net: mana: Handle Reset Request from MANA NIC

Upon receiving the Reset Request, pause the connection and clean up
queues, wait for the specified period, then resume the NIC.
In the cleanup phase, the HWC is no longer responding, so set hwc_timeout
to zero to skip waiting on the response.

Signed-off-by: Haiyang Zhang <[e-mail]>
Link: https://patch.msgid.link/[e-mail]
Signed-off-by: Jakub Kicinski <[e-mail]>

drivers/net/ethernet/microsoft/mana/gdma_main.c

···
 #include <linux/irqdomain.h>
 
 #include <net/mana/mana.h>
+#include <net/mana/hw_channel.h>
 
 struct dentry *mana_debugfs_root;
 
···
 		mana_gd_init_pf_regs(pdev);
 	else
 		mana_gd_init_vf_regs(pdev);
+}
+
+/* Suppress logging when we set timeout to zero */
+bool mana_need_log(struct gdma_context *gc, int err)
+{
+	struct hw_channel_context *hwc;
+
+	if (err != -ETIMEDOUT)
+		return true;
+
+	if (!gc)
+		return true;
+
+	hwc = gc->hwc.driver_data;
+	if (hwc && hwc->hwc_timeout == 0)
+		return false;
+
+	return true;
 }
 
 static int mana_gd_query_max_resources(struct pci_dev *pdev)
···
 
 	err = mana_gd_send_request(gc, sizeof(req), &req, sizeof(resp), &resp);
 	if (err || resp.hdr.status) {
-		dev_err(gc->dev, "Failed to disable queue: %d, 0x%x\n", err,
-			resp.hdr.status);
+		if (mana_need_log(gc, err))
+			dev_err(gc->dev, "Failed to disable queue: %d, 0x%x\n", err,
+				resp.hdr.status);
 		return err ? err : -EPROTO;
 	}
 
···
 
 #define MANA_SERVICE_PERIOD 10
 
-struct mana_serv_work {
-	struct work_struct serv_work;
-	struct pci_dev *pdev;
-};
-
-static void mana_serv_func(struct work_struct *w)
+static void mana_serv_fpga(struct pci_dev *pdev)
 {
-	struct mana_serv_work *mns_wk;
 	struct pci_bus *bus, *parent;
-	struct pci_dev *pdev;
-
-	mns_wk = container_of(w, struct mana_serv_work, serv_work);
-	pdev = mns_wk->pdev;
 
 	pci_lock_rescan_remove();
-
-	if (!pdev)
-		goto out;
 
 	bus = pdev->bus;
 	if (!bus) {
···
 
 out:
 	pci_unlock_rescan_remove();
+}
 
+static void mana_serv_reset(struct pci_dev *pdev)
+{
+	struct gdma_context *gc = pci_get_drvdata(pdev);
+	struct hw_channel_context *hwc;
+
+	if (!gc) {
+		dev_err(&pdev->dev, "MANA service: no GC\n");
+		return;
+	}
+
+	hwc = gc->hwc.driver_data;
+	if (!hwc) {
+		dev_err(&pdev->dev, "MANA service: no HWC\n");
+		goto out;
+	}
+
+	/* HWC is not responding in this case, so don't wait */
+	hwc->hwc_timeout = 0;
+
+	dev_info(&pdev->dev, "MANA reset cycle start\n");
+
+	mana_gd_suspend(pdev, PMSG_SUSPEND);
+
+	msleep(MANA_SERVICE_PERIOD * 1000);
+
+	mana_gd_resume(pdev);
+
+	dev_info(&pdev->dev, "MANA reset cycle completed\n");
+
+out:
+	gc->in_service = false;
+}
+
+struct mana_serv_work {
+	struct work_struct serv_work;
+	struct pci_dev *pdev;
+	enum gdma_eqe_type type;
+};
+
+static void mana_serv_func(struct work_struct *w)
+{
+	struct mana_serv_work *mns_wk;
+	struct pci_dev *pdev;
+
+	mns_wk = container_of(w, struct mana_serv_work, serv_work);
+	pdev = mns_wk->pdev;
+
+	if (!pdev)
+		goto out;
+
+	switch (mns_wk->type) {
+	case GDMA_EQE_HWC_FPGA_RECONFIG:
+		mana_serv_fpga(pdev);
+		break;
+
+	case GDMA_EQE_HWC_RESET_REQUEST:
+		mana_serv_reset(pdev);
+		break;
+
+	default:
+		dev_err(&pdev->dev, "MANA service: unknown type %d\n",
+			mns_wk->type);
+		break;
+	}
+
+out:
 	pci_dev_put(pdev);
 	kfree(mns_wk);
 	module_put(THIS_MODULE);
···
 		break;
 
