Commit 487f84f3 authored by Dave Chinner's avatar Dave Chinner Committed by Dave Chinner

xfs: introduce xfs_rw_lock() helpers for locking the inode

We need to obtain the i_mutex, i_iolock and i_ilock during the read
and write paths. Add a set of wrapper functions to neatly
encapsulate the lock ordering and shared/exclusive semantics to make
the locking easier to follow and get right.

Note that this changes some of the exclusive locking serialisation in
that serialisation will occur against the i_mutex instead of the
XFS_IOLOCK_EXCL. This does not change any behaviour, and it is
arguably more efficient to use the mutex for such serialisation than
the rw_sem.
Signed-off-by: default avatarDave Chinner <dchinner@redhat.com>
Reviewed-by: default avatarChristoph Hellwig <hch@lst.de>
parent 4c5cfd1b
...@@ -40,6 +40,40 @@ ...@@ -40,6 +40,40 @@
static const struct vm_operations_struct xfs_file_vm_ops; 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
* *
...@@ -262,22 +296,21 @@ xfs_file_aio_read( ...@@ -262,22 +296,21 @@ xfs_file_aio_read(
if (XFS_FORCED_SHUTDOWN(mp)) if (XFS_FORCED_SHUTDOWN(mp))
return -EIO; return -EIO;
if (unlikely(ioflags & IO_ISDIRECT))
mutex_lock(&inode->i_mutex);
xfs_ilock(ip, XFS_IOLOCK_SHARED);
if (unlikely(ioflags & IO_ISDIRECT)) { if (unlikely(ioflags & IO_ISDIRECT)) {
xfs_rw_ilock(ip, XFS_IOLOCK_EXCL);
if (inode->i_mapping->nrpages) { if (inode->i_mapping->nrpages) {
ret = -xfs_flushinval_pages(ip, ret = -xfs_flushinval_pages(ip,
(iocb->ki_pos & PAGE_CACHE_MASK), (iocb->ki_pos & PAGE_CACHE_MASK),
-1, FI_REMAPF_LOCKED); -1, FI_REMAPF_LOCKED);
if (ret) {
xfs_rw_iunlock(ip, XFS_IOLOCK_EXCL);
return ret;
}
} }
mutex_unlock(&inode->i_mutex); xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
if (ret) { } else
xfs_iunlock(ip, XFS_IOLOCK_SHARED); xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
return ret;
}
}
trace_xfs_file_read(ip, size, iocb->ki_pos, ioflags); trace_xfs_file_read(ip, size, iocb->ki_pos, ioflags);
...@@ -285,7 +318,7 @@ xfs_file_aio_read( ...@@ -285,7 +318,7 @@ xfs_file_aio_read(
if (ret > 0) if (ret > 0)
XFS_STATS_ADD(xs_read_bytes, ret); XFS_STATS_ADD(xs_read_bytes, ret);
xfs_iunlock(ip, XFS_IOLOCK_SHARED); xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
return ret; return ret;
} }
...@@ -309,7 +342,7 @@ xfs_file_splice_read( ...@@ -309,7 +342,7 @@ xfs_file_splice_read(
if (XFS_FORCED_SHUTDOWN(ip->i_mount)) if (XFS_FORCED_SHUTDOWN(ip->i_mount))
return -EIO; return -EIO;
xfs_ilock(ip, XFS_IOLOCK_SHARED); xfs_rw_ilock(ip, XFS_IOLOCK_SHARED);
trace_xfs_file_splice_read(ip, count, *ppos, ioflags); trace_xfs_file_splice_read(ip, count, *ppos, ioflags);
...@@ -317,7 +350,7 @@ xfs_file_splice_read( ...@@ -317,7 +350,7 @@ xfs_file_splice_read(
if (ret > 0) if (ret > 0)
XFS_STATS_ADD(xs_read_bytes, ret); XFS_STATS_ADD(xs_read_bytes, ret);
xfs_iunlock(ip, XFS_IOLOCK_SHARED); xfs_rw_iunlock(ip, XFS_IOLOCK_SHARED);
