If I had to pinpoint a culprit, it'd be in gnuradio-runtime/lib/block.cc
750 bool
751 block::finished()
752 {
753 if((detail()->ninputs() != 0) || (detail()->noutputs() != 0))
754 return false;
755 else
756 return d_finished;
757 }
758
Meaning that blocks with any stream in- or output can't be finished(),
and hence, execution will only stop if the blocks are in- or output
blocked, or the work function explicitely returned WORK_DONE(==-1),
which probably works quite well for stream-only flowgraphs.
However, that "if" was added but a year ago – and likely for a good
reason, that I don't see right now. Maybe someone else could have a look
at this?
Greetings,
Marcus
On 24.07.2016 10:55, Marcus Müller wrote:
> Hi,
>
> 'doh.
>
> Which leaves one to wonder why the finished state never gets checked.
> I'll be back after a bit of tracing.
>
> Cheers,
>
> Marcus
>
>
> On 24.07.2016 08:04, Sylvain Munaut wrote:
>> Hi,
>>
>>> 52 int pdu_to_tagged_stream_impl::calculate_output_stream_length(const
>>> gr_vector_int &)
>>> 53 {
>>> 54 if (d_curr_len == 0) {
>>> 55 /* FIXME: This blocking call is far from ideal but is the
>>> best we
>>> 56 * can do at the moment
>>> 57 */
>>> 58 pmt::pmt_t msg(delete_head_blocking(PDU_PORT_ID, 100));
>>> 59 if (msg.get() == NULL) {
>>> 60 return 0;
>>> 61 }
>> [snip]
>>
>>> Problem is that if we use the non-blocking call here, the scheduler would
>>> have a chance to process the shutdown signal, but it would be constantly
>>> asking (spinning) for the output stream length.
>>>
>>> You could try out what would happen if we'd added a timeout to the blocking
>>> cal; that way, you could reduce the spinning, and hopefully get the
>>> scheduler to check for "done" messages.
>> There _is_ a timeout ... that "100" in there is the # of millisec to
>> wait at most.
>>
>>
>> Cheers,
>>
>> Sylvain
_______________________________________________
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
