Avoid to duplicate commit message when is not encoded

When a commit message doesn't have encoding information
and encoding output is utf-8 (default) then an useless
xstrdup() of commit message is done.

If we assume most of users live in an utf-8 world, this
useless copy is the common case.

Performance issue found with KCachegrind.

Signed-off-by: Marco Costalba <mcostalba@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>

1 files changed, 4 insertions, 1 deletions

diff --git a/commit.c b/commit.c
index 4c5dfa9af0..dc5a0643f3 100644
--- a/commit.c
+++ b/commit.c
@@ -721,7 +721,10 @@ static char *logmsg_reencode(const struct commit *commit,
 	encoding = get_header(commit, "encoding");
 	use_encoding = encoding ? encoding : utf8;
 	if (!strcmp(use_encoding, output_encoding))
-		out = xstrdup(commit->buffer);
+		if (encoding) /* we'll strip encoding header later */
+			out = xstrdup(commit->buffer);
+		else
+			return NULL; /* nothing to do */
 	else
 		out = reencode_string(commit->buffer,
 				      output_encoding, use_encoding);