fork/minimap2
1
0
Fork 0
mirror of https://github.com/lh3/minimap2.git synced 2026-09-26 03:28:12 +08:00
Code Issues Packages Projects Releases Wiki Activity
Files
b2ff8fbe92ee660b543cb5814b8ea9493ce9ca77
minimap2/ksw2_extd2_avx.c
T
Saurabh 0369874d4e mm2-fast: Initial commit
2021-06-16 15:50:17 -04:00

665 lines
34 KiB
C
Raw Blame History

/* The MIT License
Copyright (c) 2018- Dana-Farber Cancer Institute
2017-2018 Broad Institute, Inc.
Permission is hereby granted, free of charge, to any person obtaining
a copy of this software and associated documentation files (the
"Software"), to deal in the Software without restriction, including
without limitation the rights to use, copy, modify, merge, publish,
distribute, sublicense, and/or sell copies of the Software, and to
permit persons to whom the Software is furnished to do so, subject to
the following conditions:
The above copyright notice and this permission notice shall be
included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
Modified Copyright (C) 2021 Intel Corporation
Contacts: Saurabh Kalikar <saurabh.kalikar@intel.com>;
Vasimuddin Md <vasimuddin.md@intel.com>; Sanchit Misra <sanchit.misra@intel.com>;
Chirag Jain <chirag@iisc.ac.in>; Heng Li <hli@jimmy.harvard.edu>
*/
#include "ksw2_extd2_avx.h"
#ifdef __AVX512BW__
void ksw_extd2_avx512(void *km, int qlen, const uint8_t *query, int tlen, const uint8_t *target, int8_t m, const int8_t *mat,
int8_t q, int8_t e, int8_t q2, int8_t e2, int w, int zdrop, int end_bonus, int flag, ksw_extz_t *ez)
{
__m512i ind512_srli = _mm512_set1_epi8(15);
__mmask8 msk_srli = 0x0003;
__m512i bt32_ = _mm512_setr_epi32(0,0,0,0,4,4,4,4,8,8,8,8,12,12,12,12);
int8_t index[64] __attribute((aligned(64)));
for (int i=0; i<64; i++)
index[i] = i%16 - 1;
index[0] = 15;
index[16] = 31;
index[32] = 47;
index[48] = 63;
__m512i shf512a, shf512b, slli512;
__m512i ind512_slli = _mm512_load_si512((__m512i*) index);
__mmask8 mska = 0x90;
__mmask64 mskb = 0x0001000100010000;
__mmask64 mskc = 0x1;
__mmask64 mskc_ar[4] = {0x1, 0x10000, 0x100000000, 0x1000000000000};
#define __dp_code_block1_pcl \
/*__mmask64 mskc_ = mskc; */ \
/*if (t == st_) */ \
__mmask64 mskc_ = (t == st_) ? mskc_ar[(st0 - t*64)/16]:mskc; \
z = _mm512_load_si512(&s[t]); \
xt1 = _mm512_load_si512(&x[t]); /* xt1 <- x[r-1][t..t+15] */ \
/* tmp = _mm_srli_si128(xt1, 15); */ /* tmp <- x[r-1][t+15] */ \
tmp = _mm512_set1_epi8(((int8_t*)x)[t*64 + 63]); \
/* xt1 = _mm_or_si128(_mm_slli_si128(xt1, 1), x1_);*/ /* xt1 <- x[r-1][t-1..t+14] */ \
shf512a = _mm512_shuffle_epi8(xt1, ind512_slli); \
shf512b = _mm512_shuffle_i32x4(shf512a, shf512a, mska); \
slli512 = _mm512_mask_blend_epi8(mskb, shf512a, shf512b); \
xt1 = _mm512_mask_blend_epi8(mskc_, slli512, x1_); \
x1_ = tmp; \
vt1 = _mm512_load_si512(&v[t]); /* vt1 <- v[r-1][t..t+15] */ \
/* tmp = _mm_srli_si128(vt1, 15); */ /* tmp <- v[r-1][t+15] */ \
tmp = _mm512_set1_epi8(((int8_t*)v)[t*64 + 63]); \
/* vt1 = _mm_or_si128(_mm_slli_si128(vt1, 1), v1_); *//* vt1 <- v[r-1][t-1..t+14] */ \
shf512a = _mm512_shuffle_epi8(vt1, ind512_slli); \
