Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Work on powerpc/beatnik/if_mve network driver (epics codeathon 2021) #4

Open
wants to merge 7 commits into
base: 5-freebsd-12
Choose a base branch
from

Conversation

till-s
Copy link

@till-s till-s commented Mar 15, 2021

This pull-request adds a nexus ethernet driver for the Marvell mv643xx chip which is present on the MVME6100 board (beatnik BSP). Note that the patch requires a companion patch to the RTEMS BSP.

First stab at mv643xx; port almost complete (mii stuff missing)
and compiles -- no tests yet!
was delayed by waf not printing compiler warnings by default!

Three little typos cost me almost a day:
  - missing 'return 0;' at the end of mve_attach()
  - two bad pointers (wrong type) used
All of these would have been found within minutes had the
warnings not been disabled :-(
The recursive lock takes care of this...
Testing (TCP with larger amounts of data while MVETH_TESTING
was defined which reduces the ring sizes to very small numbers)
revealed that the BSD stack honours the 'sendqlen' and
never hands us chains exceeding the ring size.
…r statistics

The 'legal' access by means of a special ioctl() is way too cumbersome
when you need to do quick diagnosis.
avoid using struct members directly.
@kiwichris
Copy link
Member

Is this a PCI bus device?
Are these on the bus 0 on the beatnik?

@till-s
Copy link
Author

till-s commented Mar 30, 2021

The ethernet adapter is part of the system controller and accessible directly (w/o going through PCI).

@kiwichris
Copy link
Member

Maybe I have the wrong board. Do you have a link to the board's datasheet? The beatnik I have looked at has an octet Marvel NIC and there are are 8 marvel devices appearing in the PCI probes.

The e1000 is on a secondary PCI bus.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants