Re: backporting 2.6.38 from unstable to squeeze
Hi Ben,
On Sat, 16 Apr 2011, Ben Hutchings wrote:
On Sat, 2011-04-16 at 13:23 +0530, Faheem Mitha wrote:
[...]
Ok. Is the only problem with the lenny 2.6.26 that it has
CONFIG_SYSFS_DEPRECATED enabled? Or are there other problems?
AFAIK that is the only problem.
Good to know. Thanks.
I am seeing some failures in the 2.6.38 installation. See output below.
Specifically, some problems with nvidia and virtualbox. I tried booting
with .38, and it does boot, but compiling the nvidia module fails, and m-a
is not particularly helpful when outputting errors. Should I upgrade the
nvidia stuff to that in unstable? Would the unstable nvidia compile Ok
with 2.6.32? Ie. is it backward compatible?
I don't know. I don't really care about non-free drivers.
Fair enough. I have a Nvidia card GeForce 7300 GS. I'm not aware of free
alternatives to the non-free Nvidia driver. Also, I have two screens, and
I like the dual-screen functionality.
So why not stick with 2.6.32?
Good question. I've having hardware problems with a new IDE DVD writer. I
was hoping I could make that problem go away by switching to another
kernel, but instead I'm just encountering different problems.
I guess I should report a bug about this.
Additionally, I was seeing the following errors thrown on the console.
"No human readable MCE decoding support on this CPU type Run the
message through 'mcelog --ascii' to decode."
That's not an error, the messages before it are errors.
I see. I didn't notice any other messsages. Where would these messages be
logged?
Should I file bugs against any of this?
[...]
MCEs could be triggered by driver or other bugs, though they are usually
hardware problems. So, do report a bug, but there may be nothing we can
do.
Ok. Thanks.
Faheem
Reply to: