Gpc August 2001

gpc@gnu.de

23 participants
33 discussions

compiler crash
by Adam Naumowicz 14 Aug '01

14 Aug '01

Hello, Trying to debug a program I used a conditional compilation of my source which simply includes a couple (two or three) of additional units providing some extended output to a file. Unfortunately then I got gpc: Internal compiler error: program gpc1 got fatal signal 11 gpc1: gpc exited with status 1 gpc1: gpc exited with status 1 gpc1: gpc exited with status 1 Maybe there is a limit of used units or some other entity that I exceeded? In fact I don't think if that could be that easy, but maybe ... ;-) I use the latest snapshot on Solaris. The problem existed also with the Linux version but vanished when I removed a conflict with overloaded methods. And I would like to re-ask a simple question for which I still don't know the answer: how to switch on/off the range checking with GPC ? Regards, Adam Naumowicz -------------------------------------- WWW: http://math.uwb.edu.pl/~adamn/ --------------------------------------

1 0

automake, autoconf etc.. missing
by Kocherlakota Harikrishna 09 Aug '01

09 Aug '01

Hello, while building the GPC for sun sparc solaris 2.6(native) the following messages are seen with 'configure' command checking for working autoconf missing checking for working automake missing checking for working xgettext missing checking for working bison missing will it any way cause problem in building the compiler ? As building of each of these tools requires lot of time, can i get executables for sparc solaris anywhere ? Any other alternative ? please suggest me. thanks and regards hari

3 3

request for help
by Kocherlakota Harikrishna 09 Aug '01

09 Aug '01

hello, while building the GPC for sparc sun solaris 2.6 (native) the following error is coming with 'make ' command. Can anybody help me out. make[1]: *** No rule to make target `choose-temp.o', needed by `xgpc'. Stop. make[1]: Leaving directory `/home/users/kharik/hari/gcc' make: *** [all-gcc] Error 2 thanks and regards hari

1 0

building GPC on sun-solaris
by Anuradha 07 Aug '01

07 Aug '01

Hello, I am using binutils-2.10,gcc-2.95.2 and gpc20010107 to build a GPc for sun-solaris-2.6. I did a configure with <gpc-src-dir>/configure --prefix=<objdir> --enable-languages=pascal --with-gnu-as --with-gnu-ld then gave make bootstrap It gives me an error fixinc.sh : cannot open file but the current directory is ofcourse <gpcsource direc>/gcc where fixinc is. Do you know what the error could be? Thank you. Kind Regards, Anu frank(a)g-n-u.de wrote: > Mirsad Todorovac wrote: > > > I have to report a bug in your code which prevents your PASCAL compiler to > > be built unless -DEGCS -DEGCS92 and -DEGCS95 are all defined (using 2.95.3 > > back-end). This has not been set by ``configure'', so manual intervention > > was needed. I can't tell if defining -DEGCS97 would be better than not, > > but I was satisfied (as for now) with having a PASCAL running on the > > system!!! (alpha-dec-osf4.0b box, gcc-2.95.3 compiler suite, gmake, gsed) > > > > Also, never in the installation instruction it isn't mentioned that GPC > > needs a smarter sed - this could be addressed by ``configure'' script too. > > Well, I'm not so familiar with non-GNU sed's. Since you didn't tell > us exactly fails with sed, I can only guess. Does the following > help? > > --- /home/gpc/src/p/config-lang.in.orig Mon May 7 03:54:43 2001 > +++ /home/gpc/src/p/config-lang.in Sun May 27 00:58:49 2001 > @@ -26,8 +26,8 @@ > # stagestuff - files to add to $(STAGESTUFF) > # diff_excludes - files to ignore when building diffs between two versions. > > -version=`sed -ne '/version_string/s/.*\"$[^ \"]*$[ \"].*/\1/p' "$srcdir/version.c"` > -mainversion=`echo $version | sed -e 's/.*\"$[0-9]*\.[0-9]*\.[0-9]*$.*/\1/'` > +version=`grep version_string "$srcdir/version.c" | sed -e 's/[^"]*"//;s/[ "].*//'` > +mainversion="$version" > if echo $version | grep '2\.9\|3\.0' > /dev/null; then > echo "#define EGCS" > gcc-version.h > echo "#define EGCS92" >> gcc-version.h > > (Peter, if you read this: If I'm not mistaken, mainversion is always > equal to version because the regex for mainversion contained a `"' > and version never contains any `"'. Since I don't know for which > strange e?gc[cs] version numbering schemes this might be useful, I > can't check it, either. So I'm just setting mainversion=version now. > If that's not was meant, please correct it yourself.) > > > Also, very helpful would be if the building of PASCAL compiler wouldn't > > include "fixing" of C include files again, which takes anormous amount of > > time on Alpha/Digital-Unix box. > > Building GPC always causes (most) of the C compiler to be built > because GPC shares parts with it. The fixincludes thing is probably > superfluous in this case, but I don't know if/how it can easily be > changed. Perhaps someone else can help you there... > > Maybe doing `touch stmp-fixinc' before the make helps, but that's > untested. > > 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 -- S.Anuradha Generic Data Tools, Alcatel Chennai. Alcanet : 2-757-7123

