gmc_v9_0.c 34.4 KB
Newer Older
A
Alex Xie 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
/*
 * Copyright 2016 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>
24
#include <drm/drm_cache.h>
A
Alex Xie 已提交
25 26
#include "amdgpu.h"
#include "gmc_v9_0.h"
27
#include "amdgpu_atomfirmware.h"
28
#include "amdgpu_gem.h"
A
Alex Xie 已提交
29

30 31
#include "hdp/hdp_4_0_offset.h"
#include "hdp/hdp_4_0_sh_mask.h"
32
#include "gc/gc_9_0_sh_mask.h"
33 34
#include "dce/dce_12_0_offset.h"
#include "dce/dce_12_0_sh_mask.h"
35
#include "vega10_enum.h"
36
#include "mmhub/mmhub_1_0_offset.h"
37
#include "athub/athub_1_0_offset.h"
38
#include "oss/osssys_4_0_offset.h"
A
Alex Xie 已提交
39

40
#include "soc15.h"
A
Alex Xie 已提交
41
#include "soc15_common.h"
42
#include "umc/umc_6_0_sh_mask.h"
A
Alex Xie 已提交
43 44 45

#include "gfxhub_v1_0.h"
#include "mmhub_v1_0.h"
46
#include "gfxhub_v1_1.h"
A
Alex Xie 已提交
47

48 49
#include "ivsrcid/vmc/irqsrcs_vmc_1_0.h"

X
xinhui pan 已提交
50 51
#include "amdgpu_ras.h"

52 53 54 55 56 57 58 59
/* add these here since we already include dce12 headers and these are for DCN */
#define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION                                                          0x055d
#define mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION_BASE_IDX                                                 2
#define HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION__PRI_VIEWPORT_WIDTH__SHIFT                                        0x0
#define HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION__PRI_VIEWPORT_HEIGHT__SHIFT                                       0x10
#define HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION__PRI_VIEWPORT_WIDTH_MASK                                          0x00003FFFL
#define HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION__PRI_VIEWPORT_HEIGHT_MASK                                         0x3FFF0000L

A
Alex Xie 已提交
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76
/* XXX Move this macro to VEGA10 header file, which is like vid.h for VI.*/
#define AMDGPU_NUM_OF_VMIDS			8

static const u32 golden_settings_vega10_hdp[] =
{
	0xf64, 0x0fffffff, 0x00000000,
	0xf65, 0x0fffffff, 0x00000000,
	0xf66, 0x0fffffff, 0x00000000,
	0xf67, 0x0fffffff, 0x00000000,
	0xf68, 0x0fffffff, 0x00000000,
	0xf6a, 0x0fffffff, 0x00000000,
	0xf6b, 0x0fffffff, 0x00000000,
	0xf6c, 0x0fffffff, 0x00000000,
	0xf6d, 0x0fffffff, 0x00000000,
	0xf6e, 0x0fffffff, 0x00000000,
};

77
static const struct soc15_reg_golden golden_settings_mmhub_1_0_0[] =
78
{
79 80
	SOC15_REG_GOLDEN_VALUE(MMHUB, 0, mmDAGB1_WRCLI2, 0x00000007, 0xfe5fe0fa),
	SOC15_REG_GOLDEN_VALUE(MMHUB, 0, mmMMEA1_DRAM_WR_CLI2GRP_MAP0, 0x00000030, 0x55555565)
81 82
};

83
static const struct soc15_reg_golden golden_settings_athub_1_0_0[] =
84
{
85 86
	SOC15_REG_GOLDEN_VALUE(ATHUB, 0, mmRPB_ARB_CNTL, 0x0000ff00, 0x00000800),
	SOC15_REG_GOLDEN_VALUE(ATHUB, 0, mmRPB_ARB_CNTL2, 0x00ff00ff, 0x00080008)
87 88
};

X
xinhui pan 已提交
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
static const uint32_t ecc_umc_mcumc_ctrl_addrs[] = {
	(0x000143c0 + 0x00000000),
	(0x000143c0 + 0x00000800),
	(0x000143c0 + 0x00001000),
	(0x000143c0 + 0x00001800),
	(0x000543c0 + 0x00000000),
	(0x000543c0 + 0x00000800),
	(0x000543c0 + 0x00001000),
	(0x000543c0 + 0x00001800),
	(0x000943c0 + 0x00000000),
	(0x000943c0 + 0x00000800),
	(0x000943c0 + 0x00001000),
	(0x000943c0 + 0x00001800),
	(0x000d43c0 + 0x00000000),
	(0x000d43c0 + 0x00000800),
	(0x000d43c0 + 0x00001000),
	(0x000d43c0 + 0x00001800),
	(0x001143c0 + 0x00000000),
	(0x001143c0 + 0x00000800),
	(0x001143c0 + 0x00001000),
	(0x001143c0 + 0x00001800),
	(0x001543c0 + 0x00000000),
	(0x001543c0 + 0x00000800),
	(0x001543c0 + 0x00001000),
	(0x001543c0 + 0x00001800),
	(0x001943c0 + 0x00000000),
	(0x001943c0 + 0x00000800),
	(0x001943c0 + 0x00001000),
	(0x001943c0 + 0x00001800),
	(0x001d43c0 + 0x00000000),
	(0x001d43c0 + 0x00000800),
	(0x001d43c0 + 0x00001000),
	(0x001d43c0 + 0x00001800),
122 123
};

X
xinhui pan 已提交
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
static const uint32_t ecc_umc_mcumc_ctrl_mask_addrs[] = {
	(0x000143e0 + 0x00000000),
	(0x000143e0 + 0x00000800),
	(0x000143e0 + 0x00001000),
	(0x000143e0 + 0x00001800),
	(0x000543e0 + 0x00000000),
	(0x000543e0 + 0x00000800),
	(0x000543e0 + 0x00001000),
	(0x000543e0 + 0x00001800),
	(0x000943e0 + 0x00000000),
	(0x000943e0 + 0x00000800),
	(0x000943e0 + 0x00001000),
	(0x000943e0 + 0x00001800),
	(0x000d43e0 + 0x00000000),
	(0x000d43e0 + 0x00000800),
	(0x000d43e0 + 0x00001000),
	(0x000d43e0 + 0x00001800),
	(0x001143e0 + 0x00000000),
	(0x001143e0 + 0x00000800),
	(0x001143e0 + 0x00001000),
	(0x001143e0 + 0x00001800),
	(0x001543e0 + 0x00000000),
	(0x001543e0 + 0x00000800),
	(0x001543e0 + 0x00001000),
	(0x001543e0 + 0x00001800),
	(0x001943e0 + 0x00000000),
	(0x001943e0 + 0x00000800),
	(0x001943e0 + 0x00001000),
	(0x001943e0 + 0x00001800),
	(0x001d43e0 + 0x00000000),
	(0x001d43e0 + 0x00000800),
	(0x001d43e0 + 0x00001000),
	(0x001d43e0 + 0x00001800),
157 158
};

X
xinhui pan 已提交
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
static const uint32_t ecc_umc_mcumc_status_addrs[] = {
	(0x000143c2 + 0x00000000),
	(0x000143c2 + 0x00000800),
	(0x000143c2 + 0x00001000),
	(0x000143c2 + 0x00001800),
	(0x000543c2 + 0x00000000),
	(0x000543c2 + 0x00000800),
	(0x000543c2 + 0x00001000),
	(0x000543c2 + 0x00001800),
	(0x000943c2 + 0x00000000),
	(0x000943c2 + 0x00000800),
	(0x000943c2 + 0x00001000),
	(0x000943c2 + 0x00001800),
	(0x000d43c2 + 0x00000000),
	(0x000d43c2 + 0x00000800),
	(0x000d43c2 + 0x00001000),
	(0x000d43c2 + 0x00001800),
	(0x001143c2 + 0x00000000),
	(0x001143c2 + 0x00000800),
	(0x001143c2 + 0x00001000),
	(0x001143c2 + 0x00001800),
	(0x001543c2 + 0x00000000),
	(0x001543c2 + 0x00000800),
	(0x001543c2 + 0x00001000),
	(0x001543c2 + 0x00001800),
	(0x001943c2 + 0x00000000),
	(0x001943c2 + 0x00000800),
	(0x001943c2 + 0x00001000),
	(0x001943c2 + 0x00001800),
	(0x001d43c2 + 0x00000000),
	(0x001d43c2 + 0x00000800),
	(0x001d43c2 + 0x00001000),
	(0x001d43c2 + 0x00001800),
192 193
};

X
xinhui pan 已提交
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
static int gmc_v9_0_ecc_interrupt_state(struct amdgpu_device *adev,
		struct amdgpu_irq_src *src,
		unsigned type,
		enum amdgpu_interrupt_state state)
{
	u32 bits, i, tmp, reg;

	bits = 0x7f;

	switch (state) {
	case AMDGPU_IRQ_STATE_DISABLE:
		for (i = 0; i < ARRAY_SIZE(ecc_umc_mcumc_ctrl_addrs); i++) {
			reg = ecc_umc_mcumc_ctrl_addrs[i];
			tmp = RREG32(reg);
			tmp &= ~bits;
			WREG32(reg, tmp);
		}
		for (i = 0; i < ARRAY_SIZE(ecc_umc_mcumc_ctrl_mask_addrs); i++) {
			reg = ecc_umc_mcumc_ctrl_mask_addrs[i];
			tmp = RREG32(reg);
			tmp &= ~bits;
			WREG32(reg, tmp);
		}
		break;
	case AMDGPU_IRQ_STATE_ENABLE:
		for (i = 0; i < ARRAY_SIZE(ecc_umc_mcumc_ctrl_addrs); i++) {
			reg = ecc_umc_mcumc_ctrl_addrs[i];
			tmp = RREG32(reg);
			tmp |= bits;
			WREG32(reg, tmp);
		}
		for (i = 0; i < ARRAY_SIZE(ecc_umc_mcumc_ctrl_mask_addrs); i++) {
			reg = ecc_umc_mcumc_ctrl_mask_addrs[i];
			tmp = RREG32(reg);
			tmp |= bits;
			WREG32(reg, tmp);
		}
		break;
	default:
		break;
	}

	return 0;
}

static int gmc_v9_0_process_ras_data_cb(struct amdgpu_device *adev,
		struct amdgpu_iv_entry *entry)
{
242
	kgd2kfd_set_sram_ecc_flag(adev->kfd.dev);
X
xinhui pan 已提交
243 244 245 246 247 248 249 250
	amdgpu_ras_reset_gpu(adev, 0);
	return AMDGPU_RAS_UE;
}

static int gmc_v9_0_process_ecc_irq(struct amdgpu_device *adev,
		struct amdgpu_irq_src *source,
		struct amdgpu_iv_entry *entry)
{
251
	struct ras_common_if *ras_if = adev->gmc.ras_if;
X
xinhui pan 已提交
252 253 254
	struct ras_dispatch_if ih_data = {
		.entry = entry,
	};
255 256 257 258 259 260

	if (!ras_if)
		return 0;

	ih_data.head = *ras_if;

X
xinhui pan 已提交
261 262 263 264
	amdgpu_ras_interrupt_dispatch(adev, &ih_data);
	return 0;
}

A
Alex Xie 已提交
265 266 267 268 269 270
static int gmc_v9_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;
271
	u32 tmp, reg, bits, i, j;
A
Alex Xie 已提交
272

273 274 275 276 277 278 279 280
	bits = VM_CONTEXT1_CNTL__RANGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
		VM_CONTEXT1_CNTL__DUMMY_PAGE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
		VM_CONTEXT1_CNTL__PDE0_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
		VM_CONTEXT1_CNTL__VALID_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
		VM_CONTEXT1_CNTL__READ_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
		VM_CONTEXT1_CNTL__WRITE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK |
		VM_CONTEXT1_CNTL__EXECUTE_PROTECTION_FAULT_ENABLE_INTERRUPT_MASK;

A
Alex Xie 已提交
281 282
	switch (state) {
	case AMDGPU_IRQ_STATE_DISABLE:
283 284 285 286 287 288 289 290
		for (j = 0; j < AMDGPU_MAX_VMHUBS; j++) {
			hub = &adev->vmhub[j];
			for (i = 0; i < 16; i++) {
				reg = hub->vm_context0_cntl + i;
				tmp = RREG32(reg);
				tmp &= ~bits;
				WREG32(reg, tmp);
			}
A
Alex Xie 已提交
291 292 293
		}
		break;
	case AMDGPU_IRQ_STATE_ENABLE:
294 295 296 297 298 299 300 301
		for (j = 0; j < AMDGPU_MAX_VMHUBS; j++) {
			hub = &adev->vmhub[j];
			for (i = 0; i < 16; i++) {
				reg = hub->vm_context0_cntl + i;
				tmp = RREG32(reg);
				tmp |= bits;
				WREG32(reg, tmp);
			}
A
Alex Xie 已提交
302 303 304 305 306 307 308 309 310 311 312 313
		}
	default:
		break;
	}

	return 0;
}

static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev,
				struct amdgpu_irq_src *source,
				struct amdgpu_iv_entry *entry)
{
314
	struct amdgpu_vmhub *hub = &adev->vmhub[entry->vmid_src];
315
	bool retry_fault = !!(entry->src_data[1] & 0x80);
316
	uint32_t status = 0;
A
Alex Xie 已提交
317 318 319 320 321
	u64 addr;

	addr = (u64)entry->src_data[0] << 12;
	addr |= ((u64)entry->src_data[1] & 0xf) << 44;

322 323
	if (retry_fault && amdgpu_gmc_filter_faults(adev, addr, entry->pasid,
						    entry->timestamp))
324 325
		return 1; /* This also prevents sending it to KFD */

326
	/* If it's the first fault for this address, process it normally */
327
	if (!amdgpu_sriov_vf(adev)) {
328 329
		status = RREG32(hub->vm_l2_pro_fault_status);
		WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1);
330
	}
A
Alex Xie 已提交
331

332
	if (printk_ratelimit()) {
333
		struct amdgpu_task_info task_info;
334

335
		memset(&task_info, 0, sizeof(struct amdgpu_task_info));
336 337
		amdgpu_vm_get_task_info(adev, entry->pasid, &task_info);

338
		dev_err(adev->dev,
339 340
			"[%s] %s page fault (src_id:%u ring:%u vmid:%u "
			"pasid:%u, for process %s pid %d thread %s pid %d)\n",
341
			entry->vmid_src ? "mmhub" : "gfxhub",
342
			retry_fault ? "retry" : "no-retry",
343
			entry->src_id, entry->ring_id, entry->vmid,
344 345
			entry->pasid, task_info.process_name, task_info.tgid,
			task_info.task_name, task_info.pid);
346
		dev_err(adev->dev, "  in page starting at address 0x%016llx from %d\n",
347 348 349 350 351
			addr, entry->client_id);
		if (!amdgpu_sriov_vf(adev))
			dev_err(adev->dev,
				"VM_L2_PROTECTION_FAULT_STATUS:0x%08X\n",
				status);
352
	}
A
Alex Xie 已提交
353 354 355 356 357 358 359 360 361

	return 0;
}

static const struct amdgpu_irq_src_funcs gmc_v9_0_irq_funcs = {
	.set = gmc_v9_0_vm_fault_interrupt_state,
	.process = gmc_v9_0_process_interrupt,
};

X
xinhui pan 已提交
362 363 364 365 366 367

static const struct amdgpu_irq_src_funcs gmc_v9_0_ecc_funcs = {
	.set = gmc_v9_0_ecc_interrupt_state,
	.process = gmc_v9_0_process_ecc_irq,
};

A
Alex Xie 已提交
368 369
static void gmc_v9_0_set_irq_funcs(struct amdgpu_device *adev)
{
370 371
	adev->gmc.vm_fault.num_types = 1;
	adev->gmc.vm_fault.funcs = &gmc_v9_0_irq_funcs;
X
xinhui pan 已提交
372 373 374

	adev->gmc.ecc_irq.num_types = 1;
	adev->gmc.ecc_irq.funcs = &gmc_v9_0_ecc_funcs;
A
Alex Xie 已提交
375 376
}

377 378
static uint32_t gmc_v9_0_get_invalidate_req(unsigned int vmid,
					uint32_t flush_type)
379 380 381 382
{
	u32 req = 0;

	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ,
383
			    PER_VMID_INVALIDATE_REQ, 1 << vmid);
384
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, FLUSH_TYPE, flush_type);
385 386 387 388 389 390 391 392 393 394 395
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PTES, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE0, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE1, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE2, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L1_PTES, 1);
	req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ,
			    CLEAR_PROTECTION_FAULT_STATUS_ADDR,	0);

	return req;
}

A
Alex Xie 已提交
396 397 398 399 400 401 402 403
/*
 * 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.
 */

/**
404
 * gmc_v9_0_flush_gpu_tlb - tlb flush with certain type
A
Alex Xie 已提交
405 406 407
 *
 * @adev: amdgpu_device pointer
 * @vmid: vm instance to flush
408
 * @flush_type: the flush type
A
Alex Xie 已提交
409
 *
410
 * Flush the TLB for the requested page table using certain type.
A
Alex Xie 已提交
411
 */
412
static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev,
413
				uint32_t vmid, uint32_t flush_type)
A
Alex Xie 已提交
414 415 416 417 418 419
{
	const unsigned eng = 17;
	unsigned i, j;

	for (i = 0; i < AMDGPU_MAX_VMHUBS; ++i) {
		struct amdgpu_vmhub *hub = &adev->vmhub[i];
420
		u32 tmp = gmc_v9_0_get_invalidate_req(vmid, flush_type);
A
Alex Xie 已提交
421

422 423 424 425 426 427
		/* This is necessary for a HW workaround under SRIOV as well
		 * as GFXOFF under bare metal
		 */
		if (adev->gfx.kiq.ring.sched.ready &&
		    (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) &&
		    !adev->in_gpu_reset) {
428 429 430 431 432 433
			uint32_t req = hub->vm_inv_eng0_req + eng;
			uint32_t ack = hub->vm_inv_eng0_ack + eng;

			amdgpu_virt_kiq_reg_write_reg_wait(adev, req, ack, tmp,
							   1 << vmid);
			continue;
434
		}
435 436

		spin_lock(&adev->gmc.invalidate_lock);
437
		WREG32_NO_KIQ(hub->vm_inv_eng0_req + eng, tmp);
A
Alex Xie 已提交
438
		for (j = 0; j < adev->usec_timeout; j++) {
439
			tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack + eng);
440
			if (tmp & (1 << vmid))
A
Alex Xie 已提交
441 442 443
				break;
			udelay(1);
		}
444
		spin_unlock(&adev->gmc.invalidate_lock);
445 446 447
		if (j < adev->usec_timeout)
			continue;

A
Alex Xie 已提交
448 449 450 451
		DRM_ERROR("Timeout waiting for VM flush ACK!\n");
	}
}

452
static uint64_t gmc_v9_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring,
453
					    unsigned vmid, uint64_t pd_addr)
454
{
455 456
	struct amdgpu_device *adev = ring->adev;
	struct amdgpu_vmhub *hub = &adev->vmhub[ring->funcs->vmhub];
457
	uint32_t req = gmc_v9_0_get_invalidate_req(vmid, 0);
458 459 460 461 462 463 464 465
	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));

466 467 468
	amdgpu_ring_emit_reg_write_reg_wait(ring, hub->vm_inv_eng0_req + eng,
					    hub->vm_inv_eng0_ack + eng,
					    req, 1 << vmid);
469

470 471 472
	return pd_addr;
}

473 474 475 476 477 478 479 480 481 482 483 484 485 486
static void gmc_v9_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)
		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);
}

487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505
/*
 * PTE format on VEGA 10:
 * 63:59 reserved
 * 58:57 mtype
 * 56 F
 * 55 L
 * 54 P
 * 53 SW
 * 52 T
 * 50:48 reserved
 * 47:12 4k physical page base address
 * 11:7 fragment
 * 6 write
 * 5 read
 * 4 exe
 * 3 Z
 * 2 snooped
 * 1 system
 * 0 valid
A
Alex Xie 已提交
506
 *
507 508 509 510 511 512 513 514 515 516
 * PDE format on VEGA 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
A
Alex Xie 已提交
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
 */

static uint64_t gmc_v9_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(MTYPE_NC);
		break;
	case AMDGPU_VM_MTYPE_NC:
		pte_flag |= AMDGPU_PTE_MTYPE(MTYPE_NC);
		break;
	case AMDGPU_VM_MTYPE_WC:
		pte_flag |= AMDGPU_PTE_MTYPE(MTYPE_WC);
		break;
	case AMDGPU_VM_MTYPE_CC:
		pte_flag |= AMDGPU_PTE_MTYPE(MTYPE_CC);
		break;
	case AMDGPU_VM_MTYPE_UC:
		pte_flag |= AMDGPU_PTE_MTYPE(MTYPE_UC);
		break;
	default:
		pte_flag |= AMDGPU_PTE_MTYPE(MTYPE_NC);
		break;
	}

	if (flags & AMDGPU_VM_PAGE_PRT)
		pte_flag |= AMDGPU_PTE_PRT;

	return pte_flag;
}

559 560
static void gmc_v9_0_get_vm_pde(struct amdgpu_device *adev, int level,
				uint64_t *addr, uint64_t *flags)
A
Alex Xie 已提交
561
{
562
	if (!(*flags & AMDGPU_PDE_PTE) && !(*flags & AMDGPU_PTE_SYSTEM))
563
		*addr = adev->vm_manager.vram_base_offset + *addr -
564
			adev->gmc.vram_start;
565
	BUG_ON(*addr & 0xFFFF00000000003FULL);
566

567
	if (!adev->gmc.translate_further)
568 569 570 571 572 573 574 575 576 577 578 579 580
		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;
	}
A
Alex Xie 已提交
581 582
}

583 584
static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = {
	.flush_gpu_tlb = gmc_v9_0_flush_gpu_tlb,
585
	.emit_flush_gpu_tlb = gmc_v9_0_emit_flush_gpu_tlb,
586
	.emit_pasid_mapping = gmc_v9_0_emit_pasid_mapping,
587 588
	.get_vm_pte_flags = gmc_v9_0_get_vm_pte_flags,
	.get_vm_pde = gmc_v9_0_get_vm_pde
A
Alex Xie 已提交
589 590
};

591
static void gmc_v9_0_set_gmc_funcs(struct amdgpu_device *adev)
A
Alex Xie 已提交
592
{
593
	adev->gmc.gmc_funcs = &gmc_v9_0_gmc_funcs;
A
Alex Xie 已提交
594 595 596 597 598 599
}

static int gmc_v9_0_early_init(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

600
	gmc_v9_0_set_gmc_funcs(adev);
A
Alex Xie 已提交
601 602
	gmc_v9_0_set_irq_funcs(adev);

603 604 605
	adev->gmc.shared_aperture_start = 0x2000000000000000ULL;
	adev->gmc.shared_aperture_end =
		adev->gmc.shared_aperture_start + (4ULL << 30) - 1;
606
	adev->gmc.private_aperture_start = 0x1000000000000000ULL;
607 608
	adev->gmc.private_aperture_end =
		adev->gmc.private_aperture_start + (4ULL << 30) - 1;
609

A
Alex Xie 已提交
610 611 612
	return 0;
}

613 614 615 616 617 618 619 620 621 622 623 624 625
static bool gmc_v9_0_keep_stolen_memory(struct amdgpu_device *adev)
{

	/*
	 * TODO:
	 * Currently there is a bug where some memory client outside
	 * of the driver writes to first 8M of VRAM on S3 resume,
	 * this overrides GART which by default gets placed in first 8M and
	 * causes VM_FAULTS once GTT is accessed.
	 * Keep the stolen memory reservation until the while this is not solved.
	 * Also check code in gmc_v9_0_get_vbios_fb_size and gmc_v9_0_late_init
	 */
	switch (adev->asic_type) {
626 627
	case CHIP_VEGA10:
		return true;
628 629 630 631
	case CHIP_RAVEN:
	case CHIP_VEGA12:
	case CHIP_VEGA20:
	default:
632
		return false;
633 634 635
	}
}

636
static int gmc_v9_0_allocate_vm_inv_eng(struct amdgpu_device *adev)
A
Alex Xie 已提交
637
{
638 639 640
	struct amdgpu_ring *ring;
	unsigned vm_inv_engs[AMDGPU_MAX_VMHUBS] =
		{GFXHUB_FREE_VM_INV_ENGS_BITMAP, MMHUB_FREE_VM_INV_ENGS_BITMAP};
641
	unsigned i;
642
	unsigned vmhub, inv_eng;
643

644 645 646
	for (i = 0; i < adev->num_rings; ++i) {
		ring = adev->rings[i];
		vmhub = ring->funcs->vmhub;
647

648 649 650 651 652 653 654 655
		inv_eng = ffs(vm_inv_engs[vmhub]);
		if (!inv_eng) {
			dev_err(adev->dev, "no VM inv eng for ring %s\n",
				ring->name);
			return -EINVAL;
		}

		ring->vm_inv_eng = inv_eng - 1;
656
		vm_inv_engs[vmhub] &= ~(1 << ring->vm_inv_eng);
657

658 659
		dev_info(adev->dev, "ring %s uses VM inv eng %u on hub %u\n",
			 ring->name, ring->vm_inv_eng, ring->funcs->vmhub);
660 661
	}

662 663 664
	return 0;
}

X
xinhui pan 已提交
665
static int gmc_v9_0_ecc_late_init(void *handle)
666 667
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
X
xinhui pan 已提交
668 669 670 671 672 673 674 675 676 677 678 679 680 681
	struct ras_common_if **ras_if = &adev->gmc.ras_if;
	struct ras_ih_if ih_info = {
		.cb = gmc_v9_0_process_ras_data_cb,
	};
	struct ras_fs_if fs_info = {
		.sysfs_name = "umc_err_count",
		.debugfs_name = "umc_err_inject",
	};
	struct ras_common_if ras_block = {
		.block = AMDGPU_RAS_BLOCK__UMC,
		.type = AMDGPU_RAS_ERROR__MULTI_UNCORRECTABLE,
		.sub_block_index = 0,
		.name = "umc",
	};
682 683
	int r;

X
xinhui pan 已提交
684
	if (!amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC)) {
685
		amdgpu_ras_feature_enable_on_boot(adev, &ras_block, 0);
X
xinhui pan 已提交
686 687
		return 0;
	}
X
xinhui pan 已提交
688
	/* handle resume path. */
X
xinhui pan 已提交
689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705
	if (*ras_if) {
		/* resend ras TA enable cmd during resume.
		 * prepare to handle failure.
		 */
		ih_info.head = **ras_if;
		r = amdgpu_ras_feature_enable_on_boot(adev, *ras_if, 1);
		if (r) {
			if (r == -EAGAIN) {
				/* request a gpu reset. will run again. */
				amdgpu_ras_request_reset_on_boot(adev,
						AMDGPU_RAS_BLOCK__UMC);
				return 0;
			}
			/* fail to enable ras, cleanup all. */
			goto irq;
		}
		/* enable successfully. continue. */
X
xinhui pan 已提交
706
		goto resume;
X
xinhui pan 已提交
707
	}
X
xinhui pan 已提交
708 709 710 711 712 713 714

	*ras_if = kmalloc(sizeof(**ras_if), GFP_KERNEL);
	if (!*ras_if)
		return -ENOMEM;

	**ras_if = ras_block;

715
	r = amdgpu_ras_feature_enable_on_boot(adev, *ras_if, 1);
716 717 718 719 720 721
	if (r) {
		if (r == -EAGAIN) {
			amdgpu_ras_request_reset_on_boot(adev,
					AMDGPU_RAS_BLOCK__UMC);
			r = 0;
		}
X
xinhui pan 已提交
722
		goto feature;
723
	}
X
xinhui pan 已提交
724 725 726 727 728 729 730 731 732 733 734 735 736 737 738

	ih_info.head = **ras_if;
	fs_info.head = **ras_if;

	r = amdgpu_ras_interrupt_add_handler(adev, &ih_info);
	if (r)
		goto interrupt;

	r = amdgpu_ras_debugfs_create(adev, &fs_info);
	if (r)
		goto debugfs;

	r = amdgpu_ras_sysfs_create(adev, &fs_info);
	if (r)
		goto sysfs;
X
xinhui pan 已提交
739
resume:
X
xinhui pan 已提交
740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755
	r = amdgpu_irq_get(adev, &adev->gmc.ecc_irq, 0);
	if (r)
		goto irq;

	return 0;
irq:
	amdgpu_ras_sysfs_remove(adev, *ras_if);
sysfs:
	amdgpu_ras_debugfs_remove(adev, *ras_if);
debugfs:
	amdgpu_ras_interrupt_remove_handler(adev, &ih_info);
interrupt:
	amdgpu_ras_feature_enable(adev, *ras_if, 0);
feature:
	kfree(*ras_if);
	*ras_if = NULL;
756
	return r;
X
xinhui pan 已提交
757 758 759
}


760 761 762
static int gmc_v9_0_late_init(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
763
	bool r;
764 765 766 767 768 769 770

	if (!gmc_v9_0_keep_stolen_memory(adev))
		amdgpu_bo_late_init(adev);

	r = gmc_v9_0_allocate_vm_inv_eng(adev);
	if (r)
		return r;
771 772 773 774 775 776 777 778 779 780 781 782 783
	/* Check if ecc is available */
	if (!amdgpu_sriov_vf(adev)) {
		switch (adev->asic_type) {
		case CHIP_VEGA10:
		case CHIP_VEGA20:
			r = amdgpu_atomfirmware_mem_ecc_supported(adev);
			if (!r) {
				DRM_INFO("ECC is not present.\n");
				if (adev->df_funcs->enable_ecc_force_par_wr_rmw)
					adev->df_funcs->enable_ecc_force_par_wr_rmw(adev, false);
			} else {
				DRM_INFO("ECC is active.\n");
			}
784

785 786 787 788 789 790 791 792 793
			r = amdgpu_atomfirmware_sram_ecc_supported(adev);
			if (!r) {
				DRM_INFO("SRAM ECC is not present.\n");
			} else {
				DRM_INFO("SRAM ECC is active.\n");
			}
			break;
		default:
			break;
794
		}
795 796
	}

X
xinhui pan 已提交
797 798 799 800
	r = gmc_v9_0_ecc_late_init(handle);
	if (r)
		return r;

801
	return amdgpu_irq_get(adev, &adev->gmc.vm_fault, 0);
A
Alex Xie 已提交
802 803 804
}

static void gmc_v9_0_vram_gtt_location(struct amdgpu_device *adev,
805
					struct amdgpu_gmc *mc)
A
Alex Xie 已提交
806
{
807 808 809
	u64 base = 0;
	if (!amdgpu_sriov_vf(adev))
		base = mmhub_v1_0_get_fb_location(adev);
810 811
	/* add the xgmi offset of the physical node */
	base += adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size;
812
	amdgpu_gmc_vram_location(adev, mc, base);
813
	amdgpu_gmc_gart_location(adev, mc);
814 815
	if (!amdgpu_sriov_vf(adev))
		amdgpu_gmc_agp_location(adev, mc);
816
	/* base offset of vram pages */
817
	adev->vm_manager.vram_base_offset = gfxhub_v1_0_get_mc_fb_offset(adev);
818 819 820 821

	/* XXX: add the xgmi offset of the physical node? */
	adev->vm_manager.vram_base_offset +=
		adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size;
A
Alex Xie 已提交
822 823 824 825 826 827 828 829 830 831 832 833 834 835
}

/**
 * gmc_v9_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_v9_0_mc_init(struct amdgpu_device *adev)
{
	int chansize, numchan;
836
	int r;
A
Alex Xie 已提交
837

838 839
	if (amdgpu_emu_mode != 1)
		adev->gmc.vram_width = amdgpu_atomfirmware_get_vram_width(adev);
840
	if (!adev->gmc.vram_width) {
841
		/* hbm memory channel size */
842 843 844 845
		if (adev->flags & AMD_IS_APU)
			chansize = 64;
		else
			chansize = 128;
846

847
		numchan = adev->df_funcs->get_hbm_channel_number(adev);
848
		adev->gmc.vram_width = numchan * chansize;
A
Alex Xie 已提交
849 850 851
	}

	/* size in MB on si */
852
	adev->gmc.mc_vram_size =
853
		adev->nbio_funcs->get_memsize(adev) * 1024ULL * 1024ULL;
854
	adev->gmc.real_vram_size = adev->gmc.mc_vram_size;
855 856 857 858 859 860

	if (!(adev->flags & AMD_IS_APU)) {
		r = amdgpu_device_resize_fb_bar(adev);
		if (r)
			return r;
	}
861 862
	adev->gmc.aper_base = pci_resource_start(adev->pdev, 0);
	adev->gmc.aper_size = pci_resource_len(adev->pdev, 0);
A
Alex Xie 已提交
863

864 865 866 867 868 869
#ifdef CONFIG_X86_64
	if (adev->flags & AMD_IS_APU) {
		adev->gmc.aper_base = gfxhub_v1_0_get_mc_fb_offset(adev);
		adev->gmc.aper_size = adev->gmc.real_vram_size;
	}
#endif
A
Alex Xie 已提交
870
	/* In case the PCI BAR is larger than the actual amount of vram */
871 872 873
	adev->gmc.visible_vram_size = adev->gmc.aper_size;
	if (adev->gmc.visible_vram_size > adev->gmc.real_vram_size)
		adev->gmc.visible_vram_size = adev->gmc.real_vram_size;
A
Alex Xie 已提交
874

875 876 877 878
	/* set the gart size */
	if (amdgpu_gart_size == -1) {
		switch (adev->asic_type) {
		case CHIP_VEGA10:  /* all engines support GPUVM */
879
		case CHIP_VEGA12:  /* all engines support GPUVM */
F
Feifei Xu 已提交
880
		case CHIP_VEGA20:
881
		default:
882
			adev->gmc.gart_size = 512ULL << 20;
883 884
			break;
		case CHIP_RAVEN:   /* DCE SG support */
885
			adev->gmc.gart_size = 1024ULL << 20;
886 887 888
			break;
		}
	} else {
889
		adev->gmc.gart_size = (u64)amdgpu_gart_size << 20;
890 891
	}

892
	gmc_v9_0_vram_gtt_location(adev, &adev->gmc);
A
Alex Xie 已提交
893 894 895 896 897 898 899 900

	return 0;
}

static int gmc_v9_0_gart_init(struct amdgpu_device *adev)
{
	int r;

901
	if (adev->gart.bo) {
A
Alex Xie 已提交
902 903 904 905 906 907 908 909 910 911 912 913 914
		WARN(1, "VEGA10 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(MTYPE_UC) |
				 AMDGPU_PTE_EXECUTABLE;
	return amdgpu_gart_table_vram_alloc(adev);
}

915 916 917 918 919
static unsigned gmc_v9_0_get_vbios_fb_size(struct amdgpu_device *adev)
{
	u32 d1vga_control = RREG32_SOC15(DCE, 0, mmD1VGA_CONTROL);
	unsigned size;

920 921 922 923
	/*
	 * TODO Remove once GART corruption is resolved
	 * Check related code in gmc_v9_0_sw_fini
	 * */
924 925
	if (gmc_v9_0_keep_stolen_memory(adev))
		return 9 * 1024 * 1024;
926

927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942
	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;

		switch (adev->asic_type) {
		case CHIP_RAVEN:
			viewport = RREG32_SOC15(DCE, 0, mmHUBP0_DCSURF_PRI_VIEWPORT_DIMENSION);
			size = (REG_GET_FIELD(viewport,
					      HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_HEIGHT) *
				REG_GET_FIELD(viewport,
					      HUBP0_DCSURF_PRI_VIEWPORT_DIMENSION, PRI_VIEWPORT_WIDTH) *
				4);
			break;
		case CHIP_VEGA10:
		case CHIP_VEGA12:
943
		case CHIP_VEGA20:
944 945 946 947 948 949 950 951 952 953 954
		default:
			viewport = RREG32_SOC15(DCE, 0, mmSCL0_VIEWPORT_SIZE);
			size = (REG_GET_FIELD(viewport, SCL0_VIEWPORT_SIZE, VIEWPORT_HEIGHT) *
				REG_GET_FIELD(viewport, SCL0_VIEWPORT_SIZE, VIEWPORT_WIDTH) *
				4);
			break;
		}
	}
	/* return 0 if the pre-OS buffer uses up most of vram */
	if ((adev->gmc.real_vram_size - size) < (8 * 1024 * 1024))
		return 0;
955

956 957 958
	return size;
}

A
Alex Xie 已提交
959 960 961 962 963 964
static int gmc_v9_0_sw_init(void *handle)
{
	int r;
	int dma_bits;
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

