In PR116149 we choose a wrong vector length which causes wrong values in
a reduction. The problem happens in avlprop where we choose the
number of units in the instruction's mode as vector length. For the
non-scalar variants the respective operand has the correct non-widened
mode. For the scalar variants, however, the same operand has a scalar
mode which obviously only has one unit. This makes us choose VL = 1
leaving three elements undisturbed (so potentially -1). Those end up
in the reduction causing the wrong result.
This patch adjusts the mode_idx just for the scalar variants of the
affected instruction patterns.
gcc/ChangeLog:
PR target/116149
* config/riscv/vector.md: Fix mode_idx attribute of scalar
widen add/sub variants.
gcc/testsuite/ChangeLog:
* gcc.target/riscv/rvv/autovec/pr116149.c: New test.
"TARGET_VECTOR"
"vwadd<any_extend:u>.wx\t%0,%3,%z4%p1"
[(set_attr "type" "viwalu")
+ (set_attr "mode_idx" "3")
(set_attr "mode" "<V_DOUBLE_TRUNC>")])
(define_insn "@pred_single_widen_sub<any_extend:su><mode>_extended_scalar"
"TARGET_VECTOR"
"vwsub<any_extend:u>.wx\t%0,%3,%z4%p1"
[(set_attr "type" "viwalu")
+ (set_attr "mode_idx" "3")
(set_attr "mode" "<V_DOUBLE_TRUNC>")])
(define_insn "@pred_widen_mulsu<mode>"
--- /dev/null
+/* { dg-do compile } */
+/* { dg-options "-O3 -march=rv64gcv_zvl256b -mabi=lp64d -mrvv-vector-bits=zvl" } */
+
+long a;
+short b[60000];
+short c[20];
+int main() {
+ for (short d = 0; d < 20; d += 3) {
+ c[d] = 0;
+ for (int e = 0; e < 20; e += 2)
+ for (int f = 1; f < 20; f += 2)
+ a += (unsigned)b[f + e];
+ }
+ if (a != 0)
+ __builtin_abort ();
+}
+
+/* { dg-final { scan-assembler-times "vsetivli\tzero,1" 0 } } */