summaryrefslogtreecommitdiffstats
path: root/arch/powerpc/kernel/eeh_driver.c
blob: 438568a472d03bba2ac5db805ad5ae3e13bd1921 (plain)
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
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
// SPDX-License-Identifier: GPL-2.0-or-later
/*
 * PCI Error Recovery Driver for RPA-compliant PPC64 platform.
 * Copyright IBM Corp. 2004 2005
 * Copyright Linas Vepstas <linas@linas.org> 2004, 2005
 *
 * Send comments and feedback to Linas Vepstas <linas@austin.ibm.com>
 */
#include <linux/delay.h>
#include <linux/interrupt.h>
#include <linux/irq.h>
#include <linux/module.h>
#include <linux/pci.h>
#include <linux/pci_hotplug.h>
#include <asm/eeh.h>
#include <asm/eeh_event.h>
#include <asm/ppc-pci.h>
#include <asm/pci-bridge.h>
#include <asm/rtas.h>

struct eeh_rmv_data {
	struct list_head removed_vf_list;
	int removed_dev_count;
};

static int eeh_result_priority(enum pci_ers_result result)
{
	switch (result) {
	case PCI_ERS_RESULT_NONE:
		return 1;
	case PCI_ERS_RESULT_NO_AER_DRIVER:
		return 2;
	case PCI_ERS_RESULT_RECOVERED:
		return 3;
	case PCI_ERS_RESULT_CAN_RECOVER:
		return 4;
	case PCI_ERS_RESULT_DISCONNECT:
		return 5;
	case PCI_ERS_RESULT_NEED_RESET:
		return 6;
	default:
		WARN_ONCE(1, "Unknown pci_ers_result value: %d\n", (int)result);
		return 0;
	}
};

static const char *pci_ers_result_name(enum pci_ers_result result)
{
	switch (result) {
	case PCI_ERS_RESULT_NONE:
		return "none";
	case PCI_ERS_RESULT_CAN_RECOVER:
		return "can recover";
	case PCI_ERS_RESULT_NEED_RESET:
		return "need reset";
	case PCI_ERS_RESULT_DISCONNECT:
		return "disconnect";
	case PCI_ERS_RESULT_RECOVERED:
		return "recovered";
	case PCI_ERS_RESULT_NO_AER_DRIVER:
		return "no AER driver";
	default:
		WARN_ONCE(1, "Unknown result type: %d\n", (int)result);
		return "unknown";
	}
};

static enum pci_ers_result pci_ers_merge_result(enum pci_ers_result old,
						enum pci_ers_result new)
{
	if (eeh_result_priority(new) > eeh_result_priority(old))
		return new;
	return old;
}

static bool eeh_dev_removed(struct eeh_dev *edev)
{
	return !edev || (edev->mode & EEH_DEV_REMOVED);
}

static bool eeh_edev_actionable(struct eeh_dev *edev)
{
	if (!edev->pdev)
		return false;
	if (edev->pdev->error_state == pci_channel_io_perm_failure)
		return false;
	if (eeh_dev_removed(edev))
		return false;
	if (eeh_pe_passed(edev->pe))
		return false;

	return true;
}

/**
 * eeh_pcid_get - Get the PCI device driver
 * @pdev: PCI device
 *
 * The function is used to retrieve the PCI device driver for
 * the indicated PCI device. Besides, we will increase the reference
 * of the PCI device driver to prevent that being unloaded on
 * the fly. Otherwise, kernel crash would be seen.
 */
static inline struct pci_driver *eeh_pcid_get(struct pci_dev *pdev)
{
	if (!pdev || !pdev->dev.driver)
		return NULL;

	if (!try_module_get(pdev->dev.driver->owner))
		return NULL;

	return to_pci_driver(pdev->dev.driver);
}

/**
 * eeh_pcid_put - Dereference on the PCI device driver
 * @pdev: PCI device
 *
 * The function is called to do dereference on the PCI device
 * driver of the indicated PCI device.
 */
static inline void eeh_pcid_put(struct pci_dev *pdev)
{
	if (!pdev || !pdev->dev.driver)
		return;

	module_put(pdev->dev.driver->owner);
}

/**
 * eeh_disable_irq - Disable interrupt for the recovering device
 * @dev: PCI device
 *
 * This routine must be called when reporting temporary or permanent
 * error to the particular PCI device to disable interrupt of that
 * device. If the device has enabled MSI or MSI-X interrupt, we needn't
 * do real work because EEH should freeze DMA transfers for those PCI
 * devices encountering EEH errors, which includes MSI or MSI-X.
 */
static void eeh_disable_irq(struct eeh_dev *edev)
{
	/* Don't disable MSI and MSI-X interrupts. They are
	 * effectively disabled by the DMA Stopped state
	 * when an EEH error occurs.
	 */
	if (edev->pdev->msi_enabled || edev->pdev->msix_enabled)
		return;

	if (!irq_has_action(edev->pdev->irq))
		return;

	edev->mode |= EEH_DEV_IRQ_DISABLED;
	disable_irq_nosync(edev->pdev->irq);
}

/**
 * eeh_enable_irq - Enable interrupt for the recovering device
 * @dev: PCI device
 *
 * This routine must be called to enable interrupt while failed
 * device could be resumed.
 */
static void eeh_enable_irq(struct eeh_dev *edev)
{
	if ((edev->mode) & EEH_DEV_IRQ_DISABLED) {
		edev->mode &= ~EEH_DEV_IRQ_DISABLED;
		/*
		 * FIXME !!!!!
		 *
		 * This is just ass backwards. This maze has
		 * unbalanced irq_enable/disable calls. So instead of
		 * finding the root cause it works around the warning
		 * in the irq_enable code by conditionally calling
		 * into it.
		 *
		 * That's just wrong.The warning in the core code is
		 * there to tell people to fix their asymmetries in
		 * their own code, not by abusing the core information
		 * to avoid it.
		 *
		 * I so wish that the assymetry would be the other way
		 * round and a few more irq_disable calls render that
		 * shit unusable forever.
		 *
		 *	tglx
		 */
		if (irqd_irq_disabled(irq_get_irq_data(edev->pdev->irq)))
			enable_irq(edev->pdev->irq);
	}
}

static void eeh_dev_save_state(struct eeh_dev *edev, void *userdata)
{
	struct pci_dev *pdev;

	if (!edev)
		return;

	/*
	 * We cannot access the config space on some adapters.
	 * Otherwise, it will cause fenced PHB. We don't save
	 * the content in their config space and will restore
	 * from the initial config space saved when the EEH
	 * device is created.
	 */
	if (edev->pe && (edev->pe->state & EEH_PE_CFG_RESTRICTED))
		return;

	pdev = eeh_dev_to_pci_dev(edev);
	if (!pdev)
		return;

	pci_save_state(pdev);
}

static void eeh_set_channel_state(struct eeh_pe *root, pci_channel_state_t s)
{
	struct eeh_pe *pe;
	struct eeh_dev *edev, *tmp;

	eeh_for_each_pe(root, pe)
		eeh_pe_for_each_dev(pe, edev, tmp)
			if (eeh_edev_actionable(edev))
				edev->pdev->error_state = s;
}

static void eeh_set_irq_state(struct eeh_pe *root, bool enable)
{
	struct eeh_pe *pe;
	struct eeh_dev *edev, *tmp;

	eeh_for_each_pe(root, pe) {
		eeh_pe_for_each_dev(pe, edev, tmp) {
			if (!eeh_edev_actionable(edev))
				continue;

			if (!eeh_pcid_get(edev->pdev))
				continue;

			if (enable)
				eeh_enable_irq(edev);
			else
				eeh_disable_irq(edev);

			eeh_pcid_put(edev->pdev);
		}
	}
}

typedef enum pci_ers_result (*eeh_report_fn)(struct eeh_dev *,
					     struct pci_dev *,
					     struct pci_driver *);
static void eeh_pe_report_edev(struct eeh_dev *edev, eeh_report_fn fn,
			       enum pci_ers_result *result)
{
	struct pci_dev *pdev;
	struct pci_driver *driver;
	enum pci_ers_result new_result;

	pci_lock_rescan_remove();
	pdev = edev->pdev;
	if (pdev)
		get_device(&pdev->dev);
	pci_unlock_rescan_remove();
	if (!pdev) {
		eeh_edev_info(edev, "no device");
		return;
	}
	device_lock(&pdev->dev);
	if (eeh_edev_actionable(edev)) {
		driver = eeh_pcid_get(pdev);

		if (!driver)
			eeh_edev_info(edev, "no driver");
		else if (!driver->err_handler)
			eeh_edev_info(edev, "driver not EEH aware");
		else if (edev->mode & EEH_DEV_NO_HANDLER)
			eeh_edev_info(edev, "driver bound too late");
		else {
			new_result = fn(edev, pdev, driver);
			eeh_edev_info(edev, "%s driver reports: '%s'",
				      driver->name,
				      pci_ers_result_name(new_result));
			if (result)
				*result = pci_ers_merge_result(*result,
							       new_result);
		}
		if (driver)
			eeh_pcid_put(pdev);
	} else {
		eeh_edev_info(edev, "not actionable (%d,%d,%d)", !!pdev,
			      !eeh_dev_removed(edev), !eeh_pe_passed(edev->pe));
	}
	device_unlock(&pdev->dev);
	if (edev->pdev != pdev)
		eeh_edev_warn(edev, "Device changed during processing!\n");
	put_device(&pdev->dev);
}

static void eeh_pe_report(const char *name, struct eeh_pe *root,
			  eeh_report_fn fn, enum pci_ers_result *result)
{
	struct eeh_pe *pe;
	struct eeh_dev *edev, *tmp;

	pr_info("EEH: Beginning: '%s'\n", name);
	eeh_for_each_pe(root, pe) eeh_pe_for_each_dev(pe, edev, tmp)
		eeh_pe_report_edev(edev, fn, result);
	if (result)
		pr_info("EEH: Finished:'%s' with aggregate recovery state:'%s'\n",
			name, pci_ers_result_name(*result));
	else
		pr_info("EEH: Finished:'%s'", name);
}

