Git development
 help / color / mirror / Atom feed
From: Phillip Wood <phillip.wood123@gmail.com>
To: "D. Ben Knoble" <ben.knoble@gmail.com>, git@vger.kernel.org
Cc: Thomas Bachem <mail@thomasbachem.com>,
	Eli Barzilay <eli@barzilay.org>,
	Phillip Wood <phillip.wood@dunelm.org.uk>,
	Junio C Hamano <gitster@pobox.com>
Subject: Re: [PATCH v4 4/5] t5520: don't expire reflogs where it matters
Date: Tue, 29 Sep 2026 16:46:03 +0100	[thread overview]
Message-ID: <3547f4aa-649a-4f46-868c-0e50dfa69466@gmail.com> (raw)
In-Reply-To: <2ac371d2dc1425cc47bf369e88b321d3c0c8c605.1790684309.git.ben.knoble@gmail.com>

On 29/09/2026 13:18, D. Ben Knoble wrote:
> From: Thomas Bachem <mail@thomasbachem.com>
> 
> The "--rebase -f with rebased upstream" test computes its fork point
> from the reflog of refs/remotes/me/copy, and the entry it needs is
> the one that the fetch of the test before it wrote. Like every reflog
> entry the suite writes after test_tick, it is dated 2005, so the
> first "git reflog expire --all" after that fetch removes it. Pull
> then finds no fork point and rebases onto the merge head with the
> merge head as the upstream, and the rewound commits come back as a
> conflict.
> 
> Since 452b12c2e0 (builtin/maintenance: use "geometric" strategy by
> default, 2026-02-24) auto maintenance runs that expiry once the reflog
> of HEAD holds a hundred entries it would remove, the default of
> maintenance.reflog-expire.auto. Which run crosses the threshold
> depends on the entries and maintenance runs before it, so the script
> passed by chance: a stash topic that no longer runs "git reset" from
> "stash apply --index" and a rebase topic that runs auto maintenance
> at the end of "git rebase" together move the expiry between the two
> tests.
> 
> Pin the expiry as ea7d894f44 (t34xx: don't expire reflogs where it
> matters, 2026-02-24) did for the rebase tests. That covers a "git gc"
> as well, which expires reflogs on its own, where turning off the auto
> trigger of the reflog-expire task alone would not.

I find this commit message quite hard to understand. From my perspective 
the important points are

  - "git merge" uses "git stash" to clear any uncommitted changes from
    the worktree before it tries each strategy. The stashed changes are
    popped with "--index".

  - switching "git stash pop --index" to use merge_incore_nonrecursive()
    causes "git merge" to stop writing the reflog entries that came from
    "git stash pop --index" running "git reset"

  - that combined with "git rebase" starting to run "git maintenance
    --auto" changed when we expire the reflogs which breaks the fork-
    point detection.

The changes themselves look good

Thanks

Phillip

> 
> Reported-by: Junio C Hamano <gitster@pobox.com>
> Helped-by: D. Ben Knoble <ben.knoble@gmail.com>
> Helped-by: Phillip Wood <phillip.wood@dunelm.org.uk>
> Assisted-by: Claude Fable 5.1
> Signed-off-by: Thomas Bachem <mail@thomasbachem.com>
> Signed-off-by: D. Ben Knoble <ben.knoble@gmail.com>
> ---
>   t/t5520-pull.sh | 6 ++++++
>   1 file changed, 6 insertions(+)
> 
> diff --git a/t/t5520-pull.sh b/t/t5520-pull.sh
> index 27f38ab3c8..bc818605a5 100755
> --- a/t/t5520-pull.sh
> +++ b/t/t5520-pull.sh
> @@ -35,6 +35,12 @@ test_pull_autostash_fail () {
>   }
>   
>   test_expect_success setup '
> +	# Commit dates are hardcoded to 2005, and the reflog entries will have
> +	# a matching timestamp. Maintenance may thus immediately expire
> +	# reflogs if it was running.
> +	git config set gc.reflogExpire never &&
> +	git config set gc.reflogExpireUnreachable never &&
> +
>   	echo file >file &&
>   	git add file &&
>   	git commit -a -m original


  reply	other threads:[~2026-09-29 15:46 UTC|newest]

