summaryrefslogtreecommitdiff
path: root/fs/xfs
diff options
context:
space:
mode:
Diffstat (limited to 'fs/xfs')
-rw-r--r--fs/xfs/Makefile1
-rw-r--r--fs/xfs/linux-2.6/xfs_buf.c7
-rw-r--r--fs/xfs/linux-2.6/xfs_buf.h7
-rw-r--r--fs/xfs/linux-2.6/xfs_discard.c191
-rw-r--r--fs/xfs/linux-2.6/xfs_discard.h8
-rw-r--r--fs/xfs/linux-2.6/xfs_file.c587
-rw-r--r--fs/xfs/linux-2.6/xfs_ioctl.c23
-rw-r--r--fs/xfs/linux-2.6/xfs_iops.c57
-rw-r--r--fs/xfs/linux-2.6/xfs_super.c2
-rw-r--r--fs/xfs/linux-2.6/xfs_sync.c11
-rw-r--r--fs/xfs/linux-2.6/xfs_sysctl.c23
-rw-r--r--fs/xfs/linux-2.6/xfs_trace.h33
-rw-r--r--fs/xfs/quota/xfs_qm.c46
-rw-r--r--fs/xfs/support/debug.c112
-rw-r--r--fs/xfs/support/debug.h25
-rw-r--r--fs/xfs/xfs_alloc.c10
-rw-r--r--fs/xfs/xfs_alloc.h41
-rw-r--r--fs/xfs/xfs_bmap.c61
-rw-r--r--fs/xfs/xfs_buf_item.c163
-rw-r--r--fs/xfs/xfs_error.c31
-rw-r--r--fs/xfs/xfs_error.h18
-rw-r--r--fs/xfs/xfs_extfree_item.c3
-rw-r--r--fs/xfs/xfs_fsops.c10
-rw-r--r--fs/xfs/xfs_fsops.h2
-rw-r--r--fs/xfs/xfs_iomap.c7
-rw-r--r--fs/xfs/xfs_log.c2
-rw-r--r--fs/xfs/xfs_log.h2
-rw-r--r--fs/xfs/xfs_log_cil.c15
-rw-r--r--fs/xfs/xfs_log_recover.c2
-rw-r--r--fs/xfs/xfs_trans.c43
30 files changed, 935 insertions, 608 deletions
diff --git a/fs/xfs/Makefile b/fs/xfs/Makefile
index 0dce969d6cad..faca44997099 100644
--- a/fs/xfs/Makefile
+++ b/fs/xfs/Makefile
@@ -98,6 +98,7 @@ xfs-y += $(addprefix $(XFS_LINUX)/, \
kmem.o \
xfs_aops.o \
xfs_buf.o \
+ xfs_discard.o \
xfs_export.o \
xfs_file.o \
xfs_fs_subr.o \
diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index 92f1f2acc6ab..ac1c7e8378dd 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -896,7 +896,6 @@ xfs_buf_rele(
trace_xfs_buf_rele(bp, _RET_IP_);
if (!pag) {
- ASSERT(!bp->b_relse);
ASSERT(list_empty(&bp->b_lru));
ASSERT(RB_EMPTY_NODE(&bp->b_rbnode));
if (atomic_dec_and_test(&bp->b_hold))
@@ -908,11 +907,7 @@ xfs_buf_rele(
ASSERT(atomic_read(&bp->b_hold) > 0);
if (atomic_dec_and_lock(&bp->b_hold, &pag->pag_buf_lock)) {
- if (bp->b_relse) {
- atomic_inc(&bp->b_hold);
- spin_unlock(&pag->pag_buf_lock);
- bp->b_relse(bp);
- } else if (!(bp->b_flags & XBF_STALE) &&
+ if (!(bp->b_flags & XBF_STALE) &&
atomic_read(&bp->b_lru_ref)) {
xfs_buf_lru_add(bp);
spin_unlock(&pag->pag_buf_lock);
diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h
index a76c2428faff..cbe65950e524 100644
--- a/fs/xfs/linux-2.6/xfs_buf.h
+++ b/fs/xfs/linux-2.6/xfs_buf.h
@@ -152,8 +152,6 @@ typedef struct xfs_buftarg {
struct xfs_buf;
typedef void (*xfs_buf_iodone_t)(struct xfs_buf *);
-typedef void (*xfs_buf_relse_t)(struct xfs_buf *);
-typedef int (*xfs_buf_bdstrat_t)(struct xfs_buf *);
#define XB_PAGES 2
@@ -183,7 +181,6 @@ typedef struct xfs_buf {
void *b_addr; /* virtual address of buffer */
struct work_struct b_iodone_work;
xfs_buf_iodone_t b_iodone; /* I/O completion function */
- xfs_buf_relse_t b_relse; /* releasing function */
struct completion b_iowait; /* queue for I/O waiters */
void *b_fspriv;
void *b_fspriv2;
@@ -323,7 +320,6 @@ void xfs_buf_stale(struct xfs_buf *bp);
#define XFS_BUF_FSPRIVATE2(bp, type) ((type)(bp)->b_fspriv2)
#define XFS_BUF_SET_FSPRIVATE2(bp, val) ((bp)->b_fspriv2 = (void*)(val))
#define XFS_BUF_SET_START(bp) do { } while (0)
-#define XFS_BUF_SET_BRELSE_FUNC(bp, func) ((bp)->b_relse = (func))
#define XFS_BUF_PTR(bp) (xfs_caddr_t)((bp)->b_addr)
#define XFS_BUF_SET_PTR(bp, val, cnt) xfs_buf_associate_memory(bp, val, cnt)
@@ -360,8 +356,7 @@ xfs_buf_set_ref(
static inline void xfs_buf_relse(xfs_buf_t *bp)
{
- if (!bp->b_relse)
- xfs_buf_unlock(bp);
+ xfs_buf_unlock(bp);
xfs_buf_rele(bp);
}
diff --git a/fs/xfs/linux-2.6/xfs_discard.c b/fs/xfs/linux-2.6/xfs_discard.c
new file mode 100644
index 000000000000..05201ae719e5
--- /dev/null
+++ b/fs/xfs/linux-2.6/xfs_discard.c
@@ -0,0 +1,191 @@
+/*
+ * Copyright (C) 2010 Red Hat, Inc.
+ * All Rights Reserved.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write the Free Software Foundation,
+ * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+#include "xfs.h"
+#include "xfs_sb.h"
+#include "xfs_inum.h"
+#include "xfs_log.h"
+#include "xfs_ag.h"
+#include "xfs_mount.h"
+#include "xfs_quota.h"
+#include "xfs_trans.h"
+#include "xfs_alloc_btree.h"
+#include "xfs_bmap_btree.h"
+#include "xfs_ialloc_btree.h"
+#include "xfs_btree.h"
+#include "xfs_inode.h"
+#include "xfs_alloc.h"
+#include "xfs_error.h"
+#include "xfs_discard.h"
+#include "xfs_trace.h"
+
+STATIC int
+xfs_trim_extents(
+ struct xfs_mount *mp,
+ xfs_agnumber_t agno,
+ xfs_fsblock_t start,
+ xfs_fsblock_t len,
+ xfs_fsblock_t minlen,
+ __uint64_t *blocks_trimmed)
+{
+ struct block_device *bdev = mp->m_ddev_targp->bt_bdev;
+ struct xfs_btree_cur *cur;
+ struct xfs_buf *agbp;
+ struct xfs_perag *pag;
+ int error;
+ int i;
+
+ pag = xfs_perag_get(mp, agno);
+
+ error = xfs_alloc_read_agf(mp, NULL, agno, 0, &agbp);
+ if (error || !agbp)
+ goto out_put_perag;
+
+ cur = xfs_allocbt_init_cursor(mp, NULL, agbp, agno, XFS_BTNUM_CNT);
+
+ /*
+ * Force out the log. This means any transactions that might have freed
+ * space before we took the AGF buffer lock are now on disk, and the
+ * volatile disk cache is flushed.
+ */
+ xfs_log_force(mp, XFS_LOG_SYNC);
+
+ /*
+ * Look up the longest btree in the AGF and start with it.
+ */
+ error = xfs_alloc_lookup_le(cur, 0,
+ XFS_BUF_TO_AGF(agbp)->agf_longest, &i);
+ if (error)
+ goto out_del_cursor;
+
+ /*
+ * Loop until we are done with all extents that are large
+ * enough to be worth discarding.
+ */
+ while (i) {
+ xfs_agblock_t fbno;
+ xfs_extlen_t flen;
+
+ error = xfs_alloc_get_rec(cur, &fbno, &flen, &i);
+ if (error)
+ goto out_del_cursor;
+ XFS_WANT_CORRUPTED_GOTO(i == 1, out_del_cursor);
+ ASSERT(flen <= XFS_BUF_TO_AGF(agbp)->agf_longest);
+
+ /*
+ * Too small? Give up.
+ */
+ if (flen < minlen) {
+ trace_xfs_discard_toosmall(mp, agno, fbno, flen);
+ goto out_del_cursor;
+ }
+
+ /*
+ * If the extent is entirely outside of the range we are
+ * supposed to discard skip it. Do not bother to trim
+ * down partially overlapping ranges for now.
+ */
+ if (XFS_AGB_TO_FSB(mp, agno, fbno) + flen < start ||
+ XFS_AGB_TO_FSB(mp, agno, fbno) >= start + len) {
+ trace_xfs_discard_exclude(mp, agno, fbno, flen);
+ goto next_extent;
+ }
+
+ /*
+ * If any blocks in the range are still busy, skip the
+ * discard and try again the next time.
+ */
+ if (xfs_alloc_busy_search(mp, agno, fbno, flen)) {
+ trace_xfs_discard_busy(mp, agno, fbno, flen);
+ goto next_extent;
+ }
+
+ trace_xfs_discard_extent(mp, agno, fbno, flen);
+ error = -blkdev_issue_discard(bdev,
+ XFS_AGB_TO_DADDR(mp, agno, fbno),
+ XFS_FSB_TO_BB(mp, flen),
+ GFP_NOFS, 0);
+ if (error)
+ goto out_del_cursor;
+ *blocks_trimmed += flen;
+
+next_extent:
+ error = xfs_btree_decrement(cur, 0, &i);
+ if (error)
+ goto out_del_cursor;
+ }
+
+out_del_cursor:
+ xfs_btree_del_cursor(cur, error ? XFS_BTREE_ERROR : XFS_BTREE_NOERROR);
+ xfs_buf_relse(agbp);
+out_put_perag:
+ xfs_perag_put(pag);
+ return error;
+}
+
+int
+xfs_ioc_trim(
+ struct xfs_mount *mp,
+ struct fstrim_range __user *urange)
+{
+ struct request_queue *q = mp->m_ddev_targp->bt_bdev->bd_disk->queue;
+ unsigned int granularity = q->limits.discard_granularity;
+ struct fstrim_range range;
+ xfs_fsblock_t start, len, minlen;
+ xfs_agnumber_t start_agno, end_agno, agno;
+ __uint64_t blocks_trimmed = 0;
+ int error, last_error = 0;
+
+ if (!capable(CAP_SYS_ADMIN))
+ return -XFS_ERROR(EPERM);
+ if (copy_from_user(&range, urange, sizeof(range)))
+ return -XFS_ERROR(EFAULT);
+
+ /*
+ * Truncating down the len isn't actually quite correct, but using
+ * XFS_B_TO_FSB would mean we trivially get overflows for values
+ * of ULLONG_MAX or slightly lower. And ULLONG_MAX is the default
+ * used by the fstrim application. In the end it really doesn't
+ * matter as trimming blocks is an advisory interface.
