drm/amdgpu: Add MB_REQ_MSG_READY_TO_RESET response when VF get FLR notification.

When guest received FLR notification from host, it would
lock adapter into reset state. There will be no more
job submission and hardware access after that.

Then it should send a response to host that it has prepared
for host reset.

Signed-off-by: Jiange Zhao <Jiange.Zhao@amd.com>
Signed-off-by: Peng Ju Zhou <PengJu.Zhou@amd.com>
Reviewed-by: Emily.Deng <Emily.Deng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
Jiange Zhao 2021-03-19 10:32:04 +08:00 committed by Alex Deucher
parent 1d0e622f8d
commit 3e183e2fae
2 changed files with 4 additions and 1 deletions

View File

@ -287,6 +287,8 @@ static void xgpu_nv_mailbox_flr_work(struct work_struct *work)
amdgpu_virt_fini_data_exchange(adev);
atomic_set(&adev->in_gpu_reset, 1);
xgpu_nv_mailbox_trans_msg(adev, IDH_READY_TO_RESET, 0, 0, 0);
do {
if (xgpu_nv_mailbox_peek_msg(adev) == IDH_FLR_NOTIFICATION_CMPL)
goto flr_done;

View File

@ -37,7 +37,8 @@ enum idh_request {
IDH_REQ_GPU_RESET_ACCESS,
IDH_REQ_GPU_INIT_DATA,
IDH_LOG_VF_ERROR = 200,
IDH_LOG_VF_ERROR = 200,
IDH_READY_TO_RESET = 201,
};
enum idh_event {