linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[Javier Sanz]

Hello,

Sorry about nvidia drivers issue, but nobody help me from another forums ...

I can't boot linux rt 3.4.8-rt16 with nvidia 302.17 patched for 3.x-rt
series using the famous  "patch-3.0-rt-nvidia.patch"

My Fedora 17 starts ok,  services ok, but into loading nvidia driver,
it freezes ...  no text mode, no keyboard .. all locked !

I guess some one has it working ... can you help me? which patch did
you use?  how do you do it?  where can i find it ?

I'll try with lastest 3.4.9-rt17  and  nvidia 304.37 with the same patch ..

do any clues to get it works?

I guess that info help a lot of people

I wonder ... I don't know how difficult and hard is test linux rt
patch for nvidia drivers ... but ... if someone help me, i'll test it
on my system
differents rt kernels and nvidia drivers and send an email explaining
solution and process...

I can't believe how "great company, nvidia" has not enough resources
to support linux community ... next buy ... amd card ? intel card?
who is the lesser evil?

thank you

Re: linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[Mike Galbraith]

On Fri, 2012-08-17 at 22:40 +0200, Javier Sanz wrote: 
> Hello,
> 
> Sorry about nvidia drivers issue, but nobody help me from another forums ...
> 
> I can't boot linux rt 3.4.8-rt16 with nvidia 302.17 patched for 3.x-rt
> series using the famous  "patch-3.0-rt-nvidia.patch"
> 
> My Fedora 17 starts ok,  services ok, but into loading nvidia driver,
> it freezes ...  no text mode, no keyboard .. all locked !
> 
> I guess some one has it working ... can you help me? which patch did
> you use?  how do you do it?  where can i find it ?
> 
> I'll try with lastest 3.4.9-rt17  and  nvidia 304.37 with the same patch ..

I just built 3.4.9-rt17 with nvidia 304.37.  For my 8600 GT card at
least, no patch is required, it builds and runs just fine as is.

-Mike

Re: linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[David L. Craig]

On 12Aug18:0633+0200, Mike Galbraith wrote:

> On Fri, 2012-08-17 at 22:40 +0200, Javier Sanz wrote: 
> > 
> > Sorry about nvidia drivers issue, but nobody help me from another forums ...
> > 
> > I can't boot linux rt 3.4.8-rt16 with nvidia 302.17 patched for 3.x-rt
> > series using the famous  "patch-3.0-rt-nvidia.patch"
> > 
> > My Fedora 17 starts ok,  services ok, but into loading nvidia driver,
> > it freezes ...  no text mode, no keyboard .. all locked !
> > 
> > I guess some one has it working ... can you help me? which patch did
> > you use?  how do you do it?  where can i find it ?
> > 
> > I'll try with lastest 3.4.9-rt17  and  nvidia 304.37 with the same patch ..
> 
> I just built 3.4.9-rt17 with nvidia 304.37.  For my 8600 GT card at
> least, no patch is required, it builds and runs just fine as is.

Is this bug report similar to what you're experiencing?

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=684952
-- 
May the LORD God bless you exceedingly abundantly!

Dave_Craig______________________________________________
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_________________

Re: linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[Javier Sanz]

Yes, i got the same problem, black screen, no mouse, no keyboard ...
i didnt  change any config file from X or nvidia ... only rt patch
applied  ...  :(
I'll try with nvidia 304.37 ... no patch required ? ... i'll test ...

thank you


2012/8/18 David L. Craig <dlc.usa@gmail.com>:
> On 12Aug18:0633+0200, Mike Galbraith wrote:
>
>> On Fri, 2012-08-17 at 22:40 +0200, Javier Sanz wrote:
>> >
>> > Sorry about nvidia drivers issue, but nobody help me from another forums ...
>> >
>> > I can't boot linux rt 3.4.8-rt16 with nvidia 302.17 patched for 3.x-rt
>> > series using the famous  "patch-3.0-rt-nvidia.patch"
>> >
>> > My Fedora 17 starts ok,  services ok, but into loading nvidia driver,
>> > it freezes ...  no text mode, no keyboard .. all locked !
>> >
>> > I guess some one has it working ... can you help me? which patch did
>> > you use?  how do you do it?  where can i find it ?
>> >
>> > I'll try with lastest 3.4.9-rt17  and  nvidia 304.37 with the same patch ..
>>
>> I just built 3.4.9-rt17 with nvidia 304.37.  For my 8600 GT card at
>> least, no patch is required, it builds and runs just fine as is.
>
> Is this bug report similar to what you're experiencing?
>
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=684952
> --
> May the LORD God bless you exceedingly abundantly!
>
> Dave_Craig______________________________________________
> "So the universe is not quite as you thought it was.
>  You'd better rearrange your beliefs, then.
>  Because you certainly can't rearrange the universe."
> __--from_Nightfall_by_Asimov/Silverberg_________________

