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
|
/*
* Copyright (c) 1999-2005 Mark D. Hill and David A. Wood
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are
* met: redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer;
* redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution;
* neither the name of the copyright holders nor the names of its
* contributors may be used to endorse or promote products derived from
* this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
machine(L1Cache, "MESI Directory L1 Cache CMP")
: Sequencer * sequencer,
CacheMemory * L1IcacheMemory,
CacheMemory * L1DcacheMemory,
int l2_select_num_bits,
int l1_request_latency = 2,
int l1_response_latency = 2,
int to_l2_latency = 1,
bool send_evictions
{
// NODE L1 CACHE
// From this node's L1 cache TO the network
// a local L1 -> this L2 bank, currently ordered with directory forwarded requests
MessageBuffer requestFromL1Cache, network="To", virtual_network="0", ordered="false", vnet_type="request";
// a local L1 -> this L2 bank
MessageBuffer responseFromL1Cache, network="To", virtual_network="1", ordered="false", vnet_type="response";
MessageBuffer unblockFromL1Cache, network="To", virtual_network="2", ordered="false", vnet_type="unblock";
// To this node's L1 cache FROM the network
// a L2 bank -> this L1
MessageBuffer requestToL1Cache, network="From", virtual_network="0", ordered="false", vnet_type="request";
// a L2 bank -> this L1
MessageBuffer responseToL1Cache, network="From", virtual_network="1", ordered="false", vnet_type="response";
// STATES
state_declaration(State, desc="Cache states", default="L1Cache_State_I") {
// Base states
NP, AccessPermission:Invalid, desc="Not present in either cache";
I, AccessPermission:Invalid, desc="a L1 cache entry Idle";
S, AccessPermission:Read_Only, desc="a L1 cache entry Shared";
E, AccessPermission:Read_Only, desc="a L1 cache entry Exclusive";
M, AccessPermission:Read_Write, desc="a L1 cache entry Modified", format="!b";
// Transient States
IS, AccessPermission:Busy, desc="L1 idle, issued GETS, have not seen response yet";
IM, AccessPermission:Busy, desc="L1 idle, issued GETX, have not seen response yet";
SM, AccessPermission:Read_Only, desc="L1 idle, issued GETX, have not seen response yet";
IS_I, AccessPermission:Busy, desc="L1 idle, issued GETS, saw Inv before data because directory doesn't block on GETS hit";
M_I, AccessPermission:Busy, desc="L1 replacing, waiting for ACK";
SINK_WB_ACK, AccessPermission:Busy, desc="This is to sink WB_Acks from L2";
}
// EVENTS
enumeration(Event, desc="Cache events") {
// L1 events
Load, desc="Load request from the home processor";
Ifetch, desc="I-fetch request from the home processor";
Store, desc="Store request from the home processor";
Inv, desc="Invalidate request from L2 bank";
// internal generated request
L1_Replacement, desc="L1 Replacement", format="!r";
// other requests
Fwd_GETX, desc="GETX from other processor";
Fwd_GETS, desc="GETS from other processor";
Fwd_GET_INSTR, desc="GET_INSTR from other processor";
Data, desc="Data for processor";
Data_Exclusive, desc="Data for processor";
DataS_fromL1, desc="data for GETS request, need to unblock directory";
Data_all_Acks, desc="Data for processor, all acks";
Ack, desc="Ack for processor";
Ack_all, desc="Last ack for processor";
WB_Ack, desc="Ack for replacement";
}
// TYPES
// CacheEntry
structure(Entry, desc="...", interface="AbstractCacheEntry" ) {
State CacheState, desc="cache state";
DataBlock DataBlk, desc="data for the block";
