xref: /openbmc/qemu/target/arm/tcg/mve.decode (revision 9ea2e69f)
1# M-profile MVE instruction descriptions
2#
3#  Copyright (c) 2021 Linaro, Ltd
4#
5# This library is free software; you can redistribute it and/or
6# modify it under the terms of the GNU Lesser General Public
7# License as published by the Free Software Foundation; either
8# version 2.1 of the License, or (at your option) any later version.
9#
10# This library is distributed in the hope that it will be useful,
11# but WITHOUT ANY WARRANTY; without even the implied warranty of
12# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
13# Lesser General Public License for more details.
14#
15# You should have received a copy of the GNU Lesser General Public
16# License along with this library; if not, see <http://www.gnu.org/licenses/>.
17
18#
19# This file is processed by scripts/decodetree.py
20#
21
22%qd 22:1 13:3
23%qm 5:1 1:3
24%qn 7:1 17:3
25
26# VQDMULL has size in bit 28: 0 for 16 bit, 1 for 32 bit
27%size_28 28:1 !function=plus_1
28
29# 2 operand fp insns have size in bit 20: 1 for 16 bit, 0 for 32 bit,
30# like Neon FP insns.
31%2op_fp_size 20:1 !function=neon_3same_fp_size
32# VCADD is an exception, where bit 20 is 0 for 16 bit and 1 for 32 bit
33%2op_fp_size_rev 20:1 !function=plus_1
34# FP scalars have size in bit 28, 1 for 16 bit, 0 for 32 bit
35%2op_fp_scalar_size 28:1 !function=neon_3same_fp_size
36
37# 1imm format immediate
38%imm_28_16_0 28:1 16:3 0:4
39
40&vldr_vstr rn qd imm p a w size l u
41&1op qd qm size
42&2op qd qm qn size
43&2scalar qd qn rm size
44&1imm qd imm cmode op
45&2shift qd qm shift size
46&vidup qd rn size imm
47&viwdup qd rn rm size imm
48&vcmp qm qn size mask
49&vcmp_scalar qn rm size mask
50&shl_scalar qda rm size
51&vmaxv qm rda size
52&vabav qn qm rda size
53&vldst_sg qd qm rn size msize os
54&vldst_sg_imm qd qm a w imm
55&vldst_il qd rn size pat w
56
57# scatter-gather memory size is in bits 6:4
58%sg_msize 6:1 4:1
59
60@vldr_vstr ....... . . . . l:1 rn:4 ... ...... imm:7 &vldr_vstr qd=%qd u=0
61# Note that both Rn and Qd are 3 bits only (no D bit)
62@vldst_wn ... u:1 ... . . . . l:1 . rn:3 qd:3 . ... .. imm:7 &vldr_vstr
63
64@vldst_sg .... .... .... rn:4 .... ... size:2 ... ... os:1 &vldst_sg \
65          qd=%qd qm=%qm msize=%sg_msize
66
67# Qm is in the fields usually labeled Qn
68@vldst_sg_imm .... .... a:1 . w:1 . .... .... .... . imm:7 &vldst_sg_imm \
69              qd=%qd qm=%qn
70
71# Deinterleaving load/interleaving store
72@vldst_il .... .... .. w:1 . rn:4 .... ... size:2 pat:2 ..... &vldst_il \
73          qd=%qd
74
75@1op .... .... .... size:2 .. .... .... .... .... &1op qd=%qd qm=%qm
76@1op_nosz .... .... .... .... .... .... .... .... &1op qd=%qd qm=%qm size=0
77@2op .... .... .. size:2 .... .... .... .... .... &2op qd=%qd qm=%qm qn=%qn
78@2op_nosz .... .... .... .... .... .... .... .... &2op qd=%qd qm=%qm qn=%qn size=0
79@2op_sz28 .... .... .... .... .... .... .... .... &2op qd=%qd qm=%qm qn=%qn \
80     size=%size_28
81@1imm .... .... .... .... .... cmode:4 .. op:1 . .... &1imm qd=%qd imm=%imm_28_16_0
82
83# The _rev suffix indicates that Vn and Vm are reversed. This is
84# the case for shifts. In the Arm ARM these insns are documented
85# with the Vm and Vn fields in their usual places, but in the
86# assembly the operands are listed "backwards", ie in the order
87# Qd, Qm, Qn where other insns use Qd, Qn, Qm. For QEMU we choose
88# to consider Vm and Vn as being in different fields in the insn.
89# This gives us consistency with A64 and Neon.
90@2op_rev .... .... .. size:2 .... .... .... .... .... &2op qd=%qd qm=%qn qn=%qm
91
92@2scalar .... .... .. size:2 .... .... .... .... rm:4 &2scalar qd=%qd qn=%qn
93@2scalar_nosz .... .... .... .... .... .... .... rm:4 &2scalar qd=%qd qn=%qn
94
95@2_shl_b .... .... .. 001 shift:3 .... .... .... .... &2shift qd=%qd qm=%qm size=0
96@2_shl_h .... .... .. 01  shift:4 .... .... .... .... &2shift qd=%qd qm=%qm size=1
97@2_shl_w .... .... .. 1   shift:5 .... .... .... .... &2shift qd=%qd qm=%qm size=2
98
99@2_shll_b .... .... ... 01 shift:3 .... .... .... .... &2shift qd=%qd qm=%qm size=0
100@2_shll_h .... .... ... 1  shift:4 .... .... .... .... &2shift qd=%qd qm=%qm size=1
101# VSHLL encoding T2 where shift == esize
102@2_shll_esize_b .... .... .... 00 .. .... .... .... .... &2shift \
103                qd=%qd qm=%qm size=0 shift=8
104@2_shll_esize_h .... .... .... 01 .. .... .... .... .... &2shift \
105                qd=%qd qm=%qm size=1 shift=16
106
107# Right shifts are encoded as N - shift, where N is the element size in bits.
108%rshift_i5  16:5 !function=rsub_32
109%rshift_i4  16:4 !function=rsub_16
110%rshift_i3  16:3 !function=rsub_8
111
112@2_shr_b .... .... .. 001 ... .... .... .... .... &2shift qd=%qd qm=%qm \
113         size=0 shift=%rshift_i3
114@2_shr_h .... .... .. 01 .... .... .... .... .... &2shift qd=%qd qm=%qm \
115         size=1 shift=%rshift_i4
116@2_shr_w .... .... .. 1 ..... .... .... .... .... &2shift qd=%qd qm=%qm \
117         size=2 shift=%rshift_i5
118
119@shl_scalar .... .... .... size:2 .. .... .... .... rm:4 &shl_scalar qda=%qd
120
121# Vector comparison; 4-bit Qm but 3-bit Qn
122%mask_22_13      22:1 13:3
123@vcmp    .... .... .. size:2 qn:3 . .... .... .... .... &vcmp qm=%qm mask=%mask_22_13
124@vcmp_scalar .... .... .. size:2 qn:3 . .... .... .... rm:4 &vcmp_scalar \
125             mask=%mask_22_13
126
127@vcmp_fp .... .... .... qn:3 . .... .... .... .... &vcmp \
128         qm=%qm size=%2op_fp_scalar_size mask=%mask_22_13
129
130# Bit 28 is a 2op_fp_scalar_size bit, but we do not decode it in this
131# format to avoid complicated overlapping-instruction-groups
132@vcmp_fp_scalar .... .... .... qn:3 . .... .... .... rm:4 &vcmp_scalar \
133                mask=%mask_22_13
134
135@vmaxv .... .... .... size:2 .. rda:4 .... .... .... &vmaxv qm=%qm
136
137@2op_fp .... .... .... .... .... .... .... .... &2op \
138        qd=%qd qn=%qn qm=%qm size=%2op_fp_size
139
140@2op_fp_size_rev .... .... .... .... .... .... .... .... &2op \
141                 qd=%qd qn=%qn qm=%qm size=%2op_fp_size_rev
142
143# 2-operand, but Qd and Qn share a field. Size is in bit 28, but we
144# don't decode it in this format
145@vmaxnma  .... .... .... .... .... .... .... .... &2op \
146          qd=%qd qn=%qd qm=%qm
147
148# Here also we don't decode the bit 28 size in the format to avoid
149# awkward nested overlap groups
150@vmaxnmv          .... .... .... .... rda:4 .... .... .... &vmaxv qm=%qm
151
152@2op_fp_scalar .... .... .... .... .... .... .... rm:4 &2scalar \
153               qd=%qd qn=%qn size=%2op_fp_scalar_size
154
155# Vector loads and stores
156
157# Widening loads and narrowing stores:
158# for these P=0 W=0 is 'related encoding'; sz=11 is 'related encoding'
159# This means we need to expand out to multiple patterns for P, W, SZ.
160# For stores the U bit must be 0 but we catch that in the trans_ function.
161# The naming scheme here is "VLDSTB_H == in-memory byte load/store to/from
162# signed halfword element in register", etc.
163VLDSTB_H         111 . 110 0 a:1 0 1   . 0 ... ... 0 111 01 ....... @vldst_wn \
164                 p=0 w=1 size=1
165VLDSTB_H         111 . 110 1 a:1 0 w:1 . 0 ... ... 0 111 01 ....... @vldst_wn \
166                 p=1 size=1
167VLDSTB_W         111 . 110 0 a:1 0 1   . 0 ... ... 0 111 10 ....... @vldst_wn \
168                 p=0 w=1 size=2
169VLDSTB_W         111 . 110 1 a:1 0 w:1 . 0 ... ... 0 111 10 ....... @vldst_wn \
170                 p=1 size=2
171VLDSTH_W         111 . 110 0 a:1 0 1   . 1 ... ... 0 111 10 ....... @vldst_wn \
172                 p=0 w=1 size=2
173VLDSTH_W         111 . 110 1 a:1 0 w:1 . 1 ... ... 0 111 10 ....... @vldst_wn \
174                 p=1 size=2
175
176# Non-widening loads/stores (P=0 W=0 is 'related encoding')
177VLDR_VSTR        1110110 0 a:1 . 1   . .... ... 111100 .......   @vldr_vstr \
178                 size=0 p=0 w=1
179VLDR_VSTR        1110110 0 a:1 . 1   . .... ... 111101 .......   @vldr_vstr \
180                 size=1 p=0 w=1
181VLDR_VSTR        1110110 0 a:1 . 1   . .... ... 111110 .......   @vldr_vstr \
182                 size=2 p=0 w=1
183VLDR_VSTR        1110110 1 a:1 . w:1 . .... ... 111100 .......   @vldr_vstr \
184                 size=0 p=1
185VLDR_VSTR        1110110 1 a:1 . w:1 . .... ... 111101 .......   @vldr_vstr \
186                 size=1 p=1
187VLDR_VSTR        1110110 1 a:1 . w:1 . .... ... 111110 .......   @vldr_vstr \
188                 size=2 p=1
189
190# gather loads/scatter stores
191VLDR_S_sg        111 0 1100 1 . 01 .... ... 0 111 . .... .... @vldst_sg
192VLDR_U_sg        111 1 1100 1 . 01 .... ... 0 111 . .... .... @vldst_sg
193VSTR_sg          111 0 1100 1 . 00 .... ... 0 111 . .... .... @vldst_sg
194
195VLDRW_sg_imm     111 1 1101 ... 1 ... 0 ... 1 1110 .... .... @vldst_sg_imm
196VLDRD_sg_imm     111 1 1101 ... 1 ... 0 ... 1 1111 .... .... @vldst_sg_imm
197VSTRW_sg_imm     111 1 1101 ... 0 ... 0 ... 1 1110 .... .... @vldst_sg_imm
198VSTRD_sg_imm     111 1 1101 ... 0 ... 0 ... 1 1111 .... .... @vldst_sg_imm
199
200# deinterleaving loads/interleaving stores
201VLD2             1111 1100 1 .. 1 .... ... 1 111 .. .. 00000 @vldst_il
202VLD4             1111 1100 1 .. 1 .... ... 1 111 .. .. 00001 @vldst_il
203VST2             1111 1100 1 .. 0 .... ... 1 111 .. .. 00000 @vldst_il
204VST4             1111 1100 1 .. 0 .... ... 1 111 .. .. 00001 @vldst_il
205
206# Moves between 2 32-bit vector lanes and 2 general purpose registers
207VMOV_to_2gp      1110 1100 0 . 00 rt2:4 ... 0 1111 000 idx:1 rt:4 qd=%qd
208VMOV_from_2gp    1110 1100 0 . 01 rt2:4 ... 0 1111 000 idx:1 rt:4 qd=%qd
209
210# Vector 2-op
211VAND             1110 1111 0 . 00 ... 0 ... 0 0001 . 1 . 1 ... 0 @2op_nosz
212VBIC             1110 1111 0 . 01 ... 0 ... 0 0001 . 1 . 1 ... 0 @2op_nosz
213VORR             1110 1111 0 . 10 ... 0 ... 0 0001 . 1 . 1 ... 0 @2op_nosz
214VORN             1110 1111 0 . 11 ... 0 ... 0 0001 . 1 . 1 ... 0 @2op_nosz
215VEOR             1111 1111 0 . 00 ... 0 ... 0 0001 . 1 . 1 ... 0 @2op_nosz
216
217VADD             1110 1111 0 . .. ... 0 ... 0 1000 . 1 . 0 ... 0 @2op
218VSUB             1111 1111 0 . .. ... 0 ... 0 1000 . 1 . 0 ... 0 @2op
219VMUL             1110 1111 0 . .. ... 0 ... 0 1001 . 1 . 1 ... 0 @2op
220
221# The VSHLL T2 encoding is not a @2op pattern, but is here because it
222# overlaps what would be size=0b11 VMULH/VRMULH
223{
224  VCVTB_SH       111 0 1110 0 . 11 1111 ... 0 1110 0 0 . 0 ... 1 @1op_nosz
225
226  VMAXNMA        111 0 1110 0 . 11 1111 ... 0 1110 1 0 . 0 ... 1 @vmaxnma size=2
227
228  VSHLL_BS       111 0 1110 0 . 11 .. 01 ... 0 1110 0 0 . 0 ... 1 @2_shll_esize_b
229  VSHLL_BS       111 0 1110 0 . 11 .. 01 ... 0 1110 0 0 . 0 ... 1 @2_shll_esize_h
230
231  VQMOVUNB       111 0 1110 0 . 11 .. 01 ... 0 1110 1 0 . 0 ... 1 @1op
232  VQMOVN_BS      111 0 1110 0 . 11 .. 11 ... 0 1110 0 0 . 0 ... 1 @1op
233
234  VMAXA          111 0 1110 0 . 11 .. 11 ... 0 1110 1 0 . 0 ... 1 @1op
235
236  VMULH_S        111 0 1110 0 . .. ...1 ... 0 1110 . 0 . 0 ... 1 @2op
237}
238
239{
240  VCVTB_HS       111 1 1110 0 . 11  1111 ... 0 1110 0 0 . 0 ... 1  @1op_nosz
241
242  VMAXNMA        111 1 1110 0 . 11  1111 ... 0 1110 1 0 . 0 ... 1 @vmaxnma size=1
243
244  VSHLL_BU       111 1 1110 0 . 11 .. 01 ... 0 1110 0 0 . 0 ... 1 @2_shll_esize_b
