tack-devel — Discussion for developers of the ACK.

Re: [Tack-devel] compiling on linux code for 6500

[Norayr C. <gn...@gm...>]

Thank you very much for reply :)

Norayr Chilingaryan

---
http://geocities.com/asprayama/if-so


On 1/19/06, David Given <dg...@co...> wrote:
> On Thursday 19 January 2006 14:23, Norayr Chilingaryan wrote:
> [...]
> > Is this the system you are running on? (y/n) [n]
>
> Yes, you're asking for it to be cross-compiled using the ACK itself, whic=
h is
> why it's asking for a runnable ACK tree:
>
> [...]
> > Please give the root of a runnable ACK tree,
> > an absolute path: [/usr/local/6500]
>
> Since you don't have one, it's going to fail.
>
> There are detailed instructions in the README on how to build the ACK for
> Linux --- could you give them a try and let me know if it works?
>
> (The build process is a horrible, unmaintainable mess, and I'm working on
> rewriting it. Sorry it's taking so long.)
>
> --
> +- David Given --McQ-+ "I can handle myself. I've been in a fire fight.
> |  dg...@co...    | Well, I was in a fire.... Actually I was fired.
> | (dg...@ta...) | From a fry cook opportunity." --- Firefly, _War
> +- www.cowlark.com --+ Stories_
>
>
>

Re: [Tack-devel] compiling on linux code for 6500

[David G. <dg...@co...>]

On Thursday 19 January 2006 14:23, Norayr Chilingaryan wrote:
[...]
> Is this the system you are running on? (y/n) [n]

Yes, you're asking for it to be cross-compiled using the ACK itself, which =
is=20
why it's asking for a runnable ACK tree:

[...]
> Please give the root of a runnable ACK tree,
> an absolute path: [/usr/local/6500]

Since you don't have one, it's going to fail.

There are detailed instructions in the README on how to build the ACK for=20
Linux --- could you give them a try and let me know if it works?

(The build process is a horrible, unmaintainable mess, and I'm working on=20
rewriting it. Sorry it's taking so long.)

=2D-=20
+- David Given --McQ-+ "I can handle myself. I've been in a fire fight.
|  dg...@co...    | Well, I was in a fire.... Actually I was fired.
| (dg...@ta...) | From a fry cook opportunity." --- Firefly, _War
+- www.cowlark.com --+ Stories_

[Tack-devel] Re: compiling on linux code for 6500

[Norayr C. <gn...@gm...>]

Excuse me, I forgot to mention version of gcc
3.3.6
glibc
2.3.5-r2


Thanks

[Tack-devel] compiling on linux code for 6500

[Norayr C. <gn...@gm...>]

Hello.
I downloaded ack source, and trying to build it on my gentoo linux box.
All is fine, but there are no word about how to compile it to produce
binaries for 6500.

My answers:
Give me the type of the system, the current choice is
vax_bsd4_1a     VAX11 with BSD4.1a
vax_bsd4_2      VAX11 with BSD4.2
vax_sysV_2      VAX11 with System V.2
i386            Intel 80386 system running Xenix System V
sun3            Sun 3 M68020 workstation
sun2            Sun 2 M68000 workstation
m68_sysV_0      Motorola 68000 with Uniplus System V.0 Unix
m68020          Motorola M68020 VME131 running Unix System V/68 R2V2.1
sparc           SUN SPARC workstation running SunOs 4
sparc_solaris   SUN SPARC workstation running solaris 2
ANY             Neither of the above

system type: [ANY] ANY

Is this the system you are running on? (y/n) [n]

You will now be asked for the root directory of ACK on the current machine.
This tree will not be changed by the installation process.

Please give the root of a runnable ACK tree,
an absolute path: [/usr/local/6500]

The system to install ACK for is ANY,
the root of the ACK source tree is /home/noch/install/ack/ack-5.6,
the root of the configuration tree is /tmp/ack-conf,
the root of the ACK binary tree to be created is /usr/local/ack,
and the root of a runnable ACK binary tree is /usr/local/6500.
If the machine to compile ACK for is the current machine, the last two name=
s
may be identical.
Are you satisfied with all this? (y/n) y

This installation script has no knowledge about ANY.
You will have to specify the default machine that you want ACK to
compile for. Choices:
i86 xenix3 minix i386 6500 6800 6805 6809 i80 em22
em24 em44 m68k2 pmds minixST m68k4 pmds4 sun2 mantra m68020
sun3 sparc sparc_solaris ns pdp s2650 vax4 z80 z8000 arm

Your choice: [6500]

What kind of Unix is the target system running?
Choices:
        V7      for Unix V7, BSD 2.*
        BSD4_1  for Berkeley 4.1
        BSD4_2  for Berkeley 4.2, 4.3, SunOs 3, SunOs 4
        SYS_5   for Xenix, System III, System V, SunOs 5
Your choice (V7|BSD4_1|BSD4_2|SYS_5): [SYS_5] V7

Your default machine to compile for is 6500
Installation of the complete ACK takes a long time. Limiting the
number of languages, runtime libraries, back-ends, and assemblers to be
installed may save a lot of time. If you want to install everything,
answer no to the next question. Otherwise, answer yes, and you will be
prompted for details
Do you want to limit the installation in any way? (y/n) [n] n

Which system-call library do you want to install for the VAX?
You can choose between
        libbsd4_1a      for Berkeley Unix 4.1
        libbsd4_2       for Berkeley Unix 4.2 or newer, or Ultrix
        libsysV_2       for Unix System V
Your choice: [libsysV_2]

