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

Re: [PATCH 2/5] sequencer: always roll back lock in `do_recursive_merge()`

From
MÅMartin Ågren <martin.agren@gmail.com>
Date
Feb 27, 2018, 22:08 UTC
Message-ID
<CAN0heSr_4Sd5-yAUQ-SG+=NZrwFrAFqY66aYk0DgMEZ_s53iGQ@mail.gmail.com>
In-Reply-To
<20180227214430.GC6899@sigill.intra.peff.net>
On 27 February 2018 at 22:44, Jeff King <peff@peff.net> wrote:
Show 27 quoted lines
> I want to note one thing that confused me while reviewing. While looking
> to see if there were other returns, I noticed that the lines right near
> the end of your context are funny:
>
>         if (active_cache_changed &&
>               write_locked_index(&the_index, &index_lock, COMMIT_LOCK))
>                   /*
>                    * TRANSLATORS: %s will be "revert", "cherry-pick" or
>                    * "rebase -i".
>                    */
>                   return error(_("%s: Unable to write new index file"),
>                           _(action_name(opts)));
>           rollback_lock_file(&index_lock);
>
> At first I thought that rollback was a noop, since write_locked_index()
> would always either commit or rollback. But it's needed for the case
> when we active_cache_changed isn't true.
>
> So I think it's correct as-is, but I wonder if writing it as:
>
>   if (!active_cache_changed)
>         rollback_lock_file(&index_lock);
>   else if (write_locked_index(&the_index, &index_lock, COMMIT_LOCK))
>         return error(...);
>
> might be easier to follow. I'm OK with leaving it, too, but thought I'd
> mention it in case it confused other reviewers.

I also hesitated at that one. There are some similar instances elsewhere, e.g., in builtin/merge.c. There's also rerere.c, which does a variant of your suggestion.

Martin
Previous: Jeff KingNext: Jeff King
Message 5 of 26 in “roll back locks in various code paths”
  1. 0/5 roll back locks in various code pathsMartin Ågren, Feb 27, 2018
  2. 1/5 sequencer: make lockfiles non-staticMartin Ågren, Feb 27, 2018
  3. 2/5 sequencer: always roll back lock in `do_recursive_merge()`Martin Ågren, Feb 27, 2018
  4. Jeff KingFeb 27, 2018
  5. Martin ÅgrenFeb 27, 2018
  6. Jeff KingFeb 27, 2018
  7. 3/5 merge-recursive: always roll back lock in `merge_recursive_generic()`Martin Ågren, Feb 27, 2018
  8. Martin ÅgrenFeb 28, 2018
  9. 0/5 roll back locks in various code pathsMartin Ågren, Feb 28, 2018
  10. 1/5 sequencer: make lockfiles non-staticMartin Ågren, Feb 28, 2018
  11. 2/5 sequencer: always roll back lock in `do_recursive_merge()`Martin Ågren, Feb 28, 2018
  12. 5/5 sequencer: do not roll back lockfile unnecessarilyMartin Ågren, Feb 28, 2018
  13. 4/5 merge: always roll back lock in `checkout_fast_forward()`Martin Ågren, Feb 28, 2018
  14. 3/5 merge-recursive: always roll back lock in `merge_recursive_generic()`Martin Ågren, Feb 28, 2018
  15. Martin ÅgrenFeb 28, 2018
  16. Junio C HamanoFeb 28, 2018
  17. Martin ÅgrenMar 1, 2018
  18. Jeff KingMar 1, 2018
  19. Junio C HamanoMar 1, 2018
  20. Martin ÅgrenMar 1, 2018
  21. Jeff KingMar 2, 2018
  22. Jeff KingMar 1, 2018
  23. 4/5 merge: always roll back lock in `checkout_fast_forward()`Martin Ågren, Feb 27, 2018
  24. 5/5 sequencer: do not roll back lockfile unnecessarilyMartin Ågren, Feb 27, 2018
  25. Jeff KingFeb 27, 2018
  26. Junio C HamanoFeb 27, 2018

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.