Re: [RFC PATCH 3/4] ext4: factor out write end code of inline file

[kernel test robot <lkp@intel.com>]

[-- Attachment #1: Type: text/plain, Size: 5480 bytes --]

Hi Zhang,

[FYI, it's a private test report for your RFC patch.]
[auto build test WARNING on ext4/dev]
[also build test WARNING on v5.13 next-20210701]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]

url:    https://github.com/0day-ci/linux/commits/Zhang-Yi/ext4-improve-delalloc-buffer-write-performance/20210706-103431
base:   https://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4.git dev
config: arm-randconfig-r016-20210706 (attached as .config)
compiler: clang version 13.0.0 (https://github.com/llvm/llvm-project 873e8b96b1226d64e4f95083147d8592ba7bd5d8)
reproduce (this is a W=1 build):
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # install arm cross compiling tool for clang build
        # apt-get install binutils-arm-linux-gnueabi
        # https://github.com/0day-ci/linux/commit/b6bfc15d82616d158c3b969e1796d4848b70e354
        git remote add linux-review https://github.com/0day-ci/linux
        git fetch --no-tags linux-review Zhang-Yi/ext4-improve-delalloc-buffer-write-performance/20210706-103431
        git checkout b6bfc15d82616d158c3b969e1796d4848b70e354
        # save the attached .config to linux build tree
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross ARCH=arm 

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>

All warnings (new ones prefixed by >>):

>> fs/ext4/inline.c:742:6: warning: variable 'ret' is used uninitialized whenever 'if' condition is false [-Wsometimes-uninitialized]
           if (likely(copied)) {
               ^~~~~~~~~~~~~~
   include/linux/compiler.h:77:20: note: expanded from macro 'likely'
   # define likely(x)      __builtin_expect(!!(x), 1)
                           ^~~~~~~~~~~~~~~~~~~~~~~~~~
   fs/ext4/inline.c:793:7: note: uninitialized use occurs here
           if (!ret)
                ^~~
   fs/ext4/inline.c:742:2: note: remove the 'if' if its condition is always true
           if (likely(copied)) {
           ^~~~~~~~~~~~~~~~~~~~
   fs/ext4/inline.c:737:9: note: initialize the variable 'ret' to silence this warning
           int ret, ret2;
                  ^
                   = 0
   1 warning generated.


vim +742 fs/ext4/inline.c

   728	
   729	int ext4_write_inline_data_end(struct inode *inode, loff_t pos, unsigned len,
   730				       unsigned copied, struct page *page)
   731	{
   732		handle_t *handle = ext4_journal_current_handle();
   733		int i_size_changed = 0;
   734		int no_expand;
   735		void *kaddr;
   736		struct ext4_iloc iloc;
   737		int ret, ret2;
   738	
   739		if (unlikely(copied < len) && !PageUptodate(page))
   740			copied = 0;
   741	
 > 742		if (likely(copied)) {
   743			ret = ext4_get_inode_loc(inode, &iloc);
   744			if (ret) {
   745				unlock_page(page);
   746				put_page(page);
   747				ext4_std_error(inode->i_sb, ret);
   748				goto out;
   749			}
   750			ext4_write_lock_xattr(inode, &no_expand);
   751			BUG_ON(!ext4_has_inline_data(inode));
   752	
   753			kaddr = kmap_atomic(page);
   754			ext4_write_inline_data(inode, &iloc, kaddr, pos, copied);
   755			kunmap_atomic(kaddr);
   756			SetPageUptodate(page);
   757			/* clear page dirty so that writepages wouldn't work for us. */
   758			ClearPageDirty(page);
   759	
   760			ext4_write_unlock_xattr(inode, &no_expand);
   761			brelse(iloc.bh);
   762		}
   763	
   764		/*
   765		 * It's important to update i_size while still holding page lock:
   766		 * page writeout could otherwise come in and zero beyond i_size.
   767		 */
   768		i_size_changed = ext4_update_inode_size(inode, pos + copied);
   769		if (ext4_should_journal_data(inode)) {
   770			ext4_set_inode_state(inode, EXT4_STATE_JDATA);
   771			EXT4_I(inode)->i_datasync_tid = handle->h_transaction->t_tid;
   772		}
   773		unlock_page(page);
   774		put_page(page);
   775	
   776		/*
   777		 * Don't mark the inode dirty under page lock. First, it unnecessarily
   778		 * makes the holding time of page lock longer. Second, it forces lock
   779		 * ordering of page lock and transaction start for journaling
   780		 * filesystems.
   781		 */
   782		if (likely(copied) || i_size_changed)
   783			mark_inode_dirty(inode);
   784	out:
   785		/*
   786		 * If we have allocated more blocks and copied less. We will have
   787		 * blocks allocated outside inode->i_size, so truncate them.
   788		 */
   789		if (pos + len > inode->i_size && ext4_can_truncate(inode))
   790			ext4_orphan_add(handle, inode);
   791	
   792		ret2 = ext4_journal_stop(handle);
   793		if (!ret)
   794			ret = ret2;
   795		if (pos + len > inode->i_size) {
   796			ext4_truncate_failed_write(inode);
   797			/*
   798			 * If truncate failed early the inode might still be
   799			 * on the orphan list; we need to make sure the inode
   800			 * is removed from the orphan list in that case.
   801			 */
   802			if (inode->i_nlink)
   803				ext4_orphan_del(NULL, inode);
   804		}
   805		return ret ? ret : copied;
   806	}
   807	

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org

[-- Attachment #2: config.gz --]
[-- Type: application/gzip, Size: 36949 bytes --]