Skip to contentSkip to conversation

Leadtek Winfast TV2100

DISCUSSION dz-tor 5 participants 11 messages Jan 3, 2010 – Jan 11, 2010 Media (V4L/DVB)

lore
Share & tools
lore

dz-tor

author
#0 lore raw

Hi,

I want to ask whether you are working on this 'issue'. I'm interested in the subject, as I'm owner of this card and cannot get this stuff to work - I've picture, no sound. Earlier you have written, that you can provide patches for testing. If offer is still actual, I can test them. Currently I'm using kernel 2.6.31, but if it's a problem I can switch to 2.6.32.

As I wrote earlier I have the same card - Winfast TV2100 with FM, tv norm used in my country is PAL (I'm from Poland).

Regards,
Darek

Quoted 118 lines
Quoted 118 lines

Hey guys,

Let me start by thanking you both for all your help. Unfortunately, there isn't much I can do with the data you provided - it's way too technical for me. I'd be happy to do any tests and apply any patches, but I would need some instructions - but only if you have time; I really don't want to burden you with this.

Thanks again,

Pavle.

________________________________
From: hermann pitton<[email protected]>
To: Terry Wu<[email protected]>
Cc: [email protected]; [email protected]
Sent: Sat, 28 November, 2009 4:09:28
Subject: Re: Re: Leadtek Winfast TV2100

Hi Terry Wu,

Am Donnerstag, den 26.11.2009, 10:33 +0800 schrieb Terry Wu:

Quoted 8 lines

Hi,

Here are the subsystem IDs for different TV2100 models:

     Subsystem ID:0x6f30107d, TVF8533-BDF (PAL BG/DK)
     Subsystem ID:0x6f32107d, TVF5533-MF (NTSC)
     Subsystem ID:0x6f3a107d, TVF88T5-B/DFF (PAL BG/DK, FM)

Terry

better is to become active on it.

we can most likely help about how to match such tuners fine, if any doubts left, but the best is to send just patches with having such hardware to test on.

Compared to the early tda9887 stuff on LeadTek devices, LeadTek was pioneering and we have that in mind, it should be fairly easy for those.

Let's know the other way round too, where you have still concerns doing so.

Cheers,
Hermann

Quoted 7 lines

2009/11/23 hermann pitton<[email protected]>:

Quoted 6 lines

Hi, thanks again!

Am Montag, den 23.11.2009, 09:42 +0800 schrieb Terry Wu:

Quoted 3 lines

Hi,

Please refer to the attached JPEG file for the GPIO

settings of

Quoted 8 lines
Quoted 8 lines
Quoted 6 lines

TV2100 with FM (PCB:B).

Let me know if you need the information of TV2100 without FM
(PCB:A, TVF8533-BDF).

Terry Wu

On a first look, if we start to count gpios from zero, we tell

the same.

Quoted 2 lines
Quoted 2 lines

The TVF8533_BDF I would have to look up. It is four to five

years back.

Quoted 2 lines
Quoted 2 lines

If it uses that minor number TI chip without radio support, we

treat it

Quoted 3 lines
Quoted 3 lines

as tuner=69 too currently.

For all such older can tuners, widely different about the globe,

counts,

Quoted 1 line
Quoted 1 line

that we don't have any way to detect them. So first working,

either NTSC

Quoted 3 lines
Quoted 3 lines

or PAL, sits in the pool position and others have to think twice.

OEMs do code tuners into eeprom content, some do not at all,

such doing

Quoted 6 lines
Quoted 6 lines

it are in competition and don't follow the rules of the main chip manufacturer, Philips/NXP in that case, and go their own ways.

So tuner tables are unstable across manufacturers.

We often can't help that much in such cases, but implementing

their own

Quoted 4 lines
Quoted 4 lines

tuner eeprom detection into the linux drivers is of course still welcome. Hauppauge does it very successfully since years.

We can't do much about it, if OEMs don't follow Philips or whom

ever on

Quoted 28 lines
Quoted 28 lines

such.

Thanks,
Hermann

Quoted 22 lines

2009/11/23 Terry Wu<[email protected]>:

Quoted 20 lines

Hi,

The TVF88T5-BDFF data sheet is attached.

Terry Wu

 11/17/2003  06:39 PM            72,010 TVF5531-MF.pdf
 03/12/2008  11:37 AM           555,285 TVF5533-MF-.pdf
 02/24/2004  02:19 PM           120,727 TVF5533-MF.pdf
 12/30/2003  06:59 PM            91,577 TVF5831-MFF.pdf
 09/26/2005  10:20 AM           156,853 TVF78P3-MFF.pdf
 11/17/2003  06:39 PM            67,947 TVF8531-BDF.pdf
 11/17/2003  06:39 PM            67,715 TVF8531-DIF.pdf
 03/12/2008  11:37 AM           509,340 TVF8533-BDF.pdf
 03/12/2008  11:37 AM           507,295 TVF8533-DIF.pdf
 12/30/2003  06:59 PM            87,921 TVF8831-BDFF.pdf
 12/30/2003  06:59 PM            87,624 TVF8831-DIFF.pdf
 09/26/2005  10:20 AM           176,525 TVF88P3-CFF.pdf
 03/24/2006  10:48 AM           460,941 TVF88T5-BDFF.pdf
 02/24/2004  02:19 PM           132,304 TVF9533-BDF.pdf
Quoted 8 lines
Quoted 8 lines
Quoted 6 lines
Quoted 6 lines
Quoted 3 lines
Quoted 3 lines
 02/24/2004  02:19 PM           120,940 TVF9533-DIF.pdf
 03/12/2008  11:37 AM           458,967 TVF99T5-BDFF.pdf

--
video4linux-list mailing list
Unsubscribe mailto:[email protected]?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

#1 lore raw

Hi,

Am Sonntag, den 03.01.2010, 16:37 +0100 schrieb dz-tor:

Quoted 9 lines

Hi,

I want to ask whether you are working on this 'issue'. I'm interested in the subject, as I'm owner of this card and cannot get this stuff to work
- I've picture, no sound. Earlier you have written, that you can provide patches for testing. If offer is still actual, I can test them. Currently I'm using kernel 2.6.31, but if it's a problem I can switch to 2.6.32.

Quoted 5 lines

As I wrote earlier I have the same card - Winfast TV2100 with FM, tv norm used in my country is PAL (I'm from Poland).

Regards,
Darek

please have a look at the xtal/oscillator close to the saa713x chip.

It can have two different values. See README.saa7134 in Documentation.

I'll provide a patch with best guessing based on the regspy.exe results then, assuming all is connected to LINE2 and routed through the external mux chip as a start. If that fails, we try with LINE1 next.

Cheers,
Hermann

Quoted 122 lines
Quoted 122 lines
Quoted 122 lines

Hey guys,

Let me start by thanking you both for all your help. Unfortunately, there isn't much I can do with the data you provided - it's way too technical for me. I'd be happy to do any tests and apply any patches, but I would need some instructions - but only if you have time; I really don't want to burden you with this.

Thanks again,

Pavle.

________________________________
From: hermann pitton<[email protected]>
To: Terry Wu<[email protected]>
Cc: [email protected]; [email protected]
Sent: Sat, 28 November, 2009 4:09:28
Subject: Re: Re: Leadtek Winfast TV2100

Hi Terry Wu,

Am Donnerstag, den 26.11.2009, 10:33 +0800 schrieb Terry Wu:

Quoted 8 lines

Hi,

Here are the subsystem IDs for different TV2100 models:

     Subsystem ID:0x6f30107d, TVF8533-BDF (PAL BG/DK)
     Subsystem ID:0x6f32107d, TVF5533-MF (NTSC)
     Subsystem ID:0x6f3a107d, TVF88T5-B/DFF (PAL BG/DK, FM)

Terry

better is to become active on it.

we can most likely help about how to match such tuners fine, if any doubts left, but the best is to send just patches with having such hardware to test on.

Compared to the early tda9887 stuff on LeadTek devices, LeadTek was pioneering and we have that in mind, it should be fairly easy for those.

Let's know the other way round too, where you have still concerns doing so.

Cheers,
Hermann

Quoted 7 lines

2009/11/23 hermann pitton<[email protected]>:

Quoted 6 lines

Hi, thanks again!

Am Montag, den 23.11.2009, 09:42 +0800 schrieb Terry Wu:

Quoted 3 lines

Hi,

Please refer to the attached JPEG file for the GPIO

settings of

Quoted 8 lines
Quoted 8 lines
Quoted 6 lines

TV2100 with FM (PCB:B).

Let me know if you need the information of TV2100 without FM
(PCB:A, TVF8533-BDF).

Terry Wu

On a first look, if we start to count gpios from zero, we tell

the same.

Quoted 2 lines
Quoted 2 lines

The TVF8533_BDF I would have to look up. It is four to five

years back.

Quoted 2 lines
Quoted 2 lines

If it uses that minor number TI chip without radio support, we

treat it

Quoted 3 lines
Quoted 3 lines

as tuner=69 too currently.

For all such older can tuners, widely different about the globe,

counts,

Quoted 1 line
Quoted 1 line

that we don't have any way to detect them. So first working,

either NTSC

Quoted 3 lines
Quoted 3 lines

or PAL, sits in the pool position and others have to think twice.

OEMs do code tuners into eeprom content, some do not at all,

such doing

Quoted 6 lines
Quoted 6 lines

it are in competition and don't follow the rules of the main chip manufacturer, Philips/NXP in that case, and go their own ways.

So tuner tables are unstable across manufacturers.

We often can't help that much in such cases, but implementing

their own

Quoted 4 lines
Quoted 4 lines

tuner eeprom detection into the linux drivers is of course still welcome. Hauppauge does it very successfully since years.

We can't do much about it, if OEMs don't follow Philips or whom

ever on

Quoted 28 lines
Quoted 28 lines

such.

Thanks,
Hermann

Quoted 22 lines

2009/11/23 Terry Wu<[email protected]>:

Quoted 20 lines

Hi,

The TVF88T5-BDFF data sheet is attached.

