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

[PATCH v3 10/21] mmapped_ref_iterator: add iterator over a packed-refs file

From
Michael Haggerty <mhagger@alum.mit.edu>
Date
Sep 25, 2017, 08:00 UTC
Message-ID
<29ee72916bcc64af347c6306df8c74502f2a1015.1506325610.git.mhagger@alum.mit.edu>
In-Reply-To
<cover.1506325610.git.mhagger@alum.mit.edu>

Add a new `mmapped_ref_iterator`, which can iterate over the references in an mmapped `packed-refs` file directly. Use this iterator from `read_packed_refs()` to fill the packed refs cache.

Note that we are not yet willing to promise that the new iterator generates its output in order. That doesn't matter for now, because the packed refs cache doesn't care what order it is filled.

This change adds a lot of boilerplate without providing any obvious benefits. The benefits will come soon, when we get rid of the `ref_cache` for packed references altogether.

Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
---
 refs/packed-backend.c | 207 ++++++++++++++++++++++++++++++++++++--------------
 1 file changed, 152 insertions(+), 55 deletions(-)
diff --git a/refs/packed-backend.c b/refs/packed-backend.c
index ae276f3445..312116a99d 100644
--- a/refs/packed-backend.c
+++ b/refs/packed-backend.c
@@ -163,6 +163,141 @@ static NORETURN void die_invalid_line(const char *path,
 
 }
 
+/*
+ * An iterator over a packed-refs file that is currently mmapped.
+ */
+struct mmapped_ref_iterator {
+	struct ref_iterator base;
+
+	struct packed_ref_cache *packed_refs;
+
+	/* The current position in the mmapped file: */
+	const char *pos;
+
+	/* The end of the mmapped file: */
+	const char *eof;
+
+	struct object_id oid, peeled;
+
+	struct strbuf refname_buf;
+};
+
+static int mmapped_ref_iterator_advance(struct ref_iterator *ref_iterator)
+{
+	struct mmapped_ref_iterator *iter =
+		(struct mmapped_ref_iterator *)ref_iterator;
+	const char *p = iter->pos, *eol;
+
+	strbuf_reset(&iter->refname_buf);
+
+	if (iter->pos == iter->eof)
+		return ref_iterator_abort(ref_iterator);
+
+	iter->base.flags = REF_ISPACKED;
+
+	if (iter->eof - p < GIT_SHA1_HEXSZ + 2 ||
+	    parse_oid_hex(p, &iter->oid, &p) ||
+	    !isspace(*p++))
+		die_invalid_line(iter->packed_refs->refs->path,
+				 iter->pos, iter->eof - iter->pos);
+
+	eol = memchr(p, '\n', iter->eof - p);
+	if (!eol)
+		die_unterminated_line(iter->packed_refs->refs->path,
+				      iter->pos, iter->eof - iter->pos);
+
+	strbuf_add(&iter->refname_buf, p, eol - p);
+	iter->base.refname = iter->refname_buf.buf;
+
+	if (check_refname_format(iter->base.refname, REFNAME_ALLOW_ONELEVEL)) {
+		if (!refname_is_safe(iter->base.refname))
+			die("packed refname is dangerous: %s",
+			    iter->base.refname);
+		oidclr(&iter->oid);
+		iter->base.flags |= REF_BAD_NAME | REF_ISBROKEN;
+	}
+	if (iter->packed_refs->peeled == PEELED_FULLY ||
+	    (iter->packed_refs->peeled == PEELED_TAGS &&
+	     starts_with(iter->base.refname, "refs/tags/")))
+		iter->base.flags |= REF_KNOWS_PEELED;
+
+	iter->pos = eol + 1;
+
+	if (iter->pos < iter->eof && *iter->pos == '^') {
+		p = iter->pos + 1;
+		if (iter->eof - p < GIT_SHA1_HEXSZ + 1 ||
+		    parse_oid_hex(p, &iter->peeled, &p) ||
+		    *p++ != '\n')
+			die_invalid_line(iter->packed_refs->refs->path,
+					 iter->pos, iter->eof - iter->pos);
+		iter->pos = p;
+
+		/*
+		 * Regardless of what the file header said, we
+		 * definitely know the value of *this* reference:
+		 */
+		iter->base.flags |= REF_KNOWS_PEELED;
+	} else {
+		oidclr(&iter->peeled);
+	}
+
+	return ITER_OK;
+}
+
+static int mmapped_ref_iterator_peel(struct ref_iterator *ref_iterator,
+				    struct object_id *peeled)
+{
+	struct mmapped_ref_iterator *iter =
+		(struct mmapped_ref_iterator *)ref_iterator;
+
+	if ((iter->base.flags & REF_KNOWS_PEELED)) {
+		oidcpy(peeled, &iter->peeled);
+		return is_null_oid(&iter->peeled) ? -1 : 0;
+	} else if ((iter->base.flags & (REF_ISBROKEN | REF_ISSYMREF))) {
+		return -1;
+	} else {
+		return !!peel_object(iter->oid.hash, peeled->hash);
+	}
+}
+
+static int mmapped_ref_iterator_abort(struct ref_iterator *ref_iterator)
+{
+	struct mmapped_ref_iterator *iter =
+		(struct mmapped_ref_iterator *)ref_iterator;
+
+	release_packed_ref_cache(iter->packed_refs);
+	strbuf_release(&iter->refname_buf);
+	base_ref_iterator_free(ref_iterator);
+	return ITER_DONE;
+}
+
+static struct ref_iterator_vtable mmapped_ref_iterator_vtable = {
+	mmapped_ref_iterator_advance,
+	mmapped_ref_iterator_peel,
+	mmapped_ref_iterator_abort
+};
+
+struct ref_iterator *mmapped_ref_iterator_begin(
+		const char *packed_refs_file,
+		struct packed_ref_cache *packed_refs,
+		const char *pos, const char *eof)
+{
+	struct mmapped_ref_iterator *iter = xcalloc(1, sizeof(*iter));
+	struct ref_iterator *ref_iterator = &iter->base;
+
+	base_ref_iterator_init(ref_iterator, &mmapped_ref_iterator_vtable, 0);
+
+	iter->packed_refs = packed_refs;
+	acquire_packed_ref_cache(iter->packed_refs);
+	iter->pos = pos;
+	iter->eof = eof;
+	strbuf_init(&iter->refname_buf, 0);
+
+	iter->base.oid = &iter->oid;
+
+	return ref_iterator;
+}
+
 /*
  * Read from the `packed-refs` file into a newly-allocated
  * `packed_ref_cache` and return it. The return value will already
@@ -199,9 +334,10 @@ static struct packed_ref_cache *read_packed_refs(struct packed_ref_store *refs)
 	struct stat st;
 	size_t size;
 	char *buf;
-	const char *pos, *eol, *eof;
-	struct strbuf tmp = STRBUF_INIT;
+	const char *pos, *eof;
 	struct ref_dir *dir;
+	struct ref_iterator *iter;
+	int ok;
 
