forked from apache/tvm
-
Notifications
You must be signed in to change notification settings - Fork 0
/
gemm__avx2.c
531 lines (514 loc) · 22.1 KB
/
gemm__avx2.c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
#include <stddef.h>
#include <stdint.h>
#include <stdio.h>
#include <string.h>
#include <x86intrin.h>
void sgemm_compute_4x24__avx2(int32_t k, const float *a, int32_t a_off,
const float *b, int32_t b_off, float *c,
int32_t c_off, int32_t ldc) {
a = a + a_off;
b = b + b_off;
c = c + c_off;
size_t k_size_t = k;
size_t ldc_size_t = ldc;
asm volatile("shl $0x2,%[ldc_size_t]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"vzeroall\n\t"
"LOOP_START%=:\n\t"
"vmovaps (%[b]),%%ymm3\n\t"
"vmovaps 0x20(%[b]),%%ymm2\n\t"
"vmovaps 0x40(%[b]),%%ymm1\n\t"
"add $0x60,%[b]\n\t"
"vbroadcastss (%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm8\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm9\n\t"
"vfmadd231ps %%ymm1,%%ymm0,%%ymm10\n\t"
"vbroadcastss 0x4(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm11\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm12\n\t"
"vfmadd231ps %%ymm1,%%ymm0,%%ymm13\n\t"
"vbroadcastss 0x8(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm14\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm15\n\t"
"vfmadd231ps %%ymm1,%%ymm0,%%ymm7\n\t"
"vbroadcastss 0xc(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm6\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm5\n\t"
"vfmadd231ps %%ymm1,%%ymm0,%%ymm4\n\t"
"add $0x10,%[a]\n\t"
"dec %[k_size_t]\n\t"
"jne LOOP_START%=\n\t"
"vmovups %%ymm6,(%[c])\n\t"
"vmovups %%ymm5,0x20(%[c])\n\t"
"vmovups %%ymm4,0x40(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm14,(%[c])\n\t"
"vmovups %%ymm15,0x20(%[c])\n\t"
"vmovups %%ymm7,0x40(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm11,(%[c])\n\t"
"vmovups %%ymm12,0x20(%[c])\n\t"
"vmovups %%ymm13,0x40(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm8,(%[c])\n\t"
"vmovups %%ymm9,0x20(%[c])\n\t"
"vmovups %%ymm10,0x40(%[c])\n\t"
"vzeroupper\n\t"
: [c] "+r"(c), [b] "+r"(b), [a] "+r"(a),
[k_size_t] "+r"(k_size_t), [ldc_size_t] "+r"(ldc_size_t)
:
: "cc", "memory", "%ymm0", "%ymm1", "%ymm2", "%ymm3", "%ymm4",
"%ymm5", "%ymm6", "%ymm7", "%ymm8", "%ymm9", "%ymm10",
"%ymm11", "%ymm12", "%ymm13", "%ymm14", "%ymm15");
}
void sgemm_compute_5x16__avx2(int32_t k, const float *a, int32_t a_off,
const float *b, int32_t b_off, float *c,
int32_t c_off, int32_t ldc) {
a = a + a_off;
b = b + b_off;
c = c + c_off;
size_t k_size_t = k;
size_t ldc_size_t = ldc;
asm volatile("shl $0x2,%[ldc_size_t]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"vzeroall\n\t"
"LOOP_START%=:\n\t"
"vmovaps (%[b]),%%ymm3\n\t"
"vmovaps 0x20(%[b]),%%ymm2\n\t"
"add $0x40,%[b]\n\t"
"vbroadcastss (%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm8\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm9\n\t"
"vbroadcastss 0x4(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm11\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm12\n\t"
"vbroadcastss 0x8(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm14\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm15\n\t"
