> It fails before FW loading as below.
> 
> 
>     Maybe you could do some testing with https://github.com/rastersoft/gsl1680
>     <https://github.com/rastersoft/gsl1680> and
>     see if you can get it working by skipping the GPIO stuff?
> 
> 
> I did a try with Joe's one (https://github.com/jabjoe/sunxi-gslx680) but it is
> failling with error
>
>  1. [   91.366514] incomplete xfer (0xff)
>  2. [   91.366532] reset_chip: gsl_ts_write 1 fail!
>  3. [   91.366541] init_chip: reset_chip fail: -70
>  4. [   91.366552] gslx680 1-0040: init_chip failed
> 
>  It seems it's due to I2C bus error and someone on LeMaker forum's pointed to
> me I have probably to switch to 400 khz bus.
>  Another thing I have to do, is to check to which GPIO SHUTDOWN pin is
> connected ... but for that, I need to tears down my device (no time yet : ski
> :) ).
> 
> But if I skip GPIO stuff, how the device will turn on loading phase ? Is
> writing to the device enough ?

If you're interested, there are two new drivers for Silead devices, one by
Robert Dolca (https://lkml.org/lkml/2015/8/25/738) and one by myself.
(https://github.com/onitake/gslx680-acpi) They are mainly intended for
ACPI-based platforms (i.e. Intel Baytrail tablets). I'm very interested in
getting either (or both) of them to work on DeviceTree as well, but this is
difficult without a device to test.

Do you have a DT in human-readable form available? Does it contain any
GPIO/IRQ configuration data for your device?

If you do not use power management, I think the device will respond after
powerup even without the GPIO. At least, I got that impression while working
with a GSL1680. However, if you want to enable power save, I'm pretty sure you
need the pin. Also, you definitely need the IRQ line, unless you want to poll
the chip.

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to linux-sunxi+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to