File size: 3,271 Bytes
8b7c501 |
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 |
// Auto-generated file. Do not edit!
// Template: src/f32-gemm/1x8-aarch64-neonfma-ld64.S.in
// Generator: tools/xngen
//
// Copyright 2020 Google LLC
//
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree.
#include <xnnpack/assembly.h>
# void xnn_f32_gemm_minmax_ukernel_1x8__asm_aarch64_neonfma_ld64_prfm(
# size_t mr, (x0) - unused. mr = 1
# size_t nc, x1
# size_t kc, x2 / x0
# const float* a, x3
# size_t a_stride, (x4) - unused
# const void* w, x5
# float* c, x6
# size_t cm_stride, (x7) - unused
# size_t cn_stride, [sp] -> x14
# const xnn_f32_minmax_params* params) [sp + 8] -> (x8)
# d8-d15, x19-x30 need to be preserved if used. x18 is reserved by the OS.
# Register usage
# A0 x3 v0
# B x5 v20 v21 v22 v23
# C0 x6 v16 v17
# Clamp v4 v5
BEGIN_FUNCTION xnn_f32_gemm_minmax_ukernel_1x8__asm_aarch64_neonfma_ld64_prfm
# Load cn_stride, params pointer
LDP x14, x8, [sp]
# Load min/max values
LD2R {v4.4s, v5.4s}, [x8]
0:
# Load initial bias from w into accumulators
LDP q16, q17, [x5], 32
SUBS x0, x2, 8 // k = kc - 8
# Is there at least 2 floats (8 bytes)
B.LO 3f
PRFM PLDL1KEEP, [x5]
PRFM PLDL1KEEP, [x5, 64]
PRFM PLDL1KEEP, [x5, 128]
PRFM PLDL1KEEP, [x5, 192]
# Main loop - 2 floats of A (8 bytes)
1:
LDR d0, [x3], 8
LDP q20, q21, [x5], 32 // 16 F32 weights
LDP q22, q23, [x5], 32
SUBS x0, x0, 8
FMLA v16.4s, v20.4s, v0.s[0]
PRFM PLDL1KEEP, [x5, 128]
FMLA v17.4s, v21.4s, v0.s[0]
FMLA v16.4s, v22.4s, v0.s[1]
FMLA v17.4s, v23.4s, v0.s[1]
B.HS 1b
# Is there a remainder?- 1 float of A (4 bytes)
TBNZ x0, 2, 3f
2:
SUBS x1, x1, 8
# Clamp
FMAX v16.4s, v16.4s, v4.4s
FMAX v17.4s, v17.4s, v4.4s
FMIN v16.4s, v16.4s, v5.4s
FMIN v17.4s, v17.4s, v5.4s
# Store full 1 x 8
B.LO 4f
STP q16, q17, [x6]
ADD x6, x6, x14
SUB x3, x3, x2 // a0 -= kc
B.HI 0b
RET
3:
# Remainder- 1 float of A (4 bytes)
LDR s0, [x3], 4
LDP q20, q21, [x5], 32 // 8 F32 weights
FMLA v16.4s, v20.4s, v0.s[0]
FMLA v17.4s, v21.4s, v0.s[0]
B 2b
# Store odd channels
4:
TBZ x1, 2, 5f
STR q16, [x6], 16
MOV v16.16b, v17.16b
5:
TBZ x1, 1, 6f
STR d16, [x6], 8
DUP d16, v16.d[1]
6:
TBZ x1, 0, 7f
STR s16, [x6]
7:
RET
END_FUNCTION xnn_f32_gemm_minmax_ukernel_1x8__asm_aarch64_neonfma_ld64_prfm
#ifdef __ELF__
.section ".note.GNU-stack","",%progbits
#endif
|