On Tue, 2015-03-24 at 18:02 +0700, Lars Melin wrote: > On 2015-03-24 17:20, Oliver Neukum wrote: > > On Mon, 2015-03-23 at 21:11 +0200, Sami Farin wrote: > >> I tried reverting f46ad73ac696 in 4.0.0-rc5, did not help. Have to > >> reboot first into 3.18.9. > >> > >> This happens: > >> [ 2765.268081] usb 2-1.4.4: USB disconnect, device number 16 > >> [ 2777.209388] usb 2-1.4.4: new high-speed USB device number 17 using > >> ehci-pci > >> [ 2777.303172] usb 2-1.4.4: New USB device found, idVendor=12d1, > >> idProduct=157d > >> [ 2777.303174] usb 2-1.4.4: New USB device strings: Mfr=1, Product=2, > >> SerialNumber=3 > >> [ 2777.303176] usb 2-1.4.4: Product: HUAWEI_MOBILE > >> [ 2777.303177] usb 2-1.4.4: Manufacturer: HUAWEI_MOBILE > >> [ 2777.303178] usb 2-1.4.4: SerialNumber: 0123456789ABCDEF > >> [ 2777.325198] usb-storage 2-1.4.4:1.0: USB Mass Storage device > >> detected > >> [ 2777.325282] scsi host37: usb-storage 2-1.4.4:1.0 > >> [ 2777.903039] usb 2-1.4.4: usbfs: interface 0 claimed by usb-storage > >> while 'usb_modeswitch' sets config #2 > > This fails. It comes from proc_setconfig(). It suggests that a change > > in storage causes this. To verify, could you blacklist usb-storage > > in a test with 4.0? > > > > Regards > > Oliver > 12d1:157d has 2 configurations, config #1 being the virtual cd-rom > containing windows drivers and config #2 containing cdc_mbim interfaces. > cdc_mbim is the primary protocol for this Win8 approved dongle but it > can be switched via a bulkmsg into an "old Win" compatibility mode which > usually is cdc_ether. Yes. But the dmesg makes clear that usb_modeswitch had decided to go to config#2. It just fails. So we must look for the reason of the failure. And that is likely in the storage driver binding to the virtual CD. At this point that is just theory so I would like to see this confirmed by blacklisting usb-storage. Regards Oliver -- 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