bool Dirty, default="false", desc="data is dirty";
}
// TBE fields
structure(TBE, desc="...") {
Address Address, desc="Physical address for this TBE";
State TBEState, desc="Transient state";
DataBlock DataBlk, desc="Buffer for the data block";
bool Dirty, default="false", desc="data is dirty";
bool isPrefetch, desc="Set if this was caused by a prefetch";
int pendingAcks, default="0", desc="number of pending acks";
}
structure(TBETable, external="yes") {
TBE lookup(Address);
void allocate(Address);
void deallocate(Address);
bool isPresent(Address);
}
TBETable L1_TBEs, template_hack="<L1Cache_TBE>";
MessageBuffer mandatoryQueue, ordered="false";
int l2_select_low_bit, default="RubySystem::getBlockSizeBits()";
void set_cache_entry(AbstractCacheEntry a);
void unset_cache_entry();
void set_tbe(TBE a);
void unset_tbe();
// inclusive cache returns L1 entries only
Entry getCacheEntry(Address addr), return_by_pointer="yes" {
Entry L1Dcache_entry := static_cast(Entry, "pointer", L1DcacheMemory[addr]);
if(is_valid(L1Dcache_entry)) {
return L1Dcache_entry;
}
Entry L1Icache_entry := static_cast(Entry, "pointer", L1IcacheMemory[addr]);
return L1Icache_entry;
}
Entry getL1DCacheEntry(Address addr), return_by_pointer="yes" {
Entry L1Dcache_entry := static_cast(Entry, "pointer", L1DcacheMemory[addr]);
return L1Dcache_entry;
}
Entry getL1ICacheEntry(Address addr), return_by_pointer="yes" {
Entry L1Icache_entry := static_cast(Entry, "pointer", L1IcacheMemory[addr]);
return L1Icache_entry;
}
State getState(TBE tbe, Entry cache_entry, Address addr) {
assert((L1DcacheMemory.isTagPresent(addr) && L1IcacheMemory.isTagPresent(addr)) == false);
if(is_valid(tbe)) {
return tbe.TBEState;
} else if (is_valid(cache_entry)) {
return cache_entry.CacheState;
}
return State:NP;
}
void setState(TBE tbe, Entry cache_entry, Address addr, State state) {
assert((L1DcacheMemory.isTagPresent(addr) && L1IcacheMemory.isTagPresent(addr)) == false);
// MUST CHANGE
if(is_valid(tbe)) {
tbe.TBEState := state;
}
if (is_valid(cache_entry)) {
cache_entry.CacheState := state;
}
}
AccessPermission getAccessPermission(Address addr) {
TBE tbe := L1_TBEs[addr];
if(is_valid(tbe)) {
DPRINTF(RubySlicc, "%s\n", L1Cache_State_to_permission(tbe.TBEState));
return L1Cache_State_to_permission(tbe.TBEState);
}
Entry cache_entry := getCacheEntry(addr);
if(is_valid(cache_entry)) {
DPRINTF(RubySlicc, "%s\n", L1Cache_State_to_permission(cache_entry.CacheState));
return L1Cache_State_to_permission(cache_entry.CacheState);
}
DPRINTF(RubySlicc, "%s\n", AccessPermission:NotPresent);
return AccessPermission:NotPresent;
}
DataBlock getDataBlock(Address addr), return_by_ref="yes" {
return getCacheEntry(addr).DataBlk;
}
void setAccessPermission(Entry cache_entry, Address addr, State state) {
if (is_valid(cache_entry)) {
cache_entry.changePermission(L1Cache_State_to_permission(state));
}
}
Event mandatory_request_type_to_event(RubyRequestType type) {
if (type == RubyRequestType:LD) {
return Event:Load;
} else if (type == RubyRequestType:IFETCH) {
return Event:Ifetch;
} else if ((type == RubyRequestType:ST) || (type == RubyRequestType:ATOMIC)) {
return Event:Store;
} else {
error("Invalid RubyRequestType");
}
}
int getPendingAcks(TBE tbe) {
return tbe.pendingAcks;
}
out_port(requestIntraChipL1Network_out, RequestMsg, requestFromL1Cache);
out_port(responseIntraChipL1Network_out, ResponseMsg, responseFromL1Cache);
out_port(unblockNetwork_out, ResponseMsg, unblockFromL1Cache);
// Response IntraChip L1 Network - response msg to this L1 cache
in_port(responseIntraChipL1Network_in, ResponseMsg, responseToL1Cache) {
