Re: 2.6.10-mm2: swsusp regression [update]

[Rafael J. Wysocki]

Hi,

On Wednesday, 12 of January 2005 23:46, Pavel Machek wrote:
> Hi!
> 
> > > > (for example - the second number is always negative and huge).  Would 
it 
> > mean 
> > > > that get_cmos_time() needs fixing?
> > > 
> > > get_cmos_time() looks okay, but timer){suspend,resume} looks
> > > hopelessly broken.
> > 
> > Well, why don't we convert them to noops, then, at least temporarily?
> 
> Actually, it was my analysis that was wrong. Did you try Nigel's trick
> with updating wall_jiffies?

Do you mean to add

wall_jiffies +=  (ctime - sleep_start) * HZ;

or an equivalent at the end of timer_resume()?  I did and it helped a little.  
With it, the box hangs while executing device_resume() in swsusp_write().  
Without it, the box hangs earlier.

Still, it's sleep_start that has a wrong value, apparently (it shouldn't be 
negative, at least), and I see that Nigel has a patch that changes 
__get_cmos_time() on x86_64.  I'm going to try it in a couple of minutes.

Greets,
RJW

-- 
- Would you tell me, please, which way I ought to go from here?
- That depends a good deal on where you want to get to.
		-- Lewis Carroll "Alice's Adventures in Wonderland"
-
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/