git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH v2 3/4] commit-graph.c: write non-split graphs as read-only

From
Junio C Hamano <gitster@pobox.com>
Date
Apr 27, 2020, 23:54 UTC
Message-ID
<xmqqr1w85vtq.fsf@gitster.c.googlers.com>
In-Reply-To
<86cf29ce9c1e6dc1fc881458c18850c2893b092a.1588004647.git.me@ttaylorr.com>
Taylor Blau <me@ttaylorr.com> writes:
Show 35 quoted lines
> In the previous commit, Git learned 'hold_lock_file_for_update_mode' to
> allow the caller to specify the permission bits (prior to further
> adjustment by the umask and shared repository permissions) used when
> acquiring a temporary file.
>
> Use this in the commit-graph machinery for writing a non-split graph to
> acquire an opened temporary file with permissions read-only permissions
> to match the split behavior. (In the split case, Git uses
> git_mkstemp_mode' for each of the commit-graph layers with permission
> bits '0444').
>
> One can notice this discrepancy when moving a non-split graph to be part
> of a new chain. This causes a commit-graph chain where all layers have
> read-only permission bits, except for the base layer, which is writable
> for the current user.
>
> Resolve this discrepancy by using the new
> 'hold_lock_file_for_update_mode' and passing the desired permission
> bits.
>
> Doing so causes some test fallout in t5318 and t6600. In t5318, this
> occurs in tests that corrupt a commit-graph file by writing into it. For
> these, 'chmod u+w'-ing the file beforehand resolves the issue. The
> additional spot in 'corrupt_graph_verify' is necessary because of the
> extra 'git commit-graph write' beforehand (which *does* rewrite the
> commit-graph file). In t6600, this is caused by copying a read-only
> commit-graph file into place and then trying to replace it. For these,
> make these files writable.
>
> Signed-off-by: Taylor Blau <me@ttaylorr.com>
> ---
>  commit-graph.c          |  3 ++-
>  t/t5318-commit-graph.sh | 11 ++++++++++-
>  t/t6600-test-reach.sh   |  2 ++
>  3 files changed, 14 insertions(+), 2 deletions(-)

This step, queued as 3a5c7d70 (commit-graph.c: write non-split graphs as read-only, 2020-04-27), starts failing 5318#9 at least for me. Do we need to loosen umask while running this test to something not more strict than 022 or something silly like that?

Here is what I'll use as a workaround for today's pushout.
commit f062d1c028bcc839f961e8904c38c476b9deeec3
Author: Junio C Hamano <gitster@pobox.com>
Date:   Mon Apr 27 16:50:30 2020 -0700
    SQUASH??? force known umask if you are going to check the resulting mode bits
diff --git a/t/t5318-commit-graph.sh b/t/t5318-commit-graph.sh
index fb0aae61c3..901eb3ecfb 100755
--- a/t/t5318-commit-graph.sh
+++ b/t/t5318-commit-graph.sh
@@ -12,6 +12,10 @@ test_expect_success 'setup full repo' '
 	test_oid_init
 '
 
+test_expect_success POSIXPERM 'tweak umask for modebit tests' '
+	umask 022
+'
+
 test_expect_success 'verify graph with no graph file' '
 	cd "$TRASH_DIRECTORY/full" &&
 	git commit-graph verify
Previous: Taylor BlauNext: Taylor Blau
Message 15 of 36 in “commit-graph: write non-split graphs as read-only”
  1. 0/3 commit-graph: write non-split graphs as read-onlyTaylor Blau, Apr 20, 2020
  2. 1/3 tempfile.c: introduce 'create_tempfile_mode'Taylor Blau, Apr 20, 2020
  3. Junio C HamanoApr 20, 2020
  4. 2/3 lockfile.c: introduce 'hold_lock_file_for_update_mode'Taylor Blau, Apr 20, 2020
  5. 3/3 commit-graph.c: write non-split graphs as read-onlyTaylor Blau, Apr 20, 2020
  6. Junio C HamanoApr 20, 2020
  7. Taylor BlauApr 20, 2020
  8. Junio C HamanoApr 21, 2020
  9. Jeff KingApr 21, 2020
  10. Junio C HamanoApr 21, 2020
  11. 0/4 commit-graph: write non-split graphs as read-onlyTaylor Blau, Apr 27, 2020
  12. 1/4 tempfile.c: introduce 'create_tempfile_mode'Taylor Blau, Apr 27, 2020
  13. 2/4 lockfile.c: introduce 'hold_lock_file_for_update_mode'Taylor Blau, Apr 27, 2020
  14. 3/4 commit-graph.c: write non-split graphs as read-onlyTaylor Blau, Apr 27, 2020
  15. Junio C HamanoApr 27, 2020
  16. Taylor BlauApr 27, 2020
  17. Junio C HamanoApr 28, 2020
  18. Jeff KingApr 28, 2020
  19. Junio C HamanoApr 28, 2020
  20. Jeff KingApr 28, 2020
  21. Junio C HamanoApr 28, 2020
  22. Jeff KingApr 28, 2020
  23. Taylor BlauApr 28, 2020
  24. Jeff KingApr 28, 2020
  25. Junio C HamanoApr 28, 2020
  26. Derrick StoleeApr 29, 2020
  27. 4/4 commit-graph.c: ensure graph layers respect core.sharedRepositoryTaylor Blau, Apr 27, 2020
  28. Taylor BlauApr 27, 2020
  29. Jeff KingApr 27, 2020
  30. 0/5 commit-graph: write non-split graphs as read-onlyTaylor Blau, Apr 29, 2020
  31. 1/5 tempfile.c: introduce 'create_tempfile_mode'Taylor Blau, Apr 29, 2020
  32. 2/5 lockfile.c: introduce 'hold_lock_file_for_update_mode'Taylor Blau, Apr 29, 2020
  33. 3/5 commit-graph.c: write non-split graphs as read-onlyTaylor Blau, Apr 29, 2020
  34. 4/5 commit-graph.c: ensure graph layers respect core.sharedRepositoryTaylor Blau, Apr 29, 2020
  35. 5/5 commit-graph.c: make 'commit-graph-chain's read-onlyTaylor Blau, Apr 29, 2020
  36. Jeff KingMay 1, 2020

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.