Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[Jose Alberto Reguero]

I made the patch for the af9035. I have not a af9015 whith mxl5007 and dual channel. Revert it, if it cause regresions.

Jose Alberto
¡
Enviado desde mi ASUS Pad

poma <pomidorabelisima@gmail.com> wrote:

>On 26.05.2015 14:24, poma wrote:
>> 
>> If it is not taken into account the already known problem of unreliable operation of the first tuner of the two,
>> the device works reliably within kernel 4.0.4 with mxl5007t.ko reverted to
>> http://git.linuxtv.org/cgit.cgi/media_tree.git/commit/drivers/media/tuners/mxl5007t.c?id=ccae7af
>> that is in the same state as is in the longterm kernel - 3.18.14,
>> which is in correspondence with the aforementioned results.
>> 
>> 
>
>http://git.linuxtv.org/cgit.cgi/media_tree.git/log/drivers/media/tuners/mxl5007t.c
>
>
>5. 2014-11-11   [media] [PATH,2/2] mxl5007 move loop_thru to attach
>   02f9cf9      Jose Alberto Reguero
>   
>4. 2014-11-11   [media] [PATH,1/2] mxl5007 move reset to attach
>   fe4860a      Jose Alberto Reguero
>
>3. 2013-02-08   Revert "[media] [PATH,1/2] mxl5007 move reset to attach"
>   db5c05b      Mauro Carvalho Chehab
>
>2. 2013-02-08   [media] [PATH,1/2] mxl5007 move reset to attach
>   0a32377      Jose Alberto Reguero
>
>1. 2012-08-14   [media] common: move media/common/tuners to media/tuners
>   ccae7af      Mauro Carvalho Chehab
>
>
>This is the conclusion after extensive testing,
>commitas 5. 4. and 2. produce:
>
>mxl5007t_soft_reset: 521: failed!
>mxl5007t_attach: error -121 on line 907
>
>causing the device completely unusable - AF9015 DVB-T USB2.0 stick
>
>
>Do you need a patch to revert to commita 3. or 1. - again for the third time,
>or you have a better solution?
>
>

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[poma]

On 06.07.2015 22:17, Jose Alberto Reguero wrote:
> I made the patch for the af9035. I have not a af9015 whith mxl5007 and dual channel. Revert it, if it cause regresions.
> 
> Jose Alberto
> 

Thanks.

>From e19560ea038e54dc57be717db55f19d449df63f0 Mon Sep 17 00:00:00 2001
From: poma <pomidorabelisima@gmail.com>
Date: Tue, 7 Jul 2015 10:26:13 +0200
Subject: [PATCH] Fix for AF9015 DVB-T USB2.0 stick

This reverts commitas:

- 02f9cf96df57575acea2e6eb4041e9f3ecd32548
  "[media] [PATH,2/2] mxl5007 move loop_thru to attach"
- fe4860af002a4516dd878f7297b61e186c475b35
  "[media] [PATH,1/2] mxl5007 move reset to attach"

This is the conclusion after extensive testing,
these two commitas produce:

mxl5007t_soft_reset: 521: failed!
mxl5007t_attach: error -121 on line 907

causing AF9015 DVB-T USB2.0 stick completely unusable.


Tested-by: poma <pomidorabelisima@gmail.com>
---
 drivers/media/tuners/mxl5007t.c | 30 +++++-------------------------
 1 file changed, 5 insertions(+), 25 deletions(-)

diff --git a/drivers/media/tuners/mxl5007t.c b/drivers/media/tuners/mxl5007t.c
index f4ae04c..f8c4ba2 100644
--- a/drivers/media/tuners/mxl5007t.c
+++ b/drivers/media/tuners/mxl5007t.c
@@ -374,6 +374,7 @@ static struct reg_pair_t *mxl5007t_calc_init_regs(struct mxl5007t_state *state,
 	mxl5007t_set_if_freq_bits(state, cfg->if_freq_hz, cfg->invert_if);
 	mxl5007t_set_xtal_freq_bits(state, cfg->xtal_freq_hz);
 
+	set_reg_bits(state->tab_init, 0x04, 0x01, cfg->loop_thru_enable);
 	set_reg_bits(state->tab_init, 0x03, 0x08, cfg->clk_out_enable << 3);
 	set_reg_bits(state->tab_init, 0x03, 0x07, cfg->clk_out_amp);
 
@@ -530,6 +531,10 @@ static int mxl5007t_tuner_init(struct mxl5007t_state *state,
 	struct reg_pair_t *init_regs;
 	int ret;
 
+	ret = mxl5007t_soft_reset(state);
+	if (mxl_fail(ret))
+		goto fail;
+
 	/* calculate initialization reg array */
 	init_regs = mxl5007t_calc_init_regs(state, mode);
 
@@ -895,32 +900,7 @@ struct dvb_frontend *mxl5007t_attach(struct dvb_frontend *fe,
 		/* existing tuner instance */
 		break;
 	}
-
-	if (fe->ops.i2c_gate_ctrl)
-		fe->ops.i2c_gate_ctrl(fe, 1);
-
-	ret = mxl5007t_soft_reset(state);
-
-	if (fe->ops.i2c_gate_ctrl)
-		fe->ops.i2c_gate_ctrl(fe, 0);
-
-	if (mxl_fail(ret))
-		goto fail;
-
-	if (fe->ops.i2c_gate_ctrl)
-		fe->ops.i2c_gate_ctrl(fe, 1);
-
-	ret = mxl5007t_write_reg(state, 0x04,
-		state->config->loop_thru_enable);
-
-	if (fe->ops.i2c_gate_ctrl)
-		fe->ops.i2c_gate_ctrl(fe, 0);
-
-	if (mxl_fail(ret))
-		goto fail;
-
 	fe->tuner_priv = state;
-
 	mutex_unlock(&mxl5007t_list_mutex);
 
 	memcpy(&fe->ops.tuner_ops, &mxl5007t_tuner_ops,
-- 
2.4.3

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[poma]

On 07.07.2015 10:48, poma wrote:
> On 06.07.2015 22:17, Jose Alberto Reguero wrote:
>> I made the patch for the af9035. I have not a af9015 whith mxl5007 and dual channel. Revert it, if it cause regresions.
>>
>> Jose Alberto
>>
> 
> Thanks.
> 
> From e19560ea038e54dc57be717db55f19d449df63f0 Mon Sep 17 00:00:00 2001
> From: poma <pomidorabelisima@gmail.com>
> Date: Tue, 7 Jul 2015 10:26:13 +0200
> Subject: [PATCH] Fix for AF9015 DVB-T USB2.0 stick
> 
> This reverts commitas:
> 
> - 02f9cf96df57575acea2e6eb4041e9f3ecd32548
>   "[media] [PATH,2/2] mxl5007 move loop_thru to attach"
> - fe4860af002a4516dd878f7297b61e186c475b35
>   "[media] [PATH,1/2] mxl5007 move reset to attach"
> 
> This is the conclusion after extensive testing,
> these two commitas produce:
> 
> mxl5007t_soft_reset: 521: failed!
> mxl5007t_attach: error -121 on line 907
> 
> causing AF9015 DVB-T USB2.0 stick completely unusable.
> 
> 
> Tested-by: poma <pomidorabelisima@gmail.com>
> ---
>  drivers/media/tuners/mxl5007t.c | 30 +++++-------------------------
>  1 file changed, 5 insertions(+), 25 deletions(-)
> 
> diff --git a/drivers/media/tuners/mxl5007t.c b/drivers/media/tuners/mxl5007t.c
> index f4ae04c..f8c4ba2 100644
> --- a/drivers/media/tuners/mxl5007t.c
> +++ b/drivers/media/tuners/mxl5007t.c
> @@ -374,6 +374,7 @@ static struct reg_pair_t *mxl5007t_calc_init_regs(struct mxl5007t_state *state,
>  	mxl5007t_set_if_freq_bits(state, cfg->if_freq_hz, cfg->invert_if);
>  	mxl5007t_set_xtal_freq_bits(state, cfg->xtal_freq_hz);
>  
> +	set_reg_bits(state->tab_init, 0x04, 0x01, cfg->loop_thru_enable);
>  	set_reg_bits(state->tab_init, 0x03, 0x08, cfg->clk_out_enable << 3);
>  	set_reg_bits(state->tab_init, 0x03, 0x07, cfg->clk_out_amp);
>  
> @@ -530,6 +531,10 @@ static int mxl5007t_tuner_init(struct mxl5007t_state *state,
>  	struct reg_pair_t *init_regs;
>  	int ret;
>  
> +	ret = mxl5007t_soft_reset(state);
> +	if (mxl_fail(ret))
> +		goto fail;
> +
>  	/* calculate initialization reg array */
>  	init_regs = mxl5007t_calc_init_regs(state, mode);
>  
> @@ -895,32 +900,7 @@ struct dvb_frontend *mxl5007t_attach(struct dvb_frontend *fe,
>  		/* existing tuner instance */
>  		break;
>  	}
> -
> -	if (fe->ops.i2c_gate_ctrl)
> -		fe->ops.i2c_gate_ctrl(fe, 1);
> -
> -	ret = mxl5007t_soft_reset(state);
> -
> -	if (fe->ops.i2c_gate_ctrl)
> -		fe->ops.i2c_gate_ctrl(fe, 0);
> -
> -	if (mxl_fail(ret))
> -		goto fail;
> -
> -	if (fe->ops.i2c_gate_ctrl)
> -		fe->ops.i2c_gate_ctrl(fe, 1);
> -
> -	ret = mxl5007t_write_reg(state, 0x04,
> -		state->config->loop_thru_enable);
> -
> -	if (fe->ops.i2c_gate_ctrl)
> -		fe->ops.i2c_gate_ctrl(fe, 0);
> -
> -	if (mxl_fail(ret))
> -		goto fail;
> -
>  	fe->tuner_priv = state;
> -
>  	mutex_unlock(&mxl5007t_list_mutex);
>  
>  	memcpy(&fe->ops.tuner_ops, &mxl5007t_tuner_ops,
> 


Guys, any chance to push this any time soon?

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[poma]

On 13.07.2015 11:48, poma wrote:
> On 07.07.2015 10:48, poma wrote:
>> On 06.07.2015 22:17, Jose Alberto Reguero wrote:
>>> I made the patch for the af9035. I have not a af9015 whith mxl5007 and dual channel. Revert it, if it cause regresions.
>>>
>>> Jose Alberto
>>>
>>
>> Thanks.
>>
>> From e19560ea038e54dc57be717db55f19d449df63f0 Mon Sep 17 00:00:00 2001
>> From: poma <pomidorabelisima@gmail.com>
>> Date: Tue, 7 Jul 2015 10:26:13 +0200
>> Subject: [PATCH] Fix for AF9015 DVB-T USB2.0 stick
>>
>> This reverts commitas:
>>
>> - 02f9cf96df57575acea2e6eb4041e9f3ecd32548
>>   "[media] [PATH,2/2] mxl5007 move loop_thru to attach"
>> - fe4860af002a4516dd878f7297b61e186c475b35
>>   "[media] [PATH,1/2] mxl5007 move reset to attach"
>>
>> This is the conclusion after extensive testing,
>> these two commitas produce:
>>
>> mxl5007t_soft_reset: 521: failed!
>> mxl5007t_attach: error -121 on line 907
>>
>> causing AF9015 DVB-T USB2.0 stick completely unusable.
>>
>>
>> Tested-by: poma <pomidorabelisima@gmail.com>
>> ---
>>  drivers/media/tuners/mxl5007t.c | 30 +++++-------------------------
>>  1 file changed, 5 insertions(+), 25 deletions(-)
>>
>> diff --git a/drivers/media/tuners/mxl5007t.c b/drivers/media/tuners/mxl5007t.c
>> index f4ae04c..f8c4ba2 100644
>> --- a/drivers/media/tuners/mxl5007t.c
>> +++ b/drivers/media/tuners/mxl5007t.c
>> @@ -374,6 +374,7 @@ static struct reg_pair_t *mxl5007t_calc_init_regs(struct mxl5007t_state *state,
>>  	mxl5007t_set_if_freq_bits(state, cfg->if_freq_hz, cfg->invert_if);
>>  	mxl5007t_set_xtal_freq_bits(state, cfg->xtal_freq_hz);
>>  
>> +	set_reg_bits(state->tab_init, 0x04, 0x01, cfg->loop_thru_enable);
>>  	set_reg_bits(state->tab_init, 0x03, 0x08, cfg->clk_out_enable << 3);
>>  	set_reg_bits(state->tab_init, 0x03, 0x07, cfg->clk_out_amp);
>>  
>> @@ -530,6 +531,10 @@ static int mxl5007t_tuner_init(struct mxl5007t_state *state,
>>  	struct reg_pair_t *init_regs;
>>  	int ret;
>>  
>> +	ret = mxl5007t_soft_reset(state);
>> +	if (mxl_fail(ret))
>> +		goto fail;
>> +
>>  	/* calculate initialization reg array */
>>  	init_regs = mxl5007t_calc_init_regs(state, mode);
>>  
>> @@ -895,32 +900,7 @@ struct dvb_frontend *mxl5007t_attach(struct dvb_frontend *fe,
>>  		/* existing tuner instance */
>>  		break;
>>  	}
>> -
>> -	if (fe->ops.i2c_gate_ctrl)
>> -		fe->ops.i2c_gate_ctrl(fe, 1);
>> -
>> -	ret = mxl5007t_soft_reset(state);
>> -
>> -	if (fe->ops.i2c_gate_ctrl)
>> -		fe->ops.i2c_gate_ctrl(fe, 0);
>> -
>> -	if (mxl_fail(ret))
>> -		goto fail;
>> -
>> -	if (fe->ops.i2c_gate_ctrl)
>> -		fe->ops.i2c_gate_ctrl(fe, 1);
>> -
>> -	ret = mxl5007t_write_reg(state, 0x04,
>> -		state->config->loop_thru_enable);
>> -
>> -	if (fe->ops.i2c_gate_ctrl)
>> -		fe->ops.i2c_gate_ctrl(fe, 0);
>> -
>> -	if (mxl_fail(ret))
>> -		goto fail;
>> -
>>  	fe->tuner_priv = state;
>> -
>>  	mutex_unlock(&mxl5007t_list_mutex);
>>  
>>  	memcpy(&fe->ops.tuner_ops, &mxl5007t_tuner_ops,
>>
> 
> 
> Guys, any chance to push this any time soon?
> 
> 

Will someone finally push this?

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[poma]

On 31.07.2015 17:55, Olli Salonen wrote:
> Hi Poma,
> 
> Might I suggest reading points 9, 10 and 11 from the following document:
> https://www.kernel.org/doc/Documentation/SubmittingPatches
> 
> Especially point 11 needs to be taken care of in order for your patch to be
> merged.
> 
> Cheers,
> -olli
> 

This is why top posting is so bad:

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
  

Furthermore, to fix this issue - AF9015 DVB-T USB2.0 stick brokenness - is the responsibility of developers.
I am here only proven tester.

