build broken

build broken

[mike stump]

The build for the native sparc-sun-solaris2.7 bootstrap is broken:

../../gcc/gcc/unwind-dw2-fde.c: In function `classify_object_over_fdes':
../../gcc/gcc/unwind-dw2-fde.c:650: Internal compiler error in do_SUBST, at combine.c:439
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.
make[3]: *** [libgcc/./unwind-dw2-fde.o] Error 1
make[2]: *** [stmp-multilib] Error 2
make[2]: Leaving directory `/kankakee1/tmp/mrs/net/gcc-sol/gcc'
make[1]: *** [stage1_build] Error 2
make[1]: Leaving directory `/kankakee1/tmp/mrs/net/gcc-sol/gcc'
make: *** [bootstrap] Error 2

:-(  Alexandre?

Re: build broken

[Alexandre Oliva]

On Feb 21, 2002, mike stump <mrs@windriver.com> wrote:

> The build for the native sparc-sun-solaris2.7 bootstrap is broken:

> ../../gcc/gcc/unwind-dw2-fde.c: In function `classify_object_over_fdes':
> ../../gcc/gcc/unwind-dw2-fde.c:650: Internal compiler error in do_SUBST, at combine.c:439

> :-(  Alexandre?

Yup.  The sanity check I installed exposed the bug.  I also posted a
patch that would fix it, but the patch wasn't entirely approved
because part of it will require some additional work, so broken was
the sparc port in this regard.  I hope someone with more familiarity
with sparc would take it over; if not, I'm going to get to it later
tonight, after I make some other fixes for a toolchain I'm working on
a get a long session of testing running.

-- 
Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
Free Software Evangelist                Professional serial bug killer

Re: build broken

[Mike Stump]

On Mar 6, 2007, at 11:14 AM, Matthias Klose wrote:
> Mike Stump schrieb:
>> I have a feeling sed -i isn't our friend.
>
> fixed.

I can confirm that this fixed my build.  I'm expected the regression  
tester to follow shortly.

Thanks.

Re: build broken

[Matthias Klose]

[-- Attachment #1: Type: text/plain, Size: 160 bytes --]

Mike Stump schrieb:
> It appears that one of these:
> + '[' -s .bad_compare ']'
> + exit 1
> 
> I have a feeling sed -i isn't our friend.

fixed.

   Matthias


[-- Attachment #2: cp-tools-doc.diff --]
[-- Type: text/plain, Size: 787 bytes --]

2007-03-06  Matthias Klose  <doko@ubuntu.com>

	* doc/Makefile.am(gkeytool.pod): Don't use sed -i.
	* doc/Makefile.in: Regenerate.

Index: doc/Makefile.am
===================================================================
--- doc/Makefile.am	(Revision 122631)
+++ doc/Makefile.am	(Arbeitskopie)
@@ -67,10 +67,10 @@
 
 # hack around the cross references and the enumeration
 gkeytool.pod: $(srcdir)/cp-tools.texinfo
-	-$(TEXI2POD) -D gkeytool < $< > $@
-	sed -i -e 's/^For more details.*/See I<Common Options> for more details./' \
+	-$(TEXI2POD) -D gkeytool < $< \
+	  | sed -e 's/^For more details.*/See I<Common Options> for more details./' \
 	    -e 's/1\.<\([^>]*\)>/- \1/' \
-		$@
+		> $@
 
 gnative2ascii.pod: $(srcdir)/cp-tools.texinfo
 	-$(TEXI2POD) -D gnative2ascii < $< > $@

build broken

[Mike Stump]

It appears that one of these:

------------------------------------------------------------------------
r122580 | doko | 2007-03-05 15:23:18 -0800 (Mon, 05 Mar 2007) | 6 lines

2007-03-02  Mario Torre  <neugens@limasoftware.net>

        PR classpath/31017:
        committed for Petteri R<C3><83><C2><A4>ty  
<betelgeuse@gentoo.org>
        * configure.ac: fix broken build for gcj browser plugin

------------------------------------------------------------------------
r122573 | doko | 2007-03-05 12:24:50 -0800 (Mon, 05 Mar 2007) | 5 lines

2007-03-05  Matthias Klose  <doko@ubuntu.com>

         * doc/Makefile.am(man_MANS): Add $(TOOLS_MANFILES).
         * doc/Makefile.in: Regenerate.

------------------------------------------------------------------------

Did this:

2007-03-04T22:54:20Z 2007-03-05T05:50:09Z regress-3 native 122532
2007-03-05T05:53:26Z 2007-03-05T12:49:35Z regress-3 native 122539
2007-03-05T12:52:59Z 2007-03-05T19:53:13Z regress-3 native 122546
2007-03-05T19:56:35Z 2007-03-06T02:53:47Z regress-3 native 122566
2007-03-06T02:57:13Z 2007-03-06T05:18:43Z build native 122594
2007-03-06T07:48:12Z 2007-03-06T10:11:04Z build native 122597
2007-03-06T10:14:03Z 2007-03-06T12:32:47Z build native 122618
2007-03-06T14:18:12Z 2007-03-06T16:37:20Z build native 122619

to the darwin nightly tester.  From the log file:

perl /Users/regress/tbox/svn-gcc/libjava/classpath/../../contrib/ 
texi2pod.pl -D gkeytool < /Users/regress/tbox/svn-gcc/libjava/ 
classpath/doc/cp-tools.texinfo > gkeytool.pod
perl /Users/regress/tbox/svn-gcc/libjava/classpath/../../contrib/ 
texi2pod.pl -D gnative2ascii < /Users/regress/tbox/svn-gcc/libjava/ 
classpath/doc/cp-tools.texinfo > gnative2ascii.pod
perl /Users/regress/tbox/svn-gcc/libjava/classpath/../../contrib/ 
texi2pod.pl -D gorbd < /Users/regress/tbox/svn-gcc/libjava/classpath/ 
doc/cp-tools.texinfo > gorbd.pod
sed -i -e 's/^For more details.*/See I<Common Options> for more  
details./' \
     -e 's/1\.<\([^>]*\)>/- \1/' \
	gkeytool.pod
sed: lstat: No such file or directory
make[7]: *** [gkeytool.pod] Error 1
make[7]: *** Waiting for unfinished jobs....
rm gnative2ascii.pod gorbd.pod gjar.pod gappletviewer.pod  
gkeytool.pod gjavah.pod gjarsigner.pod
make[6]: *** [all-recursive] Error 1
make[5]: *** [all-recursive] Error 1
make[4]: *** [all-recursive] Error 1
make[3]: *** [multi-do] Error 1
make[2]: *** [all-multi] Error 2
make[1]: *** [all-target-libjava] Error 2
make: *** [bootstrap] Error 2
+ '[' -s .bad_compare ']'
+ exit 1

I have a feeling sed -i isn't our friend.