Re: saa7134 (not very) new board 5168:0307

["tomlohave@gmail.com" <tomlohave@gmail.com>]

hermann pitton a écrit :
> Hi Tom,
>
> Am Mittwoch, den 23.12.2009, 13:17 +0100 schrieb tomlohave@gmail.com:
>   
>> Some news,
>>     
>>> Hi hermann,
>>>
>>> we are this results :
>>>
>>> with
>>>
>>> &tda827x_cfg_0, &tda827x_cfg_1 or &tda827x_cfg_2
>>>
>>>       
>>> we have a perfect image without sound on the analogic part (test with 
>>> mplayer),
>>> a partial result with dvb-t : we need to initialize first with 
>>> analogic (with cold boot, the card doesn't work on dvb)
>>> but only for few seconds(sound and image are ok) then re-initialize 
>>> with analogic, work for few seconds on dvb and then nothing
>>>       
>>> maybe i am wrong but, the sound part for analogic is a problem of 
>>> redirection, isn't it  ?
>>>       
>> fixed
>>     
>>> here are our configuration for this card :
>>>
>>> in saa7134-dvb.c
>>>
>>> static struct tda1004x_config tda827x_flydvbtduo_medion_config = {
>>>     .demod_address = 0x08,
>>>     .invert        = 1,
>>>     .invert_oclk   = 0,
>>>     .xtal_freq     = TDA10046_XTAL_16M,
>>>     .agc_config    = TDA10046_AGC_TDA827X,
>>>     .gpio_config   = TDA10046_GP01_I,
>>>     .if_freq       = TDA10046_FREQ_045,
>>>     .i2c_gate      = 0x4b,
>>>     .tuner_address = 0x61,
>>>     .antenna_switch = 2,
>>>     .request_firmware = philips_tda1004x_request_firmware
>>> };
>>>
>>> case SAA7134_BOARD_FLYDVBTDUO_MEDION:
>>>         if (configure_tda827x_fe(dev, &tda827x_flydvbtduo_medion_config,
>>>                      &tda827x_cfg_2) < 0)
>>>             goto dettach_frontend;
>>>         break;
>>>     default:
>>>         wprintk("Huh? unknown DVB card?\n");
>>>         break;
>>>
>>>
>>> in saa7134-cards.c
>>>
>>>    [SAA7134_BOARD_FLYDVBTDUO_MEDION] = {
>>>        .name           = "LifeView FlyDVB-T DUO Medion",
>>>       
>>>        .audio_clock    = 0x00187de7,
>>>       
>> change with  0x00200000 and there is a perfect sound :)
>>     
>
> fine. In README.saa7134 since Gerd wrote it the first time ;)
>
>   
>>>        .tuner_type     = TUNER_PHILIPS_TDA8290,
>>>        .radio_type     = UNSET,
>>>        .tuner_addr    = ADDR_UNSET,
>>>        .radio_addr    = ADDR_UNSET,
>>>        .gpiomask    = 0x00200000,
>>>        .mpeg           = SAA7134_MPEG_DVB,
>>>        .inputs         = {{
>>>            .name = name_tv,
>>>            .vmux = 1,
>>>            .amux = TV,
>>>            .gpio = 0x200000,     /* GPIO21=High for TV input */
>>>            .tv   = 1,
>>>        },{
>>>            .name = name_comp1,    /* Composite signal on S-Video input */
>>>            .vmux = 3,
>>>            .amux = LINE1,
>>>        },{
>>>            .name = name_svideo,    /* S-Video signal on S-Video input */
>>>            .vmux = 8,
>>>            .amux = LINE1,
>>>        }},
>>>        .radio = {
>>>            .name = name_radio,
>>>            .amux = TV,
>>>            .gpio = 0x000000,    /* GPIO21=Low for FM radio antenna */
>>>        },
>>>
>>>
>>> .vendor       = PCI_VENDOR_ID_PHILIPS,
>>>        .device       = PCI_DEVICE_ID_PHILIPS_SAA7133,
>>>        .subvendor    = 0x5168,               .subdevice    = 0x0307,  
>>> /* LR307-N */             .driver_data  = 
>>> SAA7134_BOARD_FLYDVBTDUO_MEDION,
>>>
>>> case SAA7134_BOARD_FLYDVBTDUO_MEDION:
>>>    {
>>>        /* this is a hybrid board, initialize to analog mode
>>>         * and configure firmware eeprom address
>>>         */
>>>        u8 data[] = { 0x3c, 0x33, 0x60};
>>>        struct i2c_msg msg = {.addr=0x08, .flags=0, .buf=data, .len = 
>>> sizeof(data)};
>>>        i2c_transfer(&dev->i2c_adap, &msg, 1);
>>>        break;
>>>
>>>
>>>
>>>
>>> What can we do to have dvb fully supported ?
>>>       
>> Can someone point me in the right direction ?
>>     
>
>   
hi hermann
> Hmmm, is there not anything with i2c_debug=1 before it fails after a few
> seconds?
>   
will post this
> Gpio pins can trigger cascades of switches, so to know the gpio status
> of the card for working DVB-T on m$ might still be a desire. Maybe chips
> power off.
>   
will post the results of dscaler
> Also, for 99.99% only a shot at the dark side of the moon before ever
> seen, but I would also try to force TS_SERIAL to have it visited.
>
> Cheers,
> Hermann
>
>
>
>   
Thanks for you answer
Cheers

Thomas