-
Notifications
You must be signed in to change notification settings - Fork 97
/
prdfP9Mca.C
1560 lines (1333 loc) · 56.7 KB
/
prdfP9Mca.C
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
/* IBM_PROLOG_BEGIN_TAG */
/* This is an automatically generated prolog. */
/* */
/* $Source: src/usr/diag/prdf/plat/mem/prdfP9Mca.C $ */
/* */
/* OpenPOWER HostBoot Project */
/* */
/* Contributors Listed Below - COPYRIGHT 2016,2020 */
/* [+] International Business Machines Corp. */
/* */
/* */
/* Licensed under the Apache License, Version 2.0 (the "License"); */
/* you may not use this file except in compliance with the License. */
/* You may obtain a copy of the License at */
/* */
/* http://www.apache.org/licenses/LICENSE-2.0 */
/* */
/* Unless required by applicable law or agreed to in writing, software */
/* distributed under the License is distributed on an "AS IS" BASIS, */
/* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or */
/* implied. See the License for the specific language governing */
/* permissions and limitations under the License. */
/* */
/* IBM_PROLOG_END_TAG */
// Framework includes
#include <iipServiceDataCollector.h>
#include <prdfExtensibleChip.H>
#include <prdfPluginMap.H>
// Platform includes
#include <prdfMemDbUtils.H>
#include <prdfMemEccAnalysis.H>
#include <prdfP9McbistDataBundle.H>
#include <prdfPlatServices.H>
#ifdef __HOSTBOOT_RUNTIME
#include <prdfMemTps.H>
#endif
#ifdef CONFIG_NVDIMM
#include <nvdimm.H>
#endif
using namespace TARGETING;
namespace PRDF
{
using namespace PlatServices;
namespace nimbus_mca
{
//##############################################################################
//
// Special plugins
//
//##############################################################################
/**
* @brief Plugin function called after analysis is complete but before PRD
* exits.
* @param i_chip An MCA chip.
* @param io_sc The step code data struct.
* @note This is especially useful for any analysis that still needs to be
* done after the framework clears the FIR bits that were at attention.
* @return SUCCESS.
*/
int32_t PostAnalysis( ExtensibleChip * i_chip, STEP_CODE_DATA_STRUCT & io_sc )
{
#define PRDF_FUNC "[nimbus_mca::PostAnalysis] "
#ifdef __HOSTBOOT_RUNTIME
// If the IUE threshold in our data bundle has been reached, we trigger
// a port fail. Once we trigger the port fail, the system may crash
// right away. Since PRD is running in the hypervisor, it is possible we
// may not get the error log. To better our chances, we trigger the port
// fail here after the error log has been committed.
if ( MemEcc::queryIueTh<TYPE_MCA>(i_chip, io_sc) )
{
if ( SUCCESS != MemEcc::triggerPortFail<TYPE_MCA>(i_chip) )
{
PRDF_ERR( PRDF_FUNC "triggerPortFail(0x%08x) failed",
i_chip->getHuid() );
}
}
#endif // __HOSTBOOT_RUNTIME
return SUCCESS; // Always return SUCCESS for this plugin.
#undef PRDF_FUNC
}
PRDF_PLUGIN_DEFINE( nimbus_mca, PostAnalysis );
//##############################################################################
//
// MCACALFIR
//
//##############################################################################
/**
* @brief MCACALFIR[4] - RCD Parity Error.
* @param i_mcaChip A P9 MCA chip.
* @param io_sc The step code data struct.
* @return SUCCESS
*/
int32_t RcdParityError( ExtensibleChip * i_mcaChip,
STEP_CODE_DATA_STRUCT & io_sc )
{
#define PRDF_FUNC "[nimbus_mca::RcdParityError] "
// The callouts have already been made in the rule code. All other actions
// documented below.
// Nothing more to do if this is a checkstop attention.
if ( CHECK_STOP == io_sc.service_data->getPrimaryAttnType() )
return SUCCESS;
uint32_t l_rc = SUCCESS;
// If MCBISTFIR[3] is found to be on at the same time, mask it so it won't
// be logged as a separate event.