Re: linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[David L. Craig]

On 12Aug18:1152+0200, Javier Sanz wrote:

> Yes, i got the same problem, black screen, no mouse, no keyboard ...
> i didnt  change any config file from X or nvidia ... only rt patch
> applied  ...  :(
> I'll try with nvidia 304.37 ... no patch required ? ... i'll test ...
> 
> thank you

Just to be sure you didn't misunderstand the situation,
I do not have any rt kernel working with nvidia
except the Linux Mint Debian XFCE 13 distribution.
Debian considers this an important bug but nVidia's
assessment is unknown to date.

Perhaps another rt-kernel lister can offer a quality
suggestion about a likely fruitful approach to finding
people able and willing to dig into what the kern.log
shows is going on and direct further resolution
activities.  Such people are probably rt-kernel and/or
nVidia developers.  I could probably do it myself after
three months of nothing but working on it (including
in my sleep ;-) ).  I'd hazard a guess the following
has a lot to do with the problem:

[    6.993495] Disabling lock debugging due to kernel taint

I don't know what the long-term outlook for nVidia market
share in the FLOSS area is.  I saw hardware acceleration
is apparently required for GNOME Unity.  That's good for
the hardware vendors, I suppose, but problematic for folks
who take the L in FLOSS very seriously.  I think we all
wish the vendors would choose to make these problems go
away by distributing their source.  As RMS says, they
won't if they don't have to, so we shouldn't be pragmatic,
just say no thank you.  Can I live without FlightGear?
Why should I have to, nVidia (BTW, I was given your
hardware--I wouldn't buy it)?  If the gamers really are
becoming Windows 8 refugees, I guess there's hope for the
future.  But if the governments start locking up systems
programming, doling out practioner licenses, and making
libre software illegal, then not.

But to hopefully get back on topic, I have the impression
I NEED the rt kernel because FG frame updates get
behind without it.  I've heard JACK audio works well
enough without rt these days (but I don't believe it
yet--soon enough I will be able to prove it to myself,
unless I get a job).
-- 
May the LORD God bless you exceedingly abundantly!

Dave_Craig______________________________________________
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_________________

Re: linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[Javier Sanz]

Hello,

I can understand that there are many problems, nvidia drivers, rt
kernel, and no collaboration between those involved, thereby
ultimately harms the opensource community, etc., And in my case, the
main thing is that I have no knowledge to fix them, or help to be
corrected, for now all I can ask for help and hope someone help me.

For example, how I can see where the problem occurs? because there is
no trace in the files of nvidia, the dmesg, nowhere ...
how can i get a document to help people to find theses problems? which
files affected, boots modes, etc ...  show to people that there are a
way to damm nvidia drivers ...

You say that there is a message,
[6.993495] Disabling lock debugging due to kernel taint

how do you get it ? have you sent this information any nvidia  forums?
no answer?  how can we fix it ?

Yesterday, testing with kernel 3.4.9-rt17 with nvidia 304.37 drivers
... nvidia gt 240 ...  all blocked .. i am at same point ...

So, what is the next step? mails? forums?

What do you suggest?

