[PATCH 1/4] drm/amdgpu: fix GDS/GWS/OA switch handling
Christian König
ckoenig.leichtzumerken at gmail.com
Mon Dec 5 10:43:30 UTC 2022
Bas pointed out that this isn't working as expected and could cause
crashes. Fix the handling by storing the marker that a switch is needed
inside the job instead.
Reported-by: Bas Nieuwenhuizen <bas at basnieuwenhuizen.nl>
Signed-off-by: Christian König <christian.koenig at amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c | 42 +++++++++++++++----
drivers/gpu/drm/amd/amdgpu/amdgpu_job.h | 1 +
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 54 +++++++++----------------
3 files changed, 54 insertions(+), 43 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
index 03d115d2b5ed..16bd12ddba85 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
@@ -165,6 +165,26 @@ bool amdgpu_vmid_had_gpu_reset(struct amdgpu_device *adev,
atomic_read(&adev->gpu_reset_counter);
}
+/* Check if we need to switch to another set of resources */
+static bool amdgpu_vmid_gds_switch_needed(struct amdgpu_vmid *id,
+ struct amdgpu_job *job)
+{
+ return id->gds_base != job->gds_base ||
+ id->gds_size != job->gds_size ||
+ id->gws_base != job->gws_base ||
+ id->gws_size != job->gws_size ||
+ id->oa_base != job->oa_base ||
+ id->oa_size != job->oa_size;
+}
+
+/* Check if the id is compatible with the job */
+static bool amdgpu_vmid_compatible(struct amdgpu_vmid *id,
+ struct amdgpu_job *job)
+{
+ return id->pd_gpu_addr == job->vm_pd_addr &&
+ !amdgpu_vmid_gds_switch_needed(id, job);
+}
+
/**
* amdgpu_vmid_grab_idle - grab idle VMID
*
@@ -266,7 +286,7 @@ static int amdgpu_vmid_grab_reserved(struct amdgpu_vm *vm,
*id = vm->reserved_vmid[vmhub];
if ((*id)->owner != vm->immediate.fence_context ||
- (*id)->pd_gpu_addr != job->vm_pd_addr ||
+ !amdgpu_vmid_compatible(*id, job) ||
(*id)->flushed_updates < updates ||
!(*id)->last_flush ||
((*id)->last_flush->context != fence_context &&
@@ -294,7 +314,6 @@ static int amdgpu_vmid_grab_reserved(struct amdgpu_vm *vm,
if (r)
return r;
- (*id)->flushed_updates = updates;
job->vm_needs_flush = needs_flush;
return 0;
}
@@ -335,7 +354,7 @@ static int amdgpu_vmid_grab_used(struct amdgpu_vm *vm,
if ((*id)->owner != vm->immediate.fence_context)
continue;
- if ((*id)->pd_gpu_addr != job->vm_pd_addr)
+ if (!amdgpu_vmid_compatible(*id, job))
continue;
if (!(*id)->last_flush ||
@@ -356,7 +375,6 @@ static int amdgpu_vmid_grab_used(struct amdgpu_vm *vm,
if (r)
return r;
- (*id)->flushed_updates = updates;
job->vm_needs_flush |= needs_flush;
return 0;
}
@@ -410,22 +428,30 @@ int amdgpu_vmid_grab(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
if (r)
goto error;
- id->flushed_updates = amdgpu_vm_tlb_seq(vm);
job->vm_needs_flush = true;
}
list_move_tail(&id->list, &id_mgr->ids_lru);
}
- id->pd_gpu_addr = job->vm_pd_addr;
- id->owner = vm->immediate.fence_context;
-
+ job->gds_switch_needed = amdgpu_vmid_gds_switch_needed(id, job);
if (job->vm_needs_flush) {
+ id->flushed_updates = amdgpu_vm_tlb_seq(vm);
dma_fence_put(id->last_flush);
id->last_flush = NULL;
}
job->vmid = id - id_mgr->ids;
job->pasid = vm->pasid;
+
+ id->gds_base = job->gds_base;
+ id->gds_size = job->gds_size;
+ id->gws_base = job->gws_base;
+ id->gws_size = job->gws_size;
+ id->oa_base = job->oa_base;
+ id->oa_size = job->oa_size;
+ id->pd_gpu_addr = job->vm_pd_addr;
+ id->owner = vm->immediate.fence_context;
+
