neon16

Author: serge-sans-paille

include/xsimd/arch/xsimd_neon.hpp

@@ -731,14 +731,28 @@ namespace xsimd
         XSIMD_INLINE void store(batch_bool<T, A> b, bool* mem, requires_arch<neon>) noexcept
         {
             uint8x16_t val = vsubq_u8(vdupq_n_u8(0), b.data);
-            vst1q_u8((unsigned char*)mem, val);
+            vst1q_u8((uint8_t*)mem, val);
         }
 
-        // template <class A>
-        // XSIMD_INLINE void store(batch_bool<float, A> b, bool* mem, requires_arch<sse2>) noexcept
-        //{
-        //     store(batch_bool<uint32_t, A>(b.data), mem, A {});
-        // }
+        template <class T, class A, detail::enable_sized_t<T, 2> = 0>
+        XSIMD_INLINE void store(batch_bool<T, A> b, bool* mem, requires_arch<neon>) noexcept
+        {
+            uint8x8_t val = vsub_u8(vdup_n_u8(0), vqmovn_u16(b.data));
+            vst1_u8((uint8_t*)mem, val);
+        }
+
+        template <class T, class A, detail::enable_sized_t<T, 4> = 0>
+        XSIMD_INLINE void store(batch_bool<T, A> b, bool* mem, requires_arch<neon>) noexcept
+        {
+            uint8x8_t val = vsub_u8(vdup_n_u8(0), vqmovn_u16(vcombine_u16(vqmovn_u32(b.data), vdup_n_u16(0))));
+            vst1_lane_u32((uint32_t*)mem, vreinterpret_u32_u8(val), 0);
+        }
+
+        template <class A>
+        XSIMD_INLINE void store(batch_bool<float, A> b, bool* mem, requires_arch<neon>) noexcept
+        {
+            store(batch_bool<uint32_t, A>(b.data), mem, A {});
+        }
 
         /*******
          * neg *