[kernel-sec-discuss] r2739 - active

[Ben Hutchings]

Author: benh
Date: 2012-08-10 01:09:50 +0000 (Fri, 10 Aug 2012)
New Revision: 2739

Modified:
   active/CVE-2012-2372
   active/CVE-2012-3412
Log:
Notes for CVE-2012-2372 and CVE-2012-3412

Modified: active/CVE-2012-2372
===================================================================
--- active/CVE-2012-2372	2012-08-03 06:38:06 UTC (rev 2738)
+++ active/CVE-2012-2372	2012-08-10 01:09:50 UTC (rev 2739)
@@ -4,6 +4,8 @@
 Notes:
  jmm> This could be a0c6ffbcfe600606b2d913dded4dc6b37b3bbbfd, but should be
  jmm> double-checked
+ bwh> Sounds more like 2e7b3b994529d4760231a45a6b88950187bda877 or
+ bwh> 77dd550e5547846604ff6f90c4dc6bba4414e485.
 Bugs:
 upstream:
 2.6.32-upstream-stable:

Modified: active/CVE-2012-3412
===================================================================
--- active/CVE-2012-3412	2012-08-03 06:38:06 UTC (rev 2738)
+++ active/CVE-2012-3412	2012-08-10 01:09:50 UTC (rev 2739)
@@ -3,6 +3,13 @@
  http://www.spinics.net/lists/netdev/msg206292.html
  https://bugzilla.redhat.com/show_bug.cgi?id=CVE-2012-3412
 Notes:
+ bwh> Critical changes are commit 30b678d844af3305cda5953467005cebb5d7b687
+ bwh> and 7e6d06f0de3f74ca929441add094518ae332257c.  The following commit
+ bwh> 7e6d06f0de3f74ca929441add094518ae332257c reduces the performance hit
+ bwh> for sfc in the extreme case that causes DoS, but it's higher risk and
+ bwh> I'm not aware of any customer hitting this.  There is an alternate
+ bwh> fix available that only touches the sfc driver, but with a greater
+ bwh> performance hit again.
 Bugs:
 upstream:
 2.6.32-upstream-stable: