md5sum makes unaligned word accesses...

Michael S. Zick mszick at morethan.org
Mon May 15 00:06:01 UTC 2006


On Sun May 14 2006 19:00, Rob Landley wrote:
> On Friday 12 May 2006 6:21 pm, Mike Frysinger wrote:
> > On Friday 12 May 2006 16:57, Rob Landley wrote:
> > > On Friday 12 May 2006 12:43 am, Mike Frysinger wrote:
> > > > On Thursday 11 May 2006 12:22, Rob Landley wrote:
> > > > > But it doesn't on any other platform.
> > > >
> > > > alignment issues arent arm-specific
> > >
> > > Yeah, but do we support another platform yet that cares?
> > >
> > > I'm all for fixing it.  I just don't want to penalize systems that don't
> > > need the fix.
> >
> > of course we do
> >
> > more systems care about alignment issues than not
> >
> > it's just that most systems will automatically clean up alignment bugs at
> > the kernel level so no one usually notices
> >
> > so, we're already penalizing systems, we just dont notice ;)
> 
> I'm talking about binary size, not performance.  And the big bottleneck is the 
> cache line fetches, not the internal processing; try actually benching stuff 
> and the alignment is likely to be noise at best.
>
Say what? Read:

linux/arch/parisc/kernel/unaligned.c which is called from a trap handler.

Are you saying that process is faster than a data cache line fetch?

Mike
> Rob



More information about the busybox mailing list