2005-04-17 00:20:36 +02:00
|
|
|
/*
|
2005-11-02 04:58:39 +01:00
|
|
|
* Copyright (c) 2000-2005 Silicon Graphics, Inc.
|
|
|
|
* All Rights Reserved.
|
2005-04-17 00:20:36 +02:00
|
|
|
*
|
2005-11-02 04:58:39 +01:00
|
|
|
* This program is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of the GNU General Public License as
|
2005-04-17 00:20:36 +02:00
|
|
|
* published by the Free Software Foundation.
|
|
|
|
*
|
2005-11-02 04:58:39 +01:00
|
|
|
* 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.
|
2005-04-17 00:20:36 +02:00
|
|
|
*
|
2005-11-02 04:58:39 +01:00
|
|
|
* 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
|
2005-04-17 00:20:36 +02:00
|
|
|
*/
|
|
|
|
#include "xfs.h"
|
|
|
|
#include "xfs_fs.h"
|
2013-10-23 01:36:05 +02:00
|
|
|
#include "xfs_shared.h"
|
2013-10-23 01:50:10 +02:00
|
|
|
#include "xfs_format.h"
|
|
|
|
#include "xfs_log_format.h"
|
|
|
|
#include "xfs_trans_resv.h"
|
2005-04-17 00:20:36 +02:00
|
|
|
#include "xfs_mount.h"
|
2013-10-15 00:17:51 +02:00
|
|
|
#include "xfs_da_format.h"
|
2005-04-17 00:20:36 +02:00
|
|
|
#include "xfs_inode.h"
|
|
|
|
#include "xfs_bmap.h"
|
2013-08-12 12:49:42 +02:00
|
|
|
#include "xfs_bmap_util.h"
|
2013-10-23 01:50:10 +02:00
|
|
|
#include "xfs_acl.h"
|
|
|
|
#include "xfs_quota.h"
|
2005-04-17 00:20:36 +02:00
|
|
|
#include "xfs_error.h"
|
|
|
|
#include "xfs_attr.h"
|
2013-10-23 01:50:10 +02:00
|
|
|
#include "xfs_trans.h"
|
2009-12-15 00:14:59 +01:00
|
|
|
#include "xfs_trace.h"
|
2012-11-06 15:50:38 +01:00
|
|
|
#include "xfs_icache.h"
|
2013-08-12 12:49:45 +02:00
|
|
|
#include "xfs_symlink.h"
|
2013-08-12 12:50:09 +02:00
|
|
|
#include "xfs_da_btree.h"
|
2014-12-03 23:43:17 +01:00
|
|
|
#include "xfs_dir2.h"
|
2013-12-18 01:22:40 +01:00
|
|
|
#include "xfs_trans_space.h"
|
2015-02-16 01:59:50 +01:00
|
|
|
#include "xfs_pnfs.h"
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2006-01-11 21:17:46 +01:00
|
|
|
#include <linux/capability.h>
|
2005-04-17 00:20:36 +02:00
|
|
|
#include <linux/xattr.h>
|
2009-06-10 17:07:47 +02:00
|
|
|
#include <linux/posix_acl.h>
|
2006-01-11 05:35:44 +01:00
|
|
|
#include <linux/security.h>
|
2008-11-28 04:23:35 +01:00
|
|
|
#include <linux/fiemap.h>
|
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 09:04:11 +01:00
|
|
|
#include <linux/slab.h>
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2014-02-27 06:51:39 +01:00
|
|
|
/*
|
|
|
|
* Directories have different lock order w.r.t. mmap_sem compared to regular
|
|
|
|
* files. This is due to readdir potentially triggering page faults on a user
|
|
|
|
* buffer inside filldir(), and this happens with the ilock on the directory
|
|
|
|
* held. For regular files, the lock order is the other way around - the
|
|
|
|
* mmap_sem is taken during the page fault, and then we lock the ilock to do
|
|
|
|
* block mapping. Hence we need a different class for the directory ilock so
|
|
|
|
* that lockdep can tell them apart.
|
|
|
|
*/
|
|
|
|
static struct lock_class_key xfs_nondir_ilock_class;
|
|
|
|
static struct lock_class_key xfs_dir_ilock_class;
|
|
|
|
|
2012-03-07 05:50:19 +01:00
|
|
|
static int
|
|
|
|
xfs_initxattrs(
|
|
|
|
struct inode *inode,
|
|
|
|
const struct xattr *xattr_array,
|
|
|
|
void *fs_info)
|
2011-06-06 21:29:25 +02:00
|
|
|
{
|
2012-03-07 05:50:19 +01:00
|
|
|
const struct xattr *xattr;
|
|
|
|
struct xfs_inode *ip = XFS_I(inode);
|
|
|
|
int error = 0;
|
2011-06-06 21:29:25 +02:00
|
|
|
|
|
|
|
for (xattr = xattr_array; xattr->name != NULL; xattr++) {
|
2014-06-25 06:58:08 +02:00
|
|
|
error = xfs_attr_set(ip, xattr->name, xattr->value,
|
2014-05-15 01:22:21 +02:00
|
|
|
xattr->value_len, ATTR_SECURE);
|
2011-06-06 21:29:25 +02:00
|
|
|
if (error < 0)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
return error;
|
|
|
|
}
|
|
|
|
|
2006-01-11 05:35:44 +01:00
|
|
|
/*
|
|
|
|
* Hook in SELinux. This is not quite correct yet, what we really need
|
|
|
|
* here (as we do for default ACLs) is a mechanism by which creation of
|
|
|
|
* these attrs can be journalled at inode creation time (along with the
|
|
|
|
* inode, of course, such that log replay can't cause these to be lost).
|
|
|
|
*/
|
2011-06-06 21:29:25 +02:00
|
|
|
|
2006-01-11 05:35:44 +01:00
|
|
|
STATIC int
|
2006-03-14 04:00:51 +01:00
|
|
|
xfs_init_security(
|
2008-03-06 03:46:43 +01:00
|
|
|
struct inode *inode,
|
2011-02-01 17:05:39 +01:00
|
|
|
struct inode *dir,
|
|
|
|
const struct qstr *qstr)
|
2006-01-11 05:35:44 +01:00
|
|
|
{
|
2014-06-25 06:58:08 +02:00
|
|
|
return security_inode_init_security(inode, dir, qstr,
|
2014-05-15 01:22:21 +02:00
|
|
|
&xfs_initxattrs, NULL);
|
2006-01-11 05:35:44 +01:00
|
|
|
}
|
|
|
|
|
2008-04-10 04:22:07 +02:00
|
|
|
static void
|
|
|
|
xfs_dentry_to_name(
|
|
|
|
struct xfs_name *namep,
|
2013-08-12 12:50:09 +02:00
|
|
|
struct dentry *dentry,
|
|
|
|
int mode)
|
2008-04-10 04:22:07 +02:00
|
|
|
{
|
|
|
|
namep->name = dentry->d_name.name;
|
|
|
|
namep->len = dentry->d_name.len;
|
2013-08-12 12:50:09 +02:00
|
|
|
namep->type = xfs_mode_to_ftype[(mode & S_IFMT) >> S_SHIFT];
|
2008-04-10 04:22:07 +02:00
|
|
|
}
|
|
|
|
|
2007-02-10 08:34:56 +01:00
|
|
|
STATIC void
|
2006-03-14 04:00:51 +01:00
|
|
|
xfs_cleanup_inode(
|
2007-08-29 02:58:01 +02:00
|
|
|
struct inode *dir,
|
2008-03-06 03:46:43 +01:00
|
|
|
struct inode *inode,
|
2008-06-23 05:25:17 +02:00
|
|
|
struct dentry *dentry)
|
2006-02-01 02:14:34 +01:00
|
|
|
{
|
2008-04-10 04:22:07 +02:00
|
|
|
struct xfs_name teardown;
|
2006-02-01 02:14:34 +01:00
|
|
|
|
|
|
|
/* Oh, the horror.
|
2006-03-14 03:33:36 +01:00
|
|
|
* If we can't add the ACL or we fail in
|
2006-03-14 04:00:51 +01:00
|
|
|
* xfs_init_security we must back out.
|
2006-02-01 02:14:34 +01:00
|
|
|
* ENOSPC can hit here, among other things.
|
|
|
|
*/
|
2013-08-12 12:50:09 +02:00
|
|
|
xfs_dentry_to_name(&teardown, dentry, 0);
|
2006-02-01 02:14:34 +01:00
|
|
|
|
2008-06-23 05:25:17 +02:00
|
|
|
xfs_remove(XFS_I(dir), &teardown, XFS_I(inode));
|
2006-02-01 02:14:34 +01:00
|
|
|
}
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
STATIC int
|
2014-05-05 23:34:28 +02:00
|
|
|
xfs_generic_create(
|
2005-04-17 00:20:36 +02:00
|
|
|
struct inode *dir,
|
|
|
|
struct dentry *dentry,
|
2011-07-26 07:52:52 +02:00
|
|
|
umode_t mode,
|
2014-05-05 23:34:28 +02:00
|
|
|
dev_t rdev,
|
|
|
|
bool tmpfile) /* unnamed file */
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
2008-03-06 03:44:35 +01:00
|
|
|
struct inode *inode;
|
2008-03-06 03:46:05 +01:00
|
|
|
struct xfs_inode *ip = NULL;
|
2013-12-20 14:16:50 +01:00
|
|
|
struct posix_acl *default_acl, *acl;
|
2008-04-10 04:22:07 +02:00
|
|
|
struct xfs_name name;
|
2005-04-17 00:20:36 +02:00
|
|
|
int error;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Irix uses Missed'em'V split, but doesn't want to see
|
|
|
|
* the upper 5 bits of (14bit) major.
|
|
|
|
*/
|
2009-02-09 08:38:02 +01:00
|
|
|
if (S_ISCHR(mode) || S_ISBLK(mode)) {
|
|
|
|
if (unlikely(!sysv_valid_dev(rdev) || MAJOR(rdev) & ~0x1ff))
|
|
|
|
return -EINVAL;
|
|
|
|
rdev = sysv_encode_dev(rdev);
|
|
|
|
} else {
|
|
|
|
rdev = 0;
|
|
|
|
}
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2013-12-20 14:16:50 +01:00
|
|
|
error = posix_acl_create(dir, &mode, &default_acl, &acl);
|
|
|
|
if (error)
|
|
|
|
return error;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2014-05-05 23:34:28 +02:00
|
|
|
if (!tmpfile) {
|
|
|
|
xfs_dentry_to_name(&name, dentry, mode);
|
|
|
|
error = xfs_create(XFS_I(dir), &name, mode, rdev, &ip);
|
|
|
|
} else {
|
|
|
|
error = xfs_create_tmpfile(XFS_I(dir), dentry, mode, &ip);
|
|
|
|
}
|
2008-03-06 03:44:35 +01:00
|
|
|
if (unlikely(error))
|
|
|
|
goto out_free_acl;
|
2006-01-11 05:35:44 +01:00
|
|
|
|
2008-08-13 07:45:15 +02:00
|
|
|
inode = VFS_I(ip);
|
2008-03-06 03:46:05 +01:00
|
|
|
|
2011-02-01 17:05:39 +01:00
|
|
|
error = xfs_init_security(inode, dir, &dentry->d_name);
|
2008-03-06 03:44:35 +01:00
|
|
|
if (unlikely(error))
|
|
|
|
goto out_cleanup_inode;
|
|
|
|
|
2013-12-20 14:16:50 +01:00
|
|
|
#ifdef CONFIG_XFS_POSIX_ACL
|
2008-03-06 03:44:35 +01:00
|
|
|
if (default_acl) {
|
2014-06-25 06:58:08 +02:00
|
|
|
error = xfs_set_acl(inode, default_acl, ACL_TYPE_DEFAULT);
|
2013-12-20 14:16:50 +01:00
|
|
|
if (error)
|
2008-03-06 03:44:35 +01:00
|
|
|
goto out_cleanup_inode;
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
2013-12-20 14:16:50 +01:00
|
|
|
if (acl) {
|
2014-06-25 06:58:08 +02:00
|
|
|
error = xfs_set_acl(inode, acl, ACL_TYPE_ACCESS);
|
2013-12-20 14:16:50 +01:00
|
|
|
if (error)
|
|
|
|
goto out_cleanup_inode;
|
|
|
|
}
|
|
|
|
#endif
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2014-05-05 23:34:28 +02:00
|
|
|
if (tmpfile)
|
|
|
|
d_tmpfile(dentry, inode);
|
|
|
|
else
|
|
|
|
d_instantiate(dentry, inode);
|
|
|
|
|
2015-02-23 12:38:08 +01:00
|
|
|
xfs_finish_inode_setup(ip);
|
|
|
|
|
2013-12-20 14:16:50 +01:00
|
|
|
out_free_acl:
|
|
|
|
if (default_acl)
|
|
|
|
posix_acl_release(default_acl);
|
|
|
|
if (acl)
|
|
|
|
posix_acl_release(acl);
|
2014-06-25 06:58:08 +02:00
|
|
|
return error;
|
2008-03-06 03:44:35 +01:00
|
|
|
|
|
|
|
out_cleanup_inode:
|
2015-02-23 12:38:08 +01:00
|
|
|
xfs_finish_inode_setup(ip);
|
2014-05-05 23:34:28 +02:00
|
|
|
if (!tmpfile)
|
|
|
|
xfs_cleanup_inode(dir, inode, dentry);
|
|
|
|
iput(inode);
|
2013-12-20 14:16:50 +01:00
|
|
|
goto out_free_acl;
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
2014-05-05 23:34:28 +02:00
|
|
|
STATIC int
|
|
|
|
xfs_vn_mknod(
|
|
|
|
struct inode *dir,
|
|
|
|
struct dentry *dentry,
|
|
|
|
umode_t mode,
|
|
|
|
dev_t rdev)
|
|
|
|
{
|
|
|
|
return xfs_generic_create(dir, dentry, mode, rdev, false);
|
|
|
|
}
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
STATIC int
|
2006-03-14 04:00:51 +01:00
|
|
|
xfs_vn_create(
|
2005-04-17 00:20:36 +02:00
|
|
|
struct inode *dir,
|
|
|
|
struct dentry *dentry,
|
2011-07-26 07:42:34 +02:00
|
|
|
umode_t mode,
|
2012-06-11 00:05:36 +02:00
|
|
|
bool flags)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
2006-03-14 04:00:51 +01:00
|
|
|
return xfs_vn_mknod(dir, dentry, mode, 0);
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
STATIC int
|
2006-03-14 04:00:51 +01:00
|
|
|
xfs_vn_mkdir(
|
2005-04-17 00:20:36 +02:00
|
|
|
struct inode *dir,
|
|
|
|
struct dentry *dentry,
|
2011-07-26 07:41:39 +02:00
|
|
|
umode_t mode)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
2006-03-14 04:00:51 +01:00
|
|
|
return xfs_vn_mknod(dir, dentry, mode|S_IFDIR, 0);
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
STATIC struct dentry *
|
2006-03-14 04:00:51 +01:00
|
|
|
xfs_vn_lookup(
|
2005-04-17 00:20:36 +02:00
|
|
|
struct inode *dir,
|
|
|
|
struct dentry *dentry,
|
2012-06-10 23:13:09 +02:00
|
|
|
unsigned int flags)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
2008-03-06 03:46:25 +01:00
|
|
|
struct xfs_inode *cip;
|
2008-04-10 04:22:07 +02:00
|
|
|
struct xfs_name name;
|
2005-04-17 00:20:36 +02:00
|
|
|
int error;
|
|
|
|
|
|
|
|
if (dentry->d_name.len >= MAXNAMELEN)
|
|
|
|
return ERR_PTR(-ENAMETOOLONG);
|
|
|
|
|
2013-08-12 12:50:09 +02:00
|
|
|
xfs_dentry_to_name(&name, dentry, 0);
|
2008-05-21 08:58:22 +02:00
|
|
|
error = xfs_lookup(XFS_I(dir), &name, &cip, NULL);
|
2006-06-09 09:00:52 +02:00
|
|
|
if (unlikely(error)) {
|
2014-06-25 06:58:08 +02:00
|
|
|
if (unlikely(error != -ENOENT))
|
|
|
|
return ERR_PTR(error);
|
2005-04-17 00:20:36 +02:00
|
|
|
d_add(dentry, NULL);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2008-08-13 07:45:15 +02:00
|
|
|
return d_splice_alias(VFS_I(cip), dentry);
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
2008-05-21 08:58:22 +02:00
|
|
|
STATIC struct dentry *
|
|
|
|
xfs_vn_ci_lookup(
|
|
|
|
struct inode *dir,
|
|
|
|
struct dentry *dentry,
|
2012-06-10 23:13:09 +02:00
|
|
|
unsigned int flags)
|
2008-05-21 08:58:22 +02:00
|
|
|
{
|
|
|
|
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);
|
|
|
|
|
2013-08-12 12:50:09 +02:00
|
|
|
xfs_dentry_to_name(&xname, dentry, 0);
|
2008-05-21 08:58:22 +02:00
|
|
|
error = xfs_lookup(XFS_I(dir), &xname, &ip, &ci_name);
|
|
|
|
if (unlikely(error)) {
|
2014-06-25 06:58:08 +02:00
|
|
|
if (unlikely(error != -ENOENT))
|
|
|
|
return ERR_PTR(error);
|
2008-05-22 09:21:40 +02:00
|
|
|
/*
|
|
|
|
* call d_add(dentry, NULL) here when d_drop_negative_children
|
|
|
|
* is called in xfs_vn_mknod (ie. allow negative dentries
|
|
|
|
* with CI filesystems).
