GNAT GPL 2009 for Mac OS X (Snow Leopard)

GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

I've just released a 32-bit build of GNAT GPL 2009 for Snow Leopard;
find it at http://sourceforge.net/projects/gnuada/files/ (navigate to
GNAT_GPL Mac OS X/2009-snow-leopard-i386).

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[(see below)]

On 22/12/2009 12:09, in article
af608388-8088-4779-a3ca-39275fd2c6cc@r24g2000yqd.googlegroups.com, "sjw"
<simon.j.wright@mac.com> wrote:

> I've just released a 32-bit build of GNAT GPL 2009 for Snow Leopard;
> find it at http://sourceforge.net/projects/gnuada/files/ (navigate to
> GNAT_GPL Mac OS X/2009-snow-leopard-i386).

Many thanks -- I'll try it ASAP.
Any chance of a 64-bit build?

-- 
Bill Findlay
<surname><forename> chez blueyonder.co.uk

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

On Dec 22, 2:23 pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
> On 22/12/2009 12:09, in article
> af608388-8088-4779-a3ca-39275fd2c...@r24g2000yqd.googlegroups.com, "sjw"
>
> <simon.j.wri...@mac.com> wrote:
> > I've just released a 32-bit build of GNAT GPL 2009 for Snow Leopard;
> > find it athttp://sourceforge.net/projects/gnuada/files/(navigate to
> > GNAT_GPL Mac OS X/2009-snow-leopard-i386).
>
> Many thanks -- I'll try it ASAP.
> Any chance of a 64-bit build?

Will have a go, but it's a cross-build so may take some getting my
head round!

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[(see below)]

On 22/12/2009 17:10, in article
5b0bf629-0935-414d-9a57-1757632840b8@d10g2000yqh.googlegroups.com, "sjw"
<simon.j.wright@mac.com> wrote:

> On Dec 22, 2:23�pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
>> On 22/12/2009 12:09, in article
>> af608388-8088-4779-a3ca-39275fd2c...@r24g2000yqd.googlegroups.com, "sjw"
>> 
>> <simon.j.wri...@mac.com> wrote:
>>> I've just released a 32-bit build of GNAT GPL 2009 for Snow Leopard;
>>> find it athttp://sourceforge.net/projects/gnuada/files/(navigate to
>>> GNAT_GPL Mac OS X/2009-snow-leopard-i386).
>> 
>> Many thanks -- I'll try it ASAP.
>> Any chance of a 64-bit build?
> 
> Will have a go, but it's a cross-build so may take some getting my
> head round!

Thanks. My two main projects run at 2.25 times the 32-bit speed in 64-bit
mode, so it is quite an issue for me.

This is what happened when I tried out the 32-bit version:

gnatmake -j2 -aI/Users/wf/emulation/Source -aO/Users/wf/emulation/MacOSX
-funwind-tables  -g -gnatfl05wawCwl -gnatoVa  -m32 -k -O1 EE9.adb    -largs
-lgcc_eh -bargs -Sev >komlog.ada

gcc -c -I./ -I/Users/wf/emulation/Source -funwind-tables -g -gnatfl05wawCwl
-gnatoVa -m32 -O1 -I- /Users/wf/emulation/Source/EE9.adb

dyld: Symbol not found: ___gmpn_bases
  Referenced from: 
/opt/gnat-gpl-2009/libexec/gcc/i386-apple-darwin10.2.0/4.3.4/gnat1
  Expected in: /usr/local/lib/libmpfr.1.dylib
 in /opt/gnat-gpl-2009/libexec/gcc/i386-apple-darwin10.2.0/4.3.4/gnat1
gcc: Internal error: Trace/BPT trap (program gnat1)
Please submit a full bug report to
<http://gcc.gnu.org/bugs.html>
with preprocessed source if appropriate.
gnatmake: "/Users/wf/emulation/Source/EE9.adb" compilation error

-- 
Bill Findlay
<surname><forename> chez blueyonder.co.uk

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

On Dec 22, 5:46 pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:

> This is what happened when I tried out the 32-bit version:
[...]
> dyld: Symbol not found: ___gmpn_bases
>   Referenced from:
> /opt/gnat-gpl-2009/libexec/gcc/i386-apple-darwin10.2.0/4.3.4/gnat1
>   Expected in: /usr/local/lib/libmpfr.1.dylib
>  in /opt/gnat-gpl-2009/libexec/gcc/i386-apple-darwin10.2.0/4.3.4/gnat1

Very sorry about that. Have uploaded a more appropriate libmpfr at the
same place as the compiler (seems to be taking its time to appear ...)
unpack /some/where and

  $ export DYLD_LIBRARY_PATH=/some/where

Am rebuilding without the dependency.

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

On Dec 22, 9:52 pm, sjw <simon.j.wri...@mac.com> wrote:
> On Dec 22, 5:46 pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
>
>
>
> > This is what happened when I tried out the 32-bit version:
> [...]
> > dyld: Symbol not found: ___gmpn_bases
> >   Referenced from:
> > /opt/gnat-gpl-2009/libexec/gcc/i386-apple-darwin10.2.0/4.3.4/gnat1
> >   Expected in: /usr/local/lib/libmpfr.1.dylib
> >  in /opt/gnat-gpl-2009/libexec/gcc/i386-apple-darwin10.2.0/4.3.4/gnat1
>
> Very sorry about that. Have uploaded a more appropriate libmpfr at the
> same place as the compiler (seems to be taking its time to appear ...)
> unpack /some/where and
>
>   $ export DYLD_LIBRARY_PATH=/some/where
>
> Am rebuilding without the dependency.

Have just uploaded gnat-gpl-2009-i386-apple-darwin10.2.0-1.tar.bz2,
without the dependency; will remove the offending files after a day or
two.

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[(see below)]

On 22/12/2009 23:37, in article
96854a61-2032-49b1-a8b2-8a5aee181701@22g2000yqr.googlegroups.com, "sjw"
<simon.j.wright@mac.com> wrote:
 
> Have just uploaded gnat-gpl-2009-i386-apple-darwin10.2.0-1.tar.bz2,
> without the dependency; will remove the offending files after a day or
> two.

Unfortunately, it goes very badly wrong at -O1 optimization level on another
example:

----------------------------------------------------------------------------
package test is
   type i64 is mod 2**64;
   for  i64'Size use 64;
   subtype a_virtual_address is i64;
   subtype an_offset is a_virtual_address;

   function fetch_byte_from (
         location : i64;
         position : an_offset )
   return i64;
end test;
--
package body test is
   type i08 is mod 2**8;
   for  i08'Size use 8;
   type an_i08_array is array (a_virtual_address range <>) of i08;
   for  an_i08_array'Component_Size use i08'Size;
   type an_i08_group is new an_i08_array (0..7);
   for  an_i08_group'Size use i64'Size;
   pragma Convention (C, an_i08_group);

   function fetch_byte_from (
         location : i64;
         position : an_offset )
   return i64 is
      the_location_as_i08s : an_i08_group;
      for the_location_as_i08s'Address use location'Address;
      pragma Import (Ada, the_location_as_i08s);
   begin
      return i64(the_location_as_i08s
                  ((position and 2#111#) xor 2#111#));
   end fetch_byte_from;
end test;
----------------------------------------------------------------------------

thus:

+===========================GNAT BUG DETECTED==============================+
| GPL 2009 (20090519) (i386-apple-darwin10.2.0) GCC error:                 |
| in gnat_to_gnu_entity, at ada/gcc-interface/decl.c:312                   |
| Error detected at test.adb:14:7                                          |
...

This error message does not appear at -O3, but the generated object program
is non-functional, although that may be for a different reason, of course.

This code compiled and ran correctly under the Leopard compilers I was using
until I stupidly installed SL.

What on earth has happened for GNAT to regress so badly in the face of what
is often described as a tidying-up and performance release of OS X?

-- 
Bill Findlay
<surname><forename> chez blueyonder.co.uk

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

On Dec 23, 3:29 am, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
> On 22/12/2009 23:37, in article
> 96854a61-2032-49b1-a8b2-8a5aee181...@22g2000yqr.googlegroups.com, "sjw"
>
> <simon.j.wri...@mac.com> wrote:
> > Have just uploaded gnat-gpl-2009-i386-apple-darwin10.2.0-1.tar.bz2,
> > without the dependency; will remove the offending files after a day or
> > two.
>
> Unfortunately, it goes very badly wrong at -O1 optimization level on another
> example:
>
> --------------------------------------------------------------------------- -
> package test is
>    type i64 is mod 2**64;
>    for  i64'Size use 64;
>    subtype a_virtual_address is i64;
>    subtype an_offset is a_virtual_address;
>
>    function fetch_byte_from (
>          location : i64;
>          position : an_offset )
>    return i64;
> end test;
> --
> package body test is
>    type i08 is mod 2**8;
>    for  i08'Size use 8;
>    type an_i08_array is array (a_virtual_address range <>) of i08;
>    for  an_i08_array'Component_Size use i08'Size;
>    type an_i08_group is new an_i08_array (0..7);
>    for  an_i08_group'Size use i64'Size;
>    pragma Convention (C, an_i08_group);
>
>    function fetch_byte_from (
>          location : i64;
>          position : an_offset )
>    return i64 is
>       the_location_as_i08s : an_i08_group;
>       for the_location_as_i08s'Address use location'Address;
>       pragma Import (Ada, the_location_as_i08s);
>    begin
>       return i64(the_location_as_i08s
>                   ((position and 2#111#) xor 2#111#));
>    end fetch_byte_from;
> end test;
> --------------------------------------------------------------------------- -
>
> thus:
>
> +===========================GNAT BUG DETECTED==============================+
> | GPL 2009 (20090519) (i386-apple-darwin10.2.0) GCC error:                 |
> | in gnat_to_gnu_entity, at ada/gcc-interface/decl.c:312                   |
> | Error detected at test.adb:14:7                                          |
> ...
>
> This error message does not appear at -O3, but the generated object program
> is non-functional, although that may be for a different reason, of course.
>
> This code compiled and ran correctly under the Leopard compilers I was using
> until I stupidly installed SL.
>
> What on earth has happened for GNAT to regress so badly in the face of what
> is often described as a tidying-up and performance release of OS X?

I don't know either, because here it compiles without complaint at -O
{0,1,2,3}!