Please give the word-size of the target-machine (sizeof(int)) in bytes: [2]

Please give the pointer-size of the target-machine (sizeof(char *)) in
bytes: [1]

A file called 'make_macros' has been created. This file defines some
'make' variables that parameterize all Makefiles in ACK. You may want
to check it before attempting to actually install ACK.
In fact, this installation script does not know much about
your target machine, so expect some things to be wrong

A shell-script called 'INSTALL' has been created. Running it
installs ACK. Note that this may take a (very) long time, so run it
in the background, with its output redirected, f.i.:
        sh INSTALL > INSTALL.out 2>&1 &

wks-nchiling first # ./INSTALL
System definition -- done
Failed for Manual pages, see man/Out
EM definition library -- done
Failed for C utilities, see util/cmisc/Out
Yacc parser generator -- done
Failed for Flex lexical analyzer generator, see util/flex/Out
Include files for modules -- done
Failed for system-call interface module, see system/Out
Failed for string routines module, see string/Out
Failed for formatted print module, see print/Out
Failed for assertion module, see assert/Out
Failed for memory allocation module, see alloc/Out
Failed for fast, linear time malloc, see malloc/Out
Failed for identifier table module, see idf/Out
Failed for input module, see input/Out
Failed for ACK-object reading and writing module, see object/Out
Failed for EM code generation module, see em_code/Out
Failed for EM messages generation module, see em_mes/Out
Failed for EM-code reading module, see read_em/Out
Failed for Floating point arithmetic module, see flt_arith/Out
Failed for Modules
Failed for LL(1) Parser generator, see util/LLgen/Out
Failed for C preprocessor, see util/cpp/Out
Failed for ACK object utilities, see util/amisc/Out
Failed for Encode/Decode, see util/misc/Out
Shell files in bin -- done
Failed for EM assembler, see util/ass/Out
Failed for EM Peephole optimizer, see util/opt/Out
Global optimizer, shared code -- done
Global optimizer, intermediate code -- done
Global optimizer, control flow -- done
Global optimizer, cross jumping -- done
Global optimizer, stack pollution -- done
Global optimizer, branch optimization -- done
Global optimizer, strength reduction -- done
Global optimizer, inline substitution -- done
Global optimizer, use-definition analysis -- done
Global optimizer, common subexpression elimination -- done
Global optimizer, live-variable analysis -- done
Global optimizer, register allocation -- done
Global optimizer, compact assembly generation -- done
Global optimizer, machine descriptions -- done
Global optimizer, driver -- done
EM Global optimizer -- done
Failed for ACK archiver, see util/arch/Out
Failed for Program 'ack', see util/ack/Out
Failed for Bootstrap for backend tables, see util/cgg/Out
Failed for Bootstrap for newest form of backend tables, see util/ncgg/Out
Code Expander Generator, EM_parser/as_EM_pars part -- done
Code Expander Generator, EM_parser/obj_EM_pars part -- done
Failed for Code Expander Generator, EM_parser/common part, see
EM_parser/common/Out
Failed for Code Expander Generator, as_parser part, see as_parser/Out
Code Expander Generator, assemble part -- done
Code Expander Generator, ce_back part -- done
Code Expander Generator, defaults part -- done
Code Expander Generator, utilities -- done
Failed for Bootstrap for code expanders
Failed for LED link editor, see util/led/Out
Failed for TOPGEN target optimizer generator, see util/topgen/Out
Failed for C frontend, see lang/cem/cemcom/Out
Failed for ANSI-C frontend, see lang/cem/cemcom.ansi/Out
Failed for ANSI-C preprocessor, see lang/cem/cpp.ansi/Out
ANSI-C header files -- done
LINT C program checker -- done
Failed for EM definition lint-library, see util/data/Out
Failed for system-call interface module lint library, see system/Out
Failed for string routines module lint library, see string/Out
Failed for formatted print module lint library, see print/Out
Failed for assertion module lint library, see assert/Out
Failed for memory allocation module lint library, see alloc/Out
fast, linear time malloc lint library -- done
identifier table module lint library -- done
Failed for input module lint library, see input/Out
Failed for ACK-object reading and writing module lint library, see object/O=
ut
Failed for EM code generation module lint library, see em_code/Out
Failed for EM messages generation module lint library, see em_mes/Out
Failed for EM-code reading module lint library, see read_em/Out
Failed for Floating point arithmetic module lint library, see flt_arith/Out
Failed for Modules lint libraries
Failed for Global optimizer lint libraries, see util/ego/share/Out
Failed for Pascal frontend, see lang/pc/comp/Out
Failed for Basic frontend, see lang/basic/src/Out
Failed for Occam frontend, see lang/occam/comp/Out
Failed for Modula-2 frontend, see lang/m2/comp/Out
Modula-2 definition modules -- done
Failed for Modula-2 makefile generator, see lang/m2/m2mm/Out
Fortran to C compiler -- done