shf512b = _mm512_shuffle_i32x4(shf512a, shf512a, mska); \
slli512 = _mm512_mask_blend_epi8(mskb, shf512a, shf512b); \
vt1 = _mm512_mask_blend_epi8(mskc_, slli512, v1_); \
v1_ = tmp; \
a = _mm512_add_epi8(xt1, vt1); /* a <- x[r-1][t-1..t+14] + v[r-1][t-1..t+14] */ \
ut = _mm512_load_si512(&u[t]); /* ut <- u[t..t+15] */ \
b = _mm512_add_epi8(_mm512_load_si512(&y[t]), ut); /* b <- y[r-1][t..t+15] + u[r-1][t..t+15] */ \
x2t1= _mm512_load_si512(&x2[t]); \
/* tmp = _mm_srli_si128(x2t1, 15);*/ \
tmp = _mm512_set1_epi8(((int8_t*)x2)[t*64 + 63]); \
/* x2t1= _mm_or_si128(_mm_slli_si128(x2t1, 1), x21_); */ \
shf512a = _mm512_shuffle_epi8(x2t1, ind512_slli); \
shf512b = _mm512_shuffle_i32x4(shf512a, shf512a, mska); \
slli512 = _mm512_mask_blend_epi8(mskb, shf512a, shf512b); \
x2t1 = _mm512_mask_blend_epi8(mskc_, slli512, x21_); \
x21_= tmp; \
a2= _mm512_add_epi8(x2t1, vt1); \
b2= _mm512_add_epi8(_mm512_load_si512(&y2[t]), ut);
#define __dp_code_block2_pcl \
_mm512_store_si512(&u[t], _mm512_sub_epi8(z, vt1)); /* u[r][t..t+15] <- z - v[r-1][t-1..t+14] */ \
_mm512_store_si512(&v[t], _mm512_sub_epi8(z, ut)); /* v[r][t..t+15] <- z - u[r-1][t..t+15] */ \
tmp = _mm512_sub_epi8(z, q_); \
a = _mm512_sub_epi8(a, tmp); \
b = _mm512_sub_epi8(b, tmp); \
tmp = _mm512_sub_epi8(z, q2_); \
a2= _mm512_sub_epi8(a2, tmp); \
b2= _mm512_sub_epi8(b2, tmp);
__mmask64 msk_ar[5] = {0xFFFF, 0xFFFFFFFF, 0xFFFFFFFFFFFF, 0xFFFFFFFFFFFFFFFF, 0xFFFFFFFFFFFFFFFF};
__mmask64 msk_ar2[5] = {0xFFFF, 0xFFFF, 0xFFFFFFFF, 0xFFFFFFFFFFFF, 0xFFFFFFFFFFFFFFFF};
int r, t, qe = q + e, n_col_, *off = 0, *off_end = 0, tlen_, qlen_, last_st, last_en, wl, wr, max_sc, min_sc, long_thres, long_diff;
int with_cigar = !(flag&KSW_EZ_SCORE_ONLY), approx_max = !!(flag&KSW_EZ_APPROX_MAX);
int32_t *H = 0, H0 = 0, last_H0_t = 0;
uint8_t *qr, *sf, *mem, *mem2 = 0;
__m512i q_, q2_, qe_, qe2_, zero_, sc_mch_, sc_mis_, m1_, sc_N_;
__m512i *u, *v, *x, *y, *x2, *y2, *s, *p = 0;
__m512i one_, two_, three_, four_, s1_, s2_, s3_, s4_;
ksw_reset_extz(ez);
if (m <= 1 || qlen <= 0 || tlen <= 0) return;
if (q2 + e2 < q + e) t = q, q = q2, q2 = t, t = e, e = e2, e2 = t; // make sure q+e no larger than q2+e2
s1_ = _mm512_set1_epi8(0x08);
s2_ = _mm512_set1_epi8(0x10);
s3_ = _mm512_set1_epi8(0x20);
s4_ = _mm512_set1_epi8(0x40);
one_ = _mm512_set1_epi8(1);
two_ = _mm512_set1_epi8(2);
three_ = _mm512_set1_epi8(3);
four_ = _mm512_set1_epi8(4);
zero_ = _mm512_set1_epi8(0);
q_ = _mm512_set1_epi8(q);
q2_ = _mm512_set1_epi8(q2);
qe_ = _mm512_set1_epi8(q + e);
qe2_ = _mm512_set1_epi8(q2 + e2);
sc_mch_ = _mm512_set1_epi8(mat[0]);
sc_mis_ = _mm512_set1_epi8(mat[1]);
sc_N_ = mat[m*m-1] == 0? _mm512_set1_epi8(-e2) : _mm512_set1_epi8(mat[m*m-1]);
m1_ = _mm512_set1_epi8(m - 1); // wildcard
if (w < 0) w = tlen > qlen? tlen : qlen;
wl = wr = w;
tlen_ = (tlen + 63) / 64;
n_col_ = qlen < tlen? qlen : tlen;
n_col_ = ((n_col_ < w + 1? n_col_ : w + 1) + 63) / 64 + 1;
qlen_ = (qlen + 63) / 64;
for (t = 1, max_sc = mat[0], min_sc = mat[1]; t < m * m; ++t) {
max_sc = max_sc > mat[t]? max_sc : mat[t];
min_sc = min_sc < mat[t]? min_sc : mat[t];
}
