This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
fix for altivec-4 signedness problem
- From: Geoff Keating <geoffk at geoffk dot org>
- To: gcc-patches at gcc dot gnu dot org
- Cc: Aldy Hernandez <aldyh at redhat dot com>
- Date: Tue, 8 Jan 2002 02:36:54 -0800
- Subject: fix for altivec-4 signedness problem
- Reply-to: Geoff Keating <geoffk at redhat dot com>
'char' is unsigned on powerpc, and all the builtins are defined with
signed types, so the compiler was (properly) complaining. I'm not
sure if this is quite correct but at least it's consistent.
Tested by checking this one testcase on powerpc-linux.
--
- Geoffrey Keating <geoffk@geoffk.org> <geoffk@redhat.com>
===File /home/geoffk/patches/cygnus/gcc-altivec4signed.patch===
2002-01-08 Geoff Keating <geoffk@redhat.com>
* gcc.dg/altivec-4.c: Make the 'char' entries signed also for
consistency.
Index: gcc.dg/altivec-4.c
===================================================================
RCS file: /cvs/gcc/gcc/gcc/testsuite/gcc.dg/altivec-4.c,v
retrieving revision 1.2
diff -p -u -u -p -r1.2 altivec-4.c
--- altivec-4.c 2002/01/08 04:52:36 1.2
+++ altivec-4.c 2002/01/08 10:32:18
@@ -7,7 +7,7 @@ int __attribute__((mode(V4SI))) x, y;
vector int i,j,k;
vector short s,t,u;
-vector char c,d,e;
+vector signed char c,d,e;
vector float f,g,h;
void
============================================================