From: Jeff King <peff@peff.net>
To: Junio C Hamano <gitster@pobox.com>
Cc: Michael Haggerty <mhagger@alum.mit.edu>,
Stefan Beller <sbeller@google.com>,
Jonathan Nieder <jrnieder@gmail.com>,
Ronnie Sahlberg <ronniesahlberg@gmail.com>,
git@vger.kernel.org
Subject: Re: [PATCH 07/23] expire_reflog(): use a lock_file for rewriting the reflog file
Date: Tue, 9 Dec 2014 13:54:34 -0500 [thread overview]
Message-ID: <20141209185433.GB31158@peff.net> (raw)
In-Reply-To: <xmqqfvcomzib.fsf@gitster.dls.corp.google.com>
On Tue, Dec 09, 2014 at 10:47:24AM -0800, Junio C Hamano wrote:
> Michael Haggerty <mhagger@alum.mit.edu> writes:
>
> > This isn't documented very well. I thought I saw a comment somewhere in
> > the code that stated it explicitly, but I can't find it now. In any
> > case, my understanding of the locking protocol for reflogs is:
> >
> > The reflog for "$refname", which is stored at
> > "$GIT_DIR/logs/$refname", is locked by holding
> > "$GIT_DIR/refs/$refname.lock", *even if the corresponding
> > reference is packed*.
> >
> > This implies that readers, who don't pay attention to locks, have to be
> > prepared for the possibility that the reflog is in the middle of an
> > update and that the last line is incomplete. This is handled by
> > show_one_reflog_ent(), which discards incomplete lines.
>
> Interesting, as I think I saw Peff did something around that area
> recently.
Yeah, and I had no idea about the truncated-line race which Michael
described at the time. It makes me glad that I took the time to do the
more careful thing[1] (fixing the reverse-reflog parser to properly
include the newline when present) and not the quick-and-easy thing[2]
(teaching show_one_reflog_ent to accept entries without newlines).
What you have queued in jk/for-each-reflog-ent-reverse should be fine,
even in light of what Michael said.
-Peff
[1] http://article.gmane.org/gmane.comp.version-control.git/260849
[2] http://article.gmane.org/gmane.comp.version-control.git/260807
next prev parent reply other threads:[~2014-12-09 18:54 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-12-04 23:08 [PATCH 00/23] Add reflog_expire() to the references API Michael Haggerty
2014-12-04 23:08 ` [PATCH 01/23] refs.c: make ref_transaction_create a wrapper for ref_transaction_update Michael Haggerty
2014-12-04 23:08 ` [PATCH 02/23] refs.c: make ref_transaction_delete " Michael Haggerty
2014-12-04 23:08 ` [PATCH 03/23] refs.c: add a function to append a reflog entry to a fd Michael Haggerty
2014-12-04 23:08 ` [PATCH 04/23] expire_reflog(): remove unused parameter Michael Haggerty
2014-12-04 23:20 ` Jonathan Nieder
2014-12-04 23:28 ` Jonathan Nieder
2014-12-05 12:43 ` Michael Haggerty
2014-12-04 23:08 ` [PATCH 05/23] expire_reflog(): rename "ref" parameter to "refname" Michael Haggerty
2014-12-04 23:44 ` Jonathan Nieder
2014-12-04 23:08 ` [PATCH 06/23] expire_reflog(): exit early if the reference has no reflog Michael Haggerty
2014-12-04 23:48 ` Jonathan Nieder
2014-12-04 23:53 ` Jonathan Nieder
2014-12-05 15:10 ` Michael Haggerty
2014-12-04 23:08 ` [PATCH 07/23] expire_reflog(): use a lock_file for rewriting the reflog file Michael Haggerty
2014-12-05 0:23 ` Jonathan Nieder
2014-12-05 2:19 ` Stefan Beller
2014-12-08 10:07 ` Michael Haggerty
2014-12-09 18:47 ` Junio C Hamano