if (-min_sc > 2 * (q + e)) return; // otherwise, we won't see any mismatches
long_thres = e != e2? (q2 - q) / (e - e2) - 1 : 0;
if (q2 + e2 + long_thres * e2 > q + e + long_thres * e)
++long_thres;
long_diff = long_thres * (e - e2) - (q2 - q) - e2;
mem = (uint8_t*)kcalloc(km, tlen_ * 8 + qlen_ + 1 + 63, 64);
u = (__m512i*)(((size_t)mem + 63) >> 6 << 6); // 16-byte aligned
v = u + tlen_, x = v + tlen_, y = x + tlen_, x2 = y + tlen_, y2 = x2 + tlen_;
s = y2 + tlen_, sf = (uint8_t*)(s + tlen_), qr = sf + tlen_ * 64;
memset(u, -q - e, tlen_ * 64);
memset(v, -q - e, tlen_ * 64);
memset(x, -q - e, tlen_ * 64);
memset(y, -q - e, tlen_ * 64);
memset(x2, -q2 - e2, tlen_ * 64);
memset(y2, -q2 - e2, tlen_ * 64);
if (!approx_max) {
H = (int32_t*)kmalloc(km, tlen_ * 64 * 4);
for (t = 0; t < tlen_ * 64; ++t) H[t] = KSW_NEG_INF;
}
if (with_cigar) {
mem2 = (uint8_t*)kmalloc(km, ((size_t)(qlen + tlen - 1) * n_col_ + 1) * 64);
p = (__m512i*)(((size_t)mem2 + 63) >> 6 << 6);
off = (int*)kmalloc(km, (qlen + tlen - 1) * sizeof(int) * 2);
off_end = off + qlen + tlen - 1;
}
for (t = 0; t < qlen; ++t) qr[t] = query[qlen - 1 - t];
memcpy(sf, target, tlen);
for (r = 0, last_st = last_en = -1; r < qlen + tlen - 1; ++r) {
int st = 0, en = tlen - 1, st0, en0, st_, en_;
int8_t x1, x21, v1;
uint8_t *qrr = qr + (qlen - 1 - r);
int8_t *u8 = (int8_t*)u, *v8 = (int8_t*)v, *x8 = (int8_t*)x, *x28 = (int8_t*)x2;
__m512i x1_, x21_, v1_;
// find the boundaries
if (st < r - qlen + 1) st = r - qlen + 1;
if (en > r) en = r;
if (st < (r-wr+1)>>1) st = (r-wr+1)>>1; // take the ceil
if (en > (r+wl)>>1) en = (r+wl)>>1; // take the floor
if (st > en) {
ez->zdropped = 1;
break;
}
st0 = st, en0 = en;
int st_new = st / 16 * 16, en_new = (en + 16) / 16 * 16 - 1;
// int st_new = st / 64 * 64, en_new = (en + 64) / 64 * 64 - 1;
int stb = st, enb = en;
st = st / 64 * 64, en = (en + 64) / 64 * 64 - 1;
int stn = stb / 16 * 16, enn = (enb + 16) / 16 * 16 - 1;
// set boundary conditions
if (st_new > 0) {
if (st_new - 1 >= last_st && st_new - 1 <= last_en) {
x1 = x8[st_new - 1], x21 = x28[st_new - 1], v1 = v8[st_new - 1]; // (r-1,s-1) calculated in the last round
} else {
x1 = -q - e, x21 = -q2 - e2;
v1 = -q - e;
}
} else {
x1 = -q - e, x21 = -q2 - e2;
v1 = r == 0? -q - e : r < long_thres? -e : r == long_thres? long_diff : -e2;
}
if (en_new >= r) {
((int8_t*)y)[r] = -q - e, ((int8_t*)y2)[r] = -q2 - e2;
u8[r] = r == 0? -q - e : r < long_thres? -e : r == long_thres? long_diff : -e2;
}
// loop fission: set scores first
if (!(flag & KSW_EZ_GENERIC_SC)) {
for (t = st0; t <= en0; t += 64) {
__m512i sq, st, tmp_512, mask_512;
__mmask64 tmp, mask;
sq = _mm512_loadu_si512((__m512i*)&sf[t]);
st = _mm512_loadu_si512((__m512i*)&qrr[t]);
// mask = _mm512_or_si512(_mm_cmpeq_epi8(sq, m1_), _mm_cmpeq_epi8(st, m1_));
mask = (_mm512_cmpeq_epi8_mask(sq, m1_) | _mm512_cmpeq_epi8_mask(st, m1_));
tmp = _mm512_cmpeq_epi8_mask(sq, st);
tmp_512 = _mm512_mask_blend_epi8(tmp, sc_mis_, sc_mch_);
tmp_512 = _mm512_mask_blend_epi8(mask, tmp_512, sc_N_);
if (t + 64 > en0)
{
__mmask64 msk;
int ind = (en0 - t + 16)/16;
//assert(ind >= 0 && ind < 5);
