NEW;
- if (!test_and_set_bit(BH_Lock, &bh->b_state)) {
+ if (trylock_buffer(bh)) {
ASSERT(buffer_mapped(bh));
if (iomap_valid)
all_bh = 1;
@@ -1130,7 +1128,7 @@ xfs_page_state_convert(
SetPageUptodate(page);
if (startio)
- xfs_start_page_writeback(page, wbc, 1, count);
+ xfs_start_page_writeback(page, 1, count);
if (ioend && iomap_valid) {
offset = (iomap.iomap_offset + iomap.iomap_bsize - 1) >>
@@ -1340,6 +1338,10 @@ __xfs_get_blocks(
offset = (xfs_off_t)iblock << inode->i_blkbits;
ASSERT(bh_result->b_size >= (1 << inode->i_blkbits));
size = bh_result->b_size;
+
+ if (!create && direct && offset >= i_size_read(inode))
+ return 0;
+
error = xfs_iomap(XFS_I(inode), offset, size,
create ? flags : BMAPI_READ, &iomap, &niomap);
if (error)
diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index 98e0e86093b..36d5fcd3f59 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -58,7 +58,7 @@ xfs_buf_trace(
bp, id,
(void *)(unsigned long)bp->b_flags,
(void *)(unsigned long)bp->b_hold.counter,
- (void *)(unsigned long)bp->b_sema.count.counter,
+ (void *)(unsigned long)bp->b_sema.count,
(void *)current,
data, ra,
(void *)(unsigned long)((bp->b_file_offset>>32) & 0xffffffff),
@@ -253,7 +253,7 @@ _xfs_buf_initialize(
memset(bp, 0, sizeof(xfs_buf_t));
atomic_set(&bp->b_hold, 1);
- init_MUTEX_LOCKED(&bp->b_iodonesema);
+ init_completion(&bp->b_iowait);
INIT_LIST_HEAD(&bp->b_list);
INIT_LIST_HEAD(&bp->b_hash_list);
init_MUTEX_LOCKED(&bp->b_sema); /* held, no waiters */
@@ -310,8 +310,7 @@ _xfs_buf_free_pages(
xfs_buf_t *bp)
{
if (bp->b_pages != bp->b_page_array) {
- kmem_free(bp->b_pages,
- bp->b_page_count * sizeof(struct page *));
+ kmem_free(bp->b_pages);
}
}
@@ -839,6 +838,7 @@ xfs_buf_rele(
return;
}
+ ASSERT(atomic_read(&bp->b_hold) > 0);
if (atomic_dec_and_lock(&bp->b_hold, &hash->bh_lock)) {
if (bp->b_relse) {
atomic_inc(&bp->b_hold);
@@ -852,11 +852,6 @@ xfs_buf_rele(
spin_unlock(&hash->bh_lock);
xfs_buf_free(bp);
}
- } else {
- /*
- * Catch reference count leaks
- */
- ASSERT(atomic_read(&bp->b_hold) >= 0);
}
}
@@ -1006,12 +1001,13 @@ xfs_buf_iodone_work(
* We can get an EOPNOTSUPP to ordered writes. Here we clear the
* ordered flag and reissue them. Because we can't tell the higher
* layers directly that they should not issue ordered I/O anymore, they
- * need to check if the ordered flag was cleared during I/O completion.
+ * need to check if the _XFS_BARRIER_FAILED flag was set during I/O completion.
*/
if ((bp->b_error == EOPNOTSUPP) &&
(bp->b_flags & (XBF_ORDERED|XBF_ASYNC)) == (XBF_ORDERED|XBF_ASYNC)) {
XB_TRACE(bp, "ordered_retry", bp->b_iodone);
bp->b_flags &= ~XBF_ORDERED;
+ bp->b_flags |= _XFS_BARRIER_FAILED;
xfs_buf_iorequest(bp);
} else if (bp->b_iodone)
(*(bp->b_iodone))(bp);
@@ -1038,7 +1034,7 @@ xfs_buf_ioend(
xfs_buf_iodone_work(&bp->b_iodone_work);
}
} else {
- up(&bp->b_iodonesema);
+ complete(&bp->b_iowait);
}
}
@@ -1276,7 +1272,7 @@ xfs_buf_iowait(
XB_TRACE(bp, "iowait", 0);
if (atomic_read(&bp->b_io_remaining))
blk_run_address_space(bp->b_target->bt_mapping);
- down(&bp->b_iodonesema);
+ wait_for_completion(&bp->b_iowait);
XB_TRACE(bp, "iowaited", (long)bp->b_error);
return bp->b_error;
}
@@ -1398,7 +1394,7 @@ STATIC void
xfs_free_bufhash(
xfs_buftarg_t *btp)
{
- kmem_free(btp->bt_hash, (1<<btp->bt_hashshift) * sizeof(xfs_bufhash_t));
+ kmem_free(btp->bt_hash);
btp->bt_hash = NULL;
}
@@ -1428,13 +1424,10 @@ xfs_unregister_buftarg(
void
xfs_free_buftarg(
- xfs_buftarg_t *btp,
- int external)
+ xfs_buftarg_t *btp)
{
xfs_flush_buftarg(btp, 1);
xfs_blkdev_issue_flush(btp);
- if (external)
- xfs_blkdev_put(btp->bt_bdev);
xfs_free_bufhash(btp);
iput(btp->bt_mapping->host);
@@ -1444,7 +1437,7 @@ xfs_free_buftarg(
xfs_unregister_buftarg(btp);
kthread_stop(btp->bt_task);
- kmem_free(btp, sizeof(*btp));
+ kmem_free(btp);
}
STATIC int
@@ -1575,7 +1568,7 @@ xfs_alloc_buftarg(
return btp;
error:
- kmem_free(btp, sizeof(*btp));
+ kmem_free(btp);
return NULL;
}
@@ -1803,7 +1796,7 @@ int __init
xfs_buf_init(void)
{
#ifdef XFS_BUF_TRACE
- xfs_buf_trace_buf = ktrace_alloc(XFS_BUF_TRACE_SIZE, KM_SLEEP);
+ xfs_buf_trace_buf = ktrace_alloc(XFS_BUF_TRACE_SIZE, KM_NOFS);
#endif
xfs_buf_zone = kmem_zone_init_flags(sizeof(xfs_buf_t), "xfs_buf",
diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h
index f948ec7ba9a..456519a088c 100644
--- a/fs/xfs/linux-2.6/xfs_buf.h
+++ b/fs/xfs/linux-2.6/xfs_buf.h
@@ -85,6 +85,14 @@ typedef enum {
* modifications being lost.
*/
_XBF_PAGE_LOCKED = (1 << 22),
+
+ /*
+ * If we try a barrier write, but it fails we have to communicate
+ * this to the upper layers. Unfortunately b_error gets overwritten
+ * when the buffer is re-issued so we have to add another flag to
+ * keep this information.
+ */
+ _XFS_BARRIER_FAILED = (1 << 23),
} xfs_buf_flags_t;
typedef enum {
@@ -157,7 +165,7 @@ typedef struct xfs_buf {
xfs_buf_iodone_t b_iodone; /* I/O completion function */
xfs_buf_relse_t b_relse; /* releasing function */
xfs_buf_bdstrat_t b_strat; /* pre-write function */
- struct semaphore b_iodonesema; /* Semaphore for I/O waiters */
+ struct completion b_iowait; /* queue for I/O waiters */
void *b_fspriv;
void *b_fspriv2;
void *b_fspriv3;
@@ -352,7 +360,7 @@ extern void xfs_buf_trace(xfs_buf_t *, char *, void *, void *);
#define XFS_BUF_CPSEMA(bp) (xfs_buf_cond_lock(bp) == 0)
#define XFS_BUF_VSEMA(bp) xfs_buf_unlock(bp)
#define XFS_BUF_PSEMA(bp,x) xfs_buf_lock(bp)
-#define XFS_BUF_V_IODONESEMA(bp) up(&bp->b_iodonesema);
+#define XFS_BUF_FINISH_IOWAIT(bp) complete(&bp->b_iowait);
#define XFS_BUF_SET_TARGET(bp, target) ((bp)->b_target = (target))
#define XFS_BUF_TARGET(bp) ((bp)->b_target)
@@ -429,7 +437,7 @@ static inline void xfs_bdwrite(void *mp, xfs_buf_t *bp)
* Handling of buftargs.
*/
extern xfs_buftarg_t *xfs_alloc_buftarg(struct block_device *, int);
-extern void xfs_free_buftarg(xfs_buftarg_t *, int);
+extern void xfs_free_buftarg(xfs_buftarg_t *);
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);
diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c
index c672b3238b1..24fd598af84 100644
--- a/fs/xfs/linux-2.6/xfs_export.c
+++ b/fs/xfs/linux-2.6/xfs_export.c
@@ -139,7 +139,7 @@ xfs_nfs_get_inode(
}
xfs_iunlock(ip, XFS_ILOCK_SHARED);
- return ip->i_vnode;
+ return VFS_I(ip);
}
STATIC struct dentry *
@@ -167,7 +167,7 @@ xfs_fs_fh_to_dentry(struct super_block *sb, struct fid *fid,
if (!inode)
return NULL;
if (IS_ERR(inode))
- return ERR_PTR(PTR_ERR(inode));
+ return ERR_CAST(inode);
result = d_alloc_anon(inode);
if (!result) {
iput(inode);
@@ -198,7 +198,7 @@ xfs_fs_fh_to_parent(struct super_block *sb, struct fid *fid,
if (!inode)
return NULL;
if (IS_ERR(inode))
- return ERR_PTR(PTR_ERR(inode));
+ return ERR_CAST(inode);
result = d_alloc_anon(inode);
if (!result) {
iput(inode);
@@ -215,13 +215,13 @@ xfs_fs_get_parent(
struct xfs_inode *cip;
struct dentry *parent;
- error = xfs_lookup(XFS_I(child->d_inode), &xfs_name_dotdot, &cip);
+ error = xfs_lookup(XFS_I(child->d_inode), &xfs_name_dotdot, &cip, NULL);
if (unlikely(error))
return ERR_PTR(-error);
- parent = d_alloc_anon(cip->i_vnode);
+ parent = d_alloc_anon(VFS_I(cip));
if (unlikely(!parent)) {
- iput(cip->i_vnode);
+ iput(VFS_I(cip));
return ERR_PTR(-ENOMEM);
}
return parent;
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index 5f60363b934..5311c1acdd4 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -475,6 +475,7 @@ const struct file_operations xfs_invis_file_operations = {
const struct file_operations xfs_dir_file_operations = {
.read = generic_read_dir,
.readdir = xfs_file_readdir,
+ .llseek = generic_file_llseek,
.unlocked_ioctl = xfs_file_ioctl,
#ifdef CONFIG_COMPAT
.compat_ioctl = xfs_file_compat_ioctl,
diff --git a/fs/xfs/linux-2.6/xfs_fs_subr.c b/fs/xfs/linux-2.6/xfs_fs_subr.c
index 1eefe61f0e1..36caa6d957d 100644
--- a/fs/xfs/linux-2.6/xfs_fs_subr.c
+++ b/fs/xfs/linux-2.6/xfs_fs_subr.c
@@ -31,7 +31,7 @@ xfs_tosspages(
xfs_off_t last,
int fiopt)
{
- struct address_space *mapping = ip->i_vnode->i_mapping;
+ struct address_space *mapping = VFS_I(ip)->i_mapping;
if (mapping->nrpages)
truncate_inode_pages(mapping, first);
@@ -44,7 +44,7 @@ xfs_flushinval_pages(
xfs_off_t last,
int fiopt)
{
- struct address_space *mapping = ip->i_vnode->i_mapping;
+ struct address_space *mapping = VFS_I(ip)->i_mapping;
int ret = 0;
if (mapping->nrpages) {
@@ -64,7 +64,7 @@ xfs_flush_pages(
uint64_t flags,
int fiopt)
{
- struct address_space *mapping = ip->i_vnode->i_mapping;
+ struct address_space *mapping = VFS_I(ip)->i_mapping;
int ret = 0;
int ret2;
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c
index a42ba9d7115..48799ba7e3e 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl.c
@@ -48,6 +48,8 @@
#include "xfs_dfrag.h"
#include "xfs_fsops.h"
#include "xfs_vnodeops.h"
+#include "xfs_quota.h"
+#include "xfs_inode_item.h"
#include <linux/capability.h>
#include <linux/dcache.h>
@@ -84,17 +86,15 @@ xfs_find_handle(
switch (cmd) {
case XFS_IOC_PATH_TO_FSHANDLE:
case XFS_IOC_PATH_TO_HANDLE: {
- struct nameidata nd;
- int error;
-
- error = user_path_walk_link((const char __user *)hreq.path, &nd);
+ struct path path;
+ int error = user_lpath((const char __user *)hreq.path, &path);
if (error)
return error;
- ASSERT(nd.path.dentry);
- ASSERT(nd.path.dentry->d_inode);
- inode = igrab(nd.path.dentry->d_inode);
- path_put(&nd.path);
+ ASSERT(path.dentry);
+ ASSERT(path.dentry->d_inode);
+ inode = igrab(path.dentry->d_inode);
+ path_put(&path);
break;
}
@@ -245,7 +245,7 @@ xfs_vget_fsop_handlereq(
xfs_iunlock(ip, XFS_ILOCK_SHARED);
- *inode = XFS_ITOV(ip);
+ *inode = VFS_I(ip);
return 0;
}
@@ -470,6 +470,12 @@ xfs_attrlist_by_handle(
if (al_hreq.buflen > XATTR_LIST_MAX)
return -XFS_ERROR(EINVAL);
+ /*
+ * Reject flags, only allow namespaces.
+ */
+ if (al_hreq.flags & ~(ATTR_ROOT | ATTR_SECURE))
+ return -XFS_ERROR(EINVAL);
+
error = xfs_vget_fsop_handlereq(mp, parinode, &al_hreq.hreq, &inode);
if (error)
goto out;
@@ -589,7 +595,7 @@ xfs_attrmulti_by_handle(
goto out;
error = E2BIG;
- size = am_hreq.opcount * sizeof(attr_multiop_t);
+ size = am_hreq.opcount * sizeof(xfs_attr_multiop_t);
if (!size || size > 16 * PAGE_SIZE)
goto out_vn_rele;
@@ -682,9 +688,9 @@ xfs_ioc_space(
return -XFS_ERROR(EFAULT);
if (filp->f_flags & (O_NDELAY|O_NONBLOCK))
- attr_flags |= ATTR_NONBLOCK;
+ attr_flags |= XFS_ATTR_NONBLOCK;
if (ioflags & IO_INVIS)
- attr_flags |= ATTR_DMI;
+ attr_flags |= XFS_ATTR_DMI;
error = xfs_change_file_space(ip, cmd, &bf, filp->f_pos,
NULL, attr_flags);
@@ -875,6 +881,322 @@ xfs_ioc_fsgetxattr(
return 0;
}
+STATIC void
+xfs_set_diflags(
+ struct xfs_inode *ip,
+ unsigned int xflags)
+{
+ unsigned int di_flags;
+
+ /* can't set PREALLOC this way, just preserve it */
+ di_flags = (ip->i_d.di_flags & XFS_DIFLAG_PREALLOC);
+ if (xflags & XFS_XFLAG_IMMUTABLE)
+ di_flags |= XFS_DIFLAG_IMMUTABLE;
+ if (xflags & XFS_XFLAG_APPEND)
+ di_flags |= XFS_DIFLAG_APPEND;
+ if (xflags & XFS_XFLAG_SYNC)
+ di_flags |= XFS_DIFLAG_SYNC;
+ if (xflags & XFS_XFLAG_NOATIME)
+ di_flags |= XFS_DIFLAG_NOATIME;
+ if (xflags & XFS_XFLAG_NODUMP)
+ di_flags |= XFS_DIFLAG_NODUMP;
+ if (xflags & XFS_XFLAG_PROJINHERIT)
+ di_flags |= XFS_DIFLAG_PROJINHERIT;
+ if (xflags & XFS_XFLAG_NODEFRAG)
+ di_flags |= XFS_DIFLAG_NODEFRAG;
+ if (xflags & XFS_XFLAG_FILESTREAM)
+ di_flags |= XFS_DIFLAG_FILESTREAM;
+ if ((ip->i_d.di_mode & S_IFMT) == S_IFDIR) {
+ if (xflags & XFS_XFLAG_RTINHERIT)
+ di_flags |= XFS_DIFLAG_RTINHERIT;
+ if (xflags & XFS_XFLAG_NOSYMLINKS)
+ di_flags |= XFS_DIFLAG_NOSYMLINKS;
+ if (xflags & XFS_XFLAG_EXTSZINHERIT)
+ di_flags |= XFS_DIFLAG_EXTSZINHERIT;
+ } else if ((ip->i_d.di_mode & S_IFMT) == S_IFREG) {
+ if (xflags & XFS_XFLAG_REALTIME)
+ di_flags |= XFS_DIFLAG_REALTIME;
+ if (xflags & XFS_XFLAG_EXTSIZE)
+ di_flags |= XFS_DIFLAG_EXTSIZE;
+ }
+
+ ip->i_d.di_flags = di_flags;
+}
+
+STATIC void
+xfs_diflags_to_linux(
+ struct xfs_inode *ip)
+{
+ struct inode *inode = VFS_I(ip);
+ unsigned int xflags = xfs_ip2xflags(ip);
+
+ if (xflags & XFS_XFLAG_IMMUTABLE)
+ inode->i_flags |= S_IMMUTABLE;
+ else
+ inode->i_flags &= ~S_IMMUTABLE;
+ if (xflags & XFS_XFLAG_APPEND)
+ inode->i_flags |= S_APPEND;
+ else
+ inode->i_flags &= ~S_APPEND;
+ if (xflags & XFS_XFLAG_SYNC)
+ inode->i_flags |= S_SYNC;
+ else
+ inode->i_flags &= ~S_SYNC;
+ if (xflags & XFS_XFLAG_NOATIME)
+ inode->i_flags |= S_NOATIME;
+ else
+ inode->i_flags &= ~S_NOATIME;
+}
+
+#define FSX_PROJID 1
+#define FSX_EXTSIZE 2
+#define FSX_XFLAGS 4
+#define FSX_NONBLOCK 8
+
+STATIC int
+xfs_ioctl_setattr(
+ xfs_inode_t *ip,
+ struct fsxattr *fa,
+ int mask)
+{
+ struct xfs_mount *mp = ip->i_mount;
+ struct xfs_trans *tp;
+ unsigned int lock_flags = 0;
+ struct xfs_dquot *udqp = NULL, *gdqp = NULL;
+ struct xfs_dquot *olddquot = NULL;
+ int code;
+
+ xfs_itrace_entry(ip);
+
+ if (mp->m_flags & XFS_MOUNT_RDONLY)
+ return XFS_ERROR(EROFS);
+ if (XFS_FORCED_SHUTDOWN(mp))
+ return XFS_ERROR(EIO);
+
+ /*
+ * If disk quotas is on, we make sure that the dquots do exist on disk,
+ * before we start any other transactions. Trying to do this later
+ * is messy. We don't care to take a readlock to look at the ids
+ * in inode here, because we can't hold it across the trans_reserve.
+ * If the IDs do change before we take the ilock, we're covered
+ * because the i_*dquot fields will get updated anyway.
+ */
+ if (XFS_IS_QUOTA_ON(mp) && (mask & FSX_PROJID)) {
+ code = XFS_QM_DQVOPALLOC(mp, ip, ip->i_d.di_uid,
+ ip->i_d.di_gid, fa->fsx_projid,
+ XFS_QMOPT_PQUOTA, &udqp, &gdqp);
+ if (code)
+ return code;
+ }
+
+ /*
+ * For the other attributes, we acquire the inode lock and
+ * first do an error checking pass.
+ */
+ tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE);
+ code = xfs_trans_reserve(tp, 0, XFS_ICHANGE_LOG_RES(mp), 0, 0, 0);
+ if (code)
+ goto error_return;
+
+ lock_flags = XFS_ILOCK_EXCL;
+ xfs_ilock(ip, lock_flags);
+
+ /*
+ * CAP_FOWNER overrides the following restrictions:
+ *
+ * The user ID of the calling process must be equal
+ * to the file owner ID, except in cases where the
+ * CAP_FSETID capability is applicable.
+ */
+ if (current->fsuid != ip->i_d.di_uid && !capable(CAP_FOWNER)) {
+ code = XFS_ERROR(EPERM);
+ goto error_return;
+ }
+
+ /*
+ * Do a quota reservation only if projid is actually going to change.
+ */
+ if (mask & FSX_PROJID) {
+ if (XFS_IS_PQUOTA_ON(mp) &&
+ ip->i_d.di_projid != fa->fsx_projid) {
+ ASSERT(tp);
+ code = XFS_QM_DQVOPCHOWNRESV(mp, tp, ip, udqp, gdqp,
+ capable(CAP_FOWNER) ?
+ XFS_QMOPT_FORCE_RES : 0);
+ if (code) /* out of quota */
+ goto error_return;
+ }
+ }
+
+ if (mask & FSX_EXTSIZE) {
+ /*
+ * Can't change extent size if any extents are allocated.
+ */
+ if (ip->i_d.di_nextents &&
+ ((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) !=
+ fa->fsx_extsize)) {
+ code = XFS_ERROR(EINVAL); /* EFBIG? */
+ goto error_return;
+ }
+
+ /*
+ * Extent size must be a multiple of the appropriate block
+ * size, if set at all.
+ */
+ if (fa->fsx_extsize != 0) {
+ xfs_extlen_t size;
+
+ if (XFS_IS_REALTIME_INODE(ip) ||
+ ((mask & FSX_XFLAGS) &&
+ (fa->fsx_xflags & XFS_XFLAG_REALTIME))) {
+ size = mp->m_sb.sb_rextsize <<
+ mp->m_sb.sb_blocklog;
+ } else {
+ size = mp->m_sb.sb_blocksize;
+ }
+
+ if (fa->fsx_extsize % size) {
+ code = XFS_ERROR(EINVAL);
+ goto error_return;
+ }
+ }
+ }
+
+
+ if (mask & FSX_XFLAGS) {
+ /*
+ * Can't change realtime flag if any extents are allocated.
+ */
+ if ((ip->i_d.di_nextents || ip->i_delayed_blks) &&
+ (XFS_IS_REALTIME_INODE(ip)) !=
+ (fa->fsx_xflags & XFS_XFLAG_REALTIME)) {
+ code = XFS_ERROR(EINVAL); /* EFBIG? */
+ goto error_return;
+ }
+
+ /*
+ * If realtime flag is set then must have realtime data.
+ */
+ if ((fa->fsx_xflags & XFS_XFLAG_REALTIME)) {
+ if ((mp->m_sb.sb_rblocks == 0) ||
+ (mp->m_sb.sb_rextsize == 0) ||
+ (ip->i_d.di_extsize % mp->m_sb.sb_rextsize)) {
+ code = XFS_ERROR(EINVAL);
+ goto error_return;
+ }
+ }
+
+ /*
+ * Can't modify an immutable/append-only file unless
+ * we have appropriate permission.
+ */
+ if ((ip->i_d.di_flags &
+ (XFS_DIFLAG_IMMUTABLE|XFS_DIFLAG_APPEND) ||
+ (fa->fsx_xflags &
+ (XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) &&
+ !capable(CAP_LINUX_IMMUTABLE)) {
+ code = XFS_ERROR(EPERM);
+ goto error_return;
+ }
+ }
+
+ xfs_trans_ijoin(tp, ip, lock_flags);
+ xfs_trans_ihold(tp, ip);
+
+ /*
+ * Change file ownership. Must be the owner or privileged.
+ * If the system was configured with the "restricted_chown"
+ * option, the owner is not permitted to give away the file,
+ * and can change the group id only to a group of which he
+ * or she is a member.
+ */
+ if (mask & FSX_PROJID) {
+ /*
+ * CAP_FSETID overrides the following restrictions:
+ *
+ * The set-user-ID and set-group-ID bits of a file will be
+ * cleared upon successful return from chown()
+ */
+ if ((ip->i_d.di_mode & (S_ISUID|S_ISGID)) &&
+ !capable(CAP_FSETID))
+ ip->i_d.di_mode &= ~(S_ISUID|S_ISGID);
+
+ /*
+ * Change the ownerships and register quota modifications
+ * in the transaction.
+ */
+ if (ip->i_d.di_projid != fa->fsx_projid) {
+ if (XFS_IS_PQUOTA_ON(mp)) {
+ olddquot = XFS_QM_DQVOPCHOWN(mp, tp, ip,
+ &ip->i_gdquot, gdqp);
+ }
+ ip->i_d.di_projid = fa->fsx_projid;
+
+ /*
+ * We may have to rev the inode as well as
+ * the superblock version number since projids didn't
+ * exist before DINODE_VERSION_2 and SB_VERSION_NLINK.
+ */
+ if (ip->i_d.di_version == XFS_DINODE_VERSION_1)
+ xfs_bump_ino_vers2(tp, ip);
+ }
+
+ }
+
+ if (mask & FSX_EXTSIZE)
+ ip->i_d.di_extsize = fa->fsx_extsize >> mp->m_sb.sb_blocklog;
+ if (mask & FSX_XFLAGS) {
+ xfs_set_diflags(ip, fa->fsx_xflags);
+ xfs_diflags_to_linux(ip);
+ }
+
+ xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+ xfs_ichgtime(ip, XFS_ICHGTIME_CHG);
+
+ XFS_STATS_INC(xs_ig_attrchg);
+
+ /*
+ * If this is a synchronous mount, make sure that the
+ * transaction goes to disk before returning to the user.
+ * This is slightly sub-optimal in that truncates require
+ * two sync transactions instead of one for wsync filesystems.
+ * One for the truncate and one for the timestamps since we
+ * don't want to change the timestamps unless we're sure the
+ * truncate worked. Truncates are less than 1% of the laddis
+ * mix so this probably isn't worth the trouble to optimize.
+ */
+ if (mp->m_flags & XFS_MOUNT_WSYNC)
+ xfs_trans_set_sync(tp);
+ code = xfs_trans_commit(tp, 0);
+ xfs_iunlock(ip, lock_flags);
+
+ /*
+ * Release any dquot(s) the inode had kept before chown.
+ */
+ XFS_QM_DQRELE(mp, olddquot);
+ XFS_QM_DQRELE(mp, udqp);
+ XFS_QM_DQRELE(mp, gdqp);
+
+ if (code)
+ return code;
+
+ if (DM_EVENT_ENABLED(ip, DM_EVENT_ATTRIBUTE)) {
+ XFS_SEND_NAMESP(mp, DM_EVENT_ATTRIBUTE, ip, DM_RIGHT_NULL,
+ NULL, DM_RIGHT_NULL, NULL, NULL, 0, 0,
+ (mask & FSX_NONBLOCK) ? DM_FLAGS_NDELAY : 0);
+ }
+
+ return 0;
+
+ error_return:
+ XFS_QM_DQRELE(mp, udqp);
+ XFS_QM_DQRELE(mp, gdqp);
+ xfs_trans_cancel(tp, 0);
+ if (lock_flags)
+ xfs_iunlock(ip, lock_flags);
+ return code;
+}
+
STATIC int
xfs_ioc_fssetxattr(
xfs_inode_t *ip,
@@ -882,31 +1204,16 @@ xfs_ioc_fssetxattr(
void __user *arg)
{
struct fsxattr fa;
- struct bhv_vattr *vattr;
- int error;
- int attr_flags;
+ unsigned int mask;
if (copy_from_user(&fa, arg, sizeof(fa)))
return -EFAULT;
- vattr = kmalloc(sizeof(*vattr), GFP_KERNEL);
- if (unlikely(!vattr))
- return -ENOMEM;
-
- attr_flags = 0;
+ mask = FSX_XFLAGS | FSX_EXTSIZE | FSX_PROJID;
if (filp->f_flags & (O_NDELAY|O_NONBLOCK))
- attr_flags |= ATTR_NONBLOCK;
-
- vattr->va_mask = XFS_AT_XFLAGS | XFS_AT_EXTSIZE | XFS_AT_PROJID;
- vattr->va_xflags = fa.fsx_xflags;
- vattr->va_extsize = fa.fsx_extsize;
- vattr->va_projid = fa.fsx_projid;
+ mask |= FSX_NONBLOCK;
- error = -xfs_setattr(ip, vattr, attr_flags, NULL);
- if (!error)
- vn_revalidate(XFS_ITOV(ip)); /* update flags */
- kfree(vattr);
- return 0;
+ return -xfs_ioctl_setattr(ip, &fa, mask);
}
STATIC int
@@ -928,10 +1235,9 @@ xfs_ioc_setxflags(
struct file *filp,
void __user *arg)
{
- struct bhv_vattr *vattr;
+ struct fsxattr fa;
unsigned int flags;
- int attr_flags;
- int error;
+ unsigned int mask;
if (copy_from_user(&flags, arg, sizeof(flags)))
return -EFAULT;
@@ -941,22 +1247,12 @@ xfs_ioc_setxflags(
FS_SYNC_FL))
return -EOPNOTSUPP;
- vattr = kmalloc(sizeof(*vattr), GFP_KERNEL);
- if (unlikely(!vattr))
- return -ENOMEM;
-
- attr_flags = 0;
+ mask = FSX_XFLAGS;
if (filp->f_flags & (O_NDELAY|O_NONBLOCK))
- attr_flags |= ATTR_NONBLOCK;
-
- vattr->va_mask = XFS_AT_XFLAGS;
- vattr->va_xflags = xfs_merge_ioc_xflags(flags, xfs_ip2xflags(ip));
+ mask |= FSX_NONBLOCK;
+ fa.fsx_xflags = xfs_merge_ioc_xflags(flags, xfs_ip2xflags(ip));
- error = -xfs_setattr(ip, vattr, attr_flags, NULL);
- if (likely(!error))
- vn_revalidate(XFS_ITOV(ip)); /* update flags */
- kfree(vattr);
- return error;
+ return -xfs_ioctl_setattr(ip, &fa, mask);
}
STATIC int
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index 2bf287ef548..095d271f343 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -62,7 +62,7 @@ void
xfs_synchronize_atime(
xfs_inode_t *ip)
{
- struct inode *inode = ip->i_vnode;
+ struct inode *inode = VFS_I(ip);
if (inode) {
ip->i_d.di_atime.t_sec = (__int32_t)inode->i_atime.tv_sec;
@@ -79,7 +79,7 @@ void
xfs_mark_inode_dirty_sync(
xfs_inode_t *ip)
{
- struct inode *inode = ip->i_vnode;
+ struct inode *inode = VFS_I(ip);
if (inode)
mark_inode_dirty_sync(inode);
@@ -89,36 +89,31 @@ xfs_mark_inode_dirty_sync(
* Change the requested timestamp in the given inode.
* We don't lock across timestamp updates, and we don't log them but
* we do record the fact that there is dirty information in core.
- *
- * NOTE -- callers MUST combine XFS_ICHGTIME_MOD or XFS_ICHGTIME_CHG
- * with XFS_ICHGTIME_ACC to be sure that access time
- * update will take. Calling first with XFS_ICHGTIME_ACC
- * and then XFS_ICHGTIME_MOD may fail to modify the access
- * timestamp if the filesystem is mounted noacctm.
*/
void
xfs_ichgtime(
xfs_inode_t *ip,
int flags)
{
- struct inode *inode = vn_to_inode(XFS_ITOV(ip));
+ struct inode *inode = VFS_I(ip);
timespec_t tv;
+ int sync_it = 0;
+
+ tv = current_fs_time(inode->i_sb);
- nanotime(&tv);
- if (flags & XFS_ICHGTIME_MOD) {
+ if ((flags & XFS_ICHGTIME_MOD) &&
+ !timespec_equal(&inode->i_mtime, &tv)) {
inode->i_mtime = tv;
ip->i_d.di_mtime.t_sec = (__int32_t)tv.tv_sec;
ip->i_d.di_mtime.t_nsec = (__int32_t)tv.tv_nsec;
+ sync_it = 1;
}
- if (flags & XFS_ICHGTIME_ACC) {
- inode->i_atime = tv;
- ip->i_d.di_atime.t_sec = (__int32_t)tv.tv_sec;
- ip->i_d.di_atime.t_nsec = (__int32_t)tv.tv_nsec;
- }
- if (flags & XFS_ICHGTIME_CHG) {
+ if ((flags & XFS_ICHGTIME_CHG) &&
+ !timespec_equal(&inode->i_ctime, &tv)) {
inode->i_ctime = tv;
ip->i_d.di_ctime.t_sec = (__int32_t)tv.tv_sec;
ip->i_d.di_ctime.t_nsec = (__int32_t)tv.tv_nsec;
+ sync_it = 1;
}
/*
@@ -130,72 +125,11 @@ xfs_ichgtime(
* ensure that the compiler does not reorder the update
* of i_update_core above the timestamp updates above.
*/
- SYNCHRONIZE();
- ip->i_update_core = 1;
- if (!(inode->i_state & I_NEW))
+ if (sync_it) {
+ SYNCHRONIZE();
+ ip->i_update_core = 1;
mark_inode_dirty_sync(inode);
-}
-
-/*
- * Variant on the above which avoids querying the system clock
- * in situations where we know the Linux inode timestamps have
- * just been updated (and so we can update our inode cheaply).
- */
-void
-xfs_ichgtime_fast(
- xfs_inode_t *ip,
- struct inode *inode,
- int flags)
-{
- timespec_t *tvp;
-
- /*
- * Atime updates for read() & friends are handled lazily now, and
- * explicit updates must go through xfs_ichgtime()
- */
- ASSERT((flags & XFS_ICHGTIME_ACC) == 0);
-
- if (flags & XFS_ICHGTIME_MOD) {
- tvp = &inode->i_mtime;
- ip->i_d.di_mtime.t_sec = (__int32_t)tvp->tv_sec;
- ip->i_d.di_mtime.t_nsec = (__int32_t)tvp->tv_nsec;
- }
- if (flags & XFS_ICHGTIME_CHG) {
- tvp = &inode->i_ctime;
- ip->i_d.di_ctime.t_sec = (__int32_t)tvp->tv_sec;
- ip->i_d.di_ctime.t_nsec = (__int32_t)tvp->tv_nsec;
}
-
- /*
- * We update the i_update_core field _after_ changing
- * the timestamps in order to coordinate properly with
- * xfs_iflush() so that we don't lose timestamp updates.
- * This keeps us from having to hold the inode lock
- * while doing this. We use the SYNCHRONIZE macro to
- * ensure that the compiler does not reorder the update
- * of i_update_core above the timestamp updates above.
- */
- SYNCHRONIZE();
- ip->i_update_core = 1;
- if (!(inode->i_state & I_NEW))
- mark_inode_dirty_sync(inode);
-}
-
-
-/*
- * Pull the link count and size up from the xfs inode to the linux inode
- */
-STATIC void
-xfs_validate_fields(
- struct inode *inode)
-{
- struct xfs_inode *ip = XFS_I(inode);
- loff_t size;
-
- /* we're under i_sem so i_size can't change under us */
- size = XFS_ISIZE(ip);
- if (i_size_read(inode) != size)
- i_size_write(inode, size);
}
/*
@@ -245,8 +179,7 @@ STATIC void
xfs_cleanup_inode(
struct inode *dir,
struct inode *inode,
- struct dentry *dentry,
- int mode)
+ struct dentry *dentry)
{
struct xfs_name teardown;
@@ -257,10 +190,7 @@ xfs_cleanup_inode(
*/
xfs_dentry_to_name(&teardown, dentry);
- if (S_ISDIR(mode))
- xfs_rmdir(XFS_I(dir), &teardown, XFS_I(inode));
- else
- xfs_remove(XFS_I(dir), &teardown, XFS_I(inode));
+ xfs_remove(XFS_I(dir), &teardown, XFS_I(inode));
iput(inode);
}
@@ -275,7 +205,7 @@ xfs_vn_mknod(
struct xfs_inode *ip = NULL;
xfs_acl_t *default_acl = NULL;
struct xfs_name name;
- attrexists_t test_default_acl = _ACL_DEFAULT_EXISTS;
+ int (*test_default_acl)(struct inode *) = _ACL_DEFAULT_EXISTS;
int error;
/*
@@ -320,7 +250,7 @@ xfs_vn_mknod(
if (unlikely(error))
goto out_free_acl;
- inode = ip->i_vnode;
+ inode = VFS_I(ip);
error = xfs_init_security(inode, dir);
if (unlikely(error))
@@ -335,14 +265,11 @@ xfs_vn_mknod(
}
- if (S_ISDIR(mode))
- xfs_validate_fields(inode);
d_instantiate(dentry, inode);
- xfs_validate_fields(dir);
return -error;
out_cleanup_inode:
- xfs_cleanup_inode(dir, inode, dentry, mode);
+ xfs_cleanup_inode(dir, inode, dentry);
out_free_acl:
if (default_acl)
_ACL_FREE(default_acl);
@@ -382,7 +309,7 @@ xfs_vn_lookup(
return ERR_PTR(-ENAMETOOLONG);
xfs_dentry_to_name(&name, dentry);
- error = xfs_lookup(XFS_I(dir), &name, &cip);
+ error = xfs_lookup(XFS_I(dir), &name, &cip, NULL);
if (unlikely(error)) {
if (unlikely(error != ENOENT))
return ERR_PTR(-error);
@@ -390,7 +317,47 @@ xfs_vn_lookup(
return NULL;
}
- return d_splice_alias(cip->i_vnode, dentry);
+ return d_splice_alias(VFS_I(cip), dentry);
+}
+
+STATIC struct dentry *
+xfs_vn_ci_lookup(
+ struct inode *dir,
+ struct dentry *dentry,
+ struct nameidata *nd)
+{
+ struct xfs_inode *ip;
+ struct xfs_name xname;
+ struct xfs_name ci_name;
+ struct qstr dname;
+ int error;
+
+ if (dentry->d_name.len >= MAXNAMELEN)
+ return ERR_PTR(-ENAMETOOLONG);
+
+ xfs_dentry_to_name(&xname, dentry);
+ error = xfs_lookup(XFS_I(dir), &xname, &ip, &ci_name);
+ if (unlikely(error)) {
+ if (unlikely(error != ENOENT))
+ return ERR_PTR(-error);
+ /*
+ * call d_add(dentry, NULL) here when d_drop_negative_children
+ * is called in xfs_vn_mknod (ie. allow negative dentries
+ * with CI filesystems).
+ */
+ return NULL;
+ }
+
+ /* if exact match, just splice and exit */
+ if (!ci_name.name)
+ return d_splice_alias(VFS_I(ip), dentry);
+
+ /* else case-insensitive match... */
+ dname.name = ci_name.name;
+ dname.len = ci_name.len;
+ dentry = d_add_ci(dentry, VFS_I(ip), &dname);
+ kmem_free(ci_name.name);
+ return dentry;
}
STATIC int
@@ -414,7 +381,6 @@ xfs_vn_link(
}
xfs_iflags_set(XFS_I(dir), XFS_IMODIFIED);
- xfs_validate_fields(inode);
d_instantiate(dentry, inode);
return 0;
}
@@ -424,19 +390,23 @@ xfs_vn_unlink(
struct inode *dir,
struct dentry *dentry)
{
- struct inode *inode;
struct xfs_name name;
int error;
- inode = dentry->d_inode;
xfs_dentry_to_name(&name, dentry);
- error = xfs_remove(XFS_I(dir), &name, XFS_I(inode));
- if (likely(!error)) {
- xfs_validate_fields(dir); /* size needs update */
- xfs_validate_fields(inode);
- }
- return -error;
+ error = -xfs_remove(XFS_I(dir), &name, XFS_I(dentry->d_inode));
+ if (error)
+ return error;
+
+ /*
+ * With unlink, the VFS makes the dentry "negative": no inode,
+ * but still hashed. This is incompatible with case-insensitive
+ * mode, so invalidate (unhash) the dentry in CI-mode.
+ */
+ if (xfs_sb_version_hasasciici(&XFS_M(dir->i_sb)->m_sb))
+ d_invalidate(dentry);
+ return 0;
}
STATIC int
@@ -459,43 +429,22 @@ xfs_vn_symlink(
if (unlikely(error))
goto out;
- inode = cip->i_vnode;
+ inode = VFS_I(cip);
error = xfs_init_security(inode, dir);
if (unlikely(error))
goto out_cleanup_inode;
d_instantiate(dentry, inode);
- xfs_validate_fields(dir);
- xfs_validate_fields(inode);
return 0;
out_cleanup_inode:
- xfs_cleanup_inode(dir, inode, dentry, 0);
+ xfs_cleanup_inode(dir, inode, dentry);
out:
return -error;
}
STATIC int
-xfs_vn_rmdir(
- struct inode *dir,
- struct dentry *dentry)
-{
- struct inode *inode = dentry->d_inode;
- struct xfs_name name;
- int error;
-
- xfs_dentry_to_name(&name, dentry);
-
- error = xfs_rmdir(XFS_I(dir), &name, XFS_I(inode));
- if (likely(!error)) {
- xfs_validate_fields(inode);
- xfs_validate_fields(dir);
- }
- return -error;
-}
-
-STATIC int
xfs_vn_rename(
struct inode *odir,
struct dentry *odentry,
@@ -505,22 +454,13 @@ xfs_vn_rename(
struct inode *new_inode = ndentry->d_inode;
struct xfs_name oname;
struct xfs_name nname;
- int error;
xfs_dentry_to_name(&oname, odentry);
xfs_dentry_to_name(&nname, ndentry);
- error = xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode),
+ return -xfs_rename(XFS_I(odir), &oname, XFS_I(odentry->d_inode),
XFS_I(ndir), &nname, new_inode ?
XFS_I(new_inode) : NULL);
- if (likely(!error)) {
- if (new_inode)
- xfs_validate_fields(new_inode);
- xfs_validate_fields(odir);
- if (ndir != odir)
- xfs_validate_fields(ndir);
- }
- return -error;
}
/*
@@ -589,8 +529,7 @@ xfs_check_acl(
STATIC int
xfs_vn_permission(
struct inode *inode,
- int mask,
- struct nameidata *nd)
+ int mask)
{
return generic_permission(inode, mask, xfs_check_acl);
}
@@ -660,57 +599,9 @@ xfs_vn_getattr(
STATIC int
xfs_vn_setattr(
struct dentry *dentry,
- struct iattr *attr)
+ struct iattr *iattr)
{
- struct inode *inode = dentry->d_inode;
- unsigned int ia_valid = attr->ia_valid;
- bhv_vattr_t vattr = { 0 };
- int flags = 0;
- int error;
-
- if (ia_valid & ATTR_UID) {
- vattr.va_mask |= XFS_AT_UID;
- vattr.va_uid = attr->ia_uid;
- }
- if (ia_valid & ATTR_GID) {
- vattr.va_mask |= XFS_AT_GID;
- vattr.va_gid = attr->ia_gid;
- }
- if (ia_valid & ATTR_SIZE) {
- vattr.va_mask |= XFS_AT_SIZE;
- vattr.va_size = attr->ia_size;
- }
- if (ia_valid & ATTR_ATIME) {
- vattr.va_mask |= XFS_AT_ATIME;
- vattr.va_atime = attr->ia_atime;
- inode->i_atime = attr->ia_atime;
- }
- if (ia_valid & ATTR_MTIME) {
- vattr.va_mask |= XFS_AT_MTIME;
- vattr.va_mtime = attr->ia_mtime;
- }
- if (ia_valid & ATTR_CTIME) {
- vattr.va_mask |= XFS_AT_CTIME;
- vattr.va_ctime = attr->ia_ctime;
- }
- if (ia_valid & ATTR_MODE) {
- vattr.va_mask |= XFS_AT_MODE;
- vattr.va_mode = attr->ia_mode;
- if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID))
- inode->i_mode &= ~S_ISGID;
- }
-
- if (ia_valid & (ATTR_MTIME_SET | ATTR_ATIME_SET))
- flags |= ATTR_UTIME;
-#ifdef ATTR_NO_BLOCK
- if ((ia_valid & ATTR_NO_BLOCK))
- flags |= ATTR_NONBLOCK;
-#endif
-
- error = xfs_setattr(XFS_I(inode), &vattr, flags, NULL);
- if (likely(!error))
- vn_revalidate(vn_from_inode(inode));
- return -error;
+ return -xfs_setattr(XFS_I(dentry->d_inode), iattr, 0, NULL);
}
/*
@@ -728,109 +619,6 @@ xfs_vn_truncate(
WARN_ON(error);
}
-STATIC int
-xfs_vn_setxattr(
- struct dentry *dentry,
- const char *name,
- const void *data,
- size_t size,
- int flags)
-{
- bhv_vnode_t *vp = vn_from_inode(dentry->d_inode);
- char *attr = (char *)name;
- attrnames_t *namesp;
- int xflags = 0;
- int error;
-
- namesp = attr_lookup_namespace(attr, attr_namespaces, ATTR_NAMECOUNT);
- if (!namesp)
- return -EOPNOTSUPP;
- attr += namesp->attr_namelen;
- error = namesp->attr_capable(vp, NULL);
- if (error)
- return error;
-
- /* Convert Linux syscall to XFS internal ATTR flags */
- if (flags & XATTR_CREATE)
- xflags |= ATTR_CREATE;
- if (flags & XATTR_REPLACE)
- xflags |= ATTR_REPLACE;
- xflags |= namesp->attr_flag;
- return namesp->attr_set(vp, attr, (void *)data, size, xflags);
-}
-
-STATIC ssize_t
-xfs_vn_getxattr(
- struct dentry *dentry,
- const char *name,
- void *data,
- size_t size)
-{
- bhv_vnode_t *vp = vn_from_inode(dentry->d_inode);
- char *attr = (char *)name;
- attrnames_t *namesp;
- int xflags = 0;
- ssize_t error;
-
- namesp = attr_lookup_namespace(attr, attr_namespaces, ATTR_NAMECOUNT);
- if (!namesp)
- return -EOPNOTSUPP;
- attr += namesp->attr_namelen;
- error = namesp->attr_capable(vp, NULL);
- if (error)
- return error;
-
- /* Convert Linux syscall to XFS internal ATTR flags */
- if (!size) {
- xflags |= ATTR_KERNOVAL;
- data = NULL;
- }
- xflags |= namesp->attr_flag;
- return namesp->attr_get(vp, attr, (void *)data, size, xflags);
-}
-
-STATIC ssize_t
-xfs_vn_listxattr(
- struct dentry *dentry,
- char *data,
- size_t size)
-{
- bhv_vnode_t *vp = vn_from_inode(dentry->d_inode);
- int error, xflags = ATTR_KERNAMELS;
- ssize_t result;
-
- if (!size)
- xflags |= ATTR_KERNOVAL;
- xflags |= capable(CAP_SYS_ADMIN) ? ATTR_KERNFULLS : ATTR_KERNORMALS;
-
- error = attr_generic_list(vp, data, size, xflags, &result);
- if (error < 0)
- return error;
- return result;
-}
-
-STATIC int
-xfs_vn_removexattr(
- struct dentry *dentry,
- const char *name)
-{
- bhv_vnode_t *vp = vn_from_inode(dentry->d_inode);
- char *attr = (char *)name;
- attrnames_t *namesp;
- int xflags = 0;
- int error;
-
- namesp = attr_lookup_namespace(attr, attr_namespaces, ATTR_NAMECOUNT);
- if (!namesp)
- return -EOPNOTSUPP;
- attr += namesp->attr_namelen;
- error = namesp->attr_capable(vp, NULL);
- if (error)
- return error;
- xflags |= namesp->attr_flag;
- return namesp->attr_remove(vp, attr, xflags);
-}
-
STATIC long
xfs_vn_fallocate(
struct inode *inode,
@@ -854,18 +642,18 @@ xfs_vn_fallocate(
xfs_ilock(ip, XFS_IOLOCK_EXCL);
error = xfs_change_file_space(ip, XFS_IOC_RESVSP, &bf,
- 0, NULL, ATTR_NOLOCK);
+ 0, NULL, XFS_ATTR_NOLOCK);
if (!error && !(mode & FALLOC_FL_KEEP_SIZE) &&
offset + len > i_size_read(inode))
new_size = offset + len;
/* Change file size if needed */
if (new_size) {
- bhv_vattr_t va;
+ struct iattr iattr;
- va.va_mask = XFS_AT_SIZE;
- va.va_size = new_size;
- error = xfs_setattr(ip, &va, ATTR_NOLOCK, NULL);
+ iattr.ia_valid = ATTR_SIZE;
+ iattr.ia_size = new_size;
+ error = xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK, NULL);
}
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
@@ -873,46 +661,172 @@ out_error:
return error;
}
-const struct inode_operations xfs_inode_operations = {
+static const struct inode_operations xfs_inode_operations = {
.permission = xfs_vn_permission,
.truncate = xfs_vn_truncate,
.getattr = xfs_vn_getattr,
.setattr = xfs_vn_setattr,
- .setxattr = xfs_vn_setxattr,
- .getxattr = xfs_vn_getxattr,
+ .setxattr = generic_setxattr,
+ .getxattr = generic_getxattr,
+ .removexattr = generic_removexattr,
.listxattr = xfs_vn_listxattr,
- .removexattr = xfs_vn_removexattr,
.fallocate = xfs_vn_fallocate,
};
-const struct inode_operations xfs_dir_inode_operations = {
+static const struct inode_operations xfs_dir_inode_operations = {
.create = xfs_vn_create,
.lookup = xfs_vn_lookup,
.link = xfs_vn_link,
.unlink = xfs_vn_unlink,
.symlink = xfs_vn_symlink,
.mkdir = xfs_vn_mkdir,
- .rmdir = xfs_vn_rmdir,
+ /*
+ * Yes, XFS uses the same method for rmdir and unlink.
+ *
+ * There are some subtile differences deeper in the code,
+ * but we use S_ISDIR to check for those.
+ */
+ .rmdir = xfs_vn_unlink,
.mknod = xfs_vn_mknod,
.rename = xfs_vn_rename,
.permission = xfs_vn_permission,
.getattr = xfs_vn_getattr,
.setattr = xfs_vn_setattr,
- .setxattr = xfs_vn_setxattr,
- .getxattr = xfs_vn_getxattr,
+ .setxattr = generic_setxattr,
+ .getxattr = generic_getxattr,
+ .removexattr = generic_removexattr,
.listxattr = xfs_vn_listxattr,
- .removexattr = xfs_vn_removexattr,
};
-const struct inode_operations xfs_symlink_inode_operations = {
+static const struct inode_operations xfs_dir_ci_inode_operations = {
+ .create = xfs_vn_create,
+ .lookup = xfs_vn_ci_lookup,
+ .link = xfs_vn_link,
+ .unlink = xfs_vn_unlink,
+ .symlink = xfs_vn_symlink,
+ .mkdir = xfs_vn_mkdir,
+ /*
+ * Yes, XFS uses the same method for rmdir and unlink.
+ *
+ * There are some subtile differences deeper in the code,
+ * but we use S_ISDIR to check for those.
+ */
+ .rmdir = xfs_vn_unlink,
+ .mknod = xfs_vn_mknod,
+ .rename = xfs_vn_rename,
+ .permission = xfs_vn_permission,
+ .getattr = xfs_vn_getattr,
+ .setattr = xfs_vn_setattr,
+ .setxattr = generic_setxattr,
+ .getxattr = generic_getxattr,
+ .removexattr = generic_removexattr,
+ .listxattr = xfs_vn_listxattr,
+};
+
+static const struct inode_operations xfs_symlink_inode_operations = {
.readlink = generic_readlink,
.follow_link = xfs_vn_follow_link,
.put_link = xfs_vn_put_link,
.permission = xfs_vn_permission,
.getattr = xfs_vn_getattr,
.setattr = xfs_vn_setattr,
- .setxattr = xfs_vn_setxattr,
- .getxattr = xfs_vn_getxattr,
+ .setxattr = generic_setxattr,
+ .getxattr = generic_getxattr,
+ .removexattr = generic_removexattr,
.listxattr = xfs_vn_listxattr,
- .removexattr = xfs_vn_removexattr,
};
+
+STATIC void
+xfs_diflags_to_iflags(
+ struct inode *inode,
+ struct xfs_inode *ip)
+{
+ if (ip->i_d.di_flags & XFS_DIFLAG_IMMUTABLE)
+ inode->i_flags |= S_IMMUTABLE;
+ else
+ inode->i_flags &= ~S_IMMUTABLE;
+ if (ip->i_d.di_flags & XFS_DIFLAG_APPEND)
+ inode->i_flags |= S_APPEND;
+ else
+ inode->i_flags &= ~S_APPEND;
+ if (ip->i_d.di_flags & XFS_DIFLAG_SYNC)
+ inode->i_flags |= S_SYNC;
+ else
+ inode->i_flags &= ~S_SYNC;
+ if (ip->i_d.di_flags & XFS_DIFLAG_NOATIME)
+ inode->i_flags |= S_NOATIME;
+ else
+ inode->i_flags &= ~S_NOATIME;
+}
+
+/*
+ * Initialize the Linux inode, set up the operation vectors and
+ * unlock the inode.
+ *
+ * When reading existing inodes from disk this is called directly
+ * from xfs_iget, when creating a new inode it is called from
+ * xfs_ialloc after setting up the inode.
+ */
+void
+xfs_setup_inode(
+ struct xfs_inode *ip)
+{
+ struct inode *inode = ip->i_vnode;
+
+ inode->i_mode = ip->i_d.di_mode;
+ inode->i_nlink = ip->i_d.di_nlink;
+ inode->i_uid = ip->i_d.di_uid;
+ inode->i_gid = ip->i_d.di_gid;
+
+ switch (inode->i_mode & S_IFMT) {
+ case S_IFBLK:
+ case S_IFCHR:
+ inode->i_rdev =
+ MKDEV(sysv_major(ip->i_df.if_u2.if_rdev) & 0x1ff,
+ sysv_minor(ip->i_df.if_u2.if_rdev));
+ break;
+ default:
+ inode->i_rdev = 0;
+ break;
+ }
+
+ inode->i_generation = ip->i_d.di_gen;
+ i_size_write(inode, ip->i_d.di_size);
+ inode->i_atime.tv_sec = ip->i_d.di_atime.t_sec;
+ inode->i_atime.tv_nsec = ip->i_d.di_atime.t_nsec;
+ inode->i_mtime.tv_sec = ip->i_d.di_mtime.t_sec;
+ inode->i_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec;
+ inode->i_ctime.tv_sec = ip->i_d.di_ctime.t_sec;
+ inode->i_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec;
+ xfs_diflags_to_iflags(inode, ip);
+ xfs_iflags_clear(ip, XFS_IMODIFIED);
+
+ switch (inode->i_mode & S_IFMT) {
+ case S_IFREG:
+ inode->i_op = &xfs_inode_operations;
+ inode->i_fop = &xfs_file_operations;
+ inode->i_mapping->a_ops = &xfs_address_space_operations;
+ break;
+ case S_IFDIR:
+ if (xfs_sb_version_hasasciici(&XFS_M(inode->i_sb)->m_sb))
+ inode->i_op = &xfs_dir_ci_inode_operations;
+ else
+ inode->i_op = &xfs_dir_inode_operations;
+ inode->i_fop = &xfs_dir_file_operations;
+ break;
+ case S_IFLNK:
+ inode->i_op = &xfs_symlink_inode_operations;
+ if (!(ip->i_df.if_flags & XFS_IFINLINE))
+ inode->i_mapping->a_ops = &xfs_address_space_operations;
+ break;
+ default:
+ inode->i_op = &xfs_inode_operations;
+ init_special_inode(inode, inode->i_mode, inode->i_rdev);
+ break;
+ }
+
+ xfs_iflags_clear(ip, XFS_INEW);
+ barrier();
+
+ unlock_new_inode(inode);
+}
diff --git a/fs/xfs/linux-2.6/xfs_iops.h b/fs/xfs/linux-2.6/xfs_iops.h
index 14d0deb7aff..8b1a1e31dc2 100644
--- a/fs/xfs/linux-2.6/xfs_iops.h
+++ b/fs/xfs/linux-2.6/xfs_iops.h
@@ -18,23 +18,14 @@
#ifndef __XFS_IOPS_H__
#define __XFS_IOPS_H__
-extern const struct inode_operations xfs_inode_operations;
-extern const struct inode_operations xfs_dir_inode_operations;
-extern const struct inode_operations xfs_symlink_inode_operations;
+struct xfs_inode;
extern const struct file_operations xfs_file_operations;
extern const struct file_operations xfs_dir_file_operations;
extern const struct file_operations xfs_invis_file_operations;
+extern ssize_t xfs_vn_listxattr(struct dentry *, char *data, size_t size);
-struct xfs_inode;
-extern void xfs_ichgtime(struct xfs_inode *, int);
-extern void xfs_ichgtime_fast(struct xfs_inode *, struct inode *, int);
-
-#define xfs_vtoi(vp) \
- ((struct xfs_inode *)vn_to_inode(vp)->i_private)
-
-#define XFS_I(inode) \
- ((struct xfs_inode *)(inode)->i_private)
+extern void xfs_setup_inode(struct xfs_inode *);
#endif /* __XFS_IOPS_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h
index 4edc46915b5..cc0f7b3a979 100644
--- a/fs/xfs/linux-2.6/xfs_linux.h
+++ b/fs/xfs/linux-2.6/xfs_linux.h
@@ -45,13 +45,13 @@
#include <mrlock.h>
#include <sv.h>
#include <mutex.h>
-#include <sema.h>
#include <time.h>
#include <support/ktrace.h>
#include <support/debug.h>
#include <support/uuid.h>
+#include <linux/semaphore.h>
#include <linux/mm.h>
#include <linux/kernel.h>
#include <linux/blkdev.h>
@@ -76,6 +76,7 @@
#include <linux/log2.h>
#include <linux/spinlock.h>
#include <linux/random.h>
+#include <linux/ctype.h>
#include <asm/page.h>
#include <asm/div64.h>
@@ -125,8 +126,6 @@
#define current_cpu() (raw_smp_processor_id())
#define current_pid() (current->pid)
-#define current_fsuid(cred) (current->fsuid)
-#define current_fsgid(cred) (current->fsgid)
#define current_test_flags(f) (current->flags & (f))
#define current_set_flags_nested(sp, f) \
(*(sp) = current->flags, current->flags |= (f))
@@ -179,7 +178,7 @@
#define xfs_sort(a,n,s,fn) sort(a,n,s,fn,NULL)
#define xfs_stack_trace() dump_stack()
#define xfs_itruncate_data(ip, off) \
- (-vmtruncate(vn_to_inode(XFS_ITOV(ip)), (off)))
+ (-vmtruncate(VFS_I(ip), (off)))
/* Move the kernel do_div definition off to one side */
@@ -299,4 +298,11 @@ static inline __uint64_t howmany_64(__uint64_t x, __uint32_t y)
return x;
}
+/* ARM old ABI has some weird alignment/padding */
+#if defined(__arm__) && !defined(__ARM_EABI__)
+#define __arch_pack __attribute__((packed))
+#else
+#define __arch_pack
+#endif
+
#endif /* __XFS_LINUX__ */
diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c
index 5e3b57516ec..1957e5357d0 100644
--- a/fs/xfs/linux-2.6/xfs_lrw.c
+++ b/fs/xfs/linux-2.6/xfs_lrw.c
@@ -137,7 +137,7 @@ xfs_iozero(
struct address_space *mapping;
int status;
- mapping = ip->i_vnode->i_mapping;
+ mapping = VFS_I(ip)->i_mapping;
do {
unsigned offset, bytes;
void *fsdata;
@@ -674,9 +674,7 @@ start:
*/
if (likely(!(ioflags & IO_INVIS) &&
!mnt_want_write(file->f_path.mnt))) {
- file_update_time(file);
- xfs_ichgtime_fast(xip, inode,
- XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
+ xfs_ichgtime(xip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
mnt_drop_write(file->f_path.mnt);
}
@@ -711,7 +709,7 @@ start:
!capable(CAP_FSETID)) {
error = xfs_write_clear_setuid(xip);
if (likely(!error))
- error = -remove_suid(file->f_path.dentry);
+ error = -file_remove_suid(file);
if (unlikely(error)) {
goto out_unlock_internal;
}
diff --git a/fs/xfs/linux-2.6/xfs_stats.c b/fs/xfs/linux-2.6/xfs_stats.c
index e480b610205..3d5b67c075c 100644
--- a/fs/xfs/linux-2.6/xfs_stats.c
+++ b/fs/xfs/linux-2.6/xfs_stats.c
@@ -98,12 +98,21 @@ xfs_read_xfsstats(
return len;
}
-void
+int
xfs_init_procfs(void)
{
if (!proc_mkdir("fs/xfs", NULL))
- return;
- create_proc_read_entry("fs/xfs/stat", 0, NULL, xfs_read_xfsstats, NULL);
+ goto out;
+
+ if (!create_proc_read_entry("fs/xfs/stat", 0, NULL,
+ xfs_read_xfsstats, NULL))
+ goto out_remove_entry;
+ return 0;
+
+ out_remove_entry:
+ remove_proc_entry("fs/xfs", NULL);
+ out:
+ return -ENOMEM;
}
void
diff --git a/fs/xfs/linux-2.6/xfs_stats.h b/fs/xfs/linux-2.6/xfs_stats.h
index afd0b0d5fdb..e83820febc9 100644
--- a/fs/xfs/linux-2.6/xfs_stats.h
+++ b/fs/xfs/linux-2.6/xfs_stats.h
@@ -134,7 +134,7 @@ DECLARE_PER_CPU(struct xfsstats, xfsstats);
#define XFS_STATS_DEC(v) (per_cpu(xfsstats, current_cpu()).v--)
#define XFS_STATS_ADD(v, inc) (per_cpu(xfsstats, current_cpu()).v += (inc))
-extern void xfs_init_procfs(void);
+extern int xfs_init_procfs(void);
extern void xfs_cleanup_procfs(void);
@@ -144,8 +144,14 @@ extern void xfs_cleanup_procfs(void);
# define XFS_STATS_DEC(count)
# define XFS_STATS_ADD(count, inc)
-static inline void xfs_init_procfs(void) { };
-static inline void xfs_cleanup_procfs(void) { };
+static inline int xfs_init_procfs(void)
+{
+ return 0;
+}
+
+static inline void xfs_cleanup_procfs(void)
+{
+}
#endif /* !CONFIG_PROC_FS */
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 742b2c7852c..7227b2efef2 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -52,6 +52,12 @@
#include "xfs_version.h"
#include "xfs_log_priv.h"
#include "xfs_trans_priv.h"
+#include "xfs_filestream.h"
+#include "xfs_da_btree.h"
+#include "xfs_dir2_trace.h"
+#include "xfs_extfree_item.h"
+#include "xfs_mru_cache.h"
+#include "xfs_inode_item.h"
#include <linux/namei.h>
#include <linux/init.h>
@@ -60,6 +66,7 @@
#include <linux/writeback.h>
#include <linux/kthread.h>
#include <linux/freezer.h>
+#include <linux/parser.h>
static struct quotactl_ops xfs_quotactl_operations;
static struct super_operations xfs_super_operations;
@@ -74,7 +81,10 @@ xfs_args_allocate(
{
struct xfs_mount_args *args;
- args = kmem_zalloc(sizeof(struct xfs_mount_args), KM_SLEEP);
+ args = kzalloc(sizeof(struct xfs_mount_args), GFP_KERNEL);
+ if (!args)
+ return NULL;
+
args->logbufs = args->logbufsize = -1;
strncpy(args->fsname, sb->s_id, MAXNAMELEN);
@@ -138,6 +148,23 @@ xfs_args_allocate(
#define MNTOPT_XDSM "xdsm" /* DMI enabled (DMAPI / XDSM) */
#define MNTOPT_DMI "dmi" /* DMI enabled (DMAPI / XDSM) */
+/*
+ * Table driven mount option parser.
+ *
+ * Currently only used for remount, but it will be used for mount
+ * in the future, too.
+ */
+enum {
+ Opt_barrier, Opt_nobarrier, Opt_err
+};
+
+static const match_table_t tokens = {
+ {Opt_barrier, "barrier"},
+ {Opt_nobarrier, "nobarrier"},
+ {Opt_err, NULL}
+};
+
+
STATIC unsigned long
suffix_strtoul(char *s, char **endp, unsigned int base)
{
@@ -314,6 +341,7 @@ xfs_parseargs(
args->flags |= XFSMNT_ATTR2;
} else if (!strcmp(this_char, MNTOPT_NOATTR2)) {
args->flags &= ~XFSMNT_ATTR2;
+ args->flags |= XFSMNT_NOATTR2;
} else if (!strcmp(this_char, MNTOPT_FILESTREAM)) {
args->flags2 |= XFSMNT2_FILESTREAMS;
} else if (!strcmp(this_char, MNTOPT_NOQUOTA)) {
@@ -553,115 +581,6 @@ xfs_max_file_offset(
return (((__uint64_t)pagefactor) << bitshift) - 1;
}
-STATIC_INLINE void
-xfs_set_inodeops(
- struct inode *inode)
-{
- switch (inode->i_mode & S_IFMT) {
- case S_IFREG:
- inode->i_op = &xfs_inode_operations;
- inode->i_fop = &xfs_file_operations;
- inode->i_mapping->a_ops = &xfs_address_space_operations;
- break;
- case S_IFDIR:
- inode->i_op = &xfs_dir_inode_operations;
- inode->i_fop = &xfs_dir_file_operations;
- break;
- case S_IFLNK:
- inode->i_op = &xfs_symlink_inode_operations;
- if (!(XFS_I(inode)->i_df.if_flags & XFS_IFINLINE))
- inode->i_mapping->a_ops = &xfs_address_space_operations;
- break;
- default:
- inode->i_op = &xfs_inode_operations;
- init_special_inode(inode, inode->i_mode, inode->i_rdev);
- break;
- }
-}
-
-STATIC_INLINE void
-xfs_revalidate_inode(
- xfs_mount_t *mp,
- bhv_vnode_t *vp,
- xfs_inode_t *ip)
-{
- struct inode *inode = vn_to_inode(vp);
-
- inode->i_mode = ip->i_d.di_mode;
- inode->i_nlink = ip->i_d.di_nlink;
- inode->i_uid = ip->i_d.di_uid;
- inode->i_gid = ip->i_d.di_gid;
-
- switch (inode->i_mode & S_IFMT) {
- case S_IFBLK:
- case S_IFCHR:
- inode->i_rdev =
- MKDEV(sysv_major(ip->i_df.if_u2.if_rdev) & 0x1ff,
- sysv_minor(ip->i_df.if_u2.if_rdev));
- break;
- default:
- inode->i_rdev = 0;
- break;
- }
-
- inode->i_generation = ip->i_d.di_gen;
- i_size_write(inode, ip->i_d.di_size);
- inode->i_atime.tv_sec = ip->i_d.di_atime.t_sec;
- inode->i_atime.tv_nsec = ip->i_d.di_atime.t_nsec;
- inode->i_mtime.tv_sec = ip->i_d.di_mtime.t_sec;
- inode->i_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec;
- inode->i_ctime.tv_sec = ip->i_d.di_ctime.t_sec;
- inode->i_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec;
- if (ip->i_d.di_flags & XFS_DIFLAG_IMMUTABLE)
- inode->i_flags |= S_IMMUTABLE;
- else
- inode->i_flags &= ~S_IMMUTABLE;
- if (ip->i_d.di_flags & XFS_DIFLAG_APPEND)
- inode->i_flags |= S_APPEND;
- else
- inode->i_flags &= ~S_APPEND;
- if (ip->i_d.di_flags & XFS_DIFLAG_SYNC)
- inode->i_flags |= S_SYNC;
- else
- inode->i_flags &= ~S_SYNC;
- if (ip->i_d.di_flags & XFS_DIFLAG_NOATIME)
- inode->i_flags |= S_NOATIME;
- else
- inode->i_flags &= ~S_NOATIME;
- xfs_iflags_clear(ip, XFS_IMODIFIED);
-}
-
-void
-xfs_initialize_vnode(
- struct xfs_mount *mp,
- bhv_vnode_t *vp,
- struct xfs_inode *ip)
-{
- struct inode *inode = vn_to_inode(vp);
-
- if (!ip->i_vnode) {
- ip->i_vnode = vp;
- inode->i_private = ip;
- }
-
- /*
- * We need to set the ops vectors, and unlock the inode, but if
- * we have been called during the new inode create process, it is
- * too early to fill in the Linux inode. We will get called a
- * second time once the inode is properly set up, and then we can
- * finish our work.
