summaryrefslogtreecommitdiffstats
path: root/drivers/irqchip/irq-apple-aic.c
blob: 3759dc36cc8f73f664e10e8ed6c8fd2ceca1754f (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
// SPDX-License-Identifier: GPL-2.0-or-later
/*
 * Copyright The Asahi Linux Contributors
 *
 * Based on irq-lpc32xx:
 *   Copyright 2015-2016 Vladimir Zapolskiy <vz@mleia.com>
 * Based on irq-bcm2836:
 *   Copyright 2015 Broadcom
 */

/*
 * AIC is a fairly simple interrupt controller with the following features:
 *
 * - 896 level-triggered hardware IRQs
 *   - Single mask bit per IRQ
 *   - Per-IRQ affinity setting
 *   - Automatic masking on event delivery (auto-ack)
 *   - Software triggering (ORed with hw line)
 * - 2 per-CPU IPIs (meant as "self" and "other", but they are
 *   interchangeable if not symmetric)
 * - Automatic prioritization (single event/ack register per CPU, lower IRQs =
 *   higher priority)
 * - Automatic masking on ack
 * - Default "this CPU" register view and explicit per-CPU views
 *
 * In addition, this driver also handles FIQs, as these are routed to the same
 * IRQ vector. These are used for Fast IPIs (TODO), the ARMv8 timer IRQs, and
 * performance counters (TODO).
 *
 * Implementation notes:
 *
 * - This driver creates two IRQ domains, one for HW IRQs and internal FIQs,
 *   and one for IPIs.
 * - Since Linux needs more than 2 IPIs, we implement a software IRQ controller
 *   and funnel all IPIs into one per-CPU IPI (the second "self" IPI is unused).
 * - FIQ hwirq numbers are assigned after true hwirqs, and are per-cpu.
 * - DT bindings use 3-cell form (like GIC):
 *   - <0 nr flags> - hwirq #nr
 *   - <1 nr flags> - FIQ #nr
 *     - nr=0  Physical HV timer
 *     - nr=1  Virtual HV timer
 *     - nr=2  Physical guest timer
 *     - nr=3  Virtual guest timer
 */

#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

#include <linux/bits.h>
#include <linux/bitfield.h>
#include <linux/cpuhotplug.h>
#include <linux/io.h>
#include <linux/irqchip.h>
#include <linux/irqchip/arm-vgic-info.h>
#include <linux/irqdomain.h>
#include <linux/limits.h>
#include <linux/of_address.h>
#include <linux/slab.h>
#include <asm/exception.h>
#include <asm/sysreg.h>
#include <asm/virt.h>

#include <dt-bindings/interrupt-controller/apple-aic.h>

/*
 * AIC registers (MMIO)
 */

#define AIC_INFO		0x0004
#define AIC_INFO_NR_HW		GENMASK(15, 0)

#define AIC_CONFIG		0x0010

#define AIC_WHOAMI		0x2000
#define AIC_EVENT		0x2004
#define AIC_EVENT_TYPE		GENMASK(31, 16)
#define AIC_EVENT_NUM		GENMASK(15, 0)

#define AIC_EVENT_TYPE_HW	1
#define AIC_EVENT_TYPE_IPI	4
#define AIC_EVENT_IPI_OTHER	1
#define AIC_EVENT_IPI_SELF	2

#define AIC_IPI_SEND		0x2008
#define AIC_IPI_ACK		0x200c
#define AIC_IPI_MASK_SET	0x2024
#define AIC_IPI_MASK_CLR	0x2028

#define AIC_IPI_SEND_CPU(cpu)	BIT(cpu)

#define AIC_IPI_OTHER		BIT(0)
#define AIC_IPI_SELF		BIT(31)

#define AIC_TARGET_CPU		0x3000
#define AIC_SW_SET		0x4000
#define AIC_SW_CLR		0x4080
#define AIC_MASK_SET		0x4100
#define AIC_MASK_CLR		0x4180

#define AIC_CPU_IPI_SET(cpu)	(0x5008 + ((cpu) << 7))
#define AIC_CPU_IPI_CLR(cpu)	(0x500c + ((cpu) << 7))
#define AIC_CPU_IPI_MASK_SET(cpu) (0x5024 + ((cpu) << 7))
#define AIC_CPU_IPI_MASK_CLR(cpu) (0x5028 + ((cpu) << 7))

#define MASK_REG(x)		(4 * ((x) >> 5))
#define MASK_BIT(x)		BIT((x) & GENMASK(4, 0))

/*
 * IMP-DEF sysregs that control FIQ sources
 * Note: sysreg-based IPIs are not supported yet.
 */

/* Core PMC control register */
#define SYS_IMP_APL_PMCR0_EL1		sys_reg(3, 1, 15, 0, 0)
#define PMCR0_IMODE			GENMASK(10, 8)
#define PMCR0_IMODE_OFF			0
#define PMCR0_IMODE_PMI			1
#define PMCR0_IMODE_AIC			2
#define PMCR0_IMODE_HALT		3
#define PMCR0_IMODE_FIQ			4
#define PMCR0_IACT			BIT(11)

/* IPI request registers */
#define SYS_IMP_APL_IPI_RR_LOCAL_EL1	sys_reg(3, 5, 15, 0, 0)
#define SYS_IMP_APL_IPI_RR_GLOBAL_EL1	sys_reg(3, 5, 15, 0, 1)
#define IPI_RR_CPU			GENMASK(7, 0)
/* Cluster only used for the GLOBAL register */
#define IPI_RR_CLUSTER			GENMASK(23, 16)
#define IPI_RR_TYPE			GENMASK(29, 28)
#define IPI_RR_IMMEDIATE		0
#define IPI_RR_RETRACT			1
#define IPI_RR_DEFERRED			2
#define IPI_RR_NOWAKE			3

/* IPI status register */
#define SYS_IMP_APL_IPI_SR_EL1		sys_reg(3, 5, 15, 1, 1)
#define IPI_SR_PENDING			BIT(0)

/* Guest timer FIQ enable register */
#define SYS_IMP_APL_VM_TMR_FIQ_ENA_EL2	sys_reg(3, 5, 15, 1, 3)
#define VM_TMR_FIQ_ENABLE_V		BIT(0)
#define VM_TMR_FIQ_ENABLE_P		BIT(1)

/* Deferred IPI countdown register */
#define SYS_IMP_APL_IPI_CR_EL1		sys_reg(3, 5, 15, 3, 1)

/* Uncore PMC control register */
#define SYS_IMP_APL_UPMCR0_EL1		sys_reg(3, 7, 15, 0, 4)
#define UPMCR0_IMODE			GENMASK(18, 16)
#define UPMCR0_IMODE_OFF		0
#define UPMCR0_IMODE_AIC		2
#define UPMCR0_IMODE_HALT		3
#define UPMCR0_IMODE_FIQ		4

/* Uncore PMC status register */
#define SYS_IMP_APL_UPMSR_EL1		sys_reg(3, 7, 15, 6, 4)
#define UPMSR_IACT			BIT(0)

#define AIC_NR_FIQ		4
#define AIC_NR_SWIPI		32

/*
 * FIQ hwirq index definitions: FIQ sources use the DT binding defines
 * directly, except that timers are special. At the irqchip level, the
 * two timer types are represented by their access method: _EL0 registers
 * or _EL02 registers. In the DT binding, the timers are represented
 * by their purpose (HV or guest). This mapping is for when the kernel is
 * running at EL2 (with VHE). When the kernel is running at EL1, the
 * mapping differs and aic_irq_domain_translate() performs the remapping.
 */

#define AIC_TMR_EL0_PHYS	AIC_TMR_HV_PHYS
#define AIC_TMR_EL0_VIRT	AIC_TMR_HV_VIRT
#define AIC_TMR_EL02_PHYS	AIC_TMR_GUEST_PHYS
#define AIC_TMR_EL02_VIRT	AIC_TMR_GUEST_VIRT

struct aic_irq_chip {
	void __iomem *base;
	struct irq_domain *hw_domain;
	struct irq_domain *ipi_domain;
	int nr_hw;
	int ipi_hwirq;
};

static DEFINE_PER_CPU(uint32_t, aic_fiq_unmasked);

static DEFINE_PER_CPU(atomic_t, aic_vipi_flag);
static DEFINE_PER_CPU(atomic_t, aic_vipi_enable);

static struct aic_irq_chip *aic_irqc;

static void aic_handle_ipi(struct pt_regs *regs);

static u32 aic_ic_read(struct aic_irq_chip *ic, u32 reg)
{
	return readl_relaxed(ic->base + reg);
}

static void aic_ic_write(struct aic_irq_chip *ic, u32 reg, u32 val)
{
	writel_relaxed(val, ic->base + reg);
}