if (responseIntraChipL1Network_in.isReady()) {
peek(responseIntraChipL1Network_in, ResponseMsg, block_on="Address") {
assert(in_msg.Destination.isElement(machineID));
Entry cache_entry := getCacheEntry(in_msg.Address);
TBE tbe := L1_TBEs[in_msg.Address];
if(in_msg.Type == CoherenceResponseType:DATA_EXCLUSIVE) {
trigger(Event:Data_Exclusive, in_msg.Address, cache_entry, tbe);
} else if(in_msg.Type == CoherenceResponseType:DATA) {
if ((getState(tbe, cache_entry, in_msg.Address) == State:IS ||
getState(tbe, cache_entry, in_msg.Address) == State:IS_I) &&
machineIDToMachineType(in_msg.Sender) == MachineType:L1Cache) {
trigger(Event:DataS_fromL1, in_msg.Address, cache_entry, tbe);
} else if ( (getPendingAcks(tbe) - in_msg.AckCount) == 0 ) {
trigger(Event:Data_all_Acks, in_msg.Address, cache_entry, tbe);
} else {
trigger(Event:Data, in_msg.Address, cache_entry, tbe);
}
} else if (in_msg.Type == CoherenceResponseType:ACK) {
if ( (getPendingAcks(tbe) - in_msg.AckCount) == 0 ) {
trigger(Event:Ack_all, in_msg.Address, cache_entry, tbe);
} else {
trigger(Event:Ack, in_msg.Address, cache_entry, tbe);
}
} else if (in_msg.Type == CoherenceResponseType:WB_ACK) {
trigger(Event:WB_Ack, in_msg.Address, cache_entry, tbe);
} else {
error("Invalid L1 response type");
}
}
}
}
// Request InterChip network - request from this L1 cache to the shared L2
in_port(requestIntraChipL1Network_in, RequestMsg, requestToL1Cache) {
if(requestIntraChipL1Network_in.isReady()) {
peek(requestIntraChipL1Network_in, RequestMsg, block_on="Address") {
assert(in_msg.Destination.isElement(machineID));
Entry cache_entry := getCacheEntry(in_msg.Address);
TBE tbe := L1_TBEs[in_msg.Address];
if (in_msg.Type == CoherenceRequestType:INV) {
trigger(Event:Inv, in_msg.Address, cache_entry, tbe);
} else if (in_msg.Type == CoherenceRequestType:GETX || in_msg.Type == CoherenceRequestType:UPGRADE) {
// upgrade transforms to GETX due to race
trigger(Event:Fwd_GETX, in_msg.Address, cache_entry, tbe);
} else if (in_msg.Type == CoherenceRequestType:GETS) {
trigger(Event:Fwd_GETS, in_msg.Address, cache_entry, tbe);
} else if (in_msg.Type == CoherenceRequestType:GET_INSTR) {
trigger(Event:Fwd_GET_INSTR, in_msg.Address, cache_entry, tbe);
} else {
error("Invalid forwarded request type");
}
}
}
}
// Mandatory Queue betweens Node's CPU and it's L1 caches
in_port(mandatoryQueue_in, RubyRequest, mandatoryQueue, desc="...") {
if (mandatoryQueue_in.isReady()) {
peek(mandatoryQueue_in, RubyRequest, block_on="LineAddress") {
// Check for data access to blocks in I-cache and ifetchs to blocks in D-cache
if (in_msg.Type == RubyRequestType:IFETCH) {
// ** INSTRUCTION ACCESS ***
Entry L1Icache_entry := getL1ICacheEntry(in_msg.LineAddress);
if (is_valid(L1Icache_entry)) {
// The tag matches for the L1, so the L1 asks the L2 for it.
trigger(mandatory_request_type_to_event(in_msg.Type), in_msg.LineAddress,
L1Icache_entry, L1_TBEs[in_msg.LineAddress]);
} else {
// Check to see if it is in the OTHER L1
Entry L1Dcache_entry := getL1DCacheEntry(in_msg.LineAddress);
if (is_valid(L1Dcache_entry)) {
// The block is in the wrong L1, put the request on the queue to the shared L2
trigger(Event:L1_Replacement, in_msg.LineAddress,
L1Dcache_entry, L1_TBEs[in_msg.LineAddress]);
}
if (L1IcacheMemory.cacheAvail(in_msg.LineAddress)) {
// L1 does't have the line, but we have space for it in the L1 so let's see if the L2 has it
trigger(mandatory_request_type_to_event(in_msg.Type), in_msg.LineAddress,
L1Icache_entry, L1_TBEs[in_msg.LineAddress]);
} else {
