Merge branch 'jk/pack-delta-reuse-with-bitmap'

When creating a thin pack, which allows objects to be made into a
delta against another object that is not in the resulting pack but
is known to be present on the receiving end, the code learned to
take advantage of the reachability bitmap; this allows the server
to send a delta against a base beyond the "boundary" commit.

* jk/pack-delta-reuse-with-bitmap:
  pack-objects: reuse on-disk deltas for thin "have" objects
  pack-bitmap: save "have" bitmap from walk
  t/perf: add perf tests for fetches from a bitmapped server
  t/perf: add infrastructure for measuring sizes
  t/perf: factor out percent calculations
  t/perf: factor boilerplate out of test_perf

4 files changed, 173 insertions, 40 deletions

diff --git a/t/perf/README b/t/perf/README
index 21321a0f36..be12090c38 100644
--- a/t/perf/README
+++ b/t/perf/README
@@ -168,3 +168,28 @@ that
   While we have tried to make sure that it can cope with embedded
   whitespace and other special characters, it will not work with
   multi-line data.
+
+Rather than tracking the performance by run-time as `test_perf` does, you
+may also track output size by using `test_size`. The stdout of the
+function should be a single numeric value, which will be captured and
+shown in the aggregated output. For example:
+
+	test_perf 'time foo' '
+		./foo >foo.out
+	'
+
+	test_size 'output size'
+		wc -c <foo.out
+	'
+
+might produce output like:
+
+	Test                origin           HEAD
+	-------------------------------------------------------------
+	1234.1 time foo     0.37(0.79+0.02)  0.26(0.51+0.02) -29.7%
+	1234.2 output size             4.3M             3.6M -14.7%
+
+The item being measured (and its units) is up to the test; the context
+and the test title should make it clear to the user whether bigger or
+smaller numbers are better. Unlike test_perf, the test code will only be
+run once, since output sizes tend to be more deterministic than timings.
diff --git a/t/perf/aggregate.perl b/t/perf/aggregate.perl
index bc865160e7..494907a892 100755
--- a/t/perf/aggregate.perl
+++ b/t/perf/aggregate.perl
@@ -13,27 +13,42 @@ sub get_times {
 	my $line = <$fh>;
 	return undef if not defined $line;
 	close $fh or die "cannot close $name: $!";
-	$line =~ /^(?:(\d+):)?(\d+):(\d+(?:\.\d+)?) (\d+(?:\.\d+)?) (\d+(?:\.\d+)?)$/
-		or die "bad input line: $line";
-	my $rt = ((defined $1 ? $1 : 0.0)*60+$2)*60+$3;
-	return ($rt, $4, $5);
+	# times
+	if ($line =~ /^(?:(\d+):)?(\d+):(\d+(?:\.\d+)?) (\d+(?:\.\d+)?) (\d+(?:\.\d+)?)$/) {
+		my $rt = ((defined $1 ? $1 : 0.0)*60+$2)*60+$3;
+		return ($rt, $4, $5);
+	# size
+	} elsif ($line =~ /^\d+$/) {
+		return $&;
+	} else {
+		die "bad input line: $line";
+	}
+}
+
+sub relative_change {
+	my ($r, $firstr) = @_;
+	if ($firstr > 0) {
+		return sprintf "%+.1f%%", 100.0*($r-$firstr)/$firstr;
+	} elsif ($r == 0) {
+		return "=";
+	} else {
+		return "+inf";
+	}
 }
 
 sub format_times {
 	my ($r, $u, $s, $firstr) = @_;
+	# no value means we did not finish the test
 	if (!defined $r) {
 		return "<missing>";
 	}
-	my $out = sprintf "%.2f(%.2f+%.2f)", $r, $u, $s;
-	if (defined $firstr) {
-		if ($firstr > 0) {
-			$out .= sprintf " %+.1f%%", 100.0*($r-$firstr)/$firstr;
-		} elsif ($r == 0) {
-			$out .= " =";
-		} else {
-			$out .= " +inf";
-		}
+	# a single value means we have a size, not times
+	if (!defined $u) {
+		return format_size($r, $firstr);
 	}
+	# otherwise, we have real/user/system times
+	my $out = sprintf "%.2f(%.2f+%.2f)", $r, $u, $s;
+	$out .= ' ' . relative_change($r, $firstr) if defined $firstr;
 	return $out;
 }
 
@@ -51,6 +66,25 @@ EOT
 	exit(1);
 }
 
+sub human_size {
+	my $n = shift;
+	my @units = ('', qw(K M G));
+	while ($n > 900 && @units > 1) {
+		$n /= 1000;
+		shift @units;
+	}
+	return $n unless length $units[0];
+	return sprintf '%.1f%s', $n, $units[0];
+}
+
+sub format_size {
+	my ($size, $first) = @_;
+	# match the width of a time: 0.00(0.00+0.00)
+	my $out = sprintf '%15s', human_size($size);
+	$out .= ' ' . relative_change($size, $first) if defined $first;
+	return $out;
+}
+
 my (@dirs, %dirnames, %dirabbrevs, %prefixes, @tests,
     $codespeed, $sortby, $subsection, $reponame);
 
