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

[PATCH v2 13/17] commit-graph: use chunk-format read API

From
Derrick Stolee via GitGitGadget <gitgitgadget@gmail.com>
Date
Jan 27, 2021, 15:01 UTC
Message-ID
<8744d2785965773a2b561e9e1e91170530d052b1.1611759716.git.gitgitgadget@gmail.com>
In-Reply-To
<pull.848.v2.git.1611759716.gitgitgadget@gmail.com>
From: Derrick Stolee <dstolee@microsoft.com>

Instead of parsing the table of contents directly, use the chunk-format API methods read_table_of_contents() and pair_chunk(). While the current implementation loses the duplicate-chunk detection, that will be added in a future change.

Signed-off-by: Derrick Stolee <dstolee@microsoft.com>
---
 commit-graph.c          | 154 ++++++++++++++--------------------------
 t/t5318-commit-graph.sh |   2 +-
 2 files changed, 53 insertions(+), 103 deletions(-)
diff --git a/commit-graph.c b/commit-graph.c
index ba33777dcb8..8aa4881d85d 100644
--- a/commit-graph.c
+++ b/commit-graph.c
@@ -299,15 +299,43 @@ static int verify_commit_graph_lite(struct commit_graph *g)
 	return 0;
 }
 
+static int graph_read_oid_lookup(const unsigned char *chunk_start,
+				 size_t chunk_size, void *data)
+{
+	struct commit_graph *g = data;
+	g->chunk_oid_lookup = chunk_start;
+	g->num_commits = chunk_size / g->hash_len;
+	return 0;
+}
+
+static int graph_read_bloom_data(const unsigned char *chunk_start,
+				  size_t chunk_size, void *data)
+{
+	struct commit_graph *g = data;
+	uint32_t hash_version;
+	g->chunk_bloom_data = chunk_start;
+	hash_version = get_be32(chunk_start);
+
+	if (hash_version != 1)
+		return 0;
+
+	g->bloom_filter_settings = xmalloc(sizeof(struct bloom_filter_settings));
+	g->bloom_filter_settings->hash_version = hash_version;
+	g->bloom_filter_settings->num_hashes = get_be32(chunk_start + 4);
+	g->bloom_filter_settings->bits_per_entry = get_be32(chunk_start + 8);
+	g->bloom_filter_settings->max_changed_paths = DEFAULT_BLOOM_MAX_CHANGES;
+
+	return 0;
+}
+
 struct commit_graph *parse_commit_graph(struct repository *r,
 					void *graph_map, size_t graph_size)
 {
-	const unsigned char *data, *chunk_lookup;
-	uint32_t i;
+	const unsigned char *data;
 	struct commit_graph *graph;
-	uint64_t next_chunk_offset;
 	uint32_t graph_signature;
 	unsigned char graph_version, hash_version;
+	struct chunkfile *cf = NULL;
 
 	if (!graph_map)
 		return NULL;
@@ -356,108 +384,28 @@ struct commit_graph *parse_commit_graph(struct repository *r,
 		return NULL;
 	}
 
-	chunk_lookup = data + 8;
-	next_chunk_offset = get_be64(chunk_lookup + 4);
-	for (i = 0; i < graph->num_chunks; i++) {
-		uint32_t chunk_id;
-		uint64_t chunk_offset = next_chunk_offset;
-		int chunk_repeated = 0;
-
-		chunk_id = get_be32(chunk_lookup + 0);
-
-		chunk_lookup += GRAPH_CHUNKLOOKUP_WIDTH;
-		next_chunk_offset = get_be64(chunk_lookup + 4);
-
-		if (chunk_offset > graph_size - the_hash_algo->rawsz) {
-			error(_("commit-graph improper chunk offset %08x%08x"), (uint32_t)(chunk_offset >> 32),
-			      (uint32_t)chunk_offset);
-			goto free_and_return;
-		}
-
-		switch (chunk_id) {
-		case GRAPH_CHUNKID_OIDFANOUT:
-			if (graph->chunk_oid_fanout)
-				chunk_repeated = 1;
-			else
-				graph->chunk_oid_fanout = (uint32_t*)(data + chunk_offset);
-			break;
-
-		case GRAPH_CHUNKID_OIDLOOKUP:
-			if (graph->chunk_oid_lookup)
-				chunk_repeated = 1;
-			else {
-				graph->chunk_oid_lookup = data + chunk_offset;
-				graph->num_commits = (next_chunk_offset - chunk_offset)
-						     / graph->hash_len;
-			}
-			break;
+	cf = init_chunkfile(NULL);
 
