Hi,
Maran reported the bug below (vanilla 2.6.30-rc8):
BUG: sleeping function called from invalid context at /mnt/s390test/linux-2.6-tip/arch/s390/include/asm/uaccess.h:234
in_atomic(): 1, irqs_disabled(): 0, pid: 3245, name: sysctl
CPU: 1 Not tainted 2.6.30-rc8-tipjun10-02053-g39ae214 #1
Process sysctl (pid: 3245, task: 000000007f675da0, ksp: 000000007eb17cf0)
0000000000000000 000000007eb17be8 0000000000000002 0000000000000000
000000007eb17c88 000000007eb17c00 000000007eb17c00 0000000000048156
00000000003e2de8 000000007f676118 000000007eb17f10 0000000000000000
0000000000000000 000000007eb17be8 000000000000000d 000000007eb17c58
00000000003e2050 000000000001635c 000000007eb17be8 000000007eb17c30
Call Trace:
(?<00000000000162e6>? show_trace+0x13a/0x148)
?<00000000000349ea>? __might_sleep+0x13a/0x164
?<0000000000050300>? proc_dostring+0x134/0x22c
?<0000000000312b70>? nf_log_proc_dostring+0xfc/0x188
?<0000000000136f5e>? proc_sys_call_handler+0xf6/0x118
?<0000000000136fda>? proc_sys_read+0x26/0x34
?<00000000000d6e9c>? vfs_read+0xac/0x158
?<00000000000d703e>? SyS_read+0x56/0x88
?<0000000000027f42>? sysc_noemu+0x10/0x16
The code that introduces the bug came in with 17625274 "netfilter:
sysctl support of logger choice".
There we have this chunk:
+ rcu_read_lock();
+ logger = rcu_dereference(nf_loggers[tindex]);
+ if (!logger)
+ table->data = "NONE";
+ else
+ table->data = logger->name;
+ r = proc_dostring(table, write, filp, buffer, lenp, ppos);
+ rcu_read_unlock();
proc_dostring() will call copy_from_user() while preemption is disabled
because of rcu_read_lock().
Looks like somebody needs to fix this ;)
diff --git a/net/netfilter/nf_log.c b/net/netfilter/nf_log.c
index beb3731..2fefe14 100644
--- a/net/netfilter/nf_log.c
+++ b/net/netfilter/nf_log.c
@@ -248,14 +248,14 @@ static int nf_log_proc_dostring(ctl_table *table, int write, struct file *filp,
rcu_assign_pointer(nf_loggers[tindex], logger);
mutex_unlock(&nf_log_mutex);
} else {
- rcu_read_lock();
- logger = rcu_dereference(nf_loggers[tindex]);
+ mutex_lock(&nf_log_mutex);
+ logger = nf_loggers[tindex];
if (!logger)
table->data = "NONE";
else
table->data = logger->name;
r = proc_dostring(table, write, filp, buffer, lenp, ppos);
- rcu_read_unlock();
+ mutex_unlock(&nf_log_mutex);
}
return r;
Maran Pakkirisamy wrote:
> I have tested the patch against Vanilla Linux 2.6.30 RC8. The issue with
> netfilter is fixed with the given patch . Thanks a lot!
Thanks, I'll send it upstream today.