trace_amdgpu_vm_grab_id(vm, ring, job);
error:
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
index ab7b150e5d50..64ff833cc95e 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
@@ -54,6 +54,7 @@ struct amdgpu_job {
uint32_t preamble_status;
uint32_t preemption_status;
bool vm_needs_flush;
+ bool gds_switch_needed;
uint64_t vm_pd_addr;
unsigned vmid;
unsigned pasid;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 003aa9e47085..c2ee2e86286d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -463,25 +463,20 @@ bool amdgpu_vm_need_pipeline_sync(struct amdgpu_ring *ring,
struct amdgpu_device *adev = ring->adev;
unsigned vmhub = ring->funcs->vmhub;
struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub];
- struct amdgpu_vmid *id;
- bool gds_switch_needed;
- bool vm_flush_needed = job->vm_needs_flush || ring->has_compute_vm_bug;
if (job->vmid == 0)
return false;
- id = &id_mgr->ids[job->vmid];
- gds_switch_needed = ring->funcs->emit_gds_switch && (
- id->gds_base != job->gds_base ||
- id->gds_size != job->gds_size ||
- id->gws_base != job->gws_base ||
- id->gws_size != job->gws_size ||
- id->oa_base != job->oa_base ||
- id->oa_size != job->oa_size);
-
- if (amdgpu_vmid_had_gpu_reset(adev, id))
+
+ if (job->vm_needs_flush || ring->has_compute_vm_bug)
+ return true;
+
+ if (ring->funcs->emit_gds_switch && job->gds_switch_needed)
return true;
- return vm_flush_needed || gds_switch_needed;
+ if (amdgpu_vmid_had_gpu_reset(adev, &id_mgr->ids[job->vmid]))
+ return true;
+
+ return false;
}
/**
@@ -503,13 +498,8 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job,
unsigned vmhub = ring->funcs->vmhub;
struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub];
struct amdgpu_vmid *id = &id_mgr->ids[job->vmid];
- bool gds_switch_needed = ring->funcs->emit_gds_switch && (
- id->gds_base != job->gds_base ||
- id->gds_size != job->gds_size ||
- id->gws_base != job->gws_base ||
- id->gws_size != job->gws_size ||
- id->oa_base != job->oa_base ||
- id->oa_size != job->oa_size);
+ bool gds_switch_needed = ring->funcs->emit_gds_switch &&
+ job->gds_switch_needed;
bool vm_flush_needed = job->vm_needs_flush;
struct dma_fence *fence = NULL;
bool pasid_mapping_needed = false;
@@ -555,6 +545,14 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job,
if (pasid_mapping_needed)
amdgpu_gmc_emit_pasid_mapping(ring, job->vmid, job->pasid);
+ if (!ring->is_mes_queue && ring->funcs->emit_gds_switch &&
+ gds_switch_needed) {
+ amdgpu_ring_emit_gds_switch(ring, job->vmid, job->gds_base,
+ job->gds_size, job->gws_base,
+ job->gws_size, job->oa_base,
+ job->oa_size);
+ }
+
if (vm_flush_needed || pasid_mapping_needed) {
r = amdgpu_fence_emit(ring, &fence, NULL, 0);
if (r)
@@ -579,20 +577,6 @@ int amdgpu_vm_flush(struct amdgpu_ring *ring, struct amdgpu_job *job,
}
dma_fence_put(fence);
- if (!ring->is_mes_queue && ring->funcs->emit_gds_switch &&
- gds_switch_needed) {
- id->gds_base = job->gds_base;
- id->gds_size = job->gds_size;
- id->gws_base = job->gws_base;
- id->gws_size = job->gws_size;
- id->oa_base = job->oa_base;
- id->oa_size = job->oa_size;
- amdgpu_ring_emit_gds_switch(ring, job->vmid, job->gds_base,
- job->gds_size, job->gws_base,
- job->gws_size, job->oa_base,
- job->oa_size);
- }
-
if (ring->funcs->patch_cond_exec)
amdgpu_ring_patch_cond_exec(ring, patch_offset);
--
2.34.1
More information about the amd-gfx
mailing list