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

[PATCH 16/19] midx: implement verification support for incremental MIDXs

From
Taylor Blau <me@ttaylorr.com>
Date
Jun 6, 2024, 23:05 UTC
Message-ID
<f729ed76121f8f38f71d5811cdeca0045e60ce8b.1717715060.git.me@ttaylorr.com>
In-Reply-To
<cover.1717715060.git.me@ttaylorr.com>

Teach the verification implementation used by `git multi-pack-index verify` to perform verification for incremental MIDX chains by independently validating each layer within the chain.

Signed-off-by: Taylor Blau <me@ttaylorr.com>
---
 midx.c | 47 ++++++++++++++++++++++++++++++-----------------
 midx.h |  2 ++
 2 files changed, 32 insertions(+), 17 deletions(-)
diff --git a/midx.c b/midx.c
index ac44fcefc2..ae3e30a062 100644
--- a/midx.c
+++ b/midx.c
@@ -467,6 +467,13 @@ int prepare_midx_pack(struct repository *r, struct multi_pack_index *m,
 	return 0;
 }
 
+struct packed_git *nth_midxed_pack(struct multi_pack_index *m,
+				   uint32_t pack_int_id)
+{
+	uint32_t local_pack_int_id = midx_for_pack(&m, pack_int_id);
+	return m->packs[local_pack_int_id];
+}
+
 #define MIDX_CHUNK_BITMAPPED_PACKS_WIDTH (2 * sizeof(uint32_t))
 
 int nth_bitmapped_pack(struct repository *r, struct multi_pack_index *m,
@@ -814,6 +821,7 @@ int verify_midx_file(struct repository *r, const char *object_dir, unsigned flag
 	uint32_t i;
 	struct progress *progress = NULL;
 	struct multi_pack_index *m = load_multi_pack_index(object_dir, 1);
+	struct multi_pack_index *curr;
 	verify_midx_error = 0;
 
 	if (!m) {
@@ -836,8 +844,8 @@ int verify_midx_file(struct repository *r, const char *object_dir, unsigned flag
 
 	if (flags & MIDX_PROGRESS)
 		progress = start_delayed_progress(_("Looking for referenced packfiles"),
-					  m->num_packs);
-	for (i = 0; i < m->num_packs; i++) {
+						  m->num_packs + m->num_packs_in_base);
+	for (i = 0; i < m->num_packs + m->num_packs_in_base; i++) {
 		if (prepare_midx_pack(r, m, i))
 			midx_report("failed to load pack in position %d", i);
 
@@ -857,17 +865,20 @@ int verify_midx_file(struct repository *r, const char *object_dir, unsigned flag
 	if (flags & MIDX_PROGRESS)
 		progress = start_sparse_progress(_("Verifying OID order in multi-pack-index"),
 						 m->num_objects - 1);
-	for (i = 0; i < m->num_objects - 1; i++) {
-		struct object_id oid1, oid2;
 
-		nth_midxed_object_oid(&oid1, m, i);
-		nth_midxed_object_oid(&oid2, m, i + 1);
+	for (curr = m; curr; curr = curr->base_midx) {
+		for (i = 0; i < m->num_objects - 1; i++) {
+			struct object_id oid1, oid2;
 
-		if (oidcmp(&oid1, &oid2) >= 0)
-			midx_report(_("oid lookup out of order: oid[%d] = %s >= %s = oid[%d]"),
-				    i, oid_to_hex(&oid1), oid_to_hex(&oid2), i + 1);
+			nth_midxed_object_oid(&oid1, m, m->num_objects_in_base + i);
+			nth_midxed_object_oid(&oid2, m, m->num_objects_in_base + i + 1);
 
-		midx_display_sparse_progress(progress, i + 1);
+			if (oidcmp(&oid1, &oid2) >= 0)
+				midx_report(_("oid lookup out of order: oid[%d] = %s >= %s = oid[%d]"),
+					    i, oid_to_hex(&oid1), oid_to_hex(&oid2), i + 1);
+
+			midx_display_sparse_progress(progress, i + 1);
+		}
 	}
 	stop_progress(&progress);
 
@@ -877,8 +888,8 @@ int verify_midx_file(struct repository *r, const char *object_dir, unsigned flag
 	 * each of the objects and only require 1 packfile to be open at a
 	 * time.
 	 */
-	ALLOC_ARRAY(pairs, m->num_objects);
-	for (i = 0; i < m->num_objects; i++) {
+	ALLOC_ARRAY(pairs, m->num_objects + m->num_objects_in_base);
+	for (i = 0; i < m->num_objects + m->num_objects_in_base; i++) {
 		pairs[i].pos = i;
 		pairs[i].pack_int_id = nth_midxed_pack_int_id(m, i);
 	}
@@ -892,16 +903,18 @@ int verify_midx_file(struct repository *r, const char *object_dir, unsigned flag
 
 	if (flags & MIDX_PROGRESS)
 		progress = start_sparse_progress(_("Verifying object offsets"), m->num_objects);
-	for (i = 0; i < m->num_objects; i++) {
+	for (i = 0; i < m->num_objects + m->num_objects_in_base; i++) {
 		struct object_id oid;
 		struct pack_entry e;
 		off_t m_offset, p_offset;
 
 		if (i > 0 && pairs[i-1].pack_int_id != pairs[i].pack_int_id &&
-		    m->packs[pairs[i-1].pack_int_id])
-		{
-			close_pack_fd(m->packs[pairs[i-1].pack_int_id]);
-			close_pack_index(m->packs[pairs[i-1].pack_int_id]);
+		    m->packs[pairs[i-1].pack_int_id]) {
+			uint32_t pack_int_id = pairs[i-1].pack_int_id;
+			struct packed_git *p = nth_midxed_pack(m, pack_int_id);
+
+			close_pack_fd(p);
+			close_pack_index(p);
 		}
 
 		nth_midxed_object_oid(&oid, m, pairs[i].pos);
diff --git a/midx.h b/midx.h
index 94de16a8c4..9d30935589 100644
--- a/midx.h
+++ b/midx.h
@@ -95,6 +95,8 @@ void get_split_midx_filename_ext(struct strbuf *buf, const char *object_dir,
 
