git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: "Nguyễn Thái Ngọc Duy" <pclouds@gmail.com>
To: git@vger.kernel.org
Cc: t.gummerer@gmail.com, larsxschneider@gmail.com,
	bmwill@google.com, peff@peff.net,
	"Junio C Hamano" <gitster@pobox.com>,
	szeder.dev@gmail.com, "Nguyễn Thái Ngọc Duy" <pclouds@gmail.com>
Subject: [PATCH 3/3] read-cache: don't write index twice if we can't write shared index
Date: Sun, 14 Jan 2018 17:18:20 +0700	[thread overview]
Message-ID: <20180114101820.4273-3-pclouds@gmail.com> (raw)
In-Reply-To: <20180114101820.4273-1-pclouds@gmail.com>

In a0a967568e ("update-index --split-index: do not split if $GIT_DIR is
read only", 2014-06-13), we tried to make sure we can still write an
index, even if the shared index can not be written.

We did so by just calling 'do_write_locked_index()' just before
'write_shared_index()'.  'do_write_locked_index()' always at least
closes the tempfile nowadays, and used to close or commit the lockfile
if COMMIT_LOCK or CLOSE_LOCK were given at the time this feature was
introduced.  COMMIT_LOCK or CLOSE_LOCK is passed in by most callers of
'write_locked_index()'.

After calling 'write_shared_index()', we call 'write_split_index()',
which calls 'do_write_locked_index()' again, which then tries to use the
closed lockfile again, but in fact fails to do so as it's already
closed. This eventually leads to a segfault.

Make sure to write the main index only once.

[nd: most of the commit message and investigation done by Thomas, I only
tweaked the solution a bit]

Helped-by: Thomas Gummerer <t.gummerer@gmail.com>
Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
---
 read-cache.c           |  5 +++--
 t/t1700-split-index.sh | 19 +++++++++++++++++++
 2 files changed, 22 insertions(+), 2 deletions(-)

diff --git a/read-cache.c b/read-cache.c
index c568643f55..c58c0a978a 100644
--- a/read-cache.c
+++ b/read-cache.c
@@ -2561,8 +2561,9 @@ int write_locked_index(struct index_state *istate, struct lock_file *lock,
 		if (!temp) {
 			hashclr(si->base_sha1);
 			ret = do_write_locked_index(istate, lock, flags);
-		} else
-			ret = write_shared_index(istate, &temp);
+			goto out;
+		}
+		ret = write_shared_index(istate, &temp);
 
 		saved_errno = errno;
 		if (is_tempfile_active(temp))
diff --git a/t/t1700-split-index.sh b/t/t1700-split-index.sh
index af9b847761..5494389dbb 100755
--- a/t/t1700-split-index.sh
+++ b/t/t1700-split-index.sh
@@ -401,4 +401,23 @@ done <<\EOF
 0642 -rw-r---w-
 EOF
 
+test_expect_success POSIXPERM 'graceful handling splitting index is not allowed' '
+	test_create_repo ro &&
+	(
+		cd ro &&
+		test_commit initial &&
+		git update-index --split-index &&
+		test -f .git/sharedindex.*
+	) &&
+	test_when_finished "chmod -R u+w ro" &&
+	chmod -R u-w ro &&
+	cp ro/.git/index new-index &&
+	GIT_INDEX_FILE="$(pwd)/new-index" git -C ro update-index --split-index &&
+	chmod -R u+w ro &&
+	rm ro/.git/sharedindex.* &&
+	GIT_INDEX_FILE=new-index git ls-files >actual &&
+	echo initial.t >expected &&
+	test_cmp expected actual
+'
+
 test_done
-- 
2.15.1.600.g899a5f85c6


  parent reply	other threads:[~2018-01-14 10:18 UTC|newest]

