The patch titled
Subject: ocfs2: fix posix_acl_create deadlock
has been added to the -mm tree. Its filename is
ocfs2-fix-posix_acl_create-deadlock.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/ocfs2-fix-posix_acl_create-deadlock.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/ocfs2-fix-posix_acl_create-deadlock.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Junxiao Bi <junxiao.bi@xxxxxxxxxx>
Subject: ocfs2: fix posix_acl_create deadlock
commit 702e5bc68ad2 ("ocfs2: use generic posix ACL infrastructure")
refactored code to use posix_acl_create. The problem with this function
is that it is not mindful of the cluster wide inode lock making it
unsuitable for use with ocfs2 inode creation with ACLs. For example, when
used in ocfs2_mknod, this function can cause deadlock as follows. The
parent dir inode lock is taken when calling posix_acl_create -> get_acl ->
ocfs2_iop_get_acl which takes the inode lock again. This can cause
deadlock if there is a blocked remote lock request waiting for the lock to
be downconverted. And same deadlock happened in ocfs2_reflink. This fix
is to revert back using ocfs2_init_acl.
Fixes: 702e5bc68ad2 ("ocfs2: use generic posix ACL infrastructure")
Signed-off-by: Tariq Saeed <tariq.x.saeed@xxxxxxxxxx>
Signed-off-by: Junxiao Bi <junxiao.bi@xxxxxxxxxx>
Cc: Mark Fasheh <mfasheh@xxxxxxx>
Cc: Joel Becker <jlbec@xxxxxxxxxxxx>
Cc: Joseph Qi <joseph.qi@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
fs/ocfs2/acl.c | 63 ++++++++++++++++++++++++++++++++++++++
fs/ocfs2/acl.h | 4 ++
fs/ocfs2/namei.c | 23 +------------
fs/ocfs2/refcounttree.c | 17 +---------
fs/ocfs2/xattr.c | 14 +++-----
fs/ocfs2/xattr.h | 4 --
6 files changed, 77 insertions(+), 48 deletions(-)
diff -puN fs/ocfs2/acl.c~ocfs2-fix-posix_acl_create-deadlock fs/ocfs2/acl.c
--- a/fs/ocfs2/acl.c~ocfs2-fix-posix_acl_create-deadlock
+++ a/fs/ocfs2/acl.c
@@ -346,3 +346,66 @@ int ocfs2_acl_chmod(struct inode *inode,
posix_acl_release(acl);
return ret;
}
+
+/*
+ * Initialize the ACLs of a new inode. If parent directory has default ACL,
+ * then clone to new inode. Called from ocfs2_mknod.
+ */
+int ocfs2_init_acl(handle_t *handle,
+ struct inode *inode,
+ struct inode *dir,
+ struct buffer_head *di_bh,
+ struct buffer_head *dir_bh,
+ struct ocfs2_alloc_context *meta_ac,
+ struct ocfs2_alloc_context *data_ac)
+{
+ struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+ struct posix_acl *acl = NULL;
+ int ret = 0, ret2;
+ umode_t mode;
+
+ if (!S_ISLNK(inode->i_mode)) {
+ if (osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL) {
+ acl = ocfs2_get_acl_nolock(dir, ACL_TYPE_DEFAULT,
+ dir_bh);
+ if (IS_ERR(acl))
+ return PTR_ERR(acl);
+ }
+ if (!acl) {
+ mode = inode->i_mode & ~current_umask();
+ ret = ocfs2_acl_set_mode(inode, di_bh, handle, mode);
+ if (ret) {
+ mlog_errno(ret);
+ goto cleanup;
+ }
+ }
+ }
+ if ((osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL) && acl) {
+ if (S_ISDIR(inode->i_mode)) {
