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

Re: [PATCH 2/5] setup: detangle loading of loose object maps

From
Patrick Steinhardt <ps@pks.im>
Date
Aug 4, 2026, 07:21 UTC
Message-ID
<anGS-IOKHMo5VUJm@pks.im>
In-Reply-To
<xmqqh5lo6xi2.fsf@gitster.g>
On Fri, Jul 24, 2026 at 11:41:41AM -0700, Junio C Hamano wrote:
Show 18 quoted lines
> Patrick Steinhardt <ps@pks.im> writes:
> 
> > When a repository is configured to use a compatibility hash function
> > then we load the loose object map when we initialize the repository.
> > This object map provides the mappings between the canonical object hash
> > and the compatibility object hash.
> >
> > Loading the object map happens in `repo_set_compat_hash_algo()`, which
> > calls `repo_read_loose_object_map()` in case the compatibility object
> > hash is non-zero. This setup sequence has two major downsides:
> >
> >   - We assume that the primary object database is the "files" object
> >     database so that we can extract its "loose" backend. This stops
> >     working with pluggable object databases.
> 
> I am not sure if I understand this sentence, especially "we can
> extract its loose backend" part.  Do you mean 'extract the object
> map from the loose backend'?  Or something else?
Yeah, this is a bit awkward. Rewritten like this:
  - We assume that the primary object database is the "files" object
    database and unconditionally downcast it. This will BUG in case a
    different object database type was used together with a compat hash
    algorithm.
Show 18 quoted lines
> > @@ -112,14 +115,10 @@ int repo_read_loose_object_map(struct repository *repo)
> >  {
> >  	struct odb_source *source;
> >  
> > -	if (!should_use_loose_object_map(repo))
> > -		return 0;
> > -
> >  	odb_prepare_alternates(repo->objects);
> > -
> >  	for (source = repo->objects->sources; source; source = source->next) {
> >  		struct odb_source_files *files = odb_source_files_downcast(source);
> > -		if (load_one_loose_object_map(files->loose) < 0)
> > +		if (loose_object_map_load(files->loose) < 0)
> >  			return -1;
> 
> If this particular source in the list of sources is not backed by
> the files backend, would downcast signal the fact (e.g., by
> returning NULL) so that we can skip the next call instead?
No, the downcast will BUG in case it's not the "files" backend.
Show 10 quoted lines
> Or would the next step in refactoring be to define "load object map"
> method that is generic to odb_source so that this part does not have
> to do any of these and instead simply do
> 
> 	for (source = ...) {
> 		if (odb_source_object_map_load(source))
>                 	return -1;
>         }
> 
> or something?

This patch series is rather moving into the direction of making the object map an internal implementation detail. Ideally, callers shouldn't even have to be aware that such an object map exists. And by making the loose object source load it automatically we get closer to that state.

There's only one more caller that calls `repo_read_loose_object_map()` directly, in "object-file-convert.c", and that caller only calls it to reload the map in case a concurrent process may have rewritten it. If we make the backends handle this via `odb_source_prepare(FLUSH_CACHES)` then we could also get rid of that caller.

Patrick
Previous: Junio C HamanoNext: Justin Tobler
Message 9 of 68 in “odb: make creation of object database pluggable”
  1. 0/5 odb: make creation of object database pluggablePatrick Steinhardt, Jul 24, 2026
  2. 1/5 loose: load loose object map for the correct sourcePatrick Steinhardt, Jul 24, 2026
  3. Junio C HamanoJul 24, 2026
  4. Justin ToblerJul 28, 2026
  5. Toon ClaesJul 30, 2026
  6. Patrick SteinhardtAug 4, 2026
  7. 2/5 setup: detangle loading of loose object mapsPatrick Steinhardt, Jul 24, 2026
  8. Junio C HamanoJul 24, 2026
  9. Patrick SteinhardtAug 4, 2026
  10. Justin ToblerJul 28, 2026
  11. Toon ClaesJul 30, 2026
  12. Patrick SteinhardtAug 4, 2026
  13. 3/5 setup: defer object database creationPatrick Steinhardt, Jul 24, 2026
  14. Junio C HamanoJul 24, 2026
  15. Patrick SteinhardtAug 4, 2026
  16. Justin ToblerJul 28, 2026
  17. Patrick SteinhardtAug 4, 2026
  18. Patrick SteinhardtAug 4, 2026
  19. 4/5 odb/source: introduce function to map source type to namePatrick Steinhardt, Jul 24, 2026
  20. Junio C HamanoJul 26, 2026
  21. Patrick SteinhardtAug 4, 2026
  22. 5/5 odb: make creation of on-disk structures pluggablePatrick Steinhardt, Jul 24, 2026
  23. Junio C HamanoJul 26, 2026
  24. Patrick SteinhardtAug 4, 2026
  25. Justin ToblerJul 28, 2026
  26. 0/5 odb: make creation of object database pluggablePatrick Steinhardt, Aug 4, 2026
  27. 1/5 loose: load loose object map for the correct sourcePatrick Steinhardt, Aug 4, 2026
  28. 2/5 setup: detangle loading of loose object mapsPatrick Steinhardt, Aug 4, 2026
  29. 3/5 setup: defer object database creationPatrick Steinhardt, Aug 4, 2026
  30. Toon ClaesAug 4, 2026
  31. Patrick SteinhardtAug 5, 2026
  32. 4/5 odb/source: introduce function to map source type to namePatrick Steinhardt, Aug 4, 2026
  33. 5/5 odb: make creation of on-disk structures pluggablePatrick Steinhardt, Aug 4, 2026
  34. Justin ToblerAug 4, 2026
  35. 0/6 odb: make creation of object database pluggablePatrick Steinhardt, Aug 5, 2026
  36. 1/6 loose: load loose object map for the correct sourcePatrick Steinhardt, Aug 5, 2026
  37. 2/6 setup: detangle loading of loose object mapsPatrick Steinhardt, Aug 5, 2026
  38. 3/6 setup: handle ODB-related environment variables in `odb_new()`Patrick Steinhardt, Aug 5, 2026
  39. Toon ClaesAug 5, 2026
  40. Patrick SteinhardtAug 6, 2026
  41. 4/6 setup: defer object database creationPatrick Steinhardt, Aug 5, 2026
  42. Toon ClaesAug 5, 2026
  43. Patrick SteinhardtAug 6, 2026
  44. 5/6 odb/source: introduce function to map source type to namePatrick Steinhardt, Aug 5, 2026
  45. 6/6 odb: make creation of on-disk structures pluggablePatrick Steinhardt, Aug 5, 2026
  46. Toon ClaesAug 5, 2026
  47. 0/6 odb: make creation of object database pluggablePatrick Steinhardt, Aug 6, 2026
  48. 1/6 loose: load loose object map for the correct sourcePatrick Steinhardt, Aug 6, 2026
  49. 2/6 setup: detangle loading of loose object mapsPatrick Steinhardt, Aug 6, 2026
  50. 3/6 setup: handle ODB-related environment variables in `odb_new()`Patrick Steinhardt, Aug 6, 2026
  51. 4/6 setup: defer object database creationPatrick Steinhardt, Aug 6, 2026
  52. Toon ClaesAug 6, 2026
  53. Patrick SteinhardtAug 6, 2026
  54. Junio C HamanoAug 6, 2026
  55. 5/6 odb/source: introduce function to map source type to namePatrick Steinhardt, Aug 6, 2026
  56. 6/6 odb: make creation of on-disk structures pluggablePatrick Steinhardt, Aug 6, 2026
  57. Toon ClaesAug 6, 2026
  58. 0/6 odb: make creation of object database pluggablePatrick Steinhardt, Aug 7, 2026
  59. 1/6 loose: load loose object map for the correct sourcePatrick Steinhardt, Aug 7, 2026
  60. 2/6 setup: detangle loading of loose object mapsPatrick Steinhardt, Aug 7, 2026
  61. 3/6 setup: handle ODB-related environment variables in `odb_new()`Patrick Steinhardt, Aug 7, 2026
  62. 4/6 setup: defer object database creationPatrick Steinhardt, Aug 7, 2026
  63. Junio C HamanoAug 7, 2026
  64. Toon ClaesAug 7, 2026
  65. 5/6 odb/source: introduce function to map source type to namePatrick Steinhardt, Aug 7, 2026
  66. 6/6 odb: make creation of on-disk structures pluggablePatrick Steinhardt, Aug 7, 2026
  67. Toon ClaesAug 7, 2026
  68. Patrick SteinhardtAug 7, 2026

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.