Greetings, and thanks for your feedback! Could you also please send
the plt.h produced in both cases? Your pcl failure report needs more
context, as the error was somehwere earlier in the log (You can tell
by the double bracket error prompt >>).
The idea of the plt stuff is to make an explicit t
Committed -- thanks!
Clemens Heitzinger <[EMAIL PROTECTED]> writes:
> Include what is needed for GET_FAULT_ADDR.
> Fixes compilation of main.c.
>
>
> Index: h/powerpc-macosx.h
> ===
> RCS file: /cvsroot/gcl/gcl/h/powerpc-macosx.h,v
Greetings! The origin of this macro is o/object.h. Certain headers
are concatenated for use by the compiler in h/cmpinclude.h, copied to
o/ for convenience, and which is stored in the final image as a
string.
Take care,
Clemens Heitzinger <[EMAIL PROTECTED]> writes:
> CORE_PAGES is defined in
Greetings!
I was just going to write to you about hash. I'd done some work on it
too, so its a good idea for us to merge. I'll take a look when I get
a moment, should be soon. In the meantime, I'll supply here the
version I've been working with for your perusal.
In sum, I'm working on CVS he
Include what is needed for GET_FAULT_ADDR.
Fixes compilation of main.c.
Index: h/powerpc-macosx.h
===
RCS file: /cvsroot/gcl/gcl/h/powerpc-macosx.h,v
retrieving revision 1.14
diff -r1.14 powerpc-macosx.h
15a16,17
> #include
> #incl
CORE_PAGES is defined in h/cmpinclude.h, h/object.h, and o/
cmpinclude.h. Is this intentional? Which file is used?
The same is true for RB_GETA.
Clemens
___
Gcl-devel mailing list
Gcl-devel@gnu.org
http://lists.gnu.org/mailman/listinfo/gcl-devel
Hello,
I switched to gcc 3.3 which is included with Tiger.
On gcl 2.6.6 with the previously mentioned patch the compilation
complete but the following ansi-test fail.
There is indeed a difference in nm plttest.o, with 3.3 :
nm plttest.o
U _acos
U _acosh
U _asin
