RE: [PATCH] idle using PNI monitor/mwait

[Mallick, Asit K]

Alan,
Mwait is not dependent directly on the processor and any bus master
write will wake up the mwait. So, your example will also work.
Thanks,
Asit

> -----Original Message-----
> From: Alan Cox [mailto:[e-mail]] 
> Sent: Wednesday, July 09, 2003 4:00 AM
> To: Nakajima, Jun
> Cc: Linus Torvalds; Linux Kernel Mailing List; Saxena, Sunil; 
> Mallick, Asit K; Pallipadi, Venkatesh
> Subject: Re: [PATCH] idle using PNI monitor/mwait
> 
> 
> On Maw, 2003-07-08 at 22:23, Nakajima, Jun wrote:
> > Hi Linus,
> > 
> > Attached is a patch that enables PNI (Prescott New Instructions)
> > monitor/mwait in kernel idle (opcodes are now public). 
> Basically MWAIT
> > is similar to hlt, but you can avoid IPI to wake up the processor
> > waiting. A write (by another processor) to the address 
> range specified
> > by MONITOR would wake up the processor waiting on MWAIT.
> 
> Is mwait dependant on cached cpu memory and the cache 
> exclusivity logic
> or directly on the processor. In other words can I use mwait in future
> to wait for DMA to hit a given location ? - Im mostly thinking about
> debugging uses 
> 
>