diff --git a/.gitignore b/.gitignore
index a70a8a5..c2de60e 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,3 +1,6 @@
 .vs/
 msvc.build/
 *.vcxproj.user
+*.obj
+*.exe
+*.lib
diff --git a/makefile.vs b/makefile.vs
index d2f407a..1dc094e 100644
--- a/makefile.vs
+++ b/makefile.vs
@@ -16,16 +16,17 @@ LIB_LIB = libic.lib
 LIB_DLL = ic.dll
 LIB_IMP = ic.lib
 
-OBJS = bitpack.obj bitunpack.obj vp4c.obj vp4d.obj transpose.obj bitutil.obj fp.obj vsimple.obj vint.obj
+OBJS = bitpack.obj bitunpack.obj bitutil.obj vp4c.obj vp4d.obj transpose.obj fp.obj trlec.obj trled.obj vint.obj vsimple.obj vsimple.obj
+OBJS_SSE = vs\bitpack_sse.obj vs\bitunpack_sse.obj vs\vp4c_sse.obj vs\vp4d_sse.obj vs\transpose_sse.obj
+OBJS_AVX2 = vs\bitpack_avx2.obj vs\bitunpack_avx2.obj vs\vp4c_avx2.obj vs\vp4d_avx2.obj vs\transpose_avx2.obj
 
 !if "$(NSIMD)" == "1"
 CFLAGS = $(CFLAGS) /DNSIMD
 !else
-OBJS = $(OBJS) transpose_sse.obj bitpack_sse.obj bitunpack_sse.obj vp4c_sse.obj vp4d_sse.obj
+OBJS = $(OBJS) $(OBJS_SSE)
 CFLAGS = $(CFLAGS) /D__SSE__ /D__SSE2__ /D__SSE3__ /D__SSSE3__ /D__SSE4_1__ /D__SSE4_2__ /DUSE_SSE
-ARCH = /arch:SSE2
 !if "$(AVX2)" == "1"
-OBJS = $(OBJS) bitpack_avx2.obj bitunpack_avx2.obj transpose_avx2.obj vp4c_avx2.obj vp4d_avx2.obj 
+OBJS = $(OBJS) $(OBJS_AVX2)
 CFLAGS = $(CFLAGS) /D__AVX2__ /DUSE_AVX2
 ARCH = /arch:AVX2
 !endif
@@ -45,7 +46,7 @@ CFLAGS = $(CFLAGS) /DBLOSC
 
 DLL_OBJS = $(OBJS:.obj=.dllobj)
 
-all: $(LIB_LIB) icbench.exe 
+all: $(LIB_LIB) icbench.exe icapp.exe
 
 #$(LIB_DLL) $(LIB_IMP) 
 
@@ -67,8 +68,11 @@ $(LIB_DLL): $(DLL_OBJS)
 
 $(LIB_IMP): $(LIB_DLL)
 
-icbench.exe: icbench.obj vs/getopt.obj plugins.obj eliasfano.obj vsimple.obj $(LIB_LIB)
+icbench.exe: icbench.obj vs\getopt.obj plugins.obj eliasfano.obj $(LIB_LIB)
+	$(LD) $(LDFLAGS) -out:$@ $**
+
+icapp.exe: icapp.obj vs\getopt.obj plugins.obj eliasfano.obj $(LIB_LIB)
 	$(LD) $(LDFLAGS) -out:$@ $**
 
 clean:
-	-del *.dll *.exe *.exp *.obj *.dllobj *.lib *.manifest 2>nul
+	-del *.obj vs\*.obj *.dll *.exe *.exp *.dllobj *.lib *.manifest 2>nul
diff --git a/plugins.cc b/plugins.cc
index aa40dd5..c692e5b 100644
--- a/plugins.cc
+++ b/plugins.cc
@@ -708,7 +708,7 @@ unsigned char *coddecomp(unsigned char *in, unsigned _n, unsigned char *_out, in
  	  #if C_TURBOPFOR256
     case TB_FOR256V: vbxget32(in, x); b = *in++;return bitfunpack256v32( in, n, out, x, b);
 
