2018-12-15 16:27:32

by Richard Weinberger

[permalink] [raw]
Subject: [PATCH] jffs2: Fix integer underflow in jffs2_rtime_compress

The rtime compressor assumes that at least two bytes are
compressed.
If we try to compress just one byte, the loop condition will
wrap around and an out-of-bounds write happens.

Cc: <[email protected]>
Signed-off-by: Richard Weinberger <[email protected]>
---
fs/jffs2/compr_rtime.c | 3 +++
1 file changed, 3 insertions(+)

diff --git a/fs/jffs2/compr_rtime.c b/fs/jffs2/compr_rtime.c
index 406d9cc84ba8..cbf700001fc9 100644
--- a/fs/jffs2/compr_rtime.c
+++ b/fs/jffs2/compr_rtime.c
@@ -39,6 +39,9 @@ static int jffs2_rtime_compress(unsigned char *data_in,

memset(positions,0,sizeof(positions));

+ if (*dstlen < 2)
+ return -1;
+
while (pos < (*sourcelen) && outpos <= (*dstlen)-2) {
int backpos, runlen=0;
unsigned char value;
--
2.20.0



2018-12-20 15:03:37

by Richard Weinberger

[permalink] [raw]
Subject: Re: [PATCH] jffs2: Fix integer underflow in jffs2_rtime_compress

Am Donnerstag, 20. Dezember 2018, 11:43:08 CET schrieb Hou Tao:
>
> On 2018/12/16 0:23, Richard Weinberger wrote:
> > The rtime compressor assumes that at least two bytes are
> > compressed.
> > If we try to compress just one byte, the loop condition will
> > wrap around and an out-of-bounds write happens.
> >
> > Cc: <[email protected]>
> > Signed-off-by: Richard Weinberger <[email protected]>
> > ---
> > fs/jffs2/compr_rtime.c | 3 +++
> > 1 file changed, 3 insertions(+)
> > It seems that it doesn't incur any harm because the minimal allocated
> size will be 8-bytes and jffs2_rtime_compress() will write 2-bytes into
> the allocated buffer.

Are you sure about that? I saw odd kernel behavior and KASAN complained too.

Thanks,
//richard



2018-12-20 15:03:37

by Hou Tao

[permalink] [raw]
Subject: Re: [PATCH] jffs2: Fix integer underflow in jffs2_rtime_compress



On 2018/12/16 0:23, Richard Weinberger wrote:
> The rtime compressor assumes that at least two bytes are
> compressed.
> If we try to compress just one byte, the loop condition will
> wrap around and an out-of-bounds write happens.
>
> Cc: <[email protected]>
> Signed-off-by: Richard Weinberger <[email protected]>
> ---
> fs/jffs2/compr_rtime.c | 3 +++
> 1 file changed, 3 insertions(+)
> It seems that it doesn't incur any harm because the minimal allocated
size will be 8-bytes and jffs2_rtime_compress() will write 2-bytes into
the allocated buffer.

> diff --git a/fs/jffs2/compr_rtime.c b/fs/jffs2/compr_rtime.c
> index 406d9cc84ba8..cbf700001fc9 100644
> --- a/fs/jffs2/compr_rtime.c
> +++ b/fs/jffs2/compr_rtime.c
> @@ -39,6 +39,9 @@ static int jffs2_rtime_compress(unsigned char *data_in,
>
> memset(positions,0,sizeof(positions));
>
> + if (*dstlen < 2)
> + return -1;
> +
> while (pos < (*sourcelen) && outpos <= (*dstlen)-2) {
> int backpos, runlen=0;
> unsigned char value;
>


2020-01-23 02:26:10

by Hou Tao

[permalink] [raw]
Subject: Re: [PATCH] jffs2: Fix integer underflow in jffs2_rtime_compress

Hi Richard,

On 2018/12/20 18:45, Richard Weinberger wrote:
> Am Donnerstag, 20. Dezember 2018, 11:43:08 CET schrieb Hou Tao:
>>
>> On 2018/12/16 0:23, Richard Weinberger wrote:
>>> The rtime compressor assumes that at least two bytes are
>>> compressed.
>>> If we try to compress just one byte, the loop condition will
>>> wrap around and an out-of-bounds write happens.
>>>
>>> Cc: <[email protected]>
>>> Signed-off-by: Richard Weinberger <[email protected]>
>>> ---
>>> fs/jffs2/compr_rtime.c | 3 +++
>>> 1 file changed, 3 insertions(+)
>>> It seems that it doesn't incur any harm because the minimal allocated
>> size will be 8-bytes and jffs2_rtime_compress() will write 2-bytes into
>> the allocated buffer.
>
> Are you sure about that? I saw odd kernel behavior and KASAN complained too.
>

Sorry for the later reply.

Yes. KASAN complains but it doesn't incur any harm because the minimal allocated
size returned by kmalloc() will be 8-bytes.

But we better fix it, because it's bad to depend on the implementation of kmalloc().

It seems that mtd-utils has already fixed it years ago. Maybe we can use it directly ?

And your fix also looks good to me, so

Reviewed-by: Hou Tao <[email protected]>

commit e8457f16306ad6e2c8708275bf42b5dfff40fffd
Author: Enrico Scholz <[email protected]>
Date: Thu Jun 24 15:02:40 2010 +0200

mkfs.jffs2: fix integer underflow in jffs2_rtime_compress()

When '*dstlen' is 0 or 1, comparison will return wrong result. Reported
by valgrind as

==5919== Invalid write of size 1
==5919== at 0x40564E: jffs2_rtime_compress (compr_rtime.c:51)
==5919== by 0x40676B: jffs2_compress (compr.c:246)
==5919== by 0x403EE4: recursive_populate_directory (mkfs.jffs2.c:884)
==5919== Address 0x4e1bdb1 is 0 bytes after a block of size 1 alloc'd
==5919== at 0x4A0515D: malloc (vg_replace_malloc.c:195)
==5919== by 0x40671C: jffs2_compress (compr.c:229)
==5919== by 0x403EE4: recursive_populate_directory (mkfs.jffs2.c:884)

Signed-off-by: Enrico Scholz <[email protected]>
Signed-off-by: Artem Bityutskiy <[email protected]>

diff --git a/compr_rtime.c b/compr_rtime.c
index 131536c..5613963 100644
--- a/compr_rtime.c
+++ b/compr_rtime.c
@@ -32,7 +32,7 @@ static int jffs2_rtime_compress(unsigned char *data_in, unsigned char *cpage_out

memset(positions,0,sizeof(positions));

- while (pos < (*sourcelen) && outpos <= (*dstlen)-2) {
+ while (pos < (*sourcelen) && outpos+2 <= (*dstlen)) {
int backpos, runlen=0;
unsigned char value;



> Thanks,
> //richard




>
>
>
> .
>