Gpc

gpc@gnu.de

5400 discussions

Byte Order checking without {$ifdef}'s...
by Orlando Llanes 05 Apr '00

05 Apr '00

This is something that is prolly considered "common-sense" but I'm posting anyway just in case someone can benefit from it. I might've gotten the idea from Frank's code posted a while back, I don't remember how he implemented Endian conversion routines. Here's the code: program Test_Endian; const ByteOrder = $01020304; function LittleEndianCPU: Boolean; var mByteOrder : Cardinal(32); mByte : array[ 1..4 ] of Cardinal(8); begin mByteOrder := ByteOrder; Move(mByteOrder, mByte, 4); LittleEndianCPU := (mByte[1] = 4) and (mByte[2] = 3) and (mByte[3] = 2) and (mByte[4] = 1); end; function BigEndianCPU: Boolean; var mByteOrder : Cardinal(32); mByte : array[ 1..4 ] of Cardinal(8); begin mByteOrder := ByteOrder; Move(mByteOrder, mByte, 4); BigEndianCPU := (mByte[1] = 1) and (mByte[2] = 2) and (mByte[3] = 3) and (mByte[4] = 4); end; begin WriteLn( 'Little Endian CPU present -> ', LittleEndianCPU ); WriteLn( ' Big Endian CPU present -> ', BigEndianCPU ); end. The above code should compile, I translated it from the BP code. The only difference is in the type declarations; mByteOrder must always be 32-bits, and mByte must always be an array of 4 elements with 8-bits each, hence the use of Cardinal(x). The way it works is simple, and the code does not depend on any compiler specific syntax (except for type declarations) so it should compile correctly in the long run. AFAIK, constants are compiled in the native byte order of the target architecture. By comparing the long value as bytes, one-by-one, the byte order can be determined, hence the endianess can also be determined. I have tested this code on a Macintosh computer (which is the only big-endian machine I currently have access to), and have of course tested this on a PC. The results displayed are correct in both architectures. See ya! Orlando Llanes "Meine Damen und Herren, Elvis hat soeben das Gebaeude verlassen!" "Look out fo' flyeeng feet" O__/ a010111t(a)bc.seflin.org \/|____. O <__. \/\> / \ ____________|___\______ http://ourworld.compuserve.com/homepages/Monkey414

2 2

Re: 6 MOD -90 Problem on Windows
by Frank Heckenbach 05 Apr '00

05 Apr '00

Tim Kaulmann wrote: > I found out that the result of 6 MOD -90 in my little Pascal Program on > Windows is -84 and on Solaris is 6 (in my opinion the right one). > Whats up there ? > BTW : My Calculator HP48 says that -84 was right ??? > A matter of definition ? Actually, it's defined as an error when the right operand is <=0. So probably we should implement such a check in GPC (perhaps together with other kinds of range or overflow checking). If in your program the right operand is a variabls which might be negative, I suggest to use `Abs' (i.e., `a mod Abs (b)'), since a mod b = a mod (-b), according to "your" definition. 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: librares for GPC and (or) GCC
by Frank Heckenbach 28 Mar '00

28 Mar '00

Robson wrote: > On 14 Mar 00, at 20:45, Robson Ferreira wrote: > > > I want to know if you help me to resolve some > > problems in to compile programs in the linux. > > I am using the compiler GPC for to compile programs > > in Pascal and my problem is that some libraries not > > compile (crt.h,dos.h,io.h) " how to resolve this?"; > > and I am using GCC to compile programs in C > > an has the equal problem (conio.h,...). Some > > commands not execute (fgets,getch,delay, > > clsrcr, gets) in the LINUX... > > Help me, please... > > GPC cannot use the C headers directly. You have to convert them > into Pascal. For more information about GPC programming under > Linux, post your questions to "gpc(a)gnu.de". You will have people > who use Linux all the time and who will be able to answer your > questions there. Some functions in these DJGPP headers headers are not portable (some are, e.g. fgets, but others like clrscr aren't), so you can't use them under Linux directly. Furthermore, as The Chief said, you can't use C headers directly... However, recent releases of GPC come with a set of units which are both portable and written in Pascal, so that's probably what you need. For the headers mentioned, you probably want to look into the GPC, CRT and maybe Dos units. (The Dos unit, despite its name, is portable as well.) 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: librares for GPC and (or) GCC
by Robson 28 Mar '00