-    case TB_PF256V : __builtin_prefetch(in+256);return n == 256?p4dec256v32(in, n, out):p4dec32(in, n, out);
+    case TB_PF256V : __builtin_prefetch((char*)in+256);return n == 256?p4dec256v32(in, n, out):p4dec32(in, n, out);
     case TB_PFN256V :      				 		return in+p4ndec256v32(in, n, out);
     case TB_BP256V: if(b < 0) b = *in++; 		return n != 256?bitunpack32(in, n, out, b):bitunpack256v32(in, n, out, b);
     case TB_BPN256V:  		                    return in+bitnunpack256v32(in, n, out);
diff --git a/bitpack_avx2.c b/vs/bitpack_avx2.c
similarity index 100%
rename from bitpack_avx2.c
rename to vs/bitpack_avx2.c
diff --git a/bitpack_sse.c b/vs/bitpack_sse.c
similarity index 100%
rename from bitpack_sse.c
rename to vs/bitpack_sse.c
diff --git a/bitunpack_avx2.c b/vs/bitunpack_avx2.c
similarity index 100%
rename from bitunpack_avx2.c
rename to vs/bitunpack_avx2.c
diff --git a/bitunpack_sse.c b/vs/bitunpack_sse.c
similarity index 100%
rename from bitunpack_sse.c
rename to vs/bitunpack_sse.c
diff --git a/transpose_avx2.c b/vs/transpose_avx2.c
similarity index 100%
rename from transpose_avx2.c
rename to vs/transpose_avx2.c
diff --git a/transpose_sse.c b/vs/transpose_sse.c
similarity index 100%
rename from transpose_sse.c
rename to vs/transpose_sse.c
diff --git a/vp4c_avx2.c b/vs/vp4c_avx2.c
similarity index 100%
rename from vp4c_avx2.c
rename to vs/vp4c_avx2.c
diff --git a/vp4c_sse.c b/vs/vp4c_sse.c
similarity index 100%
rename from vp4c_sse.c
rename to vs/vp4c_sse.c
diff --git a/vp4d_avx2.c b/vs/vp4d_avx2.c
similarity index 100%
rename from vp4d_avx2.c
rename to vs/vp4d_avx2.c
diff --git a/vp4d_sse.c b/vs/vp4d_sse.c
similarity index 100%
rename from vp4d_sse.c
rename to vs/vp4d_sse.c
diff --git a/vs/vs2017/TurboPFor.vcxproj b/vs/vs2017/TurboPFor.vcxproj
index e4fbb38..c89d5c6 100644
--- a/vs/vs2017/TurboPFor.vcxproj
+++ b/vs/vs2017/TurboPFor.vcxproj
@@ -95,6 +95,7 @@
       <RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary>
       <MultiProcessorCompilation>true</MultiProcessorCompilation>
       <AdditionalOptions>/w24003 /w24005 /w24028 /w24047 /w24090 /w24133 /w24146 /w24333 /w24789 %(AdditionalOptions)</AdditionalOptions>
+      <AdditionalIncludeDirectories>..\..</AdditionalIncludeDirectories>
     </ClCompile>
     <Link>
       <SubSystem>Console</SubSystem>
@@ -111,6 +112,7 @@
       <RuntimeLibrary>MultiThreadedDebug</RuntimeLibrary>
       <MultiProcessorCompilation>true</MultiProcessorCompilation>
       <AdditionalOptions>/w24003 /w24005 /w24028 /w24047 /w24090 /w24133 /w24146 /w24333 /w24789 %(AdditionalOptions)</AdditionalOptions>
+      <AdditionalIncludeDirectories>..\..</AdditionalIncludeDirectories>
     </ClCompile>
     <Link>
       <SubSystem>Console</SubSystem>
@@ -130,6 +132,7 @@
       <RuntimeLibrary>MultiThreaded</RuntimeLibrary>
       <MultiProcessorCompilation>true</MultiProcessorCompilation>
       <AdditionalOptions>/w24003 /w24005 /w24028 /w24047 /w24090 /w24133 /w24146 /w24333 /w24789 %(AdditionalOptions)</AdditionalOptions>
+      <AdditionalIncludeDirectories>..\..</AdditionalIncludeDirectories>
     </ClCompile>
     <Link>
       <SubSystem>Console</SubSystem>
@@ -151,6 +154,7 @@
       <RuntimeLibrary>MultiThreaded</RuntimeLibrary>
       <MultiProcessorCompilation>true</MultiProcessorCompilation>
       <AdditionalOptions>/w24003 /w24005 /w24028 /w24047 /w24090 /w24133 /w24146 /w24333 /w24789 %(AdditionalOptions)</AdditionalOptions>
+      <AdditionalIncludeDirectories>..\..</AdditionalIncludeDirectories>
     </ClCompile>
     <Link>
       <SubSystem>Console</SubSystem>
@@ -161,26 +165,26 @@
   </ItemDefinitionGroup>
   <ItemGroup>
     <ClCompile Include="..\..\bitpack.c" />
-    <ClCompile Include="..\..\bitpack_avx2.c" />
-    <ClCompile Include="..\..\bitpack_sse.c" />
     <ClCompile Include="..\..\bitunpack.c" />
-    <ClCompile Include="..\..\bitunpack_avx2.c" />
