Re: [PATCH] Added device to hid-ff

From: "Dmitry Torokhov" <dmitry.torokhov@gmail.com>
To: Anssi Hannula <anssi.hannula@gmail.com>
Cc: "STenyaK (Bruno Gonzalez)" <stenyak@gmail.com>,
	johann deneux <johann.deneux@gmail.com>,
	linux-input <linux-input@atrey.karlin.mff.cuni.cz>,
	Valentin Zagura <puthre@gmail.com>
Subject: Re: [PATCH] Added device to hid-ff
Date: Thu, 15 Feb 2007 15:09:50 -0500	[thread overview]
Message-ID: <d120d5000702151209y4b048040m85cf546129ae86b8@mail.gmail.com> (raw)
In-Reply-To: <45D4B9AB.70509@gmail.com>

On 2/15/07, Anssi Hannula <anssi.hannula@gmail.com> wrote:
> Dmitry Torokhov wrote:
> > On 2/15/07, STenyaK (Bruno Gonzalez) <stenyak@gmail.com> wrote:
> >> Sorry for not providing a proper patch, could you please also add
> >>         {0x46d, 0xc294, hid_lgff_init}, // Logitech G25 wheel
> >> to the list? I tested it here and that wheel is also working.
> >>
> >
> > OK.
> >
>
> Hmm.. Shouldn't an entry also be added to the devices[] list in
> hid-lgff.c for these devices? (and it seems one device is already
> missing from there in the current kernel).
>

Yep. It looks like they should be mapped to ff_joystick, right?

> I'm not able to check right now, but I think some earlier kernels may
> have defaulted to ff_joystick if no entry is present in devices[], hence
> these devices working for some without the entry.
> (so the alternative is to restore that behaviour)
>
> BTW, shouldn't the PID/VID check in hid_lgff_init() do endian conversion?
>

No, because input_id keeps this data in host format. See usb_to_input_id().

-- 
Dmitry