🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: comp.arch
Article #3338

Re: Integer Multiply/Divide on Sparc

#3338
From: khb Keith Bierman - SPD Advanced...
Date: Fri, 12 Jan 1990 05:23
59 lines
3107 bytes
In article <1249@otc.otca.oz> gregw@otc.otca.oz (Greg Wilkins) writes:


   in article <8840005@hpfcso.HP.COM>, dgr@hpfcso.HP.COM (Dave Roberts) says:
   > 	Now for some comments:
   > 	(1) SPARCs will get multiply and divide.  This is from a guy at
   > 	    Sun.  Coming soon to a SPARC station near you...

No, it was a guy at TI.

   OK Sun Microsystems!  we know you out there listening.  How about

Yes. Didn't you notice my posting BEFORE the fellow from TI's .... I
explained a perfectly ABI complaint way to use such instructions w/o
mandating ABI noncompliance (though one might want to go for full
speedup and not care about ABI compliance for some application).

>   interface) standard you guys are pushing.  Signing up all those
>   clone manufacturers etc etc.  So your going to spring a multiply
>   instruction on them and make it all out dated (when compared to

Ours ?? TI doesn't belong to Sun. Chip vendors sign up to build chips,
as long as they don't break specs (and there are compliant ways to add
instructions ... buy a license and learn how :>) who are we to stop
them ????

>   Even if it is not a new instruction but an integer co-processor, it
>   still needs to be included in the ABI.

ABI compliant codes can get use of additional instructions .... shared
libraries have their uses :>

>   Without an official yeh or nay on this one, nobody can have confidence in
>   Suns commitment to open systems and the ABI.

I'm no offical.

But I don't recall seeing the formal ABI publication .... there is a de
facto ABI (Solb works or haven't you noticed). Unless you have seen an
ABI spec which asserts that there is such and such an instruction,
you'd better not generate that instruction directly (if you care about
complying with the ABI). If you use the current Sun compilers, you get
.mul, .div, .rem etc. .... and if you have a chip which had those as
hw instructions you (as an end user, or software author or whathave
you) could replace those library codes with ones which employed your
nifty instruction(s). No ABI breakage.

You jump to some amazing conclusions from a posting from _TI_ about
Sun's plans and business practices.


--
Keith H. Bierman    |*My thoughts are my own. !! kbierman@sun.com
It's Not My Fault   |	MTS --Only my work belongs to Sun* 
I Voted for Bill &  | Advanced Languages/Floating Point Group            
Opus                | "When the going gets Weird .. the Weird turn PRO"

"There is NO defense against the attack of the KILLER MICROS!"
			Eugene Brooks
Technical Details
Message-ID: <KHB.90Jan11212327@chiba.kbierman@sun.com>
Path: arc1-ca.pugleaf.net!utzoo!attcan!uunet!samsung!zaphod.mps.ohio-state.edu!usc!ucsd!ames!amdcad!sun!sun-bb!khb
References: <8840005@hpfcso.HP.COM> <1249@otc.otca.oz>

Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!utzoo!attcan!uunet!samsung!zaphod.mps.ohio-state.edu!usc!ucsd!ames!amdcad!sun!sun-bb!khb
From: khb@chiba.kbierman@sun.com (Keith Bierman - SPD Advanced Languages)
Newsgroups: comp.arch
Subject: Re: Integer Multiply/Divide on Sparc
Message-ID: <KHB.90Jan11212327@chiba.kbierman@sun.com>
Date: 12 Jan 90 05:23:27 GMT
References: <8840005@hpfcso.HP.COM> <1249@otc.otca.oz>
Sender: news@sun.Eng.Sun.COM
Organization: Sun Microsystems
Lines: 59
In-reply-to: gregw@otc.otca.oz's message of 11 Jan 90 01:27:56 GMT