On 11/14/2013 02:45 PM, Peter Chen wrote: > > >>> [ 3.497987] using random self ethernet address >>> [ 3.503321] using random host ethernet address >>> [ 3.510076] g_ether gadget: Ethernet Gadget, version: Memorial Day >> 2008 >>> [ 3.516749] g_ether gadget: g_ether ready >>> [ 4.842006] irq 237: nobody cared (try booting with the "irqpoll" >> option) >>> [ 4.848855] CPU: 0 PID: 1 Comm: swapper Not tainted 3.12.0-rc6 #149 >>> [ 4.855218] [<c00144c8>] (unwind_backtrace+0x0/0xf0) from >> [<c0011ce4>] (show_stack+0x10/0x14) >>> [ 4.863808] [<c0011ce4>] (show_stack+0x10/0x14) from [<c00513a0>] >> (__report_bad_irq+0x20/0xc0) >>> [ 4.872463] [<c00513a0>] (__report_bad_irq+0x20/0xc0) from >> [<c0051848>] (note_interrupt+0x1d4/0x238) >>> [ 4.881636] [<c0051848>] (note_interrupt+0x1d4/0x238) from >> [<c004fac4>] (handle_irq_event_percpu+0xc4/0x264) >>> [ 4.891500] [<c004fac4>] (handle_irq_event_percpu+0xc4/0x264) from >> [<c004fca0>] (handle_irq_event+0x3c/0x5c) >>> [ 4.901365] [<c004fca0>] (handle_irq_event+0x3c/0x5c) from >> [<c0051fd0>] (handle_level_irq+0x8c/0xe8) >>> [ 4.910535] [<c0051fd0>] (handle_level_irq+0x8c/0xe8) from >> [<c004f358>] (generic_handle_irq+0x20/0x30) >>> [ 4.919880] [<c004f358>] (generic_handle_irq+0x20/0x30) from >> [<c000fde0>] (handle_IRQ+0x30/0x84) >>> [ 4.928705] [<c000fde0>] (handle_IRQ+0x30/0x84) from [<c0012764>] >> (__irq_svc+0x44/0x54) >>> [ 4.936757] [<c0012764>] (__irq_svc+0x44/0x54) from [<c0020248>] >> (__do_softirq+0x90/0x26c) >>> [ 4.945058] [<c0020248>] (__do_softirq+0x90/0x26c) from [<c00204f0>] >> (do_softirq+0x68/0x70) >>> [ 4.953442] [<c00204f0>] (do_softirq+0x68/0x70) from [<c00207e4>] >> (irq_exit+0xa4/0xf4) >>> [ 4.961395] [<c00207e4>] (irq_exit+0xa4/0xf4) from [<c000fde4>] >> (handle_IRQ+0x34/0x84) >>> [ 4.969353] [<c000fde4>] (handle_IRQ+0x34/0x84) from [<c0012764>] >> (__irq_svc+0x44/0x54) >>> [ 4.977396] [<c0012764>] (__irq_svc+0x44/0x54) from [<c0062c44>] >> (lock_acquire+0xac/0x104) >>> [ 4.985715] [<c0062c44>] (lock_acquire+0xac/0x104) from [<c046f76c>] >> (mutex_lock_nested+0x48/0x2d0) >>> [ 4.994815] [<c046f76c>] (mutex_lock_nested+0x48/0x2d0) from >> [<c02c211c>] (device_add+0x36c/0x608) >>> [ 5.003831] [<c02c211c>] (device_add+0x36c/0x608) from [<c035a990>] >> (mousedev_create+0x1f8/0x25c) >>> [ 5.012759] [<c035a990>] (mousedev_create+0x1f8/0x25c) from >> [<c067ca1c>] (mousedev_init+0x14/0x60) >>> [ 5.021758] [<c067ca1c>] (mousedev_init+0x14/0x60) from [<c00088ac>] >> (do_one_initcall+0xe8/0x154) >>> [ 5.030672] [<c00088ac>] (do_one_initcall+0xe8/0x154) from >> [<c0659ac8>] (kernel_init_freeable+0xec/0x1b4) >>> [ 5.040278] [<c0659ac8>] (kernel_init_freeable+0xec/0x1b4) from >> [<c0468188>] (kernel_init+0x8/0xe4) >>> [ 5.049363] [<c0468188>] (kernel_init+0x8/0xe4) from [<c000efa0>] >> (ret_from_fork+0x14/0x34) >>> [ 5.057723] handlers: >>> [ 5.060030] [<c03472b8>] ci_irq >>> [ 5.063205] Disabling IRQ #237 >> >> IRQ 237 belongs to the usb0, the OTG capable port. >> >>> grep ci_hdrc /proc/interrupts >>> 237: 100000 - 93 ci_hdrc_imx >>> 238: 0 - 92 ci_hdrc_imx >> >> I can reproduce the problem on another mx28 hardware, it happens only if >> a host type USB cable is plugged into the OTG port and connected to >> another machine's host port. > > What your otg port looks like? It is a Micro-B-Female port? > If ID pin is low, it will be considered as host role, when you > connect to another machine's host port, both sides are host, it can't > be enumerated. Yes, I know, both sides are in the host role. Granted my fault, but never the less, we get an IRQ storm and the IRQ gets disabled. You have to reboot to get the system working again. Marc -- Pengutronix e.K. | Marc Kleine-Budde | Industrial Linux Solutions | Phone: +49-231-2826-924 | Vertretung West/Dortmund | Fax: +49-5121-206917-5555 | Amtsgericht Hildesheim, HRA 2686 | http://www.pengutronix.de |
Attachment:
signature.asc
Description: OpenPGP digital signature