+ ret = ocfs2_set_acl(handle, inode, di_bh,
+ ACL_TYPE_DEFAULT, acl,
+ meta_ac, data_ac);
+ if (ret)
+ goto cleanup;
+ }
+ mode = inode->i_mode;
+ ret = __posix_acl_create(&acl, GFP_NOFS, &mode);
+ if (ret < 0)
+ return ret;
+
+ ret2 = ocfs2_acl_set_mode(inode, di_bh, handle, mode);
+ if (ret2) {
+ mlog_errno(ret2);
+ ret = ret2;
+ goto cleanup;
+ }
+ if (ret > 0) {
+ ret = ocfs2_set_acl(handle, inode,
+ di_bh, ACL_TYPE_ACCESS,
+ acl, meta_ac, data_ac);
+ }
+ }
+cleanup:
+ posix_acl_release(acl);
+ return ret;
+}
diff -puN fs/ocfs2/acl.h~ocfs2-fix-posix_acl_create-deadlock fs/ocfs2/acl.h
--- a/fs/ocfs2/acl.h~ocfs2-fix-posix_acl_create-deadlock
+++ a/fs/ocfs2/acl.h
@@ -36,5 +36,9 @@ int ocfs2_set_acl(handle_t *handle,
struct ocfs2_alloc_context *meta_ac,
struct ocfs2_alloc_context *data_ac);
extern int ocfs2_acl_chmod(struct inode *, struct buffer_head *);
+extern int ocfs2_init_acl(handle_t *, struct inode *, struct inode *,
+ struct buffer_head *, struct buffer_head *,
+ struct ocfs2_alloc_context *,
+ struct ocfs2_alloc_context *);
#endif /* OCFS2_ACL_H */
diff -puN fs/ocfs2/namei.c~ocfs2-fix-posix_acl_create-deadlock fs/ocfs2/namei.c
--- a/fs/ocfs2/namei.c~ocfs2-fix-posix_acl_create-deadlock
+++ a/fs/ocfs2/namei.c
@@ -259,7 +259,6 @@ static int ocfs2_mknod(struct inode *dir
struct ocfs2_dir_lookup_result lookup = { NULL, };
sigset_t oldset;
int did_block_signals = 0;
- struct posix_acl *default_acl = NULL, *acl = NULL;
struct ocfs2_dentry_lock *dl = NULL;
trace_ocfs2_mknod(dir, dentry, dentry->d_name.len, dentry->d_name.name,
@@ -367,12 +366,6 @@ static int ocfs2_mknod(struct inode *dir
goto leave;
}
- status = posix_acl_create(dir, &inode->i_mode, &default_acl, &acl);
- if (status) {
- mlog_errno(status);
- goto leave;
- }
-
handle = ocfs2_start_trans(osb, ocfs2_mknod_credits(osb->sb,
S_ISDIR(mode),
xattr_credits));
@@ -421,16 +414,8 @@ static int ocfs2_mknod(struct inode *dir
inc_nlink(dir);
}
- if (default_acl) {
- status = ocfs2_set_acl(handle, inode, new_fe_bh,
- ACL_TYPE_DEFAULT, default_acl,
- meta_ac, data_ac);
- }
- if (!status && acl) {
- status = ocfs2_set_acl(handle, inode, new_fe_bh,
- ACL_TYPE_ACCESS, acl,
- meta_ac, data_ac);
- }
+ status = ocfs2_init_acl(handle, inode, dir, new_fe_bh, parent_fe_bh,
+ meta_ac, data_ac);
if (status < 0) {
mlog_errno(status);
@@ -472,10 +457,6 @@ static int ocfs2_mknod(struct inode *dir
d_instantiate(dentry, inode);
status = 0;
leave:
- if (default_acl)
- posix_acl_release(default_acl);
- if (acl)
- posix_acl_release(acl);
if (status < 0 && did_quota_inode)
dquot_free_inode(inode);
if (handle)
diff -puN fs/ocfs2/refcounttree.c~ocfs2-fix-posix_acl_create-deadlock fs/ocfs2/refcounttree.c
--- a/fs/ocfs2/refcounttree.c~ocfs2-fix-posix_acl_create-deadlock
+++ a/fs/ocfs2/refcounttree.c
@@ -4248,20 +4248,12 @@ static int ocfs2_reflink(struct dentry *
struct inode *inode = d_inode(old_dentry);
struct buffer_head *old_bh = NULL;
struct inode *new_orphan_inode = NULL;
