From unknown Sat Jul 26 15:32:05 2025 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Mailer: MIME-tools 5.509 (Entity 5.509) Content-Type: text/plain; charset=utf-8 From: bug#40626 <40626@debbugs.gnu.org> To: bug#40626 <40626@debbugs.gnu.org> Subject: Status: Poor performance on low-end ARMv7 devices Reply-To: bug#40626 <40626@debbugs.gnu.org> Date: Sat, 26 Jul 2025 22:32:05 +0000 retitle 40626 Poor performance on low-end ARMv7 devices reassign 40626 guix submitter 40626 Ludovic Court=C3=A8s severity 40626 important thanks From debbugs-submit-bounces@debbugs.gnu.org Tue Apr 14 12:23:45 2020 Received: (at submit) by debbugs.gnu.org; 14 Apr 2020 16:23:45 +0000 Received: from localhost ([127.0.0.1]:34584 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jOOLh-0000zW-40 for submit@debbugs.gnu.org; Tue, 14 Apr 2020 12:23:45 -0400 Received: from lists.gnu.org ([209.51.188.17]:46028) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jOOLf-0000zO-DZ for submit@debbugs.gnu.org; Tue, 14 Apr 2020 12:23:43 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:50984) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1jOOLZ-0002pw-PY for bug-guix@gnu.org; Tue, 14 Apr 2020 12:23:43 -0400 X-Spam-Checker-Version: SpamAssassin 3.3.2 (2011-06-06) on eggs.gnu.org X-Spam-Level: X-Spam-Status: No, score=-2.9 required=5.0 tests=ALL_TRUSTED,BAYES_00, URIBL_BLOCKED autolearn=disabled version=3.3.2 Received: from fencepost.gnu.org ([2001:470:142:3::e]:41450) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1jOM62-0007co-HZ for bug-guix@gnu.org; Tue, 14 Apr 2020 09:59:26 -0400 Received: from [2a01:e0a:1d:7270:af76:b9b:ca24:c465] (port=57784 helo=ribbon) by fencepost.gnu.org with esmtpsa (TLS1.2:RSA_AES_256_CBC_SHA1:256) (Exim 4.82) (envelope-from ) id 1jOM62-0004Hf-2o for bug-guix@gnu.org; Tue, 14 Apr 2020 09:59:26 -0400 From: =?utf-8?Q?Ludovic_Court=C3=A8s?= To: bug-guix@gnu.org Subject: Poor performance on low-end ARMv7 devices X-URL: http://www.fdn.fr/~lcourtes/ X-Revolutionary-Date: 26 Germinal an 228 de la =?utf-8?Q?R=C3=A9volution?= X-PGP-Key-ID: 0x090B11993D9AEBB5 X-PGP-Key: http://www.fdn.fr/~lcourtes/ludovic.asc X-PGP-Fingerprint: 3CE4 6455 8A84 FDC6 9DB4 0CFB 090B 1199 3D9A EBB5 X-OS: x86_64-pc-linux-gnu Date: Tue, 14 Apr 2020 15:59:24 +0200 Message-ID: <87a73exj5v.fsf@inria.fr> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/26.3 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-Spam-Score: -0.7 (/) X-Debbugs-Envelope-To: submit X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: debbugs-submit-bounces@debbugs.gnu.org Sender: "Debbugs-submit" X-Spam-Score: -1.7 (-) Hello! On my Olimex OLinuXino A20, here=E2=80=99s what I get with Guix on Guile 3.= 0.2: --8<---------------cut here---------------start------------->8--- $ guix describe Generation 1 Apr 11 2020 13:26:01 (current) guix 6720616 repository URL: https://git.savannah.gnu.org/git/guix.git commit: 6720616daaf711314827c157a660339990e5cb07 $ time guix build hello -d --no-grafts /gnu/store/s0di07vhva95rl8p3gimlna5ffca7xlq-hello-2.10.drv real 0m9.964s user 0m8.660s sys 0m0.810s --8<---------------cut here---------------end--------------->8--- Most of it seems to go in loading .go files: --8<---------------cut here---------------start------------->8--- $ guix repl GNU Guile 3.0.2 Copyright (C) 1995-2020 Free Software Foundation, Inc. Guile comes with ABSOLUTELY NO WARRANTY; for details type `,show w'. This program is free software, and you are welcome to redistribute it under certain conditions; type `,show c' for details. Enter `,help' for help. scheme@(guix-user)> ,time (use-modules (guix) (gnu) (gnu packages base)) ;; 8.104000s real time, 6.879000s run time. 0.755000s spent in GC. scheme@(guix-user)> (define s (open-connection)) scheme@(guix-user)> ,time (package-derivation s hello #:graft? #f) $1 =3D # /gnu/store/z4ig2v79x3y7r5w1gs7rg7cgyvja4dn1-hello-2.10 31a5bb8> ;; 3.101000s real time, 2.493000s run time. 0.324000s spent in GC. --8<---------------cut here---------------end--------------->8--- Ludo=E2=80=99. From debbugs-submit-bounces@debbugs.gnu.org Tue Apr 14 14:08:20 2020 Received: (at 40626) by debbugs.gnu.org; 14 Apr 2020 18:08:20 +0000 Received: from localhost ([127.0.0.1]:34757 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jOPyu-0007kT-AN for submit@debbugs.gnu.org; Tue, 14 Apr 2020 14:08:20 -0400 Received: from lepiller.eu ([89.234.186.109]:33700) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jOPys-0007kJ-9c for 40626@debbugs.gnu.org; Tue, 14 Apr 2020 14:08:19 -0400 Received: from lepiller.eu (localhost [127.0.0.1]) by lepiller.eu (OpenSMTPD) with ESMTP id 91be0833; Tue, 14 Apr 2020 18:08:16 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed; d=lepiller.eu; h=date :in-reply-to:references:mime-version:content-type :content-transfer-encoding:subject:to:from:message-id; s=dkim; bh=blcnF5N47unTLUYgMGMm52DVFzRkcUTZ7ufxi5SrBjc=; b=gJ6MKpA3Poq8 RHtJLMxdruAkqy4J1JcH+gkVct6dZfvCY6Knf5XRMNdZ0EUbRs01Ol4P09VOjPrF nJOmdAF9XFDe2NrN5BDwwetGQu4d/v58O3A7vd3Ordn44UooetSI6EKILwd72qas q0MdGBz56Xt469mzguBJwH5UOEkp9R7u2pMvSMRv+DBeKq0cjiLPgVS0Gu9JIkrK gkSWh89H1E9d21t593/nC2Co9yT30eBYc6BcfPWuFQDl678Nk4zHVT5iHufu3Y6m JPriaXm22NXDAhfoH9+TVr7iN64xfOC6+bKjPUlOZpcgK/MFOO6cboBBLtZPPQ1Z JES6h3ZGQQ== Received: by lepiller.eu (OpenSMTPD) with ESMTPSA id d5e654b5 (TLSv1.2:ECDHE-RSA-AES256-GCM-SHA384:256:NO); Tue, 14 Apr 2020 18:08:15 +0000 (UTC) Date: Tue, 14 Apr 2020 14:08:00 -0400 User-Agent: K-9 Mail for Android In-Reply-To: <87a73exj5v.fsf@inria.fr> References: <87a73exj5v.fsf@inria.fr> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Subject: Re: bug#40626: Poor performance on low-end ARMv7 devices To: =?ISO-8859-1?Q?Ludovic_Court=E8s?= ,40626@debbugs.gnu.org From: Julien Lepiller Message-ID: <2AB0E1B2-4104-4378-B872-DF8E3673D15B@lepiller.eu> X-Spam-Score: 0.0 (/) X-Debbugs-Envelope-To: 40626 X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: debbugs-submit-bounces@debbugs.gnu.org Sender: "Debbugs-submit" X-Spam-Score: -1.0 (-) Le 14 avril 2020 09:59:24 GMT-04:00, "Ludovic Court=C3=A8s" a =C3=A9crit : >Hello! > >On my Olimex OLinuXino A20, here=E2=80=99s what I get with Guix on Guile = 3=2E0=2E2: > >--8<---------------cut here---------------start------------->8--- >$ guix describe >Generation 1 Apr 11 2020 13:26:01 (current) > guix 6720616 > repository URL: https://git=2Esavannah=2Egnu=2Eorg/git/guix=2Egit > commit: 6720616daaf711314827c157a660339990e5cb07 >$ time guix build hello -d --no-grafts >/gnu/store/s0di07vhva95rl8p3gimlna5ffca7xlq-hello-2=2E10=2Edrv > >real 0m9=2E964s >user 0m8=2E660s >sys 0m0=2E810s >--8<---------------cut here---------------end--------------->8--- > >Most of it seems to go in loading =2Ego files: > >--8<---------------cut here---------------start------------->8--- >$ guix repl >GNU Guile 3=2E0=2E2 >Copyright (C) 1995-2020 Free Software Foundation, Inc=2E > >Guile comes with ABSOLUTELY NO WARRANTY; for details type `,show w'=2E >This program is free software, and you are welcome to redistribute it >under certain conditions; type `,show c' for details=2E > >Enter `,help' for help=2E >scheme@(guix-user)> ,time (use-modules (guix) (gnu) (gnu packages >base)) >;; 8=2E104000s real time, 6=2E879000s run time=2E 0=2E755000s spent in G= C=2E >scheme@(guix-user)> (define s (open-connection)) >scheme@(guix-user)> ,time (package-derivation s hello #:graft? #f) >$1 =3D #/gnu/store/s0di07vhva95rl8p3gimlna5ffca7xlq-hello-2=2E10=2Edrv =3D> >/gnu/store/z4ig2v79x3y7r5w1gs7rg7cgyvja4dn1-hello-2=2E10 31a5bb8> >;; 3=2E101000s real time, 2=2E493000s run time=2E 0=2E324000s spent in G= C=2E >--8<---------------cut here---------------end--------------->8--- > >Ludo=E2=80=99=2E In case it helps, I observe the same kind of timing on my cubietruck (same= guile, but guix from april 1st)=2E It's slightly faster than you, the proc= essor is a cortex-a7=2E From debbugs-submit-bounces@debbugs.gnu.org Tue Apr 14 15:09:06 2020 Received: (at 40626) by debbugs.gnu.org; 14 Apr 2020 19:09:06 +0000 Received: from localhost ([127.0.0.1]:34837 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jOQvi-0002vI-20 for submit@debbugs.gnu.org; Tue, 14 Apr 2020 15:09:06 -0400 Received: from dd26836.kasserver.com ([85.13.145.193]:39182) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jOQvf-0002uz-3D for 40626@debbugs.gnu.org; Tue, 14 Apr 2020 15:09:04 -0400 Received: from localhost (80-110-126-163.cgn.dynamic.surfer.at [80.110.126.163]) by dd26836.kasserver.com (Postfix) with ESMTPSA id ACB13336815F; Tue, 14 Apr 2020 21:09:01 +0200 (CEST) Date: Tue, 14 Apr 2020 21:08:59 +0200 From: Danny Milosavljevic To: Ludovic =?ISO-8859-1?Q?Court=E8s?= Subject: Re: bug#40626: Poor performance on low-end ARMv7 devices Message-ID: <20200414210859.4529c305@scratchpost.org> In-Reply-To: <87a73exj5v.fsf@inria.fr> References: <87a73exj5v.fsf@inria.fr> X-Mailer: Claws Mail 3.17.5 (GTK+ 2.24.32; x86_64-unknown-linux-gnu) MIME-Version: 1.0 Content-Type: multipart/signed; boundary="Sig_/1cqL6toGj3k6gvvLIBKRfZn"; protocol="application/pgp-signature"; micalg=pgp-sha256 X-Spam-Score: -0.7 (/) X-Debbugs-Envelope-To: 40626 Cc: 40626@debbugs.gnu.org X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: debbugs-submit-bounces@debbugs.gnu.org Sender: "Debbugs-submit" X-Spam-Score: -1.7 (-) --Sig_/1cqL6toGj3k6gvvLIBKRfZn Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable For comparison, on a Banana Pi M2 Ultra I get: root@bpi-iot-ros-ai:/# time guix build hello -d --no-grafts /gnu/store/yp46hszc04dx2zy6kscy1zmjfg9y8flq-hello-2.10.drv real 0m13.481s user 0m11.230s sys 0m0.900s root@bpi-iot-ros-ai:/# time guix build hello -d --no-grafts /gnu/store/yp46hszc04dx2zy6kscy1zmjfg9y8flq-hello-2.10.drv real 0m11.471s user 0m10.820s sys 0m0.380s root@bpi-iot-ros-ai:/# guix repl GNU Guile 2.2.4 Copyright (C) 1995-2017 Free Software Foundation, Inc. Guile comes with ABSOLUTELY NO WARRANTY; for details type `,show w'. This program is free software, and you are welcome to redistribute it under certain conditions; type `,show c' for details. Enter `,help' for help. scheme@(guix-user)> ,time (use-modules (guix) (gnu) (gnu packages base)) ;; 3.360000s real time, 3.270000s run time. 0.519000s spent in GC. scheme@(guix-user)> (define s (open-connection)) scheme@(guix-user)> ,time (package-derivation s hello #:graft? #f) $1 =3D # /gnu/store/cp0qvs3vxrqiyyxfi8556n52x7ax8khf-hello-2.10 14fabb8> ;; 6.829000s real time, 6.579000s run time. 1.019000s spent in GC. Processor is a Cortex A7. Note: You can do some RAM training and if RAM is connected better on your board, you can sometimes have staggering speed gains. See also