Messages in this thread | | | Date | Mon, 22 Nov 2010 19:46:23 +0300 | From | Dan Carpenter <> | Subject | Re: [PATCH v6] Bluetooth: btwilink driver |
| |
On Mon, Nov 22, 2010 at 06:08:45AM -0500, pavan_savoy@ti.com wrote: > +static int bt_ti_probe(struct platform_device *pdev) > +{ > + static struct ti_st *hst; > + struct hci_dev *hdev; > + int err; > + > + hst = kzalloc(sizeof(struct ti_st), GFP_KERNEL); > + if (!hst) > + return -ENOMEM; > + > + /* Expose "hciX" device to user space */ > + hdev = hci_alloc_dev(); > + if (!hdev) { > + kfree(hst); > + return -ENOMEM; > + } > + > + BT_DBG("hdev %p", hdev); > + > + hst->hdev = hdev; > + hdev->bus = HCI_UART; > + hdev->driver_data = hst; > + hdev->open = ti_st_open; > + hdev->close = ti_st_close; > + hdev->flush = NULL; > + hdev->send = ti_st_send_frame; > + hdev->destruct = ti_st_destruct; > + hdev->owner = THIS_MODULE; > + > + err = hci_register_dev(hdev); > + if (err < 0) { > + BT_ERR("Can't register HCI device error %d", err); > + hci_free_dev(hdev);
Should there be a kfree(hst); here? That's how it's done in bpa10x_probe() from drivers/bluetooth/bpa10x.c
> + return err; > + } > + > + BT_DBG(" HCI device registered (hdev %p)", hdev); ^ Quite a few places have an extra space at the start.
> + > + dev_set_drvdata(&pdev->dev, hst); > + return err; > +}
regards, dan carpenter
| |