amdgpu.h 73.1 KB
Newer Older
A
Alex Deucher 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 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44
/*
 * Copyright 2008 Advanced Micro Devices, Inc.
 * Copyright 2008 Red Hat Inc.
 * Copyright 2009 Jerome Glisse.
 *
 * 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.
 *
 * Authors: Dave Airlie
 *          Alex Deucher
 *          Jerome Glisse
 */
#ifndef __AMDGPU_H__
#define __AMDGPU_H__

#include <linux/atomic.h>
#include <linux/wait.h>
#include <linux/list.h>
#include <linux/kref.h>
#include <linux/interval_tree.h>
#include <linux/hashtable.h>
#include <linux/fence.h>

#include <ttm/ttm_bo_api.h>
#include <ttm/ttm_bo_driver.h>
#include <ttm/ttm_placement.h>
#include <ttm/ttm_module.h>
#include <ttm/ttm_execbuf_util.h>

C
Chunming Zhou 已提交
45
#include <drm/drmP.h>
A
Alex Deucher 已提交
46
#include <drm/drm_gem.h>
47
#include <drm/amdgpu_drm.h>
A
Alex Deucher 已提交
48

49
#include "amd_shared.h"
A
Alex Deucher 已提交
50 51 52 53 54
#include "amdgpu_mode.h"
#include "amdgpu_ih.h"
#include "amdgpu_irq.h"
#include "amdgpu_ucode.h"
#include "amdgpu_gds.h"
55
#include "amd_powerplay.h"
56
#include "amdgpu_acp.h"
A
Alex Deucher 已提交
57

58 59
#include "gpu_scheduler.h"

A
Alex Deucher 已提交
60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82
/*
 * Modules parameters.
 */
extern int amdgpu_modeset;
extern int amdgpu_vram_limit;
extern int amdgpu_gart_size;
extern int amdgpu_benchmarking;
extern int amdgpu_testing;
extern int amdgpu_audio;
extern int amdgpu_disp_priority;
extern int amdgpu_hw_i2c;
extern int amdgpu_pcie_gen2;
extern int amdgpu_msi;
extern int amdgpu_lockup_timeout;
extern int amdgpu_dpm;
extern int amdgpu_smc_load_fw;
extern int amdgpu_aspm;
extern int amdgpu_runtime_pm;
extern unsigned amdgpu_ip_block_mask;
extern int amdgpu_bapm;
extern int amdgpu_deep_color;
extern int amdgpu_vm_size;
extern int amdgpu_vm_block_size;
83
extern int amdgpu_vm_fault_stop;
84
extern int amdgpu_vm_debug;
85
extern int amdgpu_sched_jobs;
86
extern int amdgpu_sched_hw_submission;
87
extern int amdgpu_powerplay;
88 89
extern unsigned amdgpu_pcie_gen_cap;
extern unsigned amdgpu_pcie_lane_cap;
A
Alex Deucher 已提交
90

91
#define AMDGPU_WAIT_IDLE_TIMEOUT_IN_MS	        3000
A
Alex Deucher 已提交
92 93 94 95 96 97 98 99 100 101 102 103 104 105
#define AMDGPU_MAX_USEC_TIMEOUT			100000	/* 100 ms */
#define AMDGPU_FENCE_JIFFIES_TIMEOUT		(HZ / 2)
/* AMDGPU_IB_POOL_SIZE must be a power of 2 */
#define AMDGPU_IB_POOL_SIZE			16
#define AMDGPU_DEBUGFS_MAX_COMPONENTS		32
#define AMDGPUFB_CONN_LIMIT			4
#define AMDGPU_BIOS_NUM_SCRATCH			8

/* max number of rings */
#define AMDGPU_MAX_RINGS			16
#define AMDGPU_MAX_GFX_RINGS			1
#define AMDGPU_MAX_COMPUTE_RINGS		8
#define AMDGPU_MAX_VCE_RINGS			2

106 107 108
/* max number of IP instances */
#define AMDGPU_MAX_SDMA_INSTANCES		2

A
Alex Deucher 已提交
109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147
/* hardcode that limit for now */
#define AMDGPU_VA_RESERVED_SIZE			(8 << 20)

/* hard reset data */
#define AMDGPU_ASIC_RESET_DATA                  0x39d5e86b

/* reset flags */
#define AMDGPU_RESET_GFX			(1 << 0)
#define AMDGPU_RESET_COMPUTE			(1 << 1)
#define AMDGPU_RESET_DMA			(1 << 2)
#define AMDGPU_RESET_CP				(1 << 3)
#define AMDGPU_RESET_GRBM			(1 << 4)
#define AMDGPU_RESET_DMA1			(1 << 5)
#define AMDGPU_RESET_RLC			(1 << 6)
#define AMDGPU_RESET_SEM			(1 << 7)
#define AMDGPU_RESET_IH				(1 << 8)
#define AMDGPU_RESET_VMC			(1 << 9)
#define AMDGPU_RESET_MC				(1 << 10)
#define AMDGPU_RESET_DISPLAY			(1 << 11)
#define AMDGPU_RESET_UVD			(1 << 12)
#define AMDGPU_RESET_VCE			(1 << 13)
#define AMDGPU_RESET_VCE1			(1 << 14)

/* GFX current status */
#define AMDGPU_GFX_NORMAL_MODE			0x00000000L
#define AMDGPU_GFX_SAFE_MODE			0x00000001L
#define AMDGPU_GFX_PG_DISABLED_MODE		0x00000002L
#define AMDGPU_GFX_CG_DISABLED_MODE		0x00000004L
#define AMDGPU_GFX_LBPW_DISABLED_MODE		0x00000008L

/* max cursor sizes (in pixels) */
#define CIK_CURSOR_WIDTH 128
#define CIK_CURSOR_HEIGHT 128

struct amdgpu_device;
struct amdgpu_ib;
struct amdgpu_vm;
struct amdgpu_ring;
struct amdgpu_cs_parser;
148
struct amdgpu_job;
A
Alex Deucher 已提交
149
struct amdgpu_irq_src;
150
struct amdgpu_fpriv;
A
Alex Deucher 已提交
151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180

enum amdgpu_cp_irq {
	AMDGPU_CP_IRQ_GFX_EOP = 0,
	AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP,
	AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE1_EOP,
	AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE2_EOP,
	AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE3_EOP,
	AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE0_EOP,
	AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE1_EOP,
	AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE2_EOP,
	AMDGPU_CP_IRQ_COMPUTE_MEC2_PIPE3_EOP,

	AMDGPU_CP_IRQ_LAST
};

enum amdgpu_sdma_irq {
	AMDGPU_SDMA_IRQ_TRAP0 = 0,
	AMDGPU_SDMA_IRQ_TRAP1,

	AMDGPU_SDMA_IRQ_LAST
};

enum amdgpu_thermal_irq {
	AMDGPU_THERMAL_IRQ_LOW_TO_HIGH = 0,
	AMDGPU_THERMAL_IRQ_HIGH_TO_LOW,

	AMDGPU_THERMAL_IRQ_LAST
};

int amdgpu_set_clockgating_state(struct amdgpu_device *adev,
181 182
				  enum amd_ip_block_type block_type,
				  enum amd_clockgating_state state);
A
Alex Deucher 已提交
183
int amdgpu_set_powergating_state(struct amdgpu_device *adev,
184 185
				  enum amd_ip_block_type block_type,
				  enum amd_powergating_state state);
A
Alex Deucher 已提交
186 187

struct amdgpu_ip_block_version {
188
	enum amd_ip_block_type type;
A
Alex Deucher 已提交
189 190 191
	u32 major;
	u32 minor;
	u32 rev;
192
	const struct amd_ip_funcs *funcs;
A
Alex Deucher 已提交
193 194 195
};

int amdgpu_ip_block_version_cmp(struct amdgpu_device *adev,
196
				enum amd_ip_block_type type,
A
Alex Deucher 已提交
197 198 199 200
				u32 major, u32 minor);

const struct amdgpu_ip_block_version * amdgpu_get_ip_block(
					struct amdgpu_device *adev,
201
					enum amd_ip_block_type type);
A
Alex Deucher 已提交
202 203 204 205 206 207 208 209 210 211

/* provided by hw blocks that can move/clear data.  e.g., gfx or sdma */
struct amdgpu_buffer_funcs {
	/* maximum bytes in a single operation */
	uint32_t	copy_max_bytes;

	/* number of dw to reserve per operation */
	unsigned	copy_num_dw;

	/* used for buffer migration */
212
	void (*emit_copy_buffer)(struct amdgpu_ib *ib,
A
Alex Deucher 已提交
213 214 215 216 217 218 219 220 221 222 223 224 225 226
				 /* src addr in bytes */
				 uint64_t src_offset,
				 /* dst addr in bytes */
				 uint64_t dst_offset,
				 /* number of byte to transfer */
				 uint32_t byte_count);

	/* maximum bytes in a single operation */
	uint32_t	fill_max_bytes;

	/* number of dw to reserve per operation */
	unsigned	fill_num_dw;

	/* used for buffer clearing */
227
	void (*emit_fill_buffer)(struct amdgpu_ib *ib,
A
Alex Deucher 已提交
228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243
				 /* value to write to memory */
				 uint32_t src_data,
				 /* dst addr in bytes */
				 uint64_t dst_offset,
				 /* number of byte to fill */
				 uint32_t byte_count);
};

/* provided by hw blocks that can write ptes, e.g., sdma */
struct amdgpu_vm_pte_funcs {
	/* copy pte entries from GART */
	void (*copy_pte)(struct amdgpu_ib *ib,
			 uint64_t pe, uint64_t src,
			 unsigned count);
	/* write pte one entry at a time with addr mapping */
	void (*write_pte)(struct amdgpu_ib *ib,
244
			  const dma_addr_t *pages_addr, uint64_t pe,
A
Alex Deucher 已提交
245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287
			  uint64_t addr, unsigned count,
			  uint32_t incr, uint32_t flags);
	/* for linear pte/pde updates without addr mapping */
	void (*set_pte_pde)(struct amdgpu_ib *ib,
			    uint64_t pe,
			    uint64_t addr, unsigned count,
			    uint32_t incr, uint32_t flags);
};

/* provided by the gmc block */
struct amdgpu_gart_funcs {
	/* flush the vm tlb via mmio */
	void (*flush_gpu_tlb)(struct amdgpu_device *adev,
			      uint32_t vmid);
	/* write pte/pde updates using the cpu */
	int (*set_pte_pde)(struct amdgpu_device *adev,
			   void *cpu_pt_addr, /* cpu addr of page table */
			   uint32_t gpu_page_idx, /* pte/pde to update */
			   uint64_t addr, /* addr to write into pte/pde */
			   uint32_t flags); /* access flags */
};

/* provided by the ih block */
struct amdgpu_ih_funcs {
	/* ring read/write ptr handling, called from interrupt context */
	u32 (*get_wptr)(struct amdgpu_device *adev);
	void (*decode_iv)(struct amdgpu_device *adev,
			  struct amdgpu_iv_entry *entry);
	void (*set_rptr)(struct amdgpu_device *adev);
};

/* provided by hw blocks that expose a ring buffer for commands */
struct amdgpu_ring_funcs {
	/* ring read/write ptr handling */
	u32 (*get_rptr)(struct amdgpu_ring *ring);
	u32 (*get_wptr)(struct amdgpu_ring *ring);
	void (*set_wptr)(struct amdgpu_ring *ring);
	/* validating and patching of IBs */
	int (*parse_cs)(struct amdgpu_cs_parser *p, uint32_t ib_idx);
	/* command emit functions */
	void (*emit_ib)(struct amdgpu_ring *ring,
			struct amdgpu_ib *ib);
	void (*emit_fence)(struct amdgpu_ring *ring, uint64_t addr,
288
			   uint64_t seq, unsigned flags);
289
	void (*emit_pipeline_sync)(struct amdgpu_ring *ring);
A
Alex Deucher 已提交
290 291
	void (*emit_vm_flush)(struct amdgpu_ring *ring, unsigned vm_id,
			      uint64_t pd_addr);
292
	void (*emit_hdp_flush)(struct amdgpu_ring *ring);
293
	void (*emit_hdp_invalidate)(struct amdgpu_ring *ring);
A
Alex Deucher 已提交
294 295 296 297 298 299 300
	void (*emit_gds_switch)(struct amdgpu_ring *ring, uint32_t vmid,
				uint32_t gds_base, uint32_t gds_size,
				uint32_t gws_base, uint32_t gws_size,
				uint32_t oa_base, uint32_t oa_size);
	/* testing functions */
	int (*test_ring)(struct amdgpu_ring *ring);
	int (*test_ib)(struct amdgpu_ring *ring);
301 302
	/* insert NOP packets */
	void (*insert_nop)(struct amdgpu_ring *ring, uint32_t count);
303 304
	/* pad the indirect buffer to the necessary number of dw */
	void (*pad_ib)(struct amdgpu_ring *ring, struct amdgpu_ib *ib);
M
Monk Liu 已提交
305 306
	unsigned (*init_cond_exec)(struct amdgpu_ring *ring);
	void (*patch_cond_exec)(struct amdgpu_ring *ring, unsigned offset);
A
Alex Deucher 已提交
307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351
};

/*
 * BIOS.
 */
bool amdgpu_get_bios(struct amdgpu_device *adev);
bool amdgpu_read_bios(struct amdgpu_device *adev);

/*
 * Dummy page
 */
struct amdgpu_dummy_page {
	struct page	*page;
	dma_addr_t	addr;
};
int amdgpu_dummy_page_init(struct amdgpu_device *adev);
void amdgpu_dummy_page_fini(struct amdgpu_device *adev);


