All of lore.kernel.org
 help / color / mirror / Atom feed
From: "J. Bruce Fields" <bfields-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>
To: Anand Avati <anand.avati-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Cc: Theodore Ts'o <tytso-3s7WtUTddSA@public.gmane.org>,
	Bernd Schubert
	<bernd.schubert-mPn0NPGs4xGatNDF+KUbs4QuADTiUCJX@public.gmane.org>,
	sandeen-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org,
	linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	linux-ext4-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	gluster-devel-qX2TKyscuCcdnm+yROfE0A@public.gmane.org
Subject: Re: [Gluster-devel] regressions due to 64-bit ext4 directory cookies
Date: Wed, 13 Feb 2013 18:05:11 -0500	[thread overview]
Message-ID: <20130213230511.GW14195@fieldses.org> (raw)
In-Reply-To: <CAFboF2z1akN_edrY_fT915xfehfHGioA2M=PSHv0Fp3rD-5v5A-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>

On Wed, Feb 13, 2013 at 02:57:13PM -0800, Anand Avati wrote:
> On Wed, Feb 13, 2013 at 2:47 PM, Theodore Ts'o <tytso-3s7WtUTddSA@public.gmane.org> wrote:
> 
> > On Wed, Feb 13, 2013 at 05:41:41PM -0500, J. Bruce Fields wrote:
> > > > What if we have an ioctl or a process personality flag where a broken
> > > > application can tell the file system "I'm broken, please give me a
> > > > degraded telldir/seekdir cookie"?  That way we don't penalize programs
> > > > that are doing the right thing, while providing some accomodation for
> > > > programs who are abusing the telldir cookie.
> > >
> > > Yeah, if there's a simple way to do that, maybe it would be worth it.
> >
> > Doing this as an ioctl which gets called right after opendir, i.e
> > (ignoring error checking):
> >
> >       DIR *dir = opendir("/foo/bar/baz");
> >       ioctl(dirfd(dir), EXT4_IOC_DEGRADED_READDIR, 1);
> >       ...
> >
> > should be quite easy.  It would be a very ext3/4 specific thing,
> > though.
> 
> 
> That would work, even though it would be ext3/4 specific. What is the
> recommended programmatic way to detect if the file is on ext3/4 -- we would
> not want to attempt that blindly on a non-ext3/4 FS as the numerical value
> of EXT4_IOC_DEGRADED_READDIR might get interpreted in dangerous ways?

We must have been through this before, but: is the only way to generate
a collision-free readdir cookie really to use a larger hash?

Would it be possible to make something work like, for example, a 31-bit
hash plus an offset into a hash bucket?

I have trouble thinking about this, partly because I can't remember
where to find the requirements for readdir on concurrently modified
directories....

--b.
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

WARNING: multiple messages have this Message-ID (diff)
From: "J. Bruce Fields" <bfields@fieldses.org>
To: Anand Avati <anand.avati@gmail.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>,
	Bernd Schubert <bernd.schubert@itwm.fraunhofer.de>,
	sandeen@redhat.com, linux-nfs@vger.kernel.org,
	linux-ext4@vger.kernel.org, gluster-devel@nongnu.org
Subject: Re: [Gluster-devel] regressions due to 64-bit ext4 directory cookies
Date: Wed, 13 Feb 2013 18:05:11 -0500	[thread overview]
Message-ID: <20130213230511.GW14195@fieldses.org> (raw)
In-Reply-To: <CAFboF2z1akN_edrY_fT915xfehfHGioA2M=PSHv0Fp3rD-5v5A@mail.gmail.com>

On Wed, Feb 13, 2013 at 02:57:13PM -0800, Anand Avati wrote:
> On Wed, Feb 13, 2013 at 2:47 PM, Theodore Ts'o <tytso@mit.edu> wrote:
> 
> > On Wed, Feb 13, 2013 at 05:41:41PM -0500, J. Bruce Fields wrote:
> > > > What if we have an ioctl or a process personality flag where a broken
> > > > application can tell the file system "I'm broken, please give me a
> > > > degraded telldir/seekdir cookie"?  That way we don't penalize programs
> > > > that are doing the right thing, while providing some accomodation for
> > > > programs who are abusing the telldir cookie.
> > >
> > > Yeah, if there's a simple way to do that, maybe it would be worth it.
> >
> > Doing this as an ioctl which gets called right after opendir, i.e
> > (ignoring error checking):
> >
> >       DIR *dir = opendir("/foo/bar/baz");
> >       ioctl(dirfd(dir), EXT4_IOC_DEGRADED_READDIR, 1);
> >       ...
> >
> > should be quite easy.  It would be a very ext3/4 specific thing,
> > though.
> 
> 
> That would work, even though it would be ext3/4 specific. What is the
> recommended programmatic way to detect if the file is on ext3/4 -- we would
> not want to attempt that blindly on a non-ext3/4 FS as the numerical value
> of EXT4_IOC_DEGRADED_READDIR might get interpreted in dangerous ways?

We must have been through this before, but: is the only way to generate
a collision-free readdir cookie really to use a larger hash?

Would it be possible to make something work like, for example, a 31-bit
hash plus an offset into a hash bucket?

I have trouble thinking about this, partly because I can't remember
where to find the requirements for readdir on concurrently modified
directories....

--b.

  parent reply	other threads:[~2013-02-13 23:05 UTC|newest]