/**
 * eeh_report_error - Report pci error to each device driver
 * @edev: eeh device
 * @driver: device's PCI driver
 *
 * Report an EEH error to each device driver.
 */
static enum pci_ers_result eeh_report_error(struct eeh_dev *edev,
					    struct pci_dev *pdev,
					    struct pci_driver *driver)
{
	enum pci_ers_result rc;

	if (!driver->err_handler->error_detected)
		return PCI_ERS_RESULT_NONE;

	eeh_edev_info(edev, "Invoking %s->error_detected(IO frozen)",
		      driver->name);
	rc = driver->err_handler->error_detected(pdev, pci_channel_io_frozen);

	edev->in_error = true;
	pci_uevent_ers(pdev, PCI_ERS_RESULT_NONE);
	return rc;
}

/**
 * eeh_report_mmio_enabled - Tell drivers that MMIO has been enabled
 * @edev: eeh device
 * @driver: device's PCI driver
 *
 * Tells each device driver that IO ports, MMIO and config space I/O
 * are now enabled.
 */
static enum pci_ers_result eeh_report_mmio_enabled(struct eeh_dev *edev,
						   struct pci_dev *pdev,
						   struct pci_driver *driver)
{
	if (!driver->err_handler->mmio_enabled)
		return PCI_ERS_RESULT_NONE;
	eeh_edev_info(edev, "Invoking %s->mmio_enabled()", driver->name);
	return driver->err_handler->mmio_enabled(pdev);
}

/**
 * eeh_report_reset - Tell device that slot has been reset
 * @edev: eeh device
 * @driver: device's PCI driver
 *
 * This routine must be called while EEH tries to reset particular
 * PCI device so that the associated PCI device driver could take
 * some actions, usually to save data the driver needs so that the
 * driver can work again while the device is recovered.
 */
static enum pci_ers_result eeh_report_reset(struct eeh_dev *edev,
					    struct pci_dev *pdev,
					    struct pci_driver *driver)
{
	if (!driver->err_handler->slot_reset || !edev->in_error)
		return PCI_ERS_RESULT_NONE;
	eeh_edev_info(edev, "Invoking %s->slot_reset()", driver->name);
	return driver->err_handler->slot_reset(pdev);
}

static void eeh_dev_restore_state(struct eeh_dev *edev, void *userdata)
{
	struct pci_dev *pdev;

	if (!edev)
		return;

	/*
	 * The content in the config space isn't saved because
	 * the blocked config space on some adapters. We have
	 * to restore the initial saved config space when the
	 * EEH device is created.
	 */
	if (edev->pe && (edev->pe->state & EEH_PE_CFG_RESTRICTED)) {
		if (list_is_last(&edev->entry, &edev->pe->edevs))
			eeh_pe_restore_bars(edev->pe);

		return;
	}

	pdev = eeh_dev_to_pci_dev(edev);
	if (!pdev)
		return;

	pci_restore_state(pdev);
}

/**
 * eeh_report_resume - Tell device to resume normal operations
 * @edev: eeh device
 * @driver: device's PCI driver
 *
 * This routine must be called to notify the device driver that it
 * could resume so that the device driver can do some initialization
 * to make the recovered device work again.
 */
static enum pci_ers_result eeh_report_resume(struct eeh_dev *edev,
					     struct pci_dev *pdev,
					     struct pci_driver *driver)
{
	if (!driver->err_handler->resume || !edev->in_error)
		return PCI_ERS_RESULT_NONE;

	eeh_edev_info(edev, "Invoking %s->resume()", driver->name);
	driver->err_handler->resume(pdev);

	pci_uevent_ers(edev->pdev, PCI_ERS_RESULT_RECOVERED);
#ifdef CONFIG_PCI_IOV
	if (eeh_ops->notify_resume)
		eeh_ops->notify_resume(edev);
#endif
	return PCI_ERS_RESULT_NONE;
}

/**
 * eeh_report_failure - Tell device driver that device is dead.
 * @edev: eeh device
 * @driver: device's PCI driver
 *
 * This informs the device driver that the device is permanently
 * dead, and that no further recovery attempts will be made on it.
 */
static enum pci_ers_result eeh_report_failure(struct eeh_dev *edev,
					      struct pci_dev *pdev,
					      struct pci_driver *driver)
{
	enum pci_ers_result rc;

	if (!driver->err_handler->error_detected)
		return PCI_ERS_RESULT_NONE;

	eeh_edev_info(edev, "Invoking %s->error_detected(permanent failure)",
		      driver->name);
	rc = driver->err_handler->error_detected(pdev,
						 pci_channel_io_perm_failure);

	pci_uevent_ers(pdev, PCI_ERS_RESULT_DISCONNECT);
	return rc;
}

static void *eeh_add_virt_device(struct eeh_dev *edev)
{
	struct pci_driver *driver;
	struct pci_dev *dev = eeh_dev_to_pci_dev(edev);

	if (!(edev->physfn)) {
		eeh_edev_warn(edev, "Not for VF\n");
		return NULL;
	}

	driver = eeh_pcid_get(dev);
	if (driver) {
		if (driver->err_handler) {
			eeh_pcid_put(dev);
			return NULL;
		}
		eeh_pcid_put(dev);
	}

#ifdef CONFIG_PCI_IOV
	pci_iov_add_virtfn(edev->physfn, edev->vf_index);
#endif
	return NULL;
}

static void eeh_rmv_device(struct eeh_dev *edev, void *userdata)
{
	struct pci_driver *driver;
	struct pci_dev *dev = eeh_dev_to_pci_dev(edev);
	struct eeh_rmv_data *rmv_data = (struct eeh_rmv_data *)userdata;

	/*
	 * Actually, we should remove the PCI bridges as well.
	 * However, that's lots of complexity to do that,
	 * particularly some of devices under the bridge might
	 * support EEH. So we just care about PCI devices for
	 * simplicity here.
	 */
	if (!eeh_edev_actionable(edev) ||
	    (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE))
		return;

	if (rmv_data) {
		driver = eeh_pcid_get(dev);
		if (driver) {
			if (driver->err_handler &&
			    driver->err_handler->error_detected &&
			    driver->err_handler->slot_reset) {
				eeh_pcid_put(dev);
				return;
			}
			eeh_pcid_put(dev);
		}
	}

	/* Remove it from PCI subsystem */
	pr_info("EEH: Removing %s without EEH sensitive driver\n",
		pci_name(dev));
	edev->mode |= EEH_DEV_DISCONNECTED;
	if (rmv_data)
		rmv_data->removed_dev_count++;

	if (edev->physfn) {
#ifdef CONFIG_PCI_IOV
		pci_iov_remove_virtfn(edev->physfn, edev->vf_index);
		edev->pdev = NULL;
#endif
		if (rmv_data)
			list_add(&edev->rmv_entry, &rmv_data->removed_vf_list);
	} else {
		pci_lock_rescan_remove();
		pci_stop_and_remove_bus_device(dev);
		pci_unlock_rescan_remove();
	}
}

static void *eeh_pe_detach_dev(struct eeh_pe *pe, void *userdata)
{
	struct eeh_dev *edev, *tmp;

	eeh_pe_for_each_dev(pe, edev, tmp) {
		if (!(edev->mode & EEH_DEV_DISCONNECTED))
			continue;

		edev->mode &= ~(EEH_DEV_DISCONNECTED | EEH_DEV_IRQ_DISABLED);
		eeh_pe_tree_remove(edev);
	}

	return NULL;
}

/*
 * Explicitly clear PE's frozen state for PowerNV where
 * we have frozen PE until BAR restore is completed. It's
 * harmless to clear it for pSeries. To be consistent with
 * PE reset (for 3 times), we try to clear the frozen state
 * for 3 times as well.
 */
static int eeh_clear_pe_frozen_state(struct eeh_pe *root, bool include_passed)
{
	struct eeh_pe *pe;
	int i;

	eeh_for_each_pe(root, pe) {
		if (include_passed || !eeh_pe_passed(pe)) {
			for (i = 0; i < 3; i++)
				if (!eeh_unfreeze_pe(pe))
					break;
			if (i >= 3)
				return -EIO;
		}
	}
	eeh_pe_state_clear(root, EEH_PE_ISOLATED, include_passed);
	return 0;
}

int eeh_pe_reset_and_recover(struct eeh_pe *pe)
{
	int ret;

	/* Bail if the PE is being recovered */
	if (pe->state & EEH_PE_RECOVERING)
		return 0;

	/* Put the PE into recovery mode */
	eeh_pe_state_mark(pe, EEH_PE_RECOVERING);

	/* Save states */
	eeh_pe_dev_traverse(pe, eeh_dev_save_state, NULL);

	/* Issue reset */
	ret = eeh_pe_reset_full(pe, true);
	if (ret) {
		eeh_pe_state_clear(pe, EEH_PE_RECOVERING, true);
		return ret;
	}

	/* Unfreeze the PE */
	ret = eeh_clear_pe_frozen_state(pe, true);
	if (ret) {
		eeh_pe_state_clear(pe, EEH_PE_RECOVERING, true);
		return ret;
	}

	/* Restore device state */
	eeh_pe_dev_traverse(pe, eeh_dev_restore_state, NULL);

	/* Clear recovery mode */
	eeh_pe_state_clear(pe, EEH_PE_RECOVERING, true);

	return 0;
}

/**
 * eeh_reset_device - Perform actual reset of a pci slot
 * @driver_eeh_aware: Does the device's driver provide EEH support?
 * @pe: EEH PE
 * @bus: PCI bus corresponding to the isolcated slot
 * @rmv_data: Optional, list to record removed devices
 *
 * This routine must be called to do reset on the indicated PE.
 * During the reset, udev might be invoked because those affected
 * PCI devices will be removed and then added.
 */
static int eeh_reset_device(struct eeh_pe *pe, struct pci_bus *bus,
			    struct eeh_rmv_data *rmv_data,
			    bool driver_eeh_aware)
{
	time64_t tstamp;
	int cnt, rc;
	struct eeh_dev *edev;
	struct eeh_pe *tmp_pe;
	bool any_passed = false;

	eeh_for_each_pe(pe, tmp_pe)
		any_passed |= eeh_pe_passed(tmp_pe);

	/* pcibios will clear the counter; save the value */
	cnt = pe->freeze_count;
	tstamp = pe->tstamp;