/*
 * IRQ irqchip
 */

static void aic_irq_mask(struct irq_data *d)
{
	struct aic_irq_chip *ic = irq_data_get_irq_chip_data(d);

	aic_ic_write(ic, AIC_MASK_SET + MASK_REG(irqd_to_hwirq(d)),
		     MASK_BIT(irqd_to_hwirq(d)));
}

static void aic_irq_unmask(struct irq_data *d)
{
	struct aic_irq_chip *ic = irq_data_get_irq_chip_data(d);

	aic_ic_write(ic, AIC_MASK_CLR + MASK_REG(d->hwirq),
		     MASK_BIT(irqd_to_hwirq(d)));
}

static void aic_irq_eoi(struct irq_data *d)
{
	/*
	 * Reading the interrupt reason automatically acknowledges and masks
	 * the IRQ, so we just unmask it here if needed.
	 */
	if (!irqd_irq_masked(d))
		aic_irq_unmask(d);
}

static void __exception_irq_entry aic_handle_irq(struct pt_regs *regs)
{
	struct aic_irq_chip *ic = aic_irqc;
	u32 event, type, irq;

	do {
		/*
		 * We cannot use a relaxed read here, as reads from DMA buffers
		 * need to be ordered after the IRQ fires.
		 */
		event = readl(ic->base + AIC_EVENT);
		type = FIELD_GET(AIC_EVENT_TYPE, event);
		irq = FIELD_GET(AIC_EVENT_NUM, event);

		if (type == AIC_EVENT_TYPE_HW)
			generic_handle_domain_irq(aic_irqc->hw_domain, irq);
		else if (type == AIC_EVENT_TYPE_IPI && irq == 1)
			aic_handle_ipi(regs);
		else if (event != 0)
			pr_err_ratelimited("Unknown IRQ event %d, %d\n", type, irq);
	} while (event);

	/*
	 * vGIC maintenance interrupts end up here too, so we need to check
	 * for them separately. This should never trigger if KVM is working
	 * properly, because it will have already taken care of clearing it
	 * on guest exit before this handler runs.
	 */
	if (is_kernel_in_hyp_mode() && (read_sysreg_s(SYS_ICH_HCR_EL2) & ICH_HCR_EN) &&
		read_sysreg_s(SYS_ICH_MISR_EL2) != 0) {
		pr_err_ratelimited("vGIC IRQ fired and not handled by KVM, disabling.\n");
		sysreg_clear_set_s(SYS_ICH_HCR_EL2, ICH_HCR_EN, 0);
	}
}

static int aic_irq_set_affinity(struct irq_data *d,
				const struct cpumask *mask_val, bool force)
{
	irq_hw_number_t hwirq = irqd_to_hwirq(d);
	struct aic_irq_chip *ic = irq_data_get_irq_chip_data(d);
	int cpu;

	if (force)
		cpu = cpumask_first(mask_val);
	else
		cpu = cpumask_any_and(mask_val, cpu_online_mask);

	aic_ic_write(ic, AIC_TARGET_CPU + hwirq * 4, BIT(cpu));
	irq_data_update_effective_affinity(d, cpumask_of(cpu));

	return IRQ_SET_MASK_OK;
}

static int aic_irq_set_type(struct irq_data *d, unsigned int type)
{
	/*
	 * Some IRQs (e.g. MSIs) implicitly have edge semantics, and we don't
	 * have a way to find out the type of any given IRQ, so just allow both.
	 */
	return (type == IRQ_TYPE_LEVEL_HIGH || type == IRQ_TYPE_EDGE_RISING) ? 0 : -EINVAL;
}

static struct irq_chip aic_chip = {
	.name = "AIC",
	.irq_mask = aic_irq_mask,
	.irq_unmask = aic_irq_unmask,
	.irq_eoi = aic_irq_eoi,
	.irq_set_affinity = aic_irq_set_affinity,
	.irq_set_type = aic_irq_set_type,
};

