Related articles |
---|
optimizing case-statement execution raymond@harp.ecn.purdue.edu (1992-11-22) |
Re: optimizing case-statement execution chased@rbbb.Eng.Sun.COM (1992-11-23) |
Re: optimizing case-statement execution erspert@athena.mit.edu (1992-11-25) |
Re: optimizing case-statement execution nr@volkl.Princeton.EDU (1992-11-25) |
Re: optimizing case-statement execution wtyler@adobe.com (1992-11-27) |
Re: optimizing case-statement execution pardo@cs.washington.edu (1992-12-04) |
Re: optimizing case-statement execution krste@ICSI.Berkeley.EDU (1992-12-05) |
Newsgroups: | comp.compilers |
From: | wtyler@adobe.com (William Tyler) |
Organization: | Adobe Systems Inc., Mountain View, CA |
Date: | Fri, 27 Nov 1992 04:17:44 GMT |
Followup-To: | comp.compilers |
Summary: | why not reentrant? |
References: | 92-11-126 92-11-145 |
Keywords: | C, code, optimize, comment |
erspert@athena.mit.edu (Ellen R. Spertus) writes:
>>Description of switch code using string scanning deleted<<
>... your code is not reentrant, and you write to the code segment.
>Why do I even mention it? Because self-modifying code is fun!
Is it really necessary to write to the code segment? Why not keep the
values to be scanned in the data segment, solving both the problems
you've mentioned?
Bill
--
Bill Tyler wtyler@adobe.com
[Indeed, you would normally put the data in the data segment. But this
does write to the end of an otherwise read-only table. On the '86 the
scan has a length count which avoids the need to stash the value at the
end of the table; this hack was really for the PDP-11. -John]
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.