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

["Bill Rugolsky Jr." <brugolsky@telemetry-investments.com>]

On Fri, Mar 31, 2006 at 04:35:00PM +0200, Jakob Oestergaard wrote:
> I'm wondering... Can anyone else reproduce this problem?
> 
> Just to explain quickly:
>  Running nfsbench (on the NFS client) once with LEADING_EMPTY_SPACE set
>  to 0 and then once with the option set to 1.  If there's a big change
>  in wall-clock execution time, this indicates that the problem exists.
> 
> I'd be really interested in knowing whether I'm the only one who sees
> this problem.
 
Jakob,

Your NFS setup is specific to your system.  Have you considered trying this
over loopback to narrow down the variables?  If you see similar getattr/write
behavior over loopback, it will make it easy for everyone else to test.

	-Bill
-
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/