[kbuild-all] [djwong-xfs:dax-zeroinit-clear-poison-5.15 68/68] fs/ext4/extents.c:4695:8: warning: variable 'max_blocks' is used uninitialized whenever 'if' condition is true

tree: https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux.git dax-zeroinit-clear-poison-5.15 head: 6380ce8db23c88d70b9954ff66477719ea84bf3f commit: 6380ce8db23c88d70b9954ff66477719ea84bf3f [68/68] ext4: use DAX block device zeroout for FSDAX file ZERO_RANGE operations config: x86_64-randconfig-a011-20210816 (attached as .config) compiler: clang version 14.0.0 (https://github.com/llvm/llvm-project 44d0a99a12ec7ead4d2f5ef649ba05b40f6d463d) 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 # https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux.git/comm... git remote add djwong-xfs https://git.kernel.org/pub/scm/linux/kernel/git/djwong/xfs-linux.git git fetch --no-tags djwong-xfs dax-zeroinit-clear-poison-5.15 git checkout 6380ce8db23c88d70b9954ff66477719ea84bf3f # save the attached .config to linux build tree COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross ARCH=x86_64 If you fix the issue, kindly add following tag as appropriate Reported-by: kernel test robot <lkp(a)intel.com&gt; All warnings (new ones prefixed by >>): if (ret || did_zeroout) ^~~~~~~~~~~~~~~~~~ fs/ext4/extents.c:4741:43: note: uninitialized use occurs here trace_ext4_fallocate_exit(inode, offset, max_blocks, ret); ^~~~~~~~~~ fs/ext4/extents.c:4695:4: note: remove the 'if' if its condition is always false if (ret || did_zeroout) ^~~~~~~~~~~~~~~~~~~~~~~ if (ret || did_zeroout) ^~~ fs/ext4/extents.c:4741:43: note: uninitialized use occurs here trace_ext4_fallocate_exit(inode, offset, max_blocks, ret); ^~~~~~~~~~ fs/ext4/extents.c:4695:8: note: remove the '||' if its condition is always false if (ret || did_zeroout) ^~~~~~ fs/ext4/extents.c:4638:25: note: initialize the variable 'max_blocks' to silence this warning unsigned int max_blocks; ^ = 0 2 warnings generated. vim +4695 fs/ext4/extents.c 4626 4627 /* 4628 * preallocate space for a file. This implements ext4's fallocate file 4629 * operation, which gets called from sys_fallocate system call. 4630 * For block-mapped files, posix_fallocate should fall back to the method 4631 * of writing zeroes to the required new blocks (the same behavior which is 4632 * expected for file systems which do not support fallocate() system call). 4633 */ 4634 long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len) 4635 { 4636 struct inode *inode = file_inode(file); 4637 loff_t new_size = 0; 4638 unsigned int max_blocks; 4639 int ret = 0; 4640 int flags; 4641 ext4_lblk_t lblk; 4642 unsigned int blkbits = inode->i_blkbits; 4643 4644 /* 4645 * Encrypted inodes can't handle collapse range or insert 4646 * range since we would need to re-encrypt blocks with a 4647 * different IV or XTS tweak (which are based on the logical 4648 * block number). 4649 */ 4650 if (IS_ENCRYPTED(inode) && 4651 (mode & (FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_INSERT_RANGE))) 4652 return -EOPNOTSUPP; 4653 4654 /* Return error if mode is not supported */ 4655 if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE | 4656 FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_ZERO_RANGE | 4657 FALLOC_FL_INSERT_RANGE)) 4658 return -EOPNOTSUPP; 4659 4660 ext4_fc_start_update(inode); 4661 4662 if (mode & FALLOC_FL_PUNCH_HOLE) { 4663 ret = ext4_punch_hole(inode, offset, len); 4664 goto exit; 4665 } 4666 4667 ret = ext4_convert_inline_data(inode); 4668 if (ret) 4669 goto exit; 4670 4671 if (mode & FALLOC_FL_COLLAPSE_RANGE) { 4672 ret = ext4_collapse_range(inode, offset, len); 4673 goto exit; 4674 } 4675 4676 if (mode & FALLOC_FL_INSERT_RANGE) { 4677 ret = ext4_insert_range(inode, offset, len); 4678 goto exit; 4679 } 4680 4681 if (mode & FALLOC_FL_ZERO_RANGE) { 4682 /* 4683 * If the file is in DAX mode, try to use a DAX-specific 4684 * function to zero the region. 4685 */ 4686 if (IS_DAX(inode)) { 4687 bool did_zeroout = false; 4688 4689 inode_lock(inode); 4690 4691 ret = dax_zeroinit_range(inode, offset, len, 4692 &did_zeroout, &ext4_iomap_report_ops); 4693 if (ret == -EINVAL) 4694 ret = 0; 4696 goto out; 4697 4698 inode_unlock(inode); 4699 } 4700 ret = ext4_zero_range(file, offset, len, mode); 4701 goto exit; 4702 } 4703 trace_ext4_fallocate_enter(inode, offset, len, mode); 4704 lblk = offset >> blkbits; 4705 4706 max_blocks = EXT4_MAX_BLOCKS(len, offset, blkbits); 4707 flags = EXT4_GET_BLOCKS_CREATE_UNWRIT_EXT; 4708 4709 inode_lock(inode); 4710 4711 /* 4712 * We only support preallocation for extent-based files only 4713 */ 4714 if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) { 4715 ret = -EOPNOTSUPP; 4716 goto out; 4717 } 4718 4719 if (!(mode & FALLOC_FL_KEEP_SIZE) && 4720 (offset + len > inode->i_size || 4721 offset + len > EXT4_I(inode)->i_disksize)) { 4722 new_size = offset + len; 4723 ret = inode_newsize_ok(inode, new_size); 4724 if (ret) 4725 goto out; 4726 } 4727 4728 /* Wait all existing dio workers, newcomers will block on i_mutex */ 4729 inode_dio_wait(inode); 4730 4731 ret = ext4_alloc_file_blocks(file, lblk, max_blocks, new_size, flags); 4732 if (ret) 4733 goto out; 4734 4735 if (file->f_flags & O_SYNC && EXT4_SB(inode->i_sb)->s_journal) { 4736 ret = ext4_fc_commit(EXT4_SB(inode->i_sb)->s_journal, 4737 EXT4_I(inode)->i_sync_tid); 4738 } 4739 out: 4740 inode_unlock(inode); 4741 trace_ext4_fallocate_exit(inode, offset, max_blocks, ret); 4742 exit: 4743 ext4_fc_stop_update(inode); 4744 return ret; 4745 } 4746 --- 0-DAY CI Kernel Test Service, Intel Corporation https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org