I hope we understand each other, and this problem will be resolved in good faith.

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[Antti Palosaari]

On 08/11/2015 12:46 AM, poma wrote:
> Furthermore, to fix this issue - AF9015 DVB-T USB2.0 stick brokenness - is the responsibility of developers.
> I am here only proven tester.
>
> I hope we understand each other, and this problem will be resolved in good faith.

Your patches are implemented wrong.

When I added mxl5007t support to that driver it was DigitalNow TinyTwin 
v2 I had. The rest mxl5007t dual devices using reference design IDs went 
to same due to reason driver detects used tuner. My device is still 
working fine, which means your device has different wiring. As 2nd tuner 
attach fails it means there is communication loss to tuner. Which means 
tuner is most likely hold in a reset attach time or there is some I2C 
gating which prevents communication.

Patches you sent will introduce another issue. For dual tuner 
configuration there could be antenna wired from tuner chip to another. 
After that patch you will lose antenna signal from 2nd tuner on cases 
where tuner antenna wire is loop through master tuner to slave.

So fix it correctly. Find out reason there is communication loss to 2nd 
tuner on attach time. I cannot do much as I simply don't have such 
hardware. And I really do not care to take any responsibility when that 
kind of issues happens - it is not my job to bough every single device 
from the market in able to test and fix every hardware combination.


Antti

-- 
http://palosaari.fi/

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[poma]

On 11.08.2015 00:40, Antti Palosaari wrote:
> On 08/11/2015 12:46 AM, poma wrote:
>> Furthermore, to fix this issue - AF9015 DVB-T USB2.0 stick brokenness - is the responsibility of developers.
>> I am here only proven tester.
>>
>> I hope we understand each other, and this problem will be resolved in good faith.
> 
> Your patches are implemented wrong.


Talk about deflecting.

The patch that you called "my" is nothing but two reverted commitas:

- 02f9cf96df57575acea2e6eb4041e9f3ecd32548
  "[media] [PATH,2/2] mxl5007 move loop_thru to attach"
- fe4860af002a4516dd878f7297b61e186c475b35
  "[media] [PATH,1/2] mxl5007 move reset to attach"

Besides the author, Jose Alberto Reguero does not oppose to revert them, due to the trouble they cause.

> 
> When I added mxl5007t support to that driver it was DigitalNow TinyTwin 
> v2 I had. The rest mxl5007t dual devices using reference design IDs went 
> to same due to reason driver detects used tuner. My device is still 
> working fine, which means your device has different wiring. As 2nd tuner 
> attach fails it means there is communication loss to tuner. Which means 
> tuner is most likely hold in a reset attach time or there is some I2C 
> gating which prevents communication.
> 
> Patches you sent will introduce another issue. For dual tuner 
> configuration there could be antenna wired from tuner chip to another. 
> After that patch you will lose antenna signal from 2nd tuner on cases 
> where tuner antenna wire is loop through master tuner to slave.
> 
> So fix it correctly. Find out reason there is communication loss to 2nd 
> tuner on attach time. I cannot do much as I simply don't have such 
> hardware. And I really do not care to take any responsibility when that 
> kind of issues happens - it is not my job to bough every single device 
> from the market in able to test and fix every hardware combination.
> 
> 
> Antti
> 

You very well know that I am not a developer, 
Therefore I believe this above what you write is addressed to yourself,
after all *you* are a developer, a capable one, right.

Are you?

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[poma]

On 26.05.2015 14:24, poma wrote:
> 
> If it is not taken into account the already known problem of unreliable operation of the first tuner of the two,
> the device works reliably within kernel 4.0.4 with mxl5007t.ko reverted to
> http://git.linuxtv.org/cgit.cgi/media_tree.git/commit/drivers/media/tuners/mxl5007t.c?id=ccae7af
> that is in the same state as is in the longterm kernel - 3.18.14,
> which is in correspondence with the aforementioned results.
> 
> 

http://git.linuxtv.org/cgit.cgi/media_tree.git/log/drivers/media/tuners/mxl5007t.c


5. 2014-11-11   [media] [PATH,2/2] mxl5007 move loop_thru to attach
   02f9cf9      Jose Alberto Reguero
   
4. 2014-11-11   [media] [PATH,1/2] mxl5007 move reset to attach
   fe4860a      Jose Alberto Reguero

3. 2013-02-08   Revert "[media] [PATH,1/2] mxl5007 move reset to attach"
   db5c05b      Mauro Carvalho Chehab

2. 2013-02-08   [media] [PATH,1/2] mxl5007 move reset to attach
   0a32377      Jose Alberto Reguero

1. 2012-08-14   [media] common: move media/common/tuners to media/tuners
   ccae7af      Mauro Carvalho Chehab


This is the conclusion after extensive testing,
commitas 5. 4. and 2. produce:

mxl5007t_soft_reset: 521: failed!
mxl5007t_attach: error -121 on line 907

causing the device completely unusable - AF9015 DVB-T USB2.0 stick


Do you need a patch to revert to commita 3. or 1. - again for the third time,
or you have a better solution?

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[poma]

If it is not taken into account the already known problem of unreliable operation of the first tuner of the two,
the device works reliably within kernel 4.0.4 with mxl5007t.ko reverted to
http://git.linuxtv.org/cgit.cgi/media_tree.git/commit/drivers/media/tuners/mxl5007t.c?id=ccae7af
that is in the same state as is in the longterm kernel - 3.18.14,
which is in correspondence with the aforementioned results.

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[poma]

On 18.05.2015 21:15, poma wrote:
> On 17.05.2015 01:07, Antti Palosaari wrote:
...
>> try that
>> http://git.linuxtv.org/cgit.cgi/anttip/media_tree.git/commit/?h=af9015_mxl5007t_1
>>
>> Antti
>>
> 
> Thanks.
> For now, I have noticed, unlike before the EIT program data are updated in full and promptly.
> 
> I'll be back in a week.
> 
> 

No need to continue to test this patch because this is repeated:

BOOT from S5/Soft Off:

[    1.007886] usb 1-2: Product: DVB-T 2
[    1.032430] Afatech DVB-T 2: Fixing fullspeed to highspeed interval: 10 -> 7
[    1.035255] input: Afatech DVB-T 2 as /devices/pci0000:00/0000:00:02.1/usb1/1-2/1-2:1.1/0003:15A4:9016.0001/input/input5
[    1.089852] hid-generic 0003:15A4:9016.0001: input,hidraw0: USB HID v1.01 Keyboard [Afatech DVB-T 2] on usb-0000:00:02.1-2/input1
[   13.353105] usb 1-2: dvb_usb_v2: found a 'Afatech AF9015 reference design' in warm state
[   13.557671] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
[   13.558315] DVB: registering new adapter (Afatech AF9015 reference design)
[   13.726582] usb 1-2: DVB: registering adapter 0 frontend 0 (Afatech AF9013)...
[   13.814649] mxl5007t 13-00c0: creating new instance
[   13.817588] mxl5007t_get_chip_id: MxL5007T.v4 detected @ 13-00c0
[   13.820460] mxl5007t_get_chip_id: MxL5007T.v4 detected @ 13-00c0
[   13.824594] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
[   13.825234] DVB: registering new adapter (Afatech AF9015 reference design)
[   14.040582] usb 1-2: DVB: registering adapter 1 frontend 0 (Afatech AF9013)...
[   14.040921] mxl5007t 13-00c0: attaching existing instance
[   14.043201] usb 1-2: dvb_usb_af9015: command failed=1
[   14.043429] mxl5007t_read_reg: 505: failed!
[   14.043656] mxl5007t_get_chip_id: error -121 on line 824
[   14.043880] mxl5007t_get_chip_id: unable to identify device @ 13-00c0
[   14.044698] usb 1-2: dvb_usb_af9015: command failed=1
[   14.044911] mxl5007t_soft_reset: 527: failed!
[   14.046841] mxl5007t_attach: error -121 on line 914
[   14.051434] usbcore: registered new interface driver dvb_usb_af9015

$ ls /dev/dvb/
ls: cannot access /dev/dvb/: No such file or directory

# modprobe -rv dvb_usb_af9015 mxl5007t
rmmod dvb_usb_af9015
rmmod dvb_usb_v2
rmmod rc_core
rmmod dvb_core
rmmod mxl5007t

# modprobe -v dvb_usb_af9015
insmod /lib/modules/4.0.4-502.fc21.x86_64/kernel/drivers/media/rc/rc-core.ko.xz 
insmod /lib/modules/4.0.4-502.fc21.x86_64/kernel/drivers/media/dvb-core/dvb-core.ko.xz 
insmod /lib/modules/4.0.4-502.fc21.x86_64/kernel/drivers/media/usb/dvb-usb-v2/dvb_usb_v2.ko.xz 
insmod /lib/modules/4.0.4-502.fc21.x86_64/kernel/drivers/media/usb/dvb-usb-v2/dvb-usb-af9015.ko.xz 

[ 2101.659729] usbcore: deregistering interface driver dvb_usb_af9015
[ 2117.056136] usb 1-2: dvb_usb_v2: found a 'Afatech AF9015 reference design' in warm state
[ 2117.265520] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
[ 2117.265607] DVB: registering new adapter (Afatech AF9015 reference design)
[ 2117.268348] i2c i2c-13: af9013: firmware version 5.1.0.0
[ 2117.274958] usb 1-2: DVB: registering adapter 0 frontend 0 (Afatech AF9013)...
[ 2117.280031] mxl5007t 13-00c0: creating new instance
[ 2117.281700] mxl5007t_get_chip_id: MxL5007T.v4 detected @ 13-00c0
[ 2117.284211] mxl5007t_get_chip_id: MxL5007T.v4 detected @ 13-00c0
[ 2117.287208] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
[ 2117.287245] DVB: registering new adapter (Afatech AF9015 reference design)
[ 2117.493132] i2c i2c-13: af9013: found a 'Afatech AF9013' in warm state
[ 2117.496887] i2c i2c-13: af9013: firmware version 5.1.0.0
[ 2117.504373] usb 1-2: DVB: registering adapter 1 frontend 0 (Afatech AF9013)...
[ 2117.504664] mxl5007t 13-00c0: attaching existing instance
[ 2117.506848] usb 1-2: dvb_usb_af9015: command failed=1
[ 2117.506858] mxl5007t_read_reg: 505: failed!
[ 2117.506864] mxl5007t_get_chip_id: error -121 on line 824
[ 2117.506868] mxl5007t_get_chip_id: unable to identify device @ 13-00c0
[ 2117.507344] usb 1-2: dvb_usb_af9015: command failed=1
[ 2117.507350] mxl5007t_soft_reset: 527: failed!
[ 2117.509007] mxl5007t_attach: error -121 on line 914
[ 2117.511150] usbcore: registered new interface driver dvb_usb_af9015

$ ls /dev/dvb/
ls: cannot access /dev/dvb/: No such file or directory

REBOOT:

[    1.049265] usb 1-2: Product: DVB-T 2
[    1.067810] Afatech DVB-T 2: Fixing fullspeed to highspeed interval: 10 -> 7
[    1.068145] input: Afatech DVB-T 2 as /devices/pci0000:00/0000:00:02.1/usb1/1-2/1-2:1.1/0003:15A4:9016.0001/input/input5
[    1.118889] hid-generic 0003:15A4:9016.0001: input,hidraw0: USB HID v1.01 Keyboard [Afatech DVB-T 2] on usb-0000:00:02.1-2/input1
[   12.023590] usb 1-2: dvb_usb_v2: found a 'Afatech AF9015 reference design' in warm state
[   12.225419] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
[   12.225524] DVB: registering new adapter (Afatech AF9015 reference design)
[   12.238958] usb 1-2: DVB: registering adapter 0 frontend 0 (Afatech AF9013)...
[   12.261783] mxl5007t 13-00c0: creating new instance
[   12.265081] mxl5007t_get_chip_id: MxL5007T.v4 detected @ 13-00c0
[   12.267236] mxl5007t_get_chip_id: MxL5007T.v4 detected @ 13-00c0
[   12.270335] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
[   12.270376] DVB: registering new adapter (Afatech AF9015 reference design)
[   12.484761] usb 1-2: DVB: registering adapter 1 frontend 0 (Afatech AF9013)...
[   12.485016] mxl5007t 13-00c0: attaching existing instance
[   12.487225] usb 1-2: dvb_usb_af9015: command failed=1
[   12.487251] mxl5007t_read_reg: 505: failed!
[   12.487266] mxl5007t_get_chip_id: error -121 on line 824
[   12.487281] mxl5007t_get_chip_id: unable to identify device @ 13-00c0
[   12.487716] usb 1-2: dvb_usb_af9015: command failed=1
[   12.487753] mxl5007t_soft_reset: 527: failed!
[   12.489363] mxl5007t_attach: error -121 on line 914
[   12.490589] usbcore: registered new interface driver dvb_usb_af9015

$ ls /dev/dvb
ls: cannot access /dev/dvb: No such file or directory

$ uname -r
4.0.4-502.fc21.x86_64

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[poma]

On 17.05.2015 01:07, Antti Palosaari wrote:
> 
> 
> On 05/13/2015 06:56 AM, poma wrote:
>> On 12.05.2015 16:11, Antti Palosaari wrote:
>>> On 05/11/2015 09:43 PM, poma wrote:
>>>> On 05/11/2015 12:25 AM, poma wrote:
>>>>> On 10.05.2015 12:38, poma wrote:
>>>>>> On 08.05.2015 12:59, poma wrote:
>> Is a beer keg enough as bribe? :)
>> Just do not say that you drink juice.
>>
>> After the reverting of all changes
>> http://git.linuxtv.org/cgit.cgi/media_tree.git/log/drivers/media/tuners/mxl5007t.c
>>
>> device now survives both, 'lsdvb' and rc kernels.
>>
>> Besides, despite all this, this device is already not working at its full potential.
>> One of the tuners can withstand a few hours and then hangs.
>> After that, in the application e.g. vlc is needed to select the second tuner and so continue to use the device.
>> So this is actually a "single-seater" as Formula 1.
>> Vroom vroom!
> 
> try that
> http://git.linuxtv.org/cgit.cgi/anttip/media_tree.git/commit/?h=af9015_mxl5007t_1
> 
> Antti
> 

Thanks.
For now, I have noticed, unlike before the EIT program data are updated in full and promptly.

I'll be back in a week.

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[Antti Palosaari]

On 05/13/2015 06:56 AM, poma wrote:
> On 12.05.2015 16:11, Antti Palosaari wrote:
>> On 05/11/2015 09:43 PM, poma wrote:
>>> On 05/11/2015 12:25 AM, poma wrote:
>>>> On 10.05.2015 12:38, poma wrote:
>>>>> On 08.05.2015 12:59, poma wrote:
> Is a beer keg enough as bribe? :)
> Just do not say that you drink juice.
>
> After the reverting of all changes
> http://git.linuxtv.org/cgit.cgi/media_tree.git/log/drivers/media/tuners/mxl5007t.c
>
> device now survives both, 'lsdvb' and rc kernels.
>
> Besides, despite all this, this device is already not working at its full potential.
> One of the tuners can withstand a few hours and then hangs.
> After that, in the application e.g. vlc is needed to select the second tuner and so continue to use the device.
> So this is actually a "single-seater" as Formula 1.
> Vroom vroom!

try that
http://git.linuxtv.org/cgit.cgi/anttip/media_tree.git/commit/?h=af9015_mxl5007t_1

Antti

-- 
http://palosaari.fi/

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[poma]

On 12.05.2015 16:11, Antti Palosaari wrote:
> On 05/11/2015 09:43 PM, poma wrote:
>> On 05/11/2015 12:25 AM, poma wrote:
>>> On 10.05.2015 12:38, poma wrote:
>>>> On 08.05.2015 12:59, poma wrote:
>>>>> On 08.05.2015 12:20, poma wrote:
>>>>>>
>>>>>> [    0.000000] Linux version 4.0.2-200.fc21.x86_64 ...
>>>>>>
>>>>>> [    0.870875] usb 1-2: new high-speed USB device number 2 using ehci-pci
>>>>>> [    0.990286] usb 1-2: New USB device found, idVendor=15a4, idProduct=9016
>>>>>> [    0.992575] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
>>>>>> [    0.994859] usb 1-2: Product: DVB-T 2
>>>>>>
>>>>>> [    1.001398] usb 1-2: Manufacturer: Afatech
>>>>>> [    1.003555] usb 1-2: SerialNumber: 010101010600001
>>>>>> [    1.009194] Afatech DVB-T 2: Fixing fullspeed to highspeed interval: 10 -> 7
>>>>>> [    1.011694] input: Afatech DVB-T 2 as /devices/pci0000:00/0000:00:02.1/usb1/1-2/1-2:1.1/0003:15A4:9016.0001/input/input5
>>>>>> [    1.066814] hid-generic 0003:15A4:9016.0001: input,hidraw0: USB HID v1.01 Keyboard [Afatech DVB-T 2] on usb-0000:00:02.1-2/input1
>>>>>>
>>>>>> [   11.997119] usb 1-2: dvb_usb_v2: found a 'Afatech AF9015 reference design' in warm state
>>>>>> [   12.206778] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
>>>>>> [   12.207412] DVB: registering new adapter (Afatech AF9015 reference design)
>>>>>>
>>>>>> [   12.286137] i2c i2c-13: af9013: firmware version 5.1.0.0
>>>>>> [   12.289121] usb 1-2: DVB: registering adapter 0 frontend 0 (Afatech AF9013)...
>>>>>> [   12.343650] mxl5007t 13-00c0: creating new instance
>>>>>> [   12.346003] mxl5007t_get_chip_id: unknown rev (3f)
>>>>>> [   12.346156] mxl5007t_get_chip_id: MxL5007T detected @ 13-00c0
>>>>>> [   12.350371] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
>>>>>> [   12.350649] DVB: registering new adapter (Afatech AF9015 reference design)
>>>>>> [   12.553632] i2c i2c-13: af9013: found a 'Afatech AF9013' in warm state
>>>>>> [   12.557256] i2c i2c-13: af9013: firmware version 5.1.0.0
>>>>>> [   12.563779] usb 1-2: DVB: registering adapter 1 frontend 0 (Afatech AF9013)...
>>>>>> [   12.564554] mxl5007t 13-00c0: attaching existing instance
>>>>>> [   12.567004] usb 1-2: dvb_usb_af9015: command failed=1
>>>>>> [   12.567555] mxl5007t_soft_reset: 521: failed!
>>>>>> [   12.569745] mxl5007t_attach: error -121 on line 907
>>>>>> [   12.571231] usbcore: registered new interface driver dvb_usb_af9015
>>>>>>
>>>>>>
>>>>>> $ lsdvb
>>>>>>
>>>>>> 		lsdvb: Simple utility to list PCI/PCIe DVB devices
>>>>>> 		Version: 0.0.4
>>>>>> 		Copyright (C) Manu Abraham
>>>>>> $
>>>>>>
>>>>>
>>>>>
>>>>> Afatech AF9015 reference design:
>>>>>
>>>>> 3.18.12-200.fc21.x86_64        - OK
>>>>>
>>>>> 3.19.7-200.fc21.x86_64         - KO
>>>>> 4.0.2-200.fc21.x86_64          - KO
>>>>> 4.1.0-0.rc2.git3.1.fc23.x86_64 - KO
>>>>>
>>>>>
>>>>> If you have a patch to test, shout loudly.
>>>>>
>>>>>
>>>>
>>>> Looks like the same bug:
>>>> AVerMedia HD Volar (A867) - Afatech AF9033
>>>> http://forum.sifteam.eu/sifbox-by-sif-team/125122-kernel-3-19-a867-xgaz-dove.html
>>>>
>>>> http://git.linuxtv.org/cgit.cgi/media_build.git - no positive effect.
>>>>
>>>>
>>>> Ho ho ho
>>>>
>>>
>>> It seems the 'lsdvb' is what made the device unusable after re/boot.
>>> Of course this applies to kernel >= 3.19.x
>>> The device has to be unplugged and after a minute or two plugged back to be usable again.
>>>
>>>
>>> $ lsdvb
>>>
>>> 		lsdvb: Simple utility to list PCI/PCIe DVB devices
>>> 		Version: 0.0.4
>>> 		Copyright (C) Manu Abraham
>>>
>>> usb (5:0 -868620712:32665) on PCI Domain:-874755276 Bus:32665 Device:2098 Function:0
>>> 	DEVICE:0 ADAPTER:0 FRONTEND:0 (Afatech AF9013)
>>> 		 FE_OFDM Fmin=174MHz Fmax=862MHz
>>> 	DEVICE:0 ADAPTER:1 FRONTEND:0 (Afatech AF9013)
>>> 		 FE_OFDM Fmin=174MHz Fmax=862MHz
>>>
>>> $ dmesg
>>> [   80.332837] usb 2-2: dvb_usb_af9015: command failed=1
>>> [   80.332857] i2c i2c-13: af9013: i2c wr failed=-5 reg=d607 len=1
>>> [   80.337837] usb 2-2: dvb_usb_af9015: command failed=1
>>> [   80.337848] mxl5007t_write_reg: 472: failed!
>>> [   80.337853] mxl5007t_sleep: error -121 on line 709
>>> [   80.338324] usb 2-2: dvb_usb_af9015: command failed=1
>>> [   80.338328] mxl5007t_write_reg: 472: failed!
>>> [   80.338332] mxl5007t_sleep: error -121 on line 711
>>>
>>>
>>>
>>
>>
>> Furthermore, it is sufficient to re/boot to the latest kernels,
>> and the device is K.O. again,
>>
>> e.g.
>> # modinfo -n mxl5007t
>> /lib/modules/4.1.0-0.rc3.git0.1.fc23.x86_64+debug/kernel/drivers/media/tuners/mxl5007t.ko.xz
>>
>> ...
>> [   13.874536] i2c i2c-13: af9013: found a 'Afatech AF9013' in warm state
>> [   13.878031] i2c i2c-13: af9013: firmware version 5.1.0.0
>> [   13.884691] usb 1-2: DVB: registering adapter 1 frontend 0 (Afatech AF9013)...
>> [   13.884942] mxl5007t 13-00c0: attaching existing instance
>> [   13.887166] usb 1-2: dvb_usb_af9015: command failed=1
>> [   13.887179] mxl5007t_soft_reset: 521: failed!
>> [   13.888780] mxl5007t_attach: error -121 on line 907
>> [   13.897808] usbcore: registered new interface driver dvb_usb_af9015
>> ...
>>
>> # ls -al /dev/dvb
>> ls: cannot access /dev/dvb: No such file or directory
>>
>>
>> Keep in mind, this time it has nothing to do with 'lsdvb',
>>
>> # lsdvb
>> -bash: /bin/lsdvb: Permission denied
> 
> It is that commit which causes the problem:
> commit fe4860af002a4516dd878f7297b61e186c475b35
> [media] [PATH,1/2] mxl5007 move reset to attach
> 
> ... but I am pretty sure actual root of cause something else. Likely 
> your second tuner chip is on reset/powered off and due to that it does 
> not answer. I have almost similar device which works (DigitalNow TinyTwin).
> 
> Maybe I should try to test which are that tuner GPIO reset lines... but 
> I am a bit lazy :/
> 
> regards
> Antti
> 

