vuctlsl.v
39.4 KB
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
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
// Module instances modified by /home/rws/workarea/rf/sw/bbplayer/tools/necprimfix
//
// 1 instance of an02d1h changed to j_an02.
// 1 instance of ao01d2 changed to j_ao01.
// 7 instances of mx21d1h changed to j_mx21.
// 2 instances of ni01d7 changed to j_ni01.
//
/*
*************************************************************************
* *
* Copyright (C) 1994, Silicon Graphics, Inc. *
* *
* These coded instructions, statements, and computer programs contain *
* unpublished proprietary information of Silicon Graphics, Inc., and *
* are protected by Federal copyright law. They may not be disclosed *
* to third parties or copied or duplicated in any form, in whole or *
* in part, without the prior written consent of Silicon Graphics, Inc. *
* *
*************************************************************************
*/
// $Id: vuctlsl.v,v 1.3 2003/01/24 23:07:37 berndt Exp $
/*
*************************************************************************
* *
* Project Reality *
* *
* Module: vuctlsl *
* Description: vector unit control slice for individual *
* vector unit datapath. This module is *
* instantiated eight times in vuctl, once *
* for each vector unit datapath. *
* *
* Designer: Brian Ferguson *
* Date: 8/13/94 *
* *
*************************************************************************
*/
// vuctlsl.v: RSP vector unit control
`timescale 1ns / 10ps
module vuctlsl ( clk, reset_l,
su_instvld_rd,
vct_instvld_mu,
vct_pralucmpvt_mu,
vct_addcop_mu, vct_subcop_mu,
vct_addop_mu, vct_subop_mu,
vct_vs_sgnmu_mu, vct_vt_sgnmu_mu,
vct_stltop_mu, vct_steqop_mu, vct_stneop_mu,
vct_stgeop_mu, vct_stchop_mu, vct_stclop_mu,
vct_stchrop_mu, vct_stcrop_mu,
vct_substclop_mu,
vct_absop_mu,
vct_rndop_mu,
vct_rndpop_mu, vct_rndnop_mu,
vct_mulqop_mu,
vct_mulfop_mu,
vct_muluop_mu,
vct_vseqone_mu,
vct_aluprecin_rd,
vct_cryoutld_mu, vct_cmpcdld_mu, vct_cmpcdadld_mu,
vdp_vs_zero_mu, vdp_vt_zero_mu,
vdp_vs_sign_mu,
vdp_vt_sign_mu,
vdp_aluovr_mu, vdp_aluco_mu,
vdp_aluzero_mu, vdp_aluone_mu,
vct_prsmlwrsl_mu,
vct_instvld_ac,
vct_prcslwbsl_ac,
vct_prcsupbsl_ac,
vct_prcsupcen_ac,
vct_stchop_ac, vct_stclop_ac,
vct_stnclrdop_ac, vct_stclrdop_ac,
vct_multtypop_ac,
vct_rndop_ac,
vct_macqop_ac,
vct_mudlop_ac, vct_madlop_ac,
vct_mulfop_ac, vct_muluop_ac,
vct_rndpop_ac, vct_rndnop_ac,
vct_absop_ac,
vct_vseqone_ac,
vdp_addlwco_ac, vdp_addlwov_ac,
vdp_csupco_ac, vdp_addupco_ac,
vmu_co_clal_ac, vmu_co_clah_ac,
vct_praclwsl_ac, vct_pracupsl_ac,
vct_multactyp_ac, vct_multincop_ac,
su_wrcmpcd_wb, su_wrcmpcdad_wb, su_wrcryout_wb,
su_datainlo_wb, su_datainhi_wb,
vdp_accbit15_wb, vdp_accbit21_wb,
vdp_accbit31_wb, vdp_accbit47_wb,
vdp_acchizero_wb, vdp_acchione_wb,
vdp_accmizero_wb,
vct_acchighsl_wb, vct_accmidsl_wb,
vct_acclowsl_wb, vct_accshftsl_wb,
vct_divrsltsl_wb,
vct_clpsgn16_wb, vct_clpsgn31_wb,
vct_clpsgn32_wb, vct_clpuns31_wb,
vct_adscl16op_wb,
vct_alucin_mu,
vct_alucmpvt_mu,
vct_compvt_mu,
vct_smlwrsl_mu,
vct_cryout_ac, vct_opdneql_ac,
vct_cmpcdlo_ac, vct_cmpcdhi_ac,
vct_cmpcdad_ac,
vct_rndvlu_ac,
vct_cslwbsl_ac,
vct_addlwci_ac,
vct_csupbsl_ac,
vct_csupcen_ac,
vct_incrdwn_ac, vct_incrci_ac,
vct_incrmxsl_ac,
vct_aclwsl_ac, vct_acupsl_ac,
vct_rsltsl_wb,
vct_clprslt_wb
);
/*
* The following signals are the input signals to the
* vector unit control block.
*
* The first group are input signals to the control block
* which provide general control such as clocks, reset
* hold and instruction decoding.
*
*/
input clk; /* vu clock */
input reset_l; /* active low reset signal */
/*
* The following are signals to the control slice from the scalar unit
* directly.
*/
input su_instvld_rd; /* valid CP2 instruction for vu */
/*
* The next group are input signals to the control block from
* the register file read stage of the vector unit datapaths.
*/
/*
* The next group are input signals to the control slice from
* the multiply stage of the vector unit control block.
*/
input vct_instvld_mu; /* valid instruction in multiply stage */
input vct_pralucmpvt_mu; /* complement vt for subtact in ALU in MU stage */
input vct_addcop_mu; /* ADDC instruction in MU */
input vct_subcop_mu; /* SUBC instruction in MU */
input vct_addop_mu; /* ADD instruction in MU */
input vct_subop_mu; /* SUB instruction in MU */
input vct_vs_sgnmu_mu; /* Multiply instruction with operand s signed */
input vct_vt_sgnmu_mu; /* Multiply instruction with operand t signed */
input vct_stltop_mu; /* VS < VT select instruction in MU stage */
input vct_steqop_mu; /* VS == VT select instruction in MU stage */
input vct_stneop_mu; /* VS != VT select instruction in MU stage */
input vct_stgeop_mu; /* VS >= VT select instruction in MU stage */
input vct_stchop_mu; /* -VT <= VS <= VT SP select instruction 2's comp */
input vct_stclop_mu; /* -VT <= VS <= VT DP select instruction 2's comp */
input vct_stcrop_mu; /* -VT <= VS <= VT select instruction 1's comp */
input vct_stchrop_mu; /* -VT <= VS <= VT select instruction 2's or 1's comp */
input vct_substclop_mu; /* sub or -VT<=VS<=VT select instruction 2's comp double precision*/
input vct_absop_mu; /* ABS instruction in MU stage. */
input vct_rndop_mu; /* RNDP or RNDN instruction in MU stage. */
input vct_rndpop_mu; /* RNDP instruction in MU stage. */
input vct_rndnop_mu; /* RNDN instruction in MU stage. */
input vct_mulqop_mu; /* MULQ instruction in MU stage. */
input vct_mulfop_mu; /* MULF instruction in MU stage. */
input vct_muluop_mu; /* MULU instruction in MU stage. */
input vct_vseqone_mu; /* vs field of instruction equal to 1 in MU */
input vct_aluprecin_rd; /* carry in to alu decoded subtract instructions */
input vct_cryoutld_mu; /* Load enable for Vector carry out/equal to register */
input vct_cmpcdld_mu; /* Load enable for vector compare code register */
input vct_cmpcdadld_mu; /* Load enable for vector compare add register */
/*
* The next group are input signals to the control block from
* the multiply stage of the vector unit datapaths.
