From 142262a1c02ad4d334ca1152dc4a0f6db3ef3bfc Mon Sep 17 00:00:00 2001
From: David Francis <David.Francis@amd.com>
Date: Thu, 12 Oct 2023 10:35:20 -0400
Subject: drm/amdgpu: Add EXT_COHERENT support for APU and NUMA systems
On gfx943 APU, EXT_COHERENT should give MTYPE_CC for local and
MTYPE_UC for nonlocal memory.
On NUMA systems, local memory gets the local mtype, set by an
override callback. If EXT_COHERENT is set, memory will be set as
MTYPE_UC by default, with local memory MTYPE_CC.
Add an option in the override function for this case, and
add a check to ensure it is not used on UNCACHED memory.
V2: Combined APU and NUMA code into one patch
V3: Fixed a potential nullptr in amdgpu_vm_bo_update
Signed-off-by: David Francis <David.Francis@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 17 +++++++++++------
1 file changed, 11 insertions(+), 6 deletions(-)
(limited to 'drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c')
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index f3c9f93d8899..3cd5977c0709 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -844,6 +844,7 @@ static void amdgpu_vm_tlb_seq_cb(struct dma_fence *fence,
* @immediate: immediate submission in a page fault
* @unlocked: unlocked invalidation during MM callback
* @flush_tlb: trigger tlb invalidation after update completed
+ * @allow_override: change MTYPE for local NUMA nodes
* @resv: fences we need to sync to
* @start: start of mapped range
* @last: last mapped entry
@@ -860,7 +861,7 @@ static void amdgpu_vm_tlb_seq_cb(struct dma_fence *fence,
* 0 for success, negative erro code for failure.
*/
int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm,
- bool immediate, bool unlocked, bool flush_tlb,
+ bool immediate, bool unlocked, bool flush_tlb, bool allow_override,
struct dma_resv *resv, uint64_t start, uint64_t last,
uint64_t flags, uint64_t offset, uint64_t vram_base,
struct ttm_resource *res, dma_addr_t *pages_addr,
@@ -898,6 +899,7 @@ int amdgpu_vm_update_range(struct amdgpu_device *adev, struct amdgpu_vm *vm,
params.immediate = immediate;
params.pages_addr = pages_addr;
params.unlocked = unlocked;
+ params.allow_override = allow_override;
/* Implicitly sync to command submissions in the same VM before
* unmapping. Sync to moving fences before mapping.
@@ -1073,6 +1075,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
struct ttm_resource *mem;
struct dma_fence **last_update;
bool flush_tlb = clear;
+ bool uncached;
struct dma_resv *resv;
uint64_t vram_base;
uint64_t flags;
@@ -1110,9 +1113,11 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
bo_adev = amdgpu_ttm_adev(bo->tbo.bdev);
vram_base = bo_adev->vm_manager.vram_base_offset;
+ uncached = (bo->flags & AMDGPU_GEM_CREATE_UNCACHED) != 0;
} else {
flags = 0x0;
vram_base = 0;
+ uncached = false;
}
if (clear || (bo && bo->tbo.base.resv ==
@@ -1146,7 +1151,7 @@ int amdgpu_vm_bo_update(struct amdgpu_device *adev, struct amdgpu_bo_va *bo_va,
trace_amdgpu_vm_bo_update(mapping);
r = amdgpu_vm_update_range(adev, vm, false, false, flush_tlb,
- resv, mapping->start, mapping->last,
+ !uncached, resv, mapping->start, mapping->last,
update_flags, mapping->offset,
vram_base, mem, pages_addr,
last_update);
@@ -1341,8 +1346,8 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
mapping->start < AMDGPU_GMC_HOLE_START)
init_pte_value = AMDGPU_PTE_DEFAULT_ATC;
- r = amdgpu_vm_update_range(adev, vm, false, false, true, resv,
- mapping->start, mapping->last,
+ r = amdgpu_vm_update_range(adev, vm, false, false, true, false,
+ resv, mapping->start, mapping->last,
init_pte_value, 0, 0, NULL, NULL,
&f);
amdgpu_vm_free_mapping(adev, vm, mapping, f);
@@ -2618,8 +2623,8 @@ bool amdgpu_vm_handle_fault(struct amdgpu_device *adev, u32 pasid,
goto error_unlock;
}
- r = amdgpu_vm_update_range(adev, vm, true, false, false, NULL, addr,
- addr, flags, value, 0, NULL, NULL, NULL);
+ r = amdgpu_vm_update_range(adev, vm, true, false, false, false,
+ NULL, addr, addr, flags, value, 0, NULL, NULL, NULL);
if (r)
goto error_unlock;
--
cgit v1.2.3-70-g09d2