Is a beer keg enough as bribe? :)
Just do not say that you drink juice.

After the reverting of all changes
http://git.linuxtv.org/cgit.cgi/media_tree.git/log/drivers/media/tuners/mxl5007t.c

device now survives both, 'lsdvb' and rc kernels.

Besides, despite all this, this device is already not working at its full potential.
One of the tuners can withstand a few hours and then hangs.
After that, in the application e.g. vlc is needed to select the second tuner and so continue to use the device.
So this is actually a "single-seater" as Formula 1.
Vroom vroom!


 mxl5007t.c |   28 +++++-----------------------
 1 file changed, 5 insertions(+), 23 deletions(-)

--- a/drivers/media/tuners/mxl5007t.c	2015-05-13 04:50:18.362256795 +0200
+++ b/drivers/media/tuners/mxl5007t.c	2015-05-13 05:05:15.070996638 +0200
@@ -374,6 +374,7 @@
 	mxl5007t_set_if_freq_bits(state, cfg->if_freq_hz, cfg->invert_if);
 	mxl5007t_set_xtal_freq_bits(state, cfg->xtal_freq_hz);
 
+	set_reg_bits(state->tab_init, 0x04, 0x01, cfg->loop_thru_enable);
 	set_reg_bits(state->tab_init, 0x03, 0x08, cfg->clk_out_enable << 3);
 	set_reg_bits(state->tab_init, 0x03, 0x07, cfg->clk_out_amp);
 
@@ -530,6 +531,10 @@
 	struct reg_pair_t *init_regs;
 	int ret;
 
+	ret = mxl5007t_soft_reset(state);
+	if (mxl_fail(ret))
+		goto fail;
+
 	/* calculate initialization reg array */
 	init_regs = mxl5007t_calc_init_regs(state, mode);
 
@@ -896,29 +901,6 @@
 		break;
 	}
 
-	if (fe->ops.i2c_gate_ctrl)
-		fe->ops.i2c_gate_ctrl(fe, 1);
-
-	ret = mxl5007t_soft_reset(state);
-
-	if (fe->ops.i2c_gate_ctrl)
-		fe->ops.i2c_gate_ctrl(fe, 0);
-
-	if (mxl_fail(ret))
-		goto fail;
-
-	if (fe->ops.i2c_gate_ctrl)
-		fe->ops.i2c_gate_ctrl(fe, 1);
-
-	ret = mxl5007t_write_reg(state, 0x04,
-		state->config->loop_thru_enable);
-
-	if (fe->ops.i2c_gate_ctrl)
-		fe->ops.i2c_gate_ctrl(fe, 0);
-
-	if (mxl_fail(ret))
-		goto fail;
-
 	fe->tuner_priv = state;
 
 	mutex_unlock(&mxl5007t_list_mutex);


$ modinfo mxl5007t
filename:       /lib/modules/4.1.0-0.rc3.git1.1.fc23.x86_64/updates/mxl5007t.ko
version:        0.2
license:        GPL
author:         Michael Krufky <mkrufky@linuxtv.org>
description:    MaxLinear MxL5007T Silicon IC tuner driver
srcversion:     72D27DB6EF6CEC612700FB9
depends:        
intree:         Y
vermagic:       4.1.0-0.rc3.git1.1.fc23.x86_64 SMP mod_unload 
parm:           debug:set debug level (int)


$ dmesg
[    1.114202] usb 1-2: new high-speed USB device number 2 using ehci-pci
[    1.235071] usb 1-2: New USB device found, idVendor=15a4, idProduct=9016
[    1.237463] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
[    1.239831] usb 1-2: Product: DVB-T 2
[    1.242163] usb 1-2: Manufacturer: Afatech
[    1.244422] usb 1-2: SerialNumber: 010101010600001
[    1.251607] Afatech DVB-T 2: Fixing fullspeed to highspeed interval: 10 -> 7
[    1.254602] input: Afatech DVB-T 2 as /devices/pci0000:00/0000:00:02.1/usb1/1-2/1-2:1.1/0003:15A4:9016.0001/input/input5
[    1.311052] hid-generic 0003:15A4:9016.0001: input,hidraw0: USB HID v1.01 Keyboard [Afatech DVB-T 2] on usb-0000:00:02.1-2/input1
[   14.326420] usb 1-2: dvb_usb_v2: found a 'Afatech AF9015 reference design' in warm state
[   14.563618] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
[   14.563854] DVB: registering new adapter (Afatech AF9015 reference design)
[   14.622564] usb 1-2: DVB: registering adapter 0 frontend 0 (Afatech AF9013)...
[   14.659556] mxl5007t: module verification failed: signature and/or required key missing - tainting kernel
[   14.660903] mxl5007t 13-00c0: creating new instance
[   14.663714] mxl5007t_get_chip_id: unknown rev (3f)
[   14.663727] mxl5007t_get_chip_id: MxL5007T detected @ 13-00c0
[   14.664450] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
[   14.664603] DVB: registering new adapter (Afatech AF9015 reference design)
[   14.882505] usb 1-2: DVB: registering adapter 1 frontend 0 (Afatech AF9013)...
[   14.883176] mxl5007t 13-00c0: attaching existing instance
[   14.905227] input: Afatech AF9015 reference design as /devices/pci0000:00/0000:00:02.1/usb1/1-2/rc/rc0/input14
[   14.907885] rc0: Afatech AF9015 reference design as /devices/pci0000:00/0000:00:02.1/usb1/1-2/rc/rc0
[   14.907917] usb 1-2: dvb_usb_v2: schedule remote query interval to 500 msecs
[   14.908170] usb 1-2: dvb_usb_v2: 'Afatech AF9015 reference design' successfully initialized and connected
[   14.908490] usbcore: registered new interface driver dvb_usb_af9015


$ lsdvb 

		lsdvb: Simple utility to list PCI/PCIe DVB devices
		Version: 0.0.4
		Copyright (C) Manu Abraham

usb (5:0 -956221864:32726) on PCI Domain:-962356428 Bus:32726 Device:2098 Function:0
	DEVICE:0 ADAPTER:0 FRONTEND:0 (Afatech AF9013) 
		 FE_OFDM Fmin=174MHz Fmax=862MHz
	DEVICE:0 ADAPTER:1 FRONTEND:0 (Afatech AF9013) 
		 FE_OFDM Fmin=174MHz Fmax=862MHz


