git@vger.kernel.org mailing list mirror (one of many)
 help / color / mirror / code / Atom feed
From: "René Scharfe" <l.s.r@web.de>
To: Git List <git@vger.kernel.org>
Cc: Junio C Hamano <gitster@pobox.com>
Subject: [PATCH 6/9] test-mergesort: add unriffle_skewed mode
Date: Fri, 1 Oct 2021 11:17:57 +0200	[thread overview]
Message-ID: <8763ebde-772c-500c-20fc-cbde43dcefeb@web.de> (raw)
In-Reply-To: <943b1e01-465e-5def-a766-0adf667690de@web.de>

Add a mode that turns a sorted list into adversarial input for a
bottom-up mergesort implementation that doubles the length of sorted
sublists at each level -- like our llist_mergesort().

While unriffle mode splits the list in half at each recursion step,
unriffle_skewed splits it into 2^l items and the rest, with 2^l being
the highest power of two smaller than the number of items and thus
2^l >= rest.  The rest is unriffled with the tail of the first half to
require a merge to compare the maximum number of elements.

It complements the unriffle mode, which targets balanced merges.  If
the number of elements is a power of two then both actually produce the
same result, as 2^l == rest == n/2 at each recursion step in that case.

Here are the results:

   $ t/helper/test-tool mergesort test | awk '
      $7 > max[$3] {max[$3] = $7; line[$3] = $0}
      END {for (n in line) print line[n]}
   '

distribut mode                    n        m get_next set_next  compare verdict
sawtooth  unriffle_skewed       100      128     1184      700      589 OK
sawtooth  unriffle_skewed      1023     1024    16373    10230     9207 OK
sawtooth  unriffle             1024     1024    16384    10240     9217 OK
sawtooth  unriffle_skewed      1025     2048    18454    11275    10241 OK

The sawtooth distribution with m>=n produces a sorted list and
unriffle_skewed mode turns it into adversarial input for unbalanced
merges, which it wins in all cases except for n=1024 -- the resulting
list is the same, but unriffle is tested before unriffle_skewed, so its
result is selected by the AWK script.

Signed-off-by: René Scharfe <l.s.r@web.de>
---
 t/helper/test-mergesort.c | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/t/helper/test-mergesort.c b/t/helper/test-mergesort.c
index d71ef568f3..43ec74e2d3 100644
--- a/t/helper/test-mergesort.c
+++ b/t/helper/test-mergesort.c
@@ -178,6 +178,33 @@ static void mode_unriffle(int *arr, int n)
 	free(tmp);
 }

+static unsigned int prev_pow2(unsigned int n)
+{
+	unsigned int pow2 = 1;
+	while (pow2 * 2 < n)
+		pow2 *= 2;
+	return pow2;
+}
+
+static void unriffle_recursively_skewed(int *arr, int n, int *tmp)
+{
+	if (n > 1) {
+		int pow2 = prev_pow2(n);
+		int rest = n - pow2;
+		unriffle(arr + pow2 - rest, rest * 2, tmp);
+		unriffle_recursively_skewed(arr, pow2, tmp);
+		unriffle_recursively_skewed(arr + pow2, rest, tmp);
+	}
+}
+
+static void mode_unriffle_skewed(int *arr, int n)
+{
+	int *tmp;
+	ALLOC_ARRAY(tmp, n);
+	unriffle_recursively_skewed(arr, n, tmp);
+	free(tmp);
+}
+
 #define MODE(name) { #name, mode_##name }

 static struct mode {
@@ -191,6 +218,7 @@ static struct mode {
 	MODE(sort),
 	MODE(dither),
 	MODE(unriffle),
+	MODE(unriffle_skewed),
 };

 static const struct mode *get_mode_by_name(const char *name)
--
2.33.0

  parent reply	other threads:[~2021-10-01  9:18 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-01  9:07 [PATCH 0/9] mergesort: improve tests and performance René Scharfe
2021-10-01  9:10 ` [PATCH 1/9] test-mergesort: use strbuf_getline() René Scharfe
2021-10-02  9:08   ` Ævar Arnfjörð Bjarmason
2021-10-02 16:56     ` René Scharfe
2021-10-01  9:11 ` [PATCH 2/9] test-mergesort: add sort subcommand René Scharfe
2021-10-01 20:26   ` Junio C Hamano
2021-10-01  9:12 ` [PATCH 3/9] test-mergesort: add test subcommand René Scharfe
2021-10-01 20:26   ` Junio C Hamano
2021-10-02  8:35     ` Ævar Arnfjörð Bjarmason
2021-10-03 10:15       ` René Scharfe
2021-10-03 17:33         ` Junio C Hamano
2021-10-07 20:00           ` René Scharfe
2021-10-08  4:04             ` [PATCH 10/9 v2] test-mergesort: use repeatable random numbers René Scharfe
2021-10-08  4:17               ` Jeff King
2021-10-08  7:23               ` Ævar Arnfjörð Bjarmason
2021-10-08 17:30                 ` René Scharfe
2021-10-08 19:00                   ` Ævar Arnfjörð Bjarmason
2021-10-03 10:15       ` [PATCH 3/9] test-mergesort: add test subcommand René Scharfe
2021-10-01  9:14 ` [PATCH 4/9] test-mergesort: add generate subcommand René Scharfe
2021-10-01  9:16 ` [PATCH 5/9] test-mergesort: add unriffle mode René Scharfe
2021-10-01  9:17 ` René Scharfe [this message]
2021-10-01  9:19 ` [PATCH 7/9] p0071: measure sorting of already sorted and reversed files René Scharfe
2021-10-01  9:19 ` [PATCH 8/9] p0071: test performance of llist_mergesort() René Scharfe
2021-10-01  9:22 ` [PATCH 9/9] mergesort: use ranks stack René Scharfe
2022-01-17 17:43   ` Ævar Arnfjörð Bjarmason
2022-01-17 18:22     ` René Scharfe
2022-01-18  5:07       ` René Scharfe
2022-01-18 10:40         ` Ævar Arnfjörð Bjarmason
2022-01-18 12:27           ` René Scharfe

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

  List information: http://vger.kernel.org/majordomo-info.html

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=8763ebde-772c-500c-20fc-cbde43dcefeb@web.de \
    --to=l.s.r@web.de \
    --cc=git@vger.kernel.org \
    --cc=gitster@pobox.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this public inbox

	https://80x24.org/mirrors/git.git

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).