[patch][5/5] powerpc: Add the general support for Embedded Floating-Point instructions

[Zhu Ebony-r57400]

> -----Original Message-----
> From: Kumar Gala [mailto:galak at kernel.crashing.org] 
> Sent: 2007年1月13日 05:49
> To: Benjamin Herrenschmidt
> Cc: Zhu Ebony-r57400; linuxppc-dev at ozlabs.org; paulus at samba.org
> Subject: Re: [patch][5/5] powerpc: Add the general support 
> for Embedded Floating-Point instructions
> 
> 
> On Jan 12, 2007, at 3:27 PM, Benjamin Herrenschmidt wrote:
> 
> >> I think its a fair assumption that nothing is going to 
> interrupt the 
> >> handling of the exception, so its safe to assume the 
> process causing 
> >> the interrupt will be the same one we return to.
> >
> > What about the get_user to get to the faulting instruction 
> ? I suppose 
> > if those processors are UP only and we use an exception with EE 
> > disabled, there should be no way the page has been evicted 
> since the 
> > access, so that should work, but will that ever be true ?
> 
> For UP we are ok, since nothing else can cause an invalidate.
> 
> How do we handle getting a fault when we are emulating any 
> other instruction?
> 
> - k
> 

Do you think using SRR0 is safer since it contains the exact effective address
of the instruction causing the interrupt?

B.R.
Ebony