Thread overview: 99+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-12-10 21:21 [PATCH 0/3] fixes for split index mode Thomas Gummerer
2017-12-10 21:22 ` [PATCH 1/3] repository: fix repo_read_index with submodules Thomas Gummerer
2017-12-11 18:54   ` Brandon Williams
2017-12-11 20:37     ` Thomas Gummerer
2017-12-10 21:22 ` [PATCH 2/3] prune: fix pruning with multiple worktrees and split index Thomas Gummerer
2017-12-11 19:09   ` Brandon Williams
2017-12-11 21:39     ` Thomas Gummerer
2017-12-10 21:22 ` [PATCH 3/3] travis: run tests with GIT_TEST_SPLIT_INDEX Thomas Gummerer
2017-12-10 23:37   ` Eric Sunshine
2017-12-11 21:09   ` SZEDER Gábor
2017-12-11 21:42     ` Thomas Gummerer
2017-12-12 15:54       ` Lars Schneider
2017-12-12 19:15         ` Junio C Hamano
2017-12-12 20:15           ` Thomas Gummerer
2017-12-12 20:51             ` Junio C Hamano
2017-12-13 23:21               ` Thomas Gummerer
2017-12-13 17:21           ` Lars Schneider
2017-12-13 17:38             ` Junio C Hamano
2017-12-13 17:46               ` Lars Schneider
2017-12-13 23:28                 ` Thomas Gummerer
2017-12-17 22:51 ` [PATCH v2 0/3] fixes for split index mode Thomas Gummerer
2017-12-17 22:51   ` [PATCH v2 1/3] repository: fix repo_read_index with submodules Thomas Gummerer
2017-12-18 18:01     ` Brandon Williams
2017-12-18 23:05       ` Thomas Gummerer
2017-12-18 23:05         ` Brandon Williams
2017-12-17 22:51   ` [PATCH v2 2/3] prune: fix pruning with multiple worktrees and split index Thomas Gummerer
2017-12-18 18:19     ` Brandon Williams
2018-01-03 22:18       ` Thomas Gummerer
2018-01-04 19:12         ` Junio C Hamano
2017-12-17 22:51   ` [PATCH v2 3/3] travis: run tests with GIT_TEST_SPLIT_INDEX Thomas Gummerer
2017-12-18 18:16     ` Lars Schneider
2018-01-04 20:13       ` Thomas Gummerer
2018-01-05 11:03         ` Lars Schneider
2018-01-07 20:02         ` Thomas Gummerer
2018-01-07 22:30   ` [PATCH v3 0/3] fixes for split index mode Thomas Gummerer
2018-01-07 22:30     ` [PATCH v3 1/3] read-cache: fix reading the shared index for other repos Thomas Gummerer
2018-01-08 10:41       ` Duy Nguyen
2018-01-08 22:41         ` Thomas Gummerer
2018-01-13 22:33           ` Thomas Gummerer
2018-01-08 23:38         ` Brandon Williams
2018-01-09  1:24           ` Duy Nguyen
2018-01-16 21:42       ` Brandon Williams
2018-01-17  0:16         ` Duy Nguyen
2018-01-17  0:32           ` Brandon Williams
2018-01-17 18:16           ` Jonathan Nieder
2018-01-18 10:19             ` Duy Nguyen
2018-01-19 21:57       ` Junio C Hamano
2018-01-20 11:58         ` Thomas Gummerer
2018-01-22  6:14           ` Junio C Hamano
2018-01-27 12:18             ` Thomas Gummerer
2018-02-07 22:41             ` Junio C Hamano
2018-01-07 22:30     ` [PATCH v3 2/3] split-index: don't write cache tree with null oid entries Thomas Gummerer
2018-01-07 22:30     ` [PATCH v3 3/3] travis: run tests with GIT_TEST_SPLIT_INDEX Thomas Gummerer
2018-01-13 22:37     ` [PATCH v3 4/3] read-cache: don't try to write index if we can't write shared index Thomas Gummerer
2018-01-14  9:36       ` Duy Nguyen
2018-01-14 10:18         ` [PATCH 1/3] read-cache.c: change type of "temp" in write_shared_index() Nguyễn Thái Ngọc Duy
2018-01-14 10:18           ` [PATCH 2/3] read-cache.c: move tempfile creation/cleanup out of write_shared_index Nguyễn Thái Ngọc Duy
2018-01-14 10:18           ` Nguyễn Thái Ngọc Duy [this message]
2018-01-18 11:36             ` [PATCH 3/3] read-cache: don't write index twice if we can't write shared index SZEDER Gábor
