Lines Matching +full:k +full:- +full:block
3 # This source code is licensed under the BSD-style license found in the
7 - name: xnn_f32_gemminc_minmax_ukernel_1x8__aarch64_neonfma_cortex_a53
9 k-block: 8
12 - name: xnn_f32_gemminc_minmax_ukernel_1x8__aarch64_neonfma_cortex_a75
14 k-block: 8
17 - name: xnn_f32_gemminc_minmax_ukernel_1x8__aarch64_neonfma_ld64
19 k-block: 2
21 - name: xnn_f32_gemminc_minmax_ukernel_1x8__aarch64_neonfma_prfm_cortex_a75
23 k-block: 8
26 - name: xnn_f32_gemminc_minmax_ukernel_1x12__aarch64_neonfma_cortex_a53
28 k-block: 4
31 - name: xnn_f32_gemminc_minmax_ukernel_4x8__aarch64_neonfma_cortex_a53
33 k-block: 4
36 - name: xnn_f32_gemminc_minmax_ukernel_4x8__aarch64_neonfma_cortex_a55
38 k-block: 4
41 - name: xnn_f32_gemminc_minmax_ukernel_4x8__aarch64_neonfma_cortex_a75
43 k-block: 8
46 - name: xnn_f32_gemminc_minmax_ukernel_4x8__aarch64_neonfma_ld64
48 k-block: 2
50 - name: xnn_f32_gemminc_minmax_ukernel_4x8__aarch64_neonfma_ld128
52 k-block: 4
54 - name: xnn_f32_gemminc_minmax_ukernel_4x8__aarch64_neonfma_prfm_cortex_a75
56 k-block: 8
59 - name: xnn_f32_gemminc_minmax_ukernel_4x12__aarch64_neonfma_cortex_a53
61 k-block: 4
64 - name: xnn_f32_gemminc_minmax_ukernel_5x8__aarch64_neonfma_prfm_cortex_a75
66 k-block: 8
69 - name: xnn_f32_gemminc_minmax_ukernel_6x8__aarch64_neonfma_cortex_a53
71 k-block: 4
74 - name: xnn_f32_gemminc_minmax_ukernel_6x8__aarch64_neonfma_cortex_a55
76 k-block: 4
79 - name: xnn_f32_gemminc_minmax_ukernel_6x8__aarch64_neonfma_cortex_a73
81 k-block: 8
84 - name: xnn_f32_gemminc_minmax_ukernel_6x8__aarch64_neonfma_cortex_a75
86 k-block: 4
88 - name: xnn_f32_gemminc_minmax_ukernel_6x8__aarch64_neonfma_ld64
90 k-block: 2
92 - name: xnn_f32_gemminc_minmax_ukernel_6x8__aarch64_neonfma_ld128
94 k-block: 4
96 - name: xnn_f32_gemminc_minmax_ukernel_6x8__aarch64_neonfma_prfm_cortex_a75
98 k-block: 8
102 - name: xnn_f32_gemminc_minmax_ukernel_1x8__neon_dup_ld64
104 k-block: 2
105 - name: xnn_f32_gemminc_minmax_ukernel_1x8__neon_lane_ld64
107 k-block: 2
108 - name: xnn_f32_gemminc_minmax_ukernel_1x8__neonfma_dup_ld64
110 k-block: 2
111 - name: xnn_f32_gemminc_minmax_ukernel_1x8__neonfma_lane_ld64
113 k-block: 2
115 - aarch64
116 - name: xnn_f32_gemminc_minmax_ukernel_1x8s4__neon
118 k-block: 4
119 - name: xnn_f32_gemminc_minmax_ukernel_1x8s4__neonfma
121 k-block: 4
122 - name: xnn_f32_gemminc_minmax_ukernel_4x8__neon_dup_ld64
124 k-block: 2
125 - name: xnn_f32_gemminc_minmax_ukernel_4x8__neon_dup_ld128
127 k-block: 4
128 - name: xnn_f32_gemminc_minmax_ukernel_4x8__neon_lane_ld64
130 k-block: 2
131 - name: xnn_f32_gemminc_minmax_ukernel_4x8__neon_lane_ld128
133 k-block: 4
134 - name: xnn_f32_gemminc_minmax_ukernel_4x8__neonfma_dup_ld64
136 k-block: 2
137 - name: xnn_f32_gemminc_minmax_ukernel_4x8__neonfma_dup_ld128
139 k-block: 4
140 - name: xnn_f32_gemminc_minmax_ukernel_4x8__neonfma_lane_ld64
142 k-block: 2
144 - aarch64
145 - name: xnn_f32_gemminc_minmax_ukernel_4x8__neonfma_lane_ld128
147 k-block: 4
149 - aarch64
150 - name: xnn_f32_gemminc_minmax_ukernel_4x8s4__neon
152 k-block: 4
153 - name: xnn_f32_gemminc_minmax_ukernel_4x8s4__neonfma
155 k-block: 4
156 - name: xnn_f32_gemminc_minmax_ukernel_5x8__neon_lane_ld64
158 k-block: 2
159 - name: xnn_f32_gemminc_minmax_ukernel_5x8__neonfma_lane_ld64
161 k-block: 2
163 - aarch64
164 - name: xnn_f32_gemminc_minmax_ukernel_6x8__neon_dup_ld64
166 k-block: 2
167 - name: xnn_f32_gemminc_minmax_ukernel_6x8__neon_dup_ld128
169 k-block: 4
170 - name: xnn_f32_gemminc_minmax_ukernel_6x8__neon_lane_ld64
172 k-block: 2
173 - name: xnn_f32_gemminc_minmax_ukernel_6x8__neon_lane_ld128
175 k-block: 4
176 - name: xnn_f32_gemminc_minmax_ukernel_6x8__neonfma_dup_ld64
178 k-block: 2
179 - name: xnn_f32_gemminc_minmax_ukernel_6x8__neonfma_dup_ld128
181 k-block: 4
182 - name: xnn_f32_gemminc_minmax_ukernel_6x8__neonfma_lane_ld64
184 k-block: 2
186 - aarch64
187 - name: xnn_f32_gemminc_minmax_ukernel_6x8__neonfma_lane_ld128
189 k-block: 4
191 - aarch64
192 - name: xnn_f32_gemminc_minmax_ukernel_6x8s4__neon
194 k-block: 4
195 - name: xnn_f32_gemminc_minmax_ukernel_6x8s4__neonfma
197 k-block: 4
198 - name: xnn_f32_gemminc_minmax_ukernel_8x8s4__neon
200 k-block: 4
201 - name: xnn_f32_gemminc_minmax_ukernel_8x8s4__neonfma
203 k-block: 4
205 - name: xnn_f32_gemminc_minmax_ukernel_1x8__sse_dup
207 k-block: 4
208 - name: xnn_f32_gemminc_minmax_ukernel_1x8__sse_load1
210 k-block: 1
211 - name: xnn_f32_gemminc_minmax_ukernel_1x8__sse2_dup
213 k-block: 4
214 - name: xnn_f32_gemminc_minmax_ukernel_1x8s4__sse
216 k-block: 4
217 - name: xnn_f32_gemminc_minmax_ukernel_3x8__sse_dup
219 k-block: 4
220 - name: xnn_f32_gemminc_minmax_ukernel_3x8__sse_load1
222 k-block: 1
223 - name: xnn_f32_gemminc_minmax_ukernel_3x8__sse2_dup
225 k-block: 4
226 - name: xnn_f32_gemminc_minmax_ukernel_3x8s4__sse
228 k-block: 4
229 - name: xnn_f32_gemminc_minmax_ukernel_4x8__sse_dup
231 k-block: 4
232 - name: xnn_f32_gemminc_minmax_ukernel_4x8__sse_load1
234 k-block: 1
235 - name: xnn_f32_gemminc_minmax_ukernel_4x8__sse2_dup
237 k-block: 4
238 - name: xnn_f32_gemminc_minmax_ukernel_4x8s4__sse
240 k-block: 4
241 - name: xnn_f32_gemminc_minmax_ukernel_5x8__sse_dup
243 k-block: 4
244 - name: xnn_f32_gemminc_minmax_ukernel_5x8__sse_load1
246 k-block: 1
247 - name: xnn_f32_gemminc_minmax_ukernel_5x8__sse2_dup
249 k-block: 4
250 - name: xnn_f32_gemminc_minmax_ukernel_5x8s4__sse
252 k-block: 4
254 - name: xnn_f32_gemminc_minmax_ukernel_1x8__avx_broadcast
256 k-block: 1
257 - name: xnn_f32_gemminc_minmax_ukernel_1x16__avx_broadcast
259 k-block: 1
260 - name: xnn_f32_gemminc_minmax_ukernel_3x16__avx_broadcast
262 k-block: 1
263 - name: xnn_f32_gemminc_minmax_ukernel_4x8__avx_broadcast
265 k-block: 1
266 - name: xnn_f32_gemminc_minmax_ukernel_4x16__avx_broadcast
268 k-block: 1
269 - name: xnn_f32_gemminc_minmax_ukernel_5x8__avx_broadcast
271 k-block: 1
272 - name: xnn_f32_gemminc_minmax_ukernel_5x16__avx_broadcast
274 k-block: 1
275 - name: xnn_f32_gemminc_minmax_ukernel_6x8__avx_broadcast
277 k-block: 1
278 - name: xnn_f32_gemminc_minmax_ukernel_7x8__avx_broadcast
280 k-block: 1
282 - name: xnn_f32_gemminc_minmax_ukernel_1x8__fma3_broadcast
284 k-block: 1
285 - name: xnn_f32_gemminc_minmax_ukernel_1x16__fma3_broadcast
287 k-block: 1
288 - name: xnn_f32_gemminc_minmax_ukernel_1x16s4__fma3_broadcast
290 k-block: 4
291 - name: xnn_f32_gemminc_minmax_ukernel_3x16__fma3_broadcast
293 k-block: 1
294 - name: xnn_f32_gemminc_minmax_ukernel_3x16s4__fma3_broadcast
296 k-block: 4
297 - name: xnn_f32_gemminc_minmax_ukernel_4x8__fma3_broadcast
299 k-block: 1
300 - name: xnn_f32_gemminc_minmax_ukernel_4x16__fma3_broadcast
302 k-block: 1
303 - name: xnn_f32_gemminc_minmax_ukernel_4x16s4__fma3_broadcast
305 k-block: 4
306 - name: xnn_f32_gemminc_minmax_ukernel_5x8__fma3_broadcast
308 k-block: 1
309 - name: xnn_f32_gemminc_minmax_ukernel_5x16__fma3_broadcast
311 k-block: 1
312 - name: xnn_f32_gemminc_minmax_ukernel_5x16s4__fma3_broadcast
314 k-block: 4
315 - name: xnn_f32_gemminc_minmax_ukernel_6x8__fma3_broadcast
317 k-block: 1
318 - name: xnn_f32_gemminc_minmax_ukernel_7x8__fma3_broadcast
320 k-block: 1
321 - name: xnn_f32_gemminc_minmax_ukernel_8x8__fma3_broadcast
323 k-block: 1
325 - name: xnn_f32_gemminc_minmax_ukernel_1x16__avx512f_broadcast
327 k-block: 1
328 - name: xnn_f32_gemminc_minmax_ukernel_4x16__avx512f_broadcast
330 k-block: 1
331 - name: xnn_f32_gemminc_minmax_ukernel_5x16__avx512f_broadcast
333 k-block: 1
334 - name: xnn_f32_gemminc_minmax_ukernel_6x16__avx512f_broadcast
336 k-block: 1
337 - name: xnn_f32_gemminc_minmax_ukernel_7x16__avx512f_broadcast
339 k-block: 1
340 - name: xnn_f32_gemminc_minmax_ukernel_8x16__avx512f_broadcast
342 k-block: 1
344 - name: xnn_f32_gemminc_minmax_ukernel_1x8__wasmsimd_arm_loadsplat
346 k-block: 1
347 - name: xnn_f32_gemminc_minmax_ukernel_1x8__wasmsimd_arm_splat
349 k-block: 4
350 - name: xnn_f32_gemminc_minmax_ukernel_1x8__wasmsimd_x86_loadsplat
