rt2870 (staging kernel driver) crashes fatally

rt2870 (staging kernel driver) crashes fatally

[Daniele C.]

[-- Attachment #1: Type: text/plain, Size: 722 bytes --]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello,

I am experiencing the documented crash after intense usage of the wi-fi device.

Best regards,
- --
  Daniele C.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJLjiDOAAoJEON28As9epPAacQH/igJUCDKGDLzTnZhllzpR6yn
jl7BusfXbpxwBUT6SYB4UQ8RobapmJCGpmcYMlbFTOnXRTK2WmiPrcbORqSBgZJV
MZM8kvNoF9fJOg4Yr7mvwaMAaDBhYTPJFaCDcbHeLL06Sg1XLkiCaWnAQNk7IyVp
cbVa031vh3lDrRJi7Xbcb2FLU7gxaKU9jc3dQcLWuBzu+Yte+Wn0ymbsVFDKFxQi
3ZObIaIVS3AQDS1seSielPfW3igP9blB05+S6XagGMsrolEE+yXG1SmAmk8+v7dl
XHIOQCE51yPwjFQ5ExcM3slCPBUEczEupEj/Urngck8FXnJ8jIHD/AA8z6rIhEI=
=Kees
-----END PGP SIGNATURE-----

[-- Attachment #2: rt2870crash.zip --]
[-- Type: application/zip, Size: 47283 bytes --]

Re: rt2870 (staging kernel driver) crashes fatally

[Hin-Tak Leung]

You appear to have two issues - (1) the driver doesn't
hibernate/resume well, (2) there is some bad memory leak which causes
your system to run out of memory.

You did not mention issue (1) - that you were hibernating and resuming
(and expecting it to work, which it does not). For the latter, you
probably need to track down where the memory is leaking in the driver,
which is probably easier than tracking down the hibernate/resume
issue.

Good luck with tracking down the memory leak. (you do understand that
staging drivers are DIY - i.e. you try it only if you are willing to
look into fixing it yourself - I hope )

On Wed, Mar 3, 2010 at 8:41 AM, Daniele C.
<legolas558@users.sourceforge.net> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hello,
>
> I am experiencing the documented crash after intense usage of the wi-fi device.
>
> Best regards,
> - --
>  Daniele C.
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.10 (GNU/Linux)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
>
> iQEcBAEBAgAGBQJLjiDOAAoJEON28As9epPAacQH/igJUCDKGDLzTnZhllzpR6yn
> jl7BusfXbpxwBUT6SYB4UQ8RobapmJCGpmcYMlbFTOnXRTK2WmiPrcbORqSBgZJV
> MZM8kvNoF9fJOg4Yr7mvwaMAaDBhYTPJFaCDcbHeLL06Sg1XLkiCaWnAQNk7IyVp
> cbVa031vh3lDrRJi7Xbcb2FLU7gxaKU9jc3dQcLWuBzu+Yte+Wn0ymbsVFDKFxQi
> 3ZObIaIVS3AQDS1seSielPfW3igP9blB05+S6XagGMsrolEE+yXG1SmAmk8+v7dl
> XHIOQCE51yPwjFQ5ExcM3slCPBUEczEupEj/Urngck8FXnJ8jIHD/AA8z6rIhEI=
> =Kees
> -----END PGP SIGNATURE-----
>

Re: rt2870 (staging kernel driver) crashes fatally

[Larry Finger]

On 03/03/2010 08:14 PM, Hin-Tak Leung wrote:
> You appear to have two issues - (1) the driver doesn't
> hibernate/resume well, (2) there is some bad memory leak which causes
> your system to run out of memory.
> 
> You did not mention issue (1) - that you were hibernating and resuming
> (and expecting it to work, which it does not). For the latter, you
> probably need to track down where the memory is leaking in the driver,
> which is probably easier than tracking down the hibernate/resume
> issue.
> 
> Good luck with tracking down the memory leak. (you do understand that
> staging drivers are DIY - i.e. you try it only if you are willing to
> look into fixing it yourself - I hope )

By generating your own kernel with MEMLEAK enabled, you should be able to see
which data blob is being leaked. If you post that info, someone on this list
should be able to help you find the cause.

Larry

Re: rt2870 (staging kernel driver) crashes fatally

[Daniele C.]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

- -------- Original Message  --------
Subject: Re: rt2870 (staging kernel driver) crashes fatally
From: Larry Finger <Larry.Finger@lwfinger.net>
To: Daniele C. <legolas558@users.sourceforge.net>
Date: Thu Mar 04 2010 03:35:43 GMT+0100 (CET)

> On 03/03/2010 08:14 PM, Hin-Tak Leung wrote:
>> You appear to have two issues - (1) the driver doesn't
>> hibernate/resume well, (2) there is some bad memory leak which causes
>> your system to run out of memory.
>>
>> You did not mention issue (1) - that you were hibernating and resuming
>> (and expecting it to work, which it does not). For the latter, you
>> probably need to track down where the memory is leaking in the driver,
>> which is probably easier than tracking down the hibernate/resume
>> issue.
>>
>> Good luck with tracking down the memory leak. (you do understand that
>> staging drivers are DIY - i.e. you try it only if you are willing to
>> look into fixing it yourself - I hope )
> 
> By generating your own kernel with MEMLEAK enabled, you should be able to see
> which data blob is being leaked. If you post that info, someone on this list
> should be able to help you find the cause.
> 
I will enable this feature so that next time it happens I will have such information.
However looks like bug is not triggering unless I first do a resume.

Thank you both for reviewing my logs and for the pointers.

Best regards,
- --
  Daniele C.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJLj48QAAoJEON28As9epPAqfIH/AktbatFnxAt6peijPxz64PV
jElSTuDYaieRRjKfoTQCVZoiSsf4f4sPAiqBR4E30zX7u9nWqqbaTIxLJjeC74Nr
fqfoFS2QBg8Ow5pLhs/pXUKR4c7g/x/QIQ7fzCAOUj/enirni2f6n9Gc8Tf8O1Mo
ejKXqqKeKPsnlNoBXtnB3ihE5sVp7O2PJZvxV3GCRy5KEmoIivPkrd3U4Lsc2pZG
86LOolgilto6I73/bdoU0CffiKyTENwpyJTMy1VFjqwOWLLn9ukuFbfyBd00Oo3W
uYX/SSCnouAKeAG0LzzTZMaYKOGkqJQzVLv/B0nhlcrrlk+xzWBNe04fxHamVo8=
=8q9G
-----END PGP SIGNATURE-----

Re: rt2870 (staging kernel driver) crashes fatally

[Ivo van Doorn]

On Thursday 04 March 2010, Larry Finger wrote:
> On 03/03/2010 08:14 PM, Hin-Tak Leung wrote:
> > You appear to have two issues - (1) the driver doesn't
> > hibernate/resume well, (2) there is some bad memory leak which causes
> > your system to run out of memory.
> > 
> > You did not mention issue (1) - that you were hibernating and resuming
> > (and expecting it to work, which it does not). For the latter, you
> > probably need to track down where the memory is leaking in the driver,
> > which is probably easier than tracking down the hibernate/resume
> > issue.
> > 
> > Good luck with tracking down the memory leak. (you do understand that
> > staging drivers are DIY - i.e. you try it only if you are willing to
> > look into fixing it yourself - I hope )
> 
> By generating your own kernel with MEMLEAK enabled, you should be able to see
> which data blob is being leaked. If you post that info, someone on this list
> should be able to help you find the cause.

Actually the rt2x00-users list is not intended for the Ralink staging drivers.
You have to use the staging mailing list: devel@driverdev.osuosl.org

Ivo

Re: rt2870 (staging kernel driver) crashes fatally

[Daniele C.]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

- -------- Original Message  --------
Subject: Re: rt2870 (staging kernel driver) crashes fatally
From: Ivo van Doorn <ivdoorn@gmail.com>
To: Larry Finger <Larry.Finger@lwfinger.net>
Date: Fri Mar 05 2010 17:22:41 GMT+0100 (CET)

> On Thursday 04 March 2010, Larry Finger wrote:
>> On 03/03/2010 08:14 PM, Hin-Tak Leung wrote:
>>> You appear to have two issues - (1) the driver doesn't
>>> hibernate/resume well, (2) there is some bad memory leak which causes
>>> your system to run out of memory.
>>>
>>> You did not mention issue (1) - that you were hibernating and resuming
>>> (and expecting it to work, which it does not). For the latter, you
>>> probably need to track down where the memory is leaking in the driver,
>>> which is probably easier than tracking down the hibernate/resume
>>> issue.
>>>
>>> Good luck with tracking down the memory leak. (you do understand that
>>> staging drivers are DIY - i.e. you try it only if you are willing to
>>> look into fixing it yourself - I hope )
>>
>> By generating your own kernel with MEMLEAK enabled, you should be able to see
>> which data blob is being leaked. If you post that info, someone on this list
>> should be able to help you find the cause.
> 
> Actually the rt2x00-users list is not intended for the Ralink staging drivers.
> You have to use the staging mailing list: devel@driverdev.osuosl.org
> 
Yes, they refused my message but didn't tell anything about the staging mailing list which I couldn't find either and I am now reading from you. Many thanks for pointing this out.

