This problem does not seem to be particularly related to USB. It appears to be a bug in the threaded IRQ subsystem. I don't know what I can do other than forward the message to people who might know more about it. Alan Stern On Sat, 22 Oct 2011, kanyck wrote: > Hello Alan and everybody, > > I apparently have the same problem but can't report it because the kernel.org > is still not fully up. So I apologize if I've put my report in a wrong place... > > I experienced very high iowait level during intensive disk I/O (up to 87% > per CPU) on all the kernels I tried (2.6.32 and Debian backports' 2.6.38, > 2.6.39) running on ASUS E35M1-M PRO (AMD e-350) platform > > So I installed AVLinux's 3.0.3 kernel: > Linux version 3.0.3-avl-3-pae (trulan@avlinux) (gcc version 4.6.1 (Debian > 4.6.1-4) ) #1 SMP PREEMPT Thu Aug 18 07:18:19 EDT 2011 > > It worked much better but after a couple of hours my ethernet stops working and > I see the following in my syslog: > > Oct 16 21:31:43 localhost kernel: [14328.121907] irq 18: nobody cared (try > booting with the "irqpoll" option) > Oct 16 21:31:43 localhost kernel: [14328.125161] Pid: 658, comm: > irq/18-ohci_hcd Not tainted 3.0.3-avl-3-pae #1 > Oct 16 21:31:43 localhost kernel: [14328.128863] Call Trace: > Oct 16 21:31:43 localhost kernel: [14328.132512] [<c1068fab>] ? > __report_bad_irq+0x1c/0x8d > Oct 16 21:31:43 localhost kernel: [14328.136168] [<c10692c3>] ? > note_interrupt+0x11a/0x187 > Oct 16 21:31:43 localhost kernel: [14328.139790] [<c10685df>] ? > irq_thread+0xbf/0x15c > Oct 16 21:31:43 localhost kernel: [14328.143373] [<c106869f>] ? > irq_thread_fn+0x23/0x23 > Oct 16 21:31:43 localhost kernel: [14328.146900] [<c1022289>] ? > complete+0x28/0x36 > Oct 16 21:31:43 localhost kernel: [14328.150385] [<c1068520>] ? > irq_finalize_oneshot+0x9d/0x9d > Oct 16 21:31:43 localhost kernel: [14328.153854] [<c1040587>] ? > kthread+0x63/0x68 > Oct 16 21:31:43 localhost kernel: [14328.157276] [<c1040524>] ? > kthread_worker_fn+0x101/0x101 > Oct 16 21:31:43 localhost kernel: [14328.160671] [<c1249a76>] ? > kernel_thread_helper+0x6/0xd > Oct 16 21:31:43 localhost kernel: [14328.164032] handlers: > Oct 16 21:31:43 localhost kernel: [14328.167331] [<c1068080>] > irq_default_primary_handler threaded [<ef7eccd5>] usb_hcd_irq > Oct 16 21:31:43 localhost kernel: [14328.168325] [<c1068080>] > irq_default_primary_handler threaded [<ef7eccd5>] usb_hcd_irq > Oct 16 21:31:43 localhost kernel: [14328.168325] [<c1068080>] > irq_default_primary_handler threaded [<ef7eccd5>] usb_hcd_irq > Oct 16 21:31:43 localhost kernel: [14328.168325] [<c1068080>] > irq_default_primary_handler threaded [<ef7eccd5>] usb_hcd_irq > Oct 16 21:31:43 localhost kernel: [14328.168325] [<c1068080>] > irq_default_primary_handler threaded [<ef9a54d9>] rtl8169_interrupt > Oct 16 21:31:43 localhost kernel: [14328.183633] Disabling IRQ #18 > > This happend repeatedly until I disabled threadirqs option. Now it is stable, > however I returned to high latencies and terrific iowaits... > > Please suggest how cold I get help? -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
