Skip to contentSkip to conversation

V4L/DVB fixes

PATCH Boris Cuber 4 participants 5 messages Jul 12, 2009 – Jul 13, 2009 Media (V4L/DVB)

lore
Share & tools
lore
#0 lore raw

Hi kernel folks!

Problem:
Since kernel-2.6.31-rc* my dvb-s adapter (Technisat SkyStar2 DVB card) refuses to work (worked fine in every kernel up to 2.6.30.1). So anything pulled into the new kernel seems to have broken something (at least for me :/).

I opened a detailed bug report here:
http://bugzilla.kernel.org/show_bug.cgi?id=13709
Please let me know if i can help in finding a solution or testing a patch /whatever.

Thank you for your attention ;)

PS: As i'm not subscribed to this mailing list, please answer to my address or cc me.

Signature · 1 line
http://boris64.net 20xx ;)
#1 lore raw
Quoted 8 lines · Boris Cuber

Hi kernel folks!

Problem:
Since kernel-2.6.31-rc* my dvb-s adapter (Technisat SkyStar2 DVB card) refuses to work (worked fine in every kernel up to 2.6.30.1). So anything pulled into the new kernel seems to have broken something (at least for me :/).

Quoted 4 lines

I opened a detailed bug report here:
http://bugzilla.kernel.org/show_bug.cgi?id=13709
Please let me know if i can help in finding a solution or testing a patch /whatever.

This looks like it is related to this patch:

commit d66b94b4aa2f40e134f8c07c58ae74ef3d523ee0

Author: Patrick Boettcher <[email protected]>
Date:   Wed May 20 05:08:26 2009 -0300

V4L/DVB (11829): Rewrote frontend-attach mechanism to gain noise-less
deactivation of submodules

This patch is reorganizing the frontend-attach mechanism in order to gain noise-less (superflous prints) deactivation of submodules.

Credits go to Uwe Bugla for helping to clean and test the code.

All frontend-attach related code is wrapped by ifdefs like this:
#if defined(CONFIG_DVB_MT312_MODULE) || defined(CONFIG_DVB_STV0299_MODULE) <CODE>
#endif

So this code will only be compiled if one of the two drivers is compiled as a module, having them compiled in will omit this code.

Trent Piepho seems to already have a patch for this, but it is not yet merged into the kernel.

Regards
Matthias

#2 lore raw

Am Montag 13 Juli 2009 14:13:50 schrieben Sie:

Quoted 34 lines · Boris Cuber
Quoted 8 lines

Hi kernel folks!

Problem:
Since kernel-2.6.31-rc* my dvb-s adapter (Technisat SkyStar2 DVB card) refuses to work (worked fine in every kernel up to 2.6.30.1). So anything pulled into the new kernel seems to have broken something (at least for me :/).

Quoted 4 lines

I opened a detailed bug report here:
http://bugzilla.kernel.org/show_bug.cgi?id=13709
Please let me know if i can help in finding a solution or testing a patch /whatever.

This looks like it is related to this patch:

commit d66b94b4aa2f40e134f8c07c58ae74ef3d523ee0

Author: Patrick Boettcher <[email protected]>
Date:   Wed May 20 05:08:26 2009 -0300

V4L/DVB (11829): Rewrote frontend-attach mechanism to gain noise-less
deactivation of submodules

This patch is reorganizing the frontend-attach mechanism in order to gain noise-less (superflous prints) deactivation of submodules.

Credits go to Uwe Bugla for helping to clean and test the code.

Quoted 7 lines

All frontend-attach related code is wrapped by ifdefs like this:
#if defined(CONFIG_DVB_MT312_MODULE) || defined(CONFIG_DVB_STV0299_MODULE) <CODE>
#endif

So this code will only be compiled if one of the two drivers is compiled as a module, having them compiled in will omit this code.

Yes. And that's exactly the way things were planned and should also stay, even if there exist a thousands of "Boris64" who do not have the slightest idea about what kernel compilation is or could be.....

No matter if we're talking about the main module, the frontend, the backend or whatever other part of not only a DVB driver:
None of them is permanently needed while the machine is running. So kmod can kick them out of the memory if they aren't needed, if they were compiled as module.

