Gpc July 2000

gpc@gnu.de

35 participants
109 discussions

Compiling GCC under Irix 6.5
by Jean-Marc Ottorini 24 Jul '00

24 Jul '00

While trying to build the gpc-20000719 / gcc-2.95.2 suite on top of Irix 6.5 (), the following command is issued by gmake: gcc -c -DIN_GCC -DGPC -g -O2 \ -I. -I.. \ -I../../../gcc-2.95.2/gcc/p \ -I../../../gcc-2.95.2/gcc/p/.. \ -I../../../gcc-2.95.2/gcc/p/../config \ -I../../../gcc-2.95.2/gcc/p/../../include \ ../../../gcc-2.95.2/gcc/p/../stor-layout.c This command results in a compiling error: ../../../gcc-2.95.2/gcc/p/../stor-layout.c:64: initializer element is not constant Indeed, line 64 of stor-layout.c reads: int set_words_big_endian = BYTES_BIG_ENDIAN; /* @@@ Needn't be a constant! */ Where is the truth ? I would be grateful to have a hint to possibly fix this problem Jean-Marc Ottorini _________________________________________________________________ | Jean-Marc Ottorini (MailTo://ottorini@nancy.inra.fr) | | Institut National de la Recherche Agronomique - Centre de Nancy | | 54280 Champenoux | | FRANCE | | tel. [+33] 03 83 39 40 46 | | fax 03 83 39 40 34 | |_________________________________________________________________|

1 0

Re: Copy function
by Frank Heckenbach 24 Jul '00

24 Jul '00

Couperin wrote: > After spending a long time to find why my program didn't work, I discovered > something which seems to be a bug. > > It's about the "Copy" function. > > Here is a test program : > > Program TestCopy; > > Var > Chaine : String(1296); {That's the value I need in my real program} > SousChaine : String(9); > LongueurChaine : MedWord; > IndexPosition : MedWord; > > Begin > Chaine := '1000000000'; > LongueurChaine := Length(Chaine); > > SousChaine := Copy(Chaine,LongueurChaine - 8,9); {Here is the problem} > Writeln('SousChaine : ',SousChaine); {Writes nothing !} > Writeln('Length(SousChaine) : ',Length(SousChaine)); {Writes 0} > > IndexPosition := LongueurChaine - 8; > SousChaine := Copy(Chaine,IndexPosition,9); {Now it works} > Writeln('SousChaine : ',SousChaine); {Writes 000000000} > Writeln('Length(SousChaine) : ',Length(SousChaine)); {Writes 9} > > {There is no problem with the "Delete" procedure :} > Delete(Chaine,LongueurChaine - 8,9); > Writeln('Chaine : ',Chaine) {Writes 1} > End. > > I can't understand why "Copy" works with "IndexPosition" as parameter, but not > with "LongueurChaine - 8". > Is there something wrong in : Copy(Chaine,LongueurChaine - 8,9) ? Or is this > really a bug ? > > I must confess I use GPC 2.8.1 (I need GRX and BGI2GRX that I failed to install > with the newest version). > If it's a bug, maybe it's solved now. In this case, excuse me for this somewhat > useless message. It is a GPC bug. In fact, it's a special case of the following bug listed on the To-Do list (most other cases have been fixed by now, but this one with `Copy' still exists in the latest GPC version): * integer parameters to built-in routines are not always converted to the right type (fjf349.pas) The difference between the two cases is that `IndexPosition' is of type MedWord which is probably the same size as Integer on your machine and therefore happens to work, while `LongueurChaine - 8' is internally of type LongInt (the smallest type that can hold a MedInt and negative values because the difference might be negative), and that triggers the bug. So, a work-around is to type cast the difference, i.e. `Integer (LongueurChaine - 8)'. Frank -- Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ GPC To-Do list, latest features, fixed bugs: http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

1 0

Copy function
by Couperin 24 Jul '00

24 Jul '00

After spending a long time to find why my program didn't work, I discovered something which seems to be a bug. It's about the "Copy" function. Here is a test program : Program TestCopy; Var Chaine : String(1296); {That's the value I need in my real program} SousChaine : String(9); LongueurChaine : MedWord; IndexPosition : MedWord; Begin Chaine := '1000000000'; LongueurChaine := Length(Chaine); SousChaine := Copy(Chaine,LongueurChaine - 8,9); {Here is the problem} Writeln('SousChaine : ',SousChaine); {Writes nothing !} Writeln('Length(SousChaine) : ',Length(SousChaine)); {Writes 0} IndexPosition := LongueurChaine - 8; SousChaine := Copy(Chaine,IndexPosition,9); {Now it works} Writeln('SousChaine : ',SousChaine); {Writes 000000000} Writeln('Length(SousChaine) : ',Length(SousChaine)); {Writes 9} {There is no problem with the "Delete" procedure :} Delete(Chaine,LongueurChaine - 8,9); Writeln('Chaine : ',Chaine) {Writes 1} End. I can't understand why "Copy" works with "IndexPosition" as parameter, but not with "LongueurChaine - 8". Is there something wrong in : Copy(Chaine,LongueurChaine - 8,9) ? Or is this really a bug ? I must confess I use GPC 2.8.1 (I need GRX and BGI2GRX that I failed to install with the newest version). If it's a bug, maybe it's solved now. In this case, excuse me for this somewhat useless message. -- "Couperin"

1 0

p/scripts/find-sed problem with djgpp and gpc20000723
by Joachim Falk 24 Jul '00

24 Jul '00

the find-sed script assumes that the temp dir is /tmp witch isnt always true under djgpp. instead you should use the TMPDIR var witch is set in djgpp.env MfG Joachim Falk

2 1

problem building rts gpc for gpc20000723
by Joachim Falk 24 Jul '00

24 Jul '00

The problem is the `echo $CC | sed 's/gcc/gpc/'` in p/rts/Makefile.in if $CC contains gcc in more than one position patched attached I havent tested this on the djgpp system but it schould work i have test dos style path with the sed script and it works # Generated automatically from Makefile.in by configure. # Makefile to generate the GNU Pascal Run Time System. # Copyright (C) 1995-2000 Free Software Foundation, Inc. # # Authors: J.J. van der Heijden <j.j.vanderheijden(a)student.utwente.nl> # Frank Heckenbach <frank(a)pascal.gnu.de> # # This file is part of GNU Pascal. # # GNU Pascal is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2, or (at your option) # any later version. # # GNU Pascal is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License # along with GNU Pascal; see the file COPYING. If not, write to the # Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA # 02111-1307, USA. # $(srcdir) is the RTS source directory VPATH = ../../../../gcc/p/rts srcdir = ../../../../gcc/p/rts # Reasonable default values, toplevel Makefile overrides them. # CC needs to be the GNU C compiler. CC = /cygdrive/d/cygwin/gcc-2.95.2-2/build/gcc/xgcc -B/cygdrive/d/cygwin/gcc-2.95.2-2/build/gcc/ -B/usr/bin/i586-pc-cygwin32/bin/ -I/usr/bin/i586-pc-cygwin32/include => problem here substitution gcc => gpc wont do here patched sed => commandline PC = `echo $(CC) | sed 's@^$[^ ]*[/\\][^/\\]*$gcc@\1gpc@'` CFLAGS = -O2 AR = ar AR_FLAGS = rc RANLIB = ranlib RANLIB_TEST = true DESTDIR = . --- p/rts/Makefile.in.old Mon Jul 24 01:31:26 2000 +++ p/rts/Makefile.in Mon Jul 24 01:33:16 2000 @@ -28,7 +28,7 @@ # Reasonable default values, toplevel Makefile overrides them. # CC needs to be the GNU C compiler. CC = @CC@ -PC = `echo $(CC) | sed 's/gcc/gpc/'` +PC = `echo $(CC) | sed 's@^$[^ ]*[/\\][^/\\]*$gcc@\1gpc@'` CFLAGS = -O2 AR = @AR@ AR_FLAGS = rc

1 0

build problem with gpc20000723 on cygwin rts.c
by Joachim Falk 24 Jul '00

24 Jul '00

The Problem is that the latest version of cygwin downloaded just 3 days ago doesnt have MAP_FAILED in /usr/include/sys/mman.h this prevents rts.c to compile I dont know if this is a bug in the cygwin header files or in gpc MfG Joachim Falk

1 0

gpc
by David Kredba 23 Jul '00

23 Jul '00

How can I distribute binaries of Gpc for gcc 2.95.2 (under DJGPP) ? The latest version of the Gpc at ftp.simtel.net and other mirrors is from 1996. I build version 20000707 under the gcc 19991024 and it is fully functional. Thank you for support. David Kredba

2 1

output message
by Tom Schneider 23 Jul '00

23 Jul '00

Hi! I'm going through my programs and was puzzled by this one: listerG.p: In function `Report': (irrelevant error) It turns out that the code is: procedure report( ... So gpc is calling this a function when it is really a procedure. Is this unavoidable or can it be made smarter? Of course I am still having trouble with the capitalization that gpc insists on because it prevents me from doing cut and paste to search. No other compiler has ever done this of the 4 or more others I've worked with. Thanks! Tom Dr. Thomas D. Schneider National Cancer Institute Laboratory of Experimental and Computational Biology Frederick, Maryland 21702-1201 toms(a)ncifcrf.gov permanent email: toms(a)alum.mit.edu http://www.lecb.ncifcrf.gov/~toms/

