Support for narrow operands in "dangerous" operations.
This is a revert^2 of commit 636e870d55c1739e2318c2180fac349683dbfa97.
Rationale:
Under strict conditions, even operations that are sensitive
to higher order bits can vectorize by inspecting the operands
carefully. This enables more vectorization, as demonstrated
by the removal of quite a few TODOs.
Test: test-art-target, test-art-host
Change-Id: Ic2684f771d2e36df10432286198533284acaf472
diff --git a/test/640-checker-double-simd/src/Main.java b/test/640-checker-double-simd/src/Main.java
index 0d4f87a..5709b5d 100644
--- a/test/640-checker-double-simd/src/Main.java
+++ b/test/640-checker-double-simd/src/Main.java
@@ -122,8 +122,10 @@
/// CHECK-DAG: ArraySet loop:<<Loop>> outer_loop:none
//
/// CHECK-START-ARM64: void Main.conv(long[]) loop_optimization (after)
+ /// CHECK-NOT: VecLoad
+ /// CHECK-NOT: VecStore
//
- // TODO: fill in when supported
+ // TODO: fill in when long2double is supported
static void conv(long[] b) {
for (int i = 0; i < 128; i++)
a[i] = b[i];