I am now running with MEMLEAK enabled, as soon as I will make a new "hammering session" I'll grab the debug data and post another message.

Best regards,
- --
  Daniele

> Ivo
> 

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJLkWX+AAoJEON28As9epPAI9AH/j6nspsStdgsO4AigPVlQRrb
tc7VOTmv+EwuydntCERrlFGx20sNYfKcBtyt7WHg3TWQnMr7ISgCK4fB+YDvWExb
gS4+4n7u6AgDMPkrL3tr4PBvCFQJksGPh8epaB9InRv8JnBTmhwv287Ng7PiXY6h
P0YYmcxIgQ2F9BPA5PHJ3+Y2PqIQl+2cur95rp3Q7EUESwyHSeDqQnC1l9k50FsM
M6tGyTMBYUZDHEncdu6wpDipfnTQ9gUD0j4XUlfmRhVmedJ4DTknySMOup1dgNPz
Y1UMYhyvjkibeAEcep2KAShMwEfapyUf01gaGMsRxecvexQwC1YsekGJ1/pbCBs=
=6LbC
-----END PGP SIGNATURE-----

Re: rt2870 (staging kernel driver) crashes fatally

[Ivo van Doorn]

On Friday 05 March 2010, Daniele C. wrote:
> -------- Original Message  --------
> Subject: Re: rt2870 (staging kernel driver) crashes fatally
> From: Ivo van Doorn <ivdoorn@gmail.com>
> To: Larry Finger <Larry.Finger@lwfinger.net>
> Date: Fri Mar 05 2010 17:22:41 GMT+0100 (CET)
> 
> > On Thursday 04 March 2010, Larry Finger wrote:
> >> On 03/03/2010 08:14 PM, Hin-Tak Leung wrote:
> >>> You appear to have two issues - (1) the driver doesn't
> >>> hibernate/resume well, (2) there is some bad memory leak which causes
> >>> your system to run out of memory.
> >>>
> >>> You did not mention issue (1) - that you were hibernating and resuming
> >>> (and expecting it to work, which it does not). For the latter, you
> >>> probably need to track down where the memory is leaking in the driver,
> >>> which is probably easier than tracking down the hibernate/resume
> >>> issue.
> >>>
> >>> Good luck with tracking down the memory leak. (you do understand that
> >>> staging drivers are DIY - i.e. you try it only if you are willing to
> >>> look into fixing it yourself - I hope )
> >>
> >> By generating your own kernel with MEMLEAK enabled, you should be able to see
> >> which data blob is being leaked. If you post that info, someone on this list
> >> should be able to help you find the cause.
> >
> > Actually the rt2x00-users list is not intended for the Ralink staging drivers.
> > You have to use the staging mailing list: devel@driverdev.osuosl.org
> >
> Yes, they refused my message but didn't tell anything about the staging mailing list which I couldn't find either and I am now reading from you. Many thanks for pointing this out.

Your mail wasn't refused for the rt2x00-users mailinglist, we have a filter that prevents non-subscribed members from sending mails to it.
Luis and myself are administrating the list, so your mails were approoved, but we forgot to press the button to allow all your mails
to the list to be accepted. :)

> I am now running with MEMLEAK enabled, as soon as I will make a new "hammering session" I'll grab the debug data and post another message.

Good luck. :)

Ivo

Re: rt2870 (staging kernel driver) crashes fatally

[Luis Correia]

Well,

to set the record straight I did reply privately to tell that staging
isn't our business.

But my fault was not to point out the correct mailing list for staging.