/*
 * FIQ irqchip
 */

static unsigned long aic_fiq_get_idx(struct irq_data *d)
{
	struct aic_irq_chip *ic = irq_data_get_irq_chip_data(d);

	return irqd_to_hwirq(d) - ic->nr_hw;
}

static void aic_fiq_set_mask(struct irq_data *d)
{
	/* Only the guest timers have real mask bits, unfortunately. */
	switch (aic_fiq_get_idx(d)) {
	case AIC_TMR_EL02_PHYS:
		sysreg_clear_set_s(SYS_IMP_APL_VM_TMR_FIQ_ENA_EL2, VM_TMR_FIQ_ENABLE_P, 0);
		isb();
		break;
	case AIC_TMR_EL02_VIRT:
		sysreg_clear_set_s(SYS_IMP_APL_VM_TMR_FIQ_ENA_EL2, VM_TMR_FIQ_ENABLE_V, 0);
		isb();
		break;
	default:
		break;
	}
}

static void aic_fiq_clear_mask(struct irq_data *d)
{
	switch (aic_fiq_get_idx(d)) {
	case AIC_TMR_EL02_PHYS:
		sysreg_clear_set_s(SYS_IMP_APL_VM_TMR_FIQ_ENA_EL2, 0, VM_TMR_FIQ_ENABLE_P);
		isb();
		break;
	case AIC_TMR_EL02_VIRT:
		sysreg_clear_set_s(SYS_IMP_APL_VM_TMR_FIQ_ENA_EL2, 0, VM_TMR_FIQ_ENABLE_V);
		isb();
		break;
	default:
		break;
	}
}

static void aic_fiq_mask(struct irq_data *d)
{
	aic_fiq_set_mask(d);
	__this_cpu_and(aic_fiq_unmasked, ~BIT(aic_fiq_get_idx(d)));
}

static void aic_fiq_unmask(struct irq_data *d)
{
	aic_fiq_clear_mask(d);
	__this_cpu_or(aic_fiq_unmasked, BIT(aic_fiq_get_idx(d)));
}

static void aic_fiq_eoi(struct irq_data *d)
{
	/* We mask to ack (where we can), so we need to unmask at EOI. */
	if (__this_cpu_read(aic_fiq_unmasked) & BIT(aic_fiq_get_idx(d)))
		aic_fiq_clear_mask(d);
}

#define TIMER_FIRING(x)                                                        \
	(((x) & (ARCH_TIMER_CTRL_ENABLE | ARCH_TIMER_CTRL_IT_MASK |            \
		 ARCH_TIMER_CTRL_IT_STAT)) ==                                  \
	 (ARCH_TIMER_CTRL_ENABLE | ARCH_TIMER_CTRL_IT_STAT))

static void __exception_irq_entry aic_handle_fiq(struct pt_regs *regs)
{
	/*
	 * It would be really nice if we had a system register that lets us get
	 * the FIQ source state without having to peek down into sources...
	 * but such a register does not seem to exist.
	 *
	 * So, we have these potential sources to test for:
	 *  - Fast IPIs (not yet used)
	 *  - The 4 timers (CNTP, CNTV for each of HV and guest)
	 *  - Per-core PMCs (not yet supported)
	 *  - Per-cluster uncore PMCs (not yet supported)
	 *
	 * Since not dealing with any of these results in a FIQ storm,
	 * we check for everything here, even things we don't support yet.
	 */

	if (read_sysreg_s(SYS_IMP_APL_IPI_SR_EL1) & IPI_SR_PENDING) {
		pr_err_ratelimited("Fast IPI fired. Acking.\n");
		write_sysreg_s(IPI_SR_PENDING, SYS_IMP_APL_IPI_SR_EL1);
	}

	if (TIMER_FIRING(read_sysreg(cntp_ctl_el0)))
		generic_handle_domain_irq(aic_irqc->hw_domain,
					  aic_irqc->nr_hw + AIC_TMR_EL0_PHYS);

	if (TIMER_FIRING(read_sysreg(cntv_ctl_el0)))
		generic_handle_domain_irq(aic_irqc->hw_domain,
					  aic_irqc->nr_hw + AIC_TMR_EL0_VIRT);

	if (is_kernel_in_hyp_mode()) {
		uint64_t enabled = read_sysreg_s(SYS_IMP_APL_VM_TMR_FIQ_ENA_EL2);

		if ((enabled & VM_TMR_FIQ_ENABLE_P) &&
		    TIMER_FIRING(read_sysreg_s(SYS_CNTP_CTL_EL02)))
			generic_handle_domain_irq(aic_irqc->hw_domain,
						  aic_irqc->nr_hw + AIC_TMR_EL02_PHYS);