+ */
+ start = XFS_B_TO_FSBT(mp, range.start);
+ len = XFS_B_TO_FSBT(mp, range.len);
+ minlen = XFS_B_TO_FSB(mp, max_t(u64, granularity, range.minlen));
+
+ start_agno = XFS_FSB_TO_AGNO(mp, start);
+ if (start_agno >= mp->m_sb.sb_agcount)
+ return -XFS_ERROR(EINVAL);
+
+ end_agno = XFS_FSB_TO_AGNO(mp, start + len);
+ if (end_agno >= mp->m_sb.sb_agcount)
+ end_agno = mp->m_sb.sb_agcount - 1;
+
+ for (agno = start_agno; agno <= end_agno; agno++) {
+ error = -xfs_trim_extents(mp, agno, start, len, minlen,
+ &blocks_trimmed);
+ if (error)
+ last_error = error;
+ }
+
+ if (last_error)
+ return last_error;
+
+ range.len = XFS_FSB_TO_B(mp, blocks_trimmed);
+ if (copy_to_user(urange, &range, sizeof(range)))
+ return -XFS_ERROR(EFAULT);
+ return 0;
+}
diff --git a/fs/xfs/linux-2.6/xfs_discard.h b/fs/xfs/linux-2.6/xfs_discard.h
new file mode 100644
index 000000000000..e82b6dd3e127
--- /dev/null
+++ b/fs/xfs/linux-2.6/xfs_discard.h
@@ -0,0 +1,8 @@
+#ifndef XFS_DISCARD_H
+#define XFS_DISCARD_H 1
+
+struct fstrim_range;
+
+extern int xfs_ioc_trim(struct xfs_mount *, struct fstrim_range __user *);
+
+#endif /* XFS_DISCARD_H */
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index ba8ad422a165..a55c1b46b219 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -37,10 +37,45 @@
#include "xfs_trace.h"
#include <linux/dcache.h>
+#include <linux/falloc.h>
static const struct vm_operations_struct xfs_file_vm_ops;
/*
+ * Locking primitives for read and write IO paths to ensure we consistently use
+ * and order the inode->i_mutex, ip->i_lock and ip->i_iolock.
+ */
+static inline void
+xfs_rw_ilock(
+ struct xfs_inode *ip,
+ int type)
+{
+ if (type & XFS_IOLOCK_EXCL)
+ mutex_lock(&VFS_I(ip)->i_mutex);
+ xfs_ilock(ip, type);
+}
+
+static inline void
+xfs_rw_iunlock(
+ struct xfs_inode *ip,
+ int type)
+{
+ xfs_iunlock(ip, type);
+ if (type & XFS_IOLOCK_EXCL)
+ mutex_unlock(&VFS_I(ip)->i_mutex);
+}
+
+static inline void
+xfs_rw_ilock_demote(
+ struct xfs_inode *ip,
+ int type)
+{
+ xfs_ilock_demote(ip, type);
+ if (type & XFS_IOLOCK_EXCL)
+ mutex_unlock(&VFS_I(ip)->i_mutex);
+}
+
+/*
* xfs_iozero
*
* xfs_iozero clears the specified range of buffer supplied,
@@ -262,22 +297,21 @@ xfs_file_aio_read(
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
- if (unlikely(ioflags & IO_ISDIRECT))
- mutex_lock(&inode->i_mutex);
- xfs_ilock(ip, XFS_IOLOCK_SHARED);
-
if (unlikely(ioflags & IO_ISDIRECT)) {
+ xfs_rw_ilock(ip, XFS_IOLOCK_EXCL);
+
if (inode->i_mapping->nrpages) {
ret = -xfs_flushinval_pages(ip,
(iocb->ki_pos & PAGE_CACHE_MASK),
-1, FI_REMAPF_LOCKED);
+ if (ret) {
+ xfs_rw_iunlock(ip, XFS_IOLOCK_EXCL);
+ return ret;
+ }
}
- mutex_unlock(&inode->i_mutex);
- if (ret) {
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- return ret;
- }
- }
+ xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
+ } else
+ xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
trace_xfs_file_read(ip, size, iocb->ki_pos, ioflags);
@@ -285,7 +319,7 @@ xfs_file_aio_read(
if (ret > 0)
XFS_STATS_ADD(xs_read_bytes, ret);
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
return ret;
}
@@ -309,7 +343,7 @@ xfs_file_splice_read(
if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO;
- xfs_ilock(ip, XFS_IOLOCK_SHARED);
+ xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
@@ -317,10 +351,61 @@ xfs_file_splice_read(
if (ret > 0)
XFS_STATS_ADD(xs_read_bytes, ret);
- xfs_iunlock(ip, XFS_IOLOCK_SHARED);
+ xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
return ret;
}
+STATIC void
+xfs_aio_write_isize_update(
+ struct inode *inode,
+ loff_t *ppos,
+ ssize_t bytes_written)
+{
+ struct xfs_inode *ip = XFS_I(inode);
+ xfs_fsize_t isize = i_size_read(inode);
+
+ if (bytes_written > 0)
+ XFS_STATS_ADD(xs_write_bytes, bytes_written);
+
+ if (unlikely(bytes_written < 0 && bytes_written != -EFAULT &&
+ *ppos > isize))
+ *ppos = isize;
+
+ if (*ppos > ip->i_size) {
+ xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
+ if (*ppos > ip->i_size)
+ ip->i_size = *ppos;
+ xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
+ }
+}
+
+/*
+ * If this was a direct or synchronous I/O that failed (such as ENOSPC) then
+ * part of the I/O may have been written to disk before the error occured. In
+ * this case the on-disk file size may have been adjusted beyond the in-memory
+ * file size and now needs to be truncated back.
+ */
+STATIC void
+xfs_aio_write_newsize_update(
+ struct xfs_inode *ip)
+{
+ if (ip->i_new_size) {
+ xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
+ ip->i_new_size = 0;
+ if (ip->i_d.di_size > ip->i_size)
+ ip->i_d.di_size = ip->i_size;
+ xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
+ }
+}
+
+/*
+ * xfs_file_splice_write() does not use xfs_rw_ilock() because
+ * generic_file_splice_write() takes the i_mutex itself. This, in theory,
+ * couuld cause lock inversions between the aio_write path and the splice path
+ * if someone is doing concurrent splice(2) based writes and write(2) based
+ * writes to the same inode. The only real way to fix this is to re-implement
+ * the generic code here with correct locking orders.
+ */
STATIC ssize_t
xfs_file_splice_write(
struct pipe_inode_info *pipe,
@@ -331,7 +416,7 @@ xfs_file_splice_write(
{
struct inode *inode = outfilp->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
- xfs_fsize_t isize, new_size;
+ xfs_fsize_t new_size;
int ioflags = 0;
ssize_t ret;
@@ -355,27 +440,9 @@ xfs_file_splice_write(
trace_xfs_file_splice_write(ip, count, *ppos, ioflags);
ret = generic_file_splice_write(pipe, outfilp, ppos, count, flags);
- if (ret > 0)
- XFS_STATS_ADD(xs_write_bytes, ret);
-
- isize = i_size_read(inode);
- if (unlikely(ret < 0 && ret != -EFAULT && *ppos > isize))
- *ppos = isize;
-
- if (*ppos > ip->i_size) {
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- if (*ppos > ip->i_size)
- ip->i_size = *ppos;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
- if (ip->i_new_size) {
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- ip->i_new_size = 0;
- if (ip->i_d.di_size > ip->i_size)
- ip->i_d.di_size = ip->i_size;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- }
+ xfs_aio_write_isize_update(inode, ppos, ret);
+ xfs_aio_write_newsize_update(ip);
xfs_iunlock(ip, XFS_IOLOCK_EXCL);
return ret;
}
@@ -562,247 +629,314 @@ out_lock:
return error;
}
+/*
+ * Common pre-write limit and setup checks.
+ *
+ * Returns with iolock held according to @iolock.
+ */
STATIC ssize_t
-xfs_file_aio_write(
- struct kiocb *iocb,
- const struct iovec *iovp,
- unsigned long nr_segs,
- loff_t pos)
+xfs_file_aio_write_checks(
+ struct file *file,
+ loff_t *pos,
+ size_t *count,
+ int *iolock)
{
- struct file *file = iocb->ki_filp;
- struct address_space *mapping = file->f_mapping;
- struct inode *inode = mapping->host;
+ struct inode *inode = file->f_mapping->host;
struct xfs_inode *ip = XFS_I(inode);
- struct xfs_mount *mp = ip->i_mount;
- ssize_t ret = 0, error = 0;
- int ioflags = 0;
- xfs_fsize_t isize, new_size;
- int iolock;
- size_t ocount = 0, count;
- int need_i_mutex;
+ xfs_fsize_t new_size;
+ int error = 0;
- XFS_STATS_INC(xs_write_calls);
+ error = generic_write_checks(file, pos, count, S_ISBLK(inode->i_mode));
+ if (error) {
+ xfs_rw_iunlock(ip, XFS_ILOCK_EXCL | *iolock);
+ *iolock = 0;
+ return error;
+ }
- BUG_ON(iocb->ki_pos != pos);
+ new_size = *pos + *count;
+ if (new_size > ip->i_size)
+ ip->i_new_size = new_size;
- if (unlikely(file->f_flags & O_DIRECT))
- ioflags |= IO_ISDIRECT;
- if (file->f_mode & FMODE_NOCMTIME)
- ioflags |= IO_INVIS;
+ if (likely(!(file->f_mode & FMODE_NOCMTIME)))
+ file_update_time(file);
+
+ /*
+ * If the offset is beyond the size of the file, we need to zero any
+ * blocks that fall between the existing EOF and the start of this
+ * write.
+ */
+ if (*pos > ip->i_size)
+ error = -xfs_zero_eof(ip, *pos, ip->i_size);
- error = generic_segment_checks(iovp, &nr_segs, &ocount, VERIFY_READ);
+ xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
return error;
- count = ocount;
- if (count == 0)
- return 0;
-
- xfs_wait_for_freeze(mp, SB_FREEZE_WRITE);
+ /*
+ * If we're writing the file then make sure to clear the setuid and
+ * setgid bits if the process is not being run by root. This keeps
+ * people from modifying setuid and setgid binaries.
+ */
+ return file_remove_suid(file);
- if (XFS_FORCED_SHUTDOWN(mp))
- return -EIO;
+}
-relock:
- if (ioflags & IO_ISDIRECT) {
- iolock = XFS_IOLOCK_SHARED;
- need_i_mutex = 0;
- } else {
- iolock = XFS_IOLOCK_EXCL;
- need_i_mutex = 1;
- mutex_lock(&inode->i_mutex);
+/*
+ * xfs_file_dio_aio_write - handle direct IO writes
+ *
+ * Lock the inode appropriately to prepare for and issue a direct IO write.
