-
Notifications
You must be signed in to change notification settings - Fork 5
/
Copy pathqsort_AVX3_SKL_512bit_long.s
374 lines (295 loc) · 7.13 KB
/
qsort_AVX3_SKL_512bit_long.s
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
.set pivotALU, %rcx
.set A0, %ymm0
.set A1, %ymm1
.set A2, %ymm2
.set A3, %ymm3
.set A4, %ymm4
.set A5, %ymm5
.set A6, %ymm6
.set M0, %ymm7
.set M1, %ymm8
.set M2, %ymm9
.set M3, %ymm10
.set M4, %ymm11
.set M5, %ymm12
.set M6, %ymm13
.set T, %ymm14
.set PIVOT, %ymm15
.set r0, %r8
.set r1, %r9
.set r2, %r10
.set r3, %r11
.set r4, %r15
.set r5, %r12
.set r6, %rax
# This one sorts in descending order, sort in ascending order, requires swapping the cmp operands.
.align 16
.globl qsort_AVX2
qsort_AVX2:
.set array, %rdi
.set temp_space, %rsi
.set n, %rdx
.set bottom, %r13
.set top, %r14
push %r12
push %r13
push %r14
push %r15
push %rbp
mov -8(array, n, 8), pivotALU
vpbroadcastq -8(array, n, 8), %zmm15
lea (array), bottom
lea (temp_space), top
dec n
.Lx7_loop:
cmp $56, n
jl .Lx7_loop_exit
# Load the next 28 elements
vmovdqu64 64*0(array), %zmm0
vmovdqu64 64*1(array), %zmm1
vmovdqu64 64*2(array), %zmm2
vmovdqu64 64*3(array), %zmm3
vmovdqu64 64*4(array), %zmm4
vmovdqu64 64*5(array), %zmm5
vmovdqu64 64*6(array), %zmm6
# Find elements lesser-than-equal to PIVOT
vpcmpgtq %zmm0, %zmm15, %k1
vpcmpgtq %zmm1, %zmm15, %k2
vpcmpgtq %zmm2, %zmm15, %k3
vpcmpgtq %zmm3, %zmm15, %k4
vpcmpgtq %zmm4, %zmm15, %k5
vpcmpgtq %zmm5, %zmm15, %k6
vpcmpgtq %zmm6, %zmm15, %k7
kmovq %k1, r0
kmovq %k2, r1
kmovq %k3, r2
kmovq %k4, r3
kmovq %k5, r4
kmovq %k6, r5
kmovq %k7, r6
popcnt r0, r0
popcnt r1, r1
popcnt r2, r2
popcnt r3, r3
popcnt r4, r4
popcnt r5, r5
popcnt r6, r6
shl $3, r0
shl $3, r1
shl $3, r2
shl $3, r3
shl $3, r4
shl $3, r5
shl $3, r6
vcompresspd %zmm0, (bottom){%k1}
add r0, bottom
vcompresspd %zmm1, (bottom){%k2}
add r1, bottom
vcompresspd %zmm2, (bottom){%k3}
add r2, bottom
vcompresspd %zmm3, (bottom){%k4}
add r3, bottom
vcompresspd %zmm4, (bottom){%k5}
add r4, bottom
vcompresspd %zmm5, (bottom){%k6}
add r5, bottom
vcompresspd %zmm6, (bottom){%k7}
add r6, bottom
knotb %k1, %k1
knotb %k2, %k2
knotb %k3, %k3
knotb %k4, %k4
knotb %k5, %k5
knotb %k6, %k6
knotb %k7, %k7
sub $64, r0
sub $64, r1
sub $64, r2
sub $64, r3
sub $64, r4
sub $64, r5
sub $64, r6
vcompresspd %zmm0, (top){%k1}
sub r0, top
vcompresspd %zmm1, (top){%k2}
sub r1, top
vcompresspd %zmm2, (top){%k3}
sub r2, top
vcompresspd %zmm3, (top){%k4}
sub r3, top
vcompresspd %zmm4, (top){%k5}
sub r4, top
vcompresspd %zmm5, (top){%k6}
sub r5, top
vcompresspd %zmm6, (top){%k7}
sub r6, top
sub $56, n
add $56*8, array
jmp .Lx7_loop
.Lx7_loop_exit:
.Lx3_loop:
cmp $24, n
jl .Lx3_loop_exit
# Load the next 28 elements
vmovdqu64 64*0(array), %zmm0
vmovdqu64 64*1(array), %zmm1
vmovdqu64 64*2(array), %zmm2
# Find elements lesser-than-equal to PIVOT
vpcmpgtq %zmm0, %zmm15, %k1
vpcmpgtq %zmm1, %zmm15, %k2
vpcmpgtq %zmm2, %zmm15, %k3
kmovq %k1, r0