|
|
|
|
*/
|
2008-05-21 08:58:22 +02:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* if exact match, just splice and exit */
|
|
|
|
if (!ci_name.name)
|
2008-08-13 07:45:15 +02:00
|
|
|
return d_splice_alias(VFS_I(ip), dentry);
|
2008-05-21 08:58:22 +02:00
|
|
|
|
|
|
|
/* else case-insensitive match... */
|
|
|
|
dname.name = ci_name.name;
|
|
|
|
dname.len = ci_name.len;
|
2008-08-07 23:49:07 +02:00
|
|
|
dentry = d_add_ci(dentry, VFS_I(ip), &dname);
|
2008-05-21 08:58:22 +02:00
|
|
|
kmem_free(ci_name.name);
|
|
|
|
return dentry;
|
|
|
|
}
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
STATIC int
|
2006-03-14 04:00:51 +01:00
|
|
|
xfs_vn_link(
|
2005-04-17 00:20:36 +02:00
|
|
|
struct dentry *old_dentry,
|
|
|
|
struct inode *dir,
|
|
|
|
struct dentry *dentry)
|
|
|
|
{
|
2015-03-17 23:25:59 +01:00
|
|
|
struct inode *inode = d_inode(old_dentry);
|
2008-04-10 04:22:07 +02:00
|
|
|
struct xfs_name name;
|
2005-04-17 00:20:36 +02:00
|
|
|
int error;
|
|
|
|
|
2013-08-12 12:50:09 +02:00
|
|
|
xfs_dentry_to_name(&name, dentry, inode->i_mode);
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2008-04-10 04:22:07 +02:00
|
|
|
error = xfs_link(XFS_I(dir), XFS_I(inode), &name);
|
2008-12-03 12:20:27 +01:00
|
|
|
if (unlikely(error))
|
2014-06-25 06:58:08 +02:00
|
|
|
return error;
|
2008-03-06 03:46:12 +01:00
|
|
|
|
2010-10-23 17:11:40 +02:00
|
|
|
ihold(inode);
|
2008-03-06 03:46:12 +01:00
|
|
|
d_instantiate(dentry, inode);
|
|
|
|
return 0;
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
STATIC int
|
2006-03-14 04:00:51 +01:00
|
|
|
xfs_vn_unlink(
|
2005-04-17 00:20:36 +02:00
|
|
|
struct inode *dir,
|
|
|
|
struct dentry *dentry)
|
|
|
|
{
|
2008-04-10 04:22:07 +02:00
|
|
|
struct xfs_name name;
|
2005-04-17 00:20:36 +02:00
|
|
|
int error;
|
|
|
|
|
2013-08-12 12:50:09 +02:00
|
|
|
xfs_dentry_to_name(&name, dentry, 0);
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2015-03-17 23:25:59 +01:00
|
|
|
error = xfs_remove(XFS_I(dir), &name, XFS_I(d_inode(dentry)));
|
2008-06-23 05:25:25 +02:00
|
|
|
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;
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
STATIC int
|
2006-03-14 04:00:51 +01:00
|
|
|
xfs_vn_symlink(
|
2005-04-17 00:20:36 +02:00
|
|
|
struct inode *dir,
|
|
|
|
struct dentry *dentry,
|
|
|
|
const char *symname)
|
|
|
|
{
|
2008-03-06 03:46:19 +01:00
|
|
|
struct inode *inode;
|
|
|
|
struct xfs_inode *cip = NULL;
|
2008-04-10 04:22:07 +02:00
|
|
|
struct xfs_name name;
|
2005-04-17 00:20:36 +02:00
|
|
|
int error;
|
2011-07-26 08:50:15 +02:00
|
|
|
umode_t mode;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2007-10-11 10:09:12 +02:00
|
|
|
mode = S_IFLNK |
|
2009-03-30 01:08:22 +02:00
|
|
|
(irix_symlink_mode ? 0777 & ~current_umask() : S_IRWXUGO);
|
2013-08-12 12:50:09 +02:00
|
|
|
xfs_dentry_to_name(&name, dentry, mode);
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2010-10-06 20:41:17 +02:00
|
|
|
error = xfs_symlink(XFS_I(dir), &name, symname, mode, &cip);
|
2008-03-06 03:46:19 +01:00
|
|
|
if (unlikely(error))
|
|
|
|
goto out;
|
|
|
|
|
2008-08-13 07:45:15 +02:00
|
|
|
inode = VFS_I(cip);
|
2008-03-06 03:46:19 +01:00
|
|
|
|
2011-02-01 17:05:39 +01:00
|
|
|
error = xfs_init_security(inode, dir, &dentry->d_name);
|
2008-03-06 03:46:19 +01:00
|
|
|
if (unlikely(error))
|
|
|
|
goto out_cleanup_inode;
|
|
|
|
|
|
|
|
d_instantiate(dentry, inode);
|
2015-02-23 12:38:08 +01:00
|
|
|
xfs_finish_inode_setup(cip);
|
2008-03-06 03:46:19 +01:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
out_cleanup_inode:
|
2015-02-23 12:38:08 +01:00
|
|
|
xfs_finish_inode_setup(cip);
|
2008-06-23 05:25:17 +02:00
|
|
|
xfs_cleanup_inode(dir, inode, dentry);
|
2014-05-05 23:34:28 +02:00
|
|
|
iput(inode);
|
2008-03-06 03:46:19 +01:00
|
|
|
out:
|
2014-06-25 06:58:08 +02:00
|
|
|
return error;
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
STATIC int
|
2006-03-14 04:00:51 +01:00
|
|
|
xfs_vn_rename(
|
2005-04-17 00:20:36 +02:00
|
|
|
struct inode *odir,
|
|
|
|
struct dentry *odentry,
|
|
|
|
struct inode *ndir,
|
2014-12-23 22:51:38 +01:00
|
|
|
struct dentry *ndentry,
|
|
|
|
unsigned int flags)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
2015-03-17 23:25:59 +01:00
|
|
|
struct inode *new_inode = d_inode(ndentry);
|
2014-12-23 22:51:42 +01:00
|
|
|
int omode = 0;
|
2008-04-10 04:22:07 +02:00
|
|
|
struct xfs_name oname;
|
|
|
|
struct xfs_name nname;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
xfs: add RENAME_WHITEOUT support
Whiteouts are used by overlayfs - it has a crazy convention that a
whiteout is a character device inode with a major:minor of 0:0.
Because it's not documented anywhere, here's an example of what
RENAME_WHITEOUT does on ext4:
# echo foo > /mnt/scratch/foo
# echo bar > /mnt/scratch/bar
# ls -l /mnt/scratch
total 24
-rw-r--r-- 1 root root 4 Feb 11 20:22 bar
-rw-r--r-- 1 root root 4 Feb 11 20:22 foo
drwx------ 2 root root 16384 Feb 11 20:18 lost+found
# src/renameat2 -w /mnt/scratch/foo /mnt/scratch/bar
# ls -l /mnt/scratch
total 20
-rw-r--r-- 1 root root 4 Feb 11 20:22 bar
c--------- 1 root root 0, 0 Feb 11 20:23 foo
drwx------ 2 root root 16384 Feb 11 20:18 lost+found
# cat /mnt/scratch/bar
foo
#
In XFS rename terms, the operation that has been done is that source
(foo) has been moved to the target (bar), which is like a nomal
rename operation, but rather than the source being removed, it have
been replaced with a whiteout.
We can't allocate whiteout inodes within the rename transaction due
to allocation being a multi-commit transaction: rename needs to
be a single, atomic commit. Hence we have several options here, form
most efficient to least efficient:
- use DT_WHT in the target dirent and do no whiteout inode
allocation. The main issue with this approach is that we need
hooks in lookup to create a virtual chardev inode to present
to userspace and in places where we might need to modify the
dirent e.g. unlink. Overlayfs also needs to be taught about
DT_WHT. Most invasive change, lowest overhead.
- create a special whiteout inode in the root directory (e.g. a
".wino" dirent) and then hardlink every new whiteout to it.
This means we only need to create a single whiteout inode, and
rename simply creates a hardlink to it. We can use DT_WHT for
these, though using DT_CHR means we won't have to modify
overlayfs, nor anything in userspace. Downside is we have to
look up the whiteout inode on every operation and create it if
it doesn't exist.
- copy ext4: create a special whiteout chardev inode for every
whiteout. This is more complex than the above options because
of the lack of atomicity between inode creation and the rename
operation, requiring us to create a tmpfile inode and then
linking it into the directory structure during the rename. At
least with a tmpfile inode crashes between the create and
rename doesn't leave unreferenced inodes or directory
pollution around.
By far the simplest thing to do in the short term is to copy ext4.
While it is the most inefficient way of supporting whiteouts, but as
an initial implementation we can simply reuse existing functions and
add a small amount of extra code the the rename operation.
When we get full whiteout support in the VFS (via the dentry cache)
we can then look to supporting DT_WHT method outlined as the first
method of supporting whiteouts. But until then, we'll stick with
what overlayfs expects us to be: dumb and stupid.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
2015-03-25 04:08:08 +01:00
|
|
|
if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE | RENAME_WHITEOUT))
|
2014-12-23 22:51:38 +01:00
|
|
|
return -EINVAL;
|
|
|
|
|
2014-12-23 22:51:42 +01:00
|
|
|
/* if we are exchanging files, we need to set i_mode of both files */
|
|
|
|
if (flags & RENAME_EXCHANGE)
|
2015-03-17 23:25:59 +01:00
|
|
|
omode = d_inode(ndentry)->i_mode;
|
2014-12-23 22:51:42 +01:00
|
|
|
|
|
|
|
xfs_dentry_to_name(&oname, odentry, omode);
|
2015-03-17 23:25:59 +01:00
|
|
|
xfs_dentry_to_name(&nname, ndentry, d_inode(odentry)->i_mode);
|
2008-04-10 04:22:07 +02:00
|
|
|
|
2015-03-17 23:25:59 +01:00
|
|
|
return xfs_rename(XFS_I(odir), &oname, XFS_I(d_inode(odentry)),
|
2014-12-23 22:51:38 +01:00
|
|
|
XFS_I(ndir), &nname,
|
2014-12-23 22:51:42 +01:00
|
|
|
new_inode ? XFS_I(new_inode) : NULL, flags);
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* careful here - this function can get called recursively, so
|
|
|
|
* we need to be very careful about how much stack we use.
|
|
|
|
* uio is kmalloced for this reason...
|
|
|
|
*/
|
2015-05-02 19:32:22 +02:00
|
|
|
STATIC const char *
|
2006-03-14 04:00:51 +01:00
|
|
|
xfs_vn_follow_link(
|
2005-04-17 00:20:36 +02:00
|
|
|
struct dentry *dentry,
|
2015-05-02 19:37:52 +02:00
|
|
|
void **cookie)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
|
|
|
char *link;
|
2007-08-28 05:59:03 +02:00
|
|
|
int error = -ENOMEM;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2006-09-27 10:49:39 +02:00
|
|
|
link = kmalloc(MAXPATHLEN+1, GFP_KERNEL);
|
2007-08-28 05:59:03 +02:00
|
|
|
if (!link)
|
|
|
|
goto out_err;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2015-03-17 23:25:59 +01:00
|
|
|
error = xfs_readlink(XFS_I(d_inode(dentry)), link);
|
2007-08-28 05:59:03 +02:00
|
|
|
if (unlikely(error))
|
|
|
|
goto out_kfree;
|
2005-04-17 00:20:36 +02:00
|
|
|
|
2015-05-02 19:32:22 +02:00
|
|
|
return *cookie = link;
|
2007-08-28 05:59:03 +02:00
|
|
|
|
|
|
|
out_kfree:
|
|
|
|
kfree(link);
|
|
|
|
out_err:
|
2015-05-02 19:32:22 +02:00
|
|
|
return ERR_PTR(error);
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
STATIC int
|
2006-03-14 04:00:51 +01:00
|
|
|
xfs_vn_getattr(
|
2007-10-11 09:46:39 +02:00
|
|
|
struct vfsmount *mnt,
|
|
|
|
struct dentry *dentry,
|
|
|
|
struct kstat *stat)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
2015-03-17 23:25:59 +01:00
|
|
|
struct inode *inode = d_inode(dentry);
|
2007-10-11 09:46:39 +02:00
|
|
|
struct xfs_inode *ip = XFS_I(inode);
|
|
|
|
struct xfs_mount *mp = ip->i_mount;
|
|
|
|
|
2010-06-24 03:57:09 +02:00
|
|
|
trace_xfs_getattr(ip);
|
2007-10-11 09:46:39 +02:00
|
|
|
|
|
|
|
if (XFS_FORCED_SHUTDOWN(mp))
|
2014-06-22 07:04:54 +02:00
|
|
|
return -EIO;
|
2007-10-11 09:46:39 +02:00
|
|
|
|
|
|
|
stat->size = XFS_ISIZE(ip);
|
|
|
|
stat->dev = inode->i_sb->s_dev;
|
|
|
|
stat->mode = ip->i_d.di_mode;
|
|
|
|
stat->nlink = ip->i_d.di_nlink;
|
2013-08-15 20:08:01 +02:00
|
|
|
stat->uid = inode->i_uid;
|
|
|
|
stat->gid = inode->i_gid;
|
2007-10-11 09:46:39 +02:00
|
|
|
stat->ino = ip->i_ino;
|
|
|
|
stat->atime = inode->i_atime;
|
2009-10-06 22:29:26 +02:00
|
|
|
stat->mtime = inode->i_mtime;
|
|
|
|
stat->ctime = inode->i_ctime;
|
2007-10-11 09:46:39 +02:00
|
|
|
stat->blocks =
|
|
|
|
XFS_FSB_TO_BB(mp, ip->i_d.di_nblocks + ip->i_delayed_blks);
|
|
|
|
|
|
|
|
|
|
|
|
switch (inode->i_mode & S_IFMT) {
|
|
|
|
case S_IFBLK:
|
|
|
|
case S_IFCHR:
|
|
|
|
stat->blksize = BLKDEV_IOSIZE;
|
|
|
|
stat->rdev = MKDEV(sysv_major(ip->i_df.if_u2.if_rdev) & 0x1ff,
|
|
|
|
sysv_minor(ip->i_df.if_u2.if_rdev));
|
|
|
|
break;
|
|
|
|
default:
|
2007-11-23 06:29:42 +01:00
|
|
|
if (XFS_IS_REALTIME_INODE(ip)) {
|
2007-10-11 09:46:39 +02:00
|
|
|
/*
|
|
|
|
* If the file blocks are being allocated from a
|
|
|
|
* realtime volume, then return the inode's realtime
|
|
|
|
* extent size or the realtime volume's extent size.
