Bug 12559

Reply-To: akpm@linux-foundation.org


(switched to email.  Please respond via emailed reply-to-all, not via the
bugzilla web interface).

On Wed, 28 Jan 2009 02:34:10 -0800 (PST) bugme-daemon@bugzilla.kernel.org wrote:

> http://bugzilla.kernel.org/show_bug.cgi?id=12559
> 
>            Summary: Huawei E169 doesn't work as mass storage anymore
>            Product: Drivers
>            Version: 2.5
>      KernelVersion: 2.6.28.2
>           Platform: All
>         OS/Version: Linux
>               Tree: Mainline
>             Status: NEW
>           Severity: normal
>           Priority: P1
>          Component: USB
>         AssignedTo: greg@kroah.com
>         ReportedBy: kpalberg@gmail.com
> 
> 
> Latest tested working kernel version: 2.6.27.7
> Earliest tested failing kernel version: 2.6.28.2

A regression.

> Distribution: Debian Etch
> Hardware Environment: x86
> Software Environment:
> 
> Problem Description:
> Huawei E169 HSDPA USB modem (ID 12d1:1001) also functions as a USB memory
> stick. It has a microsd memory card slot for this. This USB storage is no
> longer recognized. It also has a CD-ROM mass storage device for the windows
> drivers and this is not recognized either (doesn't matter).
> 
> Both storages used to work. The following diff is what caused the regression
> and reversing it makes 2.6.28.2 work correctly:
> 
> --- linux-2.6.27.7/drivers/usb/storage/initializers.c   2008-11-21
> 01:02:37.000000000 +0200
> +++ linux-2.6.28.2/drivers/usb/storage/initializers.c   2009-01-25
> 02:42:07.000000000 +0200
> @@ -95,11 +95,10 @@
>  {
>         int result;
> 
> -       us->iobuf[0] = 0x1;
>         result = usb_stor_control_msg(us, us->send_ctrl_pipe,
>                                       USB_REQ_SET_FEATURE,
>                                       USB_TYPE_STANDARD | USB_RECIP_DEVICE,
> -                                     0x01, 0x0, us->iobuf, 0x1, 1000);
> +                                     0x01, 0x0, NULL, 0x0, 1000);
>         US_DEBUGP("usb_control_msg performing result is %d\n", result);
>         return (result ? 0 : -1);
>  }
> 
> 
> Steps to reproduce:
> Attach the stick.
> 

Apparenty caused by

commit 1460e5e44cc5ecad7704f63b10dcb3a59d0e008b
Author:     fangxiaozhi <huananhu@huawei.com>
AuthorDate: Wed Oct 15 14:15:36 2008 +0800
Commit:     Greg Kroah-Hartman <gregkh@suse.de>
CommitDate: Wed Oct 22 10:05:28 2008 -0700

    USB: support Huawei data card product IDs

On Monday 09 February 2009, kpalberg wrote:
> On Wed, Feb 4, 2009 at 12:58 PM, Rafael J. Wysocki <rjw@sisk.pl> wrote:
> > The following bug entry is on the current list of known regressions
> > introduced between 2.6.27 and 2.6.28.  Please verify if it still should
> > be listed and let me know (either way).
> >
> >
> > Bug-Entry       : http://bugzilla.kernel.org/show_bug.cgi?id=12559
> > Subject         : Huawei E169 doesn't work as mass storage anymore
> > Submitter       : kpalberg <kpalberg@gmail.com>
> > Date            : 2009-01-28 02:34 (8 days old)
> 
> Yes, it should be listed. The problem still exists in 2.6.28.4.

On Tuesday 24 February 2009, Greg KH wrote:
> On Mon, Feb 23, 2009 at 11:00:52PM +0100, Rafael J. Wysocki wrote:
> > Bug-Entry   : http://bugzilla.kernel.org/show_bug.cgi?id=12559
> > Subject             : Huawei E169 doesn't work as mass storage anymore
> > Submitter   : kpalberg <kpalberg@gmail.com>
> > Date                : 2009-01-28 02:34 (27 days old)
> 
> I'm pretty sure this has been tracked down to a buggy firmware in the
> device itself, and the company (Huawei) is working on updating it to
> solve the issue.
> 
> This is backed up by the fact that Huawei submitted the patch that is
> said to have broken this device :)

I have E220 and E176 devices here for testing.

E220 and E176 have the same device ID (0x1003). Both have virtual mass-storage CD-ROM with Windows software. But E176 has additional microSD card slot that acts as mass-storage disk. The usb_stor_huawei_e220_init() function does a bad thing: it disables the mass-storage on Huawei devices. The old one was buggy so it did nothing with newer devices - and that's why the microSD slot worked. The patched ("new") function (introduced in 2.6.28) is fixed and works on all devices, disabling everything mass-storage.

I think that the entire usb_stor_huawei_e220_init() function in initializers.c and all Huawei entries from unusual_devs.h should be removed. At least for E220 and E176, it does not do any good thing. The "option" module works fine without this "initialization"!

Also I don't think that we should hide the virtual CD-ROM on kernel level. It's already hidden by HAL. And AFAIK, it can be overwritten with data that can be useful even in Linux.