radeon.h 60.8 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 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 __RADEON_H__
#define __RADEON_H__

/* TODO: Here are things that needs to be done :
 *	- surface allocator & initializer : (bit like scratch reg) should
 *	  initialize HDP_ stuff on RS600, R600, R700 hw, well anythings
 *	  related to surface
 *	- WB : write back stuff (do it bit like scratch reg things)
 *	- Vblank : look at Jesse's rework and what we should do
 *	- r600/r700: gart & cp
 *	- cs : clean cs ioctl use bitmap & things like that.
 *	- power management stuff
 *	- Barrier in gart code
 *	- Unmappabled vram ?
 *	- TESTING, TESTING, TESTING
 */

45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62
/* Initialization path:
 *  We expect that acceleration initialization might fail for various
 *  reasons even thought we work hard to make it works on most
 *  configurations. In order to still have a working userspace in such
 *  situation the init path must succeed up to the memory controller
 *  initialization point. Failure before this point are considered as
 *  fatal error. Here is the init callchain :
 *      radeon_device_init  perform common structure, mutex initialization
 *      asic_init           setup the GPU memory layout and perform all
 *                          one time initialization (failure in this
 *                          function are considered fatal)
 *      asic_startup        setup the GPU acceleration, in order to
 *                          follow guideline the first thing this
 *                          function should do is setting the GPU
 *                          memory controller (only MC setup failure
 *                          are considered as fatal)
 */

A
Arun Sharma 已提交
63
#include <linux/atomic.h>
64 65 66 67
#include <linux/wait.h>
#include <linux/list.h>
#include <linux/kref.h>

68 69 70 71
#include <ttm/ttm_bo_api.h>
#include <ttm/ttm_bo_driver.h>
#include <ttm/ttm_placement.h>
#include <ttm/ttm_module.h>
72
#include <ttm/ttm_execbuf_util.h>
73

74
#include "radeon_family.h"
75 76 77 78 79 80 81 82 83 84 85 86 87 88
#include "radeon_mode.h"
#include "radeon_reg.h"

/*
 * Modules parameters.
 */
extern int radeon_no_wb;
extern int radeon_modeset;
extern int radeon_dynclks;
extern int radeon_r4xx_atom;
extern int radeon_agpmode;
extern int radeon_vram_limit;
extern int radeon_gart_size;
extern int radeon_benchmarking;
89
extern int radeon_testing;
90
extern int radeon_connector_table;
91
extern int radeon_tv;
92
extern int radeon_audio;
93
extern int radeon_disp_priority;
94
extern int radeon_hw_i2c;
95
extern int radeon_pcie_gen2;
96
extern int radeon_msi;
97
extern int radeon_lockup_timeout;
98 99 100 101 102

/*
 * Copy from radeon_drv.h so we don't have to include both and have conflicting
 * symbol;
 */
103 104
#define RADEON_MAX_USEC_TIMEOUT			100000	/* 100 ms */
#define RADEON_FENCE_JIFFIES_TIMEOUT		(HZ / 2)
105
/* RADEON_IB_POOL_SIZE must be a power of 2 */
106 107 108 109
#define RADEON_IB_POOL_SIZE			16
#define RADEON_DEBUGFS_MAX_COMPONENTS		32
#define RADEONFB_CONN_LIMIT			4
#define RADEON_BIOS_NUM_SCRATCH			8
110

111
/* max number of rings */
112
#define RADEON_NUM_RINGS			5
113 114 115

/* fence seq are set to this number when signaled */
#define RADEON_FENCE_SIGNALED_SEQ		0LL
116 117 118

/* internal ring indices */
/* r1xx+ has gfx CP ring */
119
#define RADEON_RING_TYPE_GFX_INDEX		0
120 121

/* cayman has 2 compute CP rings */
122 123
#define CAYMAN_RING_TYPE_CP1_INDEX		1
#define CAYMAN_RING_TYPE_CP2_INDEX		2
124

125 126
/* R600+ has an async dma ring */
#define R600_RING_TYPE_DMA_INDEX		3
127 128
/* cayman add a second async dma ring */
#define CAYMAN_RING_TYPE_DMA1_INDEX		4
129

130
/* hardcode those limit for now */
131
#define RADEON_VA_IB_OFFSET			(1 << 20)
132 133
#define RADEON_VA_RESERVED_SIZE			(8 << 20)
#define RADEON_IB_VM_MAX_SIZE			(64 << 10)
134

A
Alex Deucher 已提交
135 136 137 138 139
/* reset flags */
#define RADEON_RESET_GFX			(1 << 0)
#define RADEON_RESET_COMPUTE			(1 << 1)
#define RADEON_RESET_DMA			(1 << 2)

140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158
/*
 * Errata workarounds.
 */
enum radeon_pll_errata {
	CHIP_ERRATA_R300_CG             = 0x00000001,
	CHIP_ERRATA_PLL_DUMMYREADS      = 0x00000002,
	CHIP_ERRATA_PLL_DELAY           = 0x00000004
};


struct radeon_device;


/*
 * BIOS.
 */
bool radeon_get_bios(struct radeon_device *rdev);

/*
159
 * Dummy page
160
 */
161 162 163 164 165 166 167
struct radeon_dummy_page {
	struct page	*page;
	dma_addr_t	addr;
};
int radeon_dummy_page_init(struct radeon_device *rdev);
void radeon_dummy_page_fini(struct radeon_device *rdev);

168

169 170 171
/*
 * Clocks
 */
172 173 174
struct radeon_clock {
	struct radeon_pll p1pll;
	struct radeon_pll p2pll;
175
	struct radeon_pll dcpll;
176 177 178 179 180
	struct radeon_pll spll;
	struct radeon_pll mpll;
	/* 10 Khz units */
	uint32_t default_mclk;
	uint32_t default_sclk;
181 182
	uint32_t default_dispclk;
	uint32_t dp_extclk;
183
	uint32_t max_pixel_clock;
184 185
};

186 187 188 189
/*
 * Power management
 */
int radeon_pm_init(struct radeon_device *rdev);
190
void radeon_pm_fini(struct radeon_device *rdev);
191
void radeon_pm_compute_clocks(struct radeon_device *rdev);
192 193
void radeon_pm_suspend(struct radeon_device *rdev);
void radeon_pm_resume(struct radeon_device *rdev);
194 195
void radeon_combios_get_power_modes(struct radeon_device *rdev);
void radeon_atombios_get_power_modes(struct radeon_device *rdev);
196
void radeon_atom_set_voltage(struct radeon_device *rdev, u16 voltage_level, u8 voltage_type);
197
void rs690_pm_info(struct radeon_device *rdev);
198 199 200 201
extern int rv6xx_get_temp(struct radeon_device *rdev);
extern int rv770_get_temp(struct radeon_device *rdev);
extern int evergreen_get_temp(struct radeon_device *rdev);
extern int sumo_get_temp(struct radeon_device *rdev);
202
extern int si_get_temp(struct radeon_device *rdev);
203 204 205
extern void evergreen_tiling_fields(unsigned tiling_flags, unsigned *bankw,
				    unsigned *bankh, unsigned *mtaspect,
				    unsigned *tile_split);
206

207 208 209 210 211
/*
 * Fences.
 */
struct radeon_fence_driver {
	uint32_t			scratch_reg;
212 213
	uint64_t			gpu_addr;
	volatile uint32_t		*cpu_addr;
214 215
	/* sync_seq is protected by ring emission lock */
	uint64_t			sync_seq[RADEON_NUM_RINGS];
216
	atomic64_t			last_seq;
217
	unsigned long			last_activity;
218
	bool				initialized;
219 220 221 222 223 224
};

struct radeon_fence {
	struct radeon_device		*rdev;
	struct kref			kref;
	/* protected by radeon_fence.lock */
225
	uint64_t			seq;
226
	/* RB, DMA, etc. */
227
	unsigned			ring;
228 229
};

230 231
int radeon_fence_driver_start_ring(struct radeon_device *rdev, int ring);
int radeon_fence_driver_init(struct radeon_device *rdev);
232
void radeon_fence_driver_fini(struct radeon_device *rdev);
233
void radeon_fence_driver_force_completion(struct radeon_device *rdev);
234
int radeon_fence_emit(struct radeon_device *rdev, struct radeon_fence **fence, int ring);
235
void radeon_fence_process(struct radeon_device *rdev, int ring);
236 237
bool radeon_fence_signaled(struct radeon_fence *fence);
int radeon_fence_wait(struct radeon_fence *fence, bool interruptible);
238
int radeon_fence_wait_next_locked(struct radeon_device *rdev, int ring);
239
int radeon_fence_wait_empty_locked(struct radeon_device *rdev, int ring);
240 241 242
int radeon_fence_wait_any(struct radeon_device *rdev,
			  struct radeon_fence **fences,
			  bool intr);
243 244
struct radeon_fence *radeon_fence_ref(struct radeon_fence *fence);
void radeon_fence_unref(struct radeon_fence **fence);
245
unsigned radeon_fence_count_emitted(struct radeon_device *rdev, int ring);
246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266
bool radeon_fence_need_sync(struct radeon_fence *fence, int ring);
void radeon_fence_note_sync(struct radeon_fence *fence, int ring);
static inline struct radeon_fence *radeon_fence_later(struct radeon_fence *a,
						      struct radeon_fence *b)
{
	if (!a) {
		return b;
	}

	if (!b) {
		return a;
	}

	BUG_ON(a->ring != b->ring);

	if (a->seq > b->seq) {
		return a;
	} else {
		return b;
	}
}
267

268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283
static inline bool radeon_fence_is_earlier(struct radeon_fence *a,
					   struct radeon_fence *b)
{
	if (!a) {
		return false;
	}

	if (!b) {
		return true;
	}

	BUG_ON(a->ring != b->ring);

	return a->seq < b->seq;
}

284 285 286 287
/*
 * Tiling registers
 */
struct radeon_surface_reg {
288
	struct radeon_bo *bo;
289 290 291
};

#define RADEON_GEM_MAX_SURFACES 8
292 293

/*
294
 * TTM.
295
 */
296 297
struct radeon_mman {
	struct ttm_bo_global_ref        bo_global_ref;
298
	struct drm_global_reference	mem_global_ref;
299
	struct ttm_bo_device		bdev;
300 301
	bool				mem_global_referenced;
	bool				initialized;
302 303
};

304 305
/* bo virtual address in a specific vm */
struct radeon_bo_va {
306
	/* protected by bo being reserved */
307 308 309 310 311
	struct list_head		bo_list;
	uint64_t			soffset;
	uint64_t			eoffset;
	uint32_t			flags;
	bool				valid;
312 313 314 315 316 317 318 319
	unsigned			ref_count;

	/* protected by vm mutex */
	struct list_head		vm_list;

	/* constant after initialization */
	struct radeon_vm		*vm;
	struct radeon_bo		*bo;
320 321
};

322 323 324 325
struct radeon_bo {
	/* Protected by gem.mutex */
	struct list_head		list;
	/* Protected by tbo.reserved */
326 327
	u32				placements[3];
	struct ttm_placement		placement;
328 329 330 331 332 333 334
	struct ttm_buffer_object	tbo;
	struct ttm_bo_kmap_obj		kmap;
	unsigned			pin_count;
	void				*kptr;
	u32				tiling_flags;
	u32				pitch;
	int				surface_reg;
335 336 337 338
	/* list of all virtual address to which this bo
	 * is associated to
	 */
	struct list_head		va;
339 340
	/* Constant after initialization */
	struct radeon_device		*rdev;
341
	struct drm_gem_object		gem_base;
342 343 344

	struct ttm_bo_kmap_obj dma_buf_vmap;
	int vmapping_count;
345
};
346
#define gem_to_radeon_bo(gobj) container_of((gobj), struct radeon_bo, gem_base)
347

348
struct radeon_bo_list {
349
	struct ttm_validate_buffer tv;
350
	struct radeon_bo	*bo;
351 352 353
	uint64_t		gpu_offset;
	unsigned		rdomain;
	unsigned		wdomain;
354
	u32			tiling_flags;
355 356
};

357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380
/* 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).
 */
struct radeon_sa_manager {
381
	wait_queue_head_t	wq;
382
	struct radeon_bo	*bo;
383 384 385
	struct list_head	*hole;
	struct list_head	flist[RADEON_NUM_RINGS];
	struct list_head	olist;
386 387 388 389 390 391 392 393 394 395
	unsigned		size;
	uint64_t		gpu_addr;
	void			*cpu_ptr;
	uint32_t		domain;
};

struct radeon_sa_bo;

/* sub-allocation buffer */
struct radeon_sa_bo {
396 397
	struct list_head		olist;
	struct list_head		flist;
398
	struct radeon_sa_manager	*manager;
399 400
	unsigned			soffset;
	unsigned			eoffset;
401
	struct radeon_fence		*fence;
402 403
};

404 405 406 407
/*
 * GEM objects.
 */
struct radeon_gem {
408
	struct mutex		mutex;
409 410 411 412 413 414
	struct list_head	objects;
};

int radeon_gem_init(struct radeon_device *rdev);
void radeon_gem_fini(struct radeon_device *rdev);
int radeon_gem_object_create(struct radeon_device *rdev, int size,
415 416 417
				int alignment, int initial_domain,
				bool discardable, bool kernel,
				struct drm_gem_object **obj);
418

419 420 421 422 423 424 425 426 427
int radeon_mode_dumb_create(struct drm_file *file_priv,
			    struct drm_device *dev,
			    struct drm_mode_create_dumb *args);
int radeon_mode_dumb_mmap(struct drm_file *filp,
			  struct drm_device *dev,
			  uint32_t handle, uint64_t *offset_p);
int radeon_mode_dumb_destroy(struct drm_file *file_priv,
			     struct drm_device *dev,
			     uint32_t handle);
428

429 430 431 432 433
/*
 * Semaphores.
 */
/* everything here is constant */
struct radeon_semaphore {
434 435
	struct radeon_sa_bo		*sa_bo;
	signed				waiters;
436 437 438 439 440 441 442 443 444
	uint64_t			gpu_addr;
};

int radeon_semaphore_create(struct radeon_device *rdev,
			    struct radeon_semaphore **semaphore);
void radeon_semaphore_emit_signal(struct radeon_device *rdev, int ring,
				  struct radeon_semaphore *semaphore);
void radeon_semaphore_emit_wait(struct radeon_device *rdev, int ring,
				struct radeon_semaphore *semaphore);
445 446
int radeon_semaphore_sync_rings(struct radeon_device *rdev,
				struct radeon_semaphore *semaphore,
447
				int signaler, int waiter);
448
void radeon_semaphore_free(struct radeon_device *rdev,
449
			   struct radeon_semaphore **semaphore,
450
			   struct radeon_fence *fence);
451

452 453 454 455 456
/*
 * GART structures, functions & helpers
 */
struct radeon_mc;

457
#define RADEON_GPU_PAGE_SIZE 4096
458
#define RADEON_GPU_PAGE_MASK (RADEON_GPU_PAGE_SIZE - 1)
459
#define RADEON_GPU_PAGE_SHIFT 12
460
#define RADEON_GPU_PAGE_ALIGN(a) (((a) + RADEON_GPU_PAGE_MASK) & ~RADEON_GPU_PAGE_MASK)
461

462 463
struct radeon_gart {
	dma_addr_t			table_addr;
464 465
	struct radeon_bo		*robj;
	void				*ptr;
466 467 468 469 470 471 472 473 474 475 476 477
	unsigned			num_gpu_pages;
	unsigned			num_cpu_pages;
	unsigned			table_size;
	struct page			**pages;
	dma_addr_t			*pages_addr;
	bool				ready;
};

int radeon_gart_table_ram_alloc(struct radeon_device *rdev);
void radeon_gart_table_ram_free(struct radeon_device *rdev);
int radeon_gart_table_vram_alloc(struct radeon_device *rdev);
void radeon_gart_table_vram_free(struct radeon_device *rdev);
478 479
int radeon_gart_table_vram_pin(struct radeon_device *rdev);
void radeon_gart_table_vram_unpin(struct radeon_device *rdev);
480 481 482 483 484
int radeon_gart_init(struct radeon_device *rdev);
void radeon_gart_fini(struct radeon_device *rdev);
void radeon_gart_unbind(struct radeon_device *rdev, unsigned offset,
			int pages);
int radeon_gart_bind(struct radeon_device *rdev, unsigned offset,
485 486
		     int pages, struct page **pagelist,
		     dma_addr_t *dma_addr);
487
void radeon_gart_restore(struct radeon_device *rdev);
488 489 490 491 492 493 494 495 496


/*
 * GPU MC structures, functions & helpers
 */
struct radeon_mc {
	resource_size_t		aper_size;
	resource_size_t		aper_base;
	resource_size_t		agp_base;
497 498
	/* for some chips with <= 32MB we need to lie
	 * about vram size near mc fb location */
499
	u64			mc_vram_size;
500
	u64			visible_vram_size;
501 502 503 504 505
	u64			gtt_size;
	u64			gtt_start;
	u64			gtt_end;
	u64			vram_start;
	u64			vram_end;
506
	unsigned		vram_width;
507
	u64			real_vram_size;
508 509
	int			vram_mtrr;
	bool			vram_is_ddr;
510
	bool			igp_sideport_enabled;
511
	u64                     gtt_base_align;
512 513
};

514 515
bool radeon_combios_sideport_present(struct radeon_device *rdev);
bool radeon_atombios_sideport_present(struct radeon_device *rdev);
516 517 518 519 520 521

/*
 * GPU scratch registers structures, functions & helpers
 */
struct radeon_scratch {
	unsigned		num_reg;
522
	uint32_t                reg_base;
523 524 525 526 527 528 529 530 531 532 533
	bool			free[32];
	uint32_t		reg[32];
};

int radeon_scratch_get(struct radeon_device *rdev, uint32_t *reg);
void radeon_scratch_free(struct radeon_device *rdev, uint32_t reg);


/*
 * IRQS.
 */
534 535 536 537 538 539 540 541 542 543 544 545 546

struct radeon_unpin_work {
	struct work_struct work;
	struct radeon_device *rdev;
	int crtc_id;
	struct radeon_fence *fence;
	struct drm_pending_vblank_event *event;
	struct radeon_bo *old_rbo;
	u64 new_crtc_base;
};

struct r500_irq_stat_regs {
	u32 disp_int;
547
	u32 hdmi0_status;
548 549 550 551 552 553 554 555
};

struct r600_irq_stat_regs {
	u32 disp_int;
	u32 disp_int_cont;
	u32 disp_int_cont2;
	u32 d1grph_int;
	u32 d2grph_int;
556 557
	u32 hdmi0_status;
	u32 hdmi1_status;
558 559 560 561 562 563 564 565 566 567 568 569 570 571 572
};

struct evergreen_irq_stat_regs {
	u32 disp_int;
	u32 disp_int_cont;
	u32 disp_int_cont2;
	u32 disp_int_cont3;
	u32 disp_int_cont4;
	u32 disp_int_cont5;
	u32 d1grph_int;
	u32 d2grph_int;
	u32 d3grph_int;
	u32 d4grph_int;
	u32 d5grph_int;
	u32 d6grph_int;
573 574 575 576 577 578
	u32 afmt_status1;
	u32 afmt_status2;
	u32 afmt_status3;
	u32 afmt_status4;
	u32 afmt_status5;
	u32 afmt_status6;
579 580 581 582 583 584 585 586
};

union radeon_irq_stat_regs {
	struct r500_irq_stat_regs r500;
	struct r600_irq_stat_regs r600;
	struct evergreen_irq_stat_regs evergreen;
};

587 588
#define RADEON_MAX_HPD_PINS 6
#define RADEON_MAX_CRTCS 6
589
#define RADEON_MAX_AFMT_BLOCKS 6
590

591
struct radeon_irq {
592 593
	bool				installed;
	spinlock_t			lock;
594
	atomic_t			ring_int[RADEON_NUM_RINGS];
595
	bool				crtc_vblank_int[RADEON_MAX_CRTCS];
596
	atomic_t			pflip[RADEON_MAX_CRTCS];
597 598 599 600
	wait_queue_head_t		vblank_queue;
	bool				hpd[RADEON_MAX_HPD_PINS];
	bool				afmt[RADEON_MAX_AFMT_BLOCKS];
	union radeon_irq_stat_regs	stat_regs;
601 602 603 604
};

int radeon_irq_kms_init(struct radeon_device *rdev);
void radeon_irq_kms_fini(struct radeon_device *rdev);
605 606
void radeon_irq_kms_sw_irq_get(struct radeon_device *rdev, int ring);
void radeon_irq_kms_sw_irq_put(struct radeon_device *rdev, int ring);
607 608
void radeon_irq_kms_pflip_irq_get(struct radeon_device *rdev, int crtc);
void radeon_irq_kms_pflip_irq_put(struct radeon_device *rdev, int crtc);
609 610 611 612
void radeon_irq_kms_enable_afmt(struct radeon_device *rdev, int block);
void radeon_irq_kms_disable_afmt(struct radeon_device *rdev, int block);
void radeon_irq_kms_enable_hpd(struct radeon_device *rdev, unsigned hpd_mask);
void radeon_irq_kms_disable_hpd(struct radeon_device *rdev, unsigned hpd_mask);
613 614

/*
615
 * CP & rings.
616
 */
617

618
struct radeon_ib {
619 620 621 622
	struct radeon_sa_bo		*sa_bo;
	uint32_t			length_dw;
	uint64_t			gpu_addr;
	uint32_t			*ptr;
623
	int				ring;
624
	struct radeon_fence		*fence;
625
	struct radeon_vm		*vm;
626
	bool				is_const_ib;
627
	struct radeon_fence		*sync_to[RADEON_NUM_RINGS];
628
	struct radeon_semaphore		*semaphore;
629 630
};

631
struct radeon_ring {
632
	struct radeon_bo	*ring_obj;
633 634
	volatile uint32_t	*ring;
	unsigned		rptr;
635 636
	unsigned		rptr_offs;
	unsigned		rptr_reg;
637
	unsigned		rptr_save_reg;
638 639
	u64			next_rptr_gpu_addr;
	volatile u32		*next_rptr_cpu_addr;
640 641
	unsigned		wptr;
	unsigned		wptr_old;
642
	unsigned		wptr_reg;
643 644 645
	unsigned		ring_size;
	unsigned		ring_free_dw;
	int			count_dw;
646 647
	unsigned long		last_activity;
	unsigned		last_rptr;
648 649 650 651
	uint64_t		gpu_addr;
	uint32_t		align_mask;
	uint32_t		ptr_mask;
	bool			ready;
652 653 654
	u32			ptr_reg_shift;
	u32			ptr_reg_mask;
	u32			nop;
655
	u32			idx;
656 657
	u64			last_semaphore_signal_addr;
	u64			last_semaphore_wait_addr;
658 659
};

660 661 662
/*
 * VM
 */
663

664
/* maximum number of VMIDs */
665 666
#define RADEON_NUM_VM	16

667 668 669 670 671 672 673 674
/* defines number of bits in page table versus page directory,
 * a page is 4KB so we have 12 bits offset, 9 bits in the page
 * table and the remaining 19 bits are in the page directory */
#define RADEON_VM_BLOCK_SIZE   9

/* number of entries in page table */
#define RADEON_VM_PTE_COUNT (1 << RADEON_VM_BLOCK_SIZE)

675 676 677
struct radeon_vm {
	struct list_head		list;
	struct list_head		va;
678
	unsigned			id;
679 680 681 682 683 684 685 686

	/* contains the page directory */
	struct radeon_sa_bo		*page_directory;
	uint64_t			pd_gpu_addr;

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

687 688 689
	struct mutex			mutex;
	/* last fence for cs using this vm */
	struct radeon_fence		*fence;
690 691
	/* last flush or NULL if we still need to flush */
	struct radeon_fence		*last_flush;
692 693 694
};

struct radeon_vm_manager {
695
	struct mutex			lock;
696
	struct list_head		lru_vm;
697
	struct radeon_fence		*active[RADEON_NUM_VM];
698 699 700 701 702 703
	struct radeon_sa_manager	sa_manager;
	uint32_t			max_pfn;
	/* number of VMIDs */
	unsigned			nvm;
	/* vram base address for page table entry  */
	u64				vram_base_offset;
704 705
	/* is vm enabled? */
	bool				enabled;
706 707 708 709 710 711 712 713 714
};

/*
 * file private structure
 */
struct radeon_fpriv {
	struct radeon_vm		vm;
};

715 716 717 718
/*
 * R6xx+ IH ring
 */
struct r600_ih {
719
	struct radeon_bo	*ring_obj;
720 721 722 723 724
	volatile uint32_t	*ring;
	unsigned		rptr;
	unsigned		ring_size;
	uint64_t		gpu_addr;
	uint32_t		ptr_mask;
725
	atomic_t		lock;
726 727 728
	bool                    enabled;
};

729 730 731 732 733 734 735 736 737 738
struct r600_blit_cp_primitives {
	void (*set_render_target)(struct radeon_device *rdev, int format,
				  int w, int h, u64 gpu_addr);
	void (*cp_set_surface_sync)(struct radeon_device *rdev,
				    u32 sync_type, u32 size,
				    u64 mc_addr);
	void (*set_shaders)(struct radeon_device *rdev);
	void (*set_vtx_resource)(struct radeon_device *rdev, u64 gpu_addr);
	void (*set_tex_resource)(struct radeon_device *rdev,
				 int format, int w, int h, int pitch,
739
				 u64 gpu_addr, u32 size);
740 741 742 743 744 745
	void (*set_scissors)(struct radeon_device *rdev, int x1, int y1,
			     int x2, int y2);
	void (*draw_auto)(struct radeon_device *rdev);
	void (*set_default_state)(struct radeon_device *rdev);
};

746
struct r600_blit {
747
	struct radeon_bo	*shader_obj;
748 749 750 751
	struct r600_blit_cp_primitives primitives;
	int max_dim;
	int ring_size_common;
	int ring_size_per_loop;
752 753 754 755 756 757
	u64 shader_gpu_addr;
	u32 vs_offset, ps_offset;
	u32 state_offset;
	u32 state_len;
};

758 759 760 761 762 763 764 765 766 767 768 769
/*
 * SI RLC stuff
 */
struct si_rlc {
	/* for power gating */
	struct radeon_bo	*save_restore_obj;
	uint64_t		save_restore_gpu_addr;
	/* for clear state */
	struct radeon_bo	*clear_state_obj;
	uint64_t		clear_state_gpu_addr;
};

770
int radeon_ib_get(struct radeon_device *rdev, int ring,
771 772
		  struct radeon_ib *ib, struct radeon_vm *vm,
		  unsigned size);
773
void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib);
774 775
int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib,
		       struct radeon_ib *const_ib);
776 777
int radeon_ib_pool_init(struct radeon_device *rdev);
void radeon_ib_pool_fini(struct radeon_device *rdev);
778
int radeon_ib_ring_tests(struct radeon_device *rdev);
779
/* Ring access between begin & end cannot sleep */
780 781
bool radeon_ring_supports_scratch_reg(struct radeon_device *rdev,
				      struct radeon_ring *ring);
782 783 784 785 786
void radeon_ring_free_size(struct radeon_device *rdev, struct radeon_ring *cp);
int radeon_ring_alloc(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ndw);
int radeon_ring_lock(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ndw);
void radeon_ring_commit(struct radeon_device *rdev, struct radeon_ring *cp);
void radeon_ring_unlock_commit(struct radeon_device *rdev, struct radeon_ring *cp);
787
void radeon_ring_undo(struct radeon_ring *ring);
788 789
void radeon_ring_unlock_undo(struct radeon_device *rdev, struct radeon_ring *cp);
int radeon_ring_test(struct radeon_device *rdev, struct radeon_ring *cp);
790
void radeon_ring_force_activity(struct radeon_device *rdev, struct radeon_ring *ring);
791 792
void radeon_ring_lockup_update(struct radeon_ring *ring);
bool radeon_ring_test_lockup(struct radeon_device *rdev, struct radeon_ring *ring);
793 794 795 796
unsigned radeon_ring_backup(struct radeon_device *rdev, struct radeon_ring *ring,
			    uint32_t **data);
int radeon_ring_restore(struct radeon_device *rdev, struct radeon_ring *ring,
			unsigned size, uint32_t *data);
797
int radeon_ring_init(struct radeon_device *rdev, struct radeon_ring *cp, unsigned ring_size,
798 799
		     unsigned rptr_offs, unsigned rptr_reg, unsigned wptr_reg,
		     u32 ptr_reg_shift, u32 ptr_reg_mask, u32 nop);
800
void radeon_ring_fini(struct radeon_device *rdev, struct radeon_ring *cp);
801 802


803 804 805 806 807
/* r600 async dma */
void r600_dma_stop(struct radeon_device *rdev);
int r600_dma_resume(struct radeon_device *rdev);
void r600_dma_fini(struct radeon_device *rdev);

808 809 810 811
void cayman_dma_stop(struct radeon_device *rdev);
int cayman_dma_resume(struct radeon_device *rdev);
void cayman_dma_fini(struct radeon_device *rdev);

812 813 814 815 816
/*
 * CS.
 */
struct radeon_cs_reloc {
	struct drm_gem_object		*gobj;
817 818
	struct radeon_bo		*robj;
	struct radeon_bo_list		lobj;
819 820 821 822 823 824 825
	uint32_t			handle;
	uint32_t			flags;
};

struct radeon_cs_chunk {
	uint32_t		chunk_id;
	uint32_t		length_dw;
826 827
	int			kpage_idx[2];
	uint32_t		*kpage[2];
828
	uint32_t		*kdata;
829 830 831
	void __user		*user_ptr;
	int			last_copied_page;
	int			last_page_index;
832 833 834
};

struct radeon_cs_parser {
835
	struct device		*dev;
836 837 838 839 840 841 842 843 844 845 846 847 848
	struct radeon_device	*rdev;
	struct drm_file		*filp;
	/* chunks */
	unsigned		nchunks;
	struct radeon_cs_chunk	*chunks;
	uint64_t		*chunks_array;
	/* IB */
	unsigned		idx;
	/* relocations */
	unsigned		nrelocs;
	struct radeon_cs_reloc	*relocs;
	struct radeon_cs_reloc	**relocs_ptr;
	struct list_head	validated;
849
	unsigned		dma_reloc_idx;
850 851 852
	/* indices of various chunks */
	int			chunk_ib_idx;
	int			chunk_relocs_idx;
853
	int			chunk_flags_idx;
854
	int			chunk_const_ib_idx;
855 856
	struct radeon_ib	ib;
	struct radeon_ib	const_ib;
857
	void			*track;
858
	unsigned		family;
859
	int			parser_error;
860 861 862
	u32			cs_flags;
	u32			ring;
	s32			priority;
863 864
};

865
extern int radeon_cs_finish_pages(struct radeon_cs_parser *p);
866
extern u32 radeon_get_ib_value(struct radeon_cs_parser *p, int idx);
867

868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887
struct radeon_cs_packet {
	unsigned	idx;
	unsigned	type;
	unsigned	reg;
	unsigned	opcode;
	int		count;
	unsigned	one_reg_wr;
};

typedef int (*radeon_packet0_check_t)(struct radeon_cs_parser *p,
				      struct radeon_cs_packet *pkt,
				      unsigned idx, unsigned reg);
typedef int (*radeon_packet3_check_t)(struct radeon_cs_parser *p,
				      struct radeon_cs_packet *pkt);


/*
 * AGP
 */
int radeon_agp_init(struct radeon_device *rdev);
888
void radeon_agp_resume(struct radeon_device *rdev);
889
void radeon_agp_suspend(struct radeon_device *rdev);
890 891 892 893 894 895 896
void radeon_agp_fini(struct radeon_device *rdev);


/*
 * Writeback
 */
struct radeon_wb {
897
	struct radeon_bo	*wb_obj;
898 899
	volatile uint32_t	*wb;
	uint64_t		gpu_addr;
900
	bool                    enabled;
901
	bool                    use_event;
902 903
};

904
#define RADEON_WB_SCRATCH_OFFSET 0
905
#define RADEON_WB_RING0_NEXT_RPTR 256
906
#define RADEON_WB_CP_RPTR_OFFSET 1024
907 908
#define RADEON_WB_CP1_RPTR_OFFSET 1280
#define RADEON_WB_CP2_RPTR_OFFSET 1536
909
#define R600_WB_DMA_RPTR_OFFSET   1792
910
#define R600_WB_IH_WPTR_OFFSET   2048
911
#define CAYMAN_WB_DMA1_RPTR_OFFSET   2304
912
#define R600_WB_EVENT_OFFSET     3072
913

914 915 916 917 918 919 920 921 922 923 924
/**
 * struct radeon_pm - power management datas
 * @max_bandwidth:      maximum bandwidth the gpu has (MByte/s)
 * @igp_sideport_mclk:  sideport memory clock Mhz (rs690,rs740,rs780,rs880)
 * @igp_system_mclk:    system clock Mhz (rs690,rs740,rs780,rs880)
 * @igp_ht_link_clk:    ht link clock Mhz (rs690,rs740,rs780,rs880)
 * @igp_ht_link_width:  ht link width in bits (rs690,rs740,rs780,rs880)
 * @k8_bandwidth:       k8 bandwidth the gpu has (MByte/s) (IGP)
 * @sideport_bandwidth: sideport bandwidth the gpu has (MByte/s) (IGP)
 * @ht_bandwidth:       ht bandwidth the gpu has (MByte/s) (IGP)
 * @core_bandwidth:     core GPU bandwidth the gpu has (MByte/s) (IGP)
L
Lucas De Marchi 已提交
925
 * @sclk:          	GPU clock Mhz (core bandwidth depends of this clock)
926 927 928
 * @needed_bandwidth:   current bandwidth needs
 *
 * It keeps track of various data needed to take powermanagement decision.
L
Lucas De Marchi 已提交
929
 * Bandwidth need is used to determine minimun clock of the GPU and memory.
930 931 932
 * Equation between gpu/memory clock and available bandwidth is hw dependent
 * (type of memory, bus size, efficiency, ...)
 */
933 934 935 936 937 938 939 940 941 942

enum radeon_pm_method {
	PM_METHOD_PROFILE,
	PM_METHOD_DYNPM,
};

enum radeon_dynpm_state {
	DYNPM_STATE_DISABLED,
	DYNPM_STATE_MINIMUM,
	DYNPM_STATE_PAUSED,
943 944
	DYNPM_STATE_ACTIVE,
	DYNPM_STATE_SUSPENDED,
945
};
946 947 948 949 950 951
enum radeon_dynpm_action {
	DYNPM_ACTION_NONE,
	DYNPM_ACTION_MINIMUM,
	DYNPM_ACTION_DOWNCLOCK,
	DYNPM_ACTION_UPCLOCK,
	DYNPM_ACTION_DEFAULT
952
};
953 954 955 956 957 958 959 960

enum radeon_voltage_type {
	VOLTAGE_NONE = 0,
	VOLTAGE_GPIO,
	VOLTAGE_VDDC,
	VOLTAGE_SW
};

961 962 963 964 965 966 967 968
enum radeon_pm_state_type {
	POWER_STATE_TYPE_DEFAULT,
	POWER_STATE_TYPE_POWERSAVE,
	POWER_STATE_TYPE_BATTERY,
	POWER_STATE_TYPE_BALANCED,
	POWER_STATE_TYPE_PERFORMANCE,
};

969 970 971 972
enum radeon_pm_profile_type {
	PM_PROFILE_DEFAULT,
	PM_PROFILE_AUTO,
	PM_PROFILE_LOW,
973
	PM_PROFILE_MID,
974 975 976 977 978
	PM_PROFILE_HIGH,
};

#define PM_PROFILE_DEFAULT_IDX 0
#define PM_PROFILE_LOW_SH_IDX  1
979 980 981 982 983 984
#define PM_PROFILE_MID_SH_IDX  2
#define PM_PROFILE_HIGH_SH_IDX 3
#define PM_PROFILE_LOW_MH_IDX  4
#define PM_PROFILE_MID_MH_IDX  5
#define PM_PROFILE_HIGH_MH_IDX 6
#define PM_PROFILE_MAX         7
985 986 987 988 989 990

struct radeon_pm_profile {
	int dpms_off_ps_idx;
	int dpms_on_ps_idx;
	int dpms_off_cm_idx;
	int dpms_on_cm_idx;
991 992
};

993 994 995 996 997
enum radeon_int_thermal_type {
	THERMAL_TYPE_NONE,
	THERMAL_TYPE_RV6XX,
	THERMAL_TYPE_RV770,
	THERMAL_TYPE_EVERGREEN,
998
	THERMAL_TYPE_SUMO,
999
	THERMAL_TYPE_NI,
1000
	THERMAL_TYPE_SI,
1001 1002
};

1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013
struct radeon_voltage {
	enum radeon_voltage_type type;
	/* gpio voltage */
	struct radeon_gpio_rec gpio;
	u32 delay; /* delay in usec from voltage drop to sclk change */
	bool active_high; /* voltage drop is active when bit is high */
	/* VDDC voltage */
	u8 vddc_id; /* index into vddc voltage table */
	u8 vddci_id; /* index into vddci voltage table */
	bool vddci_enabled;
	/* r6xx+ sw */
1014 1015 1016
	u16 voltage;
	/* evergreen+ vddci */
	u16 vddci;
1017 1018
};

1019 1020 1021
/* clock mode flags */
#define RADEON_PM_MODE_NO_DISPLAY          (1 << 0)

1022 1023 1024 1025 1026 1027 1028
struct radeon_pm_clock_info {
	/* memory clock */
	u32 mclk;
	/* engine clock */
	u32 sclk;
	/* voltage info */
	struct radeon_voltage voltage;
1029
	/* standardized clock flags */
1030 1031 1032
	u32 flags;
};

1033
/* state flags */
1034
#define RADEON_PM_STATE_SINGLE_DISPLAY_ONLY (1 << 0)
1035

1036
struct radeon_power_state {
1037
	enum radeon_pm_state_type type;
1038
	struct radeon_pm_clock_info *clock_info;
1039 1040 1041
	/* number of valid clock modes in this power state */
	int num_clock_modes;
	struct radeon_pm_clock_info *default_clock_mode;
1042 1043
	/* standardized state flags */
	u32 flags;
A
Alex Deucher 已提交
1044 1045 1046
	u32 misc; /* vbios specific flags */
	u32 misc2; /* vbios specific flags */
	int pcie_lanes; /* pcie lanes */
1047 1048
};

1049 1050 1051 1052 1053
/*
 * Some modes are overclocked by very low value, accept them
 */
#define RADEON_MODE_OVERCLOCK_MARGIN 500 /* 5 MHz */

1054
struct radeon_pm {
1055
	struct mutex		mutex;
1056 1057
	/* write locked while reprogramming mclk */
	struct rw_semaphore	mclk_lock;
1058 1059
	u32			active_crtcs;
	int			active_crtc_count;
1060
	int			req_vblank;
1061
	bool			vblank_sync;
1062 1063 1064 1065 1066 1067 1068 1069 1070 1071
	fixed20_12		max_bandwidth;
	fixed20_12		igp_sideport_mclk;
	fixed20_12		igp_system_mclk;
	fixed20_12		igp_ht_link_clk;
	fixed20_12		igp_ht_link_width;
	fixed20_12		k8_bandwidth;
	fixed20_12		sideport_bandwidth;
	fixed20_12		ht_bandwidth;
	fixed20_12		core_bandwidth;
	fixed20_12		sclk;
1072
	fixed20_12		mclk;
1073
	fixed20_12		needed_bandwidth;
1074
	struct radeon_power_state *power_state;
1075 1076
	/* number of valid power states */
	int                     num_power_states;
1077 1078 1079 1080 1081 1082 1083
	int                     current_power_state_index;
	int                     current_clock_mode_index;
	int                     requested_power_state_index;
	int                     requested_clock_mode_index;
	int                     default_power_state_index;
	u32                     current_sclk;
	u32                     current_mclk;
1084 1085
	u16                     current_vddc;
	u16                     current_vddci;
1086 1087
	u32                     default_sclk;
	u32                     default_mclk;
1088 1089
	u16                     default_vddc;
	u16                     default_vddci;
1090
	struct radeon_i2c_chan *i2c_bus;
1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103
	/* selected pm method */
	enum radeon_pm_method     pm_method;
	/* dynpm power management */
	struct delayed_work	dynpm_idle_work;
	enum radeon_dynpm_state	dynpm_state;
	enum radeon_dynpm_action	dynpm_planned_action;
	unsigned long		dynpm_action_timeout;
	bool                    dynpm_can_upclock;
	bool                    dynpm_can_downclock;
	/* profile-based power management */
	enum radeon_pm_profile_type profile;
	int                     profile_index;
	struct radeon_pm_profile profiles[PM_PROFILE_MAX];
1104 1105 1106
	/* internal thermal controller on rv6xx+ */
	enum radeon_int_thermal_type int_thermal_type;
	struct device	        *int_hwmon_dev;
1107 1108
};

1109 1110 1111
int radeon_pm_get_type_index(struct radeon_device *rdev,
			     enum radeon_pm_state_type ps_type,
			     int instance);
1112

1113 1114 1115 1116 1117 1118 1119 1120
struct r600_audio {
	int			channels;
	int			rate;
	int			bits_per_sample;
	u8			status_bits;
	u8			category_code;
};

1121 1122 1123
/*
 * Benchmarking
 */
1124
void radeon_benchmark(struct radeon_device *rdev, int test_number);
1125 1126


1127 1128 1129 1130
/*
 * Testing
 */
void radeon_test_moves(struct radeon_device *rdev);
1131
void radeon_test_ring_sync(struct radeon_device *rdev,
1132 1133
			   struct radeon_ring *cpA,
			   struct radeon_ring *cpB);
1134
void radeon_test_syncing(struct radeon_device *rdev);
1135 1136


1137 1138 1139
/*
 * Debugfs
 */
1140 1141 1142 1143 1144
struct radeon_debugfs {
	struct drm_info_list	*files;
	unsigned		num_files;
};

1145 1146 1147 1148 1149 1150 1151 1152 1153 1154
int radeon_debugfs_add_files(struct radeon_device *rdev,
			     struct drm_info_list *files,
			     unsigned nfiles);
int radeon_debugfs_fence_init(struct radeon_device *rdev);


/*
 * ASIC specific functions.
 */
struct radeon_asic {
1155
	int (*init)(struct radeon_device *rdev);
1156 1157 1158
	void (*fini)(struct radeon_device *rdev);
	int (*resume)(struct radeon_device *rdev);
	int (*suspend)(struct radeon_device *rdev);
1159
	void (*vga_set_state)(struct radeon_device *rdev, bool state);
1160
	int (*asic_reset)(struct radeon_device *rdev);
1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172
	/* ioctl hw specific callback. Some hw might want to perform special
	 * operation on specific ioctl. For instance on wait idle some hw
	 * might want to perform and HDP flush through MMIO as it seems that
	 * some R6XX/R7XX hw doesn't take HDP flush into account if programmed
	 * through ring.
	 */
	void (*ioctl_wait_idle)(struct radeon_device *rdev, struct radeon_bo *bo);
	/* check if 3D engine is idle */
	bool (*gui_idle)(struct radeon_device *rdev);
	/* wait for mc_idle */
	int (*mc_wait_for_idle)(struct radeon_device *rdev);
	/* gart */
1173 1174 1175 1176
	struct {
		void (*tlb_flush)(struct radeon_device *rdev);
		int (*set_page)(struct radeon_device *rdev, int i, uint64_t addr);
	} gart;
1177 1178 1179
	struct {
		int (*init)(struct radeon_device *rdev);
		void (*fini)(struct radeon_device *rdev);
1180 1181

		u32 pt_ring_index;
1182 1183 1184
		void (*set_page)(struct radeon_device *rdev, uint64_t pe,
				 uint64_t addr, unsigned count,
				 uint32_t incr, uint32_t flags);
1185
	} vm;
1186
	/* ring specific callbacks */
1187 1188
	struct {
		void (*ib_execute)(struct radeon_device *rdev, struct radeon_ib *ib);
1189
		int (*ib_parse)(struct radeon_device *rdev, struct radeon_ib *ib);
1190
		void (*emit_fence)(struct radeon_device *rdev, struct radeon_fence *fence);
1191
		void (*emit_semaphore)(struct radeon_device *rdev, struct radeon_ring *cp,
1192
				       struct radeon_semaphore *semaphore, bool emit_wait);
1193
		int (*cs_parse)(struct radeon_cs_parser *p);
1194 1195 1196
		void (*ring_start)(struct radeon_device *rdev, struct radeon_ring *cp);
		int (*ring_test)(struct radeon_device *rdev, struct radeon_ring *cp);
		int (*ib_test)(struct radeon_device *rdev, struct radeon_ring *cp);
1197
		bool (*is_lockup)(struct radeon_device *rdev, struct radeon_ring *cp);
1198
		void (*vm_flush)(struct radeon_device *rdev, int ridx, struct radeon_vm *vm);
1199
	} ring[RADEON_NUM_RINGS];
1200
	/* irqs */
1201 1202 1203 1204
	struct {
		int (*set)(struct radeon_device *rdev);
		int (*process)(struct radeon_device *rdev);
	} irq;
1205
	/* displays */
1206 1207 1208 1209 1210 1211 1212
	struct {
		/* display watermarks */
		void (*bandwidth_update)(struct radeon_device *rdev);
		/* get frame count */
		u32 (*get_vblank_counter)(struct radeon_device *rdev, int crtc);
		/* wait for vblank */
		void (*wait_for_vblank)(struct radeon_device *rdev, int crtc);
1213 1214
		/* set backlight level */
		void (*set_backlight_level)(struct radeon_encoder *radeon_encoder, u8 level);
1215 1216
		/* get backlight level */
		u8 (*get_backlight_level)(struct radeon_encoder *radeon_encoder);
1217
	} display;
1218
	/* copy functions for bo handling */
1219 1220 1221 1222 1223
	struct {
		int (*blit)(struct radeon_device *rdev,
			    uint64_t src_offset,
			    uint64_t dst_offset,
			    unsigned num_gpu_pages,
1224
			    struct radeon_fence **fence);
1225 1226 1227 1228 1229
		u32 blit_ring_index;
		int (*dma)(struct radeon_device *rdev,
			   uint64_t src_offset,
			   uint64_t dst_offset,
			   unsigned num_gpu_pages,
1230
			   struct radeon_fence **fence);
1231 1232 1233 1234 1235 1236
		u32 dma_ring_index;
		/* method used for bo copy */
		int (*copy)(struct radeon_device *rdev,
			    uint64_t src_offset,
			    uint64_t dst_offset,
			    unsigned num_gpu_pages,
1237
			    struct radeon_fence **fence);
1238 1239 1240
		/* ring used for bo copies */
		u32 copy_ring_index;
	} copy;
1241
	/* surfaces */
1242 1243 1244 1245 1246 1247
	struct {
		int (*set_reg)(struct radeon_device *rdev, int reg,
				       uint32_t tiling_flags, uint32_t pitch,
				       uint32_t offset, uint32_t obj_size);
		void (*clear_reg)(struct radeon_device *rdev, int reg);
	} surface;
1248
	/* hotplug detect */
1249 1250 1251 1252 1253 1254
	struct {
		void (*init)(struct radeon_device *rdev);
		void (*fini)(struct radeon_device *rdev);
		bool (*sense)(struct radeon_device *rdev, enum radeon_hpd_id hpd);
		void (*set_polarity)(struct radeon_device *rdev, enum radeon_hpd_id hpd);
	} hpd;
1255
	/* power management */
1256 1257 1258 1259 1260 1261
	struct {
		void (*misc)(struct radeon_device *rdev);
		void (*prepare)(struct radeon_device *rdev);
		void (*finish)(struct radeon_device *rdev);
		void (*init_profile)(struct radeon_device *rdev);
		void (*get_dynpm_state)(struct radeon_device *rdev);
1262 1263 1264 1265 1266 1267 1268
		uint32_t (*get_engine_clock)(struct radeon_device *rdev);
		void (*set_engine_clock)(struct radeon_device *rdev, uint32_t eng_clock);
		uint32_t (*get_memory_clock)(struct radeon_device *rdev);
		void (*set_memory_clock)(struct radeon_device *rdev, uint32_t mem_clock);
		int (*get_pcie_lanes)(struct radeon_device *rdev);
		void (*set_pcie_lanes)(struct radeon_device *rdev, int lanes);
		void (*set_clock_gating)(struct radeon_device *rdev, int enable);
1269
	} pm;
1270
	/* pageflipping */
1271 1272 1273 1274 1275
	struct {
		void (*pre_page_flip)(struct radeon_device *rdev, int crtc);
		u32 (*page_flip)(struct radeon_device *rdev, int crtc, u64 crtc_base);
		void (*post_page_flip)(struct radeon_device *rdev, int crtc);
	} pflip;
1276 1277
};

1278 1279 1280
/*
 * Asic structures
 */
1281
struct r100_asic {
1282 1283 1284
	const unsigned		*reg_safe_bm;
	unsigned		reg_safe_bm_size;
	u32			hdp_cntl;
1285 1286
};

1287
struct r300_asic {
1288 1289 1290 1291
	const unsigned		*reg_safe_bm;
	unsigned		reg_safe_bm_size;
	u32			resync_scratch;
	u32			hdp_cntl;
1292 1293 1294
};

struct r600_asic {
1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310
	unsigned		max_pipes;
	unsigned		max_tile_pipes;
	unsigned		max_simds;
	unsigned		max_backends;
	unsigned		max_gprs;
	unsigned		max_threads;
	unsigned		max_stack_entries;
	unsigned		max_hw_contexts;
	unsigned		max_gs_threads;
	unsigned		sx_max_export_size;
	unsigned		sx_max_export_pos_size;
	unsigned		sx_max_export_smx_size;
	unsigned		sq_num_cf_insts;
	unsigned		tiling_nbanks;
	unsigned		tiling_npipes;
	unsigned		tiling_group_size;
1311
	unsigned		tile_config;
1312
	unsigned		backend_map;
1313 1314 1315
};

struct rv770_asic {
1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335
	unsigned		max_pipes;
	unsigned		max_tile_pipes;
	unsigned		max_simds;
	unsigned		max_backends;
	unsigned		max_gprs;
	unsigned		max_threads;
	unsigned		max_stack_entries;
	unsigned		max_hw_contexts;
	unsigned		max_gs_threads;
	unsigned		sx_max_export_size;
	unsigned		sx_max_export_pos_size;
	unsigned		sx_max_export_smx_size;
	unsigned		sq_num_cf_insts;
	unsigned		sx_num_of_sets;
	unsigned		sc_prim_fifo_size;
	unsigned		sc_hiz_tile_fifo_size;
	unsigned		sc_earlyz_tile_fifo_fize;
	unsigned		tiling_nbanks;
	unsigned		tiling_npipes;
	unsigned		tiling_group_size;
1336
	unsigned		tile_config;
1337
	unsigned		backend_map;
1338 1339
};