$ dmesg
[  306.258474] usb 1-2: dvb_usb_af9015: command failed=1
[  306.258500] i2c i2c-13: af9013: i2c wr failed=-5 reg=d607 len=1
[  306.264805] usb 1-2: dvb_usb_af9015: command failed=1
[  306.264819] mxl5007t_write_reg: 473: failed!
[  306.264826] mxl5007t_sleep: error -121 on line 714
[  306.265298] usb 1-2: dvb_usb_af9015: command failed=1
[  306.265306] mxl5007t_write_reg: 473: failed!
[  306.265309] mxl5007t_sleep: error -121 on line 716

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[Antti Palosaari]

On 05/11/2015 09:43 PM, poma wrote:
> On 05/11/2015 12:25 AM, poma wrote:
>> On 10.05.2015 12:38, poma wrote:
>>> On 08.05.2015 12:59, poma wrote:
>>>> On 08.05.2015 12:20, poma wrote:
>>>>>
>>>>> [    0.000000] Linux version 4.0.2-200.fc21.x86_64 ...
>>>>>
>>>>> [    0.870875] usb 1-2: new high-speed USB device number 2 using ehci-pci
>>>>> [    0.990286] usb 1-2: New USB device found, idVendor=15a4, idProduct=9016
>>>>> [    0.992575] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
>>>>> [    0.994859] usb 1-2: Product: DVB-T 2
>>>>>
>>>>> [    1.001398] usb 1-2: Manufacturer: Afatech
>>>>> [    1.003555] usb 1-2: SerialNumber: 010101010600001
>>>>> [    1.009194] Afatech DVB-T 2: Fixing fullspeed to highspeed interval: 10 -> 7
>>>>> [    1.011694] input: Afatech DVB-T 2 as /devices/pci0000:00/0000:00:02.1/usb1/1-2/1-2:1.1/0003:15A4:9016.0001/input/input5
>>>>> [    1.066814] hid-generic 0003:15A4:9016.0001: input,hidraw0: USB HID v1.01 Keyboard [Afatech DVB-T 2] on usb-0000:00:02.1-2/input1
>>>>>
>>>>> [   11.997119] usb 1-2: dvb_usb_v2: found a 'Afatech AF9015 reference design' in warm state
>>>>> [   12.206778] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
>>>>> [   12.207412] DVB: registering new adapter (Afatech AF9015 reference design)
>>>>>
>>>>> [   12.286137] i2c i2c-13: af9013: firmware version 5.1.0.0
>>>>> [   12.289121] usb 1-2: DVB: registering adapter 0 frontend 0 (Afatech AF9013)...
>>>>> [   12.343650] mxl5007t 13-00c0: creating new instance
>>>>> [   12.346003] mxl5007t_get_chip_id: unknown rev (3f)
>>>>> [   12.346156] mxl5007t_get_chip_id: MxL5007T detected @ 13-00c0
>>>>> [   12.350371] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
>>>>> [   12.350649] DVB: registering new adapter (Afatech AF9015 reference design)
>>>>> [   12.553632] i2c i2c-13: af9013: found a 'Afatech AF9013' in warm state
>>>>> [   12.557256] i2c i2c-13: af9013: firmware version 5.1.0.0
>>>>> [   12.563779] usb 1-2: DVB: registering adapter 1 frontend 0 (Afatech AF9013)...
>>>>> [   12.564554] mxl5007t 13-00c0: attaching existing instance
>>>>> [   12.567004] usb 1-2: dvb_usb_af9015: command failed=1
>>>>> [   12.567555] mxl5007t_soft_reset: 521: failed!
>>>>> [   12.569745] mxl5007t_attach: error -121 on line 907
>>>>> [   12.571231] usbcore: registered new interface driver dvb_usb_af9015
>>>>>
>>>>>
>>>>> $ lsdvb
>>>>>
>>>>> 		lsdvb: Simple utility to list PCI/PCIe DVB devices
>>>>> 		Version: 0.0.4
>>>>> 		Copyright (C) Manu Abraham
>>>>> $
>>>>>
>>>>
>>>>
>>>> Afatech AF9015 reference design:
>>>>
>>>> 3.18.12-200.fc21.x86_64        - OK
>>>>
>>>> 3.19.7-200.fc21.x86_64         - KO
>>>> 4.0.2-200.fc21.x86_64          - KO
>>>> 4.1.0-0.rc2.git3.1.fc23.x86_64 - KO
>>>>
>>>>
>>>> If you have a patch to test, shout loudly.
>>>>
>>>>
>>>
>>> Looks like the same bug:
>>> AVerMedia HD Volar (A867) - Afatech AF9033
>>> http://forum.sifteam.eu/sifbox-by-sif-team/125122-kernel-3-19-a867-xgaz-dove.html
>>>
>>> http://git.linuxtv.org/cgit.cgi/media_build.git - no positive effect.
>>>
>>>
>>> Ho ho ho
>>>
>>
>> It seems the 'lsdvb' is what made the device unusable after re/boot.
>> Of course this applies to kernel >= 3.19.x
>> The device has to be unplugged and after a minute or two plugged back to be usable again.
>>
>>
>> $ lsdvb
>>
>> 		lsdvb: Simple utility to list PCI/PCIe DVB devices
>> 		Version: 0.0.4
>> 		Copyright (C) Manu Abraham
>>
>> usb (5:0 -868620712:32665) on PCI Domain:-874755276 Bus:32665 Device:2098 Function:0
>> 	DEVICE:0 ADAPTER:0 FRONTEND:0 (Afatech AF9013)
>> 		 FE_OFDM Fmin=174MHz Fmax=862MHz
>> 	DEVICE:0 ADAPTER:1 FRONTEND:0 (Afatech AF9013)
>> 		 FE_OFDM Fmin=174MHz Fmax=862MHz
>>
>> $ dmesg
>> [   80.332837] usb 2-2: dvb_usb_af9015: command failed=1
>> [   80.332857] i2c i2c-13: af9013: i2c wr failed=-5 reg=d607 len=1
>> [   80.337837] usb 2-2: dvb_usb_af9015: command failed=1
>> [   80.337848] mxl5007t_write_reg: 472: failed!
>> [   80.337853] mxl5007t_sleep: error -121 on line 709
>> [   80.338324] usb 2-2: dvb_usb_af9015: command failed=1
>> [   80.338328] mxl5007t_write_reg: 472: failed!
>> [   80.338332] mxl5007t_sleep: error -121 on line 711
>>
>>
>>
>
>
> Furthermore, it is sufficient to re/boot to the latest kernels,
> and the device is K.O. again,
>
> e.g.
> # modinfo -n mxl5007t
> /lib/modules/4.1.0-0.rc3.git0.1.fc23.x86_64+debug/kernel/drivers/media/tuners/mxl5007t.ko.xz
>
> ...
> [   13.874536] i2c i2c-13: af9013: found a 'Afatech AF9013' in warm state
> [   13.878031] i2c i2c-13: af9013: firmware version 5.1.0.0
> [   13.884691] usb 1-2: DVB: registering adapter 1 frontend 0 (Afatech AF9013)...
> [   13.884942] mxl5007t 13-00c0: attaching existing instance
> [   13.887166] usb 1-2: dvb_usb_af9015: command failed=1
> [   13.887179] mxl5007t_soft_reset: 521: failed!
> [   13.888780] mxl5007t_attach: error -121 on line 907
> [   13.897808] usbcore: registered new interface driver dvb_usb_af9015
> ...
>
> # ls -al /dev/dvb
> ls: cannot access /dev/dvb: No such file or directory
>
>
> Keep in mind, this time it has nothing to do with 'lsdvb',
>
> # lsdvb
> -bash: /bin/lsdvb: Permission denied

It is that commit which causes the problem:
commit fe4860af002a4516dd878f7297b61e186c475b35
[media] [PATH,1/2] mxl5007 move reset to attach

... but I am pretty sure actual root of cause something else. Likely 
your second tuner chip is on reset/powered off and due to that it does 
not answer. I have almost similar device which works (DigitalNow TinyTwin).