|
|
|
|
*/
|
|
|
|
stat->blksize =
|
|
|
|
xfs_get_extsz_hint(ip) << mp->m_sb.sb_blocklog;
|
|
|
|
} else
|
|
|
|
stat->blksize = xfs_preferred_iosize(mp);
|
|
|
|
stat->rdev = 0;
|
|
|
|
break;
|
2006-09-28 03:01:22 +02:00
|
|
|
}
|
2007-10-11 09:46:39 +02:00
|
|
|
|
|
|
|
return 0;
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
2013-05-27 08:38:25 +02:00
|
|
|
static void
|
|
|
|
xfs_setattr_mode(
|
|
|
|
struct xfs_inode *ip,
|
|
|
|
struct iattr *iattr)
|
|
|
|
{
|
2013-11-18 14:10:40 +01:00
|
|
|
struct inode *inode = VFS_I(ip);
|
|
|
|
umode_t mode = iattr->ia_mode;
|
2013-05-27 08:38:25 +02:00
|
|
|
|
|
|
|
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
|
|
|
|
|
|
|
|
ip->i_d.di_mode &= S_IFMT;
|
|
|
|
ip->i_d.di_mode |= mode & ~S_IFMT;
|
|
|
|
|
|
|
|
inode->i_mode &= S_IFMT;
|
|
|
|
inode->i_mode |= mode & ~S_IFMT;
|
|
|
|
}
|
|
|
|
|
2015-02-16 01:49:23 +01:00
|
|
|
void
|
2013-11-18 14:10:52 +01:00
|
|
|
xfs_setattr_time(
|
|
|
|
struct xfs_inode *ip,
|
|
|
|
struct iattr *iattr)
|
|
|
|
{
|
|
|
|
struct inode *inode = VFS_I(ip);
|
|
|
|
|
|
|
|
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
|
|
|
|
|
|
|
|
if (iattr->ia_valid & 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;
|
|
|
|
}
|
|
|
|
if (iattr->ia_valid & 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;
|
|
|
|
}
|
|
|
|
if (iattr->ia_valid & 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;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2011-07-08 14:34:23 +02:00
|
|
|
int
|
|
|
|
xfs_setattr_nonsize(
|
|
|
|
struct xfs_inode *ip,
|
|
|
|
struct iattr *iattr,
|
|
|
|
int flags)
|
|
|
|
{
|
|
|
|
xfs_mount_t *mp = ip->i_mount;
|
|
|
|
struct inode *inode = VFS_I(ip);
|
|
|
|
int mask = iattr->ia_valid;
|
|
|
|
xfs_trans_t *tp;
|
|
|
|
int error;
|
2013-08-15 20:08:01 +02:00
|
|
|
kuid_t uid = GLOBAL_ROOT_UID, iuid = GLOBAL_ROOT_UID;
|
|
|
|
kgid_t gid = GLOBAL_ROOT_GID, igid = GLOBAL_ROOT_GID;
|
2011-07-08 14:34:23 +02:00
|
|
|
struct xfs_dquot *udqp = NULL, *gdqp = NULL;
|
|
|
|
struct xfs_dquot *olddquot1 = NULL, *olddquot2 = NULL;
|
|
|
|
|
|
|
|
trace_xfs_setattr(ip);
|
|
|
|
|
2013-06-21 19:45:53 +02:00
|
|
|
/* If acls are being inherited, we already have this checked */
|
|
|
|
if (!(flags & XFS_ATTR_NOACL)) {
|
|
|
|
if (mp->m_flags & XFS_MOUNT_RDONLY)
|
2014-06-25 06:58:08 +02:00
|
|
|
return -EROFS;
|
2011-07-08 14:34:23 +02:00
|
|
|
|
2013-06-21 19:45:53 +02:00
|
|
|
if (XFS_FORCED_SHUTDOWN(mp))
|
2014-06-25 06:58:08 +02:00
|
|
|
return -EIO;
|
2011-07-08 14:34:23 +02:00
|
|
|
|
2014-06-25 06:58:08 +02:00
|
|
|
error = inode_change_ok(inode, iattr);
|
2013-06-21 19:45:53 +02:00
|
|
|
if (error)
|
2014-06-22 07:04:54 +02:00
|
|
|
return error;
|
2013-06-21 19:45:53 +02:00
|
|
|
}
|
2011-07-08 14:34:23 +02:00
|
|
|
|
|
|
|
ASSERT((mask & ATTR_SIZE) == 0);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* 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 & (ATTR_UID|ATTR_GID))) {
|
|
|
|
uint qflags = 0;
|
|
|
|
|
|
|
|
if ((mask & ATTR_UID) && XFS_IS_UQUOTA_ON(mp)) {
|
|
|
|
uid = iattr->ia_uid;
|
|
|
|
qflags |= XFS_QMOPT_UQUOTA;
|
|
|
|
} else {
|
2013-08-15 20:08:01 +02:00
|
|
|
uid = inode->i_uid;
|
2011-07-08 14:34:23 +02:00
|
|
|
}
|
|
|
|
if ((mask & ATTR_GID) && XFS_IS_GQUOTA_ON(mp)) {
|
|
|
|
gid = iattr->ia_gid;
|
|
|
|
qflags |= XFS_QMOPT_GQUOTA;
|
|
|
|
} else {
|
2013-08-15 20:08:01 +02:00
|
|
|
gid = inode->i_gid;
|
2011-07-08 14:34:23 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* We take a reference when we initialize udqp and gdqp,
|
|
|
|
* so it is important that we never blindly double trip on
|
|
|
|
* the same variable. See xfs_create() for an example.
|
|
|
|
*/
|
|
|
|
ASSERT(udqp == NULL);
|
|
|
|
ASSERT(gdqp == NULL);
|
2013-08-15 20:08:01 +02:00
|
|
|
error = xfs_qm_vop_dqalloc(ip, xfs_kuid_to_uid(uid),
|
|
|
|
xfs_kgid_to_gid(gid),
|
|
|
|
xfs_get_projid(ip),
|
|
|
|
qflags, &udqp, &gdqp, NULL);
|
2011-07-08 14:34:23 +02:00
|
|
|
if (error)
|
|
|
|
return error;
|
|
|
|
}
|
|
|
|
|
|
|
|
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_NOT_SIZE);
|
2013-08-12 12:49:59 +02:00
|
|
|
error = xfs_trans_reserve(tp, &M_RES(mp)->tr_ichange, 0, 0);
|
2011-07-08 14:34:23 +02:00
|
|
|
if (error)
|
2015-08-28 06:51:10 +02:00
|
|
|
goto out_trans_cancel;
|
2011-07-08 14:34:23 +02:00
|
|
|
|
|
|
|
xfs_ilock(ip, XFS_ILOCK_EXCL);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Change file ownership. Must be the owner or privileged.
|
|
|
|
*/
|
|
|
|
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
|
|
|
|
* while we didn't have the inode locked, inode's dquot(s)
|
|
|
|
* would have changed also.
|
|
|
|
*/
|
2013-08-15 20:08:01 +02:00
|
|
|
iuid = inode->i_uid;
|
|
|
|
igid = inode->i_gid;
|
2011-07-08 14:34:23 +02:00
|
|
|
gid = (mask & ATTR_GID) ? iattr->ia_gid : igid;
|
|
|
|
uid = (mask & ATTR_UID) ? iattr->ia_uid : iuid;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Do a quota reservation only if uid/gid is actually
|
|
|
|
* going to change.
|
|
|
|
*/
|
|
|
|
if (XFS_IS_QUOTA_RUNNING(mp) &&
|
2013-08-15 20:08:01 +02:00
|
|
|
((XFS_IS_UQUOTA_ON(mp) && !uid_eq(iuid, uid)) ||
|
|
|
|
(XFS_IS_GQUOTA_ON(mp) && !gid_eq(igid, gid)))) {
|
2011-07-08 14:34:23 +02:00
|
|
|
ASSERT(tp);
|
|
|
|
error = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp,
|
2013-07-11 07:00:40 +02:00
|
|
|
NULL, capable(CAP_FOWNER) ?
|
2011-07-08 14:34:23 +02:00
|
|
|
XFS_QMOPT_FORCE_RES : 0);
|
|
|
|
if (error) /* out of quota */
|
2015-08-28 06:51:10 +02:00
|
|
|
goto out_unlock;
|
2011-07-08 14:34:23 +02:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2011-09-19 17:00:54 +02:00
|
|
|
xfs_trans_ijoin(tp, ip, 0);
|
2011-07-08 14:34:23 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Change file ownership. Must be the owner or privileged.