kmovq %k2, r1
kmovq %k3, r2
popcnt r0, r0
popcnt r1, r1
popcnt r2, r2
shl $3, r0
shl $3, r1
shl $3, r2
vcompresspd %zmm0, (bottom){%k1}
add r0, bottom
vcompresspd %zmm1, (bottom){%k2}
add r1, bottom
vcompresspd %zmm2, (bottom){%k3}
add r2, bottom
knotb %k1, %k1
knotb %k2, %k2
knotb %k3, %k3
sub $64, r0
sub $64, r1
sub $64, r2
vcompresspd %zmm0, (top){%k1}
sub r0, top
vcompresspd %zmm1, (top){%k2}
sub r1, top
vcompresspd %zmm2, (top){%k3}
sub r2, top
sub $24, n
add $24*8, array
jmp .Lx3_loop
.Lx3_loop_exit:
.Lx1_loop:
cmp $4, n
jl .Lx1_loop_exit
# Load the next 4 elements
vmovdqu 32*0(array), A0
# Find elements lesser-than-equal to PIVOT
vpcmpgtq A0, PIVOT, M0
vmovmskpd M0, r0
shl $5, r0
vmovdqa .LPermTableLeft(,r0), T
VPERMD A0, T, M0
vmovdqa .LPermTableRight(,r0), T
VPERMD A0, T, A0
popcnt r0, r0
shl $3, r0
vmovdqu M0, (bottom)
add r0, bottom
sub $32, r0
vmovdqu A0, (top)
sub r0, top
sub $4, n
add $4*8, array
jmp .Lx1_loop
.Lx1_loop_exit:
.Ls_loop:
cmp $0, n
je .Ls_loop_exit
mov (array), %r8
cmp pivotALU, %r8
jg .Lgreater
mov %r8, (bottom)
add $8, bottom
add $8, array
dec n
jmp .Ls_loop
.Lgreater:
mov %r8, (top)
add $8, top
add $8, array
dec n
jmp .Ls_loop
.Ls_loop_exit:
mov pivotALU, (bottom)
add $8, bottom
sub temp_space, top
shr $3, top
mov top, %rax
.Lx4_copy_loop:
cmp $16, top
jl .Lx4_copy_end
vmovdqu 32*0(temp_space), %ymm0
vmovdqu 32*1(temp_space), %ymm1
vmovdqu 32*2(temp_space), %ymm2
vmovdqu 32*3(temp_space), %ymm3
vmovdqu %ymm0, 32*0(bottom)
vmovdqu %ymm1, 32*1(bottom)
vmovdqu %ymm2, 32*2(bottom)
vmovdqu %ymm3, 32*3(bottom)
lea 32*4(temp_space), temp_space
lea 32*4(bottom), bottom
sub $16, top
jmp .Lx4_copy_loop
.Lx4_copy_end:
.Lx1_copy_loop:
cmp $4, top
jl .Lx1_copy_end
vmovdqu 32*0(temp_space), %ymm0
vmovdqu %ymm0, 32*0(bottom)
lea 8*4(temp_space), temp_space
lea 8*4(bottom), bottom
sub $4, top
jmp .Lx1_copy_loop
.Lx1_copy_end:
cmp $0, top
je .Lcopy_end
mov (temp_space), %r8
mov %r8, (bottom)
lea 8(temp_space), temp_space
lea 8(bottom), bottom
sub $1, top
jmp .Lx1_copy_loop
.Lcopy_end:
pop %rbp
pop %r15
pop %r14
pop %r13
pop %r12
ret
.align 32
.LPermTableLeft:
.long 0,0,0,0,0,0,0,0 #0
.long 0,1,0,0,0,0,0,0 #1
.long 2,3,0,0,0,0,0,0 #2
.long 0,1,2,3,0,0,0,0 #3
.long 4,5,0,0,0,0,0,0 #4
.long 0,1,4,5,0,0,0,0 #5
.long 2,3,4,5,0,0,0,0 #6
.long 0,1,2,3,4,5,0,0 #7
.long 6,7,0,0,0,0,0,0 #8
.long 0,1,6,7,0,0,0,0 #9
.long 2,3,6,7,0,0,0,0 #10
.long 0,1,2,3,6,7,0,0 #11
.long 4,5,6,7,0,0,0,0 #12
.long 0,1,4,5,6,7,0,0 #13
.long 2,3,4,5,6,7,0,0 #14
.long 0,1,2,3,4,5,6,7 #15
.align 32
.LPermTableRight:
.long 0,1,2,3,4,5,6,7 #0
.long 2,3,4,5,6,7,0,0 #1
.long 0,1,4,5,6,7,0,0 #2
.long 4,5,6,7,0,0,0,0 #3
.long 0,1,2,3,6,7,0,0 #4
.long 2,3,6,7,0,0,0,0 #5
.long 0,1,6,7,0,0,0,0 #6
.long 6,7,0,0,0,0,0,0 #7
.long 0,1,2,3,4,5,0,0 #8
.long 2,3,4,5,0,0,0,0 #9
.long 0,1,4,5,0,0,0,0 #10
.long 4,5,0,0,0,0,0,0 #11
.long 0,1,2,3,0,0,0,0 #12
.long 2,3,0,0,0,0,0,0 #13
.long 0,1,0,0,0,0,0,0 #14
.long 0,0,0,0,0,0,0,0 #15