RE: mcp251xfd: rx frame truncation (1/2)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



 
> spi6 (can1): 10 MHz
> spi0 (can0): 16,67 MHz (admitted, this was not intended)

The 16,67 are expected on the pi4. There's an erratum on the chip which makes it only safe to use SPI speeds of ( sysclock/2)*0.85, so 17 MHz with a 40 MHz clock and the 16.67 is as close as the pi can get. That shouldn't be a problem though. I'm more surprised by the 10 MHz though. Is that something deliberately limited in your DT or where does that come from?
 
> Funny: the truncated message was received on can1 - with lower speed. And
AFAIK spi6 is different compared to spi0 in more than just the speed. I had my setup(spi0 and spi1)  running for a total of 16hours but didn't get anything. Will switch to spi6 instead of spi1 as well now. 

Where do the 10 MHz in your case come from? I think the spi6 can also reach the 16.67 MHz and with the throughput you're generating with the testscript that should be better.
For the record, this is how you call the test as well, right? ./sctest -b -l 990 -v cmperr,logmsg -F refilldata,lessdots,stoponerror,randomize -d 64 -n 10 -t 3 can0 can1

> I have never seen other CRC errors than for address 0x0000 and 0x0010.
CRC errors are very rare in a non-noisy environment. The two above are exceptions due to an erratum (0x10) and timing(0x0).

Thomas




[Index of Archives]     [Automotive Discussions]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Security]     [Bugtraq]     [Linux]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]     [CAN Bus]

  Powered by Linux