Improved speed of Intel Compiler for Windows
diff --git a/lib/common/mem.h b/lib/common/mem.h
index da4f592..6f31aac 100644
--- a/lib/common/mem.h
+++ b/lib/common/mem.h
@@ -110,7 +110,7 @@
#ifndef MEM_FORCE_MEMORY_ACCESS /* can be defined externally, on command line for example */
# if defined(__GNUC__) && ( defined(__ARM_ARCH_6__) || defined(__ARM_ARCH_6J__) || defined(__ARM_ARCH_6K__) || defined(__ARM_ARCH_6Z__) || defined(__ARM_ARCH_6ZK__) || defined(__ARM_ARCH_6T2__) )
# define MEM_FORCE_MEMORY_ACCESS 2
-# elif (defined(__INTEL_COMPILER) && !defined(WIN32)) || \
+# elif defined(__INTEL_COMPILER) /*|| defined(_MSC_VER)*/ || \
(defined(__GNUC__) && ( defined(__ARM_ARCH_7__) || defined(__ARM_ARCH_7A__) || defined(__ARM_ARCH_7R__) || defined(__ARM_ARCH_7M__) || defined(__ARM_ARCH_7S__) ))
# define MEM_FORCE_MEMORY_ACCESS 1
# endif
@@ -142,7 +142,13 @@
/* __pack instructions are safer, but compiler specific, hence potentially problematic for some compilers */
/* currently only defined for gcc and icc */
-typedef union { U16 u16; U32 u32; U64 u64; size_t st; } __attribute__((packed)) unalign;
+#if defined(_MSC_VER) || (defined(__INTEL_COMPILER) && defined(WIN32))
+ __pragma( pack(push, 1) )
+ typedef union { U16 u16; U32 u32; U64 u64; size_t st; } unalign;
+ __pragma( pack(pop) )
+#else
+ typedef union { U16 u16; U32 u32; U64 u64; size_t st; } __attribute__((packed)) unalign;
+#endif
MEM_STATIC U16 MEM_read16(const void* ptr) { return ((const unalign*)ptr)->u16; }
MEM_STATIC U32 MEM_read32(const void* ptr) { return ((const unalign*)ptr)->u32; }