	/*
	 * We don't remove the corresponding PE instances because
	 * we need the information afterwords. The attached EEH
	 * devices are expected to be attached soon when calling
	 * into pci_hp_add_devices().
	 */
	eeh_pe_state_mark(pe, EEH_PE_KEEP);
	if (any_passed || driver_eeh_aware || (pe->type & EEH_PE_VF)) {
		eeh_pe_dev_traverse(pe, eeh_rmv_device, rmv_data);
	} else {
		pci_lock_rescan_remove();
		pci_hp_remove_devices(bus);
		pci_unlock_rescan_remove();
	}

	/*
	 * Reset the pci controller. (Asserts RST#; resets config space).
	 * Reconfigure bridges and devices. Don't try to bring the system
	 * up if the reset failed for some reason.
	 *
	 * During the reset, it's very dangerous to have uncontrolled PCI
	 * config accesses. So we prefer to block them. However, controlled
	 * PCI config accesses initiated from EEH itself are allowed.
	 */
	rc = eeh_pe_reset_full(pe, false);
	if (rc)
		return rc;

	pci_lock_rescan_remove();

	/* Restore PE */
	eeh_ops->configure_bridge(pe);
	eeh_pe_restore_bars(pe);

	/* Clear frozen state */
	rc = eeh_clear_pe_frozen_state(pe, false);
	if (rc) {
		pci_unlock_rescan_remove();
		return rc;
	}

	/* Give the system 5 seconds to finish running the user-space
	 * hotplug shutdown scripts, e.g. ifdown for ethernet.  Yes,
	 * this is a hack, but if we don't do this, and try to bring
	 * the device up before the scripts have taken it down,
	 * potentially weird things happen.
	 */
	if (!driver_eeh_aware || rmv_data->removed_dev_count) {
		pr_info("EEH: Sleep 5s ahead of %s hotplug\n",
			(driver_eeh_aware ? "partial" : "complete"));
		ssleep(5);

		/*
		 * The EEH device is still connected with its parent
		 * PE. We should disconnect it so the binding can be
		 * rebuilt when adding PCI devices.
		 */
		edev = list_first_entry(&pe->edevs, struct eeh_dev, entry);
		eeh_pe_traverse(pe, eeh_pe_detach_dev, NULL);
		if (pe->type & EEH_PE_VF) {
			eeh_add_virt_device(edev);
		} else {
			if (!driver_eeh_aware)
				eeh_pe_state_clear(pe, EEH_PE_PRI_BUS, true);
			pci_hp_add_devices(bus);
		}
	}
	eeh_pe_state_clear(pe, EEH_PE_KEEP, true);

	pe->tstamp = tstamp;
	pe->freeze_count = cnt;

	pci_unlock_rescan_remove();
	return 0;
}

/* The longest amount of time to wait for a pci device
 * to come back on line, in seconds.
 */
#define MAX_WAIT_FOR_RECOVERY 300


/* Walks the PE tree after processing an event to remove any stale PEs.
 *
 * NB: This needs to be recursive to ensure the leaf PEs get removed
 * before their parents do. Although this is possible to do recursively
 * we don't since this is easier to read and we need to garantee
 * the leaf nodes will be handled first.
 */
static void eeh_pe_cleanup(struct eeh_pe *pe)
{
	struct eeh_pe *child_pe, *tmp;

	list_for_each_entry_safe(child_pe, tmp, &pe->child_list, child)
		eeh_pe_cleanup(child_pe);

	if (pe->state & EEH_PE_KEEP)
		return;

	if (!(pe->state & EEH_PE_INVALID))
		return;

	if (list_empty(&pe->edevs) && list_empty(&pe->child_list)) {
		list_del(&pe->child);
		kfree(pe);
	}
}

/**
 * eeh_check_slot_presence - Check if a device is still present in a slot
 * @pdev: pci_dev to check
 *
 * This function may return a false positive if we can't determine the slot's
 * presence state. This might happen for PCIe slots if the PE containing
 * the upstream bridge is also frozen, or the bridge is part of the same PE
 * as the device.
 *
 * This shouldn't happen often, but you might see it if you hotplug a PCIe
 * switch.
 */
static bool eeh_slot_presence_check(struct pci_dev *pdev)
{
	const struct hotplug_slot_ops *ops;
	struct pci_slot *slot;
	u8 state;
	int rc;

	if (!pdev)
		return false;

	if (pdev->error_state == pci_channel_io_perm_failure)
		return false;

	slot = pdev->slot;
	if (!slot || !slot->hotplug)
		return true;

	ops = slot->hotplug->ops;
	if (!ops || !ops->get_adapter_status)
		return true;

