diff --git a/common_x86.h b/common_x86.h
index f096e9074..0cb242c4e 100644
--- a/common_x86.h
+++ b/common_x86.h
@@ -56,41 +56,65 @@ static void __inline blas_lock(volatile BLASULONG *address){
   do {
     while (*address) {YIELDING;};
 
+#if defined(_MSC_VER) && !defined(__clang__)
+	// use intrinsic instead of inline assembly
+	ret = _InterlockedExchange(address, 1);
+	// inline assembly
+	/*__asm {
+		mov eax, address
+		mov ebx, 1
+		xchg [eax], ebx
+		mov ret, ebx
+	}*/
+#else
     __asm__ __volatile__(
 			 "xchgl %0, %1\n"
 			 : "=r"(ret), "=m"(*address)
 			 : "0"(1), "m"(*address)
 			 : "memory");
+#endif
 
   } while (ret);
 
 }
 
 static __inline unsigned long long rpcc(void){
+#if defined(_MSC_VER) && !defined(__clang__)
+  return __rdtsc(); // use MSVC intrinsic
+#else
   unsigned int a, d;
 
   __asm__ __volatile__ ("rdtsc" : "=a" (a), "=d" (d));
 
   return ((unsigned long long)a + ((unsigned long long)d << 32));
+#endif
 };
 
 static __inline unsigned long getstackaddr(void){
+#if defined(_MSC_VER) && !defined(__clang__)
+  return (unsigned long)_ReturnAddress(); // use MSVC intrinsic
+#else
   unsigned long addr;
 
   __asm__ __volatile__ ("mov %%esp, %0"
 			 : "=r"(addr) : : "memory");
 
   return addr;
+#endif
 };
 
 
 static __inline long double sqrt_long(long double val) {
+#if defined(_MSC_VER) && !defined(__clang__)
+  return sqrt(val); // not sure if this will use fsqrt
+#else
   long double result;
 
   __asm__ __volatile__ ("fldt %1\n"
 		    "fsqrt\n"
 		    "fstpt %0\n" : "=m" (result) : "m"(val));
   return result;
+#endif
 }
 
 #define SQRT(a)  sqrt_long(a)
@@ -146,9 +170,14 @@ static __inline int blas_quickdivide(unsigned int x, unsigned int y){
 
   y = blas_quick_divide_table[y];
 
+#if defined(_MSC_VER) && !defined(__clang__)
+  (void*)result;
+  return x*y;
+#else
   __asm__ __volatile__  ("mull %0" :"=d" (result) :"a"(x), "0" (y));
 
   return result;
+#endif
 }
 #endif
 
diff --git a/driver/others/memory.c b/driver/others/memory.c
index fa364785b..70bfa7a57 100644
--- a/driver/others/memory.c
+++ b/driver/others/memory.c
@@ -137,8 +137,13 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
 #define BITMASK(a, b, c) ((((a) >> (b)) & (c)))
 
+#if defined(_MSC_VER) && !defined(__clang__)
+#define CONSTRUCTOR __cdecl
+#define DESTRUCTOR __cdecl
+#else
 #define CONSTRUCTOR	__attribute__ ((constructor))
 #define DESTRUCTOR	__attribute__ ((destructor))
+#endif
 
 #ifdef DYNAMIC_ARCH
 gotoblas_t *gotoblas = NULL;
@@ -1360,6 +1365,28 @@ void DESTRUCTOR gotoblas_quit(void) {
    blas_shutdown();
 }
 
+#if defined(_MSC_VER) && !defined(__clang__)
+BOOL APIENTRY DllMain(HMODULE hModule, DWORD  ul_reason_for_call, LPVOID lpReserved)
+{
+  switch (ul_reason_for_call)
+  {
+    case DLL_PROCESS_ATTACH:
+      gotoblas_init();
+      break;
+    case DLL_THREAD_ATTACH:
+      break;
+    case DLL_THREAD_DETACH:
+      break;
+    case DLL_PROCESS_DETACH:
+      gotoblas_quit();
+      break;
+    default:
+      break;
+  }
+  return TRUE;
+}
+#endif
+
 #if (defined(C_PGI) || (!defined(C_SUN) && defined(F_INTERFACE_SUN))) && (defined(ARCH_X86) || defined(ARCH_X86_64))
 /* Don't call me; this is just work around for PGI / Sun bug */
 void gotoblas_dummy_for_PGI(void) {