*/
input vdp_vs_zero_mu; /* vs operand is equal to zero */
input vdp_vt_zero_mu; /* vt operand is equal to zero */
input vdp_vs_sign_mu; /* vs sign bit */
input vdp_vt_sign_mu; /* vt sign bit */
input vdp_aluovr_mu; /* overflow bit from alu */
input vdp_aluco_mu; /* carry out from alu */
input vdp_aluzero_mu; /* alu result is equal to zero */
input vdp_aluone_mu; /* alu result is equal to all ones (minus one) */
input [1:0] vct_prsmlwrsl_mu; /* pre-selects for multiply lower sum out */
/*
* The next group are input signals to the control slice from
* the accumulate stage of the vector unit control block.
*/
input [2:0] vct_prcslwbsl_ac; /* MU stage mux select for input b of lower CSA */
input [1:0] vct_prcsupbsl_ac; /* MU stage mux select for input b of upper CSA */
input vct_prcsupcen_ac; /* predecoded signal for enabling c input of upper csa */
input vct_instvld_ac; /* valid instruction in accumulate stage */
input vct_multtypop_ac; /* Multiply type op in AC */
input vct_rndop_ac; /* RNDP or RNDN instruction in AC stage. */
input vct_macqop_ac; /* Mucq instruction in AC */
input vct_mudlop_ac; /* MUDL instruction in AC stage. */
input vct_madlop_ac; /* MADL instruction in AC stage. */
input vct_mulfop_ac; /* MULF instruction in AC stage. */
input vct_muluop_ac; /* MULU instruction in AC stage. */
input vct_rndpop_ac; /* RNDP instruction in AC stage. */
input vct_rndnop_ac; /* RNDN instruction in AC stage. */
input vct_stchop_ac; /* -VT<=VS<=VT select op 2's comp single precison */
input vct_stclop_ac; /* -VT<=VS<=VT select op 2's comp double precison */
input vct_stnclrdop_ac; /* Select instruction that is not CL, CLD or CR */
input vct_stclrdop_ac; /* Select instruction that is CL, CLD or CR */
input vct_absop_ac; /* ABS instruction in AC stage. */
input vct_vseqone_ac; /* vs field of instruction equal to 1 in AC */
input [1:0] vct_praclwsl_ac; /* selects input for lower mux of accumulator */
input [1:0] vct_pracupsl_ac; /* selects input for upper mux of accumulator */
input vct_multactyp_ac; /* mux select for incrementer output all vectors */
input vct_multincop_ac; /* MACF, MACU, MADL, MADM and MADN type ops */
/*
* The next group are input signals to the control slice from
* the accumulate stage of the vector unit datapaths.
*/
input vdp_addlwco_ac; /* carry out from low adder */
input vdp_addlwov_ac; /* overflow from low adder */
input vdp_csupco_ac; /* carry out from high csa */
input vdp_addupco_ac; /* carry out from high adder */
input vmu_co_clal_ac; /* carry out from 16 bit product of multiplier */
input vmu_co_clah_ac; /* false carry out from multiplier */
/*
* The next group are input signals to the control slice from
* the Scalar unit.
*/
input su_wrcmpcd_wb; /* write vector compare code register */
input su_wrcmpcdad_wb; /* write vector compare add register */
input su_wrcryout_wb; /* write vector carry out register */
input su_datainlo_wb; /* Data in to low half of VCC, VCA and VCO registers */
input su_datainhi_wb; /* Data in to high half of VCC, VCA and VCO registers */
/*
* The next group are input signals to the control slice from
* the writeback stage of the vector unit datapaths.
*/
input vdp_accbit15_wb; /* bit 15 of accumulator used to determine sign */
input vdp_accbit21_wb; /* bit 21 of accumulator used to determine sign */
input vdp_accbit31_wb; /* bit 31 of accumulator used to determine sign */
input vdp_accbit47_wb; /* bit 47 of accumulator used to determine sign */
input vdp_acchizero_wb; /* 47:32 of accumulator equal zero */
input vdp_acchione_wb; /* 47:32 of accumulator equal one */
input vdp_accmizero_wb; /* 31:16 of accumulator equal zero */
/*
* The next group are input signals to the control slice from
* the writeback stage of the vector unit control.
*/
input vct_acchighsl_wb; /* select high portion of accumulator */
input vct_accmidsl_wb; /* select mid portion of accumulator */
input vct_acclowsl_wb; /* select low portion of accumulator */
input vct_accshftsl_wb; /* select shifted high/mid portion of accumulator */
input vct_divrsltsl_wb; /* select result from divide unit */
input vct_clpsgn16_wb; /* instruction requiring signed clamping on 15 to 0 */
input vct_clpsgn31_wb; /* instruction requiring signed clamping on 31 to MSB */
input vct_clpsgn32_wb; /* instruction requiring signed clamping on 32 to MSB */
input vct_clpuns31_wb; /* instruction requiring unsigned clamping on 31 to MSB */
input vct_adscl16op_wb; /* 16 bit ADD, SUB or ABS instruction in WB */
/*
* The following signals are the output signals for the
* vector unit control block.
*
*/
/*
* The next group are output control signals for the
* multiply stage of the vector unit datapath.
*/
output vct_alucin_mu; /* carry in to alu vector */
output vct_alucmpvt_mu; /* complement vt subtact in ALU */
output vct_compvt_mu; /* complement vt for writing -VT for CL, CLD, CR */
output [1:0] vct_smlwrsl_mu; /* selects for multiply lower sum out all vectors */
/*
* The next group are output control signals for the
* accumulate stage of the vector unit datapath.
