[gcc(refs/users/aoliva/heads/testme)] testsuite: ppc64: add -mlong-double-128 to pr82748-1.c

[gcc(refs/users/aoliva/heads/testme)] testsuite: ppc64: add -mlong-double-128 to pr82748-1.c

[Alexandre Oliva]

https://gcc.gnu.org/g:14765d058056e69a72c3fbb9912577578c88a821

commit 14765d058056e69a72c3fbb9912577578c88a821
Author: Alexandre Oliva <oliva@adacore.com>
Date:   Wed Apr 27 18:23:02 2022 -0300

    testsuite: ppc64: add -mlong-double-128 to pr82748-1.c
    
    With 64-bit long doubles, do_copysign_ld needs to convert its DFmode
    parameters to KFmode for __builtin_copysignq, and the result back to
    DFmode.  Because the expanders for the conversions are conditioned on
    TARGET_LONG_DOUBLE_128 (see PR105359), we end up issuing libcalls, but
    the test checks for no 'bl' instructions, and that fails.
    
    
    for  gcc/testsuite/ChangeLog
    
            * gcc.target/powerpc/pr82748-1.c: Add -mlong-double-128.

Diff:
---
 gcc/testsuite/gcc.target/powerpc/pr82748-1.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gcc/testsuite/gcc.target/powerpc/pr82748-1.c b/gcc/testsuite/gcc.target/powerpc/pr82748-1.c
index 15a746bcf63..a04629ce67f 100644
--- a/gcc/testsuite/gcc.target/powerpc/pr82748-1.c
+++ b/gcc/testsuite/gcc.target/powerpc/pr82748-1.c
@@ -1,6 +1,6 @@
 /* { dg-do compile { target { powerpc*-*-* && lp64 } } } */
 /* { dg-require-effective-target powerpc_p9vector_ok } */
-/* { dg-options "-mpower9-vector -O2 -mabi=ieeelongdouble -Wno-psabi" } */
+/* { dg-options "-mpower9-vector -mlong-double-128 -O2 -mabi=ieeelongdouble -Wno-psabi" } */
 
 /* Make sure the old 'q' builtin functions work correctly when the long double
    default has been changed to be IEEE 128-bit floating point.  */