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

[PATCH 3/4] traverse_bitmap_commit_list(): don't free result

From
Jeff King <peff@peff.net>
Date
Sep 1, 2018, 07:49 UTC
Message-ID
<20180901074948.GC25461@sigill.intra.peff.net>
In-Reply-To
<20180901074145.GA24023@sigill.intra.peff.net>

Since it was introduced in fff42755ef (pack-bitmap: add support for bitmap indexes, 2013-12-21), this function has freed the result after traversing it. That is an artifact of the early days of the bitmap code, when we had a single static "struct bitmap_index". Back then, it was intended that you would do:

  prepare_bitmap_walk(&revs);
  traverse_bitmap_commit_list(&revs);

Since the actual bitmap_index struct was totally behind the scenes, it was convenient for traverse_bitmap_commit_list() to clean it up, clearing the way for another traversal.

But since 3ae5fa0768 (pack-bitmap: remove bitmap_git global variable, 2018-06-07), the caller explicitly manages the bitmap_index struct itself, like this:

  b = prepare_bitmap_walk(&revs);
  traverse_bitmap_commit_list(b, &revs);
  free_bitmap_index(b);

It no longer makes sense to auto-free the result after the traversal. If you want to do another traversal, you'd just create a new bitmap_index. And while nobody tries to call traverse_bitmap_commit_list() twice, the fact that it throws away the result might be surprising, and is better avoided.

Note that in the "old" way it was possible for two walks to amortize the cost of opening the on-disk .bitmap file (since it was stored in the global bitmap_index), but we lost that in 3ae5fa0768. However, no code actually does this, so it's not worth addressing now. The solution might involve a new:

  reset_bitmap_walk(b, &revs);

call. Or we might even attach the bitmap data to its matching packed_git struct, so that multiple prepare_bitmap_walk() calls could use it. That can wait until somebody actually has need of the optimization (and until then, we'll do the correct, unsurprising thing).

Signed-off-by: Jeff King <peff@peff.net>
---
 pack-bitmap.c | 3 ---
 1 file changed, 3 deletions(-)
diff --git a/pack-bitmap.c b/pack-bitmap.c
index 76fd93a3de..8c1af1cca2 100644
--- a/pack-bitmap.c
+++ b/pack-bitmap.c
@@ -848,9 +848,6 @@ void traverse_bitmap_commit_list(struct bitmap_index *bitmap_git,
 		OBJ_TAG, show_reachable);
 
 	show_extended_objects(bitmap_git, show_reachable);
-
-	bitmap_free(bitmap_git->result);
-	bitmap_git->result = NULL;
 }
 
 static uint32_t count_object_type(struct bitmap_index *bitmap_git,
-- 
2.19.0.rc1.549.g6ce4dc0f08
Previous: Jeff KingNext: Jeff King
Message 24 of 53 in “test-tool.h: include git-compat-util.h”
  1. test-tool.h: include git-compat-util.hJeff King, Aug 21, 2018
  2. Junio C HamanoAug 21, 2018
  3. 1/6 t/perf: factor boilerplate out of test_perfJeff King, Aug 21, 2018
  4. 2/6 t/perf: factor out percent calculationsJeff King, Aug 21, 2018
  5. 3/6 t/perf: add infrastructure for measuring sizesJeff King, Aug 21, 2018
  6. Derrick StoleeAug 22, 2018
  7. Jeff KingAug 22, 2018
  8. 4/6 t/perf: add perf tests for fetches from a bitmapped serverJeff King, Aug 21, 2018
  9. 5/6 pack-bitmap: save "have" bitmap from walkJeff King, Aug 21, 2018
  10. Derrick StoleeAug 21, 2018
  11. Jeff KingAug 21, 2018
  12. Ævar Arnfjörð BjarmasonAug 31, 2018
  13. Jeff KingAug 31, 2018
  14. 0/4 un-breaking pack-objects with bitmapsJeff King, Sep 1, 2018
  15. 1/4 bitmap_has_sha1_in_uninteresting(): drop BUG checkJeff King, Sep 1, 2018
  16. 2/4 t5310: test delta reuse with bitmapsJeff King, Sep 1, 2018
  17. Jeff KingSep 1, 2018
  18. Ævar Arnfjörð BjarmasonSep 1, 2018
  19. Ben PeartSep 1, 2018
  20. Jeff KingSep 2, 2018
  21. Stefan BellerSep 4, 2018
  22. Junio C HamanoSep 4, 2018
  23. Jeff KingSep 4, 2018
  24. 3/4 traverse_bitmap_commit_list(): don't free resultJeff King, Sep 1, 2018
  25. 4/4 pack-bitmap: drop "loaded" flagJeff King, Sep 1, 2018
  26. Stefan BellerSep 4, 2018
  27. Jeff KingSep 4, 2018
  28. Ævar Arnfjörð BjarmasonSep 8, 2018
  29. Junio C HamanoSep 10, 2018
  30. Jeff KingSep 10, 2018
  31. Junio C HamanoSep 10, 2018
  32. 6/6 pack-objects: reuse on-disk deltas for thin "have" objectsJeff King, Aug 21, 2018
  33. Junio C HamanoAug 21, 2018
  34. Junio C HamanoAug 21, 2018
  35. Jeff KingAug 21, 2018
  36. Jeff KingAug 21, 2018
  37. Junio C HamanoAug 21, 2018
  38. Jeff KingAug 21, 2018
  39. Junio C HamanoAug 21, 2018
  40. Jeff KingAug 21, 2018
  41. 0/9 trailer-parsing false positivesJeff King, Aug 23, 2018
  42. 1/9 trailer: use size_t for string offsetsJeff King, Aug 23, 2018
  43. 2/9 trailer: use size_t for iterating trailer listJeff King, Aug 23, 2018
  44. 3/9 trailer: pass process_trailer_opts to trailer_info_get()Jeff King, Aug 23, 2018
  45. 4/9 interpret-trailers: tighten check for "---" patch boundaryJeff King, Aug 23, 2018
  46. 5/9 interpret-trailers: allow suppressing "---" dividerJeff King, Aug 23, 2018
  47. 6/9 pretty, ref-filter: format %(trailers) with no_divider optionJeff King, Aug 23, 2018
  48. 7/9 sequencer: ignore "---" divider when parsing trailersJeff King, Aug 23, 2018
  49. 8/9 append_signoff: use size_t for string offsetsJeff King, Aug 23, 2018
  50. 9/9 sequencer: handle ignore_footer when parsing trailersJeff King, Aug 23, 2018
  51. Junio C HamanoAug 23, 2018
  52. Jeff KingAug 24, 2018
  53. Jeff KingAug 21, 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.