This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: PATCH/Ada: Fix DIR_SEPARATOR bug on win32 targets


 --- Arnaud Charlet <charlet@ACT-Europe.FR> wrote: > > The bootstrap is successful.  However, I am
still not able to get 
> > make check-ada to work.  This is what I get:
> 
> there's not enough info here to get what is going wrong.
> I'd suggest looking at acats.log which contains more details.
> 
> Arno 

That was an excerpt from the acats.log.

I think the problem is that Ada also needs its own version of IS_DIRECTORY_SEPARATOR.
The problem with generating the support file was with gnatchop.  This
quick hack fixes that problem:

*** gnatchop.adb.orig	Thu Nov 13 23:04:57 2003
--- gnatchop.adb	Thu Nov 13 23:06:18 2003
*************** procedure Gnatchop is
*** 516,522 ****
           --  Find Start_Of_Prefix
  
           for J in reverse Current_Command'Range loop
!             if Current_Command (J) = '/' or
                Current_Command (J) = Directory_Separator or
                Current_Command (J) = ':'
              then
--- 516,522 ----
           --  Find Start_Of_Prefix
  
           for J in reverse Current_Command'Range loop
!             if Current_Command (J) = '\' or
                Current_Command (J) = Directory_Separator or
                Current_Command (J) = ':'
              then


and the testsuite goes a bit further:

		=== acats configuration ===
gcc is /develop/gcc-3.4/build/gcc/testsuite/ada/acats/gcc
Reading specs from /develop/gcc-3.4/build/gcc/specs Configured with: ../gcc/configure
--host=mingw32 --build=mingw32 --target=mingw32 --prefix=/mingw --enable-threads --disable-nls
--enable-languages=c,c++,objc,f77,ada --disable-win32-registry --disable-shared
--enable-sjlj-exceptions --enable-checking=misc Thread model: win32 gcc version 3.4 20031112
(experimental)
host=mingw32
target=mingw32
gnatmake is /develop/gcc-3.4/build/gcc/gnatmake

GNATLS 3.4 20031112 (experimental) Copyright 1997-2003 Free Software Foundation, Inc.

Source Search Path:
   <Current_Directory>
   /develop/gcc-3.4/build/gcc/ada/rts/
   /mingw/lib/gcc/mingw32/3.4/adainclude/


Object Search Path:
   <Current_Directory>
   /develop/gcc-3.4/build/gcc/ada/rts/
   /mingw/lib/gcc/mingw32/3.4/adalib/


		=== acats support ===
Generating support files...splitting
/develop/gcc-3.4/build/gcc/testsuite/ada/acats/support/checkfil.ada into:
   check_file.adb
splitting /develop/gcc-3.4/build/gcc/testsuite/ada/acats/support/enumchek.ada into:
   enum_check.ads
   enum_check.adb
 
<snip>

 done.

		=== acats tests ===
Running chapter a ...
splitting /develop/gcc-3.4/build/gcc/testsuite/ada/acats/tests/a/a22006b.ada into:
   a22006b.adb
BUILD a22006b.adb
d:\develop\gcc-3.4\build\gcc\gnatmake: error, unable to locate d:\develop\gcc-gcc
FAIL:	a22006b
splitting /develop/gcc-3.4/build/gcc/testsuite/ada/acats/tests/a/a22006c.ada into:
   a22006c.adb
BUILD a22006c.adb
d:\develop\gcc-3.4\build\gcc\gnatmake: error, unable to locate d:\develop\gcc-gcc
FAIL:	a22006c
splitting /develop/gcc-3.4/build/gcc/testsuite/ada/acats/tests/a/a22006d.ada into:
   a22006d.adb

... more of the same

So I suspect that similar problem is happening with prefix in gnatmake 

Danny

http://personals.yahoo.com.au - Yahoo! Personals
New people, new possibilities. FREE for a limited time.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]