On 3/23/23 18:30, Dmitry Torokhov wrote:
> On Fri, Feb 24, 2023 at 05:21:46PM -0800, Vicki Pfau wrote:
>> Some third-party controllers, such as the HORPIAD FPS for Nintendo Switch and
>> Gamesir-G3w, require a specific packet that the first-party XInput driver sends
>> before it will start sending reports. It's not currently known what this packet
>> does, but since the first-party driver always sends it's unlikely that this
>> could cause issues with existing controllers.
>>
>> Co-authored-by: Andrey Smirnov <andrew.smirnov@xxxxxxxxx>
>> Signed-off-by: Vicki Pfau <vi@xxxxxxxxxxx>
>> ---
>> drivers/input/joystick/xpad.c | 22 ++++++++++++++++++++++
>> 1 file changed, 22 insertions(+)
>>
>> diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c
>> index 403b57e8176b..04af2213407f 100644
>> --- a/drivers/input/joystick/xpad.c
>> +++ b/drivers/input/joystick/xpad.c
>> @@ -265,6 +265,7 @@ static const struct xpad_device {
>> { 0x0f0d, 0x0067, "HORIPAD ONE", 0, XTYPE_XBOXONE },
>> { 0x0f0d, 0x0078, "Hori Real Arcade Pro V Kai Xbox One", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOXONE },
>> { 0x0f0d, 0x00c5, "Hori Fighting Commander ONE", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOXONE },
>> + { 0x0f0d, 0x00dc, "HORIPAD FPS for Nintendo Switch", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOX360 },
>> { 0x0f30, 0x010b, "Philips Recoil", 0, XTYPE_XBOX },
>> { 0x0f30, 0x0202, "Joytech Advanced Controller", 0, XTYPE_XBOX },
>> { 0x0f30, 0x8888, "BigBen XBMiniPad Controller", 0, XTYPE_XBOX },
>> @@ -2020,6 +2021,27 @@ static int xpad_probe(struct usb_interface *intf, const struct usb_device_id *id
>> goto err_free_in_urb;
>> }
>>
>> + if (xpad->xtype == XTYPE_XBOX360) {
>> + /* Some third-party controllers Xbox 360-style controllers
>> + * require this message to finish initialization */
>> + uint8_t dummy[20];
>> + int ret;
>> +
>> + usb_control_msg_recv(udev, 0,
>> + /* bRequest */ 0x01,
>> + /* bmRequestType */
>> + USB_TYPE_VENDOR | USB_DIR_IN |
>> + USB_RECIP_INTERFACE,
>> + /* wValue */ 0x100,
>> + /* wIndex */ 0x00,
>> + dummy, sizeof(dummy),
>> + 25,
>> + GFP_KERNEL);
>> + if (ret)
>> + dev_warn(&xpad->dev->dev,
>> + "unable to receive magic message: %d\n", ret);
>
> You are not setting "ret", how was this tested?
Presumably with a stack frame that had that space unwittingly initialized to zero. Good catch, thanks. I've resubmitted with this fixed in v2.
>
>> + }
>> +
>> ep_irq_in = ep_irq_out = NULL;
>>
>> for (i = 0; i < 2; i++) {
>> --
>> 2.39.2
>>
>