Thread overview: 78+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-19 21:26 [PATCH 0/2] Hi all, D. Ben Knoble
2026-09-19 21:26 ` [PATCH 1/2] builtin/stash: remove unused header D. Ben Knoble
2026-09-21 15:10   ` Junio C Hamano
2026-09-19 21:26 ` [PATCH 2/2] builtin/stash: merge index in-core D. Ben Knoble
2026-09-21 13:17   ` Phillip Wood
2026-09-22 12:43     ` D. Ben Knoble
2026-09-22 12:51       ` D. Ben Knoble
2026-09-22 13:57       ` Phillip Wood
2026-09-22 20:34         ` D. Ben Knoble
2026-09-19 21:32 ` [PATCH 0/2] Hi all, D. Ben Knoble
2026-09-23 12:58 ` [PATCH v2 0/4] stash: clean up index-mode test merge D. Ben Knoble
2026-09-23 12:58   ` [PATCH v2 1/4] builtin/stash: remove unused header D. Ben Knoble
2026-09-23 12:58   ` [PATCH v2 2/4] stash: prepare merge options earlier D. Ben Knoble
2026-09-23 12:58   ` [PATCH v2 3/4] t: test failed "stash apply --index" D. Ben Knoble
2026-09-24  9:42     ` Phillip Wood
2026-09-25 13:36       ` D. Ben Knoble
2026-09-25 15:45         ` Phillip Wood
2026-09-26  9:53           ` Phillip Wood
2026-09-26 12:07             ` D. Ben Knoble
2026-09-23 12:58   ` [PATCH v2 4/4] builtin/stash: merge index in-core D. Ben Knoble
2026-09-24  9:42     ` Phillip Wood
2026-09-25 12:55       ` D. Ben Knoble
2026-09-25 15:58         ` Phillip Wood
2026-09-25 16:16           ` D. Ben Knoble
2026-09-24 21:59     ` Junio C Hamano
2026-09-25  4:12       ` Junio C Hamano
2026-09-25 13:00       ` D. Ben Knoble
2026-09-25 16:24         ` Junio C Hamano
2026-09-26  9:51           ` Phillip Wood
2026-09-26 12:04             ` D. Ben Knoble
2026-09-25 16:04       ` Phillip Wood
2026-09-25 16:17         ` D. Ben Knoble
2026-09-25 16:49           ` Junio C Hamano
2026-09-26 12:16   ` [PATCH v3 0/5] stash: clean up index-mode test merge D. Ben Knoble
2026-09-26 12:16     ` [PATCH v3 1/5] builtin/stash: remove unused header D. Ben Knoble
2026-09-26 12:16     ` [PATCH v3 2/5] stash: prepare merge options earlier D. Ben Knoble
2026-09-26 12:16     ` [PATCH v3 3/5] t3903: test stash --index merges D. Ben Knoble
2026-09-28 15:44       ` Phillip Wood
2026-09-28 15:55         ` D. Ben Knoble
2026-09-29  9:41           ` Phillip Wood
2026-09-26 12:16     ` [PATCH v3 4/5] t3903: test failed "stash apply --index" D. Ben Knoble
2026-09-26 12:16     ` [PATCH v3 5/5] builtin/stash: merge index in-core D. Ben Knoble
2026-09-27 18:59       ` Junio C Hamano
2026-09-28 12:02         ` D. Ben Knoble
2026-09-28  9:40       ` Junio C Hamano
2026-09-28 12:03         ` D. Ben Knoble
2026-09-28 15:32           ` Junio C Hamano
2026-09-26 12:20     ` [PATCH v3 0/5] stash: clean up index-mode test merge D. Ben Knoble
2026-09-27 19:21     ` Junio C Hamano
2026-09-28  9:50       ` Phillip Wood
2026-09-28 12:05         ` D. Ben Knoble
2026-09-28 12:33           ` D. Ben Knoble
2026-09-28 13:00             ` D. Ben Knoble
2026-09-28 13:45               ` Phillip Wood
2026-09-28 14:50                 ` Thomas Bachem
2026-09-28 15:36                   ` D. Ben Knoble
2026-09-29 11:38                     ` D. Ben Knoble
2026-09-29 15:54                     ` Phillip Wood
2026-09-28 15:40                   ` Phillip Wood
2026-09-29 12:18 ` [PATCH v4 " D. Ben Knoble
2026-09-29 12:18   ` [PATCH v4 1/5] builtin/stash: remove unused header D. Ben Knoble
2026-09-29 12:18   ` [PATCH v4 2/5] stash: prepare merge options earlier D. Ben Knoble
2026-09-29 12:18   ` [PATCH v4 3/5] t3903: test failed "stash apply --index" D. Ben Knoble
2026-09-29 12:18   ` [PATCH v4 4/5] t5520: don't expire reflogs where it matters D. Ben Knoble
2026-09-29 15:46     ` Phillip Wood [this message]
2026-09-29 12:18   ` [PATCH v4 5/5] builtin/stash: merge index in-core D. Ben Knoble
2026-09-29 20:07     ` Junio C Hamano
2026-09-30  1:29       ` D. Ben Knoble
2026-09-29 15:48   ` [PATCH v4 0/5] stash: clean up index-mode test merge Phillip Wood
2026-09-29 17:31     ` Ben Knoble
2026-09-30 21:26       ` D. Ben Knoble
2026-09-30 21:24 ` [PATCH v5 0/4] " D. Ben Knoble
2026-09-30 21:24   ` [PATCH v5 1/4] builtin/stash: remove unused header D. Ben Knoble
2026-09-30 21:24   ` [PATCH v5 2/4] stash: prepare merge options earlier D. Ben Knoble
2026-09-30 21:24   ` [PATCH v5 3/4] t3903: test failed "stash apply --index" D. Ben Knoble
2026-09-30 21:24   ` [PATCH v5 4/4] builtin/stash: merge index in-core D. Ben Knoble
2026-10-01 15:52   ` [PATCH v5 0/4] stash: clean up index-mode test merge Phillip Wood
2026-10-01 17:47     ` Junio C Hamano

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=3547f4aa-649a-4f46-868c-0e50dfa69466@gmail.com \
    --to=phillip.wood123@gmail.com \
    --cc=ben.knoble@gmail.com \
    --cc=eli@barzilay.org \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=mail@thomasbachem.com \
    --cc=phillip.wood@dunelm.org.uk \
    /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