Maybe I should try to test which are that tuner GPIO reset lines... but 
I am a bit lazy :/

regards
Antti

-- 
http://palosaari.fi/

Re: dvb_usb_af9015: command failed=1 _ kernel >= 4.1.x

[poma]

On 05/11/2015 12:25 AM, poma wrote:
> On 10.05.2015 12:38, poma wrote:
>> On 08.05.2015 12:59, poma wrote:
>>> On 08.05.2015 12:20, poma wrote:
>>>>
>>>> [    0.000000] Linux version 4.0.2-200.fc21.x86_64 ...
>>>>
>>>> [    0.870875] usb 1-2: new high-speed USB device number 2 using ehci-pci
>>>> [    0.990286] usb 1-2: New USB device found, idVendor=15a4, idProduct=9016
>>>> [    0.992575] usb 1-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
>>>> [    0.994859] usb 1-2: Product: DVB-T 2
>>>>
>>>> [    1.001398] usb 1-2: Manufacturer: Afatech
>>>> [    1.003555] usb 1-2: SerialNumber: 010101010600001
>>>> [    1.009194] Afatech DVB-T 2: Fixing fullspeed to highspeed interval: 10 -> 7
>>>> [    1.011694] input: Afatech DVB-T 2 as /devices/pci0000:00/0000:00:02.1/usb1/1-2/1-2:1.1/0003:15A4:9016.0001/input/input5
>>>> [    1.066814] hid-generic 0003:15A4:9016.0001: input,hidraw0: USB HID v1.01 Keyboard [Afatech DVB-T 2] on usb-0000:00:02.1-2/input1
>>>>
>>>> [   11.997119] usb 1-2: dvb_usb_v2: found a 'Afatech AF9015 reference design' in warm state
>>>> [   12.206778] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
>>>> [   12.207412] DVB: registering new adapter (Afatech AF9015 reference design)
>>>>
>>>> [   12.286137] i2c i2c-13: af9013: firmware version 5.1.0.0
>>>> [   12.289121] usb 1-2: DVB: registering adapter 0 frontend 0 (Afatech AF9013)...
>>>> [   12.343650] mxl5007t 13-00c0: creating new instance
>>>> [   12.346003] mxl5007t_get_chip_id: unknown rev (3f)
>>>> [   12.346156] mxl5007t_get_chip_id: MxL5007T detected @ 13-00c0
>>>> [   12.350371] usb 1-2: dvb_usb_v2: will pass the complete MPEG2 transport stream to the software demuxer
>>>> [   12.350649] DVB: registering new adapter (Afatech AF9015 reference design)
>>>> [   12.553632] i2c i2c-13: af9013: found a 'Afatech AF9013' in warm state
>>>> [   12.557256] i2c i2c-13: af9013: firmware version 5.1.0.0
>>>> [   12.563779] usb 1-2: DVB: registering adapter 1 frontend 0 (Afatech AF9013)...
>>>> [   12.564554] mxl5007t 13-00c0: attaching existing instance
>>>> [   12.567004] usb 1-2: dvb_usb_af9015: command failed=1
>>>> [   12.567555] mxl5007t_soft_reset: 521: failed!
>>>> [   12.569745] mxl5007t_attach: error -121 on line 907
>>>> [   12.571231] usbcore: registered new interface driver dvb_usb_af9015
>>>>
>>>>
>>>> $ lsdvb
>>>>
>>>> 		lsdvb: Simple utility to list PCI/PCIe DVB devices
>>>> 		Version: 0.0.4
>>>> 		Copyright (C) Manu Abraham
>>>> $ 
>>>>
>>>
>>>
>>> Afatech AF9015 reference design:
>>>
>>> 3.18.12-200.fc21.x86_64        - OK
>>>
>>> 3.19.7-200.fc21.x86_64         - KO
>>> 4.0.2-200.fc21.x86_64          - KO
>>> 4.1.0-0.rc2.git3.1.fc23.x86_64 - KO
>>>
>>>
>>> If you have a patch to test, shout loudly.
>>>
>>>
>>
>> Looks like the same bug:
>> AVerMedia HD Volar (A867) - Afatech AF9033
>> http://forum.sifteam.eu/sifbox-by-sif-team/125122-kernel-3-19-a867-xgaz-dove.html
>>
>> http://git.linuxtv.org/cgit.cgi/media_build.git - no positive effect.
>>
>>
>> Ho ho ho
>>
> 
> It seems the 'lsdvb' is what made the device unusable after re/boot.
> Of course this applies to kernel >= 3.19.x
> The device has to be unplugged and after a minute or two plugged back to be usable again.
> 
> 
> $ lsdvb
> 
> 		lsdvb: Simple utility to list PCI/PCIe DVB devices
> 		Version: 0.0.4
> 		Copyright (C) Manu Abraham
> 
> usb (5:0 -868620712:32665) on PCI Domain:-874755276 Bus:32665 Device:2098 Function:0
> 	DEVICE:0 ADAPTER:0 FRONTEND:0 (Afatech AF9013) 
> 		 FE_OFDM Fmin=174MHz Fmax=862MHz
> 	DEVICE:0 ADAPTER:1 FRONTEND:0 (Afatech AF9013) 
> 		 FE_OFDM Fmin=174MHz Fmax=862MHz
> 
> $ dmesg
> [   80.332837] usb 2-2: dvb_usb_af9015: command failed=1
> [   80.332857] i2c i2c-13: af9013: i2c wr failed=-5 reg=d607 len=1
> [   80.337837] usb 2-2: dvb_usb_af9015: command failed=1
> [   80.337848] mxl5007t_write_reg: 472: failed!
> [   80.337853] mxl5007t_sleep: error -121 on line 709
> [   80.338324] usb 2-2: dvb_usb_af9015: command failed=1
> [   80.338328] mxl5007t_write_reg: 472: failed!
> [   80.338332] mxl5007t_sleep: error -121 on line 711
> 
> 
> 


Furthermore, it is sufficient to re/boot to the latest kernels,
and the device is K.O. again,

e.g.
# modinfo -n mxl5007t
/lib/modules/4.1.0-0.rc3.git0.1.fc23.x86_64+debug/kernel/drivers/media/tuners/mxl5007t.ko.xz

...
[   13.874536] i2c i2c-13: af9013: found a 'Afatech AF9013' in warm state
[   13.878031] i2c i2c-13: af9013: firmware version 5.1.0.0
[   13.884691] usb 1-2: DVB: registering adapter 1 frontend 0 (Afatech AF9013)...
[   13.884942] mxl5007t 13-00c0: attaching existing instance
[   13.887166] usb 1-2: dvb_usb_af9015: command failed=1
[   13.887179] mxl5007t_soft_reset: 521: failed!
[   13.888780] mxl5007t_attach: error -121 on line 907
[   13.897808] usbcore: registered new interface driver dvb_usb_af9015
...

# ls -al /dev/dvb
ls: cannot access /dev/dvb: No such file or directory


Keep in mind, this time it has nothing to do with 'lsdvb',

# lsdvb
-bash: /bin/lsdvb: Permission denied