Re: [PATCH] btrfs: Fix error handling in btrfs_cleanup_ordered_extents

From: kbuild test robot <lkp@intel.com>
To: Nikolay Borisov <nborisov@suse.com>
Cc: kbuild-all@01.org, dsterba@suse.com, linux-btrfs@vger.kernel.org,
	wqu@suse.com, Nikolay Borisov <nborisov@suse.com>
Subject: Re: [PATCH] btrfs: Fix error handling in btrfs_cleanup_ordered_extents
Date: Fri, 26 Oct 2018 20:33:04 +0800	[thread overview]
Message-ID: <201810262042.w1MOC2cE%fengguang.wu@intel.com> (raw)
In-Reply-To: <1540552427-8108-1-git-send-email-nborisov@suse.com>

[-- Attachment #1: Type: text/plain, Size: 3786 bytes --]

Hi Nikolay,

Thank you for the patch! Yet something to improve:

[auto build test ERROR on v4.19]
[also build test ERROR on next-20181019]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]

url:    https://github.com/0day-ci/linux/commits/Nikolay-Borisov/btrfs-Fix-error-handling-in-btrfs_cleanup_ordered_extents/20181026-194005
config: i386-randconfig-x007-201842 (attached as .config)
compiler: gcc-7 (Debian 7.3.0-1) 7.3.0
reproduce:
        # save the attached .config to linux build tree
        make ARCH=i386 

All errors (new ones prefixed by >>):

   fs//btrfs/inode.c: In function 'btrfs_cleanup_ordered_extents':
>> fs//btrfs/inode.c:140:63: error: expected ')' before '{' token
     if (page_start >= offset && page_end <= (offset + bytes - 1) {
                                                                  ^
>> fs//btrfs/inode.c:146:1: error: expected expression before '}' token
    }
    ^

vim +140 fs//btrfs/inode.c

    86	
    87	static int btrfs_setsize(struct inode *inode, struct iattr *attr);
    88	static int btrfs_truncate(struct inode *inode, bool skip_writeback);
    89	static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent);
    90	static noinline int cow_file_range(struct inode *inode,
    91					   struct page *locked_page,
    92					   u64 start, u64 end, u64 delalloc_end,
    93					   int *page_started, unsigned long *nr_written,
    94					   int unlock, struct btrfs_dedupe_hash *hash);
    95	static struct extent_map *create_io_em(struct inode *inode, u64 start, u64 len,
    96					       u64 orig_start, u64 block_start,
    97					       u64 block_len, u64 orig_block_len,
    98					       u64 ram_bytes, int compress_type,
    99					       int type);
   100	
   101	static void __endio_write_update_ordered(struct inode *inode,
   102						 const u64 offset, const u64 bytes,
   103						 const bool uptodate);
   104	
   105	/*
   106	 * Cleanup all submitted ordered extents in specified range to handle errors
   107	 * from the fill_dellaloc() callback.
   108	 *
   109	 * NOTE: caller must ensure that when an error happens, it can not call
   110	 * extent_clear_unlock_delalloc() to clear both the bits EXTENT_DO_ACCOUNTING
   111	 * and EXTENT_DELALLOC simultaneously, because that causes the reserved metadata
   112	 * to be released, which we want to happen only when finishing the ordered
   113	 * extent (btrfs_finish_ordered_io()).
   114	 */
   115	static inline void btrfs_cleanup_ordered_extents(struct inode *inode,
   116							 struct page *locked_page,
   117							 u64 offset, u64 bytes)
   118	{
   119		unsigned long index = offset >> PAGE_SHIFT;
   120		unsigned long end_index = (offset + bytes - 1) >> PAGE_SHIFT;
   121		u64 page_start = page_offset(locked_page);
   122		u64 page_end = page_start + PAGE_SIZE - 1;
   123	
   124		struct page *page;
   125	
   126		while (index <= end_index) {
   127			page = find_get_page(inode->i_mapping, index);
   128			index++;
   129			if (!page)
   130				continue;
   131			ClearPagePrivate2(page);
   132			put_page(page);
   133		}
   134	
   135		/*
   136		 * In case this page belongs to the delalloc range being instantiated
   137		 * then skip it, since the first page of a range is going to be
   138		 * properly cleaned up by the caller of run_delalloc_range
   139		 */
 > 140		if (page_start >= offset && page_end <= (offset + bytes - 1) {
   141			offset += PAGE_SIZE;
   142			bytes -= PAGE_SIZE;
   143		}
   144	
   145		return __endio_write_update_ordered(inode, offset, bytes, false);
 > 146	}
   147	

---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 28220 bytes --]