// Work input until "buf" reaches 16 byte alignment
while ( ((intptr_t)buf) % 32 != 0 && nframes > 0) {
-
+
// Load the next float into the work buffer
work = _mm256_set1_ps(*buf);
// use 64 byte prefetch for quadruple quads:
// load each 64 bytes into cash before processing
while (nframes >= 16) {
-#if defined(COMPILER_MSVC) || defined(COMPILER_MINGW)
+#if defined(COMPILER_MSVC) || defined(COMPILER_MINGW)
_mm_prefetch(((char*)buf+64), _mm_hint(0) );
-#else
+#else
__builtin_prefetch(buf+64,0,0);
#endif
work = _mm256_load_ps(buf);