2018-01-18 12:47               ` Duy Nguyen
2018-01-18 13:29                 ` Jeff King
2018-01-18 13:36                   ` Duy Nguyen
2018-01-18 15:00                     ` Duy Nguyen
2018-01-18 21:37                       ` Jeff King
2018-01-18 22:32                         ` SZEDER Gábor
2018-01-19  0:30                           ` Duy Nguyen
2018-01-22 13:32                           ` [PATCH 0/5] Travis CI: don't run the test suite as root in the 32 bit Linux build SZEDER Gábor
2018-01-22 13:32                             ` [PATCH 1/5] travis-ci: use 'set -x' for the commands under 'su' " SZEDER Gábor
2018-01-22 13:32                             ` [PATCH 2/5] travis-ci: use 'set -e' in the 32 bit Linux build job SZEDER Gábor
2018-01-23 16:26                               ` Jeff King
2018-01-23 16:32                                 ` Jeff King
2018-01-24 12:12                                 ` SZEDER Gábor
2018-01-24 15:49                                   ` Jeff King
2018-01-22 13:32                             ` [PATCH 3/5] travis-ci: don't repeat the path of the cache directory SZEDER Gábor
2018-01-23 16:30                               ` Jeff King
2018-01-24 13:14                                 ` SZEDER Gábor
2018-01-22 13:32                             ` [PATCH 4/5] travis-ci: don't run the test suite as root in the 32 bit Linux build SZEDER Gábor
2018-01-23 16:43                               ` Jeff King
2018-01-24 13:45                                 ` SZEDER Gábor
2018-01-24 15:56                                   ` Jeff King
2018-01-24 18:01                                     ` Jeff King
2018-01-24 19:51                                       ` Jeff King
2018-01-22 13:32                             ` [PATCH 5/5] travis-ci: don't fail if user already exists on 32 bit Linux build job SZEDER Gábor
2018-01-23 16:46                               ` Jeff King
2018-01-24  0:32                                 ` Duy Nguyen
2018-01-24 19:39                                 ` SZEDER Gábor
2018-01-22 18:27                 ` [PATCH 3/3] read-cache: don't write index twice if we can't write shared index SZEDER Gábor
2018-01-22 19:46                   ` Eric Sunshine
2018-01-22 22:10                     ` SZEDER Gábor
2018-01-24  9:11                   ` Duy Nguyen
2018-01-26 22:44                   ` Lars Schneider
2018-01-14 14:29         ` [PATCH v3 4/3] read-cache: don't try to write index " Thomas Gummerer
2018-01-18 21:53     ` [PATCH v3 0/3] fixes for split index mode Thomas Gummerer
2018-01-19 18:34       ` Junio C Hamano
2018-01-19 21:11         ` Thomas Gummerer
  -- strict thread matches above, loose matches on Subject: below --
2018-01-22 11:03 [PATCH 0/3] nd/shared-index-fix update Nguyễn Thái Ngọc Duy
2018-01-22 11:03 ` [PATCH 3/3] read-cache: don't write index twice if we can't write shared index Nguyễn Thái Ngọc Duy
2018-01-22 23:09   ` Junio C Hamano
2018-01-23  4:06     ` Duy Nguyen
2018-01-24  9:38     ` [PATCH 0/3] nd/shared-index-fix updates Nguyễn Thái Ngọc Duy
2018-01-24  9:38       ` [PATCH 3/3] read-cache: don't write index twice if we can't write shared index Nguyễn Thái Ngọc Duy

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=20180114101820.4273-3-pclouds@gmail.com \
    --to=pclouds@gmail.com \
    --cc=bmwill@google.com \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    --cc=larsxschneider@gmail.com \
    --cc=peff@peff.net \
    --cc=szeder.dev@gmail.com \
    --cc=t.gummerer@gmail.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).