aiden
/
suricata
mirror of https://github.com/OISF/suricata
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

memcmp: use SCMEMCMP_BYTES everywhere; general cleanups

Browse Source
pull/7392/head
Victor Julien 4 years ago
parent 87c5d69437
commit 0fc7ba45aa
1 changed files with 36 additions and 73 deletions
Show Stats Download Patch File Download Diff File

109
src/util-memcmp.h
Unescape Escape View File

@ -55,37 +55,30 @@ MemcmpLowercase(const void *s1, const void *s2, size_t n)
}
#if defined(__SSE4_2__)
#include <nmmintrin.h>
/* No SIMD support, fall back to plain memcmp and a home grown lowercase one */
#define SCMEMCMP_BYTES 16
static inline int SCMemcmp(const void *s1, const void *s2, size_t n)
{
__m128i b1, b2;
int r;
int r = 0;
/* counter for how far we already matched in the buffer */
size_t m = 0;
do {
/* apparently we can't just read 16 bytes even though
* it almost always works fine :) */
if (likely(n - m < 16)) {
if (likely(n - m < SCMEMCMP_BYTES)) {
return memcmp(s1, s2, n - m) ? 1 : 0;
}
/* load the buffers into the 128bit vars */
b1 = _mm_loadu_si128((const __m128i *) s1);
b2 = _mm_loadu_si128((const __m128i *) s2);
/* do the actual compare */
m += (r = _mm_cmpestri(b1, n - m, b2, 16,
_SIDD_CMP_EQUAL_EACH | _SIDD_MASKED_NEGATIVE_POLARITY));
__m128i b1 = _mm_loadu_si128((const __m128i *)s1);
__m128i b2 = _mm_loadu_si128((const __m128i *)s2);
s1 += 16;
s2 += 16;
} while (r == 16);
/* do the actual compare: _mm_cmpestri() returns the number of matching bytes */
r = _mm_cmpestri(b1, SCMEMCMP_BYTES, b2, SCMEMCMP_BYTES,
_SIDD_CMP_EQUAL_EACH | _SIDD_MASKED_NEGATIVE_POLARITY);
m += r;
s1 += SCMEMCMP_BYTES;
s2 += SCMEMCMP_BYTES;
} while (r == SCMEMCMP_BYTES);
return ((m == n) ? 0 : 1);
}
@ -100,29 +93,23 @@ static char scmemcmp_uppercase[16] __attribute__((aligned(16))) = {
*/
static inline int SCMemcmpLowercase(const void *s1, const void *s2, size_t n)
{
__m128i b1, b2, mask;
int r;
/* counter for how far we already matched in the buffer */
size_t m = 0;
int r = 0;
__m128i ucase = _mm_load_si128((const __m128i *) scmemcmp_uppercase);
__m128i uplow = _mm_set1_epi8(0x20);
do {
/* apparently we can't just read 16 bytes even though
* it almost always works fine :) */
if (likely(n - m < 16)) {
return MemcmpLowercase(s1, s2, n - m);
const size_t len = n - m;
if (likely(len < SCMEMCMP_BYTES)) {
return MemcmpLowercase(s1, s2, len);
}
b1 = _mm_loadu_si128((const __m128i *) s1);
b2 = _mm_loadu_si128((const __m128i *) s2);
size_t len = n - m;
__m128i b1 = _mm_loadu_si128((const __m128i *)s1);
__m128i b2 = _mm_loadu_si128((const __m128i *)s2);
/* The first step is creating a mask that is FF for all uppercase
* characters, 00 for all others */
mask = _mm_cmpestrm(ucase, 2, b2, len, _SIDD_CMP_RANGES | _SIDD_UNIT_MASK);
__m128i mask = _mm_cmpestrm(ucase, 2, b2, len, _SIDD_CMP_RANGES | _SIDD_UNIT_MASK);
/* Next we use that mask to create a new: this one has 0x20 for
* the uppercase chars, 00 for all other. */
mask = _mm_and_si128(uplow, mask);
@ -130,41 +117,33 @@ static inline int SCMemcmpLowercase(const void *s1, const void *s2, size_t n)
* uppercase to lowercase */
b2 = _mm_add_epi8(b2, mask);
/* search using our converted buffer */
m += (r = _mm_cmpestri(b1, len, b2, 16,
_SIDD_CMP_EQUAL_EACH | _SIDD_MASKED_NEGATIVE_POLARITY));
s1 += 16;
s2 += 16;
} while (r == 16);
/* search using our converted buffer, return number of matching bytes */
r = _mm_cmpestri(b1, SCMEMCMP_BYTES, b2, SCMEMCMP_BYTES,
_SIDD_CMP_EQUAL_EACH | _SIDD_MASKED_NEGATIVE_POLARITY);
m += r;
s1 += SCMEMCMP_BYTES;
s2 += SCMEMCMP_BYTES;
} while (r == SCMEMCMP_BYTES);
return ((m == n) ? 0 : 1);
}
#elif defined(__SSE4_1__)
#include <smmintrin.h>
#define SCMEMCMP_BYTES 16