- */
- if (ip->i_d.di_mode != 0 && (inode->i_state & I_NEW)) {
- xfs_revalidate_inode(mp, vp, ip);
- xfs_set_inodeops(inode);
-
- xfs_iflags_clear(ip, XFS_INEW);
- barrier();
-
- unlock_new_inode(inode);
- }
-}
-
int
xfs_blkdev_get(
xfs_mount_t *mp,
@@ -733,14 +652,6 @@ xfs_mountfs_check_barriers(xfs_mount_t *mp)
return;
}
- if (mp->m_ddev_targp->bt_bdev->bd_disk->queue->ordered ==
- QUEUE_ORDERED_NONE) {
- xfs_fs_cmn_err(CE_NOTE, mp,
- "Disabling barriers, not supported by the underlying device");
- mp->m_flags &= ~XFS_MOUNT_BARRIER;
- return;
- }
-
if (xfs_readonly_buftarg(mp->m_ddev_targp)) {
xfs_fs_cmn_err(CE_NOTE, mp,
"Disabling barriers, underlying device is readonly");
@@ -764,6 +675,139 @@ xfs_blkdev_issue_flush(
blkdev_issue_flush(buftarg->bt_bdev, NULL);
}
+STATIC void
+xfs_close_devices(
+ struct xfs_mount *mp)
+{
+ if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) {
+ struct block_device *logdev = mp->m_logdev_targp->bt_bdev;
+ xfs_free_buftarg(mp->m_logdev_targp);
+ xfs_blkdev_put(logdev);
+ }
+ if (mp->m_rtdev_targp) {
+ struct block_device *rtdev = mp->m_rtdev_targp->bt_bdev;
+ xfs_free_buftarg(mp->m_rtdev_targp);
+ xfs_blkdev_put(rtdev);
+ }
+ xfs_free_buftarg(mp->m_ddev_targp);
+}
+
+/*
+ * The file system configurations are:
+ * (1) device (partition) with data and internal log
+ * (2) logical volume with data and log subvolumes.
+ * (3) logical volume with data, log, and realtime subvolumes.
+ *
+ * We only have to handle opening the log and realtime volumes here if
+ * they are present. The data subvolume has already been opened by
+ * get_sb_bdev() and is stored in sb->s_bdev.
+ */
+STATIC int
+xfs_open_devices(
+ struct xfs_mount *mp,
+ struct xfs_mount_args *args)
+{
+ struct block_device *ddev = mp->m_super->s_bdev;
+ struct block_device *logdev = NULL, *rtdev = NULL;
+ int error;
+
+ /*
+ * Open real time and log devices - order is important.
+ */
+ if (args->logname[0]) {
+ error = xfs_blkdev_get(mp, args->logname, &logdev);
+ if (error)
+ goto out;
+ }
+
+ if (args->rtname[0]) {
+ error = xfs_blkdev_get(mp, args->rtname, &rtdev);
+ if (error)
+ goto out_close_logdev;
+
+ if (rtdev == ddev || rtdev == logdev) {
+ cmn_err(CE_WARN,
+ "XFS: Cannot mount filesystem with identical rtdev and ddev/logdev.");
+ error = EINVAL;
+ goto out_close_rtdev;
+ }
+ }
+
+ /*
+ * Setup xfs_mount buffer target pointers
+ */
+ error = ENOMEM;
+ mp->m_ddev_targp = xfs_alloc_buftarg(ddev, 0);
+ if (!mp->m_ddev_targp)
+ goto out_close_rtdev;
+
+ if (rtdev) {
+ mp->m_rtdev_targp = xfs_alloc_buftarg(rtdev, 1);
+ if (!mp->m_rtdev_targp)
+ goto out_free_ddev_targ;
+ }
+
+ if (logdev && logdev != ddev) {
+ mp->m_logdev_targp = xfs_alloc_buftarg(logdev, 1);
+ if (!mp->m_logdev_targp)
+ goto out_free_rtdev_targ;
+ } else {
+ mp->m_logdev_targp = mp->m_ddev_targp;
+ }
+
+ return 0;
+
+ out_free_rtdev_targ:
+ if (mp->m_rtdev_targp)
+ xfs_free_buftarg(mp->m_rtdev_targp);
+ out_free_ddev_targ:
+ xfs_free_buftarg(mp->m_ddev_targp);
+ out_close_rtdev:
+ if (rtdev)
+ xfs_blkdev_put(rtdev);
+ out_close_logdev:
+ if (logdev && logdev != ddev)
+ xfs_blkdev_put(logdev);
+ out:
+ return error;
+}
+
+/*
+ * Setup xfs_mount buffer target pointers based on superblock
+ */
+STATIC int
+xfs_setup_devices(
+ struct xfs_mount *mp)
+{
+ int error;
+
+ error = xfs_setsize_buftarg(mp->m_ddev_targp, mp->m_sb.sb_blocksize,
+ mp->m_sb.sb_sectsize);
+ if (error)
+ return error;
+
+ if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) {
+ unsigned int log_sector_size = BBSIZE;
+
+ if (xfs_sb_version_hassector(&mp->m_sb))
+ log_sector_size = mp->m_sb.sb_logsectsize;
+ error = xfs_setsize_buftarg(mp->m_logdev_targp,
+ mp->m_sb.sb_blocksize,
+ log_sector_size);
+ if (error)
+ return error;
+ }
+ if (mp->m_rtdev_targp) {
+ error = xfs_setsize_buftarg(mp->m_rtdev_targp,
+ mp->m_sb.sb_blocksize,
+ mp->m_sb.sb_sectsize);
+ if (error)
+ return error;
+ }
+
+ return 0;
+}
+
/*
* XFS AIL push thread support
*/
@@ -826,63 +870,21 @@ STATIC struct inode *
xfs_fs_alloc_inode(
struct super_block *sb)
{
- bhv_vnode_t *vp;
-
- vp = kmem_zone_alloc(xfs_vnode_zone, KM_SLEEP);
- if (unlikely(!vp))
- return NULL;
- return vn_to_inode(vp);
+ return kmem_zone_alloc(xfs_vnode_zone, KM_SLEEP);
}
STATIC void
xfs_fs_destroy_inode(
struct inode *inode)
{
- kmem_zone_free(xfs_vnode_zone, vn_from_inode(inode));
+ kmem_zone_free(xfs_vnode_zone, inode);
}
STATIC void
xfs_fs_inode_init_once(
- kmem_zone_t *zonep,
void *vnode)
{
- inode_init_once(vn_to_inode((bhv_vnode_t *)vnode));
-}
-
-STATIC int __init
-xfs_init_zones(void)
-{
- xfs_vnode_zone = kmem_zone_init_flags(sizeof(bhv_vnode_t), "xfs_vnode",
- KM_ZONE_HWALIGN | KM_ZONE_RECLAIM |
- KM_ZONE_SPREAD,
- xfs_fs_inode_init_once);
- if (!xfs_vnode_zone)
- goto out;
-
- xfs_ioend_zone = kmem_zone_init(sizeof(xfs_ioend_t), "xfs_ioend");
- if (!xfs_ioend_zone)
- goto out_destroy_vnode_zone;
-
- xfs_ioend_pool = mempool_create_slab_pool(4 * MAX_BUF_PER_PAGE,
- xfs_ioend_zone);
- if (!xfs_ioend_pool)
- goto out_free_ioend_zone;
- return 0;
-
- out_free_ioend_zone:
- kmem_zone_destroy(xfs_ioend_zone);
- out_destroy_vnode_zone:
- kmem_zone_destroy(xfs_vnode_zone);
- out:
- return -ENOMEM;
-}
-
-STATIC void
-xfs_destroy_zones(void)
-{
- mempool_destroy(xfs_ioend_pool);
- kmem_zone_destroy(xfs_vnode_zone);
- kmem_zone_destroy(xfs_ioend_zone);
+ inode_init_once((struct inode *)vnode);
}
/*
@@ -987,7 +989,7 @@ void
xfs_flush_inode(
xfs_inode_t *ip)
{
- struct inode *inode = ip->i_vnode;
+ struct inode *inode = VFS_I(ip);
igrab(inode);
xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_inode_work);
@@ -1012,7 +1014,7 @@ void
xfs_flush_device(
xfs_inode_t *ip)
{
- struct inode *inode = vn_to_inode(XFS_ITOV(ip));
+ struct inode *inode = VFS_I(ip);
igrab(inode);
xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_device_work);
@@ -1074,7 +1076,7 @@ xfssyncd(
list_del(&work->w_list);
if (work == &mp->m_sync_work)
continue;
- kmem_free(work, sizeof(struct bhv_vfs_sync_work));
+ kmem_free(work);
}
}
@@ -1082,18 +1084,76 @@ xfssyncd(
}
STATIC void
+xfs_free_fsname(
+ struct xfs_mount *mp)
+{
+ kfree(mp->m_fsname);
+ kfree(mp->m_rtname);
+ kfree(mp->m_logname);
+}
+
+STATIC void
xfs_fs_put_super(
struct super_block *sb)
{
struct xfs_mount *mp = XFS_M(sb);
+ struct xfs_inode *rip = mp->m_rootip;
+ int unmount_event_flags = 0;
int error;
kthread_stop(mp->m_sync_task);
xfs_sync(mp, SYNC_ATTR | SYNC_DELWRI);
- error = xfs_unmount(mp, 0, NULL);
- if (error)
- printk("XFS: unmount got error=%d\n", error);
+
+#ifdef HAVE_DMAPI
+ if (mp->m_flags & XFS_MOUNT_DMAPI) {
+ unmount_event_flags =
+ (mp->m_dmevmask & (1 << DM_EVENT_UNMOUNT)) ?
+ 0 : DM_FLAGS_UNWANTED;
+ /*
+ * Ignore error from dmapi here, first unmount is not allowed
+ * to fail anyway, and second we wouldn't want to fail a
+ * unmount because of dmapi.
+ */
+ XFS_SEND_PREUNMOUNT(mp, rip, DM_RIGHT_NULL, rip, DM_RIGHT_NULL,
+ NULL, NULL, 0, 0, unmount_event_flags);
+ }
+#endif
+
+ /*
+ * Blow away any referenced inode in the filestreams cache.
+ * This can and will cause log traffic as inodes go inactive
+ * here.
+ */
+ xfs_filestream_unmount(mp);
+
+ XFS_bflush(mp->m_ddev_targp);
+ error = xfs_unmount_flush(mp, 0);
+ WARN_ON(error);
+
+ /*
+ * If we're forcing a shutdown, typically because of a media error,
+ * we want to make sure we invalidate dirty pages that belong to
+ * referenced vnodes as well.
+ */
+ if (XFS_FORCED_SHUTDOWN(mp)) {
+ error = xfs_sync(mp, SYNC_WAIT | SYNC_CLOSE);
+ ASSERT(error != EFSCORRUPTED);
+ }
+
+ if (mp->m_flags & XFS_MOUNT_DMAPI) {
+ XFS_SEND_UNMOUNT(mp, rip, DM_RIGHT_NULL, 0, 0,
+ unmount_event_flags);
+ }
+
+ xfs_unmountfs(mp);
+ xfs_freesb(mp);
+ xfs_icsb_destroy_counters(mp);
+ xfs_close_devices(mp);
+ xfs_qmops_put(mp);
+ xfs_dmops_put(mp);
+ xfs_free_fsname(mp);
+ kfree(mp);
}
STATIC void
@@ -1216,14 +1276,74 @@ xfs_fs_remount(
char *options)
{
struct xfs_mount *mp = XFS_M(sb);
- struct xfs_mount_args *args = xfs_args_allocate(sb, 0);
- int error;
+ substring_t args[MAX_OPT_ARGS];
+ char *p;
- error = xfs_parseargs(mp, options, args, 1);
- if (!error)
- error = xfs_mntupdate(mp, flags, args);
- kmem_free(args, sizeof(*args));
- return -error;
+ while ((p = strsep(&options, ",")) != NULL) {
+ int token;
+
+ if (!*p)
+ continue;
+
+ token = match_token(p, tokens, args);
+ switch (token) {
+ case Opt_barrier:
+ mp->m_flags |= XFS_MOUNT_BARRIER;
+
+ /*
+ * Test if barriers are actually working if we can,
+ * else delay this check until the filesystem is
+ * marked writeable.
+ */
+ if (!(mp->m_flags & XFS_MOUNT_RDONLY))
+ xfs_mountfs_check_barriers(mp);
+ break;
+ case Opt_nobarrier:
+ mp->m_flags &= ~XFS_MOUNT_BARRIER;
+ break;
+ default:
+ /*
+ * Logically we would return an error here to prevent
+ * users from believing they might have changed
+ * mount options using remount which can't be changed.
+ *
+ * But unfortunately mount(8) adds all options from
+ * mtab and fstab to the mount arguments in some cases
+ * so we can't blindly reject options, but have to
+ * check for each specified option if it actually
+ * differs from the currently set option and only
+ * reject it if that's the case.
+ *
+ * Until that is implemented we return success for
+ * every remount request, and silently ignore all
+ * options that we can't actually change.
+ */
+#if 0
+ printk(KERN_INFO
+ "XFS: mount option \"%s\" not supported for remount\n", p);
+ return -EINVAL;
+#else
+ return 0;
+#endif
+ }
+ }
+
+ /* rw/ro -> rw */
+ if ((mp->m_flags & XFS_MOUNT_RDONLY) && !(*flags & MS_RDONLY)) {
+ mp->m_flags &= ~XFS_MOUNT_RDONLY;
+ if (mp->m_flags & XFS_MOUNT_BARRIER)
+ xfs_mountfs_check_barriers(mp);
+ }
+
+ /* rw -> ro */
+ if (!(mp->m_flags & XFS_MOUNT_RDONLY) && (*flags & MS_RDONLY)) {
+ xfs_filestream_flush(mp);
+ xfs_sync(mp, SYNC_DATA_QUIESCE);
+ xfs_attr_quiesce(mp);
+ mp->m_flags |= XFS_MOUNT_RDONLY;
+ }
+
+ return 0;
}
/*
@@ -1300,6 +1420,245 @@ xfs_fs_setxquota(
Q_XSETPQLIM), id, (caddr_t)fdq);
}
+/*
+ * This function fills in xfs_mount_t fields based on mount args.
+ * Note: the superblock has _not_ yet been read in.
+ */
+STATIC int
+xfs_start_flags(
+ struct xfs_mount_args *ap,
+ struct xfs_mount *mp)
+{
+ int error;
+
+ /* Values are in BBs */
+ if ((ap->flags & XFSMNT_NOALIGN) != XFSMNT_NOALIGN) {
+ /*
+ * At this point the superblock has not been read
+ * in, therefore we do not know the block size.
+ * Before the mount call ends we will convert
+ * these to FSBs.
+ */
+ mp->m_dalign = ap->sunit;
+ mp->m_swidth = ap->swidth;
+ }
+
+ if (ap->logbufs != -1 &&
+ ap->logbufs != 0 &&
+ (ap->logbufs < XLOG_MIN_ICLOGS ||
+ ap->logbufs > XLOG_MAX_ICLOGS)) {
+ cmn_err(CE_WARN,
+ "XFS: invalid logbufs value: %d [not %d-%d]",
+ ap->logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS);
+ return XFS_ERROR(EINVAL);
+ }
+ mp->m_logbufs = ap->logbufs;
+ if (ap->logbufsize != -1 &&
+ ap->logbufsize != 0 &&
+ (ap->logbufsize < XLOG_MIN_RECORD_BSIZE ||
+ ap->logbufsize > XLOG_MAX_RECORD_BSIZE ||
+ !is_power_of_2(ap->logbufsize))) {
+ cmn_err(CE_WARN,
+ "XFS: invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]",
+ ap->logbufsize);
+ return XFS_ERROR(EINVAL);
+ }
+
+ error = ENOMEM;
+
+ mp->m_logbsize = ap->logbufsize;
+ mp->m_fsname_len = strlen(ap->fsname) + 1;
+
+ mp->m_fsname = kstrdup(ap->fsname, GFP_KERNEL);
+ if (!mp->m_fsname)
+ goto out;
+
+ if (ap->rtname[0]) {
+ mp->m_rtname = kstrdup(ap->rtname, GFP_KERNEL);
+ if (!mp->m_rtname)
+ goto out_free_fsname;
+
+ }
+
+ if (ap->logname[0]) {
+ mp->m_logname = kstrdup(ap->logname, GFP_KERNEL);
+ if (!mp->m_logname)
+ goto out_free_rtname;
+ }
+
+ if (ap->flags & XFSMNT_WSYNC)
+ mp->m_flags |= XFS_MOUNT_WSYNC;
+#if XFS_BIG_INUMS
+ if (ap->flags & XFSMNT_INO64) {
+ mp->m_flags |= XFS_MOUNT_INO64;
+ mp->m_inoadd = XFS_INO64_OFFSET;
+ }
+#endif
+ if (ap->flags & XFSMNT_RETERR)
+ mp->m_flags |= XFS_MOUNT_RETERR;
+ if (ap->flags & XFSMNT_NOALIGN)
+ mp->m_flags |= XFS_MOUNT_NOALIGN;
+ if (ap->flags & XFSMNT_SWALLOC)
+ mp->m_flags |= XFS_MOUNT_SWALLOC;
+ if (ap->flags & XFSMNT_OSYNCISOSYNC)
+ mp->m_flags |= XFS_MOUNT_OSYNCISOSYNC;
+ if (ap->flags & XFSMNT_32BITINODES)
+ mp->m_flags |= XFS_MOUNT_32BITINODES;
+
+ if (ap->flags & XFSMNT_IOSIZE) {
+ if (ap->iosizelog > XFS_MAX_IO_LOG ||
+ ap->iosizelog < XFS_MIN_IO_LOG) {
+ cmn_err(CE_WARN,
+ "XFS: invalid log iosize: %d [not %d-%d]",
+ ap->iosizelog, XFS_MIN_IO_LOG,
+ XFS_MAX_IO_LOG);
+ return XFS_ERROR(EINVAL);
+ }
+
+ mp->m_flags |= XFS_MOUNT_DFLT_IOSIZE;
+ mp->m_readio_log = mp->m_writeio_log = ap->iosizelog;
+ }
+
+ if (ap->flags & XFSMNT_IKEEP)
+ mp->m_flags |= XFS_MOUNT_IKEEP;
+ if (ap->flags & XFSMNT_DIRSYNC)
+ mp->m_flags |= XFS_MOUNT_DIRSYNC;
+ if (ap->flags & XFSMNT_ATTR2)
+ mp->m_flags |= XFS_MOUNT_ATTR2;
+ if (ap->flags & XFSMNT_NOATTR2)
+ mp->m_flags |= XFS_MOUNT_NOATTR2;
+
+ if (ap->flags2 & XFSMNT2_COMPAT_IOSIZE)
+ mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
+
+ /*
+ * no recovery flag requires a read-only mount
+ */
+ if (ap->flags & XFSMNT_NORECOVERY) {
+ if (!(mp->m_flags & XFS_MOUNT_RDONLY)) {
+ cmn_err(CE_WARN,
+ "XFS: tried to mount a FS read-write without recovery!");
+ return XFS_ERROR(EINVAL);
+ }
+ mp->m_flags |= XFS_MOUNT_NORECOVERY;
+ }
+
+ if (ap->flags & XFSMNT_NOUUID)
+ mp->m_flags |= XFS_MOUNT_NOUUID;
+ if (ap->flags & XFSMNT_BARRIER)
+ mp->m_flags |= XFS_MOUNT_BARRIER;
+ else
+ mp->m_flags &= ~XFS_MOUNT_BARRIER;
+
+ if (ap->flags2 & XFSMNT2_FILESTREAMS)
+ mp->m_flags |= XFS_MOUNT_FILESTREAMS;
+
+ if (ap->flags & XFSMNT_DMAPI)
+ mp->m_flags |= XFS_MOUNT_DMAPI;
+ return 0;
+
+
+ out_free_rtname:
+ kfree(mp->m_rtname);
+ out_free_fsname:
+ kfree(mp->m_fsname);
+ out:
+ return error;
+}
+
+/*
+ * This function fills in xfs_mount_t fields based on mount args.
+ * Note: the superblock _has_ now been read in.
+ */
+STATIC int
+xfs_finish_flags(
+ struct xfs_mount_args *ap,
+ struct xfs_mount *mp)
+{
+ int ronly = (mp->m_flags & XFS_MOUNT_RDONLY);
+
+ /* Fail a mount where the logbuf is smaller then the log stripe */
+ if (xfs_sb_version_haslogv2(&mp->m_sb)) {
+ if ((ap->logbufsize <= 0) &&
+ (mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE)) {
+ mp->m_logbsize = mp->m_sb.sb_logsunit;
+ } else if (ap->logbufsize > 0 &&
+ ap->logbufsize < mp->m_sb.sb_logsunit) {
+ cmn_err(CE_WARN,
+ "XFS: logbuf size must be greater than or equal to log stripe size");
+ return XFS_ERROR(EINVAL);
+ }
+ } else {
+ /* Fail a mount if the logbuf is larger than 32K */
+ if (ap->logbufsize > XLOG_BIG_RECORD_BSIZE) {
+ cmn_err(CE_WARN,
+ "XFS: logbuf size for version 1 logs must be 16K or 32K");
+ return XFS_ERROR(EINVAL);
+ }
+ }
+
+ /*
+ * mkfs'ed attr2 will turn on attr2 mount unless explicitly
+ * told by noattr2 to turn it off
+ */
+ if (xfs_sb_version_hasattr2(&mp->m_sb) &&
+ !(ap->flags & XFSMNT_NOATTR2))
+ mp->m_flags |= XFS_MOUNT_ATTR2;
+
+ /*
+ * prohibit r/w mounts of read-only filesystems
+ */
+ if ((mp->m_sb.sb_flags & XFS_SBF_READONLY) && !ronly) {
+ cmn_err(CE_WARN,
+ "XFS: cannot mount a read-only filesystem as read-write");
+ return XFS_ERROR(EROFS);
+ }
+
+ /*
+ * check for shared mount.
+ */
+ if (ap->flags & XFSMNT_SHARED) {
+ if (!xfs_sb_version_hasshared(&mp->m_sb))
+ return XFS_ERROR(EINVAL);
+
+ /*
+ * For IRIX 6.5, shared mounts must have the shared
+ * version bit set, have the persistent readonly
+ * field set, must be version 0 and can only be mounted
+ * read-only.
+ */
+ if (!ronly || !(mp->m_sb.sb_flags & XFS_SBF_READONLY) ||
+ (mp->m_sb.sb_shared_vn != 0))
+ return XFS_ERROR(EINVAL);
+
+ mp->m_flags |= XFS_MOUNT_SHARED;
+
+ /*
+ * Shared XFS V0 can't deal with DMI. Return EINVAL.
+ */
+ if (mp->m_sb.sb_shared_vn == 0 && (ap->flags & XFSMNT_DMAPI))
+ return XFS_ERROR(EINVAL);
+ }
+
+ if (ap->flags & XFSMNT_UQUOTA) {
+ mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE);
+ if (ap->flags & XFSMNT_UQUOTAENF)
+ mp->m_qflags |= XFS_UQUOTA_ENFD;
+ }
+
+ if (ap->flags & XFSMNT_GQUOTA) {
+ mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE);
+ if (ap->flags & XFSMNT_GQUOTAENF)
+ mp->m_qflags |= XFS_OQUOTA_ENFD;
+ } else if (ap->flags & XFSMNT_PQUOTA) {
+ mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE);
+ if (ap->flags & XFSMNT_PQUOTAENF)
+ mp->m_qflags |= XFS_OQUOTA_ENFD;
+ }
+
+ return 0;
+}
+
STATIC int
xfs_fs_fill_super(
struct super_block *sb,
@@ -1308,11 +1667,21 @@ xfs_fs_fill_super(
{
struct inode *root;
struct xfs_mount *mp = NULL;
- struct xfs_mount_args *args = xfs_args_allocate(sb, silent);
- int error;
+ struct xfs_mount_args *args;
+ int flags = 0, error = ENOMEM;
- mp = xfs_mount_init();
+ args = xfs_args_allocate(sb, silent);
+ if (!args)
+ return -ENOMEM;
+ mp = kzalloc(sizeof(struct xfs_mount), GFP_KERNEL);
+ if (!mp)
+ goto out_free_args;
+
+ spin_lock_init(&mp->m_sb_lock);
+ mutex_init(&mp->m_ilock);
+ mutex_init(&mp->m_growlock);
+ atomic_set(&mp->m_active_trans, 0);
INIT_LIST_HEAD(&mp->m_sync_list);
spin_lock_init(&mp->m_sync_lock);
init_waitqueue_head(&mp->m_wait_single_sync_task);
@@ -1325,16 +1694,60 @@ xfs_fs_fill_super(
error = xfs_parseargs(mp, (char *)data, args, 0);
if (error)
- goto fail_vfsop;
+ goto out_free_mp;
sb_min_blocksize(sb, BBSIZE);
+ sb->s_xattr = xfs_xattr_handlers;
sb->s_export_op = &xfs_export_operations;
sb->s_qcop = &xfs_quotactl_operations;
sb->s_op = &xfs_super_operations;
- error = xfs_mount(mp, args, NULL);
+ error = xfs_dmops_get(mp, args);
+ if (error)
+ goto out_free_mp;
+ error = xfs_qmops_get(mp, args);
+ if (error)
+ goto out_put_dmops;
+
+ if (args->flags & XFSMNT_QUIET)
+ flags |= XFS_MFSI_QUIET;
+
+ error = xfs_open_devices(mp, args);
+ if (error)
+ goto out_put_qmops;
+
+ if (xfs_icsb_init_counters(mp))
+ mp->m_flags |= XFS_MOUNT_NO_PERCPU_SB;
+
+ /*
+ * Setup flags based on mount(2) options and then the superblock
+ */
+ error = xfs_start_flags(args, mp);
+ if (error)
+ goto out_free_fsname;
+ error = xfs_readsb(mp, flags);
+ if (error)
+ goto out_free_fsname;
+ error = xfs_finish_flags(args, mp);
if (error)
- goto fail_vfsop;
+ goto out_free_sb;
+
+ error = xfs_setup_devices(mp);
+ if (error)
+ goto out_free_sb;
+
+ if (mp->m_flags & XFS_MOUNT_BARRIER)
+ xfs_mountfs_check_barriers(mp);
+
+ error = xfs_filestream_mount(mp);
+ if (error)
+ goto out_free_sb;
+
+ error = xfs_mountfs(mp);
+ if (error)
+ goto out_filestream_unmount;
+
+ XFS_SEND_MOUNT(mp, DM_RIGHT_NULL, args->mtpt, args->fsname);
sb->s_dirt = 1;
sb->s_magic = XFS_SB_MAGIC;
@@ -1344,7 +1757,7 @@ xfs_fs_fill_super(
sb->s_time_gran = 1;
set_posix_acl_flag(sb);
- root = igrab(mp->m_rootip->i_vnode);
+ root = igrab(VFS_I(mp->m_rootip));
if (!root) {
error = ENOENT;
goto fail_unmount;
@@ -1369,10 +1782,28 @@ xfs_fs_fill_super(
xfs_itrace_exit(XFS_I(sb->s_root->d_inode));
- kmem_free(args, sizeof(*args));
+ kfree(args);
return 0;
-fail_vnrele:
+ out_filestream_unmount:
+ xfs_filestream_unmount(mp);
+ out_free_sb:
+ xfs_freesb(mp);
+ out_free_fsname:
+ xfs_free_fsname(mp);
+ xfs_icsb_destroy_counters(mp);
+ xfs_close_devices(mp);
+ out_put_qmops:
+ xfs_qmops_put(mp);
+ out_put_dmops:
+ xfs_dmops_put(mp);
+ out_free_mp:
+ kfree(mp);
+ out_free_args:
+ kfree(args);
+ return -error;
+
+ fail_vnrele:
if (sb->s_root) {
dput(sb->s_root);
sb->s_root = NULL;
@@ -1380,12 +1811,20 @@ fail_vnrele:
iput(root);
}
-fail_unmount:
- xfs_unmount(mp, 0, NULL);
+ fail_unmount:
+ /*
+ * Blow away any referenced inode in the filestreams cache.
+ * This can and will cause log traffic as inodes go inactive
+ * here.
+ */
+ xfs_filestream_unmount(mp);
-fail_vfsop:
- kmem_free(args, sizeof(*args));
- return -error;
+ XFS_bflush(mp->m_ddev_targp);
+ error = xfs_unmount_flush(mp, 0);
+ WARN_ON(error);
+
+ xfs_unmountfs(mp);
+ goto out_free_sb;
}
STATIC int
@@ -1430,9 +1869,235 @@ static struct file_system_type xfs_fs_type = {
.fs_flags = FS_REQUIRES_DEV,
};
+STATIC int __init
+xfs_alloc_trace_bufs(void)
+{
+#ifdef XFS_ALLOC_TRACE
+ xfs_alloc_trace_buf = ktrace_alloc(XFS_ALLOC_TRACE_SIZE, KM_MAYFAIL);
+ if (!xfs_alloc_trace_buf)
+ goto out;
+#endif
+#ifdef XFS_BMAP_TRACE
+ xfs_bmap_trace_buf = ktrace_alloc(XFS_BMAP_TRACE_SIZE, KM_MAYFAIL);
+ if (!xfs_bmap_trace_buf)
+ goto out_free_alloc_trace;
+#endif
+#ifdef XFS_BMBT_TRACE
+ xfs_bmbt_trace_buf = ktrace_alloc(XFS_BMBT_TRACE_SIZE, KM_MAYFAIL);
+ if (!xfs_bmbt_trace_buf)
+ goto out_free_bmap_trace;
+#endif
+#ifdef XFS_ATTR_TRACE
+ xfs_attr_trace_buf = ktrace_alloc(XFS_ATTR_TRACE_SIZE, KM_MAYFAIL);
+ if (!xfs_attr_trace_buf)
+ goto out_free_bmbt_trace;
+#endif
+#ifdef XFS_DIR2_TRACE
+ xfs_dir2_trace_buf = ktrace_alloc(XFS_DIR2_GTRACE_SIZE, KM_MAYFAIL);
+ if (!xfs_dir2_trace_buf)
+ goto out_free_attr_trace;
+#endif
+
+ return 0;
+
+#ifdef XFS_DIR2_TRACE
+ out_free_attr_trace:
+#endif
+#ifdef XFS_ATTR_TRACE
+ ktrace_free(xfs_attr_trace_buf);
+ out_free_bmbt_trace:
+#endif
+#ifdef XFS_BMBT_TRACE
+ ktrace_free(xfs_bmbt_trace_buf);
+ out_free_bmap_trace:
+#endif
+#ifdef XFS_BMAP_TRACE
+ ktrace_free(xfs_bmap_trace_buf);
+ out_free_alloc_trace:
+#endif
+#ifdef XFS_ALLOC_TRACE
+ ktrace_free(xfs_alloc_trace_buf);
+ out:
+#endif
+ return -ENOMEM;
+}
+
+STATIC void
+xfs_free_trace_bufs(void)
+{
+#ifdef XFS_DIR2_TRACE
+ ktrace_free(xfs_dir2_trace_buf);
+#endif
+#ifdef XFS_ATTR_TRACE
+ ktrace_free(xfs_attr_trace_buf);
+#endif
+#ifdef XFS_BMBT_TRACE
+ ktrace_free(xfs_bmbt_trace_buf);
+#endif
+#ifdef XFS_BMAP_TRACE
+ ktrace_free(xfs_bmap_trace_buf);
+#endif
+#ifdef XFS_ALLOC_TRACE
+ ktrace_free(xfs_alloc_trace_buf);
+#endif
+}
+
+STATIC int __init
+xfs_init_zones(void)
+{
+ xfs_vnode_zone = kmem_zone_init_flags(sizeof(struct inode), "xfs_vnode",
+ KM_ZONE_HWALIGN | KM_ZONE_RECLAIM |
+ KM_ZONE_SPREAD,
+ xfs_fs_inode_init_once);
+ if (!xfs_vnode_zone)
+ goto out;
+
+ xfs_ioend_zone = kmem_zone_init(sizeof(xfs_ioend_t), "xfs_ioend");
+ if (!xfs_ioend_zone)
+ goto out_destroy_vnode_zone;
+
+ xfs_ioend_pool = mempool_create_slab_pool(4 * MAX_BUF_PER_PAGE,
+ xfs_ioend_zone);
+ if (!xfs_ioend_pool)
+ goto out_destroy_ioend_zone;
+
+ xfs_log_ticket_zone = kmem_zone_init(sizeof(xlog_ticket_t),
+ "xfs_log_ticket");
+ if (!xfs_log_ticket_zone)
+ goto out_destroy_ioend_pool;
+
+ xfs_bmap_free_item_zone = kmem_zone_init(sizeof(xfs_bmap_free_item_t),
+ "xfs_bmap_free_item");
+ if (!xfs_bmap_free_item_zone)
+ goto out_destroy_log_ticket_zone;
+ xfs_btree_cur_zone = kmem_zone_init(sizeof(xfs_btree_cur_t),
+ "xfs_btree_cur");
+ if (!xfs_btree_cur_zone)
+ goto out_destroy_bmap_free_item_zone;
+
+ xfs_da_state_zone = kmem_zone_init(sizeof(xfs_da_state_t),
+ "xfs_da_state");
+ if (!xfs_da_state_zone)
+ goto out_destroy_btree_cur_zone;
+
+ xfs_dabuf_zone = kmem_zone_init(sizeof(xfs_dabuf_t), "xfs_dabuf");
+ if (!xfs_dabuf_zone)
+ goto out_destroy_da_state_zone;
+
+ xfs_ifork_zone = kmem_zone_init(sizeof(xfs_ifork_t), "xfs_ifork");
+ if (!xfs_ifork_zone)
+ goto out_destroy_dabuf_zone;
+
+ xfs_trans_zone = kmem_zone_init(sizeof(xfs_trans_t), "xfs_trans");
+ if (!xfs_trans_zone)
+ goto out_destroy_ifork_zone;
+
+ /*
+ * The size of the zone allocated buf log item is the maximum
+ * size possible under XFS. This wastes a little bit of memory,
+ * but it is much faster.
+ */
+ xfs_buf_item_zone = kmem_zone_init((sizeof(xfs_buf_log_item_t) +
+ (((XFS_MAX_BLOCKSIZE / XFS_BLI_CHUNK) /
+ NBWORD) * sizeof(int))), "xfs_buf_item");
+ if (!xfs_buf_item_zone)
+ goto out_destroy_trans_zone;
+
+ xfs_efd_zone = kmem_zone_init((sizeof(xfs_efd_log_item_t) +
+ ((XFS_EFD_MAX_FAST_EXTENTS - 1) *
+ sizeof(xfs_extent_t))), "xfs_efd_item");
+ if (!xfs_efd_zone)
+ goto out_destroy_buf_item_zone;
+
+ xfs_efi_zone = kmem_zone_init((sizeof(xfs_efi_log_item_t) +
+ ((XFS_EFI_MAX_FAST_EXTENTS - 1) *
+ sizeof(xfs_extent_t))), "xfs_efi_item");
+ if (!xfs_efi_zone)
+ goto out_destroy_efd_zone;
+
+ xfs_inode_zone =
+ kmem_zone_init_flags(sizeof(xfs_inode_t), "xfs_inode",
+ KM_ZONE_HWALIGN | KM_ZONE_RECLAIM |
+ KM_ZONE_SPREAD, NULL);
+ if (!xfs_inode_zone)
+ goto out_destroy_efi_zone;
+
+ xfs_ili_zone =
+ kmem_zone_init_flags(sizeof(xfs_inode_log_item_t), "xfs_ili",
+ KM_ZONE_SPREAD, NULL);
+ if (!xfs_ili_zone)
+ goto out_destroy_inode_zone;
+
+#ifdef CONFIG_XFS_POSIX_ACL
+ xfs_acl_zone = kmem_zone_init(sizeof(xfs_acl_t), "xfs_acl");
+ if (!xfs_acl_zone)
+ goto out_destroy_ili_zone;
+#endif
+
+ return 0;
+
+#ifdef CONFIG_XFS_POSIX_ACL
+ out_destroy_ili_zone:
+#endif
+ kmem_zone_destroy(xfs_ili_zone);
+ out_destroy_inode_zone:
+ kmem_zone_destroy(xfs_inode_zone);
+ out_destroy_efi_zone:
+ kmem_zone_destroy(xfs_efi_zone);
+ out_destroy_efd_zone:
+ kmem_zone_destroy(xfs_efd_zone);
+ out_destroy_buf_item_zone:
+ kmem_zone_destroy(xfs_buf_item_zone);
+ out_destroy_trans_zone:
+ kmem_zone_destroy(xfs_trans_zone);
+ out_destroy_ifork_zone:
+ kmem_zone_destroy(xfs_ifork_zone);
+ out_destroy_dabuf_zone:
+ kmem_zone_destroy(xfs_dabuf_zone);
+ out_destroy_da_state_zone:
+ kmem_zone_destroy(xfs_da_state_zone);
+ out_destroy_btree_cur_zone:
+ kmem_zone_destroy(xfs_btree_cur_zone);
+ out_destroy_bmap_free_item_zone:
+ kmem_zone_destroy(xfs_bmap_free_item_zone);
+ out_destroy_log_ticket_zone:
+ kmem_zone_destroy(xfs_log_ticket_zone);
+ out_destroy_ioend_pool:
+ mempool_destroy(xfs_ioend_pool);
+ out_destroy_ioend_zone:
+ kmem_zone_destroy(xfs_ioend_zone);
+ out_destroy_vnode_zone:
+ kmem_zone_destroy(xfs_vnode_zone);
+ out:
+ return -ENOMEM;
+}
+
+STATIC void
+xfs_destroy_zones(void)
+{
+#ifdef CONFIG_XFS_POSIX_ACL
+ kmem_zone_destroy(xfs_acl_zone);
+#endif
+ kmem_zone_destroy(xfs_ili_zone);
+ kmem_zone_destroy(xfs_inode_zone);
+ kmem_zone_destroy(xfs_efi_zone);
+ kmem_zone_destroy(xfs_efd_zone);
+ kmem_zone_destroy(xfs_buf_item_zone);
+ kmem_zone_destroy(xfs_trans_zone);
+ kmem_zone_destroy(xfs_ifork_zone);
+ kmem_zone_destroy(xfs_dabuf_zone);
+ kmem_zone_destroy(xfs_da_state_zone);
+ kmem_zone_destroy(xfs_btree_cur_zone);
+ kmem_zone_destroy(xfs_bmap_free_item_zone);
+ kmem_zone_destroy(xfs_log_ticket_zone);
+ mempool_destroy(xfs_ioend_pool);
+ kmem_zone_destroy(xfs_ioend_zone);
+ kmem_zone_destroy(xfs_vnode_zone);
+
+}
STATIC int __init
-init_xfs_fs( void )
+init_xfs_fs(void)
{
int error;
static char message[] __initdata = KERN_INFO \
@@ -1441,42 +2106,73 @@ init_xfs_fs( void )
printk(message);
ktrace_init(64);
+ vn_init();
+ xfs_dir_startup();
error = xfs_init_zones();
- if (error < 0)
- goto undo_zones;
+ if (error)
+ goto out;
+
+ error = xfs_alloc_trace_bufs();
+ if (error)
+ goto out_destroy_zones;
+
+ error = xfs_mru_cache_init();
+ if (error)
+ goto out_free_trace_buffers;
+
+ error = xfs_filestream_init();
+ if (error)
+ goto out_mru_cache_uninit;
error = xfs_buf_init();
- if (error < 0)
- goto undo_buffers;
+ if (error)
+ goto out_filestream_uninit;
+
+ error = xfs_init_procfs();
+ if (error)
+ goto out_buf_terminate;
+
+ error = xfs_sysctl_register();
+ if (error)
+ goto out_cleanup_procfs;
- vn_init();
- xfs_init();
- uuid_init();
vfs_initquota();
error = register_filesystem(&xfs_fs_type);
if (error)
- goto undo_register;
+ goto out_sysctl_unregister;
return 0;
-undo_register:
+ out_sysctl_unregister:
+ xfs_sysctl_unregister();
+ out_cleanup_procfs:
+ xfs_cleanup_procfs();
+ out_buf_terminate:
xfs_buf_terminate();
-
-undo_buffers:
+ out_filestream_uninit:
+ xfs_filestream_uninit();
+ out_mru_cache_uninit:
+ xfs_mru_cache_uninit();
+ out_free_trace_buffers:
+ xfs_free_trace_bufs();
+ out_destroy_zones:
xfs_destroy_zones();
-
-undo_zones:
+ out:
return error;
}
STATIC void __exit
-exit_xfs_fs( void )
+exit_xfs_fs(void)
{
vfs_exitquota();
unregister_filesystem(&xfs_fs_type);
- xfs_cleanup();
+ xfs_sysctl_unregister();
+ xfs_cleanup_procfs();
xfs_buf_terminate();
+ xfs_filestream_uninit();
+ xfs_mru_cache_uninit();
+ xfs_free_trace_bufs();
xfs_destroy_zones();
ktrace_uninit();
}
diff --git a/fs/xfs/linux-2.6/xfs_super.h b/fs/xfs/linux-2.6/xfs_super.h
index 3efb7c6d330..fe2ef4e6a0f 100644
--- a/fs/xfs/linux-2.6/xfs_super.h
+++ b/fs/xfs/linux-2.6/xfs_super.h
@@ -101,18 +101,13 @@ struct block_device;
extern __uint64_t xfs_max_file_offset(unsigned int);
-extern void xfs_initialize_vnode(struct xfs_mount *mp, bhv_vnode_t *vp,
- struct xfs_inode *ip);
-
extern void xfs_flush_inode(struct xfs_inode *);
extern void xfs_flush_device(struct xfs_inode *);
-extern int xfs_blkdev_get(struct xfs_mount *, const char *,
- struct block_device **);
-extern void xfs_blkdev_put(struct block_device *);
extern void xfs_blkdev_issue_flush(struct xfs_buftarg *);
extern const struct export_operations xfs_export_operations;
+extern struct xattr_handler *xfs_xattr_handlers[];
#define XFS_M(sb) ((struct xfs_mount *)((sb)->s_fs_info))
diff --git a/fs/xfs/linux-2.6/xfs_sysctl.c b/fs/xfs/linux-2.6/xfs_sysctl.c
index bb997d75c05..7dacb5bbde3 100644
--- a/fs/xfs/linux-2.6/xfs_sysctl.c
+++ b/fs/xfs/linux-2.6/xfs_sysctl.c
@@ -259,15 +259,17 @@ static ctl_table xfs_root_table[] = {
{}
};
-void
+int
xfs_sysctl_register(void)
{
xfs_table_header = register_sysctl_table(xfs_root_table);
+ if (!xfs_table_header)
+ return -ENOMEM;
+ return 0;
}
void
xfs_sysctl_unregister(void)
{
- if (xfs_table_header)
- unregister_sysctl_table(xfs_table_header);
+ unregister_sysctl_table(xfs_table_header);
}
diff --git a/fs/xfs/linux-2.6/xfs_sysctl.h b/fs/xfs/linux-2.6/xfs_sysctl.h
index 98b97e399d6..4aadb8056c3 100644
--- a/fs/xfs/linux-2.6/xfs_sysctl.h
+++ b/fs/xfs/linux-2.6/xfs_sysctl.h
@@ -93,10 +93,10 @@ enum {
extern xfs_param_t xfs_params;
#ifdef CONFIG_SYSCTL
-extern void xfs_sysctl_register(void);
+extern int xfs_sysctl_register(void);
extern void xfs_sysctl_unregister(void);
#else
-# define xfs_sysctl_register() do { } while (0)
+# define xfs_sysctl_register() (0)
# define xfs_sysctl_unregister() do { } while (0)
#endif /* CONFIG_SYSCTL */
diff --git a/fs/xfs/linux-2.6/xfs_vnode.c b/fs/xfs/linux-2.6/xfs_vnode.c
index bc7afe00733..b52528bbbff 100644
--- a/fs/xfs/linux-2.6/xfs_vnode.c
+++ b/fs/xfs/linux-2.6/xfs_vnode.c
@@ -33,7 +33,7 @@
/*
- * Dedicated vnode inactive/reclaim sync semaphores.
+ * Dedicated vnode inactive/reclaim sync wait queues.
* Prime number of hash buckets since address is used as the key.
*/
#define NVSYNC 37
@@ -82,74 +82,6 @@ vn_ioerror(
xfs_do_force_shutdown(ip->i_mount, SHUTDOWN_DEVICE_REQ, f, l);
}
-/*
- * Revalidate the Linux inode from the XFS inode.
- * Note: i_size _not_ updated; we must hold the inode
- * semaphore when doing that - callers responsibility.
- */
-int
-vn_revalidate(
- bhv_vnode_t *vp)
-{
- struct inode *inode = vn_to_inode(vp);
- struct xfs_inode *ip = XFS_I(inode);
- struct xfs_mount *mp = ip->i_mount;
- unsigned long xflags;
-
- xfs_itrace_entry(ip);
-
- if (XFS_FORCED_SHUTDOWN(mp))
- return -EIO;
-
- xfs_ilock(ip, XFS_ILOCK_SHARED);
- inode->i_mode = ip->i_d.di_mode;
- inode->i_uid = ip->i_d.di_uid;
- inode->i_gid = ip->i_d.di_gid;
- inode->i_mtime.tv_sec = ip->i_d.di_mtime.t_sec;
- inode->i_mtime.tv_nsec = ip->i_d.di_mtime.t_nsec;
- inode->i_ctime.tv_sec = ip->i_d.di_ctime.t_sec;
- inode->i_ctime.tv_nsec = ip->i_d.di_ctime.t_nsec;
-
- xflags = xfs_ip2xflags(ip);
- if (xflags & XFS_XFLAG_IMMUTABLE)
- inode->i_flags |= S_IMMUTABLE;
- else
- inode->i_flags &= ~S_IMMUTABLE;
- if (xflags & XFS_XFLAG_APPEND)
- inode->i_flags |= S_APPEND;
- else
- inode->i_flags &= ~S_APPEND;
- if (xflags & XFS_XFLAG_SYNC)
- inode->i_flags |= S_SYNC;
- else
- inode->i_flags &= ~S_SYNC;
- if (xflags & XFS_XFLAG_NOATIME)
- inode->i_flags |= S_NOATIME;
- else
- inode->i_flags &= ~S_NOATIME;
- xfs_iunlock(ip, XFS_ILOCK_SHARED);
-
- xfs_iflags_clear(ip, XFS_IMODIFIED);
- return 0;
-}
-
-/*
- * Add a reference to a referenced vnode.
- */
-bhv_vnode_t *
-vn_hold(
- bhv_vnode_t *vp)
-{
- struct inode *inode;
-
- XFS_STATS_INC(vn_hold);
-
- inode = igrab(vn_to_inode(vp));
- ASSERT(inode);
-
- return vp;
-}
-
#ifdef XFS_INODE_TRACE
/*
@@ -158,7 +90,7 @@ vn_hold(
*/
static inline int xfs_icount(struct xfs_inode *ip)
{
- bhv_vnode_t *vp = XFS_ITOV_NULL(ip);
+ struct inode *vp = VFS_I(ip);
if (vp)
return vn_count(vp);
diff --git a/fs/xfs/linux-2.6/xfs_vnode.h b/fs/xfs/linux-2.6/xfs_vnode.h
index 25eb2a9e8d9..683ce16210f 100644
--- a/fs/xfs/linux-2.6/xfs_vnode.h
+++ b/fs/xfs/linux-2.6/xfs_vnode.h
@@ -19,24 +19,9 @@
#define __XFS_VNODE_H__
struct file;
-struct bhv_vattr;
struct xfs_iomap;
struct attrlist_cursor_kern;
-typedef struct inode bhv_vnode_t;
-
-/*
- * Vnode to Linux inode mapping.
- */
-static inline bhv_vnode_t *vn_from_inode(struct inode *inode)
-{
- return inode;
-}
-static inline struct inode *vn_to_inode(bhv_vnode_t *vnode)
-{
- return vnode;
-}
-
/*
* Return values for xfs_inactive. A return value of
* VN_INACTIVE_NOCACHE implies that the file system behavior
@@ -66,87 +51,8 @@ static inline struct inode *vn_to_inode(bhv_vnode_t *vnode)
Prevent VM access to the pages until
the operation completes. */
-/*
- * Vnode attributes. va_mask indicates those attributes the caller
- * wants to set or extract.
- */
-typedef struct bhv_vattr {
- int va_mask; /* bit-mask of attributes present */
- mode_t va_mode; /* file access mode and type */
- xfs_nlink_t va_nlink; /* number of references to file */
- uid_t va_uid; /* owner user id */
- gid_t va_gid; /* owner group id */
- xfs_ino_t va_nodeid; /* file id */
- xfs_off_t va_size; /* file size in bytes */
- u_long va_blocksize; /* blocksize preferred for i/o */
- struct timespec va_atime; /* time of last access */
- struct timespec va_mtime; /* time of last modification */
- struct timespec va_ctime; /* time file changed */
- u_int va_gen; /* generation number of file */
- xfs_dev_t va_rdev; /* device the special file represents */
- __int64_t va_nblocks; /* number of blocks allocated */
- u_long va_xflags; /* random extended file flags */
- u_long va_extsize; /* file extent size */
- u_long va_nextents; /* number of extents in file */
- u_long va_anextents; /* number of attr extents in file */
- prid_t va_projid; /* project id */
-} bhv_vattr_t;
-
-/*
- * setattr or getattr attributes
- */
-#define XFS_AT_TYPE 0x00000001
-#define XFS_AT_MODE 0x00000002
-#define XFS_AT_UID 0x00000004
-#define XFS_AT_GID 0x00000008
-#define XFS_AT_FSID 0x00000010
-#define XFS_AT_NODEID 0x00000020
-#define XFS_AT_NLINK 0x00000040
-#define XFS_AT_SIZE 0x00000080
-#define XFS_AT_ATIME 0x00000100
-#define XFS_AT_MTIME 0x00000200
-#define XFS_AT_CTIME 0x00000400
-#define XFS_AT_RDEV 0x00000800
-#define XFS_AT_BLKSIZE 0x00001000
-#define XFS_AT_NBLOCKS 0x00002000
-#define XFS_AT_VCODE 0x00004000
-#define XFS_AT_MAC 0x00008000
-#define XFS_AT_UPDATIME 0x00010000
-#define XFS_AT_UPDMTIME 0x00020000
-#define XFS_AT_UPDCTIME 0x00040000
-#define XFS_AT_ACL 0x00080000
-#define XFS_AT_CAP 0x00100000
-#define XFS_AT_INF 0x00200000
-#define XFS_AT_XFLAGS 0x00400000
-#define XFS_AT_EXTSIZE 0x00800000
-#define XFS_AT_NEXTENTS 0x01000000
-#define XFS_AT_ANEXTENTS 0x02000000
-#define XFS_AT_PROJID 0x04000000
-#define XFS_AT_SIZE_NOPERM 0x08000000
-#define XFS_AT_GENCOUNT 0x10000000
-
-#define XFS_AT_ALL (XFS_AT_TYPE|XFS_AT_MODE|XFS_AT_UID|XFS_AT_GID|\
- XFS_AT_FSID|XFS_AT_NODEID|XFS_AT_NLINK|XFS_AT_SIZE|\
- XFS_AT_ATIME|XFS_AT_MTIME|XFS_AT_CTIME|XFS_AT_RDEV|\
- XFS_AT_BLKSIZE|XFS_AT_NBLOCKS|XFS_AT_VCODE|XFS_AT_MAC|\
- XFS_AT_ACL|XFS_AT_CAP|XFS_AT_INF|XFS_AT_XFLAGS|XFS_AT_EXTSIZE|\
- XFS_AT_NEXTENTS|XFS_AT_ANEXTENTS|XFS_AT_PROJID|XFS_AT_GENCOUNT)
-
-#define XFS_AT_STAT (XFS_AT_TYPE|XFS_AT_MODE|XFS_AT_UID|XFS_AT_GID|\
- XFS_AT_FSID|XFS_AT_NODEID|XFS_AT_NLINK|XFS_AT_SIZE|\
- XFS_AT_ATIME|XFS_AT_MTIME|XFS_AT_CTIME|XFS_AT_RDEV|\
- XFS_AT_BLKSIZE|XFS_AT_NBLOCKS|XFS_AT_PROJID)
-
-#define XFS_AT_TIMES (XFS_AT_ATIME|XFS_AT_MTIME|XFS_AT_CTIME)
-
-#define XFS_AT_UPDTIMES (XFS_AT_UPDATIME|XFS_AT_UPDMTIME|XFS_AT_UPDCTIME)
-
-#define XFS_AT_NOSET (XFS_AT_NLINK|XFS_AT_RDEV|XFS_AT_FSID|XFS_AT_NODEID|\
- XFS_AT_TYPE|XFS_AT_BLKSIZE|XFS_AT_NBLOCKS|XFS_AT_VCODE|\
- XFS_AT_NEXTENTS|XFS_AT_ANEXTENTS|XFS_AT_GENCOUNT)
extern void vn_init(void);
-extern int vn_revalidate(bhv_vnode_t *);
/*
* Yeah, these don't take vnode anymore at all, all this should be
@@ -156,57 +62,52 @@ extern void vn_iowait(struct xfs_inode *ip);
extern void vn_iowake(struct xfs_inode *ip);
extern void vn_ioerror(struct xfs_inode *ip, int error, char *f, int l);
-static inline int vn_count(bhv_vnode_t *vp)
+static inline int vn_count(struct inode *vp)
{
- return atomic_read(&vn_to_inode(vp)->i_count);
+ return atomic_read(&vp->i_count);
}
-/*
- * Vnode reference counting functions (and macros for compatibility).
- */
-extern bhv_vnode_t *vn_hold(bhv_vnode_t *);
+#define IHOLD(ip) \
+do { \
+ ASSERT(atomic_read(&VFS_I(ip)->i_count) > 0) ; \
+ atomic_inc(&(VFS_I(ip)->i_count)); \
+ xfs_itrace_hold((ip), __FILE__, __LINE__, (inst_t *)__return_address); \
+} while (0)
-#if defined(XFS_INODE_TRACE)
-#define VN_HOLD(vp) \
- ((void)vn_hold(vp), \
- xfs_itrace_hold(xfs_vtoi(vp), __FILE__, __LINE__, (inst_t *)__return_address))
-#define VN_RELE(vp) \
- (xfs_itrace_rele(xfs_vtoi(vp), __FILE__, __LINE__, (inst_t *)__return_address), \
- iput(vn_to_inode(vp)))
-#else
-#define VN_HOLD(vp) ((void)vn_hold(vp))
-#define VN_RELE(vp) (iput(vn_to_inode(vp)))
-#endif
+#define IRELE(ip) \
+do { \
+ xfs_itrace_rele((ip), __FILE__, __LINE__, (inst_t *)__return_address); \
+ iput(VFS_I(ip)); \
+} while (0)
-static inline bhv_vnode_t *vn_grab(bhv_vnode_t *vp)
+static inline struct inode *vn_grab(struct inode *vp)
{
- struct inode *inode = igrab(vn_to_inode(vp));
- return inode ? vn_from_inode(inode) : NULL;
+ return igrab(vp);
}
/*
* Dealing with bad inodes
*/
-static inline int VN_BAD(bhv_vnode_t *vp)
+static inline int VN_BAD(struct inode *vp)
{
- return is_bad_inode(vn_to_inode(vp));
+ return is_bad_inode(vp);
}
/*
* Extracting atime values in various formats
*/
-static inline void vn_atime_to_bstime(bhv_vnode_t *vp, xfs_bstime_t *bs_atime)
+static inline void vn_atime_to_bstime(struct inode *vp, xfs_bstime_t *bs_atime)
{
bs_atime->tv_sec = vp->i_atime.tv_sec;
bs_atime->tv_nsec = vp->i_atime.tv_nsec;
}
-static inline void vn_atime_to_timespec(bhv_vnode_t *vp, struct timespec *ts)
+static inline void vn_atime_to_timespec(struct inode *vp, struct timespec *ts)
{
*ts = vp->i_atime;
}
-static inline void vn_atime_to_time_t(bhv_vnode_t *vp, time_t *tt)
+static inline void vn_atime_to_time_t(struct inode *vp, time_t *tt)
{
*tt = vp->i_atime.tv_sec;
}
@@ -214,20 +115,11 @@ static inline void vn_atime_to_time_t(bhv_vnode_t *vp, time_t *tt)
/*
* Some useful predicates.
*/
-#define VN_MAPPED(vp) mapping_mapped(vn_to_inode(vp)->i_mapping)
-#define VN_CACHED(vp) (vn_to_inode(vp)->i_mapping->nrpages)
-#define VN_DIRTY(vp) mapping_tagged(vn_to_inode(vp)->i_mapping, \
+#define VN_MAPPED(vp) mapping_mapped(vp->i_mapping)
+#define VN_CACHED(vp) (vp->i_mapping->nrpages)
+#define VN_DIRTY(vp) mapping_tagged(vp->i_mapping, \
PAGECACHE_TAG_DIRTY)
-/*
- * Flags to vop_setattr/getattr.
- */
-#define ATTR_UTIME 0x01 /* non-default utime(2) request */
-#define ATTR_DMI 0x08 /* invocation from a DMI function */
-#define ATTR_LAZY 0x80 /* set/get attributes lazily */
-#define ATTR_NONBLOCK 0x100 /* return EAGAIN if operation would block */
-#define ATTR_NOLOCK 0x200 /* Don't grab any conflicting locks */
-#define ATTR_NOSIZETOK 0x400 /* Don't get the SIZE token */
/*
* Tracking vnode activity.
diff --git a/fs/xfs/linux-2.6/xfs_xattr.c b/fs/xfs/linux-2.6/xfs_xattr.c
new file mode 100644
index 00000000000..964621fde6e
--- /dev/null
+++ b/fs/xfs/linux-2.6/xfs_xattr.c
@@ -0,0 +1,330 @@
+/*
+ * Copyright (C) 2008 Christoph Hellwig.
+ * Portions Copyright (C) 2000-2008 Silicon Graphics, Inc.
+ *
+ * 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_da_btree.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_inode.h"
+#include "xfs_attr.h"
+#include "xfs_attr_leaf.h"
+#include "xfs_acl.h"
+#include "xfs_vnodeops.h"
+
+#include <linux/posix_acl_xattr.h>
+#include <linux/xattr.h>
+
+
+/*
+ * ACL handling. Should eventually be moved into xfs_acl.c
+ */
+
+static int
+xfs_decode_acl(const char *name)
+{
+ if (strcmp(name, "posix_acl_access") == 0)
+ return _ACL_TYPE_ACCESS;
+ else if (strcmp(name, "posix_acl_default") == 0)
+ return _ACL_TYPE_DEFAULT;
+ return -EINVAL;
+}
+
+/*
+ * Get system extended attributes which at the moment only
+ * includes Posix ACLs.
+ */
+static int
+xfs_xattr_system_get(struct inode *inode, const char *name,
+ void *buffer, size_t size)
+{
+ int acl;
+
+ acl = xfs_decode_acl(name);
+ if (acl < 0)
+ return acl;
+
+ return xfs_acl_vget(inode, buffer, size, acl);
+}
+
+static int
+xfs_xattr_system_set(struct inode *inode, const char *name,
+ const void *value, size_t size, int flags)
+{
+ int acl;
+
+ acl = xfs_decode_acl(name);
+ if (acl < 0)
+ return acl;
+ if (flags & XATTR_CREATE)
+ return -EINVAL;
+
+ if (!value)
+ return xfs_acl_vremove(inode, acl);
+
+ return xfs_acl_vset(inode, (void *)value, size, acl);
+}
+
+static struct xattr_handler xfs_xattr_system_handler = {
+ .prefix = XATTR_SYSTEM_PREFIX,
+ .get = xfs_xattr_system_get,
+ .set = xfs_xattr_system_set,
+};
+
+
+/*
+ * Real xattr handling. The only difference between the namespaces is
+ * a flag passed to the low-level attr code.
+ */
+
+static int
+__xfs_xattr_get(struct inode *inode, const char *name,
+ void *value, size_t size, int xflags)
+{
+ struct xfs_inode *ip = XFS_I(inode);
+ int error, asize = size;
+
+ if (strcmp(name, "") == 0)
+ return -EINVAL;
+
+ /* Convert Linux syscall to XFS internal ATTR flags */
+ if (!size) {
+ xflags |= ATTR_KERNOVAL;
+ value = NULL;
+ }
+
+ error = -xfs_attr_get(ip, name, value, &asize, xflags);
+ if (error)
+ return error;
+ return asize;
+}
+
+static int
+__xfs_xattr_set(struct inode *inode, const char *name, const void *value,
+ size_t size, int flags, int xflags)
+{
+ struct xfs_inode *ip = XFS_I(inode);
+
+ if (strcmp(name, "") == 0)
+ return -EINVAL;
+
+ /* Convert Linux syscall to XFS internal ATTR flags */
+ if (flags & XATTR_CREATE)
+ xflags |= ATTR_CREATE;
+ if (flags & XATTR_REPLACE)
+ xflags |= ATTR_REPLACE;
+
+ if (!value)
+ return -xfs_attr_remove(ip, name, xflags);
+ return -xfs_attr_set(ip, name, (void *)value, size, xflags);
+}
+
+static int
+xfs_xattr_user_get(struct inode *inode, const char *name,
+ void *value, size_t size)
+{
+ return __xfs_xattr_get(inode, name, value, size, 0);
+}
+
+static int
+xfs_xattr_user_set(struct inode *inode, const char *name,
+ const void *value, size_t size, int flags)
+{
+ return __xfs_xattr_set(inode, name, value, size, flags, 0);
+}
+
+static struct xattr_handler xfs_xattr_user_handler = {
+ .prefix = XATTR_USER_PREFIX,
+ .get = xfs_xattr_user_get,
+ .set = xfs_xattr_user_set,
+};
+
+
+static int
+xfs_xattr_trusted_get(struct inode *inode, const char *name,
+ void *value, size_t size)
+{
+ return __xfs_xattr_get(inode, name, value, size, ATTR_ROOT);
+}
+
+static int
+xfs_xattr_trusted_set(struct inode *inode, const char *name,
+ const void *value, size_t size, int flags)
+{
+ return __xfs_xattr_set(inode, name, value, size, flags, ATTR_ROOT);
+}
+
+static struct xattr_handler xfs_xattr_trusted_handler = {
+ .prefix = XATTR_TRUSTED_PREFIX,
+ .get = xfs_xattr_trusted_get,
+ .set = xfs_xattr_trusted_set,
+};
+
+
+static int
+xfs_xattr_secure_get(struct inode *inode, const char *name,
+ void *value, size_t size)
+{
+ return __xfs_xattr_get(inode, name, value, size, ATTR_SECURE);
+}
+
+static int
+xfs_xattr_secure_set(struct inode *inode, const char *name,
+ const void *value, size_t size, int flags)
+{
+ return __xfs_xattr_set(inode, name, value, size, flags, ATTR_SECURE);
+}
+
+static struct xattr_handler xfs_xattr_security_handler = {
+ .prefix = XATTR_SECURITY_PREFIX,
+ .get = xfs_xattr_secure_get,
+ .set = xfs_xattr_secure_set,
+};
+
+
+struct xattr_handler *xfs_xattr_handlers[] = {
+ &xfs_xattr_user_handler,
+ &xfs_xattr_trusted_handler,
+ &xfs_xattr_security_handler,
+ &xfs_xattr_system_handler,
+ NULL
+};
+
+static unsigned int xfs_xattr_prefix_len(int flags)
+{
+ if (flags & XFS_ATTR_SECURE)
+ return sizeof("security");
+ else if (flags & XFS_ATTR_ROOT)
+ return sizeof("trusted");
+ else
+ return sizeof("user");
+}
+
+static const char *xfs_xattr_prefix(int flags)
+{
+ if (flags & XFS_ATTR_SECURE)
+ return xfs_xattr_security_handler.prefix;
+ else if (flags & XFS_ATTR_ROOT)
+ return xfs_xattr_trusted_handler.prefix;
+ else
+ return xfs_xattr_user_handler.prefix;
+}
+
+static int
+xfs_xattr_put_listent(struct xfs_attr_list_context *context, int flags,
+ char *name, int namelen, int valuelen, char *value)
+{
+ unsigned int prefix_len = xfs_xattr_prefix_len(flags);
+ char *offset;
+ int arraytop;
+
+ ASSERT(context->count >= 0);
+
+ /*
+ * Only show root namespace entries if we are actually allowed to
+ * see them.
+ */
+ if ((flags & XFS_ATTR_ROOT) && !capable(CAP_SYS_ADMIN))
+ return 0;
+
+ arraytop = context->count + prefix_len + namelen + 1;
+ if (arraytop > context->firstu) {
+ context->count = -1; /* insufficient space */
+ return 1;
+ }
+ offset = (char *)context->alist + context->count;
+ strncpy(offset, xfs_xattr_prefix(flags), prefix_len);
+ offset += prefix_len;
+ strncpy(offset, name, namelen); /* real name */
+ offset += namelen;
+ *offset = '\0';
+ context->count += prefix_len + namelen + 1;
+ return 0;
+}
+
+static int
+xfs_xattr_put_listent_sizes(struct xfs_attr_list_context *context, int flags,
+ char *name, int namelen, int valuelen, char *value)
+{
+ context->count += xfs_xattr_prefix_len(flags) + namelen + 1;
+ return 0;
+}
+
+static int
+list_one_attr(const char *name, const size_t len, void *data,
+ size_t size, ssize_t *result)
+{
+ char *p = data + *result;
+
+ *result += len;
+ if (!size)
+ return 0;
+ if (*result > size)
+ return -ERANGE;
+
+ strcpy(p, name);
+ return 0;
+}
+
+ssize_t
+xfs_vn_listxattr(struct dentry *dentry, char *data, size_t size)
+{
+ struct xfs_attr_list_context context;
+ struct attrlist_cursor_kern cursor = { 0 };
+ struct inode *inode = dentry->d_inode;
+ int error;
+
+ /*
+ * First read the regular on-disk attributes.
+ */
+ memset(&context, 0, sizeof(context));
+ context.dp = XFS_I(inode);
+ context.cursor = &cursor;
+ context.resynch = 1;
+ context.alist = data;
+ context.bufsize = size;
+ context.firstu = context.bufsize;
+
+ if (size)
+ context.put_listent = xfs_xattr_put_listent;
+ else
+ context.put_listent = xfs_xattr_put_listent_sizes;
+
+ xfs_attr_list_int(&context);
+ if (context.count < 0)
+ return -ERANGE;
+
+ /*
+ * Then add the two synthetic ACL attributes.
+ */
+ if (xfs_acl_vhasacl_access(inode)) {
+ error = list_one_attr(POSIX_ACL_XATTR_ACCESS,
+ strlen(POSIX_ACL_XATTR_ACCESS) + 1,
+ data, size, &context.count);
+ if (error)
+ return error;
+ }
+
+ if (xfs_acl_vhasacl_default(inode)) {
+ error = list_one_attr(POSIX_ACL_XATTR_DEFAULT,
+ strlen(POSIX_ACL_XATTR_DEFAULT) + 1,
+ data, size, &context.count);
+ if (error)
+ return error;
+ }
+
+ return context.count;
+}
diff --git a/fs/xfs/quota/xfs_dquot.c b/fs/xfs/quota/xfs_dquot.c
index 85df3288efd..f2705f2fd43 100644
--- a/fs/xfs/quota/xfs_dquot.c
+++ b/fs/xfs/quota/xfs_dquot.c
@@ -101,11 +101,18 @@ xfs_qm_dqinit(
if (brandnewdquot) {
dqp->dq_flnext = dqp->dq_flprev = dqp;
mutex_init(&dqp->q_qlock);
- initnsema(&dqp->q_flock, 1, "fdq");
sv_init(&dqp->q_pinwait, SV_DEFAULT, "pdq");
+ /*
+ * Because we want to use a counting completion, complete
+ * the flush completion once to allow a single access to
+ * the flush completion without blocking.
+ */
+ init_completion(&dqp->q_flush);
+ complete(&dqp->q_flush);
+
#ifdef XFS_DQUOT_TRACE
- dqp->q_trace = ktrace_alloc(DQUOT_TRACE_SIZE, KM_SLEEP);
+ dqp->q_trace = ktrace_alloc(DQUOT_TRACE_SIZE, KM_NOFS);
xfs_dqtrace_entry(dqp, "DQINIT");
#endif
} else {
@@ -150,7 +157,6 @@ xfs_qm_dqdestroy(
ASSERT(! XFS_DQ_IS_ON_FREELIST(dqp));
mutex_destroy(&dqp->q_qlock);
- freesema(&dqp->q_flock);
sv_destroy(&dqp->q_pinwait);
#ifdef XFS_DQUOT_TRACE
@@ -431,7 +437,7 @@ xfs_qm_dqalloc(
* when it unlocks the inode. Since we want to keep the quota
* inode around, we bump the vnode ref count now.
*/
- VN_HOLD(XFS_ITOV(quotip));
+ IHOLD(quotip);
xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL);
nmaps = 1;
@@ -1211,7 +1217,7 @@ xfs_qm_dqflush(
int error;
ASSERT(XFS_DQ_IS_LOCKED(dqp));
- ASSERT(XFS_DQ_IS_FLUSH_LOCKED(dqp));
+ ASSERT(!completion_done(&dqp->q_flush));
xfs_dqtrace_entry(dqp, "DQFLUSH");
/*
@@ -1348,34 +1354,18 @@ xfs_qm_dqflush_done(
xfs_dqfunlock(dqp);
}
-
-int
-xfs_qm_dqflock_nowait(
- xfs_dquot_t *dqp)
-{
- int locked;
-
- locked = cpsema(&((dqp)->q_flock));
-
- /* XXX ifdef these out */
- if (locked)
- (dqp)->dq_flags |= XFS_DQ_FLOCKED;
- return (locked);
-}
-
-
int
xfs_qm_dqlock_nowait(
xfs_dquot_t *dqp)
{
- return (mutex_trylock(&((dqp)->q_qlock)));
+ return mutex_trylock(&dqp->q_qlock);
}
void
xfs_dqlock(
xfs_dquot_t *dqp)
{
- mutex_lock(&(dqp->q_qlock));
+ mutex_lock(&dqp->q_qlock);
}
void
@@ -1435,8 +1425,7 @@ xfs_dqlock2(
/* ARGSUSED */
int
xfs_qm_dqpurge(
- xfs_dquot_t *dqp,
- uint flags)
+ xfs_dquot_t *dqp)
{
xfs_dqhash_t *thishash;
xfs_mount_t *mp = dqp->q_mount;
@@ -1469,7 +1458,7 @@ xfs_qm_dqpurge(
* if we're turning off quotas. Basically, we need this flush
* lock, and are willing to block on it.