	/* set the attention indicator while we've got the slot ops */
	if (ops->set_attention_status)
		ops->set_attention_status(slot->hotplug, 1);

	rc = ops->get_adapter_status(slot->hotplug, &state);
	if (rc)
		return true;

	return !!state;
}

static void eeh_clear_slot_attention(struct pci_dev *pdev)
{
	const struct hotplug_slot_ops *ops;
	struct pci_slot *slot;

	if (!pdev)
		return;

	if (pdev->error_state == pci_channel_io_perm_failure)
		return;

	slot = pdev->slot;
	if (!slot || !slot->hotplug)
		return;

	ops = slot->hotplug->ops;
	if (!ops || !ops->set_attention_status)
		return;

	ops->set_attention_status(slot->hotplug, 0);
}

/**
 * eeh_handle_normal_event - Handle EEH events on a specific PE
 * @pe: EEH PE - which should not be used after we return, as it may
 * have been invalidated.
 *
 * Attempts to recover the given PE.  If recovery fails or the PE has failed
 * too many times, remove the PE.
 *
 * While PHB detects address or data parity errors on particular PCI
 * slot, the associated PE will be frozen. Besides, DMA's occurring
 * to wild addresses (which usually happen due to bugs in device
 * drivers or in PCI adapter firmware) can cause EEH error. #SERR,
 * #PERR or other misc PCI-related errors also can trigger EEH errors.
 *
 * Recovery process consists of unplugging the device driver (which
 * generated hotplug events to userspace), then issuing a PCI #RST to
 * the device, then reconfiguring the PCI config space for all bridges
 * & devices under this slot, and then finally restarting the device
 * drivers (which cause a second set of hotplug events to go out to
 * userspace).
 */
void eeh_handle_normal_event(struct eeh_pe *pe)
{
	struct pci_bus *bus;
	struct eeh_dev *edev, *tmp;
	struct eeh_pe *tmp_pe;
	int rc = 0;
	enum pci_ers_result result = PCI_ERS_RESULT_NONE;
	struct eeh_rmv_data rmv_data =
		{LIST_HEAD_INIT(rmv_data.removed_vf_list), 0};
	int devices = 0;

	bus = eeh_pe_bus_get(pe);
	if (!bus) {
		pr_err("%s: Cannot find PCI bus for PHB#%x-PE#%x\n",
			__func__, pe->phb->global_number, pe->addr);
		return;
	}

	/*
	 * When devices are hot-removed we might get an EEH due to
	 * a driver attempting to touch the MMIO space of a removed
	 * device. In this case we don't have a device to recover
	 * so suppress the event if we can't find any present devices.
	 *
	 * The hotplug driver should take care of tearing down the
	 * device itself.
	 */
	eeh_for_each_pe(pe, tmp_pe)
		eeh_pe_for_each_dev(tmp_pe, edev, tmp)
			if (eeh_slot_presence_check(edev->pdev))
				devices++;

	if (!devices) {
		pr_debug("EEH: Frozen PHB#%x-PE#%x is empty!\n",
			pe->phb->global_number, pe->addr);
		goto out; /* nothing to recover */
	}

	/* Log the event */
	if (pe->type & EEH_PE_PHB) {
		pr_err("EEH: Recovering PHB#%x, location: %s\n",
			pe->phb->global_number, eeh_pe_loc_get(pe));
	} else {
		struct eeh_pe *phb_pe = eeh_phb_pe_get(pe->phb);

		pr_err("EEH: Recovering PHB#%x-PE#%x\n",
		       pe->phb->global_number, pe->addr);
		pr_err("EEH: PE location: %s, PHB location: %s\n",
		       eeh_pe_loc_get(pe), eeh_pe_loc_get(phb_pe));
	}

#ifdef CONFIG_STACKTRACE
	/*
	 * Print the saved stack trace now that we've verified there's
	 * something to recover.
	 */
	if (pe->trace_entries) {
		void **ptrs = (void **) pe->stack_trace;
		int i;

		pr_err("EEH: Frozen PHB#%x-PE#%x detected\n",
		       pe->phb->global_number, pe->addr);

		/* FIXME: Use the same format as dump_stack() */
		pr_err("EEH: Call Trace:\n");
		for (i = 0; i < pe->trace_entries; i++)
			pr_err("EEH: [%pK] %pS\n", ptrs[i], ptrs[i]);

		pe->trace_entries = 0;
	}
#endif /* CONFIG_STACKTRACE */

	eeh_for_each_pe(pe, tmp_pe)
		eeh_pe_for_each_dev(tmp_pe, edev, tmp)
			edev->mode &= ~EEH_DEV_NO_HANDLER;

	eeh_pe_update_time_stamp(pe);
	pe->freeze_count++;
	if (pe->freeze_count > eeh_max_freezes) {
		pr_err("EEH: PHB#%x-PE#%x has failed %d times in the last hour and has been permanently disabled.\n",
		       pe->phb->global_number, pe->addr,
		       pe->freeze_count);

		goto recover_failed;
	}

	/* Walk the various device drivers attached to this slot through
	 * a reset sequence, giving each an opportunity to do what it needs
	 * to accomplish the reset.  Each child gets a report of the
	 * status ... if any child can't handle the reset, then the entire
	 * slot is dlpar removed and added.
	 *
	 * When the PHB is fenced, we have to issue a reset to recover from
	 * the error. Override the result if necessary to have partially
	 * hotplug for this case.
	 */
	pr_warn("EEH: This PCI device has failed %d times in the last hour and will be permanently disabled after %d failures.\n",
		pe->freeze_count, eeh_max_freezes);
	pr_info("EEH: Notify device drivers to shutdown\n");
	eeh_set_channel_state(pe, pci_channel_io_frozen);
	eeh_set_irq_state(pe, false);
	eeh_pe_report("error_detected(IO frozen)", pe,
		      eeh_report_error, &result);
	if (result == PCI_ERS_RESULT_DISCONNECT)
		goto recover_failed;

	/*
	 * Error logged on a PHB are always fences which need a full
	 * PHB reset to clear so force that to happen.
	 */
	if ((pe->type & EEH_PE_PHB) && result != PCI_ERS_RESULT_NONE)
		result = PCI_ERS_RESULT_NEED_RESET;

	/* Get the current PCI slot state. This can take a long time,
	 * sometimes over 300 seconds for certain systems.
	 */
	rc = eeh_wait_state(pe, MAX_WAIT_FOR_RECOVERY * 1000);
	if (rc < 0 || rc == EEH_STATE_NOT_SUPPORT) {
		pr_warn("EEH: Permanent failure\n");
		goto recover_failed;
	}

	/* Since rtas may enable MMIO when posting the error log,
	 * don't post the error log until after all dev drivers
	 * have been informed.
	 */
	pr_info("EEH: Collect temporary log\n");
	eeh_slot_error_detail(pe, EEH_LOG_TEMP);

	/* If all device drivers were EEH-unaware, then shut
	 * down all of the device drivers, and hope they
	 * go down willingly, without panicing the system.
	 */
	if (result == PCI_ERS_RESULT_NONE) {
		pr_info("EEH: Reset with hotplug activity\n");
		rc = eeh_reset_device(pe, bus, NULL, false);
		if (rc) {
			pr_warn("%s: Unable to reset, err=%d\n", __func__, rc);
			goto recover_failed;
		}
	}

	/* If all devices reported they can proceed, then re-enable MMIO */
	if (result == PCI_ERS_RESULT_CAN_RECOVER) {
		pr_info("EEH: Enable I/O for affected devices\n");
		rc = eeh_pci_enable(pe, EEH_OPT_THAW_MMIO);
		if (rc < 0)
			goto recover_failed;

		if (rc) {
			result = PCI_ERS_RESULT_NEED_RESET;
		} else {
			pr_info("EEH: Notify device drivers to resume I/O\n");
			eeh_pe_report("mmio_enabled", pe,
				      eeh_report_mmio_enabled, &result);
		}
	}
	if (result == PCI_ERS_RESULT_CAN_RECOVER) {
		pr_info("EEH: Enabled DMA for affected devices\n");
		rc = eeh_pci_enable(pe, EEH_OPT_THAW_DMA);
		if (rc < 0)
			goto recover_failed;

		if (rc) {
			result = PCI_ERS_RESULT_NEED_RESET;
		} else {
			/*
			 * We didn't do PE reset for the case. The PE
			 * is still in frozen state. Clear it before
			 * resuming the PE.
			 */
			eeh_pe_state_clear(pe, EEH_PE_ISOLATED, true);
			result = PCI_ERS_RESULT_RECOVERED;
		}
	}

	/* If any device called out for a reset, then reset the slot */
	if (result == PCI_ERS_RESULT_NEED_RESET) {
		pr_info("EEH: Reset without hotplug activity\n");
		rc = eeh_reset_device(pe, bus, &rmv_data, true);
		if (rc) {
			pr_warn("%s: Cannot reset, err=%d\n", __func__, rc);
			goto recover_failed;
		}

		result = PCI_ERS_RESULT_NONE;
		eeh_set_channel_state(pe, pci_channel_io_normal);
		eeh_set_irq_state(pe, true);
		eeh_pe_report("slot_reset", pe, eeh_report_reset,
			      &result);
	}

	if ((result == PCI_ERS_RESULT_RECOVERED) ||
	    (result == PCI_ERS_RESULT_NONE)) {
		/*
		 * For those hot removed VFs, we should add back them after PF
		 * get recovered properly.
		 */
		list_for_each_entry_safe(edev, tmp, &rmv_data.removed_vf_list,
					 rmv_entry) {
			eeh_add_virt_device(edev);
			list_del(&edev->rmv_entry);
		}

		/* Tell all device drivers that they can resume operations */
		pr_info("EEH: Notify device driver to resume\n");
		eeh_set_channel_state(pe, pci_channel_io_normal);
		eeh_set_irq_state(pe, true);
		eeh_pe_report("resume", pe, eeh_report_resume, NULL);
		eeh_for_each_pe(pe, tmp_pe) {
			eeh_pe_for_each_dev(tmp_pe, edev, tmp) {
				edev->mode &= ~EEH_DEV_NO_HANDLER;
				edev->in_error = false;
			}
		}

		pr_info("EEH: Recovery successful.\n");
		goto out;
	}

recover_failed:
	/*
	 * About 90% of all real-life EEH failures in the field
	 * are due to poorly seated PCI cards. Only 10% or so are
	 * due to actual, failed cards.
	 */
	pr_err("EEH: Unable to recover from failure from PHB#%x-PE#%x.\n"
		"Please try reseating or replacing it\n",
		pe->phb->global_number, pe->addr);

	eeh_slot_error_detail(pe, EEH_LOG_PERM);

	/* Notify all devices that they're about to go down. */
	eeh_set_irq_state(pe, false);
	eeh_pe_report("error_detected(permanent failure)", pe,
		      eeh_report_failure, NULL);
	eeh_set_channel_state(pe, pci_channel_io_perm_failure);

