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
|
/*
* Copyright 2019 Advanced Micro Devices, Inc.
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
* OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
* ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
* OTHER DEALINGS IN THE SOFTWARE.
*
*/
#include <linux/firmware.h>
#include <linux/pci.h>
#include "amdgpu.h"
#include "amdgpu_atomfirmware.h"
#include "gmc_v10_0.h"
#include "hdp/hdp_5_0_0_offset.h"
#include "hdp/hdp_5_0_0_sh_mask.h"
#include "gc/gc_10_1_0_sh_mask.h"
#include "mmhub/mmhub_2_0_0_sh_mask.h"
#include "dcn/dcn_2_0_0_offset.h"
#include "dcn/dcn_2_0_0_sh_mask.h"
#include "oss/osssys_5_0_0_offset.h"
#include "ivsrcid/vmc/irqsrcs_vmc_1_0.h"
#include "navi10_enum.h"
#include "soc15.h"
#include "soc15_common.h"
#include "nbio_v2_3.h"
#include "gfxhub_v2_0.h"
#include "mmhub_v2_0.h"
#include "athub_v2_0.h"
/* XXX Move this macro to navi10 header file, which is like vid.h for VI.*/
#define AMDGPU_NUM_OF_VMIDS 8
#if 0
static const struct soc15_reg_golden golden_settings_navi10_hdp[] =
{
/* TODO add golden setting for hdp */
};
#endif
static int
gmc_v10_0_vm_fault_interrupt_state(struct amdgpu_device *adev,
struct amdgpu_irq_src *src, unsigned type,
enum amdgpu_interrupt_state state)
{
struct amdgpu_vmhub *hub;
u32 tmp, reg, bits[AMDGPU_MAX_VMHUBS], i;
bits[AMDGPU_GFXHUB_0] = GCVM_CONTEXT1_CNTL__RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
GCVM_CONTEXT1_CNTL__DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
GCVM_CONTEXT1_CNTL__PDE0_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
GCVM_CONTEXT1_CNTL__VALID_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
GCVM_CONTEXT1_CNTL__READ_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
GCVM_CONTEXT1_CNTL__WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
GCVM_CONTEXT1_CNTL__EXECUTE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK;
bits[AMDGPU_MMHUB_0] = MMVM_CONTEXT1_CNTL__RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
MMVM_CONTEXT1_CNTL__DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
MMVM_CONTEXT1_CNTL__PDE0_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
MMVM_CONTEXT1_CNTL__VALID_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
MMVM_CONTEXT1_CNTL__READ_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
MMVM_CONTEXT1_CNTL__WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
MMVM_CONTEXT1_CNTL__EXECUTE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK;
switch (state) {
case AMDGPU_IRQ_STATE_DISABLE:
/* MM HUB */
hub = &adev->vmhub[AMDGPU_MMHUB_0];
for (i = 0; i < 16; i++) {
reg = hub->vm_context0_cntl + i;
tmp = RREG32(reg);
tmp &= ~bits[AMDGPU_MMHUB_0];
WREG32(reg, tmp);
}
/* GFX HUB */
hub = &adev->vmhub[AMDGPU_GFXHUB_0];
for (i = 0; i < 16; i++) {
reg = hub->vm_context0_cntl + i;
tmp = RREG32(reg);
tmp &= ~bits[AMDGPU_GFXHUB_0];
WREG32(reg, tmp);
}
break;
case AMDGPU_IRQ_STATE_ENABLE:
/* MM HUB */
hub = &adev->vmhub[AMDGPU_MMHUB_0];
for (i = 0; i < 16; i++) {
reg = hub->vm_context0_cntl + i;
tmp = RREG32(reg);
tmp |= bits[AMDGPU_MMHUB_0];
WREG32(reg, tmp);
}
/* GFX HUB */
hub = &adev->vmhub[AMDGPU_GFXHUB_0];
for (i = 0; i < 16; i++) {
reg = hub->vm_context0_cntl + i;
tmp = RREG32(reg);
tmp |= bits[AMDGPU_GFXHUB_0];
WREG32(reg, tmp);
}
break;
default:
break;
}
return 0;
}
static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev,
struct amdgpu_irq_src *source,
struct amdgpu_iv_entry *entry)
{
struct amdgpu_vmhub *hub = &adev->vmhub[entry->vmid_src];
