3c59x with linuxppc

[Andreas Tobler]

Hi,


Donald Becker wrote:
> The (very old) 3c590 series is pretty ugly when it comes to SMP locking.
> The windowing scheme, a architectual characteristic from its history as an
> ISA design, interacts badly with SMP operations.  Combined with the single
> bus master engine, you end up with either low performance or complications.
> But since the 3c590 is old and rare, it's not worth providing more than
> functional (not high performance) support.

Well, I own 3c590, the 3c905b is only for test. So it would be nice if
we have a functional version. No, no SMP is necessary here.

> The primary "gotcha" with the 3c905 is the rarely accessed media selection
> and statistics.  Otherwise there should be no need for locks.  Given
> that the media selection is rare, using cli()/sti() is a better choice
> even if they are much more expensive than spinlocks on SMP machines.

Do you mean only in vortex_get_stats and in vortex_open there are locks
necessary?
Or where exactly does the media selection happen? (sorry newbie Q)

> The set_rx_mode() is called a two or three times at initialization.  (The
> 'ifconfig' program doesn't group all of the flags that need to be set.)
> It's also called whenever the multicast list or filtering mode changes.

This expails where the messages are coming from. Good to know.

I tried to replace the spin_locks with the original cli/sti, but I
didn't manage to get it working so far. Limited time.

I'm also not quite sure if the spinlocks are 100% the same on PowerPC as
on intel?

Thanks

Andreas
-------------------------------------------------------------------
To unsubscribe send a message body containing "unsubscribe"
to linux-vortex-request@beowulf.org