2 1

signals
by Adam Naumowicz 07 Aug '01

07 Aug '01

Hello, I have two questions: Is that the proper way to install a signal handler? program testsignal; uses gpc; Procedure CatchSignal(sig:integer); begin case sig of 2: writeln('Interrupted'); end; end; var o:TSignalHandler; b:boolean; i:longint; begin InstallSignalHandler(SigInt,CatchSignal,true,true,o,b); for i:=1 to 2000 do writeln(i); end. I intended it to be a SIGINT handler, but the Ctrl+C in the console doesn't make the CatchSignal proc executed. I also don't know why I cannot write SigInt as the case label - the compiler complains that SigInt is not a constant or a variable of an ordinal type (but it is declared as an integer in the GPC unit). Any ideas? The second question I would like to ask is about the 'segmentation fault' I get in one of my progs. I thought it can appear if I try to reference something that is already disposed, but this time my debugger says that it appears where a 'goto' statement is in the code. The executable is quite big (1,7 MB without the -s switch) and maybe that is the problem ? Anyone can tell me what actually the 'segmentation fault' is and what can be the reason of it? Thanks in advance. Regards, Adam Naumowicz -------------------------------------- WWW: http://math.uwb.edu.pl/~adamn/ --------------------------------------

4 8

Re: [Fwd: 16 bit sets?]
by da Silva, Joe 07 Aug '01

07 Aug '01

OK then, what I suggested might be suitable. All you would need is to redefine some of your type and record definitions, a bit. You wouldn't need to touch the "code proper", and the same source code could run on your older platforms too. Eg. perhaps as follows : type INT0_255 = packed 0..255; BITS = (BIT0,BIT1,BIT2,BIT3, BIT4,BIT5,BIT6,BIT7, BIT8,BIT9,BIT10,BIT11, BIT12,BIT13,BIT14,BIT15, BIT16,BIT17,BIT18,BIT19, BIT20,BIT21,BIT22,BIT23, BIT24,BIT25,BIT26,BIT27, BIT28,BIT29,BIT30,BIT31); var VARIABLE1 : record case shortint of 0: (DUMMY0, AWORD: shortint); 1: (DUMMY1: shortint; MSB, LSB: INT0_255); 2: (BT : BITS); end; Of course, DUMMY0/DUMMY1/BIT16..BIT31 would be just wasted memory, but you say that's not an issue ... You could even have some "conditional defines" to exclude them on other platforms, if required. :-) BTW, I have no idea which way the (set) bits and (16 bit) words are aligned within 32 bit words, for your intended platform(s), but I'm sure you can figure that out anyway. Joe. > -----Original Message----- > From: Sven Jauring [SMTP:sven.jauring@hiq.se] > Sent: Monday, August 06, 2001 10:02 PM > To: da Silva, Joe > Subject: Re: [Fwd: 16 bit sets?] > > Hi Joe > > Memory waste is no issue, but original code stability is. That is, I need > to > keep the original code as unmodified as possible. > > We are actually looking at how much work it would be to add the "set size" > functionality in GPC! > I would appretiate any thoghts/advice on this subject from anyone! > > /sven > > "da Silva, Joe" wrote: > > > Just an idea ... if memory (waste) isn't an issue, and if you > > can't (currently) get 16 bits sets with GPC, how about making > > everything 32 bit instead? > > > > Joe. > > > > > -----Original Message----- > > > From: Sven Jauring [SMTP:sven.jauring@hiq.se] > > > Sent: Friday, August 03, 2001 9:22 PM > > > To: Russell Whitaker > > > Cc: gpc(a)gnu.de > > > Subject: Re: [Fwd: 16 bit sets?] > > > > > > Hi Russel > > > Thanks for your help, but unfortunatly the pascal code that I'm trying > to > > > compile actually deals specifically with sets. (E.g. Variable1.BT := > > > Variable1.BT - [BIT4] + [BIT2]; and so forth...) > > > > > > Best Regards > > > Sven Jauring > > > > > > Russell Whitaker wrote: > > > > > > > On Fri, 3 Aug 2001, Sven Jauring wrote: > > > > > > > > > Hi again! > > > > > > > > > > I need to be able to specify not only 16 bit sizes but also 8 bit > and > > > 4 > > > > > bit sizes as well. > > > > > Is there anyone out there that could give me a hint of HOW to > > > implement > > > > > the functionality in GPC > > > > > > > > Patched your program a little. It *might* be what you're looking > for. > > > > > > > > > ******************************** > > > > type > > > > INT0_255 = packed 0..255; > > > > > > > > { BITS = (BIT0,BIT1,BIT2,BIT3, > > > > BIT4,BIT5,BIT6,BIT7, > > > > BIT8,BIT9,BIT10,BIT11, > > > > BIT12,BIT13,BIT14,BIT15); } > > > > > > > > bitarray = packed array [ 0..15 ] of integer ( 1 ); > > > > > > > > var > > > > VARIABLE1 : record case shortint of > > > > 0: (AWORD: shortint); > > > > 1: (MSB: INT0_255; > > > > LSB: INT0_255); > > > > 2: (BT : bitarray ); > > > > end; > > > > > > > > begin > > > > writeln( sizeof (VARIABLE1)); > > > > end. > > > > > ********************************* > > > > > > > > prints 2 (bytes) > > > > > > > > Hopes this helps > > > > Russ << File: Card for Sven Jauring >> << File: Card for Sven > Jauring >>

1 0

Re: [Fwd: 16 bit sets?]
by da Silva, Joe 06 Aug '01

06 Aug '01

Just an idea ... if memory (waste) isn't an issue, and if you can't (currently) get 16 bits sets with GPC, how about making everything 32 bit instead? Joe. > -----Original Message----- > From: Sven Jauring [SMTP:sven.jauring@hiq.se] > Sent: Friday, August 03, 2001 9:22 PM > To: Russell Whitaker > Cc: gpc(a)gnu.de > Subject: Re: [Fwd: 16 bit sets?] > > Hi Russel > Thanks for your help, but unfortunatly the pascal code that I'm trying to > compile actually deals specifically with sets. (E.g. Variable1.BT := > Variable1.BT - [BIT4] + [BIT2]; and so forth...) > > Best Regards > Sven Jauring > > Russell Whitaker wrote: > > > On Fri, 3 Aug 2001, Sven Jauring wrote: > > > > > Hi again! > > > > > > I need to be able to specify not only 16 bit sizes but also 8 bit and > 4 > > > bit sizes as well. > > > Is there anyone out there that could give me a hint of HOW to > implement > > > the functionality in GPC > > > > Patched your program a little. It *might* be what you're looking for. > > > > > ******************************** > > type > > INT0_255 = packed 0..255; > > > > { BITS = (BIT0,BIT1,BIT2,BIT3, > > BIT4,BIT5,BIT6,BIT7, > > BIT8,BIT9,BIT10,BIT11, > > BIT12,BIT13,BIT14,BIT15); } > > > > bitarray = packed array [ 0..15 ] of integer ( 1 ); > > > > var > > VARIABLE1 : record case shortint of > > 0: (AWORD: shortint); > > 1: (MSB: INT0_255; > > LSB: INT0_255); > > 2: (BT : bitarray ); > > end; > > > > begin > > writeln( sizeof (VARIABLE1)); > > end. > > > ********************************* > > > > prints 2 (bytes) > > > > Hopes this helps > > Russ << File: Card for Sven Jauring >>

