From: David Barksdale <amatus@ocgnet.org>
This patch against the 2.6.20-rc6 kernel fixes a dangling pointer bug in
ipmi_timeout_handler. A list of timedout messages is not re-initialized
before reuse, causing the head of the list to point to freed memory.
Signed-off-by: David Barksdale <amatus@ocgnet.org>
Signed-off-by: Corey Minyard <minyard@acm.org>
---
diff -ruNp linux-2.6.git.orig/drivers/char/ipmi/ipmi_msghandler.c linux-2.6.git/drivers/char/ipmi/ipmi_msghandler.c
--- linux-2.6.git.orig/drivers/char/ipmi/ipmi_msghandler.c 2007-01-30 10:32:14.000000000 -0600
+++ linux-2.6.git/drivers/char/ipmi/ipmi_msghandler.c 2007-01-30 10:33:50.000000000 -0600
@@ -3649,8 +3649,6 @@ static void ipmi_timeout_handler(long ti
unsigned long flags;
int i;
- INIT_LIST_HEAD(&timeouts);
-
rcu_read_lock();
list_for_each_entry_rcu(intf, &ipmi_interfaces, link) {
/* See if any waiting messages need to be processed. */
@@ -3671,6 +3669,7 @@ static void ipmi_timeout_handler(long ti
/* Go through the seq table and find any messages that
have timed out, putting them in the timeouts
list. */
+ INIT_LIST_HEAD(&timeouts);
spin_lock_irqsave(&intf->seq_lock, flags);
for (i = 0; i < IPMI_IPMB_NUM_SEQ; i++)
check_msg_timeout(intf, &(intf->seq_table[i]),
-
| Vladislav Bolkhovitin | Re: Integration of SCST in the mainstream Linux kernel |
| Eric Paris | [RFC 0/5] [TALPA] Intro to a linux interface for on access scanning |
| holzheu | Re: [RFC/PATCH] Documentation of kernel messages |
| debian developer | Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3 |
git: | |
| Jarek Poplawski | Re: [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock(). |
| Gerrit Renker | [PATCH 27/37] dccp: Integration of dynamic feature activation - part 2 (server side) |
| David Miller | [GIT]: Networking |
| Alan Cox | Re: [BUG] New Kernel Bugs |
