From: Luca Di Maio <luca.dimaio1@gmail.com>
To: "Darrick J. Wong" <djwong@kernel.org>
Cc: linux-xfs@vger.kernel.org, dimitri.ledkov@chainguard.dev,
smoser@chainguard.dev
Subject: Re: [PATCH RFC 2/2] proto: read origin also for directories, chardevs and symlinks. copy timestamps from origin.
Date: Thu, 17 Apr 2025 16:14:38 +0200 [thread overview]
Message-ID: <6882ed09-26ee-490b-9a07-8178651b9d8e@gmail.com> (raw)
In-Reply-To: <20250416160716.GG25675@frogsfrogsfrogs>
Thanks Darrick for the review
> On 4/16/25 18:07, Darrick J. Wong wrote: The protofile format that
> mkfs.xfs uses has been static for 52 years, I don't know that
> expending a large amount of effort on it is worth the time. If you
> really must have reproducible filesystem images, would it be easier
> to allow overriding current_time() via environment vars?
It's nice to preserve the source timestamps, for example, in case we
have also reproducible builds for packages and rootfs (in tar.gz format)
Fixing it to a set time would lose that information in the "translation"
> On 4/16/25 18:07, Darrick J. Wong wrote: (I also don't really get
> why anyone cares about bit-reproducible filesystem images; the only
> behavioral guarantees are the userspace interface contract.
> Filesystem drivers have wide latitude to do whatever they want under
> the covers.)
For distribution of said disks and avoiding re-doing work (eg: if the
checksum matches, we know there were no changes, thus we can skip the
publication CI)
I've reviewed the patch, in order to not change the prototype file
specification whatsoever.
The new patch will leverage a comment that `xfs_protofile` inserts since
its first iteration:
": Descending path [foo]"
This way, if the comment is not found, like for older files, the new
behavior will simply be ignored.
This should address the compatibility concerns.
I've tested this incoming patch with a matrix of
"old-protofiles/new-mkfs/new-protofile/old-mkfs" and it seems there are
no compatibility issues.
Thanks
L.
next prev parent reply other threads:[~2025-04-17 14:14 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-04-16 14:43 [PATCH RFC 0/2] prototype: improve timestamp handling Luca Di Maio
2025-04-16 14:43 ` [PATCH RFC 1/2] xfs_proto: add origin also for directories, chardevs and symlinks Luca Di Maio
2025-04-16 16:07 ` Darrick J. Wong
2025-04-16 14:43 ` [PATCH RFC 2/2] proto: read origin also for directories, chardevs and symlinks. copy timestamps from origin Luca Di Maio
2025-04-16 16:07 ` Darrick J. Wong
2025-04-17 14:14 ` Luca Di Maio [this message]
2025-04-22 3:10 ` [PATCH RFC 0/2] prototype: improve timestamp handling Darrick J. Wong
2025-04-22 6:16 ` Luca Di Maio
2025-04-23 14:46 ` Darrick J. Wong
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=6882ed09-26ee-490b-9a07-8178651b9d8e@gmail.com \
--to=luca.dimaio1@gmail.com \
--cc=dimitri.ledkov@chainguard.dev \
--cc=djwong@kernel.org \
--cc=linux-xfs@vger.kernel.org \
--cc=smoser@chainguard.dev \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox