From: Patrick Steinhardt <ps@pks.im>
To: git@vger.kernel.org
Subject: [PATCH 0/6] fetch: improve atomicity of `--atomic` flag
Date: Fri, 11 Feb 2022 08:46:39 +0100 [thread overview]
Message-ID: <cover.1644565025.git.ps@pks.im> (raw)
[-- Attachment #1: Type: text/plain, Size: 2269 bytes --]
Hi,
in c7b190dabd (fetch: implement support for atomic reference updates,
2021-01-12), I have added a new `--atomic` flag to git-fetch(1) which is
supposed to make it an all-or-nothing operation: either all refs are
updated, or none is. I have recently discovered though that there were
two oversights: neither pruning of refs via `--prune` nor the tag
backfill mechanism are currently covered by this flag, which breaks the
promise.
This patch series extends coverage of the `--atomic` flag to fix my
oversights. It applies on top of ps/avoid-unnecessary-hook-invocation-with-packed-refs
merged to v2.35.1 due to a semantic conflict in tests: I use the reftx
hook to verify some things work as expected, and above branch causes us
to not execute the hook for packed-refs anymore.
It is structured as follows:
- Patch 1 adds multiple tests which demonstrate issues with
backfilling and the lacking atomicity.
- Patch 2 and 3 do some preliminary refactorings which set the stage
for improved atomicity.
- Patch 4 fixes a bug with backfilling tags: we don't return an
error code to the user if the backfill fails.
- Patch 5 and 6 then finally extend the atomicity guarantees.
I'm not yet all that happy with patch 5: it currently has to reach into
the `ref_transaction` struct to be able to filter out any updates which
have already been queued such that we don't accidentally queue the same
tag update twice. I first wanted to get some feedback on this patch
series though, and in case it is agreed that it fixes a real issue I may
introduce a new non-internal API which allows iterating over queued
updates.
Patrick
Patrick Steinhardt (6):
fetch: increase test coverage of fetches
fetch: backfill tags before setting upstream
fetch: control lifecycle of FETCH_HEAD in a single place
fetch: report errors when backfilling tags fails
fetch: make `--atomic` flag cover backfilling of tags
fetch: make `--atomic` flag cover pruning of refs
builtin/fetch.c | 190 +++++++++++++++++++++++++++++--------------
t/t5503-tagfollow.sh | 78 ++++++++++++++++++
t/t5510-fetch.sh | 31 +++++++
3 files changed, 236 insertions(+), 63 deletions(-)
--
2.35.1
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
next reply other threads:[~2022-02-11 7:46 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-11 7:46 Patrick Steinhardt [this message]
2022-02-11 7:46 ` [PATCH 1/6] fetch: increase test coverage of fetches Patrick Steinhardt
2022-02-15 6:19 ` Christian Couder
2022-02-17 11:13 ` Patrick Steinhardt
2022-02-11 7:46 ` [PATCH 2/6] fetch: backfill tags before setting upstream Patrick Steinhardt
2022-02-15 6:43 ` Christian Couder
2022-02-11 7:46 ` [PATCH 3/6] fetch: control lifecycle of FETCH_HEAD in a single place Patrick Steinhardt
2022-02-15 7:32 ` Christian Couder
2022-02-17 11:18 ` Patrick Steinhardt
2022-02-11 7:46 ` [PATCH 4/6] fetch: report errors when backfilling tags fails Patrick Steinhardt
2022-02-15 7:52 ` Christian Couder
2022-02-17 11:27 ` Patrick Steinhardt
2022-02-17 12:47 ` Patrick Steinhardt
2022-02-16 23:35 ` Jonathan Tan
2022-02-17 1:31 ` Elijah Newren
2022-02-11 7:47 ` [PATCH 5/6] fetch: make `--atomic` flag cover backfilling of tags Patrick Steinhardt
2022-02-15 8:11 ` Christian Couder
2022-02-16 23:41 ` Jonathan Tan
2022-02-17 11:37 ` Patrick Steinhardt
2022-02-17 1:34 ` Elijah Newren
2022-02-17 11:58 ` Patrick Steinhardt
2022-02-11 7:47 ` [PATCH 6/6] fetch: make `--atomic` flag cover pruning of refs Patrick Steinhardt
2022-02-15 9:12 ` Christian Couder
2022-02-17 12:03 ` Patrick Steinhardt
2022-02-16 23:39 ` Jonathan Tan
2022-02-17 1:40 ` Elijah Newren
2022-02-17 12:06 ` Patrick Steinhardt
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=cover.1644565025.git.ps@pks.im \
--to=ps@pks.im \
--cc=git@vger.kernel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).