aaru-dps/Aaru.Checksums.Native
1
0
Fork 0
mirror of https://github.com/aaru-dps/Aaru.Checksums.Native.git synced 2025-12-16 11:14:29 +00:00
Code Issues Packages Projects Releases Wiki Activity

Enable ADLER32 using NEON in 32-bit Windows on ARM.

Browse Source
This commit is contained in:
Natalia Portillo
2023-09-23 01:34:29 +01:00
parent e18fe4d973
commit 1bd06bb761
1 changed files with 46 additions and 33 deletions
Download Patch File Download Diff File Expand all files Collapse all files

79
adler32_neon.c
View File

@@ -38,7 +38,7 @@
#include "adler32.h" #include "adler32.h"
#include "simd.h" #include "simd.h"
TARGET_WITH_SIMD void adler32_neon(uint16_t* sum1, uint16_t* sum2, const uint8_t* data, uint32_t len) TARGET_WITH_SIMD void adler32_neon(uint16_t *sum1, uint16_t *sum2, const uint8_t *data, uint32_t len)
{ {
/* /*
* Split Adler-32 into component sums. * Split Adler-32 into component sums.
@@ -48,26 +48,26 @@ TARGET_WITH_SIMD void adler32_neon(uint16_t* sum1, uint16_t* sum2, const uint8_t
/* /*
* Serially compute s1 & s2, until the data is 16-byte aligned. * Serially compute s1 & s2, until the data is 16-byte aligned.
*/ */
if((uintptr_t)data & 15) if ((uintptr_t) data & 15)
{ {
while((uintptr_t)data & 15) while ((uintptr_t) data & 15)
{ {
s2 += (s1 += *data++); s2 += (s1 += *data++);
--len; --len;
} }
if(s1 >= ADLER_MODULE) s1 -= ADLER_MODULE; if (s1 >= ADLER_MODULE) s1 -= ADLER_MODULE;
s2 %= ADLER_MODULE; s2 %= ADLER_MODULE;
} }
/* /*
* Process the data in blocks. * Process the data in blocks.
*/ */
const unsigned BLOCK_SIZE = 1 << 5; const unsigned BLOCK_SIZE = 1 << 5;
uint32_t blocks = len / BLOCK_SIZE; uint32_t blocks = len / BLOCK_SIZE;
len -= blocks * BLOCK_SIZE; len -= blocks * BLOCK_SIZE;
while(blocks) while (blocks)
{ {
unsigned n = NMAX / BLOCK_SIZE; /* The NMAX constraint. */ unsigned n = NMAX / BLOCK_SIZE; /* The NMAX constraint. */
if(n > blocks) n = (unsigned)blocks; if (n > blocks) n = (unsigned) blocks;
blocks -= n; blocks -= n;
/* /*
* Process n blocks of data. At most NMAX data bytes can be * Process n blocks of data. At most NMAX data bytes can be
@@ -77,19 +77,20 @@ TARGET_WITH_SIMD void adler32_neon(uint16_t* sum1, uint16_t* sum2, const uint8_t
uint32x4_t v_s2 = {.n128_u32 = {0, 0, 0, s1 * n}}; uint32x4_t v_s2 = {.n128_u32 = {0, 0, 0, s1 * n}};
uint32x4_t v_s1 = {.n128_u32 = {0, 0, 0, 0}}; uint32x4_t v_s1 = {.n128_u32 = {0, 0, 0, 0}};
#else #else
uint32x4_t v_s2 = (uint32x4_t){0, 0, 0, s1 * n}; uint32x4_t v_s2 = (uint32x4_t) {0, 0, 0, s1 * n};
uint32x4_t v_s1 = (uint32x4_t){0, 0, 0, 0}; uint32x4_t v_s1 = (uint32x4_t) {0, 0, 0, 0};
#endif #endif
uint16x8_t v_column_sum_1 = vdupq_n_u16(0); uint16x8_t v_column_sum_1 = vdupq_n_u16(0);
uint16x8_t v_column_sum_2 = vdupq_n_u16(0); uint16x8_t v_column_sum_2 = vdupq_n_u16(0);
uint16x8_t v_column_sum_3 = vdupq_n_u16(0); uint16x8_t v_column_sum_3 = vdupq_n_u16(0);
uint16x8_t v_column_sum_4 = vdupq_n_u16(0); uint16x8_t v_column_sum_4 = vdupq_n_u16(0);
do { do
{
/* /*
* Load 32 input bytes. * Load 32 input bytes.
*/ */
const uint8x16_t bytes1 = vld1q_u8((uint8_t*)(data)); const uint8x16_t bytes1 = vld1q_u8((uint8_t *) (data));
const uint8x16_t bytes2 = vld1q_u8((uint8_t*)(data + 16)); const uint8x16_t bytes2 = vld1q_u8((uint8_t *) (data + 16));
/* /*
* Add previous block byte sum to v_s2. * Add previous block byte sum to v_s2.
*/ */
@@ -106,29 +107,40 @@ TARGET_WITH_SIMD void adler32_neon(uint16_t* sum1, uint16_t* sum2, const uint8_t
v_column_sum_3 = vaddw_u8(v_column_sum_3, vget_low_u8(bytes2)); v_column_sum_3 = vaddw_u8(v_column_sum_3, vget_low_u8(bytes2));
v_column_sum_4 = vaddw_u8(v_column_sum_4, vget_high_u8(bytes2)); v_column_sum_4 = vaddw_u8(v_column_sum_4, vget_high_u8(bytes2));
data += BLOCK_SIZE; data += BLOCK_SIZE;
} while(--n); } while (--n);
v_s2 = vshlq_n_u32(v_s2, 5); v_s2 = vshlq_n_u32(v_s2, 5);
/* /*
* Multiply-add bytes by [ 32, 31, 30, ... ] for s2. * Multiply-add bytes by [ 32, 31, 30, ... ] for s2.
*/ */
#ifdef _MSC_VER #ifdef _MSC_VER
v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_1), neon_ld1m_16((uint16_t[]){32, 31, 30, 29})); #ifdef _M_ARM64
v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_1), neon_ld1m_16((uint16_t[]){28, 27, 26, 25})); v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_1), neon_ld1m_16((uint16_t[]) {32, 31, 30, 29}));