Failed for EM interpreter in C, see util/int/Out
Failed for Symbolic debugger, see util/grind/Out
Failed for Intel 8086 assembler, see as/Out
Failed for Intel 8086 backend, see ncg/Out
Failed for Intel 8086 EM library, see libem/Out
Failed for Intel 8086 etext,edata,end library, see libend/Out
Failed for Intel 8086 floating point library, see libfp/Out
Failed for Intel 8086 PC/IX systemcall library, see libsys/Out
Intel 8086 conversion program from ack.out --> PC/IX a.out -- done
Failed for i86 Modula-2 libraries, see libm2/Out
Failed for i86 Pascal libraries, see libpc/Out
Failed for i86 Basic libraries, see libbc/Out
Failed for i86 Occam libraries, see liboc/Out
Failed for i86 ANSI-C libraries, see libcc.ansi/Out
Failed for i86 C libraries, see libcc/Out
Failed for Intel 8086 support
Failed for Xenix 3.0 80286 systemcall library, see libsys/Out
Xenix 3.0 conversion program from ack.out --> Xenix 3.0 a.out -- done
Failed for xenix3 Modula-2 libraries, see libm2/Out
Failed for xenix3 Pascal libraries, see libpc/Out
Failed for xenix3 Basic libraries, see libbc/Out
Failed for xenix3 Occam libraries, see liboc/Out
Failed for xenix3 ANSI-C libraries, see libcc.ansi/Out
Failed for xenix3 C libraries, see libcc/Out
Failed for Intel 80286 support for Xenix
Failed for Intel 80386 assembler, see as/Out
Failed for Intel 80386 backend, see ncg/Out
Sorry, Intel 80386 code expander library can only be made on i386 systems
Intel 80386 Xenix System V conversion program -- done
Failed for Intel 80386 EM library, see libem/Out
Failed for Intel 80386 etext,edata,end library, see libend/Out
Failed for Intel 80836 floating point library, see libfp/Out
Failed for Intel 80836 debugger support library, see libdb/Out
Failed for Intel 80836 Xenix systemcall library, see libsys/Out
Failed for i386 Modula-2 libraries, see libm2/Out
Failed for i386 Pascal libraries, see libpc/Out
Failed for i386 Basic libraries, see libbc/Out
Failed for i386 Occam libraries, see liboc/Out
Failed for i386 ANSI-C libraries, see libcc.ansi/Out
Failed for i386 C libraries, see libcc/Out
Failed for i386 Fortran libraries, see libf77/Out
Failed for Intel 80386 support for Xenix 386 System V
Failed for MSC6500 assembler, see as/Out
Failed for MSC6500 backend, see cg/Out
MSC6500 download program(s) -- done
MSC6500 EM library -- done
MSC6500 etext,edata,end library -- done
MSC6500 MON library -- done
Failed for MSC6500 system call library, see libsys/Out
Failed for 6500 Modula-2 libraries, see libm2/Out
Failed for 6500 Pascal libraries, see libpc/Out
Failed for 6500 Basic libraries, see libbc/Out
Failed for 6500 Occam libraries, see liboc/Out
Failed for 6500 ANSI-C libraries, see libcc.ansi/Out
Failed for 6500 C libraries, see libcc/Out
Failed for 6500 Fortran libraries, see libf77/Out
Failed for MSC6500 support
Failed for Motorola 6800 assembler, see as/Out
Failed for Motorola 6800 support
Failed for Motorola 6805 assembler, see as/Out
Failed for Motorola 6805 support
Failed for Motorola 6809 assembler, see as/Out
Failed for Motorola 6809 support
Failed for Intel 8080 assembler, see as/Out
Failed for Intel 8080 code generator, see ncg/Out
Failed for Intel 8080 download programs, see dl/Out
Failed for Intel 8080 EM library, see libem/Out
Failed for Intel 8080 etext,edata,end library, see libend/Out
Failed for Intel 8080 MON library, see libmon/Out
Failed for Intel 8080 system call library, see libsys/Out
Failed for i80 Modula-2 libraries, see libm2/Out
Failed for i80 Pascal libraries, see libpc/Out
Failed for i80 Basic libraries, see libbc/Out
Failed for i80 Occam libraries, see liboc/Out
Failed for i80 ANSI-C libraries, see libcc.ansi/Out
Failed for i80 C libraries, see libcc/Out
Failed for Intel 8080 support
Failed for em22 etext,edata,end library, see libend/Out
Failed for em22 system call library, see libsys/Out
Failed for em22 Modula-2 libraries, see libm2/Out
Failed for em22 Pascal libraries, see libpc/Out
Failed for em22 Basic libraries, see libbc/Out
Failed for em22 Occam libraries, see liboc/Out
Failed for em22 ANSI-C libraries, see libcc.ansi/Out
Failed for em22 C libraries, see libcc/Out
Failed for 2-2 Interpreter support
Failed for em24 etext,edata,end library, see libend/Out
Failed for em24 system call library, see libsys/Out
Failed for em24 Modula-2 libraries, see libm2/Out
Failed for em24 Pascal libraries, see libpc/Out
Failed for em24 Basic libraries, see libbc/Out
Failed for em24 Occam libraries, see liboc/Out
Failed for em24 ANSI-C libraries, see libcc.ansi/Out
Failed for em24 C libraries, see libcc/Out
Failed for 2-4 Interpreter support
Failed for 4-4 Interpreter etext,edata,end library, see libend/Out
Failed for 4-4 Interpreter system call library, see libsys/Out
Failed for em44 Modula-2 libraries, see libm2/Out
Failed for em44 Pascal libraries, see libpc/Out
Failed for em44 Basic libraries, see libbc/Out
Failed for em44 Occam libraries, see liboc/Out
Failed for em44 ANSI-C libraries, see libcc.ansi/Out
Failed for em44 C libraries, see libcc/Out
Failed for em44 Fortran libraries, see libf77/Out
