Net Llama! wrote on Thursday 31 March 2005 13:07:
> On Thu, 31 Mar 2005, David Kewley wrote:
> > A quick question for anyone who knows: Is 4kstacks still a problem for
> > the xfs in kernel 2.6.11? If not, then at what kernel release did it
> > stop being a problem?
>
> Funny you should ask, cause there was a thread on this just last week (or
> maybe the week before, i forget). The general consensus is that 4k is ok
> for light, low intensity (no NFS, LVM, etc) usage, but for anything high
> volume, where uptime must be 5 9's, 8k is the safer bet.
Now that you pointed me to it (thanks!), I found it:
http://oss.sgi.com/archives/linux-xfs/2005-03/msg00047.html
I think your summary is correct, but I'd add one BIG modification: 4kstacks
only affects i386 (including e.g. i686) arches. For example, the arch I use,
x86_64, still uses 8k stacks, so I only need to add CONFIG_XFS_FS. I've
verified this to my satisfaction by looking at the RHEL4 kernel source, but
also see Andi Kleen's post to the above thread:
http://oss.sgi.com/archives/linux-xfs/2005-03/msg00078.html
-----------------
ONE MORE question or set of questions. :)
So now I've got my RHEL4 kernel source tree, which is based on 2.6.9 and
appears to have no patches to xfs (with the possible exception of patches
outside the xfs/ tree that affect xfs). Would it be advisable to just use
this version, or to try to update the xfs kernel code in some way (replace
the xfs/ tree with the 2.6.11 version? apply some patches?)? I see that the
xfs patches kept coming in for 2.6.10 and 2.6.11.
Thanks,
David
|