// No room in the L1, so we need to make room in the L1
trigger(Event:L1_Replacement, L1IcacheMemory.cacheProbe(in_msg.LineAddress),
getL1ICacheEntry(L1IcacheMemory.cacheProbe(in_msg.LineAddress)),
L1_TBEs[L1IcacheMemory.cacheProbe(in_msg.LineAddress)]);
}
}
} else {
// *** DATA ACCESS ***
Entry L1Dcache_entry := getL1DCacheEntry(in_msg.LineAddress);
if (is_valid(L1Dcache_entry)) {
// The tag matches for the L1, so the L1 ask the L2 for it
trigger(mandatory_request_type_to_event(in_msg.Type), in_msg.LineAddress,
L1Dcache_entry, L1_TBEs[in_msg.LineAddress]);
} else {
// Check to see if it is in the OTHER L1
Entry L1Icache_entry := getL1ICacheEntry(in_msg.LineAddress);
if (is_valid(L1Icache_entry)) {
// The block is in the wrong L1, put the request on the queue to the shared L2
trigger(Event:L1_Replacement, in_msg.LineAddress,
L1Icache_entry, L1_TBEs[in_msg.LineAddress]);
}
if (L1DcacheMemory.cacheAvail(in_msg.LineAddress)) {
// L1 does't have the line, but we have space for it in the L1 let's see if the L2 has it
trigger(mandatory_request_type_to_event(in_msg.Type), in_msg.LineAddress,
L1Dcache_entry, L1_TBEs[in_msg.LineAddress]);
} else {
// No room in the L1, so we need to make room in the L1
trigger(Event:L1_Replacement, L1DcacheMemory.cacheProbe(in_msg.LineAddress),
getL1DCacheEntry(L1DcacheMemory.cacheProbe(in_msg.LineAddress)),
L1_TBEs[L1DcacheMemory.cacheProbe(in_msg.LineAddress)]);
}
}
}
}
}
}
// ACTIONS
action(a_issueGETS, "a", desc="Issue GETS") {
peek(mandatoryQueue_in, RubyRequest) {
enqueue(requestIntraChipL1Network_out, RequestMsg, latency=l1_request_latency) {
out_msg.Address := address;
out_msg.Type := CoherenceRequestType:GETS;
out_msg.Requestor := machineID;
out_msg.Destination.add(mapAddressToRange(address, MachineType:L2Cache,
l2_select_low_bit, l2_select_num_bits));
DPRINTF(RubySlicc, "address: %s, destination: %s\n",
address, out_msg.Destination);
out_msg.MessageSize := MessageSizeType:Control;
out_msg.Prefetch := in_msg.Prefetch;
out_msg.AccessMode := in_msg.AccessMode;
}
}
}
action(ai_issueGETINSTR, "ai", desc="Issue GETINSTR") {
peek(mandatoryQueue_in, RubyRequest) {
enqueue(requestIntraChipL1Network_out, RequestMsg, latency=l1_request_latency) {
out_msg.Address := address;
out_msg.Type := CoherenceRequestType:GET_INSTR;
out_msg.Requestor := machineID;
out_msg.Destination.add(mapAddressToRange(address, MachineType:L2Cache,
l2_select_low_bit, l2_select_num_bits));
DPRINTF(RubySlicc, "address: %s, destination: %s\n",
address, out_msg.Destination);
out_msg.MessageSize := MessageSizeType:Control;
out_msg.Prefetch := in_msg.Prefetch;
out_msg.AccessMode := in_msg.AccessMode;
}
}
}
action(b_issueGETX, "b", desc="Issue GETX") {
peek(mandatoryQueue_in, RubyRequest) {
enqueue(requestIntraChipL1Network_out, RequestMsg, latency=l1_request_latency) {
out_msg.Address := address;
out_msg.Type := CoherenceRequestType:GETX;
out_msg.Requestor := machineID;
DPRINTF(RubySlicc, "%s\n", machineID);
out_msg.Destination.add(mapAddressToRange(address, MachineType:L2Cache,
l2_select_low_bit, l2_select_num_bits));
DPRINTF(RubySlicc, "address: %s, destination: %s\n",
address, out_msg.Destination);
out_msg.MessageSize := MessageSizeType:Control;
out_msg.Prefetch := in_msg.Prefetch;
out_msg.AccessMode := in_msg.AccessMode;
}
}
}
action(c_issueUPGRADE, "c", desc="Issue GETX") {
peek(mandatoryQueue_in, RubyRequest) {
enqueue(requestIntraChipL1Network_out, RequestMsg, latency= l1_request_latency) {
out_msg.Address := address;
out_msg.Type := CoherenceRequestType:UPGRADE;
out_msg.Requestor := machineID;
out_msg.Destination.add(mapAddressToRange(address, MachineType:L2Cache,
l2_select_low_bit, l2_select_num_bits));
DPRINTF(RubySlicc, "address: %s, destination: %s\n",
address, out_msg.Destination);
out_msg.MessageSize := MessageSizeType:Control;
out_msg.Prefetch := in_msg.Prefetch;
out_msg.AccessMode := in_msg.AccessMode;
}
}
}
action(d_sendDataToRequestor, "d", desc="send data to requestor") {
peek(requestIntraChipL1Network_in, RequestMsg) {
enqueue(responseIntraChipL1Network_out, ResponseMsg, latency=l1_response_latency) {
assert(is_valid(cache_entry));
out_msg.Address := address;
out_msg.Type := CoherenceResponseType:DATA;
out_msg.DataBlk := cache_entry.DataBlk;
out_msg.Dirty := cache_entry.Dirty;
out_msg.Sender := machineID;
out_msg.Destination.add(in_msg.Requestor);
out_msg.MessageSize := MessageSizeType:Response_Data;
}
}
}
action(d2_sendDataToL2, "d2", desc="send data to the L2 cache because of M downgrade") {
enqueue(responseIntraChipL1Network_out, ResponseMsg, latency=l1_response_latency) {
assert(is_valid(cache_entry));
out_msg.Address := address;
out_msg.Type := CoherenceResponseType:DATA;
out_msg.DataBlk := cache_entry.DataBlk;
out_msg.Dirty := cache_entry.Dirty;
out_msg.Sender := machineID;
out_msg.Destination.add(mapAddressToRange(address, MachineType:L2Cache,
l2_select_low_bit, l2_select_num_bits));
out_msg.MessageSize := MessageSizeType:Response_Data;
}
}
action(dt_sendDataToRequestor_fromTBE, "dt", desc="send data to requestor") {
peek(requestIntraChipL1Network_in, RequestMsg) {
enqueue(responseIntraChipL1Network_out, ResponseMsg, latency=l1_response_latency) {
assert(is_valid(tbe));
out_msg.Address := address;
out_msg.Type := CoherenceResponseType:DATA;
out_msg.DataBlk := tbe.DataBlk;
out_msg.Dirty := tbe.Dirty;
out_msg.Sender := machineID;
out_msg.Destination.add(in_msg.Requestor);
out_msg.MessageSize := MessageSizeType:Response_Data;
}
}
}
action(d2t_sendDataToL2_fromTBE, "d2t", desc="send data to the L2 cache") {
enqueue(responseIntraChipL1Network_out, ResponseMsg, latency=l1_response_latency) {
assert(is_valid(tbe));
out_msg.Address := address;
out_msg.Type := CoherenceResponseType:DATA;
out_msg.DataBlk := tbe.DataBlk;
out_msg.Dirty := tbe.Dirty;
out_msg.Sender := machineID;
out_msg.Destination.add(mapAddressToRange(address, MachineType:L2Cache,
l2_select_low_bit, l2_select_num_bits));
out_msg.MessageSize := MessageSizeType:Response_Data;
}
}
action(e_sendAckToRequestor, "e", desc="send invalidate ack to requestor (could be L2 or L1)") {
peek(requestIntraChipL1Network_in, RequestMsg) {
enqueue(responseIntraChipL1Network_out, ResponseMsg, latency=l1_response_latency) {
out_msg.Address := address;
out_msg.Type := CoherenceResponseType:ACK;
out_msg.Sender := machineID;
out_msg.Destination.add(in_msg.Requestor);
out_msg.MessageSize := MessageSizeType:Response_Control;
}
}
}
action(f_sendDataToL2, "f", desc="send data to the L2 cache") {
enqueue(responseIntraChipL1Network_out, ResponseMsg, latency=l1_response_latency) {
assert(is_valid(cache_entry));
out_msg.Address := address;
out_msg.Type := CoherenceResponseType:DATA;
out_msg.DataBlk := cache_entry.DataBlk;
out_msg.Dirty := cache_entry.Dirty;
out_msg.Sender := machineID;
out_msg.Destination.add(mapAddressToRange(address, MachineType:L2Cache,
l2_select_low_bit, l2_select_num_bits));
out_msg.MessageSize := MessageSizeType:Writeback_Data;
}
}
action(ft_sendDataToL2_fromTBE, "ft", desc="send data to the L2 cache") {
enqueue(responseIntraChipL1Network_out, ResponseMsg, latency=l1_response_latency) {
assert(is_valid(tbe));
