git/linux-net
1
0
Fork 0
mirror of https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git/ synced 2026-04-18 06:33:43 -04:00
Code Issues Packages Projects Releases Wiki Activity

drm/amdgpu: Optimize VM invalidation engine allocation and synchronize GPU TLB flush

Browse Source 
- Modify the VM invalidation engine allocation logic to handle SDMA page rings.
  SDMA page rings now share the VM invalidation engine with SDMA gfx rings instead of
  allocating a separate engine. This change ensures efficient resource management and
  avoids the issue of insufficient VM invalidation engines.

- Add synchronization for GPU TLB flush operations in gmc_v9_0.c.
  Use spin_lock and spin_unlock to ensure thread safety and prevent race conditions
  during TLB flush operations. This improves the stability and reliability of the driver,
  especially in multi-threaded environments.

 v2: replace the sdma ring check with a function `amdgpu_sdma_is_page_queue`
 to check if a ring is an SDMA page queue.(Lijo)

 v3: Add GC version check, only enabled on GC9.4.3/9.4.4/9.5.0
 v4: Fix code style and add more detailed description (Christian)
 v5: Remove dependency on vm_inv_eng loop order, explicitly lookup shared inv_eng(Christian/Lijo)
 v6: Added search shared ring function amdgpu_sdma_get_shared_ring (Lijo)

Suggested-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Jesse Zhang <jesse.zhang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
Jesse.zhang@amd.com
2025-02-25 15:25:00 +08:00
committed by Alex Deucher
parent ea6dd40caf
commit b09cdeb4d3
3 changed files with 57 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
View File

@@ -573,6 +573,7 @@ int amdgpu_gmc_allocate_vm_inv_eng(struct amdgpu_device *adev)
unsigned vm_inv_engs[AMDGPU_MAX_VMHUBS] = {0};
unsigned i;
unsigned vmhub, inv_eng;
struct amdgpu_ring *shared_ring;
/* init the vm inv eng for all vmhubs */
for_each_set_bit(i, adev->vmhubs_mask, AMDGPU_MAX_VMHUBS) {
@@ -595,6 +596,10 @@ int amdgpu_gmc_allocate_vm_inv_eng(struct amdgpu_device *adev)
ring == &adev->cper.ring_buf)
continue;
/* Skip if the ring is a shared ring */
if (amdgpu_sdma_is_shared_inv_eng(adev, ring))
continue;
inv_eng = ffs(vm_inv_engs[vmhub]);
if (!inv_eng) {
dev_err(adev->dev, "no VM inv eng for ring %s\n",
@@ -607,6 +612,21 @@ int amdgpu_gmc_allocate_vm_inv_eng(struct amdgpu_device *adev)
dev_info(adev->dev, "ring %s uses VM inv eng %u on hub %u\n",
ring->name, ring->vm_inv_eng, ring->vm_hub);
/* SDMA has a special packet which allows it to use the same
* invalidation engine for all the rings in one instance.
* Therefore, we do not allocate a separate VM invalidation engine
* for SDMA page rings. Instead, they share the VM invalidation
* engine with the SDMA gfx ring. This change ensures efficient
* resource management and avoids the issue of insufficient VM
* invalidation engines.
*/
shared_ring = amdgpu_sdma_get_shared_ring(adev, ring);
if (shared_ring) {
shared_ring->vm_inv_eng = ring->vm_inv_eng;
dev_info(adev->dev, "ring %s shares VM invalidation engine %u with ring %s on hub %u\n",
ring->name, ring->vm_inv_eng, shared_ring->name, ring->vm_hub);
continue;
}
}
return 0;