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

[PATCH/RFC v2 2/3] refs: resolve symbolic refs first

From
David Turner <dturner@twopensource.com>
Date
Jan 6, 2016, 23:41 UTC
Message-ID
<1452123719-22634-2-git-send-email-dturner@twopensource.com>
In-Reply-To
<1452123719-22634-1-git-send-email-dturner@twopensource.com>

Before committing ref updates, split symbolic ref updates into two parts: an update to the underlying ref, and a log-only update to the symbolic ref. This ensures that both references are locked correctly while their reflogs are updated.

It is still possible to confuse git by concurrent updates, since the splitting of symbolic refs does not happen under lock. So a symbolic ref could be replaced by a plain ref in the middle of this operation, which would lead to reflog discontinuities and missed old-ref checks.

Signed-off-by: David Turner <dturner@twopensource.com>
---
 refs.c               |  70 +++++++++++++++++++++++++++++
 refs/files-backend.c | 123 +++++++++++++++++++++++++--------------------------
 refs/refs-internal.h |   8 ++++
 3 files changed, 139 insertions(+), 62 deletions(-)
diff --git a/refs.c b/refs.c
index 38f29b3..2496b91 100644
--- a/refs.c
+++ b/refs.c
@@ -1126,6 +1126,72 @@ int refs_init_db(struct strbuf *err, int shared)
 	return the_refs_backend->init_db(err, shared);
 }
 
+/*
+ * Special case for symbolic refs when REF_NODEREF is not turned on.
+ * Dereference them here, mark them REF_LOG_ONLY, and add an update
+ * for the underlying ref.
+ */
+static int dereference_symrefs(struct ref_transaction *transaction,
+			       struct strbuf *err)
+{
+	int i;
+	int nr = transaction->nr;
+
+	for (i = 0; i < nr; i++) {
+		struct ref_update *update = transaction->updates[i];
+		const char *resolved;
+		unsigned char sha1[20];
+		int resolve_flags = 0;
+		int mustexist = (update->old_sha1 &&
+				 !is_null_sha1(update->old_sha1));
+		int deleting = (update->flags & REF_HAVE_NEW) &&
+			is_null_sha1(update->new_sha1);
+		struct ref_update *new_update;
+
+		if (mustexist)
+			resolve_flags |= RESOLVE_REF_READING;
+		if (deleting)
+			resolve_flags |= RESOLVE_REF_ALLOW_BAD_NAME |
+				RESOLVE_REF_NO_RECURSE;
+
+		if (strcmp(update->refname, "HEAD"))
+			update->flags |= REF_IS_NOT_HEAD;
+
+		resolved = resolve_ref_unsafe(update->refname, resolve_flags,
+					      sha1, &update->type);
+		if (!resolved) {
+			/*
+			 * We'll try again to resolve this during
+			 * commit and give a better error message
+			 * then, but we know it's not a symbolic ref
+			 * (or, indeed, any sort of ref).
+			 */
+			continue;
+		}
+
+		hashcpy(update->read_sha1, sha1);
+
+		if (update->flags & REF_NODEREF ||
+		    !(update->type & REF_ISSYMREF))
+			continue;
+
+		/* Create a new transaction for the underlying ref */
+		if (ref_transaction_update(transaction,
+					   resolved,
+					   update->new_sha1,
+					   (update->flags & REF_HAVE_OLD) ?
+					   update->old_sha1 : NULL,
+					   update->flags & ~REF_IS_NOT_HEAD,
+					   update->msg, err))
+			return -1;
+
+		/* Make the symbolic ref update log-only */
+		update->flags |= REF_LOG_ONLY | REF_NODEREF;
+	}
+
+	return 0;
+}
+
 int ref_transaction_commit(struct ref_transaction *transaction,
 			   struct strbuf *err)
 {
@@ -1142,6 +1208,10 @@ int ref_transaction_commit(struct ref_transaction *transaction,
 		return 0;
 	}
 
+	ret = dereference_symrefs(transaction, err);
+	if (ret)
+		goto done;
+
 	if (get_affected_refnames(transaction, &affected_refnames, err)) {
 		ret = TRANSACTION_GENERIC_ERROR;
 		goto done;
diff --git a/refs/files-backend.c b/refs/files-backend.c
index 0800a57..ce73bc4 100644
--- a/refs/files-backend.c
+++ b/refs/files-backend.c
@@ -7,7 +7,6 @@
 
 struct ref_lock {
 	char *ref_name;
-	char *orig_ref_name;
 	struct lock_file *lk;
 	struct object_id old_oid;
 };
@@ -1839,7 +1838,6 @@ static void unlock_ref(struct ref_lock *lock)
 	if (lock->lk)
 		rollback_lock_file(lock->lk);
 	free(lock->ref_name);
-	free(lock->orig_ref_name);
 	free(lock);
 }
 
@@ -1890,6 +1888,7 @@ static int remove_empty_directories(struct strbuf *path)
  */
 static struct ref_lock *lock_ref_sha1_basic(const char *refname,
 					    const unsigned char *old_sha1,
+					    const unsigned char *read_sha1,
 					    const struct string_list *extras,
 					    const struct string_list *skip,
 					    unsigned int flags, int *type_p,
@@ -1897,13 +1896,13 @@ static struct ref_lock *lock_ref_sha1_basic(const char *refname,
 {
 	struct strbuf ref_file = STRBUF_INIT;
 	struct strbuf orig_ref_file = STRBUF_INIT;
-	const char *orig_refname = refname;
 	struct ref_lock *lock;
 	int last_errno = 0;
 	int type, lflags;
 	int mustexist = (old_sha1 && !is_null_sha1(old_sha1));
-	int resolve_flags = 0;
+	int resolve_flags = RESOLVE_REF_NO_RECURSE;
 	int attempts_remaining = 3;
+	int resolved;
 
 	assert(err);
 
@@ -1911,67 +1910,68 @@ static struct ref_lock *lock_ref_sha1_basic(const char *refname,
 
 	if (mustexist)
 		resolve_flags |= RESOLVE_REF_READING;
-	if (flags & REF_DELETING) {
-		resolve_flags |= RESOLVE_REF_ALLOW_BAD_NAME;
-		if (flags & REF_NODEREF)
-			resolve_flags |= RESOLVE_REF_NO_RECURSE;
-	}
 
