Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 24 Jan 2007 00:59:42 +1100 (EST)
From:      Bruce Evans <bde@zeta.org.au>
To:        Kostik Belousov <kostikbel@gmail.com>
Cc:        fs@FreeBSD.org
Subject:   Re: more lost dotdot caching pessimizes nfs especially
Message-ID:  <20070124005135.J14386@besplex.bde.org>
In-Reply-To: <20070123133921.GL71333@deviant.kiev.zoral.com.ua>
References:  <20070123233124.R33734@delplex.bde.org> <20070123133921.GL71333@deviant.kiev.zoral.com.ua>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 23 Jan 2007, Kostik Belousov wrote:

> On Wed, Jan 24, 2007 at 12:18:38AM +1100, Bruce Evans wrote:
>> I suspect that recent locking fixes near dotdot are responsible for
>> slowing down builds over nfs even further by increasing the number
>> of RPCs.  Times and RPC counts for building a RELENG_4 kernel under
>> -current on a Turion X2 2GHz:

Nevermind, it was an editing error that lost all the non-hack part of
my RPC reductions.  This lost about half of them.

> could you confirm that reversal of rev. 1.97 of vfs_lookup.c would restore
> the old Lookup/Access RPC statistic ?

Thanks for the quick reply.  I tried that first.  It had no effect.

While you are here I'll ask for a non-hack to restore caching of dotdot :-).
(I just back out rev.1.103 of vfs_cache.c).

Bruce



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070124005135.J14386>