	/* Mark the PE to be removed permanently */
	eeh_pe_state_mark(pe, EEH_PE_REMOVED);

	/*
	 * Shut down the device drivers for good. We mark
	 * all removed devices correctly to avoid access
	 * the their PCI config any more.
	 */
	if (pe->type & EEH_PE_VF) {
		eeh_pe_dev_traverse(pe, eeh_rmv_device, NULL);
		eeh_pe_dev_mode_mark(pe, EEH_DEV_REMOVED);
	} else {
		eeh_pe_state_clear(pe, EEH_PE_PRI_BUS, true);
		eeh_pe_dev_mode_mark(pe, EEH_DEV_REMOVED);

		pci_lock_rescan_remove();
		pci_hp_remove_devices(bus);
		pci_unlock_rescan_remove();
		/* The passed PE should no longer be used */
		return;
	}

out:
	/*
	 * Clean up any PEs without devices. While marked as EEH_PE_RECOVERYING
	 * we don't want to modify the PE tree structure so we do it here.
	 */
	eeh_pe_cleanup(pe);

	/* clear the slot attention LED for all recovered devices */
	eeh_for_each_pe(pe, tmp_pe)
		eeh_pe_for_each_dev(tmp_pe, edev, tmp)
			eeh_clear_slot_attention(edev->pdev);

	eeh_pe_state_clear(pe, EEH_PE_RECOVERING, true);
}

/**
 * eeh_handle_special_event - Handle EEH events without a specific failing PE
 *
 * Called when an EEH event is detected but can't be narrowed down to a
 * specific PE.  Iterates through possible failures and handles them as
 * necessary.
 */
void eeh_handle_special_event(void)
{
	struct eeh_pe *pe, *phb_pe, *tmp_pe;
	struct eeh_dev *edev, *tmp_edev;
	struct pci_bus *bus;
	struct pci_controller *hose;
	unsigned long flags;
	int rc;


	do {
		rc = eeh_ops->next_error(&pe);

		switch (rc) {
		case EEH_NEXT_ERR_DEAD_IOC:
			/* Mark all PHBs in dead state */
			eeh_serialize_lock(&flags);

			/* Purge all events */
			eeh_remove_event(NULL, true);

			list_for_each_entry(hose, &hose_list, list_node) {
				phb_pe = eeh_phb_pe_get(hose);
				if (!phb_pe) continue;

				eeh_pe_mark_isolated(phb_pe);
			}

			eeh_serialize_unlock(flags);

			break;
		case EEH_NEXT_ERR_FROZEN_PE:
		case EEH_NEXT_ERR_FENCED_PHB:
		case EEH_NEXT_ERR_DEAD_PHB:
			/* Mark the PE in fenced state */
			eeh_serialize_lock(&flags);

			/* Purge all events of the PHB */
			eeh_remove_event(pe, true);

			if (rc != EEH_NEXT_ERR_DEAD_PHB)
				eeh_pe_state_mark(pe, EEH_PE_RECOVERING);
			eeh_pe_mark_isolated(pe);

			eeh_serialize_unlock(flags);

			break;
		case EEH_NEXT_ERR_NONE:
			return;
		default:
			pr_warn("%s: Invalid value %d from next_error()\n",
				__func__, rc);
			return;
		}

		/*
		 * For fenced PHB and frozen PE, it's handled as normal
		 * event. We have to remove the affected PHBs for dead
		 * PHB and IOC
		 */
		if (rc == EEH_NEXT_ERR_FROZEN_PE ||
		    rc == EEH_NEXT_ERR_FENCED_PHB) {
			eeh_pe_state_mark(pe, EEH_PE_RECOVERING);
			eeh_handle_normal_event(pe);
		} else {
			eeh_for_each_pe(pe, tmp_pe)
				eeh_pe_for_each_dev(tmp_pe, edev, tmp_edev)
					edev->mode &= ~EEH_DEV_NO_HANDLER;

			/* Notify all devices to be down */
			eeh_pe_state_clear(pe, EEH_PE_PRI_BUS, true);
			eeh_pe_report(
				"error_detected(permanent failure)", pe,
				eeh_report_failure, NULL);
			eeh_set_channel_state(pe, pci_channel_io_perm_failure);

			pci_lock_rescan_remove();
			list_for_each_entry(hose, &hose_list, list_node) {
				phb_pe = eeh_phb_pe_get(hose);
				if (!phb_pe ||
				    !(phb_pe->state & EEH_PE_ISOLATED) ||
				    (phb_pe->state & EEH_PE_RECOVERING))
					continue;

				bus = eeh_pe_bus_get(phb_pe);
				if (!bus) {
					pr_err("%s: Cannot find PCI bus for "
					       "PHB#%x-PE#%x\n",
					       __func__,
					       pe->phb->global_number,
					       pe->addr);
					break;
				}
				pci_hp_remove_devices(bus);
			}
			pci_unlock_rescan_remove();
		}

		/*
		 * If we have detected dead IOC, we needn't proceed
		 * any more since all PHBs would have been removed
		 */
		if (rc == EEH_NEXT_ERR_DEAD_IOC)
			break;
	} while (rc != EEH_NEXT_ERR_NONE);
}