diff options
| author | Xiaogang Chen <xiaogang.chen@amd.com> | 2026-01-13 20:45:14 -0600 |
|---|---|---|
| committer | Alex Deucher <alexander.deucher@amd.com> | 2026-01-14 14:28:49 -0500 |
| commit | 6cca686dfce79bf5bd5c1a680ed38a9f20669e39 (patch) | |
| tree | 85f8807d4c0382d5a0e232d81ab87b4a64570c43 /drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | |
| parent | d81e52fc61fb98dc98f9fdb166ab21f502cb701c (diff) | |
| download | linux-next-6cca686dfce79bf5bd5c1a680ed38a9f20669e39.tar.gz linux-next-6cca686dfce79bf5bd5c1a680ed38a9f20669e39.zip | |
drm/amdkfd: kfd driver supports hot unplug/replug amdgpu devices
This patch allows kfd driver function correctly when AMD gpu devices got
unplug/replug at run time.
When an AMD gpu device got unplug kfd driver gracefully terminates existing
kfd processes after stops all queues by sending SIGBUS to user process. After
that user space can still use remaining AMD gpu devices. When all AMD gpu
devices at system got removed kfd driver will not response new requests.
Unplugged AMD gpu devices can be re-plugged. kfd driver will use added devices
to function as usual.
The purpose of this patch is having kfd driver behavior as expected during and
after AMD gpu devices unplug/replug at run time.
Signed-off-by: Xiaogang Chen <Xiaogang.Chen@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h')
| -rw-r--r-- | drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h | 11 |
1 files changed, 11 insertions, 0 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index da4575676335..eba9556ece9a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -158,6 +158,7 @@ struct amdkfd_process_info { int amdgpu_amdkfd_init(void); void amdgpu_amdkfd_fini(void); +void amdgpu_amdkfd_teardown_processes(struct amdgpu_device *adev); void amdgpu_amdkfd_suspend(struct amdgpu_device *adev, bool suspend_proc); int amdgpu_amdkfd_resume(struct amdgpu_device *adev, bool resume_proc); @@ -438,6 +439,8 @@ 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); +void kgd2kfd_lock_kfd(void); +void kgd2kfd_teardown_processes(struct amdgpu_device *adev); #else static inline int kgd2kfd_init(void) @@ -550,5 +553,13 @@ static inline bool kgd2kfd_vmfault_fast_path(struct amdgpu_device *adev, struct return false; } +static inline void kgd2kfd_lock_kfd(void) +{ +} + +static inline void kgd2kfd_teardown_processes(struct amdgpu_device *adev) +{ +} + #endif #endif /* AMDGPU_AMDKFD_H_INCLUDED */ |