diff --git a/CMakeLists.txt b/CMakeLists.txt
index 612cf3b..22a365a 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -158,3 +158,43 @@ install(FILES
     ${CMAKE_CURRENT_BINARY_DIR}/meshoptimizerConfigVersion.cmake
     COMPONENT meshoptimizer
     DESTINATION ${CMAKE_INSTALL_LIBDIR}/cmake/meshoptimizer)
+
+##################################################
+# codecbench
+##################################################
+add_executable(codecbench tools/codecbench.cpp ${SOURCES})
+
+set_target_properties(codecbench PROPERTIES OUTPUT_NAME codecbench.wasm)
+
+target_compile_options(codecbench
+  PUBLIC
+    -O3 -msimd128
+    -std=c++11
+    -Wno-unused-function
+    -Wno-unused-variable
+)
+
+target_link_options(codecbench
+  PUBLIC
+    LINKER:-allow-undefined,--demangle,--export=malloc,--export=free
+)
+
+find_program(WASM_OPT
+    NAMES wasm-opt
+    PATHS /opt/binaryen-version_97/bin /opt/binaryen/bin
+)
+
+if (NOT WASM_OPT)
+  message(WARNING "can not find wasm-opt and will not optimize any wasm module")
+endif()
+
+add_custom_target(codecbench.opt ALL
+  COMMAND
+    ${WASM_OPT} -Oz --enable-simd -o codecbench.opt.wasm codecbench.wasm
+  BYPRODUCTS
+    ${CMAKE_CURRENT_BINARY_DIR}/codecbench.opt.wasm
+  WORKING_DIRECTORY
+    ${CMAKE_CURRENT_BINARY_DIR}
+)
+
+add_dependencies(codecbench.opt codecbench)
diff --git a/src/vertexcodec.cpp b/src/vertexcodec.cpp
index 4bd1112..257c258 100644
--- a/src/vertexcodec.cpp
+++ b/src/vertexcodec.cpp
@@ -89,13 +89,13 @@
 #endif
 
 #ifdef SIMD_WASM
-#define wasmx_splat_v32x4(v, i) wasm_v32x4_shuffle(v, v, i, i, i, i)
-#define wasmx_unpacklo_v8x16(a, b) wasm_v8x16_shuffle(a, b, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23)
-#define wasmx_unpackhi_v8x16(a, b) wasm_v8x16_shuffle(a, b, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31)
-#define wasmx_unpacklo_v16x8(a, b) wasm_v16x8_shuffle(a, b, 0, 8, 1, 9, 2, 10, 3, 11)
-#define wasmx_unpackhi_v16x8(a, b) wasm_v16x8_shuffle(a, b, 4, 12, 5, 13, 6, 14, 7, 15)
-#define wasmx_unpacklo_v64x2(a, b) wasm_v64x2_shuffle(a, b, 0, 2)
-#define wasmx_unpackhi_v64x2(a, b) wasm_v64x2_shuffle(a, b, 1, 3)
+#define wasmx_splat_v32x4(v, i) wasm_i32x4_shuffle(v, v, i, i, i, i)
+#define wasmx_unpacklo_v8x16(a, b) wasm_i8x16_shuffle(a, b, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23)
+#define wasmx_unpackhi_v8x16(a, b) wasm_i8x16_shuffle(a, b, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31)
+#define wasmx_unpacklo_v16x8(a, b) wasm_i16x8_shuffle(a, b, 0, 8, 1, 9, 2, 10, 3, 11)
+#define wasmx_unpackhi_v16x8(a, b) wasm_i16x8_shuffle(a, b, 4, 12, 5, 13, 6, 14, 7, 15)
+#define wasmx_unpacklo_v64x2(a, b) wasm_i64x2_shuffle(a, b, 0, 2)
+#define wasmx_unpackhi_v64x2(a, b) wasm_i64x2_shuffle(a, b, 1, 3)
 #endif
 
 namespace meshopt
@@ -757,7 +757,7 @@ static v128_t decodeShuffleMask(unsigned char mask0, unsigned char mask1)
 	v128_t sm1 = wasm_v128_load(&kDecodeBytesGroupShuffle[mask1]);
 
 	v128_t sm1off = wasm_v128_load(&kDecodeBytesGroupCount[mask0]);
-	sm1off = wasm_v8x16_shuffle(sm1off, sm1off, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
+	sm1off = wasm_i8x16_shuffle(sm1off, sm1off, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
 
 	v128_t sm1r = wasm_i8x16_add(sm1, sm1off);
 
@@ -807,7 +807,7 @@ static const unsigned char* decodeBytesGroupSimd(const unsigned char* data, unsi
 
 		v128_t shuf = decodeShuffleMask(mask0, mask1);
 
-		v128_t result = wasm_v128_bitselect(wasm_v8x16_swizzle(rest, shuf), sel, mask);
+		v128_t result = wasm_v128_bitselect(wasm_i8x16_swizzle(rest, shuf), sel, mask);
 
 		wasm_v128_store(buffer, result);
 
@@ -829,7 +829,7 @@ static const unsigned char* decodeBytesGroupSimd(const unsigned char* data, unsi
 
 		v128_t shuf = decodeShuffleMask(mask0, mask1);
 
-		v128_t result = wasm_v128_bitselect(wasm_v8x16_swizzle(rest, shuf), sel, mask);
+		v128_t result = wasm_v128_bitselect(wasm_i8x16_swizzle(rest, shuf), sel, mask);
 
 		wasm_v128_store(buffer, result);
 
diff --git a/src/vertexfilter.cpp b/src/vertexfilter.cpp
index 5c7589c..c79cad4 100644
--- a/src/vertexfilter.cpp
+++ b/src/vertexfilter.cpp
@@ -57,10 +57,10 @@
 #endif
 
 #ifdef SIMD_WASM
-#define wasmx_unpacklo_v16x8(a, b) wasm_v16x8_shuffle(a, b, 0, 8, 1, 9, 2, 10, 3, 11)
-#define wasmx_unpackhi_v16x8(a, b) wasm_v16x8_shuffle(a, b, 4, 12, 5, 13, 6, 14, 7, 15)
-#define wasmx_unziplo_v32x4(a, b) wasm_v32x4_shuffle(a, b, 0, 2, 4, 6)
-#define wasmx_unziphi_v32x4(a, b) wasm_v32x4_shuffle(a, b, 1, 3, 5, 7)
+#define wasmx_unpacklo_v16x8(a, b) wasm_i16x8_shuffle(a, b, 0, 8, 1, 9, 2, 10, 3, 11)
+#define wasmx_unpackhi_v16x8(a, b) wasm_i16x8_shuffle(a, b, 4, 12, 5, 13, 6, 14, 7, 15)
+#define wasmx_unziplo_v32x4(a, b) wasm_i32x4_shuffle(a, b, 0, 2, 4, 6)
+#define wasmx_unziphi_v32x4(a, b) wasm_i32x4_shuffle(a, b, 1, 3, 5, 7)
 #endif
 
 #ifndef __has_builtin