Re: NFS client (10x) performance regression 2.6.14.7 -> 2.6.15

[Jakob Oestergaard <jakob@unthought.net>]

On Fri, Mar 31, 2006 at 09:49:51AM -0500, Bill Rugolsky Jr. wrote:
...
> Jakob,
> 
> Your NFS setup is specific to your system. 

True. But it is my impression that this is a problem isolated on the
client side (am I wrong?)

> Have you considered trying this
> over loopback to narrow down the variables?

Do you mean NFS exporting a local filesystem, NFS mounting it again on
the local host?   Or do you mean something with loopback mounts?

(Sorry if that's a stupid question, but I just need to know what you
mean :)

> If you see similar getattr/write
> behavior over loopback, it will make it easy for everyone else to test.

True.

I've currently tried compiling the kernel with three different gcc
versions to see if that made any change. It didn't. And I tried SMP
versus UP, again no difference. I'm just about to try HIGHMEM vs no
HIGHMEM...

Thanks for your suggestion - I will try it, if you tell me precisely
what you mean in the above  :)

-- 

 / jakob

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/