 	case GDMA_EQE_HWC_FPGA_RECONFIG:
+	case GDMA_EQE_HWC_RESET_REQUEST:
 		dev_info(gc->dev, "Recv MANA service type:%d\n", type);
 
 		if (gc->in_service) {
···
 		dev_info(gc->dev, "Start MANA service type:%d\n", type);
 		gc->in_service = true;
 		mns_wk->pdev = to_pci_dev(gc->dev);
+		mns_wk->type = type;
 		pci_dev_get(mns_wk->pdev);
 		INIT_WORK(&mns_wk->serv_work, mana_serv_func);
 		schedule_work(&mns_wk->serv_work);
···
 
 	err = mana_gd_send_request(gc, sizeof(req), &req, sizeof(resp), &resp);
 	if (err) {
-		dev_err(dev, "test_eq failed: %d\n", err);
+		if (mana_need_log(gc, err))
+			dev_err(dev, "test_eq failed: %d\n", err);
 		goto out;
 	}
 
···
 
 	if (flush_evenets) {
 		err = mana_gd_test_eq(gc, queue);
-		if (err)
+		if (err && mana_need_log(gc, err))
 			dev_warn(gc->dev, "Failed to flush EQ: %d\n", err);
 	}
 
···
 
 	err = mana_gd_send_request(gc, sizeof(req), &req, sizeof(resp), &resp);
 	if (err || resp.hdr.status) {
-		dev_err(gc->dev, "Failed to destroy DMA region: %d, 0x%x\n",
-			err, resp.hdr.status);
+		if (mana_need_log(gc, err))
+			dev_err(gc->dev, "Failed to destroy DMA region: %d, 0x%x\n",
+				err, resp.hdr.status);
 		return -EPROTO;
 	}
 
···
 
 	err = mana_gd_send_request(gc, sizeof(req), &req, sizeof(resp), &resp);
 	if (err || resp.hdr.status) {
-		dev_err(gc->dev, "Failed to deregister device: %d, 0x%x\n",
-			err, resp.hdr.status);
+		if (mana_need_log(gc, err))
+			dev_err(gc->dev, "Failed to deregister device: %d, 0x%x\n",
+				err, resp.hdr.status);
 		if (!err)
 			err = -EPROTO;
 	}
···
 }
 
 /* The 'state' parameter is not used. */
-static int mana_gd_suspend(struct pci_dev *pdev, pm_message_t state)
+int mana_gd_suspend(struct pci_dev *pdev, pm_message_t state)
 {
 	struct gdma_context *gc = pci_get_drvdata(pdev);
 
···
  * fail -- if this happens, it's safer to just report an error than try to undo
  * what has been done.
  */
-static int mana_gd_resume(struct pci_dev *pdev)
+int mana_gd_resume(struct pci_dev *pdev)
 {
 	struct gdma_context *gc = pci_get_drvdata(pdev);
 	int err;

drivers/net/ethernet/microsoft/mana/hw_channel.c

···
 
 	if (!wait_for_completion_timeout(&ctx->comp_event,
 					 (msecs_to_jiffies(hwc->hwc_timeout)))) {
-		dev_err(hwc->dev, "HWC: Request timed out!\n");
+		if (hwc->hwc_timeout != 0)
+			dev_err(hwc->dev, "HWC: Request timed out!\n");
+
 		err = -ETIMEDOUT;
 		goto out;
 	}

drivers/net/ethernet/microsoft/mana/mana_en.c

···
 	.read   = mana_dbg_q_read,
 };
 
+static bool mana_en_need_log(struct mana_port_context *apc, int err)
+{
+	if (apc && apc->ac && apc->ac->gdma_dev &&
+	    apc->ac->gdma_dev->gdma_context)
+		return mana_need_log(apc->ac->gdma_dev->gdma_context, err);
+	else
+		return true;
+}
+
 /* Microsoft Azure Network Adapter (MANA) functions */
 
 static int mana_open(struct net_device *ndev)
···
 		if (err == -EOPNOTSUPP)
 			return err;
 
