Article View: comp.arch
Article #10850Re: how to advocate new software/hardware features (Re: Hardware Support for Numeric Algorithms)
From: Mark C. Carroll
Date: Sun, 15 Nov 1992 19:38
Date: Sun, 15 Nov 1992 19:38
58 lines
3233 bytes
3233 bytes
In article <Bxr8vG.IpI@mentor.cc.purdue.edu> hrubin@mentor.cc.purdue.edu (Herman Rubin) writes: >In article <mwm.2n6z@contessa.palo-alto.ca.us> mwm@contessa.palo-alto.ca.us (Mike Meyer) writes: ]]In <Bxq7y0.IJ@mentor.cc.purdue.edu>, hrubin@mentor.cc.purdue.edu (Herman Rubin) wrote: ]]] With instruction scheduling, one place for an optimizer would be in the ]]] ASSEMBLER, and even interaction between the programmer and the compiler/ ]]] assembler. I do not know of any such. ] ]]Herman, I've *never* seen you suggest something that hasn't been tried ]]already, in some form or another. This is no exception. It's been ]]tried. It didn't work very well. ] ]Trying something on a stupidly restricted basis cannot work well. It seems ]like you are saying that the analogs of the "enrichment" programs that the ]educationists have come up with for the bright children, rather than allowing ]them to advance, do not accomplish much. ] ]Nobody on this net, or others I have discussed this with, know of the ]existence of a versatile syntax converter (macro processor). Bob Silverman ]and Peter Montgomery have considerable contacts here, and they know of no ]such thing. There are many macro processors out which at least handle some ]problems with macro name first, but apart from compilers, I know of none ]which would allow someone to write (Bob Silverman's notation) ] ] c = a ** b ] ]and have it translated into ] ] c = doubmult(a,b). ] Herman, you're a liar. On at least three occasions, I have answered this ridiculous thread of yours with a recommendation that you take a look at TXL. TXL will do exactly what you want from a macro processor. TXL is a tool which takes a language, transforms it into an abstract syntax tree, and then allows you to perform arbitrary transformations on the tree, and output the results of the transformation. Every single syntax that you have ever posted to the net could be parsed by TXL and transformed into the conventional syntax of your language. Archie lists dozens of sites that have TXL available. In addition, someone recently announced a macro transformer named m5. From the description I've read, m5 will also do the transformations that you've wanted. <MC> -- || Mark Craig Carroll: <MC> ||"I prize the cloudy, tearing sky, || Univ of Delaware, Dept of CIS|| for the thoughts that flap and fly. || Grad Student/Labstaff Hacker || For staying in and reading by. || carroll@udel.edu || For sitting under" -Karen Peris
Technical Details
Message-ID:
<1992Nov15.193847.8390@udel.edu>Path:
arc1-ca.pugleaf.net!sparky!uunet!know!cass.ma02.bull.com!think.com!rpi!usc!zaphod.mps.ohio-state.edu!wupost!udel!pervert!louie!balin.cis.udel.edu!carrollReferences:
<Bxq7y0.IJ@mentor.cc.purdue.edu> <mwm.2n6z@contessa.palo-alto.ca.us> <Bxr8vG.IpI@mentor.cc.purdue.edu>Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!know!cass.ma02.bull.com!think.com!rpi!usc!zaphod.mps.ohio-state.edu!wupost!udel!pervert!louie!balin.cis.udel.edu!carroll From: carroll@balin.cis.udel.edu (Mark C. Carroll) Newsgroups: comp.arch,comp.lang.misc Subject: Re: how to advocate new software/hardware features (Re: Hardware Support for Numeric Algorithms) Message-ID: <1992Nov15.193847.8390@udel.edu> Date: 15 Nov 92 19:38:47 GMT References: <Bxq7y0.IJ@mentor.cc.purdue.edu> <mwm.2n6z@contessa.palo-alto.ca.us> <Bxr8vG.IpI@mentor.cc.purdue.edu> Sender: usenet@udel.edu (USENET News Service) Organization: University of Delaware, Newark Lines: 58 Nntp-Posting-Host: balin.cis.udel.edu