Failed for 4-4 Interpreter support
Failed for Motorola 68000 assembler, see as/Out
Failed for Motorola 68000 2-4 backend, see ncg/Out
Failed for Motorola 68000 target optimizer, see top/Out
Motorola 68000 2-4 conversion program -- done
Failed for Motorola 68000 2-4 EM library, see libem/Out
Failed for Motorola 68000 2-4 etext,edata,end library, see libend/Out
Failed for Motorola 68000 2-4 floating point library, see libfp/Out
Failed for Motorola 68000 2-4 System library, see libsys/Out
Failed for m68k2 Modula-2 libraries, see libm2/Out
Failed for m68k2 Pascal libraries, see libpc/Out
Failed for m68k2 Basic libraries, see libbc/Out
Failed for m68k2 Occam libraries, see liboc/Out
Failed for m68k2 ANSI-C libraries, see libcc.ansi/Out
Failed for m68k2 C libraries, see libcc/Out
Failed for Motorola 68000 2-4 support
Failed for M68000 4-4 backend, see ncg/Out
Failed for M68000 4-4 EM library, see libem/Out
Failed for M68000 4-4 etext,edata,end library, see libend/Out
Failed for M68000 4-4 floating point library, see libfp/Out
Failed for M68000 4-4 system call library, see libsys/Out
Failed for m68k4 Modula-2 libraries, see libm2/Out
Failed for m68k4 Pascal libraries, see libpc/Out
Failed for m68k4 Basic libraries, see libbc/Out
Failed for m68k4 Occam libraries, see liboc/Out
Failed for m68k4 ANSI-C libraries, see libcc.ansi/Out
Failed for m68k4 C libraries, see libcc/Out
Failed for m68k4 Fortran libraries, see libf77/Out
Failed for Motorola 68000 4-4 support
Failed for NS16032 assembler, see as/Out
Failed for NS16032 backend, see ncg/Out
Failed for NS16032 EM library, see libem/Out
Failed for NS16032 etext,edata,end library, see libend/Out
Failed for NS16032 MON library, see libmon/Out
Failed for NS16032 system call library, see libsys/Out
Failed for ns Modula-2 libraries, see libm2/Out
Failed for ns Pascal libraries, see libpc/Out
Failed for ns Basic libraries, see libbc/Out
Failed for ns Occam libraries, see liboc/Out
Failed for ns ANSI-C libraries, see libcc.ansi/Out
Failed for ns C libraries, see libcc/Out
Failed for ns Fortran libraries, see libf77/Out
Failed for NS16032 support
Failed for PDP 11 assembler, see as/Out
Failed for PDP 11 backend, see cg/Out
Failed for PDP 11 target optimizer, see top/Out
PDP 11 conversion program -- done
Sorry, PDP 11 interpreter can only be made on pdp* systems
Failed for PDP 11 EM library, see libem/Out
Failed for PDP 11 etext,edata,end library, see libend/Out
Failed for PDP 11 systemcall library, see libsys/Out
Failed for pdp Modula-2 libraries, see libm2/Out
Failed for pdp Pascal libraries, see libpc/Out
Failed for pdp Basic libraries, see libbc/Out
Failed for pdp Occam libraries, see liboc/Out
Failed for pdp ANSI-C libraries, see libcc.ansi/Out
Failed for pdp C libraries, see libcc/Out
Failed for PDP 11 support
Ack.out --> PMDS a.out format conversion program -- done
Failed for PMDS systemcall library, see libsys/Out
Failed for pmds Modula-2 libraries, see libm2/Out
Failed for pmds Pascal libraries, see libpc/Out
Failed for pmds Basic libraries, see libbc/Out
Failed for pmds Occam libraries, see liboc/Out
Failed for pmds ANSI-C libraries, see libcc.ansi/Out
Failed for pmds C libraries, see libcc/Out
Failed for PMDS support
Failed for PMDS 4-4 system call library, see libsys/Out
Failed for pmds4 Modula-2 libraries, see libm2/Out
Failed for pmds4 Pascal libraries, see libpc/Out
Failed for pmds4 Basic libraries, see libbc/Out
Failed for pmds4 Occam libraries, see liboc/Out
Failed for pmds4 ANSI-C libraries, see libcc.ansi/Out
Failed for pmds4 C libraries, see libcc/Out
Failed for pmds4 Fortran libraries, see libf77/Out
Failed for PMDS 4/4 support
Failed for Signetics 2650 assembler, see as/Out
Failed for Signetics 2650 support
Failed for Vax 4-4 assembler, see as/Out
Failed for Vax 4-4 backend, see cg/Out
Failed for Vax target optimizer, see top/Out
ACK a.out to VAX a.out conversion program -- done
Sorry, Vax code expander library can only be made on vax* systems
Failed for Vax 4-4 EM library, see libem/Out
Sorry, Vax 4-4 code expander support library can only be made on vax* syste=
ms
Failed for Vax 4-4 debugger support library, see libdb/Out
Failed for Vax 4-4 etext,edata,end library, see libend/Out
Failed for Vax 4-4 systemcall interface, see libbsd4_2/Out
Failed for vax4 Modula-2 libraries, see libm2/Out
Failed for vax4 Pascal libraries, see libpc/Out
Failed for vax4 Basic libraries, see libbc/Out
Failed for vax4 Occam libraries, see liboc/Out
Failed for vax4 ANSI-C libraries, see libcc.ansi/Out
Failed for vax4 C libraries, see libcc/Out
Failed for vax4 Fortran libraries, see libf77/Out
Failed for Vax 4-4 support
Failed for M68020 assembler, see as/Out
Failed for M68020 backend, see ncg/Out
Failed for M68020 target optimizer, see top/Out
Sorry, M68020 code expander library can only be made on m68020|sun3 systems
Failed for M68020 EM library, see libem/Out
Failed for M68020 etext,edata,end library, see libend/Out
Failed for M68020 floating point library, see libfp/Out
Failed for M68020 debugger support library, see libdb/Out
