[USRP-users] Out-of-order send

Michael West michael.west at ettus.com
Mon Oct 19 14:49:48 EDT 2015

Hi Dario,

No.  UHD does not do reordering.  This is intentional so that latency is
kept as low as possible.  The USRP only looks at the time of the next
packet.  A packet with a timespec earlier than the previous will be
considered a late packet and will be dropped.


On Fri, Aug 21, 2015 at 8:26 AM, Dario Fertonani via USRP-users <
usrp-users at lists.ettus.com> wrote:

> The function
> uhd::tx_streamer::send
> provides timed transmissions when
> ( has_time_spec == true )
> in the relevant metadata.
> Does this function support out-of-order requests? For example, if I have
> four times A<B<C<D, all far enough in the future but all within the UHD
> buffering capability, is the sequence (A,C,B,D) of times valid? By valid I
> mean: UHD reorders them and sends over-the-air (A,B,C,D). Or does UHD drop
> the out-of-order packet B and transmit (A,zeros,C,D)? Some background
> follows.
> In my application I have to deal with out-of-order packets, meaning that
> bursts of samples to send don't get to the transmit thread always in the
> correct order. (This is because we run B210-based remote radio heads about
> a thousand miles away from the AWS machines that run our LTE stack.)
> Currently, buffering and reordering is done in my application. I wonder if
> I even need to do that, since we're now down to the last bit of
> optimization.
> Thanks,
> Dario
> _______________________________________________
> USRP-users mailing list
> USRP-users at lists.ettus.com
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20151019/11788fa0/attachment-0002.html>

More information about the USRP-users mailing list