Hi all;

 We have a stubborn SPI problem with the DMM355. The TI Advisory
SPRZ264D dated October 2008 mentions that there is a problem with the
   CSHOLD, specifically SPI Master Mode:" Extra Step Required to Use CSHOLD".
We have followed the workaround for 8 or 16 bit writes but we still
have problems with an extra byte being written when the CSHOLD is set
to 0. The errata states that 'Write SPIDAT1.CSHOLD=0 using 8 or 16 bit
write. (do not write to SPIDAT1[15:0]). How do we avoid writing data
to this register? The SPI driver is the one written by Joshua Hintze.
We have tried reading the contents, toggling CSHOLD and writing it all
 back but this gives us a phantom byte. The classical approach of
changing CSHOLD on the last byte gives us a very destructive glitch.

We have thought of changing the driver to the original which seems to
provide for byte access to registers but this would be more work than
we have time for at the moment.

Anyone any ideas?
   Regards
   --
   --Adrian Edmonds

-- 
--Adrian Edmonds
Ramat Yishay
Home +972-(0)4-9930379
Mobile +972-54-680-0580

_______________________________________________
Davinci-linux-open-source mailing list
[email protected]
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source

Reply via email to