msk = msk_ar2[ind];
_mm512_mask_storeu_epi8((__m512i*)((int8_t*)s + t), msk, tmp_512);
}
else
_mm512_storeu_si512((__m512i*)((int8_t*)s + t), tmp_512);
}
} else {
for (t = st0; t <= en0; ++t)
((uint8_t*)s)[t] = mat[sf[t] * m + qrr[t]];
}
// core loop
// fprintf(stderr, "- r: %d, x1: %d, x21: %d, v1: %d, en_new: %d, e: %d, q: %d\n",
//r, x1, x21, v1, en_new, e, q);
x1_ = _mm512_set1_epi8((uint8_t)x1);
x21_ = _mm512_set1_epi8((uint8_t)x21);
v1_ = _mm512_set1_epi8((uint8_t)v1);
//st_ = st / 16, en_ = en / 16;
st_ = st / 64, en_ = en / 64;
//assert(en_ - st_ + 1 <= n_col_);
if (!with_cigar) { // score only
for (t = st_; t <= en_; ++t) {
__m512i z, a, b, a2, b2, xt1, x2t1, vt1, ut, tmp;
__dp_code_block1_pcl;
z = _mm512_max_epi8(z, a);
z = _mm512_max_epi8(z, b);
z = _mm512_max_epi8(z, a2);
z = _mm512_max_epi8(z, b2);
z = _mm512_min_epi8(z, sc_mch_);
// __dp_code_block2_pcl; // save u[] and v[]; update a, b, a2 and b2
if (t == en_) {
__mmask64 msk;
int ind = (en0 - t*64 + 16)/16;
msk = msk_ar2[ind];
// fprintf(stderr, "en0: %d, t: %d, ind: %d, msk: %d\n", en0, t, ind, msk);
_mm512_mask_storeu_epi8(&u[t], msk, _mm512_sub_epi8(z, vt1));
_mm512_mask_storeu_epi8(&v[t], msk, _mm512_sub_epi8(z, ut));
tmp = _mm512_sub_epi8(z, q_);
a = _mm512_sub_epi8(a, tmp);
b = _mm512_sub_epi8(b, tmp);
tmp = _mm512_sub_epi8(z, q2_);
a2= _mm512_sub_epi8(a2, tmp);
b2= _mm512_sub_epi8(b2, tmp);
}
else {
_mm512_store_si512(&u[t], _mm512_sub_epi8(z, vt1));
_mm512_store_si512(&v[t], _mm512_sub_epi8(z, ut));
tmp = _mm512_sub_epi8(z, q_);
a = _mm512_sub_epi8(a, tmp);
b = _mm512_sub_epi8(b, tmp);
tmp = _mm512_sub_epi8(z, q2_);
a2= _mm512_sub_epi8(a2, tmp);
b2= _mm512_sub_epi8(b2, tmp);
}
if (t == en_) {
__mmask64 msk;
int ind = (en0 - t*64 + 16)/16;
//assert(ind >= 0);
msk = msk_ar2[ind];
// fprintf(stderr, "en0: %d, t: %d, ind: %d, msk: %d\n", en0, t, ind, msk);
_mm512_mask_storeu_epi8(&x[t], msk, _mm512_sub_epi8(_mm512_max_epi8(a, zero_), qe_));
_mm512_mask_storeu_epi8(&y[t], msk, _mm512_sub_epi8(_mm512_max_epi8(b, zero_), qe_));
_mm512_mask_storeu_epi8(&x2[t], msk, _mm512_sub_epi8(_mm512_max_epi8(a2, zero_), qe2_));
_mm512_mask_storeu_epi8(&y2[t], msk, _mm512_sub_epi8(_mm512_max_epi8(b2, zero_), qe2_));
}
else
{
_mm512_store_si512(&x[t], _mm512_sub_epi8(_mm512_max_epi8(a, zero_), qe_));
_mm512_store_si512(&y[t], _mm512_sub_epi8(_mm512_max_epi8(b, zero_), qe_));
_mm512_store_si512(&x2[t], _mm512_sub_epi8(_mm512_max_epi8(a2, zero_), qe2_));
_mm512_store_si512(&y2[t], _mm512_sub_epi8(_mm512_max_epi8(b2, zero_), qe2_));
}
// for (int l=0; l<64; l++)
// fprintf(stderr, "%d ", ((int8_t*)x)[l]);
}
} else if (!(flag&KSW_EZ_RIGHT)) { // gap left-alignment
__m512i *pr = p + (size_t)r * n_col_ - st_;
off[r] = st, off_end[r] = en;
for (t = st_; t < en_; ++t) {
__m512i d, z, a, b, a2, b2, xt1, x2t1, vt1, ut, tmp;
__mmask64 tmp_mask;
__dp_code_block1_pcl;
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(a, z),zero_, one_); // d = a > z? 1 : 0
z = _mm512_max_epi8(z, a);
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(b, z), d, two_); // d = b > z? 2 : d