245  VSHLL_BU       111 1 1110 0 . 11 .. 01 ... 0 1110 0 0 . 0 ... 1 @2_shll_esize_h
246
247  VMOVNB         111 1 1110 0 . 11 .. 01 ... 0 1110 1 0 . 0 ... 1 @1op
248  VQMOVN_BU      111 1 1110 0 . 11 .. 11 ... 0 1110 0 0 . 0 ... 1 @1op
249
250  VMULH_U        111 1 1110 0 . .. ...1 ... 0 1110 . 0 . 0 ... 1 @2op
251}
252
253{
254  VCVTT_SH       111 0 1110 0 . 11  1111 ... 1 1110 0 0 . 0 ... 1 @1op_nosz
255
256  VMINNMA        111 0 1110 0 . 11  1111 ... 1 1110 1 0 . 0 ... 1 @vmaxnma size=2
257  VSHLL_TS       111 0 1110 0 . 11 .. 01 ... 1 1110 0 0 . 0 ... 1 @2_shll_esize_b
258  VSHLL_TS       111 0 1110 0 . 11 .. 01 ... 1 1110 0 0 . 0 ... 1 @2_shll_esize_h
259
260  VQMOVUNT       111 0 1110 0 . 11 .. 01 ... 1 1110 1 0 . 0 ... 1 @1op
261  VQMOVN_TS      111 0 1110 0 . 11 .. 11 ... 1 1110 0 0 . 0 ... 1 @1op
262
263  VMINA          111 0 1110 0 . 11 .. 11 ... 1 1110 1 0 . 0 ... 1 @1op
264
265  VRMULH_S       111 0 1110 0 . .. ...1 ... 1 1110 . 0 . 0 ... 1 @2op
266}
267
268{
269  VCVTT_HS       111 1 1110 0 . 11  1111 ... 1 1110 0 0 . 0 ... 1 @1op_nosz
270
271  VMINNMA        111 1 1110 0 . 11  1111 ... 1 1110 1 0 . 0 ... 1 @vmaxnma size=1
272  VSHLL_TU       111 1 1110 0 . 11 .. 01 ... 1 1110 0 0 . 0 ... 1 @2_shll_esize_b
273  VSHLL_TU       111 1 1110 0 . 11 .. 01 ... 1 1110 0 0 . 0 ... 1 @2_shll_esize_h
274
275  VMOVNT         111 1 1110 0 . 11 .. 01 ... 1 1110 1 0 . 0 ... 1 @1op
276  VQMOVN_TU      111 1 1110 0 . 11 .. 11 ... 1 1110 0 0 . 0 ... 1 @1op
277
278  VRMULH_U       111 1 1110 0 . .. ...1 ... 1 1110 . 0 . 0 ... 1 @2op
279}
280
281VMAX_S           111 0 1111 0 . .. ... 0 ... 0 0110 . 1 . 0 ... 0 @2op
282VMAX_U           111 1 1111 0 . .. ... 0 ... 0 0110 . 1 . 0 ... 0 @2op
283VMIN_S           111 0 1111 0 . .. ... 0 ... 0 0110 . 1 . 1 ... 0 @2op
284VMIN_U           111 1 1111 0 . .. ... 0 ... 0 0110 . 1 . 1 ... 0 @2op
285
286VABD_S           111 0 1111 0 . .. ... 0 ... 0 0111 . 1 . 0 ... 0 @2op
287VABD_U           111 1 1111 0 . .. ... 0 ... 0 0111 . 1 . 0 ... 0 @2op
288
289VHADD_S          111 0 1111 0 . .. ... 0 ... 0 0000 . 1 . 0 ... 0 @2op
290VHADD_U          111 1 1111 0 . .. ... 0 ... 0 0000 . 1 . 0 ... 0 @2op
291VHSUB_S          111 0 1111 0 . .. ... 0 ... 0 0010 . 1 . 0 ... 0 @2op
292VHSUB_U          111 1 1111 0 . .. ... 0 ... 0 0010 . 1 . 0 ... 0 @2op
293
294{
295  VMULLP_B       111 . 1110 0 . 11 ... 1 ... 0 1110 . 0 . 0 ... 0 @2op_sz28
296  VMULL_BS       111 0 1110 0 . .. ... 1 ... 0 1110 . 0 . 0 ... 0 @2op
297  VMULL_BU       111 1 1110 0 . .. ... 1 ... 0 1110 . 0 . 0 ... 0 @2op
298}
299{
300  VMULLP_T       111 . 1110 0 . 11 ... 1 ... 1 1110 . 0 . 0 ... 0 @2op_sz28
301  VMULL_TS       111 0 1110 0 . .. ... 1 ... 1 1110 . 0 . 0 ... 0 @2op
302  VMULL_TU       111 1 1110 0 . .. ... 1 ... 1 1110 . 0 . 0 ... 0 @2op
303}
304
305VQDMULH          1110 1111 0 . .. ... 0 ... 0 1011 . 1 . 0 ... 0 @2op
306VQRDMULH         1111 1111 0 . .. ... 0 ... 0 1011 . 1 . 0 ... 0 @2op
307
308VQADD_S          111 0 1111 0 . .. ... 0 ... 0 0000 . 1 . 1 ... 0 @2op
309VQADD_U          111 1 1111 0 . .. ... 0 ... 0 0000 . 1 . 1 ... 0 @2op
310VQSUB_S          111 0 1111 0 . .. ... 0 ... 0 0010 . 1 . 1 ... 0 @2op
311VQSUB_U          111 1 1111 0 . .. ... 0 ... 0 0010 . 1 . 1 ... 0 @2op
312
313VSHL_S           111 0 1111 0 . .. ... 0 ... 0 0100 . 1 . 0 ... 0 @2op_rev
314VSHL_U           111 1 1111 0 . .. ... 0 ... 0 0100 . 1 . 0 ... 0 @2op_rev
315
316VRSHL_S          111 0 1111 0 . .. ... 0 ... 0 0101 . 1 . 0 ... 0 @2op_rev
317VRSHL_U          111 1 1111 0 . .. ... 0 ... 0 0101 . 1 . 0 ... 0 @2op_rev
318
319VQSHL_S          111 0 1111 0 . .. ... 0 ... 0 0100 . 1 . 1 ... 0 @2op_rev
320VQSHL_U          111 1 1111 0 . .. ... 0 ... 0 0100 . 1 . 1 ... 0 @2op_rev
321
322VQRSHL_S         111 0 1111 0 . .. ... 0 ... 0 0101 . 1 . 1 ... 0 @2op_rev
323VQRSHL_U         111 1 1111 0 . .. ... 0 ... 0 0101 . 1 . 1 ... 0 @2op_rev
324
325{
326  VCMUL0         111 . 1110 0 . 11 ... 0 ... 0 1110 . 0 . 0 ... 0 @2op_sz28
327  VQDMLADH       1110  1110 0 . .. ... 0 ... 0 1110 . 0 . 0 ... 0 @2op
328  VQDMLSDH       1111  1110 0 . .. ... 0 ... 0 1110 . 0 . 0 ... 0 @2op
329}
330
331{
332  VCMUL180       111 . 1110 0 . 11 ... 0 ... 1 1110 . 0 . 0 ... 0 @2op_sz28
333  VQDMLADHX      111 0 1110 0 . .. ... 0 ... 1 1110 . 0 . 0 ... 0 @2op
334  VQDMLSDHX      111 1 1110 0 . .. ... 0 ... 1 1110 . 0 . 0 ... 0 @2op
335}
336
337{
338  VCMUL90        111 . 1110 0 . 11 ... 0 ... 0 1110 . 0 . 0 ... 1 @2op_sz28
339  VQRDMLADH      111 0 1110 0 . .. ... 0 ... 0 1110 . 0 . 0 ... 1 @2op
340  VQRDMLSDH      111 1 1110 0 . .. ... 0 ... 0 1110 . 0 . 0 ... 1 @2op