out_msg.Address := address;
out_msg.Type := CoherenceResponseType:DATA;
out_msg.DataBlk := tbe.DataBlk;
out_msg.Dirty := tbe.Dirty;
out_msg.Sender := machineID;
out_msg.Destination.add(mapAddressToRange(address, MachineType:L2Cache,
l2_select_low_bit, l2_select_num_bits));
out_msg.MessageSize := MessageSizeType:Writeback_Data;
}
}
action(fi_sendInvAck, "fi", desc="send data to the L2 cache") {
peek(requestIntraChipL1Network_in, RequestMsg) {
enqueue(responseIntraChipL1Network_out, ResponseMsg, latency=l1_response_latency) {
out_msg.Address := address;
out_msg.Type := CoherenceResponseType:ACK;
out_msg.Sender := machineID;
out_msg.Destination.add(in_msg.Requestor);
out_msg.MessageSize := MessageSizeType:Response_Control;
out_msg.AckCount := 1;
}
}
}
action(forward_eviction_to_cpu, "\cc", desc="sends eviction information to the processor") {
if (send_evictions) {
DPRINTF(RubySlicc, "Sending invalidation for %s to the CPU\n", address);
sequencer.evictionCallback(address);
}
}
action(g_issuePUTX, "g", desc="send data to the L2 cache") {
enqueue(requestIntraChipL1Network_out, RequestMsg, latency=l1_response_latency) {
assert(is_valid(cache_entry));
out_msg.Address := address;
out_msg.Type := CoherenceRequestType:PUTX;
out_msg.DataBlk := cache_entry.DataBlk;
out_msg.Dirty := cache_entry.Dirty;
out_msg.Requestor:= machineID;
out_msg.Destination.add(mapAddressToRange(address, MachineType:L2Cache,
l2_select_low_bit, l2_select_num_bits));
if (cache_entry.Dirty) {
out_msg.MessageSize := MessageSizeType:Writeback_Data;
} else {
out_msg.MessageSize := MessageSizeType:Writeback_Control;
}
}
}
action(j_sendUnblock, "j", desc="send unblock to the L2 cache") {
enqueue(unblockNetwork_out, ResponseMsg, latency=to_l2_latency) {
out_msg.Address := address;
out_msg.Type := CoherenceResponseType:UNBLOCK;
out_msg.Sender := machineID;
out_msg.Destination.add(mapAddressToRange(address, MachineType:L2Cache,
l2_select_low_bit, l2_select_num_bits));
out_msg.MessageSize := MessageSizeType:Response_Control;
DPRINTF(RubySlicc, "%s\n", address);
}
}
action(jj_sendExclusiveUnblock, "\j", desc="send unblock to the L2 cache") {
enqueue(unblockNetwork_out, ResponseMsg, latency=to_l2_latency) {
out_msg.Address := address;
out_msg.Type := CoherenceResponseType:EXCLUSIVE_UNBLOCK;
out_msg.Sender := machineID;
out_msg.Destination.add(mapAddressToRange(address, MachineType:L2Cache,
l2_select_low_bit, l2_select_num_bits));
out_msg.MessageSize := MessageSizeType:Response_Control;
DPRINTF(RubySlicc, "%s\n", address);
}
}
action(h_load_hit, "h", desc="If not prefetch, notify sequencer the load completed.") {
assert(is_valid(cache_entry));
DPRINTF(RubySlicc, "%s\n", cache_entry.DataBlk);
sequencer.readCallback(address, cache_entry.DataBlk);
}
action(hh_store_hit, "\h", desc="If not prefetch, notify sequencer that store completed.") {
assert(is_valid(cache_entry));
DPRINTF(RubySlicc, "%s\n", cache_entry.DataBlk);
sequencer.writeCallback(address, cache_entry.DataBlk);
cache_entry.Dirty := true;
}
action(i_allocateTBE, "i", desc="Allocate TBE (isPrefetch=0, number of invalidates=0)") {
check_allocate(L1_TBEs);
assert(is_valid(cache_entry));
L1_TBEs.allocate(address);
set_tbe(L1_TBEs[address]);
tbe.isPrefetch := false;
tbe.Dirty := cache_entry.Dirty;
tbe.DataBlk := cache_entry.DataBlk;
}
action(k_popMandatoryQueue, "k", desc="Pop mandatory queue.") {
mandatoryQueue_in.dequeue();
}
action(l_popRequestQueue, "l", desc="Pop incoming request queue and profile the delay within this virtual network") {
profileMsgDelay(2, requestIntraChipL1Network_in.dequeue_getDelayCycles());