*/
output vct_cryout_ac; /* Carry out flag from Vector Carry Out register */
output vct_opdneql_ac; /* Data equal flag from Vector Carry Out register */
output vct_cmpcdlo_ac; /* Flag for low half of Vector Compare Code register */
output vct_cmpcdhi_ac; /* Flag for high half of Vector Compare Code register */
output vct_cmpcdad_ac; /* Flag for Vector Compare Add register */
output [3:0] vct_rndvlu_ac; /* round value for multiplies */
output [1:0] vct_cslwbsl_ac; /* selects for input b of lower csa */
output vct_addlwci_ac; /* carry in to lower adder */
output [1:0] vct_csupbsl_ac; /* selects for input b of upper csa */
output vct_csupcen_ac; /* output c enable for upper csa */
output vct_incrdwn_ac; /* increment/decrement control signal */
output vct_incrci_ac; /* increment/decrement enable signal */
output vct_incrmxsl_ac; /* mux select for incrementer output */
output [1:0] vct_aclwsl_ac; /* selects input for lower mux of accumulator */
output [1:0] vct_acupsl_ac; /* selects input for upper mux of accumulator */
/*
* The next group are output control signals for the
* write back stage of the vector unit datapath.
*/
output [2:0] vct_rsltsl_wb; /* selects for result mux */
output [2:0] vct_clprslt_wb; /* clamp value for all clamping */
/*
* The following are the flip-flops used by the vector unit control block.
* The flags contain such information as the signs of the input operands,
* sign of the result of ALU or ADD/SUB operations in the mu stage and
*
*
* All flip-flops use the asdffen flip-flop from Compasses's library.
* asdffen #(size, reset_val) instance_name (q, d, load_enable, clk, reset_l);
*
*/
/*
* Operand sign flip-flops
*/
/*
* This group of signals are for the pipechain of the operand signs.
*/
wire vct_vs_sign_ac; /* sign of s operand in AC */
wire vct_vt_sign_ac; /* sign of t operand in AC */
asdffen #(1, 0) vctopssignffac (vct_vs_sign_ac, vdp_vs_sign_mu, vct_instvld_mu, clk, reset_l );
asdffen #(1, 0) vctoptsignffac (vct_vt_sign_ac, vdp_vt_sign_mu, vct_instvld_mu, clk, reset_l );
/*
* Overflow detection from ALU - used for clamping.
*/
wire vct_aluovfl_mu; /* overflow flag from ALU in MU stage */
wire vct_aluovflrg_ac; /* overflow flag from ALU piped to AC stage */
wire vct_aluovfl_ac; /* overflow flag from ALU piped and adder in AC stage */
wire vct_aluovfl_wb; /* overflow flag from ALU piped to WB stage */
assign vct_aluovfl_mu = ( vct_addop_mu || vct_subop_mu ) &&
( vdp_aluco_mu ^ vdp_aluovr_mu ) ;
asdffen #(1, 0) vctaluovflffac (vct_aluovflrg_ac, vct_aluovfl_mu, vct_instvld_mu, clk, reset_l );
/*
* Since 2's complementing of abs instruction is done in ac state we need to keep track of overflow
* from ac stage to ensure proper clamping when doing -VT when VT=8000. We want to clamp to max
* sign value of 7fff.
*/
assign vct_aluovfl_ac = ( vct_absop_ac && (vdp_addlwco_ac ^ vdp_addlwov_ac) ) ||
( vct_stclrdop_ac && (vct_vs_sign_ac ^ vct_vt_sign_ac) &&
(vdp_addlwco_ac ^ vdp_addlwov_ac)
) ||
vct_aluovflrg_ac ;
asdffen #(1, 0) vctaluovflffwb (vct_aluovfl_wb, vct_aluovfl_ac, vct_instvld_ac, clk, reset_l );
/*
* Multiplication product sign flip-flop
*/
wire vct_muvs_sign_mu; /* sign of multiply/round/vcl VS operand in MU stage */
wire vct_muvt_sign_mu; /* sign of multiply/round/vcl VT operand in MU stage */
wire vct_pdtsign_mu; /* sign of multiply/round/vcl product in MU stage */
wire vct_pdtsign_ac; /* sign of multiply/round/vcl product in AC stage */
assign vct_muvs_sign_mu = vct_vs_sgnmu_mu && vdp_vs_sign_mu ;
assign vct_muvt_sign_mu = vct_vt_sgnmu_mu && vdp_vt_sign_mu ;
assign vct_pdtsign_mu = ( vct_rndop_mu && vdp_vt_sign_mu ) ||
( vct_stclop_mu && vct_cryout_ac ) ||
( ( vct_stchop_mu || vct_stcrop_mu ) &&
( vdp_vs_sign_mu ^ vdp_vt_sign_mu )
) ||
( !vdp_vs_zero_mu && !vdp_vt_zero_mu && /* if either opnd */
(vct_muvs_sign_mu ^ vct_muvt_sign_mu) /* 0 then positive */
) ;
asdffen #(1, 0) vctpdtsignffac (vct_pdtsign_ac, vct_pdtsign_mu, vct_instvld_mu, clk, reset_l );
/*
* Vector Carry Out flip-flops
*/
wire vct_cndexrsgn_mu; /* Exor of VS and VT sign bits for use in CL and CR */
wire vct_cryout_mu; /* Data input to low bit of Vector Carry Out register */
wire vct_opdneql_mu; /* Data input to high bit of Vector Carry Out register */
assign vct_cryout_mu = ( vct_instvld_mu &&
( ( vdp_aluco_mu && vct_addcop_mu ) ||
( !vdp_aluco_mu && vct_subcop_mu ) ||
( vct_cndexrsgn_mu && vct_stchop_mu )
)
) ||
( su_datainlo_wb && su_wrcryout_wb ) ;
/* Always reset unless addc, subc or move to */
asdffen #(1, 0) vctcryoutffac (vct_cryout_ac, vct_cryout_mu, vct_cryoutld_mu, clk, reset_l );
assign vct_opdneql_mu = ( vct_instvld_mu && !vdp_aluzero_mu &&
( vct_subcop_mu ||
( !vct_cndexrsgn_mu && vct_stchop_mu ) ||
( vct_cndexrsgn_mu && vct_stchop_mu && !vdp_aluone_mu )
)
) ||
( su_datainhi_wb && su_wrcryout_wb ) ;