+ * By separating it from the buffered write path we remove all the tricky to
+ * follow locking changes and looping.
+ *
+ * If there are cached pages or we're extending the file, we need IOLOCK_EXCL
+ * until we're sure the bytes at the new EOF have been zeroed and/or the cached
+ * pages are flushed out.
+ *
+ * In most cases the direct IO writes will be done holding IOLOCK_SHARED
+ * allowing them to be done in parallel with reads and other direct IO writes.
+ * However, if the IO is not aligned to filesystem blocks, the direct IO layer
+ * needs to do sub-block zeroing and that requires serialisation against other
+ * direct IOs to the same block. In this case we need to serialise the
+ * submission of the unaligned IOs so that we don't get racing block zeroing in
+ * the dio layer. To avoid the problem with aio, we also need to wait for
+ * outstanding IOs to complete so that unwritten extent conversion is completed
+ * before we try to map the overlapping block. This is currently implemented by
+ * hitting it with a big hammer (i.e. xfs_ioend_wait()).
+ *
+ * Returns with locks held indicated by @iolock and errors indicated by
+ * negative return values.
+ */
+STATIC ssize_t
+xfs_file_dio_aio_write(
+ struct kiocb *iocb,
+ const struct iovec *iovp,
+ unsigned long nr_segs,
+ loff_t pos,
+ size_t ocount,
+ int *iolock)
+{
+ struct file *file = iocb->ki_filp;
+ struct address_space *mapping = file->f_mapping;
+ struct inode *inode = mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ struct xfs_mount *mp = ip->i_mount;
+ ssize_t ret = 0;
+ size_t count = ocount;
+ int unaligned_io = 0;
+ struct xfs_buftarg *target = XFS_IS_REALTIME_INODE(ip) ?
+ mp->m_rtdev_targp : mp->m_ddev_targp;
+
+ *iolock = 0;
+ if ((pos & target->bt_smask) || (count & target->bt_smask))
+ return -XFS_ERROR(EINVAL);
+
+ if ((pos & mp->m_blockmask) || ((pos + count) & mp->m_blockmask))
+ unaligned_io = 1;
+
+ if (unaligned_io || mapping->nrpages || pos > ip->i_size)
+ *iolock = XFS_IOLOCK_EXCL;
+ else
+ *iolock = XFS_IOLOCK_SHARED;
+ xfs_rw_ilock(ip, XFS_ILOCK_EXCL | *iolock);
+
+ ret = xfs_file_aio_write_checks(file, &pos, &count, iolock);
+ if (ret)
+ return ret;
+
+ if (mapping->nrpages) {
+ WARN_ON(*iolock != XFS_IOLOCK_EXCL);
+ ret = -xfs_flushinval_pages(ip, (pos & PAGE_CACHE_MASK), -1,
+ FI_REMAPF_LOCKED);
+ if (ret)
+ return ret;
}
- xfs_ilock(ip, XFS_ILOCK_EXCL|iolock);
-
-start:
- error = -generic_write_checks(file, &pos, &count,
- S_ISBLK(inode->i_mode));
- if (error) {
- xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock);
- goto out_unlock_mutex;
+ /*
+ * If we are doing unaligned IO, wait for all other IO to drain,
+ * otherwise demote the lock if we had to flush cached pages
+ */
+ if (unaligned_io)
+ xfs_ioend_wait(ip);
+ else if (*iolock == XFS_IOLOCK_EXCL) {
+ xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
+ *iolock = XFS_IOLOCK_SHARED;
}
- if (ioflags & IO_ISDIRECT) {
- xfs_buftarg_t *target =
- XFS_IS_REALTIME_INODE(ip) ?
- mp->m_rtdev_targp : mp->m_ddev_targp;
+ trace_xfs_file_direct_write(ip, count, iocb->ki_pos, 0);
+ ret = generic_file_direct_write(iocb, iovp,
+ &nr_segs, pos, &iocb->ki_pos, count, ocount);
- if ((pos & target->bt_smask) || (count & target->bt_smask)) {
- xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock);
- return XFS_ERROR(-EINVAL);
- }
+ /* No fallback to buffered IO on errors for XFS. */
+ ASSERT(ret < 0 || ret == count);
+ return ret;
+}
- if (!need_i_mutex && (mapping->nrpages || pos > ip->i_size)) {
- xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock);
- iolock = XFS_IOLOCK_EXCL;
- need_i_mutex = 1;
- mutex_lock(&inode->i_mutex);
- xfs_ilock(ip, XFS_ILOCK_EXCL|iolock);
- goto start;
- }
- }
+STATIC ssize_t
+xfs_file_buffered_aio_write(
+ struct kiocb *iocb,
+ const struct iovec *iovp,
+ unsigned long nr_segs,
+ loff_t pos,
+ size_t ocount,
+ int *iolock)
+{
+ struct file *file = iocb->ki_filp;
+ struct address_space *mapping = file->f_mapping;
+ struct inode *inode = mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ ssize_t ret;
+ int enospc = 0;
+ size_t count = ocount;
- new_size = pos + count;
- if (new_size > ip->i_size)
- ip->i_new_size = new_size;
+ *iolock = XFS_IOLOCK_EXCL;
+ xfs_rw_ilock(ip, XFS_ILOCK_EXCL | *iolock);
- if (likely(!(ioflags & IO_INVIS)))
- file_update_time(file);
+ ret = xfs_file_aio_write_checks(file, &pos, &count, iolock);
+ if (ret)
+ return ret;
+ /* We can write back this queue in page reclaim */
+ current->backing_dev_info = mapping->backing_dev_info;
+
+write_retry:
+ trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, 0);
+ ret = generic_file_buffered_write(iocb, iovp, nr_segs,
+ pos, &iocb->ki_pos, count, ret);
/*
- * If the offset is beyond the size of the file, we have a couple
- * of things to do. First, if there is already space allocated
- * we need to either create holes or zero the disk or ...
- *
- * If there is a page where the previous size lands, we need
- * to zero it out up to the new size.
+ * if we just got an ENOSPC, flush the inode now we aren't holding any
+ * page locks and retry *once*
*/
-
- if (pos > ip->i_size) {
- error = xfs_zero_eof(ip, pos, ip->i_size);
- if (error) {
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- goto out_unlock_internal;
- }
+ if (ret == -ENOSPC && !enospc) {
+ ret = -xfs_flush_pages(ip, 0, -1, 0, FI_NONE);
+ if (ret)
+ return ret;
+ enospc = 1;
+ goto write_retry;
}
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ current->backing_dev_info = NULL;
+ return ret;
+}
- /*
- * If we're writing the file then make sure to clear the
- * setuid and setgid bits if the process is not being run
- * by root. This keeps people from modifying setuid and
- * setgid binaries.
- */
- error = -file_remove_suid(file);
- if (unlikely(error))
- goto out_unlock_internal;
+STATIC ssize_t
+xfs_file_aio_write(
+ struct kiocb *iocb,
+ const struct iovec *iovp,
+ unsigned long nr_segs,
+ loff_t pos)
+{
+ struct file *file = iocb->ki_filp;
+ struct address_space *mapping = file->f_mapping;
+ struct inode *inode = mapping->host;
+ struct xfs_inode *ip = XFS_I(inode);
+ ssize_t ret;
+ int iolock;
+ size_t ocount = 0;
- /* We can write back this queue in page reclaim */
- current->backing_dev_info = mapping->backing_dev_info;
+ XFS_STATS_INC(xs_write_calls);
- if ((ioflags & IO_ISDIRECT)) {
- if (mapping->nrpages) {
- WARN_ON(need_i_mutex == 0);
- error = xfs_flushinval_pages(ip,
- (pos & PAGE_CACHE_MASK),
- -1, FI_REMAPF_LOCKED);
- if (error)
- goto out_unlock_internal;
- }
+ BUG_ON(iocb->ki_pos != pos);
- if (need_i_mutex) {
- /* demote the lock now the cached pages are gone */
- xfs_ilock_demote(ip, XFS_IOLOCK_EXCL);
- mutex_unlock(&inode->i_mutex);
+ ret = generic_segment_checks(iovp, &nr_segs, &ocount, VERIFY_READ);
+ if (ret)
+ return ret;
- iolock = XFS_IOLOCK_SHARED;
- need_i_mutex = 0;
- }
+ if (ocount == 0)
+ return 0;
- trace_xfs_file_direct_write(ip, count, iocb->ki_pos, ioflags);
- ret = generic_file_direct_write(iocb, iovp,
- &nr_segs, pos, &iocb->ki_pos, count, ocount);
+ xfs_wait_for_freeze(ip->i_mount, SB_FREEZE_WRITE);
- /*
- * direct-io write to a hole: fall through to buffered I/O
- * for completing the rest of the request.
