Mainline infinite recursion on ia64

[John David Anglin]

> > This also happens on hppa64-hp-hpux11.11.
> 
> Once again my apologies.  I don't suppose you could try the patch
> I posted to gcc-patches on hppa64-hp-hpux11.11?  I'm having problems
> with "cvs update", so it'll be a while before I came confirm that
> 64-bit targets are fixed with my change.

Your patch fixes the bootstrap problem.

However in the testsuite, I see the following regressions since
my last hppa64-hp-hpux11.11 build:

FAIL: gcc.c-torture/execute/pure-1.c compilation,  -O3 -fomit-frame-pointer 
UNRESOLVED: gcc.c-torture/execute/pure-1.c execution,  -O3 -fomit-frame-pointer 
FAIL: gcc.c-torture/execute/pure-1.c compilation,  -O3 -g 
UNRESOLVED: gcc.c-torture/execute/pure-1.c execution,  -O3 -g 

Over the last month or so, a number of new fails have been introduced:

FAIL: gcc.dg/Wreturn-type.c (test for excess errors)
FAIL: gcc.dg/funcorder.c scan-assembler-not link_error
FAIL: gcc.dg/old-style-asm-1.c scan-assembler L2
FAIL: gcc.dg/warn-1.c  (test for warnings, line 8)
FAIL: gcc.dg/warn-1.c  (test for warnings, line 15)
FAIL: gcc.dg/pch/inline-1.c (test for excess errors)

FAIL: pch/system-1.C -g assembly comparison
FAIL: pch/system-1.C -O2 -g assembly comparison
FAIL: g++.robertl/eb42.C (test for excess errors)

FAIL: g77.f-torture/execute/980520-1.f compilation,  -O0 

and a huge number of locale related fails in the libstdc++ testsuite.
I didn't include 20010119-1.c which is a known and understood problem.

The Wreturn-type.c fail is caused by the warning:
/xxx/gnu/gcc-3.3/gcc/gcc/testsuite/gcc.dg/Wreturn-type.c:5: warning: control rea
ches end of non-void function

The funcorder.c fail is caused by I believe an ".IMPORT" statement containing
"link_error", so this is just a test problem.

old-style-asm-1.c fails because the label is "L$0002", so again this is a test
problem.

Don't know why warn-1.c fails.

pch/inline-1.c fails because:
/xxx/gnu/gcc-3.3/gcc/gcc/testsuite/gcc.dg/pch/inline-1.c: In function `foo':
/xxx/gnu/gcc-3.3/gcc/gcc/testsuite/gcc.dg/pch/inline-1.c:5: internal compiler er
ror: in gen_subprogram_die, at dwarf2out.c:10863

The g++ pch assembly comparisons have a huge number of differences.

g++.robertl/eb42.C fails because of
/xxx/gnu/gcc-3.3/gcc/gcc/testsuite/g++.old-deja/g++.robertl/eb42.C:13: error: extra `;'

971102-1.f fails because of
ld: Unsatisfied symbol "val_" in file /var/tmp//ccqeJr2U.o
This fail is new as of March 4.

Dave
-- 
J. David Anglin                                  dave.anglin@nrc-cnrc.gc.ca
National Research Council of Canada              (613) 990-0752 (FAX: 952-6605)