But if you compile them into the kernel you are wasting system resources because the main kernel becomes too big (I'd call that a "Windoze-effect").

So compiling those drivers a module is gold, and any other choice is simply nonsense.

Quoted 2 lines

Trent Piepho seems to already have a patch for this, but it is not yet merged into the kernel.

May Trent Piepho do whatever he likes. I do not think that any further patch is necessary for that driver section.

It would rather be necessary for some quirky users to enlarge their limited brain and understand what kernel compilation means and is here for.

Quoted 2 lines

Regards
Matthias

CU

cyber.bogh

P. S.: The other part that really makes me utmost angry about the "Boris's" in that world:

If you're doing really hard for months to enhance things, and you urgently need testers to help and invest brain those Boris's aren't visible at all. Nowhere!

Once things are done they come back and all they have got to do then is to complain for stupid nonsense...

How did Lou Reed say?
"Stick a fork in their ass, turn it over and they're done!"

Quoted 3 lines

To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#3 lore raw

Hey cyber.bogh,

besides the fact that you MAY be right in your zhinking about such usage of drivers and kernel elements ...

This is not the right way to complain about users/developers not so deeply into the materia than you.
If you know more about special things in the area you work in - fine. Let others participate - and if you don't want to - be gentle and calm. If you need people testing your development output just ask. You are free to remember those e-mail addresses as of boris and ask him directly.

Beside all this: What does you mean with

"So compiling those drivers a module is gold, and any other choice is simply nonsense."

???

The concept of build in drivers and modules to my view is: use a "module" if functionality is not really needed at boot time otherwise "build in".

What does this mean in regard of your statement?

Does funtionality change if I would use "build in" instead of "modules" ????

Regards
/Jesko

-----Ursprüngliche Nachricht-----
Von: [email protected] [mailto:[email protected]] Im Auftrag von cyber.bogh
Gesendet: Montag, 13. Juli 2009 15:29
An: Matthias Schwarzott
Cc: [email protected]; [email protected]; Trent Piepho
Betreff: Re: [GIT PATCHES for 2.6.31] V4L/DVB fixes

Am Montag 13 Juli 2009 14:13:50 schrieben Sie:

Quoted 34 lines · Boris Cuber
Quoted 8 lines

Hi kernel folks!

Problem:
Since kernel-2.6.31-rc* my dvb-s adapter (Technisat SkyStar2 DVB card) refuses to work (worked fine in every kernel up to 2.6.30.1). So anything pulled into the new kernel seems to have broken something (at least for me :/).

Quoted 4 lines

I opened a detailed bug report here:
http://bugzilla.kernel.org/show_bug.cgi?id=13709
Please let me know if i can help in finding a solution or testing a patch /whatever.

This looks like it is related to this patch:

commit d66b94b4aa2f40e134f8c07c58ae74ef3d523ee0

Author: Patrick Boettcher <[email protected]>
Date:   Wed May 20 05:08:26 2009 -0300

V4L/DVB (11829): Rewrote frontend-attach mechanism to gain
noise-less deactivation of submodules

This patch is reorganizing the frontend-attach mechanism in order to gain noise-less (superflous prints) deactivation of submodules.

Credits go to Uwe Bugla for helping to clean and test the code.

Quoted 8 lines

All frontend-attach related code is wrapped by ifdefs like this:
#if defined(CONFIG_DVB_MT312_MODULE) || defined(CONFIG_DVB_STV0299_MODULE)
<CODE>
#endif

So this code will only be compiled if one of the two drivers is compiled as a module, having them compiled in will omit this code.

Yes. And that's exactly the way things were planned and should also stay, even if there exist a thousands of "Boris64" who do not have the slightest idea about what kernel compilation is or could be.....

No matter if we're talking about the main module, the frontend, the backend or whatever other part of not only a DVB driver:
None of them is permanently needed while the machine is running. So kmod can kick them out of the memory if they aren't needed, if they were compiled as module.

But if you compile them into the kernel you are wasting system resources because the main kernel becomes too big (I'd call that a "Windoze-effect").

So compiling those drivers a module is gold, and any other choice is simply nonsense.

Quoted 2 lines

Trent Piepho seems to already have a patch for this, but it is not yet merged into the kernel.

May Trent Piepho do whatever he likes. I do not think that any further patch is necessary for that driver section.

It would rather be necessary for some quirky users to enlarge their limited brain and understand what kernel compilation means and is here for.

Quoted 2 lines

Regards
Matthias

CU

cyber.bogh

P. S.: The other part that really makes me utmost angry about the "Boris's" in that world:

If you're doing really hard for months to enhance things, and you urgently need testers to help and invest brain those Boris's aren't visible at all. Nowhere!

Once things are done they come back and all they have got to do then is to complain for stupid nonsense...

How did Lou Reed say?
"Stick a fork in their ass, turn it over and they're done!"

Quoted 3 lines

To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

--
To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to [email protected] More majordomo info at
http://vger.kernel.org/majordomo-info.html

#4 lore raw
In reply to cyber.bogh

Am Montag, 13. Juli 2009 schrieben Sie:

Quoted 42 lines

Am Montag 13 Juli 2009 14:13:50 schrieben Sie:

Quoted 41 lines · Boris Cuber
Quoted 8 lines

Hi kernel folks!

Problem:
Since kernel-2.6.31-rc* my dvb-s adapter (Technisat SkyStar2 DVB card) refuses to work (worked fine in every kernel up to 2.6.30.1). So anything pulled into the new kernel seems to have broken something (at least for me :/).

Quoted 4 lines

I opened a detailed bug report here:
http://bugzilla.kernel.org/show_bug.cgi?id=13709
Please let me know if i can help in finding a solution or testing a patch /whatever.

This looks like it is related to this patch:

commit d66b94b4aa2f40e134f8c07c58ae74ef3d523ee0

Author: Patrick Boettcher <[email protected]>
Date:   Wed May 20 05:08:26 2009 -0300

V4L/DVB (11829): Rewrote frontend-attach mechanism to gain noise-less
deactivation of submodules

This patch is reorganizing the frontend-attach mechanism in order to gain noise-less (superflous prints) deactivation of submodules.

Credits go to Uwe Bugla for helping to clean and test the code.

All frontend-attach related code is wrapped by ifdefs like this:
#if defined(CONFIG_DVB_MT312_MODULE) || defined(CONFIG_DVB_STV0299_MODULE) <CODE>
#endif

So this code will only be compiled if one of the two drivers is compiled as a module, having them compiled in will omit this code.

Hi Matthias

thank you for your help. I manually edited
those #ifdefs to temporary get around my problem.
My dvb adapter is working again ;)

Quoted 4 lines

Yes. And that's exactly the way things were planned and should also stay, even if there exist a thousands of "Boris64" who do not have the slightest idea about what kernel compilation is or could be.....

Well, i always thought compiling your own kernel is about choice. And _my_ choice always was:
I don't want use modules, because booting a monolithic kernel on my computer is slightly faster.

Quoted 27 lines

No matter if we're talking about the main module, the frontend, the backend or whatever other part of not only a DVB driver:
None of them is permanently needed while the machine is running. So kmod can kick them out of the memory if they aren't needed, if they were compiled as module.

But if you compile them into the kernel you are wasting system resources because the main kernel becomes too big (I'd call that a "Windoze-effect").

So compiling those drivers a module is gold, and any other choice is simply nonsense.

Quoted 2 lines

Trent Piepho seems to already have a patch for this, but it is not yet merged into the kernel.

May Trent Piepho do whatever he likes. I do not think that any further patch is necessary for that driver section.

It would rather be necessary for some quirky users to enlarge their limited brain and understand what kernel compilation means and is here for.

Quoted 2 lines

Regards
Matthias

CU

Quoted 8 lines

cyber.bogh

P. S.: The other part that really makes me utmost angry about the "Boris's" in that world:

If you're doing really hard for months to enhance things, and you urgently need testers to help and invest brain those Boris's aren't visible at all. Nowhere!

Those "Boris's" can't remember that anyone ever asked'em to help. And "they" didn't subscribe to this list until "their" dvb card(s) didn't work anymore for some reason.

Quoted 7 lines

Once things are done they come back and all they have got to do then is to complain for stupid nonsense...

How did Lou Reed say?
"Stick a fork in their ass, turn it over and they're done!"

Have a nice day,
regards, Boris