https://linux-sunxi.org/A10_DRAM_Controller_Calibration --Sig_/1cqL6toGj3k6gvvLIBKRfZn Content-Type: application/pgp-signature Content-Description: OpenPGP digital signature -----BEGIN PGP SIGNATURE----- iQEzBAEBCAAdFiEEds7GsXJ0tGXALbPZ5xo1VCwwuqUFAl6WCksACgkQ5xo1VCww uqWCRwgAk/gZNM0K1fkKPY0GAcMJd9nmg98vl6uP2o/b68KyQOKkCAtbeZiOh0zd voN+sVLhPZ/x/V7ByWeKODL6ySL9KYrK7FBzIc4S+5g1yU0uQ4AA2kKfrZDlPQpA uKvmrN+KEbs69VLvqDPVqVq+zOpL9NwdZGshpkIBWruiOfCFFMSFYPA5UsGEKHrO lVRKBkvODoEOS+Tnq51yfo0RXxgmg4Iy4Um0C0ynZ4Bsbbs2bDIZK8AnADTNyRos yW5lra6BWpgADGJSJBfNorzVE9Nrn4ZlZR4QR0KmR0ngv7EloCyvnWbPNEPFfO3Q sGWlaQXFvx75qGp4HKHwd4Jd/qOROA== =oHvp -----END PGP SIGNATURE----- --Sig_/1cqL6toGj3k6gvvLIBKRfZn-- From debbugs-submit-bounces@debbugs.gnu.org Tue Apr 14 16:01:12 2020 Received: (at 40626) by debbugs.gnu.org; 14 Apr 2020 20:01:12 +0000 Received: from localhost ([127.0.0.1]:34868 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jORk8-0004C0-Mi for submit@debbugs.gnu.org; Tue, 14 Apr 2020 16:01:12 -0400 Received: from eggs.gnu.org ([209.51.188.92]:56883) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jORk6-0004Bm-Ex for 40626@debbugs.gnu.org; Tue, 14 Apr 2020 16:01:11 -0400 Received: from fencepost.gnu.org ([2001:470:142:3::e]:49263) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1jORk1-0003Y6-0j; Tue, 14 Apr 2020 16:01:05 -0400 Received: from [2a01:e0a:1d:7270:af76:b9b:ca24:c465] (port=60556 helo=ribbon) by fencepost.gnu.org with esmtpsa (TLS1.2:RSA_AES_256_CBC_SHA1:256) (Exim 4.82) (envelope-from ) id 1jORk0-0008FO-AJ; Tue, 14 Apr 2020 16:01:04 -0400 From: =?utf-8?Q?Ludovic_Court=C3=A8s?= To: Danny Milosavljevic Subject: Re: bug#40626: Poor performance on low-end ARMv7 devices References: <87a73exj5v.fsf@inria.fr> <20200414210859.4529c305@scratchpost.org> X-URL: http://www.fdn.fr/~lcourtes/ X-Revolutionary-Date: 26 Germinal an 228 de la =?utf-8?Q?R=C3=A9volution?= X-PGP-Key-ID: 0x090B11993D9AEBB5 X-PGP-Key: http://www.fdn.fr/~lcourtes/ludovic.asc X-PGP-Fingerprint: 3CE4 6455 8A84 FDC6 9DB4 0CFB 090B 1199 3D9A EBB5 X-OS: x86_64-pc-linux-gnu Date: Tue, 14 Apr 2020 22:01:02 +0200 In-Reply-To: <20200414210859.4529c305@scratchpost.org> (Danny Milosavljevic's message of "Tue, 14 Apr 2020 21:08:59 +0200") Message-ID: <87sgh5vnup.fsf@gnu.org> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/26.3 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-Spam-Score: -0.7 (/) X-Debbugs-Envelope-To: 40626 Cc: 40626@debbugs.gnu.org X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: debbugs-submit-bounces@debbugs.gnu.org Sender: "Debbugs-submit" X-Spam-Score: -1.7 (-) Hi Danny, Danny Milosavljevic skribis: > root@bpi-iot-ros-ai:/# guix repl > GNU Guile 2.2.4 This is a fairly old Guix though, so not directly comparable. Could you try with a current Guix? (Yeah, you=E2=80=99ll have to pull, and that takes verrrrry loong. Be sure= to pick a commit that=E2=80=99s fully built on .) Ludo=E2=80=99. From debbugs-submit-bounces@debbugs.gnu.org Thu Apr 16 04:22:44 2020 Received: (at control) by debbugs.gnu.org; 16 Apr 2020 08:22:44 +0000 Received: from localhost ([127.0.0.1]:37445 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jOznH-0000Sb-TJ for submit@debbugs.gnu.org; Thu, 16 Apr 2020 04:22:44 -0400 Received: from eggs.gnu.org ([209.51.188.92]:57832) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jOznG-0000SP-VW for control@debbugs.gnu.org; Thu, 16 Apr 2020 04:22:43 -0400 Received: from fencepost.gnu.org ([2001:470:142:3::e]:55675) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1jOznB-0003Kr-Ru for control@debbugs.gnu.org; Thu, 16 Apr 2020 04:22:37 -0400 Received: from [2a01:e0a:1d:7270:af76:b9b:ca24:c465] (port=40006 helo=ribbon) by fencepost.gnu.org with esmtpsa (TLS1.2:RSA_AES_256_CBC_SHA1:256) (Exim 4.82) (envelope-from ) id 1jOznA-0001oZ-9A for control@debbugs.gnu.org; Thu, 16 Apr 2020 04:22:37 -0400 Date: Thu, 16 Apr 2020 10:22:35 +0200 Message-Id: <87tv1jkfg4.fsf@gnu.org> To: control@debbugs.gnu.org From: =?utf-8?Q?Ludovic_Court=C3=A8s?= Subject: control message for bug #40626 MIME-version: 1.0 Content-type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-Spam-Score: -1.5 (-) X-Debbugs-Envelope-To: control X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: debbugs-submit-bounces@debbugs.gnu.org Sender: "Debbugs-submit" X-Spam-Score: -2.5 (--) severity 40626 important quit From debbugs-submit-bounces@debbugs.gnu.org Thu Apr 16 11:21:54 2020 Received: (at 40626) by debbugs.gnu.org; 16 Apr 2020 15:21:54 +0000 Received: from localhost ([127.0.0.1]:39129 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jP6Kv-0002jZ-NK for submit@debbugs.gnu.org; Thu, 16 Apr 2020 11:21:54 -0400 Received: from eggs.gnu.org ([209.51.188.92]:34380) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jP6Kt-0002jL-6n for 40626@debbugs.gnu.org; Thu, 16 Apr 2020 11:21:52 -0400 Received: from fencepost.gnu.org ([2001:470:142:3::e]:34020) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1jP6Ko-0002Cz-2g for 40626@debbugs.gnu.org; Thu, 16 Apr 2020 11:21:46 -0400 Received: from [2a01:e0a:1d:7270:af76:b9b:ca24:c465] (port=42240 helo=ribbon) by fencepost.gnu.org with esmtpsa (TLS1.2:RSA_AES_256_CBC_SHA1:256) (Exim 4.82) (envelope-from ) id 1jP6Kn-0007cx-NI for 40626@debbugs.gnu.org; Thu, 16 Apr 2020 11:21:45 -0400 From: =?utf-8?Q?Ludovic_Court=C3=A8s?= To: 40626@debbugs.gnu.org Subject: Re: bug#40626: Poor performance on low-end ARMv7 devices References: <87a73exj5v.fsf@inria.fr> Date: Thu, 16 Apr 2020 17:21:43 +0200 In-Reply-To: <87a73exj5v.fsf@inria.fr> ("Ludovic \=\?utf-8\?Q\?Court\=C3\=A8s\=22\?\= \=\?utf-8\?Q\?'s\?\= message of "Tue, 14 Apr 2020 15:59:24 +0200") Message-ID: <875zdzihh4.fsf@gnu.org> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/26.3 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-Spam-Score: -1.5 (-) X-Debbugs-Envelope-To: 40626 X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: debbugs-submit-bounces@debbugs.gnu.org Sender: "Debbugs-submit" X-Spam-Score: -2.5 (--) Ludovic Court=C3=A8s skribis: > Most of it seems to go in loading .go files: On x86_64, I=E2=80=99ve run: ./pre-inst-env perf record guile -c '(use-modules (gnu packages libreoffi= ce))' which shows mostly GC activity, along with symbol interning activity: --8<---------------cut here---------------start------------->8--- 14.71% guile libgc.so.1.3.6 [.] GC_mark_from 7.49% guile libgc.so.1.3.6 [.] GC_header_cache_miss 5.56% guile libguile-3.0.so.1.1.1 [.] vm_regular_engine 5.42% guile libgc.so.1.3.6 [.] GC_add_to_black_list_normal 2.66% guile libpthread-2.29.so [.] __pthread_mutex_unlock_usercnt 2.63% guile libgc.so.1.3.6 [.] GC_find_header 2.09% guile ld-2.29.so [.] _dl_update_slotinfo 1.88% guile libguile-3.0.so.1.1.1 [.] scm_c_weak_set_lookup 1.68% guile libguile-3.0.so.1.1.1 [.] narrow_string_hash 1.64% guile libguile-3.0.so.1.1.1 [.] scm_i_is_narrow_string 1.55% guile libguile-3.0.so.1.1.1 [.] scm_ihashq 1.52% guile libguile-3.0.so.1.1.1 [.] scm_sloppy_assq 1.32% guile libgc.so.1.3.6 [.] GC_move_disappearing_link_inn= er 1.23% guile libgc.so.1.3.6 [.] GC_malloc_kind 1.23% guile libpthread-2.29.so [.] __pthread_mutex_lock 1.22% guile libguile-3.0.so.1.1.1 [.] scm_hash_fn_get_handle 1.19% guile libpthread-2.29.so [.] __pthread_mutex_trylock 1.12% guile libguile-3.0.so.1.1.1 [.] get_callee_vcode 1.12% guile libguile-3.0.so.1.1.1 [.] scm_equal_p --8<---------------cut here---------------end--------------->8--- Back on my A20 board, I get this (unhelpful) GC profile: --8<---------------cut here---------------start------------->8--- scheme@(guix-user)> ,use(statprof) scheme@(guix-user)> (gcprof (lambda () (resolve-module '(gnu packages base)= ))) % cumulative self time seconds seconds procedure 100.00 5.13 5.13 ice-9/boot-9.scm:2201:0:%load-announce 0.00 4081.53 0.00 ice-9/boot-9.scm:220:5:map1 0.00 621.21 0.00 ice-9/threads.scm:388:4 0.00 310.61 0.00 ice-9/boot-9.scm:2803:0:save-module-excursion 0.00 310.61 0.00 anon #x1b15600 0.00 310.61 0.00 ice-9/boot-9.scm:3211:7 0.00 310.61 0.00 ice-9/boot-9.scm:3500:5 0.00 310.61 0.00 ice-9/boot-9.scm:3508:21 0.00 305.47 0.00 ice-9/boot-9.scm:3256:0:resolve-interface 0.00 295.21 0.00 ice-9/boot-9.scm:3381:5 0.00 295.21 0.00 ice-9/boot-9.scm:3351:0:define-module* 0.00 5.13 0.00 anon #xb2d8d098 0.00 5.13 0.00 anon #xb32d0098 0.00 5.13 0.00 anon #xb2fca098 0.00 5.13 0.00 anon #xb32e2098 0.00 5.13 0.00 anon #xb3343098 0.00 5.13 0.00 ice-9/boot-9.scm:2557:0:call-with-deferred-obse= rvers [=E2=80=A6] --- Sample count: 2 Total time: 5.134 seconds (0.44 seconds in GC) --8<---------------cut here---------------end--------------->8--- and this profile: --8<---------------cut here---------------start------------->8--- scheme@(guix-user)> ,pr (resolve-module '(gnu packages base)) % cumulative self time seconds seconds procedure 17.86 1.38 1.38 ice-9/boot-9.scm:2201:0:%load-announce 5.36 0.41 0.41 ice-9/boot-9.scm:3545:0:autoload-done! 4.76 0.37 0.37 ice-9/boot-9.scm:3540:0:autoload-done-or-in-pro= gress? 4.76 0.37 0.37 anon #x1a671cc 4.17 0.32 0.32 ice-9/format.scm:113:2:format:format-work 3.57 0.28 0.28 ice-9/boot-9.scm:3552:0:autoload-in-progress! 2.98 0.55 0.23 ice-9/boot-9.scm:1396:0:symbol-append 2.98 0.23 0.23 anon #x1a65c28 1.79 0.32 0.14 ice-9/boot-9.scm:2729:0:module-make-local-var! 1.79 0.14 0.14 anon #x1a671a4 1.79 0.14 0.14 anon #x1a6c284 1.79 0.14 0.14 anon #x1a67644 1.79 0.14 0.14 ice-9/boot-9.scm:3209:4 1.19 6106.96 0.09 ice-9/boot-9.scm:220:5:map1 1.19 518.54 0.09 ice-9/boot-9.scm:3211:7 1.19 511.19 0.09 ice-9/boot-9.scm:3381:5 1.19 5.69 0.09 anon #xb13f9098 1.19 0.18 0.09 anon #xad994098 1.19 0.09 0.09 anon #x1a6848c 1.19 0.09 0.09 srfi/srfi-60.scm:57:0:bitwise-if 1.19 0.09 0.09 ice-9/boot-9.scm:2468:2 1.19 0.09 0.09 ice-9/boot-9.scm:2468:2 1.19 0.09 0.09 anon #x1a6e284 1.19 0.09 0.09 anon #x1a67cd4 1.19 0.09 0.09 anon #x1a6e87c [=E2=80=A6] --- Sample count: 168 Total time: 7.714 seconds (1.279 seconds in GC) --8<---------------cut here---------------end--------------->8--- What=E2=80=99s the deal with =E2=80=98%load-announce=E2=80=99? How many ti= mes is it called? --8<---------------cut here---------------start------------->8--- $ guix repl GNU Guile 3.0.2 Copyright (C) 1995-2020 Free Software Foundation, Inc. Guile comes with ABSOLUTELY NO WARRANTY; for details type `,show w'. This program is free software, and you are welcome to redistribute it under certain conditions; type `,show c' for details. Enter `,help' for help. scheme@(guix-user)> ,use(system vm vm) scheme@(guix-user)> (set-vm-engine! 