-	refname = resolve_ref_unsafe(refname, resolve_flags,
-				     lock->old_oid.hash, &type);
-	if (!refname && errno == EISDIR) {
-		/*
-		 * we are trying to lock foo but we used to
-		 * have foo/bar which now does not exist;
-		 * it is normal for the empty directory 'foo'
-		 * to remain.
-		 */
-		strbuf_git_path(&orig_ref_file, "%s", orig_refname);
-		if (remove_empty_directories(&orig_ref_file)) {
+	if (type_p && *type_p & REF_ISSYMREF) {
+		hashcpy(lock->old_oid.hash, read_sha1);
+	} else {
+
+		if (flags & REF_DELETING)
+			resolve_flags |= RESOLVE_REF_ALLOW_BAD_NAME;
+
+		resolved = !!resolve_ref_unsafe(refname, resolve_flags,
+						lock->old_oid.hash, &type);
+		if (!resolved && errno == EISDIR) {
+			/*
+			 * we are trying to lock foo but we used to
+			 * have foo/bar which now does not exist;
+			 * it is normal for the empty directory 'foo'
+			 * to remain.
+			 */
+			strbuf_git_path(&orig_ref_file, "%s", refname);
+			if (remove_empty_directories(&orig_ref_file)) {
+				last_errno = errno;
+				if (!verify_refname_available_dir(refname, extras, skip,
+								  get_loose_refs(&ref_cache), err))
+					strbuf_addf(err, "there are still refs under '%s'",
+						    refname);
+				goto error_return;
+			}
+			resolved = !!resolve_ref_unsafe(refname, resolve_flags,
+							lock->old_oid.hash, &type);
+		}
+
+		if (type_p)
+			*type_p = type;
+		if (!resolved) {
 			last_errno = errno;
-			if (!verify_refname_available_dir(orig_refname, extras, skip,
+			if (last_errno != ENOTDIR ||
+			    !verify_refname_available_dir(refname, extras, skip,
 							  get_loose_refs(&ref_cache), err))
-				strbuf_addf(err, "there are still refs under '%s'",
-					    orig_refname);
+				strbuf_addf(err,
+					    "unable to resolve reference %s: %s",
+					    refname, strerror(last_errno));
+
+			goto error_return;
+		}
+		/*
+		 * If the ref did not exist and we are creating it, make sure
+		 * there is no existing packed ref whose name begins with our
+		 * refname, nor a packed ref whose name is a proper prefix of
+		 * our refname.
+		 */
+		if (is_null_oid(&lock->old_oid) &&
+		    verify_refname_available_dir(refname, extras, skip,
+						 get_packed_refs(&ref_cache), err)) {
+			last_errno = ENOTDIR;
 			goto error_return;
 		}
-		refname = resolve_ref_unsafe(orig_refname, resolve_flags,
-					     lock->old_oid.hash, &type);
-	}
-	if (type_p)
-	    *type_p = type;
-	if (!refname) {
-		last_errno = errno;
-		if (last_errno != ENOTDIR ||
-		    !verify_refname_available_dir(orig_refname, extras, skip,
-						  get_loose_refs(&ref_cache), err))
-			strbuf_addf(err, "unable to resolve reference %s: %s",
-				    orig_refname, strerror(last_errno));
-
-		goto error_return;
-	}
-	/*
-	 * If the ref did not exist and we are creating it, make sure
-	 * there is no existing packed ref whose name begins with our
-	 * refname, nor a packed ref whose name is a proper prefix of
-	 * our refname.
-	 */
-	if (is_null_oid(&lock->old_oid) &&
-	    verify_refname_available_dir(refname, extras, skip,
-					 get_packed_refs(&ref_cache), err)) {
-		last_errno = ENOTDIR;
-		goto error_return;
 	}
-
 	lock->lk = xcalloc(1, sizeof(struct lock_file));
 
 	lflags = 0;
-	if (flags & REF_NODEREF) {
-		refname = orig_refname;
+	if (flags & REF_NODEREF)
 		lflags |= LOCK_NO_DEREF;
-	}
 	lock->ref_name = xstrdup(refname);
-	lock->orig_ref_name = xstrdup(orig_refname);
 	strbuf_git_path(&ref_file, "%s", refname);
 
  retry:
@@ -2537,7 +2537,7 @@ static int files_rename_ref(const char *oldrefname, const char *newrefname,
 
 	logmoved = log;
 
-	lock = lock_ref_sha1_basic(newrefname, NULL, NULL, NULL, 0, NULL, &err);
+	lock = lock_ref_sha1_basic(newrefname, NULL, NULL, NULL, NULL, 0, NULL, &err);
 	if (!lock) {
 		error("unable to rename '%s' to '%s': %s", oldrefname, newrefname, err.buf);
 		strbuf_release(&err);
@@ -2555,7 +2555,7 @@ static int files_rename_ref(const char *oldrefname, const char *newrefname,
 	return 0;
 
