2.6.22 - Mission Impossible :-) -

2.6.22 - Mission Impossible :-) -

[Alessio Igor Bogani]

Hi,

I continue with my effort to port last realtime-preempt patch available to
2.6.22.

If i compile without ACPI/hrtimers i don't have any problem (performance
and stability not tested yet).

If i compile with these i have a big problem:

---[ Cutted form kernel/timer.c ]---

void msleep(unsigned int msecs)
{
	unsigned long timeout = msecs_to_jiffies(msecs) + 1;
	while (timeout)
		timeout = schedule_timeout_uninterruptible(timeout);
}

------------------------------------

The while construct never return!

I think that Hrtimer subsystem isn't initialized when msleep called from
do_probe() (ide subsystem).

Someone have suggestions?

Thanks in advance!

Ciao,
Alessio

Re: 2.6.22 - Mission Impossible :-) -

[Daniel Walker]

Why do you need to do this?

On Wed, 2007-05-30 at 01:34 +0200, Alessio Igor Bogani wrote:
> Hi,
> 
> I continue with my effort to port last realtime-preempt patch available to
> 2.6.22.
> 
> If i compile without ACPI/hrtimers i don't have any problem (performance
> and stability not tested yet).
> 
> If i compile with these i have a big problem:
> 
> ---[ Cutted form kernel/timer.c ]---
> 
> void msleep(unsigned int msecs)
> {
> 	unsigned long timeout = msecs_to_jiffies(msecs) + 1;
> 	while (timeout)
> 		timeout = schedule_timeout_uninterruptible(timeout);
> }
> 
> ------------------------------------
> 
> The while construct never return!
> 
> I think that Hrtimer subsystem isn't initialized when msleep called from
> do_probe() (ide subsystem).
> 
> Someone have suggestions?
> 
> Thanks in advance!
> 
> Ciao,
> Alessio
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: 2.6.22 - Mission Impossible :-) -

[Alessio Igor Bogani]

Hi Daniel,

> Why do you need to do this?

For two reasons:

1) For study realtime patch
2) For include it into next Ubuntu release

Ciao,
Alessio

Re: 2.6.22 - Mission Impossible :-) -

[K.R. Foley]

Alessio Igor Bogani wrote:
> Hi Daniel,
> 
>> Why do you need to do this?
> 
> For two reasons:
> 
> 1) For study realtime patch
> 2) For include it into next Ubuntu release
> 
> Ciao,
> Alessio
> 

The study part I can kind of understand, but for both reasons wouldn't
it make more sense to wait until 2.6.22 settles out?

-- 
   kr

Re: 2.6.22 - Mission Impossible :-) -

[Alessio Igor Bogani]

Hi K.R. Foley,

Excuse me for my poor english.

> The study part I can kind of understand, but for both reasons wouldn't
> it make more sense to wait until 2.6.22 settles out?

For study part:
When the things are already done it _seems_ more easy than are in
true. :-) 

For Ubuntu part:
My objective is include a realtime kernel (install & play) into the next
Ubuntu Linux distribution (will be release in October). This permit to
have it in more widespread use for testing (and finally this is very
main reason).

Ciao,
Alessio