On Thu, 26 Jun 2014, Dennis New wrote: > > Here's a revised patch for you to try. This is meant to apply on top > > of 3.15. > > > > As it turns out, when I tested this, it revealed a bug in the USB > > audio driver. So I'm including a work-around for that bug at the > > start of the patch. > > The bug seems to have been largely fixed with this patch!!1 > > Although ohci (and/or the usb audio device) does still crash, it does > so gracefully. mplayer and lsusb, etc continue working, unlike before. > I am still able to plug the usb device into another port, and it will > work, using 0000:00:13.1. I no longer seem to get the "timeout: still 2 > active urbs on EP #3" messages. > > [39209.542786] ohci-pci 0000:00:13.0: HcDoneHead not written back; > disabled > [39209.542796] ohci-pci 0000:00:13.0: HC died; cleaning up > [39209.542898] usb 2-4: USB disconnect, device number 5 That sounds like it's doing the right thing. > /sys/kernel/debug/usb/ohci/...13.0/registers when crashed says: > > bus pci, device 0000:00:13.0 > OHCI PCI host controller > ohci_hcd > OHCI 1.0, NO legacy support registers, rh state halted > control 0x283 RWC HCFS=operational CBSR=3 > cmdstatus 0x00000 SOC=0 > intrstatus 0x00000064 RHSC FNO SF > intrenable 0x00000004 SF > hcca frame 0x2e8d > fmintvl 0x27782edf FSMPS=0x2778 FI=0x2edf > fmremaining 0x000004c5 FR=0x04c5 > periodicstart 0x0000 > lsthresh 0x0628 > hub poll timer off > roothub.a 02000204 POTPGT=2 NPS NDP=4(4) > roothub.b 00000000 PPCM=0000 DR=0000 > roothub.status 00008000 DRWE > roothub.portstatus [0] 0x00000100 PPS > roothub.portstatus [1] 0x00000100 PPS > roothub.portstatus [2] 0x00000100 PPS > roothub.portstatus [3] 0x00030100 PESC CSC PPS > > (although the hcca frame counter is stopped, the fmremaining/FR values > are changing.) Doesn't matter. > echo 0000:00:13.0 >/sys/bus/pci/drivers/ohci_pci/unbind > > [39522.284135] ohci-pci 0000:00:13.0: remove, state 1 > [39522.284152] usb usb2: USB disconnect, device number 1 > [39522.285706] ohci-pci 0000:00:13.0: USB bus 2 deregistered > > echo 0000:00:13.0 >/sys/bus/pci/drivers/ohci_pci/bind > > [39530.642947] ohci-pci 0000:00:13.0: OHCI PCI host controller > [39530.643269] ohci-pci 0000:00:13.0: new USB bus registered, > assigned bus number 2 > [39530.643323] ohci-pci 0000:00:13.0: irq 19, io > mem 0xc0000000 > [other plugged-in ohci devices (i.e. 4port hub) detected] > > I wonder if this unbind'ing and bind'ing should be done automatically, > when a crash is detected? That's a separate discussion. To a large degree it doesn't matter, because most hardware is designed not to crash during use. In fact, if you wanted to, you could write a shell script that would monitor your system, looking for one of these crashes maybe once every few seconds. When a crash occurred, the script could then do the unbind and rebind. It doesn't need to be handled automatically in the kernel. Alan Stern -- 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