352 k-block: 1
353 - name: xnn_f32_gemminc_minmax_ukernel_1x8__wasmsimd_x86_splat
355 k-block: 4
356 - name: xnn_f32_gemminc_minmax_ukernel_1x8s4__wasmsimd_arm
358 k-block: 4
359 - name: xnn_f32_gemminc_minmax_ukernel_1x8s4__wasmsimd_x86
361 k-block: 4
362 - name: xnn_f32_gemminc_minmax_ukernel_3x8__wasmsimd_arm_loadsplat
364 k-block: 1
365 - name: xnn_f32_gemminc_minmax_ukernel_3x8__wasmsimd_arm_splat
367 k-block: 4
368 - name: xnn_f32_gemminc_minmax_ukernel_3x8__wasmsimd_x86_loadsplat
370 k-block: 1
371 - name: xnn_f32_gemminc_minmax_ukernel_3x8__wasmsimd_x86_splat
373 k-block: 4
374 - name: xnn_f32_gemminc_minmax_ukernel_3x8s4__wasmsimd_arm
376 k-block: 4
377 - name: xnn_f32_gemminc_minmax_ukernel_3x8s4__wasmsimd_x86
379 k-block: 4
380 - name: xnn_f32_gemminc_minmax_ukernel_4x8__wasmsimd_arm_loadsplat
382 k-block: 1
383 - name: xnn_f32_gemminc_minmax_ukernel_4x8__wasmsimd_arm_splat
385 k-block: 4
386 - name: xnn_f32_gemminc_minmax_ukernel_4x8__wasmsimd_x86_loadsplat
388 k-block: 1
389 - name: xnn_f32_gemminc_minmax_ukernel_4x8__wasmsimd_x86_splat
391 k-block: 4
392 - name: xnn_f32_gemminc_minmax_ukernel_4x8s4__wasmsimd_arm
394 k-block: 4
395 - name: xnn_f32_gemminc_minmax_ukernel_4x8s4__wasmsimd_x86
397 k-block: 4
398 - name: xnn_f32_gemminc_minmax_ukernel_5x8__wasmsimd_arm_loadsplat
400 k-block: 1
401 - name: xnn_f32_gemminc_minmax_ukernel_5x8__wasmsimd_arm_splat
403 k-block: 4
404 - name: xnn_f32_gemminc_minmax_ukernel_5x8__wasmsimd_x86_loadsplat
406 k-block: 1
407 - name: xnn_f32_gemminc_minmax_ukernel_5x8__wasmsimd_x86_splat
409 k-block: 4
410 - name: xnn_f32_gemminc_minmax_ukernel_5x8s4__wasmsimd_arm
412 k-block: 4
413 - name: xnn_f32_gemminc_minmax_ukernel_5x8s4__wasmsimd_x86
415 k-block: 4
416 - name: xnn_f32_gemminc_minmax_ukernel_6x8__wasmsimd_arm_loadsplat
418 k-block: 1
419 - name: xnn_f32_gemminc_minmax_ukernel_6x8__wasmsimd_arm_splat
421 k-block: 4
422 - name: xnn_f32_gemminc_minmax_ukernel_6x8__wasmsimd_x86_loadsplat
424 k-block: 1
425 - name: xnn_f32_gemminc_minmax_ukernel_6x8__wasmsimd_x86_splat
427 k-block: 4
428 - name: xnn_f32_gemminc_minmax_ukernel_6x8s4__wasmsimd_arm
430 k-block: 4
431 - name: xnn_f32_gemminc_minmax_ukernel_6x8s4__wasmsimd_x86
433 k-block: 4
435 - name: xnn_f32_gemminc_minmax_ukernel_1x8__wasmrelaxedsimd_fma_loadsplat
437 k-block: 1
438 - name: xnn_f32_gemminc_minmax_ukernel_1x8__wasmrelaxedsimd_fma_splat
440 k-block: 4
441 - name: xnn_f32_gemminc_minmax_ukernel_1x8__wasmrelaxedsimd_loadsplat
443 k-block: 1
444 - name: xnn_f32_gemminc_minmax_ukernel_1x8__wasmrelaxedsimd_splat
446 k-block: 4
