http://bugzilla.novell.com/show_bug.cgi?id=567137 http://bugzilla.novell.com/show_bug.cgi?id=567137#c1 Jeff Mahoney changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO CC| |jeffm@novell.com Info Provider| |drankinatty@suddenlinkmail. | |com --- Comment #1 from Jeff Mahoney 2010-01-15 21:01:54 UTC --- What are you doing to blacklist ath5k? Blacklisting it should always work. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=567137 http://bugzilla.novell.com/show_bug.cgi?id=567137#c Jeff Mahoney changed: What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|kernel-maintainers@forge.pr |jslaby@novell.com |ovo.novell.com | -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=567137 http://bugzilla.novell.com/show_bug.cgi?id=567137#c4 David Rankin changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW InfoProvider|drankinatty@suddenlinkmail. | |com | --- Comment #4 from David Rankin 2010-04-04 01:36:56 UTC --- No choice, I can't update to 11.2, I tried, but opensuse does not have a fglrx driver for 11.2 for my card (x1200) and without the fglrx driver desktop effects crashes and spontaneously shuts down. The radeonhd driver is a no-go as well due to July 09 changes that left it blackscreening for my card. So 11.0 with the 8-9 ATI driver (fglrx 8.532) is the only workable solution. I know this same problem affects 11.1, because I had to compile the madwifi on buildservice after 11.1 was released. I don't think I had the same issue with 11.2 (I just don't recall). The Arch kernel is 2.6.33-1, and the madwifi driver has worked with all Arch Kernels since I started using them about a year ago. So I know all Arch kernels 2.6.32-2 through 2.6.33-1 work fine. I am no expect of kernel module loading, but I believe the problem with the madwifi driver is the openSuSE /lib/modules/<kernel>/modules.alias file which causes the proper module to fail to load. Specifically, I think suse has a problem with module.alias because it has duplicate entries for the same madwifi device that points to different modules (that can't be right). My Atheros card has the following ID: ath_pci: r3725+AR5007EG ACPI: PCI Interrupt 0000:17:00.0[A] -> GSI 19 (level, low) -> IRQ 19 ^^^^^^^^^^^^^ PCI: Setting latency timer of device 0000:17:00.0 to 64 SuSE module.alias has: alias pci:v00001FC1d00000010sv*sd*bc*sc*i* ib_ipath alias pci:v00001FC1d0000000Dsv*sd*bc*sc*i* ib_ipath alias pci:v0000168Cd00000024sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000023sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd0000001Csv*sd*bc*sc*i* ath5k alias pci:v0000168Cd0000001Bsv*sd*bc*sc*i* ath5k alias pci:v0000168Cd0000001Asv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000019sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000018sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000017sv*sd*bc*sc*i* ath5k ^^^^^^ ^^^^^ alias pci:v0000168Cd00000016sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000015sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000014sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00001014sv*sd*bc*sc*i* ath5k alias pci:v000010B7d00000013sv*sd*bc*sc*i* ath5k alias pci:v0000A727d00000013sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000013sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000012sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000011sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000007sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000207sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00009013sv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd00000024sv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd00000023sv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd0000001Dsv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd0000001Csv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd0000001Bsv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd0000001Asv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd00000019sv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd00000018sv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd00000017sv*sd*bc*sc*i* ath_pci ^^^^^^ ^^^^^ alias pci:v0000168Cd00000016sv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd00000015sv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd0000101Asv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd00001014sv*sd*bc*sc*i* ath_pci alias pci:v000010B7d00000013sv*sd*bc*sc*i* ath_pci alias pci:v0000A727d00000013sv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd00000013sv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd00000012sv*sd*bc*sc*i* ath_pci alias pci:v0000168Cd00000007sv*sd*bc*sc*i* ath_pci That is duplicate entries for 0000:17:00.0[A] with one pointing to ath5k and one pointing at ath_pci. Consequently it doesn't know what module to load. No with Arch, the Arch module.alias only has one entry for 0000:17:00.0[A]: alias pci:v0000168Cd0000001Dsv*sd*bc*sc*i* ath5k alias pci:v0000168Cd0000001Csv*sd*bc*sc*i* ath5k alias pci:v0000168Cd0000001Bsv*sd*bc*sc*i* ath5k alias pci:v0000168Cd0000001Asv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000019sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000018sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000017sv*sd*bc*sc*i* ath5k ^^^^^^ ^^^^^ alias pci:v0000168Cd00000016sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000015sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000014sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00001014sv*sd*bc*sc*i* ath5k alias pci:v000010B7d00000013sv*sd*bc*sc*i* ath5k alias pci:v0000A727d00000013sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000013sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000012sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000011sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000007sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd00000207sv*sd*bc*sc*i* ath5k alias pci:v0000168Cd0000002Esv*sd*bc*sc*i* ath9k alias pci:v0000168Cd0000002Dsv*sd*bc*sc*i* ath9k alias pci:v0000168Cd0000002Bsv*sd*bc*sc*i* ath9k alias pci:v0000168Cd0000002Asv*sd*bc*sc*i* ath9k alias pci:v0000168Cd00000029sv*sd*bc*sc*i* ath9k alias pci:v0000168Cd00000027sv*sd*bc*sc*i* ath9k alias pci:v0000168Cd00000024sv*sd*bc*sc*i* ath9k alias pci:v0000168Cd00000023sv*sd*bc*sc*i* ath9k with the single entry, my card is detected and loads every time. If you look at the dmesg and syslog messages posted above, I think this explains why my Atheros card is nearly impossible to get working on SuSE without compiling custom madwifi packages, standing on one foot, and wiggling your nose during the driver install and boot :p Let me know if you need more info and I'm happy to chase it down. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=567137 http://bugzilla.novell.com/show_bug.cgi?id=567137#c6 --- Comment #6 from David Rankin 2010-04-06 16:37:04 UTC --- Well, Like I said, I was no expert, but in comparing the files side by side, those were the differences that stood out. As for as the actual but ID, here is what I get with lspci -nn 17:00.0 Ethernet controller [0200]: Atheros Communications Inc. AR242x 802.11abg Wireless PCI Express Adapter [168c:001c] (rev 01) As far as the card not being supported -- I'm confused? I have used this laptop with 10.2 -> 11.2 and back to 11.0. In 10.2 & 10.3, the card was no problem to configure with madwifi. Just blacklist ath5k and madwifi worked instantly. With 11.0, 11.1, you could blacklist ath5k ten different ways and you were still hosed. It was like the blacklist didn't work because even when blacklisted, the ath5k module still got loaded (this may be where the actual bug