-		case GRAPH_CHUNKID_DATA:
-			if (graph->chunk_commit_data)
-				chunk_repeated = 1;
-			else
-				graph->chunk_commit_data = data + chunk_offset;
-			break;
-
-		case GRAPH_CHUNKID_GENERATION_DATA:
-			if (graph->chunk_generation_data)
-				chunk_repeated = 1;
-			else
-				graph->chunk_generation_data = data + chunk_offset;
-			break;
-
-		case GRAPH_CHUNKID_GENERATION_DATA_OVERFLOW:
-			if (graph->chunk_generation_data_overflow)
-				chunk_repeated = 1;
-			else
-				graph->chunk_generation_data_overflow = data + chunk_offset;
-			break;
-
-		case GRAPH_CHUNKID_EXTRAEDGES:
-			if (graph->chunk_extra_edges)
-				chunk_repeated = 1;
-			else
-				graph->chunk_extra_edges = data + chunk_offset;
-			break;
-
-		case GRAPH_CHUNKID_BASE:
-			if (graph->chunk_base_graphs)
-				chunk_repeated = 1;
-			else
-				graph->chunk_base_graphs = data + chunk_offset;
-			break;
-
-		case GRAPH_CHUNKID_BLOOMINDEXES:
-			if (graph->chunk_bloom_indexes)
-				chunk_repeated = 1;
-			else if (r->settings.commit_graph_read_changed_paths)
-				graph->chunk_bloom_indexes = data + chunk_offset;
-			break;
-
-		case GRAPH_CHUNKID_BLOOMDATA:
-			if (graph->chunk_bloom_data)
-				chunk_repeated = 1;
-			else if (r->settings.commit_graph_read_changed_paths) {
-				uint32_t hash_version;
-				graph->chunk_bloom_data = data + chunk_offset;
-				hash_version = get_be32(data + chunk_offset);
-
-				if (hash_version != 1)
-					break;
+	if (read_table_of_contents(cf, graph->data, graph_size,
+				   GRAPH_HEADER_SIZE, graph->num_chunks))
+		goto free_and_return;
 
-				graph->bloom_filter_settings = xmalloc(sizeof(struct bloom_filter_settings));
-				graph->bloom_filter_settings->hash_version = hash_version;
-				graph->bloom_filter_settings->num_hashes = get_be32(data + chunk_offset + 4);
-				graph->bloom_filter_settings->bits_per_entry = get_be32(data + chunk_offset + 8);
-				graph->bloom_filter_settings->max_changed_paths = DEFAULT_BLOOM_MAX_CHANGES;
-			}
-			break;
-		}
+	pair_chunk(cf, GRAPH_CHUNKID_OIDFANOUT,
+		   (const unsigned char **)&graph->chunk_oid_fanout);
+	read_chunk(cf, GRAPH_CHUNKID_OIDLOOKUP, graph_read_oid_lookup, graph);
+	pair_chunk(cf, GRAPH_CHUNKID_DATA, &graph->chunk_commit_data);
+	pair_chunk(cf, GRAPH_CHUNKID_EXTRAEDGES, &graph->chunk_extra_edges);
+	pair_chunk(cf, GRAPH_CHUNKID_BASE, &graph->chunk_base_graphs);
+	pair_chunk(cf, GRAPH_CHUNKID_GENERATION_DATA,
+		   &graph->chunk_generation_data);
+	pair_chunk(cf, GRAPH_CHUNKID_GENERATION_DATA_OVERFLOW,
+		   &graph->chunk_generation_data_overflow);
 
-		if (chunk_repeated) {
-			error(_("commit-graph chunk id %08x appears multiple times"), chunk_id);
-			goto free_and_return;
-		}
+	if (r->settings.commit_graph_read_changed_paths) {
+		pair_chunk(cf, GRAPH_CHUNKID_BLOOMINDEXES,
+			   &graph->chunk_bloom_indexes);
+		read_chunk(cf, GRAPH_CHUNKID_BLOOMDATA,
+			   graph_read_bloom_data, graph);
 	}
 
 	if (graph->chunk_bloom_indexes && graph->chunk_bloom_data) {
@@ -474,9 +422,11 @@ struct commit_graph *parse_commit_graph(struct repository *r,
 	if (verify_commit_graph_lite(graph))
 		goto free_and_return;
 
+	free_chunkfile(cf);
 	return graph;
 
 free_and_return:
+	free_chunkfile(cf);
 	free(graph->bloom_filter_settings);
 	free(graph);
 	return NULL;
diff --git a/t/t5318-commit-graph.sh b/t/t5318-commit-graph.sh
index fa27df579a5..c7da741284e 100755
--- a/t/t5318-commit-graph.sh
+++ b/t/t5318-commit-graph.sh
@@ -564,7 +564,7 @@ test_expect_success 'detect bad hash version' '
 
 test_expect_success 'detect low chunk count' '
 	corrupt_graph_and_verify $GRAPH_BYTE_CHUNK_COUNT "\01" \
-		"missing the .* chunk"
+		"final chunk has non-zero id"
 '
 
 test_expect_success 'detect missing OID fanout chunk' '
-- 
gitgitgadget
Previous: Derrick Stolee via GitGitGadgetNext: Derrick Stolee via GitGitGadget
Message 51 of 120 in “Refactor chunk-format into an API”
  1. 00/17 Refactor chunk-format into an APIDerrick Stolee via GitGitGadget, Jan 26, 2021
  2. 02/17 chunk-format: create chunk format write APIDerrick Stolee via GitGitGadget, Jan 26, 2021
  3. Taylor BlauJan 27, 2021
  4. Derrick StoleeJan 27, 2021
  5. 01/17 commit-graph: anonymize data in chunk_write_fnDerrick Stolee via GitGitGadget, Jan 26, 2021
  6. Chris TorekJan 27, 2021
  7. Taylor BlauJan 27, 2021
  8. 05/17 midx: use context in write_midx_pack_names()Derrick Stolee via GitGitGadget, Jan 26, 2021
  9. 06/17 midx: add entries to write_midx_contextDerrick Stolee via GitGitGadget, Jan 26, 2021
  10. 04/17 midx: rename pack_info to write_midx_contextDerrick Stolee via GitGitGadget, Jan 26, 2021
  11. Taylor BlauJan 27, 2021
  12. 03/17 commit-graph: use chunk-format write APIDerrick Stolee via GitGitGadget, Jan 26, 2021
  13. Taylor BlauJan 27, 2021
  14. 14/17 midx: use chunk-format read APIDerrick Stolee via GitGitGadget, Jan 26, 2021
  15. Taylor BlauJan 27, 2021
  16. Derrick StoleeJan 27, 2021
  17. 17/17 chunk-format: add technical docsDerrick Stolee via GitGitGadget, Jan 26, 2021
  18. 11/17 midx: use chunk-format API in write_midx_internal()Derrick Stolee via GitGitGadget, Jan 26, 2021
  19. 16/17 chunk-format: restore duplicate chunk checksDerrick Stolee via GitGitGadget, Jan 26, 2021
  20. 15/17 midx: use 64-bit multiplication for chunk sizesDerrick Stolee via GitGitGadget, Jan 26, 2021
  21. 07/17 midx: add pack_perm to write_midx_contextDerrick Stolee via GitGitGadget, Jan 26, 2021
  22. 10/17 midx: drop chunk progress during writeDerrick Stolee via GitGitGadget, Jan 26, 2021
  23. 09/17 midx: return success/failure in chunk write methodsDerrick Stolee via GitGitGadget, Jan 26, 2021
  24. 08/17 midx: add num_large_offsets to write_midx_contextDerrick Stolee via GitGitGadget, Jan 26, 2021
  25. 13/17 commit-graph: use chunk-format read APIDerrick Stolee via GitGitGadget, Jan 26, 2021
  26. 12/17 chunk-format: create read chunk APIDerrick Stolee via GitGitGadget, Jan 26, 2021
  27. Taylor BlauJan 27, 2021
  28. Taylor BlauJan 27, 2021
  29. Junio C HamanoJan 26, 2021
  30. 00/17 Refactor chunk-format into an APIDerrick Stolee via GitGitGadget, Jan 27, 2021
  31. 05/17 midx: use context in write_midx_pack_names()Derrick Stolee via GitGitGadget, Jan 27, 2021
  32. 03/17 commit-graph: use chunk-format write APIDerrick Stolee via GitGitGadget, Jan 27, 2021
