Re: [RFC][PATCH -rt] irqd starvation on SMP by a single process?

[Mark Hounschell <markh@compro.net>]

Steven Rostedt wrote:
> On Fri, 12 May 2006, Mark Hounschell wrote:
> 
>>> Mark, does this fix the problem?
>>>
>>> 	Ingo
>>>
>>> Index: linux-rt.q/drivers/net/3c59x.c
>>> ===================================================================
>>> --- linux-rt.q.orig/drivers/net/3c59x.c
>>> +++ linux-rt.q/drivers/net/3c59x.c
>>> @@ -1897,7 +1897,8 @@ vortex_timer(unsigned long data)
>>>
>>>  	if (vp->medialock)
>>>  		goto leave_media_alone;
>>> -	disable_irq(dev->irq);
>>> +	/* hack! */
>>> +	disable_irq_nosync(dev->irq);
>>>  	old_window = ioread16(ioaddr + EL3_CMD) >> 13;
>>>  	EL3WINDOW(4);
>>>  	media_status = ioread16(ioaddr + Wn4_Media);
>>>
>> Yes it does.
>>
> 
> 
> It fixes it for both "complete preemption" and "normal preemption"?
> 
> -- Steve
> 
> 

Nope, I still have the complete preemption problem. My box locked up
right away. Had to reboot it.

You guys know best but I think the problem the above patch fixes is
probably not related to my 'complete preemption' problem or the problem
reported by John. But my 'complete preemption' problem may certainly be
the same as Johns.

Mark