is, perhaps the mismatch, wherever it may be, caused the blacklist to be ignored because of the card mis-ID?? (dunno, but it was hell getting madwifi configured)) The problem seems to be the binding of ath5k before ath_pci in modules.alias. I don't know if this is possible, but can that be switched, putting ath_pci before ath5k in modules.alias? That might solve the problem. IIRC, I had to physically remove /lib/modules/<kernel>/(any ath5k reference) before the init process would skip the ath5k module and actually load the ath_pci module. This wasn't a problem with 10.2-10.3. This is a problem with 11.0-11.1. On 11.2, the card is recognized and ath5k works fine. Same with Arch, ath5k just works fine. I still have madwifi installed for the encryption tools (wpa_passphrase, iwlist, etc.) modprobe -bnv ath5k, doesn't seem to do anything with the ath_pci module loaded. The ath5k module loads just fine if I boot to runlevel 1, but it will not work on 11.0, regardless of how it is configured. (wlan0 verses wifi0) Here is the output of the modprobe command while the ath_pci module is active with the "before-and-after" lsmod: [11:21 alchemy:/home/david] # lsmod | grep ath ath_rate_sample 31232 1 ath_pci 255168 0 wlan 265840 6 wlan_tkip,wlan_ccmp,wlan_scan_sta,ath_rate_sample,ath_pci ath_hal 346896 3 ath_rate_sample,ath_pci [11:21 alchemy:/home/david] # modprobe -bnv ath5k [11:21 alchemy:/home/david] # lsmod | grep ath ath_rate_sample 31232 1 ath_pci 255168 0 wlan 265840 6 wlan_tkip,wlan_ccmp,wlan_scan_sta,ath_rate_sample,ath_pci ath_hal 346896 3 ath_rate_sample,ath_pci Now I know 11.0 is reaching EOL, it has been a great release. I just wish the devs would have found a way to repackage the fglrx driver for pre-2400 series cards in 11.2 so that laptop users could move forward. (All laptops I buy henceforth will be nvidia based...) The Xorg version wasn't an issue with 11.2, so it was possible (I know ATI's screw the customer act of dropping all pre 2400 series card support in 3/09 complicated the matter, but that has left a lot of openSuSE laptop users stuck or shopping for new nvidia gpu laptops). The purpose of this report was to take a good look at a problem that still affects 11.1 users as well as the 11.0 users. For me it isn't a big deal, I have been working around it since 6/4/08, but the prospect of there being a problem with one of the base kernel module files prompted the report to help identify and fix the issue if possible. Thank you for your help. If you are satisfied that all issues concerning the load/bind order of the Atheros modules and ID of the various chipsets are where they should be for 11.1, then you can close this report. But I do think it is worth being sure of because I have had to walk a fair number of people on the user list though the blacklist (and a few through the temporary removal of /lib/modules/<kernel>/ath5k info) to get wifi working over the past year. Again, thank you for your help :-) -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=567137 http://bugzilla.novell.com/show_bug.cgi?id=567137#c Jiri Slaby changed: What |Removed |Added ---------------------------------------------------------------------------- Priority|P5 - None |P4 - Low Status|NEW |ASSIGNED -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=567137 http://bugzilla.novell.com/show_bug.cgi?id=567137#c9 --- Comment #9 from David Rankin 2010-04-19 14:06:46 UTC --- Hi Jiri, Sorry for the late reply. Right now I booted the 2.6.31.12-1-default kernel from build service. It works as it should with the ath5k kernel module, the sudo wireless interface as wlmaster0 and the actual interface as wlan0. The 11.0 kernel will not load ath5k and create any interface. I will have to get your information requested above when I boot back into the standard 11.0 kernel, but I thought I