To:	Kimberley Burchett <OKRA AT max DOT tiac DOT net>
Cc:	DJGPP Mailing List <djgpp AT sun DOT soe DOT clarkson DOT edu>
Subject:	Re: Speed tuning programs
Date:	Thu, 18 Aug 1994 10:03:10 +0100
From:	Olly Betts <olly AT mantis DOT co DOT uk>

In message <Pine DOT 3 DOT 89 DOT 9408171700 DOT B2426-0100000 AT max DOT tiac DOT net>, Kimberley Burchet
t writes:
>On Wed, 17 Aug 1994, Olly Betts wrote:
>
>> Now from my reading of this, the profiler thinks that the program took
>> 0.06 seconds, all spent in main(), which is just plain wrong.  I'm sure
>> I must just be failing to do something.
>
>> If only I could get some timings from gprof ...
>
>  I had this exact same problem and it led me to write my own profiler 
>which although it was a lot simpler, still worked for me.  Now, however, 
>gprof works for me and the only difference I can think of is that before 
>I hadn't called the setdjgpp program like it said to in the readme file.

Because my PATH is in danger of exploding, and to make sure that I get
the correct make.exe, etc (I have four C compilers installed here) I run
this batch file before doing DJGPP stuff.

PATH c:\djgpp\bin;C:\DOS;C:\QEMM;C:\4dos5.0;C:\WINDOWS;c:\utils
call c:\djgpp\bin\setdjgpp c:/djgpp c:\djgpp

Just to check, I re-ran the program with profiling on, and gprof, with
well nigh identical results.

>  Maybe that's your problem?  I'm totally guessing in the dark here...

If no one can work out why gprof isn't working for me, and your profiler
can give me the time spent in each routine, would you mind letting me
have a copy?  ftp upload to mantis.co.uk, MIME or uuencoded mail -
whatever's easier for you.

Cheers,
Olly

- Raw text -