341}
342
343{
344  VCMUL270       111 . 1110 0 . 11 ... 0 ... 1 1110 . 0 . 0 ... 1 @2op_sz28
345  VQRDMLADHX     111 0 1110 0 . .. ... 0 ... 1 1110 . 0 . 0 ... 1 @2op
346  VQRDMLSDHX     111 1 1110 0 . .. ... 0 ... 1 1110 . 0 . 0 ... 1 @2op
347}
348
349VQDMULLB         111 . 1110 0 . 11 ... 0 ... 0 1111 . 0 . 0 ... 1 @2op_sz28
350VQDMULLT         111 . 1110 0 . 11 ... 0 ... 1 1111 . 0 . 0 ... 1 @2op_sz28
351
352VRHADD_S         111 0 1111 0 . .. ... 0 ... 0 0001 . 1 . 0 ... 0 @2op
353VRHADD_U         111 1 1111 0 . .. ... 0 ... 0 0001 . 1 . 0 ... 0 @2op
354
355{
356  VADC           1110 1110 0 . 11 ... 0 ... 0 1111 . 0 . 0 ... 0 @2op_nosz
357  VADCI          1110 1110 0 . 11 ... 0 ... 1 1111 . 0 . 0 ... 0 @2op_nosz
358  VHCADD90       1110 1110 0 . .. ... 0 ... 0 1111 . 0 . 0 ... 0 @2op
359  VHCADD270      1110 1110 0 . .. ... 0 ... 1 1111 . 0 . 0 ... 0 @2op
360}
361
362{
363  VSBC           1111 1110 0 . 11 ... 0 ... 0 1111 . 0 . 0 ... 0 @2op_nosz
364  VSBCI          1111 1110 0 . 11 ... 0 ... 1 1111 . 0 . 0 ... 0 @2op_nosz
365  VCADD90        1111 1110 0 . .. ... 0 ... 0 1111 . 0 . 0 ... 0 @2op
366  VCADD270       1111 1110 0 . .. ... 0 ... 1 1111 . 0 . 0 ... 0 @2op
367}
368
369# Vector miscellaneous
370
371VCLS             1111 1111 1 . 11 .. 00 ... 0 0100 01 . 0 ... 0 @1op
372VCLZ             1111 1111 1 . 11 .. 00 ... 0 0100 11 . 0 ... 0 @1op
373
374VREV16           1111 1111 1 . 11 .. 00 ... 0 0001 01 . 0 ... 0 @1op
375VREV32           1111 1111 1 . 11 .. 00 ... 0 0000 11 . 0 ... 0 @1op
376VREV64           1111 1111 1 . 11 .. 00 ... 0 0000 01 . 0 ... 0 @1op
377
378VMVN             1111 1111 1 . 11 00 00 ... 0 0101 11 . 0 ... 0 @1op_nosz
379
380VABS             1111 1111 1 . 11 .. 01 ... 0 0011 01 . 0 ... 0 @1op
381VABS_fp          1111 1111 1 . 11 .. 01 ... 0 0111 01 . 0 ... 0 @1op
382VNEG             1111 1111 1 . 11 .. 01 ... 0 0011 11 . 0 ... 0 @1op
383VNEG_fp          1111 1111 1 . 11 .. 01 ... 0 0111 11 . 0 ... 0 @1op
384
385VQABS            1111 1111 1 . 11 .. 00 ... 0 0111 01 . 0 ... 0 @1op
386VQNEG            1111 1111 1 . 11 .. 00 ... 0 0111 11 . 0 ... 0 @1op
387
388&vdup qd rt size
389# Qd is in the fields usually named Qn
390@vdup            .... .... . . .. ... . rt:4 .... . . . . .... qd=%qn &vdup
391
392# B and E bits encode size, which we decode here to the usual size values
393VDUP             1110 1110 1 1 10 ... 0 .... 1011 . 0 0 1 0000 @vdup size=0
394VDUP             1110 1110 1 0 10 ... 0 .... 1011 . 0 1 1 0000 @vdup size=1
395VDUP             1110 1110 1 0 10 ... 0 .... 1011 . 0 0 1 0000 @vdup size=2
396
397# Incrementing and decrementing dup
398
399# VIDUP, VDDUP format immediate: 1 << (immh:imml)
400%imm_vidup 7:1 0:1 !function=vidup_imm
401
402# VIDUP, VDDUP registers: Rm bits [3:1] from insn, bit 0 is 1;
403# Rn bits [3:1] from insn, bit 0 is 0
404%vidup_rm 1:3 !function=times_2_plus_1
405%vidup_rn 17:3 !function=times_2
406
407@vidup           .... .... . . size:2 .... .... .... .... .... \
408                 qd=%qd imm=%imm_vidup rn=%vidup_rn &vidup
409@viwdup          .... .... . . size:2 .... .... .... .... .... \
410                 qd=%qd imm=%imm_vidup rm=%vidup_rm rn=%vidup_rn &viwdup
411{
412  VIDUP          1110 1110 0 . .. ... 1 ... 0 1111 . 110 111 . @vidup
413  VIWDUP         1110 1110 0 . .. ... 1 ... 0 1111 . 110 ... . @viwdup
414}
415{
416  VCMPGT_fp_scalar 1110 1110 0 . 11 ... 1 ... 1 1111  0110 .... @vcmp_fp_scalar size=2
417  VCMPLE_fp_scalar 1110 1110 0 . 11 ... 1 ... 1 1111  1110 .... @vcmp_fp_scalar size=2
418  VDDUP            1110 1110 0 . .. ... 1 ... 1 1111 . 110 111 . @vidup
419  VDWDUP           1110 1110 0 . .. ... 1 ... 1 1111 . 110 ... . @viwdup
420}
421
422# multiply-add long dual accumulate
423# rdahi: bits [3:1] from insn, bit 0 is 1
424# rdalo: bits [3:1] from insn, bit 0 is 0
425%rdahi 20:3 !function=times_2_plus_1
426%rdalo 13:3 !function=times_2
427# size bit is 0 for 16 bit, 1 for 32 bit
428%size_16 16:1 !function=plus_1
429
430&vmlaldav rdahi rdalo size qn qm x a
431&vmladav rda size qn qm x a
432
433@vmlaldav        .... .... . ... ... . ... x:1 .... .. a:1 . qm:3 . \
434                 qn=%qn rdahi=%rdahi rdalo=%rdalo size=%size_16 &vmlaldav
435@vmlaldav_nosz   .... .... . ... ... . ... x:1 .... .. a:1 . qm:3 . \
436                 qn=%qn rdahi=%rdahi rdalo=%rdalo size=0 &vmlaldav
437@vmladav         .... .... .... ... . ... x:1 .... . . a:1 . qm:3 . \
438                 qn=%qn rda=%rdalo size=%size_16 &vmladav
439@vmladav_nosz    .... .... .... ... . ... x:1 .... . . a:1 . qm:3 . \
440                 qn=%qn rda=%rdalo size=0 &vmladav
441
442{
443  VMLADAV_S      1110 1110 1111  ... . ... . 1110 . 0 . 0 ... 0 @vmladav
444  VMLALDAV_S     1110 1110 1 ... ... . ... . 1110 . 0 . 0 ... 0 @vmlaldav
445}
446{
447  VMLADAV_U      1111 1110 1111  ... . ... . 1110 . 0 . 0 ... 0 @vmladav
448  VMLALDAV_U     1111 1110 1 ... ... . ... . 1110 . 0 . 0 ... 0 @vmlaldav
449}
450
451{
452  VMLSDAV        1110 1110 1111  ... . ... . 1110 . 0 . 0 ... 1 @vmladav
453  VMLSLDAV       1110 1110 1 ... ... . ... . 1110 . 0 . 0 ... 1 @vmlaldav
454}
455
456{
457  VMLSDAV        1111 1110 1111  ... 0 ... . 1110 . 0 . 0 ... 1 @vmladav_nosz
458  VRMLSLDAVH     1111 1110 1 ... ... 0 ... . 1110 . 0 . 0 ... 1 @vmlaldav_nosz
459}
460
461VMLADAV_S        1110 1110 1111  ... 0 ... . 1111 . 0 . 0 ... 1 @vmladav_nosz
462VMLADAV_U        1111 1110 1111  ... 0 ... . 1111 . 0 . 0 ... 1 @vmladav_nosz
463
464{
465  [
466    VMAXNMAV     1110 1110 1110  11 00 ....  1111 0 0 . 0 ... 0 @vmaxnmv size=2
467    VMINNMAV     1110 1110 1110  11 00 ....  1111 1 0 . 0 ... 0 @vmaxnmv size=2
468    VMAXNMV      1110 1110 1110  11 10 ....  1111 0 0 . 0 ... 0 @vmaxnmv size=2
469    VMINNMV      1110 1110 1110  11 10 ....  1111 1 0 . 0 ... 0 @vmaxnmv size=2
470  ]
471  [
472    VMAXV_S      1110 1110 1110  .. 10 ....  1111 0 0 . 0 ... 0 @vmaxv
473    VMINV_S      1110 1110 1110  .. 10 ....  1111 1 0 . 0 ... 0 @vmaxv
474    VMAXAV       1110 1110 1110  .. 00 ....  1111 0 0 . 0 ... 0 @vmaxv
475    VMINAV       1110 1110 1110  .. 00 ....  1111 1 0 . 0 ... 0 @vmaxv
476  ]
477  VMLADAV_S      1110 1110 1111  ... 0 ... . 1111 . 0 . 0 ... 0 @vmladav_nosz
478  VRMLALDAVH_S   1110 1110 1 ... ... 0 ... . 1111 . 0 . 0 ... 0 @vmlaldav_nosz
479}
480
481{
482  [
483    VMAXNMAV     1111 1110 1110  11 00 ....  1111 0 0 . 0 ... 0 @vmaxnmv size=1
484    VMINNMAV     1111 1110 1110  11 00 ....  1111 1 0 . 0 ... 0 @vmaxnmv size=1
485    VMAXNMV      1111 1110 1110  11 10 ....  1111 0 0 . 0 ... 0 @vmaxnmv size=1
486    VMINNMV      1111 1110 1110  11 10 ....  1111 1 0 . 0 ... 0 @vmaxnmv size=1
487  ]
488  [
489    VMAXV_U      1111 1110 1110  .. 10 ....  1111 0 0 . 0 ... 0 @vmaxv
490    VMINV_U      1111 1110 1110  .. 10 ....  1111 1 0 . 0 ... 0 @vmaxv
491  ]
492  VMLADAV_U      1111 1110 1111  ... 0 ... . 1111 . 0 . 0 ... 0 @vmladav_nosz
493  VRMLALDAVH_U   1111 1110 1 ... ... 0 ... . 1111 . 0 . 0 ... 0 @vmlaldav_nosz
494}
495
496# Scalar operations
497
498{
499  VCMPEQ_fp_scalar 1110 1110 0 . 11 ... 1 ... 0 1111  0100 .... @vcmp_fp_scalar size=2
500  VCMPNE_fp_scalar 1110 1110 0 . 11 ... 1 ... 0 1111  1100 .... @vcmp_fp_scalar size=2
501  VADD_scalar      1110 1110 0 . .. ... 1 ... 0 1111 . 100 .... @2scalar
502}
503
504{
505  VCMPLT_fp_scalar 1110 1110 0 . 11 ... 1 ... 1 1111  1100 .... @vcmp_fp_scalar size=2
506  VCMPGE_fp_scalar 1110 1110 0 . 11 ... 1 ... 1 1111  0100 .... @vcmp_fp_scalar size=2
507  VSUB_scalar      1110 1110 0 . .. ... 1 ... 1 1111 . 100 .... @2scalar
508}
509
510{
511  VSHL_S_scalar   1110 1110 0 . 11 .. 01 ... 1 1110 0110 .... @shl_scalar
512  VRSHL_S_scalar  1110 1110 0 . 11 .. 11 ... 1 1110 0110 .... @shl_scalar
513  VQSHL_S_scalar  1110 1110 0 . 11 .. 01 ... 1 1110 1110 .... @shl_scalar
514  VQRSHL_S_scalar 1110 1110 0 . 11 .. 11 ... 1 1110 1110 .... @shl_scalar
515  VMUL_scalar     1110 1110 0 . .. ... 1 ... 1 1110 . 110 .... @2scalar
516}
517
518{
519  VSHL_U_scalar   1111 1110 0 . 11 .. 01 ... 1 1110 0110 .... @shl_scalar
520  VRSHL_U_scalar  1111 1110 0 . 11 .. 11 ... 1 1110 0110 .... @shl_scalar
521  VQSHL_U_scalar  1111 1110 0 . 11 .. 01 ... 1 1110 1110 .... @shl_scalar
522  VQRSHL_U_scalar 1111 1110 0 . 11 .. 11 ... 1 1110 1110 .... @shl_scalar
523  VBRSR           1111 1110 0 . .. ... 1 ... 1 1110 . 110 .... @2scalar
524}
525
526{
527  VADD_fp_scalar  111 . 1110 0 . 11 ... 0 ... 0 1111 . 100 .... @2op_fp_scalar
528  VHADD_S_scalar  1110  1110 0 . .. ... 0 ... 0 1111 . 100 .... @2scalar
529  VHADD_U_scalar  1111  1110 0 . .. ... 0 ... 0 1111 . 100 .... @2scalar
530}
531
532{
533  VSUB_fp_scalar  111 . 1110 0 . 11 ... 0 ... 1 1111 . 100 .... @2op_fp_scalar
534  VHSUB_S_scalar  1110  1110 0 . .. ... 0 ... 1 1111 . 100 .... @2scalar
535  VHSUB_U_scalar  1111  1110 0 . .. ... 0 ... 1 1111 . 100 .... @2scalar
536}
537
538{
539  VQADD_S_scalar  1110  1110 0 . .. ... 0 ... 0 1111 . 110 .... @2scalar
540  VQADD_U_scalar  1111  1110 0 . .. ... 0 ... 0 1111 . 110 .... @2scalar
541  VQDMULLB_scalar 111 . 1110 0 . 11 ... 0 ... 0 1111 . 110 .... @2scalar_nosz \
