Skip to contentSkip to conversation

ext4: refactor ext4_fallocate code

DISCUSSION Dan Carpenter 2 participants 2 messages Mar 28, 2014 – Mar 31, 2014

lore
Share & tools
lore
#0 lore raw

Hello Lukas Czerner,

The patch f6a7718e9581: "ext4: refactor ext4_fallocate code" from Mar 18, 2014, leads to the following static checker warning:

	fs/ext4/extents.c:4926 ext4_fallocate()
	warn: we tested 'ret' before and it was 'false'

fs/ext4/extents.c
4916 ret = ext4_alloc_file_blocks(file, lblk, max_blocks, flags, mode);
4917 if (ret)

                    ^^^

Checked.

4918 goto out;
4919
4920 handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
4921 if (IS_ERR(handle))
4922 goto out;
4923
4924 tv = inode->i_ctime = ext4_current_time(inode);
4925

  4926          if (!ret && new_size) {
                    ^^^^

Known.

4927 if (new_size > i_size_read(inode)) {
4928 i_size_write(inode, new_size);
4929 inode->i_mtime = tv;
4930 }
4931 if (new_size > EXT4_I(inode)->i_disksize)
4932 ext4_update_i_disksize(inode, new_size);

  4933          } else if (!ret && !new_size) {
                           ^^^^

Known.

  4934                  /*
  4935                  * Mark that we allocate beyond EOF so the subsequent truncate
  4936                  * can proceed even if the new size is the same as i_size.
  4937                  */
  4938                  if ((offset + len) > i_size_read(inode))

4939 ext4_set_inode_flag(inode, EXT4_INODE_EOFBLOCKS);
4940 }

regards,
dan carpenter

#1 lore raw
Quoted 38 lines · Dan Carpenter

Date: Fri, 28 Mar 2014 11:04:32 +0300
From: Dan Carpenter <[email protected]>
To: [email protected]
Cc: [email protected]
Subject: re: ext4: refactor ext4_fallocate code

Hello Lukas Czerner,

The patch f6a7718e9581: "ext4: refactor ext4_fallocate code" from Mar 18, 2014, leads to the following static checker warning:

	fs/ext4/extents.c:4926 ext4_fallocate()
	warn: we tested 'ret' before and it was 'false'

fs/ext4/extents.c
4916 ret = ext4_alloc_file_blocks(file, lblk, max_blocks, flags, mode);
4917 if (ret)

                    ^^^

Checked.

4918 goto out;
4919
4920 handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
4921 if (IS_ERR(handle))
4922 goto out;
4923
4924 tv = inode->i_ctime = ext4_current_time(inode);
4925

  4926          if (!ret && new_size) {
                    ^^^^

Known.

4927 if (new_size > i_size_read(inode)) {
4928 i_size_write(inode, new_size);
4929 inode->i_mtime = tv;
4930 }
4931 if (new_size > EXT4_I(inode)->i_disksize)
4932 ext4_update_i_disksize(inode, new_size);

Quoted 3 lines
  4933          } else if (!ret && !new_size) {
                           ^^^^

Known.

Right,

and this is true for the other function we;re using ext4_alloc_file_blocks().

Thanks!
-Lukas

Quoted 12 lines
  4934                  /*
  4935                  * Mark that we allocate beyond EOF so the subsequent truncate
  4936                  * can proceed even if the new size is the same as i_size.
  4937                  */
  4938                  if ((offset + len) > i_size_read(inode))

4939 ext4_set_inode_flag(inode, EXT4_INODE_EOFBLOCKS);
4940 }

regards,
dan carpenter