-		if (req->req.msg_type != MANA_QUERY_PHY_STAT)
+		if (req->req.msg_type != MANA_QUERY_PHY_STAT &&
+		    mana_need_log(gc, err))
 			dev_err(dev, "Failed to send mana message: %d, 0x%x\n",
 				err, resp->status);
 		return err ? err : -EPROTO;
···
 	err = mana_send_request(apc->ac, &req, sizeof(req), &resp,
 				sizeof(resp));
 	if (err) {
-		netdev_err(apc->ndev, "Failed to unregister hw vPort: %d\n",
-			   err);
+		if (mana_en_need_log(apc, err))
+			netdev_err(apc->ndev, "Failed to unregister hw vPort: %d\n",
+				   err);
+
 		return;
 	}
 
···
 	err = mana_send_request(apc->ac, &req, sizeof(req), &resp,
 				sizeof(resp));
 	if (err) {
-		netdev_err(apc->ndev, "Failed to unregister filter: %d\n",
-			   err);
+		if (mana_en_need_log(apc, err))
+			netdev_err(apc->ndev, "Failed to unregister filter: %d\n",
+				   err);
+
 		return;
 	}
 
···
 	err = mana_send_request(apc->ac, req, req_buf_size, &resp,
 				sizeof(resp));
 	if (err) {
-		netdev_err(ndev, "Failed to configure vPort RX: %d\n", err);
+		if (mana_en_need_log(apc, err))
+			netdev_err(ndev, "Failed to configure vPort RX: %d\n", err);
+
 		goto out;
 	}
 
···
 	err = mana_send_request(apc->ac, &req, sizeof(req), &resp,
 				sizeof(resp));
 	if (err) {
-		netdev_err(ndev, "Failed to destroy WQ object: %d\n", err);
+		if (mana_en_need_log(apc, err))
+			netdev_err(ndev, "Failed to destroy WQ object: %d\n", err);
+
 		return;
 	}
 
···
 
 	apc->rss_state = TRI_STATE_FALSE;
 	err = mana_config_rss(apc, TRI_STATE_FALSE, false, false);
-	if (err) {
+	if (err && mana_en_need_log(apc, err))
 		netdev_err(ndev, "Failed to disable vPort: %d\n", err);
-		return err;
-	}
 
+	/* Even in err case, still need to cleanup the vPort */
 	mana_destroy_vport(apc);
 
 	return 0;

include/net/mana/gdma.h

···
 	GDMA_EQE_HWC_FPGA_RECONFIG	= 132,
 	GDMA_EQE_HWC_SOC_RECONFIG_DATA	= 133,
 	GDMA_EQE_HWC_SOC_SERVICE	= 134,
+	GDMA_EQE_HWC_RESET_REQUEST	= 135,
 	GDMA_EQE_RNIC_QP_FATAL		= 176,
 };
 
···
 /* Driver supports dynamic MSI-X vector allocation */
 #define GDMA_DRV_CAP_FLAG_1_DYNAMIC_IRQ_ALLOC_SUPPORT BIT(13)
 
+/* Driver can self reset on EQE notification */
+#define GDMA_DRV_CAP_FLAG_1_SELF_RESET_ON_EQE BIT(14)
+
 /* Driver can self reset on FPGA Reconfig EQE notification */
 #define GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE BIT(17)
 
···
 	 GDMA_DRV_CAP_FLAG_1_VARIABLE_INDIRECTION_TABLE_SUPPORT | \
 	 GDMA_DRV_CAP_FLAG_1_DEV_LIST_HOLES_SUP | \
 	 GDMA_DRV_CAP_FLAG_1_DYNAMIC_IRQ_ALLOC_SUPPORT | \
+	 GDMA_DRV_CAP_FLAG_1_SELF_RESET_ON_EQE | \
 	 GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE)
 
 #define GDMA_DRV_CAP_FLAGS2 0
···
 void mana_unregister_debugfs(void);
 
 int mana_rdma_service_event(struct gdma_context *gc, enum gdma_service_type event);
+
+int mana_gd_suspend(struct pci_dev *pdev, pm_message_t state);
+int mana_gd_resume(struct pci_dev *pdev);
+
+bool mana_need_log(struct gdma_context *gc, int err);
 
 #endif /* _GDMA_H */