		if ((enabled & VM_TMR_FIQ_ENABLE_V) &&
		    TIMER_FIRING(read_sysreg_s(SYS_CNTV_CTL_EL02)))
			generic_handle_domain_irq(aic_irqc->hw_domain,
						  aic_irqc->nr_hw + AIC_TMR_EL02_VIRT);
	}

	if ((read_sysreg_s(SYS_IMP_APL_PMCR0_EL1) & (PMCR0_IMODE | PMCR0_IACT)) ==
			(FIELD_PREP(PMCR0_IMODE, PMCR0_IMODE_FIQ) | PMCR0_IACT)) {
		/*
		 * Not supported yet, let's figure out how to handle this when
		 * we implement these proprietary performance counters. For now,
		 * just mask it and move on.
		 */
		pr_err_ratelimited("PMC FIQ fired. Masking.\n");
		sysreg_clear_set_s(SYS_IMP_APL_PMCR0_EL1, PMCR0_IMODE | PMCR0_IACT,
				   FIELD_PREP(PMCR0_IMODE, PMCR0_IMODE_OFF));
	}

	if (FIELD_GET(UPMCR0_IMODE, read_sysreg_s(SYS_IMP_APL_UPMCR0_EL1)) == UPMCR0_IMODE_FIQ &&
			(read_sysreg_s(SYS_IMP_APL_UPMSR_EL1) & UPMSR_IACT)) {
		/* Same story with uncore PMCs */
		pr_err_ratelimited("Uncore PMC FIQ fired. Masking.\n");
		sysreg_clear_set_s(SYS_IMP_APL_UPMCR0_EL1, UPMCR0_IMODE,
				   FIELD_PREP(UPMCR0_IMODE, UPMCR0_IMODE_OFF));
	}
}

static int aic_fiq_set_type(struct irq_data *d, unsigned int type)
{
	return (type == IRQ_TYPE_LEVEL_HIGH) ? 0 : -EINVAL;
}

static struct irq_chip fiq_chip = {
	.name = "AIC-FIQ",
	.irq_mask = aic_fiq_mask,
	.irq_unmask = aic_fiq_unmask,
	.irq_ack = aic_fiq_set_mask,
	.irq_eoi = aic_fiq_eoi,
	.irq_set_type = aic_fiq_set_type,
};

/*
 * Main IRQ domain
 */

static int aic_irq_domain_map(struct irq_domain *id, unsigned int irq,
			      irq_hw_number_t hw)
{
	struct aic_irq_chip *ic = id->host_data;

	if (hw < ic->nr_hw) {
		irq_domain_set_info(id, irq, hw, &aic_chip, id->host_data,
				    handle_fasteoi_irq, NULL, NULL);
		irqd_set_single_target(irq_desc_get_irq_data(irq_to_desc(irq)));
	} else {
		irq_set_percpu_devid(irq);
		irq_domain_set_info(id, irq, hw, &fiq_chip, id->host_data,
				    handle_percpu_devid_irq, NULL, NULL);
	}

	return 0;
}

static int aic_irq_domain_translate(struct irq_domain *id,
				    struct irq_fwspec *fwspec,
				    unsigned long *hwirq,
				    unsigned int *type)
{
	struct aic_irq_chip *ic = id->host_data;

	if (fwspec->param_count != 3 || !is_of_node(fwspec->fwnode))
		return -EINVAL;

	switch (fwspec->param[0]) {
	case AIC_IRQ:
		if (fwspec->param[1] >= ic->nr_hw)
			return -EINVAL;
		*hwirq = fwspec->param[1];
		break;
	case AIC_FIQ:
		if (fwspec->param[1] >= AIC_NR_FIQ)
			return -EINVAL;
		*hwirq = ic->nr_hw + fwspec->param[1];

		/*
		 * In EL1 the non-redirected registers are the guest's,
		 * not EL2's, so remap the hwirqs to match.
		 */
		if (!is_kernel_in_hyp_mode()) {
			switch (fwspec->param[1]) {
			case AIC_TMR_GUEST_PHYS:
				*hwirq = ic->nr_hw + AIC_TMR_EL0_PHYS;
				break;
			case AIC_TMR_GUEST_VIRT:
				*hwirq = ic->nr_hw + AIC_TMR_EL0_VIRT;
				break;
			case AIC_TMR_HV_PHYS:
			case AIC_TMR_HV_VIRT:
				return -ENOENT;
			default:
				break;
			}
		}
		break;
	default:
		return -EINVAL;
	}

