Am 09.08.2011 15:47, schrieb Didier Raboud:
I would like to have Josua's (upstream author) input on this bug. I think
"doing nothing but closing as such" would be OKay, but his take on that would
be valuable.
Everything seems fine ... Just one thing kept bothering me:
kernel: [48972.216084] usb 6-1: USB disconnect, device number 2
kernel: [48972.952054] usb 6-1: new full speed USB device number 3 using
uhci_hcd
This is exactly the behaviour you see during a mode switch. And now I remembered
- there is a switching routine for the early Huawei modems built into the
usb-storage driver! I have all but forgotten; at the time, there was a
discussion about adding further mode switching code to the kernel.
The USB developers decided against that ("what CAN be done in userspace, SHOULD
be done in userspace) but at the time there was already code added for some few
devices. This has stayed in the kernel for compatibility reasons.
In the kernel's usb-storage folder, there is the file "unusual_devs.h" which has
this entry:
UNUSUAL_DEV( 0x12d1, 0x1001, 0x0000, 0x0000,
"HUAWEI MOBILE",
"Mass Storage",
USB_SC_DEVICE, USB_PR_DEVICE, usb_stor_huawei_e220_init,
0),
The mode-switching function is in usb_stor_huawei_e220_init, obviously. It does
exactly what usb_modeswitch's "--huawei-mode" would do.
One of the main problems of that built-in function is that you can't disable it,
except by hacking the said file and recompiling the module.
It would be impossible for you to check the content of the install storage.
Plus, we will not know if your device is actually switching modes or just
resetting once ...
Anyway, I think the bug can be closed.
Regards,
Josh
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