ExtensibleChip * mcbChip = getConnectedParent( i_mcaChip, TYPE_MCBIST );
SCAN_COMM_REGISTER_CLASS * mcbistfir = mcbChip->getRegister( "MCBISTFIR" );
l_rc = mcbistfir->Read();
if ( SUCCESS != l_rc )
{
PRDF_ERR( PRDF_FUNC "Read() failed on MCBISTFIR");
}
else if ( mcbistfir->IsBitSet(3) )
{
SCAN_COMM_REGISTER_CLASS * mcbistfir_mask_or =
mcbChip->getRegister( "MCBISTFIR_MASK_OR" );
mcbistfir_mask_or->SetBit(3);
l_rc = mcbistfir_mask_or->Write();
if ( SUCCESS != l_rc )
{
PRDF_ERR( PRDF_FUNC "Write() failed on MCBIST_MASK_OR: "
"mcbChip=0x%08x", mcbChip->getHuid() );
}
}
#ifdef __HOSTBOOT_RUNTIME // TPS only supported at runtime.
// Recovery is always enabled during runtime. If the threshold is reached,
// make the error log predictive and start TPS on all slave ranks behind
// the MCA.
if ( getMcaDataBundle(i_mcaChip)->iv_rcdParityTh.inc(io_sc) )
{
io_sc.service_data->setServiceCall();
std::vector<MemRank> list;
getSlaveRanks<TYPE_MCA>( i_mcaChip->getTrgt(), list );
PRDF_ASSERT( !list.empty() ); // target configured with no ranks
for ( auto & r : list )
{
TdEntry * entry = new TpsEvent<TYPE_MCA>( i_mcaChip, r );
MemDbUtils::pushToQueue<TYPE_MCA>( i_mcaChip, entry );
uint32_t rc = MemDbUtils::handleTdEvent<TYPE_MCA>(i_mcaChip, io_sc);
if ( SUCCESS != rc )
{
PRDF_ERR( PRDF_FUNC "handleTdEvent() failed on 0x%08x",
i_mcaChip->getHuid() );
continue; // Try the other ranks.
}
}
}
#else // IPL
SCAN_COMM_REGISTER_CLASS * farb0 = i_mcaChip->getRegister("FARB0");
if ( SUCCESS != farb0->Read() )
{
PRDF_ERR( PRDF_FUNC "Read() failed on MCAECCFIR: i_mcaChip=0x%08x",
i_mcaChip->getHuid() );
// Ensure the reg is zero so that we will use the recovery threshold and
// guarantee we don't try to do a reconfig.
farb0->clearAllBits();
}
if ( farb0->IsBitSet(54) )
{
// Recovery is disabled. Issue a reconfig loop. Make the error log
// predictive if threshold is reached.
if ( rcdParityErrorReconfigLoop(i_mcaChip->getTrgt()) )
io_sc.service_data->setServiceCall();
if ( isInMdiaMode() )
{
SCAN_COMM_REGISTER_CLASS * mask = nullptr;
// Stop any further commands on this MCBIST to avoid subsequent RCD
// errors or potential AUEs.
l_rc = mdiaSendEventMsg( mcbChip->getTrgt(), MDIA::STOP_TESTING );
if ( SUCCESS != l_rc )
{
PRDF_ERR( PRDF_FUNC "mdiaSendEventMsg(STOP_TESTING) failed" );
}
// Mask the maintenance AUE/IAUE attentions on this MCA because they
// are potential side-effects of the RCD parity errors.
mask = i_mcaChip->getRegister( "MCAECCFIR_MASK_OR" );
mask->SetBit(33); // maintenance AUE
mask->SetBit(36); // maintenance IAUE
l_rc = mask->Write();
if ( SUCCESS != l_rc )
{
PRDF_ERR( PRDF_FUNC "Write() failed on MCAECCFIR_MASK_OR: "
"i_mcaChip=0x%08x", i_mcaChip->getHuid() );
}
// Mask the maintenance command complete bits to avoid false
// attentions.
mask = mcbChip->getRegister( "MCBISTFIR_MASK_OR" );
mask->SetBit(10); // Command complete
mask->SetBit(12); // WAT workaround
l_rc = mask->Write();
if ( SUCCESS != l_rc )
{
PRDF_ERR( PRDF_FUNC "Write() failed on MCBISTFIR_MASK_OR: "
"mcbChip=0x%08x", mcbChip->getHuid() );
}
}
}
else
{
// Make the error log predictive if the recovery threshold is reached.
// Don't bother with TPS on all ranks because it is too complicated to
// handle during Memory Diagnostics and we don't have time to complete
// the procedures at any other point during the IPL. The DIMMs will be
// deconfigured during the IPL anyways. So not really much benefit
// except for extra FFDC.
if ( getMcaDataBundle(i_mcaChip)->iv_rcdParityTh.inc(io_sc) )
io_sc.service_data->setServiceCall();
}
#endif
if ( io_sc.service_data->queryServiceCall() )
{
// Mask both RCD parity error bits to prevent any flooding.