}
action(o_popIncomingResponseQueue, "o", desc="Pop Incoming Response queue and profile the delay within this virtual network") {
profileMsgDelay(3, responseIntraChipL1Network_in.dequeue_getDelayCycles());
}
action(s_deallocateTBE, "s", desc="Deallocate TBE") {
L1_TBEs.deallocate(address);
unset_tbe();
}
action(u_writeDataToL1Cache, "u", desc="Write data to cache") {
peek(responseIntraChipL1Network_in, ResponseMsg) {
assert(is_valid(cache_entry));
cache_entry.DataBlk := in_msg.DataBlk;
cache_entry.Dirty := in_msg.Dirty;
}
}
action(q_updateAckCount, "q", desc="Update ack count") {
peek(responseIntraChipL1Network_in, ResponseMsg) {
assert(is_valid(tbe));
tbe.pendingAcks := tbe.pendingAcks - in_msg.AckCount;
APPEND_TRANSITION_COMMENT(in_msg.AckCount);
APPEND_TRANSITION_COMMENT(" p: ");
APPEND_TRANSITION_COMMENT(tbe.pendingAcks);
}
}
action(z_stall, "z", desc="Stall") {
}
action(ff_deallocateL1CacheBlock, "\f", desc="Deallocate L1 cache block. Sets the cache to not present, allowing a replacement in parallel with a fetch.") {
if (L1DcacheMemory.isTagPresent(address)) {
L1DcacheMemory.deallocate(address);
} else {
L1IcacheMemory.deallocate(address);
}
unset_cache_entry();
}
action(oo_allocateL1DCacheBlock, "\o", desc="Set L1 D-cache tag equal to tag of block B.") {
if (is_invalid(cache_entry)) {
set_cache_entry(L1DcacheMemory.allocate(address, new Entry));
}
}
action(pp_allocateL1ICacheBlock, "\p", desc="Set L1 I-cache tag equal to tag of block B.") {
if (is_invalid(cache_entry)) {
set_cache_entry(L1IcacheMemory.allocate(address, new Entry));
}
}
action(zz_recycleRequestQueue, "zz", desc="recycle L1 request queue") {
requestIntraChipL1Network_in.recycle();
}
action(z_recycleMandatoryQueue, "\z", desc="recycle L1 request queue") {
mandatoryQueue_in.recycle();
}
action(uu_profileInstMiss, "\ui", desc="Profile the demand miss") {
peek(mandatoryQueue_in, RubyRequest) {
L1IcacheMemory.profileMiss(in_msg);
}
}
action(uu_profileDataMiss, "\ud", desc="Profile the demand miss") {
peek(mandatoryQueue_in, RubyRequest) {
L1DcacheMemory.profileMiss(in_msg);
}
}
//*****************************************************
// TRANSITIONS
//*****************************************************
// Transitions for Load/Store/Replacement/WriteBack from transient states
transition({IS, IM, IS_I, M_I, SM}, {Load, Ifetch, Store, L1_Replacement}) {
z_recycleMandatoryQueue;
}
// Transitions from Idle
transition({NP,I}, L1_Replacement) {
ff_deallocateL1CacheBlock;
}
transition({NP,I}, Load, IS) {
oo_allocateL1DCacheBlock;
i_allocateTBE;
a_issueGETS;
uu_profileDataMiss;
k_popMandatoryQueue;
}
transition({NP,I}, Ifetch, IS) {
pp_allocateL1ICacheBlock;
i_allocateTBE;
ai_issueGETINSTR;
uu_profileInstMiss;
k_popMandatoryQueue;
}
transition({NP,I}, Store, IM) {
oo_allocateL1DCacheBlock;
i_allocateTBE;
b_issueGETX;
uu_profileDataMiss;
k_popMandatoryQueue;
}
transition({NP, I}, Inv) {
fi_sendInvAck;
l_popRequestQueue;
}
// Transitions from Shared
transition(S, {Load,Ifetch}) {
h_load_hit;
k_popMandatoryQueue;
}
transition(S, Store, SM) {
i_allocateTBE;
c_issueUPGRADE;
uu_profileDataMiss;
k_popMandatoryQueue;
}
transition(S, L1_Replacement, I) {
forward_eviction_to_cpu;
ff_deallocateL1CacheBlock;
}
transition(S, Inv, I) {
forward_eviction_to_cpu;
fi_sendInvAck;
l_popRequestQueue;
}
// Transitions from Exclusive
transition(E, {Load, Ifetch}) {
h_load_hit;
k_popMandatoryQueue;
}
transition(E, Store, M) {
hh_store_hit;
k_popMandatoryQueue;
}
transition(E, L1_Replacement, M_I) {
// silent E replacement??
forward_eviction_to_cpu;
i_allocateTBE;
g_issuePUTX; // send data, but hold in case forwarded request
ff_deallocateL1CacheBlock;
}
transition(E, Inv, I) {
// don't send data
forward_eviction_to_cpu;
fi_sendInvAck;
l_popRequestQueue;
}
transition(E, Fwd_GETX, I) {
forward_eviction_to_cpu;
d_sendDataToRequestor;
l_popRequestQueue;
}
transition(E, {Fwd_GETS, Fwd_GET_INSTR}, S) {
d_sendDataToRequestor;
d2_sendDataToL2;
l_popRequestQueue;
}
// Transitions from Modified
transition(M, {Load, Ifetch}) {
h_load_hit;
k_popMandatoryQueue;
}
transition(M, Store) {
hh_store_hit;
k_popMandatoryQueue;
}
transition(M, L1_Replacement, M_I) {
forward_eviction_to_cpu;
i_allocateTBE;
g_issuePUTX; // send data, but hold in case forwarded request
ff_deallocateL1CacheBlock;
}
transition(M_I, WB_Ack, I) {
s_deallocateTBE;
o_popIncomingResponseQueue;
}
transition(M, Inv, I) {
forward_eviction_to_cpu;
f_sendDataToL2;
l_popRequestQueue;
}
transition(M_I, Inv, SINK_WB_ACK) {
ft_sendDataToL2_fromTBE;
l_popRequestQueue;
}
transition(M, Fwd_GETX, I) {
forward_eviction_to_cpu;
d_sendDataToRequestor;
l_popRequestQueue;
}
transition(M, {Fwd_GETS, Fwd_GET_INSTR}, S) {
d_sendDataToRequestor;
d2_sendDataToL2;
l_popRequestQueue;
}
transition(M_I, Fwd_GETX, SINK_WB_ACK) {
dt_sendDataToRequestor_fromTBE;
l_popRequestQueue;
}
transition(M_I, {Fwd_GETS, Fwd_GET_INSTR}, SINK_WB_ACK) {
dt_sendDataToRequestor_fromTBE;
d2t_sendDataToL2_fromTBE;
l_popRequestQueue;
}
// Transitions from IS
transition({IS, IS_I}, Inv, IS_I) {
fi_sendInvAck;
l_popRequestQueue;
}
transition(IS, Data_all_Acks, S) {
u_writeDataToL1Cache;
h_load_hit;
s_deallocateTBE;
o_popIncomingResponseQueue;
}
transition(IS_I, Data_all_Acks, I) {
u_writeDataToL1Cache;
h_load_hit;
s_deallocateTBE;
o_popIncomingResponseQueue;
}
transition(IS, DataS_fromL1, S) {
u_writeDataToL1Cache;
j_sendUnblock;
h_load_hit;
s_deallocateTBE;
o_popIncomingResponseQueue;
}
transition(IS_I, DataS_fromL1, I) {
u_writeDataToL1Cache;
j_sendUnblock;
h_load_hit;
s_deallocateTBE;
o_popIncomingResponseQueue;
}
// directory is blocked when sending exclusive data
transition(IS_I, Data_Exclusive, E) {
u_writeDataToL1Cache;
h_load_hit;
jj_sendExclusiveUnblock;
s_deallocateTBE;
o_popIncomingResponseQueue;
}
transition(IS, Data_Exclusive, E) {
u_writeDataToL1Cache;
h_load_hit;
jj_sendExclusiveUnblock;
s_deallocateTBE;
o_popIncomingResponseQueue;
}
// Transitions from IM
transition({IM, SM}, Inv, IM) {
fi_sendInvAck;
l_popRequestQueue;
}
transition(IM, Data, SM) {
u_writeDataToL1Cache;
q_updateAckCount;
o_popIncomingResponseQueue;
}
transition(IM, Data_all_Acks, M) {
u_writeDataToL1Cache;
hh_store_hit;
jj_sendExclusiveUnblock;
s_deallocateTBE;
o_popIncomingResponseQueue;
}
// transitions from SM
transition({SM, IM}, Ack) {
q_updateAckCount;
o_popIncomingResponseQueue;
}
transition(SM, Ack_all, M) {
jj_sendExclusiveUnblock;
hh_store_hit;
s_deallocateTBE;
o_popIncomingResponseQueue;
}
transition(SINK_WB_ACK, {Load, Store, Ifetch, L1_Replacement}){
z_recycleMandatoryQueue;
}
transition(SINK_WB_ACK, Inv){
fi_sendInvAck;
l_popRequestQueue;
}
transition(SINK_WB_ACK, WB_Ack){
s_deallocateTBE;
o_popIncomingResponseQueue;
}
}
|