diff options
Diffstat (limited to 'drivers/gpu/drm/amd/amdkfd/kfd_svm.c')
| -rw-r--r-- | drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 22 |
1 files changed, 16 insertions, 6 deletions
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index 3e2911895c74..bd3e20d981e0 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -1195,6 +1195,7 @@ svm_range_get_pte_flags(struct kfd_node *node, struct kfd_node *bo_node; uint32_t flags = prange->flags; uint32_t mapping_flags = 0; + uint32_t gc_ip_version = KFD_GC_VERSION(node); uint64_t pte_flags; bool snoop = (domain != SVM_RANGE_VRAM_DOMAIN); bool coherent = flags & (KFD_IOCTL_SVM_FLAG_COHERENT | KFD_IOCTL_SVM_FLAG_EXT_COHERENT); @@ -1204,7 +1205,7 @@ svm_range_get_pte_flags(struct kfd_node *node, if (domain == SVM_RANGE_VRAM_DOMAIN) bo_node = prange->svm_bo->node; - switch (amdgpu_ip_version(node->adev, GC_HWIP, 0)) { + switch (gc_ip_version) { case IP_VERSION(9, 4, 1): if (domain == SVM_RANGE_VRAM_DOMAIN) { if (bo_node == node) { @@ -1241,8 +1242,10 @@ svm_range_get_pte_flags(struct kfd_node *node, break; case IP_VERSION(9, 4, 3): case IP_VERSION(9, 4, 4): + case IP_VERSION(9, 5, 0): if (ext_coherent) - mtype_local = node->adev->rev_id ? AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_UC; + mtype_local = (gc_ip_version < IP_VERSION(9, 5, 0) && !node->adev->rev_id) ? + AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_CC; else mtype_local = amdgpu_mtype_local == 1 ? AMDGPU_VM_MTYPE_NC : amdgpu_mtype_local == 2 ? AMDGPU_VM_MTYPE_CC : AMDGPU_VM_MTYPE_RW; @@ -1257,9 +1260,13 @@ svm_range_get_pte_flags(struct kfd_node *node, */ else if (svm_nodes_in_same_hive(bo_node, node) && !ext_coherent) mapping_flags |= AMDGPU_VM_MTYPE_NC; - /* PCIe P2P or extended system scope coherence */ - else + /* PCIe P2P on GPUs pre-9.5.0 */ + else if (gc_ip_version < IP_VERSION(9, 5, 0) && + !svm_nodes_in_same_hive(bo_node, node)) mapping_flags |= AMDGPU_VM_MTYPE_UC; + /* Other remote memory */ + else + mapping_flags |= ext_coherent ? AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC; /* system memory accessed by the APU */ } else if (node->adev->flags & AMD_IS_APU) { /* On NUMA systems, locality is determined per-page @@ -1271,7 +1278,10 @@ svm_range_get_pte_flags(struct kfd_node *node, mapping_flags |= ext_coherent ? AMDGPU_VM_MTYPE_UC : AMDGPU_VM_MTYPE_NC; /* system memory accessed by the dGPU */ } else { - mapping_flags |= AMDGPU_VM_MTYPE_UC; + if (gc_ip_version < IP_VERSION(9, 5, 0)) + mapping_flags |= AMDGPU_VM_MTYPE_UC; + else + mapping_flags |= AMDGPU_VM_MTYPE_NC; } break; case IP_VERSION(12, 0, 0): @@ -1299,7 +1309,7 @@ svm_range_get_pte_flags(struct kfd_node *node, pte_flags = AMDGPU_PTE_VALID; pte_flags |= (domain == SVM_RANGE_VRAM_DOMAIN) ? 0 : AMDGPU_PTE_SYSTEM; pte_flags |= snoop ? AMDGPU_PTE_SNOOPED : 0; - if (KFD_GC_VERSION(node) >= IP_VERSION(12, 0, 0)) + if (gc_ip_version >= IP_VERSION(12, 0, 0)) pte_flags |= AMDGPU_PTE_IS_PTE; pte_flags |= amdgpu_gem_va_map_flags(node->adev, mapping_flags); |