Failed for M68020 system call library, see libsys/Out
M68020 VME131 System V/68 R2V2.1 conversion -- done
Failed for m68020 Modula-2 libraries, see libm2/Out
Failed for m68020 Pascal libraries, see libpc/Out
Failed for m68020 Basic libraries, see libbc/Out
Failed for m68020 Occam libraries, see liboc/Out
Failed for m68020 ANSI-C libraries, see libcc.ansi/Out
Failed for m68020 C libraries, see libcc/Out
Failed for m68020 Fortran libraries, see libf77/Out
Failed for M68020 System V/68 support
Ack.out --> Sun 3 M68020 a.out format conversion program -- done
Failed for Sun 3 M68020 systemcall library, see libsys/Out
Sorry, Sun3 code expander library can only be made on sun3 systems
Sorry, Sun3 code expander support library can only be made on sun3 systems
Failed for sun3 Modula-2 libraries, see libm2/Out
Failed for sun3 Pascal libraries, see libpc/Out
Failed for sun3 Basic libraries, see libbc/Out
Failed for sun3 Occam libraries, see liboc/Out
Failed for sun3 ANSI-C libraries, see libcc.ansi/Out
Failed for sun3 C libraries, see libcc/Out
Failed for sun3 Fortran libraries, see libf77/Out
Failed for Sun 3 M68020 support
Sorry, Sun 4 SPARC SunOs 4 support can only be made on
sparc|sparc_solaris systems
Sorry, Sun 4 SPARC Solaris support can only be made on sparc_solaris system=
s
Ack.out --> Sun 2 M68000 a.out format conversion program -- done
Failed for Sun 2 M68000 systemcall library, see libsys/Out
Failed for sun2 Modula-2 libraries, see libm2/Out
Failed for sun2 Pascal libraries, see libpc/Out
Failed for sun2 Basic libraries, see libbc/Out
Failed for sun2 Occam libraries, see liboc/Out
Failed for sun2 ANSI-C libraries, see libcc.ansi/Out
Failed for sun2 C libraries, see libcc/Out
Failed for sun2 Fortran libraries, see libf77/Out
Failed for Sun 2 M68000 support
Failed for Mantra M68000 4-4 system call library, see libsys/Out
Mantra M68000 4-4 conversion program -- done
Failed for mantra Modula-2 libraries, see libm2/Out
Failed for mantra Pascal libraries, see libpc/Out
Failed for mantra Basic libraries, see libbc/Out
Failed for mantra Occam libraries, see liboc/Out
Failed for mantra ANSI-C libraries, see libcc.ansi/Out
Failed for mantra C libraries, see libcc/Out
Failed for mantra Fortran libraries, see libf77/Out
Failed for Mantra M68000 System V.0 support
Failed for Minix PC systemcall library, see libsys/Out
Failed for minix Modula-2 libraries, see libm2/Out
Failed for minix Pascal libraries, see libpc/Out
Failed for minix Basic libraries, see libbc/Out
Failed for minix Occam libraries, see liboc/Out
Failed for minix ANSI-C libraries, see libcc.ansi/Out
Failed for minix C libraries, see libcc/Out
Failed for PC Minix support
Failed for Atari ST Minix systemcall library, see libsys/Out
Atari ST Minix conversion program from ack.out --> Minix ST a.out -- done
Failed for minixST Modula-2 libraries, see libm2/Out
Failed for minixST Pascal libraries, see libpc/Out
Failed for minixST Basic libraries, see libbc/Out
Failed for minixST Occam libraries, see liboc/Out
Failed for minixST ANSI-C libraries, see libcc.ansi/Out
Failed for minixST C libraries, see libcc/Out
Failed for Atari ST Minix support
Failed for Z80 assembler, see as/Out
Failed for Z80 code generator, see cg/Out
Failed for Z80 EM library, see libem/Out
Failed for Z80 MON library, see libmon/Out
Failed for Z80 system call library, see libsys/Out
Failed for Z80 end,etext,edata library, see libend/Out
Failed for z80 Modula-2 libraries, see libm2/Out
Failed for z80 Pascal libraries, see libpc/Out
Failed for z80 Basic libraries, see libbc/Out
Failed for z80 Occam libraries, see liboc/Out
Failed for z80 ANSI-C libraries, see libcc.ansi/Out
Failed for z80 C libraries, see libcc/Out
Failed for Z80 support
Failed for Zilog Z8000 assembler, see as/Out
Failed for Zilog Z8000 backend, see cg/Out
Zilog Z8000 EM library -- done
Zilog Z8000 MON library -- done
Failed for Zilog Z8000 system call library, see libsys/Out
Zilog Z8000 end,etext,edata library -- done
Failed for z8000 Modula-2 libraries, see libm2/Out
Failed for z8000 Pascal libraries, see libpc/Out
Failed for z8000 Basic libraries, see libbc/Out
Failed for z8000 Occam libraries, see liboc/Out
Failed for z8000 ANSI-C libraries, see libcc.ansi/Out
Failed for z8000 C libraries, see libcc/Out
Failed for Zilog Z8000 support
Failed for ARM backend, see ncg/Out
Failed for ARM assembler, see as/Out
Failed for ARM target optimiser, see top/Out
ARM cv program -- done
ARM EM library -- done
ARM etext,edata,end library -- done
ARM RiscOs library -- done
Failed for ARM system call library, see libsys/Out
Failed for ARM floating point library, see libfp/Out
Failed for arm Modula-2 libraries, see libm2/Out
Failed for arm Pascal libraries, see libpc/Out
Failed for arm Basic libraries, see libbc/Out
Failed for arm Occam libraries, see liboc/Out
Failed for arm ANSI-C libraries, see libcc.ansi/Out
Failed for arm C libraries, see libcc/Out
Failed for arm Fortran libraries, see libf77/Out
Failed for Acorn Archimedes support
Failed for Documentation, see doc/Out