I hope that my apology is accepted.
Luis Correia

On Fri, Mar 5, 2010 at 20:21, Ivo van Doorn <ivdoorn@gmail.com> wrote:
> On Friday 05 March 2010, Daniele C. wrote:
>> -------- Original Message  --------
>> Subject: Re: rt2870 (staging kernel driver) crashes fatally
>> From: Ivo van Doorn <ivdoorn@gmail.com>
>> To: Larry Finger <Larry.Finger@lwfinger.net>
>> Date: Fri Mar 05 2010 17:22:41 GMT+0100 (CET)
>>
>> > On Thursday 04 March 2010, Larry Finger wrote:
>> >> On 03/03/2010 08:14 PM, Hin-Tak Leung wrote:
>> >>> You appear to have two issues - (1) the driver doesn't
>> >>> hibernate/resume well, (2) there is some bad memory leak which causes
>> >>> your system to run out of memory.
>> >>>
>> >>> You did not mention issue (1) - that you were hibernating and resuming
>> >>> (and expecting it to work, which it does not). For the latter, you
>> >>> probably need to track down where the memory is leaking in the driver,
>> >>> which is probably easier than tracking down the hibernate/resume
>> >>> issue.
>> >>>
>> >>> Good luck with tracking down the memory leak. (you do understand that
>> >>> staging drivers are DIY - i.e. you try it only if you are willing to
>> >>> look into fixing it yourself - I hope )
>> >>
>> >> By generating your own kernel with MEMLEAK enabled, you should be able to see
>> >> which data blob is being leaked. If you post that info, someone on this list
>> >> should be able to help you find the cause.
>> >
>> > Actually the rt2x00-users list is not intended for the Ralink staging drivers.
>> > You have to use the staging mailing list: devel@driverdev.osuosl.org
>> >
>> Yes, they refused my message but didn't tell anything about the staging mailing list which I couldn't find either and I am now reading from you. Many thanks for pointing this out.
>
> Your mail wasn't refused for the rt2x00-users mailinglist, we have a filter that prevents non-subscribed members from sending mails to it.
> Luis and myself are administrating the list, so your mails were approoved, but we forgot to press the button to allow all your mails
> to the list to be accepted. :)
>
>> I am now running with MEMLEAK enabled, as soon as I will make a new "hammering session" I'll grab the debug data and post another message.
>
> Good luck. :)
>
> Ivo
> --
> To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>

Re: rt2870 (staging kernel driver) crashes fatally

[Daniele C.]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

- -------- Original Message  --------
Subject: Re: rt2870 (staging kernel driver) crashes fatally
From: Luis Correia <luis.f.correia@gmail.com>
To: Ivo van Doorn <ivdoorn@gmail.com>
Date: Fri Mar 05 2010 21:24:06 GMT+0100 (CET)

> Well,
> 
> to set the record straight I did reply privately to tell that staging
> isn't our business.
> 
> But my fault was not to point out the correct mailing list for staging.
> 
> I hope that my apology is accepted.
Of course it is Luis, perhaps I have been a bit irrational in my reply but this is what happened:

* website didn't give a bugtracker URL
* SF.net project was seemingly dead, or more probably I didn't find the new one
* I registered to forums and my registration required manual activation by some admin (a growing climax...)
* I finally wrote to the apparently correct mailing list (I was sure to have read somewhere that it offered support also for the staging driver!), and my message was refused for exceeding size and for not being on topic

At that point, even not being exactly a newbie, I was lost and a bit frustrated. Perhaps you can review the information flow to allow people reach more easily the interesting areas, or perhaps it was just a bad day for me.

Thanks for the correctness of all of you which have replied.

Best regards,
- --
  Daniele

> Luis Correia
> 
> On Fri, Mar 5, 2010 at 20:21, Ivo van Doorn <ivdoorn@gmail.com> wrote:
>> On Friday 05 March 2010, Daniele C. wrote:
>>> -------- Original Message  --------
>>> Subject: Re: rt2870 (staging kernel driver) crashes fatally
>>> From: Ivo van Doorn <ivdoorn@gmail.com>
>>> To: Larry Finger <Larry.Finger@lwfinger.net>
>>> Date: Fri Mar 05 2010 17:22:41 GMT+0100 (CET)
>>>
>>>> On Thursday 04 March 2010, Larry Finger wrote:
>>>>> On 03/03/2010 08:14 PM, Hin-Tak Leung wrote:
>>>>>> You appear to have two issues - (1) the driver doesn't
>>>>>> hibernate/resume well, (2) there is some bad memory leak which causes
>>>>>> your system to run out of memory.
>>>>>>
>>>>>> You did not mention issue (1) - that you were hibernating and resuming
>>>>>> (and expecting it to work, which it does not). For the latter, you
>>>>>> probably need to track down where the memory is leaking in the driver,
>>>>>> which is probably easier than tracking down the hibernate/resume
>>>>>> issue.
>>>>>>
>>>>>> Good luck with tracking down the memory leak. (you do understand that
>>>>>> staging drivers are DIY - i.e. you try it only if you are willing to
>>>>>> look into fixing it yourself - I hope )
>>>>>
>>>>> By generating your own kernel with MEMLEAK enabled, you should be able to see
>>>>> which data blob is being leaked. If you post that info, someone on this list
>>>>> should be able to help you find the cause.
>>>>
>>>> Actually the rt2x00-users list is not intended for the Ralink staging drivers.
>>>> You have to use the staging mailing list: devel@driverdev.osuosl.org
>>>>
>>> Yes, they refused my message but didn't tell anything about the staging mailing list which I couldn't find either and I am now reading from you. Many thanks for pointing this out.
>>
>> Your mail wasn't refused for the rt2x00-users mailinglist, we have a filter that prevents non-subscribed members from sending mails to it.
>> Luis and myself are administrating the list, so your mails were approoved, but we forgot to press the button to allow all your mails
>> to the list to be accepted. :)
>>
>>> I am now running with MEMLEAK enabled, as soon as I will make a new "hammering session" I'll grab the debug data and post another message.
>>
>> Good luck. :)
>>
>> Ivo
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>
> 