 	packed_refs->refs = refs;
 	acquire_packed_ref_cache(packed_refs);
@@ -235,7 +371,9 @@ static struct packed_ref_cache *read_packed_refs(struct packed_ref_store *refs)
 
 	/* If the file has a header line, process it: */
 	if (pos < eof && *pos == '#') {
+		struct strbuf tmp = STRBUF_INIT;
 		char *p;
+		const char *eol;
 		struct string_list traits = STRING_LIST_INIT_NODUP;
 
 		eol = memchr(pos, '\n', eof - pos);
@@ -259,69 +397,28 @@ static struct packed_ref_cache *read_packed_refs(struct packed_ref_store *refs)
 		pos = eol + 1;
 
 		string_list_clear(&traits, 0);
-		strbuf_reset(&tmp);
+		strbuf_release(&tmp);
 	}
 
 	dir = get_ref_dir(packed_refs->cache->root);
-	while (pos < eof) {
-		const char *p = pos;
-		struct object_id oid;
-		const char *refname;
-		int flag = REF_ISPACKED;
-		struct ref_entry *entry = NULL;
-
-		if (eof - pos < GIT_SHA1_HEXSZ + 2 ||
-		    parse_oid_hex(p, &oid, &p) ||
-		    !isspace(*p++))
-			die_invalid_line(refs->path, pos, eof - pos);
+	iter = mmapped_ref_iterator_begin(refs->path, packed_refs, pos, eof);
+	while ((ok = ref_iterator_advance(iter)) == ITER_OK) {
+		struct ref_entry *entry =
+			create_ref_entry(iter->refname, iter->oid, iter->flags);
 
-		eol = memchr(p, '\n', eof - p);
-		if (!eol)
-			die_unterminated_line(refs->path, pos, eof - pos);
-
-		strbuf_add(&tmp, p, eol - p);
-		refname = tmp.buf;
-
-		if (check_refname_format(refname, REFNAME_ALLOW_ONELEVEL)) {
-			if (!refname_is_safe(refname))
-				die("packed refname is dangerous: %s", refname);
-			oidclr(&oid);
-			flag |= REF_BAD_NAME | REF_ISBROKEN;
-		}
-		if (packed_refs->peeled == PEELED_FULLY ||
-		    (packed_refs->peeled == PEELED_TAGS &&
-		     starts_with(refname, "refs/tags/")))
-			flag |= REF_KNOWS_PEELED;
-		entry = create_ref_entry(refname, &oid, flag);
+		if ((iter->flags & REF_KNOWS_PEELED))
+			ref_iterator_peel(iter, &entry->u.value.peeled);
 		add_ref_entry(dir, entry);
-
-		pos = eol + 1;
-
-		if (pos < eof && *pos == '^') {
-			p = pos + 1;
-			if (eof - p < GIT_SHA1_HEXSZ + 1 ||
-			    parse_oid_hex(p, &entry->u.value.peeled, &p) ||
-			    *p++ != '\n')
-				die_invalid_line(refs->path, pos, eof - pos);
-
-			/*
-			 * Regardless of what the file header said,
-			 * we definitely know the value of *this*
-			 * reference:
-			 */
-			entry->flag |= REF_KNOWS_PEELED;
-
-			pos = p;
-		}
-
-		strbuf_reset(&tmp);
 	}
 
+	if (ok != ITER_DONE)
+		die("error reading packed-refs file %s", refs->path);
+
 	if (munmap(buf, size))
-		die_errno("error ummapping packed-refs file");
+		die_errno("error ummapping packed-refs file %s", refs->path);
+
 	close(fd);
 
-	strbuf_release(&tmp);
 	return packed_refs;
 }
 
-- 
2.14.1
Previous: Michael HaggertyNext: Michael Haggerty
Message 11 of 24 in “Read `packed-refs` using mmap()”
  1. 00/21 Read `packed-refs` using mmap()Michael Haggerty, Sep 25, 2017
  2. 01/21 ref_iterator: keep track of whether the iterator output is orderedMichael Haggerty, Sep 25, 2017
  3. 02/21 prefix_ref_iterator: break when we leave the prefixMichael Haggerty, Sep 25, 2017
  4. 03/21 packed_ref_cache: add a backlink to the associated `packed_ref_store`Michael Haggerty, Sep 25, 2017
  5. 04/21 die_unterminated_line(), die_invalid_line(): new functionsMichael Haggerty, Sep 25, 2017
  6. 05/21 read_packed_refs(): use mmap to read the `packed-refs` fileMichael Haggerty, Sep 25, 2017
  7. 06/21 read_packed_refs(): only check for a header at the top of the fileMichael Haggerty, Sep 25, 2017
  8. 07/21 read_packed_refs(): make parsing of the header line more robustMichael Haggerty, Sep 25, 2017
  9. 09/21 packed_ref_cache: remember the file-wide peeling stateMichael Haggerty, Sep 25, 2017
  10. 08/21 read_packed_refs(): read references with minimal copyingMichael Haggerty, Sep 25, 2017
  11. 10/21 mmapped_ref_iterator: add iterator over a packed-refs fileMichael Haggerty, Sep 25, 2017
  12. 12/21 packed-backend.c: reorder some definitionsMichael Haggerty, Sep 25, 2017
  13. 13/21 packed_ref_cache: keep the `packed-refs` file mmapped if possibleMichael Haggerty, Sep 25, 2017
  14. 14/21 read_packed_refs(): ensure that references are ordered when readMichael Haggerty, Sep 25, 2017
  15. 11/21 mmapped_ref_iterator_advance(): no peeled value for broken refsMichael Haggerty, Sep 25, 2017
  16. 15/21 packed_ref_iterator_begin(): iterate using `mmapped_ref_iterator`Michael Haggerty, Sep 25, 2017
  17. 16/21 packed_read_raw_ref(): read the reference from the mmapped bufferMichael Haggerty, Sep 25, 2017
  18. 17/21 ref_store: implement `refs_peel_ref()` genericallyMichael Haggerty, Sep 25, 2017
  19. 18/21 packed_ref_store: get rid of the `ref_cache` entirelyMichael Haggerty, Sep 25, 2017
  20. 20/21 mmapped_ref_iterator: inline into `packed_ref_iterator`Michael Haggerty, Sep 25, 2017
  21. 19/21 ref_cache: remove support for storing peeled valuesMichael Haggerty, Sep 25, 2017
  22. 21/21 packed-backend.c: rename a bunch of things and update commentsMichael Haggerty, Sep 25, 2017
  23. Jeff KingSep 25, 2017
  24. Junio C HamanoSep 29, 2017

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.