SCAN_COMM_REGISTER_CLASS * mask
= i_mcaChip->getRegister( "MCACALFIR_MASK_OR" );
mask->SetBit( 4);
mask->SetBit(14);
if ( SUCCESS != mask->Write() )
{
PRDF_ERR( PRDF_FUNC "Write() failed on MCACALFIR_MASK_OR: "
"i_mcaChip=0x%08x", i_mcaChip->getHuid() );
}
}
return SUCCESS;
#undef PRDF_FUNC
}
PRDF_PLUGIN_DEFINE( nimbus_mca, RcdParityError );
//------------------------------------------------------------------------------
/**
* @brief MCACALFIR[13] - Persistent RCD error, port failed.
* @param i_chip MCA chip.
* @param io_sc The step code data struct.
* @return SUCCESS
*/
int32_t MemPortFailure( ExtensibleChip * i_chip,
STEP_CODE_DATA_STRUCT & io_sc )
{
#define PRDF_FUNC "[nimbus_mca::MemPortFailure] "
if ( CHECK_STOP != io_sc.service_data->getPrimaryAttnType() )
{
// The port is dead. Mask off the entire port.
uint32_t l_rc = MemEcc::maskMemPort<TYPE_MCA>( i_chip );
if ( SUCCESS != l_rc )
{
PRDF_ERR( PRDF_FUNC "MemEcc::maskMemPort<TYPE_MCA>(0x%08x) failed",
i_chip->getHuid() );
}
}
return SUCCESS; // nothing to return to rule code
#undef PRDF_FUNC
}
PRDF_PLUGIN_DEFINE( nimbus_mca, MemPortFailure );
//##############################################################################
//
// NVDIMM
//
//##############################################################################
#ifdef CONFIG_NVDIMM
#ifdef __HOSTBOOT_RUNTIME
/**
* @brief Gets a map list of which bits are set from a uint8_t bit list (7:0)
* @param i_data uint8_t bit list (7:0)
* @return map<uint8_t, bool> with which bits were set in the bit list.
*/
std::map<uint8_t,bool> __nvdimmGetActiveBits( uint8_t i_data )
{
// NOTE: Bit position in i_data that we get from the NVDIMM status register
// will be in descending order, ie ordered 7 to 0 (left to right).
std::map<uint8_t,bool> bitList;
for ( uint8_t n = 0; n < 8; n++ )
{
if ( i_data & (0x01 << n) ) bitList[n] = true;
}
return bitList;
}
/**
* @brief Adds a callout of the NVDIMM backup power module
* @param i_dimm The target dimm.
* @param i_priority The callout priority.
* @return FAIL if unable to get the global error log, else SUCCESS
*/
uint32_t __addBpmCallout( TargetHandle_t i_dimm,
HWAS::callOutPriority i_priority )
{
#define PRDF_FUNC "[__addBpmCallout] "
uint32_t o_rc = SUCCESS;
do
{
errlHndl_t mainErrl = nullptr;
mainErrl = ServiceGeneratorClass::ThisServiceGenerator().getErrl();
if ( nullptr == mainErrl )
{
PRDF_ERR( PRDF_FUNC "Failed to get the global error log." );
o_rc = FAIL;
break;
}
// addPartCallout will default to GARD_NULL, NO_DECONFIG
mainErrl->addPartCallout( i_dimm, HWAS::BPM_PART_TYPE,
i_priority );
}while(0);
return o_rc;
#undef PRDF_FUNC
}
/**
* @brief Adds a callout of the cable connecting an NVDIMM to its
* backup power module (BPM)
* @param i_dimm The target dimm.
* @param i_priority The callout priority.
* @return FAIL if unable to get the global error log, else SUCCESS
*/
uint32_t __addNvdimmCableCallout( TargetHandle_t i_dimm,
HWAS::callOutPriority i_priority )
{
#define PRDF_FUNC "[__addNvdimmCableCallout] "
uint32_t o_rc = SUCCESS;
do
{
errlHndl_t mainErrl = nullptr;
mainErrl = ServiceGeneratorClass::ThisServiceGenerator().getErrl();
if ( nullptr == mainErrl )
{
PRDF_ERR( PRDF_FUNC "Failed to get the global error log." );
o_rc = FAIL;
break;
}
// addPartCallout will default to GARD_NULL, NO_DECONFIG
mainErrl->addPartCallout( i_dimm, HWAS::BPM_CABLE_PART_TYPE,
i_priority );
}while(0);
return o_rc;
#undef PRDF_FUNC
}
/**
* @brief If a previous error has been found, add a signature to the
* multi-signature list, else set the primary signature.