/* Always reset unless addc, subc or move to */
asdffen #(1, 0) vctopdneqlffac (vct_opdneql_ac, vct_opdneql_mu, vct_cryoutld_mu, clk, reset_l );
/*
* Vector Compare Code flip-flops
*/
wire vct_cmpcdlo_mu; /* Data input to low bit of Vector compare code register */
wire vct_cmpcdhi_mu; /* Data input to high bit of Vector compare code register */
wire vct_cndvsltvt_mu; /* Compare condition VS < VT */
wire vct_cndvseqvt_mu; /* Compare condition VS == VT */
wire vct_cndvsnevt_mu; /* Compare condition VS != VT */
wire vct_cndvsgevt_mu; /* Compare condition VS >= VT */
wire vct_cndclsplo_mu; /* Compare condition low single precision -VT<=VS<=VT 2's comp */
wire vct_cndcldplo_mu; /* Compare condition low double precision -VT<=VS<=VT 2's comp */
wire vct_cndclsphi_mu; /* Compare condition high single precision -VT<=VS<=VT 2's comp */
wire vct_cndcldphi_mu; /* Compare condition high double precision -VT<=VS<=VT 2's comp */
wire vct_cndcrlo_mu; /* Compare condition low -VT <= VS <= VT 1's complement */
wire vct_cndcrhi_mu; /* Compare condition high -VT <= VS <= VT 1's complement */
wire vct_cndvslenvt_mu; /* Compare condition VS <= VT */
assign vct_cndexrsgn_mu = vdp_vs_sign_mu ^ vdp_vt_sign_mu ;
assign vct_compvt_mu = !( vct_absop_mu && vdp_vs_sign_mu ) &&
!( ( vct_stchop_mu || vct_stcrop_mu ) && vct_cndexrsgn_mu ) &&
!( vct_stclop_mu && vct_cryout_ac ) ;
assign vct_cndvsltvt_mu = ( vdp_vs_sign_mu && !vdp_vt_sign_mu ) ||
( !vdp_aluco_mu && !vdp_aluzero_mu && !vct_cndexrsgn_mu ) ||
( vdp_aluzero_mu && !vct_cndexrsgn_mu &&
vct_opdneql_ac && vct_cryout_ac
) ;
assign vct_cndvseqvt_mu = vdp_aluzero_mu && !vct_opdneql_ac && !vct_cndexrsgn_mu ;
assign vct_cndvsnevt_mu = !vdp_aluzero_mu || vct_opdneql_ac || vct_cndexrsgn_mu ;
assign vct_cndvsgevt_mu = ( !vdp_vs_sign_mu && vdp_vt_sign_mu ) ||
( vdp_aluco_mu && !vdp_aluzero_mu && !vct_cndexrsgn_mu ) ||
( vdp_aluzero_mu && !vct_cndexrsgn_mu &&
( !vct_opdneql_ac || !vct_cryout_ac )
) ;
assign vct_cndvslenvt_mu = !vdp_aluco_mu || vdp_aluzero_mu ;
/* signs different AND VS+VT<=0 */
assign vct_cndclsplo_mu = ( !vct_cndexrsgn_mu && vdp_vt_sign_mu ) ||
( vct_cndexrsgn_mu && vct_cndvslenvt_mu ) ;
/* signs same AND VS-VT>=0 */
assign vct_cndclsphi_mu = ( vct_cndexrsgn_mu && vdp_vt_sign_mu ) ||
( !vct_cndexrsgn_mu && vdp_aluco_mu ) ||
( !vct_cndexrsgn_mu && vdp_aluzero_mu ) ;
assign vct_cndcldplo_mu = ( ( !vct_cryout_ac || vct_opdneql_ac ) && vct_cmpcdlo_ac )
||
( vct_cryout_ac && !vct_opdneql_ac &&
vct_cmpcdad_ac && !vdp_aluco_mu
)
||
( vct_cryout_ac && !vct_opdneql_ac && vdp_aluzero_mu &&
( ( !vct_cmpcdad_ac && !vdp_aluco_mu ) || vct_cmpcdad_ac )
) ;
assign vct_cndcldphi_mu = ( ( vct_cryout_ac || vct_opdneql_ac ) && vct_cmpcdhi_ac ) ||
( !vct_cryout_ac && !vct_opdneql_ac && vdp_aluco_mu ) ;
/* WHY is BORROW INVERSE OF carry??????????????
* ( !vct_cryout_ac && !vct_opdneql_ac && !vdp_aluco_mu ) ;
*/
/* signs different AND VS<=~VT */
assign vct_cndcrlo_mu = ( !vct_cndexrsgn_mu && vdp_vt_sign_mu ) ||
( vct_cndexrsgn_mu && vct_cndvslenvt_mu ) ;
/* signs same AND VS>=VT */
assign vct_cndcrhi_mu = ( vct_cndexrsgn_mu && vdp_vt_sign_mu ) ||
( !vct_cndexrsgn_mu && vdp_aluco_mu ) ||
( !vct_cndexrsgn_mu && vdp_aluzero_mu ) ;
/*
* The data to the vector unit control register is a timing critical path therefore valid
* is taken into account at the instruction decode level to eliminate it from the critical
* timing of the data.
*/
assign vct_cmpcdlo_mu = ( ( vct_cndvsltvt_mu && vct_stltop_mu ) || /* VS<VT */
( vct_cndvseqvt_mu && vct_steqop_mu ) || /* VS==VT */
( vct_cndvsnevt_mu && vct_stneop_mu ) || /* VS!=VT */
( vct_cndvsgevt_mu && vct_stgeop_mu ) || /* VS>=VT */
( vct_cndclsplo_mu && vct_stchop_mu ) || /* -VT<=VS<=VT SP 2s comp */
( vct_cndcldplo_mu && vct_stclop_mu ) || /* -VT<=VS<=VT DP 2s comp */
( vct_cndcrlo_mu && vct_stcrop_mu ) /* -VT<=VS<=VT 1s comp */
) ||
( su_datainlo_wb && su_wrcmpcd_wb ) ;
assign vct_cmpcdhi_mu = ( ( vct_cndclsphi_mu && vct_stchop_mu ) || /* -VT<=VS<=VT SP 2s comp */
( vct_cndcldphi_mu && vct_stclop_mu ) || /* -VT<=VS<=VT DP 2s comp */
( vct_cndcrhi_mu && vct_stcrop_mu ) /* -VT<=VS<=VT 1's comp */
) ||
( su_datainhi_wb && su_wrcmpcd_wb ) ;
asdffen #(1, 0) vctcmpcdloffac (vct_cmpcdlo_ac, vct_cmpcdlo_mu, vct_cmpcdld_mu, clk, reset_l );
asdffen #(1, 0) vctcmpcdhiffac (vct_cmpcdhi_ac, vct_cmpcdhi_mu, vct_cmpcdld_mu, clk, reset_l );
/*
* Vector Compare Add flip-flops
*/
wire vct_cmpcdad_mu; /* Data input to Vector Code Add register */
assign vct_cmpcdad_mu = ( vct_instvld_mu && vct_stchop_mu && /* Always reset unless VCL */
vdp_aluone_mu && vct_cndexrsgn_mu /* result of ADD is minus one */
) ||
( su_datainlo_wb && su_wrcmpcdad_wb ) ;
asdffen #(1, 0) vctcmpcdadffac (vct_cmpcdad_ac, vct_cmpcdad_mu, vct_cmpcdadld_mu, clk, reset_l );
/* ??????
