nv.c 27.6 KB
Newer Older
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
/*
 * 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/slab.h>
#include <linux/module.h>
A
Alex Deucher 已提交
26 27
#include <linux/pci.h>

28 29 30 31 32 33 34
#include "amdgpu.h"
#include "amdgpu_atombios.h"
#include "amdgpu_ih.h"
#include "amdgpu_uvd.h"
#include "amdgpu_vce.h"
#include "amdgpu_ucode.h"
#include "amdgpu_psp.h"
35
#include "amdgpu_smu.h"
36 37 38 39 40 41 42
#include "atom.h"
#include "amd_pcie.h"

#include "gc/gc_10_1_0_offset.h"
#include "gc/gc_10_1_0_sh_mask.h"
#include "hdp/hdp_5_0_0_offset.h"
#include "hdp/hdp_5_0_0_sh_mask.h"
43
#include "smuio/smuio_11_0_0_offset.h"
44
#include "mp/mp_11_0_offset.h"
45 46 47 48 49 50

#include "soc15.h"
#include "soc15_common.h"
#include "gmc_v10_0.h"
#include "gfxhub_v2_0.h"
#include "mmhub_v2_0.h"
51
#include "nbio_v2_3.h"
52 53 54 55 56
#include "nv.h"
#include "navi10_ih.h"
#include "gfx_v10_0.h"
#include "sdma_v5_0.h"
#include "vcn_v2_0.h"
57
#include "jpeg_v2_0.h"
58 59
#include "dce_virtual.h"
#include "mes_v10_1.h"
60
#include "mxgpu_nv.h"
61 62 63 64 65 66 67 68 69 70

static const struct amd_ip_funcs nv_common_ip_funcs;

/*
 * Indirect registers accessor
 */
static u32 nv_pcie_rreg(struct amdgpu_device *adev, u32 reg)
{
	unsigned long flags, address, data;
	u32 r;
71 72
	address = adev->nbio.funcs->get_pcie_index_offset(adev);
	data = adev->nbio.funcs->get_pcie_data_offset(adev);
73 74 75 76 77 78 79 80 81 82 83 84 85

	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
	WREG32(address, reg);
	(void)RREG32(address);
	r = RREG32(data);
	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
	return r;
}

static void nv_pcie_wreg(struct amdgpu_device *adev, u32 reg, u32 v)
{
	unsigned long flags, address, data;

86 87
	address = adev->nbio.funcs->get_pcie_index_offset(adev);
	data = adev->nbio.funcs->get_pcie_data_offset(adev);
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

	spin_lock_irqsave(&adev->pcie_idx_lock, flags);
	WREG32(address, reg);
	(void)RREG32(address);
	WREG32(data, v);
	(void)RREG32(data);
	spin_unlock_irqrestore(&adev->pcie_idx_lock, flags);
}

static u32 nv_didt_rreg(struct amdgpu_device *adev, u32 reg)
{
	unsigned long flags, address, data;
	u32 r;

	address = SOC15_REG_OFFSET(GC, 0, mmDIDT_IND_INDEX);
	data = SOC15_REG_OFFSET(GC, 0, mmDIDT_IND_DATA);

	spin_lock_irqsave(&adev->didt_idx_lock, flags);
	WREG32(address, (reg));
	r = RREG32(data);
	spin_unlock_irqrestore(&adev->didt_idx_lock, flags);
	return r;
}

static void nv_didt_wreg(struct amdgpu_device *adev, u32 reg, u32 v)
{
	unsigned long flags, address, data;

	address = SOC15_REG_OFFSET(GC, 0, mmDIDT_IND_INDEX);
	data = SOC15_REG_OFFSET(GC, 0, mmDIDT_IND_DATA);

	spin_lock_irqsave(&adev->didt_idx_lock, flags);
	WREG32(address, (reg));
	WREG32(data, (v));
	spin_unlock_irqrestore(&adev->didt_idx_lock, flags);
}

static u32 nv_get_config_memsize(struct amdgpu_device *adev)
{
127
	return adev->nbio.funcs->get_memsize(adev);
128 129 130 131
}

static u32 nv_get_xclk(struct amdgpu_device *adev)
{
132
	return adev->clock.spll.reference_freq;
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
}


void nv_grbm_select(struct amdgpu_device *adev,
		     u32 me, u32 pipe, u32 queue, u32 vmid)
{
	u32 grbm_gfx_cntl = 0;
	grbm_gfx_cntl = REG_SET_FIELD(grbm_gfx_cntl, GRBM_GFX_CNTL, PIPEID, pipe);
	grbm_gfx_cntl = REG_SET_FIELD(grbm_gfx_cntl, GRBM_GFX_CNTL, MEID, me);
	grbm_gfx_cntl = REG_SET_FIELD(grbm_gfx_cntl, GRBM_GFX_CNTL, VMID, vmid);
	grbm_gfx_cntl = REG_SET_FIELD(grbm_gfx_cntl, GRBM_GFX_CNTL, QUEUEID, queue);

	WREG32(SOC15_REG_OFFSET(GC, 0, mmGRBM_GFX_CNTL), grbm_gfx_cntl);
}

static void nv_vga_set_state(struct amdgpu_device *adev, bool state)
{
	/* todo */
}

static bool nv_read_disabled_bios(struct amdgpu_device *adev)
{
	/* todo */
	return false;
}

static bool nv_read_bios_from_rom(struct amdgpu_device *adev,
				  u8 *bios, u32 length_bytes)
{
162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182
	u32 *dw_ptr;
	u32 i, length_dw;

	if (bios == NULL)
		return false;
	if (length_bytes == 0)
		return false;
	/* APU vbios image is part of sbios image */
	if (adev->flags & AMD_IS_APU)
		return false;

	dw_ptr = (u32 *)bios;
	length_dw = ALIGN(length_bytes, 4) / 4;

	/* set rom index to 0 */
	WREG32(SOC15_REG_OFFSET(SMUIO, 0, mmROM_INDEX), 0);
	/* read out the rom data */
	for (i = 0; i < length_dw; i++)
		dw_ptr[i] = RREG32(SOC15_REG_OFFSET(SMUIO, 0, mmROM_DATA));

	return true;
183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
}

static struct soc15_allowed_register_entry nv_allowed_read_registers[] = {
	{ SOC15_REG_ENTRY(GC, 0, mmGRBM_STATUS)},
	{ SOC15_REG_ENTRY(GC, 0, mmGRBM_STATUS2)},
	{ SOC15_REG_ENTRY(GC, 0, mmGRBM_STATUS_SE0)},
	{ SOC15_REG_ENTRY(GC, 0, mmGRBM_STATUS_SE1)},
	{ SOC15_REG_ENTRY(GC, 0, mmGRBM_STATUS_SE2)},
	{ SOC15_REG_ENTRY(GC, 0, mmGRBM_STATUS_SE3)},
	{ SOC15_REG_ENTRY(SDMA0, 0, mmSDMA0_STATUS_REG)},
	{ SOC15_REG_ENTRY(SDMA1, 0, mmSDMA1_STATUS_REG)},
	{ SOC15_REG_ENTRY(GC, 0, mmCP_STAT)},
	{ SOC15_REG_ENTRY(GC, 0, mmCP_STALLED_STAT1)},
	{ SOC15_REG_ENTRY(GC, 0, mmCP_STALLED_STAT2)},
	{ SOC15_REG_ENTRY(GC, 0, mmCP_STALLED_STAT3)},
	{ SOC15_REG_ENTRY(GC, 0, mmCP_CPF_BUSY_STAT)},
	{ SOC15_REG_ENTRY(GC, 0, mmCP_CPF_STALLED_STAT1)},
	{ SOC15_REG_ENTRY(GC, 0, mmCP_CPF_STATUS)},
201
	{ SOC15_REG_ENTRY(GC, 0, mmCP_CPC_BUSY_STAT)},
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
	{ SOC15_REG_ENTRY(GC, 0, mmCP_CPC_STALLED_STAT1)},
	{ SOC15_REG_ENTRY(GC, 0, mmCP_CPC_STATUS)},
	{ SOC15_REG_ENTRY(GC, 0, mmGB_ADDR_CONFIG)},
};

static uint32_t nv_read_indexed_register(struct amdgpu_device *adev, u32 se_num,
					 u32 sh_num, u32 reg_offset)
{
	uint32_t val;

	mutex_lock(&adev->grbm_idx_mutex);
	if (se_num != 0xffffffff || sh_num != 0xffffffff)
		amdgpu_gfx_select_se_sh(adev, se_num, sh_num, 0xffffffff);

	val = RREG32(reg_offset);

	if (se_num != 0xffffffff || sh_num != 0xffffffff)
		amdgpu_gfx_select_se_sh(adev, 0xffffffff, 0xffffffff, 0xffffffff);
	mutex_unlock(&adev->grbm_idx_mutex);
	return val;
}

static uint32_t nv_get_register_value(struct amdgpu_device *adev,
				      bool indexed, u32 se_num,
				      u32 sh_num, u32 reg_offset)
{
	if (indexed) {
		return nv_read_indexed_register(adev, se_num, sh_num, reg_offset);
	} else {
		if (reg_offset == SOC15_REG_OFFSET(GC, 0, mmGB_ADDR_CONFIG))
			return adev->gfx.config.gb_addr_config;
		return RREG32(reg_offset);
	}
}

static int nv_read_register(struct amdgpu_device *adev, u32 se_num,
			    u32 sh_num, u32 reg_offset, u32 *value)
{
	uint32_t i;
	struct soc15_allowed_register_entry  *en;

	*value = 0;
	for (i = 0; i < ARRAY_SIZE(nv_allowed_read_registers); i++) {
		en = &nv_allowed_read_registers[i];
		if (reg_offset !=
		    (adev->reg_offset[en->hwip][en->inst][en->seg] + en->reg_offset))
			continue;

		*value = nv_get_register_value(adev,
					       nv_allowed_read_registers[i].grbm_indexed,
					       se_num, sh_num, reg_offset);
		return 0;
	}
	return -EINVAL;
}

258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279
static int nv_asic_mode1_reset(struct amdgpu_device *adev)
{
	u32 i;
	int ret = 0;

	amdgpu_atombios_scratch_regs_engine_hung(adev, true);

	dev_info(adev->dev, "GPU mode1 reset\n");

	/* disable BM */
	pci_clear_master(adev->pdev);

	pci_save_state(adev->pdev);

	ret = psp_gpu_reset(adev);
	if (ret)
		dev_err(adev->dev, "GPU mode1 reset failed\n");

	pci_restore_state(adev->pdev);

	/* wait for asic to come out of reset */
	for (i = 0; i < adev->usec_timeout; i++) {
280
		u32 memsize = adev->nbio.funcs->get_memsize(adev);
281 282 283 284 285 286 287 288 289 290

		if (memsize != 0xffffffff)
			break;
		udelay(1);
	}

	amdgpu_atombios_scratch_regs_engine_hung(adev, false);

	return ret;
}
291

292 293 294 295 296 297 298 299 300 301
static bool nv_asic_supports_baco(struct amdgpu_device *adev)
{
	struct smu_context *smu = &adev->smu;

	if (smu_baco_is_support(smu))
		return true;
	else
		return false;
}

302 303 304 305 306
static enum amd_reset_method
nv_asic_reset_method(struct amdgpu_device *adev)
{
	struct smu_context *smu = &adev->smu;

307
	if (!amdgpu_sriov_vf(adev) && smu_baco_is_support(smu))
308 309 310 311 312
		return AMD_RESET_METHOD_BACO;
	else
		return AMD_RESET_METHOD_MODE1;
}

313 314
static int nv_asic_reset(struct amdgpu_device *adev)
{
315 316
	int ret = 0;
	struct smu_context *smu = &adev->smu;
317

318
	if (nv_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) {
319 320 321 322 323 324
		ret = smu_baco_enter(smu);
		if (ret)
			return ret;
		ret = smu_baco_exit(smu);
		if (ret)
			return ret;
325
	} else {
326
		ret = nv_asic_mode1_reset(adev);
327
	}
328 329

	return ret;
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
}

static int nv_set_uvd_clocks(struct amdgpu_device *adev, u32 vclk, u32 dclk)
{
	/* todo */
	return 0;
}

static int nv_set_vce_clocks(struct amdgpu_device *adev, u32 evclk, u32 ecclk)
{
	/* todo */
	return 0;
}

static void nv_pcie_gen3_enable(struct amdgpu_device *adev)
{
	if (pci_is_root_bus(adev->pdev->bus))
		return;

	if (amdgpu_pcie_gen2 == 0)
		return;

	if (!(adev->pm.pcie_gen_mask & (CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2 |
					CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3)))
		return;

	/* todo */
}

static void nv_program_aspm(struct amdgpu_device *adev)
{

	if (amdgpu_aspm == 0)
		return;

	/* todo */
}

static void nv_enable_doorbell_aperture(struct amdgpu_device *adev,
					bool enable)
{
371 372
	adev->nbio.funcs->enable_doorbell_aperture(adev, enable);
	adev->nbio.funcs->enable_doorbell_selfring_aperture(adev, enable);
373 374 375 376 377 378 379 380 381 382 383
}

static const struct amdgpu_ip_block_version nv_common_ip_block =
{
	.type = AMD_IP_BLOCK_TYPE_COMMON,
	.major = 1,
	.minor = 0,
	.rev = 0,
	.funcs = &nv_common_ip_funcs,
};

384
static int nv_reg_base_init(struct amdgpu_device *adev)
385
{
386 387 388 389 390 391 392 393 394 395 396 397 398 399
	int r;

	if (amdgpu_discovery) {
		r = amdgpu_discovery_reg_base_init(adev);
		if (r) {
			DRM_WARN("failed to init reg base from ip discovery table, "
					"fallback to legacy init method\n");
			goto legacy_init;
		}

		return 0;
	}

legacy_init:
400 401 402 403
	switch (adev->asic_type) {
	case CHIP_NAVI10:
		navi10_reg_base_init(adev);
		break;
404 405 406
	case CHIP_NAVI14:
		navi14_reg_base_init(adev);
		break;
407 408 409
	case CHIP_NAVI12:
		navi12_reg_base_init(adev);
		break;
410 411 412
	case CHIP_SIENNA_CICHLID:
		sienna_cichlid_reg_base_init(adev);
		break;
413 414 415 416
	default:
		return -EINVAL;
	}

417 418 419 420 421 422 423
	return 0;
}

int nv_set_ip_blocks(struct amdgpu_device *adev)
{
	int r;

424 425
	adev->nbio.funcs = &nbio_v2_3_funcs;
	adev->nbio.hdp_flush_reg = &nbio_v2_3_hdp_flush_reg;
426

427
	if (amdgpu_sriov_vf(adev)) {
428
		adev->virt.ops = &xgpu_nv_virt_ops;
429 430 431 432 433 434 435 436
		/* try send GPU_INIT_DATA request to host */
		amdgpu_virt_request_init_data(adev);
	}

	/* Set IP register base before any HW register access */
	r = nv_reg_base_init(adev);
	if (r)
		return r;
437

438 439
	switch (adev->asic_type) {
	case CHIP_NAVI10:
440
	case CHIP_NAVI14:
441 442 443 444 445
		amdgpu_device_ip_block_add(adev, &nv_common_ip_block);
		amdgpu_device_ip_block_add(adev, &gmc_v10_0_ip_block);
		amdgpu_device_ip_block_add(adev, &navi10_ih_ip_block);
		amdgpu_device_ip_block_add(adev, &psp_v11_0_ip_block);
		if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP &&
446
		    !amdgpu_sriov_vf(adev))
447 448 449
			amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block);
		if (adev->enable_virtual_display || amdgpu_sriov_vf(adev))
			amdgpu_device_ip_block_add(adev, &dce_virtual_ip_block);
450
#if defined(CONFIG_DRM_AMD_DC)
451 452
		else if (amdgpu_device_has_dc_support(adev))
			amdgpu_device_ip_block_add(adev, &dm_ip_block);
453
#endif
454 455 456
		amdgpu_device_ip_block_add(adev, &gfx_v10_0_ip_block);
		amdgpu_device_ip_block_add(adev, &sdma_v5_0_ip_block);
		if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT &&
457
		    !amdgpu_sriov_vf(adev))
458 459
			amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block);
		amdgpu_device_ip_block_add(adev, &vcn_v2_0_ip_block);
460
		amdgpu_device_ip_block_add(adev, &jpeg_v2_0_ip_block);
461 462 463
		if (adev->enable_mes)
			amdgpu_device_ip_block_add(adev, &mes_v10_1_ip_block);
		break;
464 465 466 467
	case CHIP_NAVI12:
		amdgpu_device_ip_block_add(adev, &nv_common_ip_block);
		amdgpu_device_ip_block_add(adev, &gmc_v10_0_ip_block);
		amdgpu_device_ip_block_add(adev, &navi10_ih_ip_block);
468
		amdgpu_device_ip_block_add(adev, &psp_v11_0_ip_block);
469
		if (adev->firmware.load_type == AMDGPU_FW_LOAD_PSP)
470
			amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block);
471 472
		if (adev->enable_virtual_display || amdgpu_sriov_vf(adev))
			amdgpu_device_ip_block_add(adev, &dce_virtual_ip_block);
473
#if defined(CONFIG_DRM_AMD_DC)
L
Leo Li 已提交
474 475
		else if (amdgpu_device_has_dc_support(adev))
			amdgpu_device_ip_block_add(adev, &dm_ip_block);
476
#endif
477 478
		amdgpu_device_ip_block_add(adev, &gfx_v10_0_ip_block);
		amdgpu_device_ip_block_add(adev, &sdma_v5_0_ip_block);
479
		if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT &&
480
		    !amdgpu_sriov_vf(adev))
481
			amdgpu_device_ip_block_add(adev, &smu_v11_0_ip_block);
482
		amdgpu_device_ip_block_add(adev, &vcn_v2_0_ip_block);
483 484
		if (!amdgpu_sriov_vf(adev))
			amdgpu_device_ip_block_add(adev, &jpeg_v2_0_ip_block);
485
		break;
486 487
	case CHIP_SIENNA_CICHLID:
		amdgpu_device_ip_block_add(adev, &nv_common_ip_block);
488
		amdgpu_device_ip_block_add(adev, &gmc_v10_0_ip_block);
489
		break;
490 491 492 493 494 495 496 497 498
	default:
		return -EINVAL;
	}

	return 0;
}

static uint32_t nv_get_rev_id(struct amdgpu_device *adev)
{
499
	return adev->nbio.funcs->get_rev_id(adev);
500 501 502 503
}

static void nv_flush_hdp(struct amdgpu_device *adev, struct amdgpu_ring *ring)
{
504
	adev->nbio.funcs->hdp_flush(adev, ring);
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
}

static void nv_invalidate_hdp(struct amdgpu_device *adev,
				struct amdgpu_ring *ring)
{
	if (!ring || !ring->funcs->emit_wreg) {
		WREG32_SOC15_NO_KIQ(NBIO, 0, mmHDP_READ_CACHE_INVALIDATE, 1);
	} else {
		amdgpu_ring_emit_wreg(ring, SOC15_REG_OFFSET(
					HDP, 0, mmHDP_READ_CACHE_INVALIDATE), 1);
	}
}

static bool nv_need_full_reset(struct amdgpu_device *adev)
{
	return true;
}

static bool nv_need_reset_on_init(struct amdgpu_device *adev)
{
	u32 sol_reg;

	if (adev->flags & AMD_IS_APU)
		return false;

	/* Check sOS sign of life register to confirm sys driver and sOS
	 * are already been loaded.
	 */
	sol_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81);
	if (sol_reg)
		return true;
536

537 538 539
	return false;
}

540 541 542 543 544 545 546 547 548 549
static uint64_t nv_get_pcie_replay_count(struct amdgpu_device *adev)
{

	/* TODO
	 * dummy implement for pcie_replay_count sysfs interface
	 * */

	return 0;
}

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
static void nv_init_doorbell_index(struct amdgpu_device *adev)
{
	adev->doorbell_index.kiq = AMDGPU_NAVI10_DOORBELL_KIQ;
	adev->doorbell_index.mec_ring0 = AMDGPU_NAVI10_DOORBELL_MEC_RING0;
	adev->doorbell_index.mec_ring1 = AMDGPU_NAVI10_DOORBELL_MEC_RING1;
	adev->doorbell_index.mec_ring2 = AMDGPU_NAVI10_DOORBELL_MEC_RING2;
	adev->doorbell_index.mec_ring3 = AMDGPU_NAVI10_DOORBELL_MEC_RING3;
	adev->doorbell_index.mec_ring4 = AMDGPU_NAVI10_DOORBELL_MEC_RING4;
	adev->doorbell_index.mec_ring5 = AMDGPU_NAVI10_DOORBELL_MEC_RING5;
	adev->doorbell_index.mec_ring6 = AMDGPU_NAVI10_DOORBELL_MEC_RING6;
	adev->doorbell_index.mec_ring7 = AMDGPU_NAVI10_DOORBELL_MEC_RING7;
	adev->doorbell_index.userqueue_start = AMDGPU_NAVI10_DOORBELL_USERQUEUE_START;
	adev->doorbell_index.userqueue_end = AMDGPU_NAVI10_DOORBELL_USERQUEUE_END;
	adev->doorbell_index.gfx_ring0 = AMDGPU_NAVI10_DOORBELL_GFX_RING0;
	adev->doorbell_index.gfx_ring1 = AMDGPU_NAVI10_DOORBELL_GFX_RING1;
	adev->doorbell_index.sdma_engine[0] = AMDGPU_NAVI10_DOORBELL_sDMA_ENGINE0;
	adev->doorbell_index.sdma_engine[1] = AMDGPU_NAVI10_DOORBELL_sDMA_ENGINE1;
	adev->doorbell_index.ih = AMDGPU_NAVI10_DOORBELL_IH;
	adev->doorbell_index.vcn.vcn_ring0_1 = AMDGPU_NAVI10_DOORBELL64_VCN0_1;
	adev->doorbell_index.vcn.vcn_ring2_3 = AMDGPU_NAVI10_DOORBELL64_VCN2_3;
	adev->doorbell_index.vcn.vcn_ring4_5 = AMDGPU_NAVI10_DOORBELL64_VCN4_5;
	adev->doorbell_index.vcn.vcn_ring6_7 = AMDGPU_NAVI10_DOORBELL64_VCN6_7;
	adev->doorbell_index.first_non_cp = AMDGPU_NAVI10_DOORBELL64_FIRST_NON_CP;
	adev->doorbell_index.last_non_cp = AMDGPU_NAVI10_DOORBELL64_LAST_NON_CP;

	adev->doorbell_index.max_assignment = AMDGPU_NAVI10_DOORBELL_MAX_ASSIGNMENT << 1;
	adev->doorbell_index.sdma_doorbell_range = 20;
}

static const struct amdgpu_asic_funcs nv_asic_funcs =
{
	.read_disabled_bios = &nv_read_disabled_bios,
	.read_bios_from_rom = &nv_read_bios_from_rom,
	.read_register = &nv_read_register,
	.reset = &nv_asic_reset,
585
	.reset_method = &nv_asic_reset_method,
586 587 588 589 590 591 592 593 594 595
	.set_vga_state = &nv_vga_set_state,
	.get_xclk = &nv_get_xclk,
	.set_uvd_clocks = &nv_set_uvd_clocks,
	.set_vce_clocks = &nv_set_vce_clocks,
	.get_config_memsize = &nv_get_config_memsize,
	.flush_hdp = &nv_flush_hdp,
	.invalidate_hdp = &nv_invalidate_hdp,
	.init_doorbell_index = &nv_init_doorbell_index,
	.need_full_reset = &nv_need_full_reset,
	.need_reset_on_init = &nv_need_reset_on_init,
596
	.get_pcie_replay_count = &nv_get_pcie_replay_count,
597
	.supports_baco = &nv_asic_supports_baco,
598 599 600 601
};

static int nv_common_early_init(void *handle)
{
602
#define MMIO_REG_HOLE_OFFSET (0x80000 - PAGE_SIZE)
603 604
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

605 606
	adev->rmmio_remap.reg_offset = MMIO_REG_HOLE_OFFSET;
	adev->rmmio_remap.bus_addr = adev->rmmio_base + MMIO_REG_HOLE_OFFSET;
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
	adev->smc_rreg = NULL;
	adev->smc_wreg = NULL;
	adev->pcie_rreg = &nv_pcie_rreg;
	adev->pcie_wreg = &nv_pcie_wreg;

	/* TODO: will add them during VCN v2 implementation */
	adev->uvd_ctx_rreg = NULL;
	adev->uvd_ctx_wreg = NULL;

	adev->didt_rreg = &nv_didt_rreg;
	adev->didt_wreg = &nv_didt_wreg;

	adev->asic_funcs = &nv_asic_funcs;

	adev->rev_id = nv_get_rev_id(adev);
	adev->external_rev_id = 0xff;
	switch (adev->asic_type) {
	case CHIP_NAVI10:
		adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG |
			AMD_CG_SUPPORT_GFX_CGCG |
			AMD_CG_SUPPORT_IH_CG |
			AMD_CG_SUPPORT_HDP_MGCG |
			AMD_CG_SUPPORT_HDP_LS |
			AMD_CG_SUPPORT_SDMA_MGCG |
			AMD_CG_SUPPORT_SDMA_LS |
			AMD_CG_SUPPORT_MC_MGCG |
			AMD_CG_SUPPORT_MC_LS |
			AMD_CG_SUPPORT_ATHUB_MGCG |
			AMD_CG_SUPPORT_ATHUB_LS |
			AMD_CG_SUPPORT_VCN_MGCG |
L
Leo Liu 已提交
637
			AMD_CG_SUPPORT_JPEG_MGCG |
638 639
			AMD_CG_SUPPORT_BIF_MGCG |
			AMD_CG_SUPPORT_BIF_LS;
L
Leo Liu 已提交
640
		adev->pg_flags = AMD_PG_SUPPORT_VCN |
641
			AMD_PG_SUPPORT_VCN_DPG |
L
Leo Liu 已提交
642
			AMD_PG_SUPPORT_JPEG |
643
			AMD_PG_SUPPORT_ATHUB;
644 645
		adev->external_rev_id = adev->rev_id + 0x1;
		break;
646
	case CHIP_NAVI14:
647 648 649 650 651 652 653 654 655 656 657 658
		adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG |
			AMD_CG_SUPPORT_GFX_CGCG |
			AMD_CG_SUPPORT_IH_CG |
			AMD_CG_SUPPORT_HDP_MGCG |
			AMD_CG_SUPPORT_HDP_LS |
			AMD_CG_SUPPORT_SDMA_MGCG |
			AMD_CG_SUPPORT_SDMA_LS |
			AMD_CG_SUPPORT_MC_MGCG |
			AMD_CG_SUPPORT_MC_LS |
			AMD_CG_SUPPORT_ATHUB_MGCG |
			AMD_CG_SUPPORT_ATHUB_LS |
			AMD_CG_SUPPORT_VCN_MGCG |
L
Leo Liu 已提交
659
			AMD_CG_SUPPORT_JPEG_MGCG |
660 661
			AMD_CG_SUPPORT_BIF_MGCG |
			AMD_CG_SUPPORT_BIF_LS;
662
		adev->pg_flags = AMD_PG_SUPPORT_VCN |
L
Leo Liu 已提交
663
			AMD_PG_SUPPORT_JPEG |
664
			AMD_PG_SUPPORT_VCN_DPG;
665
		adev->external_rev_id = adev->rev_id + 20;
666
		break;
667
	case CHIP_NAVI12:
668 669 670 671
		adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG |
			AMD_CG_SUPPORT_GFX_MGLS |
			AMD_CG_SUPPORT_GFX_CGCG |
			AMD_CG_SUPPORT_GFX_CP_LS |
672
			AMD_CG_SUPPORT_GFX_RLC_LS |
673
			AMD_CG_SUPPORT_IH_CG |
674
			AMD_CG_SUPPORT_HDP_MGCG |
675 676
			AMD_CG_SUPPORT_HDP_LS |
			AMD_CG_SUPPORT_SDMA_MGCG |
677 678
			AMD_CG_SUPPORT_SDMA_LS |
			AMD_CG_SUPPORT_MC_MGCG |
679 680
			AMD_CG_SUPPORT_MC_LS |
			AMD_CG_SUPPORT_ATHUB_MGCG |
681
			AMD_CG_SUPPORT_ATHUB_LS |
L
Leo Liu 已提交
682 683
			AMD_CG_SUPPORT_VCN_MGCG |
			AMD_CG_SUPPORT_JPEG_MGCG;
684
		adev->pg_flags = AMD_PG_SUPPORT_VCN |
685
			AMD_PG_SUPPORT_VCN_DPG |
L
Leo Liu 已提交
686
			AMD_PG_SUPPORT_JPEG |
687
			AMD_PG_SUPPORT_ATHUB;
688 689 690 691 692 693
		/* guest vm gets 0xffffffff when reading RCC_DEV0_EPF0_STRAP0,
		 * as a consequence, the rev_id and external_rev_id are wrong.
		 * workaround it by hardcoding rev_id to 0 (default value).
		 */
		if (amdgpu_sriov_vf(adev))
			adev->rev_id = 0;
694 695
		adev->external_rev_id = adev->rev_id + 0xa;
		break;
696 697 698 699 700
	case CHIP_SIENNA_CICHLID:
		adev->cg_flags = 0;
		adev->pg_flags = 0;
		adev->external_rev_id = adev->rev_id + 0x28;
		break;
701 702 703 704 705
	default:
		/* FIXME: not supported yet */
		return -EINVAL;
	}

706 707 708 709 710
	if (amdgpu_sriov_vf(adev)) {
		amdgpu_virt_init_setting(adev);
		xgpu_nv_mailbox_set_irq_funcs(adev);
	}

711 712 713 714 715
	return 0;
}

static int nv_common_late_init(void *handle)
{
716 717 718 719 720
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	if (amdgpu_sriov_vf(adev))
		xgpu_nv_mailbox_get_irq(adev);

721 722 723 724 725
	return 0;
}

static int nv_common_sw_init(void *handle)
{
726 727 728 729 730
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	if (amdgpu_sriov_vf(adev))
		xgpu_nv_mailbox_add_irq_id(adev);

