fs: Don't pass stack garbage to filesystem's get_block() in map_bh->b_size
PATCH 1/1 fs: Don't pass stack garbage to filesystem's get_block() in map_bh->b_size
The nobh_truncate_page() function is used by ext2, exofs, and jfs. Of these three, only ext2 and jfs's get_block() function pays attention to bh->b_size --- which is normally always the filesystem blocksize except when the get_block() function is called by either mpage_readpage(), mpage_readpages(), or the direct I/O routines in fs/direct_io.c.
Unfortunately, nobh_truncate_page() does not initialize map_bh before calling the filesystem-supplied get_block() function. So ext2 and jfs will try to calculate the number of blocks to map by taking stack garbage and shifting it left by inode->i_blkbits. This should be *mostly* harmless (except the filesystem will do some unnneeded work) unless the stack garbage is less than filesystem's blocksize, in which case maxblocks will be zero, and the attempt to find out whether or not the filesystem has a hole at a given logical block will fail, and the page cache entry might not get zero'ed out.
Also if the stack garbage in in map_bh->state happens to have the BH_Mapped bit set, there could be an attempt to call readpage() on a non-existent page, which could cause nobh_truncate_page() to return an error when it should not.
Fix this by initializing map_bh->state and map_bh->size.
Fortunately, it's probably fairly unlikely that ext2 and jfs users mount with nobh these days.
--- fs/buffer.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-)
fs/buffer.c +2 -0
@@ -2736,6 +2736,8 @@ has_buffers: pos += blocksize; } + map_bh.b_size = blocksize; + map_bh.b_state =0 ; err =get_block (inode, iblock, &map_bh,0 );if (err) goto unlock;
Signature · 1 line
1.6.3.rc4.1.g3e14.dirty
Hmm, I realized that I didn't put the name of the function that was misbehaving in the summary line. So how about changing the patch summary to be:
fs: Fix nobh_truncate_page() to not pass stack garbage to get_block()
- Ted
Al Viro
maintainerQuoted 5 lines · Theodore Tso
Hmm, I realized that I didn't put the name of the function that was misbehaving in the summary line. So how about changing the patch summary to be:
fs: Fix
nobh_truncate_page()to not pass stack garbage toget_block()
OK... Merged, will push to Linus in the next pile.
Quoted 17 lines · Theodore Ts'o
The
nobh_truncate_page()function is used by ext2, exofs, and jfs. Of these three, only ext2 and jfs'sget_block()function pays attention tobh->b_size--- which is normally always the filesystem blocksize except when theget_block()function is called by eithermpage_readpage(),mpage_readpages(), or the direct I/O routines infs/direct_io.c.Unfortunately,
nobh_truncate_page()does not initializemap_bhbefore calling the filesystem-suppliedget_block()function. So ext2 and jfs will try to calculate the number of blocks to map by taking stack garbage and shifting it left byinode->i_blkbits. This should be *mostly* harmless (except the filesystem will do some unnneeded work) unless the stack garbage is less than filesystem's blocksize, in which case maxblocks will be zero, and the attempt to find out whether or not the filesystem has a hole at a given logical block will fail, and the page cache entry might not get zero'ed out.
Quoted 6 lines
Also if the stack garbage in in
map_bh->statehappens to have theBH_Mappedbit set, there could be an attempt to callreadpage()on a non-existent page, which could causenobh_truncate_page()to return an error when it should not.Fix this by initializing
map_bh->stateandmap_bh->size.
This appears obviously correct.
Quoted 2 lines
Fortunately, it's probably fairly unlikely that ext2 and jfs users mount with nobh these days.
Well, jfs doesn't have a nobh mount option. It always calls the *_nobh routines. I don't really have a good excuse why.
Quoted 2 lines
Signed-off-by "Theodore Ts'o" <[email protected]>
Quoted 18 lines
--- fs/buffer.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-)fs/buffer.c +2
-0@@ -2736,6 +2736,8 @@ has_buffers: pos += blocksize; } + map_bh.b_size = blocksize; + map_bh.b_state =0 ; err =get_block (inode, iblock, &map_bh,0 );if (err) goto unlock;
Signature · 2 lines
David Kleikamp IBM Linux Technology Center