From:	ckgp!thomas AT uunet DOT UU DOT NET
Subject:	Profiling error (BUG?)
To:	djgpp AT sun DOT soe DOT clarkson DOT edu
Date:	Wed, 20 May 92 10:05:52 EDT
Status:	O

Hi,

    Is anyone else still having problems with profiling? Last night I tried
to isolate my problem....(I moved and switched everything around until!)
finally at about 1:30am I re-installed gcc 1.39 (or the 1.05 package) everything
except go32.exe which I thought should work...well it turns out that I
still had the exception 14, error code=0 with the 1.05 stuff (which I didn't
have before?) So I got the old go32.exe and ran a program compiled with 1.06
(and gcc2.00) which would get the error and that program worked correctly...
then I got the alpha-2 release of go32.exe and ran the program and it worked
also...so this is what I think...

    I think that something happened to the go32.exe between the very first
release of djdev106.zip and the 4-3-92 release of djdev106.zip? correct me
if I am wrong (I am not talking about the beta release) but wasn't there a
small change made in between march and april of the first release? I am guessing
that most people have the first one and didn't switch to the second so they
are not getting the error (could this be?)

   DJ, what could be causing this to occur in go32.exe? I am most likely
going to re-compile go32.exe, perhaps there is a missing bit or two caused
by transmitting the file?      

   go32.exe    47550 4-13-92 9:07pm does not work (for me when profiling)
   debug32.exe 89682 4-13-92 9:07pm does not work (for me when profiling)

   go32.exe (alpha-2) 43718 3-6-92 11:16pm works great (for me when profiling)
   [...] everything older that this works great when profiling?

   If anyone out there still has a problem profiling, please try and older 
copy of go32.exe and let me know if it works for you also?

   If anyone knows what the problem is in go32.exe when profiling, please
let me know so I can fix mine... if not I will keep searching... What aspect
of go32.exe could just have a problem when profiling?

   Something else strange happened when trying to find the bug: I was doing
a listing of the asm code in debug32.exe, I had compiled with both -g & -pg
(is this okay?) anyway, after about 3 or 4 l commands to continue the listing
it slowed down hit the drive ALOT then said "out of memory" intrestingly enough
I had 7.5 meg free after getting into debug32.exe and 27 meg free on the hard
drive? there is little chance that I was out of memory? any ideas? or was it
the -g and -pg together?
      
 thanks for your time...

================================================================================
Thank you,            ||  "Sol est invisiblis in hominibus, in terra vero
Michael Thomas        ||   visibilis, tamen ex uno et eodem sole sunt ambo"
(..uunet!ckgp!thomas) ||                    -- Theatrum Chemicum (Ursel, 1602)
================================================================================

- Raw text -