* The following code defines the logic for all the output signals for this
* block.
*/
/*
* This portion controls the operation of the ALU in the MU stage.
*
*/
/*
* The complementing of vt data for subtract and alu carry in signals are timing
* critical since they dependent on signals that are not available until that clock.
*/
/*
* assign vct_aluctl_mu[0] = vct_pralucmpvt_mu ||
* ( vct_absop_mu && vdp_vs_sign_mu ) ||
* ( vct_stclop_mu && !vct_cryout_ac ) ||
* ( vct_stchrop_mu && !vct_cndexrsgn_mu ) ;
*
* vct_aluctllsb_mu ||
* ( vct_stclop_mu && !vct_cryout_ac ) ||
* ( vct_stchrop_mu && !vct_cndexrsgn_mu ) ;
*/
wire vct_aluctlmxtmp_mu; /* input to alu control mux */
wire vct_aluctlmx0_mu; /* input to alu control mux */
wire vct_aluctlmx1_mu; /* input to alu control mux */
assign vct_aluctlmxtmp_mu = vct_pralucmpvt_mu ||
( vct_stclop_mu && !vct_cryout_ac ) ;
assign vct_aluctlmx1_mu = vct_aluctlmxtmp_mu ;
assign vct_aluctlmx0_mu = vct_aluctlmxtmp_mu || vct_stchrop_mu ;
/*
* This was redundant logic that was discoverd late on in rev2.0 design.
*
* assign vct_aluctlmx1_mu = vct_aluctlmxtmp_mu ||
* ( vct_absop_mu && vdp_vs_sign_mu ) ;
*
* assign vct_aluctlmx0_mu = vct_aluctlmxtmp_mu || vct_stchrop_mu ||
* ( vct_absop_mu && vdp_vs_sign_mu ) ;
*
* ao04d1 vctaluctlin1mu ( .zn (vct_aluctlmx1_mu),
* .a1 (vct_vs_sign_mu),??????
* .a2 (vct_absop_mu),
* .b (vct_aluctlmxtmp_mu)
* );
*
* ao05d1 vctaluctlin0mu ( .zn (vct_aluctlmx0_mu),
* .a1 (vct_vs_sign_mu),??????
* .a2 (vct_absop_mu),
* .b (vct_aluctlmxtmp_mu),
* .c (vct_stchrop_mu)
* );
*/
wire vct_alucmpvtcomp_mu;
j_mx21 vctaluctlmx0mu (
.z (vct_alucmpvtcomp_mu),
.i0 (vct_aluctlmx0_mu),
.i1 (vct_aluctlmx1_mu),
.s (vct_cndexrsgn_mu)
) ;
j_ni01 vctalucmpvtinvmu ( .z (vct_alucmpvt_mu),
.i (vct_alucmpvtcomp_mu)
) ;
wire vct_aluprecin_mu; /* carry in to alu decoded subtract instructions */
asdffen #(1, 0) vctalucinffmu (vct_aluprecin_mu, vct_aluprecin_rd, su_instvld_rd, clk, reset_l );
/*
* assign vct_alucin_mu = ( vct_addop_mu && vct_cryout_ac ) ||
* ( vct_substclop_mu && !vct_cryout_ac ) ||
* ( vct_stchop_mu && !vct_cndexrsgn_mu ) ||
* ( vct_absop_mu && vdp_vs_sign_mu ) ||
* vct_aluprecin_mu ;
*/
wire vct_alucinmxtmp_mu; /* input to alu carry in mux */
wire vct_alucinmx0_mu; /* input to alu carry in mux */
wire vct_alucinmx1_mu; /* input to alu carry in mux */
assign vct_alucinmxtmp_mu = vct_aluprecin_mu ||
( vct_addop_mu && vct_cryout_ac ) ||
( vct_substclop_mu && !vct_cryout_ac ) ;
assign vct_alucinmx1_mu = vct_alucinmxtmp_mu ;
assign vct_alucinmx0_mu = vct_alucinmxtmp_mu || vct_stchop_mu ;
/*
* This was redundant logic that was discoverd late on in rev2.0 design.
*
* assign vct_alucinmx1_mu = ( vct_absop_mu && vdp_vs_sign_mu ) ||
* vct_alucinmxtmp_mu ;
*
* assign vct_alucinmx0_mu = ( vct_absop_mu && vdp_vs_sign_mu ) ||
* vct_alucinmxtmp_mu || vct_stchop_mu ;
*
* ao04d1 vctalucinin1mu ( .zn (vct_alucinmx1_mu),
* .a1 (vct_vs_sign_mu),??????
* .a2 (vct_absop_mu),
* .b (vct_alucinmxtmp_mu)
* );
*
* ao05d1 vctalucinin0mu ( .zn (vct_alucinmx0_mu),
* .a1 (vct_vs_sign_mu),??????
* .a2 (vct_absop_mu),
* .b (vct_alucinmxtmp_mu),
* .c (vct_stchop_mu)
* );
*
*/
wire vct_alucincomp_mu;
j_mx21 vctalucinmxmu (
.z (vct_alucincomp_mu),
.i0 (vct_alucinmx0_mu),
.i1 (vct_alucinmx1_mu),
.s (vct_cndexrsgn_mu)
) ;
j_ni01 vctalucininvmu ( .z (vct_alucin_mu),
.i (vct_alucincomp_mu)
) ;
assign vct_smlwrsl_mu[0] =
vct_prsmlwrsl_mu[0] ||
( vct_instvld_mu && !vct_prsmlwrsl_mu[1] &&
!( vct_absop_mu && !vdp_vs_zero_mu )
) ;
assign vct_smlwrsl_mu[1] =
( vct_instvld_mu &&
( vct_prsmlwrsl_mu[1] ||
( vct_absop_mu && !vdp_vs_zero_mu )
)
) ||
!vct_prsmlwrsl_mu[0] ;
/*
* This portion controls the value used for rounding when loading the accumulator.
* This appears extremely obtuse due to changes for timing improvements.
