On Sun, Dec 27, 2020 at 2:21 PM Roderick Colenbrander
<roderick@xxxxxxxxxx> wrote:
>
> Hi Barnabás,
>
> Thanks for you great feedback.
>
> On Sun, Dec 27, 2020 at 9:06 AM Barnabás Pőcze <pobrn@xxxxxxxxxxxxxx> wrote:
> >
> > Hi
> >
> >
> > 2020. december 19., szombat 7:23 keltezéssel, Roderick Colenbrander írta:
> >
> > > diff --git a/drivers/hid/hid-playstation.c b/drivers/hid/hid-playstation.c
> > > index a55375ac79a9..2f989da906f3 100644
> > > --- a/drivers/hid/hid-playstation.c
> > > +++ b/drivers/hid/hid-playstation.c
> > > [...]
> > > +static ssize_t ps_show_firmware_version(struct device *dev,
> > > + struct device_attribute
> > > + *attr, char *buf)
> > > +{
> > > + struct hid_device *hdev = to_hid_device(dev);
> > > + struct ps_device *ps_dev = hid_get_drvdata(hdev);
> > > +
> > > + return snprintf(buf, PAGE_SIZE, "0x%08x\n", ps_dev->fw_version);
> >
> > `sysfs_emit()` is preferred over *printf().
>
> Thanks, I wasn't aware of its existence. Looks like it was added
> recently, but yeah it is a lot nicer.
>
> >
> > > +}
> > > +
> > > +static DEVICE_ATTR(firmware_version, 0444, ps_show_firmware_version, NULL);
> > > +
> > > +static ssize_t ps_show_hardware_version(struct device *dev,
> > > + struct device_attribute
> > > + *attr, char *buf)
> > > +{
> > > + struct hid_device *hdev = to_hid_device(dev);
> > > + struct ps_device *ps_dev = hid_get_drvdata(hdev);
> > > +
> > > + return snprintf(buf, PAGE_SIZE, "0x%08x\n", ps_dev->hw_version);
> >
> > Same here.
> >
> >
> > > +}
> > > [...]
> > > +static int dualsense_get_firmware_info(struct dualsense *ds)
> > > +{
> > > + uint8_t *buf;
> > > + int ret = 0;
> >
> > Is there any reason it needs to be initialized?
> >
> >
> > > +
> > > + buf = kzalloc(DS_FEATURE_REPORT_FIRMWARE_INFO_SIZE, GFP_KERNEL);
> > > + if (!buf)
> > > + return -ENOMEM;
> > > +
> > > + ret = hid_hw_raw_request(ds->base.hdev, DS_FEATURE_REPORT_FIRMWARE_INFO, buf,
> > > + DS_FEATURE_REPORT_FIRMWARE_INFO_SIZE, HID_FEATURE_REPORT,
> > > + HID_REQ_GET_REPORT);
> > > + if (ret < 0)
> > > + goto err_free;
> > > +
> > > + ds->base.hw_version = get_unaligned_le32(&buf[24]);
> > > + ds->base.fw_version = get_unaligned_le32(&buf[28]);
> >
> > Shouldn't the size of the reply be checked?
>
> Good point, I added a check and returning -EINVAL now in case of a
> size mismatch.
>
> >
> > > +
> > > +err_free:
> > > + kfree(buf);
> > > + return ret;
> > > +}
> > > +
> > > static int dualsense_get_mac_address(struct dualsense *ds)
> > > {
> > > uint8_t *buf;
> > > @@ -1172,6 +1233,12 @@ static struct ps_device *dualsense_create(struct hid_device *hdev)
> > > }
> > > snprintf(hdev->uniq, sizeof(hdev->uniq), "%pMR", ds->base.mac_address);
> > >
> > > + ret = dualsense_get_firmware_info(ds);
> > > + if (ret < 0) {
> > > + hid_err(hdev, "Failed to get firmware info from DualSense\n");
> > > + return ERR_PTR(ret);
> > > + }
> > > +
> > > ret = ps_devices_list_add((struct ps_device *)ds);
> > > if (ret < 0)
> > > return ERR_PTR(ret);
> > > @@ -1241,6 +1308,12 @@ static struct ps_device *dualsense_create(struct hid_device *hdev)
> > > /* Set player LEDs to our player id. */
> > > dualsense_set_player_leds(ds);
> > >
> > > + /* Reporting hardware and firmware is important as there are frequent updates, which
> > > + * can change behavior.
> > > + */
> > > + hid_info(hdev, "Registered DualSense controller hw_version=%x fw_version=%x\n",
> > > + ds->base.hw_version, ds->base.fw_version);
> > > +
> > > return (struct ps_device *)ds;
> > >
> > > err:
> > > @@ -1295,6 +1368,12 @@ static int ps_probe(struct hid_device *hdev, const struct hid_device_id *id)
> > > goto err_close;
> > > }
> > >
> > > + ret = sysfs_create_group(&hdev->dev.kobj, &ps_device_attribute_group);
> >
> > It's a minor thing, but I think `device_{add,remove}_group()` would be better
> > here in the sense that it expresses the fact that the group is added to a device,
> > not just any object better.
>
> Agreed, that's nicer I wasn't aware of it. I try to follow what other
> hid drivers do and they all used the kobj directly, which honestly
> felt nasty. Will change it to this.
>
Actually devm_device_add_group seems to be even nicer. Surprisingly it
isn't widely used yet.
Roderick
