drm/amdgpu: add amdgpu_ras.c to support ras (v2)

add obj management.
add feature control.
add debugfs infrastructure.
add sysfs infrastructure.
add IH infrastructure.
add recovery infrastructure.

It is a framework. Other IPs need call amdgpu_ras_xxx function instead of
psp_ras_xxx functions.

v2: squash in warning fixes

Signed-off-by: xinhui pan <xinhui.pan@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
xinhui pan
2018-10-31 14:38:28 +08:00
committed by Alex Deucher
parent ea11421391
commit c030f2e416
5 changed files with 1475 additions and 1 deletions

View File

@@ -60,6 +60,7 @@
#include "amdgpu_pm.h"
#include "amdgpu_xgmi.h"
#include "amdgpu_ras.h"
MODULE_FIRMWARE("amdgpu/vega10_gpu_info.bin");
MODULE_FIRMWARE("amdgpu/vega12_gpu_info.bin");
@@ -1638,6 +1639,10 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
{
int i, r;
r = amdgpu_ras_init(adev);
if (r)
return r;
for (i = 0; i < adev->num_ip_blocks; i++) {
if (!adev->ip_blocks[i].status.valid)
continue;
@@ -1876,6 +1881,8 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev)
{
int i, r;
amdgpu_ras_pre_fini(adev);
if (adev->gmc.xgmi.num_physical_nodes > 1)
amdgpu_xgmi_remove_device(adev);
@@ -1945,6 +1952,8 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev)
adev->ip_blocks[i].status.late_initialized = false;
}
amdgpu_ras_fini(adev);
if (amdgpu_sriov_vf(adev))
if (amdgpu_virt_release_full_gpu(adev, false))
DRM_ERROR("failed to release exclusive mode on fini\n");