Terry Wu

 11/17/2003  06:39 PM            72,010 TVF5531-MF.pdf
 03/12/2008  11:37 AM           555,285 TVF5533-MF-.pdf
 02/24/2004  02:19 PM           120,727 TVF5533-MF.pdf
 12/30/2003  06:59 PM            91,577 TVF5831-MFF.pdf
 09/26/2005  10:20 AM           156,853 TVF78P3-MFF.pdf
 11/17/2003  06:39 PM            67,947 TVF8531-BDF.pdf
 11/17/2003  06:39 PM            67,715 TVF8531-DIF.pdf
 03/12/2008  11:37 AM           509,340 TVF8533-BDF.pdf
 03/12/2008  11:37 AM           507,295 TVF8533-DIF.pdf
 12/30/2003  06:59 PM            87,921 TVF8831-BDFF.pdf
 12/30/2003  06:59 PM            87,624 TVF8831-DIFF.pdf
 09/26/2005  10:20 AM           176,525 TVF88P3-CFF.pdf
 03/24/2006  10:48 AM           460,941 TVF88T5-BDFF.pdf
 02/24/2004  02:19 PM           132,304 TVF9533-BDF.pdf
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 6 lines
Quoted 6 lines
Quoted 3 lines
Quoted 3 lines
 02/24/2004  02:19 PM           120,940 TVF9533-DIF.pdf
 03/12/2008  11:37 AM           458,967 TVF99T5-BDFF.pdf

--
video4linux-list mailing list
Unsubscribe mailto:[email protected]?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

#2 lore raw

Blackmagic SDI card --> v4l2 ready ?

Just a quick question to see if Blackmagic's SDI card (I just have the basic one, not the studio or the extreme) is working under Video4Linux?

I have successfully tested this card using the bundled software (for Ubuntu) but I am getting this error when trying to capture from it with VLC:

*[0x2a3e828] v2l2 demux warning: FIXME: v4l2.c ControlListPrint 2738*

cheers,

forest mars

Signature · 9 lines
mnn: you're what's on!
http://mnn.org



--
video4linux-list mailing list
Unsubscribe mailto:[email protected]?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list
#3 lore raw

Re: Blackmagic SDI card --> v4l2 ready ?

Quoted 6 lines · Mars Forest

Just a quick question to see if Blackmagic's SDI card (I just have the basic one, not the studio or the extreme) is working under Video4Linux?

I have successfully tested this card using the bundled software (for Ubuntu) but I am getting this error when trying to capture from it with VLC:

    *[0x2a3e828] v2l2 demux warning: FIXME: v4l2.c ControlListPrint 2738*

Since this question was rather quickly answered on IRC ("none of Blackmagic's cards work with v4l.") I am now looking for a recommendation on a good card to replace this one. I am running Debian/Ubuntu on a Sunfire x2200.

Looking for a half-length PCI-E SDI video capture card that will work with v4l2/VLC.

What's the best choice here?

Quoted 9 lines

cheers,

forest mars

Signature · 2 lines
mnn: you're what's on!
http://mnn.org

--
video4linux-list mailing list
Unsubscribe mailto:[email protected]?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

#4 lore raw

Re: Blackmagic SDI card --> v4l2 ready ?

I think it would be interesting to write a V4L2 driver for the BlackMagic cards using the code already released by BM for the very low level kernel driver and some reverse engineering.

Rafael Diniz

Quoted 27 lines · On Friday 08 January 2010 18:32:09 Mars Forest wrote:
Quoted 8 lines · Mars Forest

Just a quick question to see if Blackmagic's SDI card (I just have the basic one, not the studio or the extreme) is working under Video4Linux?

I have successfully tested this card using the bundled software (for Ubuntu) but I am getting this error when trying to capture from it with VLC:

    *[0x2a3e828] v2l2 demux warning: FIXME: v4l2.c ControlListPrint 2738*

Since this question was rather quickly answered on IRC ("none of Blackmagic's cards work with v4l.") I am now looking for a recommendation on a good card to replace this one. I am running Debian/Ubuntu on a Sunfire x2200.

Looking for a half-length PCI-E SDI video capture card that will work with v4l2/VLC.

What's the best choice here?

Quoted 6 lines

cheers,

forest mars
--
mnn: you're what's on!
http://mnn.org

Quoted 4 lines

--
video4linux-list mailing list
Unsubscribe mailto:[email protected]?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

--
video4linux-list mailing list
Unsubscribe mailto:[email protected]?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

author
#5 lore raw
In reply to hermann pitton

Hi,

My friend helped me a little and we managed card to work, i.e. TV (only mono, I suppose that stereo is not available for this card), s-video, composite and radio works fine.

We've tried setting audio_clock to both values (32.11 MHz -> .audio_clock=0x187de7 and 24.576MHz -> .audio_clock=0x200000) and both works - for me it seems, that for 32.11 MHz sound is better (i.e. louder).

Now I have only one problem with ir. We succeeded to enable it, but reading from it is bit weird - sometimes when I press the key, no event is generated, sometimes 2 (key pressed/released) and sometimes more. Have any clue what can cause that?

Output from dmesg when module is loaded:

[32627.844221] saa7130/34: v4l2 driver version 0.2.15 loaded
[32627.844294] saa7130[0]: found at 0000:01:09.0, rev: 1, irq: 17, latency: 32, mmio: 0xe7001000
[32627.844303] saa7130[0]: subsystem: 107d:6f3a, board: Proteus Pro [philips reference design] [card=1,insmod option]
[32627.844340] saa7130[0]: board init: gpio is 60a08
[32627.844351] saa7130[0]: gpio: mode=0x0000000 in=0x0060a0c out=0x0000000 [pre-init]
[32627.844901] input: saa7134 IR (Proteus Pro [philip as /devices/pci0000:00/0000:00:08.0/0000:01:09.0/input/input15
[32627.845064] IRQ 17/saa7130[0]: IRQF_DISABLED is not guaranteed on shared IRQs
[32628.028017] saa7130[0]: i2c eeprom 00: 7d 10 3a 6f 54 20 1c 00 43 43 a9 1c 55 d2 b2 92
[32628.028029] saa7130[0]: i2c eeprom 10: 0c ff 82 0e ff 20 ff ff ff ff ff ff ff ff ff ff
[32628.028038] saa7130[0]: i2c eeprom 20: 01 40 02 03 03 02 01 03 08 ff 00 8c ff ff ff ff
[32628.028047] saa7130[0]: i2c eeprom 30: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028056] saa7130[0]: i2c eeprom 40: 50 89 00 c2 00 00 02 30 02 ff ff ff ff ff ff ff
[32628.028064] saa7130[0]: i2c eeprom 50: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028073] saa7130[0]: i2c eeprom 60: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028082] saa7130[0]: i2c eeprom 70: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028091] saa7130[0]: i2c eeprom 80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028099] saa7130[0]: i2c eeprom 90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028108] saa7130[0]: i2c eeprom a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028117] saa7130[0]: i2c eeprom b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028126] saa7130[0]: i2c eeprom c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028134] saa7130[0]: i2c eeprom d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028143] saa7130[0]: i2c eeprom e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028152] saa7130[0]: i2c eeprom f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028163] i2c i2c-2: Invalid 7-bit address 0x7a
[32628.036120] tuner 2-0061: chip found @ 0xc2 (saa7130[0])
[32628.044037] tuner-simple 2-0061: creating new instance
[32628.044044] tuner-simple 2-0061: type set to 69 (Tena TNF 5335 and similar models)
[32628.052058] saa7130[0]: gpio: mode=0x000000d in=0x0060a00 out=0x0000008 [mute]
[32628.052178] saa7130[0]: registered device video1 [v4l2]
[32628.052205] saa7130[0]: registered device vbi0
[32628.052232] saa7130[0]: registered device radio0

For testing we've used existing card (card=1 - Proteus Pro) with tuner 69. When everything will be working ok, my friend will try to add new card definition and generate the patch. Below you can find actual code modified by us:

saa7134-cards.c
struct saa7134_board saa7134_boards[] = {
...

    [SAA7134_BOARD_PROTEUS_PRO] = {
        /* /me */
        .name        = "Proteus Pro [philips reference design]",
        .audio_clock    = 0x00187de7,
        .tuner_type    = TUNER_TNF_5335MF,
        .radio_type     = UNSET,
        .tuner_addr    = ADDR_UNSET,
        .radio_addr    = ADDR_UNSET,

        .gpiomask    = 0x0d,
        .inputs         = {{
            .name = name_comp1,
            .vmux = 3,
            .amux = LINE2,
            .gpio = 0x08,
        },{
            .name = name_tv,
            .vmux = 1,
            .amux = LINE1,
            .tv   = 1,
        },{
            .name = name_tv_mono,
            .vmux = 1,
            .amux = LINE1,
            .tv   = 1,
        },{
            .name = name_svideo,
            .vmux = 8,
            .amux = LINE2,
            .gpio = 0x08,
        }},
        .radio = {
            .name = name_radio,
            .amux = LINE1,
            .gpio = 0x04,
        },
        .mute = {
            .name = name_mute,
            .amux = TV,
            .gpio = 0x08,
        },
    },

int saa7134_board_init1(struct saa7134_dev *dev)
{
...

    case SAA7134_BOARD_PROTEUS_PRO:
        dev->has_remote = SAA7134_REMOTE_GPIO;
        break;

saa7134-input.c
int saa7134_input_init1(struct saa7134_dev *dev)

{
    case SAA7134_BOARD_PROTEUS_PRO:
        ir_codes     = &ir_codes_avermedia_table;
        mask_keycode = 0x000000;
        polling      = 50; // ms

Regards,
Darek

Quoted 21 lines · hermann pitton

Hi,

Am Sonntag, den 03.01.2010, 16:37 +0100 schrieb dz-tor:

Quoted 9 lines

Hi,

I want to ask whether you are working on this 'issue'. I'm interested in the subject, as I'm owner of this card and cannot get this stuff to work
- I've picture, no sound. Earlier you have written, that you can provide patches for testing. If offer is still actual, I can test them. Currently I'm using kernel 2.6.31, but if it's a problem I can switch to 2.6.32.

Quoted 6 lines

As I wrote earlier I have the same card - Winfast TV2100 with FM, tv norm used in my country is PAL (I'm from Poland).

Regards,
Darek

please have a look at the xtal/oscillator close to the saa713x chip.

Quoted 166 lines

It can have two different values. See README.saa7134 in Documentation.

I'll provide a patch with best guessing based on the regspy.exe results then, assuming all is connected to LINE2 and routed through the external mux chip as a start. If that fails, we try with LINE1 next.

