[PATCH -mm] PM: Remove sleeping from suspend_console

[PATCH -mm] PM: Remove sleeping from suspend_console

[Rafael J. Wysocki]

Remove ssleep() from suspend_console().

Stefan thinks it is unnecessary and will slow down the suspend too much.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
---
---
 kernel/printk.c |    5 -----
 1 file changed, 5 deletions(-)

Index: linux-2.6.18-rc5-mm1/kernel/printk.c
===================================================================
--- linux-2.6.18-rc5-mm1.orig/kernel/printk.c
+++ linux-2.6.18-rc5-mm1/kernel/printk.c
@@ -713,11 +713,6 @@ void suspend_console(void)
 	printk("Suspending console(s)\n");
 	acquire_console_sem();
 	console_suspended = 1;
-	/* This is needed so that all of the messages that have already been
-	 * written to all consoles can be actually transmitted (eg. over a
-	 * network) before we try to suspend the consoles' devices.
-	 */
-	ssleep(2);
 }
 
 void resume_console(void)

Re: [PATCH -mm] PM: Remove sleeping from suspend_console

[Stefan Seyfried]

On Mon, Sep 04, 2006 at 10:50:40PM +0200, Rafael J. Wysocki wrote:
> Remove ssleep() from suspend_console().
> 
> Stefan thinks it is unnecessary and will slow down the suspend too much.

"unnecessary" is not exactly what i think, rather "unacceptable" :-)
We probably need to do something for some kinds of consoles to make sure all
characters are sent, but sleeping unconditionally is not the right thing IMO.

> --- linux-2.6.18-rc5-mm1.orig/kernel/printk.c
> +++ linux-2.6.18-rc5-mm1/kernel/printk.c
> @@ -713,11 +713,6 @@ void suspend_console(void)
>  	printk("Suspending console(s)\n");
>  	acquire_console_sem();
>  	console_suspended = 1;
> -	/* This is needed so that all of the messages that have already been
> -	 * written to all consoles can be actually transmitted (eg. over a
> -	 * network) before we try to suspend the consoles' devices.
> -	 */
> -	ssleep(2);
>  }
>  
>  void resume_console(void)

-- 
Stefan Seyfried
QA / R&D Team Mobile Devices        |              "Any ideas, John?"
SUSE LINUX Products GmbH, Nürnberg  | "Well, surrounding them's out." 

Re: [PATCH -mm] PM: Remove sleeping from suspend_console

[Rafael J. Wysocki]

On Tuesday, 5 September 2006 08:28, Stefan Seyfried wrote:
> On Mon, Sep 04, 2006 at 10:50:40PM +0200, Rafael J. Wysocki wrote:
> > Remove ssleep() from suspend_console().
> > 
> > Stefan thinks it is unnecessary and will slow down the suspend too much.
> 
> "unnecessary" is not exactly what i think, rather "unacceptable" :-)

Still that implies it's not necessary. ;-)

> We probably need to do something for some kinds of consoles to make sure all
> characters are sent, but sleeping unconditionally is not the right thing IMO.

Yup.  This was added as a result of the Laurent Riffard's report that the sleep
was necessary, but it turned out to be due to the network card problem, so we
should get rid of the ssleep().

Greetings,
Rafael


-- 
You never change things by fighting the existing reality.
		R. Buckminster Fuller