|
|
|
|
*/
|
|
|
|
if (mask & (ATTR_UID|ATTR_GID)) {
|
|
|
|
/*
|
|
|
|
* 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.
|
|
|
|
*/
|
2013-08-15 20:08:01 +02:00
|
|
|
if (!uid_eq(iuid, uid)) {
|
2011-07-08 14:34:23 +02:00
|
|
|
if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_UQUOTA_ON(mp)) {
|
|
|
|
ASSERT(mask & ATTR_UID);
|
|
|
|
ASSERT(udqp);
|
|
|
|
olddquot1 = xfs_qm_vop_chown(tp, ip,
|
|
|
|
&ip->i_udquot, udqp);
|
|
|
|
}
|
2013-08-15 20:08:01 +02:00
|
|
|
ip->i_d.di_uid = xfs_kuid_to_uid(uid);
|
2011-07-08 14:34:23 +02:00
|
|
|
inode->i_uid = uid;
|
|
|
|
}
|
2013-08-15 20:08:01 +02:00
|
|
|
if (!gid_eq(igid, gid)) {
|
2011-07-08 14:34:23 +02:00
|
|
|
if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_GQUOTA_ON(mp)) {
|
2013-11-26 14:38:34 +01:00
|
|
|
ASSERT(xfs_sb_version_has_pquotino(&mp->m_sb) ||
|
|
|
|
!XFS_IS_PQUOTA_ON(mp));
|
2011-07-08 14:34:23 +02:00
|
|
|
ASSERT(mask & ATTR_GID);
|
|
|
|
ASSERT(gdqp);
|
|
|
|
olddquot2 = xfs_qm_vop_chown(tp, ip,
|
|
|
|
&ip->i_gdquot, gdqp);
|
|
|
|
}
|
2013-08-15 20:08:01 +02:00
|
|
|
ip->i_d.di_gid = xfs_kgid_to_gid(gid);
|
2011-07-08 14:34:23 +02:00
|
|
|
inode->i_gid = gid;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2013-05-27 08:38:25 +02:00
|
|
|
if (mask & ATTR_MODE)
|
2013-11-18 14:10:40 +01:00
|
|
|
xfs_setattr_mode(ip, iattr);
|
2013-11-18 14:10:52 +01:00
|
|
|
if (mask & (ATTR_ATIME|ATTR_CTIME|ATTR_MTIME))
|
|
|
|
xfs_setattr_time(ip, iattr);
|
2011-07-08 14:34:23 +02:00
|
|
|
|
|
|
|
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
|
|
|
|
|
|
|
|
XFS_STATS_INC(xs_ig_attrchg);
|
|
|
|
|
|
|
|
if (mp->m_flags & XFS_MOUNT_WSYNC)
|
|
|
|
xfs_trans_set_sync(tp);
|
2015-06-04 05:48:08 +02:00
|
|
|
error = xfs_trans_commit(tp);
|
2011-07-08 14:34:23 +02:00
|
|
|
|
|
|
|
xfs_iunlock(ip, XFS_ILOCK_EXCL);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Release any dquot(s) the inode had kept before chown.
|
|
|
|
*/
|
|
|
|
xfs_qm_dqrele(olddquot1);
|
|
|
|
xfs_qm_dqrele(olddquot2);
|
|
|
|
xfs_qm_dqrele(udqp);
|
|
|
|
xfs_qm_dqrele(gdqp);
|
|
|
|
|
|
|
|
if (error)
|
2014-06-22 07:04:54 +02:00
|
|
|
return error;
|
2011-07-08 14:34:23 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* XXX(hch): Updating the ACL entries is not atomic vs the i_mode
|
|
|
|
* update. We could avoid this with linked transactions
|
|
|
|
* and passing down the transaction pointer all the way
|
|
|
|
* to attr_set. No previous user of the generic
|
|
|
|
* Posix ACL code seems to care about this issue either.
|
|
|
|
*/
|
|
|
|
if ((mask & ATTR_MODE) && !(flags & XFS_ATTR_NOACL)) {
|
2014-06-25 06:58:08 +02:00
|
|
|
error = posix_acl_chmod(inode, inode->i_mode);
|
2011-07-08 14:34:23 +02:00
|
|
|
if (error)
|
2014-06-22 07:04:54 +02:00
|
|
|
return error;
|
2011-07-08 14:34:23 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
2015-08-28 06:51:10 +02:00
|
|
|
out_unlock:
|
|
|
|
xfs_iunlock(ip, XFS_ILOCK_EXCL);
|
2011-07-08 14:34:23 +02:00
|
|
|
out_trans_cancel:
|
2015-06-04 05:47:56 +02:00
|
|
|
xfs_trans_cancel(tp);
|
2011-07-08 14:34:23 +02:00
|
|
|
xfs_qm_dqrele(udqp);
|
|
|
|
xfs_qm_dqrele(gdqp);
|
|
|
|
return error;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Truncate file. Must have write permission and not be a directory.
|
|
|
|
*/
|
|
|
|
int
|
|
|
|
xfs_setattr_size(
|
|
|
|
struct xfs_inode *ip,
|
2013-10-14 16:09:35 +02:00
|
|
|
struct iattr *iattr)
|
2011-07-08 14:34:23 +02:00
|
|
|
{
|
|
|
|
struct xfs_mount *mp = ip->i_mount;
|
|
|
|
struct inode *inode = VFS_I(ip);
|
2011-12-18 21:00:04 +01:00
|
|
|
xfs_off_t oldsize, newsize;
|
2011-07-08 14:34:23 +02:00
|
|
|
struct xfs_trans *tp;
|
|
|
|
int error;
|
2012-03-27 16:34:48 +02:00
|
|
|
uint lock_flags = 0;
|
2015-02-23 12:37:08 +01:00
|
|
|
bool did_zeroing = false;
|
2011-07-08 14:34:23 +02:00
|
|
|
|
|
|
|
trace_xfs_setattr(ip);
|
|
|
|
|
|
|
|
if (mp->m_flags & XFS_MOUNT_RDONLY)
|
2014-06-25 06:58:08 +02:00
|
|
|
return -EROFS;
|
2011-07-08 14:34:23 +02:00
|
|
|
|
|
|
|
if (XFS_FORCED_SHUTDOWN(mp))
|
2014-06-25 06:58:08 +02:00
|
|
|
return -EIO;
|
2011-07-08 14:34:23 +02:00
|
|
|
|
2014-06-25 06:58:08 +02:00
|
|
|
error = inode_change_ok(inode, iattr);
|
2011-07-08 14:34:23 +02:00
|
|
|
if (error)
|
2014-06-22 07:04:54 +02:00
|
|
|
return error;
|
2011-07-08 14:34:23 +02:00
|
|
|
|
2013-10-14 16:09:35 +02:00
|
|
|
ASSERT(xfs_isilocked(ip, XFS_IOLOCK_EXCL));
|
2015-02-23 11:45:32 +01:00
|
|
|
ASSERT(xfs_isilocked(ip, XFS_MMAPLOCK_EXCL));
|
2011-07-08 14:34:23 +02:00
|
|
|
ASSERT(S_ISREG(ip->i_d.di_mode));
|
2014-02-10 00:35:22 +01:00
|
|
|
ASSERT((iattr->ia_valid & (ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET|
|
|
|
|
ATTR_MTIME_SET|ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0);
|
2011-07-08 14:34:23 +02:00
|
|
|
|
2011-12-18 21:00:11 +01:00
|
|
|
oldsize = inode->i_size;
|
2011-12-18 21:00:04 +01:00
|
|
|
newsize = iattr->ia_size;
|
|
|
|
|
2011-07-08 14:34:23 +02:00
|
|
|
/*
|
|
|
|
* Short circuit the truncate case for zero length files.
|
|
|
|
*/
|
2011-12-18 21:00:04 +01:00
|
|
|
if (newsize == 0 && oldsize == 0 && ip->i_d.di_nextents == 0) {
|
2014-02-10 00:35:22 +01:00
|
|
|
if (!(iattr->ia_valid & (ATTR_CTIME|ATTR_MTIME)))
|
2013-10-14 16:09:35 +02:00
|
|
|
return 0;
|
2011-07-08 14:34:26 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Use the regular setattr path to update the timestamps.
|
|
|
|
*/
|
|
|
|
iattr->ia_valid &= ~ATTR_SIZE;
|
|
|
|
return xfs_setattr_nonsize(ip, iattr, 0);
|
2011-07-08 14:34:23 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Make sure that the dquots are attached to the inode.
|
|
|
|
*/
|
2012-03-27 16:34:48 +02:00
|
|
|
error = xfs_qm_dqattach(ip, 0);
|
2011-07-08 14:34:23 +02:00
|
|
|
if (error)
|
2013-10-14 16:09:35 +02:00
|
|
|
return error;
|
2011-07-08 14:34:23 +02:00
|
|
|
|
|
|
|
/*
|
2015-02-23 12:37:08 +01:00
|
|
|
* File data changes must be complete before we start the transaction to
|
|
|
|
* modify the inode. 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.
|
|
|
|
*
|
|
|
|
* Start with zeroing any data block beyond EOF that we may expose on
|
|
|
|
* file extension.
|
2011-07-08 14:34:23 +02:00
|
|
|
*/
|
2011-12-18 21:00:04 +01:00
|
|
|
if (newsize > oldsize) {
|
2015-02-23 12:37:08 +01:00
|
|
|
error = xfs_zero_eof(ip, newsize, oldsize, &did_zeroing);
|
2011-07-08 14:34:23 +02:00
|
|
|
if (error)
|
2013-10-14 16:09:35 +02:00
|
|
|
return error;
|
2011-07-08 14:34:23 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* We are going to log the inode size change in this transaction so
|
|
|
|
* any previous writes that are beyond the on disk EOF and the new
|
|
|
|
* EOF that have not been written out need to be written here. If we
|
|
|
|
* do not write the data out, we expose ourselves to the null files
|
2015-02-23 12:37:08 +01:00
|
|
|
* problem. Note that this includes any block zeroing we did above;
|
|
|
|
* otherwise those blocks may not be zeroed after a crash.
