[PATCH -mm] swsusp/pm: refuse to suspend devices if wrong console is active

[PATCH -mm] swsusp/pm: refuse to suspend devices if wrong console is active

[Rafael J. Wysocki]

1) Remove the console-switching code from the suspend part of the swsusp
userland interface and let the userland tools switch the console.

2) It is unsafe to suspend devices if the hardware is controlled by X.  Add
an extra check to prevent this from happening.


Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@suse.cz>
---
 drivers/base/power/suspend.c |    9 +++++++++
 kernel/power/user.c          |    3 ---
 2 files changed, 9 insertions(+), 3 deletions(-)

Index: linux-2.6.16-rc3-mm1/kernel/power/user.c
===================================================================
--- linux-2.6.16-rc3-mm1.orig/kernel/power/user.c
+++ linux-2.6.16-rc3-mm1/kernel/power/user.c
@@ -138,12 +138,10 @@ static int snapshot_ioctl(struct inode *
 		if (data->frozen)
 			break;
 		down(&pm_sem);
-		pm_prepare_console();
 		disable_nonboot_cpus();
 		if (freeze_processes()) {
 			thaw_processes();
 			enable_nonboot_cpus();
-			pm_restore_console();
 			error = -EBUSY;
 		}
 		up(&pm_sem);
@@ -157,7 +155,6 @@ static int snapshot_ioctl(struct inode *
 		down(&pm_sem);
 		thaw_processes();
 		enable_nonboot_cpus();
-		pm_restore_console();
 		up(&pm_sem);
 		data->frozen = 0;
 		break;
Index: linux-2.6.16-rc3-mm1/drivers/base/power/suspend.c
===================================================================
--- linux-2.6.16-rc3-mm1.orig/drivers/base/power/suspend.c
+++ linux-2.6.16-rc3-mm1/drivers/base/power/suspend.c
@@ -8,6 +8,9 @@
  *
  */
 
+#include <linux/vt_kern.h>
+#include <linux/kbd_kern.h>
+#include <linux/console.h>
 #include <linux/device.h>
 #include "../base.h"
 #include "power.h"
@@ -82,6 +85,12 @@ int device_suspend(pm_message_t state)
 {
 	int error = 0;
 
+	/* It is unsafe to suspend devices while X has control of the
+	 * hardware. Make sure we are running on a kernel-controlled console.
+	 */
+	if (vc_cons[fg_console].d->vc_mode != KD_TEXT)
+		return -EINVAL;
+
 	down(&dpm_sem);
 	down(&dpm_list_sem);
 	while (!list_empty(&dpm_active) && error == 0) {

Re: [PATCH -mm] swsusp/pm: refuse to suspend devices if wrong console is active

[Andrew Morton]

"Rafael J. Wysocki" <rjw@sisk.pl> wrote:
>
> 1) Remove the console-switching code from the suspend part of the swsusp
> userland interface and let the userland tools switch the console.
> 
> 2) It is unsafe to suspend devices if the hardware is controlled by X.  Add
> an extra check to prevent this from happening.
> 
> ...
> @@ -82,6 +85,12 @@ int device_suspend(pm_message_t state)
>  {
>  	int error = 0;
>  
> +	/* It is unsafe to suspend devices while X has control of the
> +	 * hardware. Make sure we are running on a kernel-controlled console.
> +	 */
> +	if (vc_cons[fg_console].d->vc_mode != KD_TEXT)
> +		return -EINVAL;
> +
>  	down(&dpm_sem);

Does this mean that swsusp simply won't work if invoked while X is running
and being displayed?  If so, that sounds like a pretty severe limitation.

Re: [PATCH -mm] swsusp/pm: refuse to suspend devices if wrong console is active

[Pavel Machek]

On Ne 19-02-06 21:35:26, Andrew Morton wrote:
> "Rafael J. Wysocki" <rjw@sisk.pl> wrote:
> >
> > 1) Remove the console-switching code from the suspend part of the swsusp
> > userland interface and let the userland tools switch the console.
> > 
> > 2) It is unsafe to suspend devices if the hardware is controlled by X.  Add
> > an extra check to prevent this from happening.
> > 
> > ...
> > @@ -82,6 +85,12 @@ int device_suspend(pm_message_t state)
> >  {
> >  	int error = 0;
> >  
> > +	/* It is unsafe to suspend devices while X has control of the
> > +	 * hardware. Make sure we are running on a kernel-controlled console.
> > +	 */
> > +	if (vc_cons[fg_console].d->vc_mode != KD_TEXT)
> > +		return -EINVAL;
> > +
> >  	down(&dpm_sem);
> 
> Does this mean that swsusp simply won't work if invoked while X is running
> and being displayed?  If so, that sounds like a pretty severe limitation.

No, no. Kernel with switch to text console before doing this in swsusp
case, and userspace has to switch to text console before doing this in
uswsusp case.

This is just a sanity check, to guard against misuse by uswsusp
program.

								Pavel
-- 
Web maintainer for suspend.sf.net (www.sf.net/projects/suspend) wanted...