* The old code is shown below which specifies functionally what is happening
* to produce the round value.
*
* assign vct_rndvlu_ac = ( vct_macqop_ac && vct_oddfyneg_ac ) ? 4'h2 :
* ( vct_macqop_ac && vct_oddfypos_ac ) ? 4'hE :
* ( vct_rndpsl_ac && !vdp_accbit47_wb ) ? 4'h8 :
* ( vct_rndnsl_ac && vdp_accbit47_wb ) ? 4'h8 :
* ( vct_mulfop_ac || vct_muluop_ac ) ? 4'h8 :
* vct_mulqsgn_ac ? 4'h1 :
* 0 ;
*
*/
wire vct_rndpsl_mu; /* selection of round value for rndp op in MU stage. */
wire vct_rndnsl_mu; /* selection of round value for rndn op in MU stage. */
assign vct_rndpsl_mu = vct_rndpop_mu && !vct_vseqone_mu && vdp_vt_sign_mu ;
assign vct_rndnsl_mu = vct_rndnop_mu && !vct_vseqone_mu && vdp_vt_sign_mu ;
wire vct_mulqsgn_mu; /* selection of round value for mulq op in MU stage. */
assign vct_mulqsgn_mu = vct_mulqop_mu && vct_pdtsign_mu ;
wire [1:0] vct_prerndvlu_mu; /* predecoded round value for multiplies */
wire [1:0] vct_prerndvlu_ac; /* predecoded round value for multiplies */
assign vct_prerndvlu_mu[0] = vct_mulqsgn_mu ;
assign vct_prerndvlu_mu[1] = vct_rndpsl_mu || vct_rndnsl_mu ||
vct_mulfop_mu || vct_muluop_mu ;
asdffen #(2, 0) vctprerndvluffac (vct_prerndvlu_ac, vct_prerndvlu_mu, vct_instvld_mu, clk, reset_l );
/*????
* Removed to improve timing of round value.
*
* assign vct_oddfypos_ac = !vdp_accbit47_wb && !vdp_accbit21_wb &&
* !(vdp_acchizero_wb && vdp_accmizero_wb) ;
*/ /* 47-21 not zero */
/*????
* assign vct_oddfyneg_ac = vdp_accbit47_wb && !vdp_accbit21_wb &&
* !(vdp_acchizero_wb && vdp_accmizero_wb) ;
*/ /* 47-21 not zero */
wire vct_oddfyposnc_ac; /* non timing critical Oddify positive accumulator for macq op */
wire vct_oddfyneg_ac; /* Oddify negative accumulator for macq op */
assign vct_oddfyposnc_ac = vct_macqop_ac && !vdp_accbit47_wb && !vdp_accbit21_wb ;
assign vct_oddfyneg_ac = vct_macqop_ac && vdp_accbit47_wb && !vdp_accbit21_wb ;
assign vct_rndvlu_ac[0] = vct_prerndvlu_ac[0] ;
wire vct_rndvlu1in_ac; /* Mux Input to round value bit mux */
wire vct_rndvlu1out_ac; /* Mux Input to round value bit mux */
wire vct_rndvlusl_ac; /* Mux select to round value bit mux */
wire vct_rndvlu2out_ac; /* Mux Input to round value bit mux */
assign vct_rndvlu1in_ac = vct_oddfyposnc_ac || vct_oddfyneg_ac ;
/*
* Replace with high performance cells
*
* assign vct_rndvlusl_ac = (vdp_acchizero_wb && vdp_accmizero_wb) ;
*/
j_an02 vctrndvlu1anac (
.z (vct_rndvlusl_ac),
.a1 (vdp_acchizero_wb),
.a2 (vdp_accmizero_wb)
) ; /* 47-21 not zero */
j_mx21 vctrndvlu1mxac (
.z (vct_rndvlu1out_ac),
.i0 (vct_rndvlu1in_ac),
.i1 (vct_oddfyneg_ac),
.s (vct_rndvlusl_ac)
) ;
assign vct_rndvlu_ac[1] = vct_rndvlu1out_ac ;
j_ao01 vctrndvlu2aoac (
.zn (vct_rndvlu2out_ac),
.a1 (!vdp_accmizero_wb),
.a2 (vct_oddfyposnc_ac),
.b1 (!vdp_acchizero_wb),
.b2 (vct_oddfyposnc_ac)
) ; /* 47-21 not zero */
assign vct_rndvlu_ac[2] = !vct_rndvlu2out_ac ;
/*
* assign vct_rndvlu_ac[3] = vct_prerndvlu_ac[1] || ( vct_oddfyposnc_ac && !vct_rndvlusl_ac ) ;
*/
wire vct_rndvlu3in_ac; /* Mux Input to round value bit mux */
assign vct_rndvlu3in_ac = vct_prerndvlu_ac[1] || vct_oddfyposnc_ac ;
j_mx21 vctrndvlu3mxac (
.z (vct_rndvlu_ac[3]),
.i0 (vct_rndvlu3in_ac),
.i1 (vct_prerndvlu_ac[1]),
.s (vct_rndvlusl_ac)
) ;
/*
* Replaced by mux structure for timing reasons. Select is controlled
* by vdp_acchizero_wb
*
* assign vct_rndvlu_ac[1] = vct_oddfypos_ac || vct_oddfyneg_ac ;
*
* assign vct_rndvlu_ac[2] = vct_oddfypos_ac ;
* assign vct_rndvlu_ac[2] = vct_oddfyposnc_ac && !vct_rndvlusl_ac ;
*
* assign vct_rndvlu_ac[3] = vct_prerndvlu_ac[1] || vct_oddfypos_ac ;
*
*/
/*
* This portion controls the muxing into the lower 3 input adder in
* the AC stage.
*/
wire vct_cslwb0mx_ac; /* mux select to low CSA inb assumes vdp_accbit47_wb=0 */
wire vct_cslwb1mx_ac; /* mux select to low CSA inb assumes vdp_accbit47_wb=1 */
assign vct_cslwb0mx_ac = vct_prcslwbsl_ac[0] ;
assign vct_cslwb1mx_ac = vct_prcslwbsl_ac[1] ;
j_mx21 vctcslwb0slmxac (
.z (vct_cslwbsl_ac[0]),
.i0 (vct_cslwb0mx_ac),
.i1 (vct_cslwb1mx_ac),
.s (vdp_accbit47_wb)
) ;
/*
* The code below was replaced by a mux structure for timing reasons
*
* assign vct_cslwbsl_ac[0] = ( vct_rndpop_ac && !vdp_accbit47_wb && !vct_vseqone_ac ) ||
* ( vct_rndnop_ac && vdp_accbit47_wb && !vct_vseqone_ac ) ||
* vct_prcslwbsl_ac[0] ;
*/
assign vct_cslwbsl_ac[1] = vct_prcslwbsl_ac[2] ;
wire vct_cndexrsgn_ac; /* Exor of VS and VT sign bits for use in CL and CR */
assign vct_cndexrsgn_ac = vct_vs_sign_ac ^ vct_vt_sign_ac ;
assign vct_addlwci_ac = ( vct_absop_ac && vct_vs_sign_ac ) ||
( ( vct_stchop_ac || vct_stclop_ac ) &&
vct_pdtsign_ac && vct_cmpcdlo_ac
) ||
( ( vct_mudlop_ac || vct_madlop_ac ) && vmu_co_clal_ac );
/*
* This portion controls the muxing into the upper 3 input adder in
* the AC stage.
