CVE-2009-4537: Input Validation

Description of problem: This was disclosed at 26c3.

Fabian also mentioned the fix for CVE-2009-1389 regarding the r8169 driver introduces a similar security problem as this: http://git.kernel.org/linus/fdd7b4c3302c93f6833e338903ea77245eb510b4

- RTLW16(RxMaxSize, 16383); + RTLW16(RxMaxSize, rxbufsz);

is actually a revert of this: http://git.kernel.org/linus/126fa4b9ca5d9d7cb7d46f779ad3bd3631ca387c

- / For gigabit rtl8169, MTU + header + CRC + VLAN / - RTLW16(RxMaxSize, tp->rxbufsz); + / Low hurts. Let's disable the filtering. / + RTLW16(RxMaxSize, 16383);

The accompanying comment for the original commit (126fa):

The size of the incoming frame is not correctly checked.

The RxMaxSize register (0xDA) does not work as expected and incoming frames whose size exceeds the MTU actually end spanning multiple descriptors. The first Rx descriptor contains the size of the whole frame (or some garbage in its place). The driver does not expect something above the space allocated to the current skb and crashes loudly when it issues a skbput.

The fix contains two parts: - disable hardware Rx size filtering: so far it only proved to be able to trigger some new fancy errors; [...]

warned of issues when using hardware Rx size filtering and Fabian claimed to be able to trigger something similar to the e1000 bug when using a specific packet size.

References: http://events.ccc.de/congress/2009/Fahrplan//events/3596.en.html http://blog.c22.cc/2009/12/27/26c3-cat-procsysnetipv4fuckups/ http://twitter.com/dakami/statuses/7104238406 https://bugzilla.redhat.com/CVE-2009-1389 http://www.securityfocus.com/bid/37521