I interrupted compilation because of too many errors...

Any suggestations howto compile ack to produce code for 6500 ???
Thanks

Re: [Tack-devel] Which version of Occam?

[David G. <dg...@co...>]

On Monday 16 January 2006 14:35, Kai Koehne wrote:
[...]
> the documentation about the occam-frontend is quite spare. Do you know
> which version of Occam (and to what extend) is supported?

No idea, I'm afraid. I've never actually used that bit, other than brief=20
playing to make sure it worked, and the last time I touched Occam was on an=
=20
ancient and ghastly Transputer setup at university, too many years ago.

How many variations of the language are there?

=2D-=20
+- David Given --McQ-+ "...it's not that well-designed GUI's are rare,
|  dg...@co...    | it's just that the three-armed users GUI's are
| (dg...@ta...) | designed for are rare." --- Mike Uhl on a.f.c
+- www.cowlark.com --+=20

[Tack-devel] Which version of Occam?

[Kai K. <ka...@ko...>]

Hi,

the documentation about the occam-frontend is quite spare. Do you know 
which version of Occam (and to what extend) is supported?

Regards,

Kai Koehne

[Tack-devel] ack testing

[Piergiorgio B. <pb...@e-...>]

Hi David, nice to see the patch useful.... ;)
in the meantime i've started to play with ack. It's not clear for me how to test it in the "em44" modality. Could you suggest something "quick-and-dirty" ??

Thanks
Piergiorgio

--  
Piergiorgio Betti <pie...@e-...>
E-TECH s.r.l.




-----

Re: [Tack-devel] fix for gcc 4.0.1

[David G. <dg...@co...>]

On Wednesday 10 August 2005 20:15, Piergiorgio Betti wrote:
]...]
> I've downloaded ack and tried it. Compilation produces some error ("Failed
> ...") due the the fact that i'm using gcc version 4.0.1 (the distro is
> Mandriva LE 2005 by the way). Included there is a patch that fix the
> procedure and also contain the conversion of mach/i80/dl/nascom.c to
> termios standard, hoping it's useful.

Woohoo! Code! Thanks very much! Particularly for the termios stuff; I have=
=20
absolutely no idea how that works, which is why I haven't tackled fixing it=
=20
yet. I'm very grateful and I've checked it in pronto.

(The other fixes will hang on a bit in my work tree first; they partially=20
overlap with things I'm working --- very slowly --- on. I'll have to get th=
is=20
all sorted before I can commit to CVS.)

Ta!

=2D-=20
"Curses! Foiled by the chilled dairy treats of righteousness!" --- Earthwor=
m=20
Jim (evil)

[Tack-devel] patch 4.0.1

[Piergiorgio B. <pb...@e-...>]

I forgot to say that the compilation is fixed but ACK itself is still untested (by me) ...

Piergiorgio

--  
Piergiorgio Betti <pie...@e-...>
E-TECH s.r.l.




-----

[Tack-devel] fix for gcc 4.0.1

[Piergiorgio B. <pb...@e-...>]

Hi folks!
I've downloaded ack and tried it. Compilation produces some error ("Failed ...") due the the fact that i'm using gcc version 4.0.1 (the distro is Mandriva LE 2005 by the way).
Included there is a patch that fix the procedure and also contain the conversion of mach/i80/dl/nascom.c to termios standard, hoping it's useful.

Regards!
Piergiorgio


--  
Piergiorgio Betti <pie...@e-...>
E-TECH s.r.l.




-----

[Tack-devel] Solaris, make, and lint

[David G. <dg...@co...>]

Sorry about the lack of updates or progress information --- I've had a majo=
r=20
attack of builders, and as a result am spending most of my time in a state =
of=20
acute exhaustion. My creative juices are pretty much nil right now. (Bet=20
that's more than you wanted to know.)

However, I have had time to have a brief play with getting the ACK to build=
 on=20
Sparc Solaris, using one of SourceForge's compile farm machines. And there'=
s=20
something really weird going on.

Whenever I try to run the makefiles, the compilation fails because it tried=
 to=20
run lint (the lintlib target for util/data, in fact), and SF's machine does=
=20
not have lint installed. That seems fair enough.

However, my home machine doesn't have lint installed either, and it works f=
ine=20
there. I tried to do some digging to find out where the lintlib target was=
=20
being invoked, and came up blank. In fact, the word 'lintlib' appears=20
*nowhere* in the entire codebase unless it's followed by a colon.

I managed to prevent it from trying to run lint by preventing the ACK from=
=20
defining the LINT make variable, but I don't know why this works.

