Fix more

Author: AntoinePrv

include/xsimd/arch/xsimd_avx.hpp

@@ -1035,20 +1035,21 @@ namespace xsimd
         XSIMD_INLINE void store_masked(T* mem, batch<T, A> const& src, batch_bool_constant<T, A, Values...> mask, Mode, requires_arch<avx>) noexcept
         {
             constexpr size_t half_size = batch<T, A>::size / 2;
+            using half_arch = avx_128;
 
             // confined to lower 128-bit half → forward to 128 bit
             XSIMD_IF_CONSTEXPR(mask.countl_zero() >= half_size)
             {
-                constexpr auto mlo = ::xsimd::detail::lower_half<sse4_2>(mask);
-                const auto lo = xsimd::batch<T, sse4_2>(detail::lower_half(src));
-                store_masked<avx_128>(mem, lo, mlo, Mode {}, sse4_2 {});
+                constexpr auto mlo = ::xsimd::detail::lower_half<half_arch>(mask);
+                const auto lo = xsimd::batch<T, half_arch>(detail::lower_half(src));
+                store_masked<half_arch>(mem, lo, mlo, Mode {}, half_arch {});
             }
             // confined to upper 128-bit half → forward to 128 bit
             else XSIMD_IF_CONSTEXPR(mask.countr_zero() >= half_size)
             {
-                constexpr auto mhi = ::xsimd::detail::upper_half<sse4_2>(mask);
-                const auto hi = xsimd::batch<T, sse4_2>(detail::upper_half(src));
-                store_masked<avx_128>(mem + half_size, hi, mhi, Mode {}, sse4_2 {});
+                constexpr auto mhi = ::xsimd::detail::upper_half<half_arch>(mask);
+                const auto hi = xsimd::batch<T, half_arch>(detail::upper_half(src));
+                store_masked<half_arch>(mem + half_size, hi, mhi, Mode {}, half_arch {});
             }
             else
             {

include/xsimd/arch/xsimd_avx2.hpp

@@ -195,20 +195,21 @@ namespace xsimd
         XSIMD_INLINE void store_masked(T* mem, batch<T, A> const& src, batch_bool_constant<T, A, Values...> mask, Mode, requires_arch<avx2>) noexcept
         {
             constexpr size_t lanes_per_half = batch<T, A>::size / 2;
+            using half_arch = avx2_128;
 
             // confined to lower 128-bit half → forward to SSE
             XSIMD_IF_CONSTEXPR(mask.countl_zero() >= lanes_per_half)
             {
-                constexpr auto mlo = ::xsimd::detail::lower_half<sse4_2>(mask);
-                const auto lo = detail::lower_half(src);
-                store_masked<sse4_2>(mem, lo, mlo, Mode {}, sse4_2 {});
+                constexpr auto mlo = ::xsimd::detail::lower_half<half_arch>(mask);
+                const auto lo = xsimd::batch<T, half_arch>(detail::lower_half(src));
+                store_masked<half_arch>(mem, lo, mlo, Mode {}, half_arch {});
             }
             // confined to upper 128-bit half → forward to SSE
             else XSIMD_IF_CONSTEXPR(mask.countr_zero() >= lanes_per_half)
             {
-                constexpr auto mhi = ::xsimd::detail::upper_half<sse4_2>(mask);
-                const auto hi = detail::upper_half(src);
-                store_masked<sse4_2>(mem + lanes_per_half, hi, mhi, Mode {}, sse4_2 {});
+                constexpr auto mhi = ::xsimd::detail::upper_half<half_arch>(mask);
+                const auto hi = xsimd::batch<T, half_arch>(detail::upper_half(src));
+                store_masked<half_arch>(mem + lanes_per_half, hi, mhi, Mode {}, half_arch {});
             }
             else
             {