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-02-12 20:28 regressions due to 64-bit ext4 directory cookies J. Bruce Fields
2013-02-12 20:56 ` Bernd Schubert
2013-02-12 21:00   ` J. Bruce Fields
2013-02-13  8:17     ` Bernd Schubert
2013-02-13 22:18       ` J. Bruce Fields
2013-02-13 13:31     ` [Gluster-devel] " Niels de Vos
2013-02-13 15:40       ` Bernd Schubert
2013-02-14  5:32         ` Dave Chinner
2013-02-13  4:00 ` Theodore Ts'o
2013-02-13 13:31   ` J. Bruce Fields
2013-02-13 15:14     ` Theodore Ts'o
2013-02-13 15:19       ` J. Bruce Fields
2013-02-13 15:36         ` Theodore Ts'o
     [not found]           ` <20130213153654.GC17431-AKGzg7BKzIDYtjvyW6yDsg@public.gmane.org>
2013-02-13 16:20             ` J. Bruce Fields
2013-02-13 16:20               ` J. Bruce Fields
     [not found]               ` <20130213162059.GL14195-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>
2013-02-13 16:43                 ` Myklebust, Trond
2013-02-13 16:43                   ` Myklebust, Trond
2013-02-13 21:33                   ` J. Bruce Fields
2013-02-14  3:59                     ` Myklebust, Trond
     [not found]                       ` <4FA345DA4F4AE44899BD2B03EEEC2FA91F3D6BAB-UCI0kNdgLrHLJmV3vhxcH3OR4cbS7gtM96Bgd4bDwmQ@public.gmane.org>
2013-02-14  5:45                         ` Dave Chinner
2013-02-14  5:45                           ` Dave Chinner
2013-02-13 21:21                 ` Anand Avati
2013-02-13 21:21                   ` Anand Avati
     [not found]                   ` <CAFboF2wXvP+vttiff8iRE9rAgvV8UWGbFprgVp8p7kE43TU=PA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-02-13 22:20                     ` [Gluster-devel] " Theodore Ts'o
2013-02-13 22:20                       ` Theodore Ts'o
2013-02-13 22:41                       ` J. Bruce Fields
     [not found]                         ` <20130213224141.GU14195-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>
2013-02-13 22:47                           ` Theodore Ts'o
2013-02-13 22:47                             ` Theodore Ts'o
     [not found]                             ` <20130213224720.GE5938-AKGzg7BKzIDYtjvyW6yDsg@public.gmane.org>
2013-02-13 22:57                               ` Anand Avati
2013-02-13 22:57                                 ` Anand Avati
     [not found]                                 ` <CAFboF2z1akN_edrY_fT915xfehfHGioA2M=PSHv0Fp3rD-5v5A-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-02-13 23:05                                   ` J. Bruce Fields [this message]
2013-02-13 23:05                                     ` [Gluster-devel] " J. Bruce Fields
     [not found]                                     ` <20130213230511.GW14195-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>
2013-02-13 23:44                                       ` Theodore Ts'o
2013-02-13 23:44                                         ` Theodore Ts'o
     [not found]                                         ` <20130213234430.GF5938-AKGzg7BKzIDYtjvyW6yDsg@public.gmane.org>
2013-02-14  0:05                                           ` Anand Avati
2013-02-14  0:05                                             ` Anand Avati
     [not found]                                             ` <CAFboF2zS+YAa0uUxMFUAbqgPh3Kb4xZu40WUjLyGn8qPoP+Oyw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-02-14 21:47                                               ` [Gluster-devel] " J. Bruce Fields
2013-02-14 21:47                                                 ` J. Bruce Fields
2013-03-26 15:23                                               ` Bernd Schubert
2013-03-26 15:23                                                 ` [Gluster-devel] " Bernd Schubert
     [not found]                                                 ` <5151BD5F.30607-mPn0NPGs4xGatNDF+KUbs4QuADTiUCJX@public.gmane.org>
2013-03-26 15:48                                                   ` Eric Sandeen
2013-03-26 15:48                                                     ` Eric Sandeen
2013-03-28 14:07                                                     ` Theodore Ts'o
2013-03-28 16:26                                                       ` Eric Sandeen
2013-03-28 17:52                                                       ` Zach Brown
     [not found]                                                         ` <20130328175205.GD16651-fypN+1c5dIyjpB87vu3CluTW4wlIGRCZ@public.gmane.org>
2013-03-28 18:05                                                           ` Anand Avati
2013-03-28 18:05                                                             ` Anand Avati
     [not found]                                                             ` <CAFboF2ztc06G00z8ga35NrxgnT2YgBiDECgU_9kvVA_Go1_Bww-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-03-28 18:31                                                               ` [Gluster-devel] " J. Bruce Fields
2013-03-28 18:31                                                                 ` J. Bruce Fields
     [not found]                                                                 ` <20130328183153.GG7080-uC3wQj2KruNg9hUCZPvPmw@public.gmane.org>
2013-03-28 18:49                                                                   ` Anand Avati
2013-03-28 18:49                                                                     ` Anand Avati
     [not found]                                                                     ` <CAFboF2w49Lc0vM0SerbJfL9_RuSHgEU+y_Yk7F4pLxeiqu+KRg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-03-28 19:43                                                                       ` [Gluster-devel] " Jeff Darcy
2013-03-28 19:43                                                                         ` Jeff Darcy
     [not found]                                                                         ` <51549D74.1060703-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2013-03-28 22:14                                                                           ` Anand Avati
2013-03-28 22:14                                                                             ` Anand Avati
     [not found]                                                                             ` <CAFboF2xkvXx9YFYxBXupwg=s=3MaeQYm2KK2m8MFtEBPsxwQ7Q-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2013-03-28 22:20                                                                               ` Anand Avati
2013-03-28 22:20                                                                                 ` Anand Avati
2013-02-14 21:46                                           ` [Gluster-devel] " J. Bruce Fields
2013-02-14 21:46                                             ` J. Bruce Fields
     [not found]                       ` <20130213222052.GD5938-AKGzg7BKzIDYtjvyW6yDsg@public.gmane.org>
2013-02-14  6:10                         ` Dave Chinner
2013-02-14  6:10                           ` Dave Chinner
2013-02-14 22:01                           ` J. Bruce Fields
2013-02-15  2:27                             ` Dave Chinner
2013-02-13  6:56 ` Andreas Dilger
2013-02-13 13:40   ` J. Bruce Fields

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20130213230511.GW14195@fieldses.org \
    --to=bfields-uc3wqj2krung9huczpvpmw@public.gmane.org \
    --cc=anand.avati-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
    --cc=bernd.schubert-mPn0NPGs4xGatNDF+KUbs4QuADTiUCJX@public.gmane.org \
    --cc=gluster-devel-qX2TKyscuCcdnm+yROfE0A@public.gmane.org \
    --cc=linux-ext4-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-nfs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=sandeen-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
    --cc=tytso-3s7WtUTddSA@public.gmane.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.