447 - name: xnn_f32_gemminc_minmax_ukernel_1x8s4__wasmrelaxedsimd
449 k-block: 4
450 - name: xnn_f32_gemminc_minmax_ukernel_1x8s4__wasmrelaxedsimd_fma
452 k-block: 4
453 - name: xnn_f32_gemminc_minmax_ukernel_3x8__wasmrelaxedsimd_fma_loadsplat
455 k-block: 1
456 - name: xnn_f32_gemminc_minmax_ukernel_3x8__wasmrelaxedsimd_fma_splat
458 k-block: 4
459 - name: xnn_f32_gemminc_minmax_ukernel_3x8__wasmrelaxedsimd_loadsplat
461 k-block: 1
462 - name: xnn_f32_gemminc_minmax_ukernel_3x8__wasmrelaxedsimd_splat
464 k-block: 4
465 - name: xnn_f32_gemminc_minmax_ukernel_3x8s4__wasmrelaxedsimd
467 k-block: 4
468 - name: xnn_f32_gemminc_minmax_ukernel_3x8s4__wasmrelaxedsimd_fma
470 k-block: 4
471 - name: xnn_f32_gemminc_minmax_ukernel_4x8__wasmrelaxedsimd_fma_loadsplat
473 k-block: 1
474 - name: xnn_f32_gemminc_minmax_ukernel_4x8__wasmrelaxedsimd_fma_splat
476 k-block: 4
477 - name: xnn_f32_gemminc_minmax_ukernel_4x8__wasmrelaxedsimd_loadsplat
479 k-block: 1
480 - name: xnn_f32_gemminc_minmax_ukernel_4x8__wasmrelaxedsimd_splat
482 k-block: 4
483 - name: xnn_f32_gemminc_minmax_ukernel_4x8s4__wasmrelaxedsimd
485 k-block: 4
486 - name: xnn_f32_gemminc_minmax_ukernel_4x8s4__wasmrelaxedsimd_fma
488 k-block: 4
489 - name: xnn_f32_gemminc_minmax_ukernel_5x8__wasmrelaxedsimd_fma_loadsplat
491 k-block: 1
492 - name: xnn_f32_gemminc_minmax_ukernel_5x8__wasmrelaxedsimd_fma_splat
494 k-block: 4
495 - name: xnn_f32_gemminc_minmax_ukernel_5x8__wasmrelaxedsimd_loadsplat
497 k-block: 1
498 - name: xnn_f32_gemminc_minmax_ukernel_5x8__wasmrelaxedsimd_splat
500 k-block: 4
501 - name: xnn_f32_gemminc_minmax_ukernel_5x8s4__wasmrelaxedsimd
503 k-block: 4
504 - name: xnn_f32_gemminc_minmax_ukernel_5x8s4__wasmrelaxedsimd_fma
506 k-block: 4
507 - name: xnn_f32_gemminc_minmax_ukernel_6x8__wasmrelaxedsimd_fma_loadsplat
509 k-block: 1
510 - name: xnn_f32_gemminc_minmax_ukernel_6x8__wasmrelaxedsimd_fma_splat
512 k-block: 4
513 - name: xnn_f32_gemminc_minmax_ukernel_6x8__wasmrelaxedsimd_loadsplat
515 k-block: 1
516 - name: xnn_f32_gemminc_minmax_ukernel_6x8__wasmrelaxedsimd_splat
518 k-block: 4
519 - name: xnn_f32_gemminc_minmax_ukernel_6x8s4__wasmrelaxedsimd
521 k-block: 4
522 - name: xnn_f32_gemminc_minmax_ukernel_6x8s4__wasmrelaxedsimd_fma
524 k-block: 4
526 - name: xnn_f32_gemminc_minmax_ukernel_1x4__wasm
528 k-block: 1
529 - name: xnn_f32_gemminc_minmax_ukernel_2x4__wasm
531 k-block: 1
532 - name: xnn_f32_gemminc_minmax_ukernel_4x4__wasm
534 k-block: 1
536 - name: xnn_f32_gemminc_minmax_ukernel_1x4__scalar
538 k-block: 1
539 - name: xnn_f32_gemminc_minmax_ukernel_2x4__scalar
541 k-block: 1
542 - name: xnn_f32_gemminc_minmax_ukernel_4x4__scalar
544 k-block: 1