|
2011-07-08 14:34:23 +02:00
|
|
|
*/
|
2015-02-23 12:37:08 +01:00
|
|
|
if (newsize > ip->i_d.di_size &&
|
|
|
|
(oldsize != ip->i_d.di_size || did_zeroing)) {
|
2014-06-25 06:58:08 +02:00
|
|
|
error = filemap_write_and_wait_range(VFS_I(ip)->i_mapping,
|
2012-11-12 12:53:56 +01:00
|
|
|
ip->i_d.di_size, newsize);
|
2011-07-08 14:34:23 +02:00
|
|
|
if (error)
|
2013-10-14 16:09:35 +02:00
|
|
|
return error;
|
2011-07-08 14:34:23 +02:00
|
|
|
}
|
|
|
|
|
2015-02-23 12:37:08 +01:00
|
|
|
/* Now wait for all direct I/O to complete. */
|
2011-08-23 10:28:13 +02:00
|
|
|
inode_dio_wait(inode);
|
2011-07-08 14:34:23 +02:00
|
|
|
|
2014-05-07 00:05:45 +02:00
|
|
|
/*
|
2015-02-23 11:46:58 +01:00
|
|
|
* We've already locked out new page faults, so now we can safely remove
|
|
|
|
* pages from the page cache knowing they won't get refaulted until we
|
|
|
|
* drop the XFS_MMAP_EXCL lock after the extent manipulations are
|
|
|
|
* complete. The truncate_setsize() call also cleans partial EOF page
|
|
|
|
* PTEs on extending truncates and hence ensures sub-page block size
|
|
|
|
* filesystems are correctly handled, too.
|
2014-05-07 00:05:45 +02:00
|
|
|
*
|
2015-02-23 11:46:58 +01:00
|
|
|
* We have to do all the page cache truncate work outside the
|
|
|
|
* transaction context as the "lock" order is page lock->log space
|
|
|
|
* reservation as defined by extent allocation in the writeback path.
|
|
|
|
* Hence a truncate can fail with ENOMEM from xfs_trans_reserve(), but
|
|
|
|
* having already truncated the in-memory version of the file (i.e. made
|
|
|
|
* user visible changes). There's not much we can do about this, except
|
|
|
|
* to hope that the caller sees ENOMEM and retries the truncate
|
|
|
|
* operation.
|
2014-05-07 00:05:45 +02:00
|
|
|
*/
|
2015-06-04 01:19:10 +02:00
|
|
|
if (IS_DAX(inode))
|
|
|
|
error = dax_truncate_page(inode, newsize, xfs_get_blocks_direct);
|
|
|
|
else
|
|
|
|
error = block_truncate_page(inode->i_mapping, newsize,
|
|
|
|
xfs_get_blocks);
|
2011-07-08 14:34:23 +02:00
|
|
|
if (error)
|
2013-10-14 16:09:35 +02:00
|
|
|
return error;
|
2014-05-07 00:05:45 +02:00
|
|
|
truncate_setsize(inode, newsize);
|
2011-07-08 14:34:23 +02:00
|
|
|
|
|
|
|
tp = xfs_trans_alloc(mp, XFS_TRANS_SETATTR_SIZE);
|
2013-08-12 12:49:59 +02:00
|
|
|
error = xfs_trans_reserve(tp, &M_RES(mp)->tr_itruncate, 0, 0);
|
2011-07-08 14:34:23 +02:00
|
|
|
if (error)
|
|
|
|
goto out_trans_cancel;
|
|
|
|
|
|
|
|
lock_flags |= XFS_ILOCK_EXCL;
|
|
|
|
xfs_ilock(ip, XFS_ILOCK_EXCL);
|
2011-09-19 17:00:54 +02:00
|
|
|
xfs_trans_ijoin(tp, ip, 0);
|
2011-07-08 14:34:23 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* 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.
|
|
|
|
*
|
|
|
|
* The regular truncate() case without ATTR_CTIME and ATTR_MTIME is a
|
|
|
|
* special case where we need to update the times despite not having
|
|
|
|
* these flags set. For all other operations the VFS set these flags
|
|
|
|
* explicitly if it wants a timestamp update.
|
|
|
|
*/
|
2014-02-10 00:35:22 +01:00
|
|
|
if (newsize != oldsize &&
|
|
|
|
!(iattr->ia_valid & (ATTR_CTIME | ATTR_MTIME))) {
|
2011-07-08 14:34:23 +02:00
|
|
|
iattr->ia_ctime = iattr->ia_mtime =
|
|
|
|
current_fs_time(inode->i_sb);
|
2014-02-10 00:35:22 +01:00
|
|
|
iattr->ia_valid |= ATTR_CTIME | ATTR_MTIME;
|
2011-07-08 14:34:23 +02:00
|
|
|
}
|
|
|
|
|
2011-12-18 21:00:04 +01:00
|
|
|
/*
|
|
|
|
* The first thing we do is set the size to new_size permanently on
|
|
|
|
* disk. This way we don't have to worry about anyone ever being able
|
|
|
|
* to look at the data being freed even in the face of a crash.
|
|
|
|
* What we're getting around here is the case where we free a block, it
|
|
|
|
* is allocated to another file, it is written to, and then we crash.
|
|
|
|
* If the new data gets written to the file but the log buffers
|
|
|
|
* containing the free and reallocation don't, then we'd end up with
|
|
|
|
* garbage in the blocks being freed. As long as we make the new size
|
|
|
|
* permanent before actually freeing any blocks it doesn't matter if
|
|
|
|
* they get written to.
|
|
|
|
*/
|
|
|
|
ip->i_d.di_size = newsize;
|
|
|
|
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
|
|
|
|
|
|
|
|
if (newsize <= oldsize) {
|
|
|
|
error = xfs_itruncate_extents(&tp, ip, XFS_DATA_FORK, newsize);
|
2011-07-08 14:34:23 +02:00
|
|
|
if (error)
|
2015-06-04 05:47:56 +02:00
|
|
|
goto out_trans_cancel;
|
2011-07-08 14:34:23 +02:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Truncated "down", so we're removing references to old data
|
|
|
|
* here - if we delay flushing for a long time, we expose
|
|
|
|
* ourselves unduly to the notorious NULL files problem. So,
|
|
|
|
* we mark this inode and flush it when the file is closed,
|
|
|
|
* and do not wait the usual (long) time for writeout.
|
|
|
|
*/
|
|
|
|
xfs_iflags_set(ip, XFS_ITRUNCATED);
|
2012-11-06 15:50:38 +01:00
|
|
|
|
|
|
|
/* A truncate down always removes post-EOF blocks. */
|
|
|
|
xfs_inode_clear_eofblocks_tag(ip);
|
2011-07-08 14:34:23 +02:00
|
|
|
}
|
|
|
|
|
2014-02-10 00:35:22 +01:00
|
|
|
if (iattr->ia_valid & ATTR_MODE)
|
2013-11-18 14:10:40 +01:00
|
|
|
xfs_setattr_mode(ip, iattr);
|
2014-02-10 00:35:22 +01:00
|
|
|
if (iattr->ia_valid & (ATTR_ATIME|ATTR_CTIME|ATTR_MTIME))
|
2013-11-18 14:10:52 +01:00
|
|
|
xfs_setattr_time(ip, iattr);
|
2011-07-08 14:34:23 +02:00
|
|
|
|
|
|
|
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
|
|
|
|
|
|
|
|
XFS_STATS_INC(xs_ig_attrchg);
|
|
|
|
|
|
|
|
if (mp->m_flags & XFS_MOUNT_WSYNC)
|
|
|
|
xfs_trans_set_sync(tp);
|
|
|
|
|
2015-06-04 05:48:08 +02:00
|
|
|
error = xfs_trans_commit(tp);
|
2011-07-08 14:34:23 +02:00
|
|
|
out_unlock:
|
|
|
|
if (lock_flags)
|
|
|
|
xfs_iunlock(ip, lock_flags);
|
|
|
|
return error;
|
|
|
|
|
|
|
|
out_trans_cancel:
|
2015-06-04 05:47:56 +02:00
|
|
|
xfs_trans_cancel(tp);
|
2011-07-08 14:34:23 +02:00
|
|
|
goto out_unlock;
|
|
|
|
}
|
|
|
|
|
2005-04-17 00:20:36 +02:00
|
|
|
STATIC int
|
2006-03-14 04:00:51 +01:00
|
|
|
xfs_vn_setattr(
|
2013-10-14 16:09:35 +02:00
|
|
|
struct dentry *dentry,
|
|
|
|
struct iattr *iattr)
|
2005-04-17 00:20:36 +02:00
|
|
|
{
|
2015-03-17 23:25:59 +01:00
|
|
|
struct xfs_inode *ip = XFS_I(d_inode(dentry));
|
2013-10-14 16:09:35 +02:00
|
|
|
int error;
|
|
|
|
|
|
|
|
if (iattr->ia_valid & ATTR_SIZE) {
|
2015-02-16 01:59:50 +01:00
|
|
|
uint iolock = XFS_IOLOCK_EXCL;
|
|
|
|
|
|
|
|
xfs_ilock(ip, iolock);
|
2015-04-27 00:48:49 +02:00
|
|
|
error = xfs_break_layouts(d_inode(dentry), &iolock, true);
|
2015-02-23 11:45:32 +01:00
|
|
|
if (!error) {
|
|
|
|
xfs_ilock(ip, XFS_MMAPLOCK_EXCL);
|
|
|
|
iolock |= XFS_MMAPLOCK_EXCL;
|
|
|
|
|
2015-02-16 01:59:50 +01:00
|
|
|
error = xfs_setattr_size(ip, iattr);
|
2015-02-23 11:45:32 +01:00
|
|
|
}
|
2015-02-16 01:59:50 +01:00
|
|
|
xfs_iunlock(ip, iolock);
|
2013-10-14 16:09:35 +02:00
|
|
|
} else {
|
|
|
|
error = xfs_setattr_nonsize(ip, iattr, 0);
|
|
|
|
}
|
|
|
|
|
2014-06-25 06:58:08 +02:00
|
|
|
return error;
|
2005-04-17 00:20:36 +02:00
|
|
|
}
|
|
|
|
|
2012-06-06 23:01:28 +02:00
|
|
|
STATIC int
|
|
|
|
xfs_vn_update_time(
|
|
|
|
struct inode *inode,
|
|
|
|
struct timespec *now,
|
|
|
|
int flags)
|
|
|
|
{
|
|
|
|
struct xfs_inode *ip = XFS_I(inode);
|
|
|
|
struct xfs_mount *mp = ip->i_mount;
|
|
|
|
struct xfs_trans *tp;
|
|
|
|
int error;
|
|
|
|
|
|
|
|
trace_xfs_update_time(ip);
|
|
|
|
|
|
|
|
tp = xfs_trans_alloc(mp, XFS_TRANS_FSYNC_TS);
|
2013-08-12 12:49:59 +02:00
|
|
|
error = xfs_trans_reserve(tp, &M_RES(mp)->tr_fsyncts, 0, 0);
|
2012-06-06 23:01:28 +02:00
|
|
|
if (error) {
|
2015-06-04 05:47:56 +02:00
|
|
|
xfs_trans_cancel(tp);
|
2014-06-25 06:58:08 +02:00
|
|
|
return error;
|
2012-06-06 23:01:28 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
xfs_ilock(ip, XFS_ILOCK_EXCL);
|
|
|
|
if (flags & S_CTIME) {
|
|
|
|
inode->i_ctime = *now;
|
|
|
|
ip->i_d.di_ctime.t_sec = (__int32_t)now->tv_sec;
|
|
|
|
ip->i_d.di_ctime.t_nsec = (__int32_t)now->tv_nsec;
|
|
|
|
}
|
|
|
|
if (flags & S_MTIME) {
|
|
|
|
inode->i_mtime = *now;
|
|
|
|
ip->i_d.di_mtime.t_sec = (__int32_t)now->tv_sec;
|
|
|
|
ip->i_d.di_mtime.t_nsec = (__int32_t)now->tv_nsec;
|
|
|
|
}
|
|
|
|
if (flags & S_ATIME) {
|
|
|
|
inode->i_atime = *now;
|
|
|
|
ip->i_d.di_atime.t_sec = (__int32_t)now->tv_sec;
|
|
|
|
ip->i_d.di_atime.t_nsec = (__int32_t)now->tv_nsec;
|
|
|
|
}
|
|
|
|
xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL);
|
|
|
|
xfs_trans_log_inode(tp, ip, XFS_ILOG_TIMESTAMP);
|
2015-06-04 05:48:08 +02:00
|
|
|
return xfs_trans_commit(tp);
|
2012-06-06 23:01:28 +02:00
|
|
|
}
|
|
|
|
|
2008-11-28 04:23:35 +01:00
|
|
|
#define XFS_FIEMAP_FLAGS (FIEMAP_FLAG_SYNC|FIEMAP_FLAG_XATTR)
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Call fiemap helper to fill in user data.