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[(see below)]

On 23/12/2009 14:26, in article
68e6e09a-9f02-4fd4-a87c-d1c43dd104e1@27g2000yqr.googlegroups.com, "sjw"
<simon.j.wright@mac.com> wrote:

> On Dec 23, 3:29�am, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
>> On 22/12/2009 23:37, in article
>> 96854a61-2032-49b1-a8b2-8a5aee181...@22g2000yqr.googlegroups.com, "sjw"
>> 
>> <simon.j.wri...@mac.com> wrote:
>>> Have just uploaded gnat-gpl-2009-i386-apple-darwin10.2.0-1.tar.bz2,
>>> without the dependency; will remove the offending files after a day or
>>> two.
>> 
>> Unfortunately, it goes very badly wrong at -O1 optimization level on another
>> example:
... 
>> thus:
>> 
>> +===========================GNAT BUG DETECTED==============================+
>> | GPL 2009 (20090519) (i386-apple-darwin10.2.0) GCC error: � � � � � � � � |
>> | in gnat_to_gnu_entity, at ada/gcc-interface/decl.c:312 � � � � � � � � � |
>> | Error detected at test.adb:14:7 � � � � � � � � � � � � � � � � � � � � �|
>> ...
>> 
>> This error message does not appear at -O3, but the generated object program
>> is non-functional, although that may be for a different reason, of course.
>> 
>> This code compiled and ran correctly under the Leopard compilers I was using
>> until I stupidly installed SL.
>> 
>> What on earth has happened for GNAT to regress so badly in the face of what
>> is often described as a tidying-up and performance release of OS X?
> 
> I don't know either, because here it compiles without complaint at -O
> {0,1,2,3}!

Hmm. Interesting ...
What could account for the difference?
Do you have any other versions of GNAT installed in that testing
environment?

-- 
Bill Findlay
<surname><forename> chez blueyonder.co.uk

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

On Dec 23, 5:18 pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
> On 23/12/2009 14:26, in article
> 68e6e09a-9f02-4fd4-a87c-d1c43dd10...@27g2000yqr.googlegroups.com, "sjw"
> <simon.j.wri...@mac.com> wrote:
> > On Dec 23, 3:29 am, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
[...]
> >> What on earth has happened for GNAT to regress so badly in the face of what
> >> is often described as a tidying-up and performance release of OS X?
>
> > I don't know either, because here it compiles without complaint at -O
> > {0,1,2,3}!
>
> Hmm. Interesting ...
> What could account for the difference?
> Do you have any other versions of GNAT installed in that testing
> environment?

Several! though I'd be surprised if any of them got invoked. Will
check further.

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[(see below)]

On 23/12/2009 23:44, in article
eeaf5cb6-a582-4c90-a494-bd17142600a8@v25g2000yqk.googlegroups.com, "sjw"
<simon.j.wright@mac.com> wrote:

> On Dec 23, 5:18�pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
>> On 23/12/2009 14:26, in article
>> 68e6e09a-9f02-4fd4-a87c-d1c43dd10...@27g2000yqr.googlegroups.com, "sjw"
>> <simon.j.wri...@mac.com> wrote:
>>> On Dec 23, 3:29�am, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
> [...]
>>>> What on earth has happened for GNAT to regress so badly in the face of what
>>>> is often described as a tidying-up and performance release of OS X?
>> 
>>> I don't know either, because here it compiles without complaint at -O
>>> {0,1,2,3}!
>> 
>> Hmm. Interesting ...
>> What could account for the difference?
>> Do you have any other versions of GNAT installed in that testing
>> environment?
> 
> Several! though I'd be surprised if any of them got invoked. Will
> check further.

What I had in mind was interference between libraries, etc.

-- 
Bill Findlay
<surname><forename> chez blueyonder.co.uk

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

On Dec 24, 12:01 am, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
> On 23/12/2009 23:44, in article
> eeaf5cb6-a582-4c90-a494-bd1714260...@v25g2000yqk.googlegroups.com, "sjw"
> <simon.j.wri...@mac.com> wrote:
> > On Dec 23, 5:18 pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
> >> On 23/12/2009 14:26, in article
[...]
> >> Do you have any other versions of GNAT installed in that testing
> >> environment?
>
> > Several! though I'd be surprised if any of them got invoked. Will
> > check further.
>
> What I had in mind was interference between libraries, etc.

Each compiler I have installed lives in a different top-level
directory. I set my path using

  $ export PATH=/opt/gnat-gpl-2009/bin:/usr/bin:/bin:/usr/sbin:/sbin

(ie, the 32-bit Snow Leopard compiler I just released on SF); now I
can compile your example without trouble. Did you use more options?

nidhoggr:findlay simon$ gnatmake -f test
gcc -c test.adb
nidhoggr:findlay simon$ gnatmake -f test -O0
gcc -c -O0 test.adb
nidhoggr:findlay simon$ gnatmake -f test -O1
gcc -c -O1 test.adb
nidhoggr:findlay simon$ gnatmake -f test -O2
gcc -c -O2 test.adb
nidhoggr:findlay simon$ gnatmake -f test -O3
gcc -c -O3 test.adb
nidhoggr:findlay

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[John B. Matthews]

In article 
<f0b47208-95f0-48ef-8184-9f36f2a2f6bb@a32g2000yqm.googlegroups.com>,
 sjw <simon.j.wright@mac.com> wrote:

