The minstrel rate controller periodically looks up rate indexes in
a sampling table. When accessing a specific row and column, minstrel
correctly does a bounds check which, on the surface, appears to handle
the case where mi->n_rates < 2. However, mi->sample_idx is actually
defined as an unsigned, so the right hand side is taken to be a huge
positive number when negative, and the check will always fail.
Consequently, the RC will overrun the array and cause random memory
corruption when communicating with a peer that has only a single rate.
The max value of mi->sample_idx is around 25 so casting to int should
have no ill effects.
Without the change, uptime is a few minutes under load with an AP
that has a single hard-coded rate, and both the AP and STA could
potentially crash. With the change, both lasted 12 hours with a
steady load.
Thanks to Ognjen Maric for providing the single-rate clue so I could
reproduce this.
This fixes http://bugzilla.kernel.org/show_bug.cgi?id=12490 on the
regression list (also http://bugzilla.kernel.org/show_bug.cgi?id=13000).
Cc: [email protected]
Reported-by: Sergey S. Kostyliov <[email protected]>
Reported-by: Ognjen Maric <[email protected]>
Signed-off-by: Bob Copeland <[email protected]>
---
John & Felix, the patch itself may be too subtle so feel free to do it a
different way. However this is as minimal as it gets so I hope it can
be applied quickly to stable, and mainline if not too late.
net/mac80211/rc80211_minstrel.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/net/mac80211/rc80211_minstrel.c b/net/mac80211/rc80211_minstrel.c
index d9233ec..2681dfa 100644
--- a/net/mac80211/rc80211_minstrel.c
+++ b/net/mac80211/rc80211_minstrel.c
@@ -216,7 +216,7 @@ minstrel_get_next_sample(struct minstrel_sta_info *mi)
unsigned int sample_ndx;
sample_ndx = SAMPLE_TBL(mi, mi->sample_idx, mi->sample_column);
mi->sample_idx++;
- if (mi->sample_idx > (mi->n_rates - 2)) {
+ if ((int) mi->sample_idx > (mi->n_rates - 2)) {
mi->sample_idx = 0;
mi->sample_column++;
if (mi->sample_column >= SAMPLE_COLUMNS)
--
1.5.4.1
Bob Copeland wrote:
> The minstrel rate controller periodically looks up rate indexes in
> a sampling table. When accessing a specific row and column, minstrel
> correctly does a bounds check which, on the surface, appears to handle
> the case where mi->n_rates < 2. However, mi->sample_idx is actually
> defined as an unsigned, so the right hand side is taken to be a huge
> positive number when negative, and the check will always fail.
>
> Consequently, the RC will overrun the array and cause random memory
> corruption when communicating with a peer that has only a single rate.
> The max value of mi->sample_idx is around 25 so casting to int should
> have no ill effects.
>
> Without the change, uptime is a few minutes under load with an AP
> that has a single hard-coded rate, and both the AP and STA could
> potentially crash. With the change, both lasted 12 hours with a
> steady load.
>
> Thanks to Ognjen Maric for providing the single-rate clue so I could
> reproduce this.
>
> This fixes http://bugzilla.kernel.org/show_bug.cgi?id=12490 on the
> regression list (also http://bugzilla.kernel.org/show_bug.cgi?id=13000).
>
> Cc: [email protected]
> Reported-by: Sergey S. Kostyliov <[email protected]>
> Reported-by: Ognjen Maric <[email protected]>
> Signed-off-by: Bob Copeland <[email protected]>
> ---
>
> John & Felix, the patch itself may be too subtle so feel free to do it a
> different way. However this is as minimal as it gets so I hope it can
> be applied quickly to stable, and mainline if not too late.
How about changing the type of sample_idx to signed instead of casting
it? It's just a cosmetic thing, so even if you leave at this you get my
Acked-by: Felix Fietkau <[email protected]>
- Felix
On Fri, Jun 5, 2009 at 10:04 AM, Felix Fietkau<[email protected]> wrote:
>> John & Felix, the patch itself may be too subtle so feel free to do it a
>> different way. ?However this is as minimal as it gets so I hope it can
>> be applied quickly to stable, and mainline if not too late.
>
> How about changing the type of sample_idx to signed instead of casting
> it? It's just a cosmetic thing, so even if you leave at this you get my
Yeah, that is a better option for 2.6.31. I would be a bit more
comfortable with this version for now though since it's exactly
the code I tested.
> Acked-by: Felix Fietkau <[email protected]>
Thanks!
--
Bob Copeland %% http://www.bobcopeland.com
> This fixes http://bugzilla.kernel.org/show_bug.cgi?id=12490 on the
> regression list (also http://bugzilla.kernel.org/show_bug.cgi?id=13000).
The patch works well. No kernel panics anymore. You made my day!
Thank you,
Thiemo