linux-kernel - Re: Kernel 5.2.8 - au0828

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]

Message-ID: <da6a1b65-cbe5-4e5e-d61c-43644a23da34@kernel.org>
Date:   Mon, 19 Aug 2019 15:44:09 -0600
From:   shuah <shuah@...nel.org>
To:     Nathan Royce <nroycea+kernel@...il.com>,
        Brad Love <brad@...tdimension.cc>
Cc:     sean@...s.org, Mauro Carvalho Chehab <mchehab@...nel.org>,
        Hans Verkuil <hverkuil@...all.nl>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        linux-media@...r.kernel.org, linux-kernel@...r.kernel.org,
        shuah <shuah@...nel.org>
Subject: Re: Kernel 5.2.8 - au0828 - Tuner Is Busy

On 8/19/19 2:49 PM, shuah wrote:
> On 8/16/19 7:15 PM, Nathan Royce wrote:
>> On Fri, Aug 16, 2019 at 1:42 PM Greg Kroah-Hartman
>> <gregkh@...uxfoundation.org> wrote:
>>> If you revert that one commit, does things start working again?
>>>
>>> thanks,
>>>
>>> greg k-h
>> Hey Greg, I just got finished building it after running "$ git revert
>> 812658d88d26" and verifying it reverted by comparing one of the files
>> from git log -p, but alas, no joy.
>>
>> On Fri, Aug 16, 2019 at 5:41 PM Brad Love <brad@...tdimension.cc> wrote:
>>>
>>> Hi Nathan,
>>>
>>> I don't have a "woodbury", but I have a Hauppauge 950Q sitting around
>>> and tested it on latest mainline kernel. w_scan is ok and streaming is
>>> fine. There's no unexpected errors. The 950Q uses the same au0828 bridge
>>> and au8522 demod as woodbury, but a different tuner. Your problem
>>> wouldn't appear to be a general au0828 issue.
>>>
> 
> Thanks Brad!
> 
>>> You might have to check out git bisect. That will be the quickest way to
>>> get to the bottom, if you've got points A and B, and are
>>> building/running your own kernel.
>>>
>>> Cheers,
>>>
>>> Brad
>> Thanks Brad, I'll explore bisecting and hopefully will be able to
>> narrow down the cause.
>> I wasn't running my own kernel, but rather using the Arch Linux kernel
>> and modding the one module and putting it in "extramodules".
>>
> 
> Hi Nathan,
> 
> Just catching up with this thread. Let me know what you find. Can you
> build your own kernel and see what you can find?
> 

You said you make changes to the

"Whenever I update my kernel, I edit the
./drivers/media/usb/au0828/au0828-cards.c file adding an entry for my
0x400 device.
I've been doing it for years and it's been working fine... until now..."

Please send me the changes you make to the file. I see the following
WOODBURY devices. I am assuming you add 0x400 entry.

{ USB_DEVICE(0x05e1, 0x0480),
                 .driver_info = AU0828_BOARD_HAUPPAUGE_WOODBURY },
         { USB_DEVICE(0x2040, 0x8200),
                 .driver_info = AU0828_BOARD_HAUPPAUGE_WOODBURY },


There is another table in sound/usb/quirks-table.h for AU0828
devices. In addition to 812658d88d26, 66354f18fe5f makes change
to this table to add a flag. I see two entries in that table:

AU0828_DEVICE(0x05e1, 0x0480, "Hauppauge", "Woodbury"),
AU0828_DEVICE(0x2040, 0x8200, "Hauppauge", "Woodbury"),

Since these drivers are now coupled doing resource sharing,
could it be that with your change to au02828 device table,
your changes are bow incomplete.

I don't have a Woodbury device though. This is something to
try.

Did you consider sending patch to add your device variant,
so you don't have to keep making this change whenever you
go to a new kernel?

thanks,
-- Shuah