> Each compiler I have installed lives in a different top-level
> directory. I set my path using
> 
>   $ export PATH=/opt/gnat-gpl-2009/bin:/usr/bin:/bin:/usr/sbin:/sbin

FWIW, here's a bash shell script I use to show the current PATH and the 
alternative. The checkPath() and removePath() functions could be used in 
managing multiple versions:

<http://home.roadrunner.com/~jbmatthews/gtk/lady.html#ada.sh>

-- 
John B. Matthews
trashgod at gmail dot com
<http://sites.google.com/site/drjohnbmatthews>

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

On Dec 24, 3:31 pm, "John B. Matthews" <nos...@nospam.invalid> wrote:
> In article
> <f0b47208-95f0-48ef-8184-9f36f2a2f...@a32g2000yqm.googlegroups.com>,
>
>  sjw <simon.j.wri...@mac.com> wrote:
> > Each compiler I have installed lives in a different top-level
> > directory. I set my path using
>
> >   $ export PATH=/opt/gnat-gpl-2009/bin:/usr/bin:/bin:/usr/sbin:/sbin
>
> FWIW, here's a bash shell script I use to show the current PATH and the
> alternative. The checkPath() and removePath() functions could be used in
> managing multiple versions:
>
> <http://home.roadrunner.com/~jbmatthews/gtk/lady.html#ada.sh>

Thanks for that.

I was meaning to reply specifically to Bill's query about possible
clashes caused by multiple compiler installations.

For normal work, I use a script gnatfe - https://sourceforge.net/projects/gnuada/files/,
navigate to gnatfe.

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[(see below)]

On 24/12/2009 17:24, in article
65af5bac-a490-4e08-b8e0-7d36030b780c@m25g2000yqc.googlegroups.com, "sjw"
<simon.j.wright@mac.com> wrote:

> On Dec 24, 3:31�pm, "John B. Matthews" <nos...@nospam.invalid> wrote:
>> In article
>> <f0b47208-95f0-48ef-8184-9f36f2a2f...@a32g2000yqm.googlegroups.com>,
>> 
>> �sjw <simon.j.wri...@mac.com> wrote:
>>> Each compiler I have installed lives in a different top-level
>>> directory. I set my path using
>> 
>>> � $ export PATH=/opt/gnat-gpl-2009/bin:/usr/bin:/bin:/usr/sbin:/sbin
>> 
>> FWIW, here's a bash shell script I use to show the current PATH and the
>> alternative. The checkPath() and removePath() functions could be used in
>> managing multiple versions:
>> 
>> <http://home.roadrunner.com/~jbmatthews/gtk/lady.html#ada.sh>
> 
> Thanks for that.
> 
> I was meaning to reply specifically to Bill's query about possible
> clashes caused by multiple compiler installations.

Well, I have deleted every copy of GNAT I could find, and downloaded
gnat-gpl-2009-i386-apple-darwin10.2.0-1.tar.bz2 afresh.

My .profile sets the PATH to:

PATH=/Users/wf/a2z/MPS/Cstage2:/Users/wf/ex-GU/HOCUS/sysdev/bin:/Users/wf/Pr
ograms:/Users/wf/KDF9/emulation/testing:/Users/wf/a2z/Testing:/Users/wf/Prog
rams/bin:/opt/gnat-gpl-2009/bin:/opt/local/sbin:/opt/local/bin:/usr/X11R6/bi
n:/usr/share:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin

which gnat gives /opt/gnat-gpl-2009/bin/gnat
which gcc gives /opt/gnat-gpl-2009/bin/gcc

The error I reported on the "test" package has not gone away. 8-(

-- 
Bill Findlay
<surname><forename> chez blueyonder.co.uk

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[(see below)]

On 24/12/2009 19:20, in article C7596F6A.12FADA%yaldnif.w@blueyonder.co.uk,
"(see below)" <yaldnif.w@blueyonder.co.uk> wrote:

> 
> Well, I have deleted every copy of GNAT I could find, and downloaded
> gnat-gpl-2009-i386-apple-darwin10.2.0-1.tar.bz2 afresh.
> 
> My .profile sets the PATH to:
> 
> PATH=/Users/wf/a2z/MPS/Cstage2:/Users/wf/ex-GU/HOCUS/sysdev/bin:/Users/wf/Pr
> ograms:/Users/wf/KDF9/emulation/testing:/Users/wf/a2z/Testing:/Users/wf/Prog
> rams/bin:/opt/gnat-gpl-2009/bin:/opt/local/sbin:/opt/local/bin:/usr/X11R6/bi
> n:/usr/share:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin
> 
> which gnat gives /opt/gnat-gpl-2009/bin/gnat
> which gcc gives /opt/gnat-gpl-2009/bin/gcc
> 
> The error I reported on the "test" package has not gone away. 8-(
  |||||||||
I.e., the compiler failure at -O1 has not gone away.
The malfunction at -O3 was my fault, and the optimised module does work.

-- 
Bill Findlay
<surname><forename> chez blueyonder.co.uk

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

On Dec 25, 4:30 am, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
> On 24/12/2009 19:20, in article C7596F6A.12FADA%yaldni...@blueyonder.co.uk,
> "(see below)" <yaldni...@blueyonder.co.uk> wrote:
[...]
> I.e., the compiler failure at -O1 has not gone away.
> The malfunction at -O3 was my fault, and the optimised module does work.