	*type = fwspec->param[2] & IRQ_TYPE_SENSE_MASK;

	return 0;
}

static int aic_irq_domain_alloc(struct irq_domain *domain, unsigned int virq,
				unsigned int nr_irqs, void *arg)
{
	unsigned int type = IRQ_TYPE_NONE;
	struct irq_fwspec *fwspec = arg;
	irq_hw_number_t hwirq;
	int i, ret;

	ret = aic_irq_domain_translate(domain, fwspec, &hwirq, &type);
	if (ret)
		return ret;

	for (i = 0; i < nr_irqs; i++) {
		ret = aic_irq_domain_map(domain, virq + i, hwirq + i);
		if (ret)
			return ret;
	}

	return 0;
}

static void aic_irq_domain_free(struct irq_domain *domain, unsigned int virq,
				unsigned int nr_irqs)
{
	int i;

	for (i = 0; i < nr_irqs; i++) {
		struct irq_data *d = irq_domain_get_irq_data(domain, virq + i);

		irq_set_handler(virq + i, NULL);
		irq_domain_reset_irq_data(d);
	}
}

static const struct irq_domain_ops aic_irq_domain_ops = {
	.translate	= aic_irq_domain_translate,
	.alloc		= aic_irq_domain_alloc,
	.free		= aic_irq_domain_free,
};

/*
 * IPI irqchip
 */

static void aic_ipi_mask(struct irq_data *d)
{
	u32 irq_bit = BIT(irqd_to_hwirq(d));

	/* No specific ordering requirements needed here. */
	atomic_andnot(irq_bit, this_cpu_ptr(&aic_vipi_enable));
}

static void aic_ipi_unmask(struct irq_data *d)
{
	struct aic_irq_chip *ic = irq_data_get_irq_chip_data(d);
	u32 irq_bit = BIT(irqd_to_hwirq(d));

	atomic_or(irq_bit, this_cpu_ptr(&aic_vipi_enable));

	/*
	 * The atomic_or() above must complete before the atomic_read()
	 * below to avoid racing aic_ipi_send_mask().
	 */
	smp_mb__after_atomic();

	/*
	 * If a pending vIPI was unmasked, raise a HW IPI to ourselves.
	 * No barriers needed here since this is a self-IPI.
	 */
	if (atomic_read(this_cpu_ptr(&aic_vipi_flag)) & irq_bit)
		aic_ic_write(ic, AIC_IPI_SEND, AIC_IPI_SEND_CPU(smp_processor_id()));
}

static void aic_ipi_send_mask(struct irq_data *d, const struct cpumask *mask)
{
	struct aic_irq_chip *ic = irq_data_get_irq_chip_data(d);
	u32 irq_bit = BIT(irqd_to_hwirq(d));
	u32 send = 0;
	int cpu;
	unsigned long pending;

	for_each_cpu(cpu, mask) {
		/*
		 * This sequence is the mirror of the one in aic_ipi_unmask();
		 * see the comment there. Additionally, release semantics
		 * ensure that the vIPI flag set is ordered after any shared
		 * memory accesses that precede it. This therefore also pairs
		 * with the atomic_fetch_andnot in aic_handle_ipi().
		 */
		pending = atomic_fetch_or_release(irq_bit, per_cpu_ptr(&aic_vipi_flag, cpu));

		/*
		 * The atomic_fetch_or_release() above must complete before the
		 * atomic_read() below to avoid racing aic_ipi_unmask().
		 */
		smp_mb__after_atomic();

		if (!(pending & irq_bit) &&
		    (atomic_read(per_cpu_ptr(&aic_vipi_enable, cpu)) & irq_bit))
			send |= AIC_IPI_SEND_CPU(cpu);
	}

	/*
	 * The flag writes must complete before the physical IPI is issued
	 * to another CPU. This is implied by the control dependency on
	 * the result of atomic_read_acquire() above, which is itself
	 * already ordered after the vIPI flag write.
	 */
	if (send)
		aic_ic_write(ic, AIC_IPI_SEND, send);
}

static struct irq_chip ipi_chip = {
	.name = "AIC-IPI",
	.irq_mask = aic_ipi_mask,
	.irq_unmask = aic_ipi_unmask,
	.ipi_send_mask = aic_ipi_send_mask,
};

/*
 * IPI IRQ domain
 */

static void aic_handle_ipi(struct pt_regs *regs)
{
	int i;
	unsigned long enabled, firing;