2014-12-09 18:54 ` Jeff King [this message]
2014-12-05 19:18 ` Stefan Beller
2014-12-05 19:32 ` Junio C Hamano
2014-12-05 19:41 ` Stefan Beller
2014-12-05 20:55 ` Junio C Hamano
2014-12-08 14:05 ` Michael Haggerty
2014-12-05 2:59 ` ronnie sahlberg
2014-12-08 10:40 ` Michael Haggerty
[not found] ` <CAN05THTTba-1n12hBszJAU-O+wsbSFd5Lt+kMk7_MU_0C=wZGQ@mail.gmail.com>
2014-12-05 17:47 ` Stefan Beller
2014-12-04 23:08 ` [PATCH 08/23] Extract function should_expire_reflog_ent() Michael Haggerty
2014-12-08 22:33 ` Stefan Beller
2014-12-04 23:08 ` [PATCH 09/23] expire_reflog(): extract two policy-related functions Michael Haggerty
2014-12-05 19:02 ` Stefan Beller
2014-12-04 23:08 ` [PATCH 10/23] expire_reflog(): add a "flags" argument Michael Haggerty
2014-12-08 22:35 ` Stefan Beller
2014-12-04 23:08 ` [PATCH 11/23] expire_reflog(): move dry_run to flags argument Michael Haggerty
2014-12-08 22:38 ` Stefan Beller
2014-12-04 23:08 ` [PATCH 12/23] expire_reflog(): move updateref " Michael Haggerty
2014-12-08 22:42 ` Stefan Beller
2014-12-04 23:08 ` [PATCH 13/23] Rename expire_reflog_cb to expire_reflog_policy_cb Michael Haggerty
2014-12-08 22:46 ` Stefan Beller
2014-12-04 23:08 ` [PATCH 14/23] struct expire_reflog_cb: a new callback data type Michael Haggerty
2014-12-08 22:49 ` Stefan Beller
2014-12-04 23:08 ` [PATCH 15/23] expire_reflog(): pass flags through to expire_reflog_ent() Michael Haggerty
2014-12-08 22:55 ` Stefan Beller
2014-12-04 23:08 ` [PATCH 16/23] expire_reflog(): move verbose to flags argument Michael Haggerty
2014-12-08 22:56 ` Stefan Beller
2014-12-04 23:08 ` [PATCH 17/23] expire_reflog(): move rewrite " Michael Haggerty
2014-12-08 22:58 ` Stefan Beller
2014-12-04 23:08 ` [PATCH 18/23] Move newlog and last_kept_sha1 to "struct expire_reflog_cb" Michael Haggerty
2014-12-08 22:59 ` Stefan Beller
2014-12-04 23:08 ` [PATCH 19/23] expire_reflog(): treat the policy callback data as opaque Michael Haggerty
2014-12-08 23:12 ` Stefan Beller
2014-12-04 23:08 ` [PATCH 20/23] reflog_expire(): new function in the reference API Michael Haggerty
2014-12-08 23:32 ` Stefan Beller
2014-12-12 8:23 ` Michael Haggerty
2014-12-12 8:50 ` Jeff King
2014-12-12 18:57 ` Junio C Hamano
2014-12-04 23:08 ` [PATCH 21/23] refs.c: remove unlock_ref/close_ref/commit_ref from the refs api Michael Haggerty
2014-12-04 23:08 ` [PATCH 22/23] lock_any_ref_for_update(): inline function Michael Haggerty
2014-12-08 23:34 ` Stefan Beller
2014-12-11 0:13 ` Michael Haggerty
2014-12-04 23:08 ` [PATCH 23/23] refs.c: don't expose the internal struct ref_lock in the header file Michael Haggerty
2014-12-04 23:47 ` [PATCH 00/23] Add reflog_expire() to the references API 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
List information: http://vger.kernel.org/majordomo-info.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20141209185433.GB31158@peff.net \
--to=peff@peff.net \
--cc=git@vger.kernel.org \
--cc=gitster@pobox.com \
--cc=jrnieder@gmail.com \
--cc=mhagger@alum.mit.edu \
--cc=ronniesahlberg@gmail.com \
--cc=sbeller@google.com \
/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.
Code repositories for project(s) associated with this public inbox
https://80x24.org/mirrors/git.git
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 read-only IMAP folder(s) and NNTP newsgroup(s).