Re: [PATCH] HID: add chardev to propagate special events of Roccathardware to userland

From: Jiri Kosina
Date: Tue May 18 2010 - 08:09:57 EST


On Sun, 9 May 2010, Stefan Achatz wrote:

> > > static void kone_set_settings_checksum(struct kone_settings *settings)
> > > @@ -630,7 +631,7 @@ static ssize_t kone_sysfs_set_startup_profile(struct device *dev,
> > > static ssize_t kone_sysfs_show_driver_version(struct device *dev,
> > > struct device_attribute *attr, char *buf)
> > > {
> > > - return snprintf(buf, PAGE_SIZE, DRIVER_VERSION "\n");
> > > + return snprintf(buf, PAGE_SIZE, ROCCAT_KONE_DRIVER_VERSION "\n");
> >
> > Do we really want to keep this kind of driver versioning? Git provided
> > much more fine-graned versioning anyway. What is it needed for?
>
> My comment to function kone_sysfs_show_driver_version states:
> /*
> * This file is used by userland software to find devices that are handled by
> * this driver. This provides a consistent way for actual and older kernels
> * where this driver replaced usbhid instead of generic-usb.
> * Driver capabilities are determined by version number.
> */

Hmm, I don't seem to get it. Is it some kind of ABI versioning? What role
does legacy usbhid play in it, as it didn't provide this file at all?

> > > +/*
> > > + * Module roccat is a char device used to report special events of roccat
> > > + * hardware to userland. These events include requests for on-screen-display of
> > > + * profile or dpi settings or requests for execution of macro sequences that are
> > > + * not stored in device. The information in these events depends on hid device
> > > + * implementation and contains data that is not available in a single hid event
> > > + * or else hidraw could have been used.
> > > + * It is inspired by hidraw, but uses only one circular buffer for all readers.
> >
> > This information could maybe go into the kernel changelog as well. For
> > someone not familiar with the driver layout, the difference between
> > 'roccat' and 'roccat-kone' might be very difficult to understand.
>
> How or where do I do this?

Just put it as an "explanation" text accompanying the patch. That will be
then propagated into the changelog once comitted.

--
Jiri Kosina
SUSE Labs, Novell Inc.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/