28 Mar '00

-----Mensagem original----- De: Dr A Olowofoyeku <laa12(a)law.keele.ac.uk> Para: Robson Ferreira <ferreira(a)ufu.br> Data: Sexta-feira, 24 de MarÃ§o de 2000 5:34 Assunto: Re: librares for GPC and (or) GCC On 14 Mar 00, at 20:45, Robson Ferreira wrote: > I want to know if you help me to resolve some > problems in to compile programs in the linux. > I am using the compiler GPC for to compile programs > in Pascal and my problem is that some libraries not > compile (crt.h,dos.h,io.h) " how to resolve this?"; > and I am using GCC to compile programs in C > an has the equal problem (conio.h,...). Some > commands not execute (fgets,getch,delay, > clsrcr, gets) in the LINUX... > Help me, please... GPC cannot use the C headers directly. You have to convert them into Pascal. For more information about GPC programming under Linux, post your questions to "gpc(a)gnu.de". You will have people who use Linux all the time and who will be able to answer your questions there. Best regards, The Chief ----- Dr Abimbola A Olowofoyeku (The African Chief) Email: African_Chief(a)bigfoot.com Author of Chief's Installer Pro v5.22 for Win32 http://www.bigfoot.com/~African_Chief/chief32.htm

1 0

Re: Probably build RTFM issue.
by Frank Heckenbach 26 Mar '00

26 Mar '00

Mariusz Zynel wrote: > > Neither GNU make, patch nor (I think) install should be required. > > ... but GNU patch is necessary. GNU make and install are optional, > at least for me. I'm using Solaris 7 and 8 on Sparc and x86. Strange, since we use context diffs which should be most portable. But anyway, sometime soon, GPC will not need any GCC patches at all, and this problem will be gone automatically... 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

2 1

RE: gpc on Sun
by Mason Ip 22 Mar '00

22 Mar '00

I have no problem with the (more-or-less) latest gcc and gpc. gpc -v Reading specs from /opt/GCC2951/lib/gcc-lib/sparc-sun-solaris2.7/2.95.1/specs gpc version 19991030, based on 2.95.1 19990816 (release) gpc --borland-pascal -O -o xx xx.pas <-- compilation OK; executable OK ---------- From: Veselic [SMTP:veselic@zagreb.FernUni-Hagen.de] Sent: Wednesday, March 22, 2000 6:26 AM To: gpc(a)gnu.de Subject: gpc on Sun Dear gpc-ers, just hearing on problems with gpc on Sun my Sun Solaris has binary-loaded gpc as follows GNU Pascal version 19990118, based on gcc-2.8.1, compiled Jan 18 1999 18:50:47. Copyright (C) 1987, 89, 92-97, 1998 Free Software Foundation, Inc. However, a lot (not all) of sources compiling smoothly under PC-Linux now halt compiling with the message gpc --borland-pascal -O -o sj sj.p gpc: Internal compiler error: program gpc1 got fatal signal 6 Sometimes there is 'fatal signal 11'. Can anybody help? Thanks, K. Veselic.

1 0

Re: Probably build RTFM issue.
by Frank Heckenbach 22 Mar '00

22 Mar '00

Mariusz Zynel wrote: > > Build platform: Sun Solaris 2.7 > > GNU 'make' version: 3.78.1 > > GNU 'patch' version: 2.5.4 > > > > There is a newer alpha GPC release, which builds less painfull on > Solaris. I had problems due to bash extensions used in Makefile-s. > > Besides, there are Solaris binaries in pkgadd form to download at: > > http://math.uwb.edu.pl/download > ftp://mizar.uwb.edu.pl/pub/misc > > Hope this helps The `if ! [ ... ]' Makefile problem will be solved in the next release. Were there any more problems left in the last release? Neither GNU make, patch nor (I think) install should be required. 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

2 1

gpc on Sun
by Veselic 22 Mar '00

22 Mar '00

Dear gpc-ers, just hearing on problems with gpc on Sun my Sun Solaris has binary-loaded gpc as follows GNU Pascal version 19990118, based on gcc-2.8.1, compiled Jan 18 1999 18:50:47. Copyright (C) 1987, 89, 92-97, 1998 Free Software Foundation, Inc. However, a lot (not all) of sources compiling smoothly under PC-Linux now halt compiling with the message gpc --borland-pascal -O -o sj sj.p gpc: Internal compiler error: program gpc1 got fatal signal 6 Sometimes there is 'fatal signal 11'. Can anybody help? Thanks, K. Veselic.