@@ -181,7 +215,14 @@ sub print_default_results {
 		my $firstr;
 		for my $i (0..$#dirs) {
 			my $d = $dirs[$i];
-			$times{$prefixes{$d}.$t} = [get_times("$resultsdir/$prefixes{$d}$t.times")];
+			my $base = "$resultsdir/$prefixes{$d}$t";
+			$times{$prefixes{$d}.$t} = [];
+			foreach my $type (qw(times size)) {
+				if (-e "$base.$type") {
+					$times{$prefixes{$d}.$t} = [get_times("$base.$type")];
+					last;
+				}
+			}
 			my ($r,$u,$s) = @{$times{$prefixes{$d}.$t}};
 			my $w = length format_times($r,$u,$s,$firstr);
 			$colwidth[$i] = $w if $w > $colwidth[$i];
diff --git a/t/perf/p5311-pack-bitmaps-fetch.sh b/t/perf/p5311-pack-bitmaps-fetch.sh
new file mode 100755
index 0000000000..b04575951f
--- /dev/null
+++ b/t/perf/p5311-pack-bitmaps-fetch.sh
@@ -0,0 +1,45 @@
+#!/bin/sh
+
+test_description='performance of fetches from bitmapped packs'
+. ./perf-lib.sh
+
+test_perf_default_repo
+
+test_expect_success 'create bitmapped server repo' '
+	git config pack.writebitmaps true &&
+	git config pack.writebitmaphashcache true &&
+	git repack -ad
+'
+
+# simulate a fetch from a repository that last fetched N days ago, for
+# various values of N. We do so by following the first-parent chain,
+# and assume the first entry in the chain that is N days older than the current
+# HEAD is where the HEAD would have been then.
+for days in 1 2 4 8 16 32 64 128; do
+	title=$(printf '%10s' "($days days)")
+	test_expect_success "setup revs from $days days ago" '
+		now=$(git log -1 --format=%ct HEAD) &&
+		then=$(($now - ($days * 86400))) &&
+		tip=$(git rev-list -1 --first-parent --until=$then HEAD) &&
+		{
+			echo HEAD &&
+			echo ^$tip
+		} >revs
+	'
+
+	test_perf "server $title" '
+		git pack-objects --stdout --revs \
+				 --thin --delta-base-offset \
+				 <revs >tmp.pack
+	'
+
+	test_size "size   $title" '
+		wc -c <tmp.pack
+	'
+
+	test_perf "client $title" '
+		git index-pack --stdin --fix-thin <tmp.pack
+	'
+done
+
+test_done
diff --git a/t/perf/perf-lib.sh b/t/perf/perf-lib.sh
index e4c343a6b7..11d1922cf5 100644
--- a/t/perf/perf-lib.sh
+++ b/t/perf/perf-lib.sh
@@ -179,8 +179,8 @@ exit $ret' >&3 2>&4
 	return "$eval_ret"
 }
 
-
-test_perf () {
+test_wrapper_ () {
+	test_wrapper_func_=$1; shift
 	test_start_
 	test "$#" = 3 && { test_prereq=$1; shift; } || test_prereq=
 	test "$#" = 2 ||
@@ -191,35 +191,57 @@ test_perf () {
 		base=$(basename "$0" .sh)
 		echo "$test_count" >>"$perf_results_dir"/$base.subtests
 		echo "$1" >"$perf_results_dir"/$base.$test_count.descr
-		if test -z "$verbose"; then
-			printf "%s" "perf $test_count - $1:"
-		else
-			echo "perf $test_count - $1:"
-		fi
-		for i in $(test_seq 1 $GIT_PERF_REPEAT_COUNT); do
-			say >&3 "running: $2"
-			if test_run_perf_ "$2"
-			then
-				if test -z "$verbose"; then
-					printf " %s" "$i"
-				else
-					echo "* timing run $i/$GIT_PERF_REPEAT_COUNT:"
-				fi
+		base="$perf_results_dir"/"$perf_results_prefix$(basename "$0" .sh)"."$test_count"
+		"$test_wrapper_func_" "$@"
+	fi
+
+	test_finish_
+}
+
+test_perf_ () {
+	if test -z "$verbose"; then
+		printf "%s" "perf $test_count - $1:"
+	else
+		echo "perf $test_count - $1:"
+	fi
+	for i in $(test_seq 1 $GIT_PERF_REPEAT_COUNT); do
+		say >&3 "running: $2"
+		if test_run_perf_ "$2"
+		then
+			if test -z "$verbose"; then
+				printf " %s" "$i"
 			else
-				test -z "$verbose" && echo
-				test_failure_ "$@"
-				break
+				echo "* timing run $i/$GIT_PERF_REPEAT_COUNT:"
 			fi
-		done
-		if test -z "$verbose"; then
-			echo " ok"
 		else
-			test_ok_ "$1"
+			test -z "$verbose" && echo
+			test_failure_ "$@"
+			break
 		fi
-		base="$perf_results_dir"/"$perf_results_prefix$(basename "$0" .sh)"."$test_count"
-		"$TEST_DIRECTORY"/perf/min_time.perl test_time.* >"$base".times
+	done
+	if test -z "$verbose"; then
+		echo " ok"
+	else
+		test_ok_ "$1"
 	fi
-	test_finish_
+	"$TEST_DIRECTORY"/perf/min_time.perl test_time.* >"$base".times
+}
+
+test_perf () {
+	test_wrapper_ test_perf_ "$@"
+}
+
+test_size_ () {
+	say >&3 "running: $2"
+	if test_eval_ "$2" 3>"$base".size; then
+		test_ok_ "$1"
+	else
+		test_failure_ "$@"
+	fi
+}
+
+test_size () {
+	test_wrapper_ test_size_ "$@"
 }
 
 # We extend test_done to print timings at the end (./run disables this