731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747
	return 0;
}

static int nv_common_sw_fini(void *handle)
{
	return 0;
}

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

	/* enable pcie gen2/3 link */
	nv_pcie_gen3_enable(adev);
	/* enable aspm */
	nv_program_aspm(adev);
	/* setup nbio registers */
748
	adev->nbio.funcs->init_registers(adev);
749 750 751 752 753 754
	/* remap HDP registers to a hole in mmio space,
	 * for the purpose of expose those registers
	 * to process space
	 */
	if (adev->nbio.funcs->remap_hdp_registers)
		adev->nbio.funcs->remap_hdp_registers(adev);
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
	/* enable the doorbell aperture */
	nv_enable_doorbell_aperture(adev, true);

	return 0;
}

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

	/* disable the doorbell aperture */
	nv_enable_doorbell_aperture(adev, false);

	return 0;
}

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

	return nv_common_hw_fini(adev);
}

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

	return nv_common_hw_init(adev);
}

static bool nv_common_is_idle(void *handle)
{
	return true;
}

static int nv_common_wait_for_idle(void *handle)
{
	return 0;
}

static int nv_common_soft_reset(void *handle)
{
	return 0;
}

static void nv_update_hdp_mem_power_gating(struct amdgpu_device *adev,
					   bool enable)
{
	uint32_t hdp_clk_cntl, hdp_clk_cntl1;
	uint32_t hdp_mem_pwr_cntl;

	if (!(adev->cg_flags & (AMD_CG_SUPPORT_HDP_LS |
				AMD_CG_SUPPORT_HDP_DS |
				AMD_CG_SUPPORT_HDP_SD)))
		return;

	hdp_clk_cntl = hdp_clk_cntl1 = RREG32_SOC15(HDP, 0, mmHDP_CLK_CNTL);
	hdp_mem_pwr_cntl = RREG32_SOC15(HDP, 0, mmHDP_MEM_POWER_CTRL);

	/* Before doing clock/power mode switch,
	 * forced on IPH & RC clock */
	hdp_clk_cntl = REG_SET_FIELD(hdp_clk_cntl, HDP_CLK_CNTL,
				     IPH_MEM_CLK_SOFT_OVERRIDE, 1);
	hdp_clk_cntl = REG_SET_FIELD(hdp_clk_cntl, HDP_CLK_CNTL,
				     RC_MEM_CLK_SOFT_OVERRIDE, 1);
	WREG32_SOC15(HDP, 0, mmHDP_CLK_CNTL, hdp_clk_cntl);

	/* HDP 5.0 doesn't support dynamic power mode switch,
	 * disable clock and power gating before any changing */
	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
					 IPH_MEM_POWER_CTRL_EN, 0);
	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
					 IPH_MEM_POWER_LS_EN, 0);
	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
					 IPH_MEM_POWER_DS_EN, 0);
	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
					 IPH_MEM_POWER_SD_EN, 0);
	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
					 RC_MEM_POWER_CTRL_EN, 0);
	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
					 RC_MEM_POWER_LS_EN, 0);
	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
					 RC_MEM_POWER_DS_EN, 0);
	hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl, HDP_MEM_POWER_CTRL,
					 RC_MEM_POWER_SD_EN, 0);
	WREG32_SOC15(HDP, 0, mmHDP_MEM_POWER_CTRL, hdp_mem_pwr_cntl);

	/* only one clock gating mode (LS/DS/SD) can be enabled */
	if (adev->cg_flags & AMD_CG_SUPPORT_HDP_LS) {
		hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl,
						 HDP_MEM_POWER_CTRL,
						 IPH_MEM_POWER_LS_EN, enable);
		hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl,
						 HDP_MEM_POWER_CTRL,
						 RC_MEM_POWER_LS_EN, enable);
	} else if (adev->cg_flags & AMD_CG_SUPPORT_HDP_DS) {
		hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl,
						 HDP_MEM_POWER_CTRL,
						 IPH_MEM_POWER_DS_EN, enable);
		hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl,
						 HDP_MEM_POWER_CTRL,
						 RC_MEM_POWER_DS_EN, enable);
	} else if (adev->cg_flags & AMD_CG_SUPPORT_HDP_SD) {
		hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl,
						 HDP_MEM_POWER_CTRL,
						 IPH_MEM_POWER_SD_EN, enable);
		/* RC should not use shut down mode, fallback to ds */
		hdp_mem_pwr_cntl = REG_SET_FIELD(hdp_mem_pwr_cntl,
						 HDP_MEM_POWER_CTRL,
						 RC_MEM_POWER_DS_EN, enable);
	}

	WREG32_SOC15(HDP, 0, mmHDP_MEM_POWER_CTRL, hdp_mem_pwr_cntl);

	/* restore IPH & RC clock override after clock/power mode changing */
	WREG32_SOC15(HDP, 0, mmHDP_CLK_CNTL, hdp_clk_cntl1);
}

static void nv_update_hdp_clock_gating(struct amdgpu_device *adev,
				       bool enable)
{
	uint32_t hdp_clk_cntl;

	if (!(adev->cg_flags & AMD_CG_SUPPORT_HDP_MGCG))
		return;

	hdp_clk_cntl = RREG32_SOC15(HDP, 0, mmHDP_CLK_CNTL);

	if (enable) {
		hdp_clk_cntl &=
			~(uint32_t)
			  (HDP_CLK_CNTL__IPH_MEM_CLK_SOFT_OVERRIDE_MASK |
			   HDP_CLK_CNTL__RC_MEM_CLK_SOFT_OVERRIDE_MASK |
			   HDP_CLK_CNTL__DBUS_CLK_SOFT_OVERRIDE_MASK |
			   HDP_CLK_CNTL__DYN_CLK_SOFT_OVERRIDE_MASK |
			   HDP_CLK_CNTL__XDP_REG_CLK_SOFT_OVERRIDE_MASK |
			   HDP_CLK_CNTL__HDP_REG_CLK_SOFT_OVERRIDE_MASK);
	} else {
		hdp_clk_cntl |= HDP_CLK_CNTL__IPH_MEM_CLK_SOFT_OVERRIDE_MASK |
			HDP_CLK_CNTL__RC_MEM_CLK_SOFT_OVERRIDE_MASK |
			HDP_CLK_CNTL__DBUS_CLK_SOFT_OVERRIDE_MASK |
			HDP_CLK_CNTL__DYN_CLK_SOFT_OVERRIDE_MASK |
			HDP_CLK_CNTL__XDP_REG_CLK_SOFT_OVERRIDE_MASK |
			HDP_CLK_CNTL__HDP_REG_CLK_SOFT_OVERRIDE_MASK;
	}

	WREG32_SOC15(HDP, 0, mmHDP_CLK_CNTL, hdp_clk_cntl);
}

static int nv_common_set_clockgating_state(void *handle,
					   enum amd_clockgating_state state)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	if (amdgpu_sriov_vf(adev))
		return 0;

	switch (adev->asic_type) {
	case CHIP_NAVI10:
914
	case CHIP_NAVI14:
915
	case CHIP_NAVI12:
916
	case CHIP_SIENNA_CICHLID:
917
		adev->nbio.funcs->update_medium_grain_clock_gating(adev,
918
				state == AMD_CG_STATE_GATE);
919
		adev->nbio.funcs->update_medium_grain_light_sleep(adev,
920
				state == AMD_CG_STATE_GATE);
921
		nv_update_hdp_mem_power_gating(adev,
922
				   state == AMD_CG_STATE_GATE);
923
		nv_update_hdp_clock_gating(adev,
924
				state == AMD_CG_STATE_GATE);
925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946
		break;
	default:
		break;
	}
	return 0;
}

static int nv_common_set_powergating_state(void *handle,
					   enum amd_powergating_state state)
{
	/* TODO */
	return 0;
}

static void nv_common_get_clockgating_state(void *handle, u32 *flags)
{
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
	uint32_t tmp;

	if (amdgpu_sriov_vf(adev))
		*flags = 0;

947
	adev->nbio.funcs->get_clockgating_state(adev, flags);
948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987

	/* AMD_CG_SUPPORT_HDP_MGCG */
	tmp = RREG32_SOC15(HDP, 0, mmHDP_CLK_CNTL);
	if (!(tmp & (HDP_CLK_CNTL__IPH_MEM_CLK_SOFT_OVERRIDE_MASK |
		     HDP_CLK_CNTL__RC_MEM_CLK_SOFT_OVERRIDE_MASK |
		     HDP_CLK_CNTL__DBUS_CLK_SOFT_OVERRIDE_MASK |
		     HDP_CLK_CNTL__DYN_CLK_SOFT_OVERRIDE_MASK |
		     HDP_CLK_CNTL__XDP_REG_CLK_SOFT_OVERRIDE_MASK |
		     HDP_CLK_CNTL__HDP_REG_CLK_SOFT_OVERRIDE_MASK)))
		*flags |= AMD_CG_SUPPORT_HDP_MGCG;

	/* AMD_CG_SUPPORT_HDP_LS/DS/SD */
	tmp = RREG32_SOC15(HDP, 0, mmHDP_MEM_POWER_CTRL);
	if (tmp & HDP_MEM_POWER_CTRL__IPH_MEM_POWER_LS_EN_MASK)
		*flags |= AMD_CG_SUPPORT_HDP_LS;
	else if (tmp & HDP_MEM_POWER_CTRL__IPH_MEM_POWER_DS_EN_MASK)
		*flags |= AMD_CG_SUPPORT_HDP_DS;
	else if (tmp & HDP_MEM_POWER_CTRL__IPH_MEM_POWER_SD_EN_MASK)
		*flags |= AMD_CG_SUPPORT_HDP_SD;

	return;
}

static const struct amd_ip_funcs nv_common_ip_funcs = {
	.name = "nv_common",
	.early_init = nv_common_early_init,
	.late_init = nv_common_late_init,
	.sw_init = nv_common_sw_init,
	.sw_fini = nv_common_sw_fini,
	.hw_init = nv_common_hw_init,
	.hw_fini = nv_common_hw_fini,
	.suspend = nv_common_suspend,
	.resume = nv_common_resume,
	.is_idle = nv_common_is_idle,
	.wait_for_idle = nv_common_wait_for_idle,
	.soft_reset = nv_common_soft_reset,
	.set_clockgating_state = nv_common_set_clockgating_state,
	.set_powergating_state = nv_common_set_powergating_state,
	.get_clockgating_state = nv_common_get_clockgating_state,
};