Can no longer use my USB mouse with my laptop

[Kevin J. Cummings]

Sam Varshavchik wrote:
> Kevin J. Cummings writes:
> 
>> Sam Varshavchik wrote:
>>> Kevin J. Cummings writes:
>>>
>>>>> Yep and this is a violation of Fedora trademarks if they are
>>>>> calling the
>>>>> OEM installation Fedora. Please let me know with more details if
>>>>> that's
>>>>> the case.
>>>>
>>>> What's the difference between the OEM packaging this kernel-suspend2 vs
>>>> me installing it myself from a 3rd party repository.  Its built from
>>>> the
>>>> official FC6 source files (isn't it?).
>>>
>>> Of course not.  Nobody would take "official FC6 source files",
>>> unmodified, and call them something else, just for the hell of it.
>>
>> I never said they did.  Stop reading in things I didn't say.  I said
>> they added something to the stock Fedora Kernel.
> 
> You have no idea how software engineering works.  You just can't "add"
> something to a very complicated piece of software, with no impact
> whatsoever on all the bits that already exist in there.  If that was the
> case, Microsoft would have no problem adding all the DRM bits to Vista,
> and everything else will still continue to work perfectly.

I have 20 years of professional software engineering experience.  Yes I
do understand how it works.

> So all the people now complaining about how copying files in Vista takes
> ten times longer than it should be, are making the whole things up.

I try not to use Microsoft Windows.  I wish I could convince my family
to do the same.  My latest contract also requires that I use Windows.  I
run it as a VM in Linux for my piece of mind.

> The modified kernel your vendor gave you might include some additional
> jigamaroo for some bizarre USB hardware in your laptop.  When the
> jigamaroo stopped working due to some change in a newer kernel, your
> mouse broke. The newer kernel might, just for the sake of argument, have
> an updated USB interrupt handler API, and that the jigamaroo was not
> updated to the newer API.  The result is that it gets something wrong,
> which results in an the error message getting logged, that you're
> reporting.

Its possible,  I've ruled nothing out.  But I started with a false
assumption, I thought most laptop users were using cubbi_suspend2
kernels.  My bad.  (We all know what happens when we Ass-U-Me).

> Do you know, for a fact, that your vendor did nothing to the USB drivers
> in the kernel?  And by that, I don't mean you calling them up, asking
> them, and they told you "no". They'll just told you whatever you needed
> to hear to get off the phone, because you're costing them money, now. 
> And it may not be the USB interrupt handler code, but some other bit in
> the kernel, dealing with interrupts, or I/O, or DMA.  A lot of moving
> parts must do their jobs right, for even the simplest things, like mice,
> to work correctly.  If your vendor touches any of them, for any reason,
> nobody can give you accurate advice any longer, only wild guesses pulled
> out of their asses.

All I know is that they claim to mirror the mhensler kernels.  I'm still
tracking down more info on this, but I won't get any answers from my OEM
until Monday at the earliest.

Actually, they've been very helpful trying to figure out why my wireless
does not connect reliably, though its pretty stable after I do get it to
connect.  I was on the phone with them for 3 hours the first day, and
another 4 hours with them doing a remote desktop debugging session the
second time.

> So please don't go around lecturing how software engineering work, to
> people who do this for a living.  You have no idea what you're talking
> about.

Quite the contrary, I do know what I'm talking about.  Principle
Software Engineer.  Please don't tell me I don't without knowing what I do.

>>                                                   Thats why the release
>> numbers have similarities!
> 
> Yes, it tells you which kernel release your vendor modified.  Which
> means absolutely nothing.  The only thing that matters is the nature of
> the actual modification.  Just knowing which specific release your
> frankenkernel mutated from isn't going to help.  Only your vendor knows
> what they did, and they are the ones who are now responsible for giving
> you the technical support you need.

Actually, it gives me a starting point to ask if this got broken my the
change from 2.6.19 to 2.6.20.  Further investigation has suggested that
the answer is no, but I still don't know for sure yet.

>>> In that case, if you're having a problem with kernel support for your
>>> hardware, why are you asking about that here?  You should be taking all
>>> these issues to your vendor.  It's their kernel.  Not Fedora's kernel.
>>
>> Prove to me that these problems don't exist in the stock Fedora kernel.
> 
> They don't.  My USB mouse works perfectly.  On a stock Fedora kernel.

Great its nice to know.  Thank-you.

> The actual question you need to be asking is whether these problems
> exist for someone else with a stock Fedora kernel, who is using hardware
> that's similar to yours: same CPU, all the motherboard chipsets are the
> same, everything you see in 'lspci', 'dmidecode', etc.  But you already
> said that you have to run your frankenkernel, in order for everything to
> work on your laptop, so you're unlikely to find anyone else running
> stock Fedora kernel on similar hardware.

I run this kernel because this vendor shipped with my laptop.  I'm
trying to gather information about what this series of kernel
modifications is supposed to provide for me, it will help me understand
why my vendor ships it.  I originally though it was a common laptop
kernel.  I was wrong.

>>  Then I'll happily go away.  I asked a question, the hostility being
>> handed back is very unusual for this forum.
> 
> When you claim to know more about how the kernel works, to people who
> are a lot more knowledgable on this subject matter than you are, and
> then you insist that they're wrong, this is one of the responses you can
> expect.

None of the responses I receive when I wrote that told me anything more
than I was asking the question in the wrong forum.  Other than
contacting my vendor, none (until Axel Thimm responded) had any
information for me that I didn't already know.

>>> The only useful information anyone can give you here is regarding stock
>>> Fedora kernels.  Since you don't use them, none of that information will
>>> be very useful for you.
>>
>> That is an over generalization.  If you have useful information give it.
>> I'll be the judge whether or not its useful to me!
> 
> Sorry, I have no useful information to give you.  Nor does anyone else,
> for that matter, unless they happen to be using the same frankenkernel
> than you are.

Bzzzt, wrong answer, but thank for replying!  People get answers about
non-Fedora provided software here all the time.  I took exception to the
attempt to deny me the same possibility.

> USB mice just don't break, randomly, on stock Fedora kernels.
> 
>>> Only your vendor can give you an accurante answer to that question.
>>> Nobody around here knows what exactly is in the kernel you're
>>> running. I'm sure that a fairly good guess can be made, but this is
>>> just one of
>>> those things where "fairly good" is not good enough.  You have to know
>>> exactly what's in there.
>>
>> And you think my vendor is going to respond to an email from me at
>> midnight on a Friday night?  Get serious.  The support from this group
>> is much better for answering simple questions.  I never said I cared if
>> the answer I got was useful or not.  I just want to know if anyone else
>> has seem this problem or not.  Sheesh!  (Obviously given the hostile
>> attitudes here, noone else has seen it.)  There, how difficult was that
>> to say?
> 
> Well, how many days now you've been looking for answers here?  Are you
> satisfied that nobody around here has any answers for you (and why that
> is), and that you'll have to talk to your vendor?

Nope. I just started and got slammed while leaving the gate.  It wasn't
pleasant, and it was un-expected.

-- 
Kevin J. Cummings
kjchome at rcn.com
cummings at kjchome.homeip.net
cummings at kjc386.framingham.ma.us
Registered Linux User #1232 (http://counter.li.org)