|
|
|
|
* Returns positive errors to xfs_getbmap.
|
|
|
|
*/
|
|
|
|
STATIC int
|
|
|
|
xfs_fiemap_format(
|
|
|
|
void **arg,
|
|
|
|
struct getbmapx *bmv,
|
|
|
|
int *full)
|
|
|
|
{
|
|
|
|
int error;
|
|
|
|
struct fiemap_extent_info *fieinfo = *arg;
|
|
|
|
u32 fiemap_flags = 0;
|
|
|
|
u64 logical, physical, length;
|
|
|
|
|
|
|
|
/* Do nothing for a hole */
|
|
|
|
if (bmv->bmv_block == -1LL)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
logical = BBTOB(bmv->bmv_offset);
|
|
|
|
physical = BBTOB(bmv->bmv_block);
|
|
|
|
length = BBTOB(bmv->bmv_length);
|
|
|
|
|
|
|
|
if (bmv->bmv_oflags & BMV_OF_PREALLOC)
|
|
|
|
fiemap_flags |= FIEMAP_EXTENT_UNWRITTEN;
|
|
|
|
else if (bmv->bmv_oflags & BMV_OF_DELALLOC) {
|
2013-06-06 15:24:01 +02:00
|
|
|
fiemap_flags |= (FIEMAP_EXTENT_DELALLOC |
|
|
|
|
FIEMAP_EXTENT_UNKNOWN);
|
2008-11-28 04:23:35 +01:00
|
|
|
physical = 0; /* no block yet */
|
|
|
|
}
|
|
|
|
if (bmv->bmv_oflags & BMV_OF_LAST)
|
|
|
|
fiemap_flags |= FIEMAP_EXTENT_LAST;
|
|
|
|
|
|
|
|
error = fiemap_fill_next_extent(fieinfo, logical, physical,
|
|
|
|
length, fiemap_flags);
|
|
|
|
if (error > 0) {
|
|
|
|
error = 0;
|
|
|
|
*full = 1; /* user array now full */
|
|
|
|
}
|
|
|
|
|
2014-06-25 06:58:08 +02:00
|
|
|
return error;
|
2008-11-28 04:23:35 +01:00
|
|
|
}
|
|
|
|
|
|
|
|
STATIC int
|
|
|
|
xfs_vn_fiemap(
|
|
|
|
struct inode *inode,
|
|
|
|
struct fiemap_extent_info *fieinfo,
|
|
|
|
u64 start,
|
|
|
|
u64 length)
|
|
|
|
{
|
|
|
|
xfs_inode_t *ip = XFS_I(inode);
|
|
|
|
struct getbmapx bm;
|
|
|
|
int error;
|
|
|
|
|
|
|
|
error = fiemap_check_flags(fieinfo, XFS_FIEMAP_FLAGS);
|
|
|
|
if (error)
|
|
|
|
return error;
|
|
|
|
|
|
|
|
/* Set up bmap header for xfs internal routine */
|
2014-08-04 03:35:35 +02:00
|
|
|
bm.bmv_offset = BTOBBT(start);
|
2008-11-28 04:23:35 +01:00
|
|
|
/* Special case for whole file */
|
|
|
|
if (length == FIEMAP_MAX_OFFSET)
|
|
|
|
bm.bmv_length = -1LL;
|
|
|
|
else
|
2014-08-04 03:35:35 +02:00
|
|
|
bm.bmv_length = BTOBB(start + length) - bm.bmv_offset;
|
2008-11-28 04:23:35 +01:00
|
|
|
|
2009-07-27 04:52:01 +02:00
|
|
|
/* We add one because in getbmap world count includes the header */
|
2010-04-29 07:13:56 +02:00
|
|
|
bm.bmv_count = !fieinfo->fi_extents_max ? MAXEXTNUM :
|
|
|
|
fieinfo->fi_extents_max + 1;
|
|
|
|
bm.bmv_count = min_t(__s32, bm.bmv_count,
|
|
|
|
(PAGE_SIZE * 16 / sizeof(struct getbmapx)));
|
xfs: Make fiemap work with sparse files
In xfs_vn_fiemap, we set bvm_count to fi_extent_max + 1 and want
to return fi_extent_max extents, but actually it won't work for
a sparse file. The reason is that in xfs_getbmap we will
calculate holes and set it in 'out', while out is malloced by
bmv_count(fi_extent_max+1) which didn't consider holes. So in the
worst case, if 'out' vector looks like
[hole, extent, hole, extent, hole, ... hole, extent, hole],
we will only return half of fi_extent_max extents.
This patch add a new parameter BMV_IF_NO_HOLES for bvm_iflags.
So with this flags, we don't use our 'out' in xfs_getbmap for
a hole. The solution is a bit ugly by just don't increasing
index of 'out' vector. I felt that it is not easy to skip it
at the very beginning since we have the complicated check and
some function like xfs_getbmapx_fix_eof_hole to adjust 'out'.
Cc: Dave Chinner <david@fromorbit.com>
Signed-off-by: Tao Ma <tao.ma@oracle.com>
Signed-off-by: Alex Elder <aelder@sgi.com>
2010-08-30 04:44:03 +02:00
|
|
|
bm.bmv_iflags = BMV_IF_PREALLOC | BMV_IF_NO_HOLES;
|
2008-11-28 04:23:35 +01:00
|
|
|
if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR)
|
|
|
|
bm.bmv_iflags |= BMV_IF_ATTRFORK;
|
|
|
|
if (!(fieinfo->fi_flags & FIEMAP_FLAG_SYNC))
|
|
|
|
bm.bmv_iflags |= BMV_IF_DELALLOC;
|
|
|
|
|
|
|
|
error = xfs_getbmap(ip, &bm, xfs_fiemap_format, fieinfo);
|
|
|
|
if (error)
|
2014-06-25 06:58:08 +02:00
|
|
|
return error;
|
2008-11-28 04:23:35 +01:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2013-12-18 01:22:40 +01:00
|
|
|
STATIC int
|
|
|
|
xfs_vn_tmpfile(
|
|
|
|
struct inode *dir,
|
|
|
|
struct dentry *dentry,
|
|
|
|
umode_t mode)
|
|
|
|
{
|
2014-05-05 23:34:28 +02:00
|
|
|
return xfs_generic_create(dir, dentry, mode, 0, true);
|
2013-12-18 01:22:40 +01:00
|
|
|
}
|
|
|
|
|
2008-08-13 08:23:13 +02:00
|
|
|
static const struct inode_operations xfs_inode_operations = {
|
2011-07-23 17:37:31 +02:00
|
|
|
.get_acl = xfs_get_acl,
|
2013-12-20 14:16:50 +01:00
|
|
|
.set_acl = xfs_set_acl,
|
2006-03-14 04:00:51 +01:00
|
|
|
.getattr = xfs_vn_getattr,
|
|
|
|
.setattr = xfs_vn_setattr,
|
2008-06-23 05:23:01 +02:00
|
|
|
.setxattr = generic_setxattr,
|
|
|
|
.getxattr = generic_getxattr,
|
|
|
|
.removexattr = generic_removexattr,
|
2006-03-14 04:00:51 +01:00
|
|
|
.listxattr = xfs_vn_listxattr,
|
2008-11-28 04:23:35 +01:00
|
|
|
.fiemap = xfs_vn_fiemap,
|
2012-06-06 23:01:28 +02:00
|
|
|
.update_time = xfs_vn_update_time,
|
2005-04-17 00:20:36 +02:00
|
|
|
};
|
|
|
|
|
2008-08-13 08:23:13 +02:00
|
|
|
static const struct inode_operations xfs_dir_inode_operations = {
|
2006-03-14 04:00:51 +01:00
|
|
|
.create = xfs_vn_create,
|
|
|
|
.lookup = xfs_vn_lookup,
|
|
|
|
.link = xfs_vn_link,
|
|
|
|
.unlink = xfs_vn_unlink,
|
|
|
|
.symlink = xfs_vn_symlink,
|
|
|
|
.mkdir = xfs_vn_mkdir,
|
2008-06-23 05:25:17 +02:00
|
|
|
/*
|
|
|
|
* 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,
|
2006-03-14 04:00:51 +01:00
|
|
|
.mknod = xfs_vn_mknod,
|
2014-12-23 22:51:38 +01:00
|
|
|
.rename2 = xfs_vn_rename,
|
2011-07-23 17:37:31 +02:00
|
|
|
.get_acl = xfs_get_acl,
|
2013-12-20 14:16:50 +01:00
|
|
|
.set_acl = xfs_set_acl,
|
2006-03-14 04:00:51 +01:00
|
|
|
.getattr = xfs_vn_getattr,
|
|
|
|
.setattr = xfs_vn_setattr,
|
2008-06-23 05:23:01 +02:00
|
|
|
.setxattr = generic_setxattr,
|
|
|
|
.getxattr = generic_getxattr,
|
|
|
|
.removexattr = generic_removexattr,
|
2006-03-14 04:00:51 +01:00
|
|
|
.listxattr = xfs_vn_listxattr,
|
2012-06-06 23:01:28 +02:00
|
|
|
.update_time = xfs_vn_update_time,
|
2013-12-18 01:22:40 +01:00
|
|
|
.tmpfile = xfs_vn_tmpfile,
|
2005-04-17 00:20:36 +02:00
|
|
|
};
|
|
|
|
|
2008-08-13 08:23:13 +02:00
|
|
|
static const struct inode_operations xfs_dir_ci_inode_operations = {
|
2008-05-21 08:58:22 +02:00
|
|
|
.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,
|
2008-06-23 05:25:17 +02:00
|
|
|
/*
|
|
|
|
* 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,
|
2008-05-21 08:58:22 +02:00
|
|
|
.mknod = xfs_vn_mknod,
|
2014-12-23 22:51:38 +01:00
|
|
|
.rename2 = xfs_vn_rename,
|
2011-07-23 17:37:31 +02:00
|
|
|
.get_acl = xfs_get_acl,
|
2013-12-20 14:16:50 +01:00
|
|
|
.set_acl = xfs_set_acl,
|
2008-05-21 08:58:22 +02:00
|
|
|
.getattr = xfs_vn_getattr,
|
|
|
|
.setattr = xfs_vn_setattr,
|
2008-06-23 05:23:01 +02:00
|
|
|
.setxattr = generic_setxattr,
|
|
|
|
.getxattr = generic_getxattr,
|
|
|
|
.removexattr = generic_removexattr,
|
2008-05-21 08:58:22 +02:00
|
|
|
.listxattr = xfs_vn_listxattr,
|
2012-06-06 23:01:28 +02:00
|
|
|
.update_time = xfs_vn_update_time,
|
2013-12-18 01:22:40 +01:00
|
|
|
.tmpfile = xfs_vn_tmpfile,
|
2008-05-21 08:58:22 +02:00
|
|
|
};
|
|
|
|
|
2008-08-13 08:23:13 +02:00
|
|
|
static const struct inode_operations xfs_symlink_inode_operations = {
|
2005-04-17 00:20:36 +02:00
|
|
|
.readlink = generic_readlink,
|
2006-03-14 04:00:51 +01:00
|
|
|
.follow_link = xfs_vn_follow_link,
|
2013-11-30 04:54:55 +01:00
|
|
|
.put_link = kfree_put_link,
|
2006-03-14 04:00:51 +01:00
|
|
|
.getattr = xfs_vn_getattr,
|
|
|
|
.setattr = xfs_vn_setattr,
|
2008-06-23 05:23:01 +02:00
|
|
|
.setxattr = generic_setxattr,
|
|
|
|
.getxattr = generic_getxattr,
|
|
|
|
.removexattr = generic_removexattr,
|
2006-03-14 04:00:51 +01:00
|
|
|
.listxattr = xfs_vn_listxattr,
|
2012-06-06 23:01:28 +02:00
|
|
|
.update_time = xfs_vn_update_time,
|
2005-04-17 00:20:36 +02:00
|
|
|
};
|
2008-08-13 08:23:13 +02:00
|
|
|
|
|
|
|
STATIC void
|
|
|
|
xfs_diflags_to_iflags(
|
|
|
|
struct inode *inode,
|
|
|
|
struct xfs_inode *ip)
|
|
|
|
{
|
2015-06-04 01:19:18 +02:00
|
|
|
uint16_t flags = ip->i_d.di_flags;
|
|
|
|
|
|
|
|
inode->i_flags &= ~(S_IMMUTABLE | S_APPEND | S_SYNC |
|
|
|
|
S_NOATIME | S_DAX);
|
|
|
|
|
|
|
|
if (flags & XFS_DIFLAG_IMMUTABLE)
|
2008-08-13 08:23:13 +02:00
|
|
|
inode->i_flags |= S_IMMUTABLE;
|
2015-06-04 01:19:18 +02:00
|
|
|
if (flags & XFS_DIFLAG_APPEND)
|
2008-08-13 08:23:13 +02:00
|
|
|
inode->i_flags |= S_APPEND;
|
2015-06-04 01:19:18 +02:00
|
|
|
if (flags & XFS_DIFLAG_SYNC)
|
2008-08-13 08:23:13 +02:00
|
|
|
inode->i_flags |= S_SYNC;
|
2015-06-04 01:19:18 +02:00
|
|
|
if (flags & XFS_DIFLAG_NOATIME)
|
2008-08-13 08:23:13 +02:00
|
|
|
inode->i_flags |= S_NOATIME;
|
2015-06-04 01:19:18 +02:00
|
|
|
/* XXX: Also needs an on-disk per inode flag! */
|
|
|
|
if (ip->i_mount->m_flags & XFS_MOUNT_DAX)
|
|
|
|
inode->i_flags |= S_DAX;
|
2008-08-13 08:23:13 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2015-02-23 12:38:08 +01:00
|
|
|
* Initialize the Linux inode and set up the operation vectors.
|
2008-10-30 07:36:14 +01:00
|
|
|
*
|
2015-02-23 12:38:08 +01:00
|
|
|
* 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. These callers have different criteria for clearing XFS_INEW, so leave
|
|
|
|
* it up to the caller to deal with unlocking the inode appropriately.
|
2008-08-13 08:23:13 +02:00
|
|
|
*/
|
|
|
|
void
|
|
|
|
xfs_setup_inode(
|
|
|
|
struct xfs_inode *ip)
|
|
|
|
{
|
2008-10-30 07:36:14 +01:00
|
|
|
struct inode *inode = &ip->i_vnode;
|
2013-10-29 12:11:57 +01:00
|
|
|
gfp_t gfp_mask;
|
2008-10-30 07:36:14 +01:00
|
|
|
|
|
|
|
inode->i_ino = ip->i_ino;
|
2009-12-17 14:25:01 +01:00
|
|
|
inode->i_state = I_NEW;
|
2010-10-23 13:15:32 +02:00
|
|
|
|
|
|
|
inode_sb_list_add(inode);
|
2010-11-06 12:43:08 +01:00
|
|
|
/* make the inode look hashed for the writeback code */
|
|
|
|
hlist_add_fake(&inode->i_hash);
|
2008-08-13 08:23:13 +02:00
|
|
|
|
|
|
|
inode->i_mode = ip->i_d.di_mode;
|
2011-10-28 14:13:29 +02:00
|
|
|
set_nlink(inode, ip->i_d.di_nlink);
|
2013-08-15 20:08:01 +02:00
|
|
|
inode->i_uid = xfs_uid_to_kuid(ip->i_d.di_uid);
|
|
|
|
inode->i_gid = xfs_gid_to_kgid(ip->i_d.di_gid);
|
2008-08-13 08:23:13 +02:00
|
|
|
|
|
|
|
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);
|
|
|
|
|
2013-10-29 12:11:51 +01:00
|
|
|
ip->d_ops = ip->i_mount->m_nondir_inode_ops;
|
2014-02-27 06:51:39 +01:00
|
|
|
lockdep_set_class(&ip->i_lock.mr_lock, &xfs_nondir_ilock_class);
|
2008-08-13 08:23:13 +02:00
|
|
|
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:
|
2014-02-27 06:51:39 +01:00
|
|
|
lockdep_set_class(&ip->i_lock.mr_lock, &xfs_dir_ilock_class);
|
2008-08-13 08:23:13 +02:00
|
|
|
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;
|
2013-10-29 12:11:46 +01:00
|
|
|
ip->d_ops = ip->i_mount->m_dir_inode_ops;
|
2008-08-13 08:23:13 +02:00
|
|
|
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;
|
|
|
|
}
|
|
|
|
|
2013-10-29 12:11:57 +01:00
|
|
|
/*
|
|
|
|
* Ensure all page cache allocations are done from GFP_NOFS context to
|
|
|
|
* prevent direct reclaim recursion back into the filesystem and blowing
|
|
|
|
* stacks or deadlocking.
|
|
|
|
*/
|
|
|
|
gfp_mask = mapping_gfp_mask(inode->i_mapping);
|
|
|
|
mapping_set_gfp_mask(inode->i_mapping, (gfp_mask & ~(__GFP_FS)));
|
|
|
|
|
2011-07-26 17:07:29 +02:00
|
|
|
/*
|
|
|
|
* If there is no attribute fork no ACL can exist on this inode,
|
|
|
|
* and it can't have any file capabilities attached to it either.
|
|
|
|
*/
|
|
|
|
if (!XFS_IFORK_Q(ip)) {
|
|
|
|
inode_has_no_xattr(inode);
|
2011-07-23 17:36:50 +02:00
|
|
|
cache_no_acl(inode);
|
2011-07-26 17:07:29 +02:00
|
|
|
}
|
2008-08-13 08:23:13 +02:00
|
|
|
}
|