- */
- if (ret >= 0 && ret != count) {
- XFS_STATS_ADD(xs_write_bytes, ret);
+ if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+ return -EIO;
- pos += ret;
- count -= ret;
+ if (unlikely(file->f_flags & O_DIRECT))
+ ret = xfs_file_dio_aio_write(iocb, iovp, nr_segs, pos,
+ ocount, &iolock);
+ else
+ ret = xfs_file_buffered_aio_write(iocb, iovp, nr_segs, pos,
+ ocount, &iolock);
- ioflags &= ~IO_ISDIRECT;
- xfs_iunlock(ip, iolock);
- goto relock;
- }
- } else {
- int enospc = 0;
- ssize_t ret2 = 0;
+ xfs_aio_write_isize_update(inode, &iocb->ki_pos, ret);
-write_retry:
- trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, ioflags);
- ret2 = generic_file_buffered_write(iocb, iovp, nr_segs,
- pos, &iocb->ki_pos, count, ret);
- /*
- * if we just got an ENOSPC, flush the inode now we
- * aren't holding any page locks and retry *once*
- */
- if (ret2 == -ENOSPC && !enospc) {
- error = xfs_flush_pages(ip, 0, -1, 0, FI_NONE);
- if (error)
- goto out_unlock_internal;
- enospc = 1;
- goto write_retry;
- }
- ret = ret2;
- }
+ if (ret <= 0)
+ goto out_unlock;
- current->backing_dev_info = NULL;
+ /* Handle various SYNC-type writes */
+ if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
+ loff_t end = pos + ret - 1;
+ int error, error2;
- isize = i_size_read(inode);
- if (unlikely(ret < 0 && ret != -EFAULT && iocb->ki_pos > isize))
- iocb->ki_pos = isize;
+ xfs_rw_iunlock(ip, iolock);
+ error = filemap_write_and_wait_range(mapping, pos, end);
+ xfs_rw_ilock(ip, iolock);
- if (iocb->ki_pos > ip->i_size) {
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- if (iocb->ki_pos > ip->i_size)
- ip->i_size = iocb->ki_pos;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ error2 = -xfs_file_fsync(file,
+ (file->f_flags & __O_SYNC) ? 0 : 1);
+ if (error)
+ ret = error;
+ else if (error2)
+ ret = error2;
}
- error = -ret;
- if (ret <= 0)
- goto out_unlock_internal;
+out_unlock:
+ xfs_aio_write_newsize_update(ip);
+ xfs_rw_iunlock(ip, iolock);
+ return ret;
+}
- XFS_STATS_ADD(xs_write_bytes, ret);
+STATIC long
+xfs_file_fallocate(
+ struct file *file,
+ int mode,
+ loff_t offset,
+ loff_t len)
+{
+ struct inode *inode = file->f_path.dentry->d_inode;
+ long error;
+ loff_t new_size = 0;
+ xfs_flock64_t bf;
+ xfs_inode_t *ip = XFS_I(inode);
+ int cmd = XFS_IOC_RESVSP;
- /* Handle various SYNC-type writes */
- if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) {
- loff_t end = pos + ret - 1;
- int error2;
+ if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
+ return -EOPNOTSUPP;
- xfs_iunlock(ip, iolock);
- if (need_i_mutex)
- mutex_unlock(&inode->i_mutex);
+ bf.l_whence = 0;
+ bf.l_start = offset;
+ bf.l_len = len;
- error2 = filemap_write_and_wait_range(mapping, pos, end);
- if (!error)
- error = error2;
- if (need_i_mutex)
- mutex_lock(&inode->i_mutex);
- xfs_ilock(ip, iolock);
+ xfs_ilock(ip, XFS_IOLOCK_EXCL);
- error2 = -xfs_file_fsync(file,
- (file->f_flags & __O_SYNC) ? 0 : 1);
- if (!error)
- error = error2;
+ if (mode & FALLOC_FL_PUNCH_HOLE)
+ cmd = XFS_IOC_UNRESVSP;
+
+ /* check the new inode size is valid before allocating */
+ if (!(mode & FALLOC_FL_KEEP_SIZE) &&
+ offset + len > i_size_read(inode)) {
+ new_size = offset + len;
+ error = inode_newsize_ok(inode, new_size);
+ if (error)
+ goto out_unlock;
}
- out_unlock_internal:
- if (ip->i_new_size) {
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- ip->i_new_size = 0;
- /*
- * If this was a direct or synchronous I/O that failed (such
- * as ENOSPC) then part of the I/O may have been written to
- * disk before the error occured. In this case the on-disk
- * file size may have been adjusted beyond the in-memory file
- * size and now needs to be truncated back.
- */
- if (ip->i_d.di_size > ip->i_size)
- ip->i_d.di_size = ip->i_size;
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ error = -xfs_change_file_space(ip, cmd, &bf, 0, XFS_ATTR_NOLOCK);
+ if (error)
+ goto out_unlock;
+
+ /* Change file size if needed */
+ if (new_size) {
+ struct iattr iattr;
+
+ iattr.ia_valid = ATTR_SIZE;
+ iattr.ia_size = new_size;
+ error = -xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK);
}
- xfs_iunlock(ip, iolock);
- out_unlock_mutex:
- if (need_i_mutex)
- mutex_unlock(&inode->i_mutex);
- return -error;
+
+out_unlock:
+ xfs_iunlock(ip, XFS_IOLOCK_EXCL);
+ return error;
}
+
STATIC int
xfs_file_open(
struct inode *inode,
@@ -921,6 +1055,7 @@ const struct file_operations xfs_file_operations = {
.open = xfs_file_open,
.release = xfs_file_release,
.fsync = xfs_file_fsync,
+ .fallocate = xfs_file_fallocate,
};
const struct file_operations xfs_dir_file_operations = {
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c
index ad442d9e392e..f5e2a19e0f8e 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl.c
@@ -39,6 +39,7 @@
#include "xfs_dfrag.h"
#include "xfs_fsops.h"
#include "xfs_vnodeops.h"
+#include "xfs_discard.h"
#include "xfs_quota.h"
#include "xfs_inode_item.h"
#include "xfs_export.h"
@@ -984,10 +985,22 @@ xfs_ioctl_setattr(
/*
* Extent size must be a multiple of the appropriate block
- * size, if set at all.
+ * size, if set at all. It must also be smaller than the
+ * maximum extent size supported by the filesystem.
+ *
+ * Also, for non-realtime files, limit the extent size hint to
+ * half the size of the AGs in the filesystem so alignment
+ * doesn't result in extents larger than an AG.
*/
if (fa->fsx_extsize != 0) {
- xfs_extlen_t size;
+ xfs_extlen_t size;
+ xfs_fsblock_t extsize_fsb;
+
+ extsize_fsb = XFS_B_TO_FSB(mp, fa->fsx_extsize);
+ if (extsize_fsb > MAXEXTLEN) {
+ code = XFS_ERROR(EINVAL);
+ goto error_return;
+ }
if (XFS_IS_REALTIME_INODE(ip) ||
((mask & FSX_XFLAGS) &&
@@ -996,6 +1009,10 @@ xfs_ioctl_setattr(
mp->m_sb.sb_blocklog;
} else {
size = mp->m_sb.sb_blocksize;
+ if (extsize_fsb > mp->m_sb.sb_agblocks / 2) {
+ code = XFS_ERROR(EINVAL);
+ goto error_return;
+ }
}
if (fa->fsx_extsize % size) {
@@ -1294,6 +1311,8 @@ xfs_file_ioctl(
trace_xfs_file_ioctl(ip);
switch (cmd) {
+ case FITRIM:
+ return xfs_ioc_trim(mp, arg);
case XFS_IOC_ALLOCSP:
case XFS_IOC_FREESP:
case XFS_IOC_RESVSP:
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index da54403633b6..bd5727852fd6 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -46,7 +46,6 @@
#include <linux/namei.h>
#include <linux/posix_acl.h>
#include <linux/security.h>
-#include <linux/falloc.h>
#include <linux/fiemap.h>
#include <linux/slab.h>
@@ -505,61 +504,6 @@ xfs_vn_setattr(
return -xfs_setattr(XFS_I(dentry->d_inode), iattr, 0);
}
-STATIC long
-xfs_vn_fallocate(
- struct inode *inode,
- int mode,
- loff_t offset,
- loff_t len)
-{
- long error;
- loff_t new_size = 0;
- xfs_flock64_t bf;
- xfs_inode_t *ip = XFS_I(inode);
- int cmd = XFS_IOC_RESVSP;
-
- /* preallocation on directories not yet supported */
- error = -ENODEV;
- if (S_ISDIR(inode->i_mode))
- goto out_error;
-
- bf.l_whence = 0;
- bf.l_start = offset;
- bf.l_len = len;
-
- xfs_ilock(ip, XFS_IOLOCK_EXCL);
-
- if (mode & FALLOC_FL_PUNCH_HOLE)
- cmd = XFS_IOC_UNRESVSP;
-
- /* check the new inode size is valid before allocating */
- if (!(mode & FALLOC_FL_KEEP_SIZE) &&
- offset + len > i_size_read(inode)) {
- new_size = offset + len;
- error = inode_newsize_ok(inode, new_size);
- if (error)
- goto out_unlock;
- }
-
- error = -xfs_change_file_space(ip, cmd, &bf, 0, XFS_ATTR_NOLOCK);
- if (error)
- goto out_unlock;
-
- /* Change file size if needed */
- if (new_size) {
- struct iattr iattr;
-
- iattr.ia_valid = ATTR_SIZE;
- iattr.ia_size = new_size;
- error = -xfs_setattr(ip, &iattr, XFS_ATTR_NOLOCK);
- }
-
-out_unlock:
- xfs_iunlock(ip, XFS_IOLOCK_EXCL);
-out_error:
- return error;
-}
-
#define XFS_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR)
/*
@@ -653,7 +597,6 @@ static const struct inode_operations xfs_inode_operations = {
.getxattr = generic_getxattr,
.removexattr = generic_removexattr,
.listxattr = xfs_vn_listxattr,
- .fallocate = xfs_vn_fallocate,
.fiemap = xfs_vn_fiemap,
};
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index bd07f7339366..9731898083ae 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -1414,7 +1414,7 @@ xfs_fs_freeze(
xfs_save_resvblks(mp);
xfs_quiesce_attr(mp);
- return -xfs_fs_log_dummy(mp, SYNC_WAIT);
+ return -xfs_fs_log_dummy(mp);
}
STATIC int
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index a02480de9759..e22f0057d21f 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -362,7 +362,7 @@ xfs_quiesce_data(
/* mark the log as covered if needed */
if (xfs_log_need_covered(mp))
- error2 = xfs_fs_log_dummy(mp, SYNC_WAIT);
+ error2 = xfs_fs_log_dummy(mp);
/* flush data-only devices */
if (mp->m_rtdev_targp)
@@ -503,13 +503,14 @@ xfs_sync_worker(
int error;
if (!(mp->m_flags & XFS_MOUNT_RDONLY)) {
- xfs_log_force(mp, 0);
- xfs_reclaim_inodes(mp, 0);
/* dgc: errors ignored here */
- error = xfs_qm_sync(mp, SYNC_TRYLOCK);
if (mp->m_super->s_frozen == SB_UNFROZEN &&
xfs_log_need_covered(mp))
- error = xfs_fs_log_dummy(mp, 0);
+ error = xfs_fs_log_dummy(mp);
+ else
+ xfs_log_force(mp, 0);
+ xfs_reclaim_inodes(mp, 0);
+ error = xfs_qm_sync(mp, SYNC_TRYLOCK);
}
mp->m_sync_seq++;