*/
- if (! xfs_qm_dqflock_nowait(dqp)) {
+ if (!xfs_dqflock_nowait(dqp)) {
/*
* Block on the flush lock after nudging dquot buffer,
* if it is incore.
diff --git a/fs/xfs/quota/xfs_dquot.h b/fs/xfs/quota/xfs_dquot.h
index 5c371a92e3e..8958d0faf8d 100644
--- a/fs/xfs/quota/xfs_dquot.h
+++ b/fs/xfs/quota/xfs_dquot.h
@@ -82,7 +82,7 @@ typedef struct xfs_dquot {
xfs_qcnt_t q_res_icount; /* total inos allocd+reserved */
xfs_qcnt_t q_res_rtbcount;/* total realtime blks used+reserved */
mutex_t q_qlock; /* quota lock */
- sema_t q_flock; /* flush lock */
+ struct completion q_flush; /* flush completion queue */
uint q_pincount; /* pin count for this dquot */
sv_t q_pinwait; /* sync var for pinning */
#ifdef XFS_DQUOT_TRACE
@@ -113,17 +113,25 @@ XFS_DQ_IS_LOCKED(xfs_dquot_t *dqp)
/*
- * The following three routines simply manage the q_flock
- * semaphore embedded in the dquot. This semaphore synchronizes
- * processes attempting to flush the in-core dquot back to disk.
+ * Manage the q_flush completion queue embedded in the dquot. This completion
+ * queue synchronizes processes attempting to flush the in-core dquot back to
+ * disk.
*/
-#define xfs_dqflock(dqp) { psema(&((dqp)->q_flock), PINOD | PRECALC);\
- (dqp)->dq_flags |= XFS_DQ_FLOCKED; }
-#define xfs_dqfunlock(dqp) { ASSERT(issemalocked(&((dqp)->q_flock))); \
- vsema(&((dqp)->q_flock)); \
- (dqp)->dq_flags &= ~(XFS_DQ_FLOCKED); }
+static inline void xfs_dqflock(xfs_dquot_t *dqp)
+{
+ wait_for_completion(&dqp->q_flush);
+}
+
+static inline int xfs_dqflock_nowait(xfs_dquot_t *dqp)
+{
+ return try_wait_for_completion(&dqp->q_flush);
+}
+
+static inline void xfs_dqfunlock(xfs_dquot_t *dqp)
+{
+ complete(&dqp->q_flush);
+}
-#define XFS_DQ_IS_FLUSH_LOCKED(dqp) (issemalocked(&((dqp)->q_flock)))
#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)
@@ -164,10 +172,9 @@ extern void xfs_qm_dqprint(xfs_dquot_t *);
extern void xfs_qm_dqdestroy(xfs_dquot_t *);
extern int xfs_qm_dqflush(xfs_dquot_t *, uint);
-extern int xfs_qm_dqpurge(xfs_dquot_t *, uint);
+extern int xfs_qm_dqpurge(xfs_dquot_t *);
extern void xfs_qm_dqunpin_wait(xfs_dquot_t *);
extern int xfs_qm_dqlock_nowait(xfs_dquot_t *);
-extern int xfs_qm_dqflock_nowait(xfs_dquot_t *);
extern void xfs_qm_dqflock_pushbuf_wait(xfs_dquot_t *dqp);
extern void xfs_qm_adjust_dqtimers(xfs_mount_t *,
xfs_disk_dquot_t *);
diff --git a/fs/xfs/quota/xfs_dquot_item.c b/fs/xfs/quota/xfs_dquot_item.c
index 36e05ca7841..f028644caa5 100644
--- a/fs/xfs/quota/xfs_dquot_item.c
+++ b/fs/xfs/quota/xfs_dquot_item.c
@@ -151,7 +151,7 @@ xfs_qm_dquot_logitem_push(
dqp = logitem->qli_dquot;
ASSERT(XFS_DQ_IS_LOCKED(dqp));
- ASSERT(XFS_DQ_IS_FLUSH_LOCKED(dqp));
+ ASSERT(!completion_done(&dqp->q_flush));
/*
* Since we were able to lock the dquot's flush lock and
@@ -245,7 +245,7 @@ xfs_qm_dquot_logitem_pushbuf(
* inode flush completed and the inode was taken off the AIL.
* So, just get out.
*/
- if (!issemalocked(&(dqp->q_flock)) ||
+ if (completion_done(&dqp->q_flush) ||
((qip->qli_item.li_flags & XFS_LI_IN_AIL) == 0)) {
qip->qli_pushbuf_flag = 0;
xfs_dqunlock(dqp);
@@ -258,7 +258,7 @@ xfs_qm_dquot_logitem_pushbuf(
if (bp != NULL) {
if (XFS_BUF_ISDELAYWRITE(bp)) {
dopush = ((qip->qli_item.li_flags & XFS_LI_IN_AIL) &&
- issemalocked(&(dqp->q_flock)));
+ !completion_done(&dqp->q_flush));
qip->qli_pushbuf_flag = 0;
xfs_dqunlock(dqp);
@@ -317,7 +317,7 @@ xfs_qm_dquot_logitem_trylock(
return (XFS_ITEM_LOCKED);
retval = XFS_ITEM_SUCCESS;
- if (! xfs_qm_dqflock_nowait(dqp)) {
+ 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
@@ -576,8 +576,8 @@ xfs_qm_qoffend_logitem_committed(
* xfs_trans_delete_ail() drops the AIL lock.
*/
xfs_trans_delete_ail(qfs->qql_item.li_mountp, (xfs_log_item_t *)qfs);
- kmem_free(qfs, sizeof(xfs_qoff_logitem_t));
- kmem_free(qfe, sizeof(xfs_qoff_logitem_t));
+ kmem_free(qfs);
+ kmem_free(qfe);
return (xfs_lsn_t)-1;
}
diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c
index d31cce1165c..df0ffef9775 100644
--- a/fs/xfs/quota/xfs_qm.c
+++ b/fs/xfs/quota/xfs_qm.c
@@ -192,8 +192,8 @@ 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, hsize * sizeof(xfs_dqhash_t));
- kmem_free(xqm->qm_grp_dqhtable, hsize * sizeof(xfs_dqhash_t));
+ kmem_free(xqm->qm_usr_dqhtable);
+ kmem_free(xqm->qm_grp_dqhtable);
xqm->qm_usr_dqhtable = NULL;
xqm->qm_grp_dqhtable = NULL;
xqm->qm_dqhashmask = 0;
@@ -201,7 +201,7 @@ xfs_qm_destroy(
#ifdef DEBUG
mutex_destroy(&qcheck_lock);
#endif
- kmem_free(xqm, sizeof(xfs_qm_t));
+ kmem_free(xqm);
}
/*
@@ -310,8 +310,7 @@ xfs_qm_unmount_quotadestroy(
*/
void
xfs_qm_mount_quotas(
- xfs_mount_t *mp,
- int mfsi_flags)
+ xfs_mount_t *mp)
{
int error = 0;
uint sbf;
@@ -346,8 +345,7 @@ xfs_qm_mount_quotas(
/*
* If any of the quotas are not consistent, do a quotacheck.
*/
- if (XFS_QM_NEED_QUOTACHECK(mp) &&
- !(mfsi_flags & XFS_MFSI_NO_QUOTACHECK)) {
+ if (XFS_QM_NEED_QUOTACHECK(mp)) {
error = xfs_qm_quotacheck(mp);
if (error) {
/* Quotacheck failed and disabled quotas. */
@@ -445,11 +443,11 @@ xfs_qm_unmount_quotas(
}
}
if (uqp) {
- XFS_PURGE_INODE(uqp);
+ IRELE(uqp);
mp->m_quotainfo->qi_uquotaip = NULL;
}
if (gqp) {
- XFS_PURGE_INODE(gqp);
+ IRELE(gqp);
mp->m_quotainfo->qi_gquotaip = NULL;
}
out:
@@ -484,7 +482,7 @@ again:
xfs_dqtrace_entry(dqp, "FLUSHALL: DQDIRTY");
/* XXX a sentinel would be better */
recl = XFS_QI_MPLRECLAIMS(mp);
- if (! xfs_qm_dqflock_nowait(dqp)) {
+ if (!xfs_dqflock_nowait(dqp)) {
/*
* If we can't grab the flush lock then check
* to see if the dquot has been flushed delayed
@@ -631,7 +629,7 @@ xfs_qm_dqpurge_int(
* freelist in INACTIVE state.
*/
nextdqp = dqp->MPL_NEXT;
- nmisses += xfs_qm_dqpurge(dqp, flags);
+ nmisses += xfs_qm_dqpurge(dqp);
dqp = nextdqp;
}
xfs_qm_mplist_unlock(mp);
@@ -1062,7 +1060,7 @@ xfs_qm_sync(
/* XXX a sentinel would be better */
recl = XFS_QI_MPLRECLAIMS(mp);
- if (! xfs_qm_dqflock_nowait(dqp)) {
+ if (!xfs_dqflock_nowait(dqp)) {
if (nowait) {
xfs_dqunlock(dqp);
continue;
@@ -1134,7 +1132,7 @@ xfs_qm_init_quotainfo(
* and change the superblock accordingly.
*/
if ((error = xfs_qm_init_quotainos(mp))) {
- kmem_free(qinf, sizeof(xfs_quotainfo_t));
+ kmem_free(qinf);
mp->m_quotainfo = NULL;
return error;
}
@@ -1240,15 +1238,15 @@ xfs_qm_destroy_quotainfo(
xfs_qm_list_destroy(&qi->qi_dqlist);
if (qi->qi_uquotaip) {
- XFS_PURGE_INODE(qi->qi_uquotaip);
+ IRELE(qi->qi_uquotaip);
qi->qi_uquotaip = NULL; /* paranoia */
}
if (qi->qi_gquotaip) {
- XFS_PURGE_INODE(qi->qi_gquotaip);
+ IRELE(qi->qi_gquotaip);
qi->qi_gquotaip = NULL;
}
mutex_destroy(&qi->qi_quotaofflock);
- kmem_free(qi, sizeof(xfs_quotainfo_t));
+ kmem_free(qi);
mp->m_quotainfo = NULL;
}
@@ -1394,7 +1392,7 @@ xfs_qm_qino_alloc(
* locked exclusively and joined to the transaction already.
*/
ASSERT(xfs_isilocked(*ip, XFS_ILOCK_EXCL));
- VN_HOLD(XFS_ITOV((*ip)));
+ IHOLD(*ip);
/*
* Make the changes in the superblock, and log those too.
@@ -1623,7 +1621,7 @@ xfs_qm_dqiterate(
break;
} while (nmaps > 0);
- kmem_free(map, XFS_DQITER_MAP_SIZE * sizeof(*map));
+ kmem_free(map);
return error;
}
@@ -2079,7 +2077,7 @@ xfs_qm_shake_freelist(
* 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_qm_dqflock_nowait(dqp)) {
+ if (!xfs_dqflock_nowait(dqp)) {
xfs_dqunlock(dqp);
dqp = dqp->dq_flnext;
continue;
@@ -2257,7 +2255,7 @@ xfs_qm_dqreclaim_one(void)
* 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_qm_dqflock_nowait(dqp)) {
+ if (!xfs_dqflock_nowait(dqp)) {
xfs_dqunlock(dqp);
continue;
}
diff --git a/fs/xfs/quota/xfs_qm.h b/fs/xfs/quota/xfs_qm.h
index cd2300e374a..44f25349e47 100644
--- a/fs/xfs/quota/xfs_qm.h
+++ b/fs/xfs/quota/xfs_qm.h
@@ -165,7 +165,7 @@ typedef struct xfs_dquot_acct {
#define XFS_QM_RELE(xqm) ((xqm)->qm_nrefs--)
extern void xfs_qm_destroy_quotainfo(xfs_mount_t *);
-extern void xfs_qm_mount_quotas(xfs_mount_t *, int);
+extern void xfs_qm_mount_quotas(xfs_mount_t *);
extern int xfs_qm_quotacheck(xfs_mount_t *);
extern void xfs_qm_unmount_quotadestroy(xfs_mount_t *);
extern int xfs_qm_unmount_quotas(xfs_mount_t *);
diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c
index f4f6c4c861d..eea2e60b456 100644
--- a/fs/xfs/quota/xfs_qm_bhv.c
+++ b/fs/xfs/quota/xfs_qm_bhv.c
@@ -162,7 +162,7 @@ xfs_qm_newmount(
* mounting, and get on with the boring life
* without disk quotas.
*/
- xfs_qm_mount_quotas(mp, 0);
+ xfs_qm_mount_quotas(mp);
} else {
/*
* Clear the quota flags, but remember them. This
@@ -184,13 +184,12 @@ STATIC int
xfs_qm_endmount(
xfs_mount_t *mp,
uint needquotamount,
- uint quotaflags,
- int mfsi_flags)
+ uint quotaflags)
{
if (needquotamount) {
ASSERT(mp->m_qflags == 0);
mp->m_qflags = quotaflags;
- xfs_qm_mount_quotas(mp, mfsi_flags);
+ xfs_qm_mount_quotas(mp);
}
#if defined(DEBUG) && defined(XFS_LOUD_RECOVERY)
diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c
index 768a3b27d2b..1a3b803dfa5 100644
--- a/fs/xfs/quota/xfs_qm_syscalls.c
+++ b/fs/xfs/quota/xfs_qm_syscalls.c
@@ -362,11 +362,11 @@ xfs_qm_scall_quotaoff(
* if we don't need them anymore.
*/
if ((dqtype & XFS_QMOPT_UQUOTA) && XFS_QI_UQIP(mp)) {
- XFS_PURGE_INODE(XFS_QI_UQIP(mp));
+ IRELE(XFS_QI_UQIP(mp));
XFS_QI_UQIP(mp) = NULL;
}
if ((dqtype & (XFS_QMOPT_GQUOTA|XFS_QMOPT_PQUOTA)) && XFS_QI_GQIP(mp)) {
- XFS_PURGE_INODE(XFS_QI_GQIP(mp));
+ IRELE(XFS_QI_GQIP(mp));
XFS_QI_GQIP(mp) = NULL;
}
out_error:
@@ -1034,7 +1034,7 @@ xfs_qm_dqrele_all_inodes(
{
xfs_inode_t *ip, *topino;
uint ireclaims;
- bhv_vnode_t *vp;
+ struct inode *vp;
boolean_t vnode_refd;
ASSERT(mp->m_quotainfo);
@@ -1059,7 +1059,7 @@ again:
ip = ip->i_mnext;
continue;
}
- vp = XFS_ITOV_NULL(ip);
+ vp = VFS_I(ip);
if (!vp) {
ASSERT(ip->i_udquot == NULL);
ASSERT(ip->i_gdquot == NULL);
@@ -1449,14 +1449,14 @@ xfs_qm_internalqcheck(
for (d = (xfs_dqtest_t *) h1->qh_next; d != NULL; ) {
xfs_dqtest_cmp(d);
e = (xfs_dqtest_t *) d->HL_NEXT;
- kmem_free(d, sizeof(xfs_dqtest_t));
+ kmem_free(d);
d = e;
}
h1 = &qmtest_gdqtab[i];
for (d = (xfs_dqtest_t *) h1->qh_next; d != NULL; ) {
xfs_dqtest_cmp(d);
e = (xfs_dqtest_t *) d->HL_NEXT;
- kmem_free(d, sizeof(xfs_dqtest_t));
+ kmem_free(d);
d = e;
}
}
@@ -1467,8 +1467,8 @@ xfs_qm_internalqcheck(
} else {
cmn_err(CE_DEBUG, "******** quotacheck successful! ********");
}
- kmem_free(qmtest_udqtab, qmtest_hashmask * sizeof(xfs_dqhash_t));
- kmem_free(qmtest_gdqtab, qmtest_hashmask * sizeof(xfs_dqhash_t));
+ kmem_free(qmtest_udqtab);
+ kmem_free(qmtest_gdqtab);
mutex_unlock(&qcheck_lock);
return (qmtest_nfails);
}
diff --git a/fs/xfs/quota/xfs_quota_priv.h b/fs/xfs/quota/xfs_quota_priv.h
index 5e4a40b1c56..c4fcea600bc 100644
--- a/fs/xfs/quota/xfs_quota_priv.h
+++ b/fs/xfs/quota/xfs_quota_priv.h
@@ -158,9 +158,6 @@ for ((dqp) = (qlist)->qh_next; (dqp) != (xfs_dquot_t *)(qlist); \
#define XFS_IS_SUSER_DQUOT(dqp) \
(!((dqp)->q_core.d_id))
-#define XFS_PURGE_INODE(ip) \
- IRELE(ip);
-
#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/support/ktrace.c b/fs/xfs/support/ktrace.c
index 0b75d302508..a34ef05489b 100644
--- a/fs/xfs/support/ktrace.c
+++ b/fs/xfs/support/ktrace.c
@@ -89,7 +89,7 @@ ktrace_alloc(int nentries, unsigned int __nocast sleep)
if (sleep & KM_SLEEP)
panic("ktrace_alloc: NULL memory on KM_SLEEP request!");
- kmem_free(ktp, sizeof(*ktp));
+ kmem_free(ktp);
return NULL;
}
@@ -126,7 +126,7 @@ ktrace_free(ktrace_t *ktp)
} else {
entries_size = (int)(ktp->kt_nentries * sizeof(ktrace_entry_t));
- kmem_free(ktp->kt_entries, entries_size);
+ kmem_free(ktp->kt_entries);
}
kmem_zone_free(ktrace_hdr_zone, ktp);
diff --git a/fs/xfs/support/uuid.c b/fs/xfs/support/uuid.c
index 493a6ecf859..5830c040ea7 100644
--- a/fs/xfs/support/uuid.c
+++ b/fs/xfs/support/uuid.c
@@ -17,7 +17,7 @@
*/
#include <xfs.h>
-static mutex_t uuid_monitor;
+static DEFINE_MUTEX(uuid_monitor);
static int uuid_table_size;
static uuid_t *uuid_table;
@@ -132,9 +132,3 @@ uuid_table_remove(uuid_t *uuid)
ASSERT(i < uuid_table_size);
mutex_unlock(&uuid_monitor);
}
-
-void __init
-uuid_init(void)
-{
- mutex_init(&uuid_monitor);
-}
diff --git a/fs/xfs/support/uuid.h b/fs/xfs/support/uuid.h
index b6f5922199b..cff5b607d44 100644
--- a/fs/xfs/support/uuid.h
+++ b/fs/xfs/support/uuid.h
@@ -22,7 +22,6 @@ typedef struct {
unsigned char __u_bits[16];
} uuid_t;
-extern void uuid_init(void);
extern void uuid_create_nil(uuid_t *uuid);
extern int uuid_is_nil(uuid_t *uuid);
extern int uuid_equal(uuid_t *uuid1, uuid_t *uuid2);
diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c
index ebee3a4f703..b2f639a1416 100644
--- a/fs/xfs/xfs_acl.c
+++ b/fs/xfs/xfs_acl.c
@@ -37,15 +37,15 @@
#include <linux/capability.h>
#include <linux/posix_acl_xattr.h>
-STATIC int xfs_acl_setmode(bhv_vnode_t *, xfs_acl_t *, int *);
+STATIC int xfs_acl_setmode(struct inode *, xfs_acl_t *, int *);
STATIC void xfs_acl_filter_mode(mode_t, xfs_acl_t *);
STATIC void xfs_acl_get_endian(xfs_acl_t *);
STATIC int xfs_acl_access(uid_t, gid_t, xfs_acl_t *, mode_t, cred_t *);
STATIC int xfs_acl_invalid(xfs_acl_t *);
STATIC void xfs_acl_sync_mode(mode_t, xfs_acl_t *);
-STATIC void xfs_acl_get_attr(bhv_vnode_t *, xfs_acl_t *, int, int, int *);
-STATIC void xfs_acl_set_attr(bhv_vnode_t *, xfs_acl_t *, int, int *);
-STATIC int xfs_acl_allow_set(bhv_vnode_t *, int);
+STATIC void xfs_acl_get_attr(struct inode *, xfs_acl_t *, int, int, int *);
+STATIC void xfs_acl_set_attr(struct inode *, xfs_acl_t *, int, int *);
+STATIC int xfs_acl_allow_set(struct inode *, int);
kmem_zone_t *xfs_acl_zone;
@@ -55,7 +55,7 @@ kmem_zone_t *xfs_acl_zone;
*/
int
xfs_acl_vhasacl_access(
- bhv_vnode_t *vp)
+ struct inode *vp)
{
int error;
@@ -68,7 +68,7 @@ xfs_acl_vhasacl_access(
*/
int
xfs_acl_vhasacl_default(
- bhv_vnode_t *vp)
+ struct inode *vp)
{
int error;
@@ -207,7 +207,7 @@ posix_acl_xfs_to_xattr(
int
xfs_acl_vget(
- bhv_vnode_t *vp,
+ struct inode *vp,
void *acl,
size_t size,
int kind)
@@ -217,7 +217,6 @@ xfs_acl_vget(
posix_acl_xattr_header *ext_acl = acl;
int flags = 0;
- VN_HOLD(vp);
if(size) {
if (!(_ACL_ALLOC(xfs_acl))) {
error = ENOMEM;
@@ -239,11 +238,10 @@ xfs_acl_vget(
goto out;
}
if (kind == _ACL_TYPE_ACCESS)
- xfs_acl_sync_mode(xfs_vtoi(vp)->i_d.di_mode, xfs_acl);
+ xfs_acl_sync_mode(XFS_I(vp)->i_d.di_mode, xfs_acl);
error = -posix_acl_xfs_to_xattr(xfs_acl, ext_acl, size);
}
out:
- VN_RELE(vp);
if(xfs_acl)
_ACL_FREE(xfs_acl);
return -error;
@@ -251,28 +249,26 @@ out:
int
xfs_acl_vremove(
- bhv_vnode_t *vp,
+ struct inode *vp,
int kind)
{
int error;
- VN_HOLD(vp);
error = xfs_acl_allow_set(vp, kind);
if (!error) {
- error = xfs_attr_remove(xfs_vtoi(vp),
+ error = xfs_attr_remove(XFS_I(vp),
kind == _ACL_TYPE_DEFAULT?
SGI_ACL_DEFAULT: SGI_ACL_FILE,
ATTR_ROOT);
if (error == ENOATTR)
error = 0; /* 'scool */
}
- VN_RELE(vp);
return -error;
}
int
xfs_acl_vset(
- bhv_vnode_t *vp,
+ struct inode *vp,
void *acl,
size_t size,
int kind)
@@ -298,7 +294,6 @@ xfs_acl_vset(
return 0;
}
- VN_HOLD(vp);
error = xfs_acl_allow_set(vp, kind);
/* Incoming ACL exists, set file mode based on its value */
@@ -321,7 +316,6 @@ xfs_acl_vset(
}
out:
- VN_RELE(vp);
_ACL_FREE(xfs_acl);
return -error;
}
@@ -341,8 +335,7 @@ xfs_acl_iaccess(
/* If the file has no ACL return -1. */
rval = sizeof(xfs_acl_t);
- if (xfs_attr_fetch(ip, &acl_name, (char *)acl, &rval,
- ATTR_ROOT | ATTR_KERNACCESS)) {
+ if (xfs_attr_fetch(ip, &acl_name, (char *)acl, &rval, ATTR_ROOT)) {
_ACL_FREE(acl);
return -1;
}
@@ -364,7 +357,7 @@ xfs_acl_iaccess(
STATIC int
xfs_acl_allow_set(
- bhv_vnode_t *vp,
+ struct inode *vp,
int kind)
{
if (vp->i_flags & (S_IMMUTABLE|S_APPEND))
@@ -373,7 +366,7 @@ xfs_acl_allow_set(
return ENOTDIR;
if (vp->i_sb->s_flags & MS_RDONLY)
return EROFS;
- if (xfs_vtoi(vp)->i_d.di_uid != current->fsuid && !capable(CAP_FOWNER))
+ if (XFS_I(vp)->i_d.di_uid != current->fsuid && !capable(CAP_FOWNER))
return EPERM;
return 0;
}
@@ -567,7 +560,7 @@ xfs_acl_get_endian(
*/
STATIC void
xfs_acl_get_attr(
- bhv_vnode_t *vp,
+ struct inode *vp,
xfs_acl_t *aclp,
int kind,
int flags,
@@ -577,7 +570,7 @@ xfs_acl_get_attr(
ASSERT((flags & ATTR_KERNOVAL) ? (aclp == NULL) : 1);
flags |= ATTR_ROOT;
- *error = xfs_attr_get(xfs_vtoi(vp),
+ *error = xfs_attr_get(XFS_I(vp),
kind == _ACL_TYPE_ACCESS ?
SGI_ACL_FILE : SGI_ACL_DEFAULT,
(char *)aclp, &len, flags);
@@ -591,7 +584,7 @@ xfs_acl_get_attr(
*/
STATIC void
xfs_acl_set_attr(
- bhv_vnode_t *vp,
+ struct inode *vp,
xfs_acl_t *aclp,
int kind,
int *error)
@@ -616,7 +609,7 @@ xfs_acl_set_attr(
INT_SET(newace->ae_perm, ARCH_CONVERT, ace->ae_perm);
}
INT_SET(newacl->acl_cnt, ARCH_CONVERT, aclp->acl_cnt);
- *error = xfs_attr_set(xfs_vtoi(vp),
+ *error = xfs_attr_set(XFS_I(vp),
kind == _ACL_TYPE_ACCESS ?
SGI_ACL_FILE: SGI_ACL_DEFAULT,
(char *)newacl, len, ATTR_ROOT);
@@ -625,7 +618,7 @@ xfs_acl_set_attr(
int
xfs_acl_vtoacl(
- bhv_vnode_t *vp,
+ struct inode *vp,
xfs_acl_t *access_acl,
xfs_acl_t *default_acl)
{
@@ -640,7 +633,7 @@ xfs_acl_vtoacl(
if (error)
access_acl->acl_cnt = XFS_ACL_NOT_PRESENT;
else /* We have a good ACL and the file mode, synchronize. */
- xfs_acl_sync_mode(xfs_vtoi(vp)->i_d.di_mode, access_acl);
+ xfs_acl_sync_mode(XFS_I(vp)->i_d.di_mode, access_acl);
}
if (default_acl) {
@@ -657,7 +650,7 @@ xfs_acl_vtoacl(
*/
int
xfs_acl_inherit(
- bhv_vnode_t *vp,
+ struct inode *vp,
mode_t mode,
xfs_acl_t *pdaclp)
{
@@ -716,11 +709,11 @@ out_error:
*/
STATIC int
xfs_acl_setmode(
- bhv_vnode_t *vp,
+ struct inode *vp,
xfs_acl_t *acl,
int *basicperms)
{
- bhv_vattr_t va;
+ struct iattr iattr;
xfs_acl_entry_t *ap;
xfs_acl_entry_t *gap = NULL;
int i, nomask = 1;
@@ -734,25 +727,25 @@ xfs_acl_setmode(
* Copy the u::, g::, o::, and m:: bits from the ACL into the
* mode. The m:: bits take precedence over the g:: bits.
*/
- va.va_mask = XFS_AT_MODE;
- va.va_mode = xfs_vtoi(vp)->i_d.di_mode;
- va.va_mode &= ~(S_IRWXU|S_IRWXG|S_IRWXO);
+ iattr.ia_valid = ATTR_MODE;
+ iattr.ia_mode = XFS_I(vp)->i_d.di_mode;
+ iattr.ia_mode &= ~(S_IRWXU|S_IRWXG|S_IRWXO);
ap = acl->acl_entry;
for (i = 0; i < acl->acl_cnt; ++i) {
switch (ap->ae_tag) {
case ACL_USER_OBJ:
- va.va_mode |= ap->ae_perm << 6;
+ iattr.ia_mode |= ap->ae_perm << 6;
break;
case ACL_GROUP_OBJ:
gap = ap;
break;
case ACL_MASK: /* more than just standard modes */
nomask = 0;
- va.va_mode |= ap->ae_perm << 3;
+ iattr.ia_mode |= ap->ae_perm << 3;
*basicperms = 0;
break;
case ACL_OTHER:
- va.va_mode |= ap->ae_perm;
+ iattr.ia_mode |= ap->ae_perm;
break;
default: /* more than just standard modes */
*basicperms = 0;
@@ -763,9 +756,9 @@ xfs_acl_setmode(
/* Set the group bits from ACL_GROUP_OBJ if there's no ACL_MASK */
if (gap && nomask)
- va.va_mode |= gap->ae_perm << 3;
+ iattr.ia_mode |= gap->ae_perm << 3;
- return xfs_setattr(xfs_vtoi(vp), &va, 0, sys_cred);
+ return xfs_setattr(XFS_I(vp), &iattr, 0, sys_cred);
}
/*
diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h
index 332a772461c..a4e293b93ef 100644
--- a/fs/xfs/xfs_acl.h
+++ b/fs/xfs/xfs_acl.h
@@ -46,6 +46,8 @@ typedef struct xfs_acl {
#define SGI_ACL_FILE_SIZE (sizeof(SGI_ACL_FILE)-1)
#define SGI_ACL_DEFAULT_SIZE (sizeof(SGI_ACL_DEFAULT)-1)
+#define _ACL_TYPE_ACCESS 1
+#define _ACL_TYPE_DEFAULT 2
#ifdef CONFIG_XFS_POSIX_ACL
@@ -57,17 +59,15 @@ extern struct kmem_zone *xfs_acl_zone;
(zone) = kmem_zone_init(sizeof(xfs_acl_t), (name))
#define xfs_acl_zone_destroy(zone) kmem_zone_destroy(zone)
-extern int xfs_acl_inherit(bhv_vnode_t *, mode_t mode, xfs_acl_t *);
+extern int xfs_acl_inherit(struct inode *, mode_t mode, xfs_acl_t *);
extern int xfs_acl_iaccess(struct xfs_inode *, mode_t, cred_t *);
-extern int xfs_acl_vtoacl(bhv_vnode_t *, xfs_acl_t *, xfs_acl_t *);
-extern int xfs_acl_vhasacl_access(bhv_vnode_t *);
-extern int xfs_acl_vhasacl_default(bhv_vnode_t *);
-extern int xfs_acl_vset(bhv_vnode_t *, void *, size_t, int);
-extern int xfs_acl_vget(bhv_vnode_t *, void *, size_t, int);
-extern int xfs_acl_vremove(bhv_vnode_t *, int);
+extern int xfs_acl_vtoacl(struct inode *, xfs_acl_t *, xfs_acl_t *);
+extern int xfs_acl_vhasacl_access(struct inode *);
+extern int xfs_acl_vhasacl_default(struct inode *);
+extern int xfs_acl_vset(struct inode *, void *, size_t, int);
+extern int xfs_acl_vget(struct inode *, void *, size_t, int);
+extern int xfs_acl_vremove(struct inode *, int);
-#define _ACL_TYPE_ACCESS 1
-#define _ACL_TYPE_DEFAULT 2
#define _ACL_PERM_INVALID(perm) ((perm) & ~(ACL_READ|ACL_WRITE|ACL_EXECUTE))
#define _ACL_INHERIT(c,m,d) (xfs_acl_inherit(c,m,d))
diff --git a/fs/xfs/xfs_arch.h b/fs/xfs/xfs_arch.h
index f9472a2076d..0b3b5efe848 100644
--- a/fs/xfs/xfs_arch.h
+++ b/fs/xfs/xfs_arch.h
@@ -92,16 +92,6 @@
((__u8*)(pointer))[1] = (((value) ) & 0xff); \
}
-/* define generic INT_ macros */
-
-#define INT_GET(reference,arch) \
- (((arch) == ARCH_NOCONVERT) \
- ? \
- (reference) \
- : \
- INT_SWAP((reference),(reference)) \
- )
-
/* does not return a value */
#define INT_SET(reference,arch,valueref) \
(__builtin_constant_p(valueref) ? \
@@ -112,64 +102,6 @@
) \
)
-/* does not return a value */
-#define INT_MOD_EXPR(reference,arch,code) \
- (((arch) == ARCH_NOCONVERT) \
- ? \
- (void)((reference) code) \
- : \
- (void)( \
- (reference) = INT_GET((reference),arch) , \
- ((reference) code), \
- INT_SET(reference, arch, reference) \
- ) \
- )
-
-/* does not return a value */
-#define INT_MOD(reference,arch,delta) \
- (void)( \
- INT_MOD_EXPR(reference,arch,+=(delta)) \
- )
-
-/*
- * INT_COPY - copy a value between two locations with the
- * _same architecture_ but _potentially different sizes_
- *
- * if the types of the two parameters are equal or they are
- * in native architecture, a simple copy is done
- *
- * otherwise, architecture conversions are done
- *
- */
-
-/* does not return a value */
-#define INT_COPY(dst,src,arch) \
- ( \
- ((sizeof(dst) == sizeof(src)) || ((arch) == ARCH_NOCONVERT)) \
- ? \
- (void)((dst) = (src)) \
- : \
- INT_SET(dst, arch, INT_GET(src, arch)) \
- )
-
-/*
- * INT_XLATE - copy a value in either direction between two locations
- * with different architectures
- *
- * dir < 0 - copy from memory to buffer (native to arch)
- * dir > 0 - copy from buffer to memory (arch to native)
- */
-
-/* does not return a value */
-#define INT_XLATE(buf,mem,dir,arch) {\
- ASSERT(dir); \
- if (dir>0) { \
- (mem)=INT_GET(buf, arch); \
- } else { \
- INT_SET(buf, arch, mem); \
- } \
-}
-
/*
* In directories inode numbers are stored as unaligned arrays of unsigned
* 8bit integers on disk.
diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c
index df151a85918..f7cdc28aff4 100644
--- a/fs/xfs/xfs_attr.c
+++ b/fs/xfs/xfs_attr.c
@@ -16,8 +16,6 @@
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
-#include <linux/capability.h>
-
#include "xfs.h"
#include "xfs_fs.h"
#include "xfs_types.h"
@@ -57,11 +55,6 @@
* Provide the external interfaces to manage attribute lists.
*/
-#define ATTR_SYSCOUNT 2
-static struct attrnames posix_acl_access;
-static struct attrnames posix_acl_default;
-static struct attrnames *attr_system_names[ATTR_SYSCOUNT];
-
/*========================================================================
* Function prototypes for the kernel.
*========================================================================*/
@@ -116,6 +109,17 @@ xfs_attr_name_to_xname(
return 0;
}
+STATIC int
+xfs_inode_hasattr(
+ struct xfs_inode *ip)
+{
+ if (!XFS_IFORK_Q(ip) ||
+ (ip->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS &&
+ ip->i_d.di_anextents == 0))
+ return 0;
+ return 1;
+}
+
/*========================================================================
* Overall external interface routines.
*========================================================================*/
@@ -127,10 +131,8 @@ xfs_attr_fetch(xfs_inode_t *ip, struct xfs_name *name,
xfs_da_args_t args;
int error;
- if ((XFS_IFORK_Q(ip) == 0) ||
- (ip->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS &&
- ip->i_d.di_anextents == 0))
- return(ENOATTR);
+ if (!xfs_inode_hasattr(ip))
+ return ENOATTR;
/*
* Fill in the arg structure for this request.
@@ -148,11 +150,7 @@ xfs_attr_fetch(xfs_inode_t *ip, struct xfs_name *name,
/*
* Decide on what work routines to call based on the inode size.
*/
- if (XFS_IFORK_Q(ip) == 0 ||
- (ip->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS &&
- ip->i_d.di_anextents == 0)) {
- error = XFS_ERROR(ENOATTR);
- } else if (ip->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
+ if (ip->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
error = xfs_attr_shortform_getvalue(&args);
} else if (xfs_bmap_one_block(ip, XFS_ATTR_FORK)) {
error = xfs_attr_leaf_get(&args);
@@ -196,6 +194,46 @@ xfs_attr_get(
return(error);
}
+/*
+ * Calculate how many blocks we need for the new attribute,
+ */
+int
+xfs_attr_calc_size(
+ struct xfs_inode *ip,
+ int namelen,
+ int valuelen,
+ int *local)
+{
+ struct xfs_mount *mp = ip->i_mount;
+ int size;
+ int nblks;
+
+ /*
+ * Determine space new attribute will use, and if it would be
+ * "local" or "remote" (note: local != inline).
+ */
+ size = xfs_attr_leaf_newentsize(namelen, valuelen,
+ mp->m_sb.sb_blocksize, local);
+
+ nblks = XFS_DAENTER_SPACE_RES(mp, XFS_ATTR_FORK);
+ if (*local) {
+ if (size > (mp->m_sb.sb_blocksize >> 1)) {
+ /* Double split possible */
+ nblks *= 2;
+ }
+ } else {
+ /*
+ * Out of line attribute, cannot double split, but
+ * make room for the attribute value itself.
+ */
+ uint dblocks = XFS_B_TO_FSB(mp, valuelen);
+ nblks += dblocks;
+ nblks += XFS_NEXTENTADD_SPACE_RES(mp, dblocks, XFS_ATTR_FORK);
+ }
+
+ return nblks;
+}
+
STATIC int
xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name,
char *value, int valuelen, int flags)
@@ -204,10 +242,9 @@ xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name,
xfs_fsblock_t firstblock;
xfs_bmap_free_t flist;
int error, err2, committed;
- int local, size;
- uint nblks;
xfs_mount_t *mp = dp->i_mount;
int rsvd = (flags & ATTR_ROOT) != 0;
+ int local;
/*
* Attach the dquots to the inode.
@@ -241,33 +278,10 @@ xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name,
args.firstblock = &firstblock;
args.flist = &flist;
args.whichfork = XFS_ATTR_FORK;
- args.addname = 1;
- args.oknoent = 1;
-
- /*
- * Determine space new attribute will use, and if it would be
- * "local" or "remote" (note: local != inline).
- */
- size = xfs_attr_leaf_newentsize(name->len, valuelen,
- mp->m_sb.sb_blocksize, &local);
-
- nblks = XFS_DAENTER_SPACE_RES(mp, XFS_ATTR_FORK);
- if (local) {
- if (size > (mp->m_sb.sb_blocksize >> 1)) {
- /* Double split possible */
- nblks <<= 1;
- }
- } else {
- uint dblocks = XFS_B_TO_FSB(mp, valuelen);
- /* Out of line attribute, cannot double split, but make
- * room for the attribute value itself.
- */
- nblks += dblocks;
- nblks += XFS_NEXTENTADD_SPACE_RES(mp, dblocks, XFS_ATTR_FORK);
- }
+ args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT;
/* Size is now blocks for attribute data */
- args.total = nblks;
+ args.total = xfs_attr_calc_size(dp, name->len, valuelen, &local);
/*
* Start our first transaction of the day.
@@ -289,18 +303,17 @@ xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name,
if (rsvd)
args.trans->t_flags |= XFS_TRANS_RESERVE;
- if ((error = xfs_trans_reserve(args.trans, (uint) nblks,
- XFS_ATTRSET_LOG_RES(mp, nblks),
- 0, XFS_TRANS_PERM_LOG_RES,
- XFS_ATTRSET_LOG_COUNT))) {
+ if ((error = xfs_trans_reserve(args.trans, args.total,
+ XFS_ATTRSET_LOG_RES(mp, args.total), 0,
+ XFS_TRANS_PERM_LOG_RES, XFS_ATTRSET_LOG_COUNT))) {
xfs_trans_cancel(args.trans, 0);
return(error);
}
xfs_ilock(dp, XFS_ILOCK_EXCL);
- error = XFS_TRANS_RESERVE_QUOTA_NBLKS(mp, args.trans, dp, nblks, 0,
- rsvd ? XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_FORCE_RES :
- XFS_QMOPT_RES_REGBLKS);
+ error = XFS_TRANS_RESERVE_QUOTA_NBLKS(mp, args.trans, dp, args.total, 0,
+ rsvd ? XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_FORCE_RES :
+ XFS_QMOPT_RES_REGBLKS);
if (error) {
xfs_iunlock(dp, XFS_ILOCK_EXCL);
xfs_trans_cancel(args.trans, XFS_TRANS_RELEASE_LOG_RES);
@@ -387,7 +400,9 @@ xfs_attr_set_int(xfs_inode_t *dp, struct xfs_name *name,
* Commit the leaf transformation. We'll need another (linked)
* transaction to add the new attribute to the leaf.
*/
- if ((error = xfs_attr_rolltrans(&args.trans, dp)))
+
+ error = xfs_trans_roll(&args.trans, dp);
+ if (error)
goto out;
}
@@ -529,9 +544,7 @@ xfs_attr_remove_int(xfs_inode_t *dp, struct xfs_name *name, int flags)
/*
* Decide on what work routines to call based on the inode size.
*/
- if (XFS_IFORK_Q(dp) == 0 ||
- (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS &&
- dp->i_d.di_anextents == 0)) {
+ if (!xfs_inode_hasattr(dp)) {
error = XFS_ERROR(ENOATTR);
goto out;
}
@@ -601,29 +614,33 @@ xfs_attr_remove(
return error;
xfs_ilock(dp, XFS_ILOCK_SHARED);
- if (XFS_IFORK_Q(dp) == 0 ||
- (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS &&
- dp->i_d.di_anextents == 0)) {
+ if (!xfs_inode_hasattr(dp)) {
xfs_iunlock(dp, XFS_ILOCK_SHARED);
- return(XFS_ERROR(ENOATTR));
+ return XFS_ERROR(ENOATTR);
}
xfs_iunlock(dp, XFS_ILOCK_SHARED);
return xfs_attr_remove_int(dp, &xname, flags);
}
-STATIC int
+int
xfs_attr_list_int(xfs_attr_list_context_t *context)
{
int error;
xfs_inode_t *dp = context->dp;
+ XFS_STATS_INC(xs_attr_list);
+
+ if (XFS_FORCED_SHUTDOWN(dp->i_mount))
+ return EIO;
+
+ xfs_ilock(dp, XFS_ILOCK_SHARED);
+ xfs_attr_trace_l_c("syscall start", context);
+
/*
* Decide on what work routines to call based on the inode size.
*/
- if (XFS_IFORK_Q(dp) == 0 ||
- (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS &&
- dp->i_d.di_anextents == 0)) {
+ if (!xfs_inode_hasattr(dp)) {
error = 0;
} else if (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
error = xfs_attr_shortform_list(context);
@@ -632,6 +649,10 @@ xfs_attr_list_int(xfs_attr_list_context_t *context)
} else {
error = xfs_attr_node_list(context);
}
+
+ xfs_iunlock(dp, XFS_ILOCK_SHARED);
+ xfs_attr_trace_l_c("syscall end", context);
+
return error;
}
@@ -648,74 +669,50 @@ xfs_attr_list_int(xfs_attr_list_context_t *context)
*/
/*ARGSUSED*/
STATIC int
-xfs_attr_put_listent(xfs_attr_list_context_t *context, attrnames_t *namesp,
+xfs_attr_put_listent(xfs_attr_list_context_t *context, int flags,
char *name, int namelen,
int valuelen, char *value)
{
+ struct attrlist *alist = (struct attrlist *)context->alist;
attrlist_ent_t *aep;
int arraytop;
ASSERT(!(context->flags & ATTR_KERNOVAL));
ASSERT(context->count >= 0);
ASSERT(context->count < (ATTR_MAX_VALUELEN/8));
- ASSERT(context->firstu >= sizeof(*context->alist));
+ ASSERT(context->firstu >= sizeof(*alist));
ASSERT(context->firstu <= context->bufsize);
- arraytop = sizeof(*context->alist) +
- context->count * sizeof(context->alist->al_offset[0]);
+ /*
+ * Only list entries in the right namespace.
+ */
+ if (((context->flags & ATTR_SECURE) == 0) !=
+ ((flags & XFS_ATTR_SECURE) == 0))
+ return 0;
+ if (((context->flags & ATTR_ROOT) == 0) !=
+ ((flags & XFS_ATTR_ROOT) == 0))
+ return 0;
+
+ arraytop = sizeof(*alist) +
+ context->count * sizeof(alist->al_offset[0]);
context->firstu -= ATTR_ENTSIZE(namelen);
if (context->firstu < arraytop) {
xfs_attr_trace_l_c("buffer full", context);
- context->alist->al_more = 1;
+ alist->al_more = 1;
context->seen_enough = 1;
return 1;
}
- aep = (attrlist_ent_t *)&(((char *)context->alist)[ context->firstu ]);
+ aep = (attrlist_ent_t *)&context->alist[context->firstu];
aep->a_valuelen = valuelen;
memcpy(aep->a_name, name, namelen);
- aep->a_name[ namelen ] = 0;
- context->alist->al_offset[ context->count++ ] = context->firstu;
- context->alist->al_count = context->count;
+ aep->a_name[namelen] = 0;
+ alist->al_offset[context->count++] = context->firstu;
+ alist->al_count = context->count;
xfs_attr_trace_l_c("add", context);
return 0;
}
-STATIC int
-xfs_attr_kern_list(xfs_attr_list_context_t *context, attrnames_t *namesp,
- char *name, int namelen,
- int valuelen, char *value)
-{
- char *offset;
- int arraytop;
-
- ASSERT(context->count >= 0);
-
- arraytop = context->count + namesp->attr_namelen + namelen + 1;
- if (arraytop > context->firstu) {
- context->count = -1; /* insufficient space */
- return 1;
- }
- offset = (char *)context->alist + context->count;
- strncpy(offset, namesp->attr_name, namesp->attr_namelen);
- offset += namesp->attr_namelen;
- strncpy(offset, name, namelen); /* real name */
- offset += namelen;
- *offset = '\0';
- context->count += namesp->attr_namelen + namelen + 1;
- return 0;
-}
-
-/*ARGSUSED*/
-STATIC int
-xfs_attr_kern_list_sizes(xfs_attr_list_context_t *context, attrnames_t *namesp,
- char *name, int namelen,
- int valuelen, char *value)
-{
- context->count += namesp->attr_namelen + namelen + 1;
- return 0;
-}
-
/*
* Generate a list of extended attribute names and optionally
* also value lengths. Positive return value follows the XFS
@@ -732,10 +729,9 @@ xfs_attr_list(
attrlist_cursor_kern_t *cursor)
{
xfs_attr_list_context_t context;
+ struct attrlist *alist;
int error;
- XFS_STATS_INC(xs_attr_list);
-
/*
* Validate the cursor.
*/
@@ -756,52 +752,23 @@ xfs_attr_list(
/*
* Initialize the output buffer.
*/
+ memset(&context, 0, sizeof(context));
context.dp = dp;
context.cursor = cursor;
- context.count = 0;
- context.dupcnt = 0;
context.resynch = 1;
context.flags = flags;
- context.seen_enough = 0;
- context.alist = (attrlist_t *)buffer;
- context.put_value = 0;
-
- if (flags & ATTR_KERNAMELS) {
- context.bufsize = bufsize;
- context.firstu = context.bufsize;
- if (flags & ATTR_KERNOVAL)
- context.put_listent = xfs_attr_kern_list_sizes;
- else
- context.put_listent = xfs_attr_kern_list;
- } else {
- context.bufsize = (bufsize & ~(sizeof(int)-1)); /* align */
- context.firstu = context.bufsize;
- context.alist->al_count = 0;
- context.alist->al_more = 0;
- context.alist->al_offset[0] = context.bufsize;
- context.put_listent = xfs_attr_put_listent;
- }
-
- if (XFS_FORCED_SHUTDOWN(dp->i_mount))
- return EIO;
+ context.alist = buffer;
+ context.bufsize = (bufsize & ~(sizeof(int)-1)); /* align */
+ context.firstu = context.bufsize;
+ context.put_listent = xfs_attr_put_listent;
- xfs_ilock(dp, XFS_ILOCK_SHARED);
- xfs_attr_trace_l_c("syscall start", &context);
+ alist = (struct attrlist *)context.alist;
+ alist->al_count = 0;
+ alist->al_more = 0;
+ alist->al_offset[0] = context.bufsize;
error = xfs_attr_list_int(&context);
-
- xfs_iunlock(dp, XFS_ILOCK_SHARED);
- xfs_attr_trace_l_c("syscall end", &context);
-
- if (context.flags & (ATTR_KERNOVAL|ATTR_KERNAMELS)) {
- /* must return negated buffer size or the error */
- if (context.count < 0)
- error = XFS_ERROR(ERANGE);
- else
- error = -context.count;
- } else
- ASSERT(error >= 0);
-
+ ASSERT(error >= 0);
return error;
}
@@ -816,12 +783,10 @@ xfs_attr_inactive(xfs_inode_t *dp)
ASSERT(! XFS_NOT_DQATTACHED(mp, dp));
xfs_ilock(dp, XFS_ILOCK_SHARED);
- if ((XFS_IFORK_Q(dp) == 0) ||
- (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) ||
- (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS &&
- dp->i_d.di_anextents == 0)) {
+ if (!xfs_inode_hasattr(dp) ||
+ dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
xfs_iunlock(dp, XFS_ILOCK_SHARED);
- return(0);
+ return 0;
}
xfs_iunlock(dp, XFS_ILOCK_SHARED);
@@ -854,10 +819,8 @@ xfs_attr_inactive(xfs_inode_t *dp)
/*
* Decide on what work routines to call based on the inode size.
*/
- if ((XFS_IFORK_Q(dp) == 0) ||
- (dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) ||
- (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS &&
- dp->i_d.di_anextents == 0)) {
+ if (!xfs_inode_hasattr(dp) ||
+ dp->i_d.di_aformat == XFS_DINODE_FMT_LOCAL) {
error = 0;
goto out;
}
@@ -974,7 +937,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
xfs_da_brelse(args->trans, bp);
return(retval);
}
- args->rename = 1; /* an atomic rename */
+ args->op_flags |= XFS_DA_OP_RENAME; /* an atomic rename */
args->blkno2 = args->blkno; /* set 2nd entry info*/
args->index2 = args->index;
args->rmtblkno2 = args->rmtblkno;
@@ -1019,7 +982,8 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
* Commit the current trans (including the inode) and start
* a new one.
*/
- if ((error = xfs_attr_rolltrans(&args->trans, dp)))
+ error = xfs_trans_roll(&args->trans, dp);
+ if (error)
return (error);
/*
@@ -1033,7 +997,8 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
* Commit the transaction that added the attr name so that
* later routines can manage their own transactions.
*/
- if ((error = xfs_attr_rolltrans(&args->trans, dp)))
+ error = xfs_trans_roll(&args->trans, dp);
+ if (error)
return (error);
/*
@@ -1054,7 +1019,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
* so that one disappears and one appears atomically. Then we
* must remove the "old" attribute/value pair.
*/
- if (args->rename) {
+ if (args->op_flags & XFS_DA_OP_RENAME) {
/*
* In a separate transaction, set the incomplete flag on the
* "old" attr and clear the incomplete flag on the "new" attr.
@@ -1122,7 +1087,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args)
/*
* Commit the remove and start the next trans in series.
*/
- error = xfs_attr_rolltrans(&args->trans, dp);
+ error = xfs_trans_roll(&args->trans, dp);
} else if (args->rmtblkno > 0) {
/*
@@ -1307,7 +1272,7 @@ restart:
} else if (retval == EEXIST) {
if (args->flags & ATTR_CREATE)
goto out;
- args->rename = 1; /* atomic rename op */
+ args->op_flags |= XFS_DA_OP_RENAME; /* atomic rename op */
args->blkno2 = args->blkno; /* set 2nd entry info*/
args->index2 = args->index;
args->rmtblkno2 = args->rmtblkno;
@@ -1353,7 +1318,8 @@ restart:
* Commit the node conversion and start the next
* trans in the chain.
*/
- if ((error = xfs_attr_rolltrans(&args->trans, dp)))
+ error = xfs_trans_roll(&args->trans, dp);
+ if (error)
goto out;
goto restart;
@@ -1404,7 +1370,8 @@ restart:
* Commit the leaf addition or btree split and start the next
* trans in the chain.
*/
- if ((error = xfs_attr_rolltrans(&args->trans, dp)))
+ error = xfs_trans_roll(&args->trans, dp);
+ if (error)
goto out;
/*
@@ -1425,7 +1392,7 @@ restart:
* so that one disappears and one appears atomically. Then we
* must remove the "old" attribute/value pair.
*/
- if (args->rename) {
+ if (args->op_flags & XFS_DA_OP_RENAME) {
/*
* In a separate transaction, set the incomplete flag on the
* "old" attr and clear the incomplete flag on the "new" attr.
@@ -1504,7 +1471,8 @@ restart:
/*
* Commit and start the next trans in the chain.
*/
- if ((error = xfs_attr_rolltrans(&args->trans, dp)))
+ error = xfs_trans_roll(&args->trans, dp);
+ if (error)
goto out;
} else if (args->rmtblkno > 0) {
@@ -1636,7 +1604,8 @@ xfs_attr_node_removename(xfs_da_args_t *args)
/*
* Commit the Btree join operation and start a new trans.
*/
- if ((error = xfs_attr_rolltrans(&args->trans, dp)))
+ error = xfs_trans_roll(&args->trans, dp);
+ if (error)
goto out;
}
@@ -2137,7 +2106,8 @@ xfs_attr_rmtval_set(xfs_da_args_t *args)
/*
* Start the next trans in the chain.
*/
- if ((error = xfs_attr_rolltrans(&args->trans, dp)))
+ error = xfs_trans_roll(&args->trans, dp);
+ if (error)
return (error);
}
@@ -2287,7 +2257,8 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args)
/*
* Close out trans and start the next one in the chain.
*/
- if ((error = xfs_attr_rolltrans(&args->trans, args->dp)))
+ error = xfs_trans_roll(&args->trans, args->dp);
+ if (error)
return (error);
}
return(0);
@@ -2300,23 +2271,7 @@ xfs_attr_rmtval_remove(xfs_da_args_t *args)
void
xfs_attr_trace_l_c(char *where, struct xfs_attr_list_context *context)
{
- xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_C, where,
- (__psunsigned_t)context->dp,
- (__psunsigned_t)context->cursor->hashval,
- (__psunsigned_t)context->cursor->blkno,
- (__psunsigned_t)context->cursor->offset,
- (__psunsigned_t)context->alist,
- (__psunsigned_t)context->bufsize,
- (__psunsigned_t)context->count,
- (__psunsigned_t)context->firstu,
- (__psunsigned_t)
- ((context->count > 0) &&
- !(context->flags & (ATTR_KERNAMELS|ATTR_KERNOVAL)))
- ? (ATTR_ENTRY(context->alist,
- context->count-1)->a_valuelen)
- : 0,
- (__psunsigned_t)context->dupcnt,
- (__psunsigned_t)context->flags,
+ xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_C, where, context,
(__psunsigned_t)NULL,
(__psunsigned_t)NULL,
(__psunsigned_t)NULL);
@@ -2329,23 +2284,7 @@ void
xfs_attr_trace_l_cn(char *where, struct xfs_attr_list_context *context,
struct xfs_da_intnode *node)
{
- xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CN, where,
- (__psunsigned_t)context->dp,
- (__psunsigned_t)context->cursor->hashval,
- (__psunsigned_t)context->cursor->blkno,
- (__psunsigned_t)context->cursor->offset,
- (__psunsigned_t)context->alist,
- (__psunsigned_t)context->bufsize,
- (__psunsigned_t)context->count,
- (__psunsigned_t)context->firstu,
- (__psunsigned_t)
- ((context->count > 0) &&
- !(context->flags & (ATTR_KERNAMELS|ATTR_KERNOVAL)))
- ? (ATTR_ENTRY(context->alist,
- context->count-1)->a_valuelen)
- : 0,
- (__psunsigned_t)context->dupcnt,
- (__psunsigned_t)context->flags,
+ xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CN, where, context,
(__psunsigned_t)be16_to_cpu(node->hdr.count),
(__psunsigned_t)be32_to_cpu(node->btree[0].hashval),
(__psunsigned_t)be32_to_cpu(node->btree[
@@ -2359,23 +2298,7 @@ void
xfs_attr_trace_l_cb(char *where, struct xfs_attr_list_context *context,
struct xfs_da_node_entry *btree)
{
- xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CB, where,
- (__psunsigned_t)context->dp,
- (__psunsigned_t)context->cursor->hashval,
- (__psunsigned_t)context->cursor->blkno,
- (__psunsigned_t)context->cursor->offset,
- (__psunsigned_t)context->alist,
- (__psunsigned_t)context->bufsize,
- (__psunsigned_t)context->count,
- (__psunsigned_t)context->firstu,
- (__psunsigned_t)
- ((context->count > 0) &&
- !(context->flags & (ATTR_KERNAMELS|ATTR_KERNOVAL)))
- ? (ATTR_ENTRY(context->alist,
- context->count-1)->a_valuelen)
- : 0,
- (__psunsigned_t)context->dupcnt,
- (__psunsigned_t)context->flags,
+ xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CB, where, context,
(__psunsigned_t)be32_to_cpu(btree->hashval),
(__psunsigned_t)be32_to_cpu(btree->before),
(__psunsigned_t)NULL);
@@ -2388,23 +2311,7 @@ void
xfs_attr_trace_l_cl(char *where, struct xfs_attr_list_context *context,
struct xfs_attr_leafblock *leaf)
{
- xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CL, where,
- (__psunsigned_t)context->dp,
- (__psunsigned_t)context->cursor->hashval,
- (__psunsigned_t)context->cursor->blkno,
- (__psunsigned_t)context->cursor->offset,
- (__psunsigned_t)context->alist,
- (__psunsigned_t)context->bufsize,
- (__psunsigned_t)context->count,
- (__psunsigned_t)context->firstu,
- (__psunsigned_t)
- ((context->count > 0) &&
- !(context->flags & (ATTR_KERNAMELS|ATTR_KERNOVAL)))
- ? (ATTR_ENTRY(context->alist,
- context->count-1)->a_valuelen)
- : 0,
- (__psunsigned_t)context->dupcnt,
- (__psunsigned_t)context->flags,
+ xfs_attr_trace_enter(XFS_ATTR_KTRACE_L_CL, where, context,
(__psunsigned_t)be16_to_cpu(leaf->hdr.count),
(__psunsigned_t)be32_to_cpu(leaf->entries[0].hashval),
(__psunsigned_t)be32_to_cpu(leaf->entries[
@@ -2417,329 +2324,24 @@ xfs_attr_trace_l_cl(char *where, struct xfs_attr_list_context *context,
*/
void
xfs_attr_trace_enter(int type, char *where,
- __psunsigned_t a2, __psunsigned_t a3,
- __psunsigned_t a4, __psunsigned_t a5,
- __psunsigned_t a6, __psunsigned_t a7,
- __psunsigned_t a8, __psunsigned_t a9,
- __psunsigned_t a10, __psunsigned_t a11,
- __psunsigned_t a12, __psunsigned_t a13,
- __psunsigned_t a14, __psunsigned_t a15)
+ struct xfs_attr_list_context *context,
+ __psunsigned_t a13, __psunsigned_t a14,
+ __psunsigned_t a15)
{
ASSERT(xfs_attr_trace_buf);
ktrace_enter(xfs_attr_trace_buf, (void *)((__psunsigned_t)type),
- (void *)where,
- (void *)a2, (void *)a3, (void *)a4,
- (void *)a5, (void *)a6, (void *)a7,
- (void *)a8, (void *)a9, (void *)a10,
- (void *)a11, (void *)a12, (void *)a13,
- (void *)a14, (void *)a15);
+ (void *)((__psunsigned_t)where),
+ (void *)((__psunsigned_t)context->dp),
+ (void *)((__psunsigned_t)context->cursor->hashval),
+ (void *)((__psunsigned_t)context->cursor->blkno),
+ (void *)((__psunsigned_t)context->cursor->offset),
+ (void *)((__psunsigned_t)context->alist),
+ (void *)((__psunsigned_t)context->bufsize),
+ (void *)((__psunsigned_t)context->count),
+ (void *)((__psunsigned_t)context->firstu),
+ NULL,
+ (void *)((__psunsigned_t)context->dupcnt),
+ (void *)((__psunsigned_t)context->flags),
+ (void *)a13, (void *)a14, (void *)a15);
}
#endif /* XFS_ATTR_TRACE */
-
-
-/*========================================================================
- * System (pseudo) namespace attribute interface routines.
- *========================================================================*/
-
-STATIC int
-posix_acl_access_set(
- bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags)
-{
- return xfs_acl_vset(vp, data, size, _ACL_TYPE_ACCESS);
-}
-
-STATIC int
-posix_acl_access_remove(
- bhv_vnode_t *vp, char *name, int xflags)
-{
- return xfs_acl_vremove(vp, _ACL_TYPE_ACCESS);
-}
-
-STATIC int
-posix_acl_access_get(
- bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags)
-{
- return xfs_acl_vget(vp, data, size, _ACL_TYPE_ACCESS);
-}
-
-STATIC int
-posix_acl_access_exists(
- bhv_vnode_t *vp)
-{
- return xfs_acl_vhasacl_access(vp);
-}
-
-STATIC int
-posix_acl_default_set(
- bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags)
-{
- return xfs_acl_vset(vp, data, size, _ACL_TYPE_DEFAULT);
-}
-
-STATIC int
-posix_acl_default_get(
- bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags)
-{
- return xfs_acl_vget(vp, data, size, _ACL_TYPE_DEFAULT);
-}
-
-STATIC int
-posix_acl_default_remove(
- bhv_vnode_t *vp, char *name, int xflags)
-{
- return xfs_acl_vremove(vp, _ACL_TYPE_DEFAULT);
-}
-
-STATIC int
-posix_acl_default_exists(
- bhv_vnode_t *vp)
-{
- return xfs_acl_vhasacl_default(vp);
-}
-
-static struct attrnames posix_acl_access = {
- .attr_name = "posix_acl_access",
- .attr_namelen = sizeof("posix_acl_access") - 1,
- .attr_get = posix_acl_access_get,
- .attr_set = posix_acl_access_set,
- .attr_remove = posix_acl_access_remove,
- .attr_exists = posix_acl_access_exists,
-};
-
-static struct attrnames posix_acl_default = {
- .attr_name = "posix_acl_default",
- .attr_namelen = sizeof("posix_acl_default") - 1,
- .attr_get = posix_acl_default_get,
- .attr_set = posix_acl_default_set,
- .attr_remove = posix_acl_default_remove,
- .attr_exists = posix_acl_default_exists,
-};
-
-static struct attrnames *attr_system_names[] =
- { &posix_acl_access, &posix_acl_default };
-
-
-/*========================================================================
- * Namespace-prefix-style attribute name interface routines.
- *========================================================================*/
-
-STATIC int
-attr_generic_set(
- bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags)
-{
- return -xfs_attr_set(xfs_vtoi(vp), name, data, size, xflags);
-}
-
-STATIC int
-attr_generic_get(
- bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags)
-{
- int error, asize = size;
-
- error = xfs_attr_get(xfs_vtoi(vp), name, data, &asize, xflags);
- if (!error)
- return asize;
- return -error;
-}
-
-STATIC int
-attr_generic_remove(
- bhv_vnode_t *vp, char *name, int xflags)
-{
- return -xfs_attr_remove(xfs_vtoi(vp), name, xflags);
-}
-
-STATIC int
-attr_generic_listadd(
- attrnames_t *prefix,
- attrnames_t *namesp,
- void *data,
- size_t size,
- ssize_t *result)
-{
- char *p = data + *result;
-
- *result += prefix->attr_namelen;
- *result += namesp->attr_namelen + 1;
- if (!size)
- return 0;
- if (*result > size)
- return -ERANGE;
- strcpy(p, prefix->attr_name);
- p += prefix->attr_namelen;
- strcpy(p, namesp->attr_name);
- p += namesp->attr_namelen + 1;
- return 0;
-}
-
-STATIC int
-attr_system_list(
- bhv_vnode_t *vp,
- void *data,
- size_t size,
- ssize_t *result)
-{
- attrnames_t *namesp;
- int i, error = 0;
-
- for (i = 0; i < ATTR_SYSCOUNT; i++) {
- namesp = attr_system_names[i];
- if (!namesp->attr_exists || !namesp->attr_exists(vp))
- continue;
- error = attr_generic_listadd(&attr_system, namesp,
- data, size, result);
- if (error)
- break;
- }
- return error;
-}
-
-int
-attr_generic_list(
- bhv_vnode_t *vp, void *data, size_t size, int xflags, ssize_t *result)
-{
- attrlist_cursor_kern_t cursor = { 0 };
- int error;
-
- error = xfs_attr_list(xfs_vtoi(vp), data, size, xflags, &cursor);
- if (error > 0)
- return -error;
- *result = -error;
- return attr_system_list(vp, data, size, result);
-}
-
-attrnames_t *
-attr_lookup_namespace(
- char *name,
- struct attrnames **names,
- int nnames)
-{
- int i;
-
- for (i = 0; i < nnames; i++)
- if (!strncmp(name, names[i]->attr_name, names[i]->attr_namelen))
- return names[i];
- return NULL;
-}
-
-/*
- * Some checks to prevent people abusing EAs to get over quota:
- * - Don't allow modifying user EAs on devices/symlinks;
- * - Don't allow modifying user EAs if sticky bit set;
- */
-STATIC int
-attr_user_capable(
- bhv_vnode_t *vp,
- cred_t *cred)
-{
- struct inode *inode = vn_to_inode(vp);
-
- if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
- return -EPERM;
- if (!S_ISREG(inode->i_mode) && !S_ISDIR(inode->i_mode) &&
- !capable(CAP_SYS_ADMIN))
- return -EPERM;
- if (S_ISDIR(inode->i_mode) && (inode->i_mode & S_ISVTX) &&
- (current_fsuid(cred) != inode->i_uid) && !capable(CAP_FOWNER))
- return -EPERM;
- return 0;
-}
-
-STATIC int
-attr_trusted_capable(
- bhv_vnode_t *vp,
- cred_t *cred)
-{
- struct inode *inode = vn_to_inode(vp);
-
- if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
- return -EPERM;
- if (!capable(CAP_SYS_ADMIN))
- return -EPERM;
- return 0;
-}
-
-STATIC int
-attr_system_set(
- bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags)
-{
- attrnames_t *namesp;
- int error;
-
- if (xflags & ATTR_CREATE)
- return -EINVAL;
-
- namesp = attr_lookup_namespace(name, attr_system_names, ATTR_SYSCOUNT);
- if (!namesp)
- return -EOPNOTSUPP;
- error = namesp->attr_set(vp, name, data, size, xflags);
- if (!error)
- error = vn_revalidate(vp);
- return error;
-}
-
-STATIC int
-attr_system_get(
- bhv_vnode_t *vp, char *name, void *data, size_t size, int xflags)
-{
- attrnames_t *namesp;
-
- namesp = attr_lookup_namespace(name, attr_system_names, ATTR_SYSCOUNT);
- if (!namesp)
- return -EOPNOTSUPP;
- return namesp->attr_get(vp, name, data, size, xflags);
-}
-
-STATIC int
-attr_system_remove(
- bhv_vnode_t *vp, char *name, int xflags)
-{
- attrnames_t *namesp;
-
- namesp = attr_lookup_namespace(name, attr_system_names, ATTR_SYSCOUNT);
- if (!namesp)
- return -EOPNOTSUPP;
- return namesp->attr_remove(vp, name, xflags);
-}
-
-struct attrnames attr_system = {
- .attr_name = "system.",
- .attr_namelen = sizeof("system.") - 1,
- .attr_flag = ATTR_SYSTEM,
- .attr_get = attr_system_get,
- .attr_set = attr_system_set,
- .attr_remove = attr_system_remove,
- .attr_capable = (attrcapable_t)fs_noerr,
-};
-
-struct attrnames attr_trusted = {
- .attr_name = "trusted.",
- .attr_namelen = sizeof("trusted.") - 1,
- .attr_flag = ATTR_ROOT,
- .attr_get = attr_generic_get,
- .attr_set = attr_generic_set,
- .attr_remove = attr_generic_remove,
- .attr_capable = attr_trusted_capable,
-};
-
-struct attrnames attr_secure = {
- .attr_name = "security.",
- .attr_namelen = sizeof("security.") - 1,
- .attr_flag = ATTR_SECURE,
- .attr_get = attr_generic_get,
- .attr_set = attr_generic_set,
- .attr_remove = attr_generic_remove,
- .attr_capable = (attrcapable_t)fs_noerr,
-};
-
-struct attrnames attr_user = {
- .attr_name = "user.",
- .attr_namelen = sizeof("user.") - 1,
- .attr_get = attr_generic_get,
- .attr_set = attr_generic_set,
- .attr_remove = attr_generic_remove,
- .attr_capable = attr_user_capable,
-};
-
-struct attrnames *attr_namespaces[] =
- { &attr_system, &attr_trusted, &attr_secure, &attr_user };
diff --git a/fs/xfs/xfs_attr.h b/fs/xfs/xfs_attr.h
index 6cfc9384fe3..fb3b2a68b9b 100644
--- a/fs/xfs/xfs_attr.h
+++ b/fs/xfs/xfs_attr.h
@@ -18,9 +18,11 @@
#ifndef __XFS_ATTR_H__
#define __XFS_ATTR_H__
+struct xfs_inode;
+struct xfs_da_args;
+struct xfs_attr_list_context;
+
/*
- * xfs_attr.h
- *
* Large attribute lists are structured around Btrees where all the data
* elements are in the leaf nodes. Attribute names are hashed into an int,
* then that int is used as the index into the Btree. Since the hashval
@@ -35,35 +37,6 @@
* External interfaces
*========================================================================*/
-struct cred;
-struct xfs_attr_list_context;
-
-typedef int (*attrset_t)(bhv_vnode_t *, char *, void *, size_t, int);
-typedef int (*attrget_t)(bhv_vnode_t *, char *, void *, size_t, int);
-typedef int (*attrremove_t)(bhv_vnode_t *, char *, int);
-typedef int (*attrexists_t)(bhv_vnode_t *);
-typedef int (*attrcapable_t)(bhv_vnode_t *, struct cred *);
-
-typedef struct attrnames {
- char * attr_name;
- unsigned int attr_namelen;
- unsigned int attr_flag;
- attrget_t attr_get;
- attrset_t attr_set;
- attrremove_t attr_remove;
- attrexists_t attr_exists;
- attrcapable_t attr_capable;
-} attrnames_t;
-
-#define ATTR_NAMECOUNT 4
-extern struct attrnames attr_user;
-extern struct attrnames attr_secure;
-extern struct attrnames attr_system;
-extern struct attrnames attr_trusted;
-extern struct attrnames *attr_namespaces[ATTR_NAMECOUNT];
-
-extern attrnames_t *attr_lookup_namespace(char *, attrnames_t **, int);
-extern int attr_generic_list(bhv_vnode_t *, void *, size_t, int, ssize_t *);
#define ATTR_DONTFOLLOW 0x0001 /* -- unused, from IRIX -- */
#define ATTR_ROOT 0x0002 /* use attrs in root (trusted) namespace */
@@ -71,16 +44,9 @@ extern int attr_generic_list(bhv_vnode_t *, void *, size_t, int, ssize_t *);
#define ATTR_SECURE 0x0008 /* use attrs in security namespace */
#define ATTR_CREATE 0x0010 /* pure create: fail if attr already exists */
#define ATTR_REPLACE 0x0020 /* pure set: fail if attr does not exist */
-#define ATTR_SYSTEM 0x0100 /* use attrs in system (pseudo) namespace */
-#define ATTR_KERNACCESS 0x0400 /* [kernel] iaccess, inode held io-locked */
#define ATTR_KERNOTIME 0x1000 /* [kernel] don't update inode timestamps */
#define ATTR_KERNOVAL 0x2000 /* [kernel] get attr size only, not value */
-#define ATTR_KERNAMELS 0x4000 /* [kernel] list attr names (simple list) */
-
-#define ATTR_KERNORMALS 0x0800 /* [kernel] normal attr list: user+secure */
-#define ATTR_KERNROOTLS 0x8000 /* [kernel] include root in the attr list */
-#define ATTR_KERNFULLS (ATTR_KERNORMALS|ATTR_KERNROOTLS)
/*
* The maximum size (into the kernel or returned from the kernel) of an
@@ -119,22 +85,6 @@ typedef struct attrlist_ent { /* data from attr_list() */
&((char *)buffer)[ ((attrlist_t *)(buffer))->al_offset[index] ])
/*
- * Multi-attribute operation vector.