uint32_t status = 0;
u64 addr;
addr = (u64)entry->src_data[0] << 12;
addr |= ((u64)entry->src_data[1] & 0xf) << 44;
if (!amdgpu_sriov_vf(adev)) {
status = RREG32(hub->vm_l2_pro_fault_status);
WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1);
}
if (printk_ratelimit()) {
dev_err(adev->dev,
"[%s] VMC page fault (src_id:%u ring:%u vmid:%u pasid:%u)\n",
entry->vmid_src ? "mmhub" : "gfxhub",
entry->src_id, entry->ring_id, entry->vmid,
entry->pasid);
dev_err(adev->dev, " at page 0x%016llx from %d\n",
addr, entry->client_id);
if (!amdgpu_sriov_vf(adev))
dev_err(adev->dev,
"VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n",
status);
}
return 0;
}
static const struct amdgpu_irq_src_funcs gmc_v10_0_irq_funcs = {
.set = gmc_v10_0_vm_fault_interrupt_state,
.process = gmc_v10_0_process_interrupt,
};
static void gmc_v10_0_set_irq_funcs(struct amdgpu_device *adev)
{
adev->gmc.vm_fault.num_types = 1;
adev->gmc.vm_fault.funcs = &gmc_v10_0_irq_funcs;
}
static uint32_t gmc_v10_0_get_invalidate_req(unsigned int vmid,
uint32_t flush_type)
{
u32 req = 0;
/* invalidate using legacy mode on vmid*/
req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ,
PER_VMID_INVALIDATE_REQ, 1 << vmid);
req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, FLUSH_TYPE, flush_type);
req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PTES, 1);
req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE0, 1);
req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE1, 1);
req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE2, 1);
req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ, INVALIDATE_L1_PTES, 1);
req = REG_SET_FIELD(req, GCVM_INVALIDATE_ENG0_REQ,
CLEAR_PROTECTION_FAULT_STATUS_ADDR, 0);
return req;
}
/*
* GART
* VMID 0 is the physical GPU addresses as used by the kernel.
* VMIDs 1-15 are used for userspace clients and are handled
* by the amdgpu vm/hsa code.
*/
static void gmc_v10_0_flush_vm_hub(struct amdgpu_device *adev, uint32_t vmid,
unsigned int vmhub, uint32_t flush_type)
{
struct amdgpu_vmhub *hub = &adev->vmhub[vmhub];
u32 tmp = gmc_v10_0_get_invalidate_req(vmid, flush_type);
/* Use register 17 for GART */
const unsigned eng = 17;
unsigned int i;
WREG32_NO_KIQ(hub->vm_inv_eng0_req + eng, tmp);
/* Wait for ACK with a delay.*/
for (i = 0; i < adev->usec_timeout; i++) {
tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack + eng);
tmp &= 1 << vmid;
if (tmp)
break;
udelay(1);
}
if (i < adev->usec_timeout)
return;
DRM_ERROR("Timeout waiting for VM flush ACK!\n");
}
/**
* gmc_v10_0_flush_gpu_tlb - gart tlb flush callback
*
* @adev: amdgpu_device pointer
* @vmid: vm instance to flush
*
* Flush the TLB for the requested page table.
*/
static void gmc_v10_0_flush_gpu_tlb(struct amdgpu_device *adev,
uint32_t vmid, uint32_t flush_type)
{
struct amdgpu_ring *ring = adev->mman.buffer_funcs_ring;
struct dma_fence *fence;
struct amdgpu_job *job;
int r;
/* flush hdp cache */
adev->nbio_funcs->hdp_flush(adev, NULL);
mutex_lock(&adev->mman.gtt_window_lock);
gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_MMHUB_0, 0);
if (!adev->mman.buffer_funcs_enabled ||
!adev->ib_pool_ready ||
adev->in_gpu_reset) {
gmc_v10_0_flush_vm_hub(adev, vmid, AMDGPU_GFXHUB_0, 0);
mutex_unlock(&adev->mman.gtt_window_lock);
return;
}
/* The SDMA on Navi has a bug which can theoretically result in memory
* corruption if an invalidation happens at the same time as an VA
* translation. Avoid this by doing the invalidation from the SDMA
* itself.
