Latest working kernel version: none I think. Before I used ath_pci and ath_hal Earliest failing kernel version: all wireless compat versions tested Distribution: Ubuntu 64 bits Hardware Environment: Samsung R20 laptop with AR242x (lspci) Software Environment: Problem Description: Wireless connection is first established (but not always) and disconnect rapidly. After it disconnected, it is no more able to restore. In the logs one can find a lot of these: Dec 6 14:45:40 portant kernel: [ 9775.586900] ------------[ cut here ]------------ Dec 6 14:45:40 portant kernel: [ 9775.586909] WARNING: at /home/antoine/compat-wireless-2008-12-02/net/mac80211/main.c:227 ieee80211_hw_config+0xac/0xb0 [mac80211]() Dec 6 14:45:40 portant kernel: [ 9775.586913] Modules linked in: nls_iso8859_1 nls_cp437 vfat fat ipv6 nfs lockd nfs_acl sunrpc af_packet radeon drm binfmt_misc ppdev acpi_cpufreq cpufreq_stats cpufreq_userspace cpufreq_powersave cpufreq_ondemand freq_table cpufreq_conservative pci_slot container wmi sbs sbshc iptable_filter ip_tables x_tables parport_pc lp parport snd_hda_intel snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_dummy snd_seq_oss snd_seq_midi psmouse snd_rawmidi serio_raw snd_seq_midi_event pcspkr joydev snd_seq snd_timer snd_seq_device snd soundcore i2c_piix4 snd_page_alloc i2c_core evdev video output arc4 ecb crypto_blkcipher ath5k mac80211 led_class cfg80211 battery ac button shpchp pci_hotplug ext3 jbd mbcache sr_mod cdrom usb_storage sd_mod crc_t10dif sg 8139cp pata_atiixp ata_generic usbhid hid libusual 8139too mii pata_acpi ahci ehci_hcd ohci_hcd libata scsi_mod dock usbcore thermal processor fan fbcon tileblit font bitblit softcursor fuse Dec 6 14:45:40 portant kernel: [ 9775.587008] Pid: 3052, comm: phy0 Tainted: G W 2.6.27-10-generic #1 Dec 6 14:45:40 portant kernel: [ 9775.587011] Dec 6 14:45:40 portant kernel: [ 9775.587012] Call Trace: Dec 6 14:45:40 portant kernel: [ 9775.587020] [<ffffffff8024e9b4>] warn_on_slowpath+0x64/0x90 Dec 6 14:45:40 portant kernel: [ 9775.587026] [<ffffffff80471d95>] ? net_ratelimit+0x15/0x20 Dec 6 14:45:40 portant kernel: [ 9775.587035] [<ffffffffa026706f>] ? ath5k_hw_reset+0xecf/0x1150 [ath5k] Dec 6 14:45:40 portant kernel: [ 9775.587041] [<ffffffff803a793c>] ? __ratelimit+0x7c/0xf0 Dec 6 14:45:40 portant kernel: [ 9775.587044] [<ffffffff80471d95>] ? net_ratelimit+0x15/0x20 Dec 6 14:45:40 portant kernel: [ 9775.587052] [<ffffffffa026a0af>] ? ath5k_reset+0x4f/0x240 [ath5k] Dec 6 14:45:40 portant kernel: [ 9775.587065] [<ffffffffa022040c>] ieee80211_hw_config+0xac/0xb0 [mac80211] Dec 6 14:45:40 portant kernel: [ 9775.587080] [<ffffffffa0224fc2>] ieee80211_scan_work+0x142/0x1b0 [mac80211] Dec 6 14:45:40 portant kernel: [ 9775.587093] [<ffffffffa0224e80>] ? ieee80211_scan_work+0x0/0x1b0 [mac80211] Dec 6 14:45:40 portant kernel: [ 9775.587098] [<ffffffff80262802>] run_workqueue+0xc2/0x1a0 Dec 6 14:45:40 portant kernel: [ 9775.587102] [<ffffffff80262b0f>] worker_thread+0xaf/0x130 Dec 6 14:45:40 portant kernel: [ 9775.587106] [<ffffffff80267050>] ? autoremove_wake_function+0x0/0x40 Dec 6 14:45:40 portant kernel: [ 9775.587110] [<ffffffff80262a60>] ? worker_thread+0x0/0x130 Dec 6 14:45:40 portant kernel: [ 9775.587114] [<ffffffff80266c1e>] kthread+0x4e/0x90 Dec 6 14:45:40 portant kernel: [ 9775.587118] [<ffffffff80213c99>] child_rip+0xa/0x11 Dec 6 14:45:40 portant kernel: [ 9775.587122] [<ffffffff80266bd0>] ? kthread+0x0/0x90 Dec 6 14:45:40 portant kernel: [ 9775.587125] [<ffffffff80213c8f>] ? child_rip+0x0/0x11 Dec 6 14:45:40 portant kernel: [ 9775.587126] Dec 6 14:45:40 portant kernel: [ 9775.587129] ---[ end trace e1b79c34e461ede4 ]--- * WEP: it happens sometimes after suspend. * WPA2: happens each time Steps to reproduce: * simply connect thanks to Network Manager * Wait some time and see that wireless is disconnected * Look at the logs and verify it is the same error
Created attachment 19196 [details] kern.log
Created attachment 19197 [details] syslog
Echoing this bug - 2.6.27-9 with wireless-testing from December 2nd, an AR5212 miniPCIe.
I think this bug is a duplicate of #12080 seen the "ath5k phy0: can't reset hardware (-11)" messages in your kernel.
Yes it seems to be *** This bug has been marked as a duplicate of bug 12068 ***