/*
 * Clocks
 */

#define AMDGPU_MAX_PPLL 3

struct amdgpu_clock {
	struct amdgpu_pll ppll[AMDGPU_MAX_PPLL];
	struct amdgpu_pll spll;
	struct amdgpu_pll mpll;
	/* 10 Khz units */
	uint32_t default_mclk;
	uint32_t default_sclk;
	uint32_t default_dispclk;
	uint32_t current_dispclk;
	uint32_t dp_extclk;
	uint32_t max_pixel_clock;
};

/*
 * Fences.
 */
struct amdgpu_fence_driver {
	uint64_t			gpu_addr;
	volatile uint32_t		*cpu_addr;
	/* sync_seq is protected by ring emission lock */
352 353
	uint32_t			sync_seq;
	atomic_t			last_seq;
A
Alex Deucher 已提交
354 355 356
	bool				initialized;
	struct amdgpu_irq_src		*irq_src;
	unsigned			irq_type;
357
	struct timer_list		fallback_timer;
358
	unsigned			num_fences_mask;
359
	spinlock_t			lock;
360
	struct fence			**fences;
A
Alex Deucher 已提交
361 362 363 364 365 366
};

/* some special values for the owner field */
#define AMDGPU_FENCE_OWNER_UNDEFINED	((void*)0ul)
#define AMDGPU_FENCE_OWNER_VM		((void*)1ul)

367 368 369
#define AMDGPU_FENCE_FLAG_64BIT         (1 << 0)
#define AMDGPU_FENCE_FLAG_INT           (1 << 1)

A
Alex Deucher 已提交
370 371 372 373 374 375 376 377 378 379 380
struct amdgpu_user_fence {
	/* write-back bo */
	struct amdgpu_bo 	*bo;
	/* write-back address offset to bo start */
	uint32_t                offset;
};

int amdgpu_fence_driver_init(struct amdgpu_device *adev);
void amdgpu_fence_driver_fini(struct amdgpu_device *adev);
void amdgpu_fence_driver_force_completion(struct amdgpu_device *adev);

381 382
int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring,
				  unsigned num_hw_submission);
A
Alex Deucher 已提交
383 384 385
int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring,
				   struct amdgpu_irq_src *irq_src,
				   unsigned irq_type);
386 387
void amdgpu_fence_driver_suspend(struct amdgpu_device *adev);
void amdgpu_fence_driver_resume(struct amdgpu_device *adev);
388
int amdgpu_fence_emit(struct amdgpu_ring *ring, struct fence **fence);
A
Alex Deucher 已提交
389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410
void amdgpu_fence_process(struct amdgpu_ring *ring);
int amdgpu_fence_wait_empty(struct amdgpu_ring *ring);
unsigned amdgpu_fence_count_emitted(struct amdgpu_ring *ring);

/*
 * TTM.
 */
struct amdgpu_mman {
	struct ttm_bo_global_ref        bo_global_ref;
	struct drm_global_reference	mem_global_ref;
	struct ttm_bo_device		bdev;
	bool				mem_global_referenced;
	bool				initialized;

#if defined(CONFIG_DEBUG_FS)
	struct dentry			*vram;
	struct dentry			*gtt;
#endif

	/* buffer handling */
	const struct amdgpu_buffer_funcs	*buffer_funcs;
	struct amdgpu_ring			*buffer_funcs_ring;
411 412
	/* Scheduler entity for buffer moves */
	struct amd_sched_entity			entity;
A
Alex Deucher 已提交
413 414 415 416 417 418 419
};

int amdgpu_copy_buffer(struct amdgpu_ring *ring,
		       uint64_t src_offset,
		       uint64_t dst_offset,
		       uint32_t byte_count,
		       struct reservation_object *resv,
420
		       struct fence **fence);
A
Alex Deucher 已提交
421 422 423 424 425 426 427
int amdgpu_mmap(struct file *filp, struct vm_area_struct *vma);

struct amdgpu_bo_list_entry {
	struct amdgpu_bo		*robj;
	struct ttm_validate_buffer	tv;
	struct amdgpu_bo_va		*bo_va;
	uint32_t			priority;
428 429
	struct page			**user_pages;
	int				user_invalidated;
A
Alex Deucher 已提交
430 431 432 433 434 435 436 437 438 439 440 441 442
};

struct amdgpu_bo_va_mapping {
	struct list_head		list;
	struct interval_tree_node	it;
	uint64_t			offset;
	uint32_t			flags;
};

/* bo virtual addresses in a specific vm */
struct amdgpu_bo_va {
	/* protected by bo being reserved */
	struct list_head		bo_list;
443
	struct fence		        *last_pt_update;
A
Alex Deucher 已提交
444 445
	unsigned			ref_count;

446
	/* protected by vm mutex and spinlock */
A
Alex Deucher 已提交
447 448
	struct list_head		vm_status;

449 450 451 452
	/* mappings for this bo_va */
	struct list_head		invalids;
	struct list_head		valids;

A
Alex Deucher 已提交
453 454 455 456 457
	/* constant after initialization */
	struct amdgpu_vm		*vm;
	struct amdgpu_bo		*bo;
};

458 459
#define AMDGPU_GEM_DOMAIN_MAX		0x3

A
Alex Deucher 已提交
460 461 462 463
struct amdgpu_bo {
	/* Protected by gem.mutex */
	struct list_head		list;
	/* Protected by tbo.reserved */
464 465
	u32				prefered_domains;
	u32				allowed_domains;
466
	struct ttm_place		placements[AMDGPU_GEM_DOMAIN_MAX + 1];
A
Alex Deucher 已提交
467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483
	struct ttm_placement		placement;
	struct ttm_buffer_object	tbo;
	struct ttm_bo_kmap_obj		kmap;
	u64				flags;
	unsigned			pin_count;
	void				*kptr;
	u64				tiling_flags;
	u64				metadata_flags;
	void				*metadata;
	u32				metadata_size;
	/* list of all virtual address to which this bo
	 * is associated to
	 */
	struct list_head		va;
	/* Constant after initialization */
	struct amdgpu_device		*adev;
	struct drm_gem_object		gem_base;
484
	struct amdgpu_bo		*parent;
A
Alex Deucher 已提交
485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534

	struct ttm_bo_kmap_obj		dma_buf_vmap;
	struct amdgpu_mn		*mn;
	struct list_head		mn_list;
};
#define gem_to_amdgpu_bo(gobj) container_of((gobj), struct amdgpu_bo, gem_base)

void amdgpu_gem_object_free(struct drm_gem_object *obj);
int amdgpu_gem_object_open(struct drm_gem_object *obj,
				struct drm_file *file_priv);
void amdgpu_gem_object_close(struct drm_gem_object *obj,
				struct drm_file *file_priv);
unsigned long amdgpu_gem_timeout(uint64_t timeout_ns);
struct sg_table *amdgpu_gem_prime_get_sg_table(struct drm_gem_object *obj);
struct drm_gem_object *amdgpu_gem_prime_import_sg_table(struct drm_device *dev,
							struct dma_buf_attachment *attach,
							struct sg_table *sg);
struct dma_buf *amdgpu_gem_prime_export(struct drm_device *dev,
					struct drm_gem_object *gobj,
					int flags);
int amdgpu_gem_prime_pin(struct drm_gem_object *obj);
void amdgpu_gem_prime_unpin(struct drm_gem_object *obj);
struct reservation_object *amdgpu_gem_prime_res_obj(struct drm_gem_object *);
void *amdgpu_gem_prime_vmap(struct drm_gem_object *obj);
void amdgpu_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr);
int amdgpu_gem_debugfs_init(struct amdgpu_device *adev);

/* sub-allocation manager, it has to be protected by another lock.
 * By conception this is an helper for other part of the driver
 * like the indirect buffer or semaphore, which both have their
 * locking.
 *
 * Principe is simple, we keep a list of sub allocation in offset
 * order (first entry has offset == 0, last entry has the highest
 * offset).
 *
 * When allocating new object we first check if there is room at
 * the end total_size - (last_object_offset + last_object_size) >=
 * alloc_size. If so we allocate new object there.
 *
 * When there is not enough room at the end, we start waiting for
 * each sub object until we reach object_offset+object_size >=
 * alloc_size, this object then become the sub object we return.
 *
 * Alignment can't be bigger than page size.
 *
 * Hole are not considered for allocation to keep things simple.
 * Assumption is that there won't be hole (all object on same
 * alignment).
 */
535 536 537

#define AMDGPU_SA_NUM_FENCE_LISTS	32

A
Alex Deucher 已提交
538 539 540 541
struct amdgpu_sa_manager {
	wait_queue_head_t	wq;
	struct amdgpu_bo	*bo;
	struct list_head	*hole;
542
	struct list_head	flist[AMDGPU_SA_NUM_FENCE_LISTS];
A
Alex Deucher 已提交
543 544 545 546 547 548 549 550 551 552 553 554 555 556 557
	struct list_head	olist;
	unsigned		size;
	uint64_t		gpu_addr;
	void			*cpu_ptr;
	uint32_t		domain;
	uint32_t		align;
};

/* sub-allocation buffer */
struct amdgpu_sa_bo {
	struct list_head		olist;
	struct list_head		flist;
	struct amdgpu_sa_manager	*manager;
	unsigned			soffset;
	unsigned			eoffset;
558
	struct fence		        *fence;
A
Alex Deucher 已提交
559 560 561 562 563
};

/*
 * GEM objects.
 */
564
void amdgpu_gem_force_release(struct amdgpu_device *adev);
A
Alex Deucher 已提交
565 566 567 568 569 570 571 572 573 574 575 576 577 578 579
int amdgpu_gem_object_create(struct amdgpu_device *adev, unsigned long size,
				int alignment, u32 initial_domain,
				u64 flags, bool kernel,
				struct drm_gem_object **obj);

int amdgpu_mode_dumb_create(struct drm_file *file_priv,
			    struct drm_device *dev,
			    struct drm_mode_create_dumb *args);
int amdgpu_mode_dumb_mmap(struct drm_file *filp,
			  struct drm_device *dev,
			  uint32_t handle, uint64_t *offset_p);
/*
 * Synchronization
 */
struct amdgpu_sync {
580
	DECLARE_HASHTABLE(fences, 4);
581
	struct fence	        *last_vm_update;
A
Alex Deucher 已提交
582 583 584
};

void amdgpu_sync_create(struct amdgpu_sync *sync);
585 586
int amdgpu_sync_fence(struct amdgpu_device *adev, struct amdgpu_sync *sync,
		      struct fence *f);
A
Alex Deucher 已提交
587 588 589 590
int amdgpu_sync_resv(struct amdgpu_device *adev,
		     struct amdgpu_sync *sync,
		     struct reservation_object *resv,
		     void *owner);
591 592 593
bool amdgpu_sync_is_idle(struct amdgpu_sync *sync);
int amdgpu_sync_cycle_fences(struct amdgpu_sync *dst, struct amdgpu_sync *src,
			     struct fence *fence);
594
struct fence *amdgpu_sync_get_fence(struct amdgpu_sync *sync);
595
int amdgpu_sync_wait(struct amdgpu_sync *sync);
596
void amdgpu_sync_free(struct amdgpu_sync *sync);
597 598
int amdgpu_sync_init(void);
void amdgpu_sync_fini(void);
A
Alex Deucher 已提交
599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616

/*
 * GART structures, functions & helpers
 */
struct amdgpu_mc;

#define AMDGPU_GPU_PAGE_SIZE 4096
#define AMDGPU_GPU_PAGE_MASK (AMDGPU_GPU_PAGE_SIZE - 1)
#define AMDGPU_GPU_PAGE_SHIFT 12
#define AMDGPU_GPU_PAGE_ALIGN(a) (((a) + AMDGPU_GPU_PAGE_MASK) & ~AMDGPU_GPU_PAGE_MASK)

struct amdgpu_gart {
	dma_addr_t			table_addr;
	struct amdgpu_bo		*robj;
	void				*ptr;
	unsigned			num_gpu_pages;
	unsigned			num_cpu_pages;
	unsigned			table_size;
617
#ifdef CONFIG_DRM_AMDGPU_GART_DEBUGFS
A
Alex Deucher 已提交
618
	struct page			**pages;
619
#endif
A
Alex Deucher 已提交
620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661
	bool				ready;
	const struct amdgpu_gart_funcs *gart_funcs;
};

int amdgpu_gart_table_ram_alloc(struct amdgpu_device *adev);
void amdgpu_gart_table_ram_free(struct amdgpu_device *adev);
int amdgpu_gart_table_vram_alloc(struct amdgpu_device *adev);
void amdgpu_gart_table_vram_free(struct amdgpu_device *adev);
int amdgpu_gart_table_vram_pin(struct amdgpu_device *adev);
void amdgpu_gart_table_vram_unpin(struct amdgpu_device *adev);
int amdgpu_gart_init(struct amdgpu_device *adev);
void amdgpu_gart_fini(struct amdgpu_device *adev);
void amdgpu_gart_unbind(struct amdgpu_device *adev, unsigned offset,
			int pages);
int amdgpu_gart_bind(struct amdgpu_device *adev, unsigned offset,
		     int pages, struct page **pagelist,
		     dma_addr_t *dma_addr, uint32_t flags);

/*
 * GPU MC structures, functions & helpers
 */
struct amdgpu_mc {
	resource_size_t		aper_size;
	resource_size_t		aper_base;
	resource_size_t		agp_base;
	/* for some chips with <= 32MB we need to lie
	 * about vram size near mc fb location */
	u64			mc_vram_size;
	u64			visible_vram_size;
	u64			gtt_size;
	u64			gtt_start;
	u64			gtt_end;
	u64			vram_start;
	u64			vram_end;
	unsigned		vram_width;
	u64			real_vram_size;
	int			vram_mtrr;
	u64                     gtt_base_align;
	u64                     mc_mask;
	const struct firmware   *fw;	/* MC firmware */
	uint32_t                fw_version;
	struct amdgpu_irq_src	vm_fault;
662
	uint32_t		vram_type;
A
Alex Deucher 已提交
663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713
};

/*
 * GPU doorbell structures, functions & helpers
 */
typedef enum _AMDGPU_DOORBELL_ASSIGNMENT
{
	AMDGPU_DOORBELL_KIQ                     = 0x000,
	AMDGPU_DOORBELL_HIQ                     = 0x001,
	AMDGPU_DOORBELL_DIQ                     = 0x002,
	AMDGPU_DOORBELL_MEC_RING0               = 0x010,
	AMDGPU_DOORBELL_MEC_RING1               = 0x011,
	AMDGPU_DOORBELL_MEC_RING2               = 0x012,
	AMDGPU_DOORBELL_MEC_RING3               = 0x013,
	AMDGPU_DOORBELL_MEC_RING4               = 0x014,
	AMDGPU_DOORBELL_MEC_RING5               = 0x015,
	AMDGPU_DOORBELL_MEC_RING6               = 0x016,
	AMDGPU_DOORBELL_MEC_RING7               = 0x017,
	AMDGPU_DOORBELL_GFX_RING0               = 0x020,
	AMDGPU_DOORBELL_sDMA_ENGINE0            = 0x1E0,
	AMDGPU_DOORBELL_sDMA_ENGINE1            = 0x1E1,
	AMDGPU_DOORBELL_IH                      = 0x1E8,
	AMDGPU_DOORBELL_MAX_ASSIGNMENT          = 0x3FF,
	AMDGPU_DOORBELL_INVALID                 = 0xFFFF
} AMDGPU_DOORBELL_ASSIGNMENT;

struct amdgpu_doorbell {
	/* doorbell mmio */
	resource_size_t		base;
	resource_size_t		size;
	u32 __iomem		*ptr;
	u32			num_doorbells;	/* Number of doorbells actually reserved for amdgpu. */
};

void amdgpu_doorbell_get_kfd_info(struct amdgpu_device *adev,
				phys_addr_t *aperture_base,
				size_t *aperture_size,
				size_t *start_offset);

/*
 * IRQS.
 */

struct amdgpu_flip_work {
	struct work_struct		flip_work;
	struct work_struct		unpin_work;
	struct amdgpu_device		*adev;
	int				crtc_id;
	uint64_t			base;
	struct drm_pending_vblank_event *event;
	struct amdgpu_bo		*old_rbo;
714 715 716
	struct fence			*excl;
	unsigned			shared_count;
	struct fence			**shared;
717
	struct fence_cb			cb;
A
Alex Deucher 已提交
718 719 720 721 722 723 724 725 726 727 728 729 730 731
};


/*
 * CP & rings.
 */

struct amdgpu_ib {
	struct amdgpu_sa_bo		*sa_bo;
	uint32_t			length_dw;
	uint64_t			gpu_addr;
	uint32_t			*ptr;
	struct amdgpu_user_fence        *user;
	struct amdgpu_vm		*vm;
732 733
	unsigned			vm_id;
	uint64_t			vm_pd_addr;
734
	struct amdgpu_ctx		*ctx;
A
Alex Deucher 已提交
735 736 737
	uint32_t			gds_base, gds_size;
	uint32_t			gws_base, gws_size;
	uint32_t			oa_base, oa_size;
738
	uint32_t			flags;
739 740
	/* resulting sequence number */
	uint64_t			sequence;
A
Alex Deucher 已提交
741 742 743 744 745 746 747 748 749 750
};

enum amdgpu_ring_type {
	AMDGPU_RING_TYPE_GFX,
	AMDGPU_RING_TYPE_COMPUTE,
	AMDGPU_RING_TYPE_SDMA,
	AMDGPU_RING_TYPE_UVD,
	AMDGPU_RING_TYPE_VCE
};

751
extern const struct amd_sched_backend_ops amdgpu_sched_ops;
752

753 754
int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs,
		     struct amdgpu_job **job);
755 756
int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev, unsigned size,
			     struct amdgpu_job **job);
M
Monk Liu 已提交
757

758
void amdgpu_job_free(struct amdgpu_job *job);
M
Monk Liu 已提交
759
void amdgpu_job_free_func(struct kref *refcount);
760
int amdgpu_job_submit(struct amdgpu_job *job, struct amdgpu_ring *ring,
761 762
		      struct amd_sched_entity *entity, void *owner,
		      struct fence **f);
763
void amdgpu_job_timeout_func(struct work_struct *work);
764

A
Alex Deucher 已提交
765 766 767 768
struct amdgpu_ring {
	struct amdgpu_device		*adev;
	const struct amdgpu_ring_funcs	*funcs;
	struct amdgpu_fence_driver	fence_drv;
769
	struct amd_gpu_scheduler 	sched;
A
Alex Deucher 已提交
770

771
	spinlock_t              fence_lock;
A
Alex Deucher 已提交
772 773 774 775 776 777 778 779
	struct amdgpu_bo	*ring_obj;
	volatile uint32_t	*ring;
	unsigned		rptr_offs;
	u64			next_rptr_gpu_addr;
	volatile u32		*next_rptr_cpu_addr;
	unsigned		wptr;
	unsigned		wptr_old;
	unsigned		ring_size;
780
	unsigned		max_dw;
A
Alex Deucher 已提交
781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796
	int			count_dw;
	uint64_t		gpu_addr;
	uint32_t		align_mask;
	uint32_t		ptr_mask;
	bool			ready;
	u32			nop;
	u32			idx;
	u32			me;
	u32			pipe;
	u32			queue;
	struct amdgpu_bo	*mqd_obj;
	u32			doorbell_index;
	bool			use_doorbell;
	unsigned		wptr_offs;
	unsigned		next_rptr_offs;
	unsigned		fence_offs;
797
	struct amdgpu_ctx	*current_ctx;
A
Alex Deucher 已提交
798 799
	enum amdgpu_ring_type	type;
	char			name[16];
M
Monk Liu 已提交
800 801 802
	unsigned		cond_exe_offs;
	u64				cond_exe_gpu_addr;
	volatile u32	*cond_exe_cpu_addr;
A
Alex Deucher 已提交
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
};

/*
 * VM
 */

/* maximum number of VMIDs */
#define AMDGPU_NUM_VM	16

/* number of entries in page table */
#define AMDGPU_VM_PTE_COUNT (1 << amdgpu_vm_block_size)

/* PTBs (Page Table Blocks) need to be aligned to 32K */
#define AMDGPU_VM_PTB_ALIGN_SIZE   32768
#define AMDGPU_VM_PTB_ALIGN_MASK (AMDGPU_VM_PTB_ALIGN_SIZE - 1)
#define AMDGPU_VM_PTB_ALIGN(a) (((a) + AMDGPU_VM_PTB_ALIGN_MASK) & ~AMDGPU_VM_PTB_ALIGN_MASK)

#define AMDGPU_PTE_VALID	(1 << 0)
#define AMDGPU_PTE_SYSTEM	(1 << 1)
#define AMDGPU_PTE_SNOOPED	(1 << 2)

/* VI only */
#define AMDGPU_PTE_EXECUTABLE	(1 << 4)

#define AMDGPU_PTE_READABLE	(1 << 5)
#define AMDGPU_PTE_WRITEABLE	(1 << 6)

/* PTE (Page Table Entry) fragment field for different page sizes */
#define AMDGPU_PTE_FRAG_4KB	(0 << 7)
#define AMDGPU_PTE_FRAG_64KB	(4 << 7)
#define AMDGPU_LOG2_PAGES_PER_FRAG 4

835 836 837 838 839
/* How to programm VM fault handling */
#define AMDGPU_VM_FAULT_STOP_NEVER	0
#define AMDGPU_VM_FAULT_STOP_FIRST	1
#define AMDGPU_VM_FAULT_STOP_ALWAYS	2

A
Alex Deucher 已提交
840
struct amdgpu_vm_pt {
841 842
	struct amdgpu_bo_list_entry	entry;
	uint64_t			addr;
A
Alex Deucher 已提交
843 844 845
};

struct amdgpu_vm {
846
	/* tree of virtual addresses mapped */
A
Alex Deucher 已提交
847 848
	struct rb_root		va;

849
	/* protecting invalidated */
A
Alex Deucher 已提交
850 851 852 853 854
	spinlock_t		status_lock;

	/* BOs moved, but not yet updated in the PT */
	struct list_head	invalidated;

855 856 857 858
	/* BOs cleared in the PT because of a move */
	struct list_head	cleared;

	/* BO mappings freed, but not yet updated in the PT */
A
Alex Deucher 已提交
859 860 861 862 863
	struct list_head	freed;

	/* contains the page directory */
	struct amdgpu_bo	*page_directory;
	unsigned		max_pde_used;
864
	struct fence		*page_directory_fence;
A
Alex Deucher 已提交
865 866 867 868 869

	/* array of page tables, one for each page directory entry */
	struct amdgpu_vm_pt	*page_tables;

	/* for id and flush management per ring */
870
	struct amdgpu_vm_id	*ids[AMDGPU_MAX_RINGS];
871

872 873
	/* protecting freed */
	spinlock_t		freed_lock;
874 875 876

	/* Scheduler entity for page table updates */
	struct amd_sched_entity	entity;
A
Alex Deucher 已提交
877 878
};

879
struct amdgpu_vm_id {
880
	struct list_head	list;
881 882
	struct fence		*first;
	struct amdgpu_sync	active;
883
	struct fence		*last_flush;
884
	atomic_long_t		owner;
885

886 887 888 889
	uint64_t		pd_gpu_addr;
	/* last flushed PD/PT update */
	struct fence		*flushed_updates;

890 891 892 893 894 895
	uint32_t		gds_base;
	uint32_t		gds_size;
	uint32_t		gws_base;
	uint32_t		gws_size;
	uint32_t		oa_base;
	uint32_t		oa_size;
896 897
};

A
Alex Deucher 已提交
898
struct amdgpu_vm_manager {
899
	/* Handling of VMIDs */
900
	struct mutex				lock;
901 902
	unsigned				num_ids;
	struct list_head			ids_lru;
903
	struct amdgpu_vm_id			ids[AMDGPU_NUM_VM];
904

905
	uint32_t				max_pfn;
A
Alex Deucher 已提交
906
	/* vram base address for page table entry  */
907
	u64					vram_base_offset;
A
Alex Deucher 已提交
908
	/* is vm enabled? */
909
	bool					enabled;
A
Alex Deucher 已提交
910 911
	/* vm pte handling */
	const struct amdgpu_vm_pte_funcs        *vm_pte_funcs;
912 913 914
	struct amdgpu_ring                      *vm_pte_rings[AMDGPU_MAX_RINGS];
	unsigned				vm_pte_num_rings;
	atomic_t				vm_pte_next_ring;
A
Alex Deucher 已提交
915 916
};

917
void amdgpu_vm_manager_init(struct amdgpu_device *adev);
918
void amdgpu_vm_manager_fini(struct amdgpu_device *adev);
919 920
int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm);
void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm);
921 922 923
void amdgpu_vm_get_pd_bo(struct amdgpu_vm *vm,
			 struct list_head *validated,
			 struct amdgpu_bo_list_entry *entry);
924
void amdgpu_vm_get_pt_bos(struct amdgpu_vm *vm, struct list_head *duplicates);
925 926
void amdgpu_vm_move_pt_bos_in_lru(struct amdgpu_device *adev,
				  struct amdgpu_vm *vm);
927
int amdgpu_vm_grab_id(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
928 929
		      struct amdgpu_sync *sync, struct fence *fence,
		      unsigned *vm_id, uint64_t *vm_pd_addr);
930 931 932 933 934
int amdgpu_vm_flush(struct amdgpu_ring *ring,
		    unsigned vm_id, uint64_t pd_addr,
		    uint32_t gds_base, uint32_t gds_size,
		    uint32_t gws_base, uint32_t gws_size,
		    uint32_t oa_base, uint32_t oa_size);
935
void amdgpu_vm_reset_id(struct amdgpu_device *adev, unsigned vm_id);
936
uint64_t amdgpu_vm_map_gart(const dma_addr_t *pages_addr, uint64_t addr);
937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962
int amdgpu_vm_update_page_directory(struct amdgpu_device *adev,
				    struct amdgpu_vm *vm);
int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
			  struct amdgpu_vm *vm);
int amdgpu_vm_clear_invalids(struct amdgpu_device *adev, struct amdgpu_vm *vm,
			     struct amdgpu_sync *sync);
int amdgpu_vm_bo_update(struct amdgpu_device *adev,
			struct amdgpu_bo_va *bo_va,
			struct ttm_mem_reg *mem);
void amdgpu_vm_bo_invalidate(struct amdgpu_device *adev,
			     struct amdgpu_bo *bo);
struct amdgpu_bo_va *amdgpu_vm_bo_find(struct amdgpu_vm *vm,
				       struct amdgpu_bo *bo);
struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
				      struct amdgpu_vm *vm,
				      struct amdgpu_bo *bo);
int amdgpu_vm_bo_map(struct amdgpu_device *adev,
		     struct amdgpu_bo_va *bo_va,
		     uint64_t addr, uint64_t offset,
		     uint64_t size, uint32_t flags);
int amdgpu_vm_bo_unmap(struct amdgpu_device *adev,
		       struct amdgpu_bo_va *bo_va,
		       uint64_t addr);
void amdgpu_vm_bo_rmv(struct amdgpu_device *adev,
		      struct amdgpu_bo_va *bo_va);

A
Alex Deucher 已提交
963 964 965 966
/*
 * context related structures
 */

967
struct amdgpu_ctx_ring {
968
	uint64_t		sequence;
969
	struct fence		**fences;
970
	struct amd_sched_entity	entity;
971 972
};

A
Alex Deucher 已提交
973
struct amdgpu_ctx {
974
	struct kref		refcount;
975
	struct amdgpu_device    *adev;
976
	unsigned		reset_counter;
977
	spinlock_t		ring_lock;
978
	struct fence            **fences;
979
	struct amdgpu_ctx_ring	rings[AMDGPU_MAX_RINGS];
A
Alex Deucher 已提交
980 981 982
};

struct amdgpu_ctx_mgr {
983 984 985 986
	struct amdgpu_device	*adev;
	struct mutex		lock;
	/* protected by lock */
	struct idr		ctx_handles;
A
Alex Deucher 已提交
987 988
};

989 990 991
struct amdgpu_ctx *amdgpu_ctx_get(struct amdgpu_fpriv *fpriv, uint32_t id);
int amdgpu_ctx_put(struct amdgpu_ctx *ctx);

992
uint64_t amdgpu_ctx_add_fence(struct amdgpu_ctx *ctx, struct amdgpu_ring *ring,
993
			      struct fence *fence);
994 995 996
struct fence *amdgpu_ctx_get_fence(struct amdgpu_ctx *ctx,
				   struct amdgpu_ring *ring, uint64_t seq);

997 998 999
int amdgpu_ctx_ioctl(struct drm_device *dev, void *data,
		     struct drm_file *filp);

1000 1001
void amdgpu_ctx_mgr_init(struct amdgpu_ctx_mgr *mgr);
void amdgpu_ctx_mgr_fini(struct amdgpu_ctx_mgr *mgr);
1002

A
Alex Deucher 已提交
1003 1004 1005 1006 1007 1008 1009 1010
/*
 * file private structure
 */

struct amdgpu_fpriv {
	struct amdgpu_vm	vm;
	struct mutex		bo_list_lock;
	struct idr		bo_list_handles;
1011
	struct amdgpu_ctx_mgr	ctx_mgr;
A
Alex Deucher 已提交
1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022
};

/*
 * residency list
 */

struct amdgpu_bo_list {
	struct mutex lock;
	struct amdgpu_bo *gds_obj;
	struct amdgpu_bo *gws_obj;
	struct amdgpu_bo *oa_obj;
1023
	unsigned first_userptr;
A
Alex Deucher 已提交
1024 1025 1026 1027 1028 1029
	unsigned num_entries;
	struct amdgpu_bo_list_entry *array;
};

struct amdgpu_bo_list *
amdgpu_bo_list_get(struct amdgpu_fpriv *fpriv, int id);
1030 1031
void amdgpu_bo_list_get_list(struct amdgpu_bo_list *list,
			     struct list_head *validated);
A
Alex Deucher 已提交
1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104
void amdgpu_bo_list_put(struct amdgpu_bo_list *list);
void amdgpu_bo_list_free(struct amdgpu_bo_list *list);

/*
 * GFX stuff
 */
#include "clearstate_defs.h"

struct amdgpu_rlc {
	/* for power gating */
	struct amdgpu_bo	*save_restore_obj;
	uint64_t		save_restore_gpu_addr;
	volatile uint32_t	*sr_ptr;
	const u32               *reg_list;
	u32                     reg_list_size;
	/* for clear state */
	struct amdgpu_bo	*clear_state_obj;
	uint64_t		clear_state_gpu_addr;
	volatile uint32_t	*cs_ptr;
	const struct cs_section_def   *cs_data;
	u32                     clear_state_size;
	/* for cp tables */
	struct amdgpu_bo	*cp_table_obj;
	uint64_t		cp_table_gpu_addr;
	volatile uint32_t	*cp_table_ptr;
	u32                     cp_table_size;
};

struct amdgpu_mec {
	struct amdgpu_bo	*hpd_eop_obj;
	u64			hpd_eop_gpu_addr;
	u32 num_pipe;
	u32 num_mec;
	u32 num_queue;
};

/*
 * GPU scratch registers structures, functions & helpers
 */
struct amdgpu_scratch {
	unsigned		num_reg;
	uint32_t                reg_base;
	bool			free[32];
	uint32_t		reg[32];
};

/*
 * GFX configurations
 */
struct amdgpu_gca_config {
	unsigned max_shader_engines;
	unsigned max_tile_pipes;
	unsigned max_cu_per_sh;
	unsigned max_sh_per_se;
	unsigned max_backends_per_se;
	unsigned max_texture_channel_caches;
	unsigned max_gprs;
	unsigned max_gs_threads;
	unsigned max_hw_contexts;
	unsigned sc_prim_fifo_size_frontend;
	unsigned sc_prim_fifo_size_backend;
	unsigned sc_hiz_tile_fifo_size;
	unsigned sc_earlyz_tile_fifo_size;

	unsigned num_tile_pipes;
	unsigned backend_enable_mask;
	unsigned mem_max_burst_length_bytes;
	unsigned mem_row_size_in_kb;
	unsigned shader_engine_tile_size;
	unsigned num_gpus;
	unsigned multi_gpu_tile_size;
	unsigned mc_arb_ramcfg;
	unsigned gb_addr_config;
1105
	unsigned num_rbs;
A
Alex Deucher 已提交
1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128

	uint32_t tile_mode_array[32];
	uint32_t macrotile_mode_array[16];
};

struct amdgpu_gfx {
	struct mutex			gpu_clock_mutex;
	struct amdgpu_gca_config	config;
	struct amdgpu_rlc		rlc;
	struct amdgpu_mec		mec;
	struct amdgpu_scratch		scratch;
	const struct firmware		*me_fw;	/* ME firmware */
	uint32_t			me_fw_version;
	const struct firmware		*pfp_fw; /* PFP firmware */
	uint32_t			pfp_fw_version;
	const struct firmware		*ce_fw;	/* CE firmware */
	uint32_t			ce_fw_version;
	const struct firmware		*rlc_fw; /* RLC firmware */
	uint32_t			rlc_fw_version;
	const struct firmware		*mec_fw; /* MEC firmware */
	uint32_t			mec_fw_version;
	const struct firmware		*mec2_fw; /* MEC2 firmware */
	uint32_t			mec2_fw_version;
1129 1130 1131
	uint32_t			me_feature_version;
	uint32_t			ce_feature_version;
	uint32_t			pfp_feature_version;
1132 1133 1134
	uint32_t			rlc_feature_version;
	uint32_t			mec_feature_version;
	uint32_t			mec2_feature_version;
A
Alex Deucher 已提交
1135 1136 1137 1138 1139 1140 1141 1142 1143
	struct amdgpu_ring		gfx_ring[AMDGPU_MAX_GFX_RINGS];
	unsigned			num_gfx_rings;
	struct amdgpu_ring		compute_ring[AMDGPU_MAX_COMPUTE_RINGS];
	unsigned			num_compute_rings;
	struct amdgpu_irq_src		eop_irq;
	struct amdgpu_irq_src		priv_reg_irq;
	struct amdgpu_irq_src		priv_inst_irq;
	/* gfx status */
	uint32_t gfx_current_status;
1144 1145
	/* ce ram size*/
	unsigned ce_ram_size;
A
Alex Deucher 已提交
1146 1147
};

1148
int amdgpu_ib_get(struct amdgpu_device *adev, struct amdgpu_vm *vm,
A
Alex Deucher 已提交
1149
		  unsigned size, struct amdgpu_ib *ib);
1150
void amdgpu_ib_free(struct amdgpu_device *adev, struct amdgpu_ib *ib, struct fence *f);
1151
int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
1152
		       struct amdgpu_ib *ib, struct fence *last_vm_update,
1153
		       struct fence **f);
A
Alex Deucher 已提交
1154 1155 1156 1157
int amdgpu_ib_pool_init(struct amdgpu_device *adev);
void amdgpu_ib_pool_fini(struct amdgpu_device *adev);
int amdgpu_ib_ring_tests(struct amdgpu_device *adev);
int amdgpu_ring_alloc(struct amdgpu_ring *ring, unsigned ndw);
1158
void amdgpu_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count);
1159
void amdgpu_ring_generic_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib);
A
Alex Deucher 已提交
1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183
void amdgpu_ring_commit(struct amdgpu_ring *ring);
void amdgpu_ring_undo(struct amdgpu_ring *ring);
unsigned amdgpu_ring_backup(struct amdgpu_ring *ring,
			    uint32_t **data);
int amdgpu_ring_restore(struct amdgpu_ring *ring,
			unsigned size, uint32_t *data);
int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
		     unsigned ring_size, u32 nop, u32 align_mask,
		     struct amdgpu_irq_src *irq_src, unsigned irq_type,
		     enum amdgpu_ring_type ring_type);
void amdgpu_ring_fini(struct amdgpu_ring *ring);

/*
 * CS.
 */
struct amdgpu_cs_chunk {
	uint32_t		chunk_id;
	uint32_t		length_dw;
	uint32_t		*kdata;
};

struct amdgpu_cs_parser {
	struct amdgpu_device	*adev;
	struct drm_file		*filp;
1184
	struct amdgpu_ctx	*ctx;
1185

A
Alex Deucher 已提交
1186 1187 1188 1189
	/* chunks */
	unsigned		nchunks;
	struct amdgpu_cs_chunk	*chunks;

1190 1191
	/* scheduler job object */
	struct amdgpu_job	*job;
A
Alex Deucher 已提交
1192

1193 1194 1195 1196 1197 1198 1199 1200
	/* buffer objects */
	struct ww_acquire_ctx		ticket;
	struct amdgpu_bo_list		*bo_list;
	struct amdgpu_bo_list_entry	vm_pd;
	struct list_head		validated;
	struct fence			*fence;
	uint64_t			bytes_moved_threshold;
	uint64_t			bytes_moved;
A
Alex Deucher 已提交
1201 1202

	/* user fence */
1203
	struct amdgpu_bo_list_entry	uf_entry;
A
Alex Deucher 已提交
1204 1205
};

1206 1207 1208
struct amdgpu_job {
	struct amd_sched_job    base;
	struct amdgpu_device	*adev;
1209
	struct amdgpu_ring	*ring;
1210
	struct amdgpu_sync	sync;
1211
	struct amdgpu_ib	*ibs;
1212
	struct fence		*fence; /* the hw fence */
1213
	uint32_t		num_ibs;
1214
	void			*owner;
1215 1216
	struct amdgpu_user_fence uf;
};
1217 1218
#define to_amdgpu_job(sched_job)		\
		container_of((sched_job), struct amdgpu_job, base)
1219

1220 1221
static inline u32 amdgpu_get_ib_value(struct amdgpu_cs_parser *p,
				      uint32_t ib_idx, int idx)
A
Alex Deucher 已提交
1222
{
1223
	return p->job->ibs[ib_idx].ptr[idx];
A
Alex Deucher 已提交
1224 1225
}

1226 1227 1228 1229
static inline void amdgpu_set_ib_value(struct amdgpu_cs_parser *p,
				       uint32_t ib_idx, int idx,
				       uint32_t value)
{
1230
	p->job->ibs[ib_idx].ptr[idx] = value;
1231 1232
}

A
Alex Deucher 已提交
1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481
/*
 * Writeback
 */
#define AMDGPU_MAX_WB 1024	/* Reserve at most 1024 WB slots for amdgpu-owned rings. */

struct amdgpu_wb {
	struct amdgpu_bo	*wb_obj;
	volatile uint32_t	*wb;
	uint64_t		gpu_addr;
	u32			num_wb;	/* Number of wb slots actually reserved for amdgpu. */
	unsigned long		used[DIV_ROUND_UP(AMDGPU_MAX_WB, BITS_PER_LONG)];
};

int amdgpu_wb_get(struct amdgpu_device *adev, u32 *wb);
void amdgpu_wb_free(struct amdgpu_device *adev, u32 wb);



enum amdgpu_int_thermal_type {
	THERMAL_TYPE_NONE,
	THERMAL_TYPE_EXTERNAL,
	THERMAL_TYPE_EXTERNAL_GPIO,
	THERMAL_TYPE_RV6XX,
	THERMAL_TYPE_RV770,
	THERMAL_TYPE_ADT7473_WITH_INTERNAL,
	THERMAL_TYPE_EVERGREEN,
	THERMAL_TYPE_SUMO,
	THERMAL_TYPE_NI,
	THERMAL_TYPE_SI,
	THERMAL_TYPE_EMC2103_WITH_INTERNAL,
	THERMAL_TYPE_CI,
	THERMAL_TYPE_KV,
};

enum amdgpu_dpm_auto_throttle_src {
	AMDGPU_DPM_AUTO_THROTTLE_SRC_THERMAL,
	AMDGPU_DPM_AUTO_THROTTLE_SRC_EXTERNAL
};

enum amdgpu_dpm_event_src {
	AMDGPU_DPM_EVENT_SRC_ANALOG = 0,
	AMDGPU_DPM_EVENT_SRC_EXTERNAL = 1,
	AMDGPU_DPM_EVENT_SRC_DIGITAL = 2,
	AMDGPU_DPM_EVENT_SRC_ANALOG_OR_EXTERNAL = 3,
	AMDGPU_DPM_EVENT_SRC_DIGIAL_OR_EXTERNAL = 4
};

#define AMDGPU_MAX_VCE_LEVELS 6

enum amdgpu_vce_level {
	AMDGPU_VCE_LEVEL_AC_ALL = 0,     /* AC, All cases */
	AMDGPU_VCE_LEVEL_DC_EE = 1,      /* DC, entropy encoding */
	AMDGPU_VCE_LEVEL_DC_LL_LOW = 2,  /* DC, low latency queue, res <= 720 */
	AMDGPU_VCE_LEVEL_DC_LL_HIGH = 3, /* DC, low latency queue, 1080 >= res > 720 */
	AMDGPU_VCE_LEVEL_DC_GP_LOW = 4,  /* DC, general purpose queue, res <= 720 */
	AMDGPU_VCE_LEVEL_DC_GP_HIGH = 5, /* DC, general purpose queue, 1080 >= res > 720 */
};

struct amdgpu_ps {
	u32 caps; /* vbios flags */
	u32 class; /* vbios flags */
	u32 class2; /* vbios flags */
	/* UVD clocks */
	u32 vclk;
	u32 dclk;
	/* VCE clocks */
	u32 evclk;
	u32 ecclk;
	bool vce_active;
	enum amdgpu_vce_level vce_level;
	/* asic priv */
	void *ps_priv;
};

struct amdgpu_dpm_thermal {
	/* thermal interrupt work */
	struct work_struct work;
	/* low temperature threshold */
	int                min_temp;
	/* high temperature threshold */
	int                max_temp;
	/* was last interrupt low to high or high to low */
	bool               high_to_low;
	/* interrupt source */
	struct amdgpu_irq_src	irq;
};

enum amdgpu_clk_action
{
	AMDGPU_SCLK_UP = 1,
	AMDGPU_SCLK_DOWN
};

struct amdgpu_blacklist_clocks
{
	u32 sclk;
	u32 mclk;
	enum amdgpu_clk_action action;
};

struct amdgpu_clock_and_voltage_limits {
	u32 sclk;
	u32 mclk;
	u16 vddc;
	u16 vddci;
};

struct amdgpu_clock_array {
	u32 count;
	u32 *values;
};

struct amdgpu_clock_voltage_dependency_entry {
	u32 clk;
	u16 v;
};

struct amdgpu_clock_voltage_dependency_table {
	u32 count;
	struct amdgpu_clock_voltage_dependency_entry *entries;
};

union amdgpu_cac_leakage_entry {
	struct {
		u16 vddc;
		u32 leakage;
	};
	struct {
		u16 vddc1;
		u16 vddc2;
		u16 vddc3;
	};
};

struct amdgpu_cac_leakage_table {
	u32 count;
	union amdgpu_cac_leakage_entry *entries;
};

struct amdgpu_phase_shedding_limits_entry {
	u16 voltage;
	u32 sclk;
	u32 mclk;
};

struct amdgpu_phase_shedding_limits_table {
	u32 count;
	struct amdgpu_phase_shedding_limits_entry *entries;
};

struct amdgpu_uvd_clock_voltage_dependency_entry {
	u32 vclk;
	u32 dclk;
	u16 v;
};

struct amdgpu_uvd_clock_voltage_dependency_table {
	u8 count;
	struct amdgpu_uvd_clock_voltage_dependency_entry *entries;
};

struct amdgpu_vce_clock_voltage_dependency_entry {
	u32 ecclk;
	u32 evclk;
	u16 v;
};

struct amdgpu_vce_clock_voltage_dependency_table {
	u8 count;
	struct amdgpu_vce_clock_voltage_dependency_entry *entries;
};

struct amdgpu_ppm_table {
	u8 ppm_design;
	u16 cpu_core_number;
	u32 platform_tdp;
	u32 small_ac_platform_tdp;
	u32 platform_tdc;
	u32 small_ac_platform_tdc;
	u32 apu_tdp;
	u32 dgpu_tdp;
	u32 dgpu_ulv_power;
	u32 tj_max;
};

struct amdgpu_cac_tdp_table {
	u16 tdp;
	u16 configurable_tdp;
	u16 tdc;
	u16 battery_power_limit;
	u16 small_power_limit;
	u16 low_cac_leakage;
	u16 high_cac_leakage;
	u16 maximum_power_delivery_limit;
};

struct amdgpu_dpm_dynamic_state {
	struct amdgpu_clock_voltage_dependency_table vddc_dependency_on_sclk;
	struct amdgpu_clock_voltage_dependency_table vddci_dependency_on_mclk;
	struct amdgpu_clock_voltage_dependency_table vddc_dependency_on_mclk;
	struct amdgpu_clock_voltage_dependency_table mvdd_dependency_on_mclk;
	struct amdgpu_clock_voltage_dependency_table vddc_dependency_on_dispclk;
	struct amdgpu_uvd_clock_voltage_dependency_table uvd_clock_voltage_dependency_table;
	struct amdgpu_vce_clock_voltage_dependency_table vce_clock_voltage_dependency_table;
	struct amdgpu_clock_voltage_dependency_table samu_clock_voltage_dependency_table;
	struct amdgpu_clock_voltage_dependency_table acp_clock_voltage_dependency_table;
	struct amdgpu_clock_voltage_dependency_table vddgfx_dependency_on_sclk;
	struct amdgpu_clock_array valid_sclk_values;
	struct amdgpu_clock_array valid_mclk_values;
	struct amdgpu_clock_and_voltage_limits max_clock_voltage_on_dc;
	struct amdgpu_clock_and_voltage_limits max_clock_voltage_on_ac;
	u32 mclk_sclk_ratio;
	u32 sclk_mclk_delta;
	u16 vddc_vddci_delta;
	u16 min_vddc_for_pcie_gen2;
	struct amdgpu_cac_leakage_table cac_leakage_table;
	struct amdgpu_phase_shedding_limits_table phase_shedding_limits_table;
	struct amdgpu_ppm_table *ppm_table;
	struct amdgpu_cac_tdp_table *cac_tdp_table;
};

struct amdgpu_dpm_fan {
	u16 t_min;
	u16 t_med;
	u16 t_high;
	u16 pwm_min;
	u16 pwm_med;
	u16 pwm_high;
	u8 t_hyst;
	u32 cycle_delay;
	u16 t_max;
	u8 control_mode;
	u16 default_max_fan_pwm;
	u16 default_fan_output_sensitivity;
	u16 fan_output_sensitivity;
	bool ucode_fan_control;
};

enum amdgpu_pcie_gen {
	AMDGPU_PCIE_GEN1 = 0,
	AMDGPU_PCIE_GEN2 = 1,
	AMDGPU_PCIE_GEN3 = 2,
	AMDGPU_PCIE_GEN_INVALID = 0xffff
};

enum amdgpu_dpm_forced_level {
	AMDGPU_DPM_FORCED_LEVEL_AUTO = 0,
	AMDGPU_DPM_FORCED_LEVEL_LOW = 1,
	AMDGPU_DPM_FORCED_LEVEL_HIGH = 2,
1482
	AMDGPU_DPM_FORCED_LEVEL_MANUAL = 3,
A
Alex Deucher 已提交
1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508
};

struct amdgpu_vce_state {
	/* vce clocks */
	u32 evclk;
	u32 ecclk;
	/* gpu clocks */
	u32 sclk;
	u32 mclk;
	u8 clk_idx;
	u8 pstate;
};

struct amdgpu_dpm_funcs {
	int (*get_temperature)(struct amdgpu_device *adev);
	int (*pre_set_power_state)(struct amdgpu_device *adev);
	int (*set_power_state)(struct amdgpu_device *adev);
	void (*post_set_power_state)(struct amdgpu_device *adev);
	void (*display_configuration_changed)(struct amdgpu_device *adev);
	u32 (*get_sclk)(struct amdgpu_device *adev, bool low);
	u32 (*get_mclk)(struct amdgpu_device *adev, bool low);
	void (*print_power_state)(struct amdgpu_device *adev, struct amdgpu_ps *ps);
	void (*debugfs_print_current_performance_level)(struct amdgpu_device *adev, struct seq_file *m);
	int (*force_performance_level)(struct amdgpu_device *adev, enum amdgpu_dpm_forced_level level);
	bool (*vblank_too_short)(struct amdgpu_device *adev);
	void (*powergate_uvd)(struct amdgpu_device *adev, bool gate);
S
Sonny Jiang 已提交
1509
	void (*powergate_vce)(struct amdgpu_device *adev, bool gate);
A
Alex Deucher 已提交
1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531
	void (*enable_bapm)(struct amdgpu_device *adev, bool enable);
	void (*set_fan_control_mode)(struct amdgpu_device *adev, u32 mode);
	u32 (*get_fan_control_mode)(struct amdgpu_device *adev);
	int (*set_fan_speed_percent)(struct amdgpu_device *adev, u32 speed);
	int (*get_fan_speed_percent)(struct amdgpu_device *adev, u32 *speed);
};

struct amdgpu_dpm {
	struct amdgpu_ps        *ps;
	/* number of valid power states */
	int                     num_ps;
	/* current power state that is active */
	struct amdgpu_ps        *current_ps;
	/* requested power state */
	struct amdgpu_ps        *requested_ps;
	/* boot up power state */
	struct amdgpu_ps        *boot_ps;
	/* default uvd power state */
	struct amdgpu_ps        *uvd_ps;
	/* vce requirements */
	struct amdgpu_vce_state vce_states[AMDGPU_MAX_VCE_LEVELS];
	enum amdgpu_vce_level vce_level;
1532 1533
	enum amd_pm_state_type state;
	enum amd_pm_state_type user_state;
A
Alex Deucher 已提交
1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580
	u32                     platform_caps;
	u32                     voltage_response_time;
	u32                     backbias_response_time;
	void                    *priv;
	u32			new_active_crtcs;
	int			new_active_crtc_count;
	u32			current_active_crtcs;
	int			current_active_crtc_count;
	struct amdgpu_dpm_dynamic_state dyn_state;
	struct amdgpu_dpm_fan fan;
	u32 tdp_limit;
	u32 near_tdp_limit;
	u32 near_tdp_limit_adjusted;
	u32 sq_ramping_threshold;
	u32 cac_leakage;
	u16 tdp_od_limit;
	u32 tdp_adjustment;
	u16 load_line_slope;
	bool power_control;
	bool ac_power;
	/* special states active */
	bool                    thermal_active;
	bool                    uvd_active;
	bool                    vce_active;
	/* thermal handling */
	struct amdgpu_dpm_thermal thermal;
	/* forced levels */
	enum amdgpu_dpm_forced_level forced_level;
};

struct amdgpu_pm {
	struct mutex		mutex;
	u32                     current_sclk;
	u32                     current_mclk;
	u32                     default_sclk;
	u32                     default_mclk;
	struct amdgpu_i2c_chan *i2c_bus;
	/* internal thermal controller on rv6xx+ */
	enum amdgpu_int_thermal_type int_thermal_type;
	struct device	        *int_hwmon_dev;
	/* fan control parameters */
	bool                    no_fan;
	u8                      fan_pulses_per_revolution;
	u8                      fan_min_rpm;
	u8                      fan_max_rpm;
	/* dpm */
	bool                    dpm_enabled;
1581
	bool                    sysfs_initialized;
A
Alex Deucher 已提交
1582 1583 1584 1585
	struct amdgpu_dpm       dpm;
	const struct firmware	*fw;	/* SMC firmware */
	uint32_t                fw_version;
	const struct amdgpu_dpm_funcs *funcs;
1586 1587
	uint32_t                pcie_gen_mask;
	uint32_t                pcie_mlw_mask;
1588
	struct amd_pp_display_configuration pm_display_cfg;/* set by DAL */
A
Alex Deucher 已提交
1589 1590
};

1591 1592
void amdgpu_get_pcie_info(struct amdgpu_device *adev);

A
Alex Deucher 已提交
1593 1594 1595
/*
 * UVD
 */
1596 1597 1598 1599 1600 1601
#define AMDGPU_DEFAULT_UVD_HANDLES	10
#define AMDGPU_MAX_UVD_HANDLES		40
#define AMDGPU_UVD_STACK_SIZE		(200*1024)
#define AMDGPU_UVD_HEAP_SIZE		(256*1024)
#define AMDGPU_UVD_SESSION_SIZE		(50*1024)
#define AMDGPU_UVD_FIRMWARE_OFFSET	256
A
Alex Deucher 已提交
1602 1603 1604 1605 1606

struct amdgpu_uvd {
	struct amdgpu_bo	*vcpu_bo;
	void			*cpu_addr;
	uint64_t		gpu_addr;
1607
	void			*saved_bo;
1608
	unsigned		max_handles;
A
Alex Deucher 已提交
1609 1610 1611 1612 1613 1614 1615
	atomic_t		handles[AMDGPU_MAX_UVD_HANDLES];
	struct drm_file		*filp[AMDGPU_MAX_UVD_HANDLES];
	struct delayed_work	idle_work;
	const struct firmware	*fw;	/* UVD firmware */
	struct amdgpu_ring	ring;
	struct amdgpu_irq_src	irq;
	bool			address_64_bit;
1616
	struct amd_sched_entity entity;
A
Alex Deucher 已提交
1617 1618 1619 1620 1621 1622 1623 1624
};

/*
 * VCE
 */
#define AMDGPU_MAX_VCE_HANDLES	16
#define AMDGPU_VCE_FIRMWARE_OFFSET 256

1625 1626 1627
#define AMDGPU_VCE_HARVEST_VCE0 (1 << 0)
#define AMDGPU_VCE_HARVEST_VCE1 (1 << 1)

