diff --git a/.appveyor.yml b/.appveyor.yml
index 90e9d4cac..182c0f2f7 100644
--- a/.appveyor.yml
+++ b/.appveyor.yml
@@ -9,6 +9,10 @@ environment:
   global:
     MINICONDA: C:\xsimd-conda
   matrix:
+    - JOB: "SSE2"
+      CXXFLAGS: ""
+      VCVARSALL: "C:\\Program Files (x86)\\Microsoft Visual Studio 14.0\\VC\\vcvarsall.bat"
+      RUNTEST: ".\\test_xsimd"
     - JOB: "AVX2"
       CXXFLAGS: "/arch:AVX2"
       VCVARSALL: "C:\\Program Files (x86)\\Microsoft Visual Studio 14.0\\VC\\vcvarsall.bat"
diff --git a/include/xsimd/arch/generic/xsimd_generic_memory.hpp b/include/xsimd/arch/generic/xsimd_generic_memory.hpp
index b0de5ff92..cfeccde1e 100644
--- a/include/xsimd/arch/generic/xsimd_generic_memory.hpp
+++ b/include/xsimd/arch/generic/xsimd_generic_memory.hpp
@@ -25,10 +25,10 @@ namespace xsimd {
 
           for (std::size_t j = 0 ; j < (size - i); ++j)
           {
-              concat_buffer[j] = self_buffer[i + j];
+              concat_buffer[j] = other_buffer[i + j];
               if(j < i)
               {
-                  concat_buffer[size - 1 - j] = other_buffer[i - 1 - j];
+                  concat_buffer[size - 1 - j] = self_buffer[i - 1 - j];
               }
           }
           return batch<T, A>::load_aligned(concat_buffer);
diff --git a/include/xsimd/arch/xsimd_neon.hpp b/include/xsimd/arch/xsimd_neon.hpp
index ac08034c2..4593ab19d 100644
--- a/include/xsimd/arch/xsimd_neon.hpp
+++ b/include/xsimd/arch/xsimd_neon.hpp
@@ -1411,7 +1411,7 @@ namespace xsimd
             {
                 if (n == I)
                 {
-                    return vextq_u8(lhs, rhs, I);
+                    return vextq_u8(rhs, lhs, I);
                 }
                 else
                 {
@@ -1424,7 +1424,7 @@ namespace xsimd
             {
                 if (n == I)
                 {
-                    return vextq_s8(lhs, rhs, I);
+                    return vextq_s8(rhs, lhs, I);
                 }
                 else
                 {
@@ -1437,7 +1437,7 @@ namespace xsimd
             {
                 if (n == I)
                 {
-                    return vextq_u16(lhs, rhs, I);
+                    return vextq_u16(rhs, lhs, I);
                 }
                 else
                 {
@@ -1450,7 +1450,7 @@ namespace xsimd
             {
                 if (n == I)
                 {
-                    return vextq_s16(lhs, rhs, I);
+                    return vextq_s16(rhs, lhs, I);
                 }
                 else
                 {
@@ -1463,7 +1463,7 @@ namespace xsimd
             {
                 if (n == I)
                 {
-                    return vextq_u32(lhs, rhs, I);
+                    return vextq_u32(rhs, lhs, I);
                 }
                 else
                 {
@@ -1476,7 +1476,7 @@ namespace xsimd
             {
                 if (n == I)
                 {
-                    return vextq_s32(lhs, rhs, I);
+                    return vextq_s32(rhs, lhs, I);
                 }
                 else
                 {
@@ -1489,7 +1489,7 @@ namespace xsimd
             {
                 if (n == I)
                 {
-                    return vextq_u64(lhs, rhs, I);
+                    return vextq_u64(rhs, lhs, I);
                 }
                 else
                 {
@@ -1502,7 +1502,7 @@ namespace xsimd
             {
                 if (n == I)
                 {
-                    return vextq_s64(lhs, rhs, I);
+                    return vextq_s64(rhs, lhs, I);
                 }
                 else
                 {
@@ -1515,7 +1515,7 @@ namespace xsimd
             {
                 if (n == I)
                 {
-                    return vextq_f32(lhs, rhs, I);
+                    return vextq_f32(rhs, lhs, I);
                 }
                 else
                 {
diff --git a/include/xsimd/arch/xsimd_neon64.hpp b/include/xsimd/arch/xsimd_neon64.hpp
index 869a7e247..0a22db16a 100644
--- a/include/xsimd/arch/xsimd_neon64.hpp
+++ b/include/xsimd/arch/xsimd_neon64.hpp
@@ -713,7 +713,7 @@ namespace xsimd
             {
                 if (n == I)
                 {
-                    return vextq_f64(lhs, rhs, I);
+                    return vextq_f64(rhs, lhs, I);
                 }
                 else
                 {
diff --git a/include/xsimd/config/xsimd_config.hpp b/include/xsimd/config/xsimd_config.hpp
index 1afbf2cfd..d7bef41ec 100644
--- a/include/xsimd/config/xsimd_config.hpp
+++ b/include/xsimd/config/xsimd_config.hpp
@@ -222,6 +222,10 @@
 #define XSIMD_WITH_SSE4_2 1
 #endif
 
+#if !defined(__clang__) && (defined(_M_X64) || (defined(_M_IX86_FP) && _M_IX86_FP >= 2))
+#define XSIMD_WITH_SSE4_2 1
+#endif
+
 #if XSIMD_WITH_SSE4_2
 #define XSIMD_WITH_SSE4_1 1
 #endif
diff --git a/test/CMakeLists.txt b/test/CMakeLists.txt
index 69f90ee84..dd8bf86a2 100644
--- a/test/CMakeLists.txt
+++ b/test/CMakeLists.txt
@@ -89,7 +89,6 @@ endif()
 
 if(CMAKE_CXX_COMPILER_ID MATCHES MSVC)
     set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /EHsc /MP /bigobj")
-    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /arch:AVX2")
     set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} /wd4244 /wd4267 /wd4005 /wd4146 /wd4800")
     set(CMAKE_EXE_LINKER_FLAGS /MANIFEST:NO)
 endif()
diff --git a/test/test_extract_pair.cpp b/test/test_extract_pair.cpp
index ec2f618bb..af64a10b1 100644
--- a/test/test_extract_pair.cpp
+++ b/test/test_extract_pair.cpp
@@ -34,10 +34,10 @@ namespace xsimd
             /* Expected shuffle data */
             for (int i = 0 ; i < (num - index); ++i)
             {
-                exped[i] = lhs_in[i + index];
+                exped[i] = rhs_in[i + index];
                 if(i < index)
                 {
-                    exped[num - 1 - i] = rhs_in[index - 1 - i];
+                    exped[num - 1 - i] = lhs_in[index - 1 - i];
                 }
             }
             vects.push_back(std::move(exped));