Cheers,
Hermann

Quoted 152 lines
Quoted 152 lines
Quoted 152 lines

Hey guys,

Let me start by thanking you both for all your help. Unfortunately, there isn't much I can do with the data you provided - it's way too technical for me. I'd be happy to do any tests and apply any patches, but I would need some instructions - but only if you have time; I really don't want to burden you with this.

Thanks again,

Pavle.

________________________________
From: hermann pitton<[email protected]>
To: Terry Wu<[email protected]>
Cc: [email protected]; [email protected]
Sent: Sat, 28 November, 2009 4:09:28
Subject: Re: Re: Leadtek Winfast TV2100

Hi Terry Wu,

Am Donnerstag, den 26.11.2009, 10:33 +0800 schrieb Terry Wu:

Quoted 9 lines

Hi,

Here are the subsystem IDs for different TV2100 models:

      Subsystem ID:0x6f30107d, TVF8533-BDF (PAL BG/DK)
      Subsystem ID:0x6f32107d, TVF5533-MF (NTSC)
      Subsystem ID:0x6f3a107d, TVF88T5-B/DFF (PAL BG/DK, FM)

Terry

better is to become active on it.

we can most likely help about how to match such tuners fine, if any doubts left, but the best is to send just patches with having such hardware to test on.

Compared to the early tda9887 stuff on LeadTek devices, LeadTek was pioneering and we have that in mind, it should be fairly easy for those.

Let's know the other way round too, where you have still concerns doing so.

Cheers,
Hermann

Quoted 10 lines

2009/11/23 hermann pitton<[email protected]>:

Quoted 8 lines

Hi, thanks again!

Am Montag, den 23.11.2009, 09:42 +0800 schrieb Terry Wu:

Quoted 4 lines

Hi,

Please refer to the attached JPEG file for the GPIO

settings of

Quoted 9 lines
Quoted 9 lines
Quoted 7 lines

TV2100 with FM (PCB:B).

Let me know if you need the information of TV2100 without FM
(PCB:A, TVF8533-BDF).

Terry Wu

On a first look, if we start to count gpios from zero, we tell

the same.

Quoted 2 lines
Quoted 2 lines

The TVF8533_BDF I would have to look up. It is four to five

years back.

Quoted 2 lines
Quoted 2 lines

If it uses that minor number TI chip without radio support, we

treat it

Quoted 4 lines
Quoted 4 lines

as tuner=69 too currently.

For all such older can tuners, widely different about the globe,

counts,

Quoted 2 lines
Quoted 2 lines

that we don't have any way to detect them. So first working,

either NTSC

Quoted 4 lines
Quoted 4 lines

or PAL, sits in the pool position and others have to think twice.

OEMs do code tuners into eeprom content, some do not at all,

such doing

Quoted 7 lines
Quoted 7 lines

it are in competition and don't follow the rules of the main chip manufacturer, Philips/NXP in that case, and go their own ways.

So tuner tables are unstable across manufacturers.

We often can't help that much in such cases, but implementing

their own

Quoted 5 lines
Quoted 5 lines

tuner eeprom detection into the linux drivers is of course still welcome. Hauppauge does it very successfully since years.

We can't do much about it, if OEMs don't follow Philips or whom

ever on

Quoted 36 lines
Quoted 36 lines

such.

Thanks,
Hermann

Quoted 26 lines

2009/11/23 Terry Wu<[email protected]>:

Quoted 24 lines

Hi,

The TVF88T5-BDFF data sheet is attached.

Terry Wu

  11/17/2003  06:39 PM            72,010 TVF5531-MF.pdf
  03/12/2008  11:37 AM           555,285 TVF5533-MF-.pdf
  02/24/2004  02:19 PM           120,727 TVF5533-MF.pdf
  12/30/2003  06:59 PM            91,577 TVF5831-MFF.pdf
  09/26/2005  10:20 AM           156,853 TVF78P3-MFF.pdf
  11/17/2003  06:39 PM            67,947 TVF8531-BDF.pdf
  11/17/2003  06:39 PM            67,715 TVF8531-DIF.pdf
  03/12/2008  11:37 AM           509,340 TVF8533-BDF.pdf
  03/12/2008  11:37 AM           507,295 TVF8533-DIF.pdf
  12/30/2003  06:59 PM            87,921 TVF8831-BDFF.pdf
  12/30/2003  06:59 PM            87,624 TVF8831-DIFF.pdf
  09/26/2005  10:20 AM           176,525 TVF88P3-CFF.pdf
  03/24/2006  10:48 AM           460,941 TVF88T5-BDFF.pdf
  02/24/2004  02:19 PM           132,304 TVF9533-BDF.pdf
  02/24/2004  02:19 PM           120,940 TVF9533-DIF.pdf
  03/12/2008  11:37 AM           458,967 TVF99T5-BDFF.pdf

--
video4linux-list mailing list
Unsubscribe mailto:[email protected]?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

#6 lore raw

Hey Darek,

Great job of making the card work. I was really thrilled when I saw your post. However, I am a total newbie, so I couldn't apply the changes you wrote about. Could you please be a bit more specific? What I did is downloaded the driver from here: http://dl.bytesex.org/releases/video4linux/saa7134-0.2.12.tar.gz and made the changes to those two files, as described. But I have no clue how to compile it. I installed linux-headers for my kernel version and tried to run make, but I'm getting an error. Are there any configuration options that I need to set in Makefile or Make.config?

Cheers,

Pavle.

PS My kernel is 2.6.26-2-686 (it's a Debian Lenny 503 system). Here is the output produced by make:

debian:/home/pavle/saa7134-0.2.12# make
make -C /lib/modules/2.6.26-2-686/build SUBDIRS=/home/pavle/saa7134-0.2.12 modules
make[1]: Entering directory `/usr/src/linux-headers-2.6.26-2-686'
  CC [M]  /home/pavle/saa7134-0.2.12/video-buf.o
In file included from /home/pavle/saa7134-0.2.12/media/video-buf.h:19,
                 from /home/pavle/saa7134-0.2.12/video-buf.c:33:
/home/pavle/saa7134-0.2.12/linux/videodev.h:68: error: field 'class_dev' has incomplete type
/home/pavle/saa7134-0.2.12/linux/videodev.h:87: warning: 'struct class_device_attribute' declared inside parameter list
/home/pavle/saa7134-0.2.12/linux/videodev.h:87: warning: its scope is only this definition or declaration, which is probably not what you want
/home/pavle/saa7134-0.2.12/linux/videodev.h: In function 'video_device_create_file':
/home/pavle/saa7134-0.2.12/linux/videodev.h:89: error: implicit declaration of function 'class_device_create_file'
/home/pavle/saa7134-0.2.12/linux/videodev.h: At top level:
/home/pavle/saa7134-0.2.12/linux/videodev.h:93: warning: 'struct class_device_attribute' declared inside parameter list
/home/pavle/saa7134-0.2.12/linux/videodev.h: In function 'video_device_remove_file':
/home/pavle/saa7134-0.2.12/linux/videodev.h:95: error: implicit declaration of function 'class_device_remove_file'
/home/pavle/saa7134-0.2.12/video-buf.c: At top level:
/home/pavle/saa7134-0.2.12/video-buf.c:46: error: expected ')' before string constant
/home/pavle/saa7134-0.2.12/video-buf.c: In function 'videobuf_vmalloc_to_sg':
/home/pavle/saa7134-0.2.12/video-buf.c:68: error: 'struct scatterlist' has no member named 'page'
/home/pavle/saa7134-0.2.12/video-buf.c: In function 'videobuf_pages_to_sg':
/home/pavle/saa7134-0.2.12/video-buf.c:96: error: 'struct scatterlist' has no member named 'page'
/home/pavle/saa7134-0.2.12/video-buf.c:104: error: 'struct scatterlist' has no member named 'page'
/home/pavle/saa7134-0.2.12/video-buf.c: In function 'videobuf_vm_nopage':
/home/pavle/saa7134-0.2.12/video-buf.c:1099: error: 'NOPAGE_SIGBUS' undeclared (first use in this function)
/home/pavle/saa7134-0.2.12/video-buf.c:1099: error: (Each undeclared identifier is reported only once
/home/pavle/saa7134-0.2.12/video-buf.c:1099: error: for each function it appears in.)
/home/pavle/saa7134-0.2.12/video-buf.c:1102: error: 'NOPAGE_OOM' undeclared (first use in this function)
/home/pavle/saa7134-0.2.12/video-buf.c: At top level:
/home/pavle/saa7134-0.2.12/video-buf.c:1119: error: unknown field 'nopage' specified in initializer
/home/pavle/saa7134-0.2.12/video-buf.c:1119: warning: initialization from incompatible pointer type
make[2]: *** [/home/pavle/saa7134-0.2.12/video-buf.o] Error 1
make[1]: *** [_module_/home/pavle/saa7134-0.2.12] Error 2
make[1]: Leaving directory `/usr/src/linux-headers-2.6.26-2-686'
make: *** [default] Error 2

________________________________
From: dz-tor <[email protected]>
To: hermann pitton <[email protected]>
Cc: [email protected]; [email protected]; [email protected]
Sent: Fri, 8 January, 2010 23:56:54
Subject: Re: Leadtek Winfast TV2100

Hi,

My friend helped me a little and we managed card to work, i.e. TV (only mono, I suppose that stereo is not available for this card), s-video, composite and radio works fine.

We've tried setting audio_clock to both values (32.11 MHz -> .audio_clock=0x187de7 and 24.576MHz -> .audio_clock=0x200000) and both works - for me it seems, that for 32.11 MHz sound is better (i.e. louder).

Now I have only one problem with ir. We succeeded to enable it, but reading from it is bit weird - sometimes when I press the key, no event is generated, sometimes 2 (key pressed/released) and sometimes more. Have any clue what can cause that?

