Hi Alex,
On 22/02/2021 10:20, Alex Coplan wrote:
For the testcase, you might want to use the one I posted most recently:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=98791#c3
which avoids the dependency on the aarch64-autovec-preference param
(which is in GCC 11 only) as this will simplify backporting.
But if it's preferable to have a testcase without SVE intrinsics for GCC
11 then we should stick with that.
I don't see any problem with having SVE intrinsics in the testcase,
committed with your other test as I agree it makes the backport easier
eventually.
Thanks for pointing that out.
diff --git a/gcc/ira-conflicts.c b/gcc/ira-conflicts.c
index
2c2234734c3166872d94d94c5960045cb89ff2a8..d83cfc1c1a708ba04f5e01a395721540e31173f0
100644
--- a/gcc/ira-conflicts.c
+++ b/gcc/ira-conflicts.c
@@ -275,7 +275,10 @@ process_regs_for_copy (rtx reg1, rtx reg2, bool
constraint_p,
ira_allocno_t a1 = ira_curr_regno_allocno_map[REGNO (reg1)];
ira_allocno_t a2 = ira_curr_regno_allocno_map[REGNO (reg2)];
- if (!allocnos_conflict_for_copy_p (a1, a2) && offset1 == offset2)
+ if (!allocnos_conflict_for_copy_p (a1, a2)
+ && offset1 == offset2
+ && ordered_p (GET_MODE_PRECISION (ALLOCNO_MODE (a1)),
+ GET_MODE_PRECISION (ALLOCNO_MODE (a2))))
{
cp = ira_add_allocno_copy (a1, a2, freq, constraint_p, insn,
ira_curr_loop_tree_node);
diff --git a/gcc/testsuite/gcc.target/aarch64/sve/pr98791.c
b/gcc/testsuite/gcc.target/aarch64/sve/pr98791.c
new file mode 100644
index
0000000000000000000000000000000000000000..cc1f1831afb68ba70016cbe26f8f9273cfceafa8
--- /dev/null
+++ b/gcc/testsuite/gcc.target/aarch64/sve/pr98791.c
@@ -0,0 +1,12 @@
+/* PR rtl-optimization/98791 */
+/* { dg-do compile } */
+/* { dg-options "-O -ftree-vectorize" } */
+#include <arm_sve.h>
+extern char a[11];
+extern long b[];
+void f() {
+ for (int d; d < 10; d++) {
+ a[d] = svaddv(svptrue_b8(), svdup_u8(0));
+ b[d] = 0;
+ }
+}
