Optimize sha1_object_info for loose objects, not concurrent repacks

When dealing with a repository with lots of loose objects, sha1_object_info would rescan the packs directory every time an unpacked object was referenced before finally giving up and looking for the loose object. This caused a lot of extra unnecessary system calls during git pack-objects; the code was rereading the entire pack directory once for each loose object file. This patch looks for a loose object before falling back to rescanning the pack directory, rather than the other way around. Signed-off-by: Steven Grimm <koreth@midwinter.com> Acked-by: Shawn O. Pearce <spearce@spearce.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Steven Grimm committed Aug 5, 2008 at 13:08 UTC ddd63e64e4ac7e455dff3e807bf6a6977bb61456

1 file changed +8 -1

sha1_file.c

+8 -1

@@ -1929,11 +1929,18 @@ static int sha1_loose_object_info(const unsigned char *sha1, unsigned long *size

1929 int sha1_object_info(const unsigned char *sha1, unsigned long *sizep)

1930 {

1931 struct pack_entry e;

1932 + int status;

1933

1934 if (!find_pack_entry(sha1, &e, NULL)) {

1935 + /* Most likely it's a loose object. */

1936 + status = sha1_loose_object_info(sha1, sizep);

1937 + if (status >= 0)

1938 + return status;

1939 +

1940 + /* Not a loose object; someone else may have just packed it. */

1941 reprepare_packed_git();

1942 if (!find_pack_entry(sha1, &e, NULL))

1936 - return sha1_loose_object_info(sha1, sizep);

1943 + return status;

1944 }

1945 return packed_object_info(e.p, e.offset, sizep);

1946 }