Output from dmesg when module is loaded:

[32627.844221] saa7130/34: v4l2 driver version 0.2.15 loaded
[32627.844294] saa7130[0]: found at 0000:01:09.0, rev: 1, irq: 17, latency: 32, mmio: 0xe7001000
[32627.844303] saa7130[0]: subsystem: 107d:6f3a, board: Proteus Pro [philips reference design] [card=1,insmod option]
[32627.844340] saa7130[0]: board init: gpio is 60a08
[32627.844351] saa7130[0]: gpio: mode=0x0000000 in=0x0060a0c out=0x0000000 [pre-init]
[32627.844901] input: saa7134 IR (Proteus Pro [philip as /devices/pci0000:00/0000:00:08.0/0000:01:09.0/input/input15
[32627.845064] IRQ 17/saa7130[0]: IRQF_DISABLED is not guaranteed on shared IRQs
[32628.028017] saa7130[0]: i2c eeprom 00: 7d 10 3a 6f 54 20 1c 00 43 43 a9 1c 55 d2 b2 92
[32628.028029] saa7130[0]: i2c eeprom 10: 0c ff 82 0e ff 20 ff ff ff ff ff ff ff ff ff ff
[32628.028038] saa7130[0]: i2c eeprom 20: 01 40 02 03 03 02 01 03 08 ff 00 8c ff ff ff ff
[32628.028047] saa7130[0]: i2c eeprom 30: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028056] saa7130[0]: i2c eeprom 40: 50 89 00 c2 00 00 02 30 02 ff ff ff ff ff ff ff
[32628.028064] saa7130[0]: i2c eeprom 50: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028073] saa7130[0]: i2c eeprom 60: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028082] saa7130[0]: i2c eeprom 70: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028091] saa7130[0]: i2c eeprom 80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028099] saa7130[0]: i2c eeprom 90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028108] saa7130[0]: i2c eeprom a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028117] saa7130[0]: i2c eeprom b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028126] saa7130[0]: i2c eeprom c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028134] saa7130[0]: i2c eeprom d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028143] saa7130[0]: i2c eeprom e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028152] saa7130[0]: i2c eeprom f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028163] i2c i2c-2: Invalid 7-bit address 0x7a
[32628.036120] tuner 2-0061: chip found @ 0xc2 (saa7130[0])
[32628.044037] tuner-simple 2-0061: creating new instance
[32628.044044] tuner-simple 2-0061: type set to 69 (Tena TNF 5335 and similar models)
[32628.052058] saa7130[0]: gpio: mode=0x000000d in=0x0060a00 out=0x0000008 [mute]
[32628.052178] saa7130[0]: registered device video1 [v4l2]
[32628.052205] saa7130[0]: registered device vbi0
[32628.052232] saa7130[0]: registered device radio0

For testing we've used existing card (card=1 - Proteus Pro) with tuner 69. When everything will be working ok, my friend will try to add new card definition and generate the patch. Below you can find actual code modified by us:

saa7134-cards.c
struct saa7134_board saa7134_boards[] = {
...

    [SAA7134_BOARD_PROTEUS_PRO] = {
        /* /me */
        .name        = "Proteus Pro [philips reference design]",
        .audio_clock    = 0x00187de7,
        .tuner_type    = TUNER_TNF_5335MF,
        .radio_type     = UNSET,
        .tuner_addr    = ADDR_UNSET,
        .radio_addr    = ADDR_UNSET,

        .gpiomask    = 0x0d,
        .inputs         = {{
            .name = name_comp1,
            .vmux = 3,
            .amux = LINE2,
            .gpio = 0x08,
        },{
            .name = name_tv,
            .vmux = 1,
            .amux = LINE1,
            .tv   = 1,
        },{
            .name = name_tv_mono,
            .vmux = 1,
            .amux = LINE1,
            .tv   = 1,
        },{
            .name = name_svideo,
            .vmux = 8,
            .amux = LINE2,
            .gpio = 0x08,
        }},
        .radio = {
            .name = name_radio,
            .amux = LINE1,
            .gpio = 0x04,
        },
        .mute = {
            .name = name_mute,
            .amux = TV,
            .gpio = 0x08,
        },
    },

int saa7134_board_init1(struct saa7134_dev *dev)
{
...

    case SAA7134_BOARD_PROTEUS_PRO:
        dev->has_remote = SAA7134_REMOTE_GPIO;
        break;

saa7134-input.c
int saa7134_input_init1(struct saa7134_dev *dev)

{
    case SAA7134_BOARD_PROTEUS_PRO:
        ir_codes     = &ir_codes_avermedia_table;
        mask_keycode = 0x000000;
        polling      = 50; // ms

Regards,
Darek

Quoted 21 lines · hermann pitton

Hi,

Am Sonntag, den 03.01.2010, 16:37 +0100 schrieb dz-tor:

Quoted 9 lines

Hi,

I want to ask whether you are working on this 'issue'. I'm interested in the subject, as I'm owner of this card and cannot get this stuff to work
- I've picture, no sound. Earlier you have written, that you can provide patches for testing. If offer is still actual, I can test them. Currently I'm using kernel 2.6.31, but if it's a problem I can switch to 2.6.32.

Quoted 6 lines

As I wrote earlier I have the same card - Winfast TV2100 with FM, tv norm used in my country is PAL (I'm from Poland).

Regards,
Darek

please have a look at the xtal/oscillator close to the saa713x chip.

Quoted 166 lines

It can have two different values. See README.saa7134 in Documentation.

I'll provide a patch with best guessing based on the regspy.exe results then, assuming all is connected to LINE2 and routed through the external mux chip as a start. If that fails, we try with LINE1 next.

Cheers,
Hermann

Quoted 152 lines
Quoted 152 lines
Quoted 152 lines

Hey guys,

Let me start by thanking you both for all your help. Unfortunately, there isn't much I can do with the data you provided - it's way too technical for me. I'd be happy to do any tests and apply any patches, but I would need some instructions - but only if you have time; I really don't want to burden you with this.

Thanks again,

Pavle.

________________________________
From: hermann pitton<[email protected]>
To: Terry Wu<[email protected]>
Cc: [email protected]; [email protected]
Sent: Sat, 28 November, 2009 4:09:28
Subject: Re: Re: Leadtek Winfast TV2100

Hi Terry Wu,

Am Donnerstag, den 26.11.2009, 10:33 +0800 schrieb Terry Wu:

Quoted 9 lines

Hi,

Here are the subsystem IDs for different TV2100 models:

      Subsystem ID:0x6f30107d, TVF8533-BDF (PAL BG/DK)
      Subsystem ID:0x6f32107d, TVF5533-MF (NTSC)
      Subsystem ID:0x6f3a107d, TVF88T5-B/DFF (PAL BG/DK, FM)

Terry

better is to become active on it.

we can most likely help about how to match such tuners fine, if any doubts left, but the best is to send just patches with having such hardware to test on.

Compared to the early tda9887 stuff on LeadTek devices, LeadTek was pioneering and we have that in mind, it should be fairly easy for those.

Let's know the other way round too, where you have still concerns doing so.

Cheers,
Hermann

Quoted 10 lines

2009/11/23 hermann pitton<[email protected]>:

Quoted 8 lines

Hi, thanks again!

Am Montag, den 23.11.2009, 09:42 +0800 schrieb Terry Wu:

Quoted 4 lines

Hi,

Please refer to the attached JPEG file for the GPIO

settings of

Quoted 9 lines
Quoted 9 lines
Quoted 7 lines

TV2100 with FM (PCB:B).

Let me know if you need the information of TV2100 without FM
(PCB:A, TVF8533-BDF).

Terry Wu

On a first look, if we start to count gpios from zero, we tell

the same.

Quoted 2 lines
Quoted 2 lines

The TVF8533_BDF I would have to look up. It is four to five

years back.

Quoted 2 lines
Quoted 2 lines

If it uses that minor number TI chip without radio support, we

treat it

Quoted 4 lines
Quoted 4 lines

as tuner=69 too currently.

For all such older can tuners, widely different about the globe,

counts,

Quoted 2 lines
Quoted 2 lines

that we don't have any way to detect them. So first working,

either NTSC

Quoted 4 lines
Quoted 4 lines

or PAL, sits in the pool position and others have to think twice.

OEMs do code tuners into eeprom content, some do not at all,

such doing

Quoted 7 lines
Quoted 7 lines

it are in competition and don't follow the rules of the main chip manufacturer, Philips/NXP in that case, and go their own ways.

So tuner tables are unstable across manufacturers.

We often can't help that much in such cases, but implementing

their own

Quoted 5 lines
Quoted 5 lines

tuner eeprom detection into the linux drivers is of course still welcome. Hauppauge does it very successfully since years.

We can't do much about it, if OEMs don't follow Philips or whom

ever on

Quoted 36 lines
Quoted 36 lines

such.

Thanks,
Hermann

Quoted 26 lines

2009/11/23 Terry Wu<[email protected]>:

Quoted 24 lines

Hi,

The TVF88T5-BDFF data sheet is attached.

Terry Wu

  11/17/2003  06:39 PM            72,010 TVF5531-MF.pdf
  03/12/2008  11:37 AM           555,285 TVF5533-MF-.pdf
  02/24/2004  02:19 PM           120,727 TVF5533-MF.pdf
  12/30/2003  06:59 PM            91,577 TVF5831-MFF.pdf
  09/26/2005  10:20 AM           156,853 TVF78P3-MFF.pdf
  11/17/2003  06:39 PM            67,947 TVF8531-BDF.pdf
  11/17/2003  06:39 PM            67,715 TVF8531-DIF.pdf
  03/12/2008  11:37 AM           509,340 TVF8533-BDF.pdf
  03/12/2008  11:37 AM           507,295 TVF8533-DIF.pdf
  12/30/2003  06:59 PM            87,921 TVF8831-BDFF.pdf
  12/30/2003  06:59 PM            87,624 TVF8831-DIFF.pdf
  09/26/2005  10:20 AM           176,525 TVF88P3-CFF.pdf
  03/24/2006  10:48 AM           460,941 TVF88T5-BDFF.pdf
  02/24/2004  02:19 PM           132,304 TVF9533-BDF.pdf
  02/24/2004  02:19 PM           120,940 TVF9533-DIF.pdf
  03/12/2008  11:37 AM           458,967 TVF99T5-BDFF.pdf

--
video4linux-list mailing list
Unsubscribe mailto:[email protected]?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

author
#7 lore raw

Hi Pavle,

Quoted 3 lines · Pavle Predic

Hey Darek,

Great job of making the card work. I was really thrilled when I saw your post. However, I am a total newbie, so I couldn't apply the changes you wrote about. Could you please be a bit more specific? What I did is downloaded the driver from here: http://dl.bytesex.org/releases/video4linux/saa7134-0.2.12.tar.gz and made the changes to those two files, as described. But I have no clue how to compile it. I installed linux-headers for my kernel version and tried to run make, but I'm getting an error. Are there any configuration options that I need to set in Makefile or Make.config?

I'm not sure whether downloading and compiling driver is good idea. v4l drivers (which includes saa7134) are included in mainline kernel, so compiling kernel is what you have to do. From what I know, in Debian there should be package in repositories with kernel sources (linux-source or similar) - this option you should use if you want to stick to the kernel version provided by your distribution. Another option is to download kernel sources from kernel.org and use them (I've done so - I'm using latest stable release). Here you have link to how-to about kernel compiling: https://help.ubuntu.com/6.10/ubuntu/installation-guide/i386/kernel-baking.html. It's for Ubuntu, but for you it should be also applicable (on bottom there is also link to Debian documentation).

Before compilation you should make changes which I gave earlier. All files which should be modified are in <kernel_src_path>/drivers/media/video/saa7134/ directory. Have in mind that what I've done is that I've changed existing card configuration - it's not proper solution. When I'll manage remote control to work, I'll try to prepare patch with new card configuration. You can apply my changes now or wait until I'll prepare the patch.

Regards,
Darek

Quoted 317 lines

Cheers,

Pavle.

PS My kernel is 2.6.26-2-686 (it's a Debian Lenny 503 system). Here is the output produced by make:

debian:/home/pavle/saa7134-0.2.12# make
make -C /lib/modules/2.6.26-2-686/build SUBDIRS=/home/pavle/saa7134-0.2.12 modules
make[1]: Entering directory `/usr/src/linux-headers-2.6.26-2-686'
  CC [M]  /home/pavle/saa7134-0.2.12/video-buf.o
In file included from /home/pavle/saa7134-0.2.12/media/video-buf.h:19,
                 from /home/pavle/saa7134-0.2.12/video-buf.c:33:
/home/pavle/saa7134-0.2.12/linux/videodev.h:68: error: field 'class_dev' has incomplete type
/home/pavle/saa7134-0.2.12/linux/videodev.h:87: warning: 'struct class_device_attribute' declared inside parameter list
/home/pavle/saa7134-0.2.12/linux/videodev.h:87: warning: its scope is only this definition or declaration, which is probably not what you want
/home/pavle/saa7134-0.2.12/linux/videodev.h: In function 'video_device_create_file':
/home/pavle/saa7134-0.2.12/linux/videodev.h:89: error: implicit declaration of function 'class_device_create_file'
/home/pavle/saa7134-0.2.12/linux/videodev.h: At top level:
/home/pavle/saa7134-0.2.12/linux/videodev.h:93: warning: 'struct class_device_attribute' declared inside parameter list
/home/pavle/saa7134-0.2.12/linux/videodev.h: In function 'video_device_remove_file':
/home/pavle/saa7134-0.2.12/linux/videodev.h:95: error: implicit declaration of function 'class_device_remove_file'
/home/pavle/saa7134-0.2.12/video-buf.c: At top level:
/home/pavle/saa7134-0.2.12/video-buf.c:46: error: expected ')' before string constant
/home/pavle/saa7134-0.2.12/video-buf.c: In function 'videobuf_vmalloc_to_sg':
/home/pavle/saa7134-0.2.12/video-buf.c:68: error: 'struct scatterlist' has no member named 'page'
/home/pavle/saa7134-0.2.12/video-buf.c: In function 'videobuf_pages_to_sg':
/home/pavle/saa7134-0.2.12/video-buf.c:96: error: 'struct scatterlist' has no member named 'page'
/home/pavle/saa7134-0.2.12/video-buf.c:104: error: 'struct scatterlist' has no member named 'page'
/home/pavle/saa7134-0.2.12/video-buf.c: In function 'videobuf_vm_nopage':
/home/pavle/saa7134-0.2.12/video-buf.c:1099: error: 'NOPAGE_SIGBUS' undeclared (first use in this function)
/home/pavle/saa7134-0.2.12/video-buf.c:1099: error: (Each undeclared identifier is reported only once
/home/pavle/saa7134-0.2.12/video-buf.c:1099: error: for each function it appears in.)
/home/pavle/saa7134-0.2.12/video-buf.c:1102: error: 'NOPAGE_OOM' undeclared (first use in this function)
/home/pavle/saa7134-0.2.12/video-buf.c: At top level:
/home/pavle/saa7134-0.2.12/video-buf.c:1119: error: unknown field 'nopage' specified in initializer
/home/pavle/saa7134-0.2.12/video-buf.c:1119: warning: initialization from incompatible pointer type
make[2]: *** [/home/pavle/saa7134-0.2.12/video-buf.o] Error 1
make[1]: *** [_module_/home/pavle/saa7134-0.2.12] Error 2
make[1]: Leaving directory `/usr/src/linux-headers-2.6.26-2-686'
make: *** [default] Error 2

------------------------------------------------------------------------
*From:* dz-tor <[email protected]>
*To:* hermann pitton <[email protected]>
*Cc:* [email protected]; [email protected]; [email protected] *Sent:* Fri, 8 January, 2010 23:56:54
*Subject:* Re: Leadtek Winfast TV2100

Hi,

My friend helped me a little and we managed card to work, i.e. TV (only mono, I suppose that stereo is not available for this card), s-video, composite and radio works fine.

We've tried setting audio_clock to both values (32.11 MHz -> .audio_clock=0x187de7 and 24.576MHz -> .audio_clock=0x200000) and both works - for me it seems, that for 32.11 MHz sound is better (i.e. louder).

Now I have only one problem with ir. We succeeded to enable it, but reading from it is bit weird - sometimes when I press the key, no event is generated, sometimes 2 (key pressed/released) and sometimes more. Have any clue what can cause that?

Output from dmesg when module is loaded:

[32627.844221] saa7130/34: v4l2 driver version 0.2.15 loaded
[32627.844294] saa7130[0]: found at 0000:01:09.0, rev: 1, irq: 17, latency: 32, mmio: 0xe7001000
[32627.844303] saa7130[0]: subsystem: 107d:6f3a, board: Proteus Pro [philips reference design] [card=1,insmod option]
[32627.844340] saa7130[0]: board init: gpio is 60a08
[32627.844351] saa7130[0]: gpio: mode=0x0000000 in=0x0060a0c out=0x0000000 [pre-init]
[32627.844901] input: saa7134 IR (Proteus Pro [philip as /devices/pci0000:00/0000:00:08.0/0000:01:09.0/input/input15
[32627.845064] IRQ 17/saa7130[0]: IRQF_DISABLED is not guaranteed on shared IRQs
[32628.028017] saa7130[0]: i2c eeprom 00: 7d 10 3a 6f 54 20 1c 00 43 43 a9 1c 55 d2 b2 92
[32628.028029] saa7130[0]: i2c eeprom 10: 0c ff 82 0e ff 20 ff ff ff ff ff ff ff ff ff ff
[32628.028038] saa7130[0]: i2c eeprom 20: 01 40 02 03 03 02 01 03 08 ff 00 8c ff ff ff ff
[32628.028047] saa7130[0]: i2c eeprom 30: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028056] saa7130[0]: i2c eeprom 40: 50 89 00 c2 00 00 02 30 02 ff ff ff ff ff ff ff
[32628.028064] saa7130[0]: i2c eeprom 50: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028073] saa7130[0]: i2c eeprom 60: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028082] saa7130[0]: i2c eeprom 70: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028091] saa7130[0]: i2c eeprom 80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028099] saa7130[0]: i2c eeprom 90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028108] saa7130[0]: i2c eeprom a0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028117] saa7130[0]: i2c eeprom b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028126] saa7130[0]: i2c eeprom c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028134] saa7130[0]: i2c eeprom d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028143] saa7130[0]: i2c eeprom e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028152] saa7130[0]: i2c eeprom f0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
[32628.028163] i2c i2c-2: Invalid 7-bit address 0x7a
[32628.036120] tuner 2-0061: chip found @ 0xc2 (saa7130[0])
[32628.044037] tuner-simple 2-0061: creating new instance
[32628.044044] tuner-simple 2-0061: type set to 69 (Tena TNF 5335 and similar models)
[32628.052058] saa7130[0]: gpio: mode=0x000000d in=0x0060a00 out=0x0000008 [mute]
[32628.052178] saa7130[0]: registered device video1 [v4l2]
[32628.052205] saa7130[0]: registered device vbi0
[32628.052232] saa7130[0]: registered device radio0

For testing we've used existing card (card=1 - Proteus Pro) with tuner 69. When everything will be working ok, my friend will try to add new card definition and generate the patch. Below you can find actual code modified by us:

saa7134-cards.c
struct saa7134_board saa7134_boards[] = {
...

    [SAA7134_BOARD_PROTEUS_PRO] = {
        /* /me */
        .name        = "Proteus Pro [philips reference design]",
        .audio_clock    = 0x00187de7,
        .tuner_type    = TUNER_TNF_5335MF,
        .radio_type    = UNSET,
        .tuner_addr    = ADDR_UNSET,
        .radio_addr    = ADDR_UNSET,

        .gpiomask    = 0x0d,
        .inputs        = {{
            .name = name_comp1,
            .vmux = 3,
            .amux = LINE2,
            .gpio = 0x08,
        },{
            .name = name_tv,
            .vmux = 1,
            .amux = LINE1,
            .tv  = 1,
        },{
            .name = name_tv_mono,
            .vmux = 1,
            .amux = LINE1,
            .tv  = 1,
        },{
            .name = name_svideo,
            .vmux = 8,
            .amux = LINE2,
            .gpio = 0x08,
        }},
        .radio = {
            .name = name_radio,
            .amux = LINE1,
            .gpio = 0x04,
        },
        .mute = {
            .name = name_mute,
            .amux = TV,
            .gpio = 0x08,
        },
    },

int saa7134_board_init1(struct saa7134_dev *dev)
{
...

    case SAA7134_BOARD_PROTEUS_PRO:
        dev->has_remote = SAA7134_REMOTE_GPIO;
        break;

saa7134-input.c
int saa7134_input_init1(struct saa7134_dev *dev)

{
    case SAA7134_BOARD_PROTEUS_PRO:
        ir_codes    = &ir_codes_avermedia_table;
        mask_keycode = 0x000000;
        polling      = 50; // ms

Regards,
Darek

Quoted 7 lines · hermann pitton

Hi,

Am Sonntag, den 03.01.2010, 16:37 +0100 schrieb dz-tor:

Quoted 3 lines

Hi,

I want to ask whether you are working on this 'issue'. I'm

interested in

Quoted 1 line
Quoted 1 line

the subject, as I'm owner of this card and cannot get this stuff to

work

Quoted 1 line
Quoted 1 line

- I've picture, no sound. Earlier you have written, that you can

provide

Quoted 2 lines
Quoted 2 lines

patches for testing. If offer is still actual, I can test them. Currently I'm using kernel 2.6.31, but if it's a problem I can

switch to

Quoted 10 lines
Quoted 8 lines

2.6.32.