The verbose results of my compilation are

$ GNAT_PREFIX=/opt/gnat-gpl-2009 gnatmake -f test -O1 -cargs -v
gcc -c -O1 -v test.adb
Using built-in specs.
Target: i386-apple-darwin10.2.0
Configured with: ../gcc-43-src/configure --prefix=/opt/gnat-gpl-2009 --
with-gmp=/opt/gnu --with-mpfr=/opt/gnu --disable-multilib --enable-
languages=c,ada --disable-libada
Thread model: posix
gcc version 4.3.4 20090511 for GNAT GPL 2009 (20090511) (GCC)
COLLECT_GCC_OPTIONS='-mmacosx-version-min=10.6.2' '-gnatea' '-c' '-O1'
'-v' '-gnatez' '-mtune=generic'
 /opt/gnat-gpl-2009/libexec/gcc/i386-apple-darwin10.2.0/4.3.4/gnat1 -
quiet -dumpbase test.adb -O1 -gnatez -mmacosx-version-min=10.6.2 -
gnatea -gnatez -mtune=generic -fPIC test.adb -o /var/folders/it/
itCqFVrG2P8Zlk+F71KE3U+++TI/-Tmp-//cccOTx7p.s
COLLECT_GCC_OPTIONS='-mmacosx-version-min=10.6.2' '-gnatea' '-c' '-O1'
'-v' '-gnatez' '-mtune=generic'
 as -arch i386 -force_cpusubtype_ALL -o test.o /var/folders/it/
itCqFVrG2P8Zlk+F71KE3U+++TI/-Tmp-//cccOTx7p.s
COMPILER_PATH=/opt/gnat-gpl-2009/libexec/gcc/i386-apple-
darwin10.2.0/4.3.4/:/opt/gnat-gpl-2009/libexec/gcc/i386-apple-
darwin10.2.0/4.3.4/:/opt/gnat-gpl-2009/libexec/gcc/i386-apple-
darwin10.2.0/:/opt/gnat-gpl-2009/lib/gcc/i386-apple-
darwin10.2.0/4.3.4/:/opt/gnat-gpl-2009/lib/gcc/i386-apple-
darwin10.2.0/
LIBRARY_PATH=/opt/gnat-gpl-2009/lib/gcc/i386-apple-
darwin10.2.0/4.3.4/:/opt/gnat-gpl-2009/lib/gcc/i386-apple-
darwin10.2.0/4.3.4/../../../:/usr/lib/
COLLECT_GCC_OPTIONS='-mmacosx-version-min=10.6.2' '-gnatea' '-c' '-O1'
'-v' '-gnatez' '-mtune=generic'
COLLECT_GCC_OPTIONS='-mmacosx-version-min=10.6.2' '-gnatea' '-c' '-O1'
'-v' '-gnatez' '-mtune=generic'

Not sure if we'll be able to glean any significant differences ...

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

On Dec 24, 7:20 pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
> On 24/12/2009 17:24, in article
> 65af5bac-a490-4e08-b8e0-7d36030b7...@m25g2000yqc.googlegroups.com, "sjw"
> <simon.j.wri...@mac.com> wrote:
> > On Dec 24, 3:31 pm, "John B. Matthews" <nos...@nospam.invalid> wrote:
> >> In article
> >> <f0b47208-95f0-48ef-8184-9f36f2a2f...@a32g2000yqm.googlegroups.com>,
>
> >>  sjw <simon.j.wri...@mac.com> wrote:
> >>> Each compiler I have installed lives in a different top-level
> >>> directory. I set my path using
>
> >>>   $ export PATH=/opt/gnat-gpl-2009/bin:/usr/bin:/bin:/usr/sbin:/sbin
>
> >> FWIW, here's a bash shell script I use to show the current PATH and the
> >> alternative. The checkPath() and removePath() functions could be used in
> >> managing multiple versions:
>
> >> <http://home.roadrunner.com/~jbmatthews/gtk/lady.html#ada.sh>
>
> > Thanks for that.
>
> > I was meaning to reply specifically to Bill's query about possible
> > clashes caused by multiple compiler installations.
>
> Well, I have deleted every copy of GNAT I could find, and downloaded
> gnat-gpl-2009-i386-apple-darwin10.2.0-1.tar.bz2 afresh.
>
> My .profile sets the PATH to:
>
> PATH=/Users/wf/a2z/MPS/Cstage2:/Users/wf/ex-GU/HOCUS/sysdev/bin:/Users/wf/P r
> ograms:/Users/wf/KDF9/emulation/testing:/Users/wf/a2z/Testing:/Users/wf/Pro g
> rams/bin:/opt/gnat-gpl-2009/bin:/opt/local/sbin:/opt/local/bin:/usr/X11R6/b i
> n:/usr/share:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin
>
> which gnat gives /opt/gnat-gpl-2009/bin/gnat
> which gcc gives /opt/gnat-gpl-2009/bin/gcc
>
> The error I reported on the "test" package has not gone away. 8-(

I said 'gnatmake -f test -O1' etc, did you have any other options?

