For the possible failure of the platform_get_irq(), the returned irq
could be error number and will finally cause the failure of the
request_irq().
Consider that platform_get_irq() can now in certain cases return
-EPROBE_DEFER, and the consequences of letting request_irq()
effectively convert that into -EINVAL, even at probe time rather than
later on. So it might be better to check just now.
Fixes: b1201e44f50b ("can: xilinx CAN controller support")
Link: https://lore.kernel.org/all/20211224021324.1447494-1-jiasheng@iscas.ac.cn
Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
        spin_lock_init(&priv->tx_lock);
 
        /* Get IRQ for the device */
-       ndev->irq = platform_get_irq(pdev, 0);
+       ret = platform_get_irq(pdev, 0);
+       if (ret < 0)
+               goto err_free;
+
+       ndev->irq = ret;
+
        ndev->flags |= IFF_ECHO;        /* We support local echo */
 
        platform_set_drvdata(pdev, ndev);