* @param io_sc The step code data struct.
* @param i_trgt The target.
* @param i_errFound Whether an error has already been found or not.
* @param i_sig The signature to be set.
*/
void __addSignature( STEP_CODE_DATA_STRUCT & io_sc, TargetHandle_t i_trgt,
bool i_errFound, uint32_t i_sig )
{
if ( i_errFound )
{
io_sc.service_data->AddSignatureList( i_trgt, i_sig );
}
else
{
io_sc.service_data->setSignature( getHuid(i_trgt), i_sig );
}
}
/**
* @brief Analyze NVDIMM Health Status0 Register for errors
* @param io_sc The step code data struct.
* @param i_dimm The target dimm.
* @param io_errFound Whether an error has already been found or not.
* @return FAIL if unable to read register, else SUCCESS
*/
uint32_t __analyzeHealthStatus0Reg(STEP_CODE_DATA_STRUCT & io_sc,
TargetHandle_t i_dimm, bool & io_errFound)
{
#define PRDF_FUNC "[__analyzeHealthStatus0Reg] "
uint32_t o_rc = SUCCESS;
uint8_t data = 0;
// Get MCA, for signatures
TargetHandle_t mca = getConnectedParent( i_dimm, TYPE_MCA );
do
{
// NVDIMM health status registers size = 1 byte
size_t NVDIMM_SIZE = 1;
// Read the Health Status0 Register (0xA1) 7:0
errlHndl_t errl = deviceRead( i_dimm, &data, NVDIMM_SIZE,
DEVICE_NVDIMM_ADDRESS(NVDIMM::i2cReg::MODULE_HEALTH_STATUS0) );
if ( errl )
{
PRDF_ERR( PRDF_FUNC "Failed to read Health Status0 Register. "
"HUID: 0x%08x", getHuid(i_dimm) );
PRDF_COMMIT_ERRL( errl, ERRL_ACTION_REPORT );
o_rc = PRD_SCANCOM_FAILURE;
break;
}
std::map<uint8_t,bool> bitList = __nvdimmGetActiveBits( data );
// BIT 0: Voltage Regulator Fail
if ( bitList.count(0) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_VoltRegFail );
// Callout NVDIMM on 1st, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_MED, NO_GARD );
io_errFound = true;
}
// BIT 1: VDD Lost
if ( bitList.count(1) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_VddLost );
// Callout NVDIMM on 1st, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_MED, NO_GARD );
io_errFound = true;
}
// BIT 2: VPP Lost
if ( bitList.count(2) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_VppLost );
// Callout NVDIMM on 1st, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_MED, NO_GARD );
io_errFound = true;
}
// BIT 3: VTT Lost
if ( bitList.count(3) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_VttLost );
// Callout NVDIMM on 1st, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_MED, NO_GARD );
io_errFound = true;
}
// BIT 4: DRAM not Self Refresh
if ( bitList.count(4) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_NotSelfRefr );
// Callout NVDIMM on 1st, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_MED, NO_GARD );
io_errFound = true;
}
// BIT 5: Controller HW Error
if ( bitList.count(5) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_CtrlHwErr );
// Callout NVDIMM on 1st, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_MED, NO_GARD );
io_errFound = true;
}
// BIT 6: NVM Controller Error
if ( bitList.count(6) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_NvmCtrlErr );
// Callout NVDIMM on 1st, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_MED, NO_GARD );
io_errFound = true;
}
// BIT 7: NVM Lifetime Error
if ( bitList.count(7) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_NvmLifeErr );
// Callout NVDIMM on 1st, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_MED, NO_GARD );
io_errFound = true;
}
}while(0);
return o_rc;
#undef PRDF_FUNC
}
/**
* @brief Analyze NVDIMM Health Status1 Register for errors
* @param io_sc The step code data struct.
* @param i_dimm The target dimm.
* @param io_errFound Whether an error has already been found or not.