*/
r = amdgpu_job_alloc_with_ib(adev, 16 * 4, &job);
if (r)
goto error_alloc;
job->vm_pd_addr = amdgpu_gmc_pd_addr(adev->gart.bo);
job->vm_needs_flush = true;
amdgpu_ring_pad_ib(ring, &job->ibs[0]);
r = amdgpu_job_submit(job, &adev->mman.entity,
AMDGPU_FENCE_OWNER_UNDEFINED, &fence);
if (r)
goto error_submit;
mutex_unlock(&adev->mman.gtt_window_lock);
dma_fence_wait(fence, false);
dma_fence_put(fence);
return;
error_submit:
amdgpu_job_free(job);
error_alloc:
mutex_unlock(&adev->mman.gtt_window_lock);
DRM_ERROR("Error flushing GPU TLB using the SDMA (%d)!\n", r);
}
static uint64_t gmc_v10_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring,
unsigned vmid, uint64_t pd_addr)
{
struct amdgpu_vmhub *hub = &ring->adev->vmhub[ring->funcs->vmhub];
uint32_t req = gmc_v10_0_get_invalidate_req(vmid, 0);
unsigned eng = ring->vm_inv_eng;
amdgpu_ring_emit_wreg(ring, hub->ctx0_ptb_addr_lo32 + (2 * vmid),
lower_32_bits(pd_addr));
amdgpu_ring_emit_wreg(ring, hub->ctx0_ptb_addr_hi32 + (2 * vmid),
upper_32_bits(pd_addr));
amdgpu_ring_emit_wreg(ring, hub->vm_inv_eng0_req + eng, req);
/* wait for the invalidate to complete */
amdgpu_ring_emit_reg_wait(ring, hub->vm_inv_eng0_ack + eng,
1 << vmid, 1 << vmid);
return pd_addr;
}
static void gmc_v10_0_emit_pasid_mapping(struct amdgpu_ring *ring, unsigned vmid,
unsigned pasid)
{
struct amdgpu_device *adev = ring->adev;
uint32_t reg;
if (ring->funcs->vmhub == AMDGPU_GFXHUB_0)
reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT) + vmid;
else
reg = SOC15_REG_OFFSET(OSSSYS, 0, mmIH_VMID_0_LUT_MM) + vmid;
amdgpu_ring_emit_wreg(ring, reg, pasid);
}
/*
* PTE format on NAVI 10:
* 63:59 reserved
* 58:57 reserved
* 56 F
* 55 L
* 54 reserved
* 53:52 SW
* 51 T
* 50:48 mtype
* 47:12 4k physical page base address
* 11:7 fragment
* 6 write
* 5 read
* 4 exe
* 3 Z
* 2 snooped
* 1 system
* 0 valid
*
* PDE format on NAVI 10:
* 63:59 block fragment size
* 58:55 reserved
* 54 P
* 53:48 reserved
* 47:6 physical base address of PD or PTE
* 5:3 reserved
* 2 C
* 1 system
* 0 valid
*/
static uint64_t gmc_v10_0_get_vm_pte_flags(struct amdgpu_device *adev,
uint32_t flags)
{
uint64_t pte_flag = 0;
if (flags & AMDGPU_VM_PAGE_EXECUTABLE)
pte_flag |= AMDGPU_PTE_EXECUTABLE;
if (flags & AMDGPU_VM_PAGE_READABLE)
pte_flag |= AMDGPU_PTE_READABLE;
if (flags & AMDGPU_VM_PAGE_WRITEABLE)
pte_flag |= AMDGPU_PTE_WRITEABLE;
switch (flags & AMDGPU_VM_MTYPE_MASK) {
case AMDGPU_VM_MTYPE_DEFAULT:
pte_flag |= AMDGPU_PTE_MTYPE_NV10(MTYPE_NC);
break;
case AMDGPU_VM_MTYPE_NC:
pte_flag |= AMDGPU_PTE_MTYPE_NV10(MTYPE_NC);
break;
case AMDGPU_VM_MTYPE_WC:
pte_flag |= AMDGPU_PTE_MTYPE_NV10(MTYPE_WC);
break;
case AMDGPU_VM_MTYPE_CC:
pte_flag |= AMDGPU_PTE_MTYPE_NV10(MTYPE_CC);
break;
case AMDGPU_VM_MTYPE_UC:
pte_flag |= AMDGPU_PTE_MTYPE_NV10(MTYPE_UC);
break;
default:
