delorie.com/archives/browse.cgi   search  
Mail Archives: djgpp/2000/01/30/02:11:57

From: dennis51 AT my-deja DOT com
Newsgroups: comp.os.msdos.djgpp
Subject: Re: 203 much slower than 202
Date: Sun, 30 Jan 2000 04:45:42 GMT
Organization: Deja.com - Before you buy.
Lines: 54
Message-ID: <870fll$3ot$1@nnrp1.deja.com>
References: <38923107 DOT F4C7C15C AT jps DOT net> <3892AC56 DOT C9577F9D AT is DOT elta DOT co DOT il>
NNTP-Posting-Host: 209.63.224.190
X-Article-Creation-Date: Sun Jan 30 04:45:42 2000 GMT
X-Http-User-Agent: Mozilla/4.7 [en] (Win95; U)
X-Http-Proxy: 1.1 smfc760-1 (NetCache 4.0R3), 1.0 x33.deja.com:80 (Squid/1.1.22) for client 216.119.53.125, 209.63.224.190
X-MyDeja-Info: XMYDJUIDdennis51
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

In article <3892AC56 DOT C9577F9D AT is DOT elta DOT co DOT il>,
  Eli Zaretskii <eliz AT is DOT elta DOT co DOT il> wrote:
> Dennis Yelle wrote:
> >
> > I just installed DJGPP version 203.
> > It seems to be much slower than 202.
> > I have a program that compiles and links in about 23 seconds
> > with 202, but it takes about 35 seconds for 203.
>
> Did you install something else beyond djdev203.zip?  Did you maybe
installed
> a new version of the compiler as well?

Well, of course I installed the whole package that
the zip picker recommended.  I am not enough of an
expert on gcc to be brave enough to just install part
of it over an old version.  In particular I got
gcc2952b.zip as well.

> Because the library cannot
possibly
> affect the linking speed (unless it grows too much, which it didn't in
this
> case), and it cannot affect the compilation speed at all.
>
> But if you installed another version of the compiler, it could well
happen,
> especially with C++ programs.  GCC 2.9x is slower in compilation than
> previous versions, because it does much more optimizations.

Yes, the slow part is CC1PLUS.
And, no, the program does not run any faster.
It is about 0.5% larger and about 1% slower.

> In any case, 23 vs 32 seconds is only 30%; it doesn't seem like "much
slower"
> is the right description of the ratio.

Well, if one of my programs suddenly went from 23 seconds
to 32 seconds you can be sure I would investigate it, and fix it.

>
> > Has anyone else noticed this?
>
> I didn't.
>
> > Is this expected?
>
> No.
>


Sent via Deja.com http://www.deja.com/
Before you buy.

- Raw text -


  webmaster     delorie software   privacy  
  Copyright © 2019   by DJ Delorie     Updated Jul 2019