Barry Brimer <lists@xxxxxxxxxx> wote:
Does anyone have any idea why the un-patched kernel from kernel.org works
> fine with ndiswrapper but both the FC6 and CentOS (RHEL) kernels lock up or
> spontaneously reboot? I built the 2.6.21 kernel by copying the config file
> for 2.6.18-8.1.1 from /boot and doing a "make oldconfig". Everything is
> x86_64.
<http://ndiswrapper.sourceforge.net/joomla/index.php?/component/option,com_openwiki/Itemid,33/id,faq/>
See Item #2.
Summary:
Red Hat/Fedora uses a 4k stack size. Doesn't work well with ndiswrapper,
or rather with the drivers you are trying to use with ndiswrapper.
Barry
That seems to be the stock response but it doesn't seem to apply to an
x86_64 kernel. 32 bit (e.g., i686) kernels have:
CONFIG_4KSTACKS=y
in the kernel config file. The equivalent beasty doesn't exist in the
x86_64 config file. Also, I built my 2.6.21 kernel by using the Red
Hat/CentOS 2.6.18-8.1.1 kernel configuration file and did a "make
oldconfig" which just prompts for how to handle changes in the kernel
config from the old config file version. Nothing came up about the
stack size (shouldn't have the way "make oldconfig" works) so I should
have the same options set for my kernel. It works but the RH/CentOS
kernel doesn't.
I'd prefer to stay with the standard kernel if possible but I need to be
able to "fix" whatever is causing the reboot. If it's as simple as
turning off some option, I can automate that. I'm just hoping it's not
some patch that applies a 4K stack limit without even having the ability
to turn it off.
Cheers,
Dave
--
Politics, n. Strife of interests masquerading as a contest of principles.
-- Ambrose Bierce
_______________________________________________
CentOS mailing list
CentOS@xxxxxxxxxx
http://lists.centos.org/mailman/listinfo/centos