* @return FAIL if unable to read register, else SUCCESS
*/
uint32_t __analyzeHealthStatus1Reg( STEP_CODE_DATA_STRUCT & io_sc,
TargetHandle_t i_dimm, bool & io_errFound )
{
#define PRDF_FUNC "[__analyzeHealthStatus1Reg] "
uint32_t o_rc = SUCCESS;
uint8_t data = 0;
// Get MCA, for signatures
TargetHandle_t mca = getConnectedParent( i_dimm, TYPE_MCA );
do
{
// NVDIMM health status registers size = 1 byte
size_t NVDIMM_SIZE = 1;
// Read the Health Status1 Register (0xA2) 7:0
errlHndl_t errl = deviceRead( i_dimm, &data, NVDIMM_SIZE,
DEVICE_NVDIMM_ADDRESS(NVDIMM::i2cReg::MODULE_HEALTH_STATUS1) );
if ( errl )
{
PRDF_ERR( PRDF_FUNC "Failed to read Health Status1 Register. "
"HUID: 0x%08x", getHuid(i_dimm) );
PRDF_COMMIT_ERRL( errl, ERRL_ACTION_REPORT );
o_rc = PRD_SCANCOM_FAILURE;
break;
}
std::map<uint8_t,bool> bitList = __nvdimmGetActiveBits( data );
// BIT 0: Insufficient Energy
if ( bitList.count(0) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_InsuffEnergy );
// Callout BPM (backup power module) high, cable high
o_rc = __addBpmCallout( i_dimm, HWAS::SRCI_PRIORITY_HIGH );
if ( SUCCESS != o_rc ) break;
o_rc = __addNvdimmCableCallout( i_dimm, HWAS::SRCI_PRIORITY_HIGH );
if ( SUCCESS != o_rc ) break;
// Callout NVDIMM low, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_LOW, NO_GARD );
io_errFound = true;
}
// BIT 1: Invalid Firmware
if ( bitList.count(1) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_InvFwErr );
// Callout NVDIMM on 1st, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_MED, NO_GARD );
io_errFound = true;
}
// BIT 2: Configuration Data Error
if ( bitList.count(2) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_CnfgDataErr );
// Callout NVDIMM on 1st, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_MED, NO_GARD );
io_errFound = true;
}
// BIT 3: No Energy Source
if ( bitList.count(3) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_NoEsPres );
// Callout BPM (backup power module) high, cable high
o_rc = __addBpmCallout( i_dimm, HWAS::SRCI_PRIORITY_HIGH );
if ( SUCCESS != o_rc ) break;
o_rc = __addNvdimmCableCallout( i_dimm, HWAS::SRCI_PRIORITY_HIGH );
if ( SUCCESS != o_rc ) break;
// Callout NVDIMM low, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_LOW, NO_GARD );
io_errFound = true;
}
// BIT 4: Energy Policy Not Set
if ( bitList.count(4) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_EsPolNotSet );
// Callout FW (Level2 Support) High
io_sc.service_data->SetCallout( LEVEL2_SUPPORT, MRU_HIGH, NO_GARD );
// Callout NVDIMM low on 1st, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_LOW, NO_GARD );
io_errFound = true;
}
// BIT 5: Energy Source HW Error
if ( bitList.count(5) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_EsHwFail );
// Callout BPM (backup power module) high, cable high
o_rc = __addBpmCallout( i_dimm, HWAS::SRCI_PRIORITY_HIGH );
if ( SUCCESS != o_rc ) break;
o_rc = __addNvdimmCableCallout( i_dimm, HWAS::SRCI_PRIORITY_HIGH );
if ( SUCCESS != o_rc ) break;
// Callout NVDIMM low, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_LOW, NO_GARD );
io_errFound = true;
}
// BIT 6: Energy Source Health Assessment Error
if ( bitList.count(6) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_EsHlthAssess);
// Callout BPM (backup power module) high, cable high
o_rc = __addBpmCallout( i_dimm, HWAS::SRCI_PRIORITY_HIGH );
if ( SUCCESS != o_rc ) break;
o_rc = __addNvdimmCableCallout( i_dimm, HWAS::SRCI_PRIORITY_HIGH );
if ( SUCCESS != o_rc ) break;
// Callout NVDIMM low, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_LOW, NO_GARD );
io_errFound = true;
}
// BIT 7: Reserved
}while(0);
return o_rc;
#undef PRDF_FUNC
}
/**
* @brief Reads and merges the data from two ES_TEMP registers to get the
* correct temperature format.
* @param i_dimm The target nvdimm.
* @param i_tempMsbReg The address of the register that contains the most
* significant byte of the temperature data.
* @param i_tempLsbReg The address of the register that contains the least
* significant byte of the temperature data.
* @param o_tempData The 16 bit temperature data.
