about summary refs log tree commit homepage
path: root/xt/git_async_cmp.t
diff options
context:
space:
mode:
authorEric Wong <e@80x24.org>2019-11-29 12:25:05 +0000
committerEric Wong <e@80x24.org>2019-12-12 03:51:14 +0000
commit46baf956987dca495ed44b1050e64939fae5c8ab (patch)
tree5ec1167e3db591de27a108f314ce8fdda8ad921b /xt/git_async_cmp.t
parent73fe3421f1ecbdc83600d5acfc643c33dbb9a0f2 (diff)
downloadpublic-inbox-46baf956987dca495ed44b1050e64939fae5c8ab.tar.gz
This is a transitionary interface which does NOT require an
event loop.  It can be plugged into in current synchronous code
without major surgery.

It allows HTTP/1.1 pipelining-like functionality by taking
advantage of predictable and well-specified POSIX pipe semantics
by stuffing multiple git cat-file requests into the --batch pipe

With xt/git_async_cmp.t and GIANT_GIT_DIR=git.git, the async
interface is 10-25% faster than the synchronous interface since
it can keep the "git cat-file" process busier.

This is expected to improve performance on systems with slower
storage (but multiple cores).
Diffstat (limited to 'xt/git_async_cmp.t')
-rw-r--r--xt/git_async_cmp.t61
1 files changed, 61 insertions, 0 deletions
diff --git a/xt/git_async_cmp.t b/xt/git_async_cmp.t
new file mode 100644
index 00000000..f8ffe3d9
--- /dev/null
+++ b/xt/git_async_cmp.t
@@ -0,0 +1,61 @@
+#!perl -w
+# Copyright (C) 2019 all contributors <meta@public-inbox.org>
+# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
+use strict;
+use Test::More;
+use Benchmark qw(:all);
+use Digest::SHA;
+use PublicInbox::TestCommon;
+my $git_dir = $ENV{GIANT_GIT_DIR};
+plan 'skip_all' => "GIANT_GIT_DIR not defined for $0" unless defined($git_dir);
+use_ok 'PublicInbox::Git';
+my $git = PublicInbox::Git->new($git_dir);
+my @cat = qw(cat-file --buffer --batch-check --batch-all-objects);
+if (require_git(2.19, 1)) {
+        push @cat, '--unordered';
+} else {
+        warn "git <2.19, cat-file lacks --unordered, locality suffers\n";
+}
+my @dig;
+my $nr = $ENV{NR} || 1;
+my $async = timeit($nr, sub {
+        my $dig = Digest::SHA->new(1);
+        my $cb = sub {
+                my ($bref) = @_;
+                $dig->add($$bref);
+        };
+        my $cat = $git->popen(@cat);
+        $git->cat_async_begin;
+
+        foreach (<$cat>) {
+                my ($oid, undef, undef) = split(/ /);
+                $git->cat_async($oid, $cb);
+        }
+        close $cat or die "cat: $?";
+        $git->cat_async_wait;
+        push @dig, ['async', $dig->hexdigest ];
+});
+
+my $sync = timeit($nr, sub {
+        my $dig = Digest::SHA->new(1);
+        my $cat = $git->popen(@cat);
+        foreach (<$cat>) {
+                my ($oid, undef, undef) = split(/ /);
+                my $bref = $git->cat_file($oid);
+                $dig->add($$bref);
+        }
+        close $cat or die "cat: $?";
+        push @dig, ['sync', $dig->hexdigest ];
+});
+
+ok(scalar(@dig) >= 2, 'got some digests');
+my $ref = shift @dig;
+my $exp = $ref->[1];
+isnt($exp, Digest::SHA->new(1)->hexdigest, 'not empty');
+foreach (@dig) {
+        is($_->[1], $exp, "digest matches $_->[0] <=> $ref->[0]");
+}
+diag "sync=".timestr($sync);
+diag "async=".timestr($async);
+done_testing;
+1;