static inline int SCMemcmp(const void *s1, const void *s2, size_t len)
{
size_t offset = 0;
__m128i b1, b2, c;
do {
/* apparently we can't just read 16 bytes even though
* it almost always works fine :) */
if (likely(len - offset < 16)) {
if (likely(len - offset < SCMEMCMP_BYTES)) {
return memcmp(s1, s2, len - offset) ? 1 : 0;
}
/* do unaligned loads using _mm_loadu_si128. On my Core2 E6600 using
* _mm_lddqu_si128 was about 2% slower even though it's supposed to
* be faster. */
b1 = _mm_loadu_si128((const __m128i *) s1);
b2 = _mm_loadu_si128((const __m128i *) s2);
c = _mm_cmpeq_epi8(b1, b2);
/* unaligned loads */
__m128i b1 = _mm_loadu_si128((const __m128i *)s1);
__m128i b2 = _mm_loadu_si128((const __m128i *)s2);
__m128i c = _mm_cmpeq_epi8(b1, b2);
if (_mm_movemask_epi8(c) != 0x0000FFFF) {
return 1;
@ -184,7 +163,7 @@ static inline int SCMemcmp(const void *s1, const void *s2, size_t len)
static inline int SCMemcmpLowercase(const void *s1, const void *s2, size_t len)
{
size_t offset = 0;
__m128i b1, b2, mask1, mask2, upper1, upper2, nulls, uplow;
__m128i b1, b2, mask1, mask2, upper1, upper2, uplow;
/* setup registers for upper to lower conversion */
upper1 = _mm_set1_epi8(UPPER_LOW);
@ -192,9 +171,7 @@ static inline int SCMemcmpLowercase(const void *s1, const void *s2, size_t len)
uplow = _mm_set1_epi8(0x20);
do {
/* apparently we can't just read 16 bytes even though
* it almost always works fine :) */
if (likely(len - offset < 16)) {
if (likely(len - offset < SCMEMCMP_BYTES)) {
return MemcmpLowercase(s1, s2, len - offset);
}
@ -211,10 +188,8 @@ static inline int SCMemcmpLowercase(const void *s1, const void *s2, size_t len)
/* Next we use that mask to create a new: this one has 0x20 for
* the uppercase chars, 00 for all other. */
mask1 = _mm_and_si128(uplow, mask1);
/* add to b2, converting uppercase to lowercase */
b2 = _mm_add_epi8(b2, mask1);
/* now all is lowercase, let's do the actual compare (reuse mask1 reg) */
mask1 = _mm_cmpeq_epi8(b1, b2);
@ -230,12 +205,8 @@ static inline int SCMemcmpLowercase(const void *s1, const void *s2, size_t len)
return 0;
}
#elif defined(__SSE3__)
#include <pmmintrin.h> /* for SSE3 */
#define SCMEMCMP_BYTES 16
static inline int SCMemcmp(const void *s1, const void *s2, size_t len)
@ -244,15 +215,11 @@ static inline int SCMemcmp(const void *s1, const void *s2, size_t len)
__m128i b1, b2, c;
do {
/* apparently we can't just read 16 bytes even though
* it almost always works fine :) */
if (likely(len - offset < 16)) {
if (likely(len - offset < SCMEMCMP_BYTES)) {
return memcmp(s1, s2, len - offset) ? 1 : 0;
}
/* do unaligned loads using _mm_loadu_si128. On my Core2 E6600 using
* _mm_lddqu_si128 was about 2% slower even though it's supposed to
* be faster. */
/* unaligned loads */
b1 = _mm_loadu_si128((const __m128i *) s1);
b2 = _mm_loadu_si128((const __m128i *) s2);
c = _mm_cmpeq_epi8(b1, b2);
@ -284,9 +251,7 @@ static inline int SCMemcmpLowercase(const void *s1, const void *s2, size_t len)
delta = _mm_set1_epi8(UPPER_DELTA);
do {
/* apparently we can't just read 16 bytes even though
* it almost always works fine :) */
if (likely(len - offset < 16)) {
if (likely(len - offset < SCMEMCMP_BYTES)) {
return MemcmpLowercase(s1, s2, len - offset);
}
@ -300,11 +265,9 @@ static inline int SCMemcmpLowercase(const void *s1, const void *s2, size_t len)
mask2 = _mm_cmplt_epi8(b2, upper2);
/* merge the two, leaving only those that are true in both */
mask1 = _mm_cmpeq_epi8(mask1, mask2);
/* sub delta leaves 0x20 only for uppercase positions, the
rest is 0x00 due to the saturation (reuse mask1 reg)*/
mask1 = _mm_subs_epu8(mask1, delta);
/* add to b2, converting uppercase to lowercase */
b2 = _mm_add_epi8(b2, mask1);

Write Preview
Loading…
Cancel
Save