  rollback:
-	lock = lock_ref_sha1_basic(oldrefname, NULL, NULL, NULL, 0, NULL, &err);
+	lock = lock_ref_sha1_basic(oldrefname, NULL, NULL, NULL, NULL, 0, NULL, &err);
 	if (!lock) {
 		error("unable to lock %s for rollback: %s", oldrefname, err.buf);
 		strbuf_release(&err);
@@ -2783,9 +2783,7 @@ static int commit_ref_update(struct ref_lock *lock,
 			     int flags, struct strbuf *err)
 {
 	clear_loose_ref_cache(&ref_cache);
-	if (log_ref_write(lock->ref_name, lock->old_oid.hash, sha1, logmsg, flags, err) < 0 ||
-	    (strcmp(lock->ref_name, lock->orig_ref_name) &&
-	     log_ref_write(lock->orig_ref_name, lock->old_oid.hash, sha1, logmsg, flags, err) < 0)) {
+	if (log_ref_write(lock->ref_name, lock->old_oid.hash, sha1, logmsg, flags, err) < 0) {
 		char *old_msg = strbuf_detach(err, NULL);
 		strbuf_addf(err, "Cannot update the ref '%s': %s",
 			    lock->ref_name, old_msg);
@@ -2793,7 +2791,7 @@ static int commit_ref_update(struct ref_lock *lock,
 		unlock_ref(lock);
 		return -1;
 	}
-	if (strcmp(lock->orig_ref_name, "HEAD") != 0) {
+	if (flags & REF_IS_NOT_HEAD) {
 		/*
 		 * Special hack: If a branch is updated directly and HEAD
 		 * points to it (may happen on the remote side of a push
@@ -3157,6 +3155,7 @@ static int files_transaction_commit(struct ref_transaction *transaction,
 				update->refname,
 				((update->flags & REF_HAVE_OLD) ?
 				 update->old_sha1 : NULL),
+				update->read_sha1,
 				affected_refnames, NULL,
 				update->flags,
 				&update->type,
@@ -3408,7 +3407,7 @@ static int files_reflog_expire(const char *refname, const unsigned char *sha1,
 	struct ref_lock *lock;
 	char *log_file;
 	int status = 0;
-	int type;
+	int type = 0;
 	struct strbuf err = STRBUF_INIT;
 
 	memset(&cb, 0, sizeof(cb));
@@ -3421,7 +3420,7 @@ static int files_reflog_expire(const char *refname, const unsigned char *sha1,
 	 * reference itself, plus we might need to update the
 	 * reference if --updateref was specified:
 	 */
-	lock = lock_ref_sha1_basic(refname, sha1, NULL, NULL, 0, &type, &err);
+	lock = lock_ref_sha1_basic(refname, sha1, NULL, NULL, NULL, 0, &type, &err);
 	if (!lock) {
 		error("cannot lock ref '%s': %s", refname, err.buf);
 		strbuf_release(&err);
diff --git a/refs/refs-internal.h b/refs/refs-internal.h
index 649ba63..971f450 100644
--- a/refs/refs-internal.h
+++ b/refs/refs-internal.h
@@ -44,6 +44,8 @@
 
 #define REF_LOG_ONLY 0x80
 
+#define REF_IS_NOT_HEAD 0x100
+
 /* Include broken references in a do_for_each_ref*() iteration */
 #define DO_FOR_EACH_INCLUDE_BROKEN 0x01
 
