RE: [Linux-ia64] High fpu register corruption (PATCH)

Yes, it can be same as 2.5.
Thanks,
Asit


> -----Original Message-----
> From: David Mosberger [mailto:davidm@napali.hpl.hp.com] 
> Sent: Wednesday, May 28, 2003 9:41 PM
> To: Bjorn Helgaas
> Cc: davidm@hpl.hp.com; David Mosberger; Mallick, Asit K; 
> linux-ia64@linuxia64.org; Chris Mason; dstownse@us.ibm.com
> Subject: Re: [Linux-ia64] High fpu register corruption (PATCH)
> 
> 
> >>>>> On Wed, 28 May 2003 22:25:37 -0600, Bjorn Helgaas 
> <bjorn_helgaas@hp.com> said:
> 
>   Bjorn> OK, ignore my questions about ptrace.c and setup.c.  
> Questions still
>   Bjorn> open:
> 
>   Bjorn> arch/ia64/kernel/signal.c:
>   Bjorn> restore_sigcontext() changes look equivalent, but have
>   Bjorn> gratuitous differences from 2.5 (reversed sense of test,
>   Bjorn> comment)
> 
>   Bjorn> arch/ia64/kernel/traps.c:
>   Bjorn> Gratuitous whitespace differences.
> 
>   Bjorn> include/asm-ia64/processor.h:
>   Bjorn> ia64_is_local_fpu_owner(), ia64_set_local_fpu_owner():
>   Bjorn> these look functionally equivalent in 2.4 and 2.5.
>   Bjorn> Can they be made identical?
> 
>   Bjorn> include/asm-ia64/system.h:
>   Bjorn> IA64_HAS_EXTRA_STATE() and switch_to() appear to be
>   Bjorn> identical in 2.4 and 2.5 except for whitespace changes.
>   Bjorn> Can they be made identical?
> 
> Not sure if those are intentional.  I of course would prefer if 2.5
> were followed. ;-) Asit?
> 
> 	--daivd
>