On 2010-11-29 01:12, David Duffy (AVD) wrote:
> Jan-Erik Soderholm wrote:
>> On 2010-11-28 23:59, David Duffy (AVD) wrote:
>>
>>> Olin Lathrop wrote:
>>>
>>>> David Duffy (AVD) wrote:
>>>>
>>>>
>>>>> I am using INDF0 for reading the receive buffer.  It is the same code
>>>>> I've used countless times before.
>>>>>
>>>>>
>>>> Are you doing this in the interrupt routine and forgot to save/restore=
 the
>>>> FSR?
>>>>
>>> I am using INDF0 in the serial receive ISR, as well as in tasks
>>> triggered by flags in the main loop.
>>>
>>> I'm not saving / restoring them in the ISR, as I understand this
>>> particular chip does that by itself.
>>> David...
>>>
>>>
>>
>> They are. But of course you have to re-init them at each
>> entry to the ISR, they are not preserved between two interrupts.
>> You didn't say you don't, but just to clearify.
>>
>> B.t.w, this is what the PIC16F193X/LF193X datasheet says :
>>
>> Upon entering an interrupt, the return PC address is
>> saved on the stack. Additionally, the following registers
>> are automatically saved in the shadow registers:
>> =95 W register
>> =95 STATUS register (except for TO and PD)
>> =95 BSR register
>> =95 FSR registers
>> =95 PCLATH register
>> Upon exit from the Interrupt Service Routine, these registers
>> are automatically restored. Any modifications to
>> these registers during the ISR will be lost. Depending
>> on the user=92s application, other registers may also need
>> to be saved.
>>
>>
>> What I do not understand is why you don't read the receive
>> buffer (RCREG) directly. Why go through the index registers ?
>> The reading will always be from the same fixed address (RCREG)
>> so I do not see the need for any "indexing" here !?
>>
>
> To clarify, I use INDF0 to store the incoming byte into my RAM receive
> buffer.
>
> Also, the use of INDF0 in the ISR&  main routines is separate. That is,
> they operate on the same buffer, but the pointers are separate variables.
> David...
>

OK. That is clearer. :-)

Then maybe "I am using INDF0 for reading the receive buffer."
should have been saying "...writing into the receive buffer". :-)

And it should be OK as long as you re-load INDF0 at each entry
of the ISR with the proper receive-position in the buffer.

The read-position in the buffer will be saved and restored
by the context save/restore the hardware does at an interrupt.

No further questions. :-)

Jan-Erik.
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.