 struct multi_pack_index *load_multi_pack_index(const char *object_dir, int local);
 int prepare_midx_pack(struct repository *r, struct multi_pack_index *m, uint32_t pack_int_id);
+struct packed_git *nth_midxed_pack(struct multi_pack_index *m,
+				   uint32_t pack_int_id);
 int nth_bitmapped_pack(struct repository *r, struct multi_pack_index *m,
 		       struct bitmapped_pack *bp, uint32_t pack_int_id);
 int bsearch_one_midx(const struct object_id *oid, struct multi_pack_index *m,
-- 
2.45.2.437.gecb9450a0e
Previous: Taylor BlauNext: Taylor Blau
Message 17 of 102 in “midx: incremental multi-pack indexes, part one”
  1. 00/19 midx: incremental multi-pack indexes, part oneTaylor Blau, Jun 6, 2024
  2. 01/19 Documentation: describe incremental MIDX formatTaylor Blau, Jun 6, 2024
  3. 02/19 midx: add new fields for incremental MIDX chainsTaylor Blau, Jun 6, 2024
  4. 03/19 midx: teach `nth_midxed_pack_int_id()` about incremental MIDXsTaylor Blau, Jun 6, 2024
  5. 04/19 midx: teach `prepare_midx_pack()` about incremental MIDXsTaylor Blau, Jun 6, 2024
  6. 05/19 midx: teach `nth_midxed_object_oid()` about incremental MIDXsTaylor Blau, Jun 6, 2024
  7. 06/19 midx: teach `nth_bitmapped_pack()` about incremental MIDXsTaylor Blau, Jun 6, 2024
  8. 07/19 midx: introduce `bsearch_one_midx()`Taylor Blau, Jun 6, 2024
  9. 08/19 midx: teach `bsearch_midx()` about incremental MIDXsTaylor Blau, Jun 6, 2024
  10. 09/19 midx: teach `nth_midxed_offset()` about incremental MIDXsTaylor Blau, Jun 6, 2024
  11. 10/19 midx: teach `fill_midx_entry()` about incremental MIDXsTaylor Blau, Jun 6, 2024
  12. 11/19 midx: remove unused `midx_locate_pack()`Taylor Blau, Jun 6, 2024
  13. 12/19 midx: teach `midx_contains_pack()` about incremental MIDXsTaylor Blau, Jun 6, 2024
  14. 13/19 midx: teach `midx_preferred_pack()` about incremental MIDXsTaylor Blau, Jun 6, 2024
  15. 14/19 midx: teach `midx_fanout_add_midx_fanout()` about incremental MIDXsTaylor Blau, Jun 6, 2024
  16. 15/19 midx: support reading incremental MIDX chainsTaylor Blau, Jun 6, 2024
  17. 16/19 midx: implement verification support for incremental MIDXsTaylor Blau, Jun 6, 2024
  18. 17/19 t: retire 'GIT_TEST_MULTI_PACK_INDEX_WRITE_BITMAP'Taylor Blau, Jun 6, 2024
  19. 18/19 t/t5313-pack-bounds-checks.sh: prepare for sub-directoriesTaylor Blau, Jun 6, 2024
  20. 19/19 midx: implement support for writing incremental MIDX chainsTaylor Blau, Jun 6, 2024
  21. Taylor BlauJun 6, 2024
  22. Junio C HamanoJun 7, 2024
  23. Taylor BlauJun 7, 2024
  24. Junio C HamanoJun 7, 2024
  25. Taylor BlauJun 7, 2024
  26. Junio C HamanoJun 25, 2024
  27. Elijah NewrenJun 26, 2024
  28. 00/19 midx: incremental multi-pack indexes, part oneTaylor Blau, Jul 17, 2024
  29. 01/19 Documentation: describe incremental MIDX formatTaylor Blau, Jul 17, 2024
  30. Jeff KingAug 1, 2024
  31. Taylor BlauAug 1, 2024
  32. 02/19 midx: add new fields for incremental MIDX chainsTaylor Blau, Jul 17, 2024
  33. Jeff KingAug 1, 2024
  34. Taylor BlauAug 1, 2024
  35. 03/19 midx: teach `nth_midxed_pack_int_id()` about incremental MIDXsTaylor Blau, Jul 17, 2024
  36. Jeff KingAug 1, 2024
  37. Taylor BlauAug 1, 2024
  38. 04/19 midx: teach `prepare_midx_pack()` about incremental MIDXsTaylor Blau, Jul 17, 2024
  39. Jeff KingAug 1, 2024
  40. Taylor BlauAug 1, 2024
  41. 05/19 midx: teach `nth_midxed_object_oid()` about incremental MIDXsTaylor Blau, Jul 17, 2024
  42. Jeff KingAug 1, 2024
  43. Taylor BlauAug 1, 2024
  44. 06/19 midx: teach `nth_bitmapped_pack()` about incremental MIDXsTaylor Blau, Jul 17, 2024
  45. Jeff KingAug 1, 2024
  46. Taylor BlauAug 1, 2024
  47. 07/19 midx: introduce `bsearch_one_midx()`Taylor Blau, Jul 17, 2024
  48. Jeff KingAug 1, 2024
  49. Taylor BlauAug 1, 2024
  50. 08/19 midx: teach `bsearch_midx()` about incremental MIDXsTaylor Blau, Jul 17, 2024
  51. Jeff KingAug 1, 2024
  52. 09/19 midx: teach `nth_midxed_offset()` about incremental MIDXsTaylor Blau, Jul 17, 2024
  53. Jeff KingAug 1, 2024
  54. 10/19 midx: teach `fill_midx_entry()` about incremental MIDXsTaylor Blau, Jul 17, 2024
  55. Jeff KingAug 1, 2024
  56. Taylor BlauAug 1, 2024
  57. 11/19 midx: remove unused `midx_locate_pack()`Taylor Blau, Jul 17, 2024
  58. Jeff KingAug 1, 2024
  59. Taylor BlauAug 1, 2024
  60. 12/19 midx: teach `midx_contains_pack()` about incremental MIDXsTaylor Blau, Jul 17, 2024
  61. Jeff KingAug 1, 2024
  62. 13/19 midx: teach `midx_preferred_pack()` about incremental MIDXsTaylor Blau, Jul 17, 2024
  63. Jeff KingAug 1, 2024
  64. Taylor BlauAug 1, 2024
  65. 14/19 midx: teach `midx_fanout_add_midx_fanout()` about incremental MIDXsTaylor Blau, Jul 17, 2024
  66. Jeff KingAug 1, 2024
  67. Taylor BlauAug 1, 2024
  68. 15/19 midx: support reading incremental MIDX chainsTaylor Blau, Jul 17, 2024
  69. Jeff KingAug 1, 2024
  70. Taylor BlauAug 1, 2024
  71. 16/19 midx: implement verification support for incremental MIDXsTaylor Blau, Jul 17, 2024
  72. Jeff KingAug 1, 2024
  73. 17/19 t: retire 'GIT_TEST_MULTI_PACK_INDEX_WRITE_BITMAP'Taylor Blau, Jul 17, 2024
  74. Jeff KingAug 1, 2024
  75. Taylor BlauAug 1, 2024
  76. 18/19 t/t5313-pack-bounds-checks.sh: prepare for sub-directoriesTaylor Blau, Jul 17, 2024
  77. 19/19 midx: implement support for writing incremental MIDX chainsTaylor Blau, Jul 17, 2024
  78. Jeff KingAug 1, 2024
  79. Taylor BlauAug 1, 2024
  80. Jeff KingAug 1, 2024
  81. Taylor BlauAug 1, 2024
  82. 00/19 midx: incremental multi-pack indexes, part oneTaylor Blau, Aug 6, 2024
  83. 01/19 Documentation: describe incremental MIDX formatTaylor Blau, Aug 6, 2024
  84. 02/19 midx: add new fields for incremental MIDX chainsTaylor Blau, Aug 6, 2024
  85. 03/19 midx: teach `nth_midxed_pack_int_id()` about incremental MIDXsTaylor Blau, Aug 6, 2024
  86. 04/19 midx: teach `prepare_midx_pack()` about incremental MIDXsTaylor Blau, Aug 6, 2024
  87. 05/19 midx: teach `nth_midxed_object_oid()` about incremental MIDXsTaylor Blau, Aug 6, 2024
  88. 06/19 midx: teach `nth_bitmapped_pack()` about incremental MIDXsTaylor Blau, Aug 6, 2024
  89. 07/19 midx: introduce `bsearch_one_midx()`Taylor Blau, Aug 6, 2024
  90. 08/19 midx: teach `bsearch_midx()` about incremental MIDXsTaylor Blau, Aug 6, 2024
  91. 09/19 midx: teach `nth_midxed_offset()` about incremental MIDXsTaylor Blau, Aug 6, 2024
  92. 10/19 midx: teach `fill_midx_entry()` about incremental MIDXsTaylor Blau, Aug 6, 2024
  93. 11/19 midx: remove unused `midx_locate_pack()`Taylor Blau, Aug 6, 2024
  94. 12/19 midx: teach `midx_contains_pack()` about incremental MIDXsTaylor Blau, Aug 6, 2024
  95. 13/19 midx: teach `midx_preferred_pack()` about incremental MIDXsTaylor Blau, Aug 6, 2024
  96. 14/19 midx: teach `midx_fanout_add_midx_fanout()` about incremental MIDXsTaylor Blau, Aug 6, 2024
  97. 15/19 midx: support reading incremental MIDX chainsTaylor Blau, Aug 6, 2024
  98. 16/19 midx: implement verification support for incremental MIDXsTaylor Blau, Aug 6, 2024
  99. 17/19 t: retire 'GIT_TEST_MULTI_PACK_INDEX_WRITE_BITMAP'Taylor Blau, Aug 6, 2024
  100. 18/19 t/t5313-pack-bounds-checks.sh: prepare for sub-directoriesTaylor Blau, Aug 6, 2024
  101. 19/19 midx: implement support for writing incremental MIDX chainsTaylor Blau, Aug 6, 2024
  102. Jeff KingAug 12, 2024

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.