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

Re: [PATCH 5/6] Teach "fsck" not to follow subproject links

From
Linus Torvalds <torvalds@linux-foundation.org>
Date
Apr 13, 2007, 15:23 UTC
Message-ID
<Pine.LNX.4.64.0704130816310.28042@woody.linux-foundation.org>
In-Reply-To
<461F46B5.2020007@dawes.za.net>
On Fri, 13 Apr 2007, Rogan Dawes wrote:
> 
> Would it not make sense for a hosting environment to say, if you are using
> alternates, or shared object directories, then you need to include *all* the
> refs in *all* the projects if you ever do an fsck?

Yes. And it shouldn't be hard to add support to do it. It's just not been done.

A lot of git programs already take refs on stdin, but fsck just doesn't do it (it can do it from the command line, but you'd run out of command line space very quickly).

More natural would be to just list all the git repos by git repo pathname (and there, usually the command line probably *is* long enough), but somebody would just have to do it. It's probably not very much code: just iterate over each repo both when adding refs and when actually doing the fsck itself.

> I'm not sure how well git will scale in this case, although it just should be
> a matter of how well git scales to dealing with a single project with tens of
> thousands of refs/tags/etc. The only problem might be in passing all those
> refs/tags to fsck in one go. STDIN, I guess?

For a real shared object directory, passing the refs to stdin (and teaching fsck about a "--stdin" flag) would be consistent with what we do for many other commands, so yes, that would work.

However, fsck actually tends to want not just the refs, but actually things like the index files and reflog files too, because those add other reachability info, which is why it's probably more natural to just give fsck the list of related repositories and let it figure them out.

That's also what you'd want to do for "alternates", since now there is no longer a single object directory either, but multiple separate (but related) ones.

Somebody would just have to write the code.. The basic rules are really all in "git/builtin-fsck.c": cmd_fsck(). Hint hint.

			Linus
Previous: Rogan DawesNext: Dana How
Message 27 of 98 in “Initial subproject support (RFC?)”
  1. 0/6 Initial subproject support (RFC?)Linus Torvalds, Apr 10, 2007
  2. 1/6 diff-lib: use ce_mode_from_stat() rather than messing with modes manuallyLinus Torvalds, Apr 10, 2007
  3. 2/6 Avoid overflowing name buffer in deep directory structuresLinus Torvalds, Apr 10, 2007
  4. 3/6 Add 'resolve_gitlink_ref()' helper functionLinus Torvalds, Apr 10, 2007
  5. Alex RiesenApr 10, 2007
  6. Linus TorvaldsApr 10, 2007
  7. Alex RiesenApr 10, 2007
  8. Linus TorvaldsApr 10, 2007
  9. Alex RiesenApr 10, 2007
  10. Linus TorvaldsApr 10, 2007
  11. Josef WeidendorferApr 10, 2007
  12. 4/6 Add "S_IFDIRLNK" file mode infrastructure for git linksLinus Torvalds, Apr 10, 2007
  13. 5/6 Teach "fsck" not to follow subproject linksLinus Torvalds, Apr 10, 2007
  14. Sam VilainApr 11, 2007
  15. Linus TorvaldsApr 11, 2007
  16. Sam VilainApr 11, 2007
  17. Linus TorvaldsApr 11, 2007
  18. David LangApr 11, 2007
  19. Linus TorvaldsApr 11, 2007
  20. David LangApr 11, 2007
  21. Linus TorvaldsApr 12, 2007
  22. Junio C HamanoApr 12, 2007
  23. David LangApr 12, 2007
  24. Dana HowApr 12, 2007
  25. Linus TorvaldsApr 12, 2007
  26. Rogan DawesApr 13, 2007
  27. Linus TorvaldsApr 13, 2007
  28. Dana HowApr 15, 2007
  29. Dana HowApr 12, 2007
  30. Sam VilainApr 12, 2007
  31. Junio C HamanoApr 12, 2007
  32. Linus TorvaldsApr 12, 2007
  33. Junio C HamanoApr 12, 2007
  34. Junio C HamanoApr 12, 2007
  35. Linus TorvaldsApr 12, 2007
  36. Dana HowApr 11, 2007
  37. 6/6 Teach core object handling functions about gitlinksLinus Torvalds, Apr 10, 2007
  38. Frank LichtenheldApr 10, 2007
  39. Alex RiesenApr 10, 2007
  40. Linus TorvaldsApr 10, 2007
  41. Josef WeidendorferApr 10, 2007
  42. Alex RiesenApr 10, 2007
  43. Josef WeidendorferApr 10, 2007
  44. Linus TorvaldsApr 10, 2007
  45. Andy ParkinsApr 10, 2007
  46. Linus TorvaldsApr 10, 2007
  47. Junio C HamanoApr 10, 2007
  48. Linus TorvaldsApr 10, 2007
  49. Sam VilainApr 12, 2007
  50. Martin WaitzApr 12, 2007
  51. Linus TorvaldsApr 12, 2007
  52. Sam VilainApr 12, 2007
  53. David LangApr 10, 2007
  54. Junio C HamanoApr 10, 2007
  55. Josef WeidendorferApr 10, 2007
  56. Linus TorvaldsApr 10, 2007
  57. Sam VilainApr 11, 2007
  58. Linus TorvaldsApr 12, 2007
  59. Torgil SvenssonApr 12, 2007
  60. Martin WaitzApr 12, 2007
  61. Torgil SvenssonApr 12, 2007
  62. Sam VilainApr 11, 2007
  63. Martin WaitzApr 11, 2007
  64. Alex RiesenApr 11, 2007
  65. Martin WaitzApr 11, 2007
  66. Alex RiesenApr 11, 2007
  67. Martin WaitzApr 11, 2007
  68. Junio C HamanoApr 11, 2007
  69. Martin WaitzApr 11, 2007
  70. Junio C HamanoApr 11, 2007
  71. Martin WaitzApr 11, 2007
  72. Linus TorvaldsApr 11, 2007
  73. Andy ParkinsApr 11, 2007
  74. Martin WaitzApr 11, 2007
  75. Linus TorvaldsApr 11, 2007
  76. Sam VilainApr 11, 2007
  77. Martin WaitzApr 11, 2007
  78. Brian GernhardtApr 12, 2007
  79. Josef WeidendorferApr 12, 2007
  80. Linus TorvaldsApr 10, 2007
  81. Alex RiesenApr 10, 2007
  82. Linus TorvaldsApr 10, 2007
  83. Alex RiesenApr 10, 2007
  84. Linus TorvaldsApr 10, 2007
  85. Alex RiesenApr 10, 2007
  86. Junio C HamanoApr 10, 2007
  87. Linus TorvaldsApr 10, 2007
  88. Junio C HamanoApr 10, 2007
  89. Sam RavnborgApr 10, 2007
  90. Junio C HamanoApr 10, 2007
  91. Nicolas PitreApr 10, 2007
  92. J. Bruce FieldsApr 15, 2007
  93. David KågedalApr 11, 2007
  94. Junio C HamanoApr 11, 2007
  95. J. Bruce FieldsApr 15, 2007
  96. Martin WaitzApr 11, 2007
  97. Alex RiesenApr 11, 2007
  98. Martin WaitzApr 11, 2007

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.