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

Re: [PATCH v4 6/6] last-modified: keep per-path Bloom filters for wildcard pathspecs

From
Patrick Steinhardt <ps@pks.im>
Date
Sep 10, 2026, 07:04 UTC
Message-ID
<aqJWihcFmX7tPio5@pks.im>
In-Reply-To
<20260901-toon-speed-up-last-modified-v4-6-a09949800404@iotcl.com>
On Tue, Sep 01, 2026 at 11:10:26AM +0200, Toon Claes wrote:
Show 19 quoted lines
> diff --git a/builtin/last-modified.c b/builtin/last-modified.c
> index 8ab7944314..bedccb3ace 100644
> --- a/builtin/last-modified.c
> +++ b/builtin/last-modified.c
> @@ -370,6 +375,14 @@ static int last_modified_run(struct last_modified *lm)
>  
>  	prepare_revision_walk(&lm->rev);
>  
> +	/*
> +	 * prepare_revision_walk() clears bloom_filter_settings for pathspecs
> +	 * without a Bloom key. Restore it so the per-path check keeps working.
> +	 */
> +	if (!lm->rev.bloom_filter_settings)
> +		lm->rev.bloom_filter_settings =
> +			get_bloom_filter_settings(lm->rev.repo);
> +
>  	max_count = lm->rev.max_count;
>  
>  	init_active_paths_for_commit(&lm->active_paths);

So the revision subsystem is unhappy, but we basically force the bloom filter settings in there anyway? That feels a bit fragile to me. Is there a reason why the revision machinery itself specifically needs to have the bloom filters populated, or do we basically just have to set up the bloom filters so that we can access them ourselves?

If the latter, can't we instead store the bloom filter settings in `struct last_modified` instead of forcing them into the revision machinery? Something like the below patch on top of tihs, which still passes all of our tests.

There might be good reasons though why we can't do it this way.
Patrick
diff --git a/builtin/last-modified.c b/builtin/last-modified.c
index bedccb3ace..dabd0b7c34 100644
--- a/builtin/last-modified.c
+++ b/builtin/last-modified.c
@@ -58,6 +58,8 @@ struct last_modified {
 	bool nul_termination;
 	int max_depth;
 
+	struct bloom_filter_settings *bloom_filter_settings;
+
 	const char **all_paths;
 	size_t all_paths_nr;
 	struct active_paths_for_commit active_paths;
@@ -117,9 +119,9 @@ static void add_path_from_diff(struct diff_queue_struct *q,
 
 		FLEX_ALLOC_STR(ent, path, path);
 		oidcpy(&ent->oid, &p->two->oid);
-		if (lm->rev.bloom_filter_settings)
+		if (lm->bloom_filter_settings)
 			bloom_key_fill(&ent->key, path, strlen(path),
-				       lm->rev.bloom_filter_settings);
+				       lm->bloom_filter_settings);
 		hashmap_entry_init(&ent->hashent, strhash(ent->path));
 		hashmap_add(&lm->paths, &ent->hashent);
 	}
@@ -265,7 +267,7 @@ static bool maybe_changed_path(struct last_modified *lm,
 	struct last_modified_entry *ent;
 	struct hashmap_iter iter;
 
-	if (!lm->rev.bloom_filter_settings)
+	if (!lm->bloom_filter_settings)
 		return true;
 
 	if (commit_graph_generation(origin) == GENERATION_NUMBER_INFINITY)
@@ -294,7 +296,7 @@ static bool maybe_changed_path(struct last_modified *lm,
 			continue;
 
 		if (bloom_filter_contains(filter, &ent->key,
-					  lm->rev.bloom_filter_settings))
+					  lm->bloom_filter_settings))
 			return true;
 	}
 	return false;
@@ -375,14 +377,6 @@ static int last_modified_run(struct last_modified *lm)
 
 	prepare_revision_walk(&lm->rev);
 
-	/*
-	 * prepare_revision_walk() clears bloom_filter_settings for pathspecs
-	 * without a Bloom key. Restore it so the per-path check keeps working.
-	 */
-	if (!lm->rev.bloom_filter_settings)
-		lm->rev.bloom_filter_settings =
-			get_bloom_filter_settings(lm->rev.repo);
-
 	max_count = lm->rev.max_count;
 
 	init_active_paths_for_commit(&lm->active_paths);
@@ -530,7 +524,7 @@ static int last_modified_init(struct last_modified *lm, struct repository *r,
 		return argc;
 	}
 
-	lm->rev.bloom_filter_settings = get_bloom_filter_settings(lm->rev.repo);
+	lm->bloom_filter_settings = get_bloom_filter_settings(lm->rev.repo);
 
 	if (populate_paths_from_revs(lm) < 0)
 		return -1;
Previous: Toon Claes
Message 22 of 22 in “last-modified: use the pathspec's Bloom key to pre-filter commits”
  1. 0/6 last-modified: use the pathspec's Bloom key to pre-filter commitsToon Claes, Aug 31, 2026
  2. 1/6 revision: move bloom keyvec precondition into functionToon Claes, Aug 31, 2026
  3. 2/6 revision: expose check for paths maybe changed in Bloom filterToon Claes, Aug 31, 2026
  4. 3/6 bloom: add helper to check if any key in a vector is presentToon Claes, Aug 31, 2026
  5. 4/6 revision: add Bloom check that includes parent directoriesToon Claes, Aug 31, 2026
  6. 5/6 last-modified: check pathspec against Bloom filter firstToon Claes, Aug 31, 2026
  7. 6/6 last-modified: keep per-path Bloom filters for wildcard pathspecsToon Claes, Aug 31, 2026
  8. Junio C HamanoSep 1, 2026
  9. Toon ClaesSep 1, 2026
  10. Junio C HamanoSep 1, 2026
  11. Junio C HamanoAug 31, 2026
  12. 0/6 last-modified: use the pathspec's Bloom key to pre-filter commitsToon Claes, Sep 1, 2026
  13. 1/6 revision: move bloom keyvec precondition into functionToon Claes, Sep 1, 2026
  14. 2/6 revision: expose check for paths maybe changed in Bloom filterToon Claes, Sep 1, 2026
  15. 3/6 bloom: add helper to check if any key in a vector is presentToon Claes, Sep 1, 2026
  16. Patrick SteinhardtSep 10, 2026
  17. 4/6 revision: add Bloom check that includes parent directoriesToon Claes, Sep 1, 2026
  18. Patrick SteinhardtSep 10, 2026
  19. 5/6 last-modified: check pathspec against Bloom filter firstToon Claes, Sep 1, 2026
  20. Patrick SteinhardtSep 10, 2026
  21. 6/6 last-modified: keep per-path Bloom filters for wildcard pathspecsToon Claes, Sep 1, 2026
  22. Patrick SteinhardtSep 10, 2026

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.