- */
-typedef struct attr_multiop {
- int am_opcode; /* operation to perform (ATTR_OP_GET, etc.) */
- int am_error; /* [out arg] result of this sub-op (an errno) */
- char *am_attrname; /* attribute name to work with */
- char *am_attrvalue; /* [in/out arg] attribute value (raw bytes) */
- int am_length; /* [in/out arg] length of value */
- int am_flags; /* bitwise OR of attr API flags defined above */
-} attr_multiop_t;
-
-#define ATTR_OP_GET 1 /* return the indicated attr's value */
-#define ATTR_OP_SET 2 /* set/create the indicated attr/value pair */
-#define ATTR_OP_REMOVE 3 /* remove the indicated attr */
-
-/*
* Kernel-internal version of the attrlist cursor.
*/
typedef struct attrlist_cursor_kern {
@@ -148,20 +98,41 @@ typedef struct attrlist_cursor_kern {
/*========================================================================
- * Function prototypes for the kernel.
+ * Structure used to pass context around among the routines.
*========================================================================*/
-struct xfs_inode;
-struct attrlist_cursor_kern;
-struct xfs_da_args;
+
+typedef int (*put_listent_func_t)(struct xfs_attr_list_context *, int,
+ char *, int, int, char *);
+
+typedef struct xfs_attr_list_context {
+ struct xfs_inode *dp; /* inode */
+ struct attrlist_cursor_kern *cursor; /* position in list */
+ char *alist; /* output buffer */
+ int seen_enough; /* T/F: seen enough of list? */
+ ssize_t count; /* num used entries */
+ int dupcnt; /* count dup hashvals seen */
+ int bufsize; /* total buffer size */
+ int firstu; /* first used byte in buffer */
+ int flags; /* from VOP call */
+ int resynch; /* T/F: resynch with cursor */
+ int put_value; /* T/F: need value for listent */
+ put_listent_func_t put_listent; /* list output fmt function */
+ int index; /* index into output buffer */
+} xfs_attr_list_context_t;
+
+
+/*========================================================================
+ * Function prototypes for the kernel.
+ *========================================================================*/
/*
* 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_shortform_getvalue(struct xfs_da_args *);
int xfs_attr_fetch(struct xfs_inode *, struct xfs_name *, char *, int *, int);
int xfs_attr_rmtval_get(struct xfs_da_args *args);
+int xfs_attr_list_int(struct xfs_attr_list_context *);
#endif /* __XFS_ATTR_H__ */
diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c
index 303d41e4217..79da6b2ea99 100644
--- a/fs/xfs/xfs_attr_leaf.c
+++ b/fs/xfs/xfs_attr_leaf.c
@@ -94,13 +94,6 @@ STATIC int xfs_attr_leaf_entsize(xfs_attr_leafblock_t *leaf, int index);
* Namespace helper routines
*========================================================================*/
-STATIC_INLINE attrnames_t *
-xfs_attr_flags_namesp(int flags)
-{
- return ((flags & XFS_ATTR_SECURE) ? &attr_secure:
- ((flags & XFS_ATTR_ROOT) ? &attr_trusted : &attr_user));
-}
-
/*
* If namespace bits don't match return 0.
* If all match then return 1.
@@ -111,25 +104,6 @@ xfs_attr_namesp_match(int arg_flags, int ondisk_flags)
return XFS_ATTR_NSP_ONDISK(ondisk_flags) == XFS_ATTR_NSP_ARGS_TO_ONDISK(arg_flags);
}
-/*
- * If namespace bits don't match and we don't have an override for it
- * then return 0.
- * If all match or are overridable then return 1.
- */
-STATIC_INLINE int
-xfs_attr_namesp_match_overrides(int arg_flags, int ondisk_flags)
-{
- if (((arg_flags & ATTR_SECURE) == 0) !=
- ((ondisk_flags & XFS_ATTR_SECURE) == 0) &&
- !(arg_flags & ATTR_KERNORMALS))
- return 0;
- if (((arg_flags & ATTR_ROOT) == 0) !=
- ((ondisk_flags & XFS_ATTR_ROOT) == 0) &&
- !(arg_flags & ATTR_KERNROOTLS))
- return 0;
- return 1;
-}
-
/*========================================================================
* External routines when attribute fork size < XFS_LITINO(mp).
@@ -369,9 +343,10 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
* Fix up the start offset of the attribute fork
*/
totsize -= size;
- if (totsize == sizeof(xfs_attr_sf_hdr_t) && !args->addname &&
- (mp->m_flags & XFS_MOUNT_ATTR2) &&
- (dp->i_d.di_format != XFS_DINODE_FMT_BTREE)) {
+ if (totsize == sizeof(xfs_attr_sf_hdr_t) &&
+ !(args->op_flags & XFS_DA_OP_ADDNAME) &&
+ (mp->m_flags & XFS_MOUNT_ATTR2) &&
+ (dp->i_d.di_format != XFS_DINODE_FMT_BTREE)) {
/*
* Last attribute now removed, revert to original
* inode format making all literal area available
@@ -389,9 +364,10 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
xfs_idata_realloc(dp, -size, XFS_ATTR_FORK);
dp->i_d.di_forkoff = xfs_attr_shortform_bytesfit(dp, totsize);
ASSERT(dp->i_d.di_forkoff);
- ASSERT(totsize > sizeof(xfs_attr_sf_hdr_t) || args->addname ||
- !(mp->m_flags & XFS_MOUNT_ATTR2) ||
- dp->i_d.di_format == XFS_DINODE_FMT_BTREE);
+ ASSERT(totsize > sizeof(xfs_attr_sf_hdr_t) ||
+ (args->op_flags & XFS_DA_OP_ADDNAME) ||
+ !(mp->m_flags & XFS_MOUNT_ATTR2) ||
+ dp->i_d.di_format == XFS_DINODE_FMT_BTREE);
dp->i_afp->if_ext_max =
XFS_IFORK_ASIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t);
dp->i_df.if_ext_max =
@@ -531,7 +507,7 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
nargs.total = args->total;
nargs.whichfork = XFS_ATTR_FORK;
nargs.trans = args->trans;
- nargs.oknoent = 1;
+ nargs.op_flags = XFS_DA_OP_OKNOENT;
sfe = &sf->list[0];
for (i = 0; i < sf->hdr.count; i++) {
@@ -555,7 +531,7 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
out:
if(bp)
xfs_da_buf_done(bp);
- kmem_free(tmpbuffer, size);
+ kmem_free(tmpbuffer);
return(error);
}
@@ -624,15 +600,8 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
(XFS_ISRESET_CURSOR(cursor) &&
(dp->i_afp->if_bytes + sf->hdr.count * 16) < context->bufsize)) {
for (i = 0, sfe = &sf->list[0]; i < sf->hdr.count; i++) {
- attrnames_t *namesp;
-
- if (!xfs_attr_namesp_match_overrides(context->flags, sfe->flags)) {
- sfe = XFS_ATTR_SF_NEXTENTRY(sfe);
- continue;
- }
- namesp = xfs_attr_flags_namesp(sfe->flags);
error = context->put_listent(context,
- namesp,
+ sfe->flags,
(char *)sfe->nameval,
(int)sfe->namelen,
(int)sfe->valuelen,
@@ -676,13 +645,10 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
XFS_ERRLEVEL_LOW,
context->dp->i_mount, sfe);
xfs_attr_trace_l_c("sf corrupted", context);
- kmem_free(sbuf, sbsize);
+ kmem_free(sbuf);
return XFS_ERROR(EFSCORRUPTED);
}
- if (!xfs_attr_namesp_match_overrides(context->flags, sfe->flags)) {
- sfe = XFS_ATTR_SF_NEXTENTRY(sfe);
- continue;
- }
+
sbp->entno = i;
sbp->hash = xfs_da_hashname((char *)sfe->nameval, sfe->namelen);
sbp->name = (char *)sfe->nameval;
@@ -717,7 +683,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
}
}
if (i == nsbuf) {
- kmem_free(sbuf, sbsize);
+ kmem_free(sbuf);
xfs_attr_trace_l_c("blk end", context);
return(0);
}
@@ -726,16 +692,12 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
* Loop putting entries into the user buffer.
*/
for ( ; i < nsbuf; i++, sbp++) {
- attrnames_t *namesp;
-
- namesp = xfs_attr_flags_namesp(sbp->flags);
-
if (cursor->hashval != sbp->hash) {
cursor->hashval = sbp->hash;
cursor->offset = 0;
}
error = context->put_listent(context,
- namesp,
+ sbp->flags,
sbp->name,
sbp->namelen,
sbp->valuelen,
@@ -747,7 +709,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
cursor->offset++;
}
- kmem_free(sbuf, sbsize);
+ kmem_free(sbuf);
xfs_attr_trace_l_c("sf E-O-F", context);
return(0);
}
@@ -853,7 +815,7 @@ xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args, int forkoff)
nargs.total = args->total;
nargs.whichfork = XFS_ATTR_FORK;
nargs.trans = args->trans;
- nargs.oknoent = 1;
+ nargs.op_flags = XFS_DA_OP_OKNOENT;
entry = &leaf->entries[0];
for (i = 0; i < be16_to_cpu(leaf->hdr.count); entry++, i++) {
if (entry->flags & XFS_ATTR_INCOMPLETE)
@@ -873,7 +835,7 @@ xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args, int forkoff)
error = 0;
out:
- kmem_free(tmpbuffer, XFS_LBSIZE(dp->i_mount));
+ kmem_free(tmpbuffer);
return(error);
}
@@ -1155,7 +1117,7 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
entry->hashval = cpu_to_be32(args->hashval);
entry->flags = tmp ? XFS_ATTR_LOCAL : 0;
entry->flags |= XFS_ATTR_NSP_ARGS_TO_ONDISK(args->flags);
- if (args->rename) {
+ if (args->op_flags & XFS_DA_OP_RENAME) {
entry->flags |= XFS_ATTR_INCOMPLETE;
if ((args->blkno2 == args->blkno) &&
(args->index2 <= args->index)) {
@@ -1271,7 +1233,7 @@ xfs_attr_leaf_compact(xfs_trans_t *trans, xfs_dabuf_t *bp)
be16_to_cpu(hdr_s->count), mp);
xfs_da_log_buf(trans, bp, 0, XFS_LBSIZE(mp) - 1);
- kmem_free(tmpbuffer, XFS_LBSIZE(mp));
+ kmem_free(tmpbuffer);
}
/*
@@ -1921,7 +1883,7 @@ xfs_attr_leaf_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
be16_to_cpu(drop_hdr->count), mp);
}
memcpy((char *)save_leaf, (char *)tmp_leaf, state->blocksize);
- kmem_free(tmpbuffer, state->blocksize);
+ kmem_free(tmpbuffer);
}
xfs_da_log_buf(state->args->trans, save_blk->bp, 0,
@@ -2400,8 +2362,6 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context)
*/
retval = 0;
for ( ; (i < be16_to_cpu(leaf->hdr.count)); entry++, i++) {
- attrnames_t *namesp;
-
if (be32_to_cpu(entry->hashval) != cursor->hashval) {
cursor->hashval = be32_to_cpu(entry->hashval);
cursor->offset = 0;
@@ -2409,17 +2369,13 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context)
if (entry->flags & XFS_ATTR_INCOMPLETE)
continue; /* skip incomplete entries */
- if (!xfs_attr_namesp_match_overrides(context->flags, entry->flags))
- continue;
-
- namesp = xfs_attr_flags_namesp(entry->flags);
if (entry->flags & XFS_ATTR_LOCAL) {
xfs_attr_leaf_name_local_t *name_loc =
XFS_ATTR_LEAF_NAME_LOCAL(leaf, i);
retval = context->put_listent(context,
- namesp,
+ entry->flags,
(char *)name_loc->nameval,
(int)name_loc->namelen,
be16_to_cpu(name_loc->valuelen),
@@ -2446,16 +2402,15 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context)
if (retval)
return retval;
retval = context->put_listent(context,
- namesp,
+ entry->flags,
(char *)name_rmt->name,
(int)name_rmt->namelen,
valuelen,
(char*)args.value);
- kmem_free(args.value, valuelen);
- }
- else {
+ kmem_free(args.value);
+ } else {
retval = context->put_listent(context,
- namesp,
+ entry->flags,
(char *)name_rmt->name,
(int)name_rmt->namelen,
valuelen,
@@ -2543,9 +2498,7 @@ xfs_attr_leaf_clearflag(xfs_da_args_t *args)
/*
* Commit the flag value change and start the next trans in series.
*/
- error = xfs_attr_rolltrans(&args->trans, args->dp);
-
- return(error);
+ return xfs_trans_roll(&args->trans, args->dp);
}
/*
@@ -2592,9 +2545,7 @@ xfs_attr_leaf_setflag(xfs_da_args_t *args)
/*
* Commit the flag value change and start the next trans in series.
*/
- error = xfs_attr_rolltrans(&args->trans, args->dp);
-
- return(error);
+ return xfs_trans_roll(&args->trans, args->dp);
}
/*
@@ -2710,7 +2661,7 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
/*
* Commit the flag value change and start the next trans in series.
*/
- error = xfs_attr_rolltrans(&args->trans, args->dp);
+ error = xfs_trans_roll(&args->trans, args->dp);
return(error);
}
@@ -2768,7 +2719,7 @@ xfs_attr_root_inactive(xfs_trans_t **trans, xfs_inode_t *dp)
/*
* Commit the invalidate and start the next transaction.
*/
- error = xfs_attr_rolltrans(trans, dp);
+ error = xfs_trans_roll(trans, dp);
return (error);
}
@@ -2870,7 +2821,8 @@ xfs_attr_node_inactive(xfs_trans_t **trans, xfs_inode_t *dp, xfs_dabuf_t *bp,
/*
* Atomically commit the whole invalidate stuff.
*/
- if ((error = xfs_attr_rolltrans(trans, dp)))
+ error = xfs_trans_roll(trans, dp);
+ if (error)
return (error);
}
@@ -2954,7 +2906,7 @@ xfs_attr_leaf_inactive(xfs_trans_t **trans, xfs_inode_t *dp, xfs_dabuf_t *bp)
error = tmp; /* save only the 1st errno */
}
- kmem_free((xfs_caddr_t)list, size);
+ kmem_free((xfs_caddr_t)list);
return(error);
}
@@ -3009,7 +2961,8 @@ xfs_attr_leaf_freextent(xfs_trans_t **trans, xfs_inode_t *dp,
/*
* Roll to next transaction.
*/
- if ((error = xfs_attr_rolltrans(trans, dp)))
+ error = xfs_trans_roll(trans, dp);
+ if (error)
return (error);
}
@@ -3019,60 +2972,3 @@ xfs_attr_leaf_freextent(xfs_trans_t **trans, xfs_inode_t *dp,
return(0);
}
-
-
-/*
- * Roll from one trans in the sequence of PERMANENT transactions to the next.
- */
-int
-xfs_attr_rolltrans(xfs_trans_t **transp, xfs_inode_t *dp)
-{
- xfs_trans_t *trans;
- unsigned int logres, count;
- int error;
-
- /*
- * Ensure that the inode is always logged.
- */
- trans = *transp;
- xfs_trans_log_inode(trans, dp, XFS_ILOG_CORE);
-
- /*
- * Copy the critical parameters from one trans to the next.
- */
- logres = trans->t_log_res;
- count = trans->t_log_count;
- *transp = xfs_trans_dup(trans);
-
- /*
- * Commit the current transaction.
- * If this commit failed, then it'd just unlock those items that
- * are not marked ihold. That also means that a filesystem shutdown
- * is in progress. The caller takes the responsibility to cancel
- * the duplicate transaction that gets returned.
- */
- if ((error = xfs_trans_commit(trans, 0)))
- return (error);
-
- trans = *transp;
-
- /*
- * Reserve space in the log for th next transaction.
- * This also pushes items in the "AIL", the list of logged items,
- * out to disk if they are taking up space at the tail of the log
- * that we want to use. This requires that either nothing be locked
- * across this call, or that anything that is locked be logged in
- * the prior and the next transactions.
- */
- error = xfs_trans_reserve(trans, 0, logres, 0,
- XFS_TRANS_PERM_LOG_RES, count);
- /*
- * Ensure that the inode is in the new transaction and locked.
- */
- if (!error) {
- xfs_trans_ijoin(trans, dp, XFS_ILOCK_EXCL);
- xfs_trans_ihold(trans, dp);
- }
- return (error);
-
-}
diff --git a/fs/xfs/xfs_attr_leaf.h b/fs/xfs/xfs_attr_leaf.h
index 040f732ce1e..83e9af417ca 100644
--- a/fs/xfs/xfs_attr_leaf.h
+++ b/fs/xfs/xfs_attr_leaf.h
@@ -30,7 +30,7 @@
struct attrlist;
struct attrlist_cursor_kern;
-struct attrnames;
+struct xfs_attr_list_context;
struct xfs_dabuf;
struct xfs_da_args;
struct xfs_da_state;
@@ -204,33 +204,6 @@ static inline int xfs_attr_leaf_entsize_local_max(int bsize)
return (((bsize) >> 1) + ((bsize) >> 2));
}
-
-/*========================================================================
- * Structure used to pass context around among the routines.
- *========================================================================*/
-
-
-struct xfs_attr_list_context;
-
-typedef int (*put_listent_func_t)(struct xfs_attr_list_context *, struct attrnames *,
- char *, int, int, char *);
-
-typedef struct xfs_attr_list_context {
- struct xfs_inode *dp; /* inode */
- struct attrlist_cursor_kern *cursor; /* position in list */
- struct attrlist *alist; /* output buffer */
- int seen_enough; /* T/F: seen enough of list? */
- int count; /* num used entries */
- int dupcnt; /* count dup hashvals seen */
- int bufsize; /* total buffer size */
- int firstu; /* first used byte in buffer */
- int flags; /* from VOP call */
- int resynch; /* T/F: resynch with cursor */
- int put_value; /* T/F: need value for listent */
- put_listent_func_t put_listent; /* list output fmt function */
- int index; /* index into output buffer */
-} xfs_attr_list_context_t;
-
/*
* Used to keep a list of "remote value" extents when unlinking an inode.
*/
@@ -301,6 +274,4 @@ int xfs_attr_leaf_order(struct xfs_dabuf *leaf1_bp,
struct xfs_dabuf *leaf2_bp);
int xfs_attr_leaf_newentsize(int namelen, int valuelen, int blocksize,
int *local);
-int xfs_attr_rolltrans(struct xfs_trans **transp, struct xfs_inode *dp);
-
#endif /* __XFS_ATTR_LEAF_H__ */
diff --git a/fs/xfs/xfs_attr_sf.h b/fs/xfs/xfs_attr_sf.h
index f67f917803b..ea22839caed 100644
--- a/fs/xfs/xfs_attr_sf.h
+++ b/fs/xfs/xfs_attr_sf.h
@@ -97,13 +97,9 @@ void xfs_attr_trace_l_cb(char *where, struct xfs_attr_list_context *context,
void xfs_attr_trace_l_cl(char *where, struct xfs_attr_list_context *context,
struct xfs_attr_leafblock *leaf);
void xfs_attr_trace_enter(int type, char *where,
- __psunsigned_t a2, __psunsigned_t a3,
- __psunsigned_t a4, __psunsigned_t a5,
- __psunsigned_t a6, __psunsigned_t a7,
- __psunsigned_t a8, __psunsigned_t a9,
- __psunsigned_t a10, __psunsigned_t a11,
- __psunsigned_t a12, __psunsigned_t a13,
- __psunsigned_t a14, __psunsigned_t a15);
+ struct xfs_attr_list_context *context,
+ __psunsigned_t a13, __psunsigned_t a14,
+ __psunsigned_t a15);
#else
#define xfs_attr_trace_l_c(w,c)
#define xfs_attr_trace_l_cn(w,c,n)
diff --git a/fs/xfs/xfs_bit.c b/fs/xfs/xfs_bit.c
index fab0b6d5a41..48228848f5a 100644
--- a/fs/xfs/xfs_bit.c
+++ b/fs/xfs/xfs_bit.c
@@ -25,109 +25,6 @@
* XFS bit manipulation routines, used in non-realtime code.
*/
-#ifndef HAVE_ARCH_HIGHBIT
-/*
- * Index of high bit number in byte, -1 for none set, 0..7 otherwise.
- */
-static const char xfs_highbit[256] = {
- -1, 0, 1, 1, 2, 2, 2, 2, /* 00 .. 07 */
- 3, 3, 3, 3, 3, 3, 3, 3, /* 08 .. 0f */
- 4, 4, 4, 4, 4, 4, 4, 4, /* 10 .. 17 */
- 4, 4, 4, 4, 4, 4, 4, 4, /* 18 .. 1f */
- 5, 5, 5, 5, 5, 5, 5, 5, /* 20 .. 27 */
- 5, 5, 5, 5, 5, 5, 5, 5, /* 28 .. 2f */
- 5, 5, 5, 5, 5, 5, 5, 5, /* 30 .. 37 */
- 5, 5, 5, 5, 5, 5, 5, 5, /* 38 .. 3f */
- 6, 6, 6, 6, 6, 6, 6, 6, /* 40 .. 47 */
- 6, 6, 6, 6, 6, 6, 6, 6, /* 48 .. 4f */
- 6, 6, 6, 6, 6, 6, 6, 6, /* 50 .. 57 */
- 6, 6, 6, 6, 6, 6, 6, 6, /* 58 .. 5f */
- 6, 6, 6, 6, 6, 6, 6, 6, /* 60 .. 67 */
- 6, 6, 6, 6, 6, 6, 6, 6, /* 68 .. 6f */
- 6, 6, 6, 6, 6, 6, 6, 6, /* 70 .. 77 */
- 6, 6, 6, 6, 6, 6, 6, 6, /* 78 .. 7f */
- 7, 7, 7, 7, 7, 7, 7, 7, /* 80 .. 87 */
- 7, 7, 7, 7, 7, 7, 7, 7, /* 88 .. 8f */
- 7, 7, 7, 7, 7, 7, 7, 7, /* 90 .. 97 */
- 7, 7, 7, 7, 7, 7, 7, 7, /* 98 .. 9f */
- 7, 7, 7, 7, 7, 7, 7, 7, /* a0 .. a7 */
- 7, 7, 7, 7, 7, 7, 7, 7, /* a8 .. af */
- 7, 7, 7, 7, 7, 7, 7, 7, /* b0 .. b7 */
- 7, 7, 7, 7, 7, 7, 7, 7, /* b8 .. bf */
- 7, 7, 7, 7, 7, 7, 7, 7, /* c0 .. c7 */
- 7, 7, 7, 7, 7, 7, 7, 7, /* c8 .. cf */
- 7, 7, 7, 7, 7, 7, 7, 7, /* d0 .. d7 */
- 7, 7, 7, 7, 7, 7, 7, 7, /* d8 .. df */
- 7, 7, 7, 7, 7, 7, 7, 7, /* e0 .. e7 */
- 7, 7, 7, 7, 7, 7, 7, 7, /* e8 .. ef */
- 7, 7, 7, 7, 7, 7, 7, 7, /* f0 .. f7 */
- 7, 7, 7, 7, 7, 7, 7, 7, /* f8 .. ff */
-};
-#endif
-
-/*
- * xfs_highbit32: get high bit set out of 32-bit argument, -1 if none set.
- */
-inline int
-xfs_highbit32(
- __uint32_t v)
-{
-#ifdef HAVE_ARCH_HIGHBIT
- return highbit32(v);
-#else
- int i;
-
- if (v & 0xffff0000)
- if (v & 0xff000000)
- i = 24;
- else
- i = 16;
- else if (v & 0x0000ffff)
- if (v & 0x0000ff00)
- i = 8;
- else
- i = 0;
- else
- return -1;
- return i + xfs_highbit[(v >> i) & 0xff];
-#endif
-}
-
-/*
- * xfs_lowbit64: get low bit set out of 64-bit argument, -1 if none set.
- */
-int
-xfs_lowbit64(
- __uint64_t v)
-{
- __uint32_t w = (__uint32_t)v;
- int n = 0;
-
- if (w) { /* lower bits */
- n = ffs(w);
- } else { /* upper bits */
- w = (__uint32_t)(v >> 32);
- if (w && (n = ffs(w)))
- n += 32;
- }
- return n - 1;
-}
-
-/*
- * xfs_highbit64: get high bit set out of 64-bit argument, -1 if none set.
- */
-int
-xfs_highbit64(
- __uint64_t v)
-{
- __uint32_t h = (__uint32_t)(v >> 32);
-
- if (h)
- return xfs_highbit32(h) + 32;
- return xfs_highbit32((__uint32_t)v);
-}
-
-
/*
* Return whether bitmap is empty.
* Size is number of words in the bitmap, which is padded to word boundary
diff --git a/fs/xfs/xfs_bit.h b/fs/xfs/xfs_bit.h
index 082641a9782..8e0e463dae2 100644
--- a/fs/xfs/xfs_bit.h
+++ b/fs/xfs/xfs_bit.h
@@ -47,13 +47,39 @@ static inline __uint64_t xfs_mask64lo(int n)
}
/* Get high bit set out of 32-bit argument, -1 if none set */
-extern int xfs_highbit32(__uint32_t v);
+static inline int xfs_highbit32(__uint32_t v)
+{
+ return fls(v) - 1;
+}
+
+/* Get high bit set out of 64-bit argument, -1 if none set */
+static inline int xfs_highbit64(__uint64_t v)
+{
+ return fls64(v) - 1;
+}
+
+/* Get low bit set out of 32-bit argument, -1 if none set */
+static inline int xfs_lowbit32(__uint32_t v)
+{
+ unsigned long t = v;
+ return (v) ? find_first_bit(&t, 32) : -1;
+}
/* Get low bit set out of 64-bit argument, -1 if none set */
-extern int xfs_lowbit64(__uint64_t v);
+static inline int xfs_lowbit64(__uint64_t v)
+{
+ __uint32_t w = (__uint32_t)v;
+ int n = 0;
-/* Get high bit set out of 64-bit argument, -1 if none set */
-extern int xfs_highbit64(__uint64_t);
+ if (w) { /* lower bits */
+ n = ffs(w);
+ } else { /* upper bits */
+ w = (__uint32_t)(v >> 32);
+ if (w && (n = ffs(w)))
+ n += 32;
+ }
+ return n - 1;
+}
/* Return whether bitmap is empty (1 == empty) */
extern int xfs_bitmap_empty(uint *map, uint size);
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c
index 53c259f5a5a..a1aab9275d5 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/xfs_bmap.c
@@ -384,14 +384,14 @@ xfs_bmap_count_tree(
int levelin,
int *count);
-STATIC int
+STATIC void
xfs_bmap_count_leaves(
xfs_ifork_t *ifp,
xfs_extnum_t idx,
int numrecs,
int *count);
-STATIC int
+STATIC void
xfs_bmap_disk_count_leaves(
xfs_extnum_t idx,
xfs_bmbt_block_t *block,
@@ -428,7 +428,8 @@ xfs_bmap_add_attrfork_btree(
cur->bc_private.b.firstblock = *firstblock;
if ((error = xfs_bmbt_lookup_ge(cur, 0, 0, 0, &stat)))
goto error0;
- ASSERT(stat == 1); /* must be at least one entry */
+ /* must be at least one entry */
+ XFS_WANT_CORRUPTED_GOTO(stat == 1, error0);
if ((error = xfs_bmbt_newroot(cur, flags, &stat)))
goto error0;
if (stat == 0) {
@@ -816,13 +817,13 @@ xfs_bmap_add_extent_delay_real(
RIGHT.br_startblock,
RIGHT.br_blockcount, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_delete(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_decrement(cur, 0, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, LEFT.br_startoff,
LEFT.br_startblock,
LEFT.br_blockcount +
@@ -860,7 +861,7 @@ xfs_bmap_add_extent_delay_real(
LEFT.br_startblock, LEFT.br_blockcount,
&i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, LEFT.br_startoff,
LEFT.br_startblock,
LEFT.br_blockcount +
@@ -895,7 +896,7 @@ xfs_bmap_add_extent_delay_real(
RIGHT.br_startblock,
RIGHT.br_blockcount, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, PREV.br_startoff,
new->br_startblock,
PREV.br_blockcount +
@@ -928,11 +929,11 @@ xfs_bmap_add_extent_delay_real(
new->br_startblock, new->br_blockcount,
&i)))
goto done;
- ASSERT(i == 0);
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
cur->bc_rec.b.br_state = XFS_EXT_NORM;
if ((error = xfs_bmbt_insert(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
*dnew = 0;
/* DELTA: The in-core extent described by new changed type. */
@@ -963,7 +964,7 @@ xfs_bmap_add_extent_delay_real(
LEFT.br_startblock, LEFT.br_blockcount,
&i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, LEFT.br_startoff,
LEFT.br_startblock,
LEFT.br_blockcount +
@@ -1004,11 +1005,11 @@ xfs_bmap_add_extent_delay_real(
new->br_startblock, new->br_blockcount,
&i)))
goto done;
- ASSERT(i == 0);
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
cur->bc_rec.b.br_state = XFS_EXT_NORM;
if ((error = xfs_bmbt_insert(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
ip->i_d.di_nextents > ip->i_df.if_ext_max) {
@@ -1054,7 +1055,7 @@ xfs_bmap_add_extent_delay_real(
RIGHT.br_startblock,
RIGHT.br_blockcount, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, new->br_startoff,
new->br_startblock,
new->br_blockcount +
@@ -1094,11 +1095,11 @@ xfs_bmap_add_extent_delay_real(
new->br_startblock, new->br_blockcount,
&i)))
goto done;
- ASSERT(i == 0);
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
cur->bc_rec.b.br_state = XFS_EXT_NORM;
if ((error = xfs_bmbt_insert(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
ip->i_d.di_nextents > ip->i_df.if_ext_max) {
@@ -1149,11 +1150,11 @@ xfs_bmap_add_extent_delay_real(
new->br_startblock, new->br_blockcount,
&i)))
goto done;
- ASSERT(i == 0);
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
cur->bc_rec.b.br_state = XFS_EXT_NORM;
if ((error = xfs_bmbt_insert(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
ip->i_d.di_nextents > ip->i_df.if_ext_max) {
@@ -1377,19 +1378,19 @@ xfs_bmap_add_extent_unwritten_real(
RIGHT.br_startblock,
RIGHT.br_blockcount, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_delete(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_decrement(cur, 0, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_delete(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_decrement(cur, 0, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, LEFT.br_startoff,
LEFT.br_startblock,
LEFT.br_blockcount + PREV.br_blockcount +
@@ -1426,13 +1427,13 @@ xfs_bmap_add_extent_unwritten_real(
PREV.br_startblock, PREV.br_blockcount,
&i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_delete(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_decrement(cur, 0, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, LEFT.br_startoff,
LEFT.br_startblock,
LEFT.br_blockcount + PREV.br_blockcount,
@@ -1469,13 +1470,13 @@ xfs_bmap_add_extent_unwritten_real(
RIGHT.br_startblock,
RIGHT.br_blockcount, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_delete(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_decrement(cur, 0, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, new->br_startoff,
new->br_startblock,
new->br_blockcount + RIGHT.br_blockcount,
@@ -1508,7 +1509,7 @@ xfs_bmap_add_extent_unwritten_real(
new->br_startblock, new->br_blockcount,
&i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, new->br_startoff,
new->br_startblock, new->br_blockcount,
newext)))
@@ -1549,7 +1550,7 @@ xfs_bmap_add_extent_unwritten_real(
PREV.br_startblock, PREV.br_blockcount,
&i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur,
PREV.br_startoff + new->br_blockcount,
PREV.br_startblock + new->br_blockcount,
@@ -1596,7 +1597,7 @@ xfs_bmap_add_extent_unwritten_real(
PREV.br_startblock, PREV.br_blockcount,
&i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur,
PREV.br_startoff + new->br_blockcount,
PREV.br_startblock + new->br_blockcount,
@@ -1606,7 +1607,7 @@ xfs_bmap_add_extent_unwritten_real(
cur->bc_rec.b = *new;
if ((error = xfs_bmbt_insert(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
/* DELTA: One in-core extent is split in two. */
temp = PREV.br_startoff;
@@ -1640,7 +1641,7 @@ xfs_bmap_add_extent_unwritten_real(
PREV.br_startblock,
PREV.br_blockcount, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, PREV.br_startoff,
PREV.br_startblock,
PREV.br_blockcount - new->br_blockcount,
@@ -1682,7 +1683,7 @@ xfs_bmap_add_extent_unwritten_real(
PREV.br_startblock, PREV.br_blockcount,
&i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, PREV.br_startoff,
PREV.br_startblock,
PREV.br_blockcount - new->br_blockcount,
@@ -1692,11 +1693,11 @@ xfs_bmap_add_extent_unwritten_real(
new->br_startblock, new->br_blockcount,
&i)))
goto done;
- ASSERT(i == 0);
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
cur->bc_rec.b.br_state = XFS_EXT_NORM;
if ((error = xfs_bmbt_insert(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
/* DELTA: One in-core extent is split in two. */
temp = PREV.br_startoff;
@@ -1732,27 +1733,34 @@ xfs_bmap_add_extent_unwritten_real(
PREV.br_startblock, PREV.br_blockcount,
&i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
/* new right extent - oldext */
if ((error = xfs_bmbt_update(cur, r[1].br_startoff,
r[1].br_startblock, r[1].br_blockcount,
r[1].br_state)))
goto done;
/* new left extent - oldext */
- PREV.br_blockcount =
- new->br_startoff - PREV.br_startoff;
cur->bc_rec.b = PREV;
+ cur->bc_rec.b.br_blockcount =
+ new->br_startoff - PREV.br_startoff;
if ((error = xfs_bmbt_insert(cur, &i)))
goto done;
- ASSERT(i == 1);
- if ((error = xfs_bmbt_increment(cur, 0, &i)))
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
+ /*
+ * Reset the cursor to the position of the new extent
+ * we are about to insert as we can't trust it after
+ * the previous insert.
+ */
+ if ((error = xfs_bmbt_lookup_eq(cur, new->br_startoff,
+ new->br_startblock, new->br_blockcount,
+ &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
/* new middle extent - newext */
- cur->bc_rec.b = *new;
+ cur->bc_rec.b.br_state = new->br_state;
if ((error = xfs_bmbt_insert(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
/* DELTA: One in-core extent is split in three. */
temp = PREV.br_startoff;
@@ -2097,13 +2105,13 @@ xfs_bmap_add_extent_hole_real(
right.br_startblock,
right.br_blockcount, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_delete(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_decrement(cur, 0, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, left.br_startoff,
left.br_startblock,
left.br_blockcount +
@@ -2139,7 +2147,7 @@ xfs_bmap_add_extent_hole_real(
left.br_startblock,
left.br_blockcount, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, left.br_startoff,
left.br_startblock,
left.br_blockcount +
@@ -2174,7 +2182,7 @@ xfs_bmap_add_extent_hole_real(
right.br_startblock,
right.br_blockcount, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
if ((error = xfs_bmbt_update(cur, new->br_startoff,
new->br_startblock,
new->br_blockcount +
@@ -2208,11 +2216,11 @@ xfs_bmap_add_extent_hole_real(
new->br_startblock,
new->br_blockcount, &i)))
goto done;
- ASSERT(i == 0);
+ XFS_WANT_CORRUPTED_GOTO(i == 0, done);
cur->bc_rec.b.br_state = new->br_state;
if ((error = xfs_bmbt_insert(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
/* DELTA: A new extent was added in a hole. */
temp = new->br_startoff;
@@ -3131,7 +3139,7 @@ xfs_bmap_del_extent(
got.br_startblock, got.br_blockcount,
&i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
}
da_old = da_new = 0;
} else {
@@ -3164,7 +3172,7 @@ xfs_bmap_del_extent(
}
if ((error = xfs_bmbt_delete(cur, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
break;
case 2:
@@ -3268,7 +3276,7 @@ xfs_bmap_del_extent(
got.br_startblock,
temp, &i)))
goto done;
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
/*
* Update the btree record back
* to the original value.
@@ -3289,7 +3297,7 @@ xfs_bmap_del_extent(
error = XFS_ERROR(ENOSPC);
goto done;
}
- ASSERT(i == 1);
+ XFS_WANT_CORRUPTED_GOTO(i == 1, done);
} else
flags |= XFS_ILOG_FEXT(whichfork);
XFS_IFORK_NEXT_SET(ip, whichfork,
@@ -3992,7 +4000,7 @@ xfs_bmap_add_attrfork(
ip->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS;
}
ASSERT(ip->i_d.di_anextents == 0);
- VN_HOLD(XFS_ITOV(ip));
+ IHOLD(ip);
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
switch (ip->i_d.di_format) {
@@ -5970,7 +5978,7 @@ unlock_and_return:
xfs_iunlock_map_shared(ip, lock);
xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- kmem_free(map, subnex * sizeof(*map));
+ kmem_free(map);
return error;
}
@@ -6088,7 +6096,7 @@ xfs_bmap_get_bp(
tp = cur->bc_tp;
licp = &tp->t_items;
while (!bp && licp != NULL) {
- if (XFS_LIC_ARE_ALL_FREE(licp)) {
+ if (xfs_lic_are_all_free(licp)) {
licp = licp->lic_next;
continue;
}
@@ -6098,11 +6106,11 @@ xfs_bmap_get_bp(
xfs_buf_log_item_t *bip;
xfs_buf_t *lbp;
- if (XFS_LIC_ISFREE(licp, i)) {
+ if (xfs_lic_isfree(licp, i)) {
continue;
}
- lidp = XFS_LIC_SLOT(licp, i);
+ lidp = xfs_lic_slot(licp, i);
lip = lidp->lid_item;
if (lip->li_type != XFS_LI_BUF)
continue;
@@ -6359,13 +6367,9 @@ xfs_bmap_count_blocks(
mp = ip->i_mount;
ifp = XFS_IFORK_PTR(ip, whichfork);
if ( XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS ) {
- if (unlikely(xfs_bmap_count_leaves(ifp, 0,
+ xfs_bmap_count_leaves(ifp, 0,
ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t),
- count) < 0)) {
- XFS_ERROR_REPORT("xfs_bmap_count_blocks(1)",
- XFS_ERRLEVEL_LOW, mp);
- return XFS_ERROR(EFSCORRUPTED);
- }
+ count);
return 0;
}
@@ -6446,13 +6450,7 @@ xfs_bmap_count_tree(
for (;;) {
nextbno = be64_to_cpu(block->bb_rightsib);
numrecs = be16_to_cpu(block->bb_numrecs);
- if (unlikely(xfs_bmap_disk_count_leaves(0,
- block, numrecs, count) < 0)) {
- xfs_trans_brelse(tp, bp);
- XFS_ERROR_REPORT("xfs_bmap_count_tree(2)",
- XFS_ERRLEVEL_LOW, mp);
- return XFS_ERROR(EFSCORRUPTED);
- }
+ xfs_bmap_disk_count_leaves(0, block, numrecs, count);
xfs_trans_brelse(tp, bp);
if (nextbno == NULLFSBLOCK)
break;
@@ -6470,7 +6468,7 @@ xfs_bmap_count_tree(
/*
* Count leaf blocks given a range of extent records.
*/
-STATIC int
+STATIC void
xfs_bmap_count_leaves(
xfs_ifork_t *ifp,
xfs_extnum_t idx,
@@ -6483,14 +6481,13 @@ xfs_bmap_count_leaves(
xfs_bmbt_rec_host_t *frp = xfs_iext_get_ext(ifp, idx + b);
*count += xfs_bmbt_get_blockcount(frp);
}
- return 0;
}
/*
* Count leaf blocks given a range of extent records originally
* in btree format.
*/
-STATIC int
+STATIC void
xfs_bmap_disk_count_leaves(
xfs_extnum_t idx,
xfs_bmbt_block_t *block,
@@ -6504,5 +6501,4 @@ xfs_bmap_disk_count_leaves(
frp = XFS_BTREE_REC_ADDR(xfs_bmbt, block, idx + b);
*count += xfs_bmbt_disk_get_blockcount(frp);
}
- return 0;
}
diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h
index 6ff70cda451..9f3e3a836d1 100644
--- a/fs/xfs/xfs_bmap.h
+++ b/fs/xfs/xfs_bmap.h
@@ -54,12 +54,23 @@ typedef struct xfs_bmap_free_item
/*
* Header for free extent list.
+ *
+ * xbf_low is used by the allocator to activate the lowspace algorithm -
+ * when free space is running low the extent allocator may choose to
+ * allocate an extent from an AG without leaving sufficient space for
+ * a btree split when inserting the new extent. In this case the allocator
+ * will enable the lowspace algorithm which is supposed to allow further
+ * allocations (such as btree splits and newroots) to allocate from
+ * sequential AGs. In order to avoid locking AGs out of order the lowspace
+ * algorithm will start searching for free space from AG 0. If the correct
+ * transaction reservations have been made then this algorithm will eventually
+ * find all the space it needs.
*/
typedef struct xfs_bmap_free
{
xfs_bmap_free_item_t *xbf_first; /* list of to-be-free extents */
int xbf_count; /* count of items on list */
- int xbf_low; /* kludge: alloc in low mode */
+ int xbf_low; /* alloc in low mode */
} xfs_bmap_free_t;
#define XFS_BMAP_MAX_NMAP 4
diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c
index 4f0e849d973..23efad29a5c 100644
--- a/fs/xfs/xfs_bmap_btree.c
+++ b/fs/xfs/xfs_bmap_btree.c
@@ -1493,12 +1493,27 @@ xfs_bmbt_split(
left = XFS_BUF_TO_BMBT_BLOCK(lbp);
args.fsbno = cur->bc_private.b.firstblock;
args.firstblock = args.fsbno;
+ args.minleft = 0;
if (args.fsbno == NULLFSBLOCK) {
args.fsbno = lbno;
args.type = XFS_ALLOCTYPE_START_BNO;
- } else
+ /*
+ * Make sure there is sufficient room left in the AG to
+ * complete a full tree split for an extent insert. If
+ * we are converting the middle part of an extent then
+ * we may need space for two tree splits.
+ *
+ * We are relying on the caller to make the correct block
+ * reservation for this operation to succeed. If the
+ * reservation amount is insufficient then we may fail a
+ * block allocation here and corrupt the filesystem.
+ */
+ args.minleft = xfs_trans_get_block_res(args.tp);
+ } else if (cur->bc_private.b.flist->xbf_low)
+ args.type = XFS_ALLOCTYPE_START_BNO;
+ else
args.type = XFS_ALLOCTYPE_NEAR_BNO;
- args.mod = args.minleft = args.alignment = args.total = args.isfl =
+ args.mod = args.alignment = args.total = args.isfl =
args.userdata = args.minalignslop = 0;
args.minlen = args.maxlen = args.prod = 1;
args.wasdel = cur->bc_private.b.flags & XFS_BTCUR_BPRV_WASDEL;
@@ -1510,6 +1525,21 @@ xfs_bmbt_split(
XFS_BMBT_TRACE_CURSOR(cur, ERROR);
return error;
}
+ if (args.fsbno == NULLFSBLOCK && args.minleft) {
+ /*
+ * Could not find an AG with enough free space to satisfy
+ * a full btree split. Try again without minleft and if
+ * successful activate the lowspace algorithm.
+ */
+ args.fsbno = 0;
+ args.type = XFS_ALLOCTYPE_FIRST_AG;
+ args.minleft = 0;
+ if ((error = xfs_alloc_vextent(&args))) {
+ XFS_BMBT_TRACE_CURSOR(cur, ERROR);
+ return error;
+ }
+ cur->bc_private.b.flist->xbf_low = 1;
+ }
if (args.fsbno == NULLFSBLOCK) {
XFS_BMBT_TRACE_CURSOR(cur, EXIT);
*stat = 0;
@@ -2029,22 +2059,8 @@ xfs_bmbt_increment(
* Insert the current record at the point referenced by cur.
*
* A multi-level split of the tree on insert will invalidate the original
- * cursor. It appears, however, that some callers assume that the cursor is
- * always valid. Hence if we do a multi-level split we need to revalidate the
- * cursor.
- *
- * When a split occurs, we will see a new cursor returned. Use that as a
- * trigger to determine if we need to revalidate the original cursor. If we get
- * a split, then use the original irec to lookup up the path of the record we
- * just inserted.
- *
- * Note that the fact that the btree root is in the inode means that we can
- * have the level of the tree change without a "split" occurring at the root
- * level. What happens is that the root is migrated to an allocated block and
- * the inode root is pointed to it. This means a single split can change the
- * level of the tree (level 2 -> level 3) and invalidate the old cursor. Hence
- * the level change should be accounted as a split so as to correctly trigger a
- * revalidation of the old cursor.
+ * cursor. All callers of this function should assume that the cursor is
+ * no longer valid and revalidate it.
*/
int /* error */
xfs_bmbt_insert(
@@ -2057,14 +2073,11 @@ xfs_bmbt_insert(
xfs_fsblock_t nbno;
xfs_btree_cur_t *ncur;
xfs_bmbt_rec_t nrec;
- xfs_bmbt_irec_t oirec; /* original irec */
xfs_btree_cur_t *pcur;
- int splits = 0;
XFS_BMBT_TRACE_CURSOR(cur, ENTRY);
level = 0;
nbno = NULLFSBLOCK;
- oirec = cur->bc_rec.b;
xfs_bmbt_disk_set_all(&nrec, &cur->bc_rec.b);
ncur = NULL;
pcur = cur;
@@ -2073,13 +2086,11 @@ xfs_bmbt_insert(
&i))) {
if (pcur != cur)
xfs_btree_del_cursor(pcur, XFS_BTREE_ERROR);
- goto error0;
+ XFS_BMBT_TRACE_CURSOR(cur, ERROR);
+ return error;
}
XFS_WANT_CORRUPTED_GOTO(i == 1, error0);
if (pcur != cur && (ncur || nbno == NULLFSBLOCK)) {
- /* allocating a new root is effectively a split */
- if (cur->bc_nlevels != pcur->bc_nlevels)
- splits++;
cur->bc_nlevels = pcur->bc_nlevels;
cur->bc_private.b.allocated +=
pcur->bc_private.b.allocated;
@@ -2093,21 +2104,10 @@ xfs_bmbt_insert(
xfs_btree_del_cursor(pcur, XFS_BTREE_NOERROR);
}
if (ncur) {
- splits++;
pcur = ncur;
ncur = NULL;
}
} while (nbno != NULLFSBLOCK);
-
- if (splits > 1) {
- /* revalidate the old cursor as we had a multi-level split */
- error = xfs_bmbt_lookup_eq(cur, oirec.br_startoff,
- oirec.br_startblock, oirec.br_blockcount, &i);
- if (error)
- goto error0;
- ASSERT(i == 1);
- }
-
XFS_BMBT_TRACE_CURSOR(cur, EXIT);
*stat = i;
return 0;
@@ -2254,7 +2254,9 @@ xfs_bmbt_newroot(
#endif
args.fsbno = be64_to_cpu(*pp);
args.type = XFS_ALLOCTYPE_START_BNO;
- } else
+ } else if (cur->bc_private.b.flist->xbf_low)
+ args.type = XFS_ALLOCTYPE_START_BNO;
+ else
args.type = XFS_ALLOCTYPE_NEAR_BNO;
if ((error = xfs_alloc_vextent(&args))) {
XFS_BMBT_TRACE_CURSOR(cur, ERROR);
diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c
index aeb87ca69fc..cc593a84c34 100644
--- a/fs/xfs/xfs_btree.c
+++ b/fs/xfs/xfs_btree.c
@@ -46,38 +46,11 @@ kmem_zone_t *xfs_btree_cur_zone;
/*
* Btree magic numbers.
*/
-const __uint32_t xfs_magics[XFS_BTNUM_MAX] =
-{
+const __uint32_t xfs_magics[XFS_BTNUM_MAX] = {
XFS_ABTB_MAGIC, XFS_ABTC_MAGIC, XFS_BMAP_MAGIC, XFS_IBT_MAGIC
};
/*
- * Prototypes for internal routines.
- */
-
-/*
- * Checking routine: return maxrecs for the block.
- */
-STATIC int /* number of records fitting in block */
-xfs_btree_maxrecs(
- xfs_btree_cur_t *cur, /* btree cursor */
- xfs_btree_block_t *block);/* generic btree block pointer */
-
-/*
- * Internal routines.
- */
-
-/*
- * Retrieve the block pointer from the cursor at the given level.
- * This may be a bmap btree root or from a buffer.
- */
-STATIC xfs_btree_block_t * /* generic btree block pointer */
-xfs_btree_get_block(
- xfs_btree_cur_t *cur, /* btree cursor */
- int level, /* level in btree */
- struct xfs_buf **bpp); /* buffer containing the block */
-
-/*
* Checking routine: return maxrecs for the block.
*/
STATIC int /* number of records fitting in block */
@@ -457,35 +430,6 @@ xfs_btree_dup_cursor(
}
/*
- * Change the cursor to point to the first record at the given level.
- * Other levels are unaffected.
- */
-int /* success=1, failure=0 */
-xfs_btree_firstrec(
- xfs_btree_cur_t *cur, /* btree cursor */
- int level) /* level to change */
-{
- xfs_btree_block_t *block; /* generic btree block pointer */
- xfs_buf_t *bp; /* buffer containing block */
-
- /*
- * Get the block pointer for this level.
- */
- block = xfs_btree_get_block(cur, level, &bp);
- xfs_btree_check_block(cur, block, level, bp);
- /*
- * It's empty, there is no such record.
- */
- if (!block->bb_h.bb_numrecs)
- return 0;
- /*
- * Set the ptr value to 1, that's the first record/key.
- */
- cur->bc_ptrs[level] = 1;
- return 1;
-}
-
-/*
* Retrieve the block pointer from the cursor at the given level.
* This may be a bmap btree root or from a buffer.
*/
@@ -626,6 +570,13 @@ xfs_btree_init_cursor(
cur->bc_private.a.agbp = agbp;
cur->bc_private.a.agno = agno;
break;
+ case XFS_BTNUM_INO:
+ /*
+ * Inode allocation btree fields.
+ */
+ cur->bc_private.a.agbp = agbp;
+ cur->bc_private.a.agno = agno;
+ break;
case XFS_BTNUM_BMAP:
/*
* Bmap btree fields.
@@ -638,13 +589,6 @@ xfs_btree_init_cursor(
cur->bc_private.b.flags = 0;
cur->bc_private.b.whichfork = whichfork;
break;
- case XFS_BTNUM_INO:
- /*
- * Inode allocation btree fields.
- */
- cur->bc_private.i.agbp = agbp;
- cur->bc_private.i.agno = agno;
- break;
default:
ASSERT(0);
}
@@ -671,6 +615,35 @@ xfs_btree_islastblock(
}
/*
+ * Change the cursor to point to the first record at the given level.
+ * Other levels are unaffected.
+ */
+int /* success=1, failure=0 */
+xfs_btree_firstrec(
+ xfs_btree_cur_t *cur, /* btree cursor */
+ int level) /* level to change */
+{
+ xfs_btree_block_t *block; /* generic btree block pointer */
+ xfs_buf_t *bp; /* buffer containing block */
+
+ /*
+ * Get the block pointer for this level.
+ */
+ block = xfs_btree_get_block(cur, level, &bp);
+ xfs_btree_check_block(cur, block, level, bp);
+ /*
+ * It's empty, there is no such record.
+ */
+ if (!block->bb_h.bb_numrecs)
+ return 0;
+ /*
+ * Set the ptr value to 1, that's the first record/key.
+ */
+ cur->bc_ptrs[level] = 1;
+ return 1;
+}
+
+/*
* Change the cursor to point to the last record in the current block
* at the given level. Other levels are unaffected.
*/
@@ -890,12 +863,12 @@ xfs_btree_readahead_core(
case XFS_BTNUM_INO:
i = XFS_BUF_TO_INOBT_BLOCK(cur->bc_bufs[lev]);
if ((lr & XFS_BTCUR_LEFTRA) && be32_to_cpu(i->bb_leftsib) != NULLAGBLOCK) {
- xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.i.agno,
+ xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.a.agno,
be32_to_cpu(i->bb_leftsib), 1);
rval++;
}
if ((lr & XFS_BTCUR_RIGHTRA) && be32_to_cpu(i->bb_rightsib) != NULLAGBLOCK) {
- xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.i.agno,
+ xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.a.agno,
be32_to_cpu(i->bb_rightsib), 1);
rval++;
}
diff --git a/fs/xfs/xfs_btree.h b/fs/xfs/xfs_btree.h
index 7440b78f9ce..1f528a2a375 100644
--- a/fs/xfs/xfs_btree.h
+++ b/fs/xfs/xfs_btree.h
@@ -158,8 +158,8 @@ typedef struct xfs_btree_cur
__uint8_t bc_blocklog; /* log2(blocksize) of btree blocks */
xfs_btnum_t bc_btnum; /* identifies which btree type */
union {
- struct { /* needed for BNO, CNT */
- struct xfs_buf *agbp; /* agf buffer pointer */
+ struct { /* needed for BNO, CNT, INO */
+ struct xfs_buf *agbp; /* agf/agi buffer pointer */
xfs_agnumber_t agno; /* ag number */
} a;
struct { /* needed for BMAP */
@@ -172,10 +172,6 @@ typedef struct xfs_btree_cur
char flags; /* flags */
#define XFS_BTCUR_BPRV_WASDEL 1 /* was delayed */
} b;
- struct { /* needed for INO */
- struct xfs_buf *agbp; /* agi buffer pointer */
- xfs_agnumber_t agno; /* ag number */
- } i;
} bc_private; /* per-btree type data */
} xfs_btree_cur_t;
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index 53a71c62025..002fc2617c8 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -732,12 +732,13 @@ xfs_buf_item_init(
bip->bli_item.li_ops = &xfs_buf_item_ops;
bip->bli_item.li_mountp = mp;
bip->bli_buf = bp;
+ xfs_buf_hold(bp);
bip->bli_format.blf_type = XFS_LI_BUF;
bip->bli_format.blf_blkno = (__int64_t)XFS_BUF_ADDR(bp);
bip->bli_format.blf_len = (ushort)BTOBB(XFS_BUF_COUNT(bp));
bip->bli_format.blf_map_size = map_size;
#ifdef XFS_BLI_TRACE
- bip->bli_trace = ktrace_alloc(XFS_BLI_TRACE_SIZE, KM_SLEEP);
+ bip->bli_trace = ktrace_alloc(XFS_BLI_TRACE_SIZE, KM_NOFS);
#endif
#ifdef XFS_TRANS_DEBUG
@@ -867,6 +868,21 @@ xfs_buf_item_dirty(
return (bip->bli_flags & XFS_BLI_DIRTY);
}
+STATIC void
+xfs_buf_item_free(
+ xfs_buf_log_item_t *bip)
+{
+#ifdef XFS_TRANS_DEBUG
+ kmem_free(bip->bli_orig);
+ kmem_free(bip->bli_logged);
+#endif /* XFS_TRANS_DEBUG */
+
+#ifdef XFS_BLI_TRACE
+ ktrace_free(bip->bli_trace);
+#endif
+ kmem_zone_free(xfs_buf_item_zone, bip);
+}
+
/*
* This is called when the buf log item is no longer needed. It should
* free the buf log item associated with the given buffer and clear
@@ -887,18 +903,8 @@ xfs_buf_item_relse(
(XFS_BUF_IODONE_FUNC(bp) != NULL)) {
XFS_BUF_CLR_IODONE_FUNC(bp);
}
-
-#ifdef XFS_TRANS_DEBUG
- kmem_free(bip->bli_orig, XFS_BUF_COUNT(bp));
- bip->bli_orig = NULL;
- kmem_free(bip->bli_logged, XFS_BUF_COUNT(bp) / NBBY);
- bip->bli_logged = NULL;
-#endif /* XFS_TRANS_DEBUG */
-
-#ifdef XFS_BLI_TRACE
- ktrace_free(bip->bli_trace);
-#endif
- kmem_zone_free(xfs_buf_item_zone, bip);
+ xfs_buf_rele(bp);
+ xfs_buf_item_free(bip);
}
@@ -1056,7 +1062,7 @@ xfs_buf_iodone_callbacks(
anyway. */
XFS_BUF_SET_BRELSE_FUNC(bp,xfs_buf_error_relse);
XFS_BUF_DONE(bp);
- XFS_BUF_V_IODONESEMA(bp);
+ XFS_BUF_FINISH_IOWAIT(bp);
}
return;
}
@@ -1120,6 +1126,7 @@ xfs_buf_iodone(
ASSERT(bip->bli_buf == bp);
+ xfs_buf_rele(bp);
mp = bip->bli_item.li_mountp;
/*
@@ -1136,18 +1143,7 @@ xfs_buf_iodone(
* xfs_trans_delete_ail() drops the AIL lock.
*/
xfs_trans_delete_ail(mp, (xfs_log_item_t *)bip);
-
-#ifdef XFS_TRANS_DEBUG
- kmem_free(bip->bli_orig, XFS_BUF_COUNT(bp));
- bip->bli_orig = NULL;
- kmem_free(bip->bli_logged, XFS_BUF_COUNT(bp) / NBBY);
- bip->bli_logged = NULL;
-#endif /* XFS_TRANS_DEBUG */
-
-#ifdef XFS_BLI_TRACE
- ktrace_free(bip->bli_trace);
-#endif
- kmem_zone_free(xfs_buf_item_zone, bip);
+ xfs_buf_item_free(bip);
}
#if defined(XFS_BLI_TRACE)
diff --git a/fs/xfs/xfs_clnt.h b/fs/xfs/xfs_clnt.h
index d5d1e60ee22..d2ce5dd70d8 100644
--- a/fs/xfs/xfs_clnt.h
+++ b/fs/xfs/xfs_clnt.h
@@ -78,6 +78,7 @@ struct xfs_mount_args {
#define XFSMNT_IOSIZE 0x00002000 /* optimize for I/O size */
#define XFSMNT_OSYNCISOSYNC 0x00004000 /* o_sync is REALLY o_sync */
/* (osyncisdsync is default) */
+#define XFSMNT_NOATTR2 0x00008000 /* turn off ATTR2 EA format */
#define XFSMNT_32BITINODES 0x00200000 /* restrict inodes to 32
* bits of address space */
#define XFSMNT_GQUOTA 0x00400000 /* group quota accounting */
diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c
index 021a8f7e563..9e561a9cefc 100644
--- a/fs/xfs/xfs_da_btree.c
+++ b/fs/xfs/xfs_da_btree.c
@@ -1431,7 +1431,7 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
}
if (level < 0) {
*result = XFS_ERROR(ENOENT); /* we're out of our tree */
- ASSERT(args->oknoent);
+ ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
return(0);
}
@@ -1530,6 +1530,28 @@ xfs_da_hashname(const uchar_t *name, int namelen)
}
}
+enum xfs_dacmp
+xfs_da_compname(
+ struct xfs_da_args *args,
+ const char *name,
+ int len)
+{
+ return (args->namelen == len && memcmp(args->name, name, len) == 0) ?
+ XFS_CMP_EXACT : XFS_CMP_DIFFERENT;
+}
+
+static xfs_dahash_t
+xfs_default_hashname(
+ struct xfs_name *name)
+{
+ return xfs_da_hashname(name->name, name->len);
+}
+
+const struct xfs_nameops xfs_default_nameops = {
+ .hashname = xfs_default_hashname,
+ .compname = xfs_da_compname
+};
+
/*
* Add a block to the btree ahead of the file.
* Return the new block number to the caller.
@@ -1598,7 +1620,7 @@ xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno)
args->firstblock, args->total,
&mapp[mapi], &nmap, args->flist,
NULL))) {
- kmem_free(mapp, sizeof(*mapp) * count);
+ kmem_free(mapp);
return error;
}
if (nmap < 1)
@@ -1620,11 +1642,11 @@ xfs_da_grow_inode(xfs_da_args_t *args, xfs_dablk_t *new_blkno)
mapp[mapi - 1].br_startoff + mapp[mapi - 1].br_blockcount !=
bno + count) {
if (mapp != &map)
- kmem_free(mapp, sizeof(*mapp) * count);
+ kmem_free(mapp);
return XFS_ERROR(ENOSPC);
}
if (mapp != &map)
- kmem_free(mapp, sizeof(*mapp) * count);
+ kmem_free(mapp);
*new_blkno = (xfs_dablk_t)bno;
return 0;
}
@@ -2090,10 +2112,10 @@ xfs_da_do_buf(
}
}
if (bplist) {
- kmem_free(bplist, sizeof(*bplist) * nmap);
+ kmem_free(bplist);
}
if (mapp != &map) {
- kmem_free(mapp, sizeof(*mapp) * nfsb);
+ kmem_free(mapp);
}
if (bpp)
*bpp = rbp;
@@ -2102,11 +2124,11 @@ exit1:
if (bplist) {
for (i = 0; i < nbplist; i++)
xfs_trans_brelse(trans, bplist[i]);
- kmem_free(bplist, sizeof(*bplist) * nmap);
+ kmem_free(bplist);
}
exit0:
if (mapp != &map)
- kmem_free(mapp, sizeof(*mapp) * nfsb);
+ kmem_free(mapp);
if (bpp)
*bpp = NULL;
return error;
@@ -2218,7 +2240,7 @@ xfs_da_state_free(xfs_da_state_t *state)
#ifdef XFS_DABUF_DEBUG
xfs_dabuf_t *xfs_dabuf_global_list;
-spinlock_t xfs_dabuf_global_lock;
+static DEFINE_SPINLOCK(xfs_dabuf_global_lock);
#endif
/*
@@ -2315,7 +2337,7 @@ xfs_da_buf_done(xfs_dabuf_t *dabuf)
if (dabuf->dirty)
xfs_da_buf_clean(dabuf);
if (dabuf->nbuf > 1)
- kmem_free(dabuf->data, BBTOB(dabuf->bbcount));
+ kmem_free(dabuf->data);
#ifdef XFS_DABUF_DEBUG
{
spin_lock(&xfs_dabuf_global_lock);
@@ -2332,7 +2354,7 @@ xfs_da_buf_done(xfs_dabuf_t *dabuf)
if (dabuf->nbuf == 1)
kmem_zone_free(xfs_dabuf_zone, dabuf);
else
- kmem_free(dabuf, XFS_DA_BUF_SIZE(dabuf->nbuf));
+ kmem_free(dabuf);
}
/*
@@ -2403,7 +2425,7 @@ xfs_da_brelse(xfs_trans_t *tp, xfs_dabuf_t *dabuf)
for (i = 0; i < nbuf; i++)
xfs_trans_brelse(tp, bplist[i]);
if (bplist != &bp)
- kmem_free(bplist, nbuf * sizeof(*bplist));
+ kmem_free(bplist);
}
/*
@@ -2429,7 +2451,7 @@ xfs_da_binval(xfs_trans_t *tp, xfs_dabuf_t *dabuf)
for (i = 0; i < nbuf; i++)
xfs_trans_binval(tp, bplist[i]);
if (bplist != &bp)
- kmem_free(bplist, nbuf * sizeof(*bplist));
+ kmem_free(bplist);
}
/*
diff --git a/fs/xfs/xfs_da_btree.h b/fs/xfs/xfs_da_btree.h
index 7facf86f74f..8be0b00ede9 100644
--- a/fs/xfs/xfs_da_btree.h
+++ b/fs/xfs/xfs_da_btree.h
@@ -99,6 +99,15 @@ typedef struct xfs_da_node_entry xfs_da_node_entry_t;
*========================================================================*/
/*
+ * Search comparison results
+ */
+enum xfs_dacmp {
+ XFS_CMP_DIFFERENT, /* names are completely different */
+ XFS_CMP_EXACT, /* names are exactly the same */
+ XFS_CMP_CASE /* names are same but differ in case */
+};
+
+/*
* Structure to ease passing around component names.
*/
typedef struct xfs_da_args {
@@ -123,13 +132,20 @@ typedef struct xfs_da_args {
int index2; /* index of 2nd attr in blk */
xfs_dablk_t rmtblkno2; /* remote attr value starting blkno */
int rmtblkcnt2; /* remote attr value block count */
- unsigned char justcheck; /* T/F: check for ok with no space */
- unsigned char rename; /* T/F: this is an atomic rename op */
- unsigned char addname; /* T/F: this is an add operation */
- unsigned char oknoent; /* T/F: ok to return ENOENT, else die */
+ int op_flags; /* operation flags */
+ enum xfs_dacmp cmpresult; /* name compare result for lookups */
} xfs_da_args_t;
/*
+ * Operation flags:
+ */
+#define XFS_DA_OP_JUSTCHECK 0x0001 /* check for ok with no space */
+#define XFS_DA_OP_RENAME 0x0002 /* this is an atomic rename op */
+#define XFS_DA_OP_ADDNAME 0x0004 /* this is an add operation */
+#define XFS_DA_OP_OKNOENT 0x0008 /* lookup/add op, ENOENT ok, else die */
+#define XFS_DA_OP_CILOOKUP 0x0010 /* lookup to return CI name if found */
+
+/*
* Structure to describe buffer(s) for a block.