- struct posix_acl *default_acl, *acl;
- umode_t mode;
if (!ocfs2_refcount_tree(OCFS2_SB(inode->i_sb)))
return -EOPNOTSUPP;
- mode = inode->i_mode;
- error = posix_acl_create(dir, &mode, &default_acl, &acl);
- if (error) {
- mlog_errno(error);
- return error;
- }
- error = ocfs2_create_inode_in_orphan(dir, mode,
+ error = ocfs2_create_inode_in_orphan(dir, inode->i_mode,
&new_orphan_inode);
if (error) {
mlog_errno(error);
@@ -4300,16 +4292,11 @@ static int ocfs2_reflink(struct dentry *
/* If the security isn't preserved, we need to re-initialize them. */
if (!preserve) {
error = ocfs2_init_security_and_acl(dir, new_orphan_inode,
- &new_dentry->d_name,
- default_acl, acl);
+ &new_dentry->d_name);
if (error)
mlog_errno(error);
}
out:
- if (default_acl)
- posix_acl_release(default_acl);
- if (acl)
- posix_acl_release(acl);
if (!error) {
error = ocfs2_mv_orphaned_inode_to_new(dir, new_orphan_inode,
new_dentry);
diff -puN fs/ocfs2/xattr.c~ocfs2-fix-posix_acl_create-deadlock fs/ocfs2/xattr.c
--- a/fs/ocfs2/xattr.c~ocfs2-fix-posix_acl_create-deadlock
+++ a/fs/ocfs2/xattr.c
@@ -7216,12 +7216,10 @@ out:
*/
int ocfs2_init_security_and_acl(struct inode *dir,
struct inode *inode,
- const struct qstr *qstr,
- struct posix_acl *default_acl,
- struct posix_acl *acl)
+ const struct qstr *qstr)
{
- struct buffer_head *dir_bh = NULL;
int ret = 0;
+ struct buffer_head *dir_bh = NULL;
ret = ocfs2_init_security_get(inode, dir, qstr, NULL);
if (ret) {
@@ -7234,11 +7232,9 @@ int ocfs2_init_security_and_acl(struct i
mlog_errno(ret);
goto leave;
}
-
- if (!ret && default_acl)
- ret = ocfs2_iop_set_acl(inode, default_acl, ACL_TYPE_DEFAULT);
- if (!ret && acl)
- ret = ocfs2_iop_set_acl(inode, acl, ACL_TYPE_ACCESS);
+ ret = ocfs2_init_acl(NULL, inode, dir, NULL, dir_bh, NULL, NULL);
+ if (ret)
+ mlog_errno(ret);
ocfs2_inode_unlock(dir, 0);
brelse(dir_bh);
diff -puN fs/ocfs2/xattr.h~ocfs2-fix-posix_acl_create-deadlock fs/ocfs2/xattr.h
--- a/fs/ocfs2/xattr.h~ocfs2-fix-posix_acl_create-deadlock
+++ a/fs/ocfs2/xattr.h
@@ -94,7 +94,5 @@ int ocfs2_reflink_xattrs(struct inode *o
bool preserve_security);
int ocfs2_init_security_and_acl(struct inode *dir,
struct inode *inode,
- const struct qstr *qstr,
- struct posix_acl *default_acl,
- struct posix_acl *acl);
+ const struct qstr *qstr);
#endif /* OCFS2_XATTR_H */
_
Patches currently in -mm which might be from junxiao.bi@xxxxxxxxxx are
ocfs2-revert-using-ocfs2_acl_chmod-to-avoid-inode-cluster-lock-hang.patch
ocfs2-fix-posix_acl_create-deadlock.patch
ocfs2-o2hb-add-negotiate-timer.patch
ocfs2-o2hb-add-negotiate-timer-v2.patch
ocfs2-o2hb-add-nego_timeout-message.patch
ocfs2-o2hb-add-nego_timeout-message-v2.patch
ocfs2-o2hb-add-negotiate_approve-message.patch
ocfs2-o2hb-add-negotiate_approve-message-v2.patch
ocfs2-o2hb-add-some-user-debug-log.patch
ocfs2-o2hb-add-some-user-debug-log-v2.patch
ocfs2-o2hb-dont-negotiate-if-last-hb-fail.patch
ocfs2-o2hb-fix-hb-hung-time.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
