Re: KVM-AMD OOPS

From: Avi Kivity
Date: Mon Jul 09 2007 - 08:03:07 EST


Avi Kivity wrote:
Jeremy Fitzhardinge wrote:
Sasa Ostrouska wrote:
root@rcserver:~# modprobe kvm-amd
int3: 0000 [1] PREEMPT SMP
CPU 1
Modules linked in: kvm_amd snd_seq_dummy snd_seq_oss
snd_seq_midi_event snd_seq snd_seq_device snd_pcm_oss snd_mixer_oss
nls_iso8859_1 ntfs nls_base usb_storage libusual capability commoncap
lp psmouse snd_hda_intel snd_hda_codec snd_pcm snd_timer ohci_hcd
ehci_hcd 8139too rtc_cmos snd soundcore snd_page_alloc usbcore k8temp
mii rtc_core rtc_lib i2c_nforce2 parport_pc parport
Pid: 2898, comm: modprobe Tainted: P 2.6.21.5 #1
RIP: 0010:[<ffffffff80572364>] [<ffffffff80572364>]
register_cpu_notifier+0x1/0x31
RSP: 0000:ffff81006e34df40 EFLAGS: 00000246
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 00000000c0010117
RDX: 000000000000ffff RSI: ffff81006e219640 RDI: ffffffff80536510
RBP: ffffffff880d8840 R08: 0000000000000000 R09: 000000000006b5f4
R10: 0000000000000000 R11: 0000000000000000 R12: 00000000005296b0
R13: 00007fff51fe55c0 R14: 0000000000000000 R15: 0000000000000000
FS: 00002b4e58e18b00(0000) GS:ffff810002e794c0(0000)
knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 000000000050df64 CR3: 000000007a0f3000 CR4: 00000000000006e0
Process modprobe (pid: 2898, threadinfo ffff81006e34c000, task
ffff81007bc48400)
Stack: ffffffff8039e024 ffff81006e34c000 ffffffff880d8840
0000000000005e19
ffffffff8024537c 0000000000000000 00007fff51fe50c0 00000000004142d0
ffffffff8020967e 0000000000000206 00000000005230e0 000000000052f4c9
Call Trace:
[<ffffffff8039e024>] kvm_init_arch+0x90/0x145
[<ffffffff8024537c>] sys_init_module+0xad/0x168
[<ffffffff8020967e>] system_call+0x7e/0x83


Code: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc
This is the init section poison pattern. Looks like an init function
was used after the code was freed.

Indeed, the function is register_cpu_notifier which is declared
__cpuinit, which, without hotplug compiled in, is simply __init.

I'll prepare a patch for this. Meanwhile, you can work around it by
enabling CONFIG_HOTPLUG_CPU.


Well, this nice patch (in 2.6.21) ought to have fixed it:

commit 47e627bc8c9a70392d2049e6af5bd55fae61fe53
Author: Avi Kivity <avi@xxxxxxxxxxxx>
Date: Mon Feb 12 00:54:43 2007 -0800

[PATCH] hotplug: Allow modules to use the cpu hotplug notifiers even if !CONFIG_HOTPLUG_CPU
The following patchset allows a host with running virtual machines to be
suspended and, on at least a subset of the machines tested, resumed. Note
that this is orthogonal to suspending and resuming an individual guest to a
file.
A side effect of implementing suspend/resume is that cpu hotplug is now
supported. This should please the owners of big iron.
This patch:
KVM wants the cpu hotplug notifications, both for cpu hotplug itself, but more
commonly for host suspend/resume.
In order to avoid extensive #ifdefs, provide stubs when CONFIG_CPU_HOTPLUG is
not defined.
In all, we have four cases:
- UP: register and unregister stubbed out
- SMP+hotplug: full register and unregister
- SMP, no hotplug, core: register as __init, unregister stubbed
(cpus are brought up during core initialization)
- SMP, no hotplug, module: register and unregister stubbed out
(cpus cannot be brought up during module lifetime)
Signed-off-by: Avi Kivity <avi@xxxxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxx>
Cc: Rusty Russell <rusty@xxxxxxxxxxxxxxx>
Cc: Oleg Nesterov <oleg@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>

diff --git a/include/linux/cpu.h b/include/linux/cpu.h
index bfb5202..769ddc6 100644
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -49,10 +49,20 @@ struct notifier_block;

#ifdef CONFIG_SMP
/* Need to know about CPUs going up/down? */
-extern int register_cpu_notifier(struct notifier_block *nb);
#ifdef CONFIG_HOTPLUG_CPU
+extern int register_cpu_notifier(struct notifier_block *nb);
extern void unregister_cpu_notifier(struct notifier_block *nb);
#else
+
+#ifndef MODULE
+extern int register_cpu_notifier(struct notifier_block *nb);
+#else
+static inline int register_cpu_notifier(struct notifier_block *nb)
+{
+ return 0;
+}
+#endif
+
static inline void unregister_cpu_notifier(struct notifier_block *nb)
{
}


Please post your .config.


--
error compiling committee.c: too many arguments to function

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/