builtin/commit-graph.c: introduce split strategy 'replace'
When using split commit-graphs, it is sometimes useful to completely replace the commit-graph chain with a new base. For example, consider a scenario in which a repository builds a new commit-graph incremental for each push. Occasionally (say, after some fixed number of pushes), they may wish to rebuild the commit-graph chain with all reachable commits. They can do so with $ git commit-graph write --reachable but this removes the chain entirely and replaces it with a single commit-graph in 'objects/info/commit-graph'. Unfortunately, this means that the next push will have to move this commit-graph into the first layer of a new chain, and then write its new commits on top. Avoid such copying entirely by allowing the caller to specify that they wish to replace the entirety of their commit-graph chain, while also specifying that the new commit-graph should become the basis of a fresh, length-one chain. This addresses the above situation by making it possible for the caller to instead write: $ git commit-graph write --reachable --split=replace which writes a new length-one chain to 'objects/info/commit-graphs', making the commit-graph incremental generated by the subsequent push relatively cheap by avoiding the aforementioned copy. In order to do this, remove an assumption in 'write_commit_graph_file' that chains are always at least two incrementals long. Signed-off-by: Taylor Blau <me@ttaylorr.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Taylor Blau committed
Apr 13, 2020 at 22:04 UTC
8a6ac287b2ba5f75bb2d9409dd97e9b501daf253
5 files changed
+66
-18
Documentation/git-commit-graph.txt
+4
-3
@@ -64,9 +64,10 @@ strategy and other splitting options. The new commits not already in the
64
commit-graph are added in a new "tip" file. This file is merged with the
65
existing file if the following merge conditions are met:
66
* If `--split=no-merge` is specified, a merge is never performed, and
67
-the remaining options are ignored. A bare `--split` defers to the
68
-remaining options. (Note that merging a chain of commit graphs replaces
69
-the existing chain with a length-1 chain where the first and only
67
+the remaining options are ignored. `--split=replace` overwrites the
68
+existing chain with a new one. A bare `--split` defers to the remaining
69
+options. (Note that merging a chain of commit graphs replaces the
70
+existing chain with a length-1 chain where the first and only
71
incremental holds the entire graph).
72
+
73
* If `--size-multiple=<X>` is not specified, let `X` equal 2. If the new
builtin/commit-graph.c
+2
@@ -126,6 +126,8 @@ static int write_option_parse_split(const struct option *opt, const char *arg,
126
127
if (!strcmp(arg, "no-merge"))
128
*flags = COMMIT_GRAPH_SPLIT_MERGE_PROHIBITED;
129
+ else if (!strcmp(arg, "replace"))
130
+ *flags = COMMIT_GRAPH_SPLIT_REPLACE;
131
else
132
die(_("unrecognized --split argument, %s"), arg);
133
commit-graph.c
+39
-14
@@ -866,7 +866,7 @@ static void write_graph_chunk_data(struct hashfile *f, int hash_len,
866
867
if (edge_value >= 0)
868
edge_value += ctx->new_num_commits_in_base;
869
- else {
869
+ else if (ctx->new_base_graph) {
870
uint32_t pos;
871
if (find_commit_in_graph(parent->item,
872
ctx->new_base_graph,
@@ -897,7 +897,7 @@ static void write_graph_chunk_data(struct hashfile *f, int hash_len,
897
898
if (edge_value >= 0)
899
edge_value += ctx->new_num_commits_in_base;
900
- else {
900
+ else if (ctx->new_base_graph) {
901
uint32_t pos;
902
if (find_commit_in_graph(parent->item,
903
ctx->new_base_graph,
@@ -964,7 +964,7 @@ static void write_graph_chunk_extra_edges(struct hashfile *f,
964
965
if (edge_value >= 0)
966
edge_value += ctx->new_num_commits_in_base;
967
- else {
967
+ else if (ctx->new_base_graph) {
968
uint32_t pos;
969
if (find_commit_in_graph(parent->item,
970
ctx->new_base_graph,
@@ -1037,6 +1037,8 @@ static void close_reachable(struct write_commit_graph_context *ctx)
1037
{
1038
int i;
1039
struct commit *commit;
1040
+ enum commit_graph_split_flags flags = ctx->split_opts ?
1041
+ ctx->split_opts->flags : COMMIT_GRAPH_SPLIT_UNSPECIFIED;
1042
1043
if (ctx->report_progress)
1044
ctx->progress = start_delayed_progress(
@@ -1066,8 +1068,9 @@ static void close_reachable(struct write_commit_graph_context *ctx)
1068
if (!commit)
1069
continue;
1070
if (ctx->split) {
1069
- if (!parse_commit(commit) &&
1070
- commit->graph_pos == COMMIT_NOT_FROM_GRAPH)
1071
+ if ((!parse_commit(commit) &&
1072
+ commit->graph_pos == COMMIT_NOT_FROM_GRAPH) ||
1073
+ flags == COMMIT_GRAPH_SPLIT_REPLACE)
1074
add_missing_parents(ctx, commit);
1075
} else if (!parse_commit_no_graph(commit))
1076
add_missing_parents(ctx, commit);
@@ -1287,6 +1290,8 @@ static uint32_t count_distinct_commits(struct write_commit_graph_context *ctx)
1290
static void copy_oids_to_commits(struct write_commit_graph_context *ctx)
1291
{
1292
uint32_t i;
1293
+ enum commit_graph_split_flags flags = ctx->split_opts ?
1294
+ ctx->split_opts->flags : COMMIT_GRAPH_SPLIT_UNSPECIFIED;
1295
1296
ctx->num_extra_edges = 0;
1297
if (ctx->report_progress)
@@ -1303,11 +1308,14 @@ static void copy_oids_to_commits(struct write_commit_graph_context *ctx)
1308
ALLOC_GROW(ctx->commits.list, ctx->commits.nr + 1, ctx->commits.alloc);
1309
ctx->commits.list[ctx->commits.nr] = lookup_commit(ctx->r, &ctx->oids.list[i]);
1310
1306
- if (ctx->split &&
1311
+ if (ctx->split && flags != COMMIT_GRAPH_SPLIT_REPLACE &&
1312
ctx->commits.list[ctx->commits.nr]->graph_pos != COMMIT_NOT_FROM_GRAPH)
1313
continue;
1314
1310
- parse_commit_no_graph(ctx->commits.list[ctx->commits.nr]);
1315
+ if (ctx->split && flags == COMMIT_GRAPH_SPLIT_REPLACE)
1316
+ parse_commit(ctx->commits.list[ctx->commits.nr]);
1317
+ else
1318
+ parse_commit_no_graph(ctx->commits.list[ctx->commits.nr]);
1319
1320
num_parents = commit_list_count(ctx->commits.list[ctx->commits.nr]->parents);
1321
if (num_parents > 2)
@@ -1488,8 +1496,12 @@ static int write_commit_graph_file(struct write_commit_graph_context *ctx)
1496
}
1497
1498
if (ctx->base_graph_name) {
1491
- const char *dest = ctx->commit_graph_filenames_after[
1492
- ctx->num_commit_graphs_after - 2];
1499
+ const char *dest;
1500
+ int idx = ctx->num_commit_graphs_after - 1;
1501
+ if (ctx->num_commit_graphs_after > 1)
1502
+ idx--;
1503
+
1504
+ dest = ctx->commit_graph_filenames_after[idx];
1505
1506
if (strcmp(ctx->base_graph_name, dest)) {
1507
result = rename(ctx->base_graph_name, dest);
@@ -1546,9 +1558,13 @@ static void split_graph_merge_strategy(struct write_commit_graph_context *ctx)
1558
1559
g = ctx->r->objects->commit_graph;
1560
num_commits = ctx->commits.nr;
1549
- ctx->num_commit_graphs_after = ctx->num_commit_graphs_before + 1;
1561
+ if (flags == COMMIT_GRAPH_SPLIT_REPLACE)
1562
+ ctx->num_commit_graphs_after = 1;
1563
+ else
1564
+ ctx->num_commit_graphs_after = ctx->num_commit_graphs_before + 1;
1565
1551
- if (flags != COMMIT_GRAPH_SPLIT_MERGE_PROHIBITED) {
1566
+ if (flags != COMMIT_GRAPH_SPLIT_MERGE_PROHIBITED &&
1567
+ flags != COMMIT_GRAPH_SPLIT_REPLACE) {
1568
while (g && (g->num_commits <= size_mult * num_commits ||
1569
(max_commits && num_commits > max_commits))) {
1570
if (g->odb != ctx->odb)
@@ -1561,7 +1577,11 @@ static void split_graph_merge_strategy(struct write_commit_graph_context *ctx)
1577
}
1578
}
1579
1564
- ctx->new_base_graph = g;
1580
+ if (flags != COMMIT_GRAPH_SPLIT_REPLACE)
1581
+ ctx->new_base_graph = g;
1582
+ else if (ctx->num_commit_graphs_after != 1)
1583
+ BUG("split_graph_merge_strategy: num_commit_graphs_after "
1584
+ "should be 1 with --split=replace");
1585
1586
if (ctx->num_commit_graphs_after == 2) {
1587
char *old_graph_name = get_commit_graph_filename(g->odb);
@@ -1772,6 +1792,7 @@ int write_commit_graph(struct object_directory *odb,
1792
struct write_commit_graph_context *ctx;
1793
uint32_t i, count_distinct = 0;
1794
int res = 0;
1795
+ int replace = 0;
1796
1797
if (!commit_graph_compatible(the_repository))
1798
return 0;
@@ -1806,6 +1827,9 @@ int write_commit_graph(struct object_directory *odb,
1827
g = g->base_graph;
1828
}
1829
}
1830
+
1831
+ if (ctx->split_opts)
1832
+ replace = ctx->split_opts->flags & COMMIT_GRAPH_SPLIT_REPLACE;
1833
}
1834
1835
ctx->approx_nr_objects = approximate_object_count();
@@ -1866,13 +1890,14 @@ int write_commit_graph(struct object_directory *odb,
1890
goto cleanup;
1891
}
1892
1869
- if (!ctx->commits.nr)
1893
+ if (!ctx->commits.nr && !replace)
1894
goto cleanup;
1895
1896
if (ctx->split) {
1897
split_graph_merge_strategy(ctx);
1898
1875
- merge_commit_graphs(ctx);
1899
+ if (!replace)
1900
+ merge_commit_graphs(ctx);
1901
} else
1902
ctx->num_commit_graphs_after = 1;
1903
commit-graph.h
+2
-1
@@ -84,7 +84,8 @@ enum commit_graph_write_flags {
84
85
enum commit_graph_split_flags {
86
COMMIT_GRAPH_SPLIT_UNSPECIFIED = 0,
87
- COMMIT_GRAPH_SPLIT_MERGE_PROHIBITED = 1
87
+ COMMIT_GRAPH_SPLIT_MERGE_PROHIBITED = 1,
88
+ COMMIT_GRAPH_SPLIT_REPLACE = 2
89
};
90
91
struct split_commit_graph_opts {
t/t5324-split-commit-graph.sh
+19
@@ -355,4 +355,23 @@ test_expect_success '--split=no-merge always writes an incremental' '
355
test_line_count = 2 $graphdir/commit-graph-chain
356
'
357
358
+test_expect_success '--split=replace replaces the chain' '
359
+ rm -rf $graphdir $infodir/commit-graph &&
360
+ git reset --hard commits/3 &&
361
+ git rev-list -1 HEAD~2 >a &&
362
+ git rev-list -1 HEAD~1 >b &&
363
+ git rev-list -1 HEAD >c &&
364
+ git commit-graph write --split=no-merge --stdin-commits <a &&
365
+ git commit-graph write --split=no-merge --stdin-commits <b &&
366
+ git commit-graph write --split=no-merge --stdin-commits <c &&
367
+ test_line_count = 3 $graphdir/commit-graph-chain &&
368
+ git commit-graph write --stdin-commits --split=replace <b &&
369
+ test_path_is_missing $infodir/commit-graph &&
370
+ test_path_is_file $graphdir/commit-graph-chain &&
371
+ ls $graphdir/graph-*.graph >graph-files &&
372
+ test_line_count = 1 graph-files &&
373
+ verify_chain_files_exist $graphdir &&
374
+ graph_read_expect 2
375
+'
376
+
377
test_done