Re: [ECOS] Assert in at91rm9200_i2c.c

["Tom Deconinck" <t dot deconinck at gmail dot com>]

On Mon, Sep 1, 2008 at 12:11 PM, Andrew Lunn <andrew@lunn.ch> wrote:
> On Mon, Sep 01, 2008 at 02:40:20AM -0700, Tim Hatton wrote:
>>
>> Hi all,
>> We've been developing an application on our own board which is based on the
>> AT91RM9200. The eCos config is closely based on that for the AT91RM9200-EK
>> and uses the I2C driver from that board to connect to an ADC (for monitoring
>> voltage, current, temp etc).
>>
>> This basically appears to work fine but when I have eCos Asserts turned on I
>> very occasionally get the
>> following assertion:
>>
>> ASSERT FAIL: <5>at91rm9200_i2c.c[101]at91rm9200_i2c_isr() I2C driver error:
>> OVRE without RXRDY
>>
>> Could anybody give me some advice about what, from an application point of
>> view, could cause this sort of behaviour, or what sort of things I should be
>> thinking about?
>>
>> Any assistance would be greatly appreciated,
>> regards,
>> Tim
>
> There is no i2c driver in the open source version of eCos, at least
> none that i know of. I suspect you have the closed source code from
> eCosCentric? That probably also means you have a support contract?  So
> i would suggest making use of your support contract, you have payed
> for it, and ask them....
>
>    Andrew
>
> --
> Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
> and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss
>
>

First of all, Andrew is right, ask eCosCentric.
Second: it's telling you there is a data overrun at the master side
(CPU side) which means there is data incoming, but the peripheral is
not ready to handle it (RXRDY), so the receive side is getting
overrun. This is an error state flagged by the peripheral, so you need
to track down your problem. Or data is coming too fast, or your
application is too slow to handle received data.

Tom

-- 
Before posting, please read the FAQ: http://ecos.sourceware.org/fom/ecos
and search the list archive: http://ecos.sourceware.org/ml/ecos-discuss