*/
assign vct_csupbsl_ac[0] = ( vct_rndpop_ac && !vdp_accbit47_wb && vct_vseqone_ac ) ||
( vct_rndnop_ac && vdp_accbit47_wb && vct_vseqone_ac ) ||
vct_prcsupbsl_ac[0] ;
assign vct_csupbsl_ac[1] = ( vct_rndpop_ac && !vdp_accbit47_wb && !vct_vseqone_ac ) ||
( vct_rndnop_ac && vdp_accbit47_wb && !vct_vseqone_ac ) ||
vct_prcsupbsl_ac[1] ;
assign vct_csupcen_ac = vct_prcsupcen_ac ;
/*
* This portion controls the incrementer/decrementer in the AC stage.
*/
assign vct_incrdwn_ac = ( vct_multtypop_ac || vct_rndpop_ac ||
vct_rndnop_ac ) && vct_pdtsign_ac ;
assign vct_incrci_ac = vct_multincop_ac ||
( vct_rndpop_ac && !vdp_accbit47_wb ) ||
( vct_rndnop_ac && vdp_accbit47_wb ) ;
/*
* Removed for area saving to allow vdpincremxac mux to be reduced from
* a 4to1 mux to a 2to1 mux.
*
* assign vct_incrmxsl_ac[0] = !vct_multactyp_ac && * not positive mulu/f result *
* !( (vct_mulfop_ac || vct_muluop_ac || vct_mudlop_ac) &&
* !vct_pdtsign_ac
* ) ;
*/
assign vct_incrmxsl_ac = vct_multactyp_ac ;
assign vct_aclwsl_ac[0] =
( vct_instvld_ac &&
( vct_praclwsl_ac[0] || vct_praclwsl_ac[1] ||
( vct_stnclrdop_ac && !vct_cmpcdlo_ac ) ||
( vct_stclrdop_ac && !vct_pdtsign_ac && vct_cmpcdhi_ac ) ||
( vct_stclrdop_ac && vct_pdtsign_ac && vct_cmpcdlo_ac )
)
) ;
/*
* Note that vct_pdtsign_ac holds the value of carryout flag for VCL instructions since
* this instruction resets carryout in the MU stage but we need to keep it around to
* determine the sign of the operation.
*/
assign vct_aclwsl_ac[1] =
( vct_instvld_ac &&
( vct_praclwsl_ac[1] ||
( vct_stnclrdop_ac && vct_cmpcdlo_ac ) ||
( vct_stclrdop_ac && !vct_pdtsign_ac && !vct_cmpcdhi_ac ) ||
( vct_stclrdop_ac && vct_pdtsign_ac && !vct_cmpcdlo_ac )
)
) ;
/*
* We only select a new value for the upper half of the accumulator if
* there is a true carry out from the csa/adder and the resulting
* product being added to the accumulator is positive or there is no
* carry out and the resulting product being added to the accumulator
* is negative.
*/
wire vct_truecoutnc_ac; /* Non timing critical True carry out from lower 32 bits */
assign vct_truecoutnc_ac = vct_pdtsign_ac ^ vdp_csupco_ac ^
(!vct_rndop_ac && vmu_co_clah_ac) ;
/*
* vmu_co_clah_ac needs qualified with round instruction since we need to ignore
* value from mult since no proper multiply is occurring but round is treated
* like a multiply op. This change was only required for 2nd tapeout since in
* first tapeout VS was zeroed in the rd stage therefore multiplier always
* multiplied vt by zero during a round instruction. This zeroing could not
* occur on 2nd tapeout due to new register file implementaiton therefore the
* above equation was required.
*/
wire vct_acup1innc_ac; /* Non-timing critical data inputs to acup0slmx */
wire vct_acup1ina_ac; /* Non-timing critical data input a to acup1slmx */
wire vct_acup1inb_ac; /* Non-timing critical data input b to acup1slmx */
wire vct_acup1outnc_ac; /* Non-timing critical data output of acup1slmx */
assign vct_acup1innc_ac =
( vct_instvld_ac && vct_pracupsl_ac[0] ) ;
assign vct_acup1ina_ac = vct_acup1innc_ac ||
( vct_instvld_ac &&
vct_multactyp_ac && vct_truecoutnc_ac
) ;
assign vct_acup1inb_ac = vct_acup1innc_ac ||
( vct_instvld_ac &&
vct_multactyp_ac && !vct_truecoutnc_ac
) ;
j_mx21 vctacup1slmxac (
.z (vct_acup1out_ac),
.i0 (vct_acup1ina_ac),
.i1 (vct_acup1inb_ac),
.s (vdp_addupco_ac)
) ;
assign vct_acupsl_ac[0] = vct_acup1out_ac ;
assign vct_acupsl_ac[1] =
( vct_instvld_ac && vct_pracupsl_ac[1] ) ||
( vct_instvld_ac &&
(vct_mulfop_ac || vct_muluop_ac || vct_mudlop_ac) &&
!vct_pdtsign_ac
) ;
/*
* This portion controls the various clamp values and when clamping is
* required.
*/
wire vct_clpsgnmu_wb; /* clamped signed multiply */
wire vct_clpunsmu_wb; /* clamped unsigned multiply */
wire vct_clpmnsgas_wb; /* 16 bit signed add/subtract underflowed */
wire vct_clpmxsgas_wb; /* 16 bit signed add/subtract overflowed */
wire vct_clpsgnmx0_wb; /* input to clamped signed control mux data 0 */
wire vct_clpsgnmx1_wb; /* input to clamped signed control mux data 1 */
assign vct_clpsgnmx0_wb = ( vct_clpsgn31_wb && !vdp_accbit31_wb ) /* clamp signed 31 to MSB */
||
( ( vct_clpsgn31_wb || vct_clpsgn32_wb ) && /* clamp on 32 to MSB */
!vdp_acchione_wb /* Not all ones */
) ;
assign vct_clpsgnmx1_wb = ( vct_clpsgn31_wb && /* clamp signed on 31 to MSB */
vdp_accbit31_wb /* Not all zeroes including mux sel */
) ;
/*
* Not all ones test not necessary since vdp_acchizero_wb we know is one
* && !( vdp_accbit31_wb && !vdp_acchione_wb ) ) ;
*/
j_mx21 vctclpsgnmuwb (
.z (vct_clpsgnmu_wb),
.i0 (vct_clpsgnmx0_wb),
.i1 (vct_clpsgnmx1_wb),
.s (vdp_acchizero_wb)
) ;
/* Modified toimprove timing of acchizero signal.