v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_2), neon_ld1m_16((uint16_t[]){24, 23, 22, 21})); v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_1), neon_ld1m_16((uint16_t[]) {28, 27, 26, 25}));
v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_2), neon_ld1m_16((uint16_t[]){20, 19, 18, 17})); v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_2), neon_ld1m_16((uint16_t[]) {24, 23, 22, 21}));
v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_3), neon_ld1m_16((uint16_t[]){16, 15, 14, 13})); v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_2), neon_ld1m_16((uint16_t[]) {20, 19, 18, 17}));
v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_3), neon_ld1m_16((uint16_t[]){12, 11, 10, 9})); v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_3), neon_ld1m_16((uint16_t[]) {16, 15, 14, 13}));
v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_4), neon_ld1m_16((uint16_t[]){8, 7, 6, 5})); v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_3), neon_ld1m_16((uint16_t[]) {12, 11, 10, 9}));
v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_4), neon_ld1m_16((uint16_t[]){4, 3, 2, 1})); v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_4), neon_ld1m_16((uint16_t[]) {8, 7, 6, 5}));
v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_4), neon_ld1m_16((uint16_t[]) {4, 3, 2, 1}));
#else #else
v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_1), (uint16x4_t){32, 31, 30, 29}); v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_1), vld1_u16(((uint16_t[]) {32, 31, 30, 29})));
v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_1), (uint16x4_t){28, 27, 26, 25}); v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_1), vld1_u16(((uint16_t[]) {28, 27, 26, 25})));
v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_2), (uint16x4_t){24, 23, 22, 21}); v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_2), vld1_u16(((uint16_t[]) {24, 23, 22, 21})));
v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_2), (uint16x4_t){20, 19, 18, 17}); v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_2), vld1_u16(((uint16_t[]) {20, 19, 18, 17})));
v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_3), (uint16x4_t){16, 15, 14, 13}); v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_3), vld1_u16(((uint16_t[]) {16, 15, 14, 13})));
v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_3), (uint16x4_t){12, 11, 10, 9}); v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_3), vld1_u16(((uint16_t[]) {12, 11, 10, 9})));
v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_4), (uint16x4_t){8, 7, 6, 5}); v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_4), vld1_u16(((uint16_t[]) {8, 7, 6, 5})));
v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_4), (uint16x4_t){4, 3, 2, 1}); v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_4), vld1_u16(((uint16_t[]) {4, 3, 2, 1})));
#endif
#else
v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_1), (uint16x4_t) {32, 31, 30, 29});
v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_1), (uint16x4_t) {28, 27, 26, 25});
v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_2), (uint16x4_t) {24, 23, 22, 21});
v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_2), (uint16x4_t) {20, 19, 18, 17});
v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_3), (uint16x4_t) {16, 15, 14, 13});
v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_3), (uint16x4_t) {12, 11, 10, 9});
v_s2 = vmlal_u16(v_s2, vget_low_u16(v_column_sum_4), (uint16x4_t) {8, 7, 6, 5});
v_s2 = vmlal_u16(v_s2, vget_high_u16(v_column_sum_4), (uint16x4_t) {4, 3, 2, 1});
#endif #endif
/* /*
* Sum epi32 ints v_s1(s2) and accumulate in s1(s2). * Sum epi32 ints v_s1(s2) and accumulate in s1(s2).
@@ -147,9 +159,9 @@ TARGET_WITH_SIMD void adler32_neon(uint16_t* sum1, uint16_t* sum2, const uint8_t
/* /*
* Handle leftover data. * Handle leftover data.
*/ */
if(len) if (len)
{ {
if(len >= 16) if (len >= 16)
{ {
s2 += (s1 += *data++); s2 += (s1 += *data++);
s2 += (s1 += *data++); s2 += (s1 += *data++);
@@ -169,8 +181,9 @@ TARGET_WITH_SIMD void adler32_neon(uint16_t* sum1, uint16_t* sum2, const uint8_t
s2 += (s1 += *data++); s2 += (s1 += *data++);
len -= 16; len -= 16;
} }
while(len--) { s2 += (s1 += *data++); } while (len--)
if(s1 >= ADLER_MODULE) s1 -= ADLER_MODULE; { s2 += (s1 += *data++); }
if (s1 >= ADLER_MODULE) s1 -= ADLER_MODULE;
s2 %= ADLER_MODULE; s2 %= ADLER_MODULE;
} }
/* /*