linux-kernel - Re: btusb "firmware request while host is not available" at resume

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Mon, 11 Sep 2017 05:49:44 +0200
From:   Gabriel C <nix.or.die@...il.com>
To:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        "Luis R. Rodriguez" <mcgrof@...nel.org>
Cc:     Tejun Heo <tj@...nel.org>, Marcel Holtmann <marcel@...tmann.org>,
        Gustavo Padovan <gustavo@...ovan.org>,
        Sukumar Ghorai <sukumar.ghorai@...el.com>,
        "Rafael J. Wysocki" <rjw@...ysocki.net>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        "bluez mailin list (linux-bluetooth@...r.kernel.org)" 
        <linux-bluetooth@...r.kernel.org>
Subject: Re: btusb "firmware request while host is not available" at resume

On 11.09.2017 05:15, Gabriel C wrote:
> On 11.09.2017 03:25, Greg Kroah-Hartman wrote:
>> On Sun, Sep 10, 2017 at 12:26:02PM -0700, Linus Torvalds wrote:
>>> This seems to be a new problem at resume for the Intel btusb driver,
>>> but I'm not seeing anything in that driver itself that looks like a
>>> likely trigger, so I wonder if it's some driver core change, a generic
>>> resume path issue, or a workqueue change that has made it trigger for
>>> me.
>>>
>>> It might also just be a timing difference, maybe it's always been there?
>>>
>>> Does anybody have any ideas? It does't happen on every resume, and the
>>> machine works despite this (but no bluetooth - the *next* resume might
>>> bring it back, though).
>>
>> Ah, it's not just me having this problem.  I don't see it happening in
>> 4.12, and haven't had the time to bisect it.  I seem to be able to
>> trigger it every suspend/resume cycle, so I don't know if it's a timing
>> issue.
> 
> 
> I see the same problem with QCA hardware.. but a bit different.
> 
> On first resume cycle the firmware call is fine but the adapter dies a bit later with :
> 
> 
>   'Bluetooth: hci0: Failed to send body at 4 of 1857 (-110)'
> 
> On second resume cycle I hit the trace too.
> 
> 
>>
>>>
>>>                   Linus
>>>
>>> -- 
>>>
>>>    ACPI: Low-level resume complete
>>>    ACPI: EC: EC started
>>>    PM: Restoring platform NVS memory
>>>    Enabling non-boot CPUs ...
>>>    x86: Booting SMP configuration:
>>>    smpboot: Booting Node 0 Processor 1 APIC 0x2
>>>     cache: parent cpu1 should not be sleeping
>>>    CPU1 is up
>>>    smpboot: Booting Node 0 Processor 2 APIC 0x1
>>>     cache: parent cpu2 should not be sleeping
>>>    CPU2 is up
>>>    smpboot: Booting Node 0 Processor 3 APIC 0x3
>>>     cache: parent cpu3 should not be sleeping
>>>    CPU3 is up
>>>    ACPI: Waking up from system sleep state S3
>>>    ACPI: EC: event unblocked
>>>    usb 1-3: reset full-speed USB device number 2 using xhci_hcd
>>>    usb 1-4: reset full-speed USB device number 3 using xhci_hcd
>>>    usb 1-5: reset high-speed USB device number 4 using xhci_hcd
>>>    usb 1-3:1.0: rebind failed: -517
>>>    usb 1-3:1.1: rebind failed: -517
>>>    Bluetooth: hci0: Bootloader revision 0.0 build 2 week 52 2014
>>>    OOM killer enabled.
>>>    Restarting tasks ...
>>>    Bluetooth: hci0: Device revision is 5
>>>    Bluetooth: hci0: Secure boot is enabled
>>>    Bluetooth: hci0: OTP lock is enabled
>>>    Bluetooth: hci0: API lock is enabled
>>>    Bluetooth: hci0: Debug lock is disabled
>>>    Bluetooth: hci0: Minimum firmware build 1 week 10 2014
>>>    firmware request while host is not available
>>>    ------------[ cut here ]------------
>>>    WARNING: CPU: 2 PID: 621 at drivers/base/firmware_class.c:1250
>>> _request_firmware+0x460/0x790
>>>    CPU: 2 PID: 621 Comm: kworker/u9:2 Not tainted 4.13.0-10313-ge860d2c904d1 #11
>>>    Hardware name: Dell Inc. XPS 13 9350/09JHRY, BIOS 1.4.17 05/10/2017
>>>    Workqueue: hci0 hci_power_on [bluetooth]
>>>    task: ffff8d3767895ac0 task.stack: ffff9d3481efc000
>>>    RIP: 0010:_request_firmware+0x460/0x790
>>>    Call Trace:
>>>     request_firmware+0x37/0x50
>>>     btusb_setup_intel_new+0x227/0x7e0 [btusb]
>>>     hci_dev_do_open+0x3da/0x570 [bluetooth]
>>>     hci_power_on+0x52/0x1f0 [bluetooth]
>>>     process_one_work+0x1db/0x3d0
>>>     worker_thread+0x47/0x3e0
>>>     kthread+0x125/0x140
>>>     ret_from_fork+0x22/0x30
>>>    ---[ end trace 007b222491432927 ]---
>>>    Bluetooth: hci0: Failed to load Intel firmware file (-112)
>>>    [drm] RC6 on
>>>    done.
>>>    thermal thermal_zone11: failed to read out thermal zone (-5)
>>>    PM: suspend exit
>>
>> Ah, I'll blame Luis for this, I think it might be due to 81f95076281f
>> ("firmware: add sanity check on shutdown/suspend")
>>
>> Luis, any ideas?  I'll try to revert this and try it out tomorrow when
>> I get a chance.
>>
> 
> I can revert it an fire up some testing..
> 

Yes 81f95076281f is to blame.. After reverting it all is fine again.

15 resume cycles on the one laptop , 10 on the other without to hit the trace.

Regards