Mail Archives: cygwin-developers/1999/01/29/21:03:33
Something I didn't take into account with this is that the configure
mechanism bypasses the headers in most cases, and checks directly by
using the linker.
looks like using the DATA keyword in cygwin.din wasn't such a good idea
:-(
_ctype_ DATA
_sys_errlist DATA
_sys_nerr DATA
configure can't find sys_errlist or sys_nerr, in .../src/libiberty.
Either the test in .../src/libiberty/configure needs to be changed to
the same as the one in .../src/bash/configure (which would also fix
mingw32) or the DATA keyword needs to be removed from these 2 in
cygwin.din.
>On Sun, Dec 27, 1998 at 11:09:18PM +0000, Mikey wrote:
>>Apparently I was mistaken, I untarred newlib, for both b20 and b20.1
>>and I do remember giving up on configure-target-newlib for 20.1 after
>>about the 30th try, ;^)
>
>Ok. This "30 times configuring newlib" sounds like another bug.
>Want to provide specifics?
>
>>I thought I had deleted the files, but if libcctype.cc has been like
>>that since 1/98 I must have left them installed.
>>
>>regardless compiling both ctype_.c and libcctype.cc and then linking
>>libcctype.o statically into every image is a waste of time and disk
>>space, and it also makes it necessary to install and compile all of
>>src/newlib unless you know ahead of time exactly what is going on.
>>
>>a simple
>>
>>#if defined(__CYGWIN__) && !defined(_COMPILING_NEWLIB) extern _CONST
>>char *_imp___ctype_[];
>>#define _ctype_ (*_imp___ctype_)
>>#else
>>extern _CONST char _ctype_[];
>>#endif in newlib/include/ctype.h
>>
>>and adding
>>
>>_ctype_ DATA to cygwin.din
>>
>>would eliminate the need No?
>
>Ok. I understand now. I'll see what I can do.
>
>-chris
- Raw text -