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[(see below)]

On 25/12/2009 17:32, in article
c7699bca-f565-4ae7-a3a1-3f5d4c4aec53@b2g2000yqi.googlegroups.com, "sjw"
<simon.j.wright@mac.com> wrote:

> On Dec 24, 7:20�pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
>> On 24/12/2009 17:24, in article
>> 65af5bac-a490-4e08-b8e0-7d36030b7...@m25g2000yqc.googlegroups.com, "sjw"
>> <simon.j.wri...@mac.com> wrote:
>>> On Dec 24, 3:31�pm, "John B. Matthews" <nos...@nospam.invalid> wrote:
>>>> In article
>>>> <f0b47208-95f0-48ef-8184-9f36f2a2f...@a32g2000yqm.googlegroups.com>,
>> 
>>>> �sjw <simon.j.wri...@mac.com> wrote:
>>>>> Each compiler I have installed lives in a different top-level
>>>>> directory. I set my path using
>> 
>>>>> � $ export PATH=/opt/gnat-gpl-2009/bin:/usr/bin:/bin:/usr/sbin:/sbin
>> 
>>>> FWIW, here's a bash shell script I use to show the current PATH and the
>>>> alternative. The checkPath() and removePath() functions could be used in
>>>> managing multiple versions:
>> 
>>>> <http://home.roadrunner.com/~jbmatthews/gtk/lady.html#ada.sh>
>> 
>>> Thanks for that.
>> 
>>> I was meaning to reply specifically to Bill's query about possible
>>> clashes caused by multiple compiler installations.
>> 
>> Well, I have deleted every copy of GNAT I could find, and downloaded
>> gnat-gpl-2009-i386-apple-darwin10.2.0-1.tar.bz2 afresh.
>> 
>> My .profile sets the PATH to:
>> 
>> PATH=/Users/wf/a2z/MPS/Cstage2:/Users/wf/ex-GU/HOCUS/sysdev/bin:/Users/wf/P r
>> ograms:/Users/wf/KDF9/emulation/testing:/Users/wf/a2z/Testing:/Users/wf/Pro g
>> rams/bin:/opt/gnat-gpl-2009/bin:/opt/local/sbin:/opt/local/bin:/usr/X11R6/b i
>> n:/usr/share:/usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin
>> 
>> which gnat gives /opt/gnat-gpl-2009/bin/gnat
>> which gcc gives /opt/gnat-gpl-2009/bin/gcc
>> 
>> The error I reported on the "test" package has not gone away. 8-(
> 
> I said 'gnatmake -f test -O1' etc, did you have any other options?

Many, many options. 8-)

gnatmake -j2 -I- -aI//Users/wf/a2z/Source/a -aI//Users/wf/Adalib/My/Bit_Sets
-aI//Users/wf/Adalib/My/Containers -aI//Users/wf/Adalib/My/Timing
-aI//Users/wf/Adalib/My/Distributor -aI//Users/wf/a2z/Source/shared
-aI//Users/wf/a2z/Source/a -aO//Users/wf/a2z/MacOSX -funwind-tables   -g
-gnatfl05wawCwl -gnatoVa -O1 test -largs -lgcc_eh -bargs -Sev >comlog.ada

-- 
Bill Findlay
<surname><forename> chez blueyonder.co.uk

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[Jerry]

On Dec 22, 8:29 pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
> On 22/12/2009 23:37, in article
> 96854a61-2032-49b1-a8b2-8a5aee181...@22g2000yqr.googlegroups.com, "sjw"
>
> <simon.j.wri...@mac.com> wrote:
> > Have just uploaded gnat-gpl-2009-i386-apple-darwin10.2.0-1.tar.bz2,
> > without the dependency; will remove the offending files after a day or
> > two.
>
> Unfortunately, it goes very badly wrong at -O1 optimization level on another
> example:
>
> --------------------------------------------------------------------------- -
> package test is
>    type i64 is mod 2**64;
>    for  i64'Size use 64;
>    subtype a_virtual_address is i64;
>    subtype an_offset is a_virtual_address;
>
>    function fetch_byte_from (
>          location : i64;
>          position : an_offset )
>    return i64;
> end test;
> --
> package body test is
>    type i08 is mod 2**8;
>    for  i08'Size use 8;
>    type an_i08_array is array (a_virtual_address range <>) of i08;
>    for  an_i08_array'Component_Size use i08'Size;
>    type an_i08_group is new an_i08_array (0..7);
>    for  an_i08_group'Size use i64'Size;
>    pragma Convention (C, an_i08_group);
>
>    function fetch_byte_from (
>          location : i64;
>          position : an_offset )
>    return i64 is
>       the_location_as_i08s : an_i08_group;
>       for the_location_as_i08s'Address use location'Address;
>       pragma Import (Ada, the_location_as_i08s);
>    begin
>       return i64(the_location_as_i08s
>                   ((position and 2#111#) xor 2#111#));
>    end fetch_byte_from;
> end test;
> --------------------------------------------------------------------------- -
>
> thus:
>
> +===========================GNAT BUG DETECTED==============================+
> | GPL 2009 (20090519) (i386-apple-darwin10.2.0) GCC error:                 |
> | in gnat_to_gnu_entity, at ada/gcc-interface/decl.c:312                   |
> | Error detected at test.adb:14:7                                          |
> ...
>
> This error message does not appear at -O3, but the generated object program
> is non-functional, although that may be for a different reason, of course.
>
> This code compiled and ran correctly under the Leopard compilers I was using
> until I stupidly installed SL.
>
> What on earth has happened for GNAT to regress so badly in the face of what
> is often described as a tidying-up and performance release of OS X?
>
> --
> Bill Findlay
> <surname><forename> chez blueyonder.co.uk

I thought OS X was officially supported by AdaCore. From the viewpoint
of a casual bystander, it looks like someone is asleep at the wheel.
This isn't support that I can believe in.