As I wrote earlier I have the same card - Winfast TV2100 with FM, tv norm used in my country is PAL (I'm from Poland).

Regards,
Darek

please have a look at the xtal/oscillator close to the saa713x chip.

Quoted 28 lines

It can have two different values. See README.saa7134 in Documentation.

I'll provide a patch with best guessing based on the regspy.exe results then, assuming all is connected to LINE2 and routed through the external mux chip as a start. If that fails, we try with LINE1 next.

Cheers,
Hermann

Quoted 17 lines
Quoted 17 lines
Quoted 17 lines

Hey guys,

Let me start by thanking you both for all your help. Unfortunately, there isn't much I can do with the data you provided - it's way too technical for me. I'd be happy to do any tests and apply any patches, but I would need some instructions - but only if you have time; I really don't want to burden you with this.

Thanks again,

Pavle.

________________________________
From: hermann pitton<[email protected]

<mailto:[email protected]>>

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

To: Terry Wu<[email protected] <mailto:[email protected]>>
Cc: [email protected] <mailto:[email protected]>;

[email protected] <mailto:[email protected]>

Quoted 16 lines
Quoted 16 lines
Quoted 16 lines
Quoted 16 lines

Sent: Sat, 28 November, 2009 4:09:28
Subject: Re: Re: Leadtek Winfast TV2100

Hi Terry Wu,

Am Donnerstag, den 26.11.2009, 10:33 +0800 schrieb Terry Wu:

Quoted 9 lines

Hi,

Here are the subsystem IDs for different TV2100 models:

     Subsystem ID:0x6f30107d, TVF8533-BDF (PAL BG/DK)
     Subsystem ID:0x6f32107d, TVF5533-MF (NTSC)
     Subsystem ID:0x6f3a107d, TVF88T5-B/DFF (PAL BG/DK, FM)

Terry

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

better is to become active on it.

we can most likely help about how to match such tuners fine, if any doubts left, but the best is to send just patches with having such hardware to test on.

Compared to the early tda9887 stuff on LeadTek devices, LeadTek was pioneering and we have that in mind, it should be fairly easy for

those.

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Let's know the other way round too, where you have still concerns

doing

Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines

so.

Cheers,
Hermann

Quoted 1 line

2009/11/23 hermann pitton<[email protected]

<mailto:[email protected]>>:

Quoted 64 lines
Quoted 64 lines
Quoted 64 lines
Quoted 64 lines
Quoted 9 lines
Quoted 9 lines

Hi, thanks again!

Am Montag, den 23.11.2009, 09:42 +0800 schrieb Terry Wu:

Quoted 4 lines

Hi,

Please refer to the attached JPEG file for the GPIO

settings of

Quoted 9 lines
Quoted 9 lines
Quoted 7 lines

TV2100 with FM (PCB:B).

Let me know if you need the information of TV2100 without FM
(PCB:A, TVF8533-BDF).

Terry Wu

On a first look, if we start to count gpios from zero, we tell

the same.

Quoted 2 lines
Quoted 2 lines

The TVF8533_BDF I would have to look up. It is four to five

years back.

Quoted 2 lines
Quoted 2 lines

If it uses that minor number TI chip without radio support, we

treat it

Quoted 4 lines
Quoted 4 lines

as tuner=69 too currently.

For all such older can tuners, widely different about the globe,

counts,

Quoted 2 lines
Quoted 2 lines

that we don't have any way to detect them. So first working,

either NTSC

Quoted 4 lines
Quoted 4 lines

or PAL, sits in the pool position and others have to think twice.

OEMs do code tuners into eeprom content, some do not at all,

such doing

Quoted 7 lines
Quoted 7 lines

it are in competition and don't follow the rules of the main chip manufacturer, Philips/NXP in that case, and go their own ways.

So tuner tables are unstable across manufacturers.

We often can't help that much in such cases, but implementing

their own

Quoted 5 lines
Quoted 5 lines

tuner eeprom detection into the linux drivers is of course still welcome. Hauppauge does it very successfully since years.

We can't do much about it, if OEMs don't follow Philips or whom

ever on

Quoted 2 lines
Quoted 2 lines

such.

Quoted 42 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines

Thanks,
Hermann

Quoted 1 line

2009/11/23 Terry Wu<[email protected]

<mailto:[email protected]>>:

Quoted 33 lines
Quoted 30 lines
Quoted 30 lines
Quoted 30 lines
Quoted 28 lines
Quoted 28 lines
Quoted 25 lines
Quoted 25 lines

Hi,

The TVF88T5-BDFF data sheet is attached.

Terry Wu

 11/17/2003  06:39 PM            72,010 TVF5531-MF.pdf
 03/12/2008  11:37 AM          555,285 TVF5533-MF-.pdf
 02/24/2004  02:19 PM          120,727 TVF5533-MF.pdf
 12/30/2003  06:59 PM            91,577 TVF5831-MFF.pdf
 09/26/2005  10:20 AM          156,853 TVF78P3-MFF.pdf
 11/17/2003  06:39 PM            67,947 TVF8531-BDF.pdf
 11/17/2003  06:39 PM            67,715 TVF8531-DIF.pdf
 03/12/2008  11:37 AM          509,340 TVF8533-BDF.pdf
 03/12/2008  11:37 AM          507,295 TVF8533-DIF.pdf
 12/30/2003  06:59 PM            87,921 TVF8831-BDFF.pdf
 12/30/2003  06:59 PM            87,624 TVF8831-DIFF.pdf
 09/26/2005  10:20 AM          176,525 TVF88P3-CFF.pdf
 03/24/2006  10:48 AM          460,941 TVF88T5-BDFF.pdf
 02/24/2004  02:19 PM          132,304 TVF9533-BDF.pdf
 02/24/2004  02:19 PM          120,940 TVF9533-DIF.pdf
 03/12/2008  11:37 AM          458,967 TVF99T5-BDFF.pdf

--
video4linux-list mailing list
Unsubscribe mailto:[email protected]?subject=unsubscribe
https://www.redhat.com/mailman/listinfo/video4linux-list

#8 lore raw

Hi,

Am Samstag, den 09.01.2010, 17:23 +0100 schrieb dz-tor:

Quoted 35 lines

Hi Pavle,

Quoted 11 lines · Pavle Predic

Hey Darek,

Great job of making the card work. I was really thrilled when I saw your post. However, I am a total newbie, so I couldn't apply the changes you wrote about. Could you please be a bit more specific? What I did is downloaded the driver from here:
http://dl.bytesex.org/releases/video4linux/saa7134-0.2.12.tar.gz and made the changes to those two files, as described. But I have no clue how to compile it. I installed linux-headers for my kernel version and tried to run make, but I'm getting an error. Are there any configuration options that I need to set in Makefile or Make.config?

I'm not sure whether downloading and compiling driver is good idea. v4l drivers (which includes saa7134) are included in mainline kernel, so compiling kernel is what you have to do. From what I know, in Debian there should be package in repositories with kernel sources (linux-source or similar) - this option you should use if you want to stick to the kernel version provided by your distribution. Another option is to download kernel sources from kernel.org and use them (I've done so - I'm using latest stable release). Here you have link to how-to about kernel compiling:
https://help.ubuntu.com/6.10/ubuntu/installation-guide/i386/kernel-baking.html.
It's for Ubuntu, but for you it should be also applicable (on bottom there is also link to Debian documentation).

Before compilation you should make changes which I gave earlier. All files which should be modified are in <kernel_src_path>/drivers/media/video/saa7134/ directory. Have in mind that what I've done is that I've changed existing card configuration - it's not proper solution. When I'll manage remote control to work, I'll try to prepare patch with new card configuration. You can apply my changes now or wait until I'll prepare the patch.

Quoted 2 lines

Regards,
Darek

Great!

So Pavle's regspy results and following my instructions did the trick.

Patches must go to LMML <[email protected]> and be against latest v4l-dvb master at linuxtv.org.

You can use what I prepared already yesterday for testing and is attached. You have only to change the clock and use LINE1 for external audio input. I suggest to use also LINE1 for mute then, gpio 0x08 is that input anyway.

We can send that modified patch already and add IR support later.

You must find the up/down gpio and with mask_keycode = 0x000000 all the other gpios which do change on keypresses and create unique keycodes. Then you either need to add a new keytable or can use an already existing one.

Yes, saa7130 chips have only mono TV sound.

Cheers,
Hermann

Quoted 398 lines
Quoted 398 lines

Cheers,

Pavle.

