Re: stmw/lmw [Was: CVS commit: src/sys/arch/powerpc/powerpc]

[Matt Thomas <matt%3am-software.com@localhost>]

On Sunday, August 3, 2003, at 06:31 PM, Simon Burge wrote:

> Matt Thomas wrote:
>
> > Module Name:    src
> > Committed By:   matt
> > Date:           Mon Aug  4 00:32:50 UTC 2003
> >
> > Modified Files:
> >
> >         src/sys/arch/powerpc/powerpc: locore_subr.S trap_subr.S
> >
> > Log Message:
> >
> > Eliminate stmw/lmw substituting the individual load/store 
> > instructions.
>
> Does this have any measurable impact on trap response and context 
> switch
> times (simply because of more instruction fetches)?  It's also larger
> code, so we have greater I$ pollution...

Actually, when I was doing my internet checksum tests, I found 
individual loads to be *faster* (by 20-25%) than lmw/stmw.  This may be 
why gcc doesn't emit them for saving/restoring callframes.
--
Matt Thomas                     email: matt%3am-software.com@localhost
3am Software Foundry              www: 
http://3am-software.com/bio/matt/ Cupertino, CA              
disclaimer: I avow all knowledge of this message.