@@ -159,6 +161,12 @@ struct ref_update {
 	 */
 	unsigned char old_sha1[20];
 	/*
+	 * During the symbolic ref split stage, we resolve refs.
+	 * We'll re-resolve non-symbolic refs once they are locked,
+	 * but we store this to avoid re-resolving symbolic refs.
+	 */
+	unsigned char read_sha1[20];
+	/*
 	 * One or more of REF_HAVE_NEW, REF_HAVE_OLD, REF_NODEREF,
 	 * REF_DELETING, and REF_ISPRUNING:
 	 */
-- 
2.4.2.749.g730654d-twtrsrc
Previous: David TurnerNext: David Turner
Message 39 of 73 in “LMDB refs backend atop pre-vtable”
  1. 00/16 LMDB refs backend atop pre-vtableDavid Turner, Dec 3, 2015
  2. 01/16 refs: add a backend method structure with transaction functionsDavid Turner, Dec 3, 2015
  3. Junio C HamanoDec 5, 2015
  4. 02/16 refs: add methods for misc ref operationsDavid Turner, Dec 3, 2015
  5. Junio C HamanoDec 11, 2015
  6. David TurnerDec 11, 2015
  7. Junio C HamanoDec 11, 2015
  8. David TurnerDec 11, 2015
  9. Junio C HamanoDec 12, 2015
  10. David TurnerDec 12, 2015
  11. Howard ChuDec 18, 2015
  12. 03/16 refs: add methods for the ref iteratorsDavid Turner, Dec 3, 2015
  13. David AguilarJan 3, 2016
  14. Junio C HamanoJan 4, 2016
  15. Michael HaggertyJan 5, 2016
  16. Junio C HamanoJan 5, 2016
  17. Ronnie SahlbergJan 4, 2016
  18. Junio C HamanoJan 4, 2016
  19. Jeff KingJan 5, 2016
  20. Junio C HamanoJan 5, 2016
  21. 04/16 refs: add do_for_each_per_worktree_refDavid Turner, Dec 3, 2015
  22. Junio C HamanoDec 11, 2015
  23. David TurnerDec 12, 2015
  24. 05/16 refs: add methods for reflogDavid Turner, Dec 3, 2015
  25. 06/16 refs: add method for initial ref transaction commitDavid Turner, Dec 3, 2015
  26. 07/16 refs: add method for delete_refsDavid Turner, Dec 3, 2015
  27. 08/16 refs: add methods to init refs backend and dbDavid Turner, Dec 3, 2015
  28. Michael HaggertyDec 23, 2015
  29. David TurnerDec 23, 2015
  30. 09/16 refs: add method to rename refsDavid Turner, Dec 3, 2015
  31. 10/16 refs: make lock genericDavid Turner, Dec 3, 2015
  32. 11/16 refs: move duplicate check to common codeDavid Turner, Dec 3, 2015
  33. Michael HaggertyDec 23, 2015
  34. David TurnerJan 5, 2016
  35. 12/16 refs: always handle non-normal refs in files backendDavid Turner, Dec 3, 2015
  36. Michael HaggertyDec 23, 2015
  37. David TurnerJan 6, 2016
  38. 1/3 refs: allow log-only updatesDavid Turner, Jan 6, 2016
  39. 2/3 refs: resolve symbolic refs firstDavid Turner, Jan 6, 2016
  40. 3/3 refs: always handle non-normal refs in files backendDavid Turner, Jan 6, 2016
  41. David TurnerJan 8, 2016
  42. David TurnerJan 6, 2016
  43. 13/16 init: allow alternate backends to be set for new reposDavid Turner, Dec 3, 2015
  44. Junio C HamanoDec 5, 2015
  45. Duy NguyenDec 5, 2015
  46. Jeff KingDec 5, 2015
  47. David TurnerDec 8, 2015
  48. Michael HaggertyDec 23, 2015
  49. Jeff KingDec 23, 2015
  50. Jeff KingDec 10, 2015
  51. David TurnerDec 10, 2015
  52. clone: use child_process for recursive checkoutsMichael Haggerty, Dec 23, 2015
  53. David TurnerJan 6, 2016
  54. Michael HaggertyDec 23, 2015
  55. David TurnerJan 5, 2016
  56. Junio C HamanoJan 5, 2016
  57. David TurnerJan 5, 2016
  58. Michael HaggertyJan 6, 2016
  59. Duy NguyenJan 6, 2016
  60. Shawn PearceJan 7, 2016
  61. 14/16 refs: allow ref backend to be set for cloneDavid Turner, Dec 3, 2015
  62. Michael HaggertyDec 23, 2015
  63. Eric SunshineDec 23, 2015
  64. 15/16 refs: add LMDB refs backendDavid Turner, Dec 3, 2015
  65. Junio C HamanoDec 5, 2015
  66. David TurnerDec 5, 2015
  67. Jonathan NiederDec 17, 2015
  68. David TurnerDec 17, 2015
  69. Jonathan NiederDec 17, 2015
  70. Michael HaggertyDec 23, 2015
  71. David TurnerJan 8, 2016
  72. 16/16 refs: tests for lmdb backendDavid Turner, Dec 3, 2015
  73. David TurnerDec 22, 2015

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.