summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu.h
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu.h')
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu.h117
1 files changed, 83 insertions, 34 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index da3d02154fa6..8ebc5f1eb4c0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -85,8 +85,12 @@ extern int amdgpu_vm_debug;
extern int amdgpu_sched_jobs;
extern int amdgpu_sched_hw_submission;
extern int amdgpu_powerplay;
+extern int amdgpu_powercontainment;
extern unsigned amdgpu_pcie_gen_cap;
extern unsigned amdgpu_pcie_lane_cap;
+extern unsigned amdgpu_cg_mask;
+extern unsigned amdgpu_pg_mask;
+extern char *amdgpu_disable_cu;
#define AMDGPU_WAIT_IDLE_TIMEOUT_IN_MS 3000
#define AMDGPU_MAX_USEC_TIMEOUT 100000 /* 100 ms */
@@ -183,6 +187,10 @@ int amdgpu_set_clockgating_state(struct amdgpu_device *adev,
int amdgpu_set_powergating_state(struct amdgpu_device *adev,
enum amd_ip_block_type block_type,
enum amd_powergating_state state);
+int amdgpu_wait_for_idle(struct amdgpu_device *adev,
+ enum amd_ip_block_type block_type);
+bool amdgpu_is_idle(struct amdgpu_device *adev,
+ enum amd_ip_block_type block_type);
struct amdgpu_ip_block_version {
enum amd_ip_block_type type;
@@ -298,13 +306,16 @@ struct amdgpu_ring_funcs {
uint32_t oa_base, uint32_t oa_size);
/* testing functions */
int (*test_ring)(struct amdgpu_ring *ring);
- int (*test_ib)(struct amdgpu_ring *ring);
+ int (*test_ib)(struct amdgpu_ring *ring, long timeout);
/* insert NOP packets */
void (*insert_nop)(struct amdgpu_ring *ring, uint32_t count);
/* pad the indirect buffer to the necessary number of dw */
void (*pad_ib)(struct amdgpu_ring *ring, struct amdgpu_ib *ib);
unsigned (*init_cond_exec)(struct amdgpu_ring *ring);
void (*patch_cond_exec)(struct amdgpu_ring *ring, unsigned offset);
+ /* note usage for clock and power gating */
+ void (*begin_use)(struct amdgpu_ring *ring);
+ void (*end_use)(struct amdgpu_ring *ring);
};
/*
@@ -594,11 +605,9 @@ int amdgpu_sync_resv(struct amdgpu_device *adev,
struct amdgpu_sync *sync,
struct reservation_object *resv,
void *owner);
-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);
+struct fence *amdgpu_sync_peek_fence(struct amdgpu_sync *sync,
+ struct amdgpu_ring *ring);
struct fence *amdgpu_sync_get_fence(struct amdgpu_sync *sync);
-int amdgpu_sync_wait(struct amdgpu_sync *sync);
void amdgpu_sync_free(struct amdgpu_sync *sync);
int amdgpu_sync_init(void);
void amdgpu_sync_fini(void);
@@ -754,12 +763,11 @@ int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs,
int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev, unsigned size,
struct amdgpu_job **job);
+void amdgpu_job_free_resources(struct amdgpu_job *job);
void amdgpu_job_free(struct amdgpu_job *job);
-void amdgpu_job_free_func(struct kref *refcount);
int amdgpu_job_submit(struct amdgpu_job *job, struct amdgpu_ring *ring,
struct amd_sched_entity *entity, void *owner,
struct fence **f);
-void amdgpu_job_timeout_func(struct work_struct *work);
struct amdgpu_ring {
struct amdgpu_device *adev;
@@ -767,12 +775,9 @@ struct amdgpu_ring {
struct amdgpu_fence_driver fence_drv;
struct amd_gpu_scheduler sched;
- spinlock_t fence_lock;
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;
@@ -791,14 +796,16 @@ struct amdgpu_ring {
u32 doorbell_index;
bool use_doorbell;
unsigned wptr_offs;
- unsigned next_rptr_offs;
unsigned fence_offs;
uint64_t current_ctx;
enum amdgpu_ring_type type;
char name[16];
unsigned cond_exe_offs;
- u64 cond_exe_gpu_addr;
- volatile u32 *cond_exe_cpu_addr;
+ u64 cond_exe_gpu_addr;
+ volatile u32 *cond_exe_cpu_addr;
+#if defined(CONFIG_DEBUG_FS)
+ struct dentry *ent;
+#endif
};
/*
@@ -861,6 +868,7 @@ struct amdgpu_vm {
struct amdgpu_bo *page_directory;
unsigned max_pde_used;
struct fence *page_directory_fence;
+ uint64_t last_eviction_counter;
/* array of page tables, one for each page directory entry */
struct amdgpu_vm_pt *page_tables;
@@ -883,13 +891,14 @@ struct amdgpu_vm_id {
struct fence *first;
struct amdgpu_sync active;
struct fence *last_flush;
- struct amdgpu_ring *last_user;
atomic64_t owner;
uint64_t pd_gpu_addr;
/* last flushed PD/PT update */
struct fence *flushed_updates;
+ uint32_t current_gpu_reset_count;
+
uint32_t gds_base;
uint32_t gds_size;
uint32_t gws_base;
@@ -905,6 +914,10 @@ struct amdgpu_vm_manager {
struct list_head ids_lru;
struct amdgpu_vm_id ids[AMDGPU_NUM_VM];
+ /* Handling of VM fences */
+ u64 fence_context;
+ unsigned seqno[AMDGPU_MAX_RINGS];
+
uint32_t max_pfn;
/* vram base address for page table entry */
u64 vram_base_offset;
@@ -926,17 +939,14 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm);
void amdgpu_vm_get_pd_bo(struct amdgpu_vm *vm,
struct list_head *validated,
struct amdgpu_bo_list_entry *entry);
-void amdgpu_vm_get_pt_bos(struct amdgpu_vm *vm, struct list_head *duplicates);
+void amdgpu_vm_get_pt_bos(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+ struct list_head *duplicates);
void amdgpu_vm_move_pt_bos_in_lru(struct amdgpu_device *adev,
struct amdgpu_vm *vm);
int amdgpu_vm_grab_id(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
struct amdgpu_sync *sync, struct fence *fence,
- unsigned *vm_id, uint64_t *vm_pd_addr);
-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);
+ struct amdgpu_job *job);
+int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job);
void amdgpu_vm_reset_id(struct amdgpu_device *adev, unsigned vm_id);
uint64_t amdgpu_vm_map_gart(const dma_addr_t *pages_addr, uint64_t addr);
int amdgpu_vm_update_page_directory(struct amdgpu_device *adev,
@@ -1142,6 +1152,12 @@ struct amdgpu_cu_info {
uint32_t bitmap[4][4];
};
+struct amdgpu_gfx_funcs {
+ /* get the gpu clock counter */
+ uint64_t (*get_gpu_clock_counter)(struct amdgpu_device *adev);
+ void (*select_se_sh)(struct amdgpu_device *adev, u32 se_num, u32 sh_num, u32 instance);
+};
+
struct amdgpu_gfx {
struct mutex gpu_clock_mutex;
struct amdgpu_gca_config config;
@@ -1178,6 +1194,7 @@ struct amdgpu_gfx {
/* ce ram size*/
unsigned ce_ram_size;
struct amdgpu_cu_info cu_info;
+ const struct amdgpu_gfx_funcs *funcs;
};
int amdgpu_ib_get(struct amdgpu_device *adev, struct amdgpu_vm *vm,
@@ -1195,10 +1212,6 @@ void amdgpu_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count);
void amdgpu_ring_generic_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib);
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,
@@ -1250,6 +1263,7 @@ struct amdgpu_job {
uint32_t num_ibs;
void *owner;
uint64_t ctx;
+ bool vm_needs_flush;
unsigned vm_id;
uint64_t vm_pd_addr;
uint32_t gds_base, gds_size;
@@ -1257,8 +1271,7 @@ struct amdgpu_job {
uint32_t oa_base, oa_size;
/* user fence handling */
- struct amdgpu_bo *uf_bo;
- uint32_t uf_offset;
+ uint64_t uf_addr;
uint64_t uf_sequence;
};
@@ -1560,6 +1573,12 @@ struct amdgpu_dpm_funcs {
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);
+ int (*force_clock_level)(struct amdgpu_device *adev, enum pp_clock_type type, uint32_t mask);
+ int (*print_clock_levels)(struct amdgpu_device *adev, enum pp_clock_type type, char *buf);
+ int (*get_sclk_od)(struct amdgpu_device *adev);
+ int (*set_sclk_od)(struct amdgpu_device *adev, uint32_t value);
+ int (*get_mclk_od)(struct amdgpu_device *adev);
+ int (*set_mclk_od)(struct amdgpu_device *adev, uint32_t value);
};
struct amdgpu_dpm {
@@ -1662,6 +1681,7 @@ struct amdgpu_uvd {
struct amdgpu_ring ring;
struct amdgpu_irq_src irq;
bool address_64_bit;
+ bool use_ctx_buf;
struct amd_sched_entity entity;
};
@@ -1683,6 +1703,7 @@ struct amdgpu_vce {
struct drm_file *filp[AMDGPU_MAX_VCE_HANDLES];
uint32_t img_size[AMDGPU_MAX_VCE_HANDLES];
struct delayed_work idle_work;
+ struct mutex idle_mutex;
const struct firmware *fw; /* VCE firmware */
struct amdgpu_ring ring[AMDGPU_MAX_VCE_RINGS];
struct amdgpu_irq_src irq;
@@ -1767,6 +1788,8 @@ int amdgpu_debugfs_init(struct drm_minor *minor);
void amdgpu_debugfs_cleanup(struct drm_minor *minor);
#endif
+int amdgpu_debugfs_firmware_init(struct amdgpu_device *adev);
+
/*
* amdgpu smumgr functions
*/
@@ -1811,15 +1834,13 @@ struct amdgpu_asic_funcs {
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);
/* 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);
+ /* query virtual capabilities */
+ u32 (*get_virtual_caps)(struct amdgpu_device *adev);
};
/*
@@ -1914,8 +1935,12 @@ void amdgpu_cgs_destroy_device(struct cgs_device *cgs_device);
/* GPU virtualization */
+#define AMDGPU_VIRT_CAPS_SRIOV_EN (1 << 0)
+#define AMDGPU_VIRT_CAPS_IS_VF (1 << 1)
struct amdgpu_virtualization {
bool supports_sr_iov;
+ bool is_virtual;
+ u32 caps;
};
/*
@@ -1997,6 +2022,10 @@ struct amdgpu_device {
spinlock_t didt_idx_lock;
amdgpu_rreg_t didt_rreg;
amdgpu_wreg_t didt_wreg;
+ /* protects concurrent gc_cac register access */
+ spinlock_t gc_cac_idx_lock;
+ amdgpu_rreg_t gc_cac_rreg;
+ amdgpu_wreg_t gc_cac_wreg;
/* protects concurrent ENDPOINT (audio) register access */
spinlock_t audio_endpt_idx_lock;
amdgpu_block_rreg_t audio_endpt_rreg;
@@ -2022,6 +2051,7 @@ struct amdgpu_device {
atomic64_t vram_vis_usage;
atomic64_t gtt_usage;
atomic64_t num_bytes_moved;
+ atomic64_t num_evictions;
atomic_t gpu_reset_counter;
/* display */
@@ -2125,6 +2155,8 @@ void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v);
#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_GC_CAC(reg) adev->gc_cac_rreg(adev, (reg))
+#define WREG32_GC_CAC(reg, v) adev->gc_cac_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) \
@@ -2200,11 +2232,10 @@ amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
*/
#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_get_virtual_caps(adev) ((adev)->asic_funcs->get_virtual_caps((adev)))
#define amdgpu_asic_read_disabled_bios(adev) (adev)->asic_funcs->read_disabled_bios((adev))
#define amdgpu_asic_read_bios_from_rom(adev, b, l) (adev)->asic_funcs->read_bios_from_rom((adev), (b), (l))
#define amdgpu_asic_read_register(adev, se, sh, offset, v)((adev)->asic_funcs->read_register((adev), (se), (sh), (offset), (v)))
@@ -2215,7 +2246,7 @@ amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
#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_test_ib(r, t) (r)->funcs->test_ib((r), (t))
#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))
@@ -2257,6 +2288,8 @@ amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
#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))
+#define amdgpu_gfx_get_gpu_clock_counter(adev) (adev)->gfx.funcs->get_gpu_clock_counter((adev))
+#define amdgpu_gfx_select_se_sh(adev, se, sh, instance) (adev)->gfx.funcs->select_se_sh((adev), (se), (sh), (instance))
#define amdgpu_dpm_get_temperature(adev) \
((adev)->pp_enabled ? \
@@ -2335,6 +2368,18 @@ amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
#define amdgpu_dpm_force_clock_level(adev, type, level) \
(adev)->powerplay.pp_funcs->force_clock_level((adev)->powerplay.pp_handle, type, level)
+#define amdgpu_dpm_get_sclk_od(adev) \
+ (adev)->powerplay.pp_funcs->get_sclk_od((adev)->powerplay.pp_handle)
+
+#define amdgpu_dpm_set_sclk_od(adev, value) \
+ (adev)->powerplay.pp_funcs->set_sclk_od((adev)->powerplay.pp_handle, value)
+
+#define amdgpu_dpm_get_mclk_od(adev) \
+ ((adev)->powerplay.pp_funcs->get_mclk_od((adev)->powerplay.pp_handle))
+
+#define amdgpu_dpm_set_mclk_od(adev, value) \
+ ((adev)->powerplay.pp_funcs->set_mclk_od((adev)->powerplay.pp_handle, value))
+
#define amdgpu_dpm_dispatch_task(adev, event_id, input, output) \
(adev)->powerplay.pp_funcs->dispatch_tasks((adev)->powerplay.pp_handle, (event_id), (input), (output))
@@ -2376,9 +2421,13 @@ bool amdgpu_device_is_px(struct drm_device *dev);
#if defined(CONFIG_VGA_SWITCHEROO)
void amdgpu_register_atpx_handler(void);
void amdgpu_unregister_atpx_handler(void);
+bool amdgpu_has_atpx_dgpu_power_cntl(void);
+bool amdgpu_is_atpx_hybrid(void);
#else
static inline void amdgpu_register_atpx_handler(void) {}
static inline void amdgpu_unregister_atpx_handler(void) {}
+static inline bool amdgpu_has_atpx_dgpu_power_cntl(void) { return false; }
+static inline bool amdgpu_is_atpx_hybrid(void) { return false; }
#endif
/*
generated by cgit v1.2.3 (git 2.39.0) at 2025-01-09 17:07:57 +0300