Jerry

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

On Dec 23, 9:16 pm, Jerry <lancebo...@qwest.net> wrote:
> On Dec 22, 8:29 pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
>
>
>
>
>
> > On 22/12/2009 23:37, in article
> > 96854a61-2032-49b1-a8b2-8a5aee181...@22g2000yqr.googlegroups.com, "sjw"
>
> > <simon.j.wri...@mac.com> wrote:
> > > Have just uploaded gnat-gpl-2009-i386-apple-darwin10.2.0-1.tar.bz2,
> > > without the dependency; will remove the offending files after a day or
> > > two.
>
> > Unfortunately, it goes very badly wrong at -O1 optimization level on another
> > example:
>
> > --------------------------------------------------------------------------- -
> > package test is
> >    type i64 is mod 2**64;
> >    for  i64'Size use 64;
> >    subtype a_virtual_address is i64;
> >    subtype an_offset is a_virtual_address;
>
> >    function fetch_byte_from (
> >          location : i64;
> >          position : an_offset )
> >    return i64;
> > end test;
> > --
> > package body test is
> >    type i08 is mod 2**8;
> >    for  i08'Size use 8;
> >    type an_i08_array is array (a_virtual_address range <>) of i08;
> >    for  an_i08_array'Component_Size use i08'Size;
> >    type an_i08_group is new an_i08_array (0..7);
> >    for  an_i08_group'Size use i64'Size;
> >    pragma Convention (C, an_i08_group);
>
> >    function fetch_byte_from (
> >          location : i64;
> >          position : an_offset )
> >    return i64 is
> >       the_location_as_i08s : an_i08_group;
> >       for the_location_as_i08s'Address use location'Address;
> >       pragma Import (Ada, the_location_as_i08s);
> >    begin
> >       return i64(the_location_as_i08s
> >                   ((position and 2#111#) xor 2#111#));
> >    end fetch_byte_from;
> > end test;
> > --------------------------------------------------------------------------- -
>
> > thus:
>
> > +===========================GNAT BUG DETECTED==============================+
> > | GPL 2009 (20090519) (i386-apple-darwin10.2.0) GCC error:                 |
> > | in gnat_to_gnu_entity, at ada/gcc-interface/decl.c:312                   |
> > | Error detected at test.adb:14:7                                          |
> > ...
>
> > This error message does not appear at -O3, but the generated object program
> > is non-functional, although that may be for a different reason, of course.
>
> > This code compiled and ran correctly under the Leopard compilers I was using
> > until I stupidly installed SL.
>
> > What on earth has happened for GNAT to regress so badly in the face of what
> > is often described as a tidying-up and performance release of OS X?
>
> > --
> > Bill Findlay
> > <surname><forename> chez blueyonder.co.uk
>
> I thought OS X was officially supported by AdaCore. From the viewpoint
> of a casual bystander, it looks like someone is asleep at the wheel.
> This isn't support that I can believe in.

Mac OS X is indeed supported by AdaCore. But we, the users of GNAT
GPL, are *not*.

I have taken the source code provided by AdaCore in GNAT GPL 2009
about 9 months before Snow Leopard was released to the world and
rebuilt it with one change (caused by Apple's removal of sigreturn()).

I don't know any of AdaCore's customers who are supported on Mac OS X
so I can't say what AdaCore's advice to them has been; possibly those
customers who mustmustmust upgrade have been provided with wavefronts.

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[(see below)]

On 22/12/2009 21:52, in article
a7b97585-f616-43e9-ac61-c468c88362fa@b2g2000yqi.googlegroups.com, "sjw"
<simon.j.wright@mac.com> wrote:

> On Dec 22, 5:46�pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
> 
>> This is what happened when I tried out the 32-bit version:
> [...]
>> dyld: Symbol not found: ___gmpn_bases
>> � Referenced from:
>> /opt/gnat-gpl-2009/libexec/gcc/i386-apple-darwin10.2.0/4.3.4/gnat1
>> � Expected in: /usr/local/lib/libmpfr.1.dylib
>> �in /opt/gnat-gpl-2009/libexec/gcc/i386-apple-darwin10.2.0/4.3.4/gnat1
> 
> Very sorry about that. Have uploaded a more appropriate libmpfr at the
> same place as the compiler (seems to be taking its time to appear ...)
> unpack /some/where and
> 
>   $ export DYLD_LIBRARY_PATH=/some/where

That works very nicely, and, unlike the other SL compilers I've tried
previously, code compiled with inlining at -O3 does not malfunction
(on the two major test cases I've tried so far, anyway).

You might be interested to see what I'm doing with it:

  <http://www.findlayw.plus.com/>.

(The KDF9 had a 48-bit word, and M is a 64-bit machine, so their emulators
run in 64-bit mode at more than twice the speed of 32-bit mode.)

Thanks for these great efforts.

-- 
Bill Findlay
<surname><forename> chez blueyonder.co.uk

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[Georg Bauhaus]

(see below) schrieb:

> That works very nicely, and, unlike the other SL compilers I've tried
> previously, code compiled with inlining at -O3 does not malfunction
> (on the two major test cases I've tried so far, anyway).