wake_up(&mp->m_wait_single_sync_task);
diff --git a/fs/xfs/linux-2.6/xfs_sysctl.c b/fs/xfs/linux-2.6/xfs_sysctl.c
index 7bb5092d6ae4..ee3cee097e7e 100644
--- a/fs/xfs/linux-2.6/xfs_sysctl.c
+++ b/fs/xfs/linux-2.6/xfs_sysctl.c
@@ -18,6 +18,7 @@
#include "xfs.h"
#include <linux/sysctl.h>
#include <linux/proc_fs.h>
+#include "xfs_error.h"
static struct ctl_table_header *xfs_table_header;
@@ -51,6 +52,26 @@ xfs_stats_clear_proc_handler(
return ret;
}
+
+STATIC int
+xfs_panic_mask_proc_handler(
+ ctl_table *ctl,
+ int write,
+ void __user *buffer,
+ size_t *lenp,
+ loff_t *ppos)
+{
+ int ret, *valp = ctl->data;
+
+ ret = proc_dointvec_minmax(ctl, write, buffer, lenp, ppos);
+ if (!ret && write) {
+ xfs_panic_mask = *valp;
+#ifdef DEBUG
+ xfs_panic_mask |= (XFS_PTAG_SHUTDOWN_CORRUPT | XFS_PTAG_LOGRES);
+#endif
+ }
+ return ret;
+}
#endif /* CONFIG_PROC_FS */
static ctl_table xfs_table[] = {
@@ -77,7 +98,7 @@ static ctl_table xfs_table[] = {
.data = &xfs_params.panic_mask.val,
.maxlen = sizeof(int),
.mode = 0644,
- .proc_handler = proc_dointvec_minmax,
+ .proc_handler = xfs_panic_mask_proc_handler,
.extra1 = &xfs_params.panic_mask.min,
.extra2 = &xfs_params.panic_mask.max
},
diff --git a/fs/xfs/linux-2.6/xfs_trace.h b/fs/xfs/linux-2.6/xfs_trace.h
index 647af2a2e7aa..2d0bcb479075 100644
--- a/fs/xfs/linux-2.6/xfs_trace.h
+++ b/fs/xfs/linux-2.6/xfs_trace.h
@@ -1759,6 +1759,39 @@ DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_recover);
DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_cancel);
DEFINE_LOG_RECOVER_INO_ITEM(xfs_log_recover_inode_skip);
+DECLARE_EVENT_CLASS(xfs_discard_class,
+ TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno,
+ xfs_agblock_t agbno, xfs_extlen_t len),
+ TP_ARGS(mp, agno, agbno, len),
+ TP_STRUCT__entry(
+ __field(dev_t, dev)
+ __field(xfs_agnumber_t, agno)
+ __field(xfs_agblock_t, agbno)
+ __field(xfs_extlen_t, len)
+ ),
+ TP_fast_assign(
+ __entry->dev = mp->m_super->s_dev;
+ __entry->agno = agno;
+ __entry->agbno = agbno;
+ __entry->len = len;
+ ),
+ TP_printk("dev %d:%d agno %u agbno %u len %u\n",
+ MAJOR(__entry->dev), MINOR(__entry->dev),
+ __entry->agno,
+ __entry->agbno,
+ __entry->len)
+)
+
+#define DEFINE_DISCARD_EVENT(name) \
+DEFINE_EVENT(xfs_discard_class, name, \
+ TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno, \
+ xfs_agblock_t agbno, xfs_extlen_t len), \
+ TP_ARGS(mp, agno, agbno, len))
+DEFINE_DISCARD_EVENT(xfs_discard_extent);
+DEFINE_DISCARD_EVENT(xfs_discard_toosmall);
+DEFINE_DISCARD_EVENT(xfs_discard_exclude);
+DEFINE_DISCARD_EVENT(xfs_discard_busy);
+
#endif /* _TRACE_XFS_H */
#undef TRACE_INCLUDE_PATH
diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c
index f8e854b4fde8..206a2815ced6 100644
--- a/fs/xfs/quota/xfs_qm.c
+++ b/fs/xfs/quota/xfs_qm.c
@@ -1863,12 +1863,14 @@ xfs_qm_dqreclaim_one(void)
xfs_dquot_t *dqpout;
xfs_dquot_t *dqp;
int restarts;
+ int startagain;
restarts = 0;
dqpout = NULL;
/* lockorder: hashchainlock, freelistlock, mplistlock, dqlock, dqflock */
-startagain:
+again:
+ startagain = 0;
mutex_lock(&xfs_Gqm->qm_dqfrlist_lock);
list_for_each_entry(dqp, &xfs_Gqm->qm_dqfrlist, q_freelist) {
@@ -1885,13 +1887,10 @@ startagain:
ASSERT(! (dqp->dq_flags & XFS_DQ_INACTIVE));
trace_xfs_dqreclaim_want(dqp);
-
- xfs_dqunlock(dqp);
- mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
- if (++restarts >= XFS_QM_RECLAIM_MAX_RESTARTS)
- return NULL;
XQM_STATS_INC(xqmstats.xs_qm_dqwants);
- goto startagain;
+ restarts++;
+ startagain = 1;
+ goto dqunlock;
}
/*
@@ -1906,23 +1905,20 @@ startagain:
ASSERT(list_empty(&dqp->q_mplist));
list_del_init(&dqp->q_freelist);
xfs_Gqm->qm_dqfrlist_cnt--;
- xfs_dqunlock(dqp);
dqpout = dqp;
XQM_STATS_INC(xqmstats.xs_qm_dqinact_reclaims);
- break;
+ goto dqunlock;
}
ASSERT(dqp->q_hash);
ASSERT(!list_empty(&dqp->q_mplist));
/*
- * Try to grab the flush lock. If this dquot is in the process of
- * getting flushed to disk, we don't want to reclaim it.
+ * Try to grab the flush lock. If this dquot is in the process
+ * of getting flushed to disk, we don't want to reclaim it.
*/
- if (!xfs_dqflock_nowait(dqp)) {
- xfs_dqunlock(dqp);
- continue;
- }
+ if (!xfs_dqflock_nowait(dqp))
+ goto dqunlock;
/*
* We have the flush lock so we know that this is not in the
@@ -1944,8 +1940,7 @@ startagain:
xfs_fs_cmn_err(CE_WARN, mp,
"xfs_qm_dqreclaim: dquot %p flush failed", dqp);
}
- xfs_dqunlock(dqp); /* dqflush unlocks dqflock */
- continue;
+ goto dqunlock;
}
/*
@@ -1967,13 +1962,8 @@ startagain:
*/
if (!mutex_trylock(&mp->m_quotainfo->qi_dqlist_lock)) {
restarts++;
- mutex_unlock(&dqp->q_hash->qh_lock);
- xfs_dqfunlock(dqp);
- xfs_dqunlock(dqp);
- mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
- if (restarts++ >= XFS_QM_RECLAIM_MAX_RESTARTS)
- return NULL;
- goto startagain;
+ startagain = 1;
+ goto qhunlock;
}
ASSERT(dqp->q_nrefs == 0);
@@ -1986,14 +1976,20 @@ startagain:
xfs_Gqm->qm_dqfrlist_cnt--;
dqpout = dqp;
mutex_unlock(&mp->m_quotainfo->qi_dqlist_lock);
+qhunlock:
mutex_unlock(&dqp->q_hash->qh_lock);
dqfunlock:
xfs_dqfunlock(dqp);
+dqunlock:
xfs_dqunlock(dqp);
if (dqpout)
break;
if (restarts >= XFS_QM_RECLAIM_MAX_RESTARTS)
- return NULL;
+ break;
+ if (startagain) {
+ mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
+ goto again;
+ }
}
mutex_unlock(&xfs_Gqm->qm_dqfrlist_lock);
return dqpout;
diff --git a/fs/xfs/support/debug.c b/fs/xfs/support/debug.c
index 975aa10e1a47..0df88897ef84 100644
--- a/fs/xfs/support/debug.c
+++ b/fs/xfs/support/debug.c
@@ -25,86 +25,78 @@
#include "xfs_mount.h"
#include "xfs_error.h"
-static char message[1024]; /* keep it off the stack */
-static DEFINE_SPINLOCK(xfs_err_lock);
-
-/* Translate from CE_FOO to KERN_FOO, err_level(CE_FOO) == KERN_FOO */
-#define XFS_MAX_ERR_LEVEL 7
-#define XFS_ERR_MASK ((1 << 3) - 1)
-static const char * const err_level[XFS_MAX_ERR_LEVEL+1] =
- {KERN_EMERG, KERN_ALERT, KERN_CRIT,
- KERN_ERR, KERN_WARNING, KERN_NOTICE,
- KERN_INFO, KERN_DEBUG};
-
void
-cmn_err(register int level, char *fmt, ...)
+cmn_err(
+ const char *lvl,
+ const char *fmt,
+ ...)
{
- char *fp = fmt;
- int len;
- ulong flags;
- va_list ap;
-
- level &= XFS_ERR_MASK;
- if (level > XFS_MAX_ERR_LEVEL)
- level = XFS_MAX_ERR_LEVEL;
- spin_lock_irqsave(&xfs_err_lock,flags);
- va_start(ap, fmt);
- if (*fmt == '!') fp++;
- len = vsnprintf(message, sizeof(message), fp, ap);
- if (len >= sizeof(message))
- len = sizeof(message) - 1;
- if (message[len-1] == '\n')
- message[len-1] = 0;
- printk("%s%s\n", err_level[level], message);
- va_end(ap);
- spin_unlock_irqrestore(&xfs_err_lock,flags);
- BUG_ON(level == CE_PANIC);
+ struct va_format vaf;
+ va_list args;
+
+ va_start(args, fmt);
+ vaf.fmt = fmt;
+ vaf.va = &args;
+
+ printk("%s%pV", lvl, &vaf);
+ va_end(args);
+
+ BUG_ON(strncmp(lvl, KERN_EMERG, strlen(KERN_EMERG)) == 0);
}
void
-xfs_fs_vcmn_err(
- int level,
+xfs_fs_cmn_err(
+ const char *lvl,
struct xfs_mount *mp,
- char *fmt,
- va_list ap)
+ const char *fmt,
+ ...)
{
- unsigned long flags;
- int len = 0;
+ struct va_format vaf;
+ va_list args;
- level &= XFS_ERR_MASK;
- if (level > XFS_MAX_ERR_LEVEL)
- level = XFS_MAX_ERR_LEVEL;
+ va_start(args, fmt);
+ vaf.fmt = fmt;
+ vaf.va = &args;
- spin_lock_irqsave(&xfs_err_lock,flags);
+ printk("%sFilesystem %s: %pV", lvl, mp->m_fsname, &vaf);
+ va_end(args);
- if (mp) {
- len = sprintf(message, "Filesystem \"%s\": ", mp->m_fsname);
+ BUG_ON(strncmp(lvl, KERN_EMERG, strlen(KERN_EMERG)) == 0);
+}
+
+/* All callers to xfs_cmn_err use CE_ALERT, so don't bother testing lvl */
+void
+xfs_cmn_err(
+ int panic_tag,
+ const char *lvl,
+ struct xfs_mount *mp,
+ const char *fmt,
+ ...)
+{
+ struct va_format vaf;
+ va_list args;
+ int do_panic = 0;
- /*
- * Skip the printk if we can't print anything useful
- * due to an over-long device name.
- */
- if (len >= sizeof(message))
- goto out;
+ if (xfs_panic_mask && (xfs_panic_mask & panic_tag)) {
+ printk(KERN_ALERT "XFS: Transforming an alert into a BUG.");
+ do_panic = 1;
}
- len = vsnprintf(message + len, sizeof(message) - len, fmt, ap);
- if (len >= sizeof(message))
- len = sizeof(message) - 1;
- if (message[len-1] == '\n')
- message[len-1] = 0;
+ va_start(args, fmt);
+ vaf.fmt = fmt;
+ vaf.va = &args;
- printk("%s%s\n", err_level[level], message);
- out:
- spin_unlock_irqrestore(&xfs_err_lock,flags);
+ printk(KERN_ALERT "Filesystem %s: %pV", mp->m_fsname, &vaf);
+ va_end(args);
- BUG_ON(level == CE_PANIC);
+ BUG_ON(do_panic);
}
void
assfail(char *expr, char *file, int line)
{
- printk("Assertion failed: %s, file: %s, line: %d\n", expr, file, line);
+ printk(KERN_CRIT "Assertion failed: %s, file: %s, line: %d\n", expr,
+ file, line);
BUG();
}
diff --git a/fs/xfs/support/debug.h b/fs/xfs/support/debug.h
index d2d20462fd4f..05699f67d475 100644
--- a/fs/xfs/support/debug.h
+++ b/fs/xfs/support/debug.h
@@ -20,15 +20,22 @@
#include <stdarg.h>
-#define CE_DEBUG 7 /* debug */
-#define CE_CONT 6 /* continuation */
-#define CE_NOTE 5 /* notice */
-#define CE_WARN 4 /* warning */
-#define CE_ALERT 1 /* alert */
-#define CE_PANIC 0 /* panic */
-
-extern void cmn_err(int, char *, ...)
