use SSSE3 consistently as opposed to SSE3

CMakeLists.txt

@@ -31,7 +31,7 @@ cmake_minimum_required(VERSION 2.8.7)
 set (randomx_sources
 src/aes_hash.cpp
 src/argon2_ref.c
-src/argon2_sse3.c
+src/argon2_ssse3.c
 src/argon2_avx2.c
 src/bytecode_machine.cpp
 src/dataset.cpp
@@ -107,7 +107,7 @@ if (ARCH_ID STREQUAL "x86_64" OR ARCH_ID STREQUAL "x86-64" OR ARCH_ID STREQUAL "
     add_flag("-maes")
     check_c_compiler_flag(-mssse3 HAVE_SSSE3)
     if(HAVE_SSSE3)
-      set_source_files_properties(src/argon2_sse3.c COMPILE_FLAGS -mssse3)
+      set_source_files_properties(src/argon2_ssse3.c COMPILE_FLAGS -mssse3)
     endif()
     check_c_compiler_flag(-mavx2 HAVE_AVX2)
     if(HAVE_AVX2)

src/argon2.h

@@ -253,7 +253,7 @@ extern "C" {
 void randomx_argon2_fill_segment_ref(const argon2_instance_t* instance,
 	argon2_position_t position);
 
-randomx_argon2_impl *randomx_argon2_impl_sse3();
+randomx_argon2_impl *randomx_argon2_impl_ssse3();
 randomx_argon2_impl *randomx_argon2_impl_avx2();
 
 #if defined(__cplusplus)

src/argon2_sse3.c → src/argon2_ssse3.c

@@ -42,12 +42,12 @@ OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 #define __SSSE3__
 #endif
 
-void randomx_argon2_fill_segment_sse3(const argon2_instance_t* instance,
+void randomx_argon2_fill_segment_ssse3(const argon2_instance_t* instance,
 	argon2_position_t position);
 
-randomx_argon2_impl* randomx_argon2_impl_sse3() {
+randomx_argon2_impl* randomx_argon2_impl_ssse3() {
 #if defined(__SSSE3__)
-	return &randomx_argon2_fill_segment_sse3;
+	return &randomx_argon2_fill_segment_ssse3;
 #endif
 	return NULL;
 }
@@ -58,7 +58,7 @@ randomx_argon2_impl* randomx_argon2_impl_sse3() {
 
 #include "argon2_core.h"
 
-#include "blake2/blamka-round-sse3.h"
+#include "blake2/blamka-round-ssse3.h"
 #include "blake2/blake2-impl.h"
 #include "blake2/blake2.h"
 
@@ -100,7 +100,7 @@ static void fill_block(__m128i* state, const block* ref_block,
 	}
 }
 
-void randomx_argon2_fill_segment_sse3(const argon2_instance_t* instance,
+void randomx_argon2_fill_segment_ssse3(const argon2_instance_t* instance,
 	argon2_position_t position) {
 	block* ref_block = NULL, * curr_block = NULL;
 	block address_block, input_block;

src/dataset.hpp

@@ -87,8 +87,8 @@ namespace randomx {
 			return &randomx_argon2_fill_segment_ref;
 		}
 		randomx_argon2_impl* impl = nullptr;
-		if ((flags & RANDOMX_FLAG_ARGON2) == RANDOMX_FLAG_ARGON2_SSE3) {
-			impl = randomx_argon2_impl_sse3();
+		if ((flags & RANDOMX_FLAG_ARGON2) == RANDOMX_FLAG_ARGON2_SSSE3) {
+			impl = randomx_argon2_impl_ssse3();
 		}
 		if ((flags & RANDOMX_FLAG_ARGON2) == RANDOMX_FLAG_ARGON2_AVX2) {
 			impl = randomx_argon2_impl_avx2();

src/randomx.h

@@ -45,7 +45,7 @@ typedef enum {
   RANDOMX_FLAG_FULL_MEM = 4,
   RANDOMX_FLAG_JIT = 8,
   RANDOMX_FLAG_SECURE = 16,
-  RANDOMX_FLAG_ARGON2_SSE3 = 32,
+  RANDOMX_FLAG_ARGON2_SSSE3 = 32,
   RANDOMX_FLAG_ARGON2_AVX2 = 64,
   RANDOMX_FLAG_ARGON2 = 96
 } randomx_flags;
@@ -66,7 +66,7 @@ extern "C" {
  *        RANDOMX_FLAG_JIT - create cache structure with JIT compilation support; this makes
  *                           subsequent Dataset initialization faster
  *        Optionally, one of these two flags may be selected:
- *        RANDOMX_FLAG_ARGON2_SSE3 - optimized Argon2 for CPUs with the SSSE3 instruction set
+ *        RANDOMX_FLAG_ARGON2_SSSE3 - optimized Argon2 for CPUs with the SSSE3 instruction set
  *                                   makes subsequent cache initialization faster
  *        RANDOMX_FLAG_ARGON2_AVX2 - optimized Argon2 for CPUs with the AVX2 instruction set
  *                                   makes subsequent cache initialization faster

src/tests/benchmark.cpp

@@ -91,7 +91,7 @@ void printUsage(const char* executable) {
 	std::cout << "  --init Q      initialize dataset with Q threads (default: 1)" << std::endl;
 	std::cout << "  --nonces N    run N nonces (default: 1000)" << std::endl;
 	std::cout << "  --seed S      seed for cache initialization (default: 0)" << std::endl;
-	std::cout << "  --sse3        use optimized Argon2 for SSSE3 CPUs" << std::endl;
+	std::cout << "  --ssse3       use optimized Argon2 for SSSE3 CPUs" << std::endl;
 	std::cout << "  --avx2        use optimized Argon2 for AVX2 CPUs" << std::endl;
 }
 