pte_flag |= AMDGPU_PTE_MTYPE_NV10(MTYPE_NC);
break;
}
if (flags & AMDGPU_VM_PAGE_PRT)
pte_flag |= AMDGPU_PTE_PRT;
return pte_flag;
}
static void gmc_v10_0_get_vm_pde(struct amdgpu_device *adev, int level,
uint64_t *addr, uint64_t *flags)
{
if (!(*flags & AMDGPU_PDE_PTE) && !(*flags & AMDGPU_PTE_SYSTEM))
*addr = adev->vm_manager.vram_base_offset + *addr -
adev->gmc.vram_start;
BUG_ON(*addr & 0xFFFF00000000003FULL);
if (!adev->gmc.translate_further)
return;
if (level == AMDGPU_VM_PDB1) {
/* Set the block fragment size */
if (!(*flags & AMDGPU_PDE_PTE))
*flags |= AMDGPU_PDE_BFS(0x9);
} else if (level == AMDGPU_VM_PDB0) {
if (*flags & AMDGPU_PDE_PTE)
*flags &= ~AMDGPU_PDE_PTE;
else
*flags |= AMDGPU_PTE_TF;
}
}
static const struct amdgpu_gmc_funcs gmc_v10_0_gmc_funcs = {
.flush_gpu_tlb = gmc_v10_0_flush_gpu_tlb,
.emit_flush_gpu_tlb = gmc_v10_0_emit_flush_gpu_tlb,
.emit_pasid_mapping = gmc_v10_0_emit_pasid_mapping,
.get_vm_pte_flags = gmc_v10_0_get_vm_pte_flags,
.get_vm_pde = gmc_v10_0_get_vm_pde
};
static void gmc_v10_0_set_gmc_funcs(struct amdgpu_device *adev)
{
if (adev->gmc.gmc_funcs == NULL)
adev->gmc.gmc_funcs = &gmc_v10_0_gmc_funcs;
}
static int gmc_v10_0_early_init(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
gmc_v10_0_set_gmc_funcs(adev);
gmc_v10_0_set_irq_funcs(adev);
adev->gmc.shared_aperture_start = 0x2000000000000000ULL;
adev->gmc.shared_aperture_end =
adev->gmc.shared_aperture_start + (4ULL << 30) - 1;
adev->gmc.private_aperture_start = 0x1000000000000000ULL;
adev->gmc.private_aperture_end =
adev->gmc.private_aperture_start + (4ULL << 30) - 1;
return 0;
}
static int gmc_v10_0_late_init(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
unsigned vm_inv_eng[AMDGPU_MAX_VMHUBS] = { 4, 4 };
unsigned i;
for(i = 0; i < adev->num_rings; ++i) {
struct amdgpu_ring *ring = adev->rings[i];
unsigned vmhub = ring->funcs->vmhub;
ring->vm_inv_eng = vm_inv_eng[vmhub]++;
dev_info(adev->dev, "ring %u(%s) uses VM inv eng %u on hub %u\n",
ring->idx, ring->name, ring->vm_inv_eng,
ring->funcs->vmhub);
}
/* Engine 17 is used for GART flushes */
for(i = 0; i < AMDGPU_MAX_VMHUBS; ++i)
BUG_ON(vm_inv_eng[i] > 17);
return amdgpu_irq_get(adev, &adev->gmc.vm_fault, 0);
}
static void gmc_v10_0_vram_gtt_location(struct amdgpu_device *adev,
struct amdgpu_gmc *mc)
{
u64 base = 0;
if (!amdgpu_sriov_vf(adev))
base = gfxhub_v2_0_get_fb_location(adev);
amdgpu_gmc_vram_location(adev, &adev->gmc, base);
amdgpu_gmc_gart_location(adev, mc);
/* base offset of vram pages */
adev->vm_manager.vram_base_offset = gfxhub_v2_0_get_mc_fb_offset(adev);
}
/**
* gmc_v10_0_mc_init - initialize the memory controller driver params
*
* @adev: amdgpu_device pointer
*
* Look up the amount of vram, vram width, and decide how to place
* vram and gart within the GPU's physical address space.
* Returns 0 for success.
*/
static int gmc_v10_0_mc_init(struct amdgpu_device *adev)
{
int chansize, numchan;
if (!amdgpu_emu_mode)
adev->gmc.vram_width = amdgpu_atomfirmware_get_vram_width(adev);
else {
/* hard code vram_width for emulation */
chansize = 128;
numchan = 1;
adev->gmc.vram_width = numchan * chansize;
}
/* Could aper size report 0 ? */
adev->gmc.aper_base = pci_resource_start(adev->pdev, 0);
adev->gmc.aper_size = pci_resource_len(adev->pdev, 0);
/* size in MB on si */
adev->gmc.mc_vram_size =
adev->nbio_funcs->get_memsize(adev) * 1024ULL * 1024ULL;
adev->gmc.real_vram_size = adev->gmc.mc_vram_size;
adev->gmc.visible_vram_size = adev->gmc.aper_size;
/* In case the PCI BAR is larger than the actual amount of vram */
if (adev->gmc.visible_vram_size > adev->gmc.real_vram_size)
adev->gmc.visible_vram_size = adev->gmc.real_vram_size;
/* set the gart size */
if (amdgpu_gart_size == -1) {
switch (adev->asic_type) {
case CHIP_NAVI10:
case CHIP_NAVI14:
case CHIP_NAVI12:
default:
adev->gmc.gart_size = 512ULL << 20;
break;
}
} else
adev->gmc.gart_size = (u64)amdgpu_gart_size << 20;
gmc_v10_0_vram_gtt_location(adev, &adev->gmc);
return 0;
}
static int gmc_v10_0_gart_init(struct amdgpu_device *adev)
{
int r;
if (adev->gart.bo) {
WARN(1, "NAVI10 PCIE GART already initialized\n");
return 0;
}
/* Initialize common gart structure */
r = amdgpu_gart_init(adev);
if (r)
return r;
adev->gart.table_size = adev->gart.num_gpu_pages * 8;
adev->gart.gart_pte_flags = AMDGPU_PTE_MTYPE_NV10(MTYPE_UC) |
AMDGPU_PTE_EXECUTABLE;
return amdgpu_gart_table_vram_alloc(adev);
}
static unsigned gmc_v10_0_get_vbios_fb_size(struct amdgpu_device *adev)
{
u32 d1vga_control = RREG32_SOC15(DCE, 0, mmD1VGA_CONTROL);
unsigned size;
if (REG_GET_FIELD(d1vga_control, D1VGA_CONTROL, D1VGA_MODE_ENABLE)) {
size = 9 * 1024 * 1024; /* reserve 8MB for vga emulator and 1 MB for FB */
} else {
u32 viewport;
u32 pitch;
viewport = RREG32_SOC15(DCE, 0, mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION);
pitch = RREG32_SOC15(DCE, 0, mmHUBPREQ0_DCSURF_SURFACE_PITCH);
size = (REG_GET_FIELD(viewport,
HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_HEIGHT) *
REG_GET_FIELD(pitch, HUBPREQ0_DCSURF_SURFACE_PITCH, PITCH) *
4);
}
/* return 0 if the pre-OS buffer uses up most of vram */
if ((adev->gmc.real_vram_size - size) < (8 * 1024 * 1024)) {
DRM_ERROR("Warning: pre-OS buffer uses most of vram, \
be aware of gart table overwrite\n");
return 0;
}
return size;
}
static int gmc_v10_0_sw_init(void *handle)
{
int r;
int dma_bits;
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
gfxhub_v2_0_init(adev);
mmhub_v2_0_init(adev);
spin_lock_init(&adev->gmc.invalidate_lock);
adev->gmc.vram_type = amdgpu_atomfirmware_get_vram_type(adev);
switch (adev->asic_type) {
case CHIP_NAVI10:
case CHIP_NAVI14:
case CHIP_NAVI12:
adev->num_vmhubs = 2;
/*
* To fulfill 4-level page support,
* vm size is 256TB (48bit), maximum size of Navi10/Navi14/Navi12,
* block size 512 (9bit)
*/
amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
break;
default:
break;
}
/* This interrupt is VMC page fault.*/
r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VMC,
VMC_1_0__SRCID__VM_FAULT,
&adev->gmc.vm_fault);
r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_UTCL2,
UTCL2_1_0__SRCID__FAULT,
&adev->gmc.vm_fault);
if (r)
return r;
/*
* Set the internal MC address mask This is the max address of the GPU's
* internal address space.
*/
adev->gmc.mc_mask = 0xffffffffffffULL; /* 48 bit MC */
/*
* Reserve 8M stolen memory for navi10 like vega10
* TODO: will check if it's really needed on asic.
*/
if (amdgpu_emu_mode == 1)
adev->gmc.stolen_size = 0;
else
adev->gmc.stolen_size = 9 * 1024 *1024;
/*
* Set DMA mask + need_dma32 flags.
* PCIE - can handle 44-bits.
* IGP - can handle 44-bits
* PCI - dma32 for legacy pci gart, 44 bits on navi10
*/
adev->need_dma32 = false;
dma_bits = adev->need_dma32 ? 32 : 44;
r = pci_set_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits));
if (r) {
adev->need_dma32 = true;
dma_bits = 32;
printk(KERN_WARNING "amdgpu: No suitable DMA available.\n");
}
r = pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(dma_bits));
if (r) {
pci_set_consistent_dma_mask(adev->pdev, DMA_BIT_MASK(32));
printk(KERN_WARNING "amdgpu: No coherent DMA available.\n");
}
r = gmc_v10_0_mc_init(adev);
if (r)
return r;
adev->gmc.stolen_size = gmc_v10_0_get_vbios_fb_size(adev);
/* Memory manager */
r = amdgpu_bo_init(adev);
if (r)
return r;
r = gmc_v10_0_gart_init(adev);
if (r)
return r;
/*
* number of VMs
* VMID 0 is reserved for System
* amdgpu graphics/compute will use VMIDs 1-7
* amdkfd will use VMIDs 8-15
*/
adev->vm_manager.id_mgr[AMDGPU_GFXHUB_0].num_ids = AMDGPU_NUM_OF_VMIDS;
adev->vm_manager.id_mgr[AMDGPU_MMHUB_0].num_ids = AMDGPU_NUM_OF_VMIDS;
amdgpu_vm_manager_init(adev);
return 0;
}
/**
* gmc_v8_0_gart_fini - vm fini callback
*
* @adev: amdgpu_device pointer
*
* Tears down the driver GART/VM setup (CIK).
*/
static void gmc_v10_0_gart_fini(struct amdgpu_device *adev)
{
amdgpu_gart_table_vram_free(adev);
amdgpu_gart_fini(adev);
}
static int gmc_v10_0_sw_fini(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
amdgpu_vm_manager_fini(adev);
gmc_v10_0_gart_fini(adev);
amdgpu_gem_force_release(adev);
amdgpu_bo_fini(adev);
return 0;
}
static void gmc_v10_0_init_golden_registers(struct amdgpu_device *adev)
{
switch (adev->asic_type) {
case CHIP_NAVI10:
case CHIP_NAVI14:
case CHIP_NAVI12:
break;
default:
break;
}
}
/**
* gmc_v10_0_gart_enable - gart enable
*
* @adev: amdgpu_device pointer
*/
static int gmc_v10_0_gart_enable(struct amdgpu_device *adev)
{
int r;
bool value;
u32 tmp;
if (adev->gart.bo == NULL) {
dev_err(adev->dev, "No VRAM object for PCIE GART.\n");
return -EINVAL;
}
r = amdgpu_gart_table_vram_pin(adev);
if (r)
return r;
r = gfxhub_v2_0_gart_enable(adev);
if (r)
return r;
r = mmhub_v2_0_gart_enable(adev);
if (r)
return r;
tmp = RREG32_SOC15(HDP, 0, mmHDP_MISC_CNTL);
tmp |= HDP_MISC_CNTL__FLUSH_INVALIDATE_CACHE_MASK;
WREG32_SOC15(HDP, 0, mmHDP_MISC_CNTL, tmp);
tmp = RREG32_SOC15(HDP, 0, mmHDP_HOST_PATH_CNTL);
WREG32_SOC15(HDP, 0, mmHDP_HOST_PATH_CNTL, tmp);
/* Flush HDP after it is initialized */
adev->nbio_funcs->hdp_flush(adev, NULL);
value = (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS) ?
