Received-SPF: pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Content-Type: text/plain;
        charset=utf-8
Mime-Version: 1.0 (Mac OS X Mail 14.0 \(3654.120.0.1.13\))
Subject: Re: [PATCH] e2fsprogs: avoid code duplication
From:   Alexey Lyahkov <alexey.lyashkov@gmail.com>
In-Reply-To: <YurTVlGWqwym2Hgg@mit.edu>
Date:   Thu, 4 Aug 2022 00:52:34 +0300
Cc:     linux-ext4 <linux-ext4@vger.kernel.org>,
        Andreas Dilger <adilger@dilger.ca>,
        Artem Blagodarenko <artem.blagodarenko@gmail.com>
Content-Transfer-Encoding: quoted-printable
Message-Id: <A850E8AB-1B4D-4956-9933-CF32DF0762BC@gmail.com>
References: <20220803075407.538398-1-alexey.lyashkov@gmail.com>
 <YurTVlGWqwym2Hgg@mit.edu>
To:     Theodore Ts'o <tytso@mit.edu>
Precedence: bulk

Thanks for pointing to the libsupport. I looking into kernel-jbd.h as =
example (it also don=E2=80=99t export outside of e2fsprogs),
but  libsupport is lost from my radar.

Lack of tag v3 is big lost also. It mean debugfs don=E2=80=99t able to =
print log records correctly if block number over 2^32.

Alex

> On 3 Aug 2022, at 22:58, Theodore Ts'o <tytso@mit.edu> wrote:
>=20
> On Wed, Aug 03, 2022 at 10:54:07AM +0300, Alexey Lyashkov wrote:
>> debugfs and e2fsck have a so much code duplication in journal =
handing.
>> debugfs have lack a many journal features handing also.
>> Let's start code merging to avoid code duplication and lack features.
>=20
> This is definitely worth doing, and as you've pointed out, there are a
> number of features which are in e2fsck/journal.c, which are not in
> debugfs/journal.c.  The most notable one which I picked up on is the
> fast_commit code --- which is in the master/next branch, but not in
> the maint branch.
>=20
> I suggest that we move the functionality into the libsupport library
> first.  I want to make sure we get the abstractions right before we
> "cast them into stone" by moving the functions to libext2fs.
> Libsupport is not exported outside of e2fsprogs, so if we decide we
> want to change function signatures, or make some functions private, we
> can do that more easily if we experiment with moving things into
> libsupport first.
>=20
> 						- Ted