vce_v3_0.c 17.7 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 26 27 28 29 30 31 32 33 34
/*
 * Copyright 2014 Advanced Micro Devices, Inc.
 * All Rights Reserved.
 *
 * 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, sub license, 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 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 NON-INFRINGEMENT. IN NO EVENT SHALL
 * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS 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.
 *
 * The above copyright notice and this permission notice (including the
 * next paragraph) shall be included in all copies or substantial portions
 * of the Software.
 *
 * Authors: Christian König <christian.koenig@amd.com>
 */

#include <linux/firmware.h>
#include <drm/drmP.h>
#include "amdgpu.h"
#include "amdgpu_vce.h"
#include "vid.h"
#include "vce/vce_3_0_d.h"
#include "vce/vce_3_0_sh_mask.h"
35 36
#include "oss/oss_3_0_d.h"
#include "oss/oss_3_0_sh_mask.h"
37
#include "gca/gfx_8_0_d.h"
38 39
#include "smu/smu_7_1_2_d.h"
#include "smu/smu_7_1_2_sh_mask.h"
40 41 42

#define GRBM_GFX_INDEX__VCE_INSTANCE__SHIFT	0x04
#define GRBM_GFX_INDEX__VCE_INSTANCE_MASK	0x10
43 44 45
#define mmVCE_LMI_VCPU_CACHE_40BIT_BAR0 	0x8616
#define mmVCE_LMI_VCPU_CACHE_40BIT_BAR1 	0x8617
#define mmVCE_LMI_VCPU_CACHE_40BIT_BAR2 	0x8618
46

47 48 49 50
#define VCE_V3_0_FW_SIZE	(384 * 1024)
#define VCE_V3_0_STACK_SIZE	(64 * 1024)
#define VCE_V3_0_DATA_SIZE	((16 * 1024 * AMDGPU_MAX_VCE_HANDLES) + (52 * 1024))

51
static void vce_v3_0_mc_resume(struct amdgpu_device *adev, int idx);
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
static void vce_v3_0_set_ring_funcs(struct amdgpu_device *adev);
static void vce_v3_0_set_irq_funcs(struct amdgpu_device *adev);

/**
 * vce_v3_0_ring_get_rptr - get read pointer
 *
 * @ring: amdgpu_ring pointer
 *
 * Returns the current hardware read pointer
 */
static uint32_t vce_v3_0_ring_get_rptr(struct amdgpu_ring *ring)
{
	struct amdgpu_device *adev = ring->adev;

	if (ring == &adev->vce.ring[0])
		return RREG32(mmVCE_RB_RPTR);
	else
		return RREG32(mmVCE_RB_RPTR2);
}

/**
 * vce_v3_0_ring_get_wptr - get write pointer
 *
 * @ring: amdgpu_ring pointer
 *
 * Returns the current hardware write pointer
 */
static uint32_t vce_v3_0_ring_get_wptr(struct amdgpu_ring *ring)
{
	struct amdgpu_device *adev = ring->adev;

	if (ring == &adev->vce.ring[0])
		return RREG32(mmVCE_RB_WPTR);
	else
		return RREG32(mmVCE_RB_WPTR2);
}

/**
 * vce_v3_0_ring_set_wptr - set write pointer
 *
 * @ring: amdgpu_ring pointer
 *
 * Commits the write pointer to the hardware
 */
static void vce_v3_0_ring_set_wptr(struct amdgpu_ring *ring)
{
	struct amdgpu_device *adev = ring->adev;

	if (ring == &adev->vce.ring[0])
		WREG32(mmVCE_RB_WPTR, ring->wptr);
	else
		WREG32(mmVCE_RB_WPTR2, ring->wptr);
}

/**
 * vce_v3_0_start - start VCE block
 *
 * @adev: amdgpu_device pointer
 *
 * Setup and start the VCE block
 */
static int vce_v3_0_start(struct amdgpu_device *adev)
{
	struct amdgpu_ring *ring;
116 117 118 119
	int idx, i, j, r;

	mutex_lock(&adev->grbm_idx_mutex);
	for (idx = 0; idx < 2; ++idx) {
120 121 122 123

		if (adev->vce.harvest_config & (1 << idx))
			continue;

124 125 126 127 128 129 130 131 132 133 134 135
		if(idx == 0)
			WREG32_P(mmGRBM_GFX_INDEX, 0,
				~GRBM_GFX_INDEX__VCE_INSTANCE_MASK);
		else
			WREG32_P(mmGRBM_GFX_INDEX,
				GRBM_GFX_INDEX__VCE_INSTANCE_MASK,
				~GRBM_GFX_INDEX__VCE_INSTANCE_MASK);

		vce_v3_0_mc_resume(adev, idx);

		/* set BUSY flag */
		WREG32_P(mmVCE_STATUS, 1, ~1);
136 137 138 139 140
		if (adev->asic_type >= CHIP_STONEY)
			WREG32_P(mmVCE_VCPU_CNTL, 1, ~0x200001);
		else
			WREG32_P(mmVCE_VCPU_CNTL, VCE_VCPU_CNTL__CLK_EN_MASK,
				~VCE_VCPU_CNTL__CLK_EN_MASK);
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

		WREG32_P(mmVCE_SOFT_RESET,
			 VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK,
			 ~VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK);

		mdelay(100);

		WREG32_P(mmVCE_SOFT_RESET, 0,
			~VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK);

		for (i = 0; i < 10; ++i) {
			uint32_t status;
			for (j = 0; j < 100; ++j) {
				status = RREG32(mmVCE_STATUS);
				if (status & 2)
					break;
				mdelay(10);
			}
			r = 0;
			if (status & 2)
				break;

			DRM_ERROR("VCE not responding, trying to reset the ECPU!!!\n");
			WREG32_P(mmVCE_SOFT_RESET,
				VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK,
				~VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK);
			mdelay(10);
			WREG32_P(mmVCE_SOFT_RESET, 0,
				~VCE_SOFT_RESET__ECPU_SOFT_RESET_MASK);
			mdelay(10);
			r = -1;
		}

		/* clear BUSY flag */
		WREG32_P(mmVCE_STATUS, 0, ~1);
176

177 178 179 180 181 182
		if (r) {
			DRM_ERROR("VCE not responding, giving up!!!\n");
			mutex_unlock(&adev->grbm_idx_mutex);
			return r;
		}
	}
183

184 185
	WREG32_P(mmGRBM_GFX_INDEX, 0, ~GRBM_GFX_INDEX__VCE_INSTANCE_MASK);
	mutex_unlock(&adev->grbm_idx_mutex);
186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203

	ring = &adev->vce.ring[0];
	WREG32(mmVCE_RB_RPTR, ring->wptr);
	WREG32(mmVCE_RB_WPTR, ring->wptr);
	WREG32(mmVCE_RB_BASE_LO, ring->gpu_addr);
	WREG32(mmVCE_RB_BASE_HI, upper_32_bits(ring->gpu_addr));
	WREG32(mmVCE_RB_SIZE, ring->ring_size / 4);

	ring = &adev->vce.ring[1];
	WREG32(mmVCE_RB_RPTR2, ring->wptr);
	WREG32(mmVCE_RB_WPTR2, ring->wptr);
	WREG32(mmVCE_RB_BASE_LO2, ring->gpu_addr);
	WREG32(mmVCE_RB_BASE_HI2, upper_32_bits(ring->gpu_addr));
	WREG32(mmVCE_RB_SIZE2, ring->ring_size / 4);

	return 0;
}

204 205 206 207 208 209 210 211 212
#define ixVCE_HARVEST_FUSE_MACRO__ADDRESS     0xC0014074
#define VCE_HARVEST_FUSE_MACRO__SHIFT       27
#define VCE_HARVEST_FUSE_MACRO__MASK        0x18000000

static unsigned vce_v3_0_get_harvest_config(struct amdgpu_device *adev)
{
	u32 tmp;
	unsigned ret;

213 214 215
	/* Fiji, Stoney are single pipe */
	if ((adev->asic_type == CHIP_FIJI) ||
	    (adev->asic_type == CHIP_STONEY)){
216 217 218 219 220
		ret = AMDGPU_VCE_HARVEST_VCE1;
		return ret;
	}

	/* Tonga and CZ are dual or single pipe */
221
	if (adev->flags & AMD_IS_APU)
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
		tmp = (RREG32_SMC(ixVCE_HARVEST_FUSE_MACRO__ADDRESS) &
		       VCE_HARVEST_FUSE_MACRO__MASK) >>
			VCE_HARVEST_FUSE_MACRO__SHIFT;
	else
		tmp = (RREG32_SMC(ixCC_HARVEST_FUSES) &
		       CC_HARVEST_FUSES__VCE_DISABLE_MASK) >>
			CC_HARVEST_FUSES__VCE_DISABLE__SHIFT;

	switch (tmp) {
	case 1:
		ret = AMDGPU_VCE_HARVEST_VCE0;
		break;
	case 2:
		ret = AMDGPU_VCE_HARVEST_VCE1;
		break;
	case 3:
		ret = AMDGPU_VCE_HARVEST_VCE0 | AMDGPU_VCE_HARVEST_VCE1;
		break;
	default:
		ret = 0;
	}

	return ret;
}

247
static int vce_v3_0_early_init(void *handle)
248
{
249 250
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

251 252 253 254 255 256 257
	adev->vce.harvest_config = vce_v3_0_get_harvest_config(adev);

	if ((adev->vce.harvest_config &
	     (AMDGPU_VCE_HARVEST_VCE0 | AMDGPU_VCE_HARVEST_VCE1)) ==
	    (AMDGPU_VCE_HARVEST_VCE0 | AMDGPU_VCE_HARVEST_VCE1))
		return -ENOENT;

258 259 260 261 262 263
	vce_v3_0_set_ring_funcs(adev);
	vce_v3_0_set_irq_funcs(adev);

	return 0;
}

264
static int vce_v3_0_sw_init(void *handle)
265
{
266
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
267 268 269 270 271 272 273 274
	struct amdgpu_ring *ring;
	int r;

	/* VCE */
	r = amdgpu_irq_add_id(adev, 167, &adev->vce.irq);
	if (r)
		return r;

275 276
	r = amdgpu_vce_sw_init(adev, VCE_V3_0_FW_SIZE +
		(VCE_V3_0_STACK_SIZE + VCE_V3_0_DATA_SIZE) * 2);
277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300
	if (r)
		return r;

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

	ring = &adev->vce.ring[0];
	sprintf(ring->name, "vce0");
	r = amdgpu_ring_init(adev, ring, 4096, VCE_CMD_NO_OP, 0xf,
			     &adev->vce.irq, 0, AMDGPU_RING_TYPE_VCE);
	if (r)
		return r;

	ring = &adev->vce.ring[1];
	sprintf(ring->name, "vce1");
	r = amdgpu_ring_init(adev, ring, 4096, VCE_CMD_NO_OP, 0xf,
			     &adev->vce.irq, 0, AMDGPU_RING_TYPE_VCE);
	if (r)
		return r;

	return r;
}

301
static int vce_v3_0_sw_fini(void *handle)
302 303
{
	int r;
304
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
305 306 307 308 309 310 311 312 313 314 315 316

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

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

	return r;
}

317
static int vce_v3_0_hw_init(void *handle)
318 319 320
{
	struct amdgpu_ring *ring;
	int r;
321
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
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

	r = vce_v3_0_start(adev);
	if (r)
		return r;

	ring = &adev->vce.ring[0];
	ring->ready = true;
	r = amdgpu_ring_test_ring(ring);
	if (r) {
		ring->ready = false;
		return r;
	}

	ring = &adev->vce.ring[1];
	ring->ready = true;
	r = amdgpu_ring_test_ring(ring);
	if (r) {
		ring->ready = false;
		return r;
	}

	DRM_INFO("VCE initialized successfully.\n");

	return 0;
}

348
static int vce_v3_0_hw_fini(void *handle)
349 350 351 352
{
	return 0;
}

353
static int vce_v3_0_suspend(void *handle)
354 355
{
	int r;
356
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
357 358 359 360 361 362 363 364 365 366 367 368

	r = vce_v3_0_hw_fini(adev);
	if (r)
		return r;

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

	return r;
}

369
static int vce_v3_0_resume(void *handle)
370 371
{
	int r;
372
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
373 374 375 376 377 378 379 380 381 382 383 384

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

	r = vce_v3_0_hw_init(adev);
	if (r)
		return r;

	return r;
}

385
static void vce_v3_0_mc_resume(struct amdgpu_device *adev, int idx)
386 387 388 389 390 391 392 393 394 395 396 397 398
{
	uint32_t offset, size;

	WREG32_P(mmVCE_CLOCK_GATING_A, 0, ~(1 << 16));
	WREG32_P(mmVCE_UENC_CLOCK_GATING, 0x1FF000, ~0xFF9FF000);
	WREG32_P(mmVCE_UENC_REG_CLOCK_GATING, 0x3F, ~0x3F);
	WREG32(mmVCE_CLOCK_GATING_B, 0xf7);

	WREG32(mmVCE_LMI_CTRL, 0x00398000);
	WREG32_P(mmVCE_LMI_CACHE_CTRL, 0x0, ~0x1);
	WREG32(mmVCE_LMI_SWAP_CNTL, 0);
	WREG32(mmVCE_LMI_SWAP_CNTL1, 0);
	WREG32(mmVCE_LMI_VM_CTRL, 0);
399 400 401 402 403 404
	if (adev->asic_type >= CHIP_STONEY) {
		WREG32(mmVCE_LMI_VCPU_CACHE_40BIT_BAR0, (adev->vce.gpu_addr >> 8));
		WREG32(mmVCE_LMI_VCPU_CACHE_40BIT_BAR1, (adev->vce.gpu_addr >> 8));
		WREG32(mmVCE_LMI_VCPU_CACHE_40BIT_BAR2, (adev->vce.gpu_addr >> 8));
	} else
		WREG32(mmVCE_LMI_VCPU_CACHE_40BIT_BAR, (adev->vce.gpu_addr >> 8));
405
	offset = AMDGPU_VCE_FIRMWARE_OFFSET;
406
	size = VCE_V3_0_FW_SIZE;
407 408 409
	WREG32(mmVCE_VCPU_CACHE_OFFSET0, offset & 0x7fffffff);
	WREG32(mmVCE_VCPU_CACHE_SIZE0, size);

410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428
	if (idx == 0) {
		offset += size;
		size = VCE_V3_0_STACK_SIZE;
		WREG32(mmVCE_VCPU_CACHE_OFFSET1, offset & 0x7fffffff);
		WREG32(mmVCE_VCPU_CACHE_SIZE1, size);
		offset += size;
		size = VCE_V3_0_DATA_SIZE;
		WREG32(mmVCE_VCPU_CACHE_OFFSET2, offset & 0x7fffffff);
		WREG32(mmVCE_VCPU_CACHE_SIZE2, size);
	} else {
		offset += size + VCE_V3_0_STACK_SIZE + VCE_V3_0_DATA_SIZE;
		size = VCE_V3_0_STACK_SIZE;
		WREG32(mmVCE_VCPU_CACHE_OFFSET1, offset & 0xfffffff);
		WREG32(mmVCE_VCPU_CACHE_SIZE1, size);
		offset += size;
		size = VCE_V3_0_DATA_SIZE;
		WREG32(mmVCE_VCPU_CACHE_OFFSET2, offset & 0xfffffff);
		WREG32(mmVCE_VCPU_CACHE_SIZE2, size);
	}
429 430 431 432 433 434 435

	WREG32_P(mmVCE_LMI_CTRL2, 0x0, ~0x100);

	WREG32_P(mmVCE_SYS_INT_EN, VCE_SYS_INT_EN__VCE_SYS_INT_TRAP_INTERRUPT_EN_MASK,
		 ~VCE_SYS_INT_EN__VCE_SYS_INT_TRAP_INTERRUPT_EN_MASK);
}

436
static bool vce_v3_0_is_idle(void *handle)
437
{
438
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
439 440
	u32 mask = 0;
	int idx;
441

442 443 444 445 446 447 448 449 450 451 452
	for (idx = 0; idx < 2; ++idx) {
		if (adev->vce.harvest_config & (1 << idx))
			continue;

		if (idx == 0)
			mask |= SRBM_STATUS2__VCE0_BUSY_MASK;
		else
			mask |= SRBM_STATUS2__VCE1_BUSY_MASK;
	}

	return !(RREG32(mmSRBM_STATUS2) & mask);
453 454
}

455
static int vce_v3_0_wait_for_idle(void *handle)
456 457
{
	unsigned i;
458
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
459 460 461 462 463 464 465 466 467 468 469 470
	u32 mask = 0;
	int idx;

	for (idx = 0; idx < 2; ++idx) {
		if (adev->vce.harvest_config & (1 << idx))
			continue;

		if (idx == 0)
			mask |= SRBM_STATUS2__VCE0_BUSY_MASK;
		else
			mask |= SRBM_STATUS2__VCE1_BUSY_MASK;
	}
471 472

	for (i = 0; i < adev->usec_timeout; i++) {
473
		if (!(RREG32(mmSRBM_STATUS2) & mask))
474 475 476 477 478
			return 0;
	}
	return -ETIMEDOUT;
}

479
static int vce_v3_0_soft_reset(void *handle)
480
{
481
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
482 483 484 485 486 487
	u32 mask = 0;
	int idx;

	for (idx = 0; idx < 2; ++idx) {
		if (adev->vce.harvest_config & (1 << idx))
			continue;
488

489 490 491 492 493 494 495 496
		if (idx == 0)
			mask |= SRBM_SOFT_RESET__SOFT_RESET_VCE0_MASK;
		else
			mask |= SRBM_SOFT_RESET__SOFT_RESET_VCE1_MASK;
	}
	WREG32_P(mmSRBM_SOFT_RESET, mask,
		 ~(SRBM_SOFT_RESET__SOFT_RESET_VCE0_MASK |
		   SRBM_SOFT_RESET__SOFT_RESET_VCE1_MASK));
497 498 499 500 501
	mdelay(5);

	return vce_v3_0_start(adev);
}

502
static void vce_v3_0_print_status(void *handle)
503
{
504 505
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

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
	dev_info(adev->dev, "VCE 3.0 registers\n");
	dev_info(adev->dev, "  VCE_STATUS=0x%08X\n",
		 RREG32(mmVCE_STATUS));
	dev_info(adev->dev, "  VCE_VCPU_CNTL=0x%08X\n",
		 RREG32(mmVCE_VCPU_CNTL));
	dev_info(adev->dev, "  VCE_VCPU_CACHE_OFFSET0=0x%08X\n",
		 RREG32(mmVCE_VCPU_CACHE_OFFSET0));
	dev_info(adev->dev, "  VCE_VCPU_CACHE_SIZE0=0x%08X\n",
		 RREG32(mmVCE_VCPU_CACHE_SIZE0));
	dev_info(adev->dev, "  VCE_VCPU_CACHE_OFFSET1=0x%08X\n",
		 RREG32(mmVCE_VCPU_CACHE_OFFSET1));
	dev_info(adev->dev, "  VCE_VCPU_CACHE_SIZE1=0x%08X\n",
		 RREG32(mmVCE_VCPU_CACHE_SIZE1));
	dev_info(adev->dev, "  VCE_VCPU_CACHE_OFFSET2=0x%08X\n",
		 RREG32(mmVCE_VCPU_CACHE_OFFSET2));
	dev_info(adev->dev, "  VCE_VCPU_CACHE_SIZE2=0x%08X\n",
		 RREG32(mmVCE_VCPU_CACHE_SIZE2));
	dev_info(adev->dev, "  VCE_SOFT_RESET=0x%08X\n",
		 RREG32(mmVCE_SOFT_RESET));
	dev_info(adev->dev, "  VCE_RB_BASE_LO2=0x%08X\n",
		 RREG32(mmVCE_RB_BASE_LO2));
	dev_info(adev->dev, "  VCE_RB_BASE_HI2=0x%08X\n",
		 RREG32(mmVCE_RB_BASE_HI2));
	dev_info(adev->dev, "  VCE_RB_SIZE2=0x%08X\n",
		 RREG32(mmVCE_RB_SIZE2));
	dev_info(adev->dev, "  VCE_RB_RPTR2=0x%08X\n",
		 RREG32(mmVCE_RB_RPTR2));
	dev_info(adev->dev, "  VCE_RB_WPTR2=0x%08X\n",
		 RREG32(mmVCE_RB_WPTR2));
	dev_info(adev->dev, "  VCE_RB_BASE_LO=0x%08X\n",
		 RREG32(mmVCE_RB_BASE_LO));
	dev_info(adev->dev, "  VCE_RB_BASE_HI=0x%08X\n",
		 RREG32(mmVCE_RB_BASE_HI));
	dev_info(adev->dev, "  VCE_RB_SIZE=0x%08X\n",
		 RREG32(mmVCE_RB_SIZE));
	dev_info(adev->dev, "  VCE_RB_RPTR=0x%08X\n",
		 RREG32(mmVCE_RB_RPTR));
	dev_info(adev->dev, "  VCE_RB_WPTR=0x%08X\n",
		 RREG32(mmVCE_RB_WPTR));
	dev_info(adev->dev, "  VCE_CLOCK_GATING_A=0x%08X\n",
		 RREG32(mmVCE_CLOCK_GATING_A));
	dev_info(adev->dev, "  VCE_CLOCK_GATING_B=0x%08X\n",
		 RREG32(mmVCE_CLOCK_GATING_B));
	dev_info(adev->dev, "  VCE_UENC_CLOCK_GATING=0x%08X\n",
		 RREG32(mmVCE_UENC_CLOCK_GATING));
	dev_info(adev->dev, "  VCE_UENC_REG_CLOCK_GATING=0x%08X\n",
		 RREG32(mmVCE_UENC_REG_CLOCK_GATING));
	dev_info(adev->dev, "  VCE_SYS_INT_EN=0x%08X\n",
		 RREG32(mmVCE_SYS_INT_EN));
	dev_info(adev->dev, "  VCE_LMI_CTRL2=0x%08X\n",
		 RREG32(mmVCE_LMI_CTRL2));
	dev_info(adev->dev, "  VCE_LMI_CTRL=0x%08X\n",
		 RREG32(mmVCE_LMI_CTRL));
	dev_info(adev->dev, "  VCE_LMI_VM_CTRL=0x%08X\n",
		 RREG32(mmVCE_LMI_VM_CTRL));
	dev_info(adev->dev, "  VCE_LMI_SWAP_CNTL=0x%08X\n",
		 RREG32(mmVCE_LMI_SWAP_CNTL));
	dev_info(adev->dev, "  VCE_LMI_SWAP_CNTL1=0x%08X\n",
		 RREG32(mmVCE_LMI_SWAP_CNTL1));
	dev_info(adev->dev, "  VCE_LMI_CACHE_CTRL=0x%08X\n",
		 RREG32(mmVCE_LMI_CACHE_CTRL));
}

static int vce_v3_0_set_interrupt_state(struct amdgpu_device *adev,
					struct amdgpu_irq_src *source,
					unsigned type,
					enum amdgpu_interrupt_state state)
{
	uint32_t val = 0;

	if (state == AMDGPU_IRQ_STATE_ENABLE)
		val |= VCE_SYS_INT_EN__VCE_SYS_INT_TRAP_INTERRUPT_EN_MASK;

	WREG32_P(mmVCE_SYS_INT_EN, val, ~VCE_SYS_INT_EN__VCE_SYS_INT_TRAP_INTERRUPT_EN_MASK);
	return 0;
}

static int vce_v3_0_process_interrupt(struct amdgpu_device *adev,
				      struct amdgpu_irq_src *source,
				      struct amdgpu_iv_entry *entry)
{
	DRM_DEBUG("IH: VCE\n");
588 589 590 591 592

	WREG32_P(mmVCE_SYS_INT_STATUS,
		VCE_SYS_INT_STATUS__VCE_SYS_INT_TRAP_INTERRUPT_INT_MASK,
		~VCE_SYS_INT_STATUS__VCE_SYS_INT_TRAP_INTERRUPT_INT_MASK);

593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608
	switch (entry->src_data) {
	case 0:
		amdgpu_fence_process(&adev->vce.ring[0]);
		break;
	case 1:
		amdgpu_fence_process(&adev->vce.ring[1]);
		break;
	default:
		DRM_ERROR("Unhandled interrupt: %d %d\n",
			  entry->src_id, entry->src_data);
		break;
	}

	return 0;
}

609 610
static int vce_v3_0_set_clockgating_state(void *handle,
					  enum amd_clockgating_state state)
611 612 613 614
{
	return 0;
}

615 616
static int vce_v3_0_set_powergating_state(void *handle,
					  enum amd_powergating_state state)
617 618 619 620 621 622 623 624
{
	/* This doesn't actually powergate the VCE block.
	 * That's done in the dpm code via the SMC.  This
	 * just re-inits the block as necessary.  The actual
	 * gating still happens in the dpm code.  We should
	 * revisit this when there is a cleaner line between
	 * the smc and the hw blocks
	 */
625 626 627
	struct amdgpu_device *adev = (struct amdgpu_device *)handle;

	if (state == AMD_PG_STATE_GATE)
628 629 630 631 632 633
		/* XXX do we need a vce_v3_0_stop()? */
		return 0;
	else
		return vce_v3_0_start(adev);
}

634
const struct amd_ip_funcs vce_v3_0_ip_funcs = {
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
	.early_init = vce_v3_0_early_init,
	.late_init = NULL,
	.sw_init = vce_v3_0_sw_init,
	.sw_fini = vce_v3_0_sw_fini,
	.hw_init = vce_v3_0_hw_init,
	.hw_fini = vce_v3_0_hw_fini,
	.suspend = vce_v3_0_suspend,
	.resume = vce_v3_0_resume,
	.is_idle = vce_v3_0_is_idle,
	.wait_for_idle = vce_v3_0_wait_for_idle,
	.soft_reset = vce_v3_0_soft_reset,
	.print_status = vce_v3_0_print_status,
	.set_clockgating_state = vce_v3_0_set_clockgating_state,
	.set_powergating_state = vce_v3_0_set_powergating_state,
};

static const struct amdgpu_ring_funcs vce_v3_0_ring_funcs = {
	.get_rptr = vce_v3_0_ring_get_rptr,
	.get_wptr = vce_v3_0_ring_get_wptr,
	.set_wptr = vce_v3_0_ring_set_wptr,
	.parse_cs = amdgpu_vce_ring_parse_cs,
	.emit_ib = amdgpu_vce_ring_emit_ib,
	.emit_fence = amdgpu_vce_ring_emit_fence,
	.emit_semaphore = amdgpu_vce_ring_emit_semaphore,
	.test_ring = amdgpu_vce_ring_test_ring,
	.test_ib = amdgpu_vce_ring_test_ib,
661
	.insert_nop = amdgpu_ring_insert_nop,
662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679
};

static void vce_v3_0_set_ring_funcs(struct amdgpu_device *adev)
{
	adev->vce.ring[0].funcs = &vce_v3_0_ring_funcs;
	adev->vce.ring[1].funcs = &vce_v3_0_ring_funcs;
}

static const struct amdgpu_irq_src_funcs vce_v3_0_irq_funcs = {
	.set = vce_v3_0_set_interrupt_state,
	.process = vce_v3_0_process_interrupt,
};

static void vce_v3_0_set_irq_funcs(struct amdgpu_device *adev)
{
	adev->vce.irq.num_types = 1;
	adev->vce.irq.funcs = &vce_v3_0_irq_funcs;
};