Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: ivtv: devel

[PULL] http://linuxtv.org/hg/~awalls/v4l-dvb-misc

 

 

ivtv devel RSS feed   Index | Next | Previous | View Threaded


awalls at radix

Nov 21, 2009, 2:26 PM

Post #1 of 5 (2102 views)
Permalink
[PULL] http://linuxtv.org/hg/~awalls/v4l-dvb-misc

Mauro,

Please pull from http://linuxtv.org/hg/~awalls/v4l-dvb-misc

for the following 7 changesets:

01/07: ivtv: Update the cards definitions to add another AVerMedia M113 variant
http://linuxtv.org/hg/~awalls/v4l-dvb-misc?cmd=changeset;node=b7e065c84fc7

02/07: ivtv: Added FM radio support to Avermedia AVerTV MCE 116 Plus card
http://linuxtv.org/hg/~awalls/v4l-dvb-misc?cmd=changeset;node=77af5cddacb5

03/07: ivtv: Add module parameter to adjust I2C SCL clock period per board
http://linuxtv.org/hg/~awalls/v4l-dvb-misc?cmd=changeset;node=ab8a84fe29ff

04/07: ivtv: Defer legacy I2C IR probing until after setup of known I2C devices
http://linuxtv.org/hg/~awalls/v4l-dvb-misc?cmd=changeset;node=0b2cae0e938f

05/07: ivtv: Add explicit IR controller initialization for the AVerTV M116
http://linuxtv.org/hg/~awalls/v4l-dvb-misc?cmd=changeset;node=9ca00802e126

06/07: cx18: Use per cx18 instance init data for ir-kbd-i2c instead of const data
http://linuxtv.org/hg/~awalls/v4l-dvb-misc?cmd=changeset;node=46e56e52b8ae

07/07: ivtv: Add more explicit detection of known IR devices for Hauppauge cards
http://linuxtv.org/hg/~awalls/v4l-dvb-misc?cmd=changeset;node=2659ce7e3123


cx18/cx18-cards.h | 3
cx18/cx18-driver.h | 7 ++
cx18/cx18-i2c.c | 31 +++------
ivtv/ivtv-cards.c | 16 +++-
ivtv/ivtv-cards.h | 87 +++++++++++++++----------
ivtv/ivtv-driver.c | 25 +++++++
ivtv/ivtv-driver.h | 10 ++
ivtv/ivtv-i2c.c | 180 ++++++++++++++++++++++++++++++++++++++++-------------
ivtv/ivtv-i2c.h | 1
9 files changed, 259 insertions(+), 101 deletions(-)

Thanks,
Andy


_______________________________________________
ivtv-devel mailing list
ivtv-devel [at] ivtvdriver
http://ivtvdriver.org/mailman/listinfo/ivtv-devel


mchehab at infradead

Jan 10, 2010, 3:35 AM

Post #2 of 5 (1735 views)
Permalink
Re: [PULL] http://linuxtv.org/hg/~awalls/v4l-dvb-misc [In reply to]

Andy Walls wrote:
> Mauro,
>
> If no one has any objections, please pull from
>
> http://linuxtv.org/hg/~awalls/v4l-dvb-misc
>
> for the following 12 changesets.
>
> Of note:
> 02-04 are from Jean Delvare and fix up the cx23885 i2c routines
> 05-17 and 12 add and use a new v4l2_subdev core op for configuring I/O pin muxes
> 08-10 are some minor cx23885 ir fixes noted when trying to get the TeVii S470 working
>
> 10/12: cx23885: Convert from struct card_ir to struct cx23885_ir_input for IR Rx
> http://linuxtv.org/hg/~awalls/v4l-dvb-misc?cmd=changeset;node=aa62944baa92

Hmm... This doesn't sound right:

+struct cx23885_ir_input {
+ struct input_dev *dev;
+ struct ir_input_state ir;
+ char name[48];
+ char phys[48];
+
+ /* Cooked code processing */
+ int start; /* Allowed start symbols */
+ u32 addr; /* Expected remote address */
+ u32 last_code; /* last good cooked code seen */
+ int key_timeout; /* ms until we force a key up */
+ struct timer_list timer_keyup; /* timer for key release */
+
+ /* Raw code collection and construction */
+ int active; /* building code */
+ int last_bit; /* last bit seen */
+ u32 code; /* code under construction */
+};

Why are you creating a name[] and phys[] chars here? It should be using the names already
defined at struct input_dev.

