1 Return-Path: <pieter@praet.org>
\r
2 X-Original-To: notmuch@notmuchmail.org
\r
3 Delivered-To: notmuch@notmuchmail.org
\r
4 Received: from localhost (localhost [127.0.0.1])
\r
5 by olra.theworths.org (Postfix) with ESMTP id 4809D429E26
\r
6 for <notmuch@notmuchmail.org>; Tue, 15 Nov 2011 08:48:42 -0800 (PST)
\r
7 X-Virus-Scanned: Debian amavisd-new at olra.theworths.org
\r
11 X-Spam-Status: No, score=-0.7 tagged_above=-999 required=5
\r
12 tests=[RCVD_IN_DNSWL_LOW=-0.7] autolearn=disabled
\r
13 Received: from olra.theworths.org ([127.0.0.1])
\r
14 by localhost (olra.theworths.org [127.0.0.1]) (amavisd-new, port 10024)
\r
15 with ESMTP id 6m9Gjjrqrkxc for <notmuch@notmuchmail.org>;
\r
16 Tue, 15 Nov 2011 08:48:41 -0800 (PST)
\r
17 Received: from mail-wy0-f181.google.com (mail-wy0-f181.google.com
\r
18 [74.125.82.181]) (using TLSv1 with cipher RC4-SHA (128/128 bits))
\r
19 (No client certificate requested)
\r
20 by olra.theworths.org (Postfix) with ESMTPS id 7AF54429E21
\r
21 for <notmuch@notmuchmail.org>; Tue, 15 Nov 2011 08:48:41 -0800 (PST)
\r
22 Received: by wyg19 with SMTP id 19so3955395wyg.26
\r
23 for <notmuch@notmuchmail.org>; Tue, 15 Nov 2011 08:48:40 -0800 (PST)
\r
24 Received: by 10.180.105.102 with SMTP id gl6mr31210724wib.46.1321375720253;
\r
25 Tue, 15 Nov 2011 08:48:40 -0800 (PST)
\r
26 Received: from localhost ([109.131.148.49])
\r
27 by mx.google.com with ESMTPS id n2sm14691483wiz.16.2011.11.15.08.48.39
\r
28 (version=TLSv1/SSLv3 cipher=OTHER);
\r
29 Tue, 15 Nov 2011 08:48:39 -0800 (PST)
\r
30 From: Pieter Praet <pieter@praet.org>
\r
31 To: Austin Clements <amdragon@MIT.EDU>
\r
32 Subject: Re: [PATCH] emacs: Use a single buffer invisibility spec to fix
\r
33 quadratic search cost.
\r
34 In-Reply-To: <20111111052716.GU2658@mit.edu>
\r
35 References: <1320807328-13728-1-git-send-email-amdragon@mit.edu>
\r
36 <877h382jax.fsf@SSpaeth.de>
\r
37 <CAPFwwQgsqe2NE9vm2RJHHK+8hWR_uMWKLHcxm0xkjduFboAfPw@mail.gmail.com>
\r
38 <87d3czxsu9.fsf@praet.org> <20111111045341.GS2658@mit.edu>
\r
39 <20111111052716.GU2658@mit.edu>
\r
40 User-Agent: Notmuch/0.9+76~g2fd88e6 (http://notmuchmail.org) Emacs/23.3.1
\r
41 (x86_64-unknown-linux-gnu)
\r
42 Date: Tue, 15 Nov 2011 17:47:46 +0100
\r
43 Message-ID: <87d3ctjpsd.fsf@praet.org>
\r
45 Content-Type: text/plain; charset=us-ascii
\r
46 Cc: notmuch@notmuchmail.org, servilio <servilio@gmail.com>
\r
47 X-BeenThere: notmuch@notmuchmail.org
\r
48 X-Mailman-Version: 2.1.13
\r
50 List-Id: "Use and development of the notmuch mail system."
\r
51 <notmuch.notmuchmail.org>
\r
52 List-Unsubscribe: <http://notmuchmail.org/mailman/options/notmuch>,
\r
53 <mailto:notmuch-request@notmuchmail.org?subject=unsubscribe>
\r
54 List-Archive: <http://notmuchmail.org/pipermail/notmuch>
\r
55 List-Post: <mailto:notmuch@notmuchmail.org>
\r
56 List-Help: <mailto:notmuch-request@notmuchmail.org?subject=help>
\r
57 List-Subscribe: <http://notmuchmail.org/mailman/listinfo/notmuch>,
\r
58 <mailto:notmuch-request@notmuchmail.org?subject=subscribe>
\r
59 X-List-Received-Date: Tue, 15 Nov 2011 16:48:42 -0000
\r
61 On Fri, 11 Nov 2011 00:27:16 -0500, Austin Clements <amdragon@MIT.EDU> wrote:
\r
62 > Quoth myself on Nov 10 at 11:53 pm:
\r
63 > > Quoth Pieter Praet on Nov 11 at 4:04 am:
\r
64 > > > I've tried getting some hard numbers using
\r
66 > > > #+begin_src sh
\r
67 > > > time emacs --eval '(progn
\r
69 > > > (notmuch-search "*")
\r
70 > > > (while (get-buffer-process (current-buffer))
\r
71 > > > (sleep-for 0.1))
\r
72 > > > (kill-emacs))'
\r
75 > > > ... but the results vary wildly on subsequent runs.
\r
77 > > For me, this doesn't actually display the results buffer (though I
\r
78 > > don't know why not), which means it won't test this, since the problem
\r
79 > > lies in the Emacs redisplay logic.
\r
81 > This may or may not actually be correct, but the following seems more
\r
82 > representative on my system:
\r
84 > time emacs --eval '(progn
\r
86 > (notmuch-search "*")
\r
87 > (while (get-buffer-process (current-buffer))
\r
93 Yup, that does the trick!
\r
95 These remained fairly consistent, even with both CPU cores enabled:
\r
98 | | before | after | before | after |
\r
99 |------+-----------+-----------+-----------+-----------|
\r
100 | real | 0m34.560s | 0m31.829s | 0m30.784s | 0m26.587s |
\r
101 | user | 0m26.188s | 0m23.672s | 0m27.028s | 0m22.889s |
\r
102 | sys | 0m0.863s | 0m0.907s | 0m1.200s | 0m1.243s |
\r
105 > This at least displays the buffer. I also tried
\r
106 > (accept-process-output) instead of the (sleep-for 0.1), which clearly
\r
107 > behaved differently, but gave only slightly higher numbers. If I
\r
108 > timed just the search part, to exclude emacs start-up, I would have a
\r
109 > better idea of which more closely matches my manual measurements.
\r
111 Oddly enough... I ran some tests using `elp-instrument-package' and
\r
112 your `time-it' macro [1] to keep emacs init out of the equation, and
\r
113 both not only produced horribly fluctuating results (even with only a
\r
114 single CPU core enabled), but often took *longer* as well!
\r
118 #+begin_src emacs-lisp
\r
121 (let ((elp-reset-after-results t))
\r
122 (elp-instrument-package "notmuch")
\r
123 (notmuch-search "*")
\r
124 (while (get-buffer-process (current-buffer))
\r
130 #+begin_src emacs-lisp
\r
131 (defmacro time-it (code)
\r
132 `(let ((start-time (get-internal-run-time)))
\r
134 (float-time (time-subtract (get-internal-run-time) start-time))))
\r
139 (notmuch-search "*")
\r
140 (while (get-buffer-process (current-buffer))
\r
151 [1] id:"20110713185721.GI25558@mit.edu"
\r