FWIW, when we tried to find the best options for the Shootout
programs, it turned out that -gnatn -O2 -funroll-loops was
about as good as -O3. Sometimes the results were even faster.

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[(see below)]

On 23/12/2009 13:48, in article
4b321fb4$0$7621$9b4e6d93@newsspool1.arcor-online.net, "Georg Bauhaus"
<rm.dash-bauhaus@futureapps.de> wrote:

> (see below) schrieb:
> 
>> That works very nicely, and, unlike the other SL compilers I've tried
>> previously, code compiled with inlining at -O3 does not malfunction
>> (on the two major test cases I've tried so far, anyway).
> 
> FWIW, when we tried to find the best options for the Shootout
> programs, it turned out that -gnatn -O2 -funroll-loops was
> about as good as -O3. Sometimes the results were even faster.

I dare say.
I have made similar trials for my own code, and the optimization
flags I use are empirically the best for its object code speed.

-- 
Bill Findlay
<surname><forename> chez blueyonder.co.uk

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

On Dec 22, 5:10 pm, sjw <simon.j.wri...@mac.com> wrote:
> On Dec 22, 2:23 pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
>
> > On 22/12/2009 12:09, in article
> > af608388-8088-4779-a3ca-39275fd2c...@r24g2000yqd.googlegroups.com, "sjw"
>
> > <simon.j.wri...@mac.com> wrote:
> > > I've just released a 32-bit build of GNAT GPL 2009 for Snow Leopard;
> > > find it athttp://sourceforge.net/projects/gnuada/files/(navigateto
> > > GNAT_GPL Mac OS X/2009-snow-leopard-i386).
>
> > Many thanks -- I'll try it ASAP.
> > Any chance of a 64-bit build?
>
> Will have a go, but it's a cross-build so may take some getting my
> head round!

The cross-build (I think I may be wrong about that; perhaps there's an
'architecture' switch?) eventually produced a compiler. Unfortunately
it wouldn't process the simplest exception properly (SIGABRT).

I then had the idea, since the problem that stops building is that
init.c calls sigreturn() which isn't present in Snow Leopard, why not
fake up our own sigreturn() that does exactly what the patch did, then
link against that?

This turned out to work as hoped, result being that we can use Apple's
GNAT-GPL-2009 on Snow Leopard; see http://sourceforge.net/projects/gnuada/files/(navigate
to GNAT_GPL Mac OS X/2009-snow-leopard, get sigreturn.tar.bz2).

Sadly, the failure to handle exceptions with the 64-bit compiler is
unchanged (the 32-bit compiler is OK):

with Ada.Text_IO; use Ada.Text_IO;
procedure Raiser is
begin
   begin
      raise Constraint_Error;
   exception
      when Constraint_Error =>
         Put_Line ("CE raised.");
   end;
end Raiser;

nidhoggr:Test simon$ GNAT_PREFIX=/opt/gnat-gpl-2009_64-leopard
gnatmake -g -f raiser -largs sigreturn-64.o
gcc -c -g raiser.adb
gnatbind -x raiser.ali
gnatlink raiser.ali -g sigreturn-64.o
nidhoggr:Test simon$ ./raiser
Segmentation fault


Rats.

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[Jerry]

On Dec 24, 8:01 am, sjw <simon.j.wri...@mac.com> wrote:
> This turned out to work as hoped, result being that we can use Apple's
> GNAT-GPL-2009 on Snow Leopard;

Is this a typo?

Jerry

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

On Dec 30, 3:12 am, Jerry <lancebo...@qwest.net> wrote:
> On Dec 24, 8:01 am, sjw <simon.j.wri...@mac.com> wrote:
>
> > This turned out to work as hoped, result being that we can use Apple's
> > GNAT-GPL-2009 on Snow Leopard;
>
> Is this a typo?

Yes, sorry; s/Apple/AdaCore/

Re: GNAT GPL 2009 for Mac OS X (Snow Leopard)

[sjw]

On Dec 22, 5:10 pm, sjw <simon.j.wri...@mac.com> wrote:
> On Dec 22, 2:23 pm, "(see below)" <yaldni...@blueyonder.co.uk> wrote:
>
> > On 22/12/2009 12:09, in article
> > af608388-8088-4779-a3ca-39275fd2c...@r24g2000yqd.googlegroups.com, "sjw"
>
> > <simon.j.wri...@mac.com> wrote:
> > > I've just released a 32-bit build of GNAT GPL 2009 for Snow Leopard;
> > > find it athttp://sourceforge.net/projects/gnuada/files/(navigateto
> > > GNAT_GPL Mac OS X/2009-snow-leopard-i386).
>
> > Many thanks -- I'll try it ASAP.
> > Any chance of a 64-bit build?
>
> Will have a go, but it's a cross-build so may take some getting my
> head round!

Not actually a cross build, which would be specified by configuring
with --target=x86_64-apple-darwin10.2.0; I don't know what the
technical name is, but configuring with --build=x86_64-apple-
darwin10.2.0 does the trick.

Unfortunately the built compiler won't handle exceptions; a severe
failing. Perhaps Apple have changed the OS features that GNAT uses to
support exception handling? (I tried configuring with --enable-sjlj-
exceptions, no joy).