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJLkWpAAAoJEON28As9epPAQq4H/RBsF7+r9EvmESHV8N8wlC+c
Sgx/M8fT1vlio7svHTfbP7YQfThOgRl/ox9mrEkdNS8QwpQN0g+D9dH8m1xYkFYO
r1MUXTVGoWHVRP4eyr0+VlFQ1XBM58ig/AojCzAf6bNx8yRIJPRimz2NcoGZT6e5
Utk9HDpBn8cT5Q7cbpM79QGIect7luGYkdhqi62q/IPyLE4MLKSoxSFVYHlbrW08
b2MtqnyWQNc7aNVWfT+ZkCkhmRClWKQSYKaahugnsgYk1fTaFLmKuAKMSqZW9Tmp
X8O9vjx30fT+ms5L/SbgmWNMa9FI4j2obrohIrLWVBjpyTYseWAQnzbue7ybfrs=
=qMs7
-----END PGP SIGNATURE-----

Re: rt2870 (staging kernel driver) crashes fatally

[Daniele C.]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

- -------- Original Message  --------
Subject: Re: rt2870 (staging kernel driver) crashes fatally
From: Hin-Tak Leung <hintak.leung@gmail.com>
To: Daniele C. <legolas558@users.sourceforge.net>
Date: Thu Mar 04 2010 03:14:38 GMT+0100 (CET)

> You appear to have two issues - (1) the driver doesn't
> hibernate/resume well, (2) there is some bad memory leak which causes
> your system to run out of memory.
> 
> You did not mention issue (1) - that you were hibernating and resuming
> (and expecting it to work, which it does not). For the latter, you
> probably need to track down where the memory is leaking in the driver,
> which is probably easier than tracking down the hibernate/resume
> issue.
> 
Sorry I thought that hibernation/resume was supported.

> Good luck with tracking down the memory leak. (you do understand that
> staging drivers are DIY - i.e. you try it only if you are willing to
> look into fixing it yourself - I hope )
> 
I wish I were that good, but I'll try.

Best regards,
- --
  Daniele

> On Wed, Mar 3, 2010 at 8:41 AM, Daniele C.
> <legolas558@users.sourceforge.net> wrote:
> Hello,
> 
> I am experiencing the documented crash after intense usage of the wi-fi device.
> 
> Best regards,
>>

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJLj4e2AAoJEON28As9epPA/+4H/jDpF8Y8FIzJ9UlDOD4yuVNb
yeRgoJW4h1fnL5atZ66Ku+5EehOsKYQEjQDa3Apwl+g1RXFPenbIm4+A/mJQ4INu
7/wIgQIGRh9mPLmRw1+ylMExBbGSm/v4yp4oUj1ZygLGbiBgV2iQMxbmYPQoCdNt
Rf6lvj20Sz0cfVZWvAw+9HfpMRHYYzQ2UgJkdl4iMaJyNIHhAHCgtMsC/00SGsCn
Wi9iQDtw4O5BbhGPx99J4OipBg0LfAihsOdzgxfv8/jNB9z6OwS4hudmwyHc4PHF
B+ZyEnEJfD3CL8qyaknHZXC18QV0NfnrM1P0NBU/E+WTlFuooywBOG+EWDRmF9E=
=NLMu
-----END PGP SIGNATURE-----