Cheers,
Mauro.

_______________________________________________
ivtv-devel mailing list
ivtv-devel [at] ivtvdriver
http://ivtvdriver.org/mailman/listinfo/ivtv-devel


awalls at radix

Jan 10, 2010, 5:35 AM

Post #3 of 5 (1732 views)
Permalink
Re: [PULL] http://linuxtv.org/hg/~awalls/v4l-dvb-misc [In reply to]

On Sun, 2010-01-10 at 09:35 -0200, Mauro Carvalho Chehab wrote:
> Andy Walls wrote:
> > Mauro,
> >
> > If no one has any objections, please pull from
> >
> > http://linuxtv.org/hg/~awalls/v4l-dvb-misc
> >
> > for the following 12 changesets.
> >
> > Of note:
> > 02-04 are from Jean Delvare and fix up the cx23885 i2c routines
> > 05-17 and 12 add and use a new v4l2_subdev core op for configuring I/O pin muxes
> > 08-10 are some minor cx23885 ir fixes noted when trying to get the TeVii S470 working
> >
> > 10/12: cx23885: Convert from struct card_ir to struct cx23885_ir_input for IR Rx
> > http://linuxtv.org/hg/~awalls/v4l-dvb-misc?cmd=changeset;node=aa62944baa92
>
> Hmm... This doesn't sound right:
>
> +struct cx23885_ir_input {
> + struct input_dev *dev;
> + struct ir_input_state ir;
> + char name[48];
> + char phys[48];
> +
> + /* Cooked code processing */
> + int start; /* Allowed start symbols */
> + u32 addr; /* Expected remote address */
> + u32 last_code; /* last good cooked code seen */
> + int key_timeout; /* ms until we force a key up */
> + struct timer_list timer_keyup; /* timer for key release */
> +
> + /* Raw code collection and construction */
> + int active; /* building code */
> + int last_bit; /* last bit seen */
> + u32 code; /* code under construction */
> +};
>
> Why are you creating a name[] and phys[] chars here? It should be using the names already
> defined at struct input_dev.

Well two reasons:

1. That's what the previous, common "card ir" struct did. (Not a good
reason of course.) When I needed to reimplement specific fields (in
anticipation of NEC decoding for the TeVii S470) I just carried them
over.

2. The strings in the old card ir struct were too short: the card names
in cx23885-cards.c are pretty long and would get truncated.


I'll reexamine if the strings in input_dev are long enough to do the
job, and get back to you.

Regards,
Andy

> Cheers,
> Mauro.
>


_______________________________________________
ivtv-devel mailing list
ivtv-devel [at] ivtvdriver
http://ivtvdriver.org/mailman/listinfo/ivtv-devel


mchehab at infradead

Jan 10, 2010, 5:52 AM

Post #4 of 5 (1726 views)
Permalink
Re: [PULL] http://linuxtv.org/hg/~awalls/v4l-dvb-misc [In reply to]

Andy Walls wrote:
> On Sun, 2010-01-10 at 09:35 -0200, Mauro Carvalho Chehab wrote:
>> Andy Walls wrote:
>>> Mauro,
>>>
>>> If no one has any objections, please pull from
>>>
>>> http://linuxtv.org/hg/~awalls/v4l-dvb-misc
>>>
>>> for the following 12 changesets.
>>>
>>> Of note:
>>> 02-04 are from Jean Delvare and fix up the cx23885 i2c routines
>>> 05-17 and 12 add and use a new v4l2_subdev core op for configuring I/O pin muxes
>>> 08-10 are some minor cx23885 ir fixes noted when trying to get the TeVii S470 working
>>>
>>> 10/12: cx23885: Convert from struct card_ir to struct cx23885_ir_input for IR Rx
>>> http://linuxtv.org/hg/~awalls/v4l-dvb-misc?cmd=changeset;node=aa62944baa92
>> Hmm... This doesn't sound right:
>>
>> +struct cx23885_ir_input {
>> + struct input_dev *dev;
>> + struct ir_input_state ir;
>> + char name[48];
>> + char phys[48];
>> +
>> + /* Cooked code processing */
>> + int start; /* Allowed start symbols */
>> + u32 addr; /* Expected remote address */
>> + u32 last_code; /* last good cooked code seen */
>> + int key_timeout; /* ms until we force a key up */
>> + struct timer_list timer_keyup; /* timer for key release */
>> +
>> + /* Raw code collection and construction */
>> + int active; /* building code */
>> + int last_bit; /* last bit seen */
>> + u32 code; /* code under construction */
>> +};
>>
>> Why are you creating a name[] and phys[] chars here? It should be using the names already
>> defined at struct input_dev.
>
> Well two reasons:
>
> 1. That's what the previous, common "card ir" struct did. (Not a good
> reason of course.) When I needed to reimplement specific fields (in
> anticipation of NEC decoding for the TeVii S470) I just carried them
> over.
>
> 2. The strings in the old card ir struct were too short: the card names
> in cx23885-cards.c are pretty long and would get truncated.
>
>
> I'll reexamine if the strings in input_dev are long enough to do the
> job, and get back to you.

The better is to rely on input_dev stuff, since they can easily be used by ir-core
sysfs to provide device naming for loading keytables from userspace during udev
handling.

Cheers,
Mauro.

_______________________________________________
ivtv-devel mailing list
ivtv-devel [at] ivtvdriver
http://ivtvdriver.org/mailman/listinfo/ivtv-devel


awalls at radix

Jan 11, 2010, 3:57 PM

Post #5 of 5 (1707 views)
Permalink
Re: [PULL] http://linuxtv.org/hg/~awalls/v4l-dvb-misc [In reply to]

On Sun, 2010-01-10 at 11:52 -0200, Mauro Carvalho Chehab wrote:
> Andy Walls wrote:
> > On Sun, 2010-01-10 at 09:35 -0200, Mauro Carvalho Chehab wrote:
> >> Andy Walls wrote:
> >>> Mauro,
> >>>
> >>> If no one has any objections, please pull from
> >>>
> >>> http://linuxtv.org/hg/~awalls/v4l-dvb-misc
> >>>
> >>> for the following 12 changesets.
> >>>
> >>> Of note:
> >>> 02-04 are from Jean Delvare and fix up the cx23885 i2c routines
> >>> 05-17 and 12 add and use a new v4l2_subdev core op for configuring I/O pin muxes
> >>> 08-10 are some minor cx23885 ir fixes noted when trying to get the TeVii S470 working
> >>>
> >>> 10/12: cx23885: Convert from struct card_ir to struct cx23885_ir_input for IR Rx
> >>> http://linuxtv.org/hg/~awalls/v4l-dvb-misc?cmd=changeset;node=aa62944baa92
> >> Hmm... This doesn't sound right:
> >>
> >> +struct cx23885_ir_input {
> >> + struct input_dev *dev;
> >> + struct ir_input_state ir;
> >> + char name[48];
> >> + char phys[48];
> >> +
> >> + /* Cooked code processing */
> >> + int start; /* Allowed start symbols */
> >> + u32 addr; /* Expected remote address */
> >> + u32 last_code; /* last good cooked code seen */
> >> + int key_timeout; /* ms until we force a key up */
> >> + struct timer_list timer_keyup; /* timer for key release */
> >> +
> >> + /* Raw code collection and construction */
> >> + int active; /* building code */
> >> + int last_bit; /* last bit seen */
> >> + u32 code; /* code under construction */
> >> +};
> >>
> >> Why are you creating a name[] and phys[] chars here? It should be using the names already
> >> defined at struct input_dev.
> >
> > Well two reasons:
> >
> > 1. That's what the previous, common "card ir" struct did. (Not a good
> > reason of course.) When I needed to reimplement specific fields (in
> > anticipation of NEC decoding for the TeVii S470) I just carried them
> > over.
> >
> > 2. The strings in the old card ir struct were too short: the card names
> > in cx23885-cards.c are pretty long and would get truncated.
> >
> >
> > I'll reexamine if the strings in input_dev are long enough to do the
> > job, and get back to you.
>
> The better is to rely on input_dev stuff, since they can easily be used by ir-core
> sysfs to provide device naming for loading keytables from userspace during udev
> handling.

OK. Hold off on that whole pull request. That whole pull request was
the stable/ready part of what is in my cx23885-ir tree.

However, I just found what was wrong with my cx23885-ir tree code for IR
from an actual CX23885 chip. I'll just rework and update the whole
series, once I also get a change prepared and tested for the TeVii S470.

Regards,
Andy

> Cheers,
> Mauro.
> --



_______________________________________________
ivtv-devel mailing list
ivtv-devel [at] ivtvdriver
http://ivtvdriver.org/mailman/listinfo/ivtv-devel

ivtv devel RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.