TMS320F280049: CAN messages receiving - which one was first - time stamp

Martin Pietka2 said:

until control system do not send many messages in the row, with no delay between

Do you mean "does send many messages in the row with no delay between"?

Yes. Does....

And when you say many messages in a row, you mean with the same CAN ID or different IDs?

With same ID, I expect messages are put in mailboxes from lowest. But it shouldnt be true in all cases as reading is not synchronised. With different ID situation is different, and here I have no idea..

Before you give yourself a lot of work, I would question why these frames (same ID) are being sent in rapid succession. If they are periodic frames they should have an agreed transmission period in the order of >1ms in my experience. I wonder if it is a bug on the transmission side that you receive a burst of frames.

Anyway, there is no message time stamp in the DCAN module as far as I can see but (I think) you can establish an order of reception by linking together multiple message objects (with the same ID) to form a FIFO with the EoB bit. The frame in the lowest message number would be the first to be received since the last processing.

I would still consider an interrupt just to copy the new data with IPC counter time stamp into your own custom buffer and process later. The CAN interrupts are fairly low priority.

Martin,

    The DCAN module does not have a time-stamp counter like the eCAN module in F28065. 

Martin Pietka2 said:

With same ID, I expect messages are put in mailboxes from lowest.

Correct.

Martin Pietka2 said:

But it shouldnt be true in all cases as reading is not synchronised.

True.

Martin Pietka2 said:

With different ID situation is different,

Yes, this is easier to handle when IDs are different.