From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mout.gmx.net ([212.227.15.19]:51558 "EHLO mout.gmx.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751999AbaHAGi1 (ORCPT ); Fri, 1 Aug 2014 02:38:27 -0400 Message-ID: <53DB35DC.6080000@rempel-privat.de> (sfid-20140801_083833_523684_0BEE7317) Date: Fri, 01 Aug 2014 08:38:20 +0200 From: Oleksij Rempel MIME-Version: 1.0 To: Julian Calaby , linux-sunxi@googlegroups.com, linux-wireless Subject: Re: [linux-sunxi] Re: A20 + AR9271 wifi problem References: <8d579ffa-80e9-470e-9c29-ecb3c4b67c15@googlegroups.com> <9652a9a0-f205-4d03-8a69-4f7eb95963da@googlegroups.com> <53DA5F11.4040306@gmail.com> In-Reply-To: Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="10eDIKv5R3pQknWRqM9PeNvW8IHHntd3F" Sender: linux-wireless-owner@vger.kernel.org List-ID: This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --10eDIKv5R3pQknWRqM9PeNvW8IHHntd3F Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Am 01.08.2014 um 02:13 schrieb Julian Calaby: > Hi Steven, >=20 > I'm CCing Linux-wireless as they might be able to help and I'm running > out of ideas =3D) >=20 > On Fri, Aug 1, 2014 at 1:21 AM, Steven wrote:= >> Julian, >> >> >> On 07/31/2014 02:28 PM, Julian Calaby wrote: >>> Hi Steven, >>> >>> On Thu, Jul 31, 2014 at 8:47 PM, wrote: >>>> Hello Tim, >>>> >>>> Thanks for your advice and sorry for delay. For now I'm trying to fi= nd out how do I know if kernel was compiled with this config. >>>> >>>> All I can see is : >>>> >>>> root@a20-OLinuXino:~# find / -iname "*cfg80211*" >>>> /sys/module/cfg80211 >>>> /sys/module/cfg80211/parameters/cfg80211_disable_40mhz_24ghz >>>> >>>> root@a20-OLinuXino:~# find / -iname "*ath9k*" >>>> /sys/bus/usb/drivers/ath9k_htc >>>> /sys/devices/platform/sw-ehci.1/usb2/2-1/2-1:1.0/leds/ath9k_htc-phy1= >>>> /sys/class/leds/ath9k_htc-phy1 >>>> /sys/module/ath/holders/ath9k_common >>>> /sys/module/ath/holders/ath9k_htc >>>> /sys/module/ath/holders/ath9k_hw >>>> /sys/module/ath9k_common >>>> /sys/module/ath9k_common/holders/ath9k_htc >>>> /sys/module/mac80211/holders/ath9k_htc >>>> /sys/module/ath9k_htc >>>> /sys/module/ath9k_htc/drivers/usb:ath9k_htc >>>> /sys/module/ath9k_hw >>>> /sys/module/ath9k_hw/holders/ath9k_common >>>> /sys/module/ath9k_hw/holders/ath9k_htc >>>> /lib/modules/3.4.67+/kernel/drivers/net/wireless/ath/ath9k >>>> /lib/modules/3.4.67+/kernel/drivers/net/wireless/ath/ath9k/ath9k_com= mon.ko >>>> /lib/modules/3.4.67+/kernel/drivers/net/wireless/ath/ath9k/ath9k.ko >>>> /lib/modules/3.4.67+/kernel/drivers/net/wireless/ath/ath9k/ath9k_htc= =2Eko >>>> /lib/modules/3.4.67+/kernel/drivers/net/wireless/ath/ath9k/ath9k_hw.= ko >>>> >>>> >>>> So I guess cfg80211 is supposed to be present in my /lib/modules dir= ectory. >>> You wouldn't have the /sys/module cfg80211 entries if it wasn't >>> present - ath9k_htc simply won't work without it. I'm guessing it's >>> built in. >>> >>> I suspect it's something else. >>> >>> Do other devices work on that USB port? What happens if you plug it i= n >>> after it's booted? >>> >>> Also, you should be using the iw command ("iw dev wlan9 scan") not >>> iwlist. Make sure you have the "iw" package installed. >>> >>> Finally, scanning doesn't always work first time (i.e. the device is >>> busy doing something else) and it doesn't work before you configure i= t >>> up ("ifconfig wlan9 up") so try it a couple of times before calling i= t >>> completely broken. >>> >>> Thanks, >>> >> >> I use as well a RT5370 based wifi card and external USB HDD without an= y >> issue. >> I played around plugging and unplugging the device, but ifconfig ends = up >> freezing. I attached dmesg output with probably interesting informatio= n, >> timeouts everywhere. >> >> After a reboot, if I plug the device, it gets correctly detected, and = I >> can bring it up, then I tried : >> >> root@a20-OLinuXino:~# iw dev wlan9 scan >> command failed: Network is down (-100) >> root@a20-OLinuXino:~# ifconfig wlan9 up >> root@a20-OLinuXino:~# iw dev wlan9 scan >> scan aborted! >> root@a20-OLinuXino:~# ifconfig wlan9 up >> root@a20-OLinuXino:~# iw dev wlan9 scan >> ^C >> root@a20-OLinuXino:~# iw dev wlan9 scan >> command failed: Device or resource busy (-16) >> root@a20-OLinuXino:~# iw dev wlan9 scan >> command failed: Device or resource busy (-16) >> root@a20-OLinuXino:~# iw dev wlan9 scan >> command failed: Device or resource busy (-16) >> >> >> When "scan aborted!" appears, dmesg shows at the same time : >> >> Jul 31 14:58:54 a20-OLinuXino kernel: [ 152.493643] ehci_irq: port >> change detect >> Jul 31 14:58:54 a20-OLinuXino kernel: [ 152.503192] hub 2-0:1.0: port= 1 >> disabled by hub (EMI?), re-enabling... >> Jul 31 14:58:54 a20-OLinuXino kernel: [ 152.513886] usb 2-1: USB >> disconnect, device number 2 >> Jul 31 14:58:56 a20-OLinuXino kernel: [ 154.386911] ath: phy0: Chip >> reset failed >> Jul 31 14:58:56 a20-OLinuXino kernel: [ 154.396668] ath: phy0: Unable= >> to reset channel (2412 Mhz) reset status -22 >> Jul 31 14:58:56 a20-OLinuXino kernel: [ 154.412196] ath: phy0: Unable= >> to set channel >> Jul 31 14:58:56 a20-OLinuXino kernel: [ 154.532207] ath: phy0: Chip >> reset failed >> Jul 31 14:58:56 a20-OLinuXino kernel: [ 154.542586] ath: phy0: Unable= >> to reset hardware; reset status -22 (freq 2422 MHz) >> Jul 31 14:58:56 a20-OLinuXino kernel: [ 154.671860] ath: phy0: Failed= >> to wakeup in 500us >> Jul 31 14:58:56 a20-OLinuXino kernel: [ 154.697274] ath: phy0: Failed= >> to wakeup in 500us >> Jul 31 14:58:57 a20-OLinuXino kernel: [ 154.721561] ath: phy0: Failed= >> to wakeup in 500us >> Jul 31 14:58:57 a20-OLinuXino kernel: [ 154.993220] usb 2-1: ath9k_ht= c: >> USB layer deinitialized >> Jul 31 14:58:57 a20-OLinuXino kernel: [ 155.278312] usb 2-1: new >> high-speed USB device number 3 using sw-ehci >> Jul 31 14:58:57 a20-OLinuXino kernel: [ 155.460699] usb 2-1: ath9k_ht= c: >> Firmware htc_9271.fw requested >> Jul 31 14:58:58 a20-OLinuXino kernel: [ 155.782216] usb 2-1: ath9k_ht= c: >> Transferred FW: htc_9271.fw, size: 51008 >> Jul 31 14:58:58 a20-OLinuXino kernel: [ 156.032307] ath9k_htc 2-1:1.0= : >> ath9k_htc: HTC initialized with 33 credits >> Jul 31 14:58:58 a20-OLinuXino kernel: [ 156.262060] ath9k_htc 2-1:1.0= : >> ath9k_htc: FW Version: 1.4 >> Jul 31 14:58:58 a20-OLinuXino kernel: [ 156.270467] ath: EEPROM >> regdomain: 0x833a >> Jul 31 14:58:58 a20-OLinuXino kernel: [ 156.279670] ath: EEPROM >> indicates we should expect a country code >> Jul 31 14:58:58 a20-OLinuXino kernel: [ 156.290086] ath: doing EEPROM= >> country->regdmn map search >> Jul 31 14:58:58 a20-OLinuXino kernel: [ 156.299279] ath: country maps= >> to regdmn code: 0x37 >> Jul 31 14:58:58 a20-OLinuXino kernel: [ 156.307499] ath: Country alph= a2 >> being used: GB >> Jul 31 14:58:58 a20-OLinuXino kernel: [ 156.314466] ath: Regpair used= : 0x37 >> Jul 31 14:58:58 a20-OLinuXino kernel: [ 156.337430] ieee80211 phy1: >> Atheros AR9271 Rev:1 >> Jul 31 14:58:58 a20-OLinuXino kernel: [ 156.352292] Registered led >> device: ath9k_htc-phy1 >> Jul 31 14:59:11 a20-OLinuXino kernel: [ 169.356075] >> ADDRCONF(NETDEV_UP): wlan9: link is not ready >> >> >> At this point all network commands are locked. Unplugging et repluggin= g >> won't change anything. Even rebooting seems to hang, unfortunately I >> don't have a screen to follow what's going on. >> >> Thanks for the advice about iwlist, indeed "iw dev" gave me a good sca= n, >> once. Actually I didn't manage to make it work again. >=20 > This really looks like a hardware issue. Maybe your ath9k_htc device > is particularly sensitive to power supply noise or something like > that. Have you tried a different power supply or connecting it through > a powered hub? That can be an issue. After firmware is loaded power consumption goes from 30mA to 100-200mA. If it won't get enough power, then it will get strange issues. Second problem is noise on data or power lines. It depend on circuit design, cable and so on. Some users reported similar issues which was fixed after cable change. > I really wish I could help out more with this =3D) >=20 > I'd be surprised if it's a software issue, but if you're compiling > your own kernels, it might be worth checking out the Linux Backports > project: >=20 > https://backports.wiki.kernel.org/index.php/Main_Page >=20 > Which is essentially the latest kernel's WiFi drivers and stuff > repackaged so it can be compiled with older kernels. Yea... i will not continue this discussion untill you get updated driver.= --=20 Regards, Oleksij --10eDIKv5R3pQknWRqM9PeNvW8IHHntd3F Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iF4EAREIAAYFAlPbNeAACgkQHwImuRkmbWkB7gD5AWDAExtzJIT4KNgwSzSD8Z3s xvZQ8AVI05yhvBTf+VEA/1RJnZ2I8zVwvkPdEL6gIor8dEb2Rpqg5iBJ1wFSP2Pc =s1yE -----END PGP SIGNATURE----- --10eDIKv5R3pQknWRqM9PeNvW8IHHntd3F--