2 1

Re: Compiling GPC under Irix 6.5
by Frank Heckenbach 22 Jul '00

22 Jul '00

Jean-Marc Ottorini wrote: > I am currently in the process of compiling GPC on an SGI O2 station > under Irix 6.5, and an error is occurring in the buiding of the > compiler (after a successful configuration). > > This compilation is based on gcc-2.95.2 (the core version) and gpc-20000719. Have you tried to build GCC without GPC first? The error message > [...] > > gmake[1]: *** [stmp-multilib] Error 1 > gmake[1]: Leaving directory `/usr/people/ottorini/pascal/gpc-build/gcc' > gmake: *** [install-gcc] Error 2 looks like the problem is not in the GPC code, but in the general GCC code. Did you check the section on Irix in the install/specific.html file, and perhaps the URL at sgi.com given there? Frank -- Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ GPC To-Do list, latest features, fixed bugs: http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

1 0

Re: gpc-20000707 djgpp native compilation
by Frank Heckenbach 22 Jul '00

22 Jul '00

Maurice Lombardi wrote: > OK I have downloaded gpc-20000719. > Everythings (demos and test suite) works now both with gmp 2.2 and gmp 3.01 downloaded from djgpp > site. > fjf119 and fjf120 also work now. Fine. > There is a problem to run the test suite on djgpp. > It is the same that forced djgpp maintainers to split the build process with a build.bat. > For a too large number of compilations, program stops or crashes do to lost dpmi handles. > This is fortunately reset when going back to dos (even within a bat file). > The solution is thus to split the running of tests in chunks (of 250 files). > The attached files dostest.bat which calls dodostst.sh and makefile.dos, > to be put in the test directory do that automatically. > type > dostest > the result are concatenated into a file test_sum.out and typed at end. I'm a bit concerned with the changes to the Makefile (two similar versions of the same file are always harder to maintain -- and there will probably more changes to the Makefile in the future). First, I think you don't have to comment out MASK because the command line setting in dodostst.sh should override it, anyway. Then, SHELL should probably be set to /bin/sh (which `make' under DJGPP will change to the correct path automagically). I think we can do this in the main Makefile if it doesn't break things on other systems. On Unix, it's no problem, but I don't know about Cygwin and Mingw. Chief, you might want to test it -- perhaps it'll be necessary to set SHELL to the full path of sh.exe before calling make (if you don't already have to do this, anyway). Finally, I think the output redicrection can also be done in the shell file, i.e.: make SHELL=sh MASK="$FILES" >> test_sum.out If you're concerned about the messages ("Running the test suite..."), I can add another target, say `pascal.check-nomsg' which will not write the messages, together with a target `msg' which will just output the messages, in case you want to run this first. -- Or wait, perhaps even better: run `pascal.check-long' in the loop, and pipe the concatenated results through test_sum at the end -- this should be more convenient as the results will be listed as a total. BTW, what does the for loop in dodostst.sh do? Doesn't make SHELL=sh MASK="`cat $1`" pascal.check-long >> test_sum.out work already? Then, you could probably get by without dodostst.sh at all, by just inserting it into the batch file (don't worry about the `` -- command.com will not expand them, of course, but sh called by make later will do so). Finally, I suggest to change test_sum.out to make.out because "sum" is then no more appropriate, and other scripts already use make.out, and the Makefile will remove make.out in the `extraclean' target. Likewise, renaming list.* to dtlist.* might avoid conflict with possible later list* test programs... This results in the following file (provided you agree to release it under the GPL). I'm testing it under DosEmu, and it seems to work. I'll upload it already -- if it doesn't work for you, let me know. @echo off rem Script to run the GPC Test Suite under DJGPP on a system that rem suffers from losing DPMI handles (such as MS-Windows). rem rem Copyright (C) 2000 Free Software Foundation, Inc. rem rem Authors: Maurice Lombardi [Maurice.Lombardi(a)ujf-grenoble.fr] rem Frank Heckenbach [frank(a)pascal.gnu.de] rem rem This file is part of GNU Pascal. rem rem GNU Pascal is free software; you can redistribute it and/or modify rem it under the terms of the GNU General Public License as published by rem the Free Software Foundation; either version 2, or (at your option) rem any later version. rem rem GNU Pascal is distributed in the hope that it will be useful, rem but WITHOUT ANY WARRANTY; without even the implied warranty of rem MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the rem GNU General Public License for more details. rem rem You should have received a copy of the GNU General Public License rem along with GNU Pascal; see the file COPYING. If not, write to the rem Free Software Foundation, 59 Temple Place - Suite 330, Boston, MA rem 02111-1307, USA. if exist make.out del make.out if exist dtlist.* del dtlist.* make msg ls -1 *.pas | split -l 250 - dtlist. for %%f in (dtlist.*) do make MASK="`cat %%f`" pascal.check-long >> make.out sh test_sum < make.out Frank -- Frank Heckenbach, frank(a)g-n-u.de, http://fjf.gnu.de/ GPC To-Do list, latest features, fixed bugs: http://agnes.dida.physik.uni-essen.de/~gnu-pascal/todo.html

3 2

← Newer
1
2
3
4
5
6
7
...
11
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

Gpc July 2000