[PATCH] ngene: Fix CI data transfer regression

[PATCH] ngene: Fix CI data transfer regression

[Oliver Endriss]

Fix CI data transfer regression introduced by previous cleanup.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
---
 drivers/media/dvb/ngene/ngene-core.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/drivers/media/dvb/ngene/ngene-core.c b/drivers/media/dvb/ngene/ngene-core.c
index 175a0f6..9630705 100644
--- a/drivers/media/dvb/ngene/ngene-core.c
+++ b/drivers/media/dvb/ngene/ngene-core.c
@@ -1520,6 +1520,7 @@ static int init_channel(struct ngene_channel *chan)
 	if (dev->ci.en && (io & NGENE_IO_TSOUT)) {
 		dvb_ca_en50221_init(adapter, dev->ci.en, 0, 1);
 		set_transfer(chan, 1);
+		chan->dev->channel[2].DataFormatFlags = DF_SWAP32;
 		set_transfer(&chan->dev->channel[2], 1);
 		dvb_register_device(adapter, &chan->ci_dev,
 				    &ngene_dvbdev_ci, (void *) chan,
-- 
1.6.5.3

-- 
----------------------------------------------------------------
VDR Remote Plugin 0.4.0: http://www.escape-edv.de/endriss/vdr/
4 MByte Mod: http://www.escape-edv.de/endriss/dvb-mem-mod/
Full-TS Mod: http://www.escape-edv.de/endriss/dvb-full-ts-mod/
----------------------------------------------------------------

Re: [PATCH] ngene: Fix CI data transfer regression

[Oliver Endriss]

On Tuesday 29 March 2011 22:35:24 Oliver Endriss wrote:
> Fix CI data transfer regression introduced by previous cleanup.
> 
> Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
> ---
>  drivers/media/dvb/ngene/ngene-core.c |    1 +
>  1 files changed, 1 insertions(+), 0 deletions(-)
> 
> diff --git a/drivers/media/dvb/ngene/ngene-core.c b/drivers/media/dvb/ngene/ngene-core.c
> index 175a0f6..9630705 100644
> --- a/drivers/media/dvb/ngene/ngene-core.c
> +++ b/drivers/media/dvb/ngene/ngene-core.c
> @@ -1520,6 +1520,7 @@ static int init_channel(struct ngene_channel *chan)
>  	if (dev->ci.en && (io & NGENE_IO_TSOUT)) {
>  		dvb_ca_en50221_init(adapter, dev->ci.en, 0, 1);
>  		set_transfer(chan, 1);
> +		chan->dev->channel[2].DataFormatFlags = DF_SWAP32;
>  		set_transfer(&chan->dev->channel[2], 1);
>  		dvb_register_device(adapter, &chan->ci_dev,
>  				    &ngene_dvbdev_ci, (void *) chan,
> -- 
> 1.6.5.3
> 

What happened to this patch? I am sure that it was in patchwork, but
patchwork apparently lost all patches between February 26th and
April 16th.

Please note that this patch must go to 2.6.39!

CU
Oliver

-- 
----------------------------------------------------------------
VDR Remote Plugin 0.4.0: http://www.escape-edv.de/endriss/vdr/
4 MByte Mod: http://www.escape-edv.de/endriss/dvb-mem-mod/
Full-TS Mod: http://www.escape-edv.de/endriss/dvb-full-ts-mod/
----------------------------------------------------------------

Re: [PATCH] ngene: Fix CI data transfer regression

[Mauro Carvalho Chehab]

Em 23-04-2011 13:31, Oliver Endriss escreveu:
> On Tuesday 29 March 2011 22:35:24 Oliver Endriss wrote:
>> Fix CI data transfer regression introduced by previous cleanup.
>>
>> Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
>> ---
>>  drivers/media/dvb/ngene/ngene-core.c |    1 +
>>  1 files changed, 1 insertions(+), 0 deletions(-)
>>
>> diff --git a/drivers/media/dvb/ngene/ngene-core.c b/drivers/media/dvb/ngene/ngene-core.c
>> index 175a0f6..9630705 100644
>> --- a/drivers/media/dvb/ngene/ngene-core.c
>> +++ b/drivers/media/dvb/ngene/ngene-core.c
>> @@ -1520,6 +1520,7 @@ static int init_channel(struct ngene_channel *chan)
>>  	if (dev->ci.en && (io & NGENE_IO_TSOUT)) {
>>  		dvb_ca_en50221_init(adapter, dev->ci.en, 0, 1);
>>  		set_transfer(chan, 1);
>> +		chan->dev->channel[2].DataFormatFlags = DF_SWAP32;
>>  		set_transfer(&chan->dev->channel[2], 1);
>>  		dvb_register_device(adapter, &chan->ci_dev,
>>  				    &ngene_dvbdev_ci, (void *) chan,
>> -- 
>> 1.6.5.3
>>
> 
> What happened to this patch? I am sure that it was in patchwork, but
> patchwork apparently lost all patches between February 26th and
> April 16th.

Yes, patchwork seemed to have a problem. They have 2 patchwork instances there,
on a cluster environment. In the past, we have the same problem when the backup
instance started with some patches out of sync on their mySQL datababse. After
I pointed the issue, they fixed it on the next day.

I sent one email last week about it to the kernel.org maintainer asking for his help, 
but didn't get an answer yet. Maybe he got some PTO days due to Easter. I'm hoping
that we'll be able to get it recovered next week.
> 
> Please note that this patch must go to 2.6.39!
> 
> CU
> Oliver
>