From: alrc AT task DOT com DOT br (Andre Chrcanovic)
Newsgroups: comp.os.msdos.djgpp
Subject: Re: gprof with all zero execution times in version 2.03
Date: 13 May 2003 07:28:24 -0700
Organization: http://groups.google.com/
Lines: 49
Message-ID: <5c9be5c2.0305130628.45188bfd@posting.google.com>
References: <200305121929 DOT h4CJTEN08938 AT delorie DOT com> <b9qgbu$ef9$1 AT nets3 DOT rz DOT RWTH-Aachen DOT DE>
NNTP-Posting-Host: 200.165.11.92
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
X-Trace: posting.google.com 1052836105 16235 127.0.0.1 (13 May 2003 14:28:25 GMT)
X-Complaints-To: groups-abuse AT google DOT com
NNTP-Posting-Date: 13 May 2003 14:28:25 GMT
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

I got the output sample after running the program for approximately 45
minutes...

The compiler run on an Windows NT machine, but I run the executable in
a machine with FreeDos (Kernel version 1.1.23 - Build 2023). I use
cwsdpr0.exe as DPMI host to run the program.

Sections of the makefile are listed below:

CC      = djredir -eo gcc
LD      = gcc
AS      = asm
RM      = rm
LINKA   = gcc

INCLUDES = -Iinclude -Inucleo\include -I..\src\protocol
-I..\src\include -I..\src\tcpip\include

CFLAGS = -mcpu=pentium -pg -c -imacros boot.def -D__DOS__ $(INCLUDES)
DFLAGS = -E -M -MM -MT $(BINDIR)/$(basename $(@F)).o -w $(INCLUDES)

LNKOPTS = -L$(NPCDIR)/bin -L$(PRODIR)/bin -L$(LANDIR)/bin
-L$(TCPDIR)/bin \
          -lnuc00 -lprotocol -llan_n -ltcpip

GERACKSOBJ = $(BINDIR)/geracrc.o $(BINDIR)/crc.o

$(BINDIR)/$(BOOT).exe:   $(BOOTOBJ) $(BOOTLIB)
	    $(LINKA) -pg -o $(BINDIR)/$(BOOT).exe $(BOOTOBJ) $(LNKOPTS)
-Wl,-Map,boot_bz.map

Hans-Bernhard Broeker <broeker AT physik DOT rwth-aachen DOT de> wrote in message news:<b9qgbu$ef9$1 AT nets3 DOT rz DOT RWTH-Aachen DOT DE>...
> alrc AT task DOT com DOT br <alrc AT task DOT com DOT br> wrote:
> 
> > I work with DJGPP 2.3 (*.ver files: Djdev203.ver, Bnu213b.ver,
> > Gcc32b.ver) and I want to use gprof, but its exit shows all
> > functions (except __dpmi_int) with 0.00 % execution time.
> 
> There's more wrong than just a high percentage to ascribed to
> __dpmi_int.  Note that your program only reports a cumulative time of
> 0.28 total, that's a measly 5 time samples (0.055 seconds each) for
> gprof to work with, for the *whole* program!
> 
> Does three tenths of a second feel like a valid figure for the total
> run time of your program?  If so, you should really try to generate a
> test case that executes longer.  If not, something must have broken
> the profile data collection methods, and you'll have to supply more
> details, like: what platform did this happen on, what were the exact
> compile and link flags you passed.