would provide this information for comparison purposes. The dmesg and syslog messages concerning wlan0 configuration were as follows: dmesg: [ 522.520089] cfg80211: Calling CRDA to update world regulatory domain [ 522.578303] ath5k 0000:17:00.0: PCI INT A -> GSI 19 (level, low) -> IRQ 19 [ 522.578315] ath5k 0000:17:00.0: setting latency timer to 64 [ 522.578405] ath5k 0000:17:00.0: registered as 'phy0' [ 523.073427] ath: EEPROM regdomain: 0x64 [ 523.073429] ath: EEPROM indicates we should expect a direct regpair map [ 523.073434] ath: Country alpha2 being used: 00 [ 523.073436] ath: Regpair used: 0x64 [ 523.194194] phy0: Selected rate control algorithm 'minstrel' [ 523.194935] ath5k phy0: Atheros AR2425 chip found (MAC: 0xe2, PHY: 0x70) [ 710.055318] r8169 Gigabit Ethernet driver 2.3LK-NAPI loaded [ 710.055347] r8169 0000:11:00.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18 [ 710.055421] r8169 0000:11:00.0: setting latency timer to 64 [ 710.055473] alloc irq_desc for 27 on node 0 [ 710.055476] alloc kstat_irqs on node 0 [ 710.055494] r8169 0000:11:00.0: irq 27 for MSI/MSI-X [ 710.056393] eth0: RTL8101e at 0xffffc90002e9c000, 00:1b:38:af:36:b8, XID 34200000 IRQ 27 [ 710.114882] r8169: eth0: link down [ 710.115766] ADDRCONF(NETDEV_UP): eth0: link is not ready [ 710.168998] ADDRCONF(NETDEV_UP): wlan0: link is not ready [ 712.435341] ip_tables: (C) 2000-2006 Netfilter Core Team [ 828.255838] r8169: eth0: link down [ 828.256707] ADDRCONF(NETDEV_UP): eth0: link is not ready [ 832.522878] ip6_tables: (C) 2000-2006 Netfilter Core Team [ 832.770856] r8169: eth0: link down [ 832.771763] ADDRCONF(NETDEV_UP): eth0: link is not ready [ 877.973676] NET: Registered protocol family 17 [ 879.007393] wlan0: authenticate with AP 00:23:69:5c:fd:b6 [ 879.008845] wlan0: authenticated [ 879.008850] wlan0: associate with AP 00:23:69:5c:fd:b6 [ 879.011126] wlan0: RX AssocResp from 00:23:69:5c:fd:b6 (capab=0x411 status=0 aid=1) [ 879.011130] wlan0: associated [ 879.012417] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready [ 879.038983] padlock: VIA PadLock not detected. [ 879.056375] Intel AES-NI instructions are not detected. Syslog: Apr 19 08:50:11 alchemy ifup: wlan0 device: Atheros Communications Inc. AR242x 802.11abg Wireless PCI Express Adapter (rev 01) Apr 19 08:50:11 alchemy ifup-wireless: wlan0 warning: WPA configured but may be unsupported Apr 19 08:50:11 alchemy ifup-wireless: wlan0 warning: by this device Apr 19 08:50:11 alchemy ifup-wireless: wlan0 starting wpa_supplicant Apr 19 08:50:11 alchemy kernel: [ 877.973676] NET: Registered protocol family 17 Apr 19 08:50:11 alchemy ifup-dhcp: Starting DHCP Client Daemon on wlan0... Apr 19 08:50:11 alchemy ifup-dhcp: . Apr 19 08:50:11 alchemy dhcpcd[8505]: wlan0: dhcpcd 3.2.3 starting Apr 19 08:50:11 alchemy dhcpcd[8505]: wlan0: hardware address = 00:1b:9e:7c:f6:e7 Apr 19 08:50:11 alchemy dhcpcd[8505]: wlan0: broadcasting for a lease Apr 19 08:50:12 alchemy kernel: [ 879.007393] wlan0: authenticate with AP 00:23:69:5c:fd:b6 Apr 19 08:50:12 alchemy kernel: [ 879.008845] wlan0: authenticated Apr 19 08:50:12 alchemy kernel: [ 879.008850] wlan0: associate with AP 00:23:69:5c:fd:b6 Apr 19 08:50:12 alchemy kernel: [ 879.011126] wlan0: RX AssocResp from 00:23:69:5c:fd:b6 (capab=0x411 status=0 aid=1) Apr 19 08:50:12 alchemy kernel: [ 879.011130] wlan0: associated Apr 19 08:50:12 alchemy kernel: [ 879.012417] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready Apr 19 08:50:12 alchemy kernel: [ 879.038983] padlock: VIA PadLock not detected. Apr 19 08:50:12 alchemy modprobe: WARNING: Error inserting padlock_aes (/lib/modules/2.6.31.12-1-default/kernel/drivers/crypto/padlock-aes.ko): No such device Apr 19 08:50:12 alchemy kernel: [ 879.056375] Intel AES-NI instructions are not detected. Apr 19 08:50:12 alchemy modprobe: WARNING: Error inserting aesni_intel (/lib/modules/2.6.31.12-1-default/kernel/arch/x86/crypto/aesni-intel.ko): No such device Apr 19 08:50:12 alchemy ifup-dhcp: . Apr 19 08:50:12 alchemy ifup-dhcp: . Apr 19 08:50:13 alchemy avahi-daemon[2354]: Registering new address record for fe80::21b:9eff:fe7c:f6e7 on wlan0.