542                  size=%size_28
543}
544
545{
546  VQSUB_S_scalar  1110  1110 0 . .. ... 0 ... 1 1111 . 110 .... @2scalar
547  VQSUB_U_scalar  1111  1110 0 . .. ... 0 ... 1 1111 . 110 .... @2scalar
548  VQDMULLT_scalar 111 . 1110 0 . 11 ... 0 ... 1 1111 . 110 .... @2scalar_nosz \
549                  size=%size_28
550}
551
552{
553  VMUL_fp_scalar  111 . 1110 0 . 11 ... 1 ... 0 1110 . 110 .... @2op_fp_scalar
554  VQDMULH_scalar  1110  1110 0 . .. ... 1 ... 0 1110 . 110 .... @2scalar
555  VQRDMULH_scalar 1111  1110 0 . .. ... 1 ... 0 1110 . 110 .... @2scalar
556}
557
558{
559  VFMA_scalar     111 . 1110 0 . 11 ... 1 ... 0 1110 . 100 .... @2op_fp_scalar
560  # The U bit (28) is don't-care because it does not affect the result
561  VMLA            111 - 1110 0 . .. ... 1 ... 0 1110 . 100 .... @2scalar
562}
563
564{
565  VFMAS_scalar    111 . 1110 0 . 11 ... 1 ... 1 1110 . 100 .... @2op_fp_scalar
566  # The U bit (28) is don't-care because it does not affect the result
567  VMLAS           111 - 1110 0 . .. ... 1 ... 1 1110 . 100 .... @2scalar
568}
569
570VQRDMLAH         1110 1110 0 . .. ... 0 ... 0 1110 . 100 .... @2scalar
571VQRDMLASH        1110 1110 0 . .. ... 0 ... 1 1110 . 100 .... @2scalar
572VQDMLAH          1110 1110 0 . .. ... 0 ... 0 1110 . 110 .... @2scalar
573VQDMLASH         1110 1110 0 . .. ... 0 ... 1 1110 . 110 .... @2scalar
574
575# Vector add across vector
576{
577  VADDV          111 u:1 1110 1111 size:2 01 ... 0 1111 0 0 a:1 0 qm:3 0 rda=%rdalo
578  VADDLV         111 u:1 1110 1 ... 1001 ... 0 1111 00 a:1 0 qm:3 0 \
579                 rdahi=%rdahi rdalo=%rdalo
580}
581
582@vabav           .... .... .. size:2 .... rda:4 .... .... .... &vabav qn=%qn qm=%qm
583
584VABAV_S          111 0 1110 10 .. ... 0 .... 1111 . 0 . 0 ... 1 @vabav
585VABAV_U          111 1 1110 10 .. ... 0 .... 1111 . 0 . 0 ... 1 @vabav
586
587# Logical immediate operations (1 reg and modified-immediate)
588
589# The cmode/op bits here decode VORR/VBIC/VMOV/VMVN, but
590# not in a way we can conveniently represent in decodetree without
591# a lot of repetition:
592# VORR: op=0, (cmode & 1) && cmode < 12
593# VBIC: op=1, (cmode & 1) && cmode < 12
594# VMOV: everything else
595# So we have a single decode line and check the cmode/op in the
596# trans function.
597Vimm_1r 111 . 1111 1 . 00 0 ... ... 0 .... 0 1 . 1 .... @1imm
598
599# Shifts by immediate
600
601VSHLI             111 0 1111 1 . ... ... ... 0 0101 0 1 . 1 ... 0 @2_shl_b
602VSHLI             111 0 1111 1 . ... ... ... 0 0101 0 1 . 1 ... 0 @2_shl_h
603VSHLI             111 0 1111 1 . ... ... ... 0 0101 0 1 . 1 ... 0 @2_shl_w
604
605VQSHLI_S          111 0 1111 1 . ... ... ... 0 0111 0 1 . 1 ... 0 @2_shl_b
606VQSHLI_S          111 0 1111 1 . ... ... ... 0 0111 0 1 . 1 ... 0 @2_shl_h
607VQSHLI_S          111 0 1111 1 . ... ... ... 0 0111 0 1 . 1 ... 0 @2_shl_w
608
609VQSHLI_U          111 1 1111 1 . ... ... ... 0 0111 0 1 . 1 ... 0 @2_shl_b
610VQSHLI_U          111 1 1111 1 . ... ... ... 0 0111 0 1 . 1 ... 0 @2_shl_h
611VQSHLI_U          111 1 1111 1 . ... ... ... 0 0111 0 1 . 1 ... 0 @2_shl_w
612
613VQSHLUI           111 1 1111 1 . ... ... ... 0 0110 0 1 . 1 ... 0 @2_shl_b
614VQSHLUI           111 1 1111 1 . ... ... ... 0 0110 0 1 . 1 ... 0 @2_shl_h
615VQSHLUI           111 1 1111 1 . ... ... ... 0 0110 0 1 . 1 ... 0 @2_shl_w
616
617VSHRI_S           111 0 1111 1 . ... ... ... 0 0000 0 1 . 1 ... 0 @2_shr_b
618VSHRI_S           111 0 1111 1 . ... ... ... 0 0000 0 1 . 1 ... 0 @2_shr_h
619VSHRI_S           111 0 1111 1 . ... ... ... 0 0000 0 1 . 1 ... 0 @2_shr_w
620
621VSHRI_U           111 1 1111 1 . ... ... ... 0 0000 0 1 . 1 ... 0 @2_shr_b
622VSHRI_U           111 1 1111 1 . ... ... ... 0 0000 0 1 . 1 ... 0 @2_shr_h
623VSHRI_U           111 1 1111 1 . ... ... ... 0 0000 0 1 . 1 ... 0 @2_shr_w
624
625VRSHRI_S          111 0 1111 1 . ... ... ... 0 0010 0 1 . 1 ... 0 @2_shr_b
626VRSHRI_S          111 0 1111 1 . ... ... ... 0 0010 0 1 . 1 ... 0 @2_shr_h
627VRSHRI_S          111 0 1111 1 . ... ... ... 0 0010 0 1 . 1 ... 0 @2_shr_w
628
629VRSHRI_U          111 1 1111 1 . ... ... ... 0 0010 0 1 . 1 ... 0 @2_shr_b
630VRSHRI_U          111 1 1111 1 . ... ... ... 0 0010 0 1 . 1 ... 0 @2_shr_h
631VRSHRI_U          111 1 1111 1 . ... ... ... 0 0010 0 1 . 1 ... 0 @2_shr_w
632
633# VSHLL T1 encoding; the T2 VSHLL encoding is elsewhere in this file
634# Note that VMOVL is encoded as "VSHLL with a zero shift count"; we
635# implement it that way rather than special-casing it in the decode.
636VSHLL_BS          111 0 1110 1 . 1 .. ... ... 0 1111 0 1 . 0 ... 0 @2_shll_b
637VSHLL_BS          111 0 1110 1 . 1 .. ... ... 0 1111 0 1 . 0 ... 0 @2_shll_h
638
639VSHLL_BU          111 1 1110 1 . 1 .. ... ... 0 1111 0 1 . 0 ... 0 @2_shll_b
640VSHLL_BU          111 1 1110 1 . 1 .. ... ... 0 1111 0 1 . 0 ... 0 @2_shll_h
641
642VSHLL_TS          111 0 1110 1 . 1 .. ... ... 1 1111 0 1 . 0 ... 0 @2_shll_b
643VSHLL_TS          111 0 1110 1 . 1 .. ... ... 1 1111 0 1 . 0 ... 0 @2_shll_h
644
645VSHLL_TU          111 1 1110 1 . 1 .. ... ... 1 1111 0 1 . 0 ... 0 @2_shll_b
646VSHLL_TU          111 1 1110 1 . 1 .. ... ... 1 1111 0 1 . 0 ... 0 @2_shll_h
647
648# Shift-and-insert
649VSRI              111 1 1111 1 . ... ... ... 0 0100 0 1 . 1 ... 0 @2_shr_b
650VSRI              111 1 1111 1 . ... ... ... 0 0100 0 1 . 1 ... 0 @2_shr_h
651VSRI              111 1 1111 1 . ... ... ... 0 0100 0 1 . 1 ... 0 @2_shr_w
652
653VSLI              111 1 1111 1 . ... ... ... 0 0101 0 1 . 1 ... 0 @2_shl_b
654VSLI              111 1 1111 1 . ... ... ... 0 0101 0 1 . 1 ... 0 @2_shl_h
655VSLI              111 1 1111 1 . ... ... ... 0 0101 0 1 . 1 ... 0 @2_shl_w
656
657# Narrowing shifts (which only support b and h sizes)
658VSHRNB            111 0 1110 1 . ... ... ... 0 1111 1 1 . 0 ... 1 @2_shr_b
659VSHRNB            111 0 1110 1 . ... ... ... 0 1111 1 1 . 0 ... 1 @2_shr_h
660VSHRNT            111 0 1110 1 . ... ... ... 1 1111 1 1 . 0 ... 1 @2_shr_b
661VSHRNT            111 0 1110 1 . ... ... ... 1 1111 1 1 . 0 ... 1 @2_shr_h
662
663VRSHRNB           111 1 1110 1 . ... ... ... 0 1111 1 1 . 0 ... 1 @2_shr_b
664VRSHRNB           111 1 1110 1 . ... ... ... 0 1111 1 1 . 0 ... 1 @2_shr_h
665VRSHRNT           111 1 1110 1 . ... ... ... 1 1111 1 1 . 0 ... 1 @2_shr_b
666VRSHRNT           111 1 1110 1 . ... ... ... 1 1111 1 1 . 0 ... 1 @2_shr_h
667
668VQSHRNB_S         111 0 1110 1 . ... ... ... 0 1111 0 1 . 0 ... 0 @2_shr_b
669VQSHRNB_S         111 0 1110 1 . ... ... ... 0 1111 0 1 . 0 ... 0 @2_shr_h
670VQSHRNT_S         111 0 1110 1 . ... ... ... 1 1111 0 1 . 0 ... 0 @2_shr_b
671VQSHRNT_S         111 0 1110 1 . ... ... ... 1 1111 0 1 . 0 ... 0 @2_shr_h
672VQSHRNB_U         111 1 1110 1 . ... ... ... 0 1111 0 1 . 0 ... 0 @2_shr_b
673VQSHRNB_U         111 1 1110 1 . ... ... ... 0 1111 0 1 . 0 ... 0 @2_shr_h
674VQSHRNT_U         111 1 1110 1 . ... ... ... 1 1111 0 1 . 0 ... 0 @2_shr_b
675VQSHRNT_U         111 1 1110 1 . ... ... ... 1 1111 0 1 . 0 ... 0 @2_shr_h
676
677VQSHRUNB          111 0 1110 1 . ... ... ... 0 1111 1 1 . 0 ... 0 @2_shr_b
678VQSHRUNB          111 0 1110 1 . ... ... ... 0 1111 1 1 . 0 ... 0 @2_shr_h
679VQSHRUNT          111 0 1110 1 . ... ... ... 1 1111 1 1 . 0 ... 0 @2_shr_b
680VQSHRUNT          111 0 1110 1 . ... ... ... 1 1111 1 1 . 0 ... 0 @2_shr_h
681
682VQRSHRNB_S        111 0 1110 1 . ... ... ... 0 1111 0 1 . 0 ... 1 @2_shr_b
683VQRSHRNB_S        111 0 1110 1 . ... ... ... 0 1111 0 1 . 0 ... 1 @2_shr_h
684VQRSHRNT_S        111 0 1110 1 . ... ... ... 1 1111 0 1 . 0 ... 1 @2_shr_b
685VQRSHRNT_S        111 0 1110 1 . ... ... ... 1 1111 0 1 . 0 ... 1 @2_shr_h
686VQRSHRNB_U        111 1 1110 1 . ... ... ... 0 1111 0 1 . 0 ... 1 @2_shr_b
687VQRSHRNB_U        111 1 1110 1 . ... ... ... 0 1111 0 1 . 0 ... 1 @2_shr_h
688VQRSHRNT_U        111 1 1110 1 . ... ... ... 1 1111 0 1 . 0 ... 1 @2_shr_b
689VQRSHRNT_U        111 1 1110 1 . ... ... ... 1 1111 0 1 . 0 ... 1 @2_shr_h
690
691VQRSHRUNB         111 1 1110 1 . ... ... ... 0 1111 1 1 . 0 ... 0 @2_shr_b
692VQRSHRUNB         111 1 1110 1 . ... ... ... 0 1111 1 1 . 0 ... 0 @2_shr_h
693VQRSHRUNT         111 1 1110 1 . ... ... ... 1 1111 1 1 . 0 ... 0 @2_shr_b
694VQRSHRUNT         111 1 1110 1 . ... ... ... 1 1111 1 1 . 0 ... 0 @2_shr_h
695
696VSHLC             111 0 1110 1 . 1 imm:5 ... 0 1111 1100 rdm:4 qd=%qd
697
698# Comparisons. We expand out the conditions which are split across
699# encodings T1, T2, T3 and the fc bits. These include VPT, which is
700# effectively "VCMP then VPST". A plain "VCMP" has a mask field of zero.
701{
702  VCMPEQ_fp       111 . 1110 0 . 11 ... 1 ... 0 1111 0 0 . 0 ... 0 @vcmp_fp
703  VCMPEQ          111 1 1110 0 . .. ... 1 ... 0 1111 0 0 . 0 ... 0 @vcmp
704}
705
706{
707  VCMPNE_fp       111 . 1110 0 . 11 ... 1 ... 0 1111 1 0 . 0 ... 0 @vcmp_fp
708  VCMPNE          111 1 1110 0 . .. ... 1 ... 0 1111 1 0 . 0 ... 0 @vcmp
709}
710
711{
712  VCMPGE_fp       111 . 1110 0 . 11 ... 1 ... 1 1111 0 0 . 0 ... 0 @vcmp_fp
713  VCMPGE          111 1 1110 0 . .. ... 1 ... 1 1111 0 0 . 0 ... 0 @vcmp
714}
715
716{
717  VCMPLT_fp       111 . 1110 0 . 11 ... 1 ... 1 1111 1 0 . 0 ... 0 @vcmp_fp
718  VCMPLT          111 1 1110 0 . .. ... 1 ... 1 1111 1 0 . 0 ... 0 @vcmp
719}
720
721{
722  VCMPGT_fp       111 . 1110 0 . 11 ... 1 ... 1 1111 0 0 . 0 ... 1 @vcmp_fp
723  VCMPGT          111 1 1110 0 . .. ... 1 ... 1 1111 0 0 . 0 ... 1 @vcmp
724}
725
726{
727  VCMPLE_fp         111 . 1110 0 . 11 ... 1 ... 1 1111 1 0 . 0 ... 1 @vcmp_fp
728  VCMPLE            1111 1110 0 . .. ... 1 ... 1 1111 1 0 . 0 ... 1 @vcmp
729}
730
731{
732  VPSEL           1111 1110 0 . 11 ... 1 ... 0 1111 . 0 . 0 ... 1 @2op_nosz
733  VCMPCS          1111 1110 0 . .. ... 1 ... 0 1111 0 0 . 0 ... 1 @vcmp