My question is: does Solaris make have some strange feature where it'll=20
automatically try to run lint if it thinks it's available?

(Does anyone use lint these days?)

=2D-=20
"Curses! Foiled by the chilled dairy treats of righteousness!" --- Earthwor=
m=20
Jim (evil)

Re: [Tack-devel] K&R vs ANSI C

[Daters R Us A. A. <ad...@da...>]

On 25 Jul 2005, at 09:58, David Given wrote:

> On Monday 25 July 2005 09:38, Simon Waite wrote:
>> I think half the point of ACK is that it can be compiled by itself.
>
> Well, the ACK does come with an ANSI C compiler, and I've had it 
> compiling on
> Minix, which uses a (ANSI-only) subset of the ACK as it's native cc.
>
> Currently the ACK won't generate runnable code for any platform that 
> can run
> the ACK, with the possible exception of SPARC Solaris (I need to try 
> that),
> so I haven't been able to try compiling the ACK with the *actual* ACK.
>
>> My last problems with OSX compiling seems to stem from it not being
>> able to link properly which is beyond my abilities to debug as I'm not
>> that up to speed with developing on OSX (though I guess if I used a 
>> gnu
>> binutils toolchain rather than using OSXs native mach-o linker I'd 
>> have
>> better luck)
>
> I've had a brief play on SourceForge's OSX compiler box. I have *no 
> idea*
> what's going on --- it's complaining about missing symbols which are 
> quite
> clearly in the libraries. Changing the library order doesn't help

Yup, the only thing I didn't try is manually link to the object files.


> Does OSX require symbols to be explicitly exported in the source 
> files, do you
> know?

Nope, and the symbols are right there in the libraries! it's probably a 
bug in the linker :(

> -- 
> +- David Given --McQ-+ "I don't like the thought of her hearing what 
> I'm
> |  dg...@co...    | thinking." "*No-one* likes the thought of 
> hearing
> | (dg...@ta...) | what you're thinking." --- Firefly, _Objects in
> +- www.cowlark.com --+ Space_

Re: [Tack-devel] K&R vs ANSI C

[David G. <dg...@co...>]

On Monday 25 July 2005 09:38, Simon Waite wrote:
> I think half the point of ACK is that it can be compiled by itself.

Well, the ACK does come with an ANSI C compiler, and I've had it compiling =
on=20
Minix, which uses a (ANSI-only) subset of the ACK as it's native cc.=20

Currently the ACK won't generate runnable code for any platform that can ru=
n=20
the ACK, with the possible exception of SPARC Solaris (I need to try that),=
=20
so I haven't been able to try compiling the ACK with the *actual* ACK.

> My last problems with OSX compiling seems to stem from it not being
> able to link properly which is beyond my abilities to debug as I'm not
> that up to speed with developing on OSX (though I guess if I used a gnu
> binutils toolchain rather than using OSXs native mach-o linker I'd have
> better luck)

I've had a brief play on SourceForge's OSX compiler box. I have *no idea*=20
what's going on --- it's complaining about missing symbols which are quite=
=20
clearly in the libraries. Changing the library order doesn't help.

Does OSX require symbols to be explicitly exported in the source files, do =
you=20
know?

=2D-=20
+- David Given --McQ-+ "I don't like the thought of her hearing what I'm
|  dg...@co...    | thinking." "*No-one* likes the thought of hearing
| (dg...@ta...) | what you're thinking." --- Firefly, _Objects in
+- www.cowlark.com --+ Space_

Re: [Tack-devel] K&R vs ANSI C

[Simon W. <ta...@rm...>]

I think half the point of ACK is that it can be compiled by itself.

My last problems with OSX compiling seems to stem from it not being 
able to link properly which is beyond my abilities to debug as I'm not 
that up to speed with developing on OSX (though I guess if I used a gnu 
binutils toolchain rather than using OSXs native mach-o linker I'd have 
better luck)

ho hum! :o)


On 23 Jul 2005, at 12:36, David Given wrote:

> Is there any real requirement these days for the ACK to be written 
> using K&R
> C?
>
> I ask because K&R C has a lot of very fuzzy semantics when it comes to
> imported and exported symbols, not to mention the lack of anything at 
> all
> resembling type safety, that I suspect is the cause of the
> not-compiling-on-OSX problem. Just clearing out all the more repulsive
> K&Risms (like arbitrarily importing symbols where you want them rather 
> than
> using headers) should sort a lot of this.
>
> I don't believe there are any K&R-only systems these days; every 
> platform I've
> tried the ACK on so far compiles it with an ANSI compiler, even Minix.
>
> -- 
> "Curses! Foiled by the chilled dairy treats of righteousness!" --- 
> Earthworm
> Jim (evil)

[Tack-devel] K&R vs ANSI C

[David G. <dg...@co...>]

Is there any real requirement these days for the ACK to be written using K&=
R=20
C?

I ask because K&R C has a lot of very fuzzy semantics when it comes to=20
imported and exported symbols, not to mention the lack of anything at all=20
resembling type safety, that I suspect is the cause of the=20
not-compiling-on-OSX problem. Just clearing out all the more repulsive=20
K&Risms (like arbitrarily importing symbols where you want them rather than=
=20
using headers) should sort a lot of this.

I don't believe there are any K&R-only systems these days; every platform I=
've=20
tried the ACK on so far compiles it with an ANSI compiler, even Minix.