*. Apr 19 08:50:14 alchemy ifup-dhcp: . Apr 19 08:50:14 alchemy dhcpcd[8505]: wlan0: offered 192.168.6.102 from 192.168.6.17 Apr 19 08:50:14 alchemy dhcpcd[8505]: wlan0: checking 192.168.6.102 is available on attached networks Apr 19 08:50:15 alchemy ifup-dhcp: . Apr 19 08:50:15 alchemy dhcpcd[8505]: wlan0: leased 192.168.6.102 for 28800 seconds Apr 19 08:50:15 alchemy dhcpcd[8505]: wlan0: adding IP address 192.168.6.102/24 Apr 19 08:50:15 alchemy dhcpcd[8505]: wlan0: adding default route via 192.168.6.13 metric 0 Apr 19 08:50:15 alchemy dhcpcd[8505]: wlan0: adding route to 169.254.0.0/16 metric 0 Apr 19 08:50:15 alchemy avahi-daemon[2354]: Joining mDNS multicast group on interface wlan0.IPv4 with address 192.168.6.102. Apr 19 08:50:15 alchemy avahi-daemon[2354]: New relevant interface wlan0.IPv4 for mDNS. Apr 19 08:50:15 alchemy avahi-daemon[2354]: Registering new address record for 192.168.6.102 on wlan0.IPv4. Apr 19 08:50:15 alchemy modify_resolvconf: Service dhcpcd modified /etc/resolv.conf. See info block in this file Apr 19 08:50:15 alchemy dhcpcd[8505]: wlan0: exiting Apr 19 08:50:15 alchemy syslog-ng[1595]: SIGHUP received, restarting syslog-ng Apr 19 08:50:15 alchemy ifdown: wlan0 device: Atheros Communications Inc. AR242x 802.11abg Wireless PCI Express Adapter (rev 01) Apr 19 08:50:16 alchemy ifup: wlan0 device: Atheros Communications Inc. AR242x 802.11abg Wireless PCI Express Adapter (rev 01) Apr 19 08:50:16 alchemy syslog-ng[1595]: new configuration initialized Apr 19 08:50:16 alchemy ifup-dhcp: IP/Netmask: '192.168.6.102' Apr 19 08:50:16 alchemy ifup-dhcp: / '255.255.255.0' NOTE: all of the above is from the 2.6-31 kernel from build service :p I'll provide the information you requested from the standard kernel when I boot back into it in a little while. Thank you for your help. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=567137 http://bugzilla.novell.com/show_bug.cgi?id=567137#c11 Jiri Slaby changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |NEEDINFO InfoProvider| |mmarek@novell.com --- Comment #11 from Jiri Slaby 2010-04-28 21:43:21 UTC --- (In reply to comment #10)

So in sum, the problem is that ath5k in the standard kernel incorrectly recognizes the hardware and blacklisting ath5k is ignored leaving the wireless user screwed unless you go through the steps outlined above.

Michal, do you see immediately any reason for this, please? This is 11.0, so I think we're not going to invest much time into that unless it affects also 11.1/11.2. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=567137 http://bugzilla.novell.com/show_bug.cgi?id=567137#c Jiri Slaby changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |NEEDINFO InfoProvider| |drankinatty@suddenlinkmail. | |com -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=567137 http://bugzilla.novell.com/show_bug.cgi?id=567137#c14 Jiri Slaby changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |NEEDINFO InfoProvider| |mmarek@novell.com --- Comment #14 from Jiri Slaby 2010-06-03 20:28:31 UTC --- Michal, do you have any other ideas? -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=567137 http://bugzilla.novell.com/show_bug.cgi?id=567137#c16 Jiri Slaby changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |RESOLVED InfoProvider|drankinatty@suddenlinkmail. | |com | Resolution| |WONTFIX --- Comment #16 from Jiri Slaby 2010-07-23 09:04:18 UTC --- 11.0 is not supported anymore. Please try some later version. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.