Re: [PATCH 3/4] liss331d1: accelerometer driver

[Ãric Piel]

On 15/04/10 01:05, Neil Brown wrote:
:
> 
> The Openmoko Freerunner has as lis302D which is an SPI version of a very
> similar device, so making the one driver work on that was well would be nice
> if possible.
Definitely!

> 
> The lis302d can detect threshold crossings and taps as well as simple
> orientation and I found it useful to include support for those in the driver
> as well.
> In particular:
>  1/ I allowed the 'data_rate' that was programmed to be any number (including 
>     fractions and 0) and the driver would report at that rate, using a timer
>     for rates below 50Hz.
>     When the app is only interested in long-term change, this can
>     significantly reduce the amount of data that has to be handled by
>     userspace.
Samu (from Nokia) has started to had such idea via the patch
lis3-setup-poll-interval-limits.patch which in now in the -mm tree. It
allows to specify the polling rate via the input interface
(/sys/devices/platform/lis3lv02d/input/input8/poll). But for now I don't
think the actual refresh rate is updated automatically, it would be a
nice addition.


> 
>  2/ I allowed a 'threshold' to be set so that changes bigger than that get
>     notified promptly even if that is faster than the requested data rate
It could be interesting too... but what interface are you using for the
userspace to specify this threshold? A sysfs file?

> 
>  3/ I allowed taps to be detected and reported as BTN_X BTN_Y and BTN_Z
>     input events.
This one should now be working with Samu's patch
lis3-interrupt-handlers-for-8bit-wakeup-and-click-events.patch also in
the -mm tree.

> 
> Would it be sensible to include that support in this driver?  The
> LIS331DL seem to have the same threshold and tap support.
So all in all, if you could have a look at what changes are necessary to
get the lis3lv02d driver work with your hardware, it would be great. (I
suspect it should be quite little.)

See you,
Eric
--
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/