false : true;
gfxhub_v2_0_set_fault_enable_default(adev, value);
mmhub_v2_0_set_fault_enable_default(adev, value);
gmc_v10_0_flush_gpu_tlb(adev, 0, 0);
DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n",
(unsigned)(adev->gmc.gart_size >> 20),
(unsigned long long)amdgpu_bo_gpu_offset(adev->gart.bo));
adev->gart.ready = true;
return 0;
}
static int gmc_v10_0_hw_init(void *handle)
{
int r;
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
/* The sequence of these two function calls matters.*/
gmc_v10_0_init_golden_registers(adev);
r = gmc_v10_0_gart_enable(adev);
if (r)
return r;
return 0;
}
/**
* gmc_v10_0_gart_disable - gart disable
*
* @adev: amdgpu_device pointer
*
* This disables all VM page table.
*/
static void gmc_v10_0_gart_disable(struct amdgpu_device *adev)
{
gfxhub_v2_0_gart_disable(adev);
mmhub_v2_0_gart_disable(adev);
amdgpu_gart_table_vram_unpin(adev);
}
static int gmc_v10_0_hw_fini(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
if (amdgpu_sriov_vf(adev)) {
/* full access mode, so don't touch any GMC register */
DRM_DEBUG("For SRIOV client, shouldn't do anything.\n");
return 0;
}
amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0);
gmc_v10_0_gart_disable(adev);
return 0;
}
static int gmc_v10_0_suspend(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
gmc_v10_0_hw_fini(adev);
return 0;
}
static int gmc_v10_0_resume(void *handle)
{
int r;
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
r = gmc_v10_0_hw_init(adev);
if (r)
return r;
amdgpu_vmid_reset_all(adev);
return 0;
}
static bool gmc_v10_0_is_idle(void *handle)
{
/* MC is always ready in GMC v10.*/
return true;
}
static int gmc_v10_0_wait_for_idle(void *handle)
{
/* There is no need to wait for MC idle in GMC v10.*/
return 0;
}
static int gmc_v10_0_soft_reset(void *handle)
{
return 0;
}
static int gmc_v10_0_set_clockgating_state(void *handle,
enum amd_clockgating_state state)
{
int r;
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
r = mmhub_v2_0_set_clockgating(adev, state);
if (r)
return r;
return athub_v2_0_set_clockgating(adev, state);
}
static void gmc_v10_0_get_clockgating_state(void *handle, u32 *flags)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
mmhub_v2_0_get_clockgating(adev, flags);
athub_v2_0_get_clockgating(adev, flags);
}
static int gmc_v10_0_set_powergating_state(void *handle,
enum amd_powergating_state state)
{
return 0;
}
const struct amd_ip_funcs gmc_v10_0_ip_funcs = {
.name = "gmc_v10_0",
.early_init = gmc_v10_0_early_init,
.late_init = gmc_v10_0_late_init,
.sw_init = gmc_v10_0_sw_init,
.sw_fini = gmc_v10_0_sw_fini,
.hw_init = gmc_v10_0_hw_init,
.hw_fini = gmc_v10_0_hw_fini,
.suspend = gmc_v10_0_suspend,
.resume = gmc_v10_0_resume,
.is_idle = gmc_v10_0_is_idle,
.wait_for_idle = gmc_v10_0_wait_for_idle,
.soft_reset = gmc_v10_0_soft_reset,
.set_clockgating_state = gmc_v10_0_set_clockgating_state,
.set_powergating_state = gmc_v10_0_set_powergating_state,
.get_clockgating_state = gmc_v10_0_get_clockgating_state,
};
const struct amdgpu_ip_block_version gmc_v10_0_ip_block =
{
.type = AMD_IP_BLOCK_TYPE_GMC,
.major = 10,
.minor = 0,
.rev = 0,
.funcs = &gmc_v10_0_ip_funcs,
};
|