[PATCH] hid: wellspring device quirks

Henrik Rydberg rydberg at euromail.se
Tue Jul 15 23:37:54 UTC 2008


On Tue, 2008-07-15 at 14:56 -0600, Tim Gardner wrote:
[...]
> >> Rather then modifying the kernel quirks list, how about updating the
> >> quirks list dynamically. It has the advantage of not causing regressions
> >> for folks that do not have LBM installe.  For example, see
> >> atp_update_quirks() in this patch. It could also work for the upstream
> >> kernel version as well.
> >>
> >> http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-hardy-lbm.git;a=commit;h=32f4209218f2b5481db8643504e9d66060fb9dae
> > 
> > Excellent - you can tell I did not know about this possibility. :)
> > 
> > The device constants are present twice, but I suppose that is fixable
> > later on, if things are already committed. The same goes for the
> > upstream version, once settled. I will have a look at the dynamic quirks
> > then as well.
[...]
> I agree with regard to the duplication of the constants. Your driver has
> not been pulled by Linus, so you may yet have time to get akpm to
> replace the existing patch with something closer to what I have in LBM.
> 
On second thought: the usbhid_modify_dquirk function has been made
static upstream, also in linux-next. It seems the quirk api now only
contains support for loadable parameters. Besides, it is slightly odd to
depend on usbhid to be loaded, isn't it?

Best regards,
Henrik Rydberg






More information about the kernel-team mailing list