do_compare_entry: use already-computed path
In traverse_trees, we generate the complete traverse path for a traverse_info. Later, in do_compare_entry, we used to go do a bunch of work to compare the traverse_info to a cache_entry's name without computing that path. But since we already have that path, we don't need to do all that work. Instead, we can just put the generated path into the traverse_info, and do the comparison more directly. We copy the path because prune_traversal might mutate `base`. This doesn't happen in any codepaths where do_compare_entry is called, but it's better to be safe. This makes git checkout much faster -- about 25% on Twitter's monorepo. Deeper directory trees are likely to benefit more than shallower ones. Signed-off-by: David Turner <dturner@twopensource.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
David Turner committed
Dec 21, 2015 at 17:34 UTC
d9c2bd560e1e7a3d4654fb6ef3f9037ad337eb01
3 files changed
+44
-2
tree-walk.c
+7
@@ -320,6 +320,7 @@ int traverse_trees(int n, struct tree_desc *t, struct traverse_info *info)
320
struct tree_desc_x *tx = xcalloc(n, sizeof(*tx));
321
struct strbuf base = STRBUF_INIT;
322
int interesting = 1;
323
+ char *traverse_path;
324
325
for (i = 0; i < n; i++)
326
tx[i].d = t[i];
@@ -329,7 +330,11 @@ int traverse_trees(int n, struct tree_desc *t, struct traverse_info *info)
330
make_traverse_path(base.buf, info->prev, &info->name);
331
base.buf[info->pathlen-1] = '/';
332
strbuf_setlen(&base, info->pathlen);
333
+ traverse_path = xstrndup(base.buf, info->pathlen);
334
+ } else {
335
+ traverse_path = xstrndup(info->name.path, info->pathlen);
336
}
337
+ info->traverse_path = traverse_path;
338
for (;;) {
339
int trees_used;
340
unsigned long mask, dirmask;
@@ -411,6 +416,8 @@ int traverse_trees(int n, struct tree_desc *t, struct traverse_info *info)
416
for (i = 0; i < n; i++)
417
free_extended_entry(tx + i);
418
free(tx);
419
+ free(traverse_path);
420
+ info->traverse_path = NULL;
421
strbuf_release(&base);
422
return error;
423
}
tree-walk.h
+1
@@ -59,6 +59,7 @@ enum follow_symlinks_result {
59
enum follow_symlinks_result get_tree_entry_follow_symlinks(unsigned char *tree_sha1, const char *name, unsigned char *result, struct strbuf *result_path, unsigned *mode);
60
61
struct traverse_info {
62
+ const char *traverse_path;
63
struct traverse_info *prev;
64
struct name_entry name;
65
int pathlen;
unpack-trees.c
+36
-2
@@ -498,13 +498,14 @@ static int traverse_trees_recursive(int n, unsigned long dirmask,
498
* itself - the caller needs to do the final check for the cache
499
* entry having more data at the end!
500
*/
501
-static int do_compare_entry(const struct cache_entry *ce, const struct traverse_info *info, const struct name_entry *n)
501
+static int do_compare_entry_piecewise(const struct cache_entry *ce, const struct traverse_info *info, const struct name_entry *n)
502
{
503
int len, pathlen, ce_len;
504
const char *ce_name;
505
506
if (info->prev) {
507
- int cmp = do_compare_entry(ce, info->prev, &info->name);
507
+ int cmp = do_compare_entry_piecewise(ce, info->prev,
508
+ &info->name);
509
if (cmp)
510
return cmp;
511
}
@@ -522,6 +523,39 @@ static int do_compare_entry(const struct cache_entry *ce, const struct traverse_
523
return df_name_compare(ce_name, ce_len, S_IFREG, n->path, len, n->mode);
524
}
525
526
+static int do_compare_entry(const struct cache_entry *ce,
527
+ const struct traverse_info *info,
528
+ const struct name_entry *n)
529
+{
530
+ int len, pathlen, ce_len;
531
+ const char *ce_name;
532
+ int cmp;
533
+
534
+ /*
535
+ * If we have not precomputed the traverse path, it is quicker
536
+ * to avoid doing so. But if we have precomputed it,
537
+ * it is quicker to use the precomputed version.
538
+ */
539
+ if (!info->traverse_path)
540
+ return do_compare_entry_piecewise(ce, info, n);
541
+
542
+ cmp = strncmp(ce->name, info->traverse_path, info->pathlen);
543
+ if (cmp)
544
+ return cmp;
545
+
546
+ pathlen = info->pathlen;
547
+ ce_len = ce_namelen(ce);
548
+
549
+ if (ce_len < pathlen)
550
+ return -1;
551
+
552
+ ce_len -= pathlen;
553
+ ce_name = ce->name + pathlen;
554
+
555
+ len = tree_entry_len(n);
556
+ return df_name_compare(ce_name, ce_len, S_IFREG, n->path, len, n->mode);
557
+}
558
+
559
static int compare_entry(const struct cache_entry *ce, const struct traverse_info *info, const struct name_entry *n)
560
{
561
int cmp = do_compare_entry(ce, info, n);