The current approach breaks S3/S4 as asic reset is needed for them.
And putting SMU out of service(via SMU_MSG_PrepareMp1ForUnload) will make
that(asic reset) failed. Considering with current designs, there is
actually also asic reset involved on driver reloading. That can make
asic back to a clean state. So, the SMU_MSG_PrepareMp1ForUnload operation
will be not so necessary. Thus we will just drop the SMU_MSG_PrepareMp1ForUnload
operation. We may revise the whole driver reloading sequences when there
is a better design.
Fixes: 72aeb6ee0c78 ("drm/amd/pm: fix driver reload SMC firmware fail issue for smu13")
Signed-off-by: Evan Quan <evan.quan@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
        switch (adev->ip_versions[MP1_HWIP][0]) {
        case IP_VERSION(13, 0, 0):
        case IP_VERSION(13, 0, 7):
-               if (!(adev->in_runpm || amdgpu_in_reset(adev))) {
-                       ret = smu_set_mp1_state(smu, PP_MP1_STATE_UNLOAD);
-                       if (ret) {
-                               dev_err(adev->dev, "Fail set mp1 state to UNLOAD!\n");
-                               return ret;
-                       }
-               }
                return 0;
        default:
                break;