hal-info: Do not mount the CD drive of the Huawei E220 USB HSDPA modem

Bastien Nocera hadess at hadess.net
Tue Mar 25 03:21:08 PDT 2008


On Sat, 2008-03-22 at 12:42 +0000, Sam Morris wrote:
> On Mon, 2008-03-10 at 21:02 +0000, Bastien Nocera wrote:
> > On Mon, 2008-03-10 at 18:13 +0000, Sam Morris wrote:
> > > Package: hal-info
> > > Version: 20080215-1
> > > Severity: normal
> > > Tags: patch
> > > 
> > > The Huawei E220 is a USB HSPDA/GPRS modem. It appears to the computer as
> > > a USB CD drive, on which are drivers for Windows systems. Of course, we
> > > need no such drivers, so the CD drive is useless and annoying as it is
> > > mounted by GNOME whenever a user logs in.
> > > 
> > > The attached patch (taken from
> > > <https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.22/+bug/105545>)
> > > causes HAL to ignore the CD drive.
> > 
> > That's wrong. There's a blacklist in the kernel for that. See:
> > http://thread.gmane.org/gmane.linux.kernel/597829
> > 
> > It's even mentioned in the bug you linked. The quirk in the kernel tree
> > is supposed to poke the device into serial mode, from storage mode. Why
> > is that not working?
> 
> Right, I spent some time looking into this. There are three commits that
> matter.
> 
> The first one use the UNUSUAL_DEV macro to tell the usb-storage code to
> ignore the device completely. This was present in 2.6.20 and can be seen
> at
> <http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=16c76865df40357027479b6d85f59a07f8d01c8d>.
> 
> The second commit reverted this patch; it was present in 2.6.22 and can
> be seen at
> <http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=862e9043776578d575494db79d9189795350ed67>.

That seems quite backwards to me...

> The third commit,
> <http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=d853d872c14b9adc4adad29e56cd378b291f86e0> is present in 2.6.24. It brings back an UNUSUAL_DEV entry for the Huawei device, but look at the final argument to the macro ('flags').
> 
> In the first commit, it was 'US_FL_IGNORE_DEVICE'. In the third commit,
> it is instead '0'. So that explains why the kernel does not ignore the
> usb-storage part of the device since 2.6.24, and why I think hal-info
> should be updated with a blacklist entry for this device.

I'm not really sure what the point is to have Linux drivers on the
device when the kernel carries it out-of-the-box, but it looks like the
HAL patch is required.

No complaints from me.



More information about the hal mailing list