Crash in timer.c
I had 128 STA vifs on an ath9k system...started deleting all but 16 of them. I did this twice, and both times, it crashes before finishing deleting all of them.
This is using latest wireless-testing plus some of my patches previously posted to this list.
------------[ cut here ]------------ kernel BUG at /home/greearb/git/linux.wireless-testing/kernel/timer.c:1005! invalid opcode: 0000 [#1] SMP last sysfs file: /sys/devices/pci0000:00/0000:00:1e.0/0000:03:00.0/net/sta111/flags Modules linked in: cryptd aes_i586 aes_generic 8021q garp stp llc michael_mic macvlan pktgen fuse coretemp hwmon nfs lockd fsc]
Pid: 9998, comm: wpa_supplicant Not tainted 2.6.38-rc3-wl+ #42 To be filled by O.E.M./To Be Filled By O.E.M.
EIP: 0060:[<c0443d87>] EFLAGS: 00210092 CPU: 1
EIP is at cascade+0x45/0x63
EAX: 00060008 EBX: 00000008 ECX: f54d8934 EDX: f123c908
ESI: 00000000 EDI: f54d8000 EBP: f54e9f7c ESP: f54e9f64
DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
Process wpa_supplicant (pid: 9998, ti=f54e8000 task=f17ee720 task.ti=f081e000)
Stack: f54e9f68 f1158064 f4a2d664 f54d8000 00000000 00000004 f54e9fd0 c04451dd f54e9fbc f54e9fac 00000286 f54d8e2c f54d8c2c f54d8a2c f54d882c f54e9fc8 c0486a6b 00000003 f17ee720 c043fb2f c043fade f54e9fc8 c0460dec 00000202 Call Trace: [<c04451dd>] run_timer_softirq+0x95/0x232 [<c0486a6b>] ? __rcu_process_callbacks+0x6b/0x26f [<c043fb2f>] ? __do_softirq+0x51/0x184 [<c043fade>] ? __do_softirq+0x0/0x184 [<c0460dec>] ? trace_hardirqs_on_caller+0xf5/0x12f [<c043fb94>] __do_softirq+0xb6/0x184 [<c043fade>] ? __do_softirq+0x0/0x184 <IRQ> [<c043f976>] ? irq_exit+0x3a/0x6e [<c041a713>] ? smp_apic_timer_interrupt+0x73/0x81 [<c07e2e0f>] ? apic_timer_interrupt+0x2f/0x40 [<c04f2a5f>] ? dput+0x1/0xde [<c04ea6c0>] ? path_put+0x13/0x1d [<c047fd0b>] ? audit_free_names+0x45/0x5e [<c047feb2>] ? audit_syscall_exit+0x96/0x114 [<c04031ea>] ? sysexit_audit+0x29/0x4f Code: ca 89 75 ec 89 46 04 8b 74 da 04 89 75 f0 89 06 89 0c da 89 4c da 04 8b 55 ec 8b 32 89 45 e8 eb 19 8b 42 0c 83 e0 fe 39 EIP: [<c0443d87>] cascade+0x45/0x63 SS:ESP 0068:f54e9f64
Signature · 2 lines
Ben Greear <[email protected]> Candela Technologies Inc http://www.candelatech.com
Quoted 6 lines · Ben Greear
I had 128 STA vifs on an ath9k system...started deleting all but 16 of them. I did this twice, and both times, it crashes before finishing deleting all of them.
This is using latest wireless-testing plus some of my patches previously posted to this list.
Timers should have debug objects support -- that would be helpful.
johannes
Quoted 9 lines · Johannes Berg
Quoted 6 lines · Ben Greear
I had 128 STA vifs on an ath9k system...started deleting all but 16 of them. I did this twice, and both times, it crashes before finishing deleting all of them.
This is using latest wireless-testing plus some of my patches previously posted to this list.
Timers should have debug objects support -- that would be helpful.
Yeah, I was hitting a similar crash before..though I had more debugging enabled then. Someone posted ath9k patches that they thought would fix it..but not sure they did.
I also don't remember seeing this while I had copybreak in my tree (I recently removed it to make way for the paged skb code...) That could just be a coincidence though....
Thanks,
Ben
Quoted 2 lines
johannes
Signature · 2 lines
Ben Greear <[email protected]> Candela Technologies Inc http://www.candelatech.com