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

[Jakob Oestergaard]

I've run a git bisect from 2.6.14 to 2.6.15 to find what broke the NFS
client.

It seems to be the GIT patch: 24aa1fe6779eaddb3e0b1b802585dcf6faf9cc44
that introduces the problem.

Trond, it took me a lot of tries with GIT to narrow this down, because
the problem does not show up consistently. Some times nfsbench would
complete very quickly, but then a second (or third or ...) run would run
slow.

Could I ask you to try: 
 for i in `seq 1 100`; do time ./nfsbench; done
or something like that?

I'm looking through the patch, but to be honest the NFS cache internals
isn't really something I understand... :)

-- 

 / jakob

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