[PATCH] Btrfs: send, fix corrupted paths strings for long paths

[PATCH] Btrfs: send, fix corrupted paths strings for long paths

[Filipe David Borba Manana]

If a path has more than 230 characters, we allocate a new buffer to
use for the path, but we were forgotting to copy the contents of the
previous buffer into the new one, which has random content from the
kmalloc call.

Test:

    mkfs.btrfs -f /dev/sdd
    mount /dev/sdd /mnt

    TEST_PATH="/mnt/fdmanana/.config/google-chrome-mysetup/Default/Pepper_Data/Shockwave_Flash/WritableRoot/#SharedObjects/JSHJ4ZKN/s.wsj.net/[[IMPORT]]/players.edgesuite.net/flash/plugins/osmf/advanced-streaming-plugin/v2.7/osmf1.6/Ak#"
    mkdir -p $TEST_PATH
    echo "hello world" > $TEST_PATH/amaiAdvancedStreamingPlugin.txt

    btrfs subvolume snapshot -r /mnt /mnt/mysnap1
    btrfs send /mnt/mysnap1 -f /tmp/1.snap

A test for xfstests follows.

Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Cc: Marc Merlin <marc@merlins.org>
---
 fs/btrfs/send.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index f6bbc1e..70c5e8c 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -368,10 +368,13 @@ static int fs_path_ensure_buf(struct fs_path *p, int len)
 	/*
 	 * First time the inline_buf does not suffice
 	 */
-	if (p->buf == p->inline_buf)
+	if (p->buf == p->inline_buf) {
 		tmp_buf = kmalloc(len, GFP_NOFS);
-	else
+		if (tmp_buf)
+			memcpy(tmp_buf, p->buf, old_buf_len);
+	} else {
 		tmp_buf = krealloc(p->buf, len, GFP_NOFS);
+	}
 	if (!tmp_buf)
 		return -ENOMEM;
 	p->buf = tmp_buf;
-- 
1.9.1

[PATCH v2] Btrfs: send, fix corrupted path strings for long paths

[Filipe David Borba Manana]

If a path has more than 230 characters, we allocate a new buffer to
use for the path, but we were forgotting to copy the contents of the
previous buffer into the new one, which has random content from the
kmalloc call.

Test:

    mkfs.btrfs -f /dev/sdd
    mount /dev/sdd /mnt

    TEST_PATH="/mnt/fdmanana/.config/google-chrome-mysetup/Default/Pepper_Data/Shockwave_Flash/WritableRoot/#SharedObjects/JSHJ4ZKN/s.wsj.net/[[IMPORT]]/players.edgesuite.net/flash/plugins/osmf/advanced-streaming-plugin/v2.7/osmf1.6/Ak#"
    mkdir -p $TEST_PATH
    echo "hello world" > $TEST_PATH/amaiAdvancedStreamingPlugin.txt

    btrfs subvolume snapshot -r /mnt /mnt/mysnap1
    btrfs send /mnt/mysnap1 -f /tmp/1.snap

A test for xfstests follows.

Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Cc: Marc Merlin <marc@merlins.org>
---

V2: Fix change title, "paths" to "path".

 fs/btrfs/send.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
index f6bbc1e..70c5e8c 100644
--- a/fs/btrfs/send.c
+++ b/fs/btrfs/send.c
@@ -368,10 +368,13 @@ static int fs_path_ensure_buf(struct fs_path *p, int len)
 	/*
 	 * First time the inline_buf does not suffice
 	 */
-	if (p->buf == p->inline_buf)
+	if (p->buf == p->inline_buf) {
 		tmp_buf = kmalloc(len, GFP_NOFS);
-	else
+		if (tmp_buf)
+			memcpy(tmp_buf, p->buf, old_buf_len);
+	} else {
 		tmp_buf = krealloc(p->buf, len, GFP_NOFS);
+	}
 	if (!tmp_buf)
 		return -ENOMEM;
 	p->buf = tmp_buf;
-- 
1.9.1

Re: [PATCH v2] Btrfs: send, fix corrupted path strings for long paths

[Marc MERLIN]

On Wed, May 21, 2014 at 05:38:13PM +0100, Filipe David Borba Manana wrote:
> If a path has more than 230 characters, we allocate a new buffer to
> use for the path, but we were forgotting to copy the contents of the
> previous buffer into the new one, which has random content from the
> kmalloc call.

I've confirmed this fixes the problem I was seeing when applied to
3.15rc5.
Thanks for taking that down.

Tested-by: Marc MERLIN <marc@merlins.org>

(while we're at it, I post with my personal address, but I work at
Google, I'm supposed to state that :) )

Marc

> Test:
> 
>     mkfs.btrfs -f /dev/sdd
>     mount /dev/sdd /mnt
> 
>     TEST_PATH="/mnt/fdmanana/.config/google-chrome-mysetup/Default/Pepper_Data/Shockwave_Flash/WritableRoot/#SharedObjects/JSHJ4ZKN/s.wsj.net/[[IMPORT]]/players.edgesuite.net/flash/plugins/osmf/advanced-streaming-plugin/v2.7/osmf1.6/Ak#"
>     mkdir -p $TEST_PATH
>     echo "hello world" > $TEST_PATH/amaiAdvancedStreamingPlugin.txt
> 
>     btrfs subvolume snapshot -r /mnt /mnt/mysnap1
>     btrfs send /mnt/mysnap1 -f /tmp/1.snap
> 
> A test for xfstests follows.
> 
> Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
> Cc: Marc Merlin <marc@merlins.org>
> ---
> 
> V2: Fix change title, "paths" to "path".
> 
>  fs/btrfs/send.c | 7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c
> index f6bbc1e..70c5e8c 100644
> --- a/fs/btrfs/send.c
> +++ b/fs/btrfs/send.c
> @@ -368,10 +368,13 @@ static int fs_path_ensure_buf(struct fs_path *p, int len)
>  	/*
>  	 * First time the inline_buf does not suffice
>  	 */
> -	if (p->buf == p->inline_buf)
> +	if (p->buf == p->inline_buf) {
>  		tmp_buf = kmalloc(len, GFP_NOFS);
> -	else
> +		if (tmp_buf)
> +			memcpy(tmp_buf, p->buf, old_buf_len);
> +	} else {
>  		tmp_buf = krealloc(p->buf, len, GFP_NOFS);
> +	}
>  	if (!tmp_buf)
>  		return -ENOMEM;
>  	p->buf = tmp_buf;
> -- 
> 1.9.1
> 
> 

-- 
"A mouse is a device used to point at the xterm you want to type in" - A.S.R.
Microsoft is to operating systems ....
                                      .... what McDonalds is to gourmet cooking
Home page: http://marc.merlins.org/                         | PGP 1024R/763BE901