z = _mm512_max_epi8(z, b);
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(a2, z), d, three_); // d = a2 > z? 3 : d
z = _mm512_max_epi8(z, a2);
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(b2, z), d, four_); // d = b2 > z? 4 : d
z = _mm512_max_epi8(z, b2);
z = _mm512_min_epi8(z, sc_mch_);
// __dp_code_block2_pcl;
_mm512_store_si512(&u[t], _mm512_sub_epi8(z, vt1));
_mm512_store_si512(&v[t], _mm512_sub_epi8(z, ut));
tmp = _mm512_sub_epi8(z, q_);
a = _mm512_sub_epi8(a, tmp);
b = _mm512_sub_epi8(b, tmp);
tmp = _mm512_sub_epi8(z, q2_);
a2= _mm512_sub_epi8(a2, tmp);
b2= _mm512_sub_epi8(b2, tmp);
tmp_mask = _mm512_cmpgt_epi8_mask(a, zero_);
_mm512_store_si512(&x[t], _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, zero_, a), qe_));
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, zero_, s1_)); // d = a > 0? 1<<3 : 0
tmp_mask = _mm512_cmpgt_epi8_mask(b, zero_);
_mm512_store_si512(&y[t], _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, zero_, b), qe_));
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, zero_, s2_)); // d = b > 0? 1<<4 : 0
tmp_mask = _mm512_cmpgt_epi8_mask(a2, zero_);
_mm512_store_si512(&x2[t], _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, zero_, a2), qe2_));
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, zero_, s3_)); // d = a > 0? 1<<5 : 0
tmp_mask = _mm512_cmpgt_epi8_mask(b2, zero_);
_mm512_store_si512(&y2[t], _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, zero_, b2), qe2_));
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, zero_, s4_)); // d = b > 0? 1<<6 : 0
_mm512_store_si512(&pr[t], d);
}
{
__m512i d, z, a, b, a2, b2, xt1, x2t1, vt1, ut, tmp;
__mmask64 tmp_mask;
__dp_code_block1_pcl;
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(a, z),zero_, one_); // d = a > z? 1 : 0
z = _mm512_max_epi8(z, a);
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(b, z), d, two_); // d = b > z? 2 : d
z = _mm512_max_epi8(z, b);
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(a2, z), d, three_); // d = a2 > z? 3 : d
z = _mm512_max_epi8(z, a2);
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(b2, z), d, four_); // d = b2 > z? 3 : d
z = _mm512_max_epi8(z, b2);
z = _mm512_min_epi8(z, sc_mch_);
// __dp_code_block2_pcl;
{
__mmask64 msk;
int ind = (en0 - t*64 + 16)/16;
// //assert(ind >= 0 && ind < 5);
msk = msk_ar2[ind];
_mm512_mask_storeu_epi8(&u[t], msk, _mm512_sub_epi8(z, vt1));
_mm512_mask_storeu_epi8(&v[t], msk, _mm512_sub_epi8(z, ut));
tmp = _mm512_sub_epi8(z, q_);
a = _mm512_sub_epi8(a, tmp);
b = _mm512_sub_epi8(b, tmp);
tmp = _mm512_sub_epi8(z, q2_);
a2= _mm512_sub_epi8(a2, tmp);
b2= _mm512_sub_epi8(b2, tmp);
}
{
__mmask64 msk;
int ind = (en0 - t*64 + 16)/16;
msk = msk_ar2[ind];
off_end[r] -= (4-ind)*16;
tmp_mask = _mm512_cmpgt_epi8_mask(a, zero_);
_mm512_mask_storeu_epi8(&x[t], msk, _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, zero_, a), qe_));
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, zero_, s1_)); // d = a > 0? 1<<3 : 0
tmp_mask = _mm512_cmpgt_epi8_mask(b, zero_);
