This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: fixincludes for darwin7.9.0 long double
- From: BjÃrn Haase <bjoern dot m dot haase at web dot de>
- To: gcc-patches at gcc dot gnu dot org,bkorb at gnu dot org
- Cc: Geoff Keating <geoffk at geoffk dot org>,ebotcazou at libertysurf dot fr
- Date: Sun, 7 Aug 2005 12:41:41 +0200
- Subject: Re: fixincludes for darwin7.9.0 long double
- References: <F834E49F-598A-4BE7-A075-92BB553BA1C4@geoffk.org>
Geoff Keating wrote on Freitag, 24. Juni 2005 20:27 :
> This fixincludes patch should fix the problem with long double
> functions on Darwin 7.9.0 / Mac OS 10.3.9.
Hi,
with present head 4.1 I am no longer able to run the testsuites on my linux
x86_64 host. The error message reads very much like the one Eric Botcazu has
posted:
math.h /home/bmh/gnucvs/cleanhead/gcc/fixincludes/tests/base/math.h
differieren:
Byte 501, Zeile 22.
*** math.h 2005-08-07 12:24:15.085487856 +0200
--- /home/bmh/gnucvs/cleanhead/gcc/fixincludes/tests/base/math.h
2005-06-
15 01:59:48.000000000 +0200
***************
*** 19,29 ****
struct exception;
- #if defined( AAB_DARWIN7_9_LONG_DOUBLE_FUNCS_2_CHECK )
- #include <architecture/ppc/math.h>
- #endif /* AAB_DARWIN7_9_LONG_DOUBLE_FUNCS_2_CHECK */
-
-
#if defined( BROKEN_CABS_CHECK )
#ifdef __STDC__
--- 19,24 ----
There were fixinclude test FAILURES
make[1]: *** [check] Fehler 1
make[1]: Leaving directory `/home/bmh/gnucvs/cleanhead/nbuild/fixincludes'
make: *** [check-fixincludes] Fehler 2
"make check" now fails for both systems, my cross compiler avr testsuite and
the x86_64 native testsuite. There seems to be a solution sketched by bruce
korb. I tried it but I did not get the system back functional. Probably I did
something wrong since the post of bruce korb did not provide a patch and was
not very explicit on what to do and on which file to operate (I assumed some
gcc/fixincludes/includehack.def file.?)
Since the problem is pending since June and not fixed yet, it might be
necessary to open an official bugzilla entry?
It presently stops me from testing regressions of the avr port.
Yours,
Bjoern