fast-import: count and report # of calls to diff_delta in stats

It's an interesting number, how often do we try to deltify each type of
objects and how often do we succeed. So do add it to stats.

Success doesn't mean much gain in pack size though. As we allow delta to
be as big as (data.len - 20). And delta close to data.len gains nothing
compared to no delta at all even after zlib compression (delta is pretty
much the same as data, just with few modifications).

We should try to make less attempts that result in huge deltas as these
consume more cpu than trivial small deltas. Either by choosing a better
delta base or reducing delta size upper bound or doing less delta attempts
at all.

Currently, delta base for blobs is a waste literally. Each blob delta
base is chosen as a previously stored blob. Disabling deltas for blobs
doesn't increase pack size and reduce import time, or at least doesn't
increase time for all fast-import streams I've tried.

Signed-off-by: Dmitry Ivankov <divanorama@gmail.com>
Acked-by: David Barr <davidbarr@google.com>
Acked-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>

diff --git a/fast-import.c b/fast-import.c
index 78d978684da2e2b79340c138e1b6300100df1221..c6c8cc3a6ea6b24372ae542e2d4f5f75dd981210 100644 (file)
--- a/fast-import.c
+++ b/fast-import.c
@@ -284,6 +284,7 @@ static uintmax_t marks_set_count;
 static uintmax_t object_count_by_type[1 << TYPE_BITS];
 static uintmax_t duplicate_count_by_type[1 << TYPE_BITS];
 static uintmax_t delta_count_by_type[1 << TYPE_BITS];
 static uintmax_t object_count_by_type[1 << TYPE_BITS];
 static uintmax_t duplicate_count_by_type[1 << TYPE_BITS];
 static uintmax_t delta_count_by_type[1 << TYPE_BITS];

+static uintmax_t delta_count_attempts_by_type[1 << TYPE_BITS];

 static unsigned long object_count;
 static unsigned long branch_count;
 static unsigned long branch_load_count;
 static unsigned long object_count;
 static unsigned long branch_count;
 static unsigned long branch_load_count;


@@ -1043,6 +1044,7 @@ static int store_object(
        }
 
        if (last && last->data.buf && last->depth < max_depth && dat->len > 20) {
        }
 
        if (last && last->data.buf && last->depth < max_depth && dat->len > 20) {

+               delta_count_attempts_by_type[type]++;

                delta = diff_delta(last->data.buf, last->data.len,
                        dat->buf, dat->len,
                        &deltalen, dat->len - 20);
                delta = diff_delta(last->data.buf, last->data.len,
                        dat->buf, dat->len,
                        &deltalen, dat->len - 20);


@@ -3328,10 +3330,10 @@ int main(int argc, const char **argv)
                fprintf(stderr, "---------------------------------------------------------------------\n");
                fprintf(stderr, "Alloc'd objects: %10" PRIuMAX "\n", alloc_count);
                fprintf(stderr, "Total objects:   %10" PRIuMAX " (%10" PRIuMAX " duplicates                  )\n", total_count, duplicate_count);
                fprintf(stderr, "---------------------------------------------------------------------\n");
                fprintf(stderr, "Alloc'd objects: %10" PRIuMAX "\n", alloc_count);
                fprintf(stderr, "Total objects:   %10" PRIuMAX " (%10" PRIuMAX " duplicates                  )\n", total_count, duplicate_count);

-               fprintf(stderr, "      blobs  :   %10" PRIuMAX " (%10" PRIuMAX " duplicates %10" PRIuMAX " deltas)\n", object_count_by_type[OBJ_BLOB], duplicate_count_by_type[OBJ_BLOB], delta_count_by_type[OBJ_BLOB]);
-               fprintf(stderr, "      trees  :   %10" PRIuMAX " (%10" PRIuMAX " duplicates %10" PRIuMAX " deltas)\n", object_count_by_type[OBJ_TREE], duplicate_count_by_type[OBJ_TREE], delta_count_by_type[OBJ_TREE]);
-               fprintf(stderr, "      commits:   %10" PRIuMAX " (%10" PRIuMAX " duplicates %10" PRIuMAX " deltas)\n", object_count_by_type[OBJ_COMMIT], duplicate_count_by_type[OBJ_COMMIT], delta_count_by_type[OBJ_COMMIT]);
-               fprintf(stderr, "      tags   :   %10" PRIuMAX " (%10" PRIuMAX " duplicates %10" PRIuMAX " deltas)\n", object_count_by_type[OBJ_TAG], duplicate_count_by_type[OBJ_TAG], delta_count_by_type[OBJ_TAG]);
+               fprintf(stderr, "      blobs  :   %10" PRIuMAX " (%10" PRIuMAX " duplicates %10" PRIuMAX " deltas of %10" PRIuMAX" attempts)\n", object_count_by_type[OBJ_BLOB], duplicate_count_by_type[OBJ_BLOB], delta_count_by_type[OBJ_BLOB], delta_count_attempts_by_type[OBJ_BLOB]);
+               fprintf(stderr, "      trees  :   %10" PRIuMAX " (%10" PRIuMAX " duplicates %10" PRIuMAX " deltas of %10" PRIuMAX" attempts)\n", object_count_by_type[OBJ_TREE], duplicate_count_by_type[OBJ_TREE], delta_count_by_type[OBJ_TREE], delta_count_attempts_by_type[OBJ_TREE]);
+               fprintf(stderr, "      commits:   %10" PRIuMAX " (%10" PRIuMAX " duplicates %10" PRIuMAX " deltas of %10" PRIuMAX" attempts)\n", object_count_by_type[OBJ_COMMIT], duplicate_count_by_type[OBJ_COMMIT], delta_count_by_type[OBJ_COMMIT], delta_count_attempts_by_type[OBJ_COMMIT]);
+               fprintf(stderr, "      tags   :   %10" PRIuMAX " (%10" PRIuMAX " duplicates %10" PRIuMAX " deltas of %10" PRIuMAX" attempts)\n", object_count_by_type[OBJ_TAG], duplicate_count_by_type[OBJ_TAG], delta_count_by_type[OBJ_TAG], delta_count_attempts_by_type[OBJ_TAG]);

                fprintf(stderr, "Total branches:  %10lu (%10lu loads     )\n", branch_count, branch_load_count);
                fprintf(stderr, "      marks:     %10" PRIuMAX " (%10" PRIuMAX " unique    )\n", (((uintmax_t)1) << marks->shift) * 1024, marks_set_count);
                fprintf(stderr, "      atoms:     %10u\n", atom_cnt);
                fprintf(stderr, "Total branches:  %10lu (%10lu loads     )\n", branch_count, branch_load_count);
                fprintf(stderr, "      marks:     %10" PRIuMAX " (%10" PRIuMAX " unique    )\n", (((uintmax_t)1) << marks->shift) * 1024, marks_set_count);
                fprintf(stderr, "      atoms:     %10u\n", atom_cnt);