@@ -130,7 +130,7 @@ void mine(randomx_vm* vm, std::atomic<uint32_t>& atomicNonce, AtomicHash& result
 }
 
 int main(int argc, char** argv) {
-	bool softAes, miningMode, verificationMode, help, largePages, jit, secure, sse3, avx2;
+	bool softAes, miningMode, verificationMode, help, largePages, jit, secure, ssse3, avx2;
 	int noncesCount, threadCount, initThreadCount;
 	uint64_t threadAffinity;
 	int32_t seedValue;
@@ -151,7 +151,7 @@ int main(int argc, char** argv) {
 	readOption("--jit", argc, argv, jit);
 	readOption("--help", argc, argv, help);
 	readOption("--secure", argc, argv, secure);
-	readOption("--sse3", argc, argv, sse3);
+	readOption("--ssse3", argc, argv, ssse3);
 	readOption("--avx2", argc, argv, avx2);
 
 	store32(&seed, seedValue);
@@ -171,9 +171,9 @@ int main(int argc, char** argv) {
 	randomx_cache* cache;
 	randomx_flags flags = RANDOMX_FLAG_DEFAULT;
 
-	if (sse3) {
-		flags = (randomx_flags)(flags | RANDOMX_FLAG_ARGON2_SSE3);
-		std::cout << " - Argon2 implementation: SSE3" << std::endl;
+	if (ssse3) {
+		flags = (randomx_flags)(flags | RANDOMX_FLAG_ARGON2_SSSE3);
+		std::cout << " - Argon2 implementation: SSSE3" << std::endl;
 	}
 
 	if (avx2) {

src/tests/tests.cpp

@@ -1018,7 +1018,7 @@ int main() {
 	vm = nullptr;
 
 	randomx_release_cache(cache);
-	cache = randomx_alloc_cache(RANDOMX_FLAG_ARGON2_SSE3);
+	cache = randomx_alloc_cache(RANDOMX_FLAG_ARGON2_SSSE3);
 
 	runTest("Cache initialization: SSSE3", cache != nullptr && RANDOMX_ARGON_ITERATIONS == 3 && RANDOMX_ARGON_LANES == 1 && RANDOMX_ARGON_MEMORY == 262144 && stringsEqual(RANDOMX_ARGON_SALT, "RandomX\x03"), []() {
 		initCache("test key 000");

vcxproj/randomx.vcxproj

@@ -135,7 +135,7 @@ SET ERRORLEVEL = 0</Command>
     <ClCompile Include="..\src\argon2_avx2.c" />
     <ClCompile Include="..\src\argon2_core.c" />
     <ClCompile Include="..\src\argon2_ref.c" />
-    <ClCompile Include="..\src\argon2_sse3.c" />
+    <ClCompile Include="..\src\argon2_ssse3.c" />
     <ClCompile Include="..\src\assembly_generator_x86.cpp" />
     <ClCompile Include="..\src\blake2_generator.cpp" />
     <ClCompile Include="..\src\blake2\blake2b.c" />
@@ -168,7 +168,7 @@ SET ERRORLEVEL = 0</Command>
     <ClInclude Include="..\src\blake2\blake2.h" />
     <ClInclude Include="..\src\blake2\blamka-round-avx2.h" />
     <ClInclude Include="..\src\blake2\blamka-round-ref.h" />
-    <ClInclude Include="..\src\blake2\blamka-round-sse3.h" />
+    <ClInclude Include="..\src\blake2\blamka-round-ssse3.h" />
     <ClInclude Include="..\src\blake2\endian.h" />
     <ClInclude Include="..\src\blake2_generator.hpp" />
     <ClInclude Include="..\src\bytecode_machine.hpp" />

vcxproj/randomx.vcxproj.filters

@@ -81,10 +81,10 @@
     <ClCompile Include="..\src\bytecode_machine.cpp">
       <Filter>Source Files</Filter>
     </ClCompile>
-    <ClCompile Include="..\src\argon2_sse3.c">
+    <ClCompile Include="..\src\argon2_avx2.c">
       <Filter>Source Files</Filter>
     </ClCompile>
-    <ClCompile Include="..\src\argon2_avx2.c">
+    <ClCompile Include="..\src\argon2_ssse3.c">
       <Filter>Source Files</Filter>
     </ClCompile>
   </ItemGroup>
@@ -191,10 +191,10 @@
     <ClInclude Include="..\src\bytecode_machine.hpp">
       <Filter>Header Files</Filter>
     </ClInclude>
-    <ClInclude Include="..\src\blake2\blamka-round-sse3.h">
+    <ClInclude Include="..\src\blake2\blamka-round-avx2.h">
       <Filter>Header Files</Filter>
     </ClInclude>
-    <ClInclude Include="..\src\blake2\blamka-round-avx2.h">
+    <ClInclude Include="..\src\blake2\blamka-round-ssse3.h">
       <Filter>Header Files</Filter>
     </ClInclude>
   </ItemGroup>