Re: [uCsimm] touchscreen

From: David Beckemeyer (david@corp.earthlink.net)
Date: Mon Mar 26 2001 - 22:16:06 EST


>Tom Walsh wrote:
>
> Yes, see: 'http://www.openhardware.net/ez328lcd/', I am now using the
> ADS7846 controller and it works rather well. I am also using Phillipe
> Ney's touch panel driver, in the Lineo CVS as well as my own CVS, the
> driver is called 'mc68328digi.*'. I am still having some issues with
> noise pickup from the LCD electronics, but your milage may vary.

Cool. Thanks, Tom. Excellent reference.

Doesn't it seem like the SPI timings in the driver (which does work)
don't seem to match the ADS784x datasheets? It seem like the driver
is using the following timing on the SPI lines (hex):

  DIN 00 80 00 D3 00 93 00 80 00 00
  
  DOUT xx xx xx xx 7F F8 7F F8 xx xx

All with 16-bit xfers, where the 7FF8 bits above represent the bits
coming back from the ADS784x chip. That seems like 80 clocks to
read the X and Y values. The ADS784x datasheet seems to indicate
that the data should start to come back on the DOUT line sooner.
Also the value 0x80 as a control byte (start bit withh all zeros
after that) doesn't even seem to be documented as a valid value.

What am I missing?

Also, both the 'mc68328digi.*' touchcsreen driver and my Maxim
UART driver (http://www.bdt.com/david/ucsimm/max311Xserial-0.20-R1.tar.gz)
both assume they have full control over the SPI bus. That means both
drivers can't exist/operate without some mutex type resource management
around the SPI. Has anybody worked on something like that for
ucsimm/uclinux?

Thanks again,

  - david

This message resent by the ucsimm@uclinux.com list server http://www.uClinux.com/



This archive was generated by hypermail 2b30 : Sun Apr 07 2002 - 00:01:42 EST