linux/fs/ext2, branch v5.5

Merge tag 'for_v5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs

2019-11-30T19:16:07Z

Pull ext2, quota, reiserfs cleanups and fixes from Jan Kara: - Refactor the quota on/off kernel internal interfaces (mostly for ubifs quota support as ubifs does not want to have inodes holding quota information) - A few other small quota fixes and cleanups - Various small ext2 fixes and cleanups - Reiserfs xattr fix and one cleanup * tag 'for_v5.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs: (28 commits) ext2: code cleanup for descriptor_loc() fs/quota: handle overflows of sysctl fs.quota.* and report as unsigned long ext2: fix improper function comment ext2: code cleanup for ext2_try_to_allocate() ext2: skip unnecessary operations in ext2_try_to_allocate() ext2: Simplify initialization in ext2_try_to_allocate() ext2: code cleanup by calling ext2_group_last_block_no() ext2: introduce new helper ext2_group_last_block_no() reiserfs: replace open-coded atomic_dec_and_mutex_lock() ext2: check err when partial != NULL quota: Handle quotas without quota inodes in dquot_get_state() quota: Make dquot_disable() work without quota inodes quota: Drop dquot_enable() fs: Use dquot_load_quota_inode() from filesystems quota: Rename vfs_load_quota_inode() to dquot_load_quota_inode() quota: Simplify dquot_resume() quota: Factor out setup of quota inode quota: Check that quota is not dirty before release quota: fix livelock in dquot_writeback_dquots ext2: don't set *count in the case of failure in ext2_try_to_allocate() ...

ext2: code cleanup for descriptor_loc()

2019-11-19T16:04:38Z

Code cleanup by removing unnecessary variable in descriptor_loc(). Link: https://lore.kernel.org/r/20191115224900.2613-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu Signed-off-by: Jan Kara

ext2: fix improper function comment

2019-11-07T11:35:27Z

Just fix a improper function comment. Link: https://lore.kernel.org/r/20191104114036.9893-5-cgxu519@mykernel.net Signed-off-by: Chengguang Xu Signed-off-by: Jan Kara

ext2: code cleanup for ext2_try_to_allocate()

2019-11-07T11:35:27Z

Code cleanup by removing duplicated code. Link: https://lore.kernel.org/r/20191104114036.9893-4-cgxu519@mykernel.net Signed-off-by: Chengguang Xu Signed-off-by: Jan Kara

ext2: skip unnecessary operations in ext2_try_to_allocate()

2019-11-07T11:35:27Z

Move 'repeat' tag to proper place so that we can skip unnecessary operations in ext2_try_to_allocate(). Link: https://lore.kernel.org/r/20191104114036.9893-3-cgxu519@mykernel.net Signed-off-by: Chengguang Xu Signed-off-by: Jan Kara

ext2: Simplify initialization in ext2_try_to_allocate()

2019-11-07T11:35:22Z

Somewhat simplify the logic initializing search start and end in ext2_try_to_allocate(). No functional change. Signed-off-by: Jan Kara

ext2: code cleanup by calling ext2_group_last_block_no()

2019-11-06T15:25:15Z

Call common helper ext2_group_last_block_no() to calculate group last block number. Link: https://lore.kernel.org/r/20191104114036.9893-2-cgxu519@mykernel.net Signed-off-by: Chengguang Xu Signed-off-by: Jan Kara

ext2: introduce new helper ext2_group_last_block_no()

2019-11-06T15:25:12Z

Introduce new helper ext2_group_last_block_no() to calculate last block num for specific block group, we can replace open coded logic by calling this common helper. Link: https://lore.kernel.org/r/20191104114036.9893-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu Signed-off-by: Jan Kara

ext2: check err when partial != NULL

2019-11-05T09:42:18Z

Check err when partial == NULL is meaningless because partial == NULL means getting branch successfully without error. CC: stable@vger.kernel.org Link: https://lore.kernel.org/r/20191105045100.7104-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu Signed-off-by: Jan Kara

ext2: don't set *count in the case of failure in ext2_try_to_allocate()

2019-10-31T14:40:44Z

Currently we set *count to num(value 0) in the failure of block allocation in ext2_try_to_allocate(). Without reservation, we reuse *count(value 0) to retry block allocation and wrong *count will cause only allocating maximum 1 block even though having sufficent free blocks in that block group. Finally, it probably cause significant fragmentation. Link: https://lore.kernel.org/r/20191026090721.23794-1-cgxu519@mykernel.net Signed-off-by: Chengguang Xu Signed-off-by: Jan Kara