- __attribute__ ((format (printf, 2, 3)));
+struct xfs_mount;
+
+#define CE_DEBUG KERN_DEBUG
+#define CE_CONT KERN_INFO
+#define CE_NOTE KERN_NOTICE
+#define CE_WARN KERN_WARNING
+#define CE_ALERT KERN_ALERT
+#define CE_PANIC KERN_EMERG
+
+void cmn_err(const char *lvl, const char *fmt, ...)
+ __attribute__ ((format (printf, 2, 3)));
+void xfs_fs_cmn_err( const char *lvl, struct xfs_mount *mp,
+ const char *fmt, ...) __attribute__ ((format (printf, 3, 4)));
+void xfs_cmn_err( int panic_tag, const char *lvl, struct xfs_mount *mp,
+ const char *fmt, ...) __attribute__ ((format (printf, 4, 5)));
+
extern void assfail(char *expr, char *f, int l);
#define ASSERT_ALWAYS(expr) \
diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c
index fa8723f5870a..f3227984a9bf 100644
--- a/fs/xfs/xfs_alloc.c
+++ b/fs/xfs/xfs_alloc.c
@@ -41,10 +41,6 @@
#define XFSA_FIXUP_BNO_OK 1
#define XFSA_FIXUP_CNT_OK 2
-static int
-xfs_alloc_busy_search(struct xfs_mount *mp, xfs_agnumber_t agno,
- xfs_agblock_t bno, xfs_extlen_t len);
-
/*
* Prototypes for per-ag allocation routines
*/
@@ -94,7 +90,7 @@ xfs_alloc_lookup_ge(
* Lookup the first record less than or equal to [bno, len]
* in the btree given by cur.
*/
-STATIC int /* error */
+int /* error */
xfs_alloc_lookup_le(
struct xfs_btree_cur *cur, /* btree cursor */
xfs_agblock_t bno, /* starting block of extent */
@@ -127,7 +123,7 @@ xfs_alloc_update(
/*
* Get the data from the pointed-to record.
*/
-STATIC int /* error */
+int /* error */
xfs_alloc_get_rec(
struct xfs_btree_cur *cur, /* btree cursor */
xfs_agblock_t *bno, /* output: starting block of extent */
@@ -2615,7 +2611,7 @@ restart:
* will require a synchronous transaction, but it can still be
* used to distinguish between a partial or exact match.
*/
-static int
+int
xfs_alloc_busy_search(
struct xfs_mount *mp,
xfs_agnumber_t agno,
diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/xfs_alloc.h
index 895009a97271..d0b3bc72005b 100644
--- a/fs/xfs/xfs_alloc.h
+++ b/fs/xfs/xfs_alloc.h
@@ -19,6 +19,7 @@
#define __XFS_ALLOC_H__
struct xfs_buf;
+struct xfs_btree_cur;
struct xfs_mount;
struct xfs_perag;
struct xfs_trans;
@@ -74,6 +75,22 @@ typedef unsigned int xfs_alloctype_t;
#define XFS_ALLOC_SET_ASIDE(mp) (4 + ((mp)->m_sb.sb_agcount * 4))
/*
+ * When deciding how much space to allocate out of an AG, we limit the
+ * allocation maximum size to the size the AG. However, we cannot use all the
+ * blocks in the AG - some are permanently used by metadata. These
+ * blocks are generally:
+ * - the AG superblock, AGF, AGI and AGFL
+ * - the AGF (bno and cnt) and AGI btree root blocks
+ * - 4 blocks on the AGFL according to XFS_ALLOC_SET_ASIDE() limits
+ *
+ * The AG headers are sector sized, so the amount of space they take up is
+ * dependent on filesystem geometry. The others are all single blocks.
+ */
+#define XFS_ALLOC_AG_MAX_USABLE(mp) \
+ ((mp)->m_sb.sb_agblocks - XFS_BB_TO_FSB(mp, XFS_FSS_TO_BB(mp, 4)) - 7)
+
+
+/*
* Argument structure for xfs_alloc routines.
* This is turned into a structure to avoid having 20 arguments passed
* down several levels of the stack.
@@ -118,16 +135,16 @@ xfs_alloc_longest_free_extent(struct xfs_mount *mp,
struct xfs_perag *pag);
#ifdef __KERNEL__
-
void
-xfs_alloc_busy_insert(xfs_trans_t *tp,
- xfs_agnumber_t agno,
- xfs_agblock_t bno,
- xfs_extlen_t len);
+xfs_alloc_busy_insert(struct xfs_trans *tp, xfs_agnumber_t agno,
+ xfs_agblock_t bno, xfs_extlen_t len);
void
xfs_alloc_busy_clear(struct xfs_mount *mp, struct xfs_busy_extent *busyp);
+int
+xfs_alloc_busy_search(struct xfs_mount *mp, xfs_agnumber_t agno,
+ xfs_agblock_t bno, xfs_extlen_t len);
#endif /* __KERNEL__ */
/*
@@ -205,4 +222,18 @@ xfs_free_extent(
xfs_fsblock_t bno, /* starting block number of extent */
xfs_extlen_t len); /* length of extent */
+int /* error */
+xfs_alloc_lookup_le(
+ struct xfs_btree_cur *cur, /* btree cursor */
+ xfs_agblock_t bno, /* starting block of extent */
+ xfs_extlen_t len, /* length of extent */
+ int *stat); /* success/failure */
+
+int /* error */
+xfs_alloc_get_rec(
+ struct xfs_btree_cur *cur, /* btree cursor */
+ xfs_agblock_t *bno, /* output: starting block of extent */
+ xfs_extlen_t *len, /* output: length of extent */
+ int *stat); /* output: success/failure */
+
#endif /* __XFS_ALLOC_H__ */
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c
index 4111cd3966c7..dc3afd7739ff 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/xfs_bmap.c
@@ -1038,17 +1038,34 @@ xfs_bmap_add_extent_delay_real(
* Filling in the middle part of a previous delayed allocation.
* Contiguity is impossible here.
* This case is avoided almost all the time.
+ *
+ * We start with a delayed allocation:
+ *
+ * +ddddddddddddddddddddddddddddddddddddddddddddddddddddddd+
+ * PREV @ idx
+ *
+ * and we are allocating:
+ * +rrrrrrrrrrrrrrrrr+
+ * new
+ *
+ * and we set it up for insertion as:
+ * +ddddddddddddddddddd+rrrrrrrrrrrrrrrrr+ddddddddddddddddd+
+ * new
+ * PREV @ idx LEFT RIGHT
+ * inserted at idx + 1
*/
temp = new->br_startoff - PREV.br_startoff;
- trace_xfs_bmap_pre_update(ip, idx, 0, _THIS_IP_);
- xfs_bmbt_set_blockcount(ep, temp);
- r[0] = *new;
- r[1].br_state = PREV.br_state;
- r[1].br_startblock = 0;
- r[1].br_startoff = new_endoff;
temp2 = PREV.br_startoff + PREV.br_blockcount - new_endoff;
- r[1].br_blockcount = temp2;
- xfs_iext_insert(ip, idx + 1, 2, &r[0], state);
+ trace_xfs_bmap_pre_update(ip, idx, 0, _THIS_IP_);
+ xfs_bmbt_set_blockcount(ep, temp); /* truncate PREV */
+ LEFT = *new;
+ RIGHT.br_state = PREV.br_state;
+ RIGHT.br_startblock = nullstartblock(
+ (int)xfs_bmap_worst_indlen(ip, temp2));
+ RIGHT.br_startoff = new_endoff;
+ RIGHT.br_blockcount = temp2;
+ /* insert LEFT (r[0]) and RIGHT (r[1]) at the same time */
+ xfs_iext_insert(ip, idx + 1, 2, &LEFT, state);
ip->i_df.if_lastex = idx + 1;
ip->i_d.di_nextents++;
if (cur == NULL)
@@ -2430,7 +2447,7 @@ xfs_bmap_btalloc_nullfb(
startag = ag = 0;
pag = xfs_perag_get(mp, ag);
- while (*blen < ap->alen) {
+ while (*blen < args->maxlen) {
if (!pag->pagf_init) {
error = xfs_alloc_pagf_init(mp, args->tp, ag,
XFS_ALLOC_FLAG_TRYLOCK);
@@ -2452,7 +2469,7 @@ xfs_bmap_btalloc_nullfb(
notinit = 1;
if (xfs_inode_is_filestream(ap->ip)) {
- if (*blen >= ap->alen)
+ if (*blen >= args->maxlen)
break;
if (ap->userdata) {
@@ -2498,14 +2515,14 @@ xfs_bmap_btalloc_nullfb(
* If the best seen length is less than the request
* length, use the best as the minimum.
*/
- else if (*blen < ap->alen)
+ else if (*blen < args->maxlen)
args->minlen = *blen;
/*
- * Otherwise we've seen an extent as big as alen,
+ * Otherwise we've seen an extent as big as maxlen,
* use that as the minimum.
*/
else
- args->minlen = ap->alen;
+ args->minlen = args->maxlen;
/*
* set the failure fallback case to look in the selected
@@ -2573,7 +2590,9 @@ xfs_bmap_btalloc(
args.tp = ap->tp;
args.mp = mp;
args.fsbno = ap->rval;
- args.maxlen = MIN(ap->alen, mp->m_sb.sb_agblocks);
+
+ /* Trim the allocation back to the maximum an AG can fit. */
+ args.maxlen = MIN(ap->alen, XFS_ALLOC_AG_MAX_USABLE(mp));
args.firstblock = ap->firstblock;
blen = 0;
if (nullfb) {
@@ -2621,7 +2640,7 @@ xfs_bmap_btalloc(
/*
* Adjust for alignment
*/
- if (blen > args.alignment && blen <= ap->alen)
+ if (blen > args.alignment && blen <= args.maxlen)
args.minlen = blen - args.alignment;
args.minalignslop = 0;
} else {
@@ -2640,7 +2659,7 @@ xfs_bmap_btalloc(
* of minlen+alignment+slop doesn't go up
* between the calls.
*/
- if (blen > mp->m_dalign && blen <= ap->alen)
+ if (blen > mp->m_dalign && blen <= args.maxlen)
nextminlen = blen - mp->m_dalign;
else
nextminlen = args.minlen;
@@ -4485,6 +4504,16 @@ xfs_bmapi(
/* Figure out the extent size, adjust alen */
extsz = xfs_get_extsz_hint(ip);
if (extsz) {
+ /*
+ * make sure we don't exceed a single
+ * extent length when we align the
+ * extent by reducing length we are
+ * going to allocate by the maximum
+ * amount extent size aligment may
+ * require.