* This is needed in the directory version 2 format case, when
* multiple non-contiguous fsblocks might be needed to cover one
@@ -201,6 +217,14 @@ typedef struct xfs_da_state {
(uint)(XFS_DA_LOGOFF(BASE, ADDR)), \
(uint)(XFS_DA_LOGOFF(BASE, ADDR)+(SIZE)-1)
+/*
+ * Name ops for directory and/or attr name operations
+ */
+struct xfs_nameops {
+ xfs_dahash_t (*hashname)(struct xfs_name *);
+ enum xfs_dacmp (*compname)(struct xfs_da_args *, const char *, int);
+};
+
#ifdef __KERNEL__
/*========================================================================
@@ -249,6 +273,10 @@ int xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno,
xfs_dabuf_t *dead_buf);
uint xfs_da_hashname(const uchar_t *name_string, int name_length);
+enum xfs_dacmp xfs_da_compname(struct xfs_da_args *args,
+ const char *name, int len);
+
+
xfs_da_state_t *xfs_da_state_alloc(void);
void xfs_da_state_free(xfs_da_state_t *state);
diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c
index 5f3647cb988..75b0cd4da0e 100644
--- a/fs/xfs/xfs_dfrag.c
+++ b/fs/xfs/xfs_dfrag.c
@@ -116,7 +116,7 @@ xfs_swapext(
out_put_file:
fput(file);
out_free_sxp:
- kmem_free(sxp, sizeof(xfs_swapext_t));
+ kmem_free(sxp);
out:
return error;
}
@@ -128,10 +128,8 @@ xfs_swap_extents(
xfs_swapext_t *sxp)
{
xfs_mount_t *mp;
- xfs_inode_t *ips[2];
xfs_trans_t *tp;
xfs_bstat_t *sbp = &sxp->sx_stat;
- bhv_vnode_t *vp, *tvp;
xfs_ifork_t *tempifp, *ifp, *tifp;
int ilf_fields, tilf_fields;
static uint lock_flags = XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL;
@@ -150,19 +148,15 @@ xfs_swap_extents(
}
sbp = &sxp->sx_stat;
- vp = XFS_ITOV(ip);
- tvp = XFS_ITOV(tip);
-
- /* Lock in i_ino order */
- if (ip->i_ino < tip->i_ino) {
- ips[0] = ip;
- ips[1] = tip;
- } else {
- ips[0] = tip;
- ips[1] = ip;
- }
- xfs_lock_inodes(ips, 2, lock_flags);
+ /*
+ * we have to do two separate lock calls here to keep lockdep
+ * happy. If we try to get all the locks in one call, lock will
+ * report false positives when we drop the ILOCK and regain them
+ * below.
+ */
+ xfs_lock_two_inodes(ip, tip, XFS_IOLOCK_EXCL);
+ xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
locked = 1;
/* Verify that both files have the same format */
@@ -184,7 +178,7 @@ xfs_swap_extents(
goto error0;
}
- if (VN_CACHED(tvp) != 0) {
+ if (VN_CACHED(VFS_I(tip)) != 0) {
xfs_inval_cached_trace(tip, 0, -1, 0, -1);
error = xfs_flushinval_pages(tip, 0, -1,
FI_REMAPF_LOCKED);
@@ -193,7 +187,7 @@ xfs_swap_extents(
}
/* Verify O_DIRECT for ftmp */
- if (VN_CACHED(tvp) != 0) {
+ if (VN_CACHED(VFS_I(tip)) != 0) {
error = XFS_ERROR(EINVAL);
goto error0;
}
@@ -237,7 +231,7 @@ xfs_swap_extents(
* vop_read (or write in the case of autogrow) they block on the iolock
* until we have switched the extents.
*/
- if (VN_MAPPED(vp)) {
+ if (VN_MAPPED(VFS_I(ip))) {
error = XFS_ERROR(EBUSY);
goto error0;
}
@@ -265,7 +259,7 @@ xfs_swap_extents(
locked = 0;
goto error0;
}
- xfs_lock_inodes(ips, 2, XFS_ILOCK_EXCL);
+ xfs_lock_two_inodes(ip, tip, XFS_ILOCK_EXCL);
/*
* Count the number of extended attribute blocks
@@ -350,15 +344,11 @@ xfs_swap_extents(
break;
}
- /*
- * Increment vnode ref counts since xfs_trans_commit &
- * xfs_trans_cancel will both unlock the inodes and
- * decrement the associated ref counts.
- */
- VN_HOLD(vp);
- VN_HOLD(tvp);
+ IHOLD(ip);
xfs_trans_ijoin(tp, ip, lock_flags);
+
+ IHOLD(tip);
xfs_trans_ijoin(tp, tip, lock_flags);
xfs_trans_log_inode(tp, ip, ilf_fields);
@@ -381,6 +371,6 @@ xfs_swap_extents(
xfs_iunlock(tip, lock_flags);
}
if (tempifp != NULL)
- kmem_free(tempifp, sizeof(xfs_ifork_t));
+ kmem_free(tempifp);
return error;
}
diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c
index 7cb26529766..80e0dc51361 100644
--- a/fs/xfs/xfs_dir2.c
+++ b/fs/xfs/xfs_dir2.c
@@ -46,6 +46,54 @@
struct xfs_name xfs_name_dotdot = {"..", 2};
+extern const struct xfs_nameops xfs_default_nameops;
+
+/*
+ * ASCII case-insensitive (ie. A-Z) support for directories that was
+ * used in IRIX.
+ */
+STATIC xfs_dahash_t
+xfs_ascii_ci_hashname(
+ struct xfs_name *name)
+{
+ xfs_dahash_t hash;
+ int i;
+
+ for (i = 0, hash = 0; i < name->len; i++)
+ hash = tolower(name->name[i]) ^ rol32(hash, 7);
+
+ return hash;
+}
+
+STATIC enum xfs_dacmp
+xfs_ascii_ci_compname(
+ struct xfs_da_args *args,
+ const char *name,
+ int len)
+{
+ enum xfs_dacmp result;
+ int i;
+
+ if (args->namelen != len)
+ return XFS_CMP_DIFFERENT;
+
+ result = XFS_CMP_EXACT;
+ for (i = 0; i < len; i++) {
+ if (args->name[i] == name[i])
+ continue;
+ if (tolower(args->name[i]) != tolower(name[i]))
+ return XFS_CMP_DIFFERENT;
+ result = XFS_CMP_CASE;
+ }
+
+ return result;
+}
+
+static struct xfs_nameops xfs_ascii_ci_nameops = {
+ .hashname = xfs_ascii_ci_hashname,
+ .compname = xfs_ascii_ci_compname,
+};
+
void
xfs_dir_mount(
xfs_mount_t *mp)
@@ -65,6 +113,10 @@ xfs_dir_mount(
(mp->m_dirblksize - (uint)sizeof(xfs_da_node_hdr_t)) /
(uint)sizeof(xfs_da_node_entry_t);
mp->m_dir_magicpct = (mp->m_dirblksize * 37) / 100;
+ if (xfs_sb_version_hasasciici(&mp->m_sb))
+ mp->m_dirnameops = &xfs_ascii_ci_nameops;
+ else
+ mp->m_dirnameops = &xfs_default_nameops;
}
/*
@@ -162,9 +214,10 @@ xfs_dir_createname(
return rval;
XFS_STATS_INC(xs_dir_create);
+ memset(&args, 0, sizeof(xfs_da_args_t));
args.name = name->name;
args.namelen = name->len;
- args.hashval = xfs_da_hashname(name->name, name->len);
+ args.hashval = dp->i_mount->m_dirnameops->hashname(name);
args.inumber = inum;
args.dp = dp;
args.firstblock = first;
@@ -172,8 +225,7 @@ xfs_dir_createname(
args.total = total;
args.whichfork = XFS_DATA_FORK;
args.trans = tp;
- args.justcheck = 0;
- args.addname = args.oknoent = 1;
+ args.op_flags = XFS_DA_OP_ADDNAME | XFS_DA_OP_OKNOENT;
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
rval = xfs_dir2_sf_addname(&args);
@@ -191,14 +243,43 @@ xfs_dir_createname(
}
/*
+ * If doing a CI lookup and case-insensitive match, dup actual name into
+ * args.value. Return EEXIST for success (ie. name found) or an error.
+ */
+int
+xfs_dir_cilookup_result(
+ struct xfs_da_args *args,
+ const char *name,
+ int len)
+{
+ if (args->cmpresult == XFS_CMP_DIFFERENT)
+ return ENOENT;
+ if (args->cmpresult != XFS_CMP_CASE ||
+ !(args->op_flags & XFS_DA_OP_CILOOKUP))
+ return EEXIST;
+
+ args->value = kmem_alloc(len, KM_MAYFAIL);
+ if (!args->value)
+ return ENOMEM;
+
+ memcpy(args->value, name, len);
+ args->valuelen = len;
+ return EEXIST;
+}
+
+/*
* Lookup a name in a directory, give back the inode number.
+ * If ci_name is not NULL, returns the actual name in ci_name if it differs
+ * to name, or ci_name->name is set to NULL for an exact match.
*/
+
int
xfs_dir_lookup(
xfs_trans_t *tp,
xfs_inode_t *dp,
struct xfs_name *name,
- xfs_ino_t *inum) /* out: inode number */
+ xfs_ino_t *inum, /* out: inode number */
+ struct xfs_name *ci_name) /* out: actual name if CI match */
{
xfs_da_args_t args;
int rval;
@@ -206,15 +287,17 @@ xfs_dir_lookup(
ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR);
XFS_STATS_INC(xs_dir_lookup);
- memset(&args, 0, sizeof(xfs_da_args_t));
+ memset(&args, 0, sizeof(xfs_da_args_t));
args.name = name->name;
args.namelen = name->len;
- args.hashval = xfs_da_hashname(name->name, name->len);
+ args.hashval = dp->i_mount->m_dirnameops->hashname(name);
args.dp = dp;
args.whichfork = XFS_DATA_FORK;
args.trans = tp;
- args.oknoent = 1;
+ args.op_flags = XFS_DA_OP_OKNOENT;
+ if (ci_name)
+ args.op_flags |= XFS_DA_OP_CILOOKUP;
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
rval = xfs_dir2_sf_lookup(&args);
@@ -230,8 +313,13 @@ xfs_dir_lookup(
rval = xfs_dir2_node_lookup(&args);
if (rval == EEXIST)
rval = 0;
- if (rval == 0)
+ if (!rval) {
*inum = args.inumber;
+ if (ci_name) {
+ ci_name->name = args.value;
+ ci_name->len = args.valuelen;
+ }
+ }
return rval;
}
@@ -255,9 +343,10 @@ xfs_dir_removename(
ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR);
XFS_STATS_INC(xs_dir_remove);
+ memset(&args, 0, sizeof(xfs_da_args_t));
args.name = name->name;
args.namelen = name->len;
- args.hashval = xfs_da_hashname(name->name, name->len);
+ args.hashval = dp->i_mount->m_dirnameops->hashname(name);
args.inumber = ino;
args.dp = dp;
args.firstblock = first;
@@ -265,7 +354,6 @@ xfs_dir_removename(
args.total = total;
args.whichfork = XFS_DATA_FORK;
args.trans = tp;
- args.justcheck = args.addname = args.oknoent = 0;
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
rval = xfs_dir2_sf_removename(&args);
@@ -338,9 +426,10 @@ xfs_dir_replace(
if ((rval = xfs_dir_ino_validate(tp->t_mountp, inum)))
return rval;
+ memset(&args, 0, sizeof(xfs_da_args_t));
args.name = name->name;
args.namelen = name->len;
- args.hashval = xfs_da_hashname(name->name, name->len);
+ args.hashval = dp->i_mount->m_dirnameops->hashname(name);
args.inumber = inum;
args.dp = dp;
args.firstblock = first;
@@ -348,7 +437,6 @@ xfs_dir_replace(
args.total = total;
args.whichfork = XFS_DATA_FORK;
args.trans = tp;
- args.justcheck = args.addname = args.oknoent = 0;
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
rval = xfs_dir2_sf_replace(&args);
@@ -384,15 +472,16 @@ xfs_dir_canenter(
return 0;
ASSERT((dp->i_d.di_mode & S_IFMT) == S_IFDIR);
- memset(&args, 0, sizeof(xfs_da_args_t));
+ memset(&args, 0, sizeof(xfs_da_args_t));
args.name = name->name;
args.namelen = name->len;
- args.hashval = xfs_da_hashname(name->name, name->len);
+ args.hashval = dp->i_mount->m_dirnameops->hashname(name);
args.dp = dp;
args.whichfork = XFS_DATA_FORK;
args.trans = tp;
- args.justcheck = args.addname = args.oknoent = 1;
+ args.op_flags = XFS_DA_OP_JUSTCHECK | XFS_DA_OP_ADDNAME |
+ XFS_DA_OP_OKNOENT;
if (dp->i_d.di_format == XFS_DINODE_FMT_LOCAL)
rval = xfs_dir2_sf_addname(&args);
@@ -493,7 +582,7 @@ xfs_dir2_grow_inode(
args->firstblock, args->total,
&mapp[mapi], &nmap, args->flist,
NULL))) {
- kmem_free(mapp, sizeof(*mapp) * count);
+ kmem_free(mapp);
return error;
}
if (nmap < 1)
@@ -525,14 +614,14 @@ xfs_dir2_grow_inode(
mapp[mapi - 1].br_startoff + mapp[mapi - 1].br_blockcount !=
bno + count) {
if (mapp != &map)
- kmem_free(mapp, sizeof(*mapp) * count);
+ kmem_free(mapp);
return XFS_ERROR(ENOSPC);
}
/*
* Done with the temporary mapping table.
*/
if (mapp != &map)
- kmem_free(mapp, sizeof(*mapp) * count);
+ kmem_free(mapp);
*dbp = xfs_dir2_da_to_db(mp, (xfs_dablk_t)bno);
/*
* Update file's size if this is the data space and it grew.
diff --git a/fs/xfs/xfs_dir2.h b/fs/xfs/xfs_dir2.h
index 6392f939029..1d9ef96f33a 100644
--- a/fs/xfs/xfs_dir2.h
+++ b/fs/xfs/xfs_dir2.h
@@ -74,7 +74,8 @@ extern int xfs_dir_createname(struct xfs_trans *tp, struct xfs_inode *dp,
xfs_fsblock_t *first,
struct xfs_bmap_free *flist, xfs_extlen_t tot);
extern int xfs_dir_lookup(struct xfs_trans *tp, struct xfs_inode *dp,
- struct xfs_name *name, xfs_ino_t *inum);
+ struct xfs_name *name, xfs_ino_t *inum,
+ struct xfs_name *ci_name);
extern int xfs_dir_removename(struct xfs_trans *tp, struct xfs_inode *dp,
struct xfs_name *name, xfs_ino_t ino,
xfs_fsblock_t *first,
@@ -99,4 +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);
+
#endif /* __XFS_DIR2_H__ */
diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c
index fb5a556725b..e2fa0a1d8e9 100644
--- a/fs/xfs/xfs_dir2_block.c
+++ b/fs/xfs/xfs_dir2_block.c
@@ -215,7 +215,7 @@ xfs_dir2_block_addname(
/*
* If this isn't a real add, we're done with the buffer.
*/
- if (args->justcheck)
+ if (args->op_flags & XFS_DA_OP_JUSTCHECK)
xfs_da_brelse(tp, bp);
/*
* If we don't have space for the new entry & leaf ...
@@ -225,7 +225,7 @@ xfs_dir2_block_addname(
* Not trying to actually do anything, or don't have
* a space reservation: return no-space.
*/
- if (args->justcheck || args->total == 0)
+ if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || args->total == 0)
return XFS_ERROR(ENOSPC);
/*
* Convert to the next larger format.
@@ -240,7 +240,7 @@ xfs_dir2_block_addname(
/*
* Just checking, and it would work, so say so.
*/
- if (args->justcheck)
+ if (args->op_flags & XFS_DA_OP_JUSTCHECK)
return 0;
needlog = needscan = 0;
/*
@@ -610,14 +610,15 @@ xfs_dir2_block_lookup(
/*
* Get the offset from the leaf entry, to point to the data.
*/
- dep = (xfs_dir2_data_entry_t *)
- ((char *)block + xfs_dir2_dataptr_to_off(mp, be32_to_cpu(blp[ent].address)));
+ dep = (xfs_dir2_data_entry_t *)((char *)block +
+ xfs_dir2_dataptr_to_off(mp, be32_to_cpu(blp[ent].address)));
/*
- * Fill in inode number, release the block.
+ * Fill in inode number, CI name if appropriate, release the block.
*/
args->inumber = be64_to_cpu(dep->inumber);
+ error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
xfs_da_brelse(args->trans, bp);
- return XFS_ERROR(EEXIST);
+ return XFS_ERROR(error);
}
/*
@@ -643,6 +644,7 @@ xfs_dir2_block_lookup_int(
int mid; /* binary search current idx */
xfs_mount_t *mp; /* filesystem mount point */
xfs_trans_t *tp; /* transaction pointer */
+ enum xfs_dacmp cmp; /* comparison result */
dp = args->dp;
tp = args->trans;
@@ -673,7 +675,7 @@ xfs_dir2_block_lookup_int(
else
high = mid - 1;
if (low > high) {
- ASSERT(args->oknoent);
+ ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
xfs_da_brelse(tp, bp);
return XFS_ERROR(ENOENT);
}
@@ -697,20 +699,31 @@ xfs_dir2_block_lookup_int(
dep = (xfs_dir2_data_entry_t *)
((char *)block + xfs_dir2_dataptr_to_off(mp, addr));
/*
- * Compare, if it's right give back buffer & entry number.
+ * Compare name and if it's an exact match, return the index
+ * and buffer. If it's the first case-insensitive match, store
+ * the index and buffer and continue looking for an exact match.
*/
- if (dep->namelen == args->namelen &&
- dep->name[0] == args->name[0] &&
- memcmp(dep->name, args->name, args->namelen) == 0) {
+ cmp = mp->m_dirnameops->compname(args, dep->name, dep->namelen);
+ if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
+ args->cmpresult = cmp;
*bpp = bp;
*entno = mid;
- return 0;
+ if (cmp == XFS_CMP_EXACT)
+ return 0;
}
- } while (++mid < be32_to_cpu(btp->count) && be32_to_cpu(blp[mid].hashval) == hash);
+ } while (++mid < be32_to_cpu(btp->count) &&
+ be32_to_cpu(blp[mid].hashval) == hash);
+
+ ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
+ /*
+ * Here, we can only be doing a lookup (not a rename or replace).
+ * If a case-insensitive match was found earlier, return success.
+ */
+ if (args->cmpresult == XFS_CMP_CASE)
+ return 0;
/*
* No match, release the buffer and return ENOENT.
*/
- ASSERT(args->oknoent);
xfs_da_brelse(tp, bp);
return XFS_ERROR(ENOENT);
}
@@ -1033,6 +1046,7 @@ xfs_dir2_sf_to_block(
xfs_dir2_sf_t *sfp; /* shortform structure */
__be16 *tagp; /* end of data entry */
xfs_trans_t *tp; /* transaction pointer */
+ struct xfs_name name;
xfs_dir2_trace_args("sf_to_block", args);
dp = args->dp;
@@ -1071,7 +1085,7 @@ xfs_dir2_sf_to_block(
*/
error = xfs_dir2_grow_inode(args, XFS_DIR2_DATA_SPACE, &blkno);
if (error) {
- kmem_free(buf, buf_len);
+ kmem_free(buf);
return error;
}
/*
@@ -1079,7 +1093,7 @@ xfs_dir2_sf_to_block(
*/
error = xfs_dir2_data_init(args, blkno, &bp);
if (error) {
- kmem_free(buf, buf_len);
+ kmem_free(buf);
return error;
}
block = bp->data;
@@ -1187,8 +1201,10 @@ xfs_dir2_sf_to_block(
tagp = xfs_dir2_data_entry_tag_p(dep);
*tagp = cpu_to_be16((char *)dep - (char *)block);
xfs_dir2_data_log_entry(tp, bp, dep);
- blp[2 + i].hashval = cpu_to_be32(xfs_da_hashname(
- (char *)sfep->name, sfep->namelen));
+ name.name = sfep->name;
+ name.len = sfep->namelen;
+ blp[2 + i].hashval = cpu_to_be32(mp->m_dirnameops->
+ hashname(&name));
blp[2 + i].address = cpu_to_be32(xfs_dir2_byte_to_dataptr(mp,
(char *)dep - (char *)block));
offset = (int)((char *)(tagp + 1) - (char *)block);
@@ -1198,7 +1214,7 @@ xfs_dir2_sf_to_block(
sfep = xfs_dir2_sf_nextentry(sfp, sfep);
}
/* Done with the temporary buffer */
- kmem_free(buf, buf_len);
+ kmem_free(buf);
/*
* Sort the leaf entries by hash value.
*/
diff --git a/fs/xfs/xfs_dir2_data.c b/fs/xfs/xfs_dir2_data.c
index fb8c9e08b23..498f8d69433 100644
--- a/fs/xfs/xfs_dir2_data.c
+++ b/fs/xfs/xfs_dir2_data.c
@@ -65,6 +65,7 @@ xfs_dir2_data_check(
xfs_mount_t *mp; /* filesystem mount point */
char *p; /* current data position */
int stale; /* count of stale leaves */
+ struct xfs_name name;
mp = dp->i_mount;
d = bp->data;
@@ -140,7 +141,9 @@ xfs_dir2_data_check(
addr = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
(xfs_dir2_data_aoff_t)
((char *)dep - (char *)d));
- hash = xfs_da_hashname((char *)dep->name, dep->namelen);
+ name.name = dep->name;
+ name.len = dep->namelen;
+ hash = mp->m_dirnameops->hashname(&name);
for (i = 0; i < be32_to_cpu(btp->count); i++) {
if (be32_to_cpu(lep[i].address) == addr &&
be32_to_cpu(lep[i].hashval) == hash)
diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c
index bc52b803d79..93535992cb6 100644
--- a/fs/xfs/xfs_dir2_leaf.c
+++ b/fs/xfs/xfs_dir2_leaf.c
@@ -263,20 +263,21 @@ xfs_dir2_leaf_addname(
* If we don't have enough free bytes but we can make enough
* by compacting out stale entries, we'll do that.
*/
- if ((char *)bestsp - (char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] < needbytes &&
- be16_to_cpu(leaf->hdr.stale) > 1) {
+ if ((char *)bestsp - (char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] <
+ needbytes && be16_to_cpu(leaf->hdr.stale) > 1) {
compact = 1;
}
/*
* Otherwise if we don't have enough free bytes we need to
* convert to node form.
*/
- else if ((char *)bestsp - (char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] <
- needbytes) {
+ else if ((char *)bestsp - (char *)&leaf->ents[be16_to_cpu(
+ leaf->hdr.count)] < needbytes) {
/*
* Just checking or no space reservation, give up.
*/
- if (args->justcheck || args->total == 0) {
+ if ((args->op_flags & XFS_DA_OP_JUSTCHECK) ||
+ args->total == 0) {
xfs_da_brelse(tp, lbp);
return XFS_ERROR(ENOSPC);
}
@@ -301,7 +302,7 @@ xfs_dir2_leaf_addname(
* If just checking, then it will fit unless we needed to allocate
* a new data block.
*/
- if (args->justcheck) {
+ if (args->op_flags & XFS_DA_OP_JUSTCHECK) {
xfs_da_brelse(tp, lbp);
return use_block == -1 ? XFS_ERROR(ENOSPC) : 0;
}
@@ -1110,7 +1111,7 @@ xfs_dir2_leaf_getdents(
*offset = XFS_DIR2_MAX_DATAPTR;
else
*offset = xfs_dir2_byte_to_dataptr(mp, curoff);
- kmem_free(map, map_size * sizeof(*map));
+ kmem_free(map);
if (bp)
xfs_da_brelse(NULL, bp);
return error;
@@ -1298,12 +1299,13 @@ xfs_dir2_leaf_lookup(
((char *)dbp->data +
xfs_dir2_dataptr_to_off(dp->i_mount, be32_to_cpu(lep->address)));
/*
- * Return the found inode number.
+ * Return the found inode number & CI name if appropriate
*/
args->inumber = be64_to_cpu(dep->inumber);
+ error = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
xfs_da_brelse(tp, dbp);
xfs_da_brelse(tp, lbp);
- return XFS_ERROR(EEXIST);
+ return XFS_ERROR(error);
}
/*
@@ -1319,8 +1321,8 @@ xfs_dir2_leaf_lookup_int(
int *indexp, /* out: index in leaf block */
xfs_dabuf_t **dbpp) /* out: data buffer */
{
- xfs_dir2_db_t curdb; /* current data block number */
- xfs_dabuf_t *dbp; /* data buffer */
+ xfs_dir2_db_t curdb = -1; /* current data block number */
+ xfs_dabuf_t *dbp = NULL; /* data buffer */
xfs_dir2_data_entry_t *dep; /* data entry */
xfs_inode_t *dp; /* incore directory inode */
int error; /* error return code */
@@ -1331,6 +1333,8 @@ xfs_dir2_leaf_lookup_int(
xfs_mount_t *mp; /* filesystem mount point */
xfs_dir2_db_t newdb; /* new data block number */
xfs_trans_t *tp; /* transaction pointer */
+ xfs_dir2_db_t cidb = -1; /* case match data block no. */
+ enum xfs_dacmp cmp; /* name compare result */
dp = args->dp;
tp = args->trans;
@@ -1338,11 +1342,10 @@ xfs_dir2_leaf_lookup_int(
/*
* Read the leaf block into the buffer.
*/
- if ((error =
- xfs_da_read_buf(tp, dp, mp->m_dirleafblk, -1, &lbp,
- XFS_DATA_FORK))) {
+ error = xfs_da_read_buf(tp, dp, mp->m_dirleafblk, -1, &lbp,
+ XFS_DATA_FORK);
+ if (error)
return error;
- }
*lbpp = lbp;
leaf = lbp->data;
xfs_dir2_leaf_check(dp, lbp);
@@ -1354,9 +1357,9 @@ xfs_dir2_leaf_lookup_int(
* Loop over all the entries with the right hash value
* looking to match the name.
*/
- for (lep = &leaf->ents[index], dbp = NULL, curdb = -1;
- index < be16_to_cpu(leaf->hdr.count) && be32_to_cpu(lep->hashval) == args->hashval;
- lep++, index++) {
+ for (lep = &leaf->ents[index]; index < be16_to_cpu(leaf->hdr.count) &&
+ be32_to_cpu(lep->hashval) == args->hashval;
+ lep++, index++) {
/*
* Skip over stale leaf entries.
*/
@@ -1373,10 +1376,10 @@ xfs_dir2_leaf_lookup_int(
if (newdb != curdb) {
if (dbp)
xfs_da_brelse(tp, dbp);
- if ((error =
- xfs_da_read_buf(tp, dp,
- xfs_dir2_db_to_da(mp, newdb), -1, &dbp,
- XFS_DATA_FORK))) {
+ error = xfs_da_read_buf(tp, dp,
+ xfs_dir2_db_to_da(mp, newdb),
+ -1, &dbp, XFS_DATA_FORK);
+ if (error) {
xfs_da_brelse(tp, lbp);
return error;
}
@@ -1386,24 +1389,50 @@ xfs_dir2_leaf_lookup_int(
/*
* Point to the data entry.
*/
- dep = (xfs_dir2_data_entry_t *)
- ((char *)dbp->data +
- xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
+ dep = (xfs_dir2_data_entry_t *)((char *)dbp->data +
+ xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
/*
- * If it matches then return it.
+ * Compare name and if it's an exact match, return the index
+ * and buffer. If it's the first case-insensitive match, store
+ * the index and buffer and continue looking for an exact match.
*/
- if (dep->namelen == args->namelen &&
- dep->name[0] == args->name[0] &&
- memcmp(dep->name, args->name, args->namelen) == 0) {
- *dbpp = dbp;
+ cmp = mp->m_dirnameops->compname(args, dep->name, dep->namelen);
+ if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
+ args->cmpresult = cmp;
*indexp = index;
- return 0;
+ /* case exact match: return the current buffer. */
+ if (cmp == XFS_CMP_EXACT) {
+ *dbpp = dbp;
+ return 0;
+ }
+ cidb = curdb;
}
}
+ ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
+ /*
+ * Here, we can only be doing a lookup (not a rename or remove).
+ * If a case-insensitive match was found earlier, re-read the
+ * appropriate data block if required and return it.
+ */
+ if (args->cmpresult == XFS_CMP_CASE) {
+ ASSERT(cidb != -1);
+ if (cidb != curdb) {
+ xfs_da_brelse(tp, dbp);
+ error = xfs_da_read_buf(tp, dp,
+ xfs_dir2_db_to_da(mp, cidb),
+ -1, &dbp, XFS_DATA_FORK);
+ if (error) {
+ xfs_da_brelse(tp, lbp);
+ return error;
+ }
+ }
+ *dbpp = dbp;
+ return 0;
+ }
/*
* No match found, return ENOENT.
*/
- ASSERT(args->oknoent);
+ ASSERT(cidb == -1);
if (dbp)
xfs_da_brelse(tp, dbp);
xfs_da_brelse(tp, lbp);
diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c
index 8dade711f09..fa6c3a5ddbc 100644
--- a/fs/xfs/xfs_dir2_node.c
+++ b/fs/xfs/xfs_dir2_node.c
@@ -226,7 +226,7 @@ xfs_dir2_leafn_add(
ASSERT(index == be16_to_cpu(leaf->hdr.count) ||
be32_to_cpu(leaf->ents[index].hashval) >= args->hashval);
- if (args->justcheck)
+ if (args->op_flags & XFS_DA_OP_JUSTCHECK)
return 0;
/*
@@ -387,28 +387,26 @@ xfs_dir2_leafn_lasthash(
}
/*
- * Look up a leaf entry in a node-format leaf block.
- * If this is an addname then the extrablk in state is a freespace block,
- * otherwise it's a data block.
+ * Look up a leaf entry for space to add a name in a node-format leaf block.
+ * The extrablk in state is a freespace block.
*/
-int
-xfs_dir2_leafn_lookup_int(
+STATIC int
+xfs_dir2_leafn_lookup_for_addname(
xfs_dabuf_t *bp, /* leaf buffer */
xfs_da_args_t *args, /* operation arguments */
int *indexp, /* out: leaf entry index */
xfs_da_state_t *state) /* state to fill in */
{
- xfs_dabuf_t *curbp; /* current data/free buffer */
- xfs_dir2_db_t curdb; /* current data block number */
- xfs_dir2_db_t curfdb; /* current free block number */
- xfs_dir2_data_entry_t *dep; /* data block entry */
+ xfs_dabuf_t *curbp = NULL; /* current data/free buffer */
+ xfs_dir2_db_t curdb = -1; /* current data block number */
+ xfs_dir2_db_t curfdb = -1; /* current free block number */
xfs_inode_t *dp; /* incore directory inode */
int error; /* error return value */
int fi; /* free entry index */
- xfs_dir2_free_t *free=NULL; /* free block structure */
+ xfs_dir2_free_t *free = NULL; /* free block structure */
int index; /* leaf entry index */
xfs_dir2_leaf_t *leaf; /* leaf structure */
- int length=0; /* length of new data entry */
+ int length; /* length of new data entry */
xfs_dir2_leaf_entry_t *lep; /* leaf entry */
xfs_mount_t *mp; /* filesystem mount point */
xfs_dir2_db_t newdb; /* new data block number */
@@ -431,33 +429,20 @@ xfs_dir2_leafn_lookup_int(
/*
* Do we have a buffer coming in?
*/
- if (state->extravalid)
+ if (state->extravalid) {
+ /* If so, it's a free block buffer, get the block number. */
curbp = state->extrablk.bp;
- else
- curbp = NULL;
- /*
- * For addname, it's a free block buffer, get the block number.
- */
- if (args->addname) {
- curfdb = curbp ? state->extrablk.blkno : -1;
- curdb = -1;
- length = xfs_dir2_data_entsize(args->namelen);
- if ((free = (curbp ? curbp->data : NULL)))
- ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
- }
- /*
- * For others, it's a data block buffer, get the block number.
- */
- else {
- curfdb = -1;
- curdb = curbp ? state->extrablk.blkno : -1;
+ curfdb = state->extrablk.blkno;
+ free = curbp->data;
+ ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
}
+ length = xfs_dir2_data_entsize(args->namelen);
/*
* Loop over leaf entries with the right hash value.
*/
- for (lep = &leaf->ents[index];
- index < be16_to_cpu(leaf->hdr.count) && be32_to_cpu(lep->hashval) == args->hashval;
- lep++, index++) {
+ for (lep = &leaf->ents[index]; index < be16_to_cpu(leaf->hdr.count) &&
+ be32_to_cpu(lep->hashval) == args->hashval;
+ lep++, index++) {
/*
* Skip stale leaf entries.
*/
@@ -471,161 +456,244 @@ xfs_dir2_leafn_lookup_int(
* For addname, we're looking for a place to put the new entry.
* We want to use a data block with an entry of equal
* hash value to ours if there is one with room.
+ *
+ * If this block isn't the data block we already have
+ * in hand, take a look at it.
*/
- if (args->addname) {
+ if (newdb != curdb) {
+ curdb = newdb;
/*
- * If this block isn't the data block we already have
- * in hand, take a look at it.
+ * Convert the data block to the free block
+ * holding its freespace information.
*/
- if (newdb != curdb) {
- curdb = newdb;
- /*
- * Convert the data block to the free block
- * holding its freespace information.
- */
- newfdb = xfs_dir2_db_to_fdb(mp, newdb);
- /*
- * If it's not the one we have in hand,
- * read it in.
- */
- if (newfdb != curfdb) {
- /*
- * If we had one before, drop it.
- */
- if (curbp)
- xfs_da_brelse(tp, curbp);
- /*
- * Read the free block.
- */
- if ((error = xfs_da_read_buf(tp, dp,
- xfs_dir2_db_to_da(mp,
- newfdb),
- -1, &curbp,
- XFS_DATA_FORK))) {
- return error;
- }
- free = curbp->data;
- ASSERT(be32_to_cpu(free->hdr.magic) ==
- XFS_DIR2_FREE_MAGIC);
- ASSERT((be32_to_cpu(free->hdr.firstdb) %
- XFS_DIR2_MAX_FREE_BESTS(mp)) ==
- 0);
- ASSERT(be32_to_cpu(free->hdr.firstdb) <= curdb);
- ASSERT(curdb <
- be32_to_cpu(free->hdr.firstdb) +
- be32_to_cpu(free->hdr.nvalid));
- }
- /*
- * Get the index for our entry.
- */
- fi = xfs_dir2_db_to_fdindex(mp, curdb);
- /*
- * If it has room, return it.
- */
- if (unlikely(be16_to_cpu(free->bests[fi]) == NULLDATAOFF)) {
- XFS_ERROR_REPORT("xfs_dir2_leafn_lookup_int",
- XFS_ERRLEVEL_LOW, mp);
- if (curfdb != newfdb)
- xfs_da_brelse(tp, curbp);
- return XFS_ERROR(EFSCORRUPTED);
- }
- curfdb = newfdb;
- if (be16_to_cpu(free->bests[fi]) >= length) {
- *indexp = index;
- state->extravalid = 1;
- state->extrablk.bp = curbp;
- state->extrablk.blkno = curfdb;
- state->extrablk.index = fi;
- state->extrablk.magic =
- XFS_DIR2_FREE_MAGIC;
- ASSERT(args->oknoent);
- return XFS_ERROR(ENOENT);
- }
- }
- }
- /*
- * Not adding a new entry, so we really want to find
- * the name given to us.
- */
- else {
+ newfdb = xfs_dir2_db_to_fdb(mp, newdb);
/*
- * If it's a different data block, go get it.
+ * If it's not the one we have in hand, read it in.
*/
- if (newdb != curdb) {
+ if (newfdb != curfdb) {
/*
- * If we had a block before, drop it.
+ * If we had one before, drop it.
*/
if (curbp)
xfs_da_brelse(tp, curbp);
/*
- * Read the data block.
+ * Read the free block.
*/
- if ((error =
- xfs_da_read_buf(tp, dp,
- xfs_dir2_db_to_da(mp, newdb), -1,
- &curbp, XFS_DATA_FORK))) {
+ error = xfs_da_read_buf(tp, dp,
+ xfs_dir2_db_to_da(mp, newfdb),
+ -1, &curbp, XFS_DATA_FORK);
+ if (error)
return error;
- }
- xfs_dir2_data_check(dp, curbp);
- curdb = newdb;
+ free = curbp->data;
+ ASSERT(be32_to_cpu(free->hdr.magic) ==
+ XFS_DIR2_FREE_MAGIC);
+ ASSERT((be32_to_cpu(free->hdr.firstdb) %
+ XFS_DIR2_MAX_FREE_BESTS(mp)) == 0);
+ ASSERT(be32_to_cpu(free->hdr.firstdb) <= curdb);
+ ASSERT(curdb < be32_to_cpu(free->hdr.firstdb) +
+ be32_to_cpu(free->hdr.nvalid));
}
/*
- * Point to the data entry.
+ * Get the index for our entry.
*/
- dep = (xfs_dir2_data_entry_t *)
- ((char *)curbp->data +
- xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
+ fi = xfs_dir2_db_to_fdindex(mp, curdb);
/*
- * Compare the entry, return it if it matches.
+ * If it has room, return it.
*/
- if (dep->namelen == args->namelen &&
- dep->name[0] == args->name[0] &&
- memcmp(dep->name, args->name, args->namelen) == 0) {
- args->inumber = be64_to_cpu(dep->inumber);
- *indexp = index;
- state->extravalid = 1;
- state->extrablk.bp = curbp;
- state->extrablk.blkno = curdb;
- state->extrablk.index =
- (int)((char *)dep -
- (char *)curbp->data);
- state->extrablk.magic = XFS_DIR2_DATA_MAGIC;
- return XFS_ERROR(EEXIST);
+ if (unlikely(be16_to_cpu(free->bests[fi]) == NULLDATAOFF)) {
+ XFS_ERROR_REPORT("xfs_dir2_leafn_lookup_int",
+ XFS_ERRLEVEL_LOW, mp);
+ if (curfdb != newfdb)
+ xfs_da_brelse(tp, curbp);
+ return XFS_ERROR(EFSCORRUPTED);
}
+ curfdb = newfdb;
+ if (be16_to_cpu(free->bests[fi]) >= length)
+ goto out;
}
}
+ /* Didn't find any space */
+ fi = -1;
+out:
+ ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
+ if (curbp) {
+ /* Giving back a free block. */
+ state->extravalid = 1;
+ state->extrablk.bp = curbp;
+ state->extrablk.index = fi;
+ state->extrablk.blkno = curfdb;
+ state->extrablk.magic = XFS_DIR2_FREE_MAGIC;
+ } else {
+ state->extravalid = 0;
+ }
/*
- * Didn't find a match.
- * If we are holding a buffer, give it back in case our caller
- * finds it useful.
+ * Return the index, that will be the insertion point.
*/
- if ((state->extravalid = (curbp != NULL))) {
- state->extrablk.bp = curbp;
- state->extrablk.index = -1;
+ *indexp = index;
+ return XFS_ERROR(ENOENT);
+}
+
+/*
+ * Look up a leaf entry in a node-format leaf block.
+ * The extrablk in state a data block.
+ */
+STATIC int
+xfs_dir2_leafn_lookup_for_entry(
+ xfs_dabuf_t *bp, /* leaf buffer */
+ xfs_da_args_t *args, /* operation arguments */
+ int *indexp, /* out: leaf entry index */
+ xfs_da_state_t *state) /* state to fill in */
+{
+ xfs_dabuf_t *curbp = NULL; /* current data/free buffer */
+ xfs_dir2_db_t curdb = -1; /* current data block number */
+ xfs_dir2_data_entry_t *dep; /* data block entry */
+ xfs_inode_t *dp; /* incore directory inode */
+ int error; /* error return value */
+ int index; /* leaf entry index */
+ xfs_dir2_leaf_t *leaf; /* leaf structure */
+ xfs_dir2_leaf_entry_t *lep; /* leaf entry */
+ xfs_mount_t *mp; /* filesystem mount point */
+ xfs_dir2_db_t newdb; /* new data block number */
+ xfs_trans_t *tp; /* transaction pointer */
+ enum xfs_dacmp cmp; /* comparison result */
+
+ dp = args->dp;
+ tp = args->trans;
+ mp = dp->i_mount;
+ leaf = bp->data;
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+#ifdef __KERNEL__
+ ASSERT(be16_to_cpu(leaf->hdr.count) > 0);
+#endif
+ xfs_dir2_leafn_check(dp, bp);
+ /*
+ * Look up the hash value in the leaf entries.
+ */
+ index = xfs_dir2_leaf_search_hash(args, bp);
+ /*
+ * Do we have a buffer coming in?
+ */
+ if (state->extravalid) {
+ curbp = state->extrablk.bp;
+ curdb = state->extrablk.blkno;
+ }
+ /*
+ * Loop over leaf entries with the right hash value.
+ */
+ for (lep = &leaf->ents[index]; index < be16_to_cpu(leaf->hdr.count) &&
+ be32_to_cpu(lep->hashval) == args->hashval;
+ lep++, index++) {
/*
- * For addname, giving back a free block.
+ * Skip stale leaf entries.
*/
- if (args->addname) {
- state->extrablk.blkno = curfdb;
- state->extrablk.magic = XFS_DIR2_FREE_MAGIC;
+ if (be32_to_cpu(lep->address) == XFS_DIR2_NULL_DATAPTR)
+ continue;
+ /*
+ * Pull the data block number from the entry.
+ */
+ newdb = xfs_dir2_dataptr_to_db(mp, be32_to_cpu(lep->address));
+ /*
+ * Not adding a new entry, so we really want to find
+ * the name given to us.
+ *
+ * If it's a different data block, go get it.
+ */
+ if (newdb != curdb) {
+ /*
+ * If we had a block before that we aren't saving
+ * for a CI name, drop it
+ */
+ if (curbp && (args->cmpresult == XFS_CMP_DIFFERENT ||
+ curdb != state->extrablk.blkno))
+ xfs_da_brelse(tp, curbp);
+ /*
+ * If needing the block that is saved with a CI match,
+ * use it otherwise read in the new data block.
+ */
+ if (args->cmpresult != XFS_CMP_DIFFERENT &&
+ newdb == state->extrablk.blkno) {
+ ASSERT(state->extravalid);
+ curbp = state->extrablk.bp;
+ } else {
+ error = xfs_da_read_buf(tp, dp,
+ xfs_dir2_db_to_da(mp, newdb),
+ -1, &curbp, XFS_DATA_FORK);
+ if (error)
+ return error;
+ }
+ xfs_dir2_data_check(dp, curbp);
+ curdb = newdb;
}
/*
- * For other callers, giving back a data block.
+ * Point to the data entry.
*/
- else {
+ dep = (xfs_dir2_data_entry_t *)((char *)curbp->data +
+ xfs_dir2_dataptr_to_off(mp, be32_to_cpu(lep->address)));
+ /*
+ * Compare the entry and if it's an exact match, return
+ * EEXIST immediately. If it's the first case-insensitive
+ * match, store the block & inode number and continue looking.
+ */
+ cmp = mp->m_dirnameops->compname(args, dep->name, dep->namelen);
+ if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
+ /* If there is a CI match block, drop it */
+ if (args->cmpresult != XFS_CMP_DIFFERENT &&
+ curdb != state->extrablk.blkno)
+ xfs_da_brelse(tp, state->extrablk.bp);
+ args->cmpresult = cmp;
+ args->inumber = be64_to_cpu(dep->inumber);
+ *indexp = index;
+ state->extravalid = 1;
+ state->extrablk.bp = curbp;
state->extrablk.blkno = curdb;
+ state->extrablk.index = (int)((char *)dep -
+ (char *)curbp->data);
state->extrablk.magic = XFS_DIR2_DATA_MAGIC;
+ if (cmp == XFS_CMP_EXACT)
+ return XFS_ERROR(EEXIST);
}
}
- /*
- * Return the final index, that will be the insertion point.
- */
+ ASSERT(index == be16_to_cpu(leaf->hdr.count) ||
+ (args->op_flags & XFS_DA_OP_OKNOENT));
+ if (curbp) {
+ if (args->cmpresult == XFS_CMP_DIFFERENT) {
+ /* Giving back last used data block. */
+ state->extravalid = 1;
+ state->extrablk.bp = curbp;
+ state->extrablk.index = -1;
+ state->extrablk.blkno = curdb;
+ state->extrablk.magic = XFS_DIR2_DATA_MAGIC;
+ } else {
+ /* If the curbp is not the CI match block, drop it */
+ if (state->extrablk.bp != curbp)
+ xfs_da_brelse(tp, curbp);
+ }
+ } else {
+ state->extravalid = 0;
+ }
*indexp = index;
- ASSERT(index == be16_to_cpu(leaf->hdr.count) || args->oknoent);
return XFS_ERROR(ENOENT);
}
/*
+ * Look up a leaf entry in a node-format leaf block.
+ * If this is an addname then the extrablk in state is a freespace block,
+ * otherwise it's a data block.
+ */
+int
+xfs_dir2_leafn_lookup_int(
+ xfs_dabuf_t *bp, /* leaf buffer */
+ xfs_da_args_t *args, /* operation arguments */
+ int *indexp, /* out: leaf entry index */
+ xfs_da_state_t *state) /* state to fill in */
+{
+ if (args->op_flags & XFS_DA_OP_ADDNAME)
+ return xfs_dir2_leafn_lookup_for_addname(bp, args, indexp,
+ state);
+ return xfs_dir2_leafn_lookup_for_entry(bp, args, indexp, state);
+}
+
+/*
* Move count leaf entries from source to destination leaf.
* Log entries and headers. Stale entries are preserved.
*/
@@ -823,9 +891,10 @@ xfs_dir2_leafn_rebalance(
*/
if (!state->inleaf)
blk2->index = blk1->index - be16_to_cpu(leaf1->hdr.count);
-
- /*
- * Finally sanity check just to make sure we are not returning a negative index
+
+ /*
+ * Finally sanity check just to make sure we are not returning a
+ * negative index
*/
if(blk2->index < 0) {
state->inleaf = 1;
@@ -1332,7 +1401,7 @@ xfs_dir2_node_addname(
/*
* It worked, fix the hash values up the btree.
*/
- if (!args->justcheck)
+ if (!(args->op_flags & XFS_DA_OP_JUSTCHECK))
xfs_da_fixhashpath(state, &state->path);
} else {
/*
@@ -1515,7 +1584,8 @@ xfs_dir2_node_addname_int(
/*
* Not allowed to allocate, return failure.
*/
- if (args->justcheck || args->total == 0) {
+ if ((args->op_flags & XFS_DA_OP_JUSTCHECK) ||
+ args->total == 0) {
/*
* Drop the freespace buffer unless it came from our
* caller.
@@ -1661,7 +1731,7 @@ xfs_dir2_node_addname_int(
/*
* If just checking, we succeeded.
*/
- if (args->justcheck) {
+ if (args->op_flags & XFS_DA_OP_JUSTCHECK) {
if ((fblk == NULL || fblk->bp == NULL) && fbp != NULL)
xfs_da_buf_done(fbp);
return 0;
@@ -1767,6 +1837,14 @@ xfs_dir2_node_lookup(
error = xfs_da_node_lookup_int(state, &rval);
if (error)
rval = error;
+ else if (rval == ENOENT && args->cmpresult == XFS_CMP_CASE) {
+ /* If a CI match, dup the actual name and return EEXIST */
+ xfs_dir2_data_entry_t *dep;
+
+ dep = (xfs_dir2_data_entry_t *)((char *)state->extrablk.bp->
+ data + state->extrablk.index);
+ rval = xfs_dir_cilookup_result(args, dep->name, dep->namelen);
+ }
/*
* Release the btree blocks and leaf block.
*/
@@ -1810,9 +1888,8 @@ xfs_dir2_node_removename(
* Look up the entry we're deleting, set up the cursor.
*/
error = xfs_da_node_lookup_int(state, &rval);
- if (error) {
+ if (error)
rval = error;
- }
/*
* Didn't find it, upper layer screwed up.
*/
@@ -1829,9 +1906,8 @@ xfs_dir2_node_removename(
*/
error = xfs_dir2_leafn_remove(args, blk->bp, blk->index,
&state->extrablk, &rval);
- if (error) {
+ if (error)
return error;
- }
/*
* Fix the hash values up the btree.
*/
diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/xfs_dir2_sf.c
index 919d275a1ce..b46af0013ec 100644
--- a/fs/xfs/xfs_dir2_sf.c
+++ b/fs/xfs/xfs_dir2_sf.c
@@ -255,7 +255,7 @@ xfs_dir2_block_to_sf(
xfs_dir2_sf_check(args);
out:
xfs_trans_log_inode(args->trans, dp, logflags);
- kmem_free(block, mp->m_dirblksize);
+ kmem_free(block);
return error;
}
@@ -332,7 +332,7 @@ xfs_dir2_sf_addname(
/*
* Just checking or no space reservation, it doesn't fit.
*/
- if (args->justcheck || args->total == 0)
+ if ((args->op_flags & XFS_DA_OP_JUSTCHECK) || args->total == 0)
return XFS_ERROR(ENOSPC);
/*
* Convert to block form then add the name.
@@ -345,7 +345,7 @@ xfs_dir2_sf_addname(
/*
* Just checking, it fits.
*/
- if (args->justcheck)
+ if (args->op_flags & XFS_DA_OP_JUSTCHECK)
return 0;
/*
* Do it the easy way - just add it at the end.
@@ -512,7 +512,7 @@ xfs_dir2_sf_addname_hard(
sfep = xfs_dir2_sf_nextentry(sfp, sfep);
memcpy(sfep, oldsfep, old_isize - nbytes);
}
- kmem_free(buf, old_isize);
+ kmem_free(buf);
dp->i_d.di_size = new_isize;
xfs_dir2_sf_check(args);
}
@@ -812,8 +812,11 @@ xfs_dir2_sf_lookup(
{
xfs_inode_t *dp; /* incore directory inode */
int i; /* entry index */
+ int error;
xfs_dir2_sf_entry_t *sfep; /* shortform directory entry */
xfs_dir2_sf_t *sfp; /* shortform structure */
+ enum xfs_dacmp cmp; /* comparison result */
+ xfs_dir2_sf_entry_t *ci_sfep; /* case-insens. entry */
xfs_dir2_trace_args("sf_lookup", args);
xfs_dir2_sf_check(args);
@@ -836,6 +839,7 @@ xfs_dir2_sf_lookup(
*/
if (args->namelen == 1 && args->name[0] == '.') {
args->inumber = dp->i_ino;
+ args->cmpresult = XFS_CMP_EXACT;
return XFS_ERROR(EEXIST);
}
/*
@@ -844,28 +848,41 @@ xfs_dir2_sf_lookup(
if (args->namelen == 2 &&
args->name[0] == '.' && args->name[1] == '.') {
args->inumber = xfs_dir2_sf_get_inumber(sfp, &sfp->hdr.parent);
+ args->cmpresult = XFS_CMP_EXACT;
return XFS_ERROR(EEXIST);
}
/*
* Loop over all the entries trying to match ours.
*/
- for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp);
- i < sfp->hdr.count;
- i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) {
- if (sfep->namelen == args->namelen &&
- sfep->name[0] == args->name[0] &&
- memcmp(args->name, sfep->name, args->namelen) == 0) {
- args->inumber =
- xfs_dir2_sf_get_inumber(sfp,
- xfs_dir2_sf_inumberp(sfep));
- return XFS_ERROR(EEXIST);
+ ci_sfep = NULL;
+ for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->hdr.count;
+ i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) {
+ /*
+ * Compare name and if it's an exact match, return the inode
+ * number. If it's the first case-insensitive match, store the
+ * inode number and continue looking for an exact match.
+ */
+ cmp = dp->i_mount->m_dirnameops->compname(args, sfep->name,
+ sfep->namelen);
+ if (cmp != XFS_CMP_DIFFERENT && cmp != args->cmpresult) {
+ args->cmpresult = cmp;
+ args->inumber = xfs_dir2_sf_get_inumber(sfp,
+ xfs_dir2_sf_inumberp(sfep));
+ if (cmp == XFS_CMP_EXACT)
+ return XFS_ERROR(EEXIST);
+ ci_sfep = sfep;
}
}
+ ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
/*
- * Didn't find it.
+ * Here, we can only be doing a lookup (not a rename or replace).
+ * If a case-insensitive match was not found, return ENOENT.
*/
- ASSERT(args->oknoent);
- return XFS_ERROR(ENOENT);
+ if (!ci_sfep)
+ return XFS_ERROR(ENOENT);
+ /* otherwise process the CI match as required by the caller */
+ error = xfs_dir_cilookup_result(args, ci_sfep->name, ci_sfep->namelen);
+ return XFS_ERROR(error);
}
/*
@@ -904,24 +921,21 @@ xfs_dir2_sf_removename(
* Loop over the old directory entries.
* Find the one we're deleting.
*/
- for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp);
- i < sfp->hdr.count;
- i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) {
- if (sfep->namelen == args->namelen &&
- sfep->name[0] == args->name[0] &&
- memcmp(sfep->name, args->name, args->namelen) == 0) {
+ for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp); i < sfp->hdr.count;
+ i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) {
+ if (xfs_da_compname(args, sfep->name, sfep->namelen) ==
+ XFS_CMP_EXACT) {
ASSERT(xfs_dir2_sf_get_inumber(sfp,
- xfs_dir2_sf_inumberp(sfep)) ==
- args->inumber);
+ xfs_dir2_sf_inumberp(sfep)) ==
+ args->inumber);
break;
}
}
/*
* Didn't find it.
*/
- if (i == sfp->hdr.count) {
+ if (i == sfp->hdr.count)
return XFS_ERROR(ENOENT);
- }
/*
* Calculate sizes.
*/
@@ -1042,11 +1056,10 @@ xfs_dir2_sf_replace(
*/
else {
for (i = 0, sfep = xfs_dir2_sf_firstentry(sfp);
- i < sfp->hdr.count;
- i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) {
- if (sfep->namelen == args->namelen &&
- sfep->name[0] == args->name[0] &&
- memcmp(args->name, sfep->name, args->namelen) == 0) {
+ i < sfp->hdr.count;
+ i++, sfep = xfs_dir2_sf_nextentry(sfp, sfep)) {
+ if (xfs_da_compname(args, sfep->name, sfep->namelen) ==
+ XFS_CMP_EXACT) {
#if XFS_BIG_INUMS || defined(DEBUG)
ino = xfs_dir2_sf_get_inumber(sfp,
xfs_dir2_sf_inumberp(sfep));
@@ -1061,7 +1074,7 @@ xfs_dir2_sf_replace(
* Didn't find it.
*/
if (i == sfp->hdr.count) {
- ASSERT(args->oknoent);
+ ASSERT(args->op_flags & XFS_DA_OP_OKNOENT);
#if XFS_BIG_INUMS
if (i8elevated)
xfs_dir2_sf_toino4(args);
@@ -1174,7 +1187,7 @@ xfs_dir2_sf_toino4(
/*
* Clean up the inode.
*/
- kmem_free(buf, oldsize);
+ kmem_free(buf);
dp->i_d.di_size = newsize;
xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_DDATA);
}
@@ -1251,7 +1264,7 @@ xfs_dir2_sf_toino8(
/*
* Clean up the inode.
*/
- kmem_free(buf, oldsize);
+ kmem_free(buf);
dp->i_d.di_size = newsize;
xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_DDATA);
}
diff --git a/fs/xfs/xfs_dir2_sf.h b/fs/xfs/xfs_dir2_sf.h
index 005629d702d..deecc9d238f 100644
--- a/fs/xfs/xfs_dir2_sf.h
+++ b/fs/xfs/xfs_dir2_sf.h
@@ -62,7 +62,7 @@ typedef union {
* Normalized offset (in a data block) of the entry, really xfs_dir2_data_off_t.
* Only need 16 bits, this is the byte offset into the single block form.
*/
-typedef struct { __uint8_t i[2]; } xfs_dir2_sf_off_t;
+typedef struct { __uint8_t i[2]; } __arch_pack xfs_dir2_sf_off_t;
/*
* The parent directory has a dedicated field, and the self-pointer must
@@ -76,14 +76,14 @@ typedef struct xfs_dir2_sf_hdr {
__uint8_t count; /* count of entries */
__uint8_t i8count; /* count of 8-byte inode #s */
xfs_dir2_inou_t parent; /* parent dir inode number */
-} xfs_dir2_sf_hdr_t;
+} __arch_pack xfs_dir2_sf_hdr_t;
typedef struct xfs_dir2_sf_entry {
__uint8_t namelen; /* actual name length */
xfs_dir2_sf_off_t offset; /* saved offset */
__uint8_t name[1]; /* name, variable size */
xfs_dir2_inou_t inumber; /* inode number, var. offset */
-} xfs_dir2_sf_entry_t;
+} __arch_pack xfs_dir2_sf_entry_t;
typedef struct xfs_dir2_sf {
xfs_dir2_sf_hdr_t hdr; /* shortform header */
diff --git a/fs/xfs/xfs_dir2_trace.c b/fs/xfs/xfs_dir2_trace.c
index f3fb2ffd6f5..6cc7c0c681a 100644
--- a/fs/xfs/xfs_dir2_trace.c
+++ b/fs/xfs/xfs_dir2_trace.c
@@ -85,7 +85,8 @@ xfs_dir2_trace_args(
(void *)((unsigned long)(args->inumber >> 32)),
(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
(void *)args->dp, (void *)args->trans,
- (void *)(unsigned long)args->justcheck, NULL, NULL);
+ (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
+ NULL, NULL);
}
void
@@ -100,7 +101,7 @@ xfs_dir2_trace_args_b(
(void *)((unsigned long)(args->inumber >> 32)),
(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
(void *)args->dp, (void *)args->trans,
- (void *)(unsigned long)args->justcheck,
+ (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
(void *)(bp ? bp->bps[0] : NULL), NULL);
}
@@ -117,7 +118,7 @@ xfs_dir2_trace_args_bb(
(void *)((unsigned long)(args->inumber >> 32)),
(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
(void *)args->dp, (void *)args->trans,
- (void *)(unsigned long)args->justcheck,
+ (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
(void *)(lbp ? lbp->bps[0] : NULL),
(void *)(dbp ? dbp->bps[0] : NULL));
}
@@ -157,8 +158,8 @@ xfs_dir2_trace_args_db(
(void *)((unsigned long)(args->inumber >> 32)),
(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
(void *)args->dp, (void *)args->trans,
- (void *)(unsigned long)args->justcheck, (void *)(long)db,
- (void *)dbp);
+ (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
+ (void *)(long)db, (void *)dbp);
}
void
@@ -173,7 +174,7 @@ xfs_dir2_trace_args_i(
(void *)((unsigned long)(args->inumber >> 32)),
(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
(void *)args->dp, (void *)args->trans,
- (void *)(unsigned long)args->justcheck,
+ (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
(void *)((unsigned long)(i >> 32)),
(void *)((unsigned long)(i & 0xFFFFFFFF)));
}
@@ -190,7 +191,8 @@ xfs_dir2_trace_args_s(
(void *)((unsigned long)(args->inumber >> 32)),
(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
(void *)args->dp, (void *)args->trans,
- (void *)(unsigned long)args->justcheck, (void *)(long)s, NULL);
+ (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
+ (void *)(long)s, NULL);
}
void
@@ -208,7 +210,7 @@ xfs_dir2_trace_args_sb(
(void *)((unsigned long)(args->inumber >> 32)),
(void *)((unsigned long)(args->inumber & 0xFFFFFFFF)),
(void *)args->dp, (void *)args->trans,
- (void *)(unsigned long)args->justcheck, (void *)(long)s,
- (void *)dbp);
+ (void *)(unsigned long)(args->op_flags & XFS_DA_OP_JUSTCHECK),
+ (void *)(long)s, (void *)dbp);
}
#endif /* XFS_DIR2_TRACE */
diff --git a/fs/xfs/xfs_dmapi.h b/fs/xfs/xfs_dmapi.h
index f71784ab6a6..2813cdd7237 100644
--- a/fs/xfs/xfs_dmapi.h
+++ b/fs/xfs/xfs_dmapi.h
@@ -18,7 +18,6 @@
#ifndef __XFS_DMAPI_H__
#define __XFS_DMAPI_H__
-#include <linux/version.h>
/* Values used to define the on-disk version of dm_attrname_t. All
* on-disk attribute names start with the 8-byte string "SGI_DMI_".
*
@@ -166,6 +165,6 @@ typedef enum {
#define FILP_DELAY_FLAG(filp) ((filp->f_flags&(O_NDELAY|O_NONBLOCK)) ? \
DM_FLAGS_NDELAY : 0)
-#define AT_DELAY_FLAG(f) ((f&ATTR_NONBLOCK) ? DM_FLAGS_NDELAY : 0)
+#define AT_DELAY_FLAG(f) ((f & XFS_ATTR_NONBLOCK) ? DM_FLAGS_NDELAY : 0)
#endif /* __XFS_DMAPI_H__ */
diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index 05e5365d3c3..f227ecd1a29 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -58,22 +58,11 @@ xfs_error_trap(int e)
}
return e;
}
-#endif
-
-#if (defined(DEBUG) || defined(INDUCE_IO_ERROR))
int xfs_etest[XFS_NUM_INJECT_ERROR];
int64_t xfs_etest_fsid[XFS_NUM_INJECT_ERROR];
char * xfs_etest_fsname[XFS_NUM_INJECT_ERROR];
-void
-xfs_error_test_init(void)
-{
- memset(xfs_etest, 0, sizeof(xfs_etest));
- memset(xfs_etest_fsid, 0, sizeof(xfs_etest_fsid));
- memset(xfs_etest_fsname, 0, sizeof(xfs_etest_fsname));
-}
-
int
xfs_error_test(int error_tag, int *fsidp, char *expression,
int line, char *file, unsigned long randfactor)
@@ -150,8 +139,7 @@ xfs_errortag_clearall(xfs_mount_t *mp, int loud)
xfs_etest[i]);
xfs_etest[i] = 0;
xfs_etest_fsid[i] = 0LL;
- kmem_free(xfs_etest_fsname[i],
- strlen(xfs_etest_fsname[i]) + 1);
+ kmem_free(xfs_etest_fsname[i]);
xfs_etest_fsname[i] = NULL;
}
}
@@ -163,7 +151,7 @@ xfs_errortag_clearall(xfs_mount_t *mp, int loud)
return 0;
}
-#endif /* DEBUG || INDUCE_IO_ERROR */
+#endif /* DEBUG */
static void
xfs_fs_vcmn_err(int level, xfs_mount_t *mp, char *fmt, va_list ap)
@@ -175,7 +163,7 @@ xfs_fs_vcmn_err(int level, xfs_mount_t *mp, char *fmt, va_list ap)
newfmt = kmem_alloc(len, KM_SLEEP);
sprintf(newfmt, "Filesystem \"%s\": %s", mp->m_fsname, fmt);
icmn_err(level, newfmt, ap);
- kmem_free(newfmt, len);
+ kmem_free(newfmt);
} else {
icmn_err(level, fmt, ap);
}
diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h
index 6490d2a9f8e..11543f10b0c 100644
--- a/fs/xfs/xfs_error.h
+++ b/fs/xfs/xfs_error.h
@@ -125,23 +125,14 @@ extern void xfs_corruption_error(char *tag, int level, struct xfs_mount *mp,
#define XFS_RANDOM_DIOWRITE_IOERR (XFS_RANDOM_DEFAULT/10)
#define XFS_RANDOM_BMAPIFORMAT XFS_RANDOM_DEFAULT
-#if (defined(DEBUG) || defined(INDUCE_IO_ERROR))
+#ifdef DEBUG
extern int xfs_error_test(int, int *, char *, int, char *, unsigned long);
-extern void xfs_error_test_init(void);
#define XFS_NUM_INJECT_ERROR 10
-
-#ifdef __ANSI_CPP__
-#define XFS_TEST_ERROR(expr, mp, tag, rf) \
- ((expr) || \
- xfs_error_test((tag), (mp)->m_fixedfsid, #expr, __LINE__, __FILE__, \
- (rf)))
-#else
#define XFS_TEST_ERROR(expr, mp, tag, rf) \
((expr) || \
xfs_error_test((tag), (mp)->m_fixedfsid, "expr", __LINE__, __FILE__, \
(rf)))
-#endif /* __ANSI_CPP__ */
extern int xfs_errortag_add(int error_tag, xfs_mount_t *mp);
extern int xfs_errortag_clearall(xfs_mount_t *mp, int loud);
@@ -149,7 +140,7 @@ extern int xfs_errortag_clearall(xfs_mount_t *mp, int loud);
#define XFS_TEST_ERROR(expr, mp, tag, rf) (expr)
#define xfs_errortag_add(tag, mp) (ENOSYS)
#define xfs_errortag_clearall(mp, loud) (ENOSYS)
-#endif /* (DEBUG || INDUCE_IO_ERROR) */
+#endif /* DEBUG */
/*
* XFS panic tags -- allow a call to xfs_cmn_err() be turned into
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index 132bd07b9bb..8aa28f751b2 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -41,8 +41,7 @@ xfs_efi_item_free(xfs_efi_log_item_t *efip)
int nexts = efip->efi_format.efi_nextents;
if (nexts > XFS_EFI_MAX_FAST_EXTENTS) {
- kmem_free(efip, sizeof(xfs_efi_log_item_t) +
- (nexts - 1) * sizeof(xfs_extent_t));
+ kmem_free(efip);
} else {
kmem_zone_free(xfs_efi_zone, efip);
}
@@ -374,8 +373,7 @@ xfs_efd_item_free(xfs_efd_log_item_t *efdp)
int nexts = efdp->efd_format.efd_nextents;
if (nexts > XFS_EFD_MAX_FAST_EXTENTS) {
- kmem_free(efdp, sizeof(xfs_efd_log_item_t) +
- (nexts - 1) * sizeof(xfs_extent_t));
+ kmem_free(efdp);
} else {
kmem_zone_free(xfs_efd_zone, efdp);
}
diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c
index 3f3785b1080..f3bb75da384 100644
--- a/fs/xfs/xfs_filestream.c
+++ b/fs/xfs/xfs_filestream.c
@@ -397,10 +397,12 @@ int
xfs_filestream_init(void)
{
item_zone = kmem_zone_init(sizeof(fstrm_item_t), "fstrm_item");
+ if (!item_zone)
+ return -ENOMEM;
#ifdef XFS_FILESTREAMS_TRACE
- xfs_filestreams_trace_buf = ktrace_alloc(XFS_FSTRM_KTRACE_SIZE, KM_SLEEP);
+ xfs_filestreams_trace_buf = ktrace_alloc(XFS_FSTRM_KTRACE_SIZE, KM_NOFS);
#endif
- return item_zone ? 0 : -ENOMEM;
+ return 0;
}
/*
diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h
index 3bed6433d05..01c0cc88d3f 100644
--- a/fs/xfs/xfs_fs.h
+++ b/fs/xfs/xfs_fs.h
@@ -239,6 +239,7 @@ typedef struct xfs_fsop_resblks {
#define XFS_FSOP_GEOM_FLAGS_LOGV2 0x0100 /* log format version 2 */
#define XFS_FSOP_GEOM_FLAGS_SECTOR 0x0200 /* sector sizes >1BB */
#define XFS_FSOP_GEOM_FLAGS_ATTR2 0x0400 /* inline attributes rework */
+#define XFS_FSOP_GEOM_FLAGS_DIRV2CI 0x1000 /* ASCII only CI names */
#define XFS_FSOP_GEOM_FLAGS_LAZYSB 0x4000 /* lazy superblock counters */
@@ -371,6 +372,9 @@ typedef struct xfs_fsop_attrlist_handlereq {
typedef struct xfs_attr_multiop {
__u32 am_opcode;
+#define ATTR_OP_GET 1 /* return the indicated attr's value */
+#define ATTR_OP_SET 2 /* set/create the indicated attr/value pair */
+#define ATTR_OP_REMOVE 3 /* remove the indicated attr */
__s32 am_error;
void __user *am_attrname;
void __user *am_attrvalue;
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index 381ebda4f7b..84583cf73db 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -95,6 +95,8 @@ xfs_fs_geometry(
XFS_FSOP_GEOM_FLAGS_DIRV2 : 0) |
(xfs_sb_version_hassector(&mp->m_sb) ?
XFS_FSOP_GEOM_FLAGS_SECTOR : 0) |
+ (xfs_sb_version_hasasciici(&mp->m_sb) ?
+ XFS_FSOP_GEOM_FLAGS_DIRV2CI : 0) |
(xfs_sb_version_haslazysbcount(&mp->m_sb) ?
XFS_FSOP_GEOM_FLAGS_LAZYSB : 0) |
(xfs_sb_version_hasattr2(&mp->m_sb) ?
@@ -625,7 +627,7 @@ xfs_fs_goingdown(
xfs_force_shutdown(mp, SHUTDOWN_FORCE_UMOUNT);
thaw_bdev(sb->s_bdev, sb);
}
-
+
break;
}
case XFS_FSOP_GOING_FLAGS_LOGFLUSH:
diff --git a/fs/xfs/xfs_ialloc_btree.c b/fs/xfs/xfs_ialloc_btree.c
index e5310c90e50..83502f3edef 100644
--- a/fs/xfs/xfs_ialloc_btree.c
+++ b/fs/xfs/xfs_ialloc_btree.c
@@ -181,7 +181,7 @@ xfs_inobt_delrec(
* then we can get rid of this level.
