Re: polyorb / gnat-4.6 problem
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Le 01/02/2012 14:07, Ludovic Brenta a écrit :
> Xavier Grave wrote:
>> I have reduced the problem out of polyorb :(see attached files)
>> gnatmake -P test.gpr => runs fine
>> gnatmake -a -P test.gpr =>
>> gcc-4.6 -c -I- -gnatA
>> /home/xavier/labo/test_gnat_expect/test_gnat_expect.adb
>> gnatmake: external source (g-expect.adb) is not part of any project;
>> cannot be compiled without gnatmake switch -x
>>
>> So I see two quick fix : remove -a or add -x options, but this seems
>> "dirty" to me because I'm not fully understand why this problem arise
>> and what others problem will be induced by these changes.
>
> I think I'm beginning to understand. With -a, gnatmake considers whether
> or not to recompile g-expect.adb and for some unknown reason decides that
> it must. But g-expect.adb is not in the Source_Dirs of any project file
> because it is part of the Ada run-time library.
>
> Maybe try gnatmake -a -v -P test.gpr and see why gnatmake thinks it must
> recompile g-expect.adb; this might be a bug in gnat-4.6, possibly because
> of time stamps on g-expect.adb and g-expect.ali.
>
> Also try gnatmake -a -m -v -P test.gpr (minimal recompilation) to see if
> it makes a difference.
>
Using these flags it's building fine :
(sid)xavier@rangifer-tarandus:~/labo/test_gnat_expect$ gnatmake -a -m -v
- -P test.gpr
Parsing project file "test.gpr".
Processing_Naming_Scheme for project "test"
4 lines: No errors
Parsing of project file "test.gpr" is finished.
Adding source directory "/home/xavier/labo/test_gnat_expect/".
Adding object directory "/home/xavier/labo/test_gnat_expect/".
GNATMAKE 4.6
Copyright (C) 1995-2010, Free Software Foundation, Inc.
Changing to object directory of "test":
"/home/xavier/labo/test_gnat_expect/"
Main source file: "test_gnat_expect.adb".
Created configuration file
"/home/xavier/labo/test_gnat_expect/GNAT-TEMP-000001.TMP"
"/home/xavier/labo/test_gnat_expect/test_gnat_expect.ali" being
checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-stalib.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/gnat.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/g-expect.ali" being checked ...
s-oscons.ads: up to date, different timestamps but same checksum
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/system.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-memory.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-soflin.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/ada.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-calend.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-except.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-tags.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/g-io.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/g-os_lib.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/g-regpat.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-oscons.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-secsta.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-stalib.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-filico.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-stream.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-exctab.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-finimp.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-finroo.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-stoele.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-stratt.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-crtl.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-parame.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-stache.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/interfac.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-osprim.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-elchha.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-except.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-imgint.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-traceb.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-unstyp.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-wchcon.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-wchstw.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-traent.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-htable.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-valllu.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-os_lib.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-regpat.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/i-c.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-finali.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-restri.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-ioexce.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-wchcnv.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-strhas.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-valuti.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-casuti.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-string.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-charac.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-chahan.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-imgcha.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-imenne.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-io.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-valint.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-wchjis.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-chlat1.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-string.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-strmap.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/a-stmaco.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-valuns.ali" being checked ...
"/usr/lib/gcc/x86_64-linux-gnu/4.6/adalib/s-bitops.ali" being checked ...
End of compilation
"/home/xavier/labo/test_gnat_expect/test_gnat_expect" missing.
gnatbind -I- -x /home/xavier/labo/test_gnat_expect/test_gnat_expect.ali
gnatlink /home/xavier/labo/test_gnat_expect/test_gnat_expect.ali -o
/home/xavier/labo/test_gnat_expect/test_gnat_expect
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
iEYEARECAAYFAk8pP80ACgkQVIZi0A5BZF7zogCgk99U95BJw8J8x+NvIXjVxSQi
LuEAoK3fHvI/B8SuBgPD/XKLU8PMIDaE
=lEfx
-----END PGP SIGNATURE-----
Reply to: