From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ian Kent Subject: Re: [PATCH 0/4] more on vfs-scale and vfs-automount Date: Tue, 01 Mar 2011 13:14:22 +0800 Message-ID: <1298956462.3726.33.camel@perseus> References: <20110301044457.4615.40333.stgit@localhost6.localdomain6> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: Nick Piggin , David Howells , Kernel Mailing List , linux-fsdevel , Linus Torvalds , Andrew Morton To: Al Viro Return-path: In-Reply-To: <20110301044457.4615.40333.stgit@localhost6.localdomain6> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Tue, 2011-03-01 at 12:56 +0800, Ian Kent wrote: > I've put quite a bit of time into testing 2.6.38-rc and, given the > time frame, and update is needed. The included patch series much > improves the behaviour of autofs under load. Sorry Al, I didn't yet test against the namei patches you provided. This was against rc6. > > I first thought the dentry leak you found was not the cause of the > BUG() I was seeing but that appears to not be the case. I'm not > seeing the BUG() at shutdown when umounting any more. > > I am still seeing occassional incorrect ENOENT returns. They must be > comming from the VFS or the daemon as I've changed almost all the > ENOENT returns in the autofs module to identify where it's comming > from. > > Anyway, all, please review. > > --- > > Ian Kent (4): > autofs4 - fix autofs4_expire_indirect() traversal > autofs4 - fix dentry leak in autofs4_expire_direct() > autofs4 - fix rootless multi-mount race > vfs - check non-mountpoint dentry might block in __follow_mount_rcu() > > > fs/autofs4/expire.c | 72 ++++++++++++++++++++++++++++++++++++++++----------- > fs/autofs4/root.c | 20 +++----------- > fs/namei.c | 24 ++++++++++++++--- > 3 files changed, 80 insertions(+), 36 deletions(-) >