@@ -176,6 +176,9 @@ int hns_roce_cmd_mbox(struct hns_roce_dev *hr_dev, u64 in_param, u64 out_param,
unsigned long in_modifier, u8 op_modifier, u16 op,
unsigned long timeout)
{
+ if (hr_dev->is_reset)
+ return 0;
+
if (hr_dev->cmd.use_events)
return hns_roce_cmd_mbox_wait(hr_dev, in_param, out_param,
in_modifier, op_modifier, op,
@@ -774,6 +774,8 @@ struct hns_roce_dev {
const char *irq_names[HNS_ROCE_MAX_IRQ_NUM];
spinlock_t sm_lock;
spinlock_t bt_cmd_lock;
+ bool active;
+ bool is_reset;
struct hns_roce_ib_iboe iboe;
struct list_head pgdir_list;
@@ -768,6 +768,9 @@ static int hns_roce_cmq_send(struct hns_roce_dev *hr_dev,
int ret = 0;
int ntc;
+ if (hr_dev->is_reset)
+ return 0;
+
spin_lock_bh(&csq->lock);
if (num > hns_roce_cmq_space(csq)) {
@@ -4790,14 +4793,87 @@ static void hns_roce_hw_v2_uninit_instance(struct hnae3_handle *handle,
{
struct hns_roce_dev *hr_dev = (struct hns_roce_dev *)handle->priv;
+ if (!hr_dev)
+ return;
+
hns_roce_exit(hr_dev);
kfree(hr_dev->priv);
ib_dealloc_device(&hr_dev->ib_dev);
}
+static int hns_roce_hw_v2_reset_notify_down(struct hnae3_handle *handle)
+{
+ struct hns_roce_dev *hr_dev = (struct hns_roce_dev *)handle->priv;
+ struct ib_event event;
+
+ if (!hr_dev) {
+ dev_err(&handle->pdev->dev,
+ "Input parameter handle->priv is NULL!\n");
+ return -EINVAL;
+ }
+
+ hr_dev->active = false;
+ hr_dev->is_reset = true;
+
+ event.event = IB_EVENT_DEVICE_FATAL;
+ event.device = &hr_dev->ib_dev;
+ event.element.port_num = 1;
+ ib_dispatch_event(&event);
+
+ return 0;
+}
+
+static int hns_roce_hw_v2_reset_notify_init(struct hnae3_handle *handle)
+{
+ int ret;
+
+ ret = hns_roce_hw_v2_init_instance(handle);
+ if (ret) {
+ /* when reset notify type is HNAE3_INIT_CLIENT In reset notify
+ * callback function, RoCE Engine reinitialize. If RoCE reinit
+ * failed, we should inform NIC driver.
+ */
+ handle->priv = NULL;
+ dev_err(&handle->pdev->dev,
+ "In reset process RoCE reinit failed %d.\n", ret);
+ }
+
+ return ret;
+}
+
+static int hns_roce_hw_v2_reset_notify_uninit(struct hnae3_handle *handle)
+{
+ msleep(100);
+ hns_roce_hw_v2_uninit_instance(handle, false);
+ return 0;
+}
+
+static int hns_roce_hw_v2_reset_notify(struct hnae3_handle *handle,
+ enum hnae3_reset_notify_type type)
+{
+ int ret = 0;
+
+ switch (type) {
+ case HNAE3_DOWN_CLIENT:
+ ret = hns_roce_hw_v2_reset_notify_down(handle);
+ break;
+ case HNAE3_INIT_CLIENT:
+ ret = hns_roce_hw_v2_reset_notify_init(handle);
+ break;
+ case HNAE3_UNINIT_CLIENT:
+ ret = hns_roce_hw_v2_reset_notify_uninit(handle);
+ break;
+ default:
+ break;
+ }
+
+ return ret;
+}
+
static const struct hnae3_client_ops hns_roce_hw_v2_ops = {
.init_instance = hns_roce_hw_v2_init_instance,
.uninit_instance = hns_roce_hw_v2_uninit_instance,
+ .reset_notify = hns_roce_hw_v2_reset_notify,
};
static struct hnae3_client hns_roce_hw_v2_client = {
@@ -332,6 +332,9 @@ static struct ib_ucontext *hns_roce_alloc_ucontext(struct ib_device *ib_dev,
struct hns_roce_ib_alloc_ucontext_resp resp = {};
struct hns_roce_dev *hr_dev = to_hr_dev(ib_dev);
+ if (!hr_dev->active)
+ return ERR_PTR(-EAGAIN);
+
resp.qp_tab_size = hr_dev->caps.num_qps;
context = kmalloc(sizeof(*context), GFP_KERNEL);
@@ -425,6 +428,7 @@ static void hns_roce_unregister_device(struct hns_roce_dev *hr_dev)
{
struct hns_roce_ib_iboe *iboe = &hr_dev->iboe;
+ hr_dev->active = false;
unregister_netdevice_notifier(&iboe->nb);
ib_unregister_device(&hr_dev->ib_dev);
}
@@ -536,6 +540,7 @@ static int hns_roce_register_device(struct hns_roce_dev *hr_dev)
goto error_failed_setup_mtu_mac;
}
+ hr_dev->active = true;
return 0;
error_failed_setup_mtu_mac:
@@ -728,6 +733,7 @@ int hns_roce_init(struct hns_roce_dev *hr_dev)
return ret;
}
}
+ hr_dev->is_reset = false;
if (hr_dev->hw->cmq_init) {
ret = hr_dev->hw->cmq_init(hr_dev);
@@ -827,6 +833,7 @@ int hns_roce_init(struct hns_roce_dev *hr_dev)
void hns_roce_exit(struct hns_roce_dev *hr_dev)
{
hns_roce_unregister_device(hr_dev);
+
if (hr_dev->hw->hw_exit)
hr_dev->hw->hw_exit(hr_dev);
hns_roce_cleanup_bitmap(hr_dev);
This patch added reset process for RoCE in hip08. Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com> -- v2->v3: no change. v1->v2: 1.Delete handle->priv = NULL in hns_roce_hw_v2_uninit_instance. 2.Add hns_roce_hw_v2_reset_notify_init callback function, When RoCE reinit failed in this function, inform NIC driver. The related link of Jason's commets: https://www.spinics.net/lists/linux-rdma/msg65009.html --- drivers/infiniband/hw/hns/hns_roce_cmd.c | 3 ++ drivers/infiniband/hw/hns/hns_roce_device.h | 2 + drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 76 +++++++++++++++++++++++++++++ drivers/infiniband/hw/hns/hns_roce_main.c | 7 +++ 4 files changed, 88 insertions(+)