"vbroadcastss 0xc(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm6\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm5\n\t"
"vbroadcastss 0x10(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm13\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm10\n\t"
"add $0x14,%[a]\n\t"
"dec %[k_size_t]\n\t"
"jne LOOP_START%=\n\t"
"vmovups %%ymm13,(%[c])\n\t"
"vmovups %%ymm10,0x20(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm6,(%[c])\n\t"
"vmovups %%ymm5,0x20(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm14,(%[c])\n\t"
"vmovups %%ymm15,0x20(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm11,(%[c])\n\t"
"vmovups %%ymm12,0x20(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm8,(%[c])\n\t"
"vmovups %%ymm9,0x20(%[c])\n\t"
"vzeroupper\n\t"
: [c] "+r"(c), [b] "+r"(b), [a] "+r"(a),
[k_size_t] "+r"(k_size_t), [ldc_size_t] "+r"(ldc_size_t)
:
: "cc", "memory", "%ymm0", "%ymm1", "%ymm2", "%ymm3", "%ymm4",
"%ymm5", "%ymm6", "%ymm7", "%ymm8", "%ymm9", "%ymm10",
"%ymm11", "%ymm12", "%ymm13", "%ymm14", "%ymm15");
}
void sgemm_compute_6x16__avx2(int32_t k, const float *a, int32_t a_off,
const float *b, int32_t b_off, float *c,
int32_t c_off, int32_t ldc) {
a = a + a_off;
b = b + b_off;
c = c + c_off;
size_t k_size_t = k;
size_t ldc_size_t = ldc;
asm volatile("shl $0x2,%[ldc_size_t]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"vzeroall\n\t"
"LOOP_START%=:\n\t"
"vmovaps (%[b]),%%ymm3\n\t"
"vmovaps 0x20(%[b]),%%ymm2\n\t"
"add $0x40,%[b]\n\t"
"vbroadcastss (%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm8\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm9\n\t"
"vbroadcastss 0x4(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm11\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm12\n\t"
"vbroadcastss 0x8(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm14\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm15\n\t"
"vbroadcastss 0xc(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm6\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm5\n\t"
"vbroadcastss 0x10(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm13\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm10\n\t"
"vbroadcastss 0x14(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm1\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm4\n\t"
"add $0x18,%[a]\n\t"
"dec %[k_size_t]\n\t"
"jne LOOP_START%=\n\t"
"vmovups %%ymm1,(%[c])\n\t"
"vmovups %%ymm4,0x20(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm13,(%[c])\n\t"
"vmovups %%ymm10,0x20(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm6,(%[c])\n\t"
"vmovups %%ymm5,0x20(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm14,(%[c])\n\t"
"vmovups %%ymm15,0x20(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm11,(%[c])\n\t"
"vmovups %%ymm12,0x20(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm8,(%[c])\n\t"
"vmovups %%ymm9,0x20(%[c])\n\t"
"vzeroupper\n\t"
: [c] "+r"(c), [b] "+r"(b), [a] "+r"(a),
[k_size_t] "+r"(k_size_t), [ldc_size_t] "+r"(ldc_size_t)
:
: "cc", "memory", "%ymm0", "%ymm1", "%ymm2", "%ymm3", "%ymm4",
"%ymm5", "%ymm6", "%ymm7", "%ymm8", "%ymm9", "%ymm10",
"%ymm11", "%ymm12", "%ymm13", "%ymm14", "%ymm15");
}