PS My kernel is 2.6.26-2-686 (it's a Debian Lenny 503 system). Here is the output produced by make:

debian:/home/pavle/saa7134-0.2.12# make
make -C /lib/modules/2.6.26-2-686/build 
SUBDIRS=/home/pavle/saa7134-0.2.12 modules
make[1]: Entering directory `/usr/src/linux-headers-2.6.26-2-686'
  CC [M]  /home/pavle/saa7134-0.2.12/video-buf.o
In file included from /home/pavle/saa7134-0.2.12/media/video-buf.h:19,
                 from /home/pavle/saa7134-0.2.12/video-buf.c:33:
/home/pavle/saa7134-0.2.12/linux/videodev.h:68: error: field 
'class_dev' has incomplete type
/home/pavle/saa7134-0.2.12/linux/videodev.h:87: warning: 'struct 
class_device_attribute' declared inside parameter list
/home/pavle/saa7134-0.2.12/linux/videodev.h:87: warning: its scope is 
only this definition or declaration, which is probably not what you want
/home/pavle/saa7134-0.2.12/linux/videodev.h: In function 
'video_device_create_file':
/home/pavle/saa7134-0.2.12/linux/videodev.h:89: error: implicit 
declaration of function 'class_device_create_file'
/home/pavle/saa7134-0.2.12/linux/videodev.h: At top level:
/home/pavle/saa7134-0.2.12/linux/videodev.h:93: warning: 'struct 
class_device_attribute' declared inside parameter list
/home/pavle/saa7134-0.2.12/linux/videodev.h: In function 
'video_device_remove_file':
/home/pavle/saa7134-0.2.12/linux/videodev.h:95: error: implicit 
declaration of function 'class_device_remove_file'
/home/pavle/saa7134-0.2.12/video-buf.c: At top level:
/home/pavle/saa7134-0.2.12/video-buf.c:46: error: expected ')' before 
string constant
/home/pavle/saa7134-0.2.12/video-buf.c: In function 
'videobuf_vmalloc_to_sg':
/home/pavle/saa7134-0.2.12/video-buf.c:68: error: 'struct scatterlist' 
has no member named 'page'
/home/pavle/saa7134-0.2.12/video-buf.c: In function 
'videobuf_pages_to_sg':
/home/pavle/saa7134-0.2.12/video-buf.c:96: error: 'struct scatterlist' 
has no member named 'page'
/home/pavle/saa7134-0.2.12/video-buf.c:104: error: 'struct 
scatterlist' has no member named 'page'
/home/pavle/saa7134-0.2.12/video-buf.c: In function 'videobuf_vm_nopage':
/home/pavle/saa7134-0.2.12/video-buf.c:1099: error: 'NOPAGE_SIGBUS' 
undeclared (first use in this function)
/home/pavle/saa7134-0.2.12/video-buf.c:1099: error: (Each undeclared 
identifier is reported only once
/home/pavle/saa7134-0.2.12/video-buf.c:1099: error: for each function 
it appears in.)
/home/pavle/saa7134-0.2.12/video-buf.c:1102: error: 'NOPAGE_OOM' 
undeclared (first use in this function)
/home/pavle/saa7134-0.2.12/video-buf.c: At top level:
/home/pavle/saa7134-0.2.12/video-buf.c:1119: error: unknown field 
'nopage' specified in initializer
/home/pavle/saa7134-0.2.12/video-buf.c:1119: warning: initialization 
from incompatible pointer type
make[2]: *** [/home/pavle/saa7134-0.2.12/video-buf.o] Error 1
make[1]: *** [_module_/home/pavle/saa7134-0.2.12] Error 2
make[1]: Leaving directory `/usr/src/linux-headers-2.6.26-2-686'
make: *** [default] Error 2

------------------------------------------------------------------------
*From:* dz-tor <[email protected]>
*To:* hermann pitton <[email protected]> *Cc:* [email protected]; [email protected]; [email protected]
*Sent:* Fri, 8 January, 2010 23:56:54 *Subject:* Re: Leadtek Winfast TV2100

Hi,

My friend helped me a little and we managed card to work, i.e. TV (only mono, I suppose that stereo is not available for this card), s-video, composite and radio works fine.

We've tried setting audio_clock to both values (32.11 MHz -> .audio_clock=0x187de7 and 24.576MHz -> .audio_clock=0x200000) and both works - for me it seems, that for 32.11 MHz sound is better (i.e. louder).

Now I have only one problem with ir. We succeeded to enable it, but reading from it is bit weird - sometimes when I press the key, no event is generated, sometimes 2 (key pressed/released) and sometimes more. Have any clue what can cause that?

Output from dmesg when module is loaded:

[32627.844221] saa7130/34: v4l2 driver version 0.2.15 loaded
[32627.844294] saa7130[0]: found at 0000:01:09.0, rev: 1, irq: 17, 
latency: 32, mmio: 0xe7001000
[32627.844303] saa7130[0]: subsystem: 107d:6f3a, board: Proteus Pro 
[philips reference design] [card=1,insmod option]
[32627.844340] saa7130[0]: board init: gpio is 60a08
[32627.844351] saa7130[0]: gpio: mode=0x0000000 in=0x0060a0c 
out=0x0000000 [pre-init]
[32627.844901] input: saa7134 IR (Proteus Pro [philip as 
/devices/pci0000:00/0000:00:08.0/0000:01:09.0/input/input15
[32627.845064] IRQ 17/saa7130[0]: IRQF_DISABLED is not guaranteed on 
shared IRQs
[32628.028017] saa7130[0]: i2c eeprom 00: 7d 10 3a 6f 54 20 1c 00 43 
43 a9 1c 55 d2 b2 92
[32628.028029] saa7130[0]: i2c eeprom 10: 0c ff 82 0e ff 20 ff ff ff 
ff ff ff ff ff ff ff
[32628.028038] saa7130[0]: i2c eeprom 20: 01 40 02 03 03 02 01 03 08 
ff 00 8c ff ff ff ff
[32628.028047] saa7130[0]: i2c eeprom 30: ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff
[32628.028056] saa7130[0]: i2c eeprom 40: 50 89 00 c2 00 00 02 30 02 
ff ff ff ff ff ff ff
[32628.028064] saa7130[0]: i2c eeprom 50: ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff
[32628.028073] saa7130[0]: i2c eeprom 60: ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff
[32628.028082] saa7130[0]: i2c eeprom 70: ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff
[32628.028091] saa7130[0]: i2c eeprom 80: ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff
[32628.028099] saa7130[0]: i2c eeprom 90: ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff
[32628.028108] saa7130[0]: i2c eeprom a0: ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff
[32628.028117] saa7130[0]: i2c eeprom b0: ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff
[32628.028126] saa7130[0]: i2c eeprom c0: ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff
[32628.028134] saa7130[0]: i2c eeprom d0: ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff
[32628.028143] saa7130[0]: i2c eeprom e0: ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff
[32628.028152] saa7130[0]: i2c eeprom f0: ff ff ff ff ff ff ff ff ff 
ff ff ff ff ff ff ff
[32628.028163] i2c i2c-2: Invalid 7-bit address 0x7a
[32628.036120] tuner 2-0061: chip found @ 0xc2 (saa7130[0])
[32628.044037] tuner-simple 2-0061: creating new instance
[32628.044044] tuner-simple 2-0061: type set to 69 (Tena TNF 5335 and 
similar models)
[32628.052058] saa7130[0]: gpio: mode=0x000000d in=0x0060a00 
out=0x0000008 [mute]
[32628.052178] saa7130[0]: registered device video1 [v4l2]
[32628.052205] saa7130[0]: registered device vbi0
[32628.052232] saa7130[0]: registered device radio0

For testing we've used existing card (card=1 - Proteus Pro) with tuner
69. When everything will be working ok, my friend will try to add new card definition and generate the patch. Below you can find actual code modified by us:

saa7134-cards.c
struct saa7134_board saa7134_boards[] = {
...

    [SAA7134_BOARD_PROTEUS_PRO] = {
        /* /me */
        .name        = "Proteus Pro [philips reference design]",
        .audio_clock    = 0x00187de7,
        .tuner_type    = TUNER_TNF_5335MF,
        .radio_type    = UNSET,
        .tuner_addr    = ADDR_UNSET,
        .radio_addr    = ADDR_UNSET,

        .gpiomask    = 0x0d,
        .inputs        = {{
            .name = name_comp1,
            .vmux = 3,
            .amux = LINE2,
            .gpio = 0x08,
        },{
            .name = name_tv,
            .vmux = 1,
            .amux = LINE1,
            .tv  = 1,
        },{
            .name = name_tv_mono,
            .vmux = 1,
            .amux = LINE1,
            .tv  = 1,
        },{
            .name = name_svideo,
            .vmux = 8,
            .amux = LINE2,
            .gpio = 0x08,
        }},
        .radio = {
            .name = name_radio,
            .amux = LINE1,
            .gpio = 0x04,
        },
        .mute = {
            .name = name_mute,
            .amux = TV,
            .gpio = 0x08,
        },
    },

int saa7134_board_init1(struct saa7134_dev *dev)
{
...

    case SAA7134_BOARD_PROTEUS_PRO:
        dev->has_remote = SAA7134_REMOTE_GPIO;
        break;

saa7134-input.c
int saa7134_input_init1(struct saa7134_dev *dev)

{
    case SAA7134_BOARD_PROTEUS_PRO:
        ir_codes    = &ir_codes_avermedia_table;
        mask_keycode = 0x000000;
        polling      = 50; // ms

Regards,
Darek

Quoted 7 lines · hermann pitton

Hi,

Am Sonntag, den 03.01.2010, 16:37 +0100 schrieb dz-tor:

Quoted 3 lines

Hi,

I want to ask whether you are working on this 'issue'. I'm

interested in

Quoted 1 line
Quoted 1 line

the subject, as I'm owner of this card and cannot get this stuff to

work

Quoted 1 line
Quoted 1 line

- I've picture, no sound. Earlier you have written, that you can

provide

Quoted 2 lines
Quoted 2 lines

patches for testing. If offer is still actual, I can test them. Currently I'm using kernel 2.6.31, but if it's a problem I can

switch to

Quoted 10 lines
Quoted 8 lines

2.6.32.

As I wrote earlier I have the same card - Winfast TV2100 with FM, tv norm used in my country is PAL (I'm from Poland).

Regards,
Darek

please have a look at the xtal/oscillator close to the saa713x chip.

Quoted 28 lines

It can have two different values. See README.saa7134 in Documentation.

I'll provide a patch with best guessing based on the regspy.exe results then, assuming all is connected to LINE2 and routed through the external mux chip as a start. If that fails, we try with LINE1 next.

Cheers,
Hermann

Quoted 17 lines
Quoted 17 lines
Quoted 17 lines

Hey guys,

Let me start by thanking you both for all your help. Unfortunately, there isn't much I can do with the data you provided - it's way too technical for me. I'd be happy to do any tests and apply any patches, but I would need some instructions - but only if you have time; I really don't want to burden you with this.

Thanks again,

Pavle.

________________________________
From: hermann pitton<[email protected]

<mailto:[email protected]>>

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

To: Terry Wu<[email protected] <mailto:[email protected]>>
Cc: [email protected] <mailto:[email protected]>;

[email protected] <mailto:[email protected]>

Quoted 16 lines
Quoted 16 lines
Quoted 16 lines
Quoted 16 lines

Sent: Sat, 28 November, 2009 4:09:28
Subject: Re: Re: Leadtek Winfast TV2100

Hi Terry Wu,

Am Donnerstag, den 26.11.2009, 10:33 +0800 schrieb Terry Wu:

Quoted 9 lines

Hi,

Here are the subsystem IDs for different TV2100 models:

     Subsystem ID:0x6f30107d, TVF8533-BDF (PAL BG/DK)
     Subsystem ID:0x6f32107d, TVF5533-MF (NTSC)
     Subsystem ID:0x6f3a107d, TVF88T5-B/DFF (PAL BG/DK, FM)

Terry

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

better is to become active on it.

we can most likely help about how to match such tuners fine, if any doubts left, but the best is to send just patches with having such hardware to test on.

