MIME-Version: 1.0
In-Reply-To: <CALYGNiMXhvGOq0P-NG_1jMyN7qp+1+UJ8qihp6wVr0xAKjNHsQ@mail.gmail.com>
References: <1477380887-21333-1-git-send-email-mszeredi@redhat.com>
 <1477380887-21333-4-git-send-email-mszeredi@redhat.com> <20161025115748.ydhkkp5cfcdnjzwn@home.ouaza.com>
 <CAJfpegu97ce6LtMeKSHqDxZGAtCJUGgYUwOwYm9P6p8XtNXf_A@mail.gmail.com>
 <CALYGNiN_Hj+4Kf4HJihu5OEkNFU+zzioz5Cc1oHpJ8Wggi0M8Q@mail.gmail.com>
 <CAJfpegvzsjWBL5nu7SQt2kp7efjGMpb9L2JMxzXPXUTdDnft0Q@mail.gmail.com>
 <CALYGNiOfcuM55458mZoUd4r789z7nVs7Yj0vAA5Qo=BetTJA0Q@mail.gmail.com>
 <CAJfpeguB-RRFbuwy=csPxoByKC=FTxYyuvnm+rmfTUmkVX+K9w@mail.gmail.com> <CALYGNiMXhvGOq0P-NG_1jMyN7qp+1+UJ8qihp6wVr0xAKjNHsQ@mail.gmail.com>
From: Amir Goldstein <amir73il@gmail.com>
Date: Mon, 7 Nov 2016 15:38:19 +0200
Message-ID: <CAOQ4uxgzwh55eM19ebfKv6+F8gyHJTN1QRK6R9qALUGmFd1k0g@mail.gmail.com>
Subject: Re: [PATCH 3/3] ovl: redirect on rename-dir
To: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Miklos Szeredi <miklos@szeredi.hu>,
        Raphael Hertzog <hertzog@debian.org>,
        Miklos Szeredi <mszeredi@redhat.com>,
        "linux-unionfs@vger.kernel.org" <linux-unionfs@vger.kernel.org>,
        Guillem Jover <guillem@debian.org>,
        linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset=UTF-8
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1426
Lines: 31

On Mon, Nov 7, 2016 at 12:08 PM, Konstantin Khlebnikov <koct9i@gmail.com> wrote:
> On Mon, Nov 7, 2016 at 1:04 PM, Miklos Szeredi <miklos@szeredi.hu> wrote:
>> On Mon, Nov 7, 2016 at 10:58 AM, Konstantin Khlebnikov <koct9i@gmail.com> wrote:
>>
>>> I've stumbled on somehow related problem - concurrent copy-ups are
>>> strictly serialized by rename locks.
>>> Obviously, file copying could be done in parallel: locks are required
>>> only for final rename.
>>> Because of that overlay slower that aufs for some workloads.
>>
>> Easy to fix: for each copy up create a separate subdir of "work".
>> Then the contention is only for the time of creating the subdir, which
>> is very short.
>
> Yeah, but lock_rename() also takes per-sb s_vfs_rename_mutex (kludge by Al Viro)
> I think proper synchronization for concurrent copy-up (for example
> round flag on ovl_entry) and  locking rename only for rename could be
> better.

Removing s_vfs_rename_mutex from copy-up path is something I have been
pondering about.
Assuming that I understand Al's comment above vfs_rename() correctly,
the sole purpose of per-sb serialization is to prevent loop creations.
However, how can one create a loop by moving a non-directory?
So it looks like at least for the non-dir copy up case, a much finer grained
lock is in order.

Anyway, it's on my todo list, as concurrent operation performance on overlayfs
is important to out use case.

Amir.