void sgemm_compute_4x16__avx2(int32_t k, const float *a, int32_t a_off,
const float *b, int32_t b_off, float *c,
int32_t c_off, int32_t ldc) {
a = a + a_off;
b = b + b_off;
c = c + c_off;
size_t k_size_t = k;
size_t ldc_size_t = ldc;
asm volatile("shl $0x2,%[ldc_size_t]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"vzeroall\n\t"
"LOOP_START%=:\n\t"
"vmovaps (%[b]),%%ymm3\n\t"
"vmovaps 0x20(%[b]),%%ymm2\n\t"
"add $0x40,%[b]\n\t"
"vbroadcastss (%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm8\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm9\n\t"
"vbroadcastss 0x4(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm11\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm12\n\t"
"vbroadcastss 0x8(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm14\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm15\n\t"
"vbroadcastss 0xc(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm6\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm5\n\t"
"add $0x10,%[a]\n\t"
"dec %[k_size_t]\n\t"
"jne LOOP_START%=\n\t"
"vmovups %%ymm6,(%[c])\n\t"
"vmovups %%ymm5,0x20(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm14,(%[c])\n\t"
"vmovups %%ymm15,0x20(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm11,(%[c])\n\t"
"vmovups %%ymm12,0x20(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm8,(%[c])\n\t"
"vmovups %%ymm9,0x20(%[c])\n\t"
"vzeroupper\n\t"
: [c] "+r"(c), [b] "+r"(b), [a] "+r"(a),
[k_size_t] "+r"(k_size_t), [ldc_size_t] "+r"(ldc_size_t)
:
: "cc", "memory", "%ymm0", "%ymm1", "%ymm2", "%ymm3", "%ymm4",
"%ymm5", "%ymm6", "%ymm7", "%ymm8", "%ymm9", "%ymm10",
"%ymm11", "%ymm12", "%ymm13", "%ymm14", "%ymm15");
}
void sgemm_compute_4x8__avx2(int32_t k, const float *a, int32_t a_off,
const float *b, int32_t b_off, float *c,
int32_t c_off, int32_t ldc) {
a = a + a_off;
b = b + b_off;
c = c + c_off;
size_t k_size_t = k;
size_t ldc_size_t = ldc;
asm volatile("shl $0x2,%[ldc_size_t]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"vzeroall\n\t"
"LOOP_START%=:\n\t"
"vmovaps (%[b]),%%ymm3\n\t"
"add $0x20,%[b]\n\t"
"vbroadcastss (%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm8\n\t"
"vbroadcastss 0x4(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm11\n\t"
"vbroadcastss 0x8(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm14\n\t"
"vbroadcastss 0xc(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm6\n\t"
"add $0x10,%[a]\n\t"
"dec %[k_size_t]\n\t"
"jne LOOP_START%=\n\t"
"vmovups %%ymm6,(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm14,(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm11,(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm8,(%[c])\n\t"
"vzeroupper\n\t"
: [c] "+r"(c), [b] "+r"(b), [a] "+r"(a),
[k_size_t] "+r"(k_size_t), [ldc_size_t] "+r"(ldc_size_t)
:
: "cc", "memory", "%ymm0", "%ymm1", "%ymm2", "%ymm3", "%ymm4",
"%ymm5", "%ymm6", "%ymm7", "%ymm8", "%ymm9", "%ymm10",
"%ymm11", "%ymm12", "%ymm13", "%ymm14", "%ymm15");
}
void sgemm_reset_4x24__avx2(float *c, int32_t c_off, int32_t ldc) {
c = c + c_off;
size_t ldc_size_t = ldc;
asm volatile("shl $0x2,%[ldc_size_t]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"vzeroall\n\t"
"vmovups %%ymm6,(%[c])\n\t"