=2D-=20
"Curses! Foiled by the chilled dairy treats of righteousness!" --- Earthwor=
m=20
Jim (evil)

Re: [Tack-devel] test

[Simon W. <ta...@rm...>]

On 26 Jun 2005, at 11:44, David Given wrote:

> On Sunday 26 June 2005 04:31, Simon Waite wrote:
>
[snip]
> Meh. Could you send me a copy of <confdir>/mach/$MACH/cg/Out, please?
I found the "problem" and have a suggested fix below:
==================
ack-5.6/util/ncgg/
==================

problem:
   ld: Undefined symbols: _end

defined in:
   subr.c line 282, function freevi(vip)

added: to main.c at line 15, before main()
   char *end; /* FIXME - intialise? */

notes:
   Possibly a quick an dirty hack to gain a NULL pointer for comparison.
   Lack of commenting makes it awkward for this developer to determine
   purpose of freevi() for testing.

   Above change allows compilation (under gcc 3.3 under OSX 10.3)



> [...]
>> this time the missing item is $BIN/lib.bin/em_opt
>
> Likewise, modules/src/em_opt/Out.

Nothing found :-( But I'm concentrating on getting it working on OSX 
first :-)



> Unfortunately, I only have Linux systems. I've already received one 
> set of bug
> reports about it not building with gcc 4, and hopefully I've sorted 
> most of
> that, but it needs lots more testing on different platforms...
>

I have a solaris machine, ppc linux, that could be pressganged into 
service, the sf compile farm, if it exists can be pressganged into 
service for other platforms.

Other fails...

Failed for Pascal frontend, see lang/pc/comp/Out
ld: Undefined symbols:
_free_scopelist
_free_withdesig
_new_scopelist
_new_withdesig
_new_lab
_free_forwtype
_free_node
_new_node
_new_type
_new_def
_new_paramlist
_new_forwtype
_new_scope
_free_def

Failed for Modula-2 frontend, see lang/m2/comp/Out
ld: Undefined symbols:
_new_real
_free_def
_free_paramlist
_free_type
_new_array
_new_enume
_new_subrange
_new_type
_new_def
_new_node
_new_paramlist
_free_node
_free_real

Failed for EM interpreter in C, see util/int/Out
cc -c -O -D_EM_WSIZE=4 -D_EM_PSIZE=4 -D__XXX__ 
-I/Users/simon/Projects/ACK/bin/h 
-I/Users/simon/Projects/ACK/bin/config -I. 
/Users/simon/Projects/ACK/ack-5.6/util/int/moncalls.c
util/int/moncalls.c: In function `moncall':
util/int/moncalls.c:747: error: invalid operands to binary /
util/int/moncalls.c:748: error: `daylight' undeclared (first use in 
this function)
util/int/moncalls.c:748: error: (Each undeclared identifier is reported 
only once
util/int/moncalls.c:748: error: for each function it appears in.)
make: *** [moncalls.o] Error 1

I'm going to have a look at the EM interpreter problem shortly, looks 
like I'm compiling for SYS_V instead of BSD_X (oops!)

Finally it seems march doesn't make bsd ar compatible archives either, 
more to look into later!

Re: [Tack-devel] test

[David G. <dg...@co...>]

On Sunday 26 June 2005 04:31, Simon Waite wrote:
[...]
> "\\000"  rather than "\000" at lines 18 and 25

Fixed. Ta.

[...]
> Pascal, Modula-2, front ends fail to build (linker error)
> EM interpreter also fails to build for similar reasons
> Most, if not all, Libraries fail to build - not looked at all of them
> but from the arm target it looks like $BIN/lib.bin/$MACH/cg hasn't been
> built

Meh. Could you send me a copy of <confdir>/mach/$MACH/cg/Out, please?

[...]
> this time the missing item is $BIN/lib.bin/em_opt

Likewise, modules/src/em_opt/Out.

Unfortunately, I only have Linux systems. I've already received one set of bug 
reports about it not building with gcc 4, and hopefully I've sorted most of 
that, but it needs lots more testing on different platforms...

-- 
"Curses! Foiled by the chilled dairy treats of righteousness!" --- Earthworm 
Jim (evil)

Re: [Tack-devel] test

[Simon W. <ta...@rm...>]

Now with the correct email address.

Trying to compile on OSX Machine (Panther) with gcc 3.3 so far I've 
found and squashed one bug.

When util/opt/pop_push.awk is processed with bsd awk instead of the 
following sequence

/* opcode */ '^@',

the following is output

/*opcode */ ',

Solution:

List gawk aliased as awk as a dependancy OR fix pop_push.awk to read

"\\000"  rather than "\000" at lines 18 and 25


Apart from that ...

Pascal, Modula-2, front ends fail to build (linker error)
EM interpreter also fails to build for similar reasons
Most, if not all, Libraries fail to build - not looked at all of them 
but from the arm target it looks like $BIN/lib.bin/$MACH/cg hasn't been 
built

It'll be quite interesting to get this built on something more than 
just linux, but a quick compile on a debian 3.0 system (yes I know, I 
should upgrade) appears to build everything, but fails on the libraries 
again, this time the missing item is $BIN/lib.bin/em_opt

Well thats enough for tonight !

Regards,

Simon

On 26 Jun 2005, at 03:29, Simon Waite wrote:

> test.

[Tack-devel] test

[Simon W. <ta...@rm...>]

test.

[Tack-devel] Test message.

[David G. <dg...@co...>]

-- 
"Curses! Foiled by the chilled dairy treats of righteousness!" --- Earthworm 
Jim (evil)