On 03/13/2010 09:29 PM, Jonas Gorski wrote:
> I am currently adapting the ar8216 driver to support the ar8316[1],
> and therefore it probably the best to combine the effort (and there
> won't be a race whose patch gets accepted first ;-).
Must have missed this thread searching for ar7240 only. I'll try
to base any change on top of your patch.
> As far as I can see, the internal switch of the ar7240 is basically an
> ar8216, but with a gigabit mii connection on the cpu port.
Yes. According to the source code the register definitions seems to be almost
identical with
the DSA driver for ar7240 defining more (MIB counter, MDIO access, ...).
> I am currently identifying the chip through the version and revision
> fields of the control register (0x0000), where the ar8216 is 1/1, the
> ar8316 is 16/1.
> As far as I can see in the dsa driver, the version of the ar7240 is
> also 1, but I lack the revision information.
> ('Though it could be "assumed" through the phy_device being connected
> with a gigabit mii connection).
The revision of the AR7240 is 2. Another way to identify the device may be the
ID of the
integrated PHYs (AR7240 004d:d041)
> OTOH, the DSA switching architecture looks like the "right way" to do
> it, but currently lacks too many features.
> There is a patch for hardware bridging support for DSA[2], but its
> already over a year old, and got never applied. But this might be
> worth a shot to try to add that to OpenWRT and adapt the ar7240 dsa
> driver to it.
Thanks for the pointer. I'll look into this patch.
Regards,
Jörg.
_______________________________________________
openwrt-devel mailing list
[email protected]
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
