While chasing pr42333, I stumbled that using
"RUNTESTFLAGS=--target_board=unix'{-m32,}'" or
"RUNTESTFLAGS=--target_board=unix'{-m32,-m64}'" gave different results for two
obj-c++ tests (const-str-9.mm and template-4.mm) on x86_64-apple-darwin10. The
reason is that their dg-skip-if expect explicitly the -m64 option, not needed
for 64 bit platforms. Does the following patch makes sense?
diff -up ../_clean/gcc/testsuite/obj-c++.dg/const-str-9.mm
../work/gcc/testsuite/obj-c++.dg/const-str-9.mm
--- ../_clean/gcc/testsuite/obj-c++.dg/const-str-9.mm 2009-11-25
18:16:34.000000000 +0100
+++ ../work/gcc/testsuite/obj-c++.dg/const-str-9.mm 2009-12-10
11:13:03.000000000 +0100
@@ -3,7 +3,7 @@
/* { dg-options "-fnext-runtime" } */
/* { dg-do compile { target *-*-darwin* } } */
-/* { dg-skip-if "" { *-*-darwin* } { "-m64" } { "" } } */
+/* { dg-skip-if "" { *-*-darwin* && lp64 } { "*" } { "" } } */
#include <objc/Object.h>
diff -up ../_clean/gcc/testsuite/obj-c++.dg/template-4.mm
../work/gcc/testsuite/obj-c++.dg/template-4.mm
--- ../_clean/gcc/testsuite/obj-c++.dg/template-4.mm 2009-11-25
18:16:34.000000000 +0100
+++ ../work/gcc/testsuite/obj-c++.dg/template-4.mm 2009-12-10
11:14:02.000000000 +0100
@@ -1,7 +1,7 @@
/* Author: Ziemowit Laski <zla...@apple.com>. */
/* { dg-do run } */
-/* { dg-skip-if "" { *-*-darwin* } { "-m64" } { "" } } */
+/* { dg-skip-if "" { *-*-darwin* && lp64 } { "*" } { "" } } */
#include <objc/Object.h>
#include <stdarg.h>
--
Summary: Syntax of dg-skip-if in two obj-c++ tests
Product: gcc
Version: 4.5.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: testsuite
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: dominiq at lps dot ens dot fr
GCC build triplet: *-apple-darwin*
GCC host triplet: *-apple-darwin*
GCC target triplet: *-apple-darwin*
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42348