+ */
+ alen = XFS_FILBLKS_MIN(len,
+ MAXEXTLEN - (2 * extsz - 1));
error = xfs_bmap_extsize_align(mp,
&got, &prev, extsz,
rt, eof,
diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c
index ed2b65f3f8b9..6f8c21ce0d6d 100644
--- a/fs/xfs/xfs_buf_item.c
+++ b/fs/xfs/xfs_buf_item.c
@@ -141,7 +141,6 @@ xfs_buf_item_log_check(
#define xfs_buf_item_log_check(x)
#endif
-STATIC void xfs_buf_error_relse(xfs_buf_t *bp);
STATIC void xfs_buf_do_callbacks(struct xfs_buf *bp);
/*
@@ -428,13 +427,15 @@ xfs_buf_item_unpin(
if (remove) {
/*
- * We have to remove the log item from the transaction
- * as we are about to release our reference to the
- * buffer. If we don't, the unlock that occurs later
- * in xfs_trans_uncommit() will ry to reference the
+ * If we are in a transaction context, we have to
+ * remove the log item from the transaction as we are
+ * about to release our reference to the buffer. If we
+ * don't, the unlock that occurs later in
+ * xfs_trans_uncommit() will try to reference the
* buffer which we no longer have a hold on.
*/
- xfs_trans_del_item(lip);
+ if (lip->li_desc)
+ xfs_trans_del_item(lip);
/*
* Since the transaction no longer refers to the buffer,
@@ -959,128 +960,76 @@ xfs_buf_do_callbacks(
*/
void
xfs_buf_iodone_callbacks(
- xfs_buf_t *bp)
+ struct xfs_buf *bp)
{
- xfs_log_item_t *lip;
- static ulong lasttime;
- static xfs_buftarg_t *lasttarg;
- xfs_mount_t *mp;
+ struct xfs_log_item *lip = bp->b_fspriv;
+ struct xfs_mount *mp = lip->li_mountp;
+ static ulong lasttime;
+ static xfs_buftarg_t *lasttarg;
- ASSERT(XFS_BUF_FSPRIVATE(bp, void *) != NULL);
- lip = XFS_BUF_FSPRIVATE(bp, xfs_log_item_t *);
+ if (likely(!XFS_BUF_GETERROR(bp)))
+ goto do_callbacks;
- if (XFS_BUF_GETERROR(bp) != 0) {
- /*
- * If we've already decided to shutdown the filesystem
- * because of IO errors, there's no point in giving this
- * a retry.
- */
- mp = lip->li_mountp;
- if (XFS_FORCED_SHUTDOWN(mp)) {
- ASSERT(XFS_BUF_TARGET(bp) == mp->m_ddev_targp);
- XFS_BUF_SUPER_STALE(bp);
- trace_xfs_buf_item_iodone(bp, _RET_IP_);
- xfs_buf_do_callbacks(bp);
- XFS_BUF_SET_FSPRIVATE(bp, NULL);
- XFS_BUF_CLR_IODONE_FUNC(bp);
- xfs_buf_ioend(bp, 0);
- return;
- }
+ /*
+ * If we've already decided to shutdown the filesystem because of
+ * I/O errors, there's no point in giving this a retry.
+ */
+ if (XFS_FORCED_SHUTDOWN(mp)) {
+ XFS_BUF_SUPER_STALE(bp);
+ trace_xfs_buf_item_iodone(bp, _RET_IP_);
+ goto do_callbacks;
+ }
- if ((XFS_BUF_TARGET(bp) != lasttarg) ||
- (time_after(jiffies, (lasttime + 5*HZ)))) {
- lasttime = jiffies;
- cmn_err(CE_ALERT, "Device %s, XFS metadata write error"
- " block 0x%llx in %s",
- XFS_BUFTARG_NAME(XFS_BUF_TARGET(bp)),
- (__uint64_t)XFS_BUF_ADDR(bp), mp->m_fsname);
- }
- lasttarg = XFS_BUF_TARGET(bp);
+ if (XFS_BUF_TARGET(bp) != lasttarg ||
+ time_after(jiffies, (lasttime + 5*HZ))) {
+ lasttime = jiffies;
+ cmn_err(CE_ALERT, "Device %s, XFS metadata write error"
+ " block 0x%llx in %s",
+ XFS_BUFTARG_NAME(XFS_BUF_TARGET(bp)),
+ (__uint64_t)XFS_BUF_ADDR(bp), mp->m_fsname);
+ }
+ lasttarg = XFS_BUF_TARGET(bp);
- if (XFS_BUF_ISASYNC(bp)) {
- /*
- * If the write was asynchronous then noone will be
- * looking for the error. Clear the error state
- * and write the buffer out again delayed write.
- *
- * XXXsup This is OK, so long as we catch these
- * before we start the umount; we don't want these
- * DELWRI metadata bufs to be hanging around.
- */
- XFS_BUF_ERROR(bp,0); /* errno of 0 unsets the flag */
-
- if (!(XFS_BUF_ISSTALE(bp))) {
- XFS_BUF_DELAYWRITE(bp);
- XFS_BUF_DONE(bp);
- XFS_BUF_SET_START(bp);
- }
- ASSERT(XFS_BUF_IODONE_FUNC(bp));
- trace_xfs_buf_item_iodone_async(bp, _RET_IP_);
- xfs_buf_relse(bp);
- } else {
- /*
- * If the write of the buffer was not asynchronous,
- * then we want to make sure to return the error
- * to the caller of bwrite(). Because of this we
- * cannot clear the B_ERROR state at this point.
- * Instead we install a callback function that
- * will be called when the buffer is released, and
- * that routine will clear the error state and
- * set the buffer to be written out again after
- * some delay.
- */
- /* We actually overwrite the existing b-relse
- function at times, but we're gonna be shutting down
- anyway. */
- XFS_BUF_SET_BRELSE_FUNC(bp,xfs_buf_error_relse);
+ /*
+ * If the write was asynchronous then noone will be looking for the
+ * error. Clear the error state and write the buffer out again.
+ *
+ * During sync or umount we'll write all pending buffers again
+ * synchronous, which will catch these errors if they keep hanging
+ * around.
+ */
+ if (XFS_BUF_ISASYNC(bp)) {
+ XFS_BUF_ERROR(bp, 0); /* errno of 0 unsets the flag */
+
+ if (!XFS_BUF_ISSTALE(bp)) {
+ XFS_BUF_DELAYWRITE(bp);
XFS_BUF_DONE(bp);
- XFS_BUF_FINISH_IOWAIT(bp);
+ XFS_BUF_SET_START(bp);
}
+ ASSERT(XFS_BUF_IODONE_FUNC(bp));
+ trace_xfs_buf_item_iodone_async(bp, _RET_IP_);
+ xfs_buf_relse(bp);
return;
}
- xfs_buf_do_callbacks(bp);
- XFS_BUF_SET_FSPRIVATE(bp, NULL);
- XFS_BUF_CLR_IODONE_FUNC(bp);
- xfs_buf_ioend(bp, 0);
-}
-
-/*
- * This is a callback routine attached to a buffer which gets an error
- * when being written out synchronously.
- */
-STATIC void
-xfs_buf_error_relse(
- xfs_buf_t *bp)
-{
- xfs_log_item_t *lip;
- xfs_mount_t *mp;
-
- lip = XFS_BUF_FSPRIVATE(bp, xfs_log_item_t *);
- mp = (xfs_mount_t *)lip->li_mountp;
- ASSERT(XFS_BUF_TARGET(bp) == mp->m_ddev_targp);
-
+ /*
+ * If the write of the buffer was synchronous, we want to make
+ * sure to return the error to the caller of xfs_bwrite().
+ */
XFS_BUF_STALE(bp);
XFS_BUF_DONE(bp);
XFS_BUF_UNDELAYWRITE(bp);
- XFS_BUF_ERROR(bp,0);
trace_xfs_buf_error_relse(bp, _RET_IP_);
+ xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
- if (! XFS_FORCED_SHUTDOWN(mp))
- xfs_force_shutdown(mp, SHUTDOWN_META_IO_ERROR);
- /*
- * We have to unpin the pinned buffers so do the
- * callbacks.
- */
+do_callbacks:
xfs_buf_do_callbacks(bp);
XFS_BUF_SET_FSPRIVATE(bp, NULL);
XFS_BUF_CLR_IODONE_FUNC(bp);
- XFS_BUF_SET_BRELSE_FUNC(bp,NULL);
- xfs_buf_relse(bp);
+ xfs_buf_ioend(bp, 0);
}
-
/*
* This is the iodone() function for buffers which have been
* logged. It is called when they are eventually flushed out.
diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c
index c78cc6a3d87c..4c7db74a05f7 100644
--- a/fs/xfs/xfs_error.c
+++ b/fs/xfs/xfs_error.c
@@ -152,37 +152,6 @@ xfs_errortag_clearall(xfs_mount_t *mp, int loud)
}
#endif /* DEBUG */
-
-void
-xfs_fs_cmn_err(int level, xfs_mount_t *mp, char *fmt, ...)
-{
- va_list ap;
-
- va_start(ap, fmt);
- xfs_fs_vcmn_err(level, mp, fmt, ap);
- va_end(ap);
-}
-
-void
-xfs_cmn_err(int panic_tag, int level, xfs_mount_t *mp, char *fmt, ...)
-{
- va_list ap;
-
-#ifdef DEBUG
- xfs_panic_mask |= (XFS_PTAG_SHUTDOWN_CORRUPT | XFS_PTAG_LOGRES);
-#endif
-
- if (xfs_panic_mask && (xfs_panic_mask & panic_tag)
- && (level & CE_ALERT)) {
- level &= ~CE_ALERT;
- level |= CE_PANIC;
- cmn_err(CE_ALERT, "XFS: Transforming an alert into a BUG.");
- }
- va_start(ap, fmt);
- xfs_fs_vcmn_err(level, mp, fmt, ap);
- va_end(ap);
-}
-
void
xfs_error_report(
const char *tag,
diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h
index f338847f80b8..10dce5475f02 100644
--- a/fs/xfs/xfs_error.h
+++ b/fs/xfs/xfs_error.h
@@ -136,8 +136,8 @@ extern int xfs_error_test(int, int *, char *, int, char *, unsigned long);
xfs_error_test((tag), (mp)->m_fixedfsid, "expr", __LINE__, __FILE__, \
(rf))))
-extern int xfs_errortag_add(int error_tag, xfs_mount_t *mp);
-extern int xfs_errortag_clearall(xfs_mount_t *mp, int loud);
+extern int xfs_errortag_add(int error_tag, struct xfs_mount *mp);
+extern int xfs_errortag_clearall(struct xfs_mount *mp, int loud);
#else
#define XFS_TEST_ERROR(expr, mp, tag, rf) (expr)
#define xfs_errortag_add(tag, mp) (ENOSYS)
@@ -162,21 +162,15 @@ extern int xfs_errortag_clearall(xfs_mount_t *mp, int loud);
struct xfs_mount;
-extern void xfs_fs_vcmn_err(int level, struct xfs_mount *mp,
- char *fmt, va_list ap)
- __attribute__ ((format (printf, 3, 0)));
-extern void xfs_cmn_err(int panic_tag, int level, struct xfs_mount *mp,
- char *fmt, ...)
