Return-path: Received: from mail.candelatech.com ([208.74.158.172]:52339 "EHLO ns3.lanforge.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754001Ab1EITFm (ORCPT ); Mon, 9 May 2011 15:05:42 -0400 Message-ID: <4DC83B03.5070002@candelatech.com> (sfid-20110509_210549_910436_49C27E5E) Date: Mon, 09 May 2011 12:05:39 -0700 From: Ben Greear MIME-Version: 1.0 To: Adrian Chadd CC: "linux-wireless@vger.kernel.org" Subject: Re: Looking for 3x3 Atheros NICs using ath9k driver. References: <4DBF1967.9010604@candelatech.com> <4DC042A5.9050806@candelatech.com> <4DC17DAE.9020808@candelatech.com> <4DC47C04.1080604@candelatech.com> In-Reply-To: <4DC47C04.1080604@candelatech.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Sender: linux-wireless-owner@vger.kernel.org List-ID: On 05/06/2011 03:53 PM, Ben Greear wrote: > On 05/04/2011 09:24 AM, Ben Greear wrote: >> On 05/04/2011 09:16 AM, Adrian Chadd wrote: >>> On 4 May 2011 02:00, Ben Greear wrote: >>> >>>> Here's the lspci for the SR71-A that doesn't work with ath9k: >>>> >>>> 03:00.0 Ethernet controller: Atheros Communications Inc. AR5008 >>>> Wireless >>>> Network Adapter (rev 01) >>>> Subsystem: Atheros Communications Inc. Device ee1c >>> >>> ... ee1c? Err, that isn't right. Not by a very, very long shot. >>> >>> What board are you putting this into? >>> >>>> From FreeBSD on a RSPRO (ar71xx mips): >>> >>> ath0@pci0:0:17:0: class=0x028000 card=0x40820777 chip=0x0027168c >>> rev=0x01 hdr=0x00 >>> vendor = 'Atheros Communications Inc.' >>> class = network >>> >>> That definitely, positively is the same AC1A revision that you have. >>> It even does 11n(ag) too. Honest. I've seen it. :-) >>> >>> This and the previous AR93xx "0xabcd" device id post is starting to >>> sound a lot like something's not waking up the PCI bus and/or chip >>> "right" in order to get the PCI vendor id init'ed from the EEPROM.. >> >> Well, I put in one NIC, it works fine, I put in the other, and it >> doesn't. >> >> Same system, reproducible on other systems from same manufacturer >> (Lanner, LEC 2010E). >> >> Haven't tried it on a totally different motherboard. I'll send you this >> broken NIC if you want...I'm too tight on time to hack on it now... > > So, you were right. I put this NIC in another system (VIA C3) and it > comes up > as AR9160 and loads ath9k driver just fine. > > 00:0c.0 Network controller: Atheros Communications Inc. AR9160 > 802.11abgn Wireless PCI Adapter (rev 01) > Subsystem: Device 0777:4082 > Flags: bus master, 66MHz, medium devsel, latency 168, IRQ 18 > Memory at f6000000 (32-bit, non-prefetchable) [size=64K] > Capabilities: [44] Power Management version 2 > Kernel driver in use: ath9k > Kernel modules: ath9k The NIC is still showing up funky on the other system (Atom), even with the same 39-rc6+ kernel. This NIC is reliably showing up wrong on in this system, and the other is reliably showing up well, so I think the problem must be with the NIC or perhaps the ath9k driver. If anyone has any suggestions for how to debug this further, I'm interested... Thanks, Ben > > Thanks, > Ben > >> >> Ben >> >> > > -- Ben Greear Candela Technologies Inc http://www.candelatech.com