OT Linux (RE: [rescue] OT: Stuffed Proliant?)

[Joshua D Boyd]

On Sun, Dec 23, 2001 at 03:55:41PM -0500, Dave McGuire wrote:
> On December 22, Joshua D Boyd wrote:
>> Besides, lack of rotate isn't all that bad.  Instead of rotating left by 12 
>> bits, you shift left by 12, shift right by 20, then sum the two shifts.  The
>> two shifts can be done simultaneously on a super-pipelined processor, so you
>> need 3 cycles to do a rotate instead of one. While this seems bad, I find it
>> hard to believe that this is the sole reason for distributd net to be slower
>> on SPARCs than intel boxes.  We loose 2 cylces on a rotate.  So what.  We
>> make it up by doing less loads.
> 
>   Well, that's what the distributed.net people say.  And the rotate
> instruction /no rotate instruction performance difference holds true
> for every processor for which there's a distributed.net client.
> 
>   That's all the info I need...if it's not enough for you, grab the
> sources and try it.

Other than x86, what other platforms have a hardware rotate?

-- 
Joshua D. Boyd