Use float64x2 for _mm_stream_pd temporal store

Replace previously used float32x4 which leads to compilation error when using clang++.
JishinMaster · Apr 21, 2022 · 72a657b · 72a657b
1 parent 11bc6dc
commit 72a657b
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/sse2neon.h b/sse2neon.h
@@ -5989,7 +5989,7 @@ FORCE_INLINE void _mm_storeu_si32(void *p, __m128i a)
 FORCE_INLINE void _mm_stream_pd(double *p, __m128d a)
 {
 #if __has_builtin(__builtin_nontemporal_store)
-    __builtin_nontemporal_store(a, (float32x4_t *) p);
+    __builtin_nontemporal_store(a, (float64x2_t *) p);
 #elif defined(__aarch64__)
     vst1q_f64(p, vreinterpretq_f64_m128d(a));
 #else