*/
if (numrecs == 1 && level > 0) {
- agbp = cur->bc_private.i.agbp;
+ agbp = cur->bc_private.a.agbp;
agi = XFS_BUF_TO_AGI(agbp);
/*
* pp is still set to the first pointer in the block.
@@ -194,7 +194,7 @@ xfs_inobt_delrec(
* Free the block.
*/
if ((error = xfs_free_extent(cur->bc_tp,
- XFS_AGB_TO_FSB(mp, cur->bc_private.i.agno, bno), 1)))
+ XFS_AGB_TO_FSB(mp, cur->bc_private.a.agno, bno), 1)))
return error;
xfs_trans_binval(cur->bc_tp, bp);
xfs_ialloc_log_agi(cur->bc_tp, agbp,
@@ -379,7 +379,7 @@ xfs_inobt_delrec(
rrecs = be16_to_cpu(right->bb_numrecs);
rbp = bp;
if ((error = xfs_btree_read_bufs(mp, cur->bc_tp,
- cur->bc_private.i.agno, lbno, 0, &lbp,
+ cur->bc_private.a.agno, lbno, 0, &lbp,
XFS_INO_BTREE_REF)))
return error;
left = XFS_BUF_TO_INOBT_BLOCK(lbp);
@@ -401,7 +401,7 @@ xfs_inobt_delrec(
lrecs = be16_to_cpu(left->bb_numrecs);
lbp = bp;
if ((error = xfs_btree_read_bufs(mp, cur->bc_tp,
- cur->bc_private.i.agno, rbno, 0, &rbp,
+ cur->bc_private.a.agno, rbno, 0, &rbp,
XFS_INO_BTREE_REF)))
return error;
right = XFS_BUF_TO_INOBT_BLOCK(rbp);
@@ -484,7 +484,7 @@ xfs_inobt_delrec(
xfs_buf_t *rrbp;
if ((error = xfs_btree_read_bufs(mp, cur->bc_tp,
- cur->bc_private.i.agno, be32_to_cpu(left->bb_rightsib), 0,
+ cur->bc_private.a.agno, be32_to_cpu(left->bb_rightsib), 0,
&rrbp, XFS_INO_BTREE_REF)))
return error;
rrblock = XFS_BUF_TO_INOBT_BLOCK(rrbp);
@@ -497,7 +497,7 @@ xfs_inobt_delrec(
* Free the deleting block.
*/
if ((error = xfs_free_extent(cur->bc_tp, XFS_AGB_TO_FSB(mp,
- cur->bc_private.i.agno, rbno), 1)))
+ cur->bc_private.a.agno, rbno), 1)))
return error;
xfs_trans_binval(cur->bc_tp, rbp);
/*
@@ -854,7 +854,7 @@ xfs_inobt_lookup(
{
xfs_agi_t *agi; /* a.g. inode header */
- agi = XFS_BUF_TO_AGI(cur->bc_private.i.agbp);
+ agi = XFS_BUF_TO_AGI(cur->bc_private.a.agbp);
agno = be32_to_cpu(agi->agi_seqno);
agbno = be32_to_cpu(agi->agi_root);
}
@@ -1089,7 +1089,7 @@ xfs_inobt_lshift(
* Set up the left neighbor as "left".
*/
if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp,
- cur->bc_private.i.agno, be32_to_cpu(right->bb_leftsib),
+ cur->bc_private.a.agno, be32_to_cpu(right->bb_leftsib),
0, &lbp, XFS_INO_BTREE_REF)))
return error;
left = XFS_BUF_TO_INOBT_BLOCK(lbp);
@@ -1207,10 +1207,10 @@ xfs_inobt_newroot(
/*
* Get a block & a buffer.
*/
- agi = XFS_BUF_TO_AGI(cur->bc_private.i.agbp);
+ agi = XFS_BUF_TO_AGI(cur->bc_private.a.agbp);
args.tp = cur->bc_tp;
args.mp = cur->bc_mp;
- args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_private.i.agno,
+ args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_private.a.agno,
be32_to_cpu(agi->agi_root));
args.mod = args.minleft = args.alignment = args.total = args.wasdel =
args.isfl = args.userdata = args.minalignslop = 0;
@@ -1233,7 +1233,7 @@ xfs_inobt_newroot(
*/
agi->agi_root = cpu_to_be32(args.agbno);
be32_add_cpu(&agi->agi_level, 1);
- xfs_ialloc_log_agi(args.tp, cur->bc_private.i.agbp,
+ xfs_ialloc_log_agi(args.tp, cur->bc_private.a.agbp,
XFS_AGI_ROOT | XFS_AGI_LEVEL);
/*
* At the previous root level there are now two blocks: the old
@@ -1376,7 +1376,7 @@ xfs_inobt_rshift(
* Set up the right neighbor as "right".
*/
if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp,
- cur->bc_private.i.agno, be32_to_cpu(left->bb_rightsib),
+ cur->bc_private.a.agno, be32_to_cpu(left->bb_rightsib),
0, &rbp, XFS_INO_BTREE_REF)))
return error;
right = XFS_BUF_TO_INOBT_BLOCK(rbp);
@@ -1492,7 +1492,7 @@ xfs_inobt_split(
* Allocate the new block.
* If we can't do it, we're toast. Give up.
*/
- args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_private.i.agno, lbno);
+ args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_private.a.agno, lbno);
args.mod = args.minleft = args.alignment = args.total = args.wasdel =
args.isfl = args.userdata = args.minalignslop = 0;
args.minlen = args.maxlen = args.prod = 1;
@@ -1725,7 +1725,7 @@ xfs_inobt_decrement(
agbno = be32_to_cpu(*XFS_INOBT_PTR_ADDR(block, cur->bc_ptrs[lev], cur));
if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp,
- cur->bc_private.i.agno, agbno, 0, &bp,
+ cur->bc_private.a.agno, agbno, 0, &bp,
XFS_INO_BTREE_REF)))
return error;
lev--;
@@ -1897,7 +1897,7 @@ xfs_inobt_increment(
agbno = be32_to_cpu(*XFS_INOBT_PTR_ADDR(block, cur->bc_ptrs[lev], cur));
if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp,
- cur->bc_private.i.agno, agbno, 0, &bp,
+ cur->bc_private.a.agno, agbno, 0, &bp,
XFS_INO_BTREE_REF)))
return error;
lev--;
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index b07604b94d9..e229e9e001c 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -216,7 +216,14 @@ finish_inode:
mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino);
init_waitqueue_head(&ip->i_ipin_wait);
atomic_set(&ip->i_pincount, 0);
- initnsema(&ip->i_flock, 1, "xfsfino");
+
+ /*
+ * Because we want to use a counting completion, complete
+ * the flush completion once to allow a single access to
+ * the flush completion without blocking.
+ */
+ init_completion(&ip->i_flush);
+ complete(&ip->i_flush);
if (lock_flags)
xfs_ilock(ip, lock_flags);
@@ -288,10 +295,17 @@ finish_inode:
*ipp = ip;
/*
+ * Set up the Linux with the Linux inode.
+ */
+ ip->i_vnode = inode;
+ inode->i_private = ip;
+
+ /*
* If we have a real type for an on-disk inode, we can set ops(&unlock)
* now. If it's a new inode being created, xfs_ialloc will handle it.
*/
- xfs_initialize_vnode(mp, inode, ip);
+ if (ip->i_d.di_mode != 0)
+ xfs_setup_inode(ip);
return 0;
}
@@ -411,10 +425,11 @@ xfs_iput(xfs_inode_t *ip,
* Special iput for brand-new inodes that are still locked
*/
void
-xfs_iput_new(xfs_inode_t *ip,
- uint lock_flags)
+xfs_iput_new(
+ xfs_inode_t *ip,
+ uint lock_flags)
{
- struct inode *inode = ip->i_vnode;
+ struct inode *inode = VFS_I(ip);
xfs_itrace_entry(ip);
@@ -775,26 +790,3 @@ xfs_isilocked(
}
#endif
-/*
- * The following three routines simply manage the i_flock
- * semaphore embedded in the inode. This semaphore synchronizes
- * processes attempting to flush the in-core inode back to disk.
- */
-void
-xfs_iflock(xfs_inode_t *ip)
-{
- psema(&(ip->i_flock), PINOD|PLTWAIT);
-}
-
-int
-xfs_iflock_nowait(xfs_inode_t *ip)
-{
- return (cpsema(&(ip->i_flock)));
-}
-
-void
-xfs_ifunlock(xfs_inode_t *ip)
-{
- ASSERT(issemalocked(&(ip->i_flock)));
- vsema(&(ip->i_flock));
-}
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index e569bf5d6cf..dbd9cef852e 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -580,8 +580,8 @@ xfs_iformat_extents(
xfs_validate_extents(ifp, nex, XFS_EXTFMT_INODE(ip));
for (i = 0; i < nex; i++, dp++) {
xfs_bmbt_rec_host_t *ep = xfs_iext_get_ext(ifp, i);
- ep->l0 = be64_to_cpu(get_unaligned(&dp->l0));
- ep->l1 = be64_to_cpu(get_unaligned(&dp->l1));
+ ep->l0 = get_unaligned_be64(&dp->l0);
+ ep->l1 = get_unaligned_be64(&dp->l1);
}
XFS_BMAP_TRACE_EXLIST(ip, nex, whichfork);
if (whichfork != XFS_DATA_FORK ||
@@ -835,22 +835,22 @@ xfs_iread(
* Do this before xfs_iformat in case it adds entries.
*/
#ifdef XFS_INODE_TRACE
- ip->i_trace = ktrace_alloc(INODE_TRACE_SIZE, KM_SLEEP);
+ ip->i_trace = ktrace_alloc(INODE_TRACE_SIZE, KM_NOFS);
#endif
#ifdef XFS_BMAP_TRACE
- ip->i_xtrace = ktrace_alloc(XFS_BMAP_KTRACE_SIZE, KM_SLEEP);
+ ip->i_xtrace = ktrace_alloc(XFS_BMAP_KTRACE_SIZE, KM_NOFS);
#endif
#ifdef XFS_BMBT_TRACE
- ip->i_btrace = ktrace_alloc(XFS_BMBT_KTRACE_SIZE, KM_SLEEP);
+ ip->i_btrace = ktrace_alloc(XFS_BMBT_KTRACE_SIZE, KM_NOFS);
#endif
#ifdef XFS_RW_TRACE
- ip->i_rwtrace = ktrace_alloc(XFS_RW_KTRACE_SIZE, KM_SLEEP);
+ ip->i_rwtrace = ktrace_alloc(XFS_RW_KTRACE_SIZE, KM_NOFS);
#endif
#ifdef XFS_ILOCK_TRACE
- ip->i_lock_trace = ktrace_alloc(XFS_ILOCK_KTRACE_SIZE, KM_SLEEP);
+ ip->i_lock_trace = ktrace_alloc(XFS_ILOCK_KTRACE_SIZE, KM_NOFS);
#endif
#ifdef XFS_DIR2_TRACE
- ip->i_dir_trace = ktrace_alloc(XFS_DIR2_KTRACE_SIZE, KM_SLEEP);
+ ip->i_dir_trace = ktrace_alloc(XFS_DIR2_KTRACE_SIZE, KM_NOFS);
#endif
/*
@@ -1046,9 +1046,9 @@ xfs_ialloc(
{
xfs_ino_t ino;
xfs_inode_t *ip;
- bhv_vnode_t *vp;
uint flags;
int error;
+ timespec_t tv;
/*
* Call the space management code to pick
@@ -1077,13 +1077,12 @@ xfs_ialloc(
}
ASSERT(ip != NULL);
- vp = XFS_ITOV(ip);
ip->i_d.di_mode = (__uint16_t)mode;
ip->i_d.di_onlink = 0;
ip->i_d.di_nlink = nlink;
ASSERT(ip->i_d.di_nlink == nlink);
- ip->i_d.di_uid = current_fsuid(cr);
- ip->i_d.di_gid = current_fsgid(cr);
+ ip->i_d.di_uid = current_fsuid();
+ ip->i_d.di_gid = current_fsgid();
ip->i_d.di_projid = prid;
memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad));
@@ -1130,7 +1129,13 @@ xfs_ialloc(
ip->i_size = 0;
ip->i_d.di_nextents = 0;
ASSERT(ip->i_d.di_nblocks == 0);
- xfs_ichgtime(ip, XFS_ICHGTIME_CHG|XFS_ICHGTIME_ACC|XFS_ICHGTIME_MOD);
+
+ nanotime(&tv);
+ ip->i_d.di_mtime.t_sec = (__int32_t)tv.tv_sec;
+ ip->i_d.di_mtime.t_nsec = (__int32_t)tv.tv_nsec;
+ ip->i_d.di_atime = ip->i_d.di_mtime;
+ ip->i_d.di_ctime = ip->i_d.di_mtime;
+
/*
* di_gen will have been taken care of in xfs_iread.
*/
@@ -1220,7 +1225,7 @@ xfs_ialloc(
xfs_trans_log_inode(tp, ip, flags);
/* now that we have an i_mode we can setup inode ops and unlock */
- xfs_initialize_vnode(tp->t_mountp, vp, ip);
+ xfs_setup_inode(ip);
*ipp = ip;
return 0;
@@ -1399,7 +1404,6 @@ xfs_itruncate_start(
xfs_fsize_t last_byte;
xfs_off_t toss_start;
xfs_mount_t *mp;
- bhv_vnode_t *vp;
int error = 0;
ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
@@ -1408,7 +1412,6 @@ xfs_itruncate_start(
(flags == XFS_ITRUNC_MAYBE));
mp = ip->i_mount;
- vp = XFS_ITOV(ip);
/* wait for the completion of any pending DIOs */
if (new_size < ip->i_size)
@@ -1457,7 +1460,7 @@ xfs_itruncate_start(
#ifdef DEBUG
if (new_size == 0) {
- ASSERT(VN_CACHED(vp) == 0);
+ ASSERT(VN_CACHED(VFS_I(ip)) == 0);
}
#endif
return error;
@@ -1763,67 +1766,6 @@ xfs_itruncate_finish(
return 0;
}
-
-/*
- * xfs_igrow_start
- *
- * Do the first part of growing a file: zero any data in the last
- * block that is beyond the old EOF. We need to do this before
- * the inode is joined to the transaction to modify the i_size.
- * That way we can drop the inode lock and call into the buffer
- * cache to get the buffer mapping the EOF.
- */
-int
-xfs_igrow_start(
- xfs_inode_t *ip,
- xfs_fsize_t new_size,
- cred_t *credp)
-{
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
- ASSERT(new_size > ip->i_size);
-
- /*
- * Zero any pages that may have been created by
- * xfs_write_file() beyond the end of the file
- * and any blocks between the old and new file sizes.
- */
- return xfs_zero_eof(ip, new_size, ip->i_size);
-}
-
-/*
- * xfs_igrow_finish
- *
- * This routine is called to extend the size of a file.
- * The inode must have both the iolock and the ilock locked
- * for update and it must be a part of the current transaction.
- * The xfs_igrow_start() function must have been called previously.
- * If the change_flag is not zero, the inode change timestamp will
- * be updated.
- */
-void
-xfs_igrow_finish(
- xfs_trans_t *tp,
- xfs_inode_t *ip,
- xfs_fsize_t new_size,
- int change_flag)
-{
- ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_IOLOCK_EXCL));
- ASSERT(ip->i_transp == tp);
- ASSERT(new_size > ip->i_size);
-
- /*
- * Update the file size. Update the inode change timestamp
- * if change_flag set.
- */
- ip->i_d.di_size = new_size;
- ip->i_size = new_size;
- if (change_flag)
- xfs_ichgtime(ip, XFS_ICHGTIME_CHG);
- xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
-
-}
-
-
/*
* This is called when the inode's link count goes to 0.
* We place the on-disk inode on a list in the AGI. It
@@ -2258,7 +2200,7 @@ xfs_ifree_cluster(
xfs_trans_binval(tp, bp);
}
- kmem_free(ip_found, ninodes * sizeof(xfs_inode_t *));
+ kmem_free(ip_found);
xfs_put_perag(mp, pag);
}
@@ -2470,7 +2412,7 @@ xfs_iroot_realloc(
(int)new_size);
memcpy(np, op, new_max * (uint)sizeof(xfs_dfsbno_t));
}
- kmem_free(ifp->if_broot, ifp->if_broot_bytes);
+ kmem_free(ifp->if_broot);
ifp->if_broot = new_broot;
ifp->if_broot_bytes = (int)new_size;
ASSERT(ifp->if_broot_bytes <=
@@ -2514,7 +2456,7 @@ xfs_idata_realloc(
if (new_size == 0) {
if (ifp->if_u1.if_data != ifp->if_u2.if_inline_data) {
- kmem_free(ifp->if_u1.if_data, ifp->if_real_bytes);
+ kmem_free(ifp->if_u1.if_data);
}
ifp->if_u1.if_data = NULL;
real_size = 0;
@@ -2529,7 +2471,7 @@ xfs_idata_realloc(
ASSERT(ifp->if_real_bytes != 0);
memcpy(ifp->if_u2.if_inline_data, ifp->if_u1.if_data,
new_size);
- kmem_free(ifp->if_u1.if_data, ifp->if_real_bytes);
+ kmem_free(ifp->if_u1.if_data);
ifp->if_u1.if_data = ifp->if_u2.if_inline_data;
}
real_size = 0;
@@ -2636,7 +2578,7 @@ xfs_idestroy_fork(
ifp = XFS_IFORK_PTR(ip, whichfork);
if (ifp->if_broot != NULL) {
- kmem_free(ifp->if_broot, ifp->if_broot_bytes);
+ kmem_free(ifp->if_broot);
ifp->if_broot = NULL;
}
@@ -2650,7 +2592,7 @@ xfs_idestroy_fork(
if ((ifp->if_u1.if_data != ifp->if_u2.if_inline_data) &&
(ifp->if_u1.if_data != NULL)) {
ASSERT(ifp->if_real_bytes != 0);
- kmem_free(ifp->if_u1.if_data, ifp->if_real_bytes);
+ kmem_free(ifp->if_u1.if_data);
ifp->if_u1.if_data = NULL;
ifp->if_real_bytes = 0;
}
@@ -2691,7 +2633,6 @@ xfs_idestroy(
xfs_idestroy_fork(ip, XFS_ATTR_FORK);
mrfree(&ip->i_lock);
mrfree(&ip->i_iolock);
- freesema(&ip->i_flock);
#ifdef XFS_INODE_TRACE
ktrace_free(ip->i_trace);
@@ -3058,7 +2999,7 @@ xfs_iflush_cluster(
out_free:
read_unlock(&pag->pag_ici_lock);
- kmem_free(ilist, ilist_size);
+ kmem_free(ilist);
return 0;
@@ -3102,17 +3043,17 @@ cluster_corrupt_out:
* Unlocks the flush lock
*/
xfs_iflush_abort(iq);
- kmem_free(ilist, ilist_size);
+ kmem_free(ilist);
return XFS_ERROR(EFSCORRUPTED);
}
/*
* xfs_iflush() will write a modified inode's changes out to the
* inode's on disk home. The caller must have the inode lock held
- * in at least shared mode and the inode flush semaphore must be
- * held as well. The inode lock will still be held upon return from
+ * in at least shared mode and the inode flush completion must be
+ * active as well. The inode lock will still be held upon return from
* the call and the caller is free to unlock it.
- * The inode flush lock will be unlocked when the inode reaches the disk.
+ * The inode flush will be completed when the inode reaches the disk.
* The flags indicate how the inode's buffer should be written out.
*/
int
@@ -3131,7 +3072,7 @@ xfs_iflush(
XFS_STATS_INC(xs_iflush_count);
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
- ASSERT(issemalocked(&(ip->i_flock)));
+ ASSERT(!completion_done(&ip->i_flush));
ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
ip->i_d.di_nextents > ip->i_df.if_ext_max);
@@ -3143,8 +3084,6 @@ xfs_iflush(
* flush lock and do nothing.
*/
if (xfs_inode_clean(ip)) {
- ASSERT((iip != NULL) ?
- !(iip->ili_item.li_flags & XFS_LI_IN_AIL) : 1);
xfs_ifunlock(ip);
return 0;
}
@@ -3296,7 +3235,7 @@ xfs_iflush_int(
#endif
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED));
- ASSERT(issemalocked(&(ip->i_flock)));
+ ASSERT(!completion_done(&ip->i_flush));
ASSERT(ip->i_d.di_format != XFS_DINODE_FMT_BTREE ||
ip->i_d.di_nextents > ip->i_df.if_ext_max);
@@ -3528,7 +3467,6 @@ xfs_iflush_all(
xfs_mount_t *mp)
{
xfs_inode_t *ip;
- bhv_vnode_t *vp;
again:
XFS_MOUNT_ILOCK(mp);
@@ -3543,14 +3481,13 @@ xfs_iflush_all(
continue;
}
- vp = XFS_ITOV_NULL(ip);
- if (!vp) {
+ if (!VFS_I(ip)) {
XFS_MOUNT_IUNLOCK(mp);
xfs_finish_reclaim(ip, 0, XFS_IFLUSH_ASYNC);
goto again;
}
- ASSERT(vn_count(vp) == 0);
+ ASSERT(vn_count(VFS_I(ip)) == 0);
ip = ip->i_mnext;
} while (ip != mp->m_inodes);
@@ -3770,7 +3707,7 @@ xfs_iext_add_indirect_multi(
* (all extents past */
if (nex2) {
byte_diff = nex2 * sizeof(xfs_bmbt_rec_t);
- nex2_ep = (xfs_bmbt_rec_t *) kmem_alloc(byte_diff, KM_SLEEP);
+ nex2_ep = (xfs_bmbt_rec_t *) kmem_alloc(byte_diff, KM_NOFS);
memmove(nex2_ep, &erp->er_extbuf[idx], byte_diff);
erp->er_extcount -= nex2;
xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, -nex2);
@@ -3836,7 +3773,7 @@ xfs_iext_add_indirect_multi(
erp = xfs_iext_irec_new(ifp, erp_idx);
}
memmove(&erp->er_extbuf[i], nex2_ep, byte_diff);
- kmem_free(nex2_ep, byte_diff);
+ kmem_free(nex2_ep);
erp->er_extcount += nex2;
xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, nex2);
}
@@ -4070,8 +4007,7 @@ xfs_iext_realloc_direct(
ifp->if_u1.if_extents =
kmem_realloc(ifp->if_u1.if_extents,
rnew_size,
- ifp->if_real_bytes,
- KM_SLEEP);
+ ifp->if_real_bytes, KM_NOFS);
}
if (rnew_size > ifp->if_real_bytes) {
memset(&ifp->if_u1.if_extents[ifp->if_bytes /
@@ -4112,7 +4048,7 @@ xfs_iext_direct_to_inline(
*/
memcpy(ifp->if_u2.if_inline_ext, ifp->if_u1.if_extents,
nextents * sizeof(xfs_bmbt_rec_t));
- kmem_free(ifp->if_u1.if_extents, ifp->if_real_bytes);
+ kmem_free(ifp->if_u1.if_extents);
ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext;
ifp->if_real_bytes = 0;
}
@@ -4130,7 +4066,7 @@ xfs_iext_inline_to_direct(
xfs_ifork_t *ifp, /* inode fork pointer */
int new_size) /* number of extents in file */
{
- ifp->if_u1.if_extents = kmem_alloc(new_size, KM_SLEEP);
+ ifp->if_u1.if_extents = kmem_alloc(new_size, KM_NOFS);
memset(ifp->if_u1.if_extents, 0, new_size);
if (ifp->if_bytes) {
memcpy(ifp->if_u1.if_extents, ifp->if_u2.if_inline_ext,
@@ -4162,7 +4098,7 @@ xfs_iext_realloc_indirect(
} else {
ifp->if_u1.if_ext_irec = (xfs_ext_irec_t *)
kmem_realloc(ifp->if_u1.if_ext_irec,
- new_size, size, KM_SLEEP);
+ new_size, size, KM_NOFS);
}
}
@@ -4182,11 +4118,11 @@ xfs_iext_indirect_to_direct(
ASSERT(nextents <= XFS_LINEAR_EXTS);
size = nextents * sizeof(xfs_bmbt_rec_t);
- xfs_iext_irec_compact_full(ifp);
+ xfs_iext_irec_compact_pages(ifp);
ASSERT(ifp->if_real_bytes == XFS_IEXT_BUFSZ);
ep = ifp->if_u1.if_ext_irec->er_extbuf;
- kmem_free(ifp->if_u1.if_ext_irec, sizeof(xfs_ext_irec_t));
+ kmem_free(ifp->if_u1.if_ext_irec);
ifp->if_flags &= ~XFS_IFEXTIREC;
ifp->if_u1.if_extents = ep;
ifp->if_bytes = size;
@@ -4212,7 +4148,7 @@ xfs_iext_destroy(
}
ifp->if_flags &= ~XFS_IFEXTIREC;
} else if (ifp->if_real_bytes) {
- kmem_free(ifp->if_u1.if_extents, ifp->if_real_bytes);
+ kmem_free(ifp->if_u1.if_extents);
} else if (ifp->if_bytes) {
memset(ifp->if_u2.if_inline_ext, 0, XFS_INLINE_EXTS *
sizeof(xfs_bmbt_rec_t));
@@ -4404,11 +4340,10 @@ xfs_iext_irec_init(
nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
ASSERT(nextents <= XFS_LINEAR_EXTS);
- erp = (xfs_ext_irec_t *)
- kmem_alloc(sizeof(xfs_ext_irec_t), KM_SLEEP);
+ erp = kmem_alloc(sizeof(xfs_ext_irec_t), KM_NOFS);
if (nextents == 0) {
- ifp->if_u1.if_extents = kmem_alloc(XFS_IEXT_BUFSZ, KM_SLEEP);
+ ifp->if_u1.if_extents = kmem_alloc(XFS_IEXT_BUFSZ, KM_NOFS);
} else if (!ifp->if_real_bytes) {
xfs_iext_inline_to_direct(ifp, XFS_IEXT_BUFSZ);
} else if (ifp->if_real_bytes < XFS_IEXT_BUFSZ) {
@@ -4456,7 +4391,7 @@ xfs_iext_irec_new(
/* Initialize new extent record */
erp = ifp->if_u1.if_ext_irec;
- erp[erp_idx].er_extbuf = kmem_alloc(XFS_IEXT_BUFSZ, KM_SLEEP);
+ erp[erp_idx].er_extbuf = kmem_alloc(XFS_IEXT_BUFSZ, KM_NOFS);
ifp->if_real_bytes = nlists * XFS_IEXT_BUFSZ;
memset(erp[erp_idx].er_extbuf, 0, XFS_IEXT_BUFSZ);
erp[erp_idx].er_extcount = 0;
@@ -4483,7 +4418,7 @@ xfs_iext_irec_remove(
if (erp->er_extbuf) {
xfs_iext_irec_update_extoffs(ifp, erp_idx + 1,
-erp->er_extcount);
- kmem_free(erp->er_extbuf, XFS_IEXT_BUFSZ);
+ kmem_free(erp->er_extbuf);
}
/* Compact extent records */
erp = ifp->if_u1.if_ext_irec;
@@ -4501,8 +4436,7 @@ xfs_iext_irec_remove(
xfs_iext_realloc_indirect(ifp,
nlists * sizeof(xfs_ext_irec_t));
} else {
- kmem_free(ifp->if_u1.if_ext_irec,
- sizeof(xfs_ext_irec_t));
+ kmem_free(ifp->if_u1.if_ext_irec);
}
ifp->if_real_bytes = nlists * XFS_IEXT_BUFSZ;
}
@@ -4515,8 +4449,7 @@ xfs_iext_irec_remove(
* compaction policy is as follows:
*
* Full Compaction: Extents fit into a single page (or inline buffer)
- * Full Compaction: Extents occupy less than 10% of allocated space
- * Partial Compaction: Extents occupy > 10% and < 50% of allocated space
+ * Partial Compaction: Extents occupy less than 50% of allocated space
* No Compaction: Extents occupy at least 50% of allocated space
*/
void
@@ -4537,8 +4470,6 @@ xfs_iext_irec_compact(
xfs_iext_direct_to_inline(ifp, nextents);
} else if (nextents <= XFS_LINEAR_EXTS) {
xfs_iext_indirect_to_direct(ifp);
- } else if (nextents < (nlists * XFS_LINEAR_EXTS) >> 3) {
- xfs_iext_irec_compact_full(ifp);
} else if (nextents < (nlists * XFS_LINEAR_EXTS) >> 1) {
xfs_iext_irec_compact_pages(ifp);
}
@@ -4562,7 +4493,7 @@ xfs_iext_irec_compact_pages(
erp_next = erp + 1;
if (erp_next->er_extcount <=
(XFS_LINEAR_EXTS - erp->er_extcount)) {
- memmove(&erp->er_extbuf[erp->er_extcount],
+ memcpy(&erp->er_extbuf[erp->er_extcount],
erp_next->er_extbuf, erp_next->er_extcount *
sizeof(xfs_bmbt_rec_t));
erp->er_extcount += erp_next->er_extcount;
@@ -4571,75 +4502,13 @@ xfs_iext_irec_compact_pages(
* so er_extoffs don't get modified in
* xfs_iext_irec_remove.
*/
- kmem_free(erp_next->er_extbuf, XFS_IEXT_BUFSZ);
- erp_next->er_extbuf = NULL;
- xfs_iext_irec_remove(ifp, erp_idx + 1);
- nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
- } else {
- erp_idx++;
- }
- }
-}
-
-/*
- * Fully compact the extent records managed by the indirection array.
- */
-void
-xfs_iext_irec_compact_full(
- xfs_ifork_t *ifp) /* inode fork pointer */
-{
- xfs_bmbt_rec_host_t *ep, *ep_next; /* extent record pointers */
- xfs_ext_irec_t *erp, *erp_next; /* extent irec pointers */
- int erp_idx = 0; /* extent irec index */
- int ext_avail; /* empty entries in ex list */
- int ext_diff; /* number of exts to add */
- int nlists; /* number of irec's (ex lists) */
-
- ASSERT(ifp->if_flags & XFS_IFEXTIREC);
- nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
- erp = ifp->if_u1.if_ext_irec;
- ep = &erp->er_extbuf[erp->er_extcount];
- erp_next = erp + 1;
- ep_next = erp_next->er_extbuf;
- while (erp_idx < nlists - 1) {
- ext_avail = XFS_LINEAR_EXTS - erp->er_extcount;
- ext_diff = MIN(ext_avail, erp_next->er_extcount);
- memcpy(ep, ep_next, ext_diff * sizeof(xfs_bmbt_rec_t));
- erp->er_extcount += ext_diff;
- erp_next->er_extcount -= ext_diff;
- /* Remove next page */
- if (erp_next->er_extcount == 0) {
- /*
- * Free page before removing extent record
- * so er_extoffs don't get modified in
- * xfs_iext_irec_remove.
- */
- kmem_free(erp_next->er_extbuf,
- erp_next->er_extcount * sizeof(xfs_bmbt_rec_t));
+ kmem_free(erp_next->er_extbuf);
erp_next->er_extbuf = NULL;
xfs_iext_irec_remove(ifp, erp_idx + 1);
- erp = &ifp->if_u1.if_ext_irec[erp_idx];
nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
- /* Update next page */
} else {
- /* Move rest of page up to become next new page */
- memmove(erp_next->er_extbuf, ep_next,
- erp_next->er_extcount * sizeof(xfs_bmbt_rec_t));
- ep_next = erp_next->er_extbuf;
- memset(&ep_next[erp_next->er_extcount], 0,
- (XFS_LINEAR_EXTS - erp_next->er_extcount) *
- sizeof(xfs_bmbt_rec_t));
- }
- if (erp->er_extcount == XFS_LINEAR_EXTS) {
erp_idx++;
- if (erp_idx < nlists)
- erp = &ifp->if_u1.if_ext_irec[erp_idx];
- else
- break;
}
- ep = &erp->er_extbuf[erp->er_extcount];
- erp_next = erp + 1;
- ep_next = erp_next->er_extbuf;
}
}
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 0a999fee4f0..1420c49674d 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -87,8 +87,7 @@ typedef struct xfs_ifork {
* Flags for xfs_ichgtime().
*/
#define XFS_ICHGTIME_MOD 0x1 /* data fork modification timestamp */
-#define XFS_ICHGTIME_ACC 0x2 /* data fork access timestamp */
-#define XFS_ICHGTIME_CHG 0x4 /* inode field change timestamp */
+#define XFS_ICHGTIME_CHG 0x2 /* inode field change timestamp */
/*
* Per-fork incore inode flags.
@@ -204,7 +203,7 @@ typedef struct xfs_inode {
struct xfs_inode *i_mprev; /* ptr to prev inode */
struct xfs_mount *i_mount; /* fs mount struct ptr */
struct list_head i_reclaim; /* reclaim list */
- bhv_vnode_t *i_vnode; /* vnode backpointer */
+ struct inode *i_vnode; /* vnode backpointer */
struct xfs_dquot *i_udquot; /* user dquot */
struct xfs_dquot *i_gdquot; /* group dquot */
@@ -223,7 +222,7 @@ typedef struct xfs_inode {
struct xfs_inode_log_item *i_itemp; /* logging information */
mrlock_t i_lock; /* inode lock */
mrlock_t i_iolock; /* inode IO lock */
- sema_t i_flock; /* inode flush lock */
+ struct completion i_flush; /* inode flush completion q */
atomic_t i_pincount; /* inode pin count */
wait_queue_head_t i_ipin_wait; /* inode pinning wait queue */
spinlock_t i_flags_lock; /* inode i_flags lock */
@@ -263,6 +262,18 @@ typedef struct xfs_inode {
#define XFS_ISIZE(ip) (((ip)->i_d.di_mode & S_IFMT) == S_IFREG) ? \
(ip)->i_size : (ip)->i_d.di_size;
+/* Convert from vfs inode to xfs inode */
+static inline struct xfs_inode *XFS_I(struct inode *inode)
+{
+ return (struct xfs_inode *)inode->i_private;
+}
+
+/* convert from xfs inode to vfs inode */
+static inline struct inode *VFS_I(struct xfs_inode *ip)
+{
+ return (struct inode *)ip->i_vnode;
+}
+
/*
* i_flags helper functions
*/
@@ -439,9 +450,6 @@ xfs_iflags_test_and_clear(xfs_inode_t *ip, unsigned short flags)
#define XFS_ITRUNC_DEFINITE 0x1
#define XFS_ITRUNC_MAYBE 0x2
-#define XFS_ITOV(ip) ((ip)->i_vnode)
-#define XFS_ITOV_NULL(ip) ((ip)->i_vnode)
-
/*
* For multiple groups support: if S_ISGID bit is set in the parent
* directory, group of new file is set to that of the parent, and
@@ -473,11 +481,8 @@ int xfs_ilock_nowait(xfs_inode_t *, uint);
void xfs_iunlock(xfs_inode_t *, uint);
void xfs_ilock_demote(xfs_inode_t *, uint);
int xfs_isilocked(xfs_inode_t *, uint);
-void xfs_iflock(xfs_inode_t *);
-int xfs_iflock_nowait(xfs_inode_t *);
uint xfs_ilock_map_shared(xfs_inode_t *);
void xfs_iunlock_map_shared(xfs_inode_t *, uint);
-void xfs_ifunlock(xfs_inode_t *);
void xfs_ireclaim(xfs_inode_t *);
int xfs_finish_reclaim(xfs_inode_t *, int, int);
int xfs_finish_reclaim_all(struct xfs_mount *, int);
@@ -507,9 +512,6 @@ int xfs_itruncate_start(xfs_inode_t *, uint, xfs_fsize_t);
int xfs_itruncate_finish(struct xfs_trans **, xfs_inode_t *,
xfs_fsize_t, int, int);
int xfs_iunlink(struct xfs_trans *, xfs_inode_t *);
-int xfs_igrow_start(xfs_inode_t *, xfs_fsize_t, struct cred *);
-void xfs_igrow_finish(struct xfs_trans *, xfs_inode_t *,
- xfs_fsize_t, int);
void xfs_idestroy_fork(xfs_inode_t *, int);
void xfs_idestroy(xfs_inode_t *);
@@ -525,6 +527,7 @@ void xfs_iflush_all(struct xfs_mount *);
void xfs_ichgtime(xfs_inode_t *, int);
xfs_fsize_t xfs_file_last_byte(xfs_inode_t *);
void xfs_lock_inodes(xfs_inode_t **, int, uint);
+void xfs_lock_two_inodes(xfs_inode_t *, xfs_inode_t *, uint);
void xfs_synchronize_atime(xfs_inode_t *);
void xfs_mark_inode_dirty_sync(xfs_inode_t *);
@@ -573,6 +576,26 @@ extern struct kmem_zone *xfs_ifork_zone;
extern struct kmem_zone *xfs_inode_zone;
extern struct kmem_zone *xfs_ili_zone;
+/*
+ * Manage the i_flush queue embedded in the inode. This completion
+ * queue synchronizes processes attempting to flush the in-core
+ * inode back to disk.
+ */
+static inline void xfs_iflock(xfs_inode_t *ip)
+{
+ wait_for_completion(&ip->i_flush);
+}
+
+static inline int xfs_iflock_nowait(xfs_inode_t *ip)
+{
+ return try_wait_for_completion(&ip->i_flush);
+}
+
+static inline void xfs_ifunlock(xfs_inode_t *ip)
+{
+ complete(&ip->i_flush);
+}
+
#endif /* __KERNEL__ */
#endif /* __XFS_INODE_H__ */
diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c
index 167b33f1577..97c7452e262 100644
--- a/fs/xfs/xfs_inode_item.c
+++ b/fs/xfs/xfs_inode_item.c
@@ -686,7 +686,7 @@ xfs_inode_item_unlock(
ASSERT(ip->i_d.di_nextents > 0);
ASSERT(iip->ili_format.ilf_fields & XFS_ILOG_DEXT);
ASSERT(ip->i_df.if_bytes > 0);
- kmem_free(iip->ili_extents_buf, ip->i_df.if_bytes);
+ kmem_free(iip->ili_extents_buf);
iip->ili_extents_buf = NULL;
}
if (iip->ili_aextents_buf != NULL) {
@@ -694,7 +694,7 @@ xfs_inode_item_unlock(
ASSERT(ip->i_d.di_anextents > 0);
ASSERT(iip->ili_format.ilf_fields & XFS_ILOG_AEXT);
ASSERT(ip->i_afp->if_bytes > 0);
- kmem_free(iip->ili_aextents_buf, ip->i_afp->if_bytes);
+ kmem_free(iip->ili_aextents_buf);
iip->ili_aextents_buf = NULL;
}
@@ -779,11 +779,10 @@ xfs_inode_item_pushbuf(
ASSERT(iip->ili_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 a flush is not in progress anymore, chances are that the
+ * inode was taken off the AIL. So, just get out.
*/
- if (!issemalocked(&(ip->i_flock)) ||
+ 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);
@@ -805,7 +804,7 @@ xfs_inode_item_pushbuf(
* If not, we can flush it async.
*/
dopush = ((iip->ili_item.li_flags & XFS_LI_IN_AIL) &&
- issemalocked(&(ip->i_flock)));
+ !completion_done(&ip->i_flush));
iip->ili_pushbuf_flag = 0;
xfs_iunlock(ip, XFS_ILOCK_SHARED);
xfs_buftrace("INODE ITEM PUSH", bp);
@@ -858,7 +857,7 @@ xfs_inode_item_push(
ip = iip->ili_inode;
ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED));
- ASSERT(issemalocked(&(ip->i_flock)));
+ ASSERT(!completion_done(&ip->i_flush));
/*
* Since we were able to lock the inode's flush lock and
* we found it on the AIL, the inode must be dirty. This
@@ -957,8 +956,7 @@ xfs_inode_item_destroy(
{
#ifdef XFS_TRANS_DEBUG
if (ip->i_itemp->ili_root_size != 0) {
- kmem_free(ip->i_itemp->ili_orig_root,
- ip->i_itemp->ili_root_size);
+ kmem_free(ip->i_itemp->ili_orig_root);
}
#endif
kmem_zone_free(xfs_ili_zone, ip->i_itemp);
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 7edcde691d1..67f22b2b44b 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -889,6 +889,16 @@ xfs_iomap_write_unwritten(
count_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
count_fsb = (xfs_filblks_t)(count_fsb - offset_fsb);
+ /*
+ * Reserve enough blocks in this transaction for two complete extent
+ * btree splits. We may be converting the middle part of an unwritten
+ * extent and in this case we will insert two new extents in the btree
+ * each of which could cause a full split.
+ *
+ * This reservation amount will be used in the first call to
+ * xfs_bmbt_split() to select an AG with enough space to satisfy the
+ * rest of the operation.
+ */
resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1;
do {
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index 419de15aeb4..cf6754a3c5b 100644
--- a/fs/xfs/xfs_itable.c
+++ b/fs/xfs/xfs_itable.c
@@ -59,7 +59,6 @@ xfs_bulkstat_one_iget(
{
xfs_icdinode_t *dic; /* dinode core info pointer */
xfs_inode_t *ip; /* incore inode pointer */
- bhv_vnode_t *vp;
int error;
error = xfs_iget(mp, NULL, ino,
@@ -72,7 +71,6 @@ xfs_bulkstat_one_iget(
ASSERT(ip != NULL);
ASSERT(ip->i_blkno != (xfs_daddr_t)0);
- vp = XFS_ITOV(ip);
dic = &ip->i_d;
/* xfs_iget returns the following without needing
@@ -85,7 +83,7 @@ xfs_bulkstat_one_iget(
buf->bs_uid = dic->di_uid;
buf->bs_gid = dic->di_gid;
buf->bs_size = dic->di_size;
- vn_atime_to_bstime(vp, &buf->bs_atime);
+ vn_atime_to_bstime(VFS_I(ip), &buf->bs_atime);
buf->bs_mtime.tv_sec = dic->di_mtime.t_sec;
buf->bs_mtime.tv_nsec = dic->di_mtime.t_nsec;
buf->bs_ctime.tv_sec = dic->di_ctime.t_sec;
@@ -257,7 +255,7 @@ xfs_bulkstat_one(
*ubused = error;
out_free:
- kmem_free(buf, sizeof(*buf));
+ kmem_free(buf);
return error;
}
@@ -708,7 +706,7 @@ xfs_bulkstat(
/*
* Done, we're either out of filesystem or space to put the data.
*/
- kmem_free(irbuf, irbsize);
+ kmem_free(irbuf);
*ubcountp = ubelem;
/*
* Found some inodes, return them now and return the error next time.
@@ -914,7 +912,7 @@ xfs_inumbers(
}
*lastino = XFS_AGINO_TO_INO(mp, agno, agino);
}
- kmem_free(buffer, bcount * sizeof(*buffer));
+ kmem_free(buffer);
if (cur)
xfs_btree_del_cursor(cur, (error ? XFS_BTREE_ERROR :
XFS_BTREE_NOERROR));
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index afaee301b0e..0b02c644355 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -124,16 +124,27 @@ STATIC void xlog_verify_tail_lsn(xlog_t *log, xlog_in_core_t *iclog,
STATIC int xlog_iclogs_empty(xlog_t *log);
#if defined(XFS_LOG_TRACE)
+
+#define XLOG_TRACE_LOGGRANT_SIZE 2048
+#define XLOG_TRACE_ICLOG_SIZE 256
+
+void
+xlog_trace_loggrant_alloc(xlog_t *log)
+{
+ log->l_grant_trace = ktrace_alloc(XLOG_TRACE_LOGGRANT_SIZE, KM_NOFS);
+}
+
+void
+xlog_trace_loggrant_dealloc(xlog_t *log)
+{
+ ktrace_free(log->l_grant_trace);
+}
+
void
xlog_trace_loggrant(xlog_t *log, xlog_ticket_t *tic, xfs_caddr_t string)
{
unsigned long cnts;
- if (!log->l_grant_trace) {
- log->l_grant_trace = ktrace_alloc(2048, KM_NOSLEEP);
- if (!log->l_grant_trace)
- return;
- }
/* ticket counts are 1 byte each */
cnts = ((unsigned long)tic->t_ocnt) | ((unsigned long)tic->t_cnt) << 8;
@@ -157,10 +168,20 @@ xlog_trace_loggrant(xlog_t *log, xlog_ticket_t *tic, xfs_caddr_t string)
}
void
+xlog_trace_iclog_alloc(xlog_in_core_t *iclog)
+{
+ iclog->ic_trace = ktrace_alloc(XLOG_TRACE_ICLOG_SIZE, KM_NOFS);
+}
+
+void
+xlog_trace_iclog_dealloc(xlog_in_core_t *iclog)
+{
+ ktrace_free(iclog->ic_trace);
+}
+
+void
xlog_trace_iclog(xlog_in_core_t *iclog, uint state)
{
- if (!iclog->ic_trace)
- iclog->ic_trace = ktrace_alloc(256, KM_SLEEP);
ktrace_enter(iclog->ic_trace,
(void *)((unsigned long)state),
(void *)((unsigned long)current_pid()),
@@ -170,8 +191,15 @@ xlog_trace_iclog(xlog_in_core_t *iclog, uint state)
(void *)NULL, (void *)NULL);
}
#else
+
+#define xlog_trace_loggrant_alloc(log)
+#define xlog_trace_loggrant_dealloc(log)
#define xlog_trace_loggrant(log,tic,string)
+
+#define xlog_trace_iclog_alloc(iclog)
+#define xlog_trace_iclog_dealloc(iclog)
#define xlog_trace_iclog(iclog,state)
+
#endif /* XFS_LOG_TRACE */
@@ -226,20 +254,24 @@ xlog_grant_sub_space(struct log *log, int bytes)
static void
xlog_grant_add_space_write(struct log *log, int bytes)
{
- log->l_grant_write_bytes += bytes;
- if (log->l_grant_write_bytes > log->l_logsize) {
- log->l_grant_write_bytes -= log->l_logsize;
+ int tmp = log->l_logsize - log->l_grant_write_bytes;
+ if (tmp > bytes)
+ log->l_grant_write_bytes += bytes;
+ else {
log->l_grant_write_cycle++;
+ log->l_grant_write_bytes = bytes - tmp;
}
}
static void
xlog_grant_add_space_reserve(struct log *log, int bytes)
{
- log->l_grant_reserve_bytes += bytes;
- if (log->l_grant_reserve_bytes > log->l_logsize) {
- log->l_grant_reserve_bytes -= log->l_logsize;
+ int tmp = log->l_logsize - log->l_grant_reserve_bytes;
+ if (tmp > bytes)
+ log->l_grant_reserve_bytes += bytes;
+ else {
log->l_grant_reserve_cycle++;
+ log->l_grant_reserve_bytes = bytes - tmp;
}
}
@@ -332,15 +364,12 @@ xfs_log_done(xfs_mount_t *mp,
} else {
xlog_trace_loggrant(log, ticket, "xfs_log_done: (permanent)");
xlog_regrant_reserve_log_space(log, ticket);
- }
-
- /* If this ticket was a permanent reservation and we aren't
- * trying to release it, reset the inited flags; so next time
- * we write, a start record will be written out.
- */
- if ((ticket->t_flags & XLOG_TIC_PERM_RESERV) &&
- (flags & XFS_LOG_REL_PERM_RESERV) == 0)
+ /* If this ticket was a permanent reservation and we aren't
+ * trying to release it, reset the inited flags; so next time
+ * we write, a start record will be written out.
+ */
ticket->t_flags |= XLOG_TIC_INITED;
+ }
return lsn;
} /* xfs_log_done */
@@ -353,11 +382,11 @@ xfs_log_done(xfs_mount_t *mp,
* 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 semaphore. All callers
- * to force a given lsn to disk will wait on a semaphore attached to the
+ * 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
- * semaphore.
+ * sv.
*/
int
_xfs_log_force(
@@ -584,12 +613,12 @@ error:
* mp - ubiquitous xfs mount point structure
*/
int
-xfs_log_mount_finish(xfs_mount_t *mp, int mfsi_flags)
+xfs_log_mount_finish(xfs_mount_t *mp)
{
int error;
if (!(mp->m_flags & XFS_MOUNT_NORECOVERY))
- error = xlog_recover_finish(mp->m_log, mfsi_flags);
+ error = xlog_recover_finish(mp->m_log);
else {
error = 0;
ASSERT(mp->m_flags & XFS_MOUNT_RDONLY);
@@ -703,7 +732,7 @@ xfs_log_unmount_write(xfs_mount_t *mp)
if (!(iclog->ic_state == XLOG_STATE_ACTIVE ||
iclog->ic_state == XLOG_STATE_DIRTY)) {
if (!XLOG_FORCED_SHUTDOWN(log)) {
- sv_wait(&iclog->ic_forcesema, PMEM,
+ sv_wait(&iclog->ic_force_wait, PMEM,
&log->l_icloglock, s);
} else {
spin_unlock(&log->l_icloglock);
@@ -744,7 +773,7 @@ xfs_log_unmount_write(xfs_mount_t *mp)
|| iclog->ic_state == XLOG_STATE_DIRTY
|| iclog->ic_state == XLOG_STATE_IOERROR) ) {
- sv_wait(&iclog->ic_forcesema, PMEM,
+ sv_wait(&iclog->ic_force_wait, PMEM,
&log->l_icloglock, s);
} else {
spin_unlock(&log->l_icloglock);
@@ -834,7 +863,7 @@ xfs_log_move_tail(xfs_mount_t *mp,
break;
tail_lsn = 0;
free_bytes -= tic->t_unit_res;
- sv_signal(&tic->t_sema);
+ sv_signal(&tic->t_wait);
tic = tic->t_next;
} while (tic != log->l_write_headq);
}
@@ -855,7 +884,7 @@ xfs_log_move_tail(xfs_mount_t *mp,
break;
tail_lsn = 0;
free_bytes -= need_bytes;
- sv_signal(&tic->t_sema);
+ sv_signal(&tic->t_wait);
tic = tic->t_next;
} while (tic != log->l_reserve_headq);
}
@@ -1004,11 +1033,12 @@ xlog_iodone(xfs_buf_t *bp)
l = iclog->ic_log;
/*
- * If the ordered flag has been removed by a lower
- * layer, it means the underlyin device no longer supports
+ * If the _XFS_BARRIER_FAILED flag was set by a lower
+ * layer, it means the underlying device no longer supports
* barrier I/O. Warn loudly and turn off barriers.
*/
- if ((l->l_mp->m_flags & XFS_MOUNT_BARRIER) && !XFS_BUF_ORDERED(bp)) {
+ if (bp->b_flags & _XFS_BARRIER_FAILED) {
+ bp->b_flags &= ~_XFS_BARRIER_FAILED;
l->l_mp->m_flags &= ~XFS_MOUNT_BARRIER;
xfs_fs_cmn_err(CE_WARN, l->l_mp,
"xlog_iodone: Barriers are no longer supported"
@@ -1228,8 +1258,9 @@ xlog_alloc_log(xfs_mount_t *mp,
spin_lock_init(&log->l_icloglock);
spin_lock_init(&log->l_grant_lock);
- initnsema(&log->l_flushsema, 0, "ic-flush");
+ sv_init(&log->l_flush_wait, 0, "flush_wait");
+ xlog_trace_loggrant_alloc(log);
/* log record size must be multiple of BBSIZE; see xlog_rec_header_t */
ASSERT((XFS_BUF_SIZE(bp) & BBMASK) == 0);
@@ -1281,8 +1312,10 @@ xlog_alloc_log(xfs_mount_t *mp,
ASSERT(XFS_BUF_ISBUSY(iclog->ic_bp));
ASSERT(XFS_BUF_VALUSEMA(iclog->ic_bp) <= 0);
- sv_init(&iclog->ic_forcesema, SV_DEFAULT, "iclog-force");
- sv_init(&iclog->ic_writesema, SV_DEFAULT, "iclog-write");
+ sv_init(&iclog->ic_force_wait, SV_DEFAULT, "iclog-force");
+ sv_init(&iclog->ic_write_wait, SV_DEFAULT, "iclog-write");
+
+ xlog_trace_iclog_alloc(iclog);
iclogp = &iclog->ic_next;
}
@@ -1561,33 +1594,21 @@ xlog_dealloc_log(xlog_t *log)
iclog = log->l_iclog;
for (i=0; i<log->l_iclog_bufs; i++) {
- sv_destroy(&iclog->ic_forcesema);
- sv_destroy(&iclog->ic_writesema);
+ sv_destroy(&iclog->ic_force_wait);
+ sv_destroy(&iclog->ic_write_wait);
xfs_buf_free(iclog->ic_bp);
-#ifdef XFS_LOG_TRACE
- if (iclog->ic_trace != NULL) {
- ktrace_free(iclog->ic_trace);
- }
-#endif
+ xlog_trace_iclog_dealloc(iclog);
next_iclog = iclog->ic_next;
- kmem_free(iclog, sizeof(xlog_in_core_t));
+ kmem_free(iclog);
iclog = next_iclog;
}
- freesema(&log->l_flushsema);
spinlock_destroy(&log->l_icloglock);
spinlock_destroy(&log->l_grant_lock);
xfs_buf_free(log->l_xbuf);
-#ifdef XFS_LOG_TRACE
- if (log->l_trace != NULL) {
- ktrace_free(log->l_trace);
- }
- if (log->l_grant_trace != NULL) {
- ktrace_free(log->l_grant_trace);
- }
-#endif
+ xlog_trace_loggrant_dealloc(log);
log->l_mp->m_log = NULL;
- kmem_free(log, sizeof(xlog_t));
+ kmem_free(log);
} /* xlog_dealloc_log */
/*
@@ -1973,7 +1994,7 @@ xlog_write(xfs_mount_t * mp,
/* Clean iclogs starting from the head. This ordering must be
* maintained, so an iclog doesn't become ACTIVE beyond one that
* is SYNCING. This is also required to maintain the notion that we use
- * a counting semaphore to hold off would be writers to the log when every
+ * a ordered wait queue to hold off would be writers to the log when every
* iclog is trying to sync to disk.
*
* State Change: DIRTY -> ACTIVE
@@ -2097,6 +2118,7 @@ xlog_state_do_callback(
int funcdidcallbacks; /* flag: function did callbacks */
int repeats; /* for issuing console warnings if
* looping too many times */
+ int wake = 0;
spin_lock(&log->l_icloglock);
first_iclog = iclog = log->l_iclog;
@@ -2236,7 +2258,7 @@ xlog_state_do_callback(
xlog_state_clean_log(log);
/* wake up threads waiting in xfs_log_force() */
- sv_broadcast(&iclog->ic_forcesema);
+ sv_broadcast(&iclog->ic_force_wait);
iclog = iclog->ic_next;
} while (first_iclog != iclog);
@@ -2278,15 +2300,13 @@ xlog_state_do_callback(
}
#endif
- flushcnt = 0;
- if (log->l_iclog->ic_state & (XLOG_STATE_ACTIVE|XLOG_STATE_IOERROR)) {
- flushcnt = log->l_flushcnt;
- log->l_flushcnt = 0;
- }
+ if (log->l_iclog->ic_state & (XLOG_STATE_ACTIVE|XLOG_STATE_IOERROR))
+ wake = 1;
spin_unlock(&log->l_icloglock);
- while (flushcnt--)
- vsema(&log->l_flushsema);
-} /* xlog_state_do_callback */
+
+ if (wake)
+ sv_broadcast(&log->l_flush_wait);
+}
/*
@@ -2300,8 +2320,7 @@ xlog_state_do_callback(
* the second completion goes through.
*
* Callbacks could take time, so they are done outside the scope of the
- * global state machine log lock. Assume that the calls to cvsema won't
- * take a long time. At least we know it won't sleep.
+ * global state machine log lock.
*/
STATIC void
xlog_state_done_syncing(
@@ -2337,7 +2356,7 @@ xlog_state_done_syncing(
* iclog buffer, we wake them all, one will get to do the
* I/O, the others get to wait for the result.
*/
- sv_broadcast(&iclog->ic_writesema);
+ sv_broadcast(&iclog->ic_write_wait);
spin_unlock(&log->l_icloglock);
xlog_state_do_callback(log, aborted, iclog); /* also cleans log */
} /* xlog_state_done_syncing */
@@ -2345,11 +2364,9 @@ xlog_state_done_syncing(
/*
* If the head of the in-core log ring is not (ACTIVE or DIRTY), then we must
- * sleep. The flush semaphore is set to the number of in-core buffers and
- * decremented around disk syncing. Therefore, if all buffers are syncing,
- * this semaphore will cause new writes to sleep until a sync completes.
- * Otherwise, this code just does p() followed by v(). This approximates
- * a sleep/wakeup except we can't race.
+ * sleep. We wait on the flush queue on the head iclog as that should be
+ * the first iclog to complete flushing. Hence if all iclogs are syncing,
+ * we will wait here and all new writes will sleep until a sync completes.
*
* The in-core logs are used in a circular fashion. They are not used
* out-of-order even when an iclog past the head is free.
@@ -2384,16 +2401,15 @@ restart:
}
iclog = log->l_iclog;
- if (! (iclog->ic_state == XLOG_STATE_ACTIVE)) {
- log->l_flushcnt++;
- spin_unlock(&log->l_icloglock);
+ if (iclog->ic_state != XLOG_STATE_ACTIVE) {
xlog_trace_iclog(iclog, XLOG_TRACE_SLEEP_FLUSH);
XFS_STATS_INC(xs_log_noiclogs);
- /* Ensure that log writes happen */
- psema(&log->l_flushsema, PINOD);
+
+ /* Wait for log writes to have flushed */
+ sv_wait(&log->l_flush_wait, 0, &log->l_icloglock, 0);
goto restart;
}
- ASSERT(iclog->ic_state == XLOG_STATE_ACTIVE);
+
head = &iclog->ic_header;
atomic_inc(&iclog->ic_refcnt); /* prevents sync */
@@ -2427,13 +2443,20 @@ restart:
if (iclog->ic_size - iclog->ic_offset < 2*sizeof(xlog_op_header_t)) {
xlog_state_switch_iclogs(log, iclog, iclog->ic_size);
- /* If I'm the only one writing to this iclog, sync it to disk */
- if (atomic_read(&iclog->ic_refcnt) == 1) {
+ /*
+ * If I'm the only one writing to this iclog, sync it to disk.
+ * We need to do an atomic compare and decrement here to avoid
+ * racing with concurrent atomic_dec_and_lock() calls in
+ * xlog_state_release_iclog() when there is more than one
+ * reference to the iclog.
+ */
+ if (!atomic_add_unless(&iclog->ic_refcnt, -1, 1)) {
+ /* we are the only one */
spin_unlock(&log->l_icloglock);
- if ((error = xlog_state_release_iclog(log, iclog)))
+ error = xlog_state_release_iclog(log, iclog);
+ if (error)
return error;
} else {
- atomic_dec(&iclog->ic_refcnt);
spin_unlock(&log->l_icloglock);
}
goto restart;
@@ -2500,7 +2523,7 @@ xlog_grant_log_space(xlog_t *log,
goto error_return;
XFS_STATS_INC(xs_sleep_logspace);
- sv_wait(&tic->t_sema, PINOD|PLTWAIT, &log->l_grant_lock, s);
+ sv_wait(&tic->t_wait, PINOD|PLTWAIT, &log->l_grant_lock, s);
/*
* If we got an error, and the filesystem is shutting down,
* we'll catch it down below. So just continue...
@@ -2526,7 +2549,7 @@ redo:
xlog_trace_loggrant(log, tic,
"xlog_grant_log_space: sleep 2");
XFS_STATS_INC(xs_sleep_logspace);
- sv_wait(&tic->t_sema, PINOD|PLTWAIT, &log->l_grant_lock, s);
+ sv_wait(&tic->t_wait, PINOD|PLTWAIT, &log->l_grant_lock, s);
if (XLOG_FORCED_SHUTDOWN(log)) {
spin_lock(&log->l_grant_lock);
@@ -2625,7 +2648,7 @@ xlog_regrant_write_log_space(xlog_t *log,
if (free_bytes < ntic->t_unit_res)
break;
free_bytes -= ntic->t_unit_res;
- sv_signal(&ntic->t_sema);
+ sv_signal(&ntic->t_wait);
ntic = ntic->t_next;
} while (ntic != log->l_write_headq);
@@ -2636,7 +2659,7 @@ xlog_regrant_write_log_space(xlog_t *log,
xlog_trace_loggrant(log, tic,
"xlog_regrant_write_log_space: sleep 1");
XFS_STATS_INC(xs_sleep_logspace);
- sv_wait(&tic->t_sema, PINOD|PLTWAIT,
+ sv_wait(&tic->t_wait, PINOD|PLTWAIT,
&log->l_grant_lock, s);
/* If we're shutting down, this tic is already
@@ -2665,7 +2688,7 @@ redo:
if ((tic->t_flags & XLOG_TIC_IN_Q) == 0)
xlog_ins_ticketq(&log->l_write_headq, tic);
XFS_STATS_INC(xs_sleep_logspace);
- sv_wait(&tic->t_sema, PINOD|PLTWAIT, &log->l_grant_lock, s);
+ sv_wait(&tic->t_wait, PINOD|PLTWAIT, &log->l_grant_lock, s);
/* If we're shutting down, this tic is already off the queue */
if (XLOG_FORCED_SHUTDOWN(log)) {
@@ -2908,7 +2931,7 @@ xlog_state_switch_iclogs(xlog_t *log,
* 2. the current iclog is drity, and the previous iclog is in the
* active or dirty state.
*
- * We may sleep (call psema) if:
+ * We may sleep if:
*
* 1. the current iclog is not in the active nor dirty state.
* 2. the current iclog dirty, and the previous iclog is not in the
@@ -3005,7 +3028,7 @@ maybe_sleep:
return XFS_ERROR(EIO);
}
XFS_STATS_INC(xs_log_force_sleep);
- sv_wait(&iclog->ic_forcesema, PINOD, &log->l_icloglock, s);
+ sv_wait(&iclog->ic_force_wait, PINOD, &log->l_icloglock, s);
/*
* No need to grab the log lock here since we're
* only deciding whether or not to return EIO
@@ -3088,7 +3111,7 @@ try_again:
XLOG_STATE_SYNCING))) {
ASSERT(!(iclog->ic_state & XLOG_STATE_IOERROR));
XFS_STATS_INC(xs_log_force_sleep);
- sv_wait(&iclog->ic_prev->ic_writesema, PSWP,
+ sv_wait(&iclog->ic_prev->ic_write_wait, PSWP,
&log->l_icloglock, s);
*log_flushed = 1;
already_slept = 1;
@@ -3108,7 +3131,7 @@ try_again:
!(iclog->ic_state & (XLOG_STATE_ACTIVE | XLOG_STATE_DIRTY))) {
/*
- * Don't wait on the forcesema if we know that we've
+ * Don't wait on completion if we know that we've
* gotten a log write error.
*/
if (iclog->ic_state & XLOG_STATE_IOERROR) {
@@ -3116,7 +3139,7 @@ try_again:
return XFS_ERROR(EIO);
}
XFS_STATS_INC(xs_log_force_sleep);
- sv_wait(&iclog->ic_forcesema, PSWP, &log->l_icloglock, s);
+ 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
@@ -3172,7 +3195,7 @@ STATIC void
xlog_ticket_put(xlog_t *log,
xlog_ticket_t *ticket)
{
- sv_destroy(&ticket->t_sema);
+ sv_destroy(&ticket->t_wait);
kmem_zone_free(xfs_log_ticket_zone, ticket);
} /* xlog_ticket_put */
@@ -3262,7 +3285,7 @@ xlog_ticket_get(xlog_t *log,
tic->t_trans_type = 0;
if (xflags & XFS_LOG_PERM_RESERV)
tic->t_flags |= XLOG_TIC_PERM_RESERV;
- sv_init(&(tic->t_sema), SV_DEFAULT, "logtick");
+ sv_init(&(tic->t_wait), SV_DEFAULT, "logtick");
xlog_tic_reset_res(tic);
@@ -3549,14 +3572,14 @@ xfs_log_force_umount(
*/
if ((tic = log->l_reserve_headq)) {
do {
- sv_signal(&tic->t_sema);
+ sv_signal(&tic->t_wait);
tic = tic->t_next;
} while (tic != log->l_reserve_headq);
}
if ((tic = log->l_write_headq)) {
do {
- sv_signal(&tic->t_sema);
+ sv_signal(&tic->t_wait);
tic = tic->t_next;
} while (tic != log->l_write_headq);
}
diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h
index d1d678ecb63..d47b91f1082 100644
--- a/fs/xfs/xfs_log.h
+++ b/fs/xfs/xfs_log.h
@@ -149,7 +149,7 @@ int xfs_log_mount(struct xfs_mount *mp,
struct xfs_buftarg *log_target,
xfs_daddr_t start_block,
int num_bblocks);
-int xfs_log_mount_finish(struct xfs_mount *mp, int);
+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,
diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h
index 8952a392b5f..e7d8f84443f 100644
--- a/fs/xfs/xfs_log_priv.h
+++ b/fs/xfs/xfs_log_priv.h
@@ -241,7 +241,7 @@ typedef struct xlog_res {
} xlog_res_t;
typedef struct xlog_ticket {
- sv_t t_sema; /* sleep on this semaphore : 20 */
+ sv_t t_wait; /* ticket wait queue : 20 */
struct xlog_ticket *t_next; /* :4|8 */
struct xlog_ticket *t_prev; /* :4|8 */
xlog_tid_t t_tid; /* transaction identifier : 4 */
@@ -314,7 +314,7 @@ typedef struct xlog_rec_ext_header {
* xlog_rec_header_t into the reserved space.
* - ic_data follows, so a write to disk can start at the beginning of
* the iclog.
- * - ic_forcesema is used to implement synchronous forcing of the iclog to disk.
+ * - ic_forcewait is used to implement synchronous forcing of the iclog to disk.
* - ic_next is the pointer to the next iclog in the ring.
* - ic_bp is a pointer to the buffer used to write this incore log to disk.
* - ic_log is a pointer back to the global log structure.
@@ -339,8 +339,8 @@ typedef struct xlog_rec_ext_header {
* and move everything else out to subsequent cachelines.
*/
typedef struct xlog_iclog_fields {
- sv_t ic_forcesema;
- sv_t ic_writesema;
+ sv_t ic_force_wait;
+ sv_t ic_write_wait;
struct xlog_in_core *ic_next;
struct xlog_in_core *ic_prev;
struct xfs_buf *ic_bp;
@@ -377,8 +377,8 @@ typedef struct xlog_in_core {
/*
* Defines to save our code from this glop.
*/
-#define ic_forcesema hic_fields.ic_forcesema
-#define ic_writesema hic_fields.ic_writesema
+#define ic_force_wait hic_fields.ic_force_wait
+#define ic_write_wait hic_fields.ic_write_wait
#define ic_next hic_fields.ic_next
#define ic_prev hic_fields.ic_prev
#define ic_bp hic_fields.ic_bp
@@ -423,10 +423,8 @@ typedef struct log {
int l_logBBsize; /* size of log in BB chunks */
/* The following block of fields are changed while holding icloglock */
- sema_t l_flushsema ____cacheline_aligned_in_smp;
- /* iclog flushing semaphore */
- int l_flushcnt; /* # of procs waiting on this
- * sema */
+ sv_t l_flush_wait ____cacheline_aligned_in_smp;
+ /* waiting for iclog flush */
int l_covered_state;/* state of "covering disk
* log entries" */
xlog_in_core_t *l_iclog; /* head log queue */
@@ -450,7 +448,6 @@ typedef struct log {
int l_grant_write_bytes;
#ifdef XFS_LOG_TRACE
- struct ktrace *l_trace;
struct ktrace *l_grant_trace;
#endif
@@ -470,7 +467,7 @@ 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, int mfsi_flags);
+extern int xlog_recover_finish(xlog_t *log);
extern void xlog_pack_data(xlog_t *log, xlog_in_core_t *iclog, int);
extern void xlog_recover_process_iunlinks(xlog_t *log);
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index e65ab4af095..82d46ce69d5 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -1715,8 +1715,7 @@ xlog_check_buffer_cancelled(
} else {
prevp->bc_next = bcp->bc_next;
}
- kmem_free(bcp,
- sizeof(xfs_buf_cancel_t));
+ kmem_free(bcp);
}
}
return 1;
@@ -2519,7 +2518,7 @@ write_inode_buffer:
error:
if (need_free)
- kmem_free(in_f, sizeof(*in_f));
+ kmem_free(in_f);
return XFS_ERROR(error);
}
@@ -2830,16 +2829,14 @@ xlog_recover_free_trans(
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,
- free_item->ri_buf[i].i_len);
+ kmem_free(free_item->ri_buf[i].i_addr);
}
/* Free the item itself */
- kmem_free(free_item->ri_buf,
- (free_item->ri_total * sizeof(xfs_log_iovec_t)));
- kmem_free(free_item, sizeof(xlog_recover_item_t));
+ kmem_free(free_item->ri_buf);
+ kmem_free(free_item);
} while (first_item != item);
/* Free the transaction recover structure */
- kmem_free(trans, sizeof(xlog_recover_t));
+ kmem_free(trans);
}
STATIC int
@@ -3786,8 +3783,7 @@ xlog_do_log_recovery(
error = xlog_do_recovery_pass(log, head_blk, tail_blk,
XLOG_RECOVER_PASS1);
if (error != 0) {
- kmem_free(log->l_buf_cancel_table,
- XLOG_BC_TABLE_SIZE * sizeof(xfs_buf_cancel_t*));
+ kmem_free(log->l_buf_cancel_table);
log->l_buf_cancel_table = NULL;
return error;
}
@@ -3806,8 +3802,7 @@ xlog_do_log_recovery(
}
#endif /* DEBUG */
- kmem_free(log->l_buf_cancel_table,
- XLOG_BC_TABLE_SIZE * sizeof(xfs_buf_cancel_t*));
+ kmem_free(log->l_buf_cancel_table);
log->l_buf_cancel_table = NULL;
return error;
@@ -3945,8 +3940,7 @@ xlog_recover(
*/
int
xlog_recover_finish(
- xlog_t *log,
- int mfsi_flags)
+ xlog_t *log)
{
/*
* Now we're ready to do the transactions needed for the
@@ -3974,9 +3968,7 @@ xlog_recover_finish(
xfs_log_force(log->l_mp, (xfs_lsn_t)0,
(XFS_LOG_FORCE | XFS_LOG_SYNC));
- if ( (mfsi_flags & XFS_MFSI_NOUNLINK) == 0 ) {
- xlog_recover_process_iunlinks(log);
- }
+ xlog_recover_process_iunlinks(log);
xlog_recover_check_summary(log);
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index da3988453b7..a4503f5e949 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -47,12 +47,10 @@
STATIC int xfs_mount_log_sb(xfs_mount_t *, __int64_t);
STATIC int xfs_uuid_mount(xfs_mount_t *);
-STATIC void xfs_uuid_unmount(xfs_mount_t *mp);
STATIC void xfs_unmountfs_wait(xfs_mount_t *);
#ifdef HAVE_PERCPU_SB
-STATIC void xfs_icsb_destroy_counters(xfs_mount_t *);
STATIC void xfs_icsb_balance_counter(xfs_mount_t *, xfs_sb_field_t,
int);
STATIC void xfs_icsb_balance_counter_locked(xfs_mount_t *, xfs_sb_field_t,
@@ -63,7 +61,6 @@ STATIC void xfs_icsb_disable_counter(xfs_mount_t *, xfs_sb_field_t);
#else
-#define xfs_icsb_destroy_counters(mp) do { } while (0)
#define xfs_icsb_balance_counter(mp, a, b) do { } while (0)
#define xfs_icsb_balance_counter_locked(mp, a, b) do { } while (0)
#define xfs_icsb_modify_counters(mp, a, b, c) do { } while (0)
@@ -126,34 +123,12 @@ static const struct {
};
/*
- * Return a pointer to an initialized xfs_mount structure.
