drm/amdgpu: added support to get mGPU DRAM base
resolves issue with RAS error injection in mGPU configuration
Reviewed-by: Hawking Zhang <[email protected]>
Signed-off-by: John Clements <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 766be7f1..cef94e2 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -742,6 +742,20 @@ int amdgpu_ras_error_query(struct amdgpu_device *adev,
return 0;
}
+uint64_t get_xgmi_relative_phy_addr(struct amdgpu_device *adev, uint64_t addr)
+{
+ uint32_t df_inst_id;
+
+ if ((!adev->df.funcs) ||
+ (!adev->df.funcs->get_df_inst_id) ||
+ (!adev->df.funcs->get_dram_base_addr))
+ return addr;
+
+ df_inst_id = adev->df.funcs->get_df_inst_id(adev);
+
+ return addr + adev->df.funcs->get_dram_base_addr(adev, df_inst_id);
+}
+
/* wrapper of psp_ras_trigger_error */
int amdgpu_ras_error_inject(struct amdgpu_device *adev,
struct ras_inject_if *info)
@@ -759,6 +773,12 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev,
if (!obj)
return -EINVAL;
+ /* Calculate XGMI relative offset */
+ if (adev->gmc.xgmi.num_physical_nodes > 1) {
+ block_info.address = get_xgmi_relative_phy_addr(adev,
+ block_info.address);
+ }
+
switch (info->head.block) {
case AMDGPU_RAS_BLOCK__GFX:
if (adev->gfx.funcs->ras_error_inject)