TX


2012/8/18 David L. Craig <dlc.usa@gmail.com>:
> On 12Aug18:1152+0200, Javier Sanz wrote:
>
>> Yes, i got the same problem, black screen, no mouse, no keyboard ...
>> i didnt  change any config file from X or nvidia ... only rt patch
>> applied  ...  :(
>> I'll try with nvidia 304.37 ... no patch required ? ... i'll test ...
>>
>> thank you
>
> Just to be sure you didn't misunderstand the situation,
> I do not have any rt kernel working with nvidia
> except the Linux Mint Debian XFCE 13 distribution.
> Debian considers this an important bug but nVidia's
> assessment is unknown to date.
>
> Perhaps another rt-kernel lister can offer a quality
> suggestion about a likely fruitful approach to finding
> people able and willing to dig into what the kern.log
> shows is going on and direct further resolution
> activities.  Such people are probably rt-kernel and/or
> nVidia developers.  I could probably do it myself after
> three months of nothing but working on it (including
> in my sleep ;-) ).  I'd hazard a guess the following
> has a lot to do with the problem:
>
> [    6.993495] Disabling lock debugging due to kernel taint
>
> I don't know what the long-term outlook for nVidia market
> share in the FLOSS area is.  I saw hardware acceleration
> is apparently required for GNOME Unity.  That's good for
> the hardware vendors, I suppose, but problematic for folks
> who take the L in FLOSS very seriously.  I think we all
> wish the vendors would choose to make these problems go
> away by distributing their source.  As RMS says, they
> won't if they don't have to, so we shouldn't be pragmatic,
> just say no thank you.  Can I live without FlightGear?
> Why should I have to, nVidia (BTW, I was given your
> hardware--I wouldn't buy it)?  If the gamers really are
> becoming Windows 8 refugees, I guess there's hope for the
> future.  But if the governments start locking up systems
> programming, doling out practioner licenses, and making
> libre software illegal, then not.
>
> But to hopefully get back on topic, I have the impression
> I NEED the rt kernel because FG frame updates get
> behind without it.  I've heard JACK audio works well
> enough without rt these days (but I don't believe it
> yet--soon enough I will be able to prove it to myself,
> unless I get a job).
> --
> May the LORD God bless you exceedingly abundantly!
>
> Dave_Craig______________________________________________
> "So the universe is not quite as you thought it was.
>  You'd better rearrange your beliefs, then.
>  Because you certainly can't rearrange the universe."
> __--from_Nightfall_by_Asimov/Silverberg_________________

Re: linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[Sven-Thorsten Dietrich]

Can you use a product other than NVidia?

It's a free market:

Use what works, reject what doesn't.

Sven-Thorsten Dietrich 
415 4804180

On Aug 23, 2012, at 1:47, Javier Sanz <jsanza@gmail.com> wrote:

