Lines Matching +full:5 +full:v0
152 half_predicates[i] = (i % 3 == 0 || i % 5 == 0) ? ~0 : 0; in create_offsets_values_preds_16()
227 "v0 = vmem(%2 + #0)\n\t" in vector_scatter_16()
229 "vscatter(%0, m0, v0.h).h = v1\n\t" in vector_scatter_16()
232 : "m0", "v0", "v1", "memory"); in vector_scatter_16()
241 "v0 = vmem(%2 + #0)\n\t" in vector_scatter_16_acc()
243 "vscatter(%0, m0, v0.h).h += v1\n\t" in vector_scatter_16_acc()
246 : "m0", "v0", "v1", "memory"); in vector_scatter_16_acc()
255 "v0 = vmem(%0 + #0)\n\t" in vector_scatter_16_masked()
256 "q0 = vand(v0, r1)\n\t" in vector_scatter_16_masked()
258 "v0 = vmem(%3 + #0)\n\t" in vector_scatter_16_masked()
260 "if (q0) vscatter(%1, m0, v0.h).h = v1\n\t" in vector_scatter_16_masked()
263 : "r1", "q0", "m0", "q0", "v0", "v1", "memory"); in vector_scatter_16_masked()
277 "v0 = vmem(%2 + #0)\n\t" in vector_scatter_32()
279 "vscatter(%0, m0, v0.w).w = v1\n\t" in vector_scatter_32()
282 : "m0", "v0", "v1", "memory"); in vector_scatter_32()
284 "v0 = vmem(%2 + #0)\n\t" in vector_scatter_32()
286 "vscatter(%0, m0, v0.w).w = v1\n\t" in vector_scatter_32()
289 : "m0", "v0", "v1", "memory"); in vector_scatter_32()
303 "v0 = vmem(%2 + #0)\n\t" in vector_scatter_32_acc()
305 "vscatter(%0, m0, v0.w).w += v1\n\t" in vector_scatter_32_acc()
308 : "m0", "v0", "v1", "memory"); in vector_scatter_32_acc()
310 "v0 = vmem(%2 + #0)\n\t" in vector_scatter_32_acc()
312 "vscatter(%0, m0, v0.w).w += v1\n\t" in vector_scatter_32_acc()
315 : "m0", "v0", "v1", "memory"); in vector_scatter_32_acc()
331 "v0 = vmem(%0 + #0)\n\t" in vector_scatter_32_masked()
332 "q0 = vand(v0, r1)\n\t" in vector_scatter_32_masked()
334 "v0 = vmem(%3 + #0)\n\t" in vector_scatter_32_masked()
336 "if (q0) vscatter(%1, m0, v0.w).w = v1\n\t" in vector_scatter_32_masked()
339 : "r1", "q0", "m0", "q0", "v0", "v1", "memory"); in vector_scatter_32_masked()
341 "v0 = vmem(%0 + #0)\n\t" in vector_scatter_32_masked()
342 "q0 = vand(v0, r1)\n\t" in vector_scatter_32_masked()
344 "v0 = vmem(%3 + #0)\n\t" in vector_scatter_32_masked()
346 "if (q0) vscatter(%1, m0, v0.w).w = v1\n\t" in vector_scatter_32_masked()
349 : "r1", "q0", "m0", "q0", "v0", "v1", "memory"); in vector_scatter_32_masked()
358 "v0 = vmem(%2 + #0)\n\t" in vector_scatter_16_32()
365 : "m0", "v0", "v1", "v2", "memory"); in vector_scatter_16_32()
374 "v0 = vmem(%2 + #0)\n\t" in vector_scatter_16_32_acc()
381 : "m0", "v0", "v1", "v2", "memory"); in vector_scatter_16_32_acc()
390 "v0 = vmem(%0 + #0)\n\t" in vector_scatter_16_32_masked()
391 "v0.h = vshuff(v0.h)\n\t" /* shuffle the predicates */ in vector_scatter_16_32_masked()
392 "q0 = vand(v0, r1)\n\t" in vector_scatter_16_32_masked()
394 "v0 = vmem(%3 + #0)\n\t" in vector_scatter_16_32_masked()
401 : "r1", "q0", "m0", "v0", "v1", "v2", "memory"); in vector_scatter_16_32_masked()
410 "v0 = vmem(%2 + #0)\n\t" in vector_gather_16()
411 "{ vtmp.h = vgather(%0, m0, v0.h).h\n\t" in vector_gather_16()
415 : "m0", "v0", "memory"); in vector_gather_16()
431 asm ("v0.h = vsplat(%5)\n\t" in vector_gather_16_masked()
432 "vmem(%4 + #0) = v0\n\t" /* initialize the write area */ in vector_gather_16_masked()
434 "v0 = vmem(%0 + #0)\n\t" in vector_gather_16_masked()
435 "q0 = vand(v0, r1)\n\t" in vector_gather_16_masked()
437 "v0 = vmem(%3 + #0)\n\t" in vector_gather_16_masked()
438 "{ if (q0) vtmp.h = vgather(%1, m0, v0.h).h\n\t" in vector_gather_16_masked()
442 : "r1", "q0", "m0", "v0", "memory"); in vector_gather_16_masked()
456 "v0 = vmem(%2 + #0)\n\t" in vector_gather_32()
457 "{ vtmp.w = vgather(%0, m0, v0.w).w\n\t" in vector_gather_32()
461 : "m0", "v0", "memory"); in vector_gather_32()
463 "v0 = vmem(%2 + #0)\n\t" in vector_gather_32()
464 "{ vtmp.w = vgather(%0, m0, v0.w).w\n\t" in vector_gather_32()
468 : "m0", "v0", "memory"); in vector_gather_32()
491 asm ("v0.h = vsplat(%5)\n\t" in vector_gather_32_masked()
492 "vmem(%4 + #0) = v0\n\t" /* initialize the write area */ in vector_gather_32_masked()
494 "v0 = vmem(%0 + #0)\n\t" in vector_gather_32_masked()
495 "q0 = vand(v0, r1)\n\t" in vector_gather_32_masked()
497 "v0 = vmem(%3 + #0)\n\t" in vector_gather_32_masked()
498 "{ if (q0) vtmp.w = vgather(%1, m0, v0.w).w\n\t" in vector_gather_32_masked()
502 : "r1", "q0", "m0", "v0", "memory"); in vector_gather_32_masked()
503 asm ("v0.h = vsplat(%5)\n\t" in vector_gather_32_masked()
504 "vmem(%4 + #0) = v0\n\t" /* initialize the write area */ in vector_gather_32_masked()
506 "v0 = vmem(%0 + #0)\n\t" in vector_gather_32_masked()
507 "q0 = vand(v0, r1)\n\t" in vector_gather_32_masked()
509 "v0 = vmem(%3 + #0)\n\t" in vector_gather_32_masked()
510 "{ if (q0) vtmp.w = vgather(%1, m0, v0.w).w\n\t" in vector_gather_32_masked()
514 : "r1", "q0", "m0", "v0", "memory"); in vector_gather_32_masked()
524 "v0 = vmem(%2 + #0)\n\t" in vector_gather_16_32()
528 "v0 = vmem(%3 + #0)\n\t" in vector_gather_16_32()
529 "v0.h = vdeal(v0.h)\n\t" /* deal the elements to get the order back */ in vector_gather_16_32()
530 "vmem(%3 + #0) = v0\n\t" in vector_gather_16_32()
533 : "m0", "v0", "v1", "memory"); in vector_gather_16_32()
543 asm ("v0.h = vsplat(%5)\n\t" in vector_gather_16_32_masked()
544 "vmem(%4 + #0) = v0\n\t" /* initialize the write area */ in vector_gather_16_32_masked()
546 "v0 = vmem(%0 + #0)\n\t" in vector_gather_16_32_masked()
547 "v0.h = vshuff(v0.h)\n\t" /* shuffle the predicates */ in vector_gather_16_32_masked()
548 "q0 = vand(v0, r1)\n\t" in vector_gather_16_32_masked()
550 "v0 = vmem(%3 + #0)\n\t" in vector_gather_16_32_masked()
554 "v0 = vmem(%4 + #0)\n\t" in vector_gather_16_32_masked()
555 "v0.h = vdeal(v0.h)\n\t" /* deal the elements to get the order back */ in vector_gather_16_32_masked()
556 "vmem(%4 + #0) = v0\n\t" in vector_gather_16_32_masked()
559 : "r1", "q0", "m0", "v0", "v1", "memory"); in vector_gather_16_32_masked()