- */
-xfs_mount_t *
-xfs_mount_init(void)
-{
- xfs_mount_t *mp;
-
- mp = kmem_zalloc(sizeof(xfs_mount_t), KM_SLEEP);
-
- if (xfs_icsb_init_counters(mp)) {
- mp->m_flags |= XFS_MOUNT_NO_PERCPU_SB;
- }
-
- spin_lock_init(&mp->m_sb_lock);
- mutex_init(&mp->m_ilock);
- mutex_init(&mp->m_growlock);
- atomic_set(&mp->m_active_trans, 0);
-
- return mp;
-}
-
-/*
* Free up the resources associated with a mount structure. Assume that
* the structure was initially zeroed, so we can tell which fields got
* initialized.
*/
-void
-xfs_mount_free(
+STATIC void
+xfs_free_perag(
xfs_mount_t *mp)
{
if (mp->m_perag) {
@@ -161,28 +136,9 @@ xfs_mount_free(
for (agno = 0; agno < mp->m_maxagi; agno++)
if (mp->m_perag[agno].pagb_list)
- kmem_free(mp->m_perag[agno].pagb_list,
- sizeof(xfs_perag_busy_t) *
- XFS_PAGB_NUM_SLOTS);
- kmem_free(mp->m_perag,
- sizeof(xfs_perag_t) * mp->m_sb.sb_agcount);
+ kmem_free(mp->m_perag[agno].pagb_list);
+ kmem_free(mp->m_perag);
}
-
- spinlock_destroy(&mp->m_ail_lock);
- spinlock_destroy(&mp->m_sb_lock);
- mutex_destroy(&mp->m_ilock);
- mutex_destroy(&mp->m_growlock);
- if (mp->m_quotainfo)
- XFS_QM_DONE(mp);
-
- if (mp->m_fsname != NULL)
- kmem_free(mp->m_fsname, mp->m_fsname_len);
- if (mp->m_rtname != NULL)
- kmem_free(mp->m_rtname, strlen(mp->m_rtname) + 1);
- if (mp->m_logname != NULL)
- kmem_free(mp->m_logname, strlen(mp->m_logname) + 1);
-
- xfs_icsb_destroy_counters(mp);
}
/*
@@ -288,6 +244,19 @@ xfs_mount_validate_sb(
return XFS_ERROR(EFSCORRUPTED);
}
+ /*
+ * Until this is fixed only page-sized or smaller data blocks work.
+ */
+ if (unlikely(sbp->sb_blocksize > PAGE_SIZE)) {
+ xfs_fs_mount_cmn_err(flags,
+ "file system with blocksize %d bytes",
+ sbp->sb_blocksize);
+ xfs_fs_mount_cmn_err(flags,
+ "only pagesize (%ld) or less will currently work.",
+ PAGE_SIZE);
+ return XFS_ERROR(ENOSYS);
+ }
+
if (xfs_sb_validate_fsb_count(sbp, sbp->sb_dblocks) ||
xfs_sb_validate_fsb_count(sbp, sbp->sb_rblocks)) {
xfs_fs_mount_cmn_err(flags,
@@ -309,19 +278,6 @@ xfs_mount_validate_sb(
return XFS_ERROR(ENOSYS);
}
- /*
- * Until this is fixed only page-sized or smaller data blocks work.
- */
- if (unlikely(sbp->sb_blocksize > PAGE_SIZE)) {
- xfs_fs_mount_cmn_err(flags,
- "file system with blocksize %d bytes",
- sbp->sb_blocksize);
- xfs_fs_mount_cmn_err(flags,
- "only pagesize (%ld) or less will currently work.",
- PAGE_SIZE);
- return XFS_ERROR(ENOSYS);
- }
-
return 0;
}
@@ -734,11 +690,11 @@ xfs_initialize_perag_data(xfs_mount_t *mp, xfs_agnumber_t agcount)
* Update alignment values based on mount options and sb values
*/
STATIC int
-xfs_update_alignment(xfs_mount_t *mp, int mfsi_flags, __uint64_t *update_flags)
+xfs_update_alignment(xfs_mount_t *mp, __uint64_t *update_flags)
{
xfs_sb_t *sbp = &(mp->m_sb);
- if (mp->m_dalign && !(mfsi_flags & XFS_MFSI_SECOND)) {
+ if (mp->m_dalign) {
/*
* If stripe unit and stripe width are not multiples
* of the fs blocksize turn off alignment.
@@ -894,7 +850,7 @@ xfs_set_inoalignment(xfs_mount_t *mp)
* Check that the data (and log if separate) are an ok size.
*/
STATIC int
-xfs_check_sizes(xfs_mount_t *mp, int mfsi_flags)
+xfs_check_sizes(xfs_mount_t *mp)
{
xfs_buf_t *bp;
xfs_daddr_t d;
@@ -917,8 +873,7 @@ xfs_check_sizes(xfs_mount_t *mp, int mfsi_flags)
return error;
}
- if (((mfsi_flags & XFS_MFSI_CLIENT) == 0) &&
- mp->m_logdev_targp != mp->m_ddev_targp) {
+ if (mp->m_logdev_targp != mp->m_ddev_targp) {
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");
@@ -953,15 +908,13 @@ xfs_check_sizes(xfs_mount_t *mp, int mfsi_flags)
*/
int
xfs_mountfs(
- xfs_mount_t *mp,
- int mfsi_flags)
+ xfs_mount_t *mp)
{
xfs_sb_t *sbp = &(mp->m_sb);
xfs_inode_t *rip;
__uint64_t resblks;
__int64_t update_flags = 0LL;
uint quotamount, quotaflags;
- int agno;
int uuid_mounted = 0;
int error = 0;
@@ -994,9 +947,19 @@ xfs_mountfs(
* Re-check for ATTR2 in case it was found in bad_features2
* slot.
*/
- if (xfs_sb_version_hasattr2(&mp->m_sb))
+ if (xfs_sb_version_hasattr2(&mp->m_sb) &&
+ !(mp->m_flags & XFS_MOUNT_NOATTR2))
mp->m_flags |= XFS_MOUNT_ATTR2;
+ }
+
+ if (xfs_sb_version_hasattr2(&mp->m_sb) &&
+ (mp->m_flags & XFS_MOUNT_NOATTR2)) {
+ xfs_sb_version_removeattr2(&mp->m_sb);
+ update_flags |= XFS_SB_FEATURES2;
+ /* update sb_versionnum for the clearing of the morebits */
+ if (!sbp->sb_features2)
+ update_flags |= XFS_SB_VERSIONNUM;
}
/*
@@ -1005,7 +968,7 @@ xfs_mountfs(
* allocator alignment is within an ag, therefore ag has
* to be aligned at stripe boundary.
*/
- error = xfs_update_alignment(mp, mfsi_flags, &update_flags);
+ error = xfs_update_alignment(mp, &update_flags);
if (error)
goto error1;
@@ -1024,8 +987,7 @@ xfs_mountfs(
* since a single partition filesystem is identical to a single
* partition volume/filesystem.
*/
- if ((mfsi_flags & XFS_MFSI_SECOND) == 0 &&
- (mp->m_flags & XFS_MOUNT_NOUUID) == 0) {
+ if ((mp->m_flags & XFS_MOUNT_NOUUID) == 0) {
if (xfs_uuid_mount(mp)) {
error = XFS_ERROR(EINVAL);
goto error1;
@@ -1053,7 +1015,7 @@ xfs_mountfs(
/*
* Check that the data (and log if separate) are an ok size.
*/
- error = xfs_check_sizes(mp, mfsi_flags);
+ error = xfs_check_sizes(mp);
if (error)
goto error1;
@@ -1067,13 +1029,6 @@ xfs_mountfs(
}
/*
- * For client case we are done now
- */
- if (mfsi_flags & XFS_MFSI_CLIENT) {
- return 0;
- }
-
- /*
* Copies the low order bits of the timestamp and the randomly
* set "sequence" number out of a UUID.
*/
@@ -1097,8 +1052,10 @@ 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_SLEEP);
+ mp->m_perag = kmem_zalloc(sbp->sb_agcount * sizeof(xfs_perag_t),
+ KM_MAYFAIL);
+ if (!mp->m_perag)
+ goto error1;
mp->m_maxagi = xfs_initialize_perag(mp, sbp->sb_agcount);
@@ -1210,7 +1167,7 @@ xfs_mountfs(
* delayed until after the root and real-time bitmap inodes
* were consistently read in.
*/
- error = xfs_log_mount_finish(mp, mfsi_flags);
+ error = xfs_log_mount_finish(mp);
if (error) {
cmn_err(CE_WARN, "XFS: log mount finish failed");
goto error4;
@@ -1219,7 +1176,7 @@ xfs_mountfs(
/*
* Complete the quota initialisation, post-log-replay component.
*/
- error = XFS_QM_MOUNT(mp, quotamount, quotaflags, mfsi_flags);
+ error = XFS_QM_MOUNT(mp, quotamount, quotaflags);
if (error)
goto error4;
@@ -1253,31 +1210,25 @@ xfs_mountfs(
error3:
xfs_log_unmount_dealloc(mp);
error2:
- for (agno = 0; agno < sbp->sb_agcount; agno++)
- if (mp->m_perag[agno].pagb_list)
- kmem_free(mp->m_perag[agno].pagb_list,
- sizeof(xfs_perag_busy_t) * XFS_PAGB_NUM_SLOTS);
- kmem_free(mp->m_perag, sbp->sb_agcount * sizeof(xfs_perag_t));
- mp->m_perag = NULL;
- /* FALLTHROUGH */
+ xfs_free_perag(mp);
error1:
if (uuid_mounted)
- xfs_uuid_unmount(mp);
- xfs_freesb(mp);
+ uuid_table_remove(&mp->m_sb.sb_uuid);
return error;
}
/*
- * xfs_unmountfs
- *
* This flushes out the inodes,dquots and the superblock, unmounts the
* log and makes sure that incore structures are freed.
*/
-int
-xfs_unmountfs(xfs_mount_t *mp, struct cred *cr)
+void
+xfs_unmountfs(
+ struct xfs_mount *mp)
{
- __uint64_t resblks;
- int error = 0;
+ __uint64_t resblks;
+ int error;
+
+ IRELE(mp->m_rootip);
/*
* We can potentially deadlock here if we have an inode cluster
@@ -1334,32 +1285,20 @@ xfs_unmountfs(xfs_mount_t *mp, struct cred *cr)
xfs_unmountfs_wait(mp); /* wait for async bufs */
xfs_log_unmount(mp); /* Done! No more fs ops. */
- xfs_freesb(mp);
-
/*
* All inodes from this mount point should be freed.
*/
ASSERT(mp->m_inodes == NULL);
- xfs_unmountfs_close(mp, cr);
if ((mp->m_flags & XFS_MOUNT_NOUUID) == 0)
- xfs_uuid_unmount(mp);
+ uuid_table_remove(&mp->m_sb.sb_uuid);
-#if defined(DEBUG) || defined(INDUCE_IO_ERROR)
+#if defined(DEBUG)
xfs_errortag_clearall(mp, 0);
#endif
- xfs_mount_free(mp);
- return 0;
-}
-
-void
-xfs_unmountfs_close(xfs_mount_t *mp, struct cred *cr)
-{
- if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp)
- xfs_free_buftarg(mp->m_logdev_targp, 1);
- if (mp->m_rtdev_targp)
- xfs_free_buftarg(mp->m_rtdev_targp, 1);
- xfs_free_buftarg(mp->m_ddev_targp, 0);
+ xfs_free_perag(mp);
+ if (mp->m_quotainfo)
+ XFS_QM_DONE(mp);
}
STATIC void
@@ -1905,16 +1844,6 @@ xfs_uuid_mount(
}
/*
- * Remove filesystem from the UUID table.
- */
-STATIC void
-xfs_uuid_unmount(
- xfs_mount_t *mp)
-{
- uuid_table_remove(&mp->m_sb.sb_uuid);
-}
-
-/*
* Used to log changes to the superblock unit and width fields which could
* be altered by the mount options, as well as any potential sb_features2
* fixup. Only the first superblock is updated.
@@ -1928,7 +1857,8 @@ xfs_mount_log_sb(
int error;
ASSERT(fields & (XFS_SB_UNIT | XFS_SB_WIDTH | XFS_SB_UUID |
- XFS_SB_FEATURES2 | XFS_SB_BAD_FEATURES2));
+ XFS_SB_FEATURES2 | XFS_SB_BAD_FEATURES2 |
+ XFS_SB_VERSIONNUM));
tp = xfs_trans_alloc(mp, XFS_TRANS_SB_UNIT);
error = xfs_trans_reserve(tp, 0, mp->m_sb.sb_sectsize + 128, 0, 0,
@@ -2109,7 +2039,7 @@ xfs_icsb_reinit_counters(
xfs_icsb_unlock(mp);
}
-STATIC void
+void
xfs_icsb_destroy_counters(
xfs_mount_t *mp)
{
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index 63e0693a358..f3c1024b124 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -61,6 +61,7 @@ struct xfs_bmap_free;
struct xfs_extdelta;
struct xfs_swapext;
struct xfs_mru_cache;
+struct xfs_nameops;
/*
* Prototypes and functions for the Data Migration subsystem.
@@ -113,7 +114,7 @@ struct xfs_dqtrxops;
struct xfs_quotainfo;
typedef int (*xfs_qminit_t)(struct xfs_mount *, uint *, uint *);
-typedef int (*xfs_qmmount_t)(struct xfs_mount *, uint, uint, int);
+typedef int (*xfs_qmmount_t)(struct xfs_mount *, uint, uint);
typedef int (*xfs_qmunmount_t)(struct xfs_mount *);
typedef void (*xfs_qmdone_t)(struct xfs_mount *);
typedef void (*xfs_dqrele_t)(struct xfs_dquot *);
@@ -157,8 +158,8 @@ typedef struct xfs_qmops {
#define XFS_QM_INIT(mp, mnt, fl) \
(*(mp)->m_qm_ops->xfs_qminit)(mp, mnt, fl)
-#define XFS_QM_MOUNT(mp, mnt, fl, mfsi_flags) \
- (*(mp)->m_qm_ops->xfs_qmmount)(mp, mnt, fl, mfsi_flags)
+#define XFS_QM_MOUNT(mp, mnt, fl) \
+ (*(mp)->m_qm_ops->xfs_qmmount)(mp, mnt, fl)
#define XFS_QM_UNMOUNT(mp) \
(*(mp)->m_qm_ops->xfs_qmunmount)(mp)
#define XFS_QM_DONE(mp) \
@@ -210,12 +211,14 @@ typedef struct xfs_icsb_cnts {
extern int xfs_icsb_init_counters(struct xfs_mount *);
extern void xfs_icsb_reinit_counters(struct xfs_mount *);
+extern void xfs_icsb_destroy_counters(struct xfs_mount *);
extern void xfs_icsb_sync_counters(struct xfs_mount *, int);
extern void xfs_icsb_sync_counters_locked(struct xfs_mount *, int);
#else
-#define xfs_icsb_init_counters(mp) (0)
-#define xfs_icsb_reinit_counters(mp) do { } while (0)
+#define xfs_icsb_init_counters(mp) (0)
+#define xfs_icsb_destroy_counters(mp) do { } while (0)
+#define xfs_icsb_reinit_counters(mp) do { } while (0)
#define xfs_icsb_sync_counters(mp, flags) do { } while (0)
#define xfs_icsb_sync_counters_locked(mp, flags) do { } while (0)
#endif
@@ -313,6 +316,7 @@ typedef struct xfs_mount {
__uint8_t m_inode_quiesce;/* call quiesce on new inodes.
field governed by m_ilock */
__uint8_t m_sectbb_log; /* sectlog - BBSHIFT */
+ const struct xfs_nameops *m_dirnameops; /* vector of dir name ops */
int m_dirblksize; /* directory block sz--bytes */
int m_dirblkfsbs; /* directory block sz--fsbs */
xfs_dablk_t m_dirdatablk; /* blockno of dir data v2 */
@@ -378,6 +382,7 @@ typedef struct xfs_mount {
counters */
#define XFS_MOUNT_FILESTREAMS (1ULL << 24) /* enable the filestreams
allocator */
+#define XFS_MOUNT_NOATTR2 (1ULL << 25) /* disable use of attr2 format */
/*
@@ -437,13 +442,6 @@ void xfs_do_force_shutdown(struct xfs_mount *mp, int flags, char *fname,
/*
* Flags for xfs_mountfs
*/
-#define XFS_MFSI_SECOND 0x01 /* Secondary mount -- skip stuff */
-#define XFS_MFSI_CLIENT 0x02 /* Is a client -- skip lots of stuff */
-/* XFS_MFSI_RRINODES */
-#define XFS_MFSI_NOUNLINK 0x08 /* Skip unlinked inode processing in */
- /* log recovery */
-#define XFS_MFSI_NO_QUOTACHECK 0x10 /* Skip quotacheck processing */
-/* XFS_MFSI_CONVERT_SUNIT */
#define XFS_MFSI_QUIET 0x40 /* Be silent if mount errors found */
#define XFS_DADDR_TO_AGNO(mp,d) xfs_daddr_to_agno(mp,d)
@@ -510,15 +508,12 @@ typedef struct xfs_mod_sb {
#define XFS_MOUNT_ILOCK(mp) mutex_lock(&((mp)->m_ilock))
#define XFS_MOUNT_IUNLOCK(mp) mutex_unlock(&((mp)->m_ilock))
-extern xfs_mount_t *xfs_mount_init(void);
extern void xfs_mod_sb(xfs_trans_t *, __int64_t);
extern int xfs_log_sbcount(xfs_mount_t *, uint);
-extern void xfs_mount_free(xfs_mount_t *mp);
-extern int xfs_mountfs(xfs_mount_t *mp, int);
+extern int xfs_mountfs(xfs_mount_t *mp);
extern void xfs_mountfs_check_barriers(xfs_mount_t *mp);
-extern int xfs_unmountfs(xfs_mount_t *, struct cred *);
-extern void xfs_unmountfs_close(xfs_mount_t *, struct cred *);
+extern void xfs_unmountfs(xfs_mount_t *);
extern int xfs_unmountfs_writesb(xfs_mount_t *);
extern int xfs_unmount_flush(xfs_mount_t *, int);
extern int xfs_mod_incore_sb(xfs_mount_t *, xfs_sb_field_t, int64_t, int);
@@ -544,9 +539,6 @@ extern void xfs_qmops_put(struct xfs_mount *);
extern struct xfs_dmops xfs_dmcore_xfs;
-extern int xfs_init(void);
-extern void xfs_cleanup(void);
-
#endif /* __KERNEL__ */
#endif /* __XFS_MOUNT_H__ */
diff --git a/fs/xfs/xfs_mru_cache.c b/fs/xfs/xfs_mru_cache.c
index a0b2c0a2589..afee7eb2432 100644
--- a/fs/xfs/xfs_mru_cache.c
+++ b/fs/xfs/xfs_mru_cache.c
@@ -307,15 +307,18 @@ xfs_mru_cache_init(void)
xfs_mru_elem_zone = kmem_zone_init(sizeof(xfs_mru_cache_elem_t),
"xfs_mru_cache_elem");
if (!xfs_mru_elem_zone)
- return ENOMEM;
+ goto out;
xfs_mru_reap_wq = create_singlethread_workqueue("xfs_mru_cache");
- if (!xfs_mru_reap_wq) {
- kmem_zone_destroy(xfs_mru_elem_zone);
- return ENOMEM;
- }
+ if (!xfs_mru_reap_wq)
+ goto out_destroy_mru_elem_zone;
return 0;
+
+ out_destroy_mru_elem_zone:
+ kmem_zone_destroy(xfs_mru_elem_zone);
+ out:
+ return -ENOMEM;
}
void
@@ -382,9 +385,9 @@ xfs_mru_cache_create(
exit:
if (err && mru && mru->lists)
- kmem_free(mru->lists, mru->grp_count * sizeof(*mru->lists));
+ kmem_free(mru->lists);
if (err && mru)
- kmem_free(mru, sizeof(*mru));
+ kmem_free(mru);
return err;
}
@@ -424,8 +427,8 @@ xfs_mru_cache_destroy(
xfs_mru_cache_flush(mru);
- kmem_free(mru->lists, mru->grp_count * sizeof(*mru->lists));
- kmem_free(mru, sizeof(*mru));
+ kmem_free(mru->lists);
+ kmem_free(mru);
}
/*
diff --git a/fs/xfs/xfs_rename.c b/fs/xfs/xfs_rename.c
index d8063e1ad29..d700dacdb10 100644
--- a/fs/xfs/xfs_rename.c
+++ b/fs/xfs/xfs_rename.c
@@ -336,22 +336,18 @@ xfs_rename(
ASSERT(error != EEXIST);
if (error)
goto abort_return;
- xfs_ichgtime(src_ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
-
- } else {
- /*
- * We always want to hit the ctime on the source inode.
- * We do it in the if clause above for the 'new_parent &&
- * src_is_directory' case, and here we get all the other
- * cases. This isn't strictly required by the standards
- * since the source inode isn't really being changed,
- * but old unix file systems did it and some incremental
- * backup programs won't work without it.
- */
- xfs_ichgtime(src_ip, XFS_ICHGTIME_CHG);
}
/*
+ * We always want to hit the ctime on the source inode.
+ *
+ * This isn't strictly required by the standards since the source
+ * inode isn't really being changed, but old unix file systems did
+ * it and some incremental backup programs won't work without it.
+ */
+ xfs_ichgtime(src_ip, XFS_ICHGTIME_CHG);
+
+ /*
* Adjust the link count on src_dp. This is necessary when
* renaming a directory, either within one parent when
* the target existed, or across two parent directories.
diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index a0dc6e5bc5b..e2f68de1615 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -74,18 +74,6 @@ STATIC int xfs_rtmodify_summary(xfs_mount_t *, xfs_trans_t *, int,
*/
/*
- * xfs_lowbit32: get low bit set out of 32-bit argument, -1 if none set.
- */
-STATIC int
-xfs_lowbit32(
- __uint32_t v)
-{
- if (v)
- return ffs(v) - 1;
- return -1;
-}
-
-/*
* Allocate space to the bitmap or summary file, and zero it, for growfs.
*/
STATIC int /* error */
@@ -450,6 +438,7 @@ xfs_rtallocate_extent_near(
}
bbno = XFS_BITTOBLOCK(mp, bno);
i = 0;
+ ASSERT(minlen != 0);
log2len = xfs_highbit32(minlen);
/*
* Loop over all bitmap blocks (bbno + i is current block).
@@ -618,6 +607,8 @@ xfs_rtallocate_extent_size(
xfs_suminfo_t sum; /* summary information for extents */
ASSERT(minlen % prod == 0 && maxlen % prod == 0);
+ ASSERT(maxlen != 0);
+
/*
* Loop over all the levels starting with maxlen.
* At each level, look at all the bitmap blocks, to see if there
@@ -675,6 +666,9 @@ xfs_rtallocate_extent_size(
*rtblock = NULLRTBLOCK;
return 0;
}
+ ASSERT(minlen != 0);
+ ASSERT(maxlen != 0);
+
/*
* Loop over sizes, from maxlen down to minlen.
* This time, when we do the allocations, allow smaller ones
@@ -1961,6 +1955,7 @@ xfs_growfs_rt(
nsbp->sb_blocksize * nsbp->sb_rextsize);
nsbp->sb_rextents = nsbp->sb_rblocks;
do_div(nsbp->sb_rextents, nsbp->sb_rextsize);
+ ASSERT(nsbp->sb_rextents != 0);
nsbp->sb_rextslog = xfs_highbit32(nsbp->sb_rextents);
nrsumlevels = nmp->m_rsumlevels = nsbp->sb_rextslog + 1;
nrsumsize =
@@ -2062,7 +2057,7 @@ xfs_growfs_rt(
/*
* Free the fake mp structure.
*/
- kmem_free(nmp, sizeof(*nmp));
+ kmem_free(nmp);
return error;
}
diff --git a/fs/xfs/xfs_rw.c b/fs/xfs/xfs_rw.c
index b0f31c09a76..3a82576dde9 100644
--- a/fs/xfs/xfs_rw.c
+++ b/fs/xfs/xfs_rw.c
@@ -314,7 +314,7 @@ xfs_bioerror_relse(
* ASYNC buffers.
*/
XFS_BUF_ERROR(bp, EIO);
- XFS_BUF_V_IODONESEMA(bp);
+ XFS_BUF_FINISH_IOWAIT(bp);
} else {
xfs_buf_relse(bp);
}
diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/xfs_sb.h
index d904efe7f87..3f8cf1587f4 100644
--- a/fs/xfs/xfs_sb.h
+++ b/fs/xfs/xfs_sb.h
@@ -46,10 +46,12 @@ struct xfs_mount;
#define XFS_SB_VERSION_SECTORBIT 0x0800
#define XFS_SB_VERSION_EXTFLGBIT 0x1000
#define XFS_SB_VERSION_DIRV2BIT 0x2000
+#define XFS_SB_VERSION_BORGBIT 0x4000 /* ASCII only case-insens. */
#define XFS_SB_VERSION_MOREBITSBIT 0x8000
#define XFS_SB_VERSION_OKSASHFBITS \
(XFS_SB_VERSION_EXTFLGBIT | \
- XFS_SB_VERSION_DIRV2BIT)
+ XFS_SB_VERSION_DIRV2BIT | \
+ XFS_SB_VERSION_BORGBIT)
#define XFS_SB_VERSION_OKREALFBITS \
(XFS_SB_VERSION_ATTRBIT | \
XFS_SB_VERSION_NLINKBIT | \
@@ -437,6 +439,12 @@ static inline int xfs_sb_version_hassector(xfs_sb_t *sbp)
((sbp)->sb_versionnum & XFS_SB_VERSION_SECTORBIT);
}
+static inline int xfs_sb_version_hasasciici(xfs_sb_t *sbp)
+{
+ return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \
+ (sbp->sb_versionnum & XFS_SB_VERSION_BORGBIT);
+}
+
static inline int xfs_sb_version_hasmorebits(xfs_sb_t *sbp)
{
return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \
@@ -473,6 +481,13 @@ static inline void xfs_sb_version_addattr2(xfs_sb_t *sbp)
((sbp)->sb_features2 | XFS_SB_VERSION2_ATTR2BIT)));
}
+static inline void xfs_sb_version_removeattr2(xfs_sb_t *sbp)
+{
+ sbp->sb_features2 &= ~XFS_SB_VERSION2_ATTR2BIT;
+ if (!sbp->sb_features2)
+ sbp->sb_versionnum &= ~XFS_SB_VERSION_MOREBITSBIT;
+}
+
/*
* end of superblock version macros
*/
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index 140386434aa..4e1c22a23be 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -43,6 +43,7 @@
#include "xfs_quota.h"
#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 *);
@@ -253,7 +254,7 @@ _xfs_trans_alloc(
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_lic_init(&(tp->t_items));
XFS_LBC_INIT(&(tp->t_busy));
return tp;
}
@@ -282,7 +283,7 @@ xfs_trans_dup(
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_lic_init(&(ntp->t_items));
XFS_LBC_INIT(&(ntp->t_busy));
ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES);
@@ -889,7 +890,7 @@ shut_us_down:
tp->t_commit_lsn = commit_lsn;
if (nvec > XFS_TRANS_LOGVEC_COUNT) {
- kmem_free(log_vector, nvec * sizeof(xfs_log_iovec_t));
+ kmem_free(log_vector);
}
/*
@@ -1169,7 +1170,7 @@ xfs_trans_cancel(
while (licp != NULL) {
lidp = licp->lic_descs;
for (i = 0; i < licp->lic_unused; i++, lidp++) {
- if (XFS_LIC_ISFREE(licp, i)) {
+ if (xfs_lic_isfree(licp, i)) {
continue;
}
@@ -1216,6 +1217,68 @@ xfs_trans_free(
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
+ * committed with XFS_TRANS_RELEASE_LOG_RES, but we still want as soon
+ * as possible to let chunks of it go to the log. So we commit the
+ * chunk we've been working on and get a new transaction to continue.
+ */
+int
+xfs_trans_roll(
+ struct xfs_trans **tpp,
+ struct xfs_inode *dp)
+{
+ struct xfs_trans *trans;
+ unsigned int logres, count;
+ int error;
+
+ /*
+ * Ensure that the inode is always logged.
+ */
+ trans = *tpp;
+ xfs_trans_log_inode(trans, dp, XFS_ILOG_CORE);
+
+ /*
+ * Copy the critical parameters from one trans to the next.
+ */
+ logres = trans->t_log_res;
+ count = trans->t_log_count;
+ *tpp = xfs_trans_dup(trans);
+
+ /*
+ * Commit the current transaction.
+ * If this commit failed, then it'd just unlock those items that
+ * are not marked ihold. That also means that a filesystem shutdown
+ * is in progress. The caller takes the responsibility to cancel
+ * the duplicate transaction that gets returned.
+ */
+ error = xfs_trans_commit(trans, 0);
+ if (error)
+ return (error);
+
+ trans = *tpp;
+
+ /*
+ * Reserve space in the log for th next transaction.
+ * This also pushes items in the "AIL", the list of logged items,
+ * out to disk if they are taking up space at the tail of the log
+ * that we want to use. This requires that either nothing be locked
+ * across this call, or that anything that is locked be logged in
+ * the prior and the next transactions.
+ */
+ error = xfs_trans_reserve(trans, 0, logres, 0,
+ XFS_TRANS_PERM_LOG_RES, count);
+ /*
+ * Ensure that the inode is in the new transaction and locked.
+ */
+ if (error)
+ return error;
+
+ xfs_trans_ijoin(trans, dp, XFS_ILOCK_EXCL);
+ xfs_trans_ihold(trans, dp);
+ return 0;
+}
/*
* THIS SHOULD BE REWRITTEN TO USE xfs_trans_next_item().
@@ -1253,7 +1316,7 @@ xfs_trans_committed(
* Special case the chunk embedded in the transaction.
*/
licp = &(tp->t_items);
- if (!(XFS_LIC_ARE_ALL_FREE(licp))) {
+ if (!(xfs_lic_are_all_free(licp))) {
xfs_trans_chunk_committed(licp, tp->t_lsn, abortflag);
}
@@ -1262,10 +1325,10 @@ xfs_trans_committed(
*/
licp = licp->lic_next;
while (licp != NULL) {
- ASSERT(!XFS_LIC_ARE_ALL_FREE(licp));
+ ASSERT(!xfs_lic_are_all_free(licp));
xfs_trans_chunk_committed(licp, tp->t_lsn, abortflag);
next_licp = licp->lic_next;
- kmem_free(licp, sizeof(xfs_log_item_chunk_t));
+ kmem_free(licp);
licp = next_licp;
}
@@ -1325,7 +1388,7 @@ xfs_trans_chunk_committed(
lidp = licp->lic_descs;
for (i = 0; i < licp->lic_unused; i++, lidp++) {
- if (XFS_LIC_ISFREE(licp, i)) {
+ if (xfs_lic_isfree(licp, i)) {
continue;
}
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index 0804207c739..74c80bd2b0e 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -210,62 +210,52 @@ typedef struct xfs_log_item_chunk {
* lic_unused to the right value (0 matches all free). The
* lic_descs.lid_index values are set up as each desc is allocated.
*/
-#define XFS_LIC_INIT(cp) xfs_lic_init(cp)
static inline void xfs_lic_init(xfs_log_item_chunk_t *cp)
{
cp->lic_free = XFS_LIC_FREEMASK;
}
-#define XFS_LIC_INIT_SLOT(cp,slot) xfs_lic_init_slot(cp, slot)
static inline void xfs_lic_init_slot(xfs_log_item_chunk_t *cp, int slot)
{
cp->lic_descs[slot].lid_index = (unsigned char)(slot);
}
-#define XFS_LIC_VACANCY(cp) xfs_lic_vacancy(cp)
static inline int xfs_lic_vacancy(xfs_log_item_chunk_t *cp)
{
return cp->lic_free & XFS_LIC_FREEMASK;
}
-#define XFS_LIC_ALL_FREE(cp) xfs_lic_all_free(cp)
static inline void xfs_lic_all_free(xfs_log_item_chunk_t *cp)
{
cp->lic_free = XFS_LIC_FREEMASK;
}
-#define XFS_LIC_ARE_ALL_FREE(cp) xfs_lic_are_all_free(cp)
static inline int xfs_lic_are_all_free(xfs_log_item_chunk_t *cp)
{
return ((cp->lic_free & XFS_LIC_FREEMASK) == XFS_LIC_FREEMASK);
}
-#define XFS_LIC_ISFREE(cp,slot) xfs_lic_isfree(cp,slot)
static inline int xfs_lic_isfree(xfs_log_item_chunk_t *cp, int slot)
{
return (cp->lic_free & (1 << slot));
}
-#define XFS_LIC_CLAIM(cp,slot) xfs_lic_claim(cp,slot)
static inline void xfs_lic_claim(xfs_log_item_chunk_t *cp, int slot)
{
cp->lic_free &= ~(1 << slot);
}
-#define XFS_LIC_RELSE(cp,slot) xfs_lic_relse(cp,slot)
static inline void xfs_lic_relse(xfs_log_item_chunk_t *cp, int slot)
{
cp->lic_free |= 1 << slot;
}
-#define XFS_LIC_SLOT(cp,slot) xfs_lic_slot(cp,slot)
static inline xfs_log_item_desc_t *
xfs_lic_slot(xfs_log_item_chunk_t *cp, int slot)
{
return &(cp->lic_descs[slot]);
}
-#define XFS_LIC_DESC_TO_SLOT(dp) xfs_lic_desc_to_slot(dp)
static inline int xfs_lic_desc_to_slot(xfs_log_item_desc_t *dp)
{
return (uint)dp->lid_index;
@@ -278,7 +268,6 @@ static inline int xfs_lic_desc_to_slot(xfs_log_item_desc_t *dp)
* All of this yields the address of the chunk, which is
* cast to a chunk pointer.
*/
-#define XFS_LIC_DESC_TO_CHUNK(dp) xfs_lic_desc_to_chunk(dp)
static inline xfs_log_item_chunk_t *
xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp)
{
@@ -986,6 +975,7 @@ int _xfs_trans_commit(xfs_trans_t *,
int *);
#define xfs_trans_commit(tp, flags) _xfs_trans_commit(tp, flags, NULL)
void xfs_trans_cancel(xfs_trans_t *, int);
+int xfs_trans_roll(struct xfs_trans **, struct xfs_inode *);
int xfs_trans_ail_init(struct xfs_mount *);
void xfs_trans_ail_destroy(struct xfs_mount *);
void xfs_trans_push_ail(struct xfs_mount *, xfs_lsn_t);
diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c
index cb0c5839154..4e855b5ced6 100644
--- a/fs/xfs/xfs_trans_buf.c
+++ b/fs/xfs/xfs_trans_buf.c
@@ -1021,16 +1021,16 @@ xfs_trans_buf_item_match(
bp = NULL;
len = BBTOB(len);
licp = &tp->t_items;
- if (!XFS_LIC_ARE_ALL_FREE(licp)) {
+ if (!xfs_lic_are_all_free(licp)) {
for (i = 0; i < licp->lic_unused; i++) {
/*
* Skip unoccupied slots.
*/
- if (XFS_LIC_ISFREE(licp, i)) {
+ if (xfs_lic_isfree(licp, i)) {
continue;
}
- lidp = XFS_LIC_SLOT(licp, i);
+ 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;
@@ -1074,7 +1074,7 @@ xfs_trans_buf_item_match_all(
bp = NULL;
len = BBTOB(len);
for (licp = &tp->t_items; licp != NULL; licp = licp->lic_next) {
- if (XFS_LIC_ARE_ALL_FREE(licp)) {
+ if (xfs_lic_are_all_free(licp)) {
ASSERT(licp == &tp->t_items);
ASSERT(licp->lic_next == NULL);
return NULL;
@@ -1083,11 +1083,11 @@ xfs_trans_buf_item_match_all(
/*
* Skip unoccupied slots.
*/
- if (XFS_LIC_ISFREE(licp, i)) {
+ if (xfs_lic_isfree(licp, i)) {
continue;
}
- lidp = XFS_LIC_SLOT(licp, i);
+ 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;
diff --git a/fs/xfs/xfs_trans_inode.c b/fs/xfs/xfs_trans_inode.c
index 4c70bf5e998..2a1c0f071f9 100644
--- a/fs/xfs/xfs_trans_inode.c
+++ b/fs/xfs/xfs_trans_inode.c
@@ -291,7 +291,7 @@ xfs_trans_inode_broot_debug(
iip = ip->i_itemp;
if (iip->ili_root_size != 0) {
ASSERT(iip->ili_orig_root != NULL);
- kmem_free(iip->ili_orig_root, iip->ili_root_size);
+ kmem_free(iip->ili_orig_root);
iip->ili_root_size = 0;
iip->ili_orig_root = NULL;
}
diff --git a/fs/xfs/xfs_trans_item.c b/fs/xfs/xfs_trans_item.c
index 66a09f0d894..3c666e8317f 100644
--- a/fs/xfs/xfs_trans_item.c
+++ b/fs/xfs/xfs_trans_item.c
@@ -53,11 +53,11 @@ xfs_trans_add_item(xfs_trans_t *tp, xfs_log_item_t *lip)
* Initialize the chunk, and then
* claim the first slot in the newly allocated chunk.
*/
- XFS_LIC_INIT(licp);
- XFS_LIC_CLAIM(licp, 0);
+ xfs_lic_init(licp);
+ xfs_lic_claim(licp, 0);
licp->lic_unused = 1;
- XFS_LIC_INIT_SLOT(licp, 0);
- lidp = XFS_LIC_SLOT(licp, 0);
+ xfs_lic_init_slot(licp, 0);
+ lidp = xfs_lic_slot(licp, 0);
/*
* Link in the new chunk and update the free count.
@@ -88,14 +88,14 @@ xfs_trans_add_item(xfs_trans_t *tp, xfs_log_item_t *lip)
*/
licp = &tp->t_items;
while (licp != NULL) {
- if (XFS_LIC_VACANCY(licp)) {
+ if (xfs_lic_vacancy(licp)) {
if (licp->lic_unused <= XFS_LIC_MAX_SLOT) {
i = licp->lic_unused;
- ASSERT(XFS_LIC_ISFREE(licp, i));
+ ASSERT(xfs_lic_isfree(licp, i));
break;
}
for (i = 0; i <= XFS_LIC_MAX_SLOT; i++) {
- if (XFS_LIC_ISFREE(licp, i))
+ if (xfs_lic_isfree(licp, i))
break;
}
ASSERT(i <= XFS_LIC_MAX_SLOT);
@@ -108,12 +108,12 @@ xfs_trans_add_item(xfs_trans_t *tp, xfs_log_item_t *lip)
* If we find a free descriptor, claim it,
* initialize it, and return it.
*/
- XFS_LIC_CLAIM(licp, i);
+ xfs_lic_claim(licp, i);
if (licp->lic_unused <= i) {
licp->lic_unused = i + 1;
- XFS_LIC_INIT_SLOT(licp, i);
+ xfs_lic_init_slot(licp, i);
}
- lidp = XFS_LIC_SLOT(licp, i);
+ lidp = xfs_lic_slot(licp, i);
tp->t_items_free--;
lidp->lid_item = lip;
lidp->lid_flags = 0;
@@ -136,9 +136,9 @@ xfs_trans_free_item(xfs_trans_t *tp, xfs_log_item_desc_t *lidp)
xfs_log_item_chunk_t *licp;
xfs_log_item_chunk_t **licpp;
- slot = XFS_LIC_DESC_TO_SLOT(lidp);
- licp = XFS_LIC_DESC_TO_CHUNK(lidp);
- XFS_LIC_RELSE(licp, slot);
+ slot = xfs_lic_desc_to_slot(lidp);
+ licp = xfs_lic_desc_to_chunk(lidp);
+ xfs_lic_relse(licp, slot);
lidp->lid_item->li_desc = NULL;
tp->t_items_free++;
@@ -154,14 +154,14 @@ xfs_trans_free_item(xfs_trans_t *tp, xfs_log_item_desc_t *lidp)
* Also decrement the transaction structure's count of free items
* by the number in a chunk since we are freeing an empty chunk.
*/
- if (XFS_LIC_ARE_ALL_FREE(licp) && (licp != &(tp->t_items))) {
+ if (xfs_lic_are_all_free(licp) && (licp != &(tp->t_items))) {
licpp = &(tp->t_items.lic_next);
while (*licpp != licp) {
ASSERT(*licpp != NULL);
licpp = &((*licpp)->lic_next);
}
*licpp = licp->lic_next;
- kmem_free(licp, sizeof(xfs_log_item_chunk_t));
+ kmem_free(licp);
tp->t_items_free -= XFS_LIC_NUM_SLOTS;
}
}
@@ -207,20 +207,20 @@ xfs_trans_first_item(xfs_trans_t *tp)
/*
* If it's not in the first chunk, skip to the second.
*/
- if (XFS_LIC_ARE_ALL_FREE(licp)) {
+ if (xfs_lic_are_all_free(licp)) {
licp = licp->lic_next;
}
/*
* Return the first non-free descriptor in the chunk.
*/
- ASSERT(!XFS_LIC_ARE_ALL_FREE(licp));
+ ASSERT(!xfs_lic_are_all_free(licp));
for (i = 0; i < licp->lic_unused; i++) {
- if (XFS_LIC_ISFREE(licp, i)) {
+ if (xfs_lic_isfree(licp, i)) {
continue;
}
- return XFS_LIC_SLOT(licp, i);
+ return xfs_lic_slot(licp, i);
}
cmn_err(CE_WARN, "xfs_trans_first_item() -- no first item");
return NULL;
@@ -242,18 +242,18 @@ xfs_trans_next_item(xfs_trans_t *tp, xfs_log_item_desc_t *lidp)
xfs_log_item_chunk_t *licp;
int i;
- licp = XFS_LIC_DESC_TO_CHUNK(lidp);
+ licp = xfs_lic_desc_to_chunk(lidp);
/*
* First search the rest of the chunk. The for loop keeps us
* from referencing things beyond the end of the chunk.
*/
- for (i = (int)XFS_LIC_DESC_TO_SLOT(lidp) + 1; i < licp->lic_unused; i++) {
- if (XFS_LIC_ISFREE(licp, i)) {
+ for (i = (int)xfs_lic_desc_to_slot(lidp) + 1; i < licp->lic_unused; i++) {
+ if (xfs_lic_isfree(licp, i)) {
continue;
}
- return XFS_LIC_SLOT(licp, i);
+ return xfs_lic_slot(licp, i);
}
/*
@@ -266,13 +266,13 @@ xfs_trans_next_item(xfs_trans_t *tp, xfs_log_item_desc_t *lidp)
}
licp = licp->lic_next;
- ASSERT(!XFS_LIC_ARE_ALL_FREE(licp));
+ ASSERT(!xfs_lic_are_all_free(licp));
for (i = 0; i < licp->lic_unused; i++) {
- if (XFS_LIC_ISFREE(licp, i)) {
+ if (xfs_lic_isfree(licp, i)) {
continue;
}
- return XFS_LIC_SLOT(licp, i);
+ return xfs_lic_slot(licp, i);
}
ASSERT(0);
/* NOTREACHED */
@@ -300,9 +300,9 @@ xfs_trans_free_items(
/*
* Special case the embedded chunk so we don't free it below.
*/
- if (!XFS_LIC_ARE_ALL_FREE(licp)) {
+ if (!xfs_lic_are_all_free(licp)) {
(void) xfs_trans_unlock_chunk(licp, 1, abort, NULLCOMMITLSN);
- XFS_LIC_ALL_FREE(licp);
+ xfs_lic_all_free(licp);
licp->lic_unused = 0;
}
licp = licp->lic_next;
@@ -311,10 +311,10 @@ xfs_trans_free_items(
* Unlock each item in each chunk and free the chunks.
*/
while (licp != NULL) {
- ASSERT(!XFS_LIC_ARE_ALL_FREE(licp));
+ ASSERT(!xfs_lic_are_all_free(licp));
(void) xfs_trans_unlock_chunk(licp, 1, abort, NULLCOMMITLSN);
next_licp = licp->lic_next;
- kmem_free(licp, sizeof(xfs_log_item_chunk_t));
+ kmem_free(licp);
licp = next_licp;
}
@@ -347,7 +347,7 @@ xfs_trans_unlock_items(xfs_trans_t *tp, xfs_lsn_t commit_lsn)
/*
* Special case the embedded chunk so we don't free.
*/
- if (!XFS_LIC_ARE_ALL_FREE(licp)) {
+ if (!xfs_lic_are_all_free(licp)) {
freed = xfs_trans_unlock_chunk(licp, 0, 0, commit_lsn);
}
licpp = &(tp->t_items.lic_next);
@@ -358,12 +358,12 @@ xfs_trans_unlock_items(xfs_trans_t *tp, xfs_lsn_t commit_lsn)
* and free empty chunks.
*/
while (licp != NULL) {
- ASSERT(!XFS_LIC_ARE_ALL_FREE(licp));
+ ASSERT(!xfs_lic_are_all_free(licp));
freed += xfs_trans_unlock_chunk(licp, 0, 0, commit_lsn);
next_licp = licp->lic_next;
- if (XFS_LIC_ARE_ALL_FREE(licp)) {
+ if (xfs_lic_are_all_free(licp)) {
*licpp = next_licp;
- kmem_free(licp, sizeof(xfs_log_item_chunk_t));
+ kmem_free(licp);
freed -= XFS_LIC_NUM_SLOTS;
} else {
licpp = &(licp->lic_next);
@@ -402,7 +402,7 @@ xfs_trans_unlock_chunk(
freed = 0;
lidp = licp->lic_descs;
for (i = 0; i < licp->lic_unused; i++, lidp++) {
- if (XFS_LIC_ISFREE(licp, i)) {
+ if (xfs_lic_isfree(licp, i)) {
continue;
}
lip = lidp->lid_item;
@@ -421,7 +421,7 @@ xfs_trans_unlock_chunk(
*/
if (!(freeing_chunk) &&
(!(lidp->lid_flags & XFS_LID_DIRTY) || abort)) {
- XFS_LIC_RELSE(licp, i);
+ xfs_lic_relse(licp, i);
freed++;
}
}
@@ -530,7 +530,7 @@ xfs_trans_free_busy(xfs_trans_t *tp)
lbcp = tp->t_busy.lbc_next;
while (lbcp != NULL) {
lbcq = lbcp->lbc_next;
- kmem_free(lbcp, sizeof(xfs_log_busy_chunk_t));
+ kmem_free(lbcp);
lbcp = lbcq;
}
diff --git a/fs/xfs/xfs_utils.c b/fs/xfs/xfs_utils.c
index 98e5f110ba5..35d4d414bcc 100644
--- a/fs/xfs/xfs_utils.c
+++ b/fs/xfs/xfs_utils.c
@@ -237,7 +237,7 @@ xfs_droplink(
ASSERT (ip->i_d.di_nlink > 0);
ip->i_d.di_nlink--;
- drop_nlink(ip->i_vnode);
+ drop_nlink(VFS_I(ip));
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
error = 0;
@@ -301,7 +301,7 @@ xfs_bumplink(
ASSERT(ip->i_d.di_nlink > 0);
ip->i_d.di_nlink++;
- inc_nlink(ip->i_vnode);
+ inc_nlink(VFS_I(ip));
if ((ip->i_d.di_version == XFS_DINODE_VERSION_1) &&
(ip->i_d.di_nlink > XFS_MAXLINK_1)) {
/*
diff --git a/fs/xfs/xfs_utils.h b/fs/xfs/xfs_utils.h
index f316cb85d8e..ef321225d26 100644
--- a/fs/xfs/xfs_utils.h
+++ b/fs/xfs/xfs_utils.h
@@ -18,9 +18,6 @@
#ifndef __XFS_UTILS_H__
#define __XFS_UTILS_H__
-#define IRELE(ip) VN_RELE(XFS_ITOV(ip))
-#define IHOLD(ip) VN_HOLD(XFS_ITOV(ip))
-
extern int xfs_truncate_file(xfs_mount_t *, xfs_inode_t *);
extern int xfs_dir_ialloc(xfs_trans_t **, xfs_inode_t *, mode_t, xfs_nlink_t,
xfs_dev_t, cred_t *, prid_t, int,
diff --git a/fs/xfs/xfs_vfsops.c b/fs/xfs/xfs_vfsops.c
index 30bacd8bb0e..439dd3939dd 100644
--- a/fs/xfs/xfs_vfsops.c
+++ b/fs/xfs/xfs_vfsops.c
@@ -58,586 +58,6 @@
#include "xfs_utils.h"
-int __init
-xfs_init(void)
-{
-#ifdef XFS_DABUF_DEBUG
- extern spinlock_t xfs_dabuf_global_lock;
- spin_lock_init(&xfs_dabuf_global_lock);
-#endif
-
- /*
- * Initialize all of the zone allocators we use.
- */
- xfs_log_ticket_zone = kmem_zone_init(sizeof(xlog_ticket_t),
- "xfs_log_ticket");
- xfs_bmap_free_item_zone = kmem_zone_init(sizeof(xfs_bmap_free_item_t),
- "xfs_bmap_free_item");
- xfs_btree_cur_zone = kmem_zone_init(sizeof(xfs_btree_cur_t),
- "xfs_btree_cur");
- xfs_da_state_zone = kmem_zone_init(sizeof(xfs_da_state_t),
- "xfs_da_state");
- xfs_dabuf_zone = kmem_zone_init(sizeof(xfs_dabuf_t), "xfs_dabuf");
- xfs_ifork_zone = kmem_zone_init(sizeof(xfs_ifork_t), "xfs_ifork");
- xfs_trans_zone = kmem_zone_init(sizeof(xfs_trans_t), "xfs_trans");
- xfs_acl_zone_init(xfs_acl_zone, "xfs_acl");
- xfs_mru_cache_init();
- xfs_filestream_init();
-
- /*
- * The size of the zone allocated buf log item is the maximum
- * size possible under XFS. This wastes a little bit of memory,
- * but it is much faster.
- */
- xfs_buf_item_zone =
- kmem_zone_init((sizeof(xfs_buf_log_item_t) +
- (((XFS_MAX_BLOCKSIZE / XFS_BLI_CHUNK) /
- NBWORD) * sizeof(int))),
- "xfs_buf_item");
- xfs_efd_zone =
- kmem_zone_init((sizeof(xfs_efd_log_item_t) +
- ((XFS_EFD_MAX_FAST_EXTENTS - 1) *
- sizeof(xfs_extent_t))),
- "xfs_efd_item");
- xfs_efi_zone =
- kmem_zone_init((sizeof(xfs_efi_log_item_t) +
- ((XFS_EFI_MAX_FAST_EXTENTS - 1) *
- sizeof(xfs_extent_t))),
- "xfs_efi_item");
-
- /*
- * These zones warrant special memory allocator hints
- */
- xfs_inode_zone =
- kmem_zone_init_flags(sizeof(xfs_inode_t), "xfs_inode",
- KM_ZONE_HWALIGN | KM_ZONE_RECLAIM |
- KM_ZONE_SPREAD, NULL);
- xfs_ili_zone =
- kmem_zone_init_flags(sizeof(xfs_inode_log_item_t), "xfs_ili",
- KM_ZONE_SPREAD, NULL);
-
- /*
- * Allocate global trace buffers.
- */
-#ifdef XFS_ALLOC_TRACE
- xfs_alloc_trace_buf = ktrace_alloc(XFS_ALLOC_TRACE_SIZE, KM_SLEEP);
-#endif
-#ifdef XFS_BMAP_TRACE
- xfs_bmap_trace_buf = ktrace_alloc(XFS_BMAP_TRACE_SIZE, KM_SLEEP);
-#endif
-#ifdef XFS_BMBT_TRACE
- xfs_bmbt_trace_buf = ktrace_alloc(XFS_BMBT_TRACE_SIZE, KM_SLEEP);
-#endif
-#ifdef XFS_ATTR_TRACE
- xfs_attr_trace_buf = ktrace_alloc(XFS_ATTR_TRACE_SIZE, KM_SLEEP);
-#endif
-#ifdef XFS_DIR2_TRACE
- xfs_dir2_trace_buf = ktrace_alloc(XFS_DIR2_GTRACE_SIZE, KM_SLEEP);
-#endif
-
- xfs_dir_startup();
-
-#if (defined(DEBUG) || defined(INDUCE_IO_ERROR))
- xfs_error_test_init();
-#endif /* DEBUG || INDUCE_IO_ERROR */
-
- xfs_init_procfs();
- xfs_sysctl_register();
- return 0;
-}
-
-void __exit
-xfs_cleanup(void)
-{
- extern kmem_zone_t *xfs_inode_zone;
- extern kmem_zone_t *xfs_efd_zone;
- extern kmem_zone_t *xfs_efi_zone;
-
- xfs_cleanup_procfs();
- xfs_sysctl_unregister();
- xfs_filestream_uninit();
- xfs_mru_cache_uninit();
- xfs_acl_zone_destroy(xfs_acl_zone);
-
-#ifdef XFS_DIR2_TRACE
- ktrace_free(xfs_dir2_trace_buf);
-#endif
-#ifdef XFS_ATTR_TRACE
- ktrace_free(xfs_attr_trace_buf);
-#endif
-#ifdef XFS_BMBT_TRACE
- ktrace_free(xfs_bmbt_trace_buf);
-#endif
-#ifdef XFS_BMAP_TRACE
- ktrace_free(xfs_bmap_trace_buf);
-#endif
-#ifdef XFS_ALLOC_TRACE
- ktrace_free(xfs_alloc_trace_buf);
-#endif
-
- kmem_zone_destroy(xfs_bmap_free_item_zone);
- kmem_zone_destroy(xfs_btree_cur_zone);
- kmem_zone_destroy(xfs_inode_zone);
- kmem_zone_destroy(xfs_trans_zone);
- kmem_zone_destroy(xfs_da_state_zone);
- kmem_zone_destroy(xfs_dabuf_zone);
- kmem_zone_destroy(xfs_buf_item_zone);
- kmem_zone_destroy(xfs_efd_zone);
- kmem_zone_destroy(xfs_efi_zone);
- kmem_zone_destroy(xfs_ifork_zone);
- kmem_zone_destroy(xfs_ili_zone);
- kmem_zone_destroy(xfs_log_ticket_zone);
-}
-
-/*
- * xfs_start_flags
- *
- * This function fills in xfs_mount_t fields based on mount args.
- * Note: the superblock has _not_ yet been read in.
- */
-STATIC int
-xfs_start_flags(
- struct xfs_mount_args *ap,
- struct xfs_mount *mp)
-{
- /* Values are in BBs */
- if ((ap->flags & XFSMNT_NOALIGN) != XFSMNT_NOALIGN) {
- /*
- * At this point the superblock has not been read
- * in, therefore we do not know the block size.
- * Before the mount call ends we will convert
- * these to FSBs.
- */
- mp->m_dalign = ap->sunit;
- mp->m_swidth = ap->swidth;
- }
-
- if (ap->logbufs != -1 &&
- ap->logbufs != 0 &&
- (ap->logbufs < XLOG_MIN_ICLOGS ||
- ap->logbufs > XLOG_MAX_ICLOGS)) {
- cmn_err(CE_WARN,
- "XFS: invalid logbufs value: %d [not %d-%d]",
- ap->logbufs, XLOG_MIN_ICLOGS, XLOG_MAX_ICLOGS);
- return XFS_ERROR(EINVAL);
- }
- mp->m_logbufs = ap->logbufs;
- if (ap->logbufsize != -1 &&
- ap->logbufsize != 0 &&
- (ap->logbufsize < XLOG_MIN_RECORD_BSIZE ||
- ap->logbufsize > XLOG_MAX_RECORD_BSIZE ||
- !is_power_of_2(ap->logbufsize))) {
- cmn_err(CE_WARN,
- "XFS: invalid logbufsize: %d [not 16k,32k,64k,128k or 256k]",
- ap->logbufsize);
- return XFS_ERROR(EINVAL);
- }
- mp->m_logbsize = ap->logbufsize;
- mp->m_fsname_len = strlen(ap->fsname) + 1;
- mp->m_fsname = kmem_alloc(mp->m_fsname_len, KM_SLEEP);
- strcpy(mp->m_fsname, ap->fsname);
- if (ap->rtname[0]) {
- mp->m_rtname = kmem_alloc(strlen(ap->rtname) + 1, KM_SLEEP);
- strcpy(mp->m_rtname, ap->rtname);
- }
- if (ap->logname[0]) {
- mp->m_logname = kmem_alloc(strlen(ap->logname) + 1, KM_SLEEP);
- strcpy(mp->m_logname, ap->logname);
- }
-
- if (ap->flags & XFSMNT_WSYNC)
- mp->m_flags |= XFS_MOUNT_WSYNC;
-#if XFS_BIG_INUMS
- if (ap->flags & XFSMNT_INO64) {
- mp->m_flags |= XFS_MOUNT_INO64;
- mp->m_inoadd = XFS_INO64_OFFSET;
- }
-#endif
- if (ap->flags & XFSMNT_RETERR)
- mp->m_flags |= XFS_MOUNT_RETERR;
- if (ap->flags & XFSMNT_NOALIGN)
- mp->m_flags |= XFS_MOUNT_NOALIGN;
- if (ap->flags & XFSMNT_SWALLOC)
- mp->m_flags |= XFS_MOUNT_SWALLOC;
- if (ap->flags & XFSMNT_OSYNCISOSYNC)
- mp->m_flags |= XFS_MOUNT_OSYNCISOSYNC;
- if (ap->flags & XFSMNT_32BITINODES)
- mp->m_flags |= XFS_MOUNT_32BITINODES;
-
- if (ap->flags & XFSMNT_IOSIZE) {
- if (ap->iosizelog > XFS_MAX_IO_LOG ||
- ap->iosizelog < XFS_MIN_IO_LOG) {
- cmn_err(CE_WARN,
- "XFS: invalid log iosize: %d [not %d-%d]",
- ap->iosizelog, XFS_MIN_IO_LOG,
- XFS_MAX_IO_LOG);
- return XFS_ERROR(EINVAL);
- }
-
- mp->m_flags |= XFS_MOUNT_DFLT_IOSIZE;
- mp->m_readio_log = mp->m_writeio_log = ap->iosizelog;
- }
-
- if (ap->flags & XFSMNT_IKEEP)
- mp->m_flags |= XFS_MOUNT_IKEEP;
- if (ap->flags & XFSMNT_DIRSYNC)
- mp->m_flags |= XFS_MOUNT_DIRSYNC;
- if (ap->flags & XFSMNT_ATTR2)
- mp->m_flags |= XFS_MOUNT_ATTR2;
-
- if (ap->flags2 & XFSMNT2_COMPAT_IOSIZE)
- mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE;
-
- /*
- * no recovery flag requires a read-only mount
- */
- if (ap->flags & XFSMNT_NORECOVERY) {
- if (!(mp->m_flags & XFS_MOUNT_RDONLY)) {
- cmn_err(CE_WARN,
- "XFS: tried to mount a FS read-write without recovery!");
- return XFS_ERROR(EINVAL);
- }
- mp->m_flags |= XFS_MOUNT_NORECOVERY;
- }
-
- if (ap->flags & XFSMNT_NOUUID)
- mp->m_flags |= XFS_MOUNT_NOUUID;
- if (ap->flags & XFSMNT_BARRIER)
- mp->m_flags |= XFS_MOUNT_BARRIER;
- else
- mp->m_flags &= ~XFS_MOUNT_BARRIER;
-
- if (ap->flags2 & XFSMNT2_FILESTREAMS)
- mp->m_flags |= XFS_MOUNT_FILESTREAMS;
-
- if (ap->flags & XFSMNT_DMAPI)
- mp->m_flags |= XFS_MOUNT_DMAPI;
- return 0;
-}
-
-/*
- * This function fills in xfs_mount_t fields based on mount args.
- * Note: the superblock _has_ now been read in.
- */
-STATIC int
-xfs_finish_flags(
- struct xfs_mount_args *ap,
- struct xfs_mount *mp)
-{
- int ronly = (mp->m_flags & XFS_MOUNT_RDONLY);
-
- /* Fail a mount where the logbuf is smaller then the log stripe */
- if (xfs_sb_version_haslogv2(&mp->m_sb)) {
- if ((ap->logbufsize <= 0) &&
- (mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE)) {
- mp->m_logbsize = mp->m_sb.sb_logsunit;
- } else if (ap->logbufsize > 0 &&
- ap->logbufsize < mp->m_sb.sb_logsunit) {
- cmn_err(CE_WARN,
- "XFS: logbuf size must be greater than or equal to log stripe size");
- return XFS_ERROR(EINVAL);
- }
- } else {
- /* Fail a mount if the logbuf is larger than 32K */
- if (ap->logbufsize > XLOG_BIG_RECORD_BSIZE) {
- cmn_err(CE_WARN,
- "XFS: logbuf size for version 1 logs must be 16K or 32K");
- return XFS_ERROR(EINVAL);
- }
- }
-
- if (xfs_sb_version_hasattr2(&mp->m_sb))
- mp->m_flags |= XFS_MOUNT_ATTR2;
-
- /*
- * prohibit r/w mounts of read-only filesystems
- */
- if ((mp->m_sb.sb_flags & XFS_SBF_READONLY) && !ronly) {
- cmn_err(CE_WARN,
- "XFS: cannot mount a read-only filesystem as read-write");
- return XFS_ERROR(EROFS);
- }
-
- /*
- * check for shared mount.
- */
- if (ap->flags & XFSMNT_SHARED) {
- if (!xfs_sb_version_hasshared(&mp->m_sb))
- return XFS_ERROR(EINVAL);
-
- /*
- * For IRIX 6.5, shared mounts must have the shared
- * version bit set, have the persistent readonly
- * field set, must be version 0 and can only be mounted
- * read-only.
- */
- if (!ronly || !(mp->m_sb.sb_flags & XFS_SBF_READONLY) ||
- (mp->m_sb.sb_shared_vn != 0))
- return XFS_ERROR(EINVAL);
-
- mp->m_flags |= XFS_MOUNT_SHARED;
-
- /*
- * Shared XFS V0 can't deal with DMI. Return EINVAL.
- */
- if (mp->m_sb.sb_shared_vn == 0 && (ap->flags & XFSMNT_DMAPI))
- return XFS_ERROR(EINVAL);
- }
-
- if (ap->flags & XFSMNT_UQUOTA) {
- mp->m_qflags |= (XFS_UQUOTA_ACCT | XFS_UQUOTA_ACTIVE);
- if (ap->flags & XFSMNT_UQUOTAENF)
- mp->m_qflags |= XFS_UQUOTA_ENFD;
- }
-
- if (ap->flags & XFSMNT_GQUOTA) {
- mp->m_qflags |= (XFS_GQUOTA_ACCT | XFS_GQUOTA_ACTIVE);
- if (ap->flags & XFSMNT_GQUOTAENF)
- mp->m_qflags |= XFS_OQUOTA_ENFD;
- } else if (ap->flags & XFSMNT_PQUOTA) {
- mp->m_qflags |= (XFS_PQUOTA_ACCT | XFS_PQUOTA_ACTIVE);
- if (ap->flags & XFSMNT_PQUOTAENF)
- mp->m_qflags |= XFS_OQUOTA_ENFD;
- }
-
- return 0;
-}
-
-/*
- * xfs_mount
- *
- * The file system configurations are:
- * (1) device (partition) with data and internal log
- * (2) logical volume with data and log subvolumes.
- * (3) logical volume with data, log, and realtime subvolumes.
- *
- * We only have to handle opening the log and realtime volumes here if
- * they are present. The data subvolume has already been opened by
- * get_sb_bdev() and is stored in vfsp->vfs_super->s_bdev.
- */
-int
-xfs_mount(
- struct xfs_mount *mp,
- struct xfs_mount_args *args,
- cred_t *credp)
-{
- struct block_device *ddev, *logdev, *rtdev;
- int flags = 0, error;
-
- ddev = mp->m_super->s_bdev;
- logdev = rtdev = NULL;
-
- error = xfs_dmops_get(mp, args);
- if (error)
- return error;
- error = xfs_qmops_get(mp, args);
- if (error)
- return error;
-
- if (args->flags & XFSMNT_QUIET)
- flags |= XFS_MFSI_QUIET;
-
- /*
- * Open real time and log devices - order is important.
- */
- if (args->logname[0]) {
- error = xfs_blkdev_get(mp, args->logname, &logdev);
- if (error)
- return error;
- }
- if (args->rtname[0]) {
- error = xfs_blkdev_get(mp, args->rtname, &rtdev);
- if (error) {
- xfs_blkdev_put(logdev);
- return error;
- }
-
- if (rtdev == ddev || rtdev == logdev) {
- cmn_err(CE_WARN,
- "XFS: Cannot mount filesystem with identical rtdev and ddev/logdev.");
- xfs_blkdev_put(logdev);
- xfs_blkdev_put(rtdev);
- return EINVAL;
- }
- }
-
- /*
- * Setup xfs_mount buffer target pointers
- */
- error = ENOMEM;
- mp->m_ddev_targp = xfs_alloc_buftarg(ddev, 0);
- if (!mp->m_ddev_targp) {
- xfs_blkdev_put(logdev);
- xfs_blkdev_put(rtdev);
- return error;
- }
- if (rtdev) {
- mp->m_rtdev_targp = xfs_alloc_buftarg(rtdev, 1);
- if (!mp->m_rtdev_targp) {
- xfs_blkdev_put(logdev);
- xfs_blkdev_put(rtdev);
- goto error0;
- }
- }
- mp->m_logdev_targp = (logdev && logdev != ddev) ?