734  VCMPHI          1111 1110 0 . .. ... 1 ... 0 1111 1 0 . 0 ... 1 @vcmp
735}
736
737{
738  VPNOT            1111 1110 0 0 11 000 1 000 0 1111 0100 1101
739  VPST             1111 1110 0 . 11 000 1 ... 0 1111 0100 1101 mask=%mask_22_13
740  VCMPEQ_fp_scalar 1111 1110 0 . 11 ... 1 ... 0 1111 0100 .... @vcmp_fp_scalar size=1
741  VCMPEQ_scalar    1111 1110 0 . .. ... 1 ... 0 1111 0100 .... @vcmp_scalar
742}
743
744{
745  VCMPNE_fp_scalar 1111 1110 0 . 11 ... 1 ... 0 1111 1100 .... @vcmp_fp_scalar size=1
746  VCMPNE_scalar    1111 1110 0 . .. ... 1 ... 0 1111 1100 .... @vcmp_scalar
747}
748
749{
750  VCMPGT_fp_scalar 1111 1110 0 . 11 ... 1 ... 1 1111 0110 .... @vcmp_fp_scalar size=1
751  VCMPGT_scalar    1111 1110 0 . .. ... 1 ... 1 1111 0110 .... @vcmp_scalar
752}
753
754{
755  VCMPLE_fp_scalar 1111 1110 0 . 11 ... 1 ... 1 1111 1110 .... @vcmp_fp_scalar size=1
756  VCMPLE_scalar    1111 1110 0 . .. ... 1 ... 1 1111 1110 .... @vcmp_scalar
757}
758
759{
760  VCMPGE_fp_scalar 1111 1110 0 . 11 ... 1 ... 1 1111 0100 .... @vcmp_fp_scalar size=1
761  VCMPGE_scalar    1111 1110 0 . .. ... 1 ... 1 1111 0100 .... @vcmp_scalar
762}
763{
764  VCMPLT_fp_scalar 1111 1110 0 . 11 ... 1 ... 1 1111 1100 .... @vcmp_fp_scalar size=1
765  VCMPLT_scalar    1111 1110 0 . .. ... 1 ... 1 1111 1100 .... @vcmp_scalar
766}
767
768VCMPCS_scalar     1111 1110 0 . .. ... 1 ... 0 1111 0 1 1 0 .... @vcmp_scalar
769VCMPHI_scalar     1111 1110 0 . .. ... 1 ... 0 1111 1 1 1 0 .... @vcmp_scalar
770
771# 2-operand FP
772VADD_fp           1110 1111 0 . 0 . ... 0 ... 0 1101 . 1 . 0 ... 0 @2op_fp
773VSUB_fp           1110 1111 0 . 1 . ... 0 ... 0 1101 . 1 . 0 ... 0 @2op_fp
774VMUL_fp           1111 1111 0 . 0 . ... 0 ... 0 1101 . 1 . 1 ... 0 @2op_fp
775VABD_fp           1111 1111 0 . 1 . ... 0 ... 0 1101 . 1 . 0 ... 0 @2op_fp
776
777VMAXNM            1111 1111 0 . 0 . ... 0 ... 0 1111 . 1 . 1 ... 0 @2op_fp
778VMINNM            1111 1111 0 . 1 . ... 0 ... 0 1111 . 1 . 1 ... 0 @2op_fp
779
780VCADD90_fp        1111 1100 1 . 0 . ... 0 ... 0 1000 . 1 . 0 ... 0 @2op_fp_size_rev
781VCADD270_fp       1111 1101 1 . 0 . ... 0 ... 0 1000 . 1 . 0 ... 0 @2op_fp_size_rev
782
783VFMA              1110 1111 0 . 0 . ... 0 ... 0 1100 . 1 . 1 ... 0 @2op_fp
784VFMS              1110 1111 0 . 1 . ... 0 ... 0 1100 . 1 . 1 ... 0 @2op_fp
785
786VCMLA0            1111 110 00 . 1 . ... 0 ... 0 1000 . 1 . 0 ... 0 @2op_fp_size_rev
787VCMLA90           1111 110 01 . 1 . ... 0 ... 0 1000 . 1 . 0 ... 0 @2op_fp_size_rev
788VCMLA180          1111 110 10 . 1 . ... 0 ... 0 1000 . 1 . 0 ... 0 @2op_fp_size_rev
789VCMLA270          1111 110 11 . 1 . ... 0 ... 0 1000 . 1 . 0 ... 0 @2op_fp_size_rev
790
791# floating-point <-> fixed-point conversions. Naming convention:
792# VCVT_<from><to>, S = signed int, U = unsigned int, H = halfprec, F = singleprec
793@vcvt             .... .... .. 1 ..... .... .. 1 . .... .... &2shift \
794                  qd=%qd qm=%qm shift=%rshift_i5 size=2
795@vcvt_f16         .... .... .. 11 .... .... .. 0 . .... .... &2shift \
796                  qd=%qd qm=%qm shift=%rshift_i4 size=1
797
798VCVT_SH_fixed     1110 1111 1 . ...... ... 0 11 . 0 01 . 1 ... 0 @vcvt_f16
799VCVT_UH_fixed     1111 1111 1 . ...... ... 0 11 . 0 01 . 1 ... 0 @vcvt_f16
800
801VCVT_HS_fixed     1110 1111 1 . ...... ... 0 11 . 1 01 . 1 ... 0 @vcvt_f16
802VCVT_HU_fixed     1111 1111 1 . ...... ... 0 11 . 1 01 . 1 ... 0 @vcvt_f16
803
804VCVT_SF_fixed     1110 1111 1 . ...... ... 0 11 . 0 01 . 1 ... 0 @vcvt
805VCVT_UF_fixed     1111 1111 1 . ...... ... 0 11 . 0 01 . 1 ... 0 @vcvt
806
807VCVT_FS_fixed     1110 1111 1 . ...... ... 0 11 . 1 01 . 1 ... 0 @vcvt
808VCVT_FU_fixed     1111 1111 1 . ...... ... 0 11 . 1 01 . 1 ... 0 @vcvt
809
810# VCVT between floating point and integer (halfprec and single);
811# VCVT_<from><to>, S = signed int, U = unsigned int, F = float
812VCVT_SF           1111 1111 1 . 11 .. 11 ... 0 011 00 1 . 0 ... 0 @1op
813VCVT_UF           1111 1111 1 . 11 .. 11 ... 0 011 01 1 . 0 ... 0 @1op
814VCVT_FS           1111 1111 1 . 11 .. 11 ... 0 011 10 1 . 0 ... 0 @1op
815VCVT_FU           1111 1111 1 . 11 .. 11 ... 0 011 11 1 . 0 ... 0 @1op
816
817# VCVT from floating point to integer with specified rounding mode
818VCVTAS            1111 1111 1 . 11 .. 11 ... 000 00 0 1 . 0 ... 0 @1op
819VCVTAU            1111 1111 1 . 11 .. 11 ... 000 00 1 1 . 0 ... 0 @1op
820VCVTNS            1111 1111 1 . 11 .. 11 ... 000 01 0 1 . 0 ... 0 @1op
821VCVTNU            1111 1111 1 . 11 .. 11 ... 000 01 1 1 . 0 ... 0 @1op
822VCVTPS            1111 1111 1 . 11 .. 11 ... 000 10 0 1 . 0 ... 0 @1op
823VCVTPU            1111 1111 1 . 11 .. 11 ... 000 10 1 1 . 0 ... 0 @1op
824VCVTMS            1111 1111 1 . 11 .. 11 ... 000 11 0 1 . 0 ... 0 @1op
825VCVTMU            1111 1111 1 . 11 .. 11 ... 000 11 1 1 . 0 ... 0 @1op
826
827VRINTN            1111 1111 1 . 11 .. 10 ... 001 000 1 . 0 ... 0 @1op
828VRINTX            1111 1111 1 . 11 .. 10 ... 001 001 1 . 0 ... 0 @1op
829VRINTA            1111 1111 1 . 11 .. 10 ... 001 010 1 . 0 ... 0 @1op
830VRINTZ            1111 1111 1 . 11 .. 10 ... 001 011 1 . 0 ... 0 @1op
831VRINTM            1111 1111 1 . 11 .. 10 ... 001 101 1 . 0 ... 0 @1op
832VRINTP            1111 1111 1 . 11 .. 10 ... 001 111 1 . 0 ... 0 @1op
833