A
Alex Deucher 已提交
1628 1629 1630 1631 1632 1633 1634
struct amdgpu_vce {
	struct amdgpu_bo	*vcpu_bo;
	uint64_t		gpu_addr;
	unsigned		fw_version;
	unsigned		fb_version;
	atomic_t		handles[AMDGPU_MAX_VCE_HANDLES];
	struct drm_file		*filp[AMDGPU_MAX_VCE_HANDLES];
1635
	uint32_t		img_size[AMDGPU_MAX_VCE_HANDLES];
A
Alex Deucher 已提交
1636 1637 1638 1639
	struct delayed_work	idle_work;
	const struct firmware	*fw;	/* VCE firmware */
	struct amdgpu_ring	ring[AMDGPU_MAX_VCE_RINGS];
	struct amdgpu_irq_src	irq;
1640
	unsigned		harvest_config;
1641
	struct amd_sched_entity	entity;
A
Alex Deucher 已提交
1642 1643 1644 1645 1646
};

/*
 * SDMA
 */
A
Alex Deucher 已提交
1647
struct amdgpu_sdma_instance {
A
Alex Deucher 已提交
1648 1649 1650
	/* SDMA firmware */
	const struct firmware	*fw;
	uint32_t		fw_version;
1651
	uint32_t		feature_version;
A
Alex Deucher 已提交
1652 1653

	struct amdgpu_ring	ring;
1654
	bool			burst_nop;
A
Alex Deucher 已提交
1655 1656
};

A
Alex Deucher 已提交
1657 1658 1659 1660 1661 1662 1663
struct amdgpu_sdma {
	struct amdgpu_sdma_instance instance[AMDGPU_MAX_SDMA_INSTANCES];
	struct amdgpu_irq_src	trap_irq;
	struct amdgpu_irq_src	illegal_inst_irq;
	int 			num_instances;
};

A
Alex Deucher 已提交
1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695
/*
 * Firmware
 */
struct amdgpu_firmware {
	struct amdgpu_firmware_info ucode[AMDGPU_UCODE_ID_MAXIMUM];
	bool smu_load;
	struct amdgpu_bo *fw_buf;
	unsigned int fw_size;
};

/*
 * Benchmarking
 */
void amdgpu_benchmark(struct amdgpu_device *adev, int test_number);


/*
 * Testing
 */
void amdgpu_test_moves(struct amdgpu_device *adev);
void amdgpu_test_ring_sync(struct amdgpu_device *adev,
			   struct amdgpu_ring *cpA,
			   struct amdgpu_ring *cpB);
void amdgpu_test_syncing(struct amdgpu_device *adev);

/*
 * MMU Notifier
 */
#if defined(CONFIG_MMU_NOTIFIER)
int amdgpu_mn_register(struct amdgpu_bo *bo, unsigned long addr);
void amdgpu_mn_unregister(struct amdgpu_bo *bo);
#else
1696
static inline int amdgpu_mn_register(struct amdgpu_bo *bo, unsigned long addr)
A
Alex Deucher 已提交
1697 1698 1699
{
	return -ENODEV;
}
1700
static inline void amdgpu_mn_unregister(struct amdgpu_bo *bo) {}
A
Alex Deucher 已提交
1701 1702 1703 1704 1705 1706
#endif

/*
 * Debugfs
 */
struct amdgpu_debugfs {
1707
	const struct drm_info_list	*files;
A
Alex Deucher 已提交
1708 1709 1710 1711
	unsigned		num_files;
};

int amdgpu_debugfs_add_files(struct amdgpu_device *adev,
1712
			     const struct drm_info_list *files,
A
Alex Deucher 已提交
1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765
			     unsigned nfiles);
int amdgpu_debugfs_fence_init(struct amdgpu_device *adev);

#if defined(CONFIG_DEBUG_FS)
int amdgpu_debugfs_init(struct drm_minor *minor);
void amdgpu_debugfs_cleanup(struct drm_minor *minor);
#endif

/*
 * amdgpu smumgr functions
 */
struct amdgpu_smumgr_funcs {
	int (*check_fw_load_finish)(struct amdgpu_device *adev, uint32_t fwtype);
	int (*request_smu_load_fw)(struct amdgpu_device *adev);
	int (*request_smu_specific_fw)(struct amdgpu_device *adev, uint32_t fwtype);
};

/*
 * amdgpu smumgr
 */
struct amdgpu_smumgr {
	struct amdgpu_bo *toc_buf;
	struct amdgpu_bo *smu_buf;
	/* asic priv smu data */
	void *priv;
	spinlock_t smu_lock;
	/* smumgr functions */
	const struct amdgpu_smumgr_funcs *smumgr_funcs;
	/* ucode loading complete flag */
	uint32_t fw_flags;
};

/*
 * ASIC specific register table accessible by UMD
 */
struct amdgpu_allowed_register_entry {
	uint32_t reg_offset;
	bool untouched;
	bool grbm_indexed;
};

struct amdgpu_cu_info {
	uint32_t number; /* total active CU number */
	uint32_t ao_cu_mask;
	uint32_t bitmap[4][4];
};


/*
 * ASIC specific functions.
 */
struct amdgpu_asic_funcs {
	bool (*read_disabled_bios)(struct amdgpu_device *adev);
1766 1767
	bool (*read_bios_from_rom)(struct amdgpu_device *adev,
				   u8 *bios, u32 length_bytes);
A
Alex Deucher 已提交
1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867
	int (*read_register)(struct amdgpu_device *adev, u32 se_num,
			     u32 sh_num, u32 reg_offset, u32 *value);
	void (*set_vga_state)(struct amdgpu_device *adev, bool state);
	int (*reset)(struct amdgpu_device *adev);
	/* wait for mc_idle */
	int (*wait_for_mc_idle)(struct amdgpu_device *adev);
	/* get the reference clock */
	u32 (*get_xclk)(struct amdgpu_device *adev);
	/* get the gpu clock counter */
	uint64_t (*get_gpu_clock_counter)(struct amdgpu_device *adev);
	int (*get_cu_info)(struct amdgpu_device *adev, struct amdgpu_cu_info *info);
	/* MM block clocks */
	int (*set_uvd_clocks)(struct amdgpu_device *adev, u32 vclk, u32 dclk);
	int (*set_vce_clocks)(struct amdgpu_device *adev, u32 evclk, u32 ecclk);
};

/*
 * IOCTL.
 */
int amdgpu_gem_create_ioctl(struct drm_device *dev, void *data,
			    struct drm_file *filp);
int amdgpu_bo_list_ioctl(struct drm_device *dev, void *data,
				struct drm_file *filp);

int amdgpu_gem_info_ioctl(struct drm_device *dev, void *data,
			  struct drm_file *filp);
int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
			struct drm_file *filp);
int amdgpu_gem_mmap_ioctl(struct drm_device *dev, void *data,
			  struct drm_file *filp);
int amdgpu_gem_wait_idle_ioctl(struct drm_device *dev, void *data,
			      struct drm_file *filp);
int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
			  struct drm_file *filp);
int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data,
			struct drm_file *filp);
int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp);
int amdgpu_cs_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *filp);

int amdgpu_gem_metadata_ioctl(struct drm_device *dev, void *data,
				struct drm_file *filp);

/* VRAM scratch page for HDP bug, default vram page */
struct amdgpu_vram_scratch {
	struct amdgpu_bo		*robj;
	volatile uint32_t		*ptr;
	u64				gpu_addr;
};

/*
 * ACPI
 */
struct amdgpu_atif_notification_cfg {
	bool enabled;
	int command_code;
};

struct amdgpu_atif_notifications {
	bool display_switch;
	bool expansion_mode_change;
	bool thermal_state;
	bool forced_power_state;
	bool system_power_state;
	bool display_conf_change;
	bool px_gfx_switch;
	bool brightness_change;
	bool dgpu_display_event;
};

struct amdgpu_atif_functions {
	bool system_params;
	bool sbios_requests;
	bool select_active_disp;
	bool lid_state;
	bool get_tv_standard;
	bool set_tv_standard;
	bool get_panel_expansion_mode;
	bool set_panel_expansion_mode;
	bool temperature_change;
	bool graphics_device_types;
};

struct amdgpu_atif {
	struct amdgpu_atif_notifications notifications;
	struct amdgpu_atif_functions functions;
	struct amdgpu_atif_notification_cfg notification_cfg;
	struct amdgpu_encoder *encoder_for_bl;
};

struct amdgpu_atcs_functions {
	bool get_ext_state;
	bool pcie_perf_req;
	bool pcie_dev_rdy;
	bool pcie_bus_width;
};

struct amdgpu_atcs {
	struct amdgpu_atcs_functions functions;
};

C
Chunming Zhou 已提交
1868 1869 1870
/*
 * CGS
 */
1871 1872
struct cgs_device *amdgpu_cgs_create_device(struct amdgpu_device *adev);
void amdgpu_cgs_destroy_device(struct cgs_device *cgs_device);
1873 1874


1875 1876 1877 1878 1879
/* GPU virtualization */
struct amdgpu_virtualization {
	bool supports_sr_iov;
};

A
Alex Deucher 已提交
1880 1881 1882 1883 1884 1885 1886 1887 1888
/*
 * Core structure, functions and helpers.
 */
typedef uint32_t (*amdgpu_rreg_t)(struct amdgpu_device*, uint32_t);
typedef void (*amdgpu_wreg_t)(struct amdgpu_device*, uint32_t, uint32_t);

typedef uint32_t (*amdgpu_block_rreg_t)(struct amdgpu_device*, uint32_t, uint32_t);
typedef void (*amdgpu_block_wreg_t)(struct amdgpu_device*, uint32_t, uint32_t, uint32_t);

1889 1890 1891 1892 1893 1894
struct amdgpu_ip_block_status {
	bool valid;
	bool sw;
	bool hw;
};

A
Alex Deucher 已提交
1895 1896 1897 1898 1899
struct amdgpu_device {
	struct device			*dev;
	struct drm_device		*ddev;
	struct pci_dev			*pdev;

1900 1901 1902 1903
#ifdef CONFIG_DRM_AMD_ACP
	struct amdgpu_acp		acp;
#endif

A
Alex Deucher 已提交
1904
	/* ASIC */
1905
	enum amd_asic_type		asic_type;
A
Alex Deucher 已提交
1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983
	uint32_t			family;
	uint32_t			rev_id;
	uint32_t			external_rev_id;
	unsigned long			flags;
	int				usec_timeout;
	const struct amdgpu_asic_funcs	*asic_funcs;
	bool				shutdown;
	bool				need_dma32;
	bool				accel_working;
	struct work_struct 		reset_work;
	struct notifier_block		acpi_nb;
	struct amdgpu_i2c_chan		*i2c_bus[AMDGPU_MAX_I2C_BUS];
	struct amdgpu_debugfs		debugfs[AMDGPU_DEBUGFS_MAX_COMPONENTS];
	unsigned 			debugfs_count;
#if defined(CONFIG_DEBUG_FS)
	struct dentry			*debugfs_regs;
#endif
	struct amdgpu_atif		atif;
	struct amdgpu_atcs		atcs;
	struct mutex			srbm_mutex;
	/* GRBM index mutex. Protects concurrent access to GRBM index */
	struct mutex                    grbm_idx_mutex;
	struct dev_pm_domain		vga_pm_domain;
	bool				have_disp_power_ref;

	/* BIOS */
	uint8_t				*bios;
	bool				is_atom_bios;
	struct amdgpu_bo		*stollen_vga_memory;
	uint32_t			bios_scratch[AMDGPU_BIOS_NUM_SCRATCH];

	/* Register/doorbell mmio */
	resource_size_t			rmmio_base;
	resource_size_t			rmmio_size;
	void __iomem			*rmmio;
	/* protects concurrent MM_INDEX/DATA based register access */
	spinlock_t mmio_idx_lock;
	/* protects concurrent SMC based register access */
	spinlock_t smc_idx_lock;
	amdgpu_rreg_t			smc_rreg;
	amdgpu_wreg_t			smc_wreg;
	/* protects concurrent PCIE register access */
	spinlock_t pcie_idx_lock;
	amdgpu_rreg_t			pcie_rreg;
	amdgpu_wreg_t			pcie_wreg;
	/* protects concurrent UVD register access */
	spinlock_t uvd_ctx_idx_lock;
	amdgpu_rreg_t			uvd_ctx_rreg;
	amdgpu_wreg_t			uvd_ctx_wreg;
	/* protects concurrent DIDT register access */
	spinlock_t didt_idx_lock;
	amdgpu_rreg_t			didt_rreg;
	amdgpu_wreg_t			didt_wreg;
	/* protects concurrent ENDPOINT (audio) register access */
	spinlock_t audio_endpt_idx_lock;
	amdgpu_block_rreg_t		audio_endpt_rreg;
	amdgpu_block_wreg_t		audio_endpt_wreg;
	void __iomem                    *rio_mem;
	resource_size_t			rio_mem_size;
	struct amdgpu_doorbell		doorbell;

	/* clock/pll info */
	struct amdgpu_clock            clock;

	/* MC */
	struct amdgpu_mc		mc;
	struct amdgpu_gart		gart;
	struct amdgpu_dummy_page	dummy_page;
	struct amdgpu_vm_manager	vm_manager;

	/* memory management */
	struct amdgpu_mman		mman;
	struct amdgpu_vram_scratch	vram_scratch;
	struct amdgpu_wb		wb;
	atomic64_t			vram_usage;
	atomic64_t			vram_vis_usage;
	atomic64_t			gtt_usage;
	atomic64_t			num_bytes_moved;
1984
	atomic_t			gpu_reset_counter;
A
Alex Deucher 已提交
1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

	/* display */
	struct amdgpu_mode_info		mode_info;
	struct work_struct		hotplug_work;
	struct amdgpu_irq_src		crtc_irq;
	struct amdgpu_irq_src		pageflip_irq;
	struct amdgpu_irq_src		hpd_irq;

	/* rings */
	unsigned			fence_context;
	unsigned			num_rings;
	struct amdgpu_ring		*rings[AMDGPU_MAX_RINGS];
	bool				ib_pool_ready;
	struct amdgpu_sa_manager	ring_tmp_bo;

	/* interrupts */
	struct amdgpu_irq		irq;

2003 2004
	/* powerplay */
	struct amd_powerplay		powerplay;
2005
	bool				pp_enabled;
2006
	bool				pp_force_state_enabled;
2007

A
Alex Deucher 已提交
2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
	/* dpm */
	struct amdgpu_pm		pm;
	u32				cg_flags;
	u32				pg_flags;

	/* amdgpu smumgr */
	struct amdgpu_smumgr smu;

	/* gfx */
	struct amdgpu_gfx		gfx;

	/* sdma */
A
Alex Deucher 已提交
2020
	struct amdgpu_sdma		sdma;
A
Alex Deucher 已提交
2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035

	/* uvd */
	struct amdgpu_uvd		uvd;

	/* vce */
	struct amdgpu_vce		vce;

	/* firmwares */
	struct amdgpu_firmware		firmware;

	/* GDS */
	struct amdgpu_gds		gds;

	const struct amdgpu_ip_block_version *ip_blocks;
	int				num_ip_blocks;
2036
	struct amdgpu_ip_block_status	*ip_block_status;
A
Alex Deucher 已提交
2037 2038 2039 2040 2041 2042
	struct mutex	mn_lock;
	DECLARE_HASHTABLE(mn_hash, 7);

	/* tracking pinned memory */
	u64 vram_pin_size;
	u64 gart_pin_size;
2043 2044 2045

	/* amdkfd interface */
	struct kfd_dev          *kfd;
2046

2047
	struct amdgpu_virtualization virtualization;
A
Alex Deucher 已提交
2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133
};

bool amdgpu_device_is_px(struct drm_device *dev);
int amdgpu_device_init(struct amdgpu_device *adev,
		       struct drm_device *ddev,
		       struct pci_dev *pdev,
		       uint32_t flags);
void amdgpu_device_fini(struct amdgpu_device *adev);
int amdgpu_gpu_wait_for_idle(struct amdgpu_device *adev);

uint32_t amdgpu_mm_rreg(struct amdgpu_device *adev, uint32_t reg,
			bool always_indirect);
void amdgpu_mm_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v,
		    bool always_indirect);
u32 amdgpu_io_rreg(struct amdgpu_device *adev, u32 reg);
void amdgpu_io_wreg(struct amdgpu_device *adev, u32 reg, u32 v);

u32 amdgpu_mm_rdoorbell(struct amdgpu_device *adev, u32 index);
void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v);

/*
 * Registers read & write functions.
 */
#define RREG32(reg) amdgpu_mm_rreg(adev, (reg), false)
#define RREG32_IDX(reg) amdgpu_mm_rreg(adev, (reg), true)
#define DREG32(reg) printk(KERN_INFO "REGISTER: " #reg " : 0x%08X\n", amdgpu_mm_rreg(adev, (reg), false))
#define WREG32(reg, v) amdgpu_mm_wreg(adev, (reg), (v), false)
#define WREG32_IDX(reg, v) amdgpu_mm_wreg(adev, (reg), (v), true)
#define REG_SET(FIELD, v) (((v) << FIELD##_SHIFT) & FIELD##_MASK)
#define REG_GET(FIELD, v) (((v) << FIELD##_SHIFT) & FIELD##_MASK)
#define RREG32_PCIE(reg) adev->pcie_rreg(adev, (reg))
#define WREG32_PCIE(reg, v) adev->pcie_wreg(adev, (reg), (v))
#define RREG32_SMC(reg) adev->smc_rreg(adev, (reg))
#define WREG32_SMC(reg, v) adev->smc_wreg(adev, (reg), (v))
#define RREG32_UVD_CTX(reg) adev->uvd_ctx_rreg(adev, (reg))
#define WREG32_UVD_CTX(reg, v) adev->uvd_ctx_wreg(adev, (reg), (v))
#define RREG32_DIDT(reg) adev->didt_rreg(adev, (reg))
#define WREG32_DIDT(reg, v) adev->didt_wreg(adev, (reg), (v))
#define RREG32_AUDIO_ENDPT(block, reg) adev->audio_endpt_rreg(adev, (block), (reg))
#define WREG32_AUDIO_ENDPT(block, reg, v) adev->audio_endpt_wreg(adev, (block), (reg), (v))
#define WREG32_P(reg, val, mask)				\
	do {							\
		uint32_t tmp_ = RREG32(reg);			\
		tmp_ &= (mask);					\
		tmp_ |= ((val) & ~(mask));			\
		WREG32(reg, tmp_);				\
	} while (0)
#define WREG32_AND(reg, and) WREG32_P(reg, 0, and)
#define WREG32_OR(reg, or) WREG32_P(reg, or, ~(or))
#define WREG32_PLL_P(reg, val, mask)				\
	do {							\
		uint32_t tmp_ = RREG32_PLL(reg);		\
		tmp_ &= (mask);					\
		tmp_ |= ((val) & ~(mask));			\
		WREG32_PLL(reg, tmp_);				\
	} while (0)
#define DREG32_SYS(sqf, adev, reg) seq_printf((sqf), #reg " : 0x%08X\n", amdgpu_mm_rreg((adev), (reg), false))
#define RREG32_IO(reg) amdgpu_io_rreg(adev, (reg))
#define WREG32_IO(reg, v) amdgpu_io_wreg(adev, (reg), (v))

#define RDOORBELL32(index) amdgpu_mm_rdoorbell(adev, (index))
#define WDOORBELL32(index, v) amdgpu_mm_wdoorbell(adev, (index), (v))

#define REG_FIELD_SHIFT(reg, field) reg##__##field##__SHIFT
#define REG_FIELD_MASK(reg, field) reg##__##field##_MASK

#define REG_SET_FIELD(orig_val, reg, field, field_val)			\
	(((orig_val) & ~REG_FIELD_MASK(reg, field)) |			\
	 (REG_FIELD_MASK(reg, field) & ((field_val) << REG_FIELD_SHIFT(reg, field))))

#define REG_GET_FIELD(value, reg, field)				\
	(((value) & REG_FIELD_MASK(reg, field)) >> REG_FIELD_SHIFT(reg, field))

/*
 * BIOS helpers.
 */
#define RBIOS8(i) (adev->bios[i])
#define RBIOS16(i) (RBIOS8(i) | (RBIOS8((i)+1) << 8))
#define RBIOS32(i) ((RBIOS16(i)) | (RBIOS16((i)+2) << 16))

/*
 * RING helpers.
 */
static inline void amdgpu_ring_write(struct amdgpu_ring *ring, uint32_t v)
{
	if (ring->count_dw <= 0)
2134
		DRM_ERROR("amdgpu: writing more dwords to the ring than expected!\n");
A
Alex Deucher 已提交
2135 2136 2137 2138 2139
	ring->ring[ring->wptr++] = v;
	ring->wptr &= ring->ptr_mask;
	ring->count_dw--;
}

A
Alex Deucher 已提交
2140 2141
static inline struct amdgpu_sdma_instance *
amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
2142 2143 2144 2145
{
	struct amdgpu_device *adev = ring->adev;
	int i;

A
Alex Deucher 已提交
2146 2147
	for (i = 0; i < adev->sdma.num_instances; i++)
		if (&adev->sdma.instance[i].ring == ring)
2148 2149 2150
			break;

	if (i < AMDGPU_MAX_SDMA_INSTANCES)
A
Alex Deucher 已提交
2151
		return &adev->sdma.instance[i];
2152 2153 2154 2155
	else
		return NULL;
}

A
Alex Deucher 已提交
2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166
/*
 * ASICs macro.
 */
#define amdgpu_asic_set_vga_state(adev, state) (adev)->asic_funcs->set_vga_state((adev), (state))
#define amdgpu_asic_reset(adev) (adev)->asic_funcs->reset((adev))
#define amdgpu_asic_wait_for_mc_idle(adev) (adev)->asic_funcs->wait_for_mc_idle((adev))
#define amdgpu_asic_get_xclk(adev) (adev)->asic_funcs->get_xclk((adev))
#define amdgpu_asic_set_uvd_clocks(adev, v, d) (adev)->asic_funcs->set_uvd_clocks((adev), (v), (d))
#define amdgpu_asic_set_vce_clocks(adev, ev, ec) (adev)->asic_funcs->set_vce_clocks((adev), (ev), (ec))
#define amdgpu_asic_get_gpu_clock_counter(adev) (adev)->asic_funcs->get_gpu_clock_counter((adev))
#define amdgpu_asic_read_disabled_bios(adev) (adev)->asic_funcs->read_disabled_bios((adev))
2167
#define amdgpu_asic_read_bios_from_rom(adev, b, l) (adev)->asic_funcs->read_bios_from_rom((adev), (b), (l))
A
Alex Deucher 已提交
2168 2169 2170 2171 2172
#define amdgpu_asic_read_register(adev, se, sh, offset, v)((adev)->asic_funcs->read_register((adev), (se), (sh), (offset), (v)))
#define amdgpu_asic_get_cu_info(adev, info) (adev)->asic_funcs->get_cu_info((adev), (info))
#define amdgpu_gart_flush_gpu_tlb(adev, vmid) (adev)->gart.gart_funcs->flush_gpu_tlb((adev), (vmid))
#define amdgpu_gart_set_pte_pde(adev, pt, idx, addr, flags) (adev)->gart.gart_funcs->set_pte_pde((adev), (pt), (idx), (addr), (flags))
#define amdgpu_vm_copy_pte(adev, ib, pe, src, count) ((adev)->vm_manager.vm_pte_funcs->copy_pte((ib), (pe), (src), (count)))
2173
#define amdgpu_vm_write_pte(adev, ib, pa, pe, addr, count, incr, flags) ((adev)->vm_manager.vm_pte_funcs->write_pte((ib), (pa), (pe), (addr), (count), (incr), (flags)))
A
Alex Deucher 已提交
2174 2175 2176 2177 2178 2179 2180 2181
#define amdgpu_vm_set_pte_pde(adev, ib, pe, addr, count, incr, flags) ((adev)->vm_manager.vm_pte_funcs->set_pte_pde((ib), (pe), (addr), (count), (incr), (flags)))
#define amdgpu_ring_parse_cs(r, p, ib) ((r)->funcs->parse_cs((p), (ib)))
#define amdgpu_ring_test_ring(r) (r)->funcs->test_ring((r))
#define amdgpu_ring_test_ib(r) (r)->funcs->test_ib((r))
#define amdgpu_ring_get_rptr(r) (r)->funcs->get_rptr((r))
#define amdgpu_ring_get_wptr(r) (r)->funcs->get_wptr((r))
#define amdgpu_ring_set_wptr(r) (r)->funcs->set_wptr((r))
#define amdgpu_ring_emit_ib(r, ib) (r)->funcs->emit_ib((r), (ib))
2182
#define amdgpu_ring_emit_pipeline_sync(r) (r)->funcs->emit_pipeline_sync((r))
A
Alex Deucher 已提交
2183
#define amdgpu_ring_emit_vm_flush(r, vmid, addr) (r)->funcs->emit_vm_flush((r), (vmid), (addr))
2184
#define amdgpu_ring_emit_fence(r, addr, seq, flags) (r)->funcs->emit_fence((r), (addr), (seq), (flags))
A
Alex Deucher 已提交
2185
#define amdgpu_ring_emit_gds_switch(r, v, db, ds, wb, ws, ab, as) (r)->funcs->emit_gds_switch((r), (v), (db), (ds), (wb), (ws), (ab), (as))
2186
#define amdgpu_ring_emit_hdp_flush(r) (r)->funcs->emit_hdp_flush((r))
2187
#define amdgpu_ring_emit_hdp_invalidate(r) (r)->funcs->emit_hdp_invalidate((r))
2188
#define amdgpu_ring_pad_ib(r, ib) ((r)->funcs->pad_ib((r), (ib)))
M
Monk Liu 已提交
2189 2190
#define amdgpu_ring_init_cond_exec(r) (r)->funcs->init_cond_exec((r))
#define amdgpu_ring_patch_cond_exec(r,o) (r)->funcs->patch_cond_exec((r),(o))
A
Alex Deucher 已提交
2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209
#define amdgpu_ih_get_wptr(adev) (adev)->irq.ih_funcs->get_wptr((adev))
#define amdgpu_ih_decode_iv(adev, iv) (adev)->irq.ih_funcs->decode_iv((adev), (iv))
#define amdgpu_ih_set_rptr(adev) (adev)->irq.ih_funcs->set_rptr((adev))
#define amdgpu_display_set_vga_render_state(adev, r) (adev)->mode_info.funcs->set_vga_render_state((adev), (r))
#define amdgpu_display_vblank_get_counter(adev, crtc) (adev)->mode_info.funcs->vblank_get_counter((adev), (crtc))
#define amdgpu_display_vblank_wait(adev, crtc) (adev)->mode_info.funcs->vblank_wait((adev), (crtc))
#define amdgpu_display_is_display_hung(adev) (adev)->mode_info.funcs->is_display_hung((adev))
#define amdgpu_display_backlight_set_level(adev, e, l) (adev)->mode_info.funcs->backlight_set_level((e), (l))
#define amdgpu_display_backlight_get_level(adev, e) (adev)->mode_info.funcs->backlight_get_level((e))
#define amdgpu_display_hpd_sense(adev, h) (adev)->mode_info.funcs->hpd_sense((adev), (h))
#define amdgpu_display_hpd_set_polarity(adev, h) (adev)->mode_info.funcs->hpd_set_polarity((adev), (h))
#define amdgpu_display_hpd_get_gpio_reg(adev) (adev)->mode_info.funcs->hpd_get_gpio_reg((adev))
#define amdgpu_display_bandwidth_update(adev) (adev)->mode_info.funcs->bandwidth_update((adev))
#define amdgpu_display_page_flip(adev, crtc, base) (adev)->mode_info.funcs->page_flip((adev), (crtc), (base))
#define amdgpu_display_page_flip_get_scanoutpos(adev, crtc, vbl, pos) (adev)->mode_info.funcs->page_flip_get_scanoutpos((adev), (crtc), (vbl), (pos))
#define amdgpu_display_add_encoder(adev, e, s, c) (adev)->mode_info.funcs->add_encoder((adev), (e), (s), (c))
#define amdgpu_display_add_connector(adev, ci, sd, ct, ib, coi, h, r) (adev)->mode_info.funcs->add_connector((adev), (ci), (sd), (ct), (ib), (coi), (h), (r))
#define amdgpu_display_stop_mc_access(adev, s) (adev)->mode_info.funcs->stop_mc_access((adev), (s))
#define amdgpu_display_resume_mc_access(adev, s) (adev)->mode_info.funcs->resume_mc_access((adev), (s))
2210
#define amdgpu_emit_copy_buffer(adev, ib, s, d, b) (adev)->mman.buffer_funcs->emit_copy_buffer((ib),  (s), (d), (b))
2211
#define amdgpu_emit_fill_buffer(adev, ib, s, d, b) (adev)->mman.buffer_funcs->emit_fill_buffer((ib), (s), (d), (b))
A
Alex Deucher 已提交
2212 2213 2214 2215 2216 2217 2218
#define amdgpu_dpm_pre_set_power_state(adev) (adev)->pm.funcs->pre_set_power_state((adev))
#define amdgpu_dpm_set_power_state(adev) (adev)->pm.funcs->set_power_state((adev))
#define amdgpu_dpm_post_set_power_state(adev) (adev)->pm.funcs->post_set_power_state((adev))
#define amdgpu_dpm_display_configuration_changed(adev) (adev)->pm.funcs->display_configuration_changed((adev))
#define amdgpu_dpm_print_power_state(adev, ps) (adev)->pm.funcs->print_power_state((adev), (ps))
#define amdgpu_dpm_vblank_too_short(adev) (adev)->pm.funcs->vblank_too_short((adev))
#define amdgpu_dpm_enable_bapm(adev, e) (adev)->pm.funcs->enable_bapm((adev), (e))
2219 2220

#define amdgpu_dpm_get_temperature(adev) \
2221
	((adev)->pp_enabled ?						\
2222
	      (adev)->powerplay.pp_funcs->get_temperature((adev)->powerplay.pp_handle) : \
2223
	      (adev)->pm.funcs->get_temperature((adev)))
2224 2225

#define amdgpu_dpm_set_fan_control_mode(adev, m) \
2226
	((adev)->pp_enabled ?						\
2227
	      (adev)->powerplay.pp_funcs->set_fan_control_mode((adev)->powerplay.pp_handle, (m)) : \
2228
	      (adev)->pm.funcs->set_fan_control_mode((adev), (m)))
2229 2230

#define amdgpu_dpm_get_fan_control_mode(adev) \
2231
	((adev)->pp_enabled ?						\
2232
	      (adev)->powerplay.pp_funcs->get_fan_control_mode((adev)->powerplay.pp_handle) : \
2233
	      (adev)->pm.funcs->get_fan_control_mode((adev)))
2234 2235

#define amdgpu_dpm_set_fan_speed_percent(adev, s) \
2236
	((adev)->pp_enabled ?						\
2237
	      (adev)->powerplay.pp_funcs->set_fan_speed_percent((adev)->powerplay.pp_handle, (s)) : \
2238
	      (adev)->pm.funcs->set_fan_speed_percent((adev), (s)))
2239 2240

#define amdgpu_dpm_get_fan_speed_percent(adev, s) \
2241
	((adev)->pp_enabled ?						\
2242
	      (adev)->powerplay.pp_funcs->get_fan_speed_percent((adev)->powerplay.pp_handle, (s)) : \
2243
	      (adev)->pm.funcs->get_fan_speed_percent((adev), (s)))
A
Alex Deucher 已提交
2244

2245
#define amdgpu_dpm_get_sclk(adev, l) \
2246
	((adev)->pp_enabled ?						\
2247
	      (adev)->powerplay.pp_funcs->get_sclk((adev)->powerplay.pp_handle, (l)) : \
2248
		(adev)->pm.funcs->get_sclk((adev), (l)))
2249 2250

#define amdgpu_dpm_get_mclk(adev, l)  \
2251
	((adev)->pp_enabled ?						\
2252
	      (adev)->powerplay.pp_funcs->get_mclk((adev)->powerplay.pp_handle, (l)) : \
2253
	      (adev)->pm.funcs->get_mclk((adev), (l)))
2254 2255 2256


#define amdgpu_dpm_force_performance_level(adev, l) \
2257
	((adev)->pp_enabled ?						\
2258
	      (adev)->powerplay.pp_funcs->force_performance_level((adev)->powerplay.pp_handle, (l)) : \
2259
	      (adev)->pm.funcs->force_performance_level((adev), (l)))
2260 2261

#define amdgpu_dpm_powergate_uvd(adev, g) \
2262
	((adev)->pp_enabled ?						\
2263
	      (adev)->powerplay.pp_funcs->powergate_uvd((adev)->powerplay.pp_handle, (g)) : \
2264
	      (adev)->pm.funcs->powergate_uvd((adev), (g)))
2265 2266

#define amdgpu_dpm_powergate_vce(adev, g) \
2267
	((adev)->pp_enabled ?						\
2268
	      (adev)->powerplay.pp_funcs->powergate_vce((adev)->powerplay.pp_handle, (g)) : \
2269
	      (adev)->pm.funcs->powergate_vce((adev), (g)))
2270 2271

#define amdgpu_dpm_debugfs_print_current_performance_level(adev, m) \
2272
	((adev)->pp_enabled ?						\
2273
	      (adev)->powerplay.pp_funcs->print_current_performance_level((adev)->powerplay.pp_handle, (m)) : \
2274
	      (adev)->pm.funcs->debugfs_print_current_performance_level((adev), (m)))
2275 2276

#define amdgpu_dpm_get_current_power_state(adev) \
2277
	(adev)->powerplay.pp_funcs->get_current_power_state((adev)->powerplay.pp_handle)
2278 2279

#define amdgpu_dpm_get_performance_level(adev) \
2280
	(adev)->powerplay.pp_funcs->get_performance_level((adev)->powerplay.pp_handle)
2281

2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296
#define amdgpu_dpm_get_pp_num_states(adev, data) \
	(adev)->powerplay.pp_funcs->get_pp_num_states((adev)->powerplay.pp_handle, data)

#define amdgpu_dpm_get_pp_table(adev, table) \
	(adev)->powerplay.pp_funcs->get_pp_table((adev)->powerplay.pp_handle, table)

#define amdgpu_dpm_set_pp_table(adev, buf, size) \
	(adev)->powerplay.pp_funcs->set_pp_table((adev)->powerplay.pp_handle, buf, size)

#define amdgpu_dpm_print_clock_levels(adev, type, buf) \
	(adev)->powerplay.pp_funcs->print_clock_levels((adev)->powerplay.pp_handle, type, buf)

#define amdgpu_dpm_force_clock_level(adev, type, level) \
		(adev)->powerplay.pp_funcs->force_clock_level((adev)->powerplay.pp_handle, type, level)

2297
#define amdgpu_dpm_dispatch_task(adev, event_id, input, output)		\
2298
	(adev)->powerplay.pp_funcs->dispatch_tasks((adev)->powerplay.pp_handle, (event_id), (input), (output))
A
Alex Deucher 已提交
2299 2300 2301 2302 2303 2304 2305 2306

#define amdgpu_gds_switch(adev, r, v, d, w, a) (adev)->gds.funcs->patch_gds_switch((r), (v), (d), (w), (a))

/* Common functions */
int amdgpu_gpu_reset(struct amdgpu_device *adev);
void amdgpu_pci_config_reset(struct amdgpu_device *adev);
bool amdgpu_card_posted(struct amdgpu_device *adev);
void amdgpu_update_display_priority(struct amdgpu_device *adev);
C
Chunming Zhou 已提交
2307

A
Alex Deucher 已提交
2308 2309 2310 2311 2312 2313
int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data);
int amdgpu_cs_get_ring(struct amdgpu_device *adev, u32 ip_type,
		       u32 ip_instance, u32 ring,
		       struct amdgpu_ring **out_ring);
void amdgpu_ttm_placement_from_domain(struct amdgpu_bo *rbo, u32 domain);
bool amdgpu_ttm_bo_is_amdgpu_bo(struct ttm_buffer_object *bo);
2314
int amdgpu_ttm_tt_get_user_pages(struct ttm_tt *ttm, struct page **pages);
A
Alex Deucher 已提交
2315 2316 2317
int amdgpu_ttm_tt_set_userptr(struct ttm_tt *ttm, uint64_t addr,
				     uint32_t flags);
bool amdgpu_ttm_tt_has_userptr(struct ttm_tt *ttm);
2318
struct mm_struct *amdgpu_ttm_tt_get_usermm(struct ttm_tt *ttm);
2319 2320
bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start,
				  unsigned long end);
2321 2322
bool amdgpu_ttm_tt_userptr_invalidated(struct ttm_tt *ttm,
				       int *last_invalidated);
A
Alex Deucher 已提交
2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346
bool amdgpu_ttm_tt_is_readonly(struct ttm_tt *ttm);
uint32_t amdgpu_ttm_tt_pte_flags(struct amdgpu_device *adev, struct ttm_tt *ttm,
				 struct ttm_mem_reg *mem);
void amdgpu_vram_location(struct amdgpu_device *adev, struct amdgpu_mc *mc, u64 base);
void amdgpu_gtt_location(struct amdgpu_device *adev, struct amdgpu_mc *mc);
void amdgpu_ttm_set_active_vram_size(struct amdgpu_device *adev, u64 size);
void amdgpu_program_register_sequence(struct amdgpu_device *adev,
					     const u32 *registers,
					     const u32 array_size);

bool amdgpu_device_is_px(struct drm_device *dev);
/* atpx handler */
#if defined(CONFIG_VGA_SWITCHEROO)
void amdgpu_register_atpx_handler(void);
void amdgpu_unregister_atpx_handler(void);
#else
static inline void amdgpu_register_atpx_handler(void) {}
static inline void amdgpu_unregister_atpx_handler(void) {}
#endif

/*
 * KMS
 */
extern const struct drm_ioctl_desc amdgpu_ioctls_kms[];
2347
extern const int amdgpu_max_kms_ioctl;
A
Alex Deucher 已提交
2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358

int amdgpu_driver_load_kms(struct drm_device *dev, unsigned long flags);
int amdgpu_driver_unload_kms(struct drm_device *dev);
void amdgpu_driver_lastclose_kms(struct drm_device *dev);
int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv);
void amdgpu_driver_postclose_kms(struct drm_device *dev,
				 struct drm_file *file_priv);
void amdgpu_driver_preclose_kms(struct drm_device *dev,
				struct drm_file *file_priv);
int amdgpu_suspend_kms(struct drm_device *dev, bool suspend, bool fbcon);
int amdgpu_resume_kms(struct drm_device *dev, bool resume, bool fbcon);
2359 2360 2361 2362
u32 amdgpu_get_vblank_counter_kms(struct drm_device *dev, unsigned int pipe);
int amdgpu_enable_vblank_kms(struct drm_device *dev, unsigned int pipe);
void amdgpu_disable_vblank_kms(struct drm_device *dev, unsigned int pipe);
int amdgpu_get_vblank_timestamp_kms(struct drm_device *dev, unsigned int pipe,
A
Alex Deucher 已提交
2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406
				    int *max_error,
				    struct timeval *vblank_time,
				    unsigned flags);
long amdgpu_kms_compat_ioctl(struct file *filp, unsigned int cmd,
			     unsigned long arg);

/*
 * functions used by amdgpu_encoder.c
 */
struct amdgpu_afmt_acr {
	u32 clock;

	int n_32khz;
	int cts_32khz;

	int n_44_1khz;
	int cts_44_1khz;

	int n_48khz;
	int cts_48khz;

};

struct amdgpu_afmt_acr amdgpu_afmt_acr(uint32_t clock);

/* amdgpu_acpi.c */
#if defined(CONFIG_ACPI)
int amdgpu_acpi_init(struct amdgpu_device *adev);
void amdgpu_acpi_fini(struct amdgpu_device *adev);
bool amdgpu_acpi_is_pcie_performance_request_supported(struct amdgpu_device *adev);
int amdgpu_acpi_pcie_performance_request(struct amdgpu_device *adev,
						u8 perf_req, bool advertise);
int amdgpu_acpi_pcie_notify_device_ready(struct amdgpu_device *adev);
#else
static inline int amdgpu_acpi_init(struct amdgpu_device *adev) { return 0; }
static inline void amdgpu_acpi_fini(struct amdgpu_device *adev) { }
#endif

struct amdgpu_bo_va_mapping *
amdgpu_cs_find_mapping(struct amdgpu_cs_parser *parser,
		       uint64_t addr, struct amdgpu_bo **bo);

#include "amdgpu_object.h"
#endif