FS#15151 - [kernel26] kernel panic on 2.6.30 with rtl8187se

Attached to Project: Arch Linux
Opened by Vasiliy V. (thriller) - Wednesday, 17 June 2009, 17:09 GMT
Last edited by Tobias Powalowski (tpowa) - Friday, 31 July 2009, 20:19 GMT
Task Type Bug Report
Category Kernel
Status Closed
Assigned To Tobias Powalowski (tpowa)
Thomas Bächler (brain0)
Architecture i686
Severity Medium
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 0
Private No

Details

Description:

When wicd is trying to connect to the wireless network with rtl8187se the kernel panics with “not syncing: stack-protector: Kernel stack is corrupted in: c033ca3d”. This panic was never happen on 2.6.2x kernels.

Additional info:
* kernel26 2.6.30-4 from testing
* wicd-testing 1.6.0b3-1 from AUR
* according to lspci, wireless adapter is
02:00.0 Network controller: Realtek Semiconductor Co., Ltd. RTL8187SE Wireless LAN Controller (rev 22)
(which can be found on MSI Wind U90x subnotebook)

Steps to reproduce:
* Start Arch with disabled wireless adapter
* Enable adapter
* Open wicd cliend (or start wicd-curses) and choose any wireless network
* Observe the panic
This task depends upon

Closed by  Tobias Powalowski (tpowa)
Friday, 31 July 2009, 20:19 GMT
Reason for closing:  Fixed
Additional comments about closing:  2.6.30.4-1
Comment by Vasiliy V. (thriller) - Wednesday, 17 June 2009, 17:14 GMT
Owh, i've missed. This must be in “Packages: Testing”.
Comment by Tobias Powalowski (tpowa) - Thursday, 18 June 2009, 19:04 GMT
this is a staging driver, it may work but don't blame us if id doesn't work it's kernel playground.
Please fill a upstream bug report.
Comment by tuxce (tuxce) - Tuesday, 07 July 2009, 23:42 GMT
  • Field changed: Percent Complete (100% → 0%)
I found a working patch:
http://patchwork.kernel.org/patch/31559/
and I made an AUR package (to test) http://aur.archlinux.org/packages.php?ID=28053

But, there is no upstream bug report, I don't have enough information to open one.
Comment by Bogdan Szczurek (thebodzio) - Saturday, 11 July 2009, 01:03 GMT
@tuxce: I don't think that opening mainstream bug is necessary since there's already a patch for it. I believe we should wait for patched vanilla kernel release and meantime downgrade to 2.6.2x.

*sigh* Looking at the patch this bug is both trivial and extremely inconvenient at the same time (especially for “winds” users as myself).
Comment by tuxce (tuxce) - Saturday, 11 July 2009, 13:00 GMT
" and meantime downgrade to 2.6.2x."
no need to downgrade, just use rtl8187se-patched, it works for at least 3 wind's users
Comment by Bogdan Szczurek (thebodzio) - Saturday, 11 July 2009, 19:27 GMT
Yeah, I meant “downgrade” if one's using stock kernel ;).
Comment by Thomas Bächler (brain0) - Saturday, 11 July 2009, 19:47 GMT
That patch makes no changes outside staging/, so we can apply it to the next kernel26 package.
Comment by Bogdan Szczurek (thebodzio) - Saturday, 11 July 2009, 19:51 GMT
That would be great! Thanks! :)
Comment by Gerardo Exequiel Pozzi (djgera) - Friday, 31 July 2009, 03:34 GMT
Try 2.6.30.4 when will be available in repo. There is a fix for 8187 but maybe not related with this.

Dan Aloni (1):
Staging: prevent rtl8187se from crashing dev_ioctl() in SIOCGIWNAME
Comment by Bogdan Szczurek (thebodzio) - Friday, 31 July 2009, 08:01 GMT
Will do! Thanks!
Comment by tuxce (tuxce) - Friday, 31 July 2009, 20:17 GMT
It works :)

Loading...