1 0

Re: ISO standards (was Good version for mostly IT activity)
by da Silva, Joe 06 Aug '01

06 Aug '01

See below ... Joe. > -----Original Message----- > From: Bohdan Vlasyuk [SMTP:bohdan@kivc.vstu.vinnica.ua] > Sent: Saturday, July 21, 2001 12:10 AM > To: gpc(a)gnu.de > Subject: Good version for mostly IT activity > > Hi ! > > I need your advice -- I want to find stable gpc version suitable for > my needs.. > > I'm run Linux [2.4.6 as for now, and I plan to further upgrades :-)]. > > I need gpc to compile user programs, which would be likely developed > under mics. environment, but generally under DOS with old borland > pascal {They think that BP IDE is the best}. Programs are expected to > be relatively small, and I don't think that anyone would use > OO-extensions more that for few strings, it would at all. > > Actually, BP compatibility is not REQUIRED, but would be nice. Are > there any standards in Pascal like ISO C ?? > [Joe da Silva] Yes, there are actually two ISO Pascal standards : 1. ISO 7185 - Non-extended (aka Classic) Pascal 2. ISO 10206 - Extended Pascal (superset) You can obtain a free copy of ISO 10206, in either A4 or US-Legal document size, at : http://ftp.digital.com/pub/Digital/Pascal/iso10206_a4.ps.gz http://ftp.digital.com/pub/Digital/Pascal/iso10206.ps.gz You can also find an easy-to-read introduction to Extended Pascal (by Prospero Software) at : http://www.prosperosoftware.com/epintro.html > Personally I don't program Pascal, so I can't judge which one is > better. > > What can you tell me about FPC? Is it worth looking ?? > > thank you for your opinion. > > P.s.: If that's wrong list for that type of questions, I'm terrible > sorry, could you please direct me to right source? > > -- > "I once witnessed a long-winded, month-long flamewar over the use of > mice vs. trackballs...It was very silly." > (By Matt Welsh)

1 0

GPC on Mac OS X
by chewie＠mac.com 05 Aug '01

05 Aug '01

Hello, Has anyone been successful compiling GPC for Mac OS X. I am having no luck. Thanks, J

1 0

Re: Automatic gpc installation
by Ernst-Ludwig Bohnen 03 Aug '01

03 Aug '01

On Thu, 2 Aug 2001, Stephen Hurd wrote: > Yeah, gmake is in the path if it's installed from the ports collection or as > a package. It's not in the default install though, so maybe a test and > warning if it doesn't exist... doesn't really matter though, I'd think > anyone would pick up on that pretty quick. Removed the 'alias make=..' and made a check wether gmake can be used. This should work reliable on all? Unix/Linux Systems. Gmake is mandatory, not make. See: http://www.desy.de/~bohnen/gpc/install.gpc > > I wasn't sure if you wanted the output from ftp put into a logfile or not, I > noticed that wget was outputting to outfile (If I understand wget properly - > never used it) but I left it to the screen... mostly because it has a nice > status bar. The reason to hide the output (and not throwaway) of wget is to allow together with the -v 0 switch a batch version, which makes possible a remote triggered 'Hands off installation' for many machines on a bigger site. As far as I anderstand various ftp solutions, ftp works fine in interactive mode but not in batch. Do be done: Correct output of wget under control of the variable VERBOSITY !! Thank you for suggestions > It would have required less changes if it was left using wget, but since I > don't use it, I figured I'd set it up so other people could not use it too. > :-)

2 1

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 August 2001