* @return FAIL if unable to read register, else SUCCESS
*/
uint32_t __readTemp( TargetHandle_t i_dimm, uint16_t i_tempMsbReg,
uint16_t i_tempLsbReg, uint16_t & o_tempData )
{
#define PRDF_FUNC "[__readTemp] "
/*
* -NOTE: Example showing how to read the temperature format:
* ES_TEMP1 = 0x03 (MSB: bits 15-8)
* ES_TEMP0 = 0x48 (LSB: bits 7-0)
*
* 0x0348 = 0000 0011 0100 1000 = 52.5 C
*
* -NOTE: bit definition:
* [15:13]Reserved
* [12]Sign 0 = positive, 1 = negative; 0°C should be expressed as positive
* [11] 128°C
* [10] 64°C
* [9] 32°C
* [8] 16°C
* [7] 8°C
* [6] 4°C
* [5] 2°C
* [4] 1°C
* [3] 0.5°C
* [2] 0.25°C
* [1] 0.125°C Optional for temp fields; not used for temp th fields
* [0]0.0625°C Optional for temp fields; not used for temp th fields
*/
uint32_t o_rc = SUCCESS;
do
{
// NVDIMM health status registers size = 1 byte
size_t NVDIMM_SIZE = 1;
uint8_t msbData = 0;
uint8_t lsbData = 0;
// Read the two inputted temperature registers.
errlHndl_t errl = deviceRead( i_dimm, &msbData, NVDIMM_SIZE,
DEVICE_NVDIMM_ADDRESS(i_tempMsbReg) );
if ( errl )
{
PRDF_ERR( PRDF_FUNC "Failed to read ES Temperature MSB Register. "
"HUID: 0x%08x", getHuid(i_dimm) );
PRDF_COMMIT_ERRL( errl, ERRL_ACTION_REPORT );
o_rc = PRD_SCANCOM_FAILURE;
break;
}
errl = deviceRead( i_dimm, &lsbData, NVDIMM_SIZE,
DEVICE_NVDIMM_ADDRESS(i_tempLsbReg) );
if ( errl )
{
PRDF_ERR( PRDF_FUNC "Failed to read ES Temperature LSB Register. "
"HUID: 0x%08x", getHuid(i_dimm) );
PRDF_COMMIT_ERRL( errl, ERRL_ACTION_REPORT );
o_rc = PRD_SCANCOM_FAILURE;
break;
}
o_tempData = ((uint16_t)msbData << 8) | lsbData;
}while(0);
return o_rc;
#undef PRDF_FUNC
}
/**
* @brief Analyze NVDIMM Error Threshold Status Register for errors
* @param io_sc The step code data struct.
* @param i_dimm The target dimm.
* @param io_errFound Whether an error has already been found or not.
* @param o_esTempErr A flag for whether we hit an ES TEMP error or not.
* @return FAIL if unable to read register, else SUCCESS
*/
uint32_t __analyzeErrorThrStatusReg( STEP_CODE_DATA_STRUCT & io_sc,
TargetHandle_t i_dimm, bool & io_errFound,
bool & o_esTempErr )
{
#define PRDF_FUNC "[__analyzeErrorThrStatusReg] "
uint32_t o_rc = SUCCESS;
uint8_t data = 0;
o_esTempErr = false;
// Get MCA, for signatures
TargetHandle_t mca = getConnectedParent( i_dimm, TYPE_MCA );
do
{
// NVDIMM health status registers size = 1 byte
size_t NVDIMM_SIZE = 1;
// Read the Error Threshold Status Register (0xA5) 7:0
errlHndl_t errl = deviceRead( i_dimm, &data, NVDIMM_SIZE,
DEVICE_NVDIMM_ADDRESS(NVDIMM::i2cReg::ERROR_THRESHOLD_STATUS) );
if ( errl )
{
PRDF_ERR( PRDF_FUNC "Failed to read Error Threshold Status Reg. "
"HUID: 0x%08x", getHuid(i_dimm) );
PRDF_COMMIT_ERRL( errl, ERRL_ACTION_REPORT );
o_rc = PRD_SCANCOM_FAILURE;
break;
}
std::map<uint8_t,bool> bitList = __nvdimmGetActiveBits( data );
// BIT 0: NVM Lifetime Error -- ignore
// BIT 1: ES Lifetime Error
if ( bitList.count(1) )
{
__addSignature( io_sc, mca, io_errFound, PRDFSIG_EsLifeErr );
// Callout BPM (backup power module) high
o_rc = __addBpmCallout( i_dimm, HWAS::SRCI_PRIORITY_HIGH );
if ( SUCCESS != o_rc ) break;
// Callout NVDIMM low, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_LOW, NO_GARD );
io_errFound = true;
}
// BIT 2: ES Temperature Error
if ( bitList.count(2) )
{
// Sleep two seconds to avoid exiting PRD analysis faster than the
// ES_TEMP sample rate.