- __attribute__ ((format (printf, 4, 5)));
-extern void xfs_fs_cmn_err(int level, struct xfs_mount *mp, char *fmt, ...)
- __attribute__ ((format (printf, 3, 4)));
-
extern void xfs_hex_dump(void *p, int length);
#define xfs_fs_repair_cmn_err(level, mp, fmt, args...) \
xfs_fs_cmn_err(level, mp, fmt " Unmount and run xfs_repair.", ## args)
#define xfs_fs_mount_cmn_err(f, fmt, args...) \
- ((f & XFS_MFSI_QUIET)? (void)0 : cmn_err(CE_WARN, "XFS: " fmt, ## args))
+ do { \
+ if (!(f & XFS_MFSI_QUIET)) \
+ cmn_err(CE_WARN, "XFS: " fmt, ## args); \
+ } while (0)
#endif /* __XFS_ERROR_H__ */
diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c
index 75f2ef60e579..d22e62623437 100644
--- a/fs/xfs/xfs_extfree_item.c
+++ b/fs/xfs/xfs_extfree_item.c
@@ -138,7 +138,8 @@ xfs_efi_item_unpin(
if (remove) {
ASSERT(!(lip->li_flags & XFS_LI_IN_AIL));
- xfs_trans_del_item(lip);
+ if (lip->li_desc)
+ xfs_trans_del_item(lip);
xfs_efi_item_free(efip);
return;
}
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index f56d30e8040c..cec89dd5d7d2 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -612,12 +612,13 @@ out:
*
* We cannot use an inode here for this - that will push dirty state back up
* into the VFS and then periodic inode flushing will prevent log covering from
- * making progress. Hence we log a field in the superblock instead.
+ * making progress. Hence we log a field in the superblock instead and use a
+ * synchronous transaction to ensure the superblock is immediately unpinned
+ * and can be written back.
*/
int
xfs_fs_log_dummy(
- xfs_mount_t *mp,
- int flags)
+ xfs_mount_t *mp)
{
xfs_trans_t *tp;
int error;
@@ -632,8 +633,7 @@ xfs_fs_log_dummy(
/* log the UUID because it is an unchanging field */
xfs_mod_sb(tp, XFS_SB_UUID);
- if (flags & SYNC_WAIT)
- xfs_trans_set_sync(tp);
+ xfs_trans_set_sync(tp);
return xfs_trans_commit(tp, 0);
}
diff --git a/fs/xfs/xfs_fsops.h b/fs/xfs/xfs_fsops.h
index a786c5212c1e..1b6a98b66886 100644
--- a/fs/xfs/xfs_fsops.h
+++ b/fs/xfs/xfs_fsops.h
@@ -25,6 +25,6 @@ extern int xfs_fs_counts(xfs_mount_t *mp, xfs_fsop_counts_t *cnt);
extern int xfs_reserve_blocks(xfs_mount_t *mp, __uint64_t *inval,
xfs_fsop_resblks_t *outval);
extern int xfs_fs_goingdown(xfs_mount_t *mp, __uint32_t inflags);
-extern int xfs_fs_log_dummy(xfs_mount_t *mp, int flags);
+extern int xfs_fs_log_dummy(struct xfs_mount *mp);
#endif /* __XFS_FSOPS_H__ */
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 55582bd66659..8a0f044750c3 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -337,7 +337,12 @@ xfs_iomap_prealloc_size(
int shift = 0;
int64_t freesp;
- alloc_blocks = XFS_B_TO_FSB(mp, ip->i_size);
+ /*
+ * rounddown_pow_of_two() returns an undefined result
+ * if we pass in alloc_blocks = 0. Hence the "+ 1" to
+ * ensure we always pass in a non-zero value.
+ */
+ alloc_blocks = XFS_B_TO_FSB(mp, ip->i_size) + 1;
alloc_blocks = XFS_FILEOFF_MIN(MAXEXTLEN,
rounddown_pow_of_two(alloc_blocks));
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 0bf24b11d0c4..ae6fef1ff563 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -377,7 +377,7 @@ xfs_log_mount(
cmn_err(CE_NOTE, "XFS mounting filesystem %s", mp->m_fsname);
else {
cmn_err(CE_NOTE,
- "!Mounting filesystem \"%s\" in no-recovery mode. Filesystem will be inconsistent.",
+ "Mounting filesystem \"%s\" in no-recovery mode. Filesystem will be inconsistent.",
mp->m_fsname);
ASSERT(mp->m_flags & XFS_MOUNT_RDONLY);
}
diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h
index 916eb7db14d9..3bd3291ef8d2 100644
--- a/fs/xfs/xfs_log.h
+++ b/fs/xfs/xfs_log.h
@@ -191,7 +191,7 @@ void xfs_log_ticket_put(struct xlog_ticket *ticket);
xlog_tid_t xfs_log_get_trans_ident(struct xfs_trans *tp);
-int xfs_log_commit_cil(struct xfs_mount *mp, struct xfs_trans *tp,
+void xfs_log_commit_cil(struct xfs_mount *mp, struct xfs_trans *tp,
struct xfs_log_vec *log_vector,
xfs_lsn_t *commit_lsn, int flags);
bool xfs_log_item_in_current_chkpt(struct xfs_log_item *lip);
diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c
index 9dc8125d04e5..9ca59be08977 100644
--- a/fs/xfs/xfs_log_cil.c
+++ b/fs/xfs/xfs_log_cil.c
@@ -543,7 +543,7 @@ xlog_cil_push(
error = xlog_write(log, &lvhdr, tic, &ctx->start_lsn, NULL, 0);
if (error)
- goto out_abort;
+ goto out_abort_free_ticket;
/*
* now that we've written the checkpoint into the log, strictly
@@ -569,8 +569,9 @@ restart:
}
spin_unlock(&cil->xc_cil_lock);
+ /* xfs_log_done always frees the ticket on error. */
commit_lsn = xfs_log_done(log->l_mp, tic, &commit_iclog, 0);
- if (error || commit_lsn == -1)
+ if (commit_lsn == -1)
goto out_abort;
/* attach all the transactions w/ busy extents to iclog */
@@ -600,6 +601,8 @@ out_free_ticket:
kmem_free(new_ctx);
return 0;
+out_abort_free_ticket:
+ xfs_log_ticket_put(tic);
out_abort:
xlog_cil_committed(ctx, XFS_LI_ABORTED);
return XFS_ERROR(EIO);
@@ -622,7 +625,7 @@ out_abort:
* background commit, returns without it held once background commits are
* allowed again.
*/
-int
+void
xfs_log_commit_cil(
struct xfs_mount *mp,
struct xfs_trans *tp,
@@ -637,11 +640,6 @@ xfs_log_commit_cil(
if (flags & XFS_TRANS_RELEASE_LOG_RES)
log_flags = XFS_LOG_REL_PERM_RESERV;
- if (XLOG_FORCED_SHUTDOWN(log)) {
- xlog_cil_free_logvec(log_vector);
- return XFS_ERROR(EIO);
- }
-
/*
* do all the hard work of formatting items (including memory
* allocation) outside the CIL context lock. This prevents stalling CIL
@@ -701,7 +699,6 @@ xfs_log_commit_cil(
*/
if (push)
xlog_cil_push(log, 0);
- return 0;
}
/*
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 204d8e5fa7fa..aa0ebb776903 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -3800,7 +3800,7 @@ xlog_recover_finish(
log->l_flags &= ~XLOG_RECOVERY_NEEDED;
} else {
cmn_err(CE_DEBUG,
- "!Ending clean XFS mount for filesystem: %s\n",
+ "Ending clean XFS mount for filesystem: %s\n",
log->l_mp->m_fsname);
}
return 0;
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index f80a067a4658..76922793f64f 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -1137,7 +1137,7 @@ out_undo_fdblocks:
if (blkdelta)
xfs_icsb_modify_counters(mp, XFS_SBS_FDBLOCKS, -blkdelta, rsvd);
out:
- ASSERT(error = 0);
+ ASSERT(error == 0);
return;
}
@@ -1446,6 +1446,14 @@ xfs_log_item_batch_insert(
* Bulk operation version of xfs_trans_committed that takes a log vector of
* items to insert into the AIL. This uses bulk AIL insertion techniques to
* minimise lock traffic.
+ *
+ * If we are called with the aborted flag set, it is because a log write during
+ * a CIL checkpoint commit has failed. In this case, all the items in the
+ * checkpoint have already gone through IOP_COMMITED and IOP_UNLOCK, which
+ * means that checkpoint commit abort handling is treated exactly the same
+ * as an iclog write error even though we haven't started any IO yet. Hence in
+ * this case all we need to do is IOP_COMMITTED processing, followed by an
+ * IOP_UNPIN(aborted) call.
*/
void
xfs_trans_committed_bulk(
@@ -1472,6 +1480,16 @@ xfs_trans_committed_bulk(
if (XFS_LSN_CMP(item_lsn, (xfs_lsn_t)-1) == 0)
continue;
+ /*
+ * if we are aborting the operation, no point in inserting the
+ * object into the AIL as we are in a shutdown situation.
+ */
+ if (aborted) {
+ ASSERT(XFS_FORCED_SHUTDOWN(ailp->xa_mount));
+ IOP_UNPIN(lip, 1);
+ continue;
+ }
+
if (item_lsn != commit_lsn) {
/*
@@ -1503,20 +1521,24 @@ xfs_trans_committed_bulk(
}
/*
- * Called from the trans_commit code when we notice that
- * the filesystem is in the middle of a forced shutdown.
+ * Called from the trans_commit code when we notice that the filesystem is in
+ * the middle of a forced shutdown.
+ *
+ * When we are called here, we have already pinned all the items in the
+ * transaction. However, neither IOP_COMMITTING or IOP_UNLOCK has been called
+ * so we can simply walk the items in the transaction, unpin them with an abort
+ * flag and then free the items. Note that unpinning the items can result in
+ * them being freed immediately, so we need to use a safe list traversal method
+ * here.
*/
STATIC void
xfs_trans_uncommit(
struct xfs_trans *tp,
uint flags)
{
- struct xfs_log_item_desc *lidp;
+ struct xfs_log_item_desc *lidp, *n;
- list_for_each_entry(lidp, &tp->t_items, lid_trans) {
- /*
- * Unpin all but those that aren't dirty.
- */
+ list_for_each_entry_safe(lidp, n, &tp->t_items, lid_trans) {
if (lidp->lid_flags & XFS_LID_DIRTY)
IOP_UNPIN(lidp->lid_item, 1);
}
@@ -1733,7 +1755,6 @@ xfs_trans_commit_cil(
int flags)
{
struct xfs_log_vec *log_vector;
- int error;
/*
* Get each log item to allocate a vector structure for
@@ -1744,9 +1765,7 @@ xfs_trans_commit_cil(
if (!log_vector)
return ENOMEM;
- error = xfs_log_commit_cil(mp, tp, log_vector, commit_lsn, flags);
- if (error)
- return error;
+ xfs_log_commit_cil(mp, tp, log_vector, commit_lsn, flags);
current_restore_flags_nested(&tp->t_pflags, PF_FSTRANS);
xfs_trans_free(tp);