[pvrusb2] pvrusb2: fix 7MHz & 8MHz DVB-T tuner support for HVR1900 rev D1F5

[Michael Krufky]

On Mon, Feb 27, 2012 at 6:44 AM, Mauro Carvalho Chehab
<mchehab at redhat.com> wrote:
> Em 27-02-2012 07:15, Michael Krufky escreveu:
>> On Fri, Feb 24, 2012 at 2:27 PM, Mauro Carvalho Chehab
>> <mchehab at redhat.com> wrote:
>>> Em 07-02-2012 15:08, Michael Krufky escreveu:
>>>> There are some new revisions of the HVR-1900 around whose DVB-T
>>>> support is broken without this small bug-fix.  Please merge asap -
>>>> this fix needs to go to stable kernels as well.  It applies cleanly
>>>> against *all* recent kernels.
>>>>
>>>> The following changes since commit 805a6af8dba5dfdd35ec35dc52ec0122400b2610:
>>>>
>>>>   Linux 3.2 (2012-01-04 15:55:44 -0800)
>>>>
>>>> are available in the git repository at:
>>>>   git://linuxtv.org/mkrufky/hauppauge surrey
>>>>
>>>> Michael Krufky (1):
>>>>       pvrusb2: fix 7MHz & 8MHz DVB-T tuner support for HVR1900 rev D1F5
>>>>
>>>>  drivers/media/video/pvrusb2/pvrusb2-devattr.c |   10 ++++++++++
>>>>  1 files changed, 10 insertions(+), 0 deletions(-)
>>>
>>>> The D1F5 revision of the WinTV HVR-1900 uses a tda18271c2 tuner
>>>> instead of a tda18271c1 tuner as used in revision D1E9. To
>>>> account for this, we must hardcode the frontend configuration
>>>> to use the same IF frequency configuration for both revisions
>>>> of the device.
>>>
>>> No, you don't need to hardcode the IF. Just use the get_if_frequency
>>> callback at the demod, and it will work with whatever frequency you
>>> use at the tuner.
>>>
>>> Regards,
>>> Mauro
>>
>> (apologies for delayed reply - I am travelling in Europe this month
>> and not checking all my email every day)
>>
>> Mauro,
>>
>> get_if_frequency will work, but for *optimal* settings for the
>> hardware that my company manufactures, this is it.  The hardcoded IF
>> is the correct setting for *this* hardware.  There are more factors
>> than just a tuner and demodulator, here.  I know the board layout and
>> the crystal configurations, and we did extensive testing.  Sure, you
>> can use get_if_frequency and it will "work" ... but when I have
>> information about how it will work *best* then you should merge my
>> patch.
>
> Well, your patch description doesn't say that. Instead, it makes anyone
> reading it that you'll be using a sub-optimal configuration due to some
> Linux driver limitation.
>
>> This is the fix.  You can merge it or you will not merge it.  I did my
>> part, Mauro.  My customers need this fix and I will not argue.
>
>> ALSO, for -stable, there is no get_if_frequency.  I cannot have the
>> patch merged into stable unless you merge it into mainline. There is
>> no time for opinion - only facts.
>
> No. get_if_frequency is on Kernel 3.2.
>
>> Please merge.
>
> Please fix the description and I'll apply it.


This isn't for only 3.2, this needs to go to stable for kernels dating
back ever since the addition of pvrusb2-dvb.c

The description is correct as-is.  Why does merging a bug-fix cause
such heartache?

Please just merge the fix, Mauro.  The description should not change.
The correct description is as follows:

pvrusb2: fix 7MHz & 8MHz DVB-T tuner support for HVR1900 rev D1F5

The D1F5 revision of the WinTV HVR-1900 uses a tda18271c2 tuner
instead of a tda18271c1 tuner as used in revision D1E9. To
account for this, we must hardcode the frontend configuration
to use the same IF frequency configuration for both revisions
of the device.

6MHz DVB-T is unaffected by this issue, as the recommended
IF Frequency configuration for 6MHz DVB-T is the same on both
c1 and c2 revisions of the tda18271 tuner.

Signed-off-by: Michael Krufky <mkrufky at linuxtv.org>
Cc: Mike Isely <isely at pobox.com>
Cc: stable at kernel.org

This is the patch and description that I authored and sign-off on.  As
you can see, the description is specific to one revision of a
hauppauge device model number.  It is not a linux driver limitation,
this is about optimal device configuration for a specific hardware.

-Mike