_mm512_mask_storeu_epi8(&y[t], msk, _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, zero_, b), qe_));
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, zero_, s2_)); // d = b > 0? 1<<4 : 0
tmp_mask = _mm512_cmpgt_epi8_mask(a2, zero_);
_mm512_mask_storeu_epi8(&x2[t], msk, _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, zero_, a2), qe2_));
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, zero_, s3_)); // d = a > 0? 1<<5 : 0
tmp_mask = _mm512_cmpgt_epi8_mask(b2, zero_);
_mm512_mask_storeu_epi8(&y2[t], msk, _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, zero_, b2), qe2_));
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, zero_, s4_)); // d = b > 0? 1<<6 : 0
//_mm512_store_si512(&pr[t], d);
_mm512_mask_storeu_epi8(&pr[t], msk, d);
}
}
} else { // gap right-alignment
__m512i *pr = p + (size_t)r * n_col_ - st_;
off[r] = st, off_end[r] = en;
// off[r] = stn, off_end[r] = enn;
// fprintf(stderr, "t: %d, st0: %d\n", st_, st0);
for (t = st_; t < en_; ++t) {
__m512i d, z, a, b, a2, b2, xt1, x2t1, vt1, ut, tmp;
__mmask64 tmp_mask;
__dp_code_block1_pcl;
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(z, a), one_, zero_);
z = _mm512_max_epi8(z, a);
// d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(z, b), _mm512_set1_epi8(2), d);
// d = z > b? d : 2
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(z, b), two_, d); // d = z > b? d : 2
z = _mm512_max_epi8(z, b);
// d = z > a2? d : 3
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(z, a2), three_, d); // d = z > a2? d : 3
z = _mm512_max_epi8(z, a2);
// d = z > b2? d : 4
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(z, b2), four_, d); // d = z > b2? d : 4
z = _mm512_max_epi8(z, b2);
z = _mm512_min_epi8(z, sc_mch_);
// __dp_code_block2_pcl;
__mmask64 msk;
{
_mm512_store_si512(&u[t], _mm512_sub_epi8(z, vt1));
_mm512_store_si512(&v[t], _mm512_sub_epi8(z, ut));
tmp = _mm512_sub_epi8(z, q_);
a = _mm512_sub_epi8(a, tmp);
b = _mm512_sub_epi8(b, tmp);
tmp = _mm512_sub_epi8(z, q2_);
a2= _mm512_sub_epi8(a2, tmp);
b2= _mm512_sub_epi8(b2, tmp);
}
{
tmp_mask = _mm512_cmpgt_epi8_mask(zero_, a);
_mm512_store_si512(&x[t], _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, a, zero_), qe_));
// d = a > 0? 1<<3 : 0
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, s1_, zero_)); // d = a > 0? 1<<3 : 0
tmp_mask = _mm512_cmpgt_epi8_mask(zero_, b);
_mm512_store_si512(&y[t], _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, b, zero_), qe_));
// d = b > 0? 1<<4 : 0
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, s2_, zero_)); // d = b > 0? 1<<4 : 0
tmp_mask = _mm512_cmpgt_epi8_mask(zero_, a2);
_mm512_store_si512(&x2[t], _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, a2, zero_), qe2_));
// d = a > 0? 1<<5 : 0
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, s3_, zero_)); // d = a > 0? 1<<5 : 0
tmp_mask = _mm512_cmpgt_epi8_mask(zero_, b2);
_mm512_store_si512(&y2[t], _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, b2, zero_), qe2_));
// d = b > 0? 1<<6 : 0
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, s4_, zero_)); // d = b > 0? 1<<6 : 0
_mm512_store_si512(&pr[t], d);
