On 9/16/20 1:04 PM, Vladimir Oltean wrote: > [ resending, forgot to copy Wolfgang ] > > On Wed 9/2/2020 10:09 AM, Wolfgang Grandegger wrote: >> canfdtest normally runs on the DUT *and* a the host. The DUT receives >> the messages from the host, increments the frame data bytes and then >> sends them back to the host. With "loopback" mode, the data bytes are >> not incremented and that's what you see above. >> >> Wolfgang > > Wolfgang is of course right, but we're nonetheless investigating what > seems to be a real problem, and what Pankaj had seen was a red herring. > > So currently what I suspect is going on, when I am running canfdtest > between 2 LS1028A-RDB boards, is that the DUT is reordering frames on > TX. > > See, for example, the screenshot below: > https://drive.google.com/file/d/1rOeW3aXh3kPh1CJ39lCccRfjFz5JN5I6/view?usp=sharing > > I have added trace points to the end of the flexcan_start_xmit function, > which print the entire skb, and the frames appear to be written to the > TX message buffer in the correct order. They are seen, however, in the > incorrect order on the wire.
Which driver are you using? The mainline driver only uses one TX buffer. Marc -- Pengutronix e.K. | Marc Kleine-Budde | Embedded Linux | https://www.pengutronix.de | Vertretung West/Dortmund | Phone: +49-231-2826-924 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
signature.asc
Description: OpenPGP digital signature
