All of lore.kernel.org
 help / color / mirror / Atom feed
From: Hans de Goede <hdegoede@redhat.com>
To: Jan Kiszka <jan.kiszka@web.de>,
	linux-usb <linux-usb@vger.kernel.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Subject: Re: uas: module not loaded automatically
Date: Sun, 05 Oct 2014 11:54:17 +0200	[thread overview]
Message-ID: <54311549.4070808@redhat.com> (raw)
In-Reply-To: <543113FF.7050900@web.de>

Hi,

On 10/05/2014 11:48 AM, Jan Kiszka wrote:
> On 2014-10-05 11:43, Hans de Goede wrote:
>> Hi,
>>
>> On 10/05/2014 11:31 AM, Jan Kiszka wrote:
>>> On 2014-10-05 11:23, Hans de Goede wrote:
>>>> Hi,
>>>>
>>>> On 10/05/2014 11:14 AM, Jan Kiszka wrote:
>>>>> On 2014-10-05 11:08, Hans de Goede wrote:
>>>>>> Hi,
>>>>>>
>>>>>> On 10/04/2014 08:35 PM, Jan Kiszka wrote:
>>>>>>> Hi,
>>>>>>>
>>>>>>> my Delock external USB drive stopped working after updating from a
>>>>>>> UAS-disabled distro kernel to latest 3.17-rc7 with UAS on. That UAS was
>>>>>>> key became clear to me only after looking at storage_probe(): the device
>>>>>>> is ignored by usb-storage if it is UAS-capable. However, nothing causes
>>>>>>> uas.ko to be loaded when the drive is plugged here. How is this supposed
>>>>>>> to work in the normal case?
>>>>>>
>>>>>> Is the uas.ko module installed, and was depmod run after installing it ?
>>>>>
>>>>> Definitely. Just retried after another depmod -a, and only modprobe uas
>>>>> made it work.
>>>>>
>>>>>>
>>>>>> uas.c has the following:
>>>>>>
>>>>>>         { USB_INTERFACE_INFO(USB_CLASS_MASS_STORAGE, USB_SC_SCSI, USB_PR_BULK) },
>>>>>>         { USB_INTERFACE_INFO(USB_CLASS_MASS_STORAGE, USB_SC_SCSI, USB_PR_UAS) },
>>>>>>
>>>>>> Which should make it load automatically on your device.
>>>>>
>>>>> Should this match with what lsusb -v reports for the device? 
>>>>
>>>> Yes, and it does, for both alt settings of your device:
>>>>
>>>>       bInterfaceClass         8 Mass Storage
>>>>       bInterfaceSubClass      6 SCSI
>>>>       bInterfaceProtocol     80 Bulk-Only
>>>>
>>>>       bInterfaceClass         8 Mass Storage
>>>>       bInterfaceSubClass      6 SCSI
>>>>       bInterfaceProtocol     98
>>>>
>>>> Where 98 == USB_PR_UAS, seems you have an quite old lsusb if it does not
>>>
>>> It says "007".
>>>
>>>> know that though. Could it be the rest of your userspace is old too, and is
>>>> not smart enough to load all matching drivers, instead only loading the first matching
>>>> driver (which happens to be usb-storage) ?
>>>
>>> OpenSUSE 13.1, all updates installed. Which components are involved?
>>> udev - Version 208? Below is the udevadm monitor output.
>>
>> Hmm, that is not all that old, I would expect that to work.
>>
>> Here is how things "work" on my system:
>>
>> [hans@shalem ~]$ lsmod | grep uas
>> uas                    22414  0
>> usb_storage            65065  1 uas
>> [hans@shalem ~]$ sudo rmmod uas
>> [hans@shalem ~]$ lsmod | grep uas
>> [hans@shalem ~]$ sudo modprobe usb:v174Cp5136d0100dc00dsc00dp00ic08isc06ip50in00
>> [hans@shalem ~]$ lsmod | grep uas
>> uas                    22414  0
>> usb_storage            65065  1 uas
>>
>> So as you can see the modalias taken from your udev debug output causes
>> uas to get loaded, can you try the above ?
> 
> That command sequence doesn't cause uas to be reloaded. Anything I need
> to customize for my setup?

Not that I know of, so this seems to be a modprobe issue, and this should
probably be taken to the relevant list for modprobe. Please put me in the CC
when you send a mail there, and you likely will want to include the output
of the above commands + the modinfo output for both uas and usb-storage.

> 
>>
>> Also what does "modinfo uas" say? For me it says:
>>
>> [hans@shalem ~]$ modinfo uas
>> filename:       /lib/modules/3.17.0-rc6+/kernel/drivers/usb/storage/uas.ko
>> author:         Hans de Goede <hdegoede@redhat.com>, Matthew Wilcox and Sarah Sharp
>> license:        GPL
>> alias:          usb:v*p*d*dc*dsc*dp*ic08isc06ip62in*
>> alias:          usb:v*p*d*dc*dsc*dp*ic08isc06ip50in*
>> alias:          usb:v174Cp5106d*dc*dsc*dp*ic*isc*ip*in*
>> alias:          usb:v152Dp0567d*dc*dsc*dp*ic*isc*ip*in*
>> alias:          usb:v0BC2pAB20d*dc*dsc*dp*ic*isc*ip*in*
>> alias:          usb:v0BC2p3312d*dc*dsc*dp*ic*isc*ip*in*
>> alias:          usb:v0BC2p2312d*dc*dsc*dp*ic*isc*ip*in*
>> depends:        usb-storage
>> vermagic:       3.17.0-rc6+ SMP mod_unload
>> signer:         Magrathea: Glacier signing key
>> sig_key:        9B:56:00:B2:C4:97:8D:4A:A9:B3:0B:54:32:F7:B7:B2:2F:3E:FB:D8
>> sig_hashalgo:   sha256
>>
>> Note the alias-es with vendor and product ids are from quirks, and your version
>> will likely not have these. But the first 2 generic ones should be there, and
>> match the modalias from the udev output.
> 
> filename:       /lib/modules/3.17.0-rc7-homebrewed+/kernel/drivers/usb/storage/uas.ko
> author:         Hans de Goede <hdegoede@redhat.com>, Matthew Wilcox and Sarah Sharp
> license:        GPL
> srcversion:     597B27EF314ADC559827CBD
> alias:          usb:v*p*d*dc*dsc*dp*ic08isc06ipAAin*
> alias:          usb:v*p*d*dc*dsc*dp*ic08isc06ip62in*
> alias:          usb:v*p*d*dc*dsc*dp*ic08isc06ip50in*
> depends:        usb-storage
> intree:         Y
> vermagic:       3.17.0-rc7-homebrewed+ SMP preempt mod_unload modversions 

That looks good, so no idea why modprobe is not loading it.

Regards,

Hans



  reply	other threads:[~2014-10-05  9:54 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-10-04 18:35 uas: module not loaded automatically Jan Kiszka
2014-10-05  9:08 ` Hans de Goede
2014-10-05  9:14   ` Jan Kiszka
2014-10-05  9:23     ` Hans de Goede
2014-10-05  9:31       ` Jan Kiszka
2014-10-05  9:43         ` Hans de Goede
2014-10-05  9:48           ` Jan Kiszka
2014-10-05  9:54             ` Hans de Goede [this message]
2014-10-05 11:03               ` uas " Jan Kiszka
2014-10-06  6:39                 ` [solved] " Jan Kiszka
2014-10-06 22:43                   ` Lucas De Marchi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=54311549.4070808@redhat.com \
    --to=hdegoede@redhat.com \
    --cc=jan.kiszka@web.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-usb@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.