  33. 07/17 midx: add pack_perm to write_midx_contextDerrick Stolee via GitGitGadget, Jan 27, 2021
  34. 02/17 chunk-format: create chunk format write APIDerrick Stolee via GitGitGadget, Jan 27, 2021
  35. Junio C HamanoFeb 4, 2021
  36. Junio C HamanoFeb 4, 2021
  37. Derrick StoleeFeb 5, 2021
  38. Junio C HamanoFeb 5, 2021
  39. 15/17 midx: use 64-bit multiplication for chunk sizesDerrick Stolee via GitGitGadget, Jan 27, 2021
  40. Junio C HamanoFeb 5, 2021
  41. Chris TorekFeb 5, 2021
  42. Junio C HamanoFeb 5, 2021
  43. Chris TorekFeb 6, 2021
  44. Derrick StoleeFeb 5, 2021
  45. Junio C HamanoFeb 5, 2021
  46. SZEDER GáborFeb 7, 2021
  47. Junio C HamanoFeb 8, 2021
  48. 17/17 chunk-format: add technical docsDerrick Stolee via GitGitGadget, Jan 27, 2021
  49. Junio C HamanoFeb 5, 2021
  50. 14/17 midx: use chunk-format read APIDerrick Stolee via GitGitGadget, Jan 27, 2021
  51. 13/17 commit-graph: use chunk-format read APIDerrick Stolee via GitGitGadget, Jan 27, 2021
  52. 16/17 chunk-format: restore duplicate chunk checksDerrick Stolee via GitGitGadget, Jan 27, 2021
  53. Junio C HamanoFeb 5, 2021
  54. Derrick StoleeFeb 5, 2021
  55. 11/17 midx: use chunk-format API in write_midx_internal()Derrick Stolee via GitGitGadget, Jan 27, 2021
  56. 12/17 chunk-format: create read chunk APIDerrick Stolee via GitGitGadget, Jan 27, 2021
  57. Junio C HamanoFeb 4, 2021
  58. Derrick StoleeFeb 5, 2021
  59. Junio C HamanoFeb 5, 2021
  60. Junio C HamanoFeb 8, 2021
  61. Derrick StoleeFeb 9, 2021
  62. Junio C HamanoFeb 9, 2021
  63. 09/17 midx: return success/failure in chunk write methodsDerrick Stolee via GitGitGadget, Jan 27, 2021
  64. Junio C HamanoFeb 4, 2021
  65. Derrick StoleeFeb 5, 2021
  66. 10/17 midx: drop chunk progress during writeDerrick Stolee via GitGitGadget, Jan 27, 2021
  67. 06/17 midx: add entries to write_midx_contextDerrick Stolee via GitGitGadget, Jan 27, 2021
  68. 08/17 midx: add num_large_offsets to write_midx_contextDerrick Stolee via GitGitGadget, Jan 27, 2021
  69. 04/17 midx: rename pack_info to write_midx_contextDerrick Stolee via GitGitGadget, Jan 27, 2021
  70. 01/17 commit-graph: anonymize data in chunk_write_fnDerrick Stolee via GitGitGadget, Jan 27, 2021
  71. Taylor BlauJan 27, 2021
  72. Junio C HamanoFeb 5, 2021
  73. Derrick StoleeFeb 5, 2021
  74. 00/17 Refactor chunk-format into an APIDerrick Stolee via GitGitGadget, Feb 5, 2021
  75. 10/17 midx: drop chunk progress during writeDerrick Stolee via GitGitGadget, Feb 5, 2021
  76. 17/17 chunk-format: add technical docsDerrick Stolee via GitGitGadget, Feb 5, 2021
  77. 09/17 midx: return success/failure in chunk write methodsDerrick Stolee via GitGitGadget, Feb 5, 2021
  78. 15/17 midx: use 64-bit multiplication for chunk sizesDerrick Stolee via GitGitGadget, Feb 5, 2021
  79. 02/17 chunk-format: create chunk format write APIDerrick Stolee via GitGitGadget, Feb 5, 2021
  80. SZEDER GáborFeb 7, 2021
  81. Derrick StoleeFeb 8, 2021
  82. SZEDER GáborFeb 11, 2021
  83. 08/17 midx: add num_large_offsets to write_midx_contextDerrick Stolee via GitGitGadget, Feb 5, 2021
  84. 05/17 midx: use context in write_midx_pack_names()Derrick Stolee via GitGitGadget, Feb 5, 2021
  85. 14/17 midx: use chunk-format read APIDerrick Stolee via GitGitGadget, Feb 5, 2021
  86. 07/17 midx: add pack_perm to write_midx_contextDerrick Stolee via GitGitGadget, Feb 5, 2021
  87. 06/17 midx: add entries to write_midx_contextDerrick Stolee via GitGitGadget, Feb 5, 2021
  88. 16/17 chunk-format: restore duplicate chunk checksDerrick Stolee via GitGitGadget, Feb 5, 2021
  89. 13/17 commit-graph: use chunk-format read APIDerrick Stolee via GitGitGadget, Feb 5, 2021
  90. 01/17 commit-graph: anonymize data in chunk_write_fnDerrick Stolee via GitGitGadget, Feb 5, 2021
  91. 03/17 commit-graph: use chunk-format write APIDerrick Stolee via GitGitGadget, Feb 5, 2021
  92. 04/17 midx: rename pack_info to write_midx_contextDerrick Stolee via GitGitGadget, Feb 5, 2021
  93. 11/17 midx: use chunk-format API in write_midx_internal()Derrick Stolee via GitGitGadget, Feb 5, 2021
  94. 12/17 chunk-format: create read chunk APIDerrick Stolee via GitGitGadget, Feb 5, 2021
  95. SZEDER GáborFeb 7, 2021
  96. Derrick StoleeFeb 8, 2021
  97. 00/17 Refactor chunk-format into an APIDerrick Stolee via GitGitGadget, Feb 18, 2021
  98. 07/17 midx: add pack_perm to write_midx_contextDerrick Stolee via GitGitGadget, Feb 18, 2021
  99. 05/17 midx: use context in write_midx_pack_names()Derrick Stolee via GitGitGadget, Feb 18, 2021
  100. 03/17 commit-graph: use chunk-format write APIDerrick Stolee via GitGitGadget, Feb 18, 2021
  101. SZEDER GáborFeb 24, 2021
  102. Taylor BlauFeb 24, 2021
  103. Derrick StoleeFeb 24, 2021
  104. Junio C HamanoFeb 24, 2021
  105. 04/17 midx: rename pack_info to write_midx_contextDerrick Stolee via GitGitGadget, Feb 18, 2021
  106. 06/17 midx: add entries to write_midx_contextDerrick Stolee via GitGitGadget, Feb 18, 2021
  107. 02/17 chunk-format: create chunk format write APIDerrick Stolee via GitGitGadget, Feb 18, 2021
  108. 01/17 commit-graph: anonymize data in chunk_write_fnDerrick Stolee via GitGitGadget, Feb 18, 2021
  109. 08/17 midx: add num_large_offsets to write_midx_contextDerrick Stolee via GitGitGadget, Feb 18, 2021
  110. 09/17 midx: return success/failure in chunk write methodsDerrick Stolee via GitGitGadget, Feb 18, 2021
  111. 10/17 midx: drop chunk progress during writeDerrick Stolee via GitGitGadget, Feb 18, 2021
  112. 11/17 midx: use chunk-format API in write_midx_internal()Derrick Stolee via GitGitGadget, Feb 18, 2021
  113. 12/17 chunk-format: create read chunk APIDerrick Stolee via GitGitGadget, Feb 18, 2021
  114. 17/17 chunk-format: add technical docsDerrick Stolee via GitGitGadget, Feb 18, 2021
  115. Junio C HamanoFeb 18, 2021
  116. Derrick StoleeFeb 19, 2021
  117. 15/17 midx: use 64-bit multiplication for chunk sizesDerrick Stolee via GitGitGadget, Feb 18, 2021
  118. 14/17 midx: use chunk-format read APIDerrick Stolee via GitGitGadget, Feb 18, 2021
  119. 13/17 commit-graph: use chunk-format read APIDerrick Stolee via GitGitGadget, Feb 18, 2021
  120. 16/17 chunk-format: restore duplicate chunk checksDerrick Stolee via GitGitGadget, Feb 18, 2021

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.