<div dir="ltr"><div><div><div>I just sent the patch (the thread title would be different). The patch provides a function that can be polled to know if the domains have been received. You can then simply sleep until it becomes true.<br>
<br></div>Interesting quotes from a certain developer by the way (who was strictly against this harmless additional functionality):<br><br>> For me, there is no point of busily waiting till your packet arrives, instead of being relaxed and receiving the packet next cycle.<br>
<br></div>and in the same email:<br><br>> The drawback is that your propagation time from input change to output reaction is<b> 2 cycles</b>, instead of only 1.<br><br></div>and again in the same email:<br><br>> I do not think that you have a problem. Draw your operations on a time line and convince yourself that once you are in in the loop, you have<b> max 1 cycle</b> delay from input to output.</div>
<div class="gmail_extra"><br><br><div class="gmail_quote">On Wed, Nov 20, 2013 at 6:18 PM, Ian Prochazka <span dir="ltr"><<a href="mailto:jprochazka@persimmontech.com" target="_blank">jprochazka@persimmontech.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Yes, I would also be extremely interested in the patch.<br>
It would actually be very(!!!) useful to have available variation of ecrt_master_receive() that would block and wait till the transaction is done and then returns ASAP when it can.<br>
The other option would be some notification from master or NIC that it received and so ecrt_master_receive() call is safe now.<br>
<br>
Thanks,<br>
Ian<div class="HOEnZb"><div class="h5"><br>
<br>
<br>
On 11/20/2013 11:39, Jeroen Van den Keybus wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Someone actually did that and I extensively tested it. The patch<br>
was very simple.<br>
<br>
<br>
How was this packet arrival time measured ?<br>
<br>
...the ethercat master developers said you don't need to know when a<br>
packet arrives.<br>
<br>
<br>
I could understand their point if the patch would hurt performance too much.<br>
<br>
I still have the patch although it needs a bit of adaptation, in<br>
case they come to their senses. Perhaps if others also mention it<br>
would be useful they would accept that function.<br>
<br>
Well, if it's not too much asked, I'd like to see it.<br>
<br>
Thanks,<br>
<br>
<br>
J.<br>
<br>
</blockquote>
<br>
</div></div></blockquote></div><br></div>