965
	gfxhub_v1_0_init(adev);
966
	mmhub_v1_0_init(adev);
967

968
	spin_lock_init(&adev->gmc.invalidate_lock);
A
Alex Xie 已提交
969

970
	adev->gmc.vram_type = amdgpu_atomfirmware_get_vram_type(adev);
971 972
	switch (adev->asic_type) {
	case CHIP_RAVEN:
973
		if (adev->rev_id == 0x0 || adev->rev_id == 0x1) {
974
			amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
975 976 977
		} else {
			/* vm_size is 128TB + 512GB for legacy 3-level page support */
			amdgpu_vm_adjust_size(adev, 128 * 1024 + 512, 9, 2, 48);
978
			adev->gmc.translate_further =
979 980
				adev->vm_manager.num_level > 1;
		}
981 982
		break;
	case CHIP_VEGA10:
983
	case CHIP_VEGA12:
F
Feifei Xu 已提交
984
	case CHIP_VEGA20:
985 986 987 988 989
		/*
		 * To fulfill 4-level page support,
		 * vm size is 256TB (48bit), maximum size of Vega10,
		 * block size 512 (9bit)
		 */
990 991 992 993 994
		/* sriov restrict max_pfn below AMDGPU_GMC_HOLE */
		if (amdgpu_sriov_vf(adev))
			amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 47);
		else
			amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48);
995 996 997
		break;
	default:
		break;
A
Alex Xie 已提交
998 999 1000
	}

	/* This interrupt is VMC page fault.*/
1001
	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VMC, VMC_1_0__SRCID__VM_FAULT,
1002
				&adev->gmc.vm_fault);
1003 1004 1005
	if (r)
		return r;

1006
	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_UTCL2, UTCL2_1_0__SRCID__FAULT,
1007
				&adev->gmc.vm_fault);
A
Alex Xie 已提交
1008 1009 1010 1011

	if (r)
		return r;

X
xinhui pan 已提交
1012 1013 1014 1015 1016 1017
	/* interrupt sent to DF. */
	r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_DF, 0,
			&adev->gmc.ecc_irq);
	if (r)
		return r;

A
Alex Xie 已提交
1018 1019 1020 1021
	/* Set the internal MC address mask
	 * This is the max address of the GPU's
	 * internal address space.
	 */
1022
	adev->gmc.mc_mask = 0xffffffffffffULL; /* 48 bit MC */
A
Alex Xie 已提交
1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041

	/* 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 vega10
	 */
	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");
	}
1042
	adev->need_swiotlb = drm_need_swiotlb(dma_bits);
A
Alex Xie 已提交
1043

1044
	if (adev->gmc.xgmi.supported) {
1045 1046 1047 1048 1049
		r = gfxhub_v1_1_get_xgmi_info(adev);
		if (r)
			return r;
	}

A
Alex Xie 已提交
1050 1051 1052 1053
	r = gmc_v9_0_mc_init(adev);
	if (r)
		return r;

1054 1055
	adev->gmc.stolen_size = gmc_v9_0_get_vbios_fb_size(adev);

A
Alex Xie 已提交
1056 1057 1058 1059 1060 1061 1062 1063 1064
	/* Memory manager */
	r = amdgpu_bo_init(adev);
	if (r)
		return r;

	r = gmc_v9_0_gart_init(adev);
	if (r)
		return r;

1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076
	/*
	 * 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].num_ids = AMDGPU_NUM_OF_VMIDS;
	adev->vm_manager.id_mgr[AMDGPU_MMHUB].num_ids = AMDGPU_NUM_OF_VMIDS;

	amdgpu_vm_manager_init(adev);

	return 0;
A
Alex Xie 已提交
1077 1078 1079 1080 1081 1082
}

static int gmc_v9_0_sw_fini(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

X
xinhui pan 已提交
1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098
	if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC) &&
			adev->gmc.ras_if) {
		struct ras_common_if *ras_if = adev->gmc.ras_if;
		struct ras_ih_if ih_info = {
			.head = *ras_if,
		};

		/*remove fs first*/
		amdgpu_ras_debugfs_remove(adev, ras_if);
		amdgpu_ras_sysfs_remove(adev, ras_if);
		/*remove the IH*/
		amdgpu_ras_interrupt_remove_handler(adev, &ih_info);
		amdgpu_ras_feature_enable(adev, ras_if, 0);
		kfree(ras_if);
	}

1099
	amdgpu_gem_force_release(adev);
1100
	amdgpu_vm_manager_fini(adev);
1101

1102 1103
	if (gmc_v9_0_keep_stolen_memory(adev))
		amdgpu_bo_free_kernel(&adev->stolen_vga_memory, NULL, NULL);
1104

1105
	amdgpu_gart_table_vram_free(adev);
A
Alex Xie 已提交
1106
	amdgpu_bo_fini(adev);
1107
	amdgpu_gart_fini(adev);
A
Alex Xie 已提交
1108 1109 1110 1111 1112 1113

	return 0;
}

static void gmc_v9_0_init_golden_registers(struct amdgpu_device *adev)
{
1114

A
Alex Xie 已提交
1115 1116
	switch (adev->asic_type) {
	case CHIP_VEGA10:
1117 1118 1119
		if (amdgpu_virt_support_skip_setting(adev))
			break;
		/* fall through */
F
Feifei Xu 已提交
1120
	case CHIP_VEGA20:
1121
		soc15_program_register_sequence(adev,
1122
						golden_settings_mmhub_1_0_0,
1123
						ARRAY_SIZE(golden_settings_mmhub_1_0_0));
1124
		soc15_program_register_sequence(adev,
1125
						golden_settings_athub_1_0_0,
1126
						ARRAY_SIZE(golden_settings_athub_1_0_0));
A
Alex Xie 已提交
1127
		break;
1128 1129
	case CHIP_VEGA12:
		break;
1130
	case CHIP_RAVEN:
1131
		soc15_program_register_sequence(adev,
1132
						golden_settings_athub_1_0_0,
1133
						ARRAY_SIZE(golden_settings_athub_1_0_0));
1134
		break;
A
Alex Xie 已提交
1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150
	default:
		break;
	}
}

/**
 * gmc_v9_0_gart_enable - gart enable
 *
 * @adev: amdgpu_device pointer
 */
static int gmc_v9_0_gart_enable(struct amdgpu_device *adev)
{
	int r;
	bool value;
	u32 tmp;

1151 1152 1153
	amdgpu_device_program_register_sequence(adev,
						golden_settings_vega10_hdp,
						ARRAY_SIZE(golden_settings_vega10_hdp));
A
Alex Xie 已提交
1154

1155
	if (adev->gart.bo == NULL) {
A
Alex Xie 已提交
1156 1157 1158
		dev_err(adev->dev, "No VRAM object for PCIE GART.\n");
		return -EINVAL;
	}
1159 1160 1161
	r = amdgpu_gart_table_vram_pin(adev);
	if (r)
		return r;
A
Alex Xie 已提交
1162

1163 1164
	switch (adev->asic_type) {
	case CHIP_RAVEN:
1165
		mmhub_v1_0_update_power_gating(adev, true);
1166 1167 1168 1169 1170
		break;
	default:
		break;
	}

A
Alex Xie 已提交
1171 1172 1173 1174 1175 1176 1177 1178
	r = gfxhub_v1_0_gart_enable(adev);
	if (r)
		return r;

	r = mmhub_v1_0_gart_enable(adev);
	if (r)
		return r;

1179
	WREG32_FIELD15(HDP, 0, HDP_MISC_CNTL, FLUSH_INVALIDATE_CACHE, 1);
A
Alex Xie 已提交
1180

1181 1182
	tmp = RREG32_SOC15(HDP, 0, mmHDP_HOST_PATH_CNTL);
	WREG32_SOC15(HDP, 0, mmHDP_HOST_PATH_CNTL, tmp);
A
Alex Xie 已提交
1183

1184
	/* After HDP is initialized, flush HDP.*/
1185
	adev->nbio_funcs->hdp_flush(adev, NULL);
1186

A
Alex Xie 已提交
1187 1188 1189 1190 1191 1192 1193
	if (amdgpu_vm_fault_stop == AMDGPU_VM_FAULT_STOP_ALWAYS)
		value = false;
	else
		value = true;

	gfxhub_v1_0_set_fault_enable_default(adev, value);
	mmhub_v1_0_set_fault_enable_default(adev, value);
1194
	gmc_v9_0_flush_gpu_tlb(adev, 0, 0);
A
Alex Xie 已提交
1195 1196

	DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n",
1197
		 (unsigned)(adev->gmc.gart_size >> 20),
1198
		 (unsigned long long)amdgpu_bo_gpu_offset(adev->gart.bo));
A
Alex Xie 已提交
1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210
	adev->gart.ready = true;
	return 0;
}

static int gmc_v9_0_hw_init(void *handle)
{
	int r;
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	/* The sequence of these two function calls matters.*/
	gmc_v9_0_init_golden_registers(adev);

1211 1212
	if (adev->mode_info.num_crtc) {
		/* Lockout access through VGA aperture*/
1213
		WREG32_FIELD15(DCE, 0, VGA_HDP_CONTROL, VGA_MEMORY_DISABLE, 1);
1214 1215

		/* disable VGA render */
1216
		WREG32_FIELD15(DCE, 0, VGA_RENDER_CONTROL, VGA_VSTATUS_CNTL, 0);
1217 1218
	}

A
Alex Xie 已提交
1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234
	r = gmc_v9_0_gart_enable(adev);

	return r;
}

/**
 * gmc_v9_0_gart_disable - gart disable
 *
 * @adev: amdgpu_device pointer
 *
 * This disables all VM page table.
 */
static void gmc_v9_0_gart_disable(struct amdgpu_device *adev)
{
	gfxhub_v1_0_gart_disable(adev);
	mmhub_v1_0_gart_disable(adev);
1235
	amdgpu_gart_table_vram_unpin(adev);
A
Alex Xie 已提交
1236 1237 1238 1239 1240 1241
}

static int gmc_v9_0_hw_fini(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

1242 1243 1244 1245 1246 1247
	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;
	}

X
xinhui pan 已提交
1248
	amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0);
1249
	amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0);
A
Alex Xie 已提交
1250 1251 1252 1253 1254 1255 1256 1257 1258
	gmc_v9_0_gart_disable(adev);

	return 0;
}

static int gmc_v9_0_suspend(void *handle)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

1259
	return gmc_v9_0_hw_fini(adev);
A
Alex Xie 已提交
1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270
}

static int gmc_v9_0_resume(void *handle)
{
	int r;
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	r = gmc_v9_0_hw_init(adev);
	if (r)
		return r;

1271
	amdgpu_vmid_reset_all(adev);
A
Alex Xie 已提交
1272

1273
	return 0;
A
Alex Xie 已提交
1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296
}

static bool gmc_v9_0_is_idle(void *handle)
{
	/* MC is always ready in GMC v9.*/
	return true;
}

static int gmc_v9_0_wait_for_idle(void *handle)
{
	/* There is no need to wait for MC idle in GMC v9.*/
	return 0;
}

static int gmc_v9_0_soft_reset(void *handle)
{
	/* XXX for emulation.*/
	return 0;
}

static int gmc_v9_0_set_clockgating_state(void *handle,
					enum amd_clockgating_state state)
{
1297 1298 1299
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	return mmhub_v1_0_set_clockgating(adev, state);
A
Alex Xie 已提交
1300 1301
}

1302 1303 1304 1305 1306 1307 1308
static void gmc_v9_0_get_clockgating_state(void *handle, u32 *flags)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	mmhub_v1_0_get_clockgating(adev, flags);
}

A
Alex Xie 已提交
1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329
static int gmc_v9_0_set_powergating_state(void *handle,
					enum amd_powergating_state state)
{
	return 0;
}

const struct amd_ip_funcs gmc_v9_0_ip_funcs = {
	.name = "gmc_v9_0",
	.early_init = gmc_v9_0_early_init,
	.late_init = gmc_v9_0_late_init,
	.sw_init = gmc_v9_0_sw_init,
	.sw_fini = gmc_v9_0_sw_fini,
	.hw_init = gmc_v9_0_hw_init,
	.hw_fini = gmc_v9_0_hw_fini,
	.suspend = gmc_v9_0_suspend,
	.resume = gmc_v9_0_resume,
	.is_idle = gmc_v9_0_is_idle,
	.wait_for_idle = gmc_v9_0_wait_for_idle,
	.soft_reset = gmc_v9_0_soft_reset,
	.set_clockgating_state = gmc_v9_0_set_clockgating_state,
	.set_powergating_state = gmc_v9_0_set_powergating_state,
1330
	.get_clockgating_state = gmc_v9_0_get_clockgating_state,
A
Alex Xie 已提交
1331 1332 1333 1334 1335 1336 1337 1338 1339 1340
};

const struct amdgpu_ip_block_version gmc_v9_0_ip_block =
{
	.type = AMD_IP_BLOCK_TYPE_GMC,
	.major = 9,
	.minor = 0,
	.rev = 0,
	.funcs = &gmc_v9_0_ip_funcs,
};