-    <ClCompile Include="..\..\bitunpack_sse.c" />
     <ClCompile Include="..\..\bitutil.c" />
     <ClCompile Include="..\..\fp.c" />
     <ClCompile Include="..\..\transpose.c" />
-    <ClCompile Include="..\..\transpose_avx2.c" />
-    <ClCompile Include="..\..\transpose_sse.c" />
     <ClCompile Include="..\..\trlec.c" />
     <ClCompile Include="..\..\trled.c" />
     <ClCompile Include="..\..\vint.c" />
     <ClCompile Include="..\..\vp4c.c" />
-    <ClCompile Include="..\..\vp4c_avx2.c" />
-    <ClCompile Include="..\..\vp4c_sse.c" />
     <ClCompile Include="..\..\vp4d.c" />
-    <ClCompile Include="..\..\vp4d_avx2.c" />
-    <ClCompile Include="..\..\vp4d_sse.c" />
     <ClCompile Include="..\..\vsimple.c" />
+    <ClCompile Include="..\bitpack_avx2.c" />
+    <ClCompile Include="..\bitpack_sse.c" />
+    <ClCompile Include="..\bitunpack_avx2.c" />
+    <ClCompile Include="..\bitunpack_sse.c" />
+    <ClCompile Include="..\transpose_avx2.c" />
+    <ClCompile Include="..\transpose_sse.c" />
+    <ClCompile Include="..\vp4c_avx2.c" />
+    <ClCompile Include="..\vp4c_sse.c" />
+    <ClCompile Include="..\vp4d_avx2.c" />
+    <ClCompile Include="..\vp4d_sse.c" />
   </ItemGroup>
   <ItemGroup>
     <ClInclude Include="..\..\bitpack.h" />
diff --git a/vs/vs2017/TurboPFor.vcxproj.filters b/vs/vs2017/TurboPFor.vcxproj.filters
index 3c94a99..cbaf23a 100644
--- a/vs/vs2017/TurboPFor.vcxproj.filters
+++ b/vs/vs2017/TurboPFor.vcxproj.filters
@@ -38,42 +38,42 @@
     <ClCompile Include="..\..\vsimple.c">
       <Filter>Source Files</Filter>
     </ClCompile>
-    <ClCompile Include="..\..\bitpack_avx2.c">
-      <Filter>Source Files</Filter>
-    </ClCompile>
-    <ClCompile Include="..\..\bitpack_sse.c">
-      <Filter>Source Files</Filter>
-    </ClCompile>
-    <ClCompile Include="..\..\bitunpack_avx2.c">
-      <Filter>Source Files</Filter>
-    </ClCompile>
-    <ClCompile Include="..\..\bitunpack_sse.c">
-      <Filter>Source Files</Filter>
-    </ClCompile>
-    <ClCompile Include="..\..\transpose_avx2.c">
-      <Filter>Source Files</Filter>
-    </ClCompile>
-    <ClCompile Include="..\..\transpose_sse.c">
-      <Filter>Source Files</Filter>
-    </ClCompile>
-    <ClCompile Include="..\..\vp4c_avx2.c">
-      <Filter>Source Files</Filter>
-    </ClCompile>
-    <ClCompile Include="..\..\vp4c_sse.c">
-      <Filter>Source Files</Filter>
-    </ClCompile>
-    <ClCompile Include="..\..\vp4d_avx2.c">
-      <Filter>Source Files</Filter>
-    </ClCompile>
-    <ClCompile Include="..\..\vp4d_sse.c">
-      <Filter>Source Files</Filter>
-    </ClCompile>
     <ClCompile Include="..\..\trlec.c">
       <Filter>Source Files</Filter>
     </ClCompile>
     <ClCompile Include="..\..\trled.c">
       <Filter>Source Files</Filter>
     </ClCompile>
+    <ClCompile Include="..\bitpack_avx2.c">
+      <Filter>Source Files</Filter>
+    </ClCompile>
+    <ClCompile Include="..\bitpack_sse.c">
+      <Filter>Source Files</Filter>
+    </ClCompile>
+    <ClCompile Include="..\bitunpack_avx2.c">
+      <Filter>Source Files</Filter>
+    </ClCompile>
+    <ClCompile Include="..\bitunpack_sse.c">
+      <Filter>Source Files</Filter>
+    </ClCompile>
+    <ClCompile Include="..\transpose_avx2.c">
+      <Filter>Source Files</Filter>
+    </ClCompile>
+    <ClCompile Include="..\transpose_sse.c">
+      <Filter>Source Files</Filter>
+    </ClCompile>
+    <ClCompile Include="..\vp4c_avx2.c">
+      <Filter>Source Files</Filter>
+    </ClCompile>
+    <ClCompile Include="..\vp4c_sse.c">
+      <Filter>Source Files</Filter>
+    </ClCompile>
+    <ClCompile Include="..\vp4d_avx2.c">
+      <Filter>Source Files</Filter>
+    </ClCompile>
+    <ClCompile Include="..\vp4d_sse.c">
+      <Filter>Source Files</Filter>
+    </ClCompile>
   </ItemGroup>
   <ItemGroup>
     <ClInclude Include="..\..\bitpack.h">