[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <alpine.LFD.2.00.1303131830570.18319@dhcp-1-104.brq.redhat.com>
Date: Wed, 13 Mar 2013 18:31:25 +0100 (CET)
From: Lukáš Czerner <lczerner@...hat.com>
To: "Theodore Ts'o" <tytso@....edu>
cc: Lukáš Czerner <lczerner@...hat.com>,
linux-ext4@...r.kernel.org
Subject: Re: Metadata reservation for unwritten extent conversion
On Wed, 13 Mar 2013, Theodore Ts'o wrote:
> Date: Wed, 13 Mar 2013 13:25:21 -0400
> From: Theodore Ts'o <tytso@....edu>
> To: Lukáš Czerner <lczerner@...hat.com>
> Cc: linux-ext4@...r.kernel.org
> Subject: Re: Metadata reservation for unwritten extent conversion
>
> On Wed, Mar 13, 2013 at 05:09:51PM +0100, Lukáš Czerner wrote:
> >
> > Well, we can not actually revert that patch because it introduces
> > da_reserve_metadata() which is used elsewhere in the commit:
> >
> > ext4: reserve metadata block for every delayed write
> >
> > I can resend the patch which uses da_reserve_metadata() so you can use
> > it instead the old version
>
> That sounds like the best thing to do. I assume you will be simply
> taking the da_reserve_metadata() function introduced in the
> to-be-dropped commit:
>
> > ext4: Reserve metadata if writing into uninitialized
>
> ... and moving it into the "reserve metadata block for every delayed
> write" patch?
Exactly.
-Lukas
>
> - Ted
Powered by blists - more mailing lists