BUG() during suspend to disk (2.6.21-rc2, x86_64)

BUG() during suspend to disk (2.6.21-rc2, x86_64)

[Vivek Goyal]

Hi,

I see following BUG() on serial console while hibernating on a x86_64
machine. I am using 2.6.21-rc2 kernel.


BUG: at arch/x86_64/kernel/acpi/sleep.c:70 init_low_mapping()

Call Trace:
 [<ffffffff80214b9a>] acpi_save_state_mem+0x70/0xd6
 [<ffffffff8036e90e>] acpi_pm_enter+0x23/0xc1
 [<ffffffff8024eec4>] pm_suspend_disk+0x1ac/0x228
 [<ffffffff8024dc9b>] enter_state+0x50/0x1e6
 [<ffffffff8036ecce>] acpi_system_write_sleep+0x5c/0x79
 [<ffffffff8027b767>] vfs_write+0xad/0x136
 [<ffffffff8027bca4>] sys_write+0x45/0x6e
 [<ffffffff80209bbe>] system_call+0x7e/0x83

Thanks
Vivek

Re: BUG() during suspend to disk (2.6.21-rc2, x86_64)

[Rafael J. Wysocki]

Hi,

On Tuesday, 6 March 2007 11:32, Vivek Goyal wrote:
> Hi,
> 
> I see following BUG() on serial console while hibernating on a x86_64
> machine. I am using 2.6.21-rc2 kernel.

I see it too.
 
> BUG: at arch/x86_64/kernel/acpi/sleep.c:70 init_low_mapping()
> 
> Call Trace:
>  [<ffffffff80214b9a>] acpi_save_state_mem+0x70/0xd6
>  [<ffffffff8036e90e>] acpi_pm_enter+0x23/0xc1
>  [<ffffffff8024eec4>] pm_suspend_disk+0x1ac/0x228
>  [<ffffffff8024dc9b>] enter_state+0x50/0x1e6
>  [<ffffffff8036ecce>] acpi_system_write_sleep+0x5c/0x79
>  [<ffffffff8027b767>] vfs_write+0xad/0x136
>  [<ffffffff8027bca4>] sys_write+0x45/0x6e
>  [<ffffffff80209bbe>] system_call+0x7e/0x83

Hm, it doesn't like the fact that nonboot CPUs are online at that point, but
we don't do anything to disable them.  Should we?

I think the appended patch might work.

Rafael

---
 kernel/power/disk.c |    1 +
 kernel/power/user.c |    1 +
 2 files changed, 2 insertions(+)

Index: linux-2.6.21-rc2/kernel/power/disk.c
===================================================================
--- linux-2.6.21-rc2.orig/kernel/power/disk.c
+++ linux-2.6.21-rc2/kernel/power/disk.c
@@ -61,6 +61,7 @@ static void power_down(suspend_disk_meth
 	switch(mode) {
 	case PM_DISK_PLATFORM:
 		if (pm_ops && pm_ops->enter) {
+			disable_nonboot_cpus();
 			kernel_shutdown_prepare(SYSTEM_SUSPEND_DISK);
 			pm_ops->enter(PM_SUSPEND_DISK);
 			break;
Index: linux-2.6.21-rc2/kernel/power/user.c
===================================================================
--- linux-2.6.21-rc2.orig/kernel/power/user.c
+++ linux-2.6.21-rc2/kernel/power/user.c
@@ -398,6 +398,7 @@ static int snapshot_ioctl(struct inode *
 
 		case PMOPS_ENTER:
 			if (data->platform_suspend) {
+				disable_nonboot_cpus();
 				kernel_shutdown_prepare(SYSTEM_SUSPEND_DISK);
 				error = pm_ops->enter(PM_SUSPEND_DISK);
 				error = 0;

Re: BUG() during suspend to disk (2.6.21-rc2, x86_64)

[Vivek Goyal]

On Tue, Mar 06, 2007 at 09:56:46PM +0100, Rafael J. Wysocki wrote:
> Hi,
> 
> On Tuesday, 6 March 2007 11:32, Vivek Goyal wrote:
> > Hi,
> > 
> > I see following BUG() on serial console while hibernating on a x86_64
> > machine. I am using 2.6.21-rc2 kernel.
> 
> I see it too.
> 
> > BUG: at arch/x86_64/kernel/acpi/sleep.c:70 init_low_mapping()
> > 
> > Call Trace:
> >  [<ffffffff80214b9a>] acpi_save_state_mem+0x70/0xd6
> >  [<ffffffff8036e90e>] acpi_pm_enter+0x23/0xc1
> >  [<ffffffff8024eec4>] pm_suspend_disk+0x1ac/0x228
> >  [<ffffffff8024dc9b>] enter_state+0x50/0x1e6
> >  [<ffffffff8036ecce>] acpi_system_write_sleep+0x5c/0x79
> >  [<ffffffff8027b767>] vfs_write+0xad/0x136
> >  [<ffffffff8027bca4>] sys_write+0x45/0x6e
> >  [<ffffffff80209bbe>] system_call+0x7e/0x83
> 
> Hm, it doesn't like the fact that nonboot CPUs are online at that point, but
> we don't do anything to disable them.  Should we?
> 
> I think the appended patch might work.
> 

Yes. It does work for me. Thanks

Vivek