David Brownell wrote: > On Monday 14 November 2005 12:58 pm, Jan Kiszka wrote: > >>David Brownell wrote: >> >>>... >>>EBUSY status code happens in a few cases, and in this path I'd suspect >>>it means the URB is already being returned. Could you try the patch >>>I've attached, to see if it prints anything? >>> >> >>Additional question regarding this check: should this warning never show >>up or just not during that unwanted disconnect? I already have two of >>them in my log without any disconnects. > > > Sounds like you're getting more of them than you should. It's > something that ought to be almost vanishingly rare, which may > suggest something else is going wrong. > > - Dave >
These are two of now four dumps of my current 2.6.14.2 kernel: kernel: Badness in qh_link_async at drivers/usb/host/ehci-q.c:787 kernel: [pg0+241856579/1070355456] qh_link_async+0x53/0xb1 [ehci_hcd] kernel: [pg0+241857089/1070355456] submit_async+0x50/0x7f [ehci_hcd] kernel: [pg0+241871022/1070355456] ehci_urb_enqueue+0x49/0x99 [ehci_hcd] kernel: [pg0+242856890/1070355456] hcd_submit_urb+0x15b/0x1ae [usbcore] kernel: [pg0+242860976/1070355456] usb_start_wait_urb+0x53/0x166 [usbcore] kernel: [pg0+242861241/1070355456] usb_start_wait_urb+0x15c/0x166 [usbcore] kernel: [pg0+242860888/1070355456] timeout_kill+0x0/0x5 [usbcore] kernel: [pg0+242861369/1070355456] usb_internal_control_msg+0x76/0x87 [usbcore] kernel: [pg0+242861505/1070355456] usb_control_msg+0x77/0x8c [usbcore] kernel: [pg0+247333016/1070355456] RTUSB_VendorRequest+0xae/0x150 [rt2570] kernel: [pg0+247332221/1070355456] RTUSBMultiReadMAC+0x2a/0x2e [rt2570] kernel: [pg0+247244505/1070355456] PeriodicExec+0x44/0x784 [rt2570] kernel: [deactivate_task+22/36] deactivate_task+0x16/0x24 kernel: [_spin_unlock_irq+6/9] _spin_unlock_irq+0x6/0x9 kernel: [schedule+1150/1234] schedule+0x47e/0x4d2 kernel: [pg0+247235979/1070355456] CMDHandler+0x332/0xe14 [rt2570] kernel: [__wake_up_locked+16/21] __wake_up_locked+0x10/0x15 kernel: [__down+159/213] __down+0x9f/0xd5 kernel: [default_wake_function+0/18] default_wake_function+0x0/0x12 kernel: [pg0+247239968/1070355456] RTUSBCmdThread+0x74/0x8a [rt2570] kernel: [pg0+247239852/1070355456] RTUSBCmdThread+0x0/0x8a [rt2570] kernel: [kernel_thread_helper+5/11] kernel_thread_helper+0x5/0xb kernel: Badness in qh_link_async at drivers/usb/host/ehci-q.c:787 kernel: [pg0+241856579/1070355456] qh_link_async+0x53/0xb1 [ehci_hcd] kernel: [pg0+241857089/1070355456] submit_async+0x50/0x7f [ehci_hcd] kernel: [pg0+241871022/1070355456] ehci_urb_enqueue+0x49/0x99 [ehci_hcd] kernel: [pg0+242856890/1070355456] hcd_submit_urb+0x15b/0x1ae [usbcore] kernel: [pg0+247259468/1070355456] RTUSBKickBulkOut+0x80/0xdd [rt2570] kernel: [pg0+247271181/1070355456] PeerProbeReqAction+0x3d3/0x3de [rt2570] kernel: [deactivate_task+22/36] deactivate_task+0x16/0x24 kernel: [pg0+247241789/1070355456] MlmeHandler+0x104/0x25a [rt2570] kernel: [pg0+247239830/1070355456] MlmeThread+0x71/0x87 [rt2570] kernel: [pg0+247239717/1070355456] MlmeThread+0x0/0x87 [rt2570] kernel: [kernel_thread_helper+5/11] kernel_thread_helper+0x5/0xb Something rather for the rt25xx developers or actually a core issue? Jan
signature.asc
Description: OpenPGP digital signature