	/*
	 * Ack the IPI. We need to order this after the AIC event read, but
	 * that is enforced by normal MMIO ordering guarantees.
	 */
	aic_ic_write(aic_irqc, AIC_IPI_ACK, AIC_IPI_OTHER);

	/*
	 * The mask read does not need to be ordered. Only we can change
	 * our own mask anyway, so no races are possible here, as long as
	 * we are properly in the interrupt handler (which is covered by
	 * the barrier that is part of the top-level AIC handler's readl()).
	 */
	enabled = atomic_read(this_cpu_ptr(&aic_vipi_enable));

	/*
	 * Clear the IPIs we are about to handle. This pairs with the
	 * atomic_fetch_or_release() in aic_ipi_send_mask(), and needs to be
	 * ordered after the aic_ic_write() above (to avoid dropping vIPIs) and
	 * before IPI handling code (to avoid races handling vIPIs before they
	 * are signaled). The former is taken care of by the release semantics
	 * of the write portion, while the latter is taken care of by the
	 * acquire semantics of the read portion.
	 */
	firing = atomic_fetch_andnot(enabled, this_cpu_ptr(&aic_vipi_flag)) & enabled;

	for_each_set_bit(i, &firing, AIC_NR_SWIPI)
		generic_handle_domain_irq(aic_irqc->ipi_domain, i);

	/*
	 * No ordering needed here; at worst this just changes the timing of
	 * when the next IPI will be delivered.
	 */
	aic_ic_write(aic_irqc, AIC_IPI_MASK_CLR, AIC_IPI_OTHER);
}

static int aic_ipi_alloc(struct irq_domain *d, unsigned int virq,
			 unsigned int nr_irqs, void *args)
{
	int i;

	for (i = 0; i < nr_irqs; i++) {
		irq_set_percpu_devid(virq + i);
		irq_domain_set_info(d, virq + i, i, &ipi_chip, d->host_data,
				    handle_percpu_devid_irq, NULL, NULL);
	}

	return 0;
}

static void aic_ipi_free(struct irq_domain *d, unsigned int virq, unsigned int nr_irqs)
{
	/* Not freeing IPIs */
}

static const struct irq_domain_ops aic_ipi_domain_ops = {
	.alloc = aic_ipi_alloc,
	.free = aic_ipi_free,
};

static int aic_init_smp(struct aic_irq_chip *irqc, struct device_node *node)
{
	struct irq_domain *ipi_domain;
	int base_ipi;

	ipi_domain = irq_domain_create_linear(irqc->hw_domain->fwnode, AIC_NR_SWIPI,
					      &aic_ipi_domain_ops, irqc);
	if (WARN_ON(!ipi_domain))
		return -ENODEV;

	ipi_domain->flags |= IRQ_DOMAIN_FLAG_IPI_SINGLE;
	irq_domain_update_bus_token(ipi_domain, DOMAIN_BUS_IPI);

	base_ipi = __irq_domain_alloc_irqs(ipi_domain, -1, AIC_NR_SWIPI,
					   NUMA_NO_NODE, NULL, false, NULL);

	if (WARN_ON(!base_ipi)) {
		irq_domain_remove(ipi_domain);
		return -ENODEV;
	}

	set_smp_ipi_range(base_ipi, AIC_NR_SWIPI);

	irqc->ipi_domain = ipi_domain;

	return 0;
}

static int aic_init_cpu(unsigned int cpu)
{
	/* Mask all hard-wired per-CPU IRQ/FIQ sources */

	/* Pending Fast IPI FIQs */
	write_sysreg_s(IPI_SR_PENDING, SYS_IMP_APL_IPI_SR_EL1);

	/* Timer FIQs */
	sysreg_clear_set(cntp_ctl_el0, 0, ARCH_TIMER_CTRL_IT_MASK);
	sysreg_clear_set(cntv_ctl_el0, 0, ARCH_TIMER_CTRL_IT_MASK);

	/* EL2-only (VHE mode) IRQ sources */
	if (is_kernel_in_hyp_mode()) {
		/* Guest timers */
		sysreg_clear_set_s(SYS_IMP_APL_VM_TMR_FIQ_ENA_EL2,
				   VM_TMR_FIQ_ENABLE_V | VM_TMR_FIQ_ENABLE_P, 0);

