Andrea Adami wrote:
> Update:
> PATCH] input: schedule corgi_ssp and corgi_ts to be removed

> +What:  corgi_ssp and corgi_ts driver
> +When:  2.6.35
> +Files: arch/arm/mach-pxa/corgi_ssp.c, drivers/input/touchscreen/corgi_ts.c
> +Why:   The corgi touchscreen is now deprecated in favour of the generic
> +       ads7846.c driver. The noise reduction technique used in corgi_ts.c
> +       that to wait til vsync before ADC sampling is also inte grated into
> +       ads7846 driver now. Provided that the original driver is not generic
> +       and difficult to maintain, will be removed later.
> +Who:   Eric Miao <>

I agree. Maybe all deprecated Zaurus drivers should be removed. I can
confirm, that all new drivers with exception of SPI work correctly on
Spitz. (I guess there is still spitzkbd.)

Somebody should update default configuration. Here is my one:
I guess that all (maybe without the most common) options for externally
supported devices should be removed (e. g. keep only USB host config,
remove all usb devices options, keep CF/SD, remove all CF/SD driver

Regarding the SPI: The old driver supported reading MAX1111 data while
suspended. The new does not.

I guess that it is caused by hardcoded zaurus specific reference inside
MAX1111 driver code instead of correct solution. According to answer of
my "How to write driver that needs to call another driver?" question to
LKML it should be:

John Stoffel wrote:
... from what I recall in
the past when such questions came up is for you to write a low level
core driver, which binds to the device, and then exposes that various
interfaces you want to bind to for other devices to talk to.

This gets the ref counting right, and lets you know when something is
accessing the MAX1111, so that suspend and such happen in the correct

Look through the lkml archives for previous answers.  I'm sure Greg
will have a better answer and a pointer to an example driver to look

Stanislav Brabec

Zaurus-devel mailing list

Reply via email to