linux-kernel - Re: [media] duplicate code in media drivers

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] [day] [month] [year] [list]

Date:   Tue, 22 May 2018 12:11:26 -0500
From:   "Gustavo A. R. Silva" <gustavo@...eddedor.com>
To:     Devin Heitmueller <dheitmueller@...nellabs.com>,
        Mauro Carvalho Chehab <mchehab+samsung@...nel.org>
Cc:     Mauro Carvalho Chehab <mchehab@...nel.org>,
        Linux Media Mailing List <linux-media@...r.kernel.org>,
        Linux Kernel <linux-kernel@...r.kernel.org>,
        Devin Heitmueller <dheitmueller@...uxtv.org>
Subject: Re: [media] duplicate code in media drivers



On 05/21/2018 03:44 PM, Devin Heitmueller wrote:
>>> diff -u -p drivers/media/dvb-frontends/au8522_decoder.c /tmp/nothing/media/dvb-frontends/au8522_decoder.c
>>> --- drivers/media/dvb-frontends/au8522_decoder.c
>>> +++ /tmp/nothing/media/dvb-frontends/au8522_decoder.c
>>> @@ -280,14 +280,9 @@ static void setup_decoder_defaults(struc
>>>                          AU8522_TOREGAAGC_REG0E5H_CVBS);
>>>          au8522_writereg(state, AU8522_REG016H, AU8522_REG016H_CVBS);
>>>
>>> -       if (is_svideo) {
>>>                  /* Despite what the table says, for the HVR-950q we still need
>>>                     to be in CVBS mode for the S-Video input (reason unknown). */
>>>                  /* filter_coef_type = 3; */
>>> -               filter_coef_type = 5;
>>> -       } else {
>>> -               filter_coef_type = 5;
>>> -       }
>>
>> Better ask Devin about this (c/c).
> 
> This was a case where the implementation didn't match the datasheet,
> and it wasn't clear why the filter coefficients weren't working
> properly.  Essentially I should have labeled that as a TODO or FIXME
> when I disabled the "right" value and forced it to always be five.  It
> was also likely that the filter coefficients would need to differ if
> taking video over the IF interface as opposed to CVBS/S-video, which
> is why I didn't want to get rid of the logic entirely.  That said, the
> only product I've ever seen with the tda18271 mated to the au8522 will
> likely never be supported for analog video under Linux for unrelated
> reasons.
> 
> That said, it's worked "good enough" since I wrote the code nine years
> ago, so if somebody wants to submit a patch to either get rid of the
> if() statement or mark it as a FIXME that will likely never actually
> get fixed, I wouldn't have an objection to either.
> 

Devin,

I've sent a patch based on your feedback.

Thanks!
--
Gustavo