return ret; return ret;
} }
...@@ -338,10 +371,10 @@ xfs_aio_write_isize_update( ...@@ -338,10 +371,10 @@ xfs_aio_write_isize_update(
*ppos = isize; *ppos = isize;
if (*ppos > ip->i_size) { if (*ppos > ip->i_size) {
xfs_ilock(ip, XFS_ILOCK_EXCL); xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
if (*ppos > ip->i_size) if (*ppos > ip->i_size)
ip->i_size = *ppos; ip->i_size = *ppos;
xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
} }
} }
...@@ -356,14 +389,22 @@ xfs_aio_write_newsize_update( ...@@ -356,14 +389,22 @@ xfs_aio_write_newsize_update(
struct xfs_inode *ip) struct xfs_inode *ip)
{ {
if (ip->i_new_size) { if (ip->i_new_size) {
xfs_ilock(ip, XFS_ILOCK_EXCL); xfs_rw_ilock(ip, XFS_ILOCK_EXCL);
ip->i_new_size = 0; ip->i_new_size = 0;
if (ip->i_d.di_size > ip->i_size) if (ip->i_d.di_size > ip->i_size)
ip->i_d.di_size = ip->i_size; ip->i_d.di_size = ip->i_size;
xfs_iunlock(ip, XFS_ILOCK_EXCL); 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 STATIC ssize_t
xfs_file_splice_write( xfs_file_splice_write(
struct pipe_inode_info *pipe, struct pipe_inode_info *pipe,
...@@ -604,7 +645,6 @@ xfs_file_aio_write( ...@@ -604,7 +645,6 @@ xfs_file_aio_write(
xfs_fsize_t new_size; xfs_fsize_t new_size;
int iolock; int iolock;
size_t ocount = 0, count; size_t ocount = 0, count;
int need_i_mutex;
XFS_STATS_INC(xs_write_calls); XFS_STATS_INC(xs_write_calls);
...@@ -631,21 +671,17 @@ xfs_file_aio_write( ...@@ -631,21 +671,17 @@ xfs_file_aio_write(
relock: relock:
if (ioflags & IO_ISDIRECT) { if (ioflags & IO_ISDIRECT) {
iolock = XFS_IOLOCK_SHARED; iolock = XFS_IOLOCK_SHARED;
need_i_mutex = 0;
} else { } else {
iolock = XFS_IOLOCK_EXCL; iolock = XFS_IOLOCK_EXCL;
need_i_mutex = 1;
mutex_lock(&inode->i_mutex);
} }
xfs_ilock(ip, XFS_ILOCK_EXCL|iolock);
start: start:
xfs_rw_ilock(ip, XFS_ILOCK_EXCL|iolock);
ret = generic_write_checks(file, &pos, &count, ret = generic_write_checks(file, &pos, &count,
S_ISBLK(inode->i_mode)); S_ISBLK(inode->i_mode));
if (ret) { if (ret) {
xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock); xfs_rw_iunlock(ip, XFS_ILOCK_EXCL|iolock);
goto out_unlock_mutex; return ret;
} }
if (ioflags & IO_ISDIRECT) { if (ioflags & IO_ISDIRECT) {
...@@ -654,16 +690,20 @@ xfs_file_aio_write( ...@@ -654,16 +690,20 @@ xfs_file_aio_write(
mp->m_rtdev_targp : mp->m_ddev_targp; mp->m_rtdev_targp : mp->m_ddev_targp;
if ((pos & target->bt_smask) || (count & target->bt_smask)) { if ((pos & target->bt_smask) || (count & target->bt_smask)) {
xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock); xfs_rw_iunlock(ip, XFS_ILOCK_EXCL|iolock);
return XFS_ERROR(-EINVAL); return XFS_ERROR(-EINVAL);
} }
if (!need_i_mutex && (mapping->nrpages || pos > ip->i_size)) { /*
xfs_iunlock(ip, XFS_ILOCK_EXCL|iolock); * For direct I/O, 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. Upgrade the I/O lock and start again.
*/
if (iolock != XFS_IOLOCK_EXCL &&
(mapping->nrpages || pos > ip->i_size)) {
xfs_rw_iunlock(ip, XFS_ILOCK_EXCL|iolock);
iolock = XFS_IOLOCK_EXCL; iolock = XFS_IOLOCK_EXCL;
need_i_mutex = 1;
mutex_lock(&inode->i_mutex);
xfs_ilock(ip, XFS_ILOCK_EXCL|iolock);
goto start; goto start;
} }
} }
...@@ -687,11 +727,11 @@ xfs_file_aio_write( ...@@ -687,11 +727,11 @@ xfs_file_aio_write(
if (pos > ip->i_size) { if (pos > ip->i_size) {
ret = -xfs_zero_eof(ip, pos, ip->i_size); ret = -xfs_zero_eof(ip, pos, ip->i_size);
if (ret) { if (ret) {
xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
goto out_unlock_internal; goto out_unlock_internal;
} }
} }
xfs_iunlock(ip, XFS_ILOCK_EXCL); xfs_rw_iunlock(ip, XFS_ILOCK_EXCL);
/* /*
* If we're writing the file then make sure to clear the * If we're writing the file then make sure to clear the
...@@ -708,7 +748,7 @@ xfs_file_aio_write( ...@@ -708,7 +748,7 @@ xfs_file_aio_write(
if ((ioflags & IO_ISDIRECT)) { if ((ioflags & IO_ISDIRECT)) {
if (mapping->nrpages) { if (mapping->nrpages) {
WARN_ON(need_i_mutex == 0); WARN_ON(iolock != XFS_IOLOCK_EXCL);
ret = -xfs_flushinval_pages(ip, ret = -xfs_flushinval_pages(ip,
(pos & PAGE_CACHE_MASK), (pos & PAGE_CACHE_MASK),
-1, FI_REMAPF_LOCKED); -1, FI_REMAPF_LOCKED);
...@@ -716,13 +756,10 @@ xfs_file_aio_write( ...@@ -716,13 +756,10 @@ xfs_file_aio_write(
goto out_unlock_internal; goto out_unlock_internal;
} }
if (need_i_mutex) { if (iolock == XFS_IOLOCK_EXCL) {
/* demote the lock now the cached pages are gone */ /* demote the lock now the cached pages are gone */
xfs_ilock_demote(ip, XFS_IOLOCK_EXCL); xfs_rw_ilock_demote(ip, XFS_IOLOCK_EXCL);
mutex_unlock(&inode->i_mutex);
iolock = XFS_IOLOCK_SHARED; iolock = XFS_IOLOCK_SHARED;
need_i_mutex = 0;
} }
trace_xfs_file_direct_write(ip, count, iocb->ki_pos, ioflags); trace_xfs_file_direct_write(ip, count, iocb->ki_pos, ioflags);
...@@ -740,7 +777,7 @@ xfs_file_aio_write( ...@@ -740,7 +777,7 @@ xfs_file_aio_write(
count -= ret; count -= ret;
ioflags &= ~IO_ISDIRECT; ioflags &= ~IO_ISDIRECT;
xfs_iunlock(ip, iolock); xfs_rw_iunlock(ip, iolock);
goto relock; goto relock;
} }
} else { } else {
...@@ -775,14 +812,9 @@ xfs_file_aio_write( ...@@ -775,14 +812,9 @@ xfs_file_aio_write(
loff_t end = pos + ret - 1; loff_t end = pos + ret - 1;
int error, error2; int error, error2;
xfs_iunlock(ip, iolock); xfs_rw_iunlock(ip, iolock);
if (need_i_mutex)
mutex_unlock(&inode->i_mutex);
error = filemap_write_and_wait_range(mapping, pos, end); error = filemap_write_and_wait_range(mapping, pos, end);
if (need_i_mutex) xfs_rw_ilock(ip, iolock);
mutex_lock(&inode->i_mutex);
xfs_ilock(ip, iolock);
error2 = -xfs_file_fsync(file, error2 = -xfs_file_fsync(file,
(file->f_flags & __O_SYNC) ? 0 : 1); (file->f_flags & __O_SYNC) ? 0 : 1);
...@@ -794,10 +826,7 @@ xfs_file_aio_write( ...@@ -794,10 +826,7 @@ xfs_file_aio_write(
out_unlock_internal: out_unlock_internal:
xfs_aio_write_newsize_update(ip); xfs_aio_write_newsize_update(ip);
xfs_iunlock(ip, iolock); xfs_rw_iunlock(ip, iolock);
out_unlock_mutex:
if (need_i_mutex)
mutex_unlock(&inode->i_mutex);
return ret; return ret;
} }
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment