Re: [PATCH v2 03/11] drm/amdgpu: use dedicated SPM VMID from the reserved path

[Khatri, Sunil]

On 13-10-2026 12:43 pm, Christian König wrote:

On 10/12/26 11:24, Sunil Khatri wrote:

Use the spm vmid which is dedicated for the SPM only and update
the function paths accordingly.

Signed-off-by: Sunil Khatri <sunil.kha...@amd.com>
---
  drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c | 19 ++++++++++++++-----
  1 file changed, 14 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c 
b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
index ed1efde29a9a5..3617de2368d5a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c
@@ -551,9 +551,12 @@ int amdgpu_vmid_alloc_reserved(struct amdgpu_device *adev, 
struct amdgpu_vm *vm,
                r = -ENOENT;
                goto unlock;
        }
-       /* Remove from normal round robin handling */
-       id = list_first_entry(&id_mgr->ids_lru, struct amdgpu_vmid, list);
-       list_del_init(&id->list);
+       if (!adev->vm_manager.spm_vmid) {
+               r = -ENOENT;
+               goto unlock;
+       }
+
+       id = &id_mgr->ids[adev->vm_manager.spm_vmid];
        vm->reserved_vmid[vmhub] = id;
        id_mgr->reserved_vmid = true;
        mutex_unlock(&id_mgr->lock);
@@ -577,8 +580,6 @@ void amdgpu_vmid_free_reserved(struct amdgpu_device *adev, 
struct amdgpu_vm *vm,

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;
                id_mgr->reserved_vmid = false;
        }
@@ -627,6 +628,10 @@ void amdgpu_vmid_reset_all(struct amdgpu_device *adev)
                        &adev->vm_manager.id_mgr[i];
                for_each_set_bit(j, id_mgr->vmid_mask, AMDGPU_NUM_VMID)
                        amdgpu_vmid_reset(adev, i, j);
+
+               /* out-of-pool VMID, not covered by vmid_mask */

Better write "The SPM VMID is not part of the normal kernel pool, reset it anyway in 
case it is used with kernel queues.".

Noted

+               if (adev->vm_manager.spm_vmid && AMDGPU_IS_GFXHUB(i))

If would be nice if we could rather come up with a loop over the GFXHUBs only.

let me see to that.

E.g. something like this:

for (i = AMDGPU_GFXHUB_START; i < AMDGPU_MMHUB0_START; ++i)

+                       amdgpu_vmid_reset(adev, i, adev->vm_manager.spm_vmid);
        }
  }

@@ -711,6 +716,10 @@ void amdgpu_vmid_mgr_init(struct amdgpu_device *adev)

                        amdgpu_sync_create(&id_mgr->ids[j].active);
                        list_add_tail(&id_mgr->ids[j].list, &id_mgr->ids_lru);
                }
+
+               /* out-of-pool VMID; create its sync here */

Something similar here.

Noted

+               if (adev->vm_manager.spm_vmid && AMDGPU_IS_GFXHUB(i))

Same for loop here as well.

Noted
Regards
Sunil Khatri

Regards,
Christian.

+                       
amdgpu_sync_create(&id_mgr->ids[adev->vm_manager.spm_vmid].active);
        }
  }