Re: 21.2.11 wont compile with mule
Hi Gunnar
Gunnar Evermann writes:
Jan Hebler <jan(a)hebler.de> writes:
Hi Jan,
thanks for your report.
> if i try to compile xemacs 21.2.11 with mule i get the following error
>
> insert-file-contents-internal("/usr/src/Xemacs/lisp/finder-inf.el" t nil
nil nil undecided used-codesys)
does that file exist and is it readable?
No, that file doesnt exist and so, it is
not readable ;-)
But IMHO this is not the Problem.
If i start the new generated xemacs, everything works fine until
i want to load a nonexistend file. _then_ i will got this
error. it seems to me that there is something broken in the area
of locales LC_XXXX and friends. To be more precisiously the Bug
occurs _only_ if i have set my LC_ALL to "DE_de" if i set them
to "C" everything works fine (well, except that my LOCALES are
now broken)
In the File ChangeLog i found:
-1998-12-28 Martin Buchholz <martin(a)xemacs.org>
-
- * XEmacs 21.2.8 is released.
-
-1998-12-28 Martin Buchholz <martin(a)xemacs.org>
-
- * PROBLEMS: Document Linux GNU Libc 2.0 I18N crashes.
-
-1998-12-24 Martin Buchholz <martin(a)xemacs.org>
-
- * XEmacs 21.2.7 is released.
I think this is the real problem.
too bad, the patch is more than 100k, so there are many points to
hide.
*please* tell me you are not building with an gcc/egcs snapshot using
a glibc snapshot on an unstable kernel release using -O3! If you do,
then you got what you deserve. :-)
Well, lets see
Kernel is 2.1.132 this should be stable, since he is the last before the
2.2 Kernels
glibc is (from version.h)
-/* This file just defines the current version number of libc. */
-
-#define RELEASE "production"
-#define VERSION "2.0.7"
gcc is :
Reading specs from /usr/lib/gcc-lib/i486-linux/egcs-2.91.60/specs
gcc version egcs-2.91.60 19981201 (egcs-1.1.1 release)
The whole System is an SUSE 6.0 german
> used_codesys=137990816)
> at fileio.c:148
^^^^
I *hate* inlined functions -- which call to report_file_error is this?
Umpf, what do you mean with "wich call" ? If you tell me how i should look,
maybe i kan figure it out.
seriously, I really can't see how XEmacs can be blamed for strerror(2)
segfault'ing.
could you try rebuilding without -O3?
Of course, but the problem is still the
same.
ru Jan
PS sorry for my nasty english, but it is not my native language