Split off from my entry of 3/12/13 in thread RE:AM335x RTC:
I have the same problem with WinCE 7 on AM335X. I can read the registers via testi2c but I am unable to write to, for example, 0x3f: testi2c reports success but the value does not change, and I am unable to enable the RTC as a result. I tried writing many other registers but no success there either.
Using the correct channel number for testi2c I obtain the same data as Laurent reported on Nov 1 2012. I still have this problem.
I am surprised that I am able to read successfully (which requires the slave to drive the line) but not write, even though the AM335x I2C master can drive the line well enough to address the PMIC, provide clock for the read, etc. Since the original thread reports the problem as being solved in Linux this presumably did not involve the silicon erratum, correct? It is not indicated how Yegor (=rfc2217) was able to change the values.
Thanks Biser for your prompt response- I will create new posts in the future. Sorry I was being confusing rather than helpful.
Bruce
