btrfs: Defer setting new inode mode until after do_set_acl succeeds
PATCH v2 1/1 btrfs: Defer setting new inode mode until after do_set_acl succeeds
Currently a reference to inode->i_mode is passed directly to posix_acl_update_mode when setting an ACL which results in the inode's mode always being changed. In case of errors (e.g. in do_set_acl or even starting a transaction) the old mode needs to be re-assigned to ->i_mode. This mode recovery is done only in case do_set_acl fails, which leads to buggy behavior in case btrfs_start_transaction fails.
Fix it by simply setting the new mode to a temporary variable which is assigned to inode->i_mode's only when do_set_acl succeeds. This covers both failure cases explained above.
---
Changes since v1:
* Eliminated unrealted newline change (Johannes)
* Fixed logic of when i_mode is set, v1 was causing an uninitialised mode to be assigned to inode->i_mode resulting in fs consistency problems. Fix this by introducing a variable which is set to true when the i_mode needs to be changed. I know this variable could be eliminated by simply initialising mode = inode->i_mode and always assigning it in the if (!ret) branch but I find this somewhat subtle and rather be explicit with the boolean variable.
fs/btrfs/acl.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-)
fs/btrfs/acl.c +7 -5
@@ -112,14 +112,16 @@ static int do_set_acl(struct btrfs_trans_handle *trans, struct inode *inode,int btrfs_set_acl (struct inode *inode,struct posix_acl *acl,int type) {int ret; -umode_t old_mode = inode->i_mode; +umode_t mode; +bool change_mode = false;struct btrfs_trans_handle *trans;struct btrfs_root *root =BTRFS_I (inode)->root;if (type ==ACL_TYPE_ACCESS && acl) { - ret =posix_acl_update_mode (inode, &inode->i_mode, &acl); + ret =posix_acl_update_mode (inode, &mode, &acl);if (ret)return ret; + change_mode = true; } trans =btrfs_start_transaction (root,2 );
@@ -127,9 +129,9 @@ int btrfs_set_acl(struct inode *inode, struct posix_acl *acl, int type)return PTR_ERR (trans); ret =do_set_acl (trans, inode, acl, type); -if (ret) { - inode->i_mode = old_mode; - }else { +if (!ret) { +if (change_mode) + inode->i_mode = mode;inode_inc_iversion (inode); inode->i_ctime =current_time (inode);set_bit (BTRFS_INODE_COPY_EVERYTHING , &BTRFS_I (inode)->runtime_flags);
Signature · 1 line
2.17.1
On 21/03/2019 16:57, Nikolay Borisov wrote:
[...]
Quoted 6 lines
* Fixed logic of when
i_modeis set, v1 was causing an uninitialised mode to be assigned toinode->i_moderesulting in fs consistency problems. Fix this by introducing a variable which is set to true when thei_modeneeds to be changed. I know this variable could be eliminated by simply initialising mode =inode->i_modeand always assigning it in the if (!ret) branch but I find this somewhat subtle and rather be explicit with the boolean variable.
[...]
fs/btrfs/acl.c int btrfs_set_acl(struct inode *inode, struct posix_acl *acl, int type) · line 134
Quoted 3 lines
fs/btrfs/acl.c +3
-0+if (!ret) { +if (change_mode) + inode->i_mode = mode;
But what about initializing mode by inode->i_mode *and* adding a comment that this is either the saved state or the new one, depending on posix_acl_update_mode()'s outcome?
Signature · 6 lines
Johannes Thumshirn SUSE Labs Filesystems [email protected] +49 911 74053 689 SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg GF: Felix Imendörffer, Mary Higgins, Sri Rasiah HRB 21284 (AG Nürnberg) Key fingerprint = EC38 9CAB C2C4 F25D 8600 D0D0 0393 969D 2D76 0850
Quoted 15 lines · Nikolay Borisov
Currently a reference to
inode->i_modeis passed directly toposix_acl_update_modewhen setting an ACL which results in the inode's mode always being changed. In case of errors (e.g. indo_set_aclor even starting a transaction) the old mode needs to be re-assigned to ->i_mode. This mode recovery is done only in casedo_set_aclfails, which leads to buggy behavior in casebtrfs_start_transactionfails.Fix it by simply setting the new mode to a temporary variable which is assigned to inode->i_mode's only when
do_set_aclsucceeds. This covers both failure cases explained above.Fixes db0f220e98eb ("btrfs: start transaction in btrfs_set_acl")Signed-off-by Nikolay Borisov <[email protected]>---
Quoted 8 lines
Changes since v1:
* Eliminated unrealted newline change (Johannes)
* Fixed logic of wheni_modeis set, v1 was causing an uninitialised mode to be assigned toinode->i_moderesulting in fs consistency problems. Fix this by introducing a variable which is set to true when thei_modeneeds to be changed. I know this variable could be eliminated by simply initialising mode =inode->i_modeand always assigning it in the if (!ret) branch but I find this somewhat subtle and rather be explicit with the boolean variable.
Agreed. I folded the missing mode reset to the patch that introduced it as the Fixes: reference would not work, and adjusted your patch to simplify it again. Changelog adapted accordingly, only that it's not a faix but simplification. Please check it once it's pushed to misc-next.