From: Skye <s_c AT iname DOT com>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: what is __djgpp_exception_processor??
Date: Wed, 16 Apr 1997 14:19:10 -0700
Organization: Uniserve
Lines: 28
Message-ID: <3355424E.4076@iname.com>
References: <Pine DOT A41 DOT 3 DOT 96 DOT 970412191602 DOT 124522A-100000 AT asterix DOT uni-muenster DOT de> <5iub8s$28n4 AT elmo DOT cadvision DOT com>
Reply-To: s_c AT iname DOT com
NNTP-Posting-Host: dy2-34.van.tvs.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Precedence: bulk

Calvin French wrote:
> 
> "Guido Bursch" <bursch AT uni-muenster DOT de> wrote:
> 
> >gprof reported me that 72% (22% self) of the runtime were wasted by
> >the function (or macro or whatever) "__djgpp_exception_processor"
...
> >To my program is to say that FP- operations are heavyly used and I
> >have to use an FP- emulator because I dont't own an 387 FPU.

> On the odd chance that noone answers this who can give you a definite
> answer, I would say it's the FPU emulation that is killing you. I have

I can't say for sure but that is my geus too. Trying to access the FPU
in PMode when there isn't one gerates an exeption (if the emulation bit
in the CR0 register is set it generates an INT 7 I think) that the DPMI
server can catch and provide FPU emulation. 

The two fixes I can sugest based on this geuss is: 1)get a better mboard
and processor, a basic P100 is cheep now, and 2) use fixed point math.
You'll want to make your lib fixed point some time anyway for speed
reasons.

> Oh, and let me also make a small suggestion if I may; Allegro.
> Excellent fixed point routines, not to mention polygon stuff.
I'll second that.

-SC