Merge tag 'drm-next-2022-12-23' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Dave Airlie:
 "Holiday fixes!

  Two batches from amd, and one group of i915 changes.

  amdgpu:
   - Spelling fix
   - BO pin fix
   - Properly handle polaris 10/11 overlap asics
   - GMC9 fix
   - SR-IOV suspend fix
   - DCN 3.1.4 fix
   - KFD userptr locking fix
   - SMU13.x fixes
   - GDS/GWS/OA handling fix
   - Reserved VMID handling fixes
   - FRU EEPROM fix
   - BO validation fixes
   - Avoid large variable on the stack
   - S0ix fixes
   - SMU 13.x fixes
   - VCN fix
   - Add missing fence reference

  amdkfd:
   - Fix init vm error handling
   - Fix double release of compute pasid

  i915
   - Documentation fixes
   - OA-perf related fix
   - VLV/CHV HDMI/DP audio fix
   - Display DDI/Transcoder fix
   - Migrate fixes"

* tag 'drm-next-2022-12-23' of git://anongit.freedesktop.org/drm/drm: (39 commits)
  drm/amdgpu: grab extra fence reference for drm_sched_job_add_dependency
  drm/amdgpu: enable VCN DPG for GC IP v11.0.4
  drm/amdgpu: skip mes self test after s0i3 resume for MES IP v11.0
  drm/amd/pm: correct the fan speed retrieving in PWM for some SMU13 asics
  drm/amd/pm: bump SMU13.0.0 driver_if header to version 0x34
  drm/amdgpu: skip MES for S0ix as well since it's part of GFX
  drm/amd/pm: avoid large variable on kernel stack
  drm/amdkfd: Fix double release compute pasid
  drm/amdkfd: Fix kfd_process_device_init_vm error handling
  drm/amd/pm: update SMU13.0.0 reported maximum shader clock
  drm/amd/pm: correct SMU13.0.0 pstate profiling clock settings
  drm/amd/pm: enable GPO dynamic control support for SMU13.0.7
  drm/amd/pm: enable GPO dynamic control support for SMU13.0.0
  drm/amdgpu: revert "generally allow over-commit during BO allocation"
  drm/amdgpu: Remove unnecessary domain argument
  drm/amdgpu: Fix size validation for non-exclusive domains (v4)
  drm/amdgpu: Check if fru_addr is not NULL (v2)
  drm/i915/ttm: consider CCS for backup objects
  drm/i915/migrate: fix corner case in CCS aux copying
  drm/amdgpu: rework reserved VMID handling
  ...

1 files changed, 56 insertions, 32 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
index 2a9a2593dc18..fcb711a11a5b 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
  *
@@ -258,14 +278,15 @@ static int amdgpu_vmid_grab_reserved(struct amdgpu_vm *vm,
 {
 	struct amdgpu_device *adev = ring->adev;
 	unsigned vmhub = ring->funcs->vmhub;
+	struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub];
 	uint64_t fence_context = adev->fence_context + ring->idx;
 	bool needs_flush = vm->use_cpu_for_update;
 	uint64_t updates = amdgpu_vm_tlb_seq(vm);
 	int r;
 
-	*id = vm->reserved_vmid[vmhub];
+	*id = id_mgr->reserved;
 	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 &&
@@ -294,8 +315,8 @@ static int amdgpu_vmid_grab_reserved(struct amdgpu_vm *vm,
 	if (r)
 		return r;
 
-	(*id)->flushed_updates = updates;
 	job->vm_needs_flush = needs_flush;
+	job->spm_update_needed = true;
 	return 0;
 }
 
@@ -333,7 +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 ||
@@ -355,7 +376,6 @@ 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;
 	}
@@ -408,22 +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:
@@ -435,31 +463,27 @@ int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev,
 			       struct amdgpu_vm *vm,
 			       unsigned vmhub)
 {
-	struct amdgpu_vmid_mgr *id_mgr;
-	struct amdgpu_vmid *idle;
-	int r = 0;
+	struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub];
 
-	id_mgr = &adev->vm_manager.id_mgr[vmhub];
 	mutex_lock(&id_mgr->lock);
 	if (vm->reserved_vmid[vmhub])
 		goto unlock;
-	if (atomic_inc_return(&id_mgr->reserved_vmid_num) >
-	    AMDGPU_VM_MAX_RESERVED_VMID) {
-		DRM_ERROR("Over limitation of reserved vmid\n");
-		atomic_dec(&id_mgr->reserved_vmid_num);
-		r = -EINVAL;
-		goto unlock;
+
+	++id_mgr->reserved_use_count;
+	if (!id_mgr->reserved) {
+		struct amdgpu_vmid *id;
+
+		id = list_first_entry(&id_mgr->ids_lru, struct amdgpu_vmid,
+				      list);
+		/* Remove from normal round robin handling */
+		list_del_init(&id->list);
+		id_mgr->reserved = id;
 	}
-	/* Select the first entry VMID */
-	idle = list_first_entry(&id_mgr->ids_lru, struct amdgpu_vmid, list);
-	list_del_init(&idle->list);
-	vm->reserved_vmid[vmhub] = idle;
-	mutex_unlock(&id_mgr->lock);
+	vm->reserved_vmid[vmhub] = true;
 
-	return 0;
 unlock:
 	mutex_unlock(&id_mgr->lock);
-	return r;
+	return 0;
 }
 
 void amdgpu_vmid_free_reserved(struct amdgpu_device *adev,
@@ -469,12 +493,12 @@ void amdgpu_vmid_free_reserved(struct amdgpu_device *adev,
 	struct amdgpu_vmid_mgr *id_mgr = &adev->vm_manager.id_mgr[vmhub];
 
 	mutex_lock(&id_mgr->lock);
-	if (vm->reserved_vmid[vmhub]) {
-		list_add(&vm->reserved_vmid[vmhub]->list,
-			&id_mgr->ids_lru);
-		vm->reserved_vmid[vmhub] = NULL;
-		atomic_dec(&id_mgr->reserved_vmid_num);
+	if (vm->reserved_vmid[vmhub] &&
+	    !--id_mgr->reserved_use_count) {
+		/* give the reserved ID back to normal round robin */
+		list_add(&id_mgr->reserved->list, &id_mgr->ids_lru);
 	}
+	vm->reserved_vmid[vmhub] = false;
 	mutex_unlock(&id_mgr->lock);
 }
 
@@ -541,7 +565,7 @@ void amdgpu_vmid_mgr_init(struct amdgpu_device *adev)
 
 		mutex_init(&id_mgr->lock);
 		INIT_LIST_HEAD(&id_mgr->ids_lru);
-		atomic_set(&id_mgr->reserved_vmid_num, 0);
+		id_mgr->reserved_use_count = 0;
 
 		/* manage only VMIDs not used by KFD */
 		id_mgr->num_ids = adev->vm_manager.first_kfd_vmid;