		/* vGIC maintenance IRQ */
		sysreg_clear_set_s(SYS_ICH_HCR_EL2, ICH_HCR_EN, 0);
	}

	/* PMC FIQ */
	sysreg_clear_set_s(SYS_IMP_APL_PMCR0_EL1, PMCR0_IMODE | PMCR0_IACT,
			   FIELD_PREP(PMCR0_IMODE, PMCR0_IMODE_OFF));

	/* Uncore PMC FIQ */
	sysreg_clear_set_s(SYS_IMP_APL_UPMCR0_EL1, UPMCR0_IMODE,
			   FIELD_PREP(UPMCR0_IMODE, UPMCR0_IMODE_OFF));

	/* Commit all of the above */
	isb();

	/*
	 * Make sure the kernel's idea of logical CPU order is the same as AIC's
	 * If we ever end up with a mismatch here, we will have to introduce
	 * a mapping table similar to what other irqchip drivers do.
	 */
	WARN_ON(aic_ic_read(aic_irqc, AIC_WHOAMI) != smp_processor_id());

	/*
	 * Always keep IPIs unmasked at the hardware level (except auto-masking
	 * by AIC during processing). We manage masks at the vIPI level.
	 */
	aic_ic_write(aic_irqc, AIC_IPI_ACK, AIC_IPI_SELF | AIC_IPI_OTHER);
	aic_ic_write(aic_irqc, AIC_IPI_MASK_SET, AIC_IPI_SELF);
	aic_ic_write(aic_irqc, AIC_IPI_MASK_CLR, AIC_IPI_OTHER);

	/* Initialize the local mask state */
	__this_cpu_write(aic_fiq_unmasked, 0);

	return 0;
}

static struct gic_kvm_info vgic_info __initdata = {
	.type			= GIC_V3,
	.no_maint_irq_mask	= true,
	.no_hw_deactivation	= true,
};

static int __init aic_of_ic_init(struct device_node *node, struct device_node *parent)
{
	int i;
	void __iomem *regs;
	u32 info;
	struct aic_irq_chip *irqc;

	regs = of_iomap(node, 0);
	if (WARN_ON(!regs))
		return -EIO;

	irqc = kzalloc(sizeof(*irqc), GFP_KERNEL);
	if (!irqc)
		return -ENOMEM;

	aic_irqc = irqc;
	irqc->base = regs;

	info = aic_ic_read(irqc, AIC_INFO);
	irqc->nr_hw = FIELD_GET(AIC_INFO_NR_HW, info);

	irqc->hw_domain = irq_domain_create_linear(of_node_to_fwnode(node),
						   irqc->nr_hw + AIC_NR_FIQ,
						   &aic_irq_domain_ops, irqc);
	if (WARN_ON(!irqc->hw_domain)) {
		iounmap(irqc->base);
		kfree(irqc);
		return -ENODEV;
	}

	irq_domain_update_bus_token(irqc->hw_domain, DOMAIN_BUS_WIRED);

	if (aic_init_smp(irqc, node)) {
		irq_domain_remove(irqc->hw_domain);
		iounmap(irqc->base);
		kfree(irqc);
		return -ENODEV;
	}

	set_handle_irq(aic_handle_irq);
	set_handle_fiq(aic_handle_fiq);

	for (i = 0; i < BITS_TO_U32(irqc->nr_hw); i++)
		aic_ic_write(irqc, AIC_MASK_SET + i * 4, U32_MAX);
	for (i = 0; i < BITS_TO_U32(irqc->nr_hw); i++)
		aic_ic_write(irqc, AIC_SW_CLR + i * 4, U32_MAX);
	for (i = 0; i < irqc->nr_hw; i++)
		aic_ic_write(irqc, AIC_TARGET_CPU + i * 4, 1);

	if (!is_kernel_in_hyp_mode())
		pr_info("Kernel running in EL1, mapping interrupts");

	cpuhp_setup_state(CPUHP_AP_IRQ_APPLE_AIC_STARTING,
			  "irqchip/apple-aic/ipi:starting",
			  aic_init_cpu, NULL);

	vgic_set_kvm_info(&vgic_info);

	pr_info("Initialized with %d IRQs, %d FIQs, %d vIPIs\n",
		irqc->nr_hw, AIC_NR_FIQ, AIC_NR_SWIPI);

	return 0;
}

IRQCHIP_DECLARE(apple_m1_aic, "apple,aic", aic_of_ic_init);