[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: New testing kernel. 2.6.10-1.727_FC3



Dave Jones wrote:
On Fri, Jan 07, 2005 at 09:34:02AM +0000, Andy Green wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> > On Friday 07 January 2005 02:02, Dave Jones wrote:
> > On Thu, Jan 06, 2005 at 02:16:19PM +0000, Andy Green wrote:
> > > Just in case it is the same problem -- I am noting strange pauses in the
> > > current devel kernel 2.6.10-1.1056_FC4 periodically, interrupting video
> > > playback and audio for a second or two at intervals, even when the
> > > machine is fairly quietscant.
> >
> > There's a lot of heavyweight debug code in the rawhide kernels right now
> > which could account for this. It's the usual stuff that gets turned off
> > just before release (checking for reuse of free'd memory etc..)
> > Thanks, that can easily explain it if something allocated and freed a big > chunk of memory and then it is checked for being touched. I'll wait and see > if it goes away when a stable 2.6.10 kernel turns up then.
As a heads-up, todays rawhide has an extra debugging check that
checks those memory allocations periodically. This should catch
stuff that stomps past the end of its allocation, but never
free's it. Hopefully it won't turn up anything new, but it's
worth having it on for now just for a look-see..


Dave


On my laptop:
I'm not having troubles with the 727 kernel. My DVD is working, the box powers off properly, X is alright if I don't change to a virtual terminal. I did try a rawhide kernel for the first 2.6.10 kernel that I saw in rawhide. It needed to pull in newer kernel related packages from rawhide, in addition to the kernel.
Is it possible that some of the oops' that are hapnening are related to the 2.6.10 krnel not requiring the newer kernel related packages? I don't know because I have the kernel related packages installed and then installed kernel-2.6.10-1.727_FC3 afterwards.


On my desktop:
The new kernel in rawhide shows the debugging code, then oops' during further booting. This is on a rawhide system that is fully updated. I just used the previous rawhide kernel and did not investigate the oops.


Jim


[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]