PlatServices::milliSleep( 2, 0 );
// Read the ES_TEMP and ES_TEMP_ERROR_HIGH_THRESHOLD values
uint16_t msbEsTempReg = NVDIMM::i2cReg::ES_TEMP1;
uint16_t lsbEsTempReg = NVDIMM::i2cReg::ES_TEMP0;
uint16_t esTemp = 0;
o_rc = __readTemp( i_dimm, msbEsTempReg, lsbEsTempReg, esTemp );
if ( SUCCESS != o_rc ) break;
uint16_t msbThReg = NVDIMM::i2cReg::ES_TEMP_ERROR_HIGH_THRESHOLD1;
uint16_t lsbThReg = NVDIMM::i2cReg::ES_TEMP_ERROR_HIGH_THRESHOLD0;
uint16_t esTempHighTh = 0;
o_rc = __readTemp( i_dimm, msbThReg, lsbThReg, esTempHighTh );
if ( SUCCESS != o_rc ) break;
msbThReg = NVDIMM::i2cReg::ES_TEMP_ERROR_LOW_THRESHOLD1;
lsbThReg = NVDIMM::i2cReg::ES_TEMP_ERROR_LOW_THRESHOLD0;
uint16_t esTempLowTh = 0;
o_rc = __readTemp( i_dimm, msbThReg, lsbThReg, esTempLowTh );
if ( SUCCESS != o_rc ) break;
// Check to see if the ES_TEMP is negative (bit 12)
bool esTempNeg = false;
if ( esTemp & 0x1000 ) esTempNeg = true;
// If ES_TEMP is equal or above ES_TEMP_ERROR_HIGH_THRESHOLD
// Just in case ES_TEMP has moved before we read it out, we'll add
// a 2°C margin when comparing to the threshold.
if ( (esTemp >= (esTempHighTh - 0x0020)) && !esTempNeg )
{
__addSignature( io_sc, mca, io_errFound,
PRDFSIG_EsTmpErrHigh );
}
// Else check if the error hit the low threshold, again with the
// same 2°C margin.
else if ( (esTemp <= (esTempLowTh + 0x0020)) || esTempNeg )
{
__addSignature( io_sc, mca, io_errFound,
PRDFSIG_EsTmpErrLow );
}
// Else the temperature must have gone back to a normal value, so
// we will label this as a false alarm case.
else
{
__addSignature( io_sc, mca, io_errFound,
PRDFSIG_EsTmpErrFa );
}
// Callout BPM (backup power module) high
o_rc = __addBpmCallout( i_dimm, HWAS::SRCI_PRIORITY_HIGH );
if ( SUCCESS != o_rc ) break;
// Callout NVDIMM low, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_LOW, NO_GARD );
o_esTempErr = true;
io_errFound = true;
}
// BIT 3:7: Reserved
}while(0);
return o_rc;
#undef PRDF_FUNC
}
/**
* @brief Adjusts the warning threshold so that future warnings are allowed
* to report.
* @param io_sc The step code data struct.
* @param i_dimm The target nvdimm.
* @param i_warnThReg The address of the relevant warning threshold register.
* @param i_errThReg The address of the relevant error threshold register.
* @param o_firstWarn Flag if this is the first warning of this type.
* @param o_statusErr Flag to tell if we found an error from checking the
* notification status register.
* @return FAIL if unable to read register, else SUCCESS
*/
uint32_t __adjustThreshold( STEP_CODE_DATA_STRUCT & io_sc,
TargetHandle_t i_dimm, uint16_t i_warnThReg,
uint16_t i_errThReg, bool & o_firstWarn,
bool & o_statusErr )
{
#define PRDF_FUNC "[__adjustThreshold] "
uint32_t o_rc = SUCCESS;
uint16_t notifCmdReg = NVDIMM::i2cReg::SET_EVENT_NOTIFICATION_CMD;
uint16_t notifStatusReg = NVDIMM::i2cReg::SET_EVENT_NOTIFICATION_STATUS;
o_firstWarn = false;
o_statusErr = false;
do
{
// NVDIMM health status registers size = 1 byte
size_t NVDIMM_SIZE = 1;
// Read the corresponding warning threshold
uint8_t warnTh = 0;
errlHndl_t errl = deviceRead( i_dimm, &warnTh, NVDIMM_SIZE,
DEVICE_NVDIMM_ADDRESS(i_warnThReg) );
if ( errl )
{
PRDF_ERR( PRDF_FUNC "Failed to read Warning Threshold Reg. HUID: "
"0x%08x", getHuid(i_dimm) );
PRDF_COMMIT_ERRL( errl, ERRL_ACTION_REPORT );
o_rc = PRD_SCANCOM_FAILURE;
break;
}
// Read the corresponding error threshold
uint8_t errTh = 0;
errl = deviceRead( i_dimm, &errTh, NVDIMM_SIZE,
DEVICE_NVDIMM_ADDRESS(i_errThReg) );
if ( errl )
{
PRDF_ERR( PRDF_FUNC "Failed to read Error Threshold Reg. HUID: "
"0x%08x", getHuid(i_dimm) );
PRDF_COMMIT_ERRL( errl, ERRL_ACTION_REPORT );
o_rc = PRD_SCANCOM_FAILURE;
break;
}
// If the warning threshold is not set to the error threshold+1,
// move the threshold.
if ( warnTh != (errTh+1) )
{
o_firstWarn = true;
// SET_EVENT_NOTIFICATION_CMD is a write only register that is
// used to change the SET_EVENT_NOTIFICATION_STATUS register.
// The only bits within it that are used are bits 0 and 1, as such
// we can safely set the rest to 0. It is defined as:
// [0]: Persistency Notification
// [1]: Warning Threshold Notification
// [2]: Obsolete
// [3]: Firmware Activation Notification (Not Used)
// [4:7]: Reserved
// Clear SET_EVENT_NOTIFICATION_CMD bit 1 and keep bit 0 set
uint8_t notifCmd = 0x01;
errl = deviceWrite( i_dimm, ¬ifCmd, NVDIMM_SIZE,
DEVICE_NVDIMM_ADDRESS(notifCmdReg) );
if ( errl )
{
PRDF_ERR( PRDF_FUNC "Failed to clear Set Event Notification "
"Cmd Reg. HUID: 0x%08x", getHuid(i_dimm) );
PRDF_COMMIT_ERRL( errl, ERRL_ACTION_REPORT );
o_rc = PRD_SCANCOM_FAILURE;
break;
}
// Check SET_EVENT_NOTIFICATION_STATUS to ensure everything is set
// as we expect and we don't see any errors.
uint8_t notifStat = 0;
errl = deviceRead( i_dimm, ¬ifStat, NVDIMM_SIZE,
DEVICE_NVDIMM_ADDRESS(notifStatusReg) );
if ( errl )
{
PRDF_ERR( PRDF_FUNC "Failed to read Set Event Notification "
"Status Reg. HUID: 0x%08x", getHuid(i_dimm) );
PRDF_COMMIT_ERRL( errl, ERRL_ACTION_REPORT );
o_rc = PRD_SCANCOM_FAILURE;
break;
}
std::map<uint8_t,bool> bitList = __nvdimmGetActiveBits( notifStat );
// if Bit [1]: SET_EVENT_NOTIFICATION_ERROR = 1
// or Bit [2]: PERSISTENCY_ENABLED = 0
// or Bit [3]: WARNING_THRESHOLD_ENABLED = 1
if ( bitList.count(1) || !bitList.count(2) || bitList.count(3) )
{
o_statusErr = true;
// Make the log predictive and mask the fir
io_sc.service_data->SetThresholdMaskId(0);
// Callout the NVDIMM, no gard
io_sc.service_data->SetCallout( i_dimm, MRU_MED, NO_GARD );
// Send message to PHYP that save/restore may work
o_rc = PlatServices::nvdimmNotifyProtChange( i_dimm,
NVDIMM::NVDIMM_RISKY_HW_ERROR );
if ( SUCCESS != o_rc ) break;
break;
}
// Set the warning threshold to error threshold + 1
warnTh = errTh+1;
errl = deviceWrite( i_dimm, &warnTh, NVDIMM_SIZE,
DEVICE_NVDIMM_ADDRESS(i_warnThReg) );
if ( errl )
{
PRDF_ERR( PRDF_FUNC "Failed to write Warning Threshold Reg. "
"HUID: 0x%08x", getHuid(i_dimm) );
PRDF_COMMIT_ERRL( errl, ERRL_ACTION_REPORT );
o_rc = PRD_SCANCOM_FAILURE;
break;
}
// Set SET_EVENT_NOTIFICATION_CMD bit 1 and keep bit 0 set
notifCmd = 0x03;
errl = deviceWrite( i_dimm, ¬ifCmd, NVDIMM_SIZE,