*
* assign vct_clpsgnmu_wb = ( vct_clpsgn31_wb &&
* !( !vdp_accbit31_wb && vdp_acchizero_wb ) &&
* !( vdp_accbit31_wb && vdp_acchione_wb )
* )
* ||
* ( vct_clpsgn32_wb && //
* !vdp_acchizero_wb && !vdp_acchione_wb //
* ) ;
*/
/* Replaced to improve timing of acchihero signal.
*
* assign vct_clpunsmu_wb = ( vct_clpsgn16_wb &&
* !( !vdp_accbit31_wb && vdp_acchizero_wb ) &&
* !( vdp_accbit31_wb && vdp_acchione_wb )
* ) ||
* ( vct_clpuns31_wb &&
* !( !vdp_accbit31_wb && vdp_acchizero_wb )
* ) ;
*/
assign vct_clpunsmu_wb = !( !vdp_accbit31_wb && vdp_acchizero_wb ) && /* Not all zeroes */
(
( vct_clpsgn16_wb && /* clamp unsigned on 31 to MSB */
!( vdp_accbit31_wb && vdp_acchione_wb ) /* Not all ones */
)
|| vct_clpuns31_wb
) ;
/*
* ???? simplified unsigned clamping to help timing of vdp_accbit47_wb.
* ( vct_clpuns31_wb && vdp_accbit47_wb ) ||
* ( vct_clpuns31_wb && !vdp_accbit47_wb &&
* (vdp_accbit31_wb || !vdp_acchizero_wb)
* ) ;
*/
assign vct_clpmnsgas_wb = vct_adscl16op_wb && vct_aluovfl_wb && !vdp_accbit15_wb ;
assign vct_clpmxsgas_wb = vct_adscl16op_wb && vct_aluovfl_wb && vdp_accbit15_wb ;
wire vct_clprsltsl_wb; /* clamping required for result */
assign vct_clprsltsl_wb = vct_clpsgnmu_wb || vct_clpunsmu_wb || vct_clpmnsgas_wb ||
vct_clpmxsgas_wb ;
wire vct_clpminsgn_wb; /* clamped result to 16 bit signed minimum */
wire vct_clpmaxsgn_wb; /* clamped result to 16 bit signed maximum */
wire vct_clpmxun16_wb; /* clamped result to 16 bit unsigned maximum */
wire vct_clpmxsgn32_wb; /* clamped result to 16 bit signed maximum for mulq/macq */
wire vct_clpmnsgn32_wb; /* clamped result to 16 bit signed minimum for mulq/macq */
/*
* 16 bit signed underflow
*
*/
assign vct_clpminsgn_wb = ( vct_clpsgn31_wb && vdp_accbit47_wb && /* ACC -ve */
(!vdp_accbit31_wb || !vdp_acchione_wb) /* Not all ones */
) ;
/*
* 16 bit signed overflow
*/
assign vct_clpmaxsgn_wb = ( vct_clpsgn31_wb && !vdp_accbit47_wb && /* ACC +ve */
(vdp_accbit31_wb || !vdp_acchizero_wb) /* Not all zeroes */
) ;
/*
* 16 bit unsigned overflow
*/
assign vct_clpmxun16_wb = ( ( vct_clpuns31_wb || vct_clpsgn16_wb ) &&
!vdp_accbit47_wb && /* ACC +ve */
!(!vdp_accbit31_wb && vdp_acchizero_wb) /* Not all zeroes */
) ;
assign vct_clpmxsgn32_wb = ( vct_clpsgn32_wb && !vdp_accbit47_wb && /* clamp on 32 to MSB */
!vdp_acchizero_wb /* Not all zeroes */
) ;
assign vct_clpmnsgn32_wb = ( vct_clpsgn32_wb && vdp_accbit47_wb && /* clamp on 32 to MSB */
!vdp_acchione_wb /* Not all ones */
) ;
/*
* Replaced following for timing reasons since synopsys really did not do a
* good job with it.
*
* assign vct_clprslt_wb = (vct_clpminsgn_wb || vct_clpmnsgn32_wb) ? 3'h4 :
* vct_clpmaxsgn_wb ? 3'h3 :
* vct_clpmxun16_wb ? 3'h7 :
* vct_clpmxsgn32_wb ? 3'h2 : 3'h0 ;
*/
assign vct_clprslt_wb[0] = vct_clpmxsgas_wb || vct_clpmaxsgn_wb || vct_clpmxun16_wb ;
assign vct_clprslt_wb[1] = vct_clpmxsgas_wb || vct_clpmaxsgn_wb ||
vct_clpmxun16_wb || vct_clpmxsgn32_wb ;
assign vct_clprslt_wb[2] = vct_clpmnsgas_wb || vct_clpminsgn_wb ||
vct_clpmnsgn32_wb || vct_clpmxun16_wb ;
/*
* This portion controls the muxing into the result bus to be written back
* into the VU register file in the Wb stage.
*????? other terms for low portion of accumulator need to go in here.
*/
assign vct_rsltsl_wb[0] = ( vct_acchighsl_wb && !vct_clprsltsl_wb && !vct_divrsltsl_wb ) ||
( vct_acclowsl_wb && !vct_clprsltsl_wb && !vct_divrsltsl_wb ) ||
( vct_clprsltsl_wb && !vct_divrsltsl_wb ) ;
assign vct_rsltsl_wb[1] = ( vct_accmidsl_wb && !vct_clprsltsl_wb && !vct_divrsltsl_wb ) ||
( vct_acclowsl_wb && !vct_clprsltsl_wb && !vct_divrsltsl_wb ) ||
vct_divrsltsl_wb ;
assign vct_rsltsl_wb[2] = ( vct_accshftsl_wb && !vct_clprsltsl_wb && !vct_divrsltsl_wb ) ||
( vct_clprsltsl_wb && !vct_divrsltsl_wb ) ||
vct_divrsltsl_wb ;
endmodule