Created
May 19, 2011 16:09
-
-
Save imasahiro/981138 to your computer and use it in GitHub Desktop.
SIMD memset
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#ifdef __SSE__ | |
#include <emmintrin.h> | |
static void __memset(void *ptr, int c, size_t len) | |
{ | |
int i; | |
register char *p = (char *) ptr; | |
for (i = 0; i < len; i++) { | |
p[i] = c; | |
} | |
} | |
void mymemset(void *ptr, int v, size_t len) | |
{ | |
register size_t i; | |
size_t loff = ((intptr_t) ptr) % 16; | |
size_t l16 = (len - loff) / 16; | |
size_t lrem = len - l16 * 16 - loff; | |
register char *p = (char *) ptr; | |
char c = (char) v; | |
__m128i c16 = _mm_set_epi8(c, c, c, c, c, c, c, c, c, c, c, c, c, c, c, c); | |
__memset(p, c, loff); | |
p += loff; | |
for (i = 0; i < l16; ++i) { | |
_mm_store_si128((__m128i*)p, c16); | |
p += 16; | |
} | |
__memset(p, c, lrem); | |
} | |
#endif |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment