summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h')
-rw-r--r--drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h229
1 files changed, 173 insertions, 56 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 647220a8762d..8bdfcde2029b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -25,14 +25,18 @@
#ifndef AMDGPU_AMDKFD_H_INCLUDED
#define AMDGPU_AMDKFD_H_INCLUDED
+#include <linux/list.h>
#include <linux/types.h>
#include <linux/mm.h>
#include <linux/kthread.h>
#include <linux/workqueue.h>
+#include <linux/mmu_notifier.h>
+#include <linux/memremap.h>
#include <kgd_kfd_interface.h>
-#include <drm/ttm/ttm_execbuf_util.h>
+#include <drm/drm_client.h>
#include "amdgpu_sync.h"
#include "amdgpu_vm.h"
+#include "amdgpu_xcp.h"
extern uint64_t amdgpu_amdkfd_total_mem_size;
@@ -43,6 +47,8 @@ enum TLB_FLUSH_TYPE {
};
struct amdgpu_device;
+struct kfd_process_device;
+struct amdgpu_reset_context;
enum kfd_mem_attachment_type {
KFD_MEM_ATT_SHARED, /* Share kgd_mem->bo or another attachment's */
@@ -65,21 +71,22 @@ struct kgd_mem {
struct mutex lock;
struct amdgpu_bo *bo;
struct dma_buf *dmabuf;
+ struct amdgpu_hmm_range *range;
struct list_head attachments;
/* protected by amdkfd_process_info.lock */
- struct ttm_validate_buffer validate_list;
- struct ttm_validate_buffer resv_list;
+ struct list_head validate_list;
uint32_t domain;
unsigned int mapped_to_gpu_memory;
uint64_t va;
uint32_t alloc_flags;
- atomic_t invalid;
+ uint32_t invalid;
struct amdkfd_process_info *process_info;
struct amdgpu_sync sync;
+ uint32_t gem_handle;
bool aql_queue;
bool is_imported;
};
@@ -95,10 +102,18 @@ struct amdgpu_amdkfd_fence {
struct amdgpu_kfd_dev {
struct kfd_dev *dev;
- uint64_t vram_used;
- uint64_t vram_used_aligned;
+ int64_t vram_used[MAX_XCP];
+ uint64_t vram_used_aligned[MAX_XCP];
bool init_complete;
struct work_struct reset_work;
+
+ /* Client for KFD BO GEM handle allocations */
+ struct drm_client_dev client;
+
+ /* HMM page migration MEMORY_DEVICE_PRIVATE mapping
+ * Must be last --ends in a flexible-array member.
+ */
+ struct dev_pagemap pgmap;
};
enum kgd_engine_type {
@@ -131,7 +146,8 @@ struct amdkfd_process_info {
struct amdgpu_amdkfd_fence *eviction_fence;
/* MMU-notifier related fields */
- atomic_t evicted_bos;
+ struct mutex notifier_lock;
+ uint32_t evicted_bos;
struct delayed_work restore_userptr_work;
struct pid *pid;
bool block_mmu_notifications;
@@ -140,28 +156,28 @@ struct amdkfd_process_info {
int amdgpu_amdkfd_init(void);
void amdgpu_amdkfd_fini(void);
-void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool run_pm);
-int amdgpu_amdkfd_resume_iommu(struct amdgpu_device *adev);
-int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool run_pm);
+void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool suspend_proc);
+int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool resume_proc);
+void amdgpu_amdkfd_suspend_process(struct amdgpu_device *adev);
+int amdgpu_amdkfd_resume_process(struct amdgpu_device *adev);
void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev,
const void *ih_ring_entry);
void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev);
void amdgpu_amdkfd_device_init(struct amdgpu_device *adev);
void amdgpu_amdkfd_device_fini_sw(struct amdgpu_device *adev);
+int amdgpu_amdkfd_check_and_lock_kfd(struct amdgpu_device *adev);
+void amdgpu_amdkfd_unlock_kfd(struct amdgpu_device *adev);
int amdgpu_amdkfd_submit_ib(struct amdgpu_device *adev,
enum kgd_engine_type engine,
uint32_t vmid, uint64_t gpu_addr,
uint32_t *ib_cmd, uint32_t ib_len);
void amdgpu_amdkfd_set_compute_idle(struct amdgpu_device *adev, bool idle);
bool amdgpu_amdkfd_have_atomics_support(struct amdgpu_device *adev);
-int amdgpu_amdkfd_flush_gpu_tlb_vmid(struct amdgpu_device *adev,
- uint16_t vmid);
-int amdgpu_amdkfd_flush_gpu_tlb_pasid(struct amdgpu_device *adev,
- uint16_t pasid, enum TLB_FLUSH_TYPE flush_type);
bool amdgpu_amdkfd_is_kfd_vmid(struct amdgpu_device *adev, u32 vmid);
-int amdgpu_amdkfd_pre_reset(struct amdgpu_device *adev);
+int amdgpu_amdkfd_pre_reset(struct amdgpu_device *adev,
+ struct amdgpu_reset_context *reset_context);
int amdgpu_amdkfd_post_reset(struct amdgpu_device *adev);
@@ -173,14 +189,20 @@ int amdgpu_queue_mask_bit_to_set_resource_bit(struct amdgpu_device *adev,
struct amdgpu_amdkfd_fence *amdgpu_amdkfd_fence_create(u64 context,
struct mm_struct *mm,
struct svm_range_bo *svm_bo);
+
+int amdgpu_amdkfd_drm_client_create(struct amdgpu_device *adev);
#if defined(CONFIG_DEBUG_FS)
int kfd_debugfs_kfd_mem_limits(struct seq_file *m, void *data);
#endif
#if IS_ENABLED(CONFIG_HSA_AMD)
bool amdkfd_fence_check_mm(struct dma_fence *f, struct mm_struct *mm);
struct amdgpu_amdkfd_fence *to_amdgpu_amdkfd_fence(struct dma_fence *f);
-int amdgpu_amdkfd_remove_fence_on_pt_pd_bos(struct amdgpu_bo *bo);
-int amdgpu_amdkfd_evict_userptr(struct kgd_mem *mem, struct mm_struct *mm);
+void amdgpu_amdkfd_remove_all_eviction_fences(struct amdgpu_bo *bo);
+int amdgpu_amdkfd_evict_userptr(struct mmu_interval_notifier *mni,
+ unsigned long cur_seq, struct kgd_mem *mem);
+int amdgpu_amdkfd_bo_validate_and_fence(struct amdgpu_bo *bo,
+ uint32_t domain,
+ struct dma_fence *fence);
#else
static inline
bool amdkfd_fence_check_mm(struct dma_fence *f, struct mm_struct *mm)
@@ -195,13 +217,20 @@ struct amdgpu_amdkfd_fence *to_amdgpu_amdkfd_fence(struct dma_fence *f)
}
static inline
-int amdgpu_amdkfd_remove_fence_on_pt_pd_bos(struct amdgpu_bo *bo)
+void amdgpu_amdkfd_remove_all_eviction_fences(struct amdgpu_bo *bo)
{
- return 0;
}
static inline
-int amdgpu_amdkfd_evict_userptr(struct kgd_mem *mem, struct mm_struct *mm)
+int amdgpu_amdkfd_evict_userptr(struct mmu_interval_notifier *mni,
+ unsigned long cur_seq, struct kgd_mem *mem)
+{
+ return 0;
+}
+static inline
+int amdgpu_amdkfd_bo_validate_and_fence(struct amdgpu_bo *bo,
+ uint32_t domain,
+ struct dma_fence *fence)
{
return 0;
}
@@ -210,7 +239,7 @@ int amdgpu_amdkfd_evict_userptr(struct kgd_mem *mem, struct mm_struct *mm)
int amdgpu_amdkfd_alloc_gtt_mem(struct amdgpu_device *adev, size_t size,
void **mem_obj, uint64_t *gpu_addr,
void **cpu_ptr, bool mqd_gfx9);
-void amdgpu_amdkfd_free_gtt_mem(struct amdgpu_device *adev, void *mem_obj);
+void amdgpu_amdkfd_free_gtt_mem(struct amdgpu_device *adev, void **mem_obj);
int amdgpu_amdkfd_alloc_gws(struct amdgpu_device *adev, size_t size,
void **mem_obj);
void amdgpu_amdkfd_free_gws(struct amdgpu_device *adev, void *mem_obj);
@@ -219,23 +248,27 @@ int amdgpu_amdkfd_remove_gws_from_process(void *info, void *mem);
uint32_t amdgpu_amdkfd_get_fw_version(struct amdgpu_device *adev,
enum kgd_engine_type type);
void amdgpu_amdkfd_get_local_mem_info(struct amdgpu_device *adev,
- struct kfd_local_mem_info *mem_info);
+ struct kfd_local_mem_info *mem_info,
+ struct amdgpu_xcp *xcp);
uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct amdgpu_device *adev);
uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct amdgpu_device *adev);
-void amdgpu_amdkfd_get_cu_info(struct amdgpu_device *adev,
- struct kfd_cu_info *cu_info);
int amdgpu_amdkfd_get_dmabuf_info(struct amdgpu_device *adev, int dma_buf_fd,
struct amdgpu_device **dmabuf_adev,
uint64_t *bo_size, void *metadata_buffer,
size_t buffer_size, uint32_t *metadata_size,
- uint32_t *flags);
-uint8_t amdgpu_amdkfd_get_xgmi_hops_count(struct amdgpu_device *dst,
- struct amdgpu_device *src);
-int amdgpu_amdkfd_get_xgmi_bandwidth_mbytes(struct amdgpu_device *dst,
- struct amdgpu_device *src,
- bool is_min);
+ uint32_t *flags, int8_t *xcp_id);
int amdgpu_amdkfd_get_pcie_bandwidth_mbytes(struct amdgpu_device *adev, bool is_min);
+int amdgpu_amdkfd_send_close_event_drain_irq(struct amdgpu_device *adev,
+ uint32_t *payload);
+int amdgpu_amdkfd_unmap_hiq(struct amdgpu_device *adev, u32 doorbell_off,
+ u32 inst);
+int amdgpu_amdkfd_start_sched(struct amdgpu_device *adev, uint32_t node_id);
+int amdgpu_amdkfd_stop_sched(struct amdgpu_device *adev, uint32_t node_id);
+int amdgpu_amdkfd_config_sq_perfmon(struct amdgpu_device *adev, uint32_t xcp_id,
+ bool core_override_enable, bool reg_override_enable, bool perfmon_override_enable);
+bool amdgpu_amdkfd_compute_active(struct amdgpu_device *adev, uint32_t node_id);
+
/* Read user wptr from a specified user address space with page fault
* disabled. The memory must be pinned and mapped to the hardware when
@@ -266,13 +299,12 @@ int amdgpu_amdkfd_get_pcie_bandwidth_mbytes(struct amdgpu_device *adev, bool is_
((struct drm_file *)(drm_priv))->driver_priv)->vm)
int amdgpu_amdkfd_gpuvm_acquire_process_vm(struct amdgpu_device *adev,
- struct file *filp, u32 pasid,
+ struct amdgpu_vm *avm,
void **process_info,
struct dma_fence **ef);
-void amdgpu_amdkfd_gpuvm_release_process_vm(struct amdgpu_device *adev,
- void *drm_priv);
uint64_t amdgpu_amdkfd_gpuvm_get_process_page_dir(void *drm_priv);
-size_t amdgpu_amdkfd_get_available_memory(struct amdgpu_device *adev);
+size_t amdgpu_amdkfd_get_available_memory(struct amdgpu_device *adev,
+ uint8_t xcp_id);
int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
struct amdgpu_device *adev, uint64_t va, uint64_t size,
void *drm_priv, struct kgd_mem **mem,
@@ -284,35 +316,52 @@ int amdgpu_amdkfd_gpuvm_map_memory_to_gpu(struct amdgpu_device *adev,
struct kgd_mem *mem, void *drm_priv);
int amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu(
struct amdgpu_device *adev, struct kgd_mem *mem, void *drm_priv);
+int amdgpu_amdkfd_gpuvm_dmaunmap_mem(struct kgd_mem *mem, void *drm_priv);
int amdgpu_amdkfd_gpuvm_sync_memory(
struct amdgpu_device *adev, struct kgd_mem *mem, bool intr);
int amdgpu_amdkfd_gpuvm_map_gtt_bo_to_kernel(struct kgd_mem *mem,
void **kptr, uint64_t *size);
void amdgpu_amdkfd_gpuvm_unmap_gtt_bo_from_kernel(struct kgd_mem *mem);
-int amdgpu_amdkfd_map_gtt_bo_to_gart(struct amdgpu_device *adev, struct amdgpu_bo *bo);
+int amdgpu_amdkfd_map_gtt_bo_to_gart(struct amdgpu_bo *bo, struct amdgpu_bo **bo_gart);
int amdgpu_amdkfd_gpuvm_restore_process_bos(void *process_info,
- struct dma_fence **ef);
+ struct dma_fence __rcu **ef);
int amdgpu_amdkfd_gpuvm_get_vm_fault_info(struct amdgpu_device *adev,
struct kfd_vm_fault_info *info);
-int amdgpu_amdkfd_gpuvm_import_dmabuf(struct amdgpu_device *adev,
- struct dma_buf *dmabuf,
- uint64_t va, void *drm_priv,
- struct kgd_mem **mem, uint64_t *size,
- uint64_t *mmap_offset);
+int amdgpu_amdkfd_gpuvm_import_dmabuf_fd(struct amdgpu_device *adev, int fd,
+ uint64_t va, void *drm_priv,
+ struct kgd_mem **mem, uint64_t *size,
+ uint64_t *mmap_offset);
+int amdgpu_amdkfd_gpuvm_export_dmabuf(struct kgd_mem *mem,
+ struct dma_buf **dmabuf);
+void amdgpu_amdkfd_debug_mem_fence(struct amdgpu_device *adev);
int amdgpu_amdkfd_get_tile_config(struct amdgpu_device *adev,
struct tile_config *config);
void amdgpu_amdkfd_ras_poison_consumption_handler(struct amdgpu_device *adev,
- bool reset);
-bool amdgpu_amdkfd_bo_mapped_to_dev(struct amdgpu_device *adev, struct kgd_mem *mem);
+ enum amdgpu_ras_block block, uint32_t reset);
+
+void amdgpu_amdkfd_ras_pasid_poison_consumption_handler(struct amdgpu_device *adev,
+ enum amdgpu_ras_block block, uint16_t pasid,
+ pasid_notify pasid_fn, void *data, uint32_t reset);
+
+bool amdgpu_amdkfd_is_fed(struct amdgpu_device *adev);
+bool amdgpu_amdkfd_bo_mapped_to_dev(void *drm_priv, struct kgd_mem *mem);
void amdgpu_amdkfd_block_mmu_notifications(void *p);
int amdgpu_amdkfd_criu_resume(void *p);
-bool amdgpu_amdkfd_ras_query_utcl2_poison_status(struct amdgpu_device *adev);
int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev,
- uint64_t size, u32 alloc_flag);
+ uint64_t size, u32 alloc_flag, int8_t xcp_id);
void amdgpu_amdkfd_unreserve_mem_limit(struct amdgpu_device *adev,
- uint64_t size, u32 alloc_flag);
+ uint64_t size, u32 alloc_flag, int8_t xcp_id);
+
+u64 amdgpu_amdkfd_xcp_memory_size(struct amdgpu_device *adev, int xcp_id);
+
+#define KFD_XCP_MEM_ID(adev, xcp_id) \
+ ((adev)->xcp_mgr && (xcp_id) >= 0 ?\
+ (adev)->xcp_mgr->xcp[(xcp_id)].mem_id : -1)
+
+#define KFD_XCP_MEMORY_SIZE(adev, xcp_id) amdgpu_amdkfd_xcp_memory_size((adev), (xcp_id))
+
#if IS_ENABLED(CONFIG_HSA_AMD)
void amdgpu_amdkfd_gpuvm_init_mem_limits(void);
@@ -343,6 +392,17 @@ void amdgpu_amdkfd_release_notify(struct amdgpu_bo *bo)
{
}
#endif
+
+#if IS_ENABLED(CONFIG_HSA_AMD_SVM)
+int kgd2kfd_init_zone_device(struct amdgpu_device *adev);
+#else
+static inline
+int kgd2kfd_init_zone_device(struct amdgpu_device *adev)
+{
+ return 0;
+}
+#endif
+
/* KGD2KFD callbacks */
int kgd2kfd_quiesce_mm(struct mm_struct *mm, uint32_t trigger);
int kgd2kfd_resume_mm(struct mm_struct *mm);
@@ -353,17 +413,28 @@ int kgd2kfd_init(void);
void kgd2kfd_exit(void);
struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf);
bool kgd2kfd_device_init(struct kfd_dev *kfd,
- struct drm_device *ddev,
const struct kgd2kfd_shared_resources *gpu_resources);
void kgd2kfd_device_exit(struct kfd_dev *kfd);
-void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm);
-int kgd2kfd_resume_iommu(struct kfd_dev *kfd);
-int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm);
-int kgd2kfd_pre_reset(struct kfd_dev *kfd);
+void kgd2kfd_suspend(struct kfd_dev *kfd, bool suspend_proc);
+int kgd2kfd_resume(struct kfd_dev *kfd, bool resume_proc);
+void kgd2kfd_suspend_process(struct kfd_dev *kfd);
+int kgd2kfd_resume_process(struct kfd_dev *kfd);
+int kgd2kfd_pre_reset(struct kfd_dev *kfd,
+ struct amdgpu_reset_context *reset_context);
int kgd2kfd_post_reset(struct kfd_dev *kfd);
void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry);
void kgd2kfd_set_sram_ecc_flag(struct kfd_dev *kfd);
void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask);
+int kgd2kfd_check_and_lock_kfd(struct kfd_dev *kfd);
+void kgd2kfd_unlock_kfd(struct kfd_dev *kfd);
+int kgd2kfd_start_sched(struct kfd_dev *kfd, uint32_t node_id);
+int kgd2kfd_start_sched_all_nodes(struct kfd_dev *kfd);
+int kgd2kfd_stop_sched(struct kfd_dev *kfd, uint32_t node_id);
+int kgd2kfd_stop_sched_all_nodes(struct kfd_dev *kfd);
+bool kgd2kfd_compute_active(struct kfd_dev *kfd, uint32_t node_id);
+bool kgd2kfd_vmfault_fast_path(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry,
+ bool retry_fault);
+
#else
static inline int kgd2kfd_init(void)
{
@@ -381,7 +452,7 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf)
}
static inline
-bool kgd2kfd_device_init(struct kfd_dev *kfd, struct drm_device *ddev,
+bool kgd2kfd_device_init(struct kfd_dev *kfd,
const struct kgd2kfd_shared_resources *gpu_resources)
{
return false;
@@ -391,21 +462,26 @@ static inline void kgd2kfd_device_exit(struct kfd_dev *kfd)
{
}
-static inline void kgd2kfd_suspend(struct kfd_dev *kfd, bool run_pm)
+static inline void kgd2kfd_suspend(struct kfd_dev *kfd, bool suspend_proc)
{
}
-static int __maybe_unused kgd2kfd_resume_iommu(struct kfd_dev *kfd)
+static inline int kgd2kfd_resume(struct kfd_dev *kfd, bool resume_proc)
{
return 0;
}
-static inline int kgd2kfd_resume(struct kfd_dev *kfd, bool run_pm)
+static inline void kgd2kfd_suspend_process(struct kfd_dev *kfd)
+{
+}
+
+static inline int kgd2kfd_resume_process(struct kfd_dev *kfd)
{
return 0;
}
-static inline int kgd2kfd_pre_reset(struct kfd_dev *kfd)
+static inline int kgd2kfd_pre_reset(struct kfd_dev *kfd,
+ struct amdgpu_reset_context *reset_context)
{
return 0;
}
@@ -429,5 +505,46 @@ static inline
void kgd2kfd_smi_event_throttle(struct kfd_dev *kfd, uint64_t throttle_bitmask)
{
}
+
+static inline int kgd2kfd_check_and_lock_kfd(struct kfd_dev *kfd)
+{
+ return 0;
+}
+
+static inline void kgd2kfd_unlock_kfd(struct kfd_dev *kfd)
+{
+}
+
+static inline int kgd2kfd_start_sched(struct kfd_dev *kfd, uint32_t node_id)
+{
+ return 0;
+}
+
+static inline int kgd2kfd_start_sched_all_nodes(struct kfd_dev *kfd)
+{
+ return 0;
+}
+
+static inline int kgd2kfd_stop_sched(struct kfd_dev *kfd, uint32_t node_id)
+{
+ return 0;
+}
+
+static inline int kgd2kfd_stop_sched_all_nodes(struct kfd_dev *kfd)
+{
+ return 0;
+}
+
+static inline bool kgd2kfd_compute_active(struct kfd_dev *kfd, uint32_t node_id)
+{
+ return false;
+}
+
+static inline bool kgd2kfd_vmfault_fast_path(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry,
+ bool retry_fault)
+{
+ return false;
+}
+
#endif
#endif /* AMDGPU_AMDKFD_H_INCLUDED */
generated by cgit (git 2.34.1) at 2025-12-24 23:21:49 +0000