}
}
//for (t = st_; t <= en_; ++t)// Last iteration unrolled
{
__m512i d, z, a, b, a2, b2, xt1, x2t1, vt1, ut, tmp;
__mmask64 tmp_mask;
__dp_code_block1_pcl;
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(z, a), one_, zero_);
z = _mm512_max_epi8(z, a);
// d = z > b? d : 2
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(z, b), two_, d); // d = z > b? d : 2
z = _mm512_max_epi8(z, b);
// d = z > a2? d : 3
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(z, a2), three_, d); // d = z > a2? d : 3
z = _mm512_max_epi8(z, a2);
// d = z > b2? d : 4
d = _mm512_mask_blend_epi8(_mm512_cmpgt_epi8_mask(z, b2), four_, d); // d = z > b2? d : 4
z = _mm512_max_epi8(z, b2);
z = _mm512_min_epi8(z, sc_mch_);
// __dp_code_block2_pcl;
__mmask64 msk;
{
// __mmask64 msk;
int ind = (en0 - t*64 + 16)/16;
msk = msk_ar2[ind];
off_end[r] -= (4-ind)*16;
_mm512_mask_storeu_epi8(&u[t], msk, _mm512_sub_epi8(z, vt1));
_mm512_mask_storeu_epi8(&v[t], msk, _mm512_sub_epi8(z, ut));
tmp = _mm512_sub_epi8(z, q_);
a = _mm512_sub_epi8(a, tmp);
b = _mm512_sub_epi8(b, tmp);
tmp = _mm512_sub_epi8(z, q2_);
a2= _mm512_sub_epi8(a2, tmp);
b2= _mm512_sub_epi8(b2, tmp);
}
{
tmp_mask = _mm512_cmpgt_epi8_mask(zero_, a);
_mm512_mask_storeu_epi8(&x[t], msk, _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, a, zero_), qe_));
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, s1_, zero_)); // d = a > 0? 1<<3 : 0
tmp_mask = _mm512_cmpgt_epi8_mask(zero_, b);
_mm512_mask_storeu_epi8(&y[t], msk, _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, b, zero_), qe_));
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, s2_, zero_)); // d = b > 0? 1<<4 : 0
tmp_mask = _mm512_cmpgt_epi8_mask(zero_, a2);
_mm512_mask_storeu_epi8(&x2[t], msk, _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, a2, zero_), qe2_));
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, s3_, zero_)); // d = a > 0? 1<<5 : 0
tmp_mask = _mm512_cmpgt_epi8_mask(zero_, b2);
_mm512_mask_storeu_epi8(&y2[t], msk, _mm512_sub_epi8(_mm512_mask_blend_epi8(tmp_mask, b2, zero_), qe2_));
d = _mm512_or_si512(d, _mm512_mask_blend_epi8(tmp_mask, s4_, zero_)); // d = b > 0? 1<<6 : 0
// _mm512_store_si512(&pr[t], d);
_mm512_mask_storeu_epi8(&pr[t], msk, d);
}
}
}
if (!approx_max) { // find the exact max with a 32-bit score array
int32_t max_H, max_t;
// compute H[], max_H and max_t
if (r > 0) {
int32_t HH[16], tt[16], en1 = st0 + (en0 - st0) / 16 * 16, i;
__m512i max_H_, max_t_;
max_H = H[en0] = en0 > 0? H[en0-1] + u8[en0] : H[en0] + v8[en0]; // special casing the last element
max_t = en0;
max_H_ = _mm512_set1_epi32(max_H);
max_t_ = _mm512_set1_epi32(max_t);
for (t = st0; t < en1; t += /*4*/16) { // this implements: H[t]+=v8[t]-qe; if(H[t]>max_H) max_H=H[t],max_t=t;
__m512i H1, tmp, t_;
__mmask16 tmp_mask;
H1 = _mm512_loadu_si512((__m512i*)&H[t]);
__m128i t__ = _mm_load_si128((__m128i*) &v8[t]);
t_ = _mm512_cvtepi8_epi32(t__);
H1 = _mm512_add_epi32(H1, t_);
_mm512_storeu_si512((__m512i*)&H[t], H1);
// making it 4 lanes to match accuracy
__m512i shfH, shft, max1, max2;
t_ = _mm512_set1_epi32(t);
