[PATCH 130/159] drm/amdgpu: Placement of gart and vram in sysvm aperture

[Alex Deucher]

From: Oak Zeng <Oak.Zeng at amd.com>

If use GART for FB translation, place both vram and gart to sysvm
aperture. AGP aperture is not set up in this case because it
is not used

Signed-off-by: Oak Zeng <Oak.Zeng at amd.com>
Reviewed-by: Christian Konig <christian.koenig at amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling at amd.com>
Signed-off-by: Alex Deucher <alexander.deucher at amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 34 ++++++++++++++++++++++++-
 drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h |  1 +
 drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c   | 10 +++++---
 3 files changed, 41 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
index ab1c2b85695d..f492ff8facd3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c
@@ -162,6 +162,39 @@ void amdgpu_gmc_vram_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc,
 			mc->vram_end, mc->real_vram_size >> 20);
 }
 
+/** amdgpu_gmc_sysvm_location - place vram and gart in sysvm aperture
+ *
+ * @adev: amdgpu device structure holding all necessary information
+ * @mc: memory controller structure holding memory information
+ *
+ * This function is only used if use GART for FB translation. In such
+ * case, we use sysvm aperture (vmid0 page tables) for both vram
+ * and gart (aka system memory) access.
+ *
+ * GPUVM (and our organization of vmid0 page tables) require sysvm
+ * aperture to be placed at a location aligned with 8 times of native
+ * page size. For example, if vm_context0_cntl.page_table_block_size
+ * is 12, then native page size is 8G (2M*2^12), sysvm should start
+ * with a 64G aligned address. For simplicity, we just put sysvm at
+ * address 0. So vram start at address 0 and gart is right after vram.
+ */
+void amdgpu_gmc_sysvm_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc)
+{
+	u64 hive_vram_start = 0;
+	u64 hive_vram_end = mc->xgmi.node_segment_size * mc->xgmi.num_physical_nodes - 1;
+	mc->vram_start = mc->xgmi.node_segment_size * mc->xgmi.physical_node_id;
+	mc->vram_end = mc->vram_start + mc->xgmi.node_segment_size - 1;
+	mc->gart_start = hive_vram_end + 1;
+	mc->gart_end = mc->gart_start + mc->gart_size - 1;
+	mc->fb_start = hive_vram_start;
+	mc->fb_end = hive_vram_end;
+	dev_info(adev->dev, "VRAM: %lluM 0x%016llX - 0x%016llX (%lluM used)\n",
+			mc->mc_vram_size >> 20, mc->vram_start,
+			mc->vram_end, mc->real_vram_size >> 20);
+	dev_info(adev->dev, "GART: %lluM 0x%016llX - 0x%016llX\n",
+			mc->gart_size >> 20, mc->gart_start, mc->gart_end);
+}
+
 /**
  * amdgpu_gmc_gart_location - try to find GART location
  *
@@ -169,7 +202,6 @@ void amdgpu_gmc_vram_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc,
  * @mc: memory controller structure holding memory information
  *
  * Function will place try to place GART before or after VRAM.
- *
  * If GART size is bigger than space left then we ajust GART size.
  * Thus function will never fails.
  */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index f1a21e110610..a061a5632445 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -292,6 +292,7 @@ int amdgpu_gmc_set_pte_pde(struct amdgpu_device *adev, void *cpu_pt_addr,
 				uint64_t flags);
 uint64_t amdgpu_gmc_pd_addr(struct amdgpu_bo *bo);
 uint64_t amdgpu_gmc_agp_addr(struct ttm_buffer_object *bo);
+void amdgpu_gmc_sysvm_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc);
 void amdgpu_gmc_vram_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc,
 			      u64 base);
 void amdgpu_gmc_gart_location(struct amdgpu_device *adev,
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 563a8bc478c9..0fce8506fc0d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1257,9 +1257,13 @@ static void gmc_v9_0_vram_gtt_location(struct amdgpu_device *adev,
 
 	/* add the xgmi offset of the physical node */
 	base += adev->gmc.xgmi.physical_node_id * adev->gmc.xgmi.node_segment_size;
-	amdgpu_gmc_vram_location(adev, mc, base);
-	amdgpu_gmc_gart_location(adev, mc);
-	amdgpu_gmc_agp_location(adev, mc);
+	if (adev->gmc.xgmi.connected_to_cpu) {
+		amdgpu_gmc_sysvm_location(adev, mc);
+	} else {
+		amdgpu_gmc_vram_location(adev, mc, base);
+		amdgpu_gmc_gart_location(adev, mc);
+		amdgpu_gmc_agp_location(adev, mc);
+	}
 	/* base offset of vram pages */
 	adev->vm_manager.vram_base_offset = adev->gfxhub.funcs->get_mc_fb_offset(adev);
 
-- 
2.29.2