Re: [PATCH v8 0/7] CTU CAN FD open-source IP core SocketCAN driver, PCI, platform integration and documentation

On 30.03.2022 09:46:08, Ondrej Ille wrote:

sorry for the late reply, my work duties are keeping me very busy.

Known problem :)

Let me just quickly comment on topics discussed in the emails above.

*1. Separation of PROP and TSEG1*

IMHO there is no real benefit. The reason why CTU CAN FD has this
split is legacy. First implementation back in 2015 had this split
since I wanted to follow the standard. In HW, the first thing done in
bit timing logic (prescaler module), these two numbers are added, and
all further resynchronization/hard-synchronization is done with TSEG1
value...

Thanks for the insight. It's not easy to get in touch with the
developers of the proprietary IP cores :)

Never the less, there's another IP core which has different sizes for
the prop and tseg1 register. So an update of the bit timing constant
would help both.

*2. Number of TXT Buffers and RX Buffer size:*

Pavel already replied with TXTB_INFO. The same role has the RX_MEM_INFO
register, when it comes to RX side.

Thanks for the clarification.

regards,
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 |