Re: [PATCH 00/22] Staging: rt{286,287,307}0: various fixes/cleanups

[Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>]

On Monday 17 August 2009 05:47:45 Mike Galbraith wrote:
> On Sun, 2009-08-16 at 21:29 +0200, Bartlomiej Zolnierkiewicz wrote:
> > Hi,
> > 
> > Some fixes, some cleanups and a lot of preparations for finally merging
> > rt{28,30}70 drivers together.
> > 
> > Tested with RT2860 (Asus Eee 901) and RT3070 (el cheapo USB dongle that
> > I've bought recently to speed up the development, I'm posting this mail
> > using it right now) chipsets.
> 
> I plugged these into tip this morning (3, 5 and 21 needed hammering) and
> tried my box's onboard USB gizmo:
> 
> Bus 007 Device 003: ID 13d3:3247 IMC Networks 802.11 n/g/b Wireless LAN Adapter
> 
> Works fine here.

Thanks a lot for extra testing, could you also check what MAC version is
this beast using?  I'm now working on merging rt2870 and rt3070 together,
and I would need somebody with 2870 flavor to verify the end result.. :)

The USB dongle that I'm doing tests with is:

	Bus 004 Device 003: ID 148f:3070 Ralink Technology, Corp.

and is identified by rt2800usb as 3070 flavor:

	phy0 -> rt2x00_set_chip: Info - Chipset detected - rt: 1600, rf: 0005, rev: 30700200.

> > PS I tried using rt2800usb driver found in 2.6.31-rc6 (with overdue fix
> > for memory corruption in rf cache from Pavel Roskin) for RT3070 chipset
> > but it doesn't seem to work yet (iwlist wlan0 scan => no stations)..
> 
> Yeah, I can ping/traceroute with it, but that's about it, seems to choke
> as soon as you need to xfer non-dinky packets.  (I would have thought
> that the corruption fix would swim upstream given it's utterly deadly)

It it in upstream now, a bit of competition (even if the end goal is
exactly the same) is never a bad thing.. ;)