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

Re: [PATCH v6 28/32] config: read ref storage config on startup

From
David Turner <dturner@twopensource.com>
Date
Feb 29, 2016, 21:57 UTC
Message-ID
<1456783062.18017.65.camel@twopensource.com>
In-Reply-To
<20160227035634.GA10829@sigill.intra.peff.net>
On Fri, 2016-02-26 at 22:56 -0500, Jeff King wrote:
Show 73 quoted lines
> On Wed, Feb 24, 2016 at 05:59:00PM -0500, David Turner wrote:
> 
> > @@ -1207,6 +1208,29 @@ int git_config_early(config_fn_t fn, void
> > *data, const char *repo_config)
> >  	}
> >  
> >  	if (repo_config && !access_or_die(repo_config, R_OK, 0)) {
> > +		char *storage = NULL;
> > +
> > +		/*
> > +		 * make sure we always read the ref storage config
> > +		 * from the extensions section on startup
> > +		 */
> > +		ret +=
> > git_config_from_file(ref_storage_backend_config,
> > +					    repo_config,
> > &storage);
> > +
> > +		register_ref_storage_backends();
> > +		if (!storage)
> > +			storage = xstrdup("");
> > +
> > +		if (!*storage ||
> > +		    !strcmp(storage, "files")) {
> > +			/* default backend, nothing to do */
> > +			free(storage);
> > +		} else {
> > +			if
> > (set_ref_storage_backend(ref_storage_backend))
> > +				die(_("Unknown ref storage backend
> > %s"),
> > +				    ref_storage_backend);
> > +		}
> > +
> 
> Coverity complains that "storage" leaks here, and I think it does in
> the
> case that we non-default storage, and we successfully set up the
> backend. That's a pretty minor point.
>
> However, after looking at this code, I'm rather confused about a few
> things.
> 
> One is that we read the config and use ref_storage_backend_config[1]
> to
> store the string value in "storage", and then we check whether that
> is
> non-default, and if it is, I'd expect us to feed it to
> set_ref_storage_backend. But we don't; we feed ref_storage_backend
> instead!
> 
> What is that value? It looks like it is the string we set in
> check_repo_format when we load the extensions list there. So why
> are we re-reading the config here at all? Couldn't we just use
> ref_storage_backend in the first place?
>
> My second confusion is why this is happening in git_config_early().
> That
> function is called during the setup of
> check_repository_format_gently(),
> which is why I think you wanted to put the code here. But it's _also_
> called as part of a regular git_config(). Which means we're parsing
> the
> repo config and setting the ref backend all over again, every time we
> look at config for other reasons.
> 
> So I think this setup probably should be in
> check_repository_format_gently(), and should be able to trigger off
> of
> the existing ref_storage_backend string we've already saved (and we
> should bail immediately there if we don't know about the backend, as
> it
> means we _don't_ match the repo's extensions and cannot proceed).

We apparently don't always call check_repo_format before calling git_config_early -- or, more to the point, before doing ref operations. So I think we need this in git_config_early.

I'll fix up the memory safety/leak issues tho.
Previous: Jeff KingNext: Jeff King
Message 41 of 56 in “refs backend”
  1. 00/32 refs backendDavid Turner, Feb 24, 2016
  2. 01/32 refs: move head_ref{,_submodule} to the common codeDavid Turner, Feb 24, 2016
  3. 02/32 refs: move for_each_*ref* functions into common codeDavid Turner, Feb 24, 2016
  4. 03/32 files-backend: break out ref readingDavid Turner, Feb 24, 2016
  5. Junio C HamanoFeb 25, 2016
  6. Junio C HamanoFeb 25, 2016
  7. David TurnerFeb 25, 2016
  8. Junio C HamanoFeb 25, 2016
  9. 04/32 refs: move resolve_ref_unsafe into common codeDavid Turner, Feb 24, 2016
  10. Junio C HamanoFeb 25, 2016
  11. 05/32 refs: add a backend method structure with transaction functionsDavid Turner, Feb 24, 2016
  12. Jeff KingFeb 27, 2016
  13. David TurnerFeb 29, 2016
  14. 06/32 refs: add methods for misc ref operationsDavid Turner, Feb 24, 2016
  15. 07/32 refs: add method for do_for_each_refDavid Turner, Feb 24, 2016
  16. 08/32 refs: reduce the visibility of do_for_each_ref()David Turner, Feb 24, 2016
  17. 09/32 refs: add do_for_each_per_worktree_refDavid Turner, Feb 24, 2016
  18. 10/32 refs: add methods for reflogDavid Turner, Feb 24, 2016
  19. 11/32 refs: add method for initial ref transaction commitDavid Turner, Feb 24, 2016
  20. 12/32 refs: add method for delete_refsDavid Turner, Feb 24, 2016
  21. 13/32 refs: add methods to init refs dbDavid Turner, Feb 24, 2016
  22. 14/32 refs: add method to rename refsDavid Turner, Feb 24, 2016
  23. 15/32 refs: handle non-normal ref renamesDavid Turner, Feb 24, 2016
  24. 16/32 refs: make lock genericDavid Turner, Feb 24, 2016
  25. 17/32 refs: move duplicate check to common codeDavid Turner, Feb 24, 2016
  26. 18/32 refs: allow log-only updatesDavid Turner, Feb 24, 2016
  27. 19/32 refs: don't dereference on renameDavid Turner, Feb 24, 2016
  28. 20/32 refs: on symref reflog expire, lock symref not referrentDavid Turner, Feb 24, 2016
  29. 21/32 refs: resolve symbolic refs firstDavid Turner, Feb 24, 2016
  30. 22/32 refs: always handle non-normal refs in files backendDavid Turner, Feb 24, 2016
  31. 23/32 init: allow alternate ref strorage to be set for new reposDavid Turner, Feb 24, 2016
  32. 24/32 refs: check submodules' ref storage configDavid Turner, Feb 24, 2016
  33. 25/32 clone: allow ref storage backend to be set for cloneDavid Turner, Feb 24, 2016
  34. 26/32 svn: learn ref-storage argumentDavid Turner, Feb 24, 2016
  35. 27/32 refs: register ref storage backendsDavid Turner, Feb 24, 2016
  36. 28/32 config: read ref storage config on startupDavid Turner, Feb 24, 2016
  37. Jeff KingFeb 27, 2016
  38. Jeff KingFeb 27, 2016
  39. David TurnerFeb 29, 2016
  40. Jeff KingFeb 29, 2016
  41. David TurnerFeb 29, 2016
  42. Jeff KingFeb 29, 2016
  43. Jeff KingFeb 29, 2016
  44. David TurnerFeb 29, 2016
  45. 29/32 refs: break out resolve_ref_unsafe_submoduleDavid Turner, Feb 24, 2016
  46. 30/32 refs: add LMDB refs storage backendDavid Turner, Feb 24, 2016
  47. 31/32 refs: tests for lmdb backendDavid Turner, Feb 24, 2016
  48. 32/32 tests: add ref-storage argumentDavid Turner, Feb 24, 2016
  49. SZEDER GáborFeb 29, 2016
  50. David TurnerFeb 29, 2016
  51. Ramsay JonesFeb 25, 2016
  52. David TurnerFeb 25, 2016
  53. Duy NguyenFeb 25, 2016
  54. David TurnerFeb 25, 2016
  55. Stefan BellerFeb 25, 2016
  56. Duy NguyenFeb 25, 2016

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.