Re: Denial of denial of service

[prev] [thread] [next] [lurker] [Date index for 2007/01/29]

From: Yossi Kreinin
Subject: Re: Denial of denial of service
Date: 12:10 on 29 Jan 2007
Nicholas Clark wrote:
>>Now you're talking :)
>>
>>I was lying awake last night formulating a thesis that ARM26 is the  
>>absolute pinnacle of instruction set design.
> 
> 
> You're forgetting the 2**24 variants of the NOOP instruction.
> Although whether this is software is doubtful, and therefore the on-topicness
> here is doubtful too.

I can try to move this from hates-hardware to hates-software by pointing out 
that compilers have hard time dealing with ARM, which has only 16 registers (it 
prefers to spend the encoding on condition bits and shifter operands). For 
random C code, compilers produce much better code given a lot of registers since 
the stupidity of register spilling heuristics is crucial less frequently.

In fact I don't know if it's the compilers or the questionable choice to save 
registers and instead add optional bells and whistles to each and every instruction.

Of course any RISC architecture, even an "Advanced" RISC Machine, is way easier 
to deal with than any CISC one.
There's stuff above here

Generated at 23:01 on 06 Feb 2007 by mariachi 0.52