From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:56503) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gK2XQ-0007fY-6X for guix-patches@gnu.org; Tue, 06 Nov 2018 09:41:04 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1gK2XP-0002zQ-GI for guix-patches@gnu.org; Tue, 06 Nov 2018 09:41:04 -0500 Received: from debbugs.gnu.org ([208.118.235.43]:60176) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1gK2XP-0002yV-Cr for guix-patches@gnu.org; Tue, 06 Nov 2018 09:41:03 -0500 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1gK2XP-0001eR-65 for guix-patches@gnu.org; Tue, 06 Nov 2018 09:41:03 -0500 Subject: [bug#33210] Cuirass: Use a SQLite in single-thread mode Resent-Message-ID: From: ludo@gnu.org (Ludovic =?UTF-8?Q?Court=C3=A8s?=) References: <87a7mvqikl.fsf@lassieur.org> <871s80o2zc.fsf@gnu.org> <87ftwgq7da.fsf@lassieur.org> Date: Tue, 06 Nov 2018 15:40:11 +0100 In-Reply-To: <87ftwgq7da.fsf@lassieur.org> ("=?UTF-8?Q?Cl=C3=A9ment?= Lassieur"'s message of "Mon, 05 Nov 2018 09:02:57 +0100") Message-ID: <87ftwez2us.fsf@gnu.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-patches-bounces+kyle=kyleam.com@gnu.org Sender: "Guix-patches" To: =?UTF-8?Q?Cl=C3=A9ment?= Lassieur Cc: 33210@debbugs.gnu.org Hi Cl=C3=A9ment, Cl=C3=A9ment Lassieur skribis: > Ludovic Court=C3=A8s writes: > >> Hello, >> >> Cl=C3=A9ment Lassieur skribis: >> >>> These patches are supposed to slightly improve Cuirass' performances, >>> because it doesn't use the multi-threading features. >> >> Did you notice a measurable difference? > > I haven't done any measurement yet, but according to the SQLite > documentation: > > Setting -DSQLITE_THREADSAFE=3D0 causes all of the mutex and > thread-safety logic in SQLite to be omitted. This is the single > compile-time option that makes the most difference in optimizing the > performance of SQLite. > > So even if the optimization is small, it's the option that has the > biggest impact on performance. > >> We could do it, but IMO that should be a last resort because I=E2=80=99d= expect >> it to be a micro-optimization. > > Lots of micro-optimizations lead to an overall faster application ;-). > And this one doesn't make the code more complicated. To me it's just a > bonus. I agree it doesn=E2=80=99t complicate the code; still, that=E2=80=99s a cou= ple of additional package variants to deal with, for hardly measurable benefits I suspect. I think we should focus on higher-level optimizations at this development stage of Cuirass. For instance I have been meaning to patch it so that it doesn=E2=80=99t have to process all the build logs, since it doesn=E2=80=99t do anything with those logs and processing them involves to= ns of syscalls and string processing and introduces latency in fiber scheduling. This is a simple change that could have a more visible impact I believe. Hopefully I=E2=80=99ll get there real soon=E2=80=A6 WDYT? Ludo=E2=80=99.