- xfs_alloc_buftarg(logdev, 1) : mp->m_ddev_targp;
- if (!mp->m_logdev_targp) {
- xfs_blkdev_put(logdev);
- xfs_blkdev_put(rtdev);
- goto error0;
- }
-
- /*
- * Setup flags based on mount(2) options and then the superblock
- */
- error = xfs_start_flags(args, mp);
- if (error)
- goto error1;
- error = xfs_readsb(mp, flags);
- if (error)
- goto error1;
- error = xfs_finish_flags(args, mp);
- if (error)
- goto error2;
-
- /*
- * Setup xfs_mount buffer target pointers based on superblock
- */
- error = xfs_setsize_buftarg(mp->m_ddev_targp, mp->m_sb.sb_blocksize,
- mp->m_sb.sb_sectsize);
- if (!error && logdev && logdev != ddev) {
- unsigned int log_sector_size = BBSIZE;
-
- if (xfs_sb_version_hassector(&mp->m_sb))
- log_sector_size = mp->m_sb.sb_logsectsize;
- error = xfs_setsize_buftarg(mp->m_logdev_targp,
- mp->m_sb.sb_blocksize,
- log_sector_size);
- }
- if (!error && rtdev)
- error = xfs_setsize_buftarg(mp->m_rtdev_targp,
- mp->m_sb.sb_blocksize,
- mp->m_sb.sb_sectsize);
- if (error)
- goto error2;
-
- if (mp->m_flags & XFS_MOUNT_BARRIER)
- xfs_mountfs_check_barriers(mp);
-
- if ((error = xfs_filestream_mount(mp)))
- goto error2;
-
- error = xfs_mountfs(mp, flags);
- if (error)
- goto error2;
-
- XFS_SEND_MOUNT(mp, DM_RIGHT_NULL, args->mtpt, args->fsname);
-
- return 0;
-
-error2:
- if (mp->m_sb_bp)
- xfs_freesb(mp);
-error1:
- xfs_binval(mp->m_ddev_targp);
- if (logdev && logdev != ddev)
- xfs_binval(mp->m_logdev_targp);
- if (rtdev)
- xfs_binval(mp->m_rtdev_targp);
-error0:
- xfs_unmountfs_close(mp, credp);
- xfs_qmops_put(mp);
- xfs_dmops_put(mp);
- return error;
-}
-
-int
-xfs_unmount(
- xfs_mount_t *mp,
- int flags,
- cred_t *credp)
-{
- xfs_inode_t *rip;
- bhv_vnode_t *rvp;
- int unmount_event_wanted = 0;
- int unmount_event_flags = 0;
- int xfs_unmountfs_needed = 0;
- int error;
-
- rip = mp->m_rootip;
- rvp = XFS_ITOV(rip);
-
-#ifdef HAVE_DMAPI
- if (mp->m_flags & XFS_MOUNT_DMAPI) {
- error = XFS_SEND_PREUNMOUNT(mp,
- rip, DM_RIGHT_NULL, rip, DM_RIGHT_NULL,
- NULL, NULL, 0, 0,
- (mp->m_dmevmask & (1<<DM_EVENT_PREUNMOUNT))?
- 0:DM_FLAGS_UNWANTED);
- if (error)
- return XFS_ERROR(error);
- unmount_event_wanted = 1;
- unmount_event_flags = (mp->m_dmevmask & (1<<DM_EVENT_UNMOUNT))?
- 0 : DM_FLAGS_UNWANTED;
- }
-#endif
-
- /*
- * Blow away any referenced inode in the filestreams cache.
- * This can and will cause log traffic as inodes go inactive
- * here.
- */
- xfs_filestream_unmount(mp);
-
- XFS_bflush(mp->m_ddev_targp);
- error = xfs_unmount_flush(mp, 0);
- if (error)
- goto out;
-
- ASSERT(vn_count(rvp) == 1);
-
- /*
- * Drop the reference count
- */
- IRELE(rip);
-
- /*
- * If we're forcing a shutdown, typically because of a media error,
- * we want to make sure we invalidate dirty pages that belong to
- * referenced vnodes as well.
- */
- if (XFS_FORCED_SHUTDOWN(mp)) {
- error = xfs_sync(mp, SYNC_WAIT | SYNC_CLOSE);
- ASSERT(error != EFSCORRUPTED);
- }
- xfs_unmountfs_needed = 1;
-
-out:
- /* Send DMAPI event, if required.
- * Then do xfs_unmountfs() if needed.
- * Then return error (or zero).
- */
- if (unmount_event_wanted) {
- /* Note: mp structure must still exist for
- * XFS_SEND_UNMOUNT() call.
- */
- XFS_SEND_UNMOUNT(mp, error == 0 ? rip : NULL,
- DM_RIGHT_NULL, 0, error, unmount_event_flags);
- }
- if (xfs_unmountfs_needed) {
- /*
- * Call common unmount function to flush to disk
- * and free the super block buffer & mount structures.
- */
- xfs_unmountfs(mp, credp);
- xfs_qmops_put(mp);
- xfs_dmops_put(mp);
- kmem_free(mp, sizeof(xfs_mount_t));
- }
-
- return XFS_ERROR(error);
-}
-
STATIC void
xfs_quiesce_fs(
xfs_mount_t *mp)
@@ -694,30 +114,6 @@ xfs_attr_quiesce(
xfs_unmountfs_writesb(mp);
}
-int
-xfs_mntupdate(
- struct xfs_mount *mp,
- int *flags,
- struct xfs_mount_args *args)
-{
- if (!(*flags & MS_RDONLY)) { /* rw/ro -> rw */
- if (mp->m_flags & XFS_MOUNT_RDONLY)
- mp->m_flags &= ~XFS_MOUNT_RDONLY;
- if (args->flags & XFSMNT_BARRIER) {
- mp->m_flags |= XFS_MOUNT_BARRIER;
- xfs_mountfs_check_barriers(mp);
- } else {
- mp->m_flags &= ~XFS_MOUNT_BARRIER;
- }
- } else if (!(mp->m_flags & XFS_MOUNT_RDONLY)) { /* rw -> ro */
- xfs_filestream_flush(mp);
- xfs_sync(mp, SYNC_DATA_QUIESCE);
- xfs_attr_quiesce(mp);
- mp->m_flags |= XFS_MOUNT_RDONLY;
- }
- return 0;
-}
-
/*
* xfs_unmount_flush implements a set of flush operation on special
* inodes, which are needed as a separate set of operations so that
@@ -732,7 +128,6 @@ xfs_unmount_flush(
xfs_inode_t *rip = mp->m_rootip;
xfs_inode_t *rbmip;
xfs_inode_t *rsumip = NULL;
- bhv_vnode_t *rvp = XFS_ITOV(rip);
int error;
xfs_ilock(rip, XFS_ILOCK_EXCL | XFS_ILOCK_PARENT);
@@ -750,7 +145,7 @@ xfs_unmount_flush(
if (error == EFSCORRUPTED)
goto fscorrupt_out;
- ASSERT(vn_count(XFS_ITOV(rbmip)) == 1);
+ ASSERT(vn_count(VFS_I(rbmip)) == 1);
rsumip = mp->m_rsumip;
xfs_ilock(rsumip, XFS_ILOCK_EXCL);
@@ -761,7 +156,7 @@ xfs_unmount_flush(
if (error == EFSCORRUPTED)
goto fscorrupt_out;
- ASSERT(vn_count(XFS_ITOV(rsumip)) == 1);
+ ASSERT(vn_count(VFS_I(rsumip)) == 1);
}
/*
@@ -771,7 +166,7 @@ xfs_unmount_flush(
if (error == EFSCORRUPTED)
goto fscorrupt_out2;
- if (vn_count(rvp) != 1 && !relocation) {
+ if (vn_count(VFS_I(rip)) != 1 && !relocation) {
xfs_iunlock(rip, XFS_ILOCK_EXCL);
return XFS_ERROR(EBUSY);
}
@@ -888,7 +283,7 @@ xfs_sync_inodes(
int *bypassed)
{
xfs_inode_t *ip = NULL;
- bhv_vnode_t *vp = NULL;
+ struct inode *vp = NULL;
int error;
int last_error;
uint64_t fflag;
@@ -1008,7 +403,7 @@ xfs_sync_inodes(
continue;
}
- vp = XFS_ITOV_NULL(ip);
+ vp = VFS_I(ip);
/*
* If the vnode is gone then this is being torn down,
@@ -1048,7 +443,7 @@ xfs_sync_inodes(
if (XFS_FORCED_SHUTDOWN(mp) && !(flags & SYNC_CLOSE)) {
XFS_MOUNT_IUNLOCK(mp);
- kmem_free(ipointer, sizeof(xfs_iptr_t));
+ kmem_free(ipointer);
return 0;
}
@@ -1083,7 +478,7 @@ xfs_sync_inodes(
IPOINTER_INSERT(ip, mp);
xfs_ilock(ip, lock_flags);
- ASSERT(vp == XFS_ITOV(ip));
+ ASSERT(vp == VFS_I(ip));
ASSERT(ip->i_mount == mp);
vnode_refed = B_TRUE;
@@ -1194,7 +589,7 @@ xfs_sync_inodes(
}
XFS_MOUNT_IUNLOCK(mp);
ASSERT(ipointer_in == B_FALSE);
- kmem_free(ipointer, sizeof(xfs_iptr_t));
+ kmem_free(ipointer);
return XFS_ERROR(error);
}
@@ -1224,7 +619,7 @@ xfs_sync_inodes(
ASSERT(ipointer_in == B_FALSE);
- kmem_free(ipointer, sizeof(xfs_iptr_t));
+ kmem_free(ipointer);
return XFS_ERROR(last_error);
}
diff --git a/fs/xfs/xfs_vfsops.h b/fs/xfs/xfs_vfsops.h
index 1688817c55e..a74b05087da 100644
--- a/fs/xfs/xfs_vfsops.h
+++ b/fs/xfs/xfs_vfsops.h
@@ -8,11 +8,6 @@ struct kstatfs;
struct xfs_mount;
struct xfs_mount_args;
-int xfs_mount(struct xfs_mount *mp, struct xfs_mount_args *args,
- struct cred *credp);
-int xfs_unmount(struct xfs_mount *mp, int flags, struct cred *credp);
-int xfs_mntupdate(struct xfs_mount *mp, int *flags,
- struct xfs_mount_args *args);
int xfs_sync(struct xfs_mount *mp, int flags);
void xfs_do_force_shutdown(struct xfs_mount *mp, int flags, char *fname,
int lnnum);
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index e475e3717eb..8b6812f66a1 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -75,26 +75,23 @@ xfs_open(
return 0;
}
-/*
- * xfs_setattr
- */
int
xfs_setattr(
- xfs_inode_t *ip,
- bhv_vattr_t *vap,
+ struct xfs_inode *ip,
+ struct iattr *iattr,
int flags,
cred_t *credp)
{
xfs_mount_t *mp = ip->i_mount;
+ struct inode *inode = VFS_I(ip);
+ int mask = iattr->ia_valid;
xfs_trans_t *tp;
- int mask;
int code;
uint lock_flags;
uint commit_flags=0;
uid_t uid=0, iuid=0;
gid_t gid=0, igid=0;
int timeflags = 0;
- xfs_prid_t projid=0, iprojid=0;
struct xfs_dquot *udqp, *gdqp, *olddquot1, *olddquot2;
int file_owner;
int need_iolock = 1;
@@ -104,30 +101,9 @@ xfs_setattr(
if (mp->m_flags & XFS_MOUNT_RDONLY)
return XFS_ERROR(EROFS);
- /*
- * Cannot set certain attributes.
- */
- mask = vap->va_mask;
- if (mask & XFS_AT_NOSET) {
- return XFS_ERROR(EINVAL);
- }
-
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
- /*
- * Timestamps do not need to be logged and hence do not
- * need to be done within a transaction.
- */
- if (mask & XFS_AT_UPDTIMES) {
- ASSERT((mask & ~XFS_AT_UPDTIMES) == 0);
- timeflags = ((mask & XFS_AT_UPDATIME) ? XFS_ICHGTIME_ACC : 0) |
- ((mask & XFS_AT_UPDCTIME) ? XFS_ICHGTIME_CHG : 0) |
- ((mask & XFS_AT_UPDMTIME) ? XFS_ICHGTIME_MOD : 0);
- xfs_ichgtime(ip, timeflags);
- return 0;
- }
-
olddquot1 = olddquot2 = NULL;
udqp = gdqp = NULL;
@@ -139,28 +115,22 @@ xfs_setattr(
* If the IDs do change before we take the ilock, we're covered
* because the i_*dquot fields will get updated anyway.
*/
- if (XFS_IS_QUOTA_ON(mp) &&
- (mask & (XFS_AT_UID|XFS_AT_GID|XFS_AT_PROJID))) {
+ if (XFS_IS_QUOTA_ON(mp) && (mask & (ATTR_UID|ATTR_GID))) {
uint qflags = 0;
- if ((mask & XFS_AT_UID) && XFS_IS_UQUOTA_ON(mp)) {
- uid = vap->va_uid;
+ if ((mask & ATTR_UID) && XFS_IS_UQUOTA_ON(mp)) {
+ uid = iattr->ia_uid;
qflags |= XFS_QMOPT_UQUOTA;
} else {
uid = ip->i_d.di_uid;
}
- if ((mask & XFS_AT_GID) && XFS_IS_GQUOTA_ON(mp)) {
- gid = vap->va_gid;
+ if ((mask & ATTR_GID) && XFS_IS_GQUOTA_ON(mp)) {
+ gid = iattr->ia_gid;
qflags |= XFS_QMOPT_GQUOTA;
} else {
gid = ip->i_d.di_gid;
}
- if ((mask & XFS_AT_PROJID) && XFS_IS_PQUOTA_ON(mp)) {
- projid = vap->va_projid;
- qflags |= XFS_QMOPT_PQUOTA;
- } else {
- projid = ip->i_d.di_projid;
- }
+
/*
* We take a reference when we initialize udqp and gdqp,
* so it is important that we never blindly double trip on
@@ -168,8 +138,8 @@ xfs_setattr(
*/
ASSERT(udqp == NULL);
ASSERT(gdqp == NULL);
- code = XFS_QM_DQVOPALLOC(mp, ip, uid, gid, projid, qflags,
- &udqp, &gdqp);
+ code = XFS_QM_DQVOPALLOC(mp, ip, uid, gid, ip->i_d.di_projid,
+ qflags, &udqp, &gdqp);
if (code)
return code;
}
@@ -180,10 +150,10 @@ xfs_setattr(
*/
tp = NULL;
lock_flags = XFS_ILOCK_EXCL;
- if (flags & ATTR_NOLOCK)
+ if (flags & XFS_ATTR_NOLOCK)
need_iolock = 0;
- if (!(mask & XFS_AT_SIZE)) {
- if ((mask != (XFS_AT_CTIME|XFS_AT_ATIME|XFS_AT_MTIME)) ||
+ if (!(mask & ATTR_SIZE)) {
+ if ((mask != (ATTR_CTIME|ATTR_ATIME|ATTR_MTIME)) ||
(mp->m_flags & XFS_MOUNT_WSYNC)) {
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE);
commit_flags = 0;
@@ -196,10 +166,10 @@ xfs_setattr(
}
} else {
if (DM_EVENT_ENABLED(ip, DM_EVENT_TRUNCATE) &&
- !(flags & ATTR_DMI)) {
+ !(flags & XFS_ATTR_DMI)) {
int dmflags = AT_DELAY_FLAG(flags) | DM_SEM_FLAG_WR;
code = XFS_SEND_DATA(mp, DM_EVENT_TRUNCATE, ip,
- vap->va_size, 0, dmflags, NULL);
+ iattr->ia_size, 0, dmflags, NULL);
if (code) {
lock_flags = 0;
goto error_return;
@@ -212,16 +182,14 @@ xfs_setattr(
xfs_ilock(ip, lock_flags);
/* boolean: are we the file owner? */
- file_owner = (current_fsuid(credp) == ip->i_d.di_uid);
+ file_owner = (current_fsuid() == ip->i_d.di_uid);
/*
* Change various properties of a file.
* Only the owner or users with CAP_FOWNER
* capability may do these things.
*/
- if (mask &
- (XFS_AT_MODE|XFS_AT_XFLAGS|XFS_AT_EXTSIZE|XFS_AT_UID|
- XFS_AT_GID|XFS_AT_PROJID)) {
+ if (mask & (ATTR_MODE|ATTR_UID|ATTR_GID)) {
/*
* CAP_FOWNER overrides the following restrictions:
*
@@ -245,21 +213,21 @@ xfs_setattr(
* IDs of the calling process shall match the group owner of
* the file when setting the set-group-ID bit on that file
*/
- if (mask & XFS_AT_MODE) {
+ if (mask & ATTR_MODE) {
mode_t m = 0;
- if ((vap->va_mode & S_ISUID) && !file_owner)
+ if ((iattr->ia_mode & S_ISUID) && !file_owner)
m |= S_ISUID;
- if ((vap->va_mode & S_ISGID) &&
+ if ((iattr->ia_mode & S_ISGID) &&
!in_group_p((gid_t)ip->i_d.di_gid))
m |= S_ISGID;
#if 0
/* Linux allows this, Irix doesn't. */
- if ((vap->va_mode & S_ISVTX) && !S_ISDIR(ip->i_d.di_mode))
+ if ((iattr->ia_mode & S_ISVTX) && !S_ISDIR(ip->i_d.di_mode))
m |= S_ISVTX;
#endif
if (m && !capable(CAP_FSETID))
- vap->va_mode &= ~m;
+ iattr->ia_mode &= ~m;
}
}
@@ -270,7 +238,7 @@ xfs_setattr(
* and can change the group id only to a group of which he
* or she is a member.
*/
- if (mask & (XFS_AT_UID|XFS_AT_GID|XFS_AT_PROJID)) {
+ if (mask & (ATTR_UID|ATTR_GID)) {
/*
* These IDs could have changed since we last looked at them.
* But, we're assured that if the ownership did change
@@ -278,12 +246,9 @@ xfs_setattr(
* would have changed also.
*/
iuid = ip->i_d.di_uid;
- iprojid = ip->i_d.di_projid;
igid = ip->i_d.di_gid;
- gid = (mask & XFS_AT_GID) ? vap->va_gid : igid;
- uid = (mask & XFS_AT_UID) ? vap->va_uid : iuid;
- projid = (mask & XFS_AT_PROJID) ? (xfs_prid_t)vap->va_projid :
- iprojid;
+ gid = (mask & ATTR_GID) ? iattr->ia_gid : igid;
+ uid = (mask & ATTR_UID) ? iattr->ia_uid : iuid;
/*
* CAP_CHOWN overrides the following restrictions:
@@ -303,11 +268,10 @@ xfs_setattr(
goto error_return;
}
/*
- * Do a quota reservation only if uid/projid/gid is actually
+ * Do a quota reservation only if uid/gid is actually
* going to change.
*/
if ((XFS_IS_UQUOTA_ON(mp) && iuid != uid) ||
- (XFS_IS_PQUOTA_ON(mp) && iprojid != projid) ||
(XFS_IS_GQUOTA_ON(mp) && igid != gid)) {
ASSERT(tp);
code = XFS_QM_DQVOPCHOWNRESV(mp, tp, ip, udqp, gdqp,
@@ -321,13 +285,13 @@ xfs_setattr(
/*
* Truncate file. Must have write permission and not be a directory.
*/
- if (mask & XFS_AT_SIZE) {
+ if (mask & ATTR_SIZE) {
/* Short circuit the truncate case for zero length files */
- if ((vap->va_size == 0) &&
- (ip->i_size == 0) && (ip->i_d.di_nextents == 0)) {
+ if (iattr->ia_size == 0 &&
+ ip->i_size == 0 && ip->i_d.di_nextents == 0) {
xfs_iunlock(ip, XFS_ILOCK_EXCL);
lock_flags &= ~XFS_ILOCK_EXCL;
- if (mask & XFS_AT_CTIME)
+ if (mask & ATTR_CTIME)
xfs_ichgtime(ip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
code = 0;
goto error_return;
@@ -350,9 +314,9 @@ xfs_setattr(
/*
* Change file access or modified times.
*/
- if (mask & (XFS_AT_ATIME|XFS_AT_MTIME)) {
+ if (mask & (ATTR_ATIME|ATTR_MTIME)) {
if (!file_owner) {
- if ((flags & ATTR_UTIME) &&
+ if ((mask & (ATTR_MTIME_SET|ATTR_ATIME_SET)) &&
!capable(CAP_FOWNER)) {
code = XFS_ERROR(EPERM);
goto error_return;
@@ -361,90 +325,23 @@ xfs_setattr(
}
/*
- * Change extent size or realtime flag.
- */
- if (mask & (XFS_AT_EXTSIZE|XFS_AT_XFLAGS)) {
- /*
- * Can't change extent size if any extents are allocated.
- */
- if (ip->i_d.di_nextents && (mask & XFS_AT_EXTSIZE) &&
- ((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) !=
- vap->va_extsize) ) {
- code = XFS_ERROR(EINVAL); /* EFBIG? */
- goto error_return;
- }
-
- /*
- * Can't change realtime flag if any extents are allocated.
- */
- if ((ip->i_d.di_nextents || ip->i_delayed_blks) &&
- (mask & XFS_AT_XFLAGS) &&
- (XFS_IS_REALTIME_INODE(ip)) !=
- (vap->va_xflags & XFS_XFLAG_REALTIME)) {
- code = XFS_ERROR(EINVAL); /* EFBIG? */
- goto error_return;
- }
- /*
- * Extent size must be a multiple of the appropriate block
- * size, if set at all.
- */
- if ((mask & XFS_AT_EXTSIZE) && vap->va_extsize != 0) {
- xfs_extlen_t size;
-
- if (XFS_IS_REALTIME_INODE(ip) ||
- ((mask & XFS_AT_XFLAGS) &&
- (vap->va_xflags & XFS_XFLAG_REALTIME))) {
- size = mp->m_sb.sb_rextsize <<
- mp->m_sb.sb_blocklog;
- } else {
- size = mp->m_sb.sb_blocksize;
- }
- if (vap->va_extsize % size) {
- code = XFS_ERROR(EINVAL);
- goto error_return;
- }
- }
- /*
- * If realtime flag is set then must have realtime data.
- */
- if ((mask & XFS_AT_XFLAGS) &&
- (vap->va_xflags & XFS_XFLAG_REALTIME)) {
- if ((mp->m_sb.sb_rblocks == 0) ||
- (mp->m_sb.sb_rextsize == 0) ||
- (ip->i_d.di_extsize % mp->m_sb.sb_rextsize)) {
- code = XFS_ERROR(EINVAL);
- goto error_return;
- }
- }
-
- /*
- * Can't modify an immutable/append-only file unless
- * we have appropriate permission.
- */
- if ((mask & XFS_AT_XFLAGS) &&
- (ip->i_d.di_flags &
- (XFS_DIFLAG_IMMUTABLE|XFS_DIFLAG_APPEND) ||
- (vap->va_xflags &
- (XFS_XFLAG_IMMUTABLE | XFS_XFLAG_APPEND))) &&
- !capable(CAP_LINUX_IMMUTABLE)) {
- code = XFS_ERROR(EPERM);
- goto error_return;
- }
- }
-
- /*
* Now we can make the changes. Before we join the inode
- * to the transaction, if XFS_AT_SIZE is set then take care of
+ * to the transaction, if ATTR_SIZE is set then take care of
* the part of the truncation that must be done without the
* inode lock. This needs to be done before joining the inode
* to the transaction, because the inode cannot be unlocked
* once it is a part of the transaction.
*/
- if (mask & XFS_AT_SIZE) {
+ if (mask & ATTR_SIZE) {
code = 0;
- if ((vap->va_size > ip->i_size) &&
- (flags & ATTR_NOSIZETOK) == 0) {
- code = xfs_igrow_start(ip, vap->va_size, credp);
+ if (iattr->ia_size > ip->i_size) {
+ /*
+ * Do the first part of growing a file: zero any data
+ * in the last block that is beyond the old EOF. We
+ * need to do this before the inode is joined to the
+ * transaction to modify the i_size.
+ */
+ code = xfs_zero_eof(ip, iattr->ia_size, ip->i_size);
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
@@ -461,10 +358,10 @@ xfs_setattr(
* not within the range we care about here.
*/
if (!code &&
- (ip->i_size != ip->i_d.di_size) &&
- (vap->va_size > ip->i_d.di_size)) {
+ ip->i_size != ip->i_d.di_size &&
+ iattr->ia_size > ip->i_d.di_size) {
code = xfs_flush_pages(ip,
- ip->i_d.di_size, vap->va_size,
+ ip->i_d.di_size, iattr->ia_size,
XFS_B_ASYNC, FI_NONE);
}
@@ -472,7 +369,7 @@ xfs_setattr(
vn_iowait(ip);
if (!code)
- code = xfs_itruncate_data(ip, vap->va_size);
+ code = xfs_itruncate_data(ip, iattr->ia_size);
if (code) {
ASSERT(tp == NULL);
lock_flags &= ~XFS_ILOCK_EXCL;
@@ -501,28 +398,30 @@ xfs_setattr(
/*
* Truncate file. Must have write permission and not be a directory.
*/
- if (mask & XFS_AT_SIZE) {
+ if (mask & ATTR_SIZE) {
/*
* Only change the c/mtime if we are changing the size
* or we are explicitly asked to change it. This handles
* the semantic difference between truncate() and ftruncate()
* as implemented in the VFS.
*/
- if (vap->va_size != ip->i_size || (mask & XFS_AT_CTIME))
+ if (iattr->ia_size != ip->i_size || (mask & ATTR_CTIME))
timeflags |= XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG;
- if (vap->va_size > ip->i_size) {
- xfs_igrow_finish(tp, ip, vap->va_size,
- !(flags & ATTR_DMI));
- } else if ((vap->va_size <= ip->i_size) ||
- ((vap->va_size == 0) && ip->i_d.di_nextents)) {
+ if (iattr->ia_size > ip->i_size) {
+ ip->i_d.di_size = iattr->ia_size;
+ ip->i_size = iattr->ia_size;
+ if (!(flags & XFS_ATTR_DMI))
+ xfs_ichgtime(ip, XFS_ICHGTIME_CHG);
+ xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
+ } else if (iattr->ia_size <= ip->i_size ||
+ (iattr->ia_size == 0 && ip->i_d.di_nextents)) {
/*
* signal a sync transaction unless
* we're truncating an already unlinked
* file on a wsync filesystem
*/
- code = xfs_itruncate_finish(&tp, ip,
- (xfs_fsize_t)vap->va_size,
+ code = xfs_itruncate_finish(&tp, ip, iattr->ia_size,
XFS_DATA_FORK,
((ip->i_d.di_nlink != 0 ||
!(mp->m_flags & XFS_MOUNT_WSYNC))
@@ -544,9 +443,12 @@ xfs_setattr(
/*
* Change file access modes.
*/
- if (mask & XFS_AT_MODE) {
+ if (mask & ATTR_MODE) {
ip->i_d.di_mode &= S_IFMT;
- ip->i_d.di_mode |= vap->va_mode & ~S_IFMT;
+ ip->i_d.di_mode |= iattr->ia_mode & ~S_IFMT;
+
+ inode->i_mode &= S_IFMT;
+ inode->i_mode |= iattr->ia_mode & ~S_IFMT;
xfs_trans_log_inode (tp, ip, XFS_ILOG_CORE);
timeflags |= XFS_ICHGTIME_CHG;
@@ -559,7 +461,7 @@ xfs_setattr(
* and can change the group id only to a group of which he
* or she is a member.
*/
- if (mask & (XFS_AT_UID|XFS_AT_GID|XFS_AT_PROJID)) {
+ if (mask & (ATTR_UID|ATTR_GID)) {
/*
* CAP_FSETID overrides the following restrictions:
*
@@ -577,39 +479,24 @@ xfs_setattr(
*/
if (iuid != uid) {
if (XFS_IS_UQUOTA_ON(mp)) {
- ASSERT(mask & XFS_AT_UID);
+ ASSERT(mask & ATTR_UID);
ASSERT(udqp);
olddquot1 = XFS_QM_DQVOPCHOWN(mp, tp, ip,
&ip->i_udquot, udqp);
}
ip->i_d.di_uid = uid;
+ inode->i_uid = uid;
}
if (igid != gid) {
if (XFS_IS_GQUOTA_ON(mp)) {
ASSERT(!XFS_IS_PQUOTA_ON(mp));
- ASSERT(mask & XFS_AT_GID);
+ ASSERT(mask & ATTR_GID);
ASSERT(gdqp);
olddquot2 = XFS_QM_DQVOPCHOWN(mp, tp, ip,
&ip->i_gdquot, gdqp);
}
ip->i_d.di_gid = gid;
- }
- if (iprojid != projid) {
- if (XFS_IS_PQUOTA_ON(mp)) {
- ASSERT(!XFS_IS_GQUOTA_ON(mp));
- ASSERT(mask & XFS_AT_PROJID);
- ASSERT(gdqp);
- olddquot2 = XFS_QM_DQVOPCHOWN(mp, tp, ip,
- &ip->i_gdquot, gdqp);
- }
- ip->i_d.di_projid = projid;
- /*
- * We may have to rev the inode as well as
- * the superblock version number since projids didn't
- * exist before DINODE_VERSION_2 and SB_VERSION_NLINK.
- */
- if (ip->i_d.di_version == XFS_DINODE_VERSION_1)
- xfs_bump_ino_vers2(tp, ip);
+ inode->i_gid = gid;
}
xfs_trans_log_inode (tp, ip, XFS_ILOG_CORE);
@@ -620,82 +507,33 @@ xfs_setattr(
/*
* Change file access or modified times.
*/
- if (mask & (XFS_AT_ATIME|XFS_AT_MTIME)) {
- if (mask & XFS_AT_ATIME) {
- ip->i_d.di_atime.t_sec = vap->va_atime.tv_sec;
- ip->i_d.di_atime.t_nsec = vap->va_atime.tv_nsec;
+ if (mask & (ATTR_ATIME|ATTR_MTIME)) {
+ if (mask & ATTR_ATIME) {
+ inode->i_atime = iattr->ia_atime;
+ ip->i_d.di_atime.t_sec = iattr->ia_atime.tv_sec;
+ ip->i_d.di_atime.t_nsec = iattr->ia_atime.tv_nsec;
ip->i_update_core = 1;
- timeflags &= ~XFS_ICHGTIME_ACC;
}
- if (mask & XFS_AT_MTIME) {
- ip->i_d.di_mtime.t_sec = vap->va_mtime.tv_sec;
- ip->i_d.di_mtime.t_nsec = vap->va_mtime.tv_nsec;
+ if (mask & ATTR_MTIME) {
+ inode->i_mtime = iattr->ia_mtime;
+ ip->i_d.di_mtime.t_sec = iattr->ia_mtime.tv_sec;
+ ip->i_d.di_mtime.t_nsec = iattr->ia_mtime.tv_nsec;
timeflags &= ~XFS_ICHGTIME_MOD;
timeflags |= XFS_ICHGTIME_CHG;
}
- if (tp && (flags & ATTR_UTIME))
+ if (tp && (mask & (ATTR_MTIME_SET|ATTR_ATIME_SET)))
xfs_trans_log_inode (tp, ip, XFS_ILOG_CORE);
}
/*
- * Change XFS-added attributes.
- */
- if (mask & (XFS_AT_EXTSIZE|XFS_AT_XFLAGS)) {
- if (mask & XFS_AT_EXTSIZE) {
- /*
- * Converting bytes to fs blocks.
- */
- ip->i_d.di_extsize = vap->va_extsize >>
- mp->m_sb.sb_blocklog;
- }
- if (mask & XFS_AT_XFLAGS) {
- uint di_flags;
-
- /* can't set PREALLOC this way, just preserve it */
- di_flags = (ip->i_d.di_flags & XFS_DIFLAG_PREALLOC);
- if (vap->va_xflags & XFS_XFLAG_IMMUTABLE)
- di_flags |= XFS_DIFLAG_IMMUTABLE;
- if (vap->va_xflags & XFS_XFLAG_APPEND)
- di_flags |= XFS_DIFLAG_APPEND;
- if (vap->va_xflags & XFS_XFLAG_SYNC)
- di_flags |= XFS_DIFLAG_SYNC;
- if (vap->va_xflags & XFS_XFLAG_NOATIME)
- di_flags |= XFS_DIFLAG_NOATIME;
- if (vap->va_xflags & XFS_XFLAG_NODUMP)
- di_flags |= XFS_DIFLAG_NODUMP;
- if (vap->va_xflags & XFS_XFLAG_PROJINHERIT)
- di_flags |= XFS_DIFLAG_PROJINHERIT;
- if (vap->va_xflags & XFS_XFLAG_NODEFRAG)
- di_flags |= XFS_DIFLAG_NODEFRAG;
- if (vap->va_xflags & XFS_XFLAG_FILESTREAM)
- di_flags |= XFS_DIFLAG_FILESTREAM;
- if ((ip->i_d.di_mode & S_IFMT) == S_IFDIR) {
- if (vap->va_xflags & XFS_XFLAG_RTINHERIT)
- di_flags |= XFS_DIFLAG_RTINHERIT;
- if (vap->va_xflags & XFS_XFLAG_NOSYMLINKS)
- di_flags |= XFS_DIFLAG_NOSYMLINKS;
- if (vap->va_xflags & XFS_XFLAG_EXTSZINHERIT)
- di_flags |= XFS_DIFLAG_EXTSZINHERIT;
- } else if ((ip->i_d.di_mode & S_IFMT) == S_IFREG) {
- if (vap->va_xflags & XFS_XFLAG_REALTIME)
- di_flags |= XFS_DIFLAG_REALTIME;
- if (vap->va_xflags & XFS_XFLAG_EXTSIZE)
- di_flags |= XFS_DIFLAG_EXTSIZE;
- }
- ip->i_d.di_flags = di_flags;
- }
- xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
- timeflags |= XFS_ICHGTIME_CHG;
- }
-
- /*
- * Change file inode change time only if XFS_AT_CTIME set
+ * Change file inode change time only if ATTR_CTIME set
* AND we have been called by a DMI function.
*/
- if ( (flags & ATTR_DMI) && (mask & XFS_AT_CTIME) ) {
- ip->i_d.di_ctime.t_sec = vap->va_ctime.tv_sec;
- ip->i_d.di_ctime.t_nsec = vap->va_ctime.tv_nsec;
+ if ((flags & XFS_ATTR_DMI) && (mask & ATTR_CTIME)) {
+ inode->i_ctime = iattr->ia_ctime;
+ ip->i_d.di_ctime.t_sec = iattr->ia_ctime.tv_sec;
+ ip->i_d.di_ctime.t_nsec = iattr->ia_ctime.tv_nsec;
ip->i_update_core = 1;
timeflags &= ~XFS_ICHGTIME_CHG;
}
@@ -704,7 +542,7 @@ xfs_setattr(
* Send out timestamp changes that need to be set to the
* current time. Not done when called by a DMI function.
*/
- if (timeflags && !(flags & ATTR_DMI))
+ if (timeflags && !(flags & XFS_ATTR_DMI))
xfs_ichgtime(ip, timeflags);
XFS_STATS_INC(xs_ig_attrchg);
@@ -742,7 +580,7 @@ xfs_setattr(
}
if (DM_EVENT_ENABLED(ip, DM_EVENT_ATTRIBUTE) &&
- !(flags & ATTR_DMI)) {
+ !(flags & XFS_ATTR_DMI)) {
(void) XFS_SEND_NAMESP(mp, DM_EVENT_ATTRIBUTE, ip, DM_RIGHT_NULL,
NULL, DM_RIGHT_NULL, NULL, NULL,
0, 0, AT_DELAY_FLAG(flags));
@@ -875,7 +713,7 @@ xfs_fsync(
return XFS_ERROR(EIO);
/* capture size updates in I/O completion before writing the inode. */
- error = filemap_fdatawait(vn_to_inode(XFS_ITOV(ip))->i_mapping);
+ error = filemap_fdatawait(VFS_I(ip)->i_mapping);
if (error)
return XFS_ERROR(error);
@@ -1321,7 +1159,6 @@ int
xfs_release(
xfs_inode_t *ip)
{
- bhv_vnode_t *vp = XFS_ITOV(ip);
xfs_mount_t *mp = ip->i_mount;
int error;
@@ -1356,13 +1193,13 @@ xfs_release(
* be exposed to that problem.
*/
truncated = xfs_iflags_test_and_clear(ip, XFS_ITRUNCATED);
- if (truncated && VN_DIRTY(vp) && ip->i_delayed_blks > 0)
+ if (truncated && VN_DIRTY(VFS_I(ip)) && ip->i_delayed_blks > 0)
xfs_flush_pages(ip, 0, -1, XFS_B_ASYNC, FI_NONE);
}
if (ip->i_d.di_nlink != 0) {
if ((((ip->i_d.di_mode & S_IFMT) == S_IFREG) &&
- ((ip->i_size > 0) || (VN_CACHED(vp) > 0 ||
+ ((ip->i_size > 0) || (VN_CACHED(VFS_I(ip)) > 0 ||
ip->i_delayed_blks > 0)) &&
(ip->i_df.if_flags & XFS_IFEXTENTS)) &&
(!(ip->i_d.di_flags &
@@ -1388,7 +1225,6 @@ int
xfs_inactive(
xfs_inode_t *ip)
{
- bhv_vnode_t *vp = XFS_ITOV(ip);
xfs_bmap_free_t free_list;
xfs_fsblock_t first_block;
int committed;
@@ -1403,7 +1239,7 @@ xfs_inactive(
* If the inode is already free, then there can be nothing
* to clean up here.
*/
- if (ip->i_d.di_mode == 0 || VN_BAD(vp)) {
+ if (ip->i_d.di_mode == 0 || VN_BAD(VFS_I(ip))) {
ASSERT(ip->i_df.if_real_bytes == 0);
ASSERT(ip->i_df.if_broot_bytes == 0);
return VN_INACTIVE_CACHE;
@@ -1433,7 +1269,7 @@ xfs_inactive(
if (ip->i_d.di_nlink != 0) {
if ((((ip->i_d.di_mode & S_IFMT) == S_IFREG) &&
- ((ip->i_size > 0) || (VN_CACHED(vp) > 0 ||
+ ((ip->i_size > 0) || (VN_CACHED(VFS_I(ip)) > 0 ||
ip->i_delayed_blks > 0)) &&
(ip->i_df.if_flags & XFS_IFEXTENTS) &&
(!(ip->i_d.di_flags &
@@ -1601,12 +1437,18 @@ xfs_inactive(
return VN_INACTIVE_CACHE;
}
-
+/*
+ * Lookups up an inode from "name". If ci_name is not NULL, then a CI match
+ * is allowed, otherwise it has to be an exact match. If a CI match is found,
+ * ci_name->name will point to a the actual name (caller must free) or
+ * will be set to NULL if an exact match is found.
+ */
int
xfs_lookup(
xfs_inode_t *dp,
struct xfs_name *name,
- xfs_inode_t **ipp)
+ xfs_inode_t **ipp,
+ struct xfs_name *ci_name)
{
xfs_ino_t inum;
int error;
@@ -1618,7 +1460,7 @@ xfs_lookup(
return XFS_ERROR(EIO);
lock_mode = xfs_ilock_map_shared(dp);
- error = xfs_dir_lookup(NULL, dp, name, &inum);
+ error = xfs_dir_lookup(NULL, dp, name, &inum, ci_name);
xfs_iunlock_map_shared(dp, lock_mode);
if (error)
@@ -1626,12 +1468,15 @@ xfs_lookup(
error = xfs_iget(dp->i_mount, NULL, inum, 0, 0, ipp, 0);
if (error)
- goto out;
+ goto out_free_name;
xfs_itrace_ref(*ipp);
return 0;
- out:
+out_free_name:
+ if (ci_name)
+ kmem_free(ci_name->name);
+out:
*ipp = NULL;
return error;
}
@@ -1688,7 +1533,7 @@ xfs_create(
* Make sure that we have allocated dquot(s) on disk.
*/
error = XFS_QM_DQVOPALLOC(mp, dp,
- current_fsuid(credp), current_fsgid(credp), prid,
+ current_fsuid(), current_fsgid(), prid,
XFS_QMOPT_QUOTALL|XFS_QMOPT_INHERIT, &udqp, &gdqp);
if (error)
goto std_return;
@@ -1860,111 +1705,6 @@ std_return:
}
#ifdef DEBUG
-/*
- * Some counters to see if (and how often) we are hitting some deadlock
- * prevention code paths.
- */
-
-int xfs_rm_locks;
-int xfs_rm_lock_delays;
-int xfs_rm_attempts;
-#endif
-
-/*
- * The following routine will lock the inodes associated with the
- * directory and the named entry in the directory. The locks are
- * acquired in increasing inode number.
- *
- * If the entry is "..", then only the directory is locked. The
- * vnode ref count will still include that from the .. entry in
- * this case.
- *
- * There is a deadlock we need to worry about. If the locked directory is
- * in the AIL, it might be blocking up the log. The next inode we lock
- * could be already locked by another thread waiting for log space (e.g
- * a permanent log reservation with a long running transaction (see
- * xfs_itruncate_finish)). To solve this, we must check if the directory
- * is in the ail and use lock_nowait. If we can't lock, we need to
- * drop the inode lock on the directory and try again. xfs_iunlock will
- * potentially push the tail if we were holding up the log.
- */
-STATIC int
-xfs_lock_dir_and_entry(
- xfs_inode_t *dp,
- xfs_inode_t *ip) /* inode of entry 'name' */
-{
- int attempts;
- xfs_ino_t e_inum;
- xfs_inode_t *ips[2];
- xfs_log_item_t *lp;
-
-#ifdef DEBUG
- xfs_rm_locks++;
-#endif
- attempts = 0;
-
-again:
- xfs_ilock(dp, XFS_ILOCK_EXCL | XFS_ILOCK_PARENT);
-
- e_inum = ip->i_ino;
-
- xfs_itrace_ref(ip);
-
- /*
- * We want to lock in increasing inum. Since we've already
- * acquired the lock on the directory, we may need to release
- * if if the inum of the entry turns out to be less.
- */
- if (e_inum > dp->i_ino) {
- /*
- * We are already in the right order, so just
- * lock on the inode of the entry.
- * We need to use nowait if dp is in the AIL.
- */
-
- lp = (xfs_log_item_t *)dp->i_itemp;
- if (lp && (lp->li_flags & XFS_LI_IN_AIL)) {
- if (!xfs_ilock_nowait(ip, XFS_ILOCK_EXCL)) {
- attempts++;
-#ifdef DEBUG
- xfs_rm_attempts++;
-#endif
-
- /*
- * Unlock dp and try again.
- * xfs_iunlock will try to push the tail
- * if the inode is in the AIL.
- */
-
- xfs_iunlock(dp, XFS_ILOCK_EXCL);
-
- if ((attempts % 5) == 0) {
- delay(1); /* Don't just spin the CPU */
-#ifdef DEBUG
- xfs_rm_lock_delays++;
-#endif
- }
- goto again;
- }
- } else {
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- }
- } else if (e_inum < dp->i_ino) {
- xfs_iunlock(dp, XFS_ILOCK_EXCL);
-
- ips[0] = ip;
- ips[1] = dp;
- xfs_lock_inodes(ips, 2, XFS_ILOCK_EXCL);
- }
- /* else e_inum == dp->i_ino */
- /* This can happen if we're asked to lock /x/..
- * the entry is "..", which is also the parent directory.
- */
-
- return 0;
-}
-
-#ifdef DEBUG
int xfs_locked_n;
int xfs_small_retries;
int xfs_middle_retries;
@@ -2098,12 +1838,52 @@ again:
#endif
}
-#ifdef DEBUG
-#define REMOVE_DEBUG_TRACE(x) {remove_which_error_return = (x);}
-int remove_which_error_return = 0;
-#else /* ! DEBUG */
-#define REMOVE_DEBUG_TRACE(x)
-#endif /* ! DEBUG */
+/*
+ * xfs_lock_two_inodes() can only be used to lock one type of lock
+ * at a time - the iolock or the ilock, but not both at once. If
+ * we lock both at once, lockdep will report false positives saying
+ * we have violated locking orders.
+ */
+void
+xfs_lock_two_inodes(
+ xfs_inode_t *ip0,
+ xfs_inode_t *ip1,
+ uint lock_mode)
+{
+ xfs_inode_t *temp;
+ int attempts = 0;
+ xfs_log_item_t *lp;
+
+ if (lock_mode & (XFS_IOLOCK_SHARED|XFS_IOLOCK_EXCL))
+ ASSERT((lock_mode & (XFS_ILOCK_SHARED|XFS_ILOCK_EXCL)) == 0);
+ ASSERT(ip0->i_ino != ip1->i_ino);
+
+ if (ip0->i_ino > ip1->i_ino) {
+ temp = ip0;
+ ip0 = ip1;
+ ip1 = temp;
+ }
+
+ again:
+ xfs_ilock(ip0, xfs_lock_inumorder(lock_mode, 0));
+
+ /*
+ * If the first lock we have locked is in the AIL, we must TRY to get
+ * the second lock. If we can't get it, we must release the first one
+ * and try again.
+ */
+ lp = (xfs_log_item_t *)ip0->i_itemp;
+ if (lp && (lp->li_flags & XFS_LI_IN_AIL)) {
+ if (!xfs_ilock_nowait(ip1, xfs_lock_inumorder(lock_mode, 1))) {
+ xfs_iunlock(ip0, lock_mode);
+ if ((++attempts % 5) == 0)
+ delay(1); /* Don't just spin the CPU */
+ goto again;
+ }
+ } else {
+ xfs_ilock(ip1, xfs_lock_inumorder(lock_mode, 1));
+ }
+}
int
xfs_remove(
@@ -2113,6 +1893,7 @@ xfs_remove(
{
xfs_mount_t *mp = dp->i_mount;
xfs_trans_t *tp = NULL;
+ int is_dir = S_ISDIR(ip->i_d.di_mode);
int error = 0;
xfs_bmap_free_t free_list;
xfs_fsblock_t first_block;
@@ -2120,8 +1901,10 @@ xfs_remove(
int committed;
int link_zero;
uint resblks;
+ uint log_count;
xfs_itrace_entry(dp);
+ xfs_itrace_entry(ip);
if (XFS_FORCED_SHUTDOWN(mp))
return XFS_ERROR(EIO);
@@ -2134,19 +1917,23 @@ xfs_remove(
return error;
}
- xfs_itrace_entry(ip);
- xfs_itrace_ref(ip);
-
error = XFS_QM_DQATTACH(mp, dp, 0);
- if (!error)
- error = XFS_QM_DQATTACH(mp, ip, 0);
- if (error) {
- REMOVE_DEBUG_TRACE(__LINE__);
+ if (error)
goto std_return;
- }
- tp = xfs_trans_alloc(mp, XFS_TRANS_REMOVE);
+ error = XFS_QM_DQATTACH(mp, ip, 0);
+ if (error)
+ goto std_return;
+
+ if (is_dir) {
+ tp = xfs_trans_alloc(mp, XFS_TRANS_RMDIR);
+ log_count = XFS_DEFAULT_LOG_COUNT;
+ } else {
+ tp = xfs_trans_alloc(mp, XFS_TRANS_REMOVE);
+ log_count = XFS_REMOVE_LOG_COUNT;
+ }
cancel_flags = XFS_TRANS_RELEASE_LOG_RES;
+
/*
* We try to get the real space reservation first,
* allowing for directory btree deletion(s) implying
@@ -2158,25 +1945,19 @@ xfs_remove(
*/
resblks = XFS_REMOVE_SPACE_RES(mp);
error = xfs_trans_reserve(tp, resblks, XFS_REMOVE_LOG_RES(mp), 0,
- XFS_TRANS_PERM_LOG_RES, XFS_REMOVE_LOG_COUNT);
+ XFS_TRANS_PERM_LOG_RES, log_count);
if (error == ENOSPC) {
resblks = 0;
error = xfs_trans_reserve(tp, 0, XFS_REMOVE_LOG_RES(mp), 0,
- XFS_TRANS_PERM_LOG_RES, XFS_REMOVE_LOG_COUNT);
+ XFS_TRANS_PERM_LOG_RES, log_count);
}
if (error) {
ASSERT(error != ENOSPC);
- REMOVE_DEBUG_TRACE(__LINE__);
- xfs_trans_cancel(tp, 0);
- return error;
+ cancel_flags = 0;
+ goto out_trans_cancel;
}
- error = xfs_lock_dir_and_entry(dp, ip);
- if (error) {
- REMOVE_DEBUG_TRACE(__LINE__);
- xfs_trans_cancel(tp, cancel_flags);
- goto std_return;
- }
+ xfs_lock_two_inodes(dp, ip, XFS_ILOCK_EXCL);
/*
* At this point, we've gotten both the directory and the entry
@@ -2189,46 +1970,83 @@ xfs_remove(
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
/*
- * Entry must exist since we did a lookup in xfs_lock_dir_and_entry.
+ * If we're removing a directory perform some additional validation.
*/
+ if (is_dir) {
+ ASSERT(ip->i_d.di_nlink >= 2);
+ if (ip->i_d.di_nlink != 2) {
+ error = XFS_ERROR(ENOTEMPTY);
+ goto out_trans_cancel;
+ }
+ if (!xfs_dir_isempty(ip)) {
+ error = XFS_ERROR(ENOTEMPTY);
+ goto out_trans_cancel;
+ }
+ }
+
XFS_BMAP_INIT(&free_list, &first_block);
error = xfs_dir_removename(tp, dp, name, ip->i_ino,
&first_block, &free_list, resblks);
if (error) {
ASSERT(error != ENOENT);
- REMOVE_DEBUG_TRACE(__LINE__);
- goto error1;
+ goto out_bmap_cancel;
}
xfs_ichgtime(dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
+ /*
+ * Bump the in memory generation count on the parent
+ * directory so that other can know that it has changed.
+ */
dp->i_gen++;
xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE);
- error = xfs_droplink(tp, ip);
- if (error) {
- REMOVE_DEBUG_TRACE(__LINE__);
- goto error1;
+ if (is_dir) {
+ /*
+ * Drop the link from ip's "..".
+ */
+ error = xfs_droplink(tp, dp);
+ if (error)
+ goto out_bmap_cancel;
+
+ /*
+ * Drop the link from dp to ip.
+ */
+ error = xfs_droplink(tp, ip);
+ if (error)
+ goto out_bmap_cancel;
+ } else {
+ /*
+ * When removing a non-directory we need to log the parent
+ * inode here for the i_gen update. For a directory this is
+ * done implicitly by the xfs_droplink call for the ".." entry.
+ */
+ xfs_trans_log_inode(tp, dp, XFS_ILOG_CORE);
}
- /* Determine if this is the last link while
+ /*
+ * Drop the "." link from ip to self.
+ */
+ error = xfs_droplink(tp, ip);
+ if (error)
+ goto out_bmap_cancel;
+
+ /*
+ * Determine if this is the last link while
* we are in the transaction.
*/
- link_zero = (ip)->i_d.di_nlink==0;
+ link_zero = (ip->i_d.di_nlink == 0);
/*
* If this is a synchronous mount, make sure that the
* remove transaction goes to disk before returning to
* the user.
*/
- if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) {
+ if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC))
xfs_trans_set_sync(tp);
- }
error = xfs_bmap_finish(&tp, &free_list, &committed);
- if (error) {
- REMOVE_DEBUG_TRACE(__LINE__);
- goto error_rele;
- }
+ if (error)
+ goto out_bmap_cancel;
error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
if (error)
@@ -2240,38 +2058,26 @@ xfs_remove(
* will get killed on last close in xfs_close() so we don't
* have to worry about that.
*/
- if (link_zero && xfs_inode_is_filestream(ip))
+ if (!is_dir && link_zero && xfs_inode_is_filestream(ip))
xfs_filestream_deassociate(ip);
xfs_itrace_exit(ip);
+ xfs_itrace_exit(dp);
-/* Fall through to std_return with error = 0 */
std_return:
if (DM_EVENT_ENABLED(dp, DM_EVENT_POSTREMOVE)) {
- (void) XFS_SEND_NAMESP(mp, DM_EVENT_POSTREMOVE,
- dp, DM_RIGHT_NULL,
- NULL, DM_RIGHT_NULL,
- name->name, NULL, ip->i_d.di_mode, error, 0);
+ XFS_SEND_NAMESP(mp, DM_EVENT_POSTREMOVE, dp, DM_RIGHT_NULL,
+ NULL, DM_RIGHT_NULL, name->name, NULL,
+ ip->i_d.di_mode, error, 0);
}
- return error;
- error1:
- xfs_bmap_cancel(&free_list);
- cancel_flags |= XFS_TRANS_ABORT;
- xfs_trans_cancel(tp, cancel_flags);
- goto std_return;
+ return error;
- error_rele:
- /*
- * In this case make sure to not release the inode until after
- * the current transaction is aborted. Releasing it beforehand
- * can cause us to go to xfs_inactive and start a recursive
- * transaction which can easily deadlock with the current one.
- */
+ out_bmap_cancel:
xfs_bmap_cancel(&free_list);
cancel_flags |= XFS_TRANS_ABORT;
+ out_trans_cancel:
xfs_trans_cancel(tp, cancel_flags);
-
goto std_return;
}
@@ -2283,7 +2089,6 @@ xfs_link(
{
xfs_mount_t *mp = tdp->i_mount;
xfs_trans_t *tp;
- xfs_inode_t *ips[2];
int error;
xfs_bmap_free_t free_list;
xfs_fsblock_t first_block;
@@ -2331,15 +2136,7 @@ xfs_link(
goto error_return;
}
- if (sip->i_ino < tdp->i_ino) {
- ips[0] = sip;
- ips[1] = tdp;
- } else {
- ips[0] = tdp;
- ips[1] = sip;
- }
-
- xfs_lock_inodes(ips, 2, XFS_ILOCK_EXCL);
+ xfs_lock_two_inodes(sip, tdp, XFS_ILOCK_EXCL);
/*
* Increment vnode ref counts since xfs_trans_commit &
@@ -2480,7 +2277,7 @@ xfs_mkdir(
* Make sure that we have allocated dquot(s) on disk.
*/
error = XFS_QM_DQVOPALLOC(mp, dp,
- current_fsuid(credp), current_fsgid(credp), prid,
+ current_fsuid(), current_fsgid(), prid,
XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp);
if (error)
goto std_return;
@@ -2638,186 +2435,6 @@ std_return:
}
int
-xfs_rmdir(
- xfs_inode_t *dp,
- struct xfs_name *name,
- xfs_inode_t *cdp)
-{
- xfs_mount_t *mp = dp->i_mount;
- xfs_trans_t *tp;
- int error;
- xfs_bmap_free_t free_list;
- xfs_fsblock_t first_block;
- int cancel_flags;
- int committed;
- int last_cdp_link;
- uint resblks;
-
- xfs_itrace_entry(dp);
-
- if (XFS_FORCED_SHUTDOWN(mp))
- return XFS_ERROR(EIO);
-
- if (DM_EVENT_ENABLED(dp, DM_EVENT_REMOVE)) {
- error = XFS_SEND_NAMESP(mp, DM_EVENT_REMOVE,
- dp, DM_RIGHT_NULL,
- NULL, DM_RIGHT_NULL, name->name,
- NULL, cdp->i_d.di_mode, 0, 0);
- if (error)
- return XFS_ERROR(error);
- }
-
- /*
- * Get the dquots for the inodes.
- */
- error = XFS_QM_DQATTACH(mp, dp, 0);
- if (!error)
- error = XFS_QM_DQATTACH(mp, cdp, 0);
- if (error) {
- REMOVE_DEBUG_TRACE(__LINE__);
- goto std_return;
- }
-
- tp = xfs_trans_alloc(mp, XFS_TRANS_RMDIR);
- cancel_flags = XFS_TRANS_RELEASE_LOG_RES;
- /*
- * We try to get the real space reservation first,
- * allowing for directory btree deletion(s) implying
- * possible bmap insert(s). If we can't get the space
- * reservation then we use 0 instead, and avoid the bmap
- * btree insert(s) in the directory code by, if the bmap
- * insert tries to happen, instead trimming the LAST
- * block from the directory.
- */
- resblks = XFS_REMOVE_SPACE_RES(mp);
- error = xfs_trans_reserve(tp, resblks, XFS_REMOVE_LOG_RES(mp), 0,
- XFS_TRANS_PERM_LOG_RES, XFS_DEFAULT_LOG_COUNT);
- if (error == ENOSPC) {
- resblks = 0;
- error = xfs_trans_reserve(tp, 0, XFS_REMOVE_LOG_RES(mp), 0,
- XFS_TRANS_PERM_LOG_RES, XFS_DEFAULT_LOG_COUNT);
- }
- if (error) {
- ASSERT(error != ENOSPC);
- cancel_flags = 0;
- goto error_return;
- }
- XFS_BMAP_INIT(&free_list, &first_block);
-
- /*
- * Now lock the child directory inode and the parent directory
- * inode in the proper order. This will take care of validating
- * that the directory entry for the child directory inode has
- * not changed while we were obtaining a log reservation.
- */
- error = xfs_lock_dir_and_entry(dp, cdp);
- if (error) {
- xfs_trans_cancel(tp, cancel_flags);
- goto std_return;
- }
-
- IHOLD(dp);
- xfs_trans_ijoin(tp, dp, XFS_ILOCK_EXCL);
-
- IHOLD(cdp);
- xfs_trans_ijoin(tp, cdp, XFS_ILOCK_EXCL);
-
- ASSERT(cdp->i_d.di_nlink >= 2);
- if (cdp->i_d.di_nlink != 2) {
- error = XFS_ERROR(ENOTEMPTY);
- goto error_return;
- }
- if (!xfs_dir_isempty(cdp)) {
- error = XFS_ERROR(ENOTEMPTY);
- goto error_return;
- }
-
- error = xfs_dir_removename(tp, dp, name, cdp->i_ino,
- &first_block, &free_list, resblks);
- if (error)
- goto error1;
-
- xfs_ichgtime(dp, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG);
-
- /*
- * Bump the in memory generation count on the parent
- * directory so that other can know that it has changed.
- */
- dp->i_gen++;
-
- /*
- * Drop the link from cdp's "..".
- */
- error = xfs_droplink(tp, dp);
- if (error) {
- goto error1;
- }
-
- /*
- * Drop the link from dp to cdp.
- */
- error = xfs_droplink(tp, cdp);
- if (error) {
- goto error1;
- }
-
- /*
- * Drop the "." link from cdp to self.
- */
- error = xfs_droplink(tp, cdp);
- if (error) {
- goto error1;
- }
-
- /* Determine these before committing transaction */
- last_cdp_link = (cdp)->i_d.di_nlink==0;
-
- /*
- * If this is a synchronous mount, make sure that the
- * rmdir transaction goes to disk before returning to
- * the user.
- */
- if (mp->m_flags & (XFS_MOUNT_WSYNC|XFS_MOUNT_DIRSYNC)) {
- xfs_trans_set_sync(tp);
- }
-
- error = xfs_bmap_finish (&tp, &free_list, &committed);
- if (error) {
- xfs_bmap_cancel(&free_list);
- xfs_trans_cancel(tp, (XFS_TRANS_RELEASE_LOG_RES |
- XFS_TRANS_ABORT));
- goto std_return;
- }
-
- error = xfs_trans_commit(tp, XFS_TRANS_RELEASE_LOG_RES);
- if (error) {
- goto std_return;
- }
-
-
- /* Fall through to std_return with error = 0 or the errno
- * from xfs_trans_commit. */
- std_return:
- if (DM_EVENT_ENABLED(dp, DM_EVENT_POSTREMOVE)) {
- (void) XFS_SEND_NAMESP(mp, DM_EVENT_POSTREMOVE,
- dp, DM_RIGHT_NULL,
- NULL, DM_RIGHT_NULL,
- name->name, NULL, cdp->i_d.di_mode,
- error, 0);
- }
- return error;
-
- error1:
- xfs_bmap_cancel(&free_list);
- cancel_flags |= XFS_TRANS_ABORT;
- /* FALLTHROUGH */
-
- error_return:
- xfs_trans_cancel(tp, cancel_flags);
- goto std_return;
-}
-
-int
xfs_symlink(
xfs_inode_t *dp,
struct xfs_name *link_name,
@@ -2886,7 +2503,7 @@ xfs_symlink(
* Make sure that we have allocated dquot(s) on disk.
*/
error = XFS_QM_DQVOPALLOC(mp, dp,
- current_fsuid(credp), current_fsgid(credp), prid,
+ current_fsuid(), current_fsgid(), prid,
XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp);
if (error)
goto std_return;
@@ -3181,14 +2798,13 @@ int
xfs_reclaim(
xfs_inode_t *ip)
{
- bhv_vnode_t *vp = XFS_ITOV(ip);
xfs_itrace_entry(ip);
- ASSERT(!VN_MAPPED(vp));
+ ASSERT(!VN_MAPPED(VFS_I(ip)));
/* bad inode, get out here ASAP */
- if (VN_BAD(vp)) {
+ if (VN_BAD(VFS_I(ip))) {
xfs_ireclaim(ip);
return 0;
}
@@ -3225,7 +2841,7 @@ xfs_reclaim(
XFS_MOUNT_ILOCK(mp);
spin_lock(&ip->i_flags_lock);
__xfs_iflags_set(ip, XFS_IRECLAIMABLE);
- vn_to_inode(vp)->i_private = NULL;
+ VFS_I(ip)->i_private = NULL;
ip->i_vnode = NULL;
spin_unlock(&ip->i_flags_lock);
list_add_tail(&ip->i_reclaim, &mp->m_del_inodes);
@@ -3241,8 +2857,7 @@ xfs_finish_reclaim(
int sync_mode)
{
xfs_perag_t *pag = xfs_get_perag(ip->i_mount, ip->i_ino);
- bhv_vnode_t *vp = XFS_ITOV_NULL(ip);
- int error;
+ struct inode *vp = VFS_I(ip);
if (vp && VN_BAD(vp))
goto reclaim;
@@ -3285,29 +2900,16 @@ xfs_finish_reclaim(
xfs_iflock(ip);
}
- if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) {
- if (ip->i_update_core ||
- ((ip->i_itemp != NULL) &&
- (ip->i_itemp->ili_format.ilf_fields != 0))) {
- error = xfs_iflush(ip, sync_mode);
- /*
- * If we hit an error, typically because of filesystem
- * shutdown, we don't need to let vn_reclaim to know
- * because we're gonna reclaim the inode anyway.
- */
- if (error) {
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- goto reclaim;
- }
- xfs_iflock(ip); /* synchronize with xfs_iflush_done */
- }
-
- ASSERT(ip->i_update_core == 0);
- ASSERT(ip->i_itemp == NULL ||
- ip->i_itemp->ili_format.ilf_fields == 0);
+ /*
+ * In the case of a forced shutdown we rely on xfs_iflush() to
+ * wait for the inode to be unpinned before returning an error.
+ */
+ if (xfs_iflush(ip, sync_mode) == 0) {
+ /* synchronize with xfs_iflush_done */
+ xfs_iflock(ip);
+ xfs_ifunlock(ip);
}
- xfs_ifunlock(ip);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
reclaim:
@@ -3418,7 +3020,7 @@ xfs_alloc_file_space(
/* Generate a DMAPI event if needed. */
if (alloc_type != 0 && offset < ip->i_size &&
- (attr_flags&ATTR_DMI) == 0 &&
+ (attr_flags & XFS_ATTR_DMI) == 0 &&
DM_EVENT_ENABLED(ip, DM_EVENT_WRITE)) {
xfs_off_t end_dmi_offset;
@@ -3532,7 +3134,7 @@ retry:
allocatesize_fsb -= allocated_fsb;
}
dmapi_enospc_check:
- if (error == ENOSPC && (attr_flags & ATTR_DMI) == 0 &&
+ if (error == ENOSPC && (attr_flags & XFS_ATTR_DMI) == 0 &&
DM_EVENT_ENABLED(ip, DM_EVENT_NOSPACE)) {
error = XFS_SEND_NAMESP(mp, DM_EVENT_NOSPACE,
ip, DM_RIGHT_NULL,
@@ -3558,6 +3160,13 @@ error1: /* Just cancel transaction */
/*
* Zero file bytes between startoff and endoff inclusive.
* The iolock is held exclusive and no blocks are buffered.
+ *
+ * This function is used by xfs_free_file_space() to zero
+ * partial blocks when the range to free is not block aligned.
+ * When unreserving space with boundaries that are not block
+ * aligned we round up the start and round down the end
+ * boundaries and then use this function to zero the parts of
+ * the blocks that got dropped during the rounding.
*/
STATIC int
xfs_zero_remaining_bytes(
@@ -3574,6 +3183,17 @@ xfs_zero_remaining_bytes(
int nimap;
int error = 0;
+ /*
+ * Avoid doing I/O beyond eof - it's not necessary
+ * since nothing can read beyond eof. The space will
+ * be zeroed when the file is extended anyway.
+ */
+ if (startoff >= ip->i_size)
+ return 0;
+
+ if (endoff > ip->i_size)
+ endoff = ip->i_size;
+
bp = xfs_buf_get_noaddr(mp->m_sb.sb_blocksize,
XFS_IS_REALTIME_INODE(ip) ?
mp->m_rtdev_targp : mp->m_ddev_targp);
@@ -3643,7 +3263,6 @@ xfs_free_file_space(
xfs_off_t len,
int attr_flags)
{
- bhv_vnode_t *vp;
int committed;
int done;
xfs_off_t end_dmi_offset;
@@ -3663,7 +3282,6 @@ xfs_free_file_space(
xfs_trans_t *tp;
int need_iolock = 1;
- vp = XFS_ITOV(ip);
mp = ip->i_mount;
xfs_itrace_entry(ip);
@@ -3679,7 +3297,7 @@ xfs_free_file_space(
end_dmi_offset = offset + len;
endoffset_fsb = XFS_B_TO_FSBT(mp, end_dmi_offset);
- if (offset < ip->i_size && (attr_flags & ATTR_DMI) == 0 &&
+ if (offset < ip->i_size && (attr_flags & XFS_ATTR_DMI) == 0 &&
DM_EVENT_ENABLED(ip, DM_EVENT_WRITE)) {
if (end_dmi_offset > ip->i_size)
end_dmi_offset = ip->i_size;
@@ -3690,7 +3308,7 @@ xfs_free_file_space(
return error;
}
- if (attr_flags & ATTR_NOLOCK)
+ if (attr_flags & XFS_ATTR_NOLOCK)
need_iolock = 0;
if (need_iolock) {
xfs_ilock(ip, XFS_IOLOCK_EXCL);
@@ -3700,7 +3318,7 @@ xfs_free_file_space(
rounding = max_t(uint, 1 << mp->m_sb.sb_blocklog, PAGE_CACHE_SIZE);
ioffset = offset & ~(rounding - 1);
- if (VN_CACHED(vp) != 0) {
+ if (VN_CACHED(VFS_I(ip)) != 0) {
xfs_inval_cached_trace(ip, ioffset, -1, ioffset, -1);
error = xfs_flushinval_pages(ip, ioffset, -1, FI_REMAPF_LOCKED);
if (error)
@@ -3867,7 +3485,7 @@ xfs_change_file_space(
xfs_off_t startoffset;
xfs_off_t llen;
xfs_trans_t *tp;
- bhv_vattr_t va;
+ struct iattr iattr;
xfs_itrace_entry(ip);
@@ -3941,10 +3559,10 @@ xfs_change_file_space(
break;
}
- va.va_mask = XFS_AT_SIZE;
- va.va_size = startoffset;
+ iattr.ia_valid = ATTR_SIZE;
+ iattr.ia_size = startoffset;
- error = xfs_setattr(ip, &va, attr_flags, credp);
+ error = xfs_setattr(ip, &iattr, attr_flags, credp);
if (error)
return error;
@@ -3974,7 +3592,7 @@ xfs_change_file_space(
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
xfs_trans_ihold(tp, ip);
- if ((attr_flags & ATTR_DMI) == 0) {
+ if ((attr_flags & XFS_ATTR_DMI) == 0) {
ip->i_d.di_mode &= ~S_ISUID;
/*
diff --git a/fs/xfs/xfs_vnodeops.h b/fs/xfs/xfs_vnodeops.h
index 57335ba4ce5..e932a96bec5 100644
--- a/fs/xfs/xfs_vnodeops.h
+++ b/fs/xfs/xfs_vnodeops.h
@@ -2,9 +2,9 @@
#define _XFS_VNODEOPS_H 1
struct attrlist_cursor_kern;
-struct bhv_vattr;
struct cred;
struct file;
+struct iattr;
struct inode;
struct iovec;
struct kiocb;
@@ -15,14 +15,18 @@ struct xfs_iomap;
int xfs_open(struct xfs_inode *ip);
-int xfs_setattr(struct xfs_inode *ip, struct bhv_vattr *vap, int flags,
+int xfs_setattr(struct xfs_inode *ip, struct iattr *vap, int flags,
struct cred *credp);
+#define XFS_ATTR_DMI 0x01 /* invocation from a DMI function */
+#define XFS_ATTR_NONBLOCK 0x02 /* return EAGAIN if operation would block */
+#define XFS_ATTR_NOLOCK 0x04 /* Don't grab any conflicting locks */
+
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,
- struct xfs_inode **ipp);
+ struct xfs_inode **ipp, struct xfs_name *ci_name);
int xfs_create(struct xfs_inode *dp, struct xfs_name *name, mode_t mode,
xfs_dev_t rdev, struct xfs_inode **ipp, struct cred *credp);
int xfs_remove(struct xfs_inode *dp, struct xfs_name *name,
@@ -31,8 +35,6 @@ int xfs_link(struct xfs_inode *tdp, struct xfs_inode *sip,
struct xfs_name *target_name);
int xfs_mkdir(struct xfs_inode *dp, struct xfs_name *dir_name,
mode_t mode, struct xfs_inode **ipp, struct cred *credp);
-int xfs_rmdir(struct xfs_inode *dp, struct xfs_name *name,
- struct xfs_inode *cdp);
int xfs_readdir(struct xfs_inode *dp, void *dirent, size_t bufsize,
xfs_off_t *offset, filldir_t filldir);
int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name,