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

Re: [RFC PATCH v1 0/6] stash: drop usage of a second index

From
Alban Gruin <alban.gruin@gmail.com>
Date
Jun 17, 2020, 21:31 UTC
Message-ID
<b1da563c-6191-963e-6829-d30a15233d29@gmail.com>
In-Reply-To
<xmqq1rmdv4q9.fsf@gitster.c.googlers.com>
Hi,
Le 17/06/2020 à 22:04, Junio C Hamano a écrit :
Show 10 quoted lines
> SZEDER Gábor <szeder.dev@gmail.com> writes:
> 
>> On Mon, Jun 15, 2020 at 11:50:20PM +0200, SZEDER Gábor wrote:
>>> On Mon, Jun 15, 2020 at 05:27:15PM +0200, SZEDER Gábor wrote:
>>>>       - Should we even allow 'splitIndex.sharedIndexExpire=now'?
> 
> Good analysis.  The most conservative might end up to be to disable
> splitindex altogether but perhaps we can first set a reasonable
> minimum to the expiration to say 10min?
>

I tried to understand what write_locked_index() and its callees does, and I think there is an issue regarding how split-index files are handled:

 1. The new "main" index is written in a temporary file.  This temporary
file is either renamed to "index" if the COMMIT_LOCK flag is set, or
left as-is (cf. do_write_locked_index()).
 2. The split-index files are removed if they are too old, without
checking if the COMMIT_LOCK flag is set (cf. write_shared_index()).

This could lead to a situation where a split-index file is removed because it is deemed too old, but the main index is left as-is, still pointing to this file. I am afraid that this can be an issue, even when `splitIndex.sharedIndexExpire' is set to a "sane" value (which could obviously be exacerbated by `now').

Alban
Previous: Junio C HamanoNext: Alban Gruin
Message 23 of 44 in “stash: drop usage of a second index”
  1. 0/6 stash: drop usage of a second indexAlban Gruin, May 5, 2020
  2. 2/6 stash: remove the second index in stash_working_tree()Alban Gruin, May 5, 2020
  3. Christian CouderJun 13, 2020
  4. Alban GruinJun 13, 2020
  5. Christian CouderJun 15, 2020
  6. 1/6 stash: mark `i_tree' in reset_tree() constAlban Gruin, May 5, 2020
  7. Christian CouderJun 13, 2020
  8. 4/6 stash: remove the second index in save_untracked_files()Alban Gruin, May 5, 2020
  9. Christian CouderJun 13, 2020
  10. 5/6 stash: remove the second index in restore_untracked()Alban Gruin, May 5, 2020
  11. Christian CouderJun 13, 2020
  12. 3/6 stash: remove the second index in stash_patch()Alban Gruin, May 5, 2020
  13. Christian CouderJun 13, 2020
  14. Christian CouderJun 13, 2020
  15. 6/6 stash: remove `stash_index_path'Alban Gruin, May 5, 2020
  16. Alban GruinJun 4, 2020
  17. Christian CouderJun 13, 2020
  18. Alban GruinJun 25, 2020
  19. SZEDER GáborJun 15, 2020
  20. SZEDER GáborJun 15, 2020
  21. SZEDER GáborJun 16, 2020
  22. Junio C HamanoJun 17, 2020
  23. Alban GruinJun 17, 2020
  24. 0/6 stash: drop usage of a second indexAlban Gruin, Jun 30, 2020
  25. 1/6 stash: mark `i_tree' in reset_tree() constAlban Gruin, Jun 30, 2020
  26. 2/6 stash: remove the second index in stash_working_tree()Alban Gruin, Jun 30, 2020
  27. 4/6 stash: remove the second index in save_untracked_files()Alban Gruin, Jun 30, 2020
  28. 3/6 stash: remove the second index in stash_patch()Alban Gruin, Jun 30, 2020
  29. 5/6 stash: remove the second index in restore_untracked()Alban Gruin, Jun 30, 2020
  30. Christian CouderJul 31, 2020
  31. Alban GruinJul 31, 2020
  32. 6/6 stash: remove `stash_index_path'Alban Gruin, Jun 30, 2020
  33. Christian CouderJul 31, 2020
  34. 0/6 stash: drop usage of a second indexAlban Gruin, Jul 31, 2020
  35. 1/6 stash: mark `i_tree' in reset_tree() constAlban Gruin, Jul 31, 2020
  36. Junio C HamanoJul 31, 2020
  37. 2/6 stash: remove the second index in stash_working_tree()Alban Gruin, Jul 31, 2020
  38. Junio C HamanoJul 31, 2020
  39. Junio C HamanoAug 2, 2020
  40. 4/6 stash: remove the second index in save_untracked_files()Alban Gruin, Jul 31, 2020
  41. 5/6 stash: remove the second index in restore_untracked()Alban Gruin, Jul 31, 2020
  42. 3/6 stash: remove the second index in stash_patch()Alban Gruin, Jul 31, 2020
  43. 6/6 stash: remove `stash_index_path'Alban Gruin, Jul 31, 2020
  44. Junio C HamanoJul 31, 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.