Mail Archives: djgpp-workers/1996/04/13/10:05:08
Date: Sat, 13 Apr 1996 06:53:53 -0400
From: Richard Stallman <rms AT gnu DOT ai DOT mit DOT edu>
To: dj AT delorie DOT com
Subject: [storm AT olicom DOT dk: Re: New pretest]
Item 3 seems like something that could be worth fixing in DJGPP.
------- Start of forwarded message -------
From: storm AT olicom DOT dk
To: Richard Stallman <rms AT gnu DOT ai DOT mit DOT edu>
Date: Fri, 12 Apr 1996 13:59 MET
Subject: Re: New pretest
Content-Type: text/plain
I am in the process of compiling the pretest with DJGPP V2 for msdos.
So far, I have encountered the following problems:
1) The msdoss/mainmake.v2 file is missing.
To get on, I fetched that file manually from your source.
2) The config.bat script does not work with the 4DOS shell, since
it does not like dashes in environment variable names.
To fix this, perform the following global change on config.bat:
"djgpp-ver" -> "djgpp_ver"
3) I am now running under Windows95 which has long file name support.
When I unpacked the tar file using a standard tar, the file names
were "long", i.e. not the 8.3 DOS format.
Unfortunately, the DJGPP compiler cannot handle long file names
in source files, so the make failed already on the first file:
test-distrib.c
I got a tar named "ntar" from a collegue (I don't known the origin
of this tar) which will "properly" rename files to the 8.3 form.
So I was able to continue.
But this will become a problem (I don't have a solution...)
4) The "hexl" program does not compile with V2 since it references
_IOTEXT which is not defined by V2 stdio.h
I guess it could just define it to whatever value is needed (I have
not yet found the proper value) if it is DJGPP_V2 and _IOTEXT is not
previously defined - or perhaps it is defined in another header file.
I'll let you know if I find a solution.
5) There is an unresolved reference to "init_gettimeofday" in emacs.c
I guess it must be guarded by DJGPP_V2, but I haven't tried that yet,
but I'm going to look at it right now and let you know later how it
went.
++Kim
------- End of forwarded message -------
- Raw text -