Compared to the early tda9887 stuff on LeadTek devices, LeadTek was pioneering and we have that in mind, it should be fairly easy for

those.

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

Let's know the other way round too, where you have still concerns

doing

Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines

so.

Cheers,
Hermann

Quoted 1 line

2009/11/23 hermann pitton<[email protected]

<mailto:[email protected]>>:

Quoted 62 lines
Quoted 62 lines
Quoted 62 lines
Quoted 62 lines
Quoted 9 lines
Quoted 9 lines

Hi, thanks again!

Am Montag, den 23.11.2009, 09:42 +0800 schrieb Terry Wu:

Quoted 4 lines

Hi,

Please refer to the attached JPEG file for the GPIO

settings of

Quoted 9 lines
Quoted 9 lines
Quoted 7 lines

TV2100 with FM (PCB:B).

Let me know if you need the information of TV2100 without FM
(PCB:A, TVF8533-BDF).

Terry Wu

On a first look, if we start to count gpios from zero, we tell

the same.

Quoted 2 lines
Quoted 2 lines

The TVF8533_BDF I would have to look up. It is four to five

years back.

Quoted 2 lines
Quoted 2 lines

If it uses that minor number TI chip without radio support, we

treat it

Quoted 4 lines
Quoted 4 lines

as tuner=69 too currently.

For all such older can tuners, widely different about the globe,

counts,

Quoted 2 lines
Quoted 2 lines

that we don't have any way to detect them. So first working,

either NTSC

Quoted 4 lines
Quoted 4 lines

or PAL, sits in the pool position and others have to think twice.

OEMs do code tuners into eeprom content, some do not at all,

such doing

Quoted 7 lines
Quoted 7 lines

it are in competition and don't follow the rules of the main chip manufacturer, Philips/NXP in that case, and go their own ways.

So tuner tables are unstable across manufacturers.

We often can't help that much in such cases, but implementing

their own

Quoted 5 lines
Quoted 5 lines

tuner eeprom detection into the linux drivers is of course still welcome. Hauppauge does it very successfully since years.

We can't do much about it, if OEMs don't follow Philips or whom

ever on

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

such.

Thanks,
Hermann

Quoted 1 line

2009/11/23 Terry Wu<[email protected]

<mailto:[email protected]>>:

Quoted 17 lines
Quoted 17 lines
Quoted 17 lines
Quoted 17 lines
Quoted 17 lines
Quoted 17 lines
Quoted 17 lines
Quoted 17 lines

Hi,

The TVF88T5-BDFF data sheet is attached.

Terry Wu

 11/17/2003  06:39 PM            72,010 TVF5531-MF.pdf
 03/12/2008  11:37 AM          555,285 TVF5533-MF-.pdf
 02/24/2004  02:19 PM          120,727 TVF5533-MF.pdf
 12/30/2003  06:59 PM            91,577 TVF5831-MFF.pdf
 09/26/2005  10:20 AM          156,853 TVF78P3-MFF.pdf
 11/17/2003  06:39 PM            67,947 TVF8531-BDF.pdf
 11/17/2003  06:39 PM            67,715 TVF8531-DIF.pdf
 03/12/2008  11:37 AM          509,340 TVF8533-BDF.pdf
 03/12/2008  11:37 AM          507,295 TVF8533-DIF.pdf
 12/30/2003  06:59 PM            87,921 TVF8831-BDFF.pdf
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
 12/30/2003  06:59 PM            87,624 TVF8831-DIFF.pdf
 09/26/2005  10:20 AM          176,525 TVF88P3-CFF.pdf
 03/24/2006  10:48 AM          460,941 TVF88T5-BDFF.pdf
 02/24/2004  02:19 PM          132,304 TVF9533-BDF.pdf
 02/24/2004  02:19 PM          120,940 TVF9533-DIF.pdf
 03/12/2008  11:37 AM          458,967 TVF99T5-BDFF.pdf
linux/drivers/media/video/saa7134/saa7134-cards.c +47 -1
diff -r b6b82258cf5e linux/drivers/media/video/saa7134/saa7134-cards.c
@@ -5359,7 +5359,47 @@
 			.vmux = 8,
 		} },
 	},
-
+	[SAA7134_BOARD_LEADTEK_WINFAST_TV2100FM] = {
+		.name           = "Leadtek Winfast 2100FM",
+		.audio_clock    = 0x00200000,
+		.tuner_type     = TUNER_TNF_5335MF,
+		.radio_type     = UNSET,
+		.tuner_addr	= ADDR_UNSET,
+		.radio_addr	= ADDR_UNSET,
+		.gpiomask       = 0xe0d,
+		.inputs         = {{
+			.name = name_tv,
+			.vmux = 1,
+			.amux = LINE2,
+			.gpio = 0x00,
+			.tv   = 1,
+		}, {
+			.name = name_comp1,
+			.vmux = 0,
+			.amux = LINE2,
+			.gpio = 0x08,
+		}, {
+			.name = name_comp2,
+			.vmux = 3,
+			.amux = LINE2,
+			.gpio = 0x08,
+		}, {
+			.name = name_svideo,
+			.vmux = 8,
+			.amux = LINE2,
+			.gpio = 0x08,
+		} },
+		.radio = {
+			.name = name_radio,
+			.amux = LINE2,
+			.gpio = 0x04,
+		},
+		.mute = {
+			.name = name_mute,
+			.amux = LINE2,
+			.gpio = 0x08,
+		},
+	},
 };
 
 const unsigned int saa7134_bcount = ARRAY_SIZE(saa7134_boards);

@@ -6510,6 +6550,12 @@
 		.subdevice    = 0x6655,
 		.driver_data  = SAA7134_BOARD_LEADTEK_WINFAST_DTV1000S,
 	}, {
+		.vendor       = PCI_VENDOR_ID_PHILIPS,
+		.device       = PCI_DEVICE_ID_PHILIPS_SAA7130,
+		.subvendor    = 0x107d,
+		.subdevice    = 0x6f3a,
+		.driver_data  = SAA7134_BOARD_LEADTEK_WINFAST_TV2100FM,
+	}, {
 		/* --- boards without eeprom + subsystem ID --- */
 		.vendor       = PCI_VENDOR_ID_PHILIPS,
 		.device       = PCI_DEVICE_ID_PHILIPS_SAA7134,
linux/drivers/media/video/saa7134/saa7134.h +1 -0
diff -r b6b82258cf5e linux/drivers/media/video/saa7134/saa7134.h
@@ -300,6 +300,7 @@
 #define SAA7134_BOARD_ZOLID_HYBRID_PCI		173
 #define SAA7134_BOARD_ASUS_EUROPA_HYBRID	174
 #define SAA7134_BOARD_LEADTEK_WINFAST_DTV1000S 175
+#define SAA7134_BOARD_LEADTEK_WINFAST_TV2100FM 176
 
 #define SAA7134_MAXBOARDS 32
 #define SAA7134_INPUT_MAX 8
#9 lore raw

Hi,

sorry, there is a typo in the gpio mask on previously attached patch you might use against current v4l-dvb with your further findings.

Mask 0x0d is sufficient and we don't need any 0xe0d :(

You might also consider to start with vmux = 3 for Composite1 and vmux = 0 for Composite2, that is expected to be over the S-Video connector and should work too.

Does save some plugging around with two composite input devices, if S-Video is not in use.

good night,
Hermann

author
#10 lore raw
In reply to hermann pitton

Hi,

Quoted 44 lines · hermann pitton

Hi,

Am Samstag, den 09.01.2010, 17:23 +0100 schrieb dz-tor:

Quoted 37 lines

Hi Pavle,

Quoted 12 lines · Pavle Predic

Hey Darek,

Great job of making the card work. I was really thrilled when I saw your post. However, I am a total newbie, so I couldn't apply the changes you wrote about. Could you please be a bit more specific? What I did is downloaded the driver from here:
http://dl.bytesex.org/releases/video4linux/saa7134-0.2.12.tar.gz and made the changes to those two files, as described. But I have no clue how to compile it. I installed linux-headers for my kernel version and tried to run make, but I'm getting an error. Are there any configuration options that I need to set in Makefile or Make.config?

I'm not sure whether downloading and compiling driver is good idea. v4l drivers (which includes saa7134) are included in mainline kernel, so compiling kernel is what you have to do. From what I know, in Debian there should be package in repositories with kernel sources (linux-source or similar) - this option you should use if you want to stick to the kernel version provided by your distribution. Another option is to download kernel sources from kernel.org and use them (I've done so - I'm using latest stable release). Here you have link to how-to about kernel compiling:
https://help.ubuntu.com/6.10/ubuntu/installation-guide/i386/kernel-baking.html.
It's for Ubuntu, but for you it should be also applicable (on bottom there is also link to Debian documentation).

Before compilation you should make changes which I gave earlier. All files which should be modified are in <kernel_src_path>/drivers/media/video/saa7134/ directory. Have in mind that what I've done is that I've changed existing card configuration - it's not proper solution. When I'll manage remote control to work, I'll try to prepare patch with new card configuration. You can apply my changes now or wait until I'll prepare the patch.

Quoted 3 lines

Regards,
Darek

Quoted 7 lines

Great!

So Pavle's regspy results and following my instructions did the trick.

Patches must go to LMML<[email protected]> and be against latest v4l-dvb master at linuxtv.org.

Done

Quoted 5 lines

You can use what I prepared already yesterday for testing and is attached. You have only to change the clock and use LINE1 for external audio input. I suggest to use also LINE1 for mute then, gpio 0x08 is that input anyway.

Quoted 7 lines

We can send that modified patch already and add IR support later.

You must find the up/down gpio and with mask_keycode = 0x000000 all the other gpios which do change on keypresses and create unique keycodes. Then you either need to add a new keytable or can use an already existing one.

I understand what have to be done to get support for remote control - problem lays somewhere else. Later I'll send more details with code and outputs from dmesg and description of the problem.

Regards,
Darek