> Hello,
> 
> I can understand that there are many problems, nvidia drivers, rt
> kernel, and no collaboration between those involved, thereby
> ultimately harms the opensource community, etc., And in my case, the
> main thing is that I have no knowledge to fix them, or help to be
> corrected, for now all I can ask for help and hope someone help me.
> 
> For example, how I can see where the problem occurs? because there is
> no trace in the files of nvidia, the dmesg, nowhere ...
> how can i get a document to help people to find theses problems? which
> files affected, boots modes, etc ...  show to people that there are a
> way to damm nvidia drivers ...
> 
> You say that there is a message,
> [6.993495] Disabling lock debugging due to kernel taint
> 
> how do you get it ? have you sent this information any nvidia  forums?
> no answer?  how can we fix it ?
> 
> Yesterday, testing with kernel 3.4.9-rt17 with nvidia 304.37 drivers
> ... nvidia gt 240 ...  all blocked .. i am at same point ...
> 
> So, what is the next step? mails? forums?
> 
> What do you suggest?
> 
> TX
> 
> 
> 2012/8/18 David L. Craig <dlc.usa@gmail.com>:
>> On 12Aug18:1152+0200, Javier Sanz wrote:
>> 
>>> Yes, i got the same problem, black screen, no mouse, no keyboard ...
>>> i didnt  change any config file from X or nvidia ... only rt patch
>>> applied  ...  :(
>>> I'll try with nvidia 304.37 ... no patch required ? ... i'll test ...
>>> 
>>> thank you
>> 
>> Just to be sure you didn't misunderstand the situation,
>> I do not have any rt kernel working with nvidia
>> except the Linux Mint Debian XFCE 13 distribution.
>> Debian considers this an important bug but nVidia's
>> assessment is unknown to date.
>> 
>> Perhaps another rt-kernel lister can offer a quality
>> suggestion about a likely fruitful approach to finding
>> people able and willing to dig into what the kern.log
>> shows is going on and direct further resolution
>> activities.  Such people are probably rt-kernel and/or
>> nVidia developers.  I could probably do it myself after
>> three months of nothing but working on it (including
>> in my sleep ;-) ).  I'd hazard a guess the following
>> has a lot to do with the problem:
>> 
>> [    6.993495] Disabling lock debugging due to kernel taint
>> 
>> I don't know what the long-term outlook for nVidia market
>> share in the FLOSS area is.  I saw hardware acceleration
>> is apparently required for GNOME Unity.  That's good for
>> the hardware vendors, I suppose, but problematic for folks
>> who take the L in FLOSS very seriously.  I think we all
>> wish the vendors would choose to make these problems go
>> away by distributing their source.  As RMS says, they
>> won't if they don't have to, so we shouldn't be pragmatic,
>> just say no thank you.  Can I live without FlightGear?
>> Why should I have to, nVidia (BTW, I was given your
>> hardware--I wouldn't buy it)?  If the gamers really are
>> becoming Windows 8 refugees, I guess there's hope for the
>> future.  But if the governments start locking up systems
>> programming, doling out practioner licenses, and making
>> libre software illegal, then not.
>> 
>> But to hopefully get back on topic, I have the impression
>> I NEED the rt kernel because FG frame updates get
>> behind without it.  I've heard JACK audio works well
>> enough without rt these days (but I don't believe it
>> yet--soon enough I will be able to prove it to myself,
>> unless I get a job).
>> --
>> May the LORD God bless you exceedingly abundantly!
>> 
>> Dave_Craig______________________________________________
>> "So the universe is not quite as you thought it was.
>> You'd better rearrange your beliefs, then.
>> Because you certainly can't rearrange the universe."
>> __--from_Nightfall_by_Asimov/Silverberg_________________
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[Javier Sanz]

Yes, it is a "free" market, but there is "no free money" to make
change as needed...


2012/8/23 Sven-Thorsten Dietrich <sven@thebigcorporation.com>:
> Can you use a product other than NVidia?
>
> It's a free market:
>
> Use what works, reject what doesn't.
>
> Sven-Thorsten Dietrich
> 415 4804180
>
> On Aug 23, 2012, at 1:47, Javier Sanz <jsanza@gmail.com> wrote:
>
>> Hello,
>>
>> I can understand that there are many problems, nvidia drivers, rt
>> kernel, and no collaboration between those involved, thereby
>> ultimately harms the opensource community, etc., And in my case, the
>> main thing is that I have no knowledge to fix them, or help to be
>> corrected, for now all I can ask for help and hope someone help me.
>>
>> For example, how I can see where the problem occurs? because there is
>> no trace in the files of nvidia, the dmesg, nowhere ...
>> how can i get a document to help people to find theses problems? which
>> files affected, boots modes, etc ...  show to people that there are a
>> way to damm nvidia drivers ...
>>
>> You say that there is a message,
>> [6.993495] Disabling lock debugging due to kernel taint
>>
>> how do you get it ? have you sent this information any nvidia  forums?
>> no answer?  how can we fix it ?
>>
>> Yesterday, testing with kernel 3.4.9-rt17 with nvidia 304.37 drivers
>> ... nvidia gt 240 ...  all blocked .. i am at same point ...
>>
>> So, what is the next step? mails? forums?
>>
>> What do you suggest?
>>
>> TX
>>
>>
>> 2012/8/18 David L. Craig <dlc.usa@gmail.com>:
>>> On 12Aug18:1152+0200, Javier Sanz wrote:
>>>
>>>> Yes, i got the same problem, black screen, no mouse, no keyboard ...
>>>> i didnt  change any config file from X or nvidia ... only rt patch
>>>> applied  ...  :(
>>>> I'll try with nvidia 304.37 ... no patch required ? ... i'll test ...
>>>>
>>>> thank you
>>>
>>> Just to be sure you didn't misunderstand the situation,
>>> I do not have any rt kernel working with nvidia
>>> except the Linux Mint Debian XFCE 13 distribution.
>>> Debian considers this an important bug but nVidia's
>>> assessment is unknown to date.
>>>
>>> Perhaps another rt-kernel lister can offer a quality
>>> suggestion about a likely fruitful approach to finding
>>> people able and willing to dig into what the kern.log
>>> shows is going on and direct further resolution
>>> activities.  Such people are probably rt-kernel and/or
>>> nVidia developers.  I could probably do it myself after
>>> three months of nothing but working on it (including
>>> in my sleep ;-) ).  I'd hazard a guess the following
>>> has a lot to do with the problem:
>>>
>>> [    6.993495] Disabling lock debugging due to kernel taint
>>>
>>> I don't know what the long-term outlook for nVidia market
>>> share in the FLOSS area is.  I saw hardware acceleration
>>> is apparently required for GNOME Unity.  That's good for
>>> the hardware vendors, I suppose, but problematic for folks
>>> who take the L in FLOSS very seriously.  I think we all
>>> wish the vendors would choose to make these problems go
>>> away by distributing their source.  As RMS says, they
>>> won't if they don't have to, so we shouldn't be pragmatic,
>>> just say no thank you.  Can I live without FlightGear?
>>> Why should I have to, nVidia (BTW, I was given your
>>> hardware--I wouldn't buy it)?  If the gamers really are
>>> becoming Windows 8 refugees, I guess there's hope for the
>>> future.  But if the governments start locking up systems
>>> programming, doling out practioner licenses, and making
>>> libre software illegal, then not.
>>>
>>> But to hopefully get back on topic, I have the impression
>>> I NEED the rt kernel because FG frame updates get
>>> behind without it.  I've heard JACK audio works well
>>> enough without rt these days (but I don't believe it
>>> yet--soon enough I will be able to prove it to myself,
>>> unless I get a job).
>>> --
>>> May the LORD God bless you exceedingly abundantly!
>>>
>>> Dave_Craig______________________________________________
>>> "So the universe is not quite as you thought it was.
>>> You'd better rearrange your beliefs, then.
>>> Because you certainly can't rearrange the universe."
>>> __--from_Nightfall_by_Asimov/Silverberg_________________
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[David L. Craig]

On 12Aug23:1047+0200, Javier Sanz wrote:
> 
> You say that there is a message,
> [6.993495] Disabling lock debugging due to kernel taint
> 
> how do you get it ?

This and other pertinent messages are logged (under
Debian) in /var/log/kern.log and are crucial for
understanding what is going on.  Compare yours to the
one I attached to the Debian bug report.

> have you sent this information any nvidia  forums?
> no answer?  how can we fix it ?
> 
> Yesterday, testing with kernel 3.4.9-rt17 with nvidia
> 304.37 drivers ... nvidia gt 240 ...  all blocked ..
> i am at same point ...
> 
> So, what is the next step? mails? forums?
> 
> What do you suggest?

The nvidia driver, not being libre source, "taints"
the kernel environment, and policy mandates disabling
spin lock debugging logic as a consequence.  Some
interaction between nvidia and the rt kernel is the
most likely cause of the blocking encountered since
it doesn't manifest with comparable non-rt kernels.
Also, the problem has been introduced since the code
levels in the Linux Mint environment I documented in
the Debian bug report.  This is an area programmed
and debugged by well-studied kernel developers.
Mere mortals are not equipped to tackle such matters.

I have already done what I suggested in my previous
messages.  If an rt and/or nvidia developer does
not take an interest in this behaviour, it must be
hoped the misbehavior induces other problems they are
motivated to investigate.  I have asked for suggestions
on this list and in the Debian bug report about
next steps but thus far have received no responses.
While I do think I could learn enough to expertly
dig into this in a couple man-months, I do not think
I will be compelled to actually make it a priority
unless my circumstances significantly change; e.g.,
somebody offers me some financial incentive to do so.

I'll ask this list again: should this become an rt
kernel or nvidia bug report somewhere or should I
just let it percolate up from Debian?

In the meantime I don't use nvidia with the rt kernel
except with my Linux Mint partition, and that only to
run FlightGear (not very often anymore).
-- 
May the LORD God bless you exceedingly abundantly!

Dave_Craig______________________________________________
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_________________

Re: linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[David L. Craig]

On 12Aug23:1850-0400, David L. Craig wrote:
> 
> I'll ask this list again: should this become an rt
> kernel or nvidia bug report somewhere or should I
> just let it percolate up from Debian?

I'm looking at the nvidia forum's guidelines for reporting
bugs and will let you know what I determine to do.
-- 
May the LORD God bless you exceedingly abundantly!

Dave_Craig______________________________________________
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_________________

Re: linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[David L. Craig]

On 12Aug23:2032-0400, David L. Craig wrote:

> On 12Aug23:1850-0400, David L. Craig wrote:
> 
> I'm looking at the nvidia forum's guidelines for reporting
> bugs and will let you know what I determine to do.

There being no obvious comparable existing report, I opened
http://www.nvnews.net/vbulletin/showthread.php?p=2571517#post2571517
which points to the Debian report (684952) for all details.  Now we
wait again.
-- 
May the LORD God bless you exceedingly abundantly!

Dave_Craig______________________________________________
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_________________

Re: linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[Mike Galbraith]

On Thu, 2012-08-23 at 18:50 -0400, David L. Craig wrote:

> I have asked for suggestions
> on this list and in the Debian bug report about
> next steps but thus far have received no responses.

Once you report a black box (proprietary) driver problem to black box
manufacturer, your next and only step is to wait for same to produce a
new black box.  Nobody else can fix the problem, even if the source of
that problem lies in the well illuminated kernel.  It's dark inside the
black box, and only it's manufacturer can flip the light switch.

-Mike

Re: linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[David L. Craig]

On 12Aug24:0826+0200, Mike Galbraith wrote:

> On Thu, 2012-08-23 at 18:50 -0400, David L. Craig wrote:
> 
> > I have asked for suggestions
> > on this list and in the Debian bug report about
> > next steps but thus far have received no responses.
> 
> Once you report a black box (proprietary) driver problem to black box
> manufacturer, your next and only step is to wait for same to produce a
> new black box.  Nobody else can fix the problem, even if the source of
> that problem lies in the well illuminated kernel.  It's dark inside the
> black box, and only it's manufacturer can flip the light switch.

Unless the black box is fine and it's the well-illuminated kernel
that needs to be patched, n'est pas?  And, yes, I'm aware of how
likely that is to be true. :-)
-- 
May the LORD God bless you exceedingly abundantly!

Dave_Craig______________________________________________
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_________________

Re: linux rt 3.4.8-rt16 and nv 302.17, and system freezes, does nvidia strike again?

[David L. Craig]

On 12Aug24:0748-0400, David L. Craig wrote:

> On 12Aug24:0826+0200, Mike Galbraith wrote:
> 
> > On Thu, 2012-08-23 at 18:50 -0400, David L. Craig wrote:
> > 
> > > I have asked for suggestions
> > > on this list and in the Debian bug report about
> > > next steps but thus far have received no responses.
> > 
> > Once you report a black box (proprietary) driver problem to black box
> > manufacturer, your next and only step is to wait for same to produce a
> > new black box.  Nobody else can fix the problem, even if the source of
> > that problem lies in the well illuminated kernel.  It's dark inside the
> > black box, and only it's manufacturer can flip the light switch.
> 
> Unless the black box is fine and it's the well-illuminated kernel
> that needs to be patched, n'est pas?  And, yes, I'm aware of how
> likely that is to be true. :-)

I'm too tired after a long night attending to many things
other than computing problems, so I should have been more
careful in answering.  My point was nvidia probably cannot
diagnose and fix problems outside their proprietary code
completely without any external help, nor can outside
expertise directly come to bear on problems inside their
proprietary code.  If a fix to the kernel is involved,
non-nvidia people will doubtless be involved.

BTW, the citation is not mine, you missed one level of
attribution in your editing.  Perhaps you also were
having a long night (but mine's LONGER! ;-) ). :-)
-- 
May the LORD God bless you exceedingly abundantly!

Dave_Craig______________________________________________
"So the universe is not quite as you thought it was.
 You'd better rearrange your beliefs, then.
 Because you certainly can't rearrange the universe."
__--from_Nightfall_by_Asimov/Silverberg_________________