1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361
struct evergreen_asic {
	unsigned num_ses;
	unsigned max_pipes;
	unsigned max_tile_pipes;
	unsigned max_simds;
	unsigned max_backends;
	unsigned max_gprs;
	unsigned max_threads;
	unsigned max_stack_entries;
	unsigned max_hw_contexts;
	unsigned max_gs_threads;
	unsigned sx_max_export_size;
	unsigned sx_max_export_pos_size;
	unsigned sx_max_export_smx_size;
	unsigned sq_num_cf_insts;
	unsigned sx_num_of_sets;
	unsigned sc_prim_fifo_size;
	unsigned sc_hiz_tile_fifo_size;
	unsigned sc_earlyz_tile_fifo_size;
	unsigned tiling_nbanks;
	unsigned tiling_npipes;
	unsigned tiling_group_size;
1362
	unsigned tile_config;
1363
	unsigned backend_map;
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
struct cayman_asic {
	unsigned max_shader_engines;
	unsigned max_pipes_per_simd;
	unsigned max_tile_pipes;
	unsigned max_simds_per_se;
	unsigned max_backends_per_se;
	unsigned max_texture_channel_caches;
	unsigned max_gprs;
	unsigned max_threads;
	unsigned max_gs_threads;
	unsigned max_stack_entries;
	unsigned sx_num_of_sets;
	unsigned sx_max_export_size;
	unsigned sx_max_export_pos_size;
	unsigned sx_max_export_smx_size;
	unsigned max_hw_contexts;
	unsigned sq_num_cf_insts;
	unsigned sc_prim_fifo_size;
	unsigned sc_hiz_tile_fifo_size;
	unsigned sc_earlyz_tile_fifo_size;

	unsigned num_shader_engines;
	unsigned num_shader_pipes_per_simd;
	unsigned num_tile_pipes;
	unsigned num_simds_per_se;
	unsigned num_backends_per_se;
	unsigned backend_disable_mask_per_asic;
	unsigned backend_map;
	unsigned num_texture_channel_caches;
	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 tile_config;
};

1404 1405 1406
struct si_asic {
	unsigned max_shader_engines;
	unsigned max_tile_pipes;
A
Alex Deucher 已提交
1407 1408
	unsigned max_cu_per_sh;
	unsigned max_sh_per_se;
1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432
	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 num_backends_per_se;
	unsigned backend_disable_mask_per_asic;
	unsigned backend_map;
	unsigned num_texture_channel_caches;
	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 tile_config;
};

1433 1434
union radeon_asic_config {
	struct r300_asic	r300;
1435
	struct r100_asic	r100;
1436 1437
	struct r600_asic	r600;
	struct rv770_asic	rv770;
1438
	struct evergreen_asic	evergreen;
1439
	struct cayman_asic	cayman;
1440
	struct si_asic		si;
1441 1442
};

D
Daniel Vetter 已提交
1443 1444 1445 1446 1447 1448
/*
 * asic initizalization from radeon_asic.c
 */
void radeon_agp_disable(struct radeon_device *rdev);
int radeon_asic_init(struct radeon_device *rdev);

1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472

/*
 * IOCTL.
 */
int radeon_gem_info_ioctl(struct drm_device *dev, void *data,
			  struct drm_file *filp);
int radeon_gem_create_ioctl(struct drm_device *dev, void *data,
			    struct drm_file *filp);
int radeon_gem_pin_ioctl(struct drm_device *dev, void *data,
			 struct drm_file *file_priv);
int radeon_gem_unpin_ioctl(struct drm_device *dev, void *data,
			   struct drm_file *file_priv);
int radeon_gem_pwrite_ioctl(struct drm_device *dev, void *data,
			    struct drm_file *file_priv);
int radeon_gem_pread_ioctl(struct drm_device *dev, void *data,
			   struct drm_file *file_priv);
int radeon_gem_set_domain_ioctl(struct drm_device *dev, void *data,
				struct drm_file *filp);
int radeon_gem_mmap_ioctl(struct drm_device *dev, void *data,
			  struct drm_file *filp);
int radeon_gem_busy_ioctl(struct drm_device *dev, void *data,
			  struct drm_file *filp);
int radeon_gem_wait_idle_ioctl(struct drm_device *dev, void *data,
			      struct drm_file *filp);
1473 1474
int radeon_gem_va_ioctl(struct drm_device *dev, void *data,
			  struct drm_file *filp);
1475
int radeon_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp);
1476 1477 1478 1479
int radeon_gem_set_tiling_ioctl(struct drm_device *dev, void *data,
				struct drm_file *filp);
int radeon_gem_get_tiling_ioctl(struct drm_device *dev, void *data,
				struct drm_file *filp);
1480

1481 1482
/* VRAM scratch page for HDP bug, default vram page */
struct r600_vram_scratch {
1483 1484
	struct radeon_bo		*robj;
	volatile uint32_t		*ptr;
1485
	u64				gpu_addr;
1486
};
1487

1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524
/*
 * ACPI
 */
struct radeon_atif_notification_cfg {
	bool enabled;
	int command_code;
};

struct radeon_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 radeon_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 radeon_atif {
	struct radeon_atif_notifications notifications;
	struct radeon_atif_functions functions;
	struct radeon_atif_notification_cfg notification_cfg;
1525
	struct radeon_encoder *encoder_for_bl;
1526
};
1527

1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538
struct radeon_atcs_functions {
	bool get_ext_state;
	bool pcie_perf_req;
	bool pcie_dev_rdy;
	bool pcie_bus_width;
};

struct radeon_atcs {
	struct radeon_atcs_functions functions;
};

1539 1540 1541 1542 1543 1544 1545
/*
 * Core structure, functions and helpers.
 */
typedef uint32_t (*radeon_rreg_t)(struct radeon_device*, uint32_t);
typedef void (*radeon_wreg_t)(struct radeon_device*, uint32_t, uint32_t);

struct radeon_device {
1546
	struct device			*dev;
1547 1548
	struct drm_device		*ddev;
	struct pci_dev			*pdev;
1549
	struct rw_semaphore		exclusive_lock;
1550
	/* ASIC */
1551
	union radeon_asic_config	config;
1552 1553 1554 1555 1556
	enum radeon_family		family;
	unsigned long			flags;
	int				usec_timeout;
	enum radeon_pll_errata		pll_errata;
	int				num_gb_pipes;
1557
	int				num_z_pipes;
1558 1559 1560 1561 1562
	int				disp_priority;
	/* BIOS */
	uint8_t				*bios;
	bool				is_atom_bios;
	uint16_t			bios_header_start;
1563
	struct radeon_bo		*stollen_vga_memory;
1564
	/* Register mmio */
1565 1566
	resource_size_t			rmmio_base;
	resource_size_t			rmmio_size;
1567 1568
	/* protects concurrent MM_INDEX/DATA based register access */
	spinlock_t mmio_idx_lock;
1569
	void __iomem			*rmmio;
1570 1571 1572 1573
	radeon_rreg_t			mc_rreg;
	radeon_wreg_t			mc_wreg;
	radeon_rreg_t			pll_rreg;
	radeon_wreg_t			pll_wreg;
1574
	uint32_t                        pcie_reg_mask;
1575 1576
	radeon_rreg_t			pciep_rreg;
	radeon_wreg_t			pciep_wreg;
1577 1578 1579
	/* io port */
	void __iomem                    *rio_mem;
	resource_size_t			rio_mem_size;
1580 1581 1582 1583 1584 1585
	struct radeon_clock             clock;
	struct radeon_mc		mc;
	struct radeon_gart		gart;
	struct radeon_mode_info		mode_info;
	struct radeon_scratch		scratch;
	struct radeon_mman		mman;
1586
	struct radeon_fence_driver	fence_drv[RADEON_NUM_RINGS];
1587
	wait_queue_head_t		fence_queue;
1588
	struct mutex			ring_lock;
1589
	struct radeon_ring		ring[RADEON_NUM_RINGS];
J
Jerome Glisse 已提交
1590 1591
	bool				ib_pool_ready;
	struct radeon_sa_manager	ring_tmp_bo;
1592 1593 1594
	struct radeon_irq		irq;
	struct radeon_asic		*asic;
	struct radeon_gem		gem;
1595
	struct radeon_pm		pm;
1596
	uint32_t			bios_scratch[RADEON_BIOS_NUM_SCRATCH];
1597
	struct radeon_wb		wb;
1598
	struct radeon_dummy_page	dummy_page;
1599 1600
	bool				shutdown;
	bool				suspend;
D
Dave Airlie 已提交
1601
	bool				need_dma32;
1602
	bool				accel_working;
1603
	struct radeon_surface_reg surface_regs[RADEON_GEM_MAX_SURFACES];
1604 1605
	const struct firmware *me_fw;	/* all family ME firmware */
	const struct firmware *pfp_fw;	/* r6/700 PFP firmware */
1606
	const struct firmware *rlc_fw;	/* r6/700 RLC firmware */
1607
	const struct firmware *mc_fw;	/* NI MC firmware */
1608
	const struct firmware *ce_fw;	/* SI CE firmware */
1609
	struct r600_blit r600_blit;
1610
	struct r600_vram_scratch vram_scratch;
A
Alex Deucher 已提交
1611
	int msi_enabled; /* msi enabled */
1612
	struct r600_ih ih; /* r6/700 interrupt ring */
1613
	struct si_rlc rlc;
A
Alex Deucher 已提交
1614
	struct work_struct hotplug_work;
1615
	struct work_struct audio_work;
1616
	int num_crtc; /* number of crtcs */
1617
	struct mutex dc_hw_i2c_mutex; /* display controller hw i2c mutex */
1618 1619
	bool audio_enabled;
	struct r600_audio audio_status; /* audio stuff */
1620
	struct notifier_block acpi_nb;
1621
	/* only one userspace can use Hyperz features or CMASK at a time */
1622
	struct drm_file *hyperz_filp;
1623
	struct drm_file *cmask_filp;
1624 1625
	/* i2c buses */
	struct radeon_i2c_chan *i2c_bus[RADEON_MAX_I2C_BUS];
1626 1627 1628
	/* debugfs */
	struct radeon_debugfs	debugfs[RADEON_DEBUGFS_MAX_COMPONENTS];
	unsigned 		debugfs_count;
1629 1630
	/* virtual memory */
	struct radeon_vm_manager	vm_manager;
1631
	struct mutex			gpu_clock_mutex;
1632 1633
	/* ACPI interface */
	struct radeon_atif		atif;
1634
	struct radeon_atcs		atcs;
1635 1636 1637 1638 1639 1640 1641 1642 1643
};

int radeon_device_init(struct radeon_device *rdev,
		       struct drm_device *ddev,
		       struct pci_dev *pdev,
		       uint32_t flags);
void radeon_device_fini(struct radeon_device *rdev);
int radeon_gpu_wait_for_idle(struct radeon_device *rdev);

1644 1645 1646 1647
uint32_t r100_mm_rreg(struct radeon_device *rdev, uint32_t reg,
		      bool always_indirect);
void r100_mm_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v,
		  bool always_indirect);
1648 1649
u32 r100_io_rreg(struct radeon_device *rdev, u32 reg);
void r100_io_wreg(struct radeon_device *rdev, u32 reg, u32 v);
1650

1651 1652 1653 1654
/*
 * Cast helper
 */
#define to_radeon_fence(p) ((struct radeon_fence *)(p))
1655 1656 1657 1658

/*
 * Registers read & write functions.
 */
1659 1660 1661 1662
#define RREG8(reg) readb((rdev->rmmio) + (reg))
#define WREG8(reg, v) writeb(v, (rdev->rmmio) + (reg))
#define RREG16(reg) readw((rdev->rmmio) + (reg))
#define WREG16(reg, v) writew(v, (rdev->rmmio) + (reg))
1663 1664 1665 1666 1667
#define RREG32(reg) r100_mm_rreg(rdev, (reg), false)
#define RREG32_IDX(reg) r100_mm_rreg(rdev, (reg), true)
#define DREG32(reg) printk(KERN_INFO "REGISTER: " #reg " : 0x%08X\n", r100_mm_rreg(rdev, (reg), false))
#define WREG32(reg, v) r100_mm_wreg(rdev, (reg), (v), false)
#define WREG32_IDX(reg, v) r100_mm_wreg(rdev, (reg), (v), true)
1668 1669 1670 1671 1672 1673
#define REG_SET(FIELD, v) (((v) << FIELD##_SHIFT) & FIELD##_MASK)
#define REG_GET(FIELD, v) (((v) << FIELD##_SHIFT) & FIELD##_MASK)
#define RREG32_PLL(reg) rdev->pll_rreg(rdev, (reg))
#define WREG32_PLL(reg, v) rdev->pll_wreg(rdev, (reg), (v))
#define RREG32_MC(reg) rdev->mc_rreg(rdev, (reg))
#define WREG32_MC(reg, v) rdev->mc_wreg(rdev, (reg), (v))
1674 1675
#define RREG32_PCIE(reg) rv370_pcie_rreg(rdev, (reg))
#define WREG32_PCIE(reg, v) rv370_pcie_wreg(rdev, (reg), (v))
1676 1677
#define RREG32_PCIE_P(reg) rdev->pciep_rreg(rdev, (reg))
#define WREG32_PCIE_P(reg, v) rdev->pciep_wreg(rdev, (reg), (v))
1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691
#define WREG32_P(reg, val, mask)				\
	do {							\
		uint32_t tmp_ = RREG32(reg);			\
		tmp_ &= (mask);					\
		tmp_ |= ((val) & ~(mask));			\
		WREG32(reg, tmp_);				\
	} while (0)
#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)
1692
#define DREG32_SYS(sqf, rdev, reg) seq_printf((sqf), #reg " : 0x%08X\n", r100_mm_rreg((rdev), (reg), false))
1693 1694
#define RREG32_IO(reg) r100_io_rreg(rdev, (reg))
#define WREG32_IO(reg, v) r100_io_wreg(rdev, (reg), (v))
1695

1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713
/*
 * Indirect registers accessor
 */
static inline uint32_t rv370_pcie_rreg(struct radeon_device *rdev, uint32_t reg)
{
	uint32_t r;

	WREG32(RADEON_PCIE_INDEX, ((reg) & rdev->pcie_reg_mask));
	r = RREG32(RADEON_PCIE_DATA);
	return r;
}

static inline void rv370_pcie_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v)
{
	WREG32(RADEON_PCIE_INDEX, ((reg) & rdev->pcie_reg_mask));
	WREG32(RADEON_PCIE_DATA, (v));
}

1714 1715 1716 1717 1718 1719
void r100_pll_errata_after_index(struct radeon_device *rdev);


/*
 * ASICs helpers.
 */
1720 1721
#define ASIC_IS_RN50(rdev) ((rdev->pdev->device == 0x515e) || \
			    (rdev->pdev->device == 0x5969))
1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737
#define ASIC_IS_RV100(rdev) ((rdev->family == CHIP_RV100) || \
		(rdev->family == CHIP_RV200) || \
		(rdev->family == CHIP_RS100) || \
		(rdev->family == CHIP_RS200) || \
		(rdev->family == CHIP_RV250) || \
		(rdev->family == CHIP_RV280) || \
		(rdev->family == CHIP_RS300))
#define ASIC_IS_R300(rdev) ((rdev->family == CHIP_R300)  ||	\
		(rdev->family == CHIP_RV350) ||			\
		(rdev->family == CHIP_R350)  ||			\
		(rdev->family == CHIP_RV380) ||			\
		(rdev->family == CHIP_R420)  ||			\
		(rdev->family == CHIP_R423)  ||			\
		(rdev->family == CHIP_RV410) ||			\
		(rdev->family == CHIP_RS400) ||			\
		(rdev->family == CHIP_RS480))
1738 1739 1740 1741 1742 1743 1744 1745
#define ASIC_IS_X2(rdev) ((rdev->ddev->pdev->device == 0x9441) || \
		(rdev->ddev->pdev->device == 0x9443) || \
		(rdev->ddev->pdev->device == 0x944B) || \
		(rdev->ddev->pdev->device == 0x9506) || \
		(rdev->ddev->pdev->device == 0x9509) || \
		(rdev->ddev->pdev->device == 0x950F) || \
		(rdev->ddev->pdev->device == 0x689C) || \
		(rdev->ddev->pdev->device == 0x689D))
1746
#define ASIC_IS_AVIVO(rdev) ((rdev->family >= CHIP_RS600))
1747 1748 1749 1750
#define ASIC_IS_DCE2(rdev) ((rdev->family == CHIP_RS600)  ||	\
			    (rdev->family == CHIP_RS690)  ||	\
			    (rdev->family == CHIP_RS740)  ||	\
			    (rdev->family >= CHIP_R600))
1751 1752
#define ASIC_IS_DCE3(rdev) ((rdev->family >= CHIP_RV620))
#define ASIC_IS_DCE32(rdev) ((rdev->family >= CHIP_RV730))
1753
#define ASIC_IS_DCE4(rdev) ((rdev->family >= CHIP_CEDAR))
1754 1755
#define ASIC_IS_DCE41(rdev) ((rdev->family >= CHIP_PALM) && \
			     (rdev->flags & RADEON_IS_IGP))
1756
#define ASIC_IS_DCE5(rdev) ((rdev->family >= CHIP_BARTS))
1757 1758 1759
#define ASIC_IS_DCE6(rdev) ((rdev->family >= CHIP_ARUBA))
#define ASIC_IS_DCE61(rdev) ((rdev->family >= CHIP_ARUBA) && \
			     (rdev->flags & RADEON_IS_IGP))
1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776

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

int radeon_combios_init(struct radeon_device *rdev);
void radeon_combios_fini(struct radeon_device *rdev);
int radeon_atombios_init(struct radeon_device *rdev);
void radeon_atombios_fini(struct radeon_device *rdev);


/*
 * RING helpers.
 */
1777
#if DRM_DEBUG_CODE == 0
1778
static inline void radeon_ring_write(struct radeon_ring *ring, uint32_t v)
1779
{
1780 1781 1782 1783
	ring->ring[ring->wptr++] = v;
	ring->wptr &= ring->ptr_mask;
	ring->count_dw--;
	ring->ring_free_dw--;
1784
}
1785 1786
#else
/* With debugging this is just too big to inline */
1787
void radeon_ring_write(struct radeon_ring *ring, uint32_t v);
1788
#endif
1789 1790 1791 1792

/*
 * ASICs macro.
 */
1793
#define radeon_init(rdev) (rdev)->asic->init((rdev))
1794 1795 1796
#define radeon_fini(rdev) (rdev)->asic->fini((rdev))
#define radeon_resume(rdev) (rdev)->asic->resume((rdev))
#define radeon_suspend(rdev) (rdev)->asic->suspend((rdev))
1797
#define radeon_cs_parse(rdev, r, p) (rdev)->asic->ring[(r)].cs_parse((p))
1798
#define radeon_vga_set_state(rdev, state) (rdev)->asic->vga_set_state((rdev), (state))
1799
#define radeon_asic_reset(rdev) (rdev)->asic->asic_reset((rdev))
1800 1801
#define radeon_gart_tlb_flush(rdev) (rdev)->asic->gart.tlb_flush((rdev))
#define radeon_gart_set_page(rdev, i, p) (rdev)->asic->gart.set_page((rdev), (i), (p))
1802 1803
#define radeon_asic_vm_init(rdev) (rdev)->asic->vm.init((rdev))
#define radeon_asic_vm_fini(rdev) (rdev)->asic->vm.fini((rdev))
1804
#define radeon_asic_vm_set_page(rdev, pe, addr, count, incr, flags) ((rdev)->asic->vm.set_page((rdev), (pe), (addr), (count), (incr), (flags)))
1805 1806 1807
#define radeon_ring_start(rdev, r, cp) (rdev)->asic->ring[(r)].ring_start((rdev), (cp))
#define radeon_ring_test(rdev, r, cp) (rdev)->asic->ring[(r)].ring_test((rdev), (cp))
#define radeon_ib_test(rdev, r, cp) (rdev)->asic->ring[(r)].ib_test((rdev), (cp))
1808
#define radeon_ring_ib_execute(rdev, r, ib) (rdev)->asic->ring[(r)].ib_execute((rdev), (ib))
1809
#define radeon_ring_ib_parse(rdev, r, ib) (rdev)->asic->ring[(r)].ib_parse((rdev), (ib))
1810
#define radeon_ring_is_lockup(rdev, r, cp) (rdev)->asic->ring[(r)].is_lockup((rdev), (cp))
1811
#define radeon_ring_vm_flush(rdev, r, vm) (rdev)->asic->ring[(r)].vm_flush((rdev), (r), (vm))
1812 1813
#define radeon_irq_set(rdev) (rdev)->asic->irq.set((rdev))
#define radeon_irq_process(rdev) (rdev)->asic->irq.process((rdev))
1814
#define radeon_get_vblank_counter(rdev, crtc) (rdev)->asic->display.get_vblank_counter((rdev), (crtc))
1815
#define radeon_set_backlight_level(rdev, e, l) (rdev)->asic->display.set_backlight_level((e), (l))
1816
#define radeon_get_backlight_level(rdev, e) (rdev)->asic->display.get_backlight_level((e))
1817 1818
#define radeon_fence_ring_emit(rdev, r, fence) (rdev)->asic->ring[(r)].emit_fence((rdev), (fence))
#define radeon_semaphore_ring_emit(rdev, r, cp, semaphore, emit_wait) (rdev)->asic->ring[(r)].emit_semaphore((rdev), (cp), (semaphore), (emit_wait))
1819 1820 1821 1822 1823 1824
#define radeon_copy_blit(rdev, s, d, np, f) (rdev)->asic->copy.blit((rdev), (s), (d), (np), (f))
#define radeon_copy_dma(rdev, s, d, np, f) (rdev)->asic->copy.dma((rdev), (s), (d), (np), (f))
#define radeon_copy(rdev, s, d, np, f) (rdev)->asic->copy.copy((rdev), (s), (d), (np), (f))
#define radeon_copy_blit_ring_index(rdev) (rdev)->asic->copy.blit_ring_index
#define radeon_copy_dma_ring_index(rdev) (rdev)->asic->copy.dma_ring_index
#define radeon_copy_ring_index(rdev) (rdev)->asic->copy.copy_ring_index
1825 1826 1827 1828 1829 1830 1831
#define radeon_get_engine_clock(rdev) (rdev)->asic->pm.get_engine_clock((rdev))
#define radeon_set_engine_clock(rdev, e) (rdev)->asic->pm.set_engine_clock((rdev), (e))
#define radeon_get_memory_clock(rdev) (rdev)->asic->pm.get_memory_clock((rdev))
#define radeon_set_memory_clock(rdev, e) (rdev)->asic->pm.set_memory_clock((rdev), (e))
#define radeon_get_pcie_lanes(rdev) (rdev)->asic->pm.get_pcie_lanes((rdev))
#define radeon_set_pcie_lanes(rdev, l) (rdev)->asic->pm.set_pcie_lanes((rdev), (l))
#define radeon_set_clock_gating(rdev, e) (rdev)->asic->pm.set_clock_gating((rdev), (e))
1832 1833
#define radeon_set_surface_reg(rdev, r, f, p, o, s) ((rdev)->asic->surface.set_reg((rdev), (r), (f), (p), (o), (s)))
#define radeon_clear_surface_reg(rdev, r) ((rdev)->asic->surface.clear_reg((rdev), (r)))
1834
#define radeon_bandwidth_update(rdev) (rdev)->asic->display.bandwidth_update((rdev))
1835 1836 1837 1838
#define radeon_hpd_init(rdev) (rdev)->asic->hpd.init((rdev))
#define radeon_hpd_fini(rdev) (rdev)->asic->hpd.fini((rdev))
#define radeon_hpd_sense(rdev, h) (rdev)->asic->hpd.sense((rdev), (h))
#define radeon_hpd_set_polarity(rdev, h) (rdev)->asic->hpd.set_polarity((rdev), (h))
1839
#define radeon_gui_idle(rdev) (rdev)->asic->gui_idle((rdev))
1840 1841 1842 1843 1844
#define radeon_pm_misc(rdev) (rdev)->asic->pm.misc((rdev))
#define radeon_pm_prepare(rdev) (rdev)->asic->pm.prepare((rdev))
#define radeon_pm_finish(rdev) (rdev)->asic->pm.finish((rdev))
#define radeon_pm_init_profile(rdev) (rdev)->asic->pm.init_profile((rdev))
#define radeon_pm_get_dynpm_state(rdev) (rdev)->asic->pm.get_dynpm_state((rdev))
1845 1846 1847 1848 1849
#define radeon_pre_page_flip(rdev, crtc) (rdev)->asic->pflip.pre_page_flip((rdev), (crtc))
#define radeon_page_flip(rdev, crtc, base) (rdev)->asic->pflip.page_flip((rdev), (crtc), (base))
#define radeon_post_page_flip(rdev, crtc) (rdev)->asic->pflip.post_page_flip((rdev), (crtc))
#define radeon_wait_for_vblank(rdev, crtc) (rdev)->asic->display.wait_for_vblank((rdev), (crtc))
#define radeon_mc_wait_for_idle(rdev) (rdev)->asic->mc_wait_for_idle((rdev))
1850

1851
/* Common functions */
1852
/* AGP */
1853
extern int radeon_gpu_reset(struct radeon_device *rdev);
1854
extern void radeon_agp_disable(struct radeon_device *rdev);
1855 1856
extern int radeon_modeset_init(struct radeon_device *rdev);
extern void radeon_modeset_fini(struct radeon_device *rdev);
1857
extern bool radeon_card_posted(struct radeon_device *rdev);
1858
extern void radeon_update_bandwidth_info(struct radeon_device *rdev);
1859
extern void radeon_update_display_priority(struct radeon_device *rdev);
1860
extern bool radeon_boot_test_post_card(struct radeon_device *rdev);
1861
extern void radeon_scratch_init(struct radeon_device *rdev);
1862 1863 1864
extern void radeon_wb_fini(struct radeon_device *rdev);
extern int radeon_wb_init(struct radeon_device *rdev);
extern void radeon_wb_disable(struct radeon_device *rdev);
1865 1866
extern void radeon_surface_init(struct radeon_device *rdev);
extern int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data);
1867
extern void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enable);
1868
extern void radeon_atom_set_clock_gating(struct radeon_device *rdev, int enable);
1869
extern void radeon_ttm_placement_from_domain(struct radeon_bo *rbo, u32 domain);
1870
extern bool radeon_ttm_bo_is_radeon_bo(struct ttm_buffer_object *bo);
1871 1872
extern void radeon_vram_location(struct radeon_device *rdev, struct radeon_mc *mc, u64 base);
extern void radeon_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc);
1873 1874
extern int radeon_resume_kms(struct drm_device *dev);
extern int radeon_suspend_kms(struct drm_device *dev, pm_message_t state);
1875
extern void radeon_ttm_set_active_vram_size(struct radeon_device *rdev, u64 size);
1876

1877 1878 1879 1880 1881
/*
 * vm
 */
int radeon_vm_manager_init(struct radeon_device *rdev);
void radeon_vm_manager_fini(struct radeon_device *rdev);
1882
void radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm);
1883
void radeon_vm_fini(struct radeon_device *rdev, struct radeon_vm *vm);
1884
int radeon_vm_alloc_pt(struct radeon_device *rdev, struct radeon_vm *vm);
1885
void radeon_vm_add_to_lru(struct radeon_device *rdev, struct radeon_vm *vm);
1886 1887 1888 1889 1890
struct radeon_fence *radeon_vm_grab_id(struct radeon_device *rdev,
				       struct radeon_vm *vm, int ring);
void radeon_vm_fence(struct radeon_device *rdev,
		     struct radeon_vm *vm,
		     struct radeon_fence *fence);
1891
uint64_t radeon_vm_map_gart(struct radeon_device *rdev, uint64_t addr);
1892 1893 1894 1895 1896 1897
int radeon_vm_bo_update_pte(struct radeon_device *rdev,
			    struct radeon_vm *vm,
			    struct radeon_bo *bo,
			    struct ttm_mem_reg *mem);
void radeon_vm_bo_invalidate(struct radeon_device *rdev,
			     struct radeon_bo *bo);
1898 1899
struct radeon_bo_va *radeon_vm_bo_find(struct radeon_vm *vm,
				       struct radeon_bo *bo);
1900 1901 1902 1903 1904 1905 1906
struct radeon_bo_va *radeon_vm_bo_add(struct radeon_device *rdev,
				      struct radeon_vm *vm,
				      struct radeon_bo *bo);
int radeon_vm_bo_set_addr(struct radeon_device *rdev,
			  struct radeon_bo_va *bo_va,
			  uint64_t offset,
			  uint32_t flags);
1907
int radeon_vm_bo_rmv(struct radeon_device *rdev,
1908
		     struct radeon_bo_va *bo_va);
1909

1910 1911
/* audio */
void r600_audio_update_hdmi(struct work_struct *work);
1912

1913 1914 1915 1916 1917 1918
/*
 * R600 vram scratch functions
 */
int r600_vram_scratch_init(struct radeon_device *rdev);
void r600_vram_scratch_fini(struct radeon_device *rdev);

1919 1920 1921 1922 1923 1924 1925 1926 1927 1928
/*
 * r600 cs checking helper
 */
unsigned r600_mip_minify(unsigned size, unsigned level);
bool r600_fmt_is_valid_color(u32 format);
bool r600_fmt_is_valid_texture(u32 format, enum radeon_family family);
int r600_fmt_get_blocksize(u32 format);
int r600_fmt_get_nblocksx(u32 format, u32 w);
int r600_fmt_get_nblocksy(u32 format, u32 h);

1929 1930 1931
/*
 * r600 functions used by radeon_encoder.c
 */
1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945
struct radeon_hdmi_acr {
	u32 clock;

	int n_32khz;
	int cts_32khz;

	int n_44_1khz;
	int cts_44_1khz;

	int n_48khz;
	int cts_48khz;

};

1946 1947
extern struct radeon_hdmi_acr r600_hdmi_acr(uint32_t clock);

1948 1949
extern void r600_hdmi_enable(struct drm_encoder *encoder);
extern void r600_hdmi_disable(struct drm_encoder *encoder);
1950
extern void r600_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mode);
1951 1952 1953 1954 1955
extern u32 r6xx_remap_render_backend(struct radeon_device *rdev,
				     u32 tiling_pipe_num,
				     u32 max_rb_num,
				     u32 total_max_rb_num,
				     u32 enabled_rb_mask);
1956

1957 1958 1959 1960 1961 1962
/*
 * evergreen functions used by radeon_encoder.c
 */

extern void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_display_mode *mode);

1963
extern int ni_init_microcode(struct radeon_device *rdev);
1964
extern int ni_mc_load_microcode(struct radeon_device *rdev);
1965

1966 1967 1968 1969 1970 1971 1972 1973
/* radeon_acpi.c */
#if defined(CONFIG_ACPI)
extern int radeon_acpi_init(struct radeon_device *rdev);
extern void radeon_acpi_fini(struct radeon_device *rdev);
#else
static inline int radeon_acpi_init(struct radeon_device *rdev) { return 0; }
static inline void radeon_acpi_fini(struct radeon_device *rdev) { }
#endif
1974

1975 1976 1977
int radeon_cs_packet_parse(struct radeon_cs_parser *p,
			   struct radeon_cs_packet *pkt,
			   unsigned idx);
1978 1979
bool radeon_cs_packet_next_is_pkt3_nop(struct radeon_cs_parser *p);

1980

1981 1982
#include "radeon_object.h"

1983
#endif