"vmovups %%ymm5,0x20(%[c])\n\t"
"vmovups %%ymm4,0x40(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm14,(%[c])\n\t"
"vmovups %%ymm15,0x20(%[c])\n\t"
"vmovups %%ymm7,0x40(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm11,(%[c])\n\t"
"vmovups %%ymm12,0x20(%[c])\n\t"
"vmovups %%ymm13,0x40(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vmovups %%ymm8,(%[c])\n\t"
"vmovups %%ymm9,0x20(%[c])\n\t"
"vmovups %%ymm10,0x40(%[c])\n\t"
"vzeroupper\n\t"
: [c] "+r"(c), [ldc_size_t] "+r"(ldc_size_t)
:
: "cc", "memory", "%ymm0", "%ymm1", "%ymm2", "%ymm3", "%ymm4",
"%ymm5", "%ymm6", "%ymm7", "%ymm8", "%ymm9", "%ymm10",
"%ymm11", "%ymm12", "%ymm13", "%ymm14", "%ymm15");
}
void sgemm_update_4x24__avx2(int32_t k, const float *a, int32_t a_off,
const float *b, int32_t b_off, float *c,
int32_t c_off, int32_t ldc) {
a = a + a_off;
b = b + b_off;
c = c + c_off;
size_t k_size_t = k;
size_t ldc_size_t = ldc;
asm volatile("shl $0x2,%[ldc_size_t]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"add %[ldc_size_t],%[c]\n\t"
"prefetcht0 (%[c])\n\t"
"vzeroall\n\t"
"LOOP_START%=:\n\t"
"vmovaps (%[b]),%%ymm3\n\t"
"vmovaps 0x20(%[b]),%%ymm2\n\t"
"vmovaps 0x40(%[b]),%%ymm1\n\t"
"add $0x60,%[b]\n\t"
"vbroadcastss (%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm8\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm9\n\t"
"vfmadd231ps %%ymm1,%%ymm0,%%ymm10\n\t"
"vbroadcastss 0x4(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm11\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm12\n\t"
"vfmadd231ps %%ymm1,%%ymm0,%%ymm13\n\t"
"vbroadcastss 0x8(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm14\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm15\n\t"
"vfmadd231ps %%ymm1,%%ymm0,%%ymm7\n\t"
"vbroadcastss 0xc(%[a]),%%ymm0\n\t"
"vfmadd231ps %%ymm3,%%ymm0,%%ymm6\n\t"
"vfmadd231ps %%ymm2,%%ymm0,%%ymm5\n\t"
"vfmadd231ps %%ymm1,%%ymm0,%%ymm4\n\t"
"add $0x10,%[a]\n\t"
"dec %[k_size_t]\n\t"
"jne LOOP_START%=\n\t"
"vaddps (%[c]),%%ymm6,%%ymm6\n\t"
"vmovups %%ymm6,(%[c])\n\t"
"vaddps 0x20(%[c]),%%ymm5,%%ymm5\n\t"
"vmovups %%ymm5,0x20(%[c])\n\t"
"vaddps 0x40(%[c]),%%ymm4,%%ymm4\n\t"
"vmovups %%ymm4,0x40(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vaddps (%[c]),%%ymm14,%%ymm14\n\t"
"vmovups %%ymm14,(%[c])\n\t"
"vaddps 0x20(%[c]),%%ymm15,%%ymm15\n\t"
"vmovups %%ymm15,0x20(%[c])\n\t"
"vaddps 0x40(%[c]),%%ymm7,%%ymm7\n\t"
"vmovups %%ymm7,0x40(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vaddps (%[c]),%%ymm11,%%ymm11\n\t"
"vmovups %%ymm11,(%[c])\n\t"
"vaddps 0x20(%[c]),%%ymm12,%%ymm12\n\t"
"vmovups %%ymm12,0x20(%[c])\n\t"
"vaddps 0x40(%[c]),%%ymm13,%%ymm13\n\t"
"vmovups %%ymm13,0x40(%[c])\n\t"
"sub %[ldc_size_t],%[c]\n\t"
"vaddps (%[c]),%%ymm8,%%ymm8\n\t"
"vmovups %%ymm8,(%[c])\n\t"
"vaddps 0x20(%[c]),%%ymm9,%%ymm9\n\t"
"vmovups %%ymm9,0x20(%[c])\n\t"
"vaddps 0x40(%[c]),%%ymm10,%%ymm10\n\t"
"vmovups %%ymm10,0x40(%[c])\n\t"
"vzeroupper\n\t"
: [c] "+r"(c), [b] "+r"(b), [a] "+r"(a),
[k_size_t] "+r"(k_size_t), [ldc_size_t] "+r"(ldc_size_t)
:
: "cc", "memory", "%ymm0", "%ymm1", "%ymm2", "%ymm3", "%ymm4",
"%ymm5", "%ymm6", "%ymm7", "%ymm8", "%ymm9", "%ymm10",
"%ymm11", "%ymm12", "%ymm13", "%ymm14", "%ymm15");
}
void qgemm_compute_8x8x4__avx2(int32_t K, const uint8_t *A, int32_t a_off,
const int8_t *B, int32_t b_off, int32_t *C,
int32_t c_off, int32_t ldc) {
A += a_off;
B += b_off;
C += c_off;
uint64_t ldc_ = ldc * sizeof(C[0]);
uint64_t k_updated = 8 * K;
int16_t vone = 1;
int16_t *vonep = &vone;
asm volatile(
#if !defined(__clang__)
"\t mov r8, %[k_updated]\n"
"\t mov r9, %[A]\n"
"\t mov rax, %[vonep]\n"
"\t mov r10, %[B]\n"
"\t mov rsi, %[B]\n"
"\t mov r12, %[C]\n"
"\t mov r13, %[ldc]\n"
#else
"\t mov %[k_updated], %%r8\n"
"\t mov %[A], %%r9\n"
"\t mov %[vonep], %%rax\n"
"\t mov %[B], %%r10\n"
"\t mov %[B], %%rsi\n"
"\t mov %[C], %%r12\n"
"\t mov %[ldc], %%r13\n"
"\t .intel_syntax noprefix\n"
#endif
// initialize all Ctile registers to zeros
"\t vzeroall\n"
// load constant "1"
"\t vpbroadcastw ymm15, WORD PTR [rax]\n"
"\t mov r14, 0\n"
"\t loop_inner%=:\n"
// <Main micro-kernel>
// load all B's
"\t mov rbx, r10\n"
"\t vmovaps ymm9,YMMWORD PTR [rbx]\n"
"\t vpbroadcastd ymm8,DWORD PTR [r9+r14 + 0]\n"
"\t vpmaddubsw ymm14, ymm8, ymm9\n"
"\t vpmaddwd ymm14, ymm15, ymm14\n"
"\t vpaddd ymm0, ymm0, ymm14\n"
"\t vpbroadcastd ymm10,DWORD PTR [r9+r14 + 4]\n"
"\t vpmaddubsw ymm14, ymm10, ymm9\n"
"\t vpmaddwd ymm14, ymm15, ymm14\n"
"\t vpaddd ymm1, ymm1, ymm14\n"
"\t vpbroadcastd ymm11,DWORD PTR [r9+r14 + 8]\n"
"\t vpmaddubsw ymm14, ymm11, ymm9\n"
"\t vpmaddwd ymm14, ymm15, ymm14\n"
"\t vpaddd ymm2, ymm2, ymm14\n"
"\t vpbroadcastd ymm12,DWORD PTR [r9+r14 + 12]\n"
"\t vpmaddubsw ymm14, ymm12, ymm9\n"
"\t vpmaddwd ymm14, ymm15, ymm14\n"
"\t vpaddd ymm3, ymm3, ymm14\n"
"\t vpbroadcastd ymm13,DWORD PTR [r9+r14 + 16]\n"
"\t vpmaddubsw ymm14, ymm13, ymm9\n"
"\t vpmaddwd ymm14, ymm15, ymm14\n"
"\t vpaddd ymm4, ymm4, ymm14\n"
"\t vpbroadcastd ymm8,DWORD PTR [r9+r14 + 20]\n"
"\t vpmaddubsw ymm14, ymm8, ymm9\n"
"\t vpmaddwd ymm14, ymm15, ymm14\n"
"\t vpaddd ymm5, ymm5, ymm14\n"
"\t vpbroadcastd ymm10,DWORD PTR [r9+r14 + 24]\n"
"\t vpmaddubsw ymm14, ymm10, ymm9\n"
"\t vpmaddwd ymm14, ymm15, ymm14\n"
"\t vpaddd ymm6, ymm6, ymm14\n"
"\t vpbroadcastd ymm11,DWORD PTR [r9+r14 + 28]\n"
"\t vpmaddubsw ymm14, ymm11, ymm9\n"
"\t vpmaddwd ymm14, ymm15, ymm14\n"
"\t vpaddd ymm7, ymm7, ymm14\n"
// loop increment
"\t add r10, 32\n"
"\t add r14, 4 * 8\n"
"\t cmp r14, r8\n"
"\t jl loop_inner%=\n"
"\t vmovups YMMWORD PTR [r12 + 0], ymm0\n"
"\t add r12, r13\n"
"\t vmovups YMMWORD PTR [r12 + 0], ymm1\n"
"\t add r12, r13\n"
"\t vmovups YMMWORD PTR [r12 + 0], ymm2\n"
"\t add r12, r13\n"
"\t vmovups YMMWORD PTR [r12 + 0], ymm3\n"
"\t add r12, r13\n"
"\t vmovups YMMWORD PTR [r12 + 0], ymm4\n"
"\t add r12, r13\n"
"\t vmovups YMMWORD PTR [r12 + 0], ymm5\n"
"\t add r12, r13\n"
"\t vmovups YMMWORD PTR [r12 + 0], ymm6\n"
"\t add r12, r13\n"
"\t vmovups YMMWORD PTR [r12 + 0], ymm7\n"
// register list
:
: [k_updated] "rm"(k_updated), [A] "rm"(A), [B] "rm"(B), [C] "rm"(C),
[vonep] "rm"(vonep), [ldc] "rm"(ldc_)
: "cc", "r8", "r9", "r10", "r11", "r12", "r13", "r14", "r15", "rax",
"rbx", "rsi", "memory", "%ymm0", "%ymm1", "%ymm2", "%ymm3", "%ymm4",
"%ymm5", "%ymm6", "%ymm7", "%ymm8", "%ymm9", "%ymm10", "%ymm11",
"%ymm12", "%ymm13", "%ymm14", "%ymm15");
}