1 0

Probably build RTFM issue.
by Randy Latimer 21 Mar '00

21 Mar '00

Hi. I've just built your package and have a possible bug report. Of course, this may be a RTFM isuue, but I felt I should offer this report none the less. Synopsis: When installing the package, the install of the units, docs, and demos causes fatal make errors. Detailed descripion: Download date: 20 Mar 2000 Download site: ftp://agnes.dida.physik.uni-essen.de/gnu-pascal/current/ Package: gpc-19990118.tar.gz Build platform: Sun Solaris 2.7 GNU 'make' version: 3.78.1 GNU 'patch' version: 2.5.4 (Note: I also built this package on a RedHat Linux v6.0 system without this error occurring. 'INSTALL_DATA', which is set to the same value as 'INSTALL' appears to have been configured with the value '/usr/bin/install -c'.) The configure script appears to have set the values of both INSTALL and INSTALL_DATA to './install.sh -c'. This causes the installation of units, demos, and docs to fail. The makefile which is generated, performs a change of directory prior to calling 'INSTALL_DATA'. I modified the Makefile and added '../../' prior to the $(INSTALL_DATA) in the offending lines (three of them). These lines appeared in the 'pascal.install-*' rules. The changes were as follows. pascal.install-units: pascal.install-dir $(srcdir)/p/units/crtc.h . . . cd $(srcdir)/p/units; \ for file in *.pas *.inc *.c *.h; do \ ../../$(INSTALL_DATA) $$file $(unitsubdir)/$$file; \ chmod a-x $(unitsubdir)/$$file; \ done pascal.install-demos: pascal.install-dir cd $(srcdir)/p/demos; \ for file in *; do \ ../../$(INSTALL_DATA) $$file $(demodir)/$$file; \ chmod a-x $(demodir)/$$file; \ done pascal.install-info: pascal.info -rm -f $(infodir)/gpc.info* ; if [ -f p/doc/gpc.info ]; then \ cd p/doc; \ else \ cd $(srcdir)/p/doc; \ fi; \ for f in gpc.info*; do \ ../../$(INSTALL_DATA) $$f $(infodir)/$$f; \ done Further investigation shows that the Linux box (which installed just fine) has the GNU version of 'install' and the 'install' on the Sun is the version that ships from Sun. Your configure script may want to use GNU 'install' and chooses to use your 'install.sh' script in place of 'install' if GNU 'install' is not found. This appears to be the case. Conclusion: This entire situation may be a RTFM issue. However, I read the README and INSTALL file and no mention was made of needing to install GNU versions of make, patch, or install. The fact that I needed to install GNU make and GNU patch were SWAG's (Scientific Wild A$$ Guess's) on my part and only during the composition of this email have I realized that there is a GNU version of install as well. Thank you for the product and sorry for this email if it's an RTFM issue. -- Randy Latimer ______|___________________ Program in Human Genetics | | || || NS | | Department of Physiology & Biophysics | | [ ] ||[ ]|| [ ] | | Vanderbilt University School of Medicine |__|_____||____||_____|__| Nashville, Tennessee, USA O<=>O O<=>O ------------------------------------------------------------------------ "A man's mind, stretched by new ideas, can never go back to its original dimensions." Oliver Wendell Holmes

2 1

Newbie question about cygwin
by Matze Braun 19 Mar '00

19 Mar '00

Hello, I'm using GPC for DJGPP at the moment, which works perfect. Now I wanted to try the cygwin port of GPC. I've installed the cygwin b20 and then the version from "the Chief". I can compile C programms, but every time I compile .PAS programms I got the message, that gpc-cpp can't be executed. I think, it's because the cygwin b20 is still an egcs version and the gpc version is 2.95.1 Is it possible to mix 2 versions, if not can someone give me a hint how to compile the cvs snapshots of cygwin? Thanks in advance Matze

2 1

Jump to page:

2025

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