t_ = _mm512_add_epi32(t_, bt32_);
shfH = _mm512_shuffle_i32x4(H1, H1, 0x31);
shft = _mm512_shuffle_i32x4(t_, t_, 0x31);
tmp_mask = _mm512_cmpgt_epi32_mask(shfH, H1);
max1 = _mm512_mask_blend_epi32(tmp_mask, H1, shfH);
max2 = _mm512_mask_blend_epi32(tmp_mask, t_, shft);
shfH = _mm512_shuffle_i32x4(max1, max1, 0x2);
shft = _mm512_shuffle_i32x4(max2, max2, 0x2);
tmp_mask = _mm512_cmpgt_epi32_mask(shfH, max1);
max1 = _mm512_mask_blend_epi32(tmp_mask, max1, shfH);
max2 = _mm512_mask_blend_epi32(tmp_mask, max2, shft);
tmp_mask = _mm512_cmpgt_epi32_mask(max1, max_H_);
max_H_ = _mm512_mask_blend_epi32(tmp_mask, max_H_, max1);
max_t_ = _mm512_mask_blend_epi32(tmp_mask, max_t_, max2);
}
_mm512_storeu_si512((__m512i*)HH, max_H_);
_mm512_storeu_si512((__m512i*)tt, max_t_);
int rem = (en0 - t) / 4;
for (int l=0; l<rem; l++) {
int bt = t;
for (int j=0; j<4; j++) {
H[t] += (int32_t)v8[t];
if (H[t] > HH[j]) {
HH[j] = H[t];
tt[j] = bt;
}
t++;
}
}
for (i = 0; i < 4; ++i)
if (max_H < HH[i]) max_H = HH[i], max_t = tt[i] + i;
for (; t < en0; ++t) { // for the rest of values that haven't been computed with SSE
H[t] += (int32_t)v8[t];
if (H[t] > max_H) {
max_H = H[t], max_t = t;
}
}
} else H[0] = v8[0] - qe, max_H = H[0], max_t = 0; // special casing r==0
// update ez
if (en0 == tlen - 1 && H[en0] > ez->mte) {
ez->mte = H[en0], ez->mte_q = r - en_new;
}
if (r - st0 == qlen - 1 && H[st0] > ez->mqe) {
ez->mqe = H[st0], ez->mqe_t = st0;
}
if (ksw_apply_zdrop(ez, 1, max_H, r, max_t, zdrop, e2)) {
break;
}
if (r == qlen + tlen - 2 && en0 == tlen - 1) {
ez->score = H[tlen - 1];
}
} else { // find approximate max; Z-drop might be inaccurate, too.
if (r > 0) {
if (last_H0_t >= st0 && last_H0_t <= en0 && last_H0_t + 1 >= st0 && last_H0_t + 1 <= en0) {
int32_t d0 = v8[last_H0_t];
int32_t d1 = u8[last_H0_t + 1];
if (d0 > d1) H0 += d0;
else H0 += d1, ++last_H0_t;
} else if (last_H0_t >= st0 && last_H0_t <= en0) {
H0 += v8[last_H0_t];
} else {
++last_H0_t, H0 += u8[last_H0_t];
}
} else H0 = v8[0] - qe, last_H0_t = 0;
if ((flag & KSW_EZ_APPROX_DROP) && ksw_apply_zdrop(ez, 1, H0, r, last_H0_t, zdrop, e2)) {
break;
}
if (r == qlen + tlen - 2 && en0 == tlen - 1) {
ez->score = H0;
}
}
// last_st = st, last_en = en;
last_st = st_new, last_en = en_new;
}
kfree(km, mem);
if (!approx_max) kfree(km, H);
if (with_cigar) { // backtrack
int rev_cigar = !!(flag & KSW_EZ_REV_CIGAR);
if (!ez->zdropped && !(flag&KSW_EZ_EXTZ_ONLY)) {
ksw_backtrack(km, 1, rev_cigar, 0, (uint8_t*)p, off, off_end, n_col_*64, tlen-1, qlen-1, &ez->m_cigar, &ez->n_cigar, &ez->cigar);
} else if (!ez->zdropped && (flag&KSW_EZ_EXTZ_ONLY) && ez->mqe + end_bonus > (int)ez->max) {
ez->reach_end = 1;
ksw_backtrack(km, 1, rev_cigar, 0, (uint8_t*)p, off, off_end, n_col_*64, ez->mqe_t, qlen-1, &ez->m_cigar, &ez->n_cigar, &ez->cigar);
} else if (ez->max_t >= 0 && ez->max_q >= 0) {
ksw_backtrack(km, 1, rev_cigar, 0, (uint8_t*)p, off, off_end, n_col_*64, ez->max_t, ez->max_q, &ez->m_cigar, &ez->n_cigar, &ez->cigar);
}
kfree(km, mem2); kfree(km, off);
}
}
#endif
Reference in New Issue View Git Blame Copy Permalink