'debug) scheme@(guix-user)> ,use(statprof) scheme@(guix-user)> (statprof (lambda () (resolve-module '(gnu packages bas= e))) #:count-calls? #t) % cumulative self time seconds seconds calls procedure 5.31 4.39 4.39 anon #x1ba81cc 3.84 98371.57 3.17 69196 ice-9/boot-9.scm:220:5:map1 3.32 2.75 2.75 57783 ice-9/boot-9.scm:2468:2 3.26 2.70 2.70 anon #x1ba81a4 3.01 2.49 2.49 60626 ice-9/boot-9.scm:1299:5 2.94 9.23 2.43 44668 ice-9/boot-9.scm:2757:0:module-add! 2.85 4.39 2.35 47021 ice-9/boot-9.scm:2549:0:module-modified 2.53 2.09 2.09 44668 ice-9/boot-9.scm:2652:0:module-obarray-= set! 2.33 1.93 1.93 anon #x1ba6c28 2.24 1.85 1.85 anon #x1baa1fc 2.21 1.83 1.83 anon #x1bad284 2.11 1.75 1.75 anon #x1ba8cd4 2.01 4.55 1.67 32603 ice-9/boot-9.scm:2790:0:module-ref-subm= odule 1.92 1.59 1.59 anon #x1ba8f34 1.89 1.56 1.56 21349 ice-9/boot-9.scm:806:0:and=3D> 1.76 1.46 1.46 32184 ice-9/boot-9.scm:2468:2 1.69 1.40 1.40 33343 ice-9/boot-9.scm:2468:2 1.69 1.40 1.40 anon #x1ba8080 1.50 15.24 1.24 19850 ice-9/boot-9.scm:3339:29 1.44 7457.26 1.19 7325 ice-9/boot-9.scm:3256:0:resolve-interfa= ce 1.44 1.19 1.19 19133 ice-9/boot-9.scm:3345:2 1.41 9.79 1.16 19217 ice-9/boot-9.scm:1396:0:symbol-append 1.37 1.14 1.14 20160 ice-9/boot-9.scm:2468:2 1.34 1.11 1.11 19850 ice-9/boot-9.scm:3303:48 1.34 1.11 1.11 14322 srfi/srfi-60.scm:57:0:bitwise-if 1.31 1.08 1.08 36200 anon #x1ba6c18 1.15 2.65 0.95 9672 ice-9/vlist.scm:544:0:vhash-assv 1.15 2.09 0.95 20671 ice-9/boot-9.scm:215:2:map 1.15 0.95 0.95 14210 ice-9/boot-9.scm:3209:4 1.12 7.09 0.93 12743 ice-9/boot-9.scm:2729:0:module-make-loc= al-var! 1.05 0.87 0.87 11594 guix/utils.scm:810:0:source-properties-= >location 0.99 1.35 0.82 11559 ice-9/boot-9.scm:2746:0:module-ensure-l= ocal-variable! 0.90 0.74 0.74 370 ice-9/boot-9.scm:2201:0:%load-announce [=E2=80=A6] --- Sample count: 3128 Total time: 82.752 seconds (16.401 seconds in GC) --8<---------------cut here---------------end--------------->8--- Only 370 times. Hmm could something be wrong with the mapping of instruction pointers to source code location? Ludo=E2=80=99. From debbugs-submit-bounces@debbugs.gnu.org Fri Jul 03 00:41:13 2020 Received: (at 40626) by debbugs.gnu.org; 3 Jul 2020 04:41:13 +0000 Received: from localhost ([127.0.0.1]:56109 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jrDVh-0006ne-GU for submit@debbugs.gnu.org; Fri, 03 Jul 2020 00:41:13 -0400 Received: from cyberdimension.org ([80.67.179.20]:42358 helo=gnutoo.cyberdimension.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jr9DI-0008E4-Rc for 40626@debbugs.gnu.org; Thu, 02 Jul 2020 20:05:57 -0400 Received: from gnutoo.cyberdimension.org (localhost [127.0.0.1]) by cyberdimension.org (OpenSMTPD) with ESMTP id e0281888 for <40626@debbugs.gnu.org>; Fri, 3 Jul 2020 00:02:39 +0000 (UTC) Received: from primarylaptop.localdomain (localhost.localdomain [::1]) by gnutoo.cyberdimension.org (OpenSMTPD) with ESMTP id 208c83e9 for <40626@debbugs.gnu.org>; Fri, 3 Jul 2020 00:02:39 +0000 (UTC) Date: Fri, 3 Jul 2020 01:57:25 +0200 From: Denis 'GNUtoo' Carikli To: 40626@debbugs.gnu.org Subject: Re: Poor performance on low-end ARMv7 devices Message-ID: <20200703015725.11a6515c@primarylaptop.localdomain> X-Mailer: Claws Mail 3.17.5 (GTK+ 2.24.32; i686-pc-linux-gnu) MIME-Version: 1.0 Content-Type: multipart/signed; boundary="Sig_/kBOEP73xSbA9HTcJOA7o/qL"; protocol="application/pgp-signature"; micalg=pgp-sha256 X-Spam-Score: -0.0 (/) X-Debbugs-Envelope-To: 40626 X-Mailman-Approved-At: Fri, 03 Jul 2020 00:41:12 -0400 X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: debbugs-submit-bounces@debbugs.gnu.org Sender: "Debbugs-submit" X-Spam-Score: -1.0 (-) --Sig_/kBOEP73xSbA9HTcJOA7o/qL Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable Hi, Many ARM Single Board Computers are commonly used with microSD for storage, and some microSD cards are extremely slow (and sometimes unreliable when they are old). Could the performance issues be related to storage device I/Os? Denis. --Sig_/kBOEP73xSbA9HTcJOA7o/qL Content-Type: application/pgp-signature Content-Description: OpenPGP digital signature -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEEeC+d2+Nrp/PU3kkGX138wUF34mMFAl7+dGUACgkQX138wUF3 4mN9Ag//dM8BxgVhrFiAr4LDqWdAaRwyu0XfWTzfGZNIfFSXQKHk5wzRwlvVmQTs zi0xEs8bP7tMJ3Op2ZxrM8lOJvYXSzBclnC8Vh8xKHcD+99IzE79bNKX0UYOSNLM X4gI6jfIJZyVe13IGUSWhCxbZTfR0/cHgegonsJ+Ajr8FH2Qvv6kvgsAqio74ujK +iQwjTWCT4+K3VGfyGhjTcfRmoSV6DfqhOP12mMevf6YYwv6iu2a7f9vE1FxibrV ckwYQsXryfmHMzOUMuqbqolwOQu8xdGURHFn14RcYibeYemrryWQpkNaN1SPk5T7 svL3kQ1uGH774arVGc8ULozDr8TpdkTD8iCTF457Zm5F42MT5tYuKqGSselvtE3K 50Lw6yYSravSl2XjUWwUs+MxCKD0OYb2/S9e2QoY1M/vQeefX/YVX3s1ZsHMsUZk LDE40I9NVk6MnCSWILlMq257r/n7OP/vt6tGnmxRpAnCczAz5Iyv1nWU2OuuFwSO 5T81LGcZuT0RTPk4GyttybmQdXihNpk/1yRfacnFZ+QNzsK2AaSBgLiO2ZJOPACL XddcqUW+vZJF2kuYoQ6Hv+bND3hMrtWToKUzjSYKgC4iurBtTqlM0v0kz6hd8M28 2Lj4kAEefihgQUt7S3f77sjNptQxlviUyEX4/2PUh3nlDAsB4Vc= =Zxkt -----END PGP SIGNATURE----- --Sig_/kBOEP73xSbA9HTcJOA7o/qL-- From debbugs-submit-bounces@debbugs.gnu.org Fri Jul 03 03:22:59 2020 Received: (at 40626) by debbugs.gnu.org; 3 Jul 2020 07:23:00 +0000 Received: from localhost ([127.0.0.1]:56331 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jrG2F-0002iK-OW for submit@debbugs.gnu.org; Fri, 03 Jul 2020 03:22:59 -0400 Received: from eggs.gnu.org ([209.51.188.92]:50194) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jrG2C-0002i6-3N for 40626@debbugs.gnu.org; Fri, 03 Jul 2020 03:22:58 -0400 Received: from fencepost.gnu.org ([2001:470:142:3::e]:49222) by eggs.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1jrG26-0006O3-Ec; Fri, 03 Jul 2020 03:22:50 -0400 Received: from [2a01:e0a:fa:a50:9a7:de3d:abc2:6570] (port=35776 helo=meru) by fencepost.gnu.org with esmtpsa (TLS1.2:RSA_AES_256_CBC_SHA1:256) (Exim 4.82) (envelope-from ) id 1jrG25-000437-Vf; Fri, 03 Jul 2020 03:22:50 -0400 From: Mathieu Othacehe To: Denis 'GNUtoo' Carikli Subject: Re: bug#40626: Poor performance on low-end ARMv7 devices References: <87a73exj5v.fsf@inria.fr> <20200703015725.11a6515c@primarylaptop.localdomain> Date: Fri, 03 Jul 2020 09:22:48 +0200 In-Reply-To: <20200703015725.11a6515c@primarylaptop.localdomain> (Denis Carikli's message of "Fri, 3 Jul 2020 01:57:25 +0200") Message-ID: <873669kqpj.fsf@gnu.org> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/26.3 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain X-Spam-Score: -2.3 (--) X-Debbugs-Envelope-To: 40626 Cc: 40626@debbugs.gnu.org X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: debbugs-submit-bounces@debbugs.gnu.org Sender: "Debbugs-submit" X-Spam-Score: -3.3 (---) Hello, > Many ARM Single Board Computers are commonly used with microSD for > storage, and some microSD cards are extremely slow (and sometimes > unreliable when they are old). > > Could the performance issues be related to storage device I/Os? For those ARMv7 devices, which have few substitutes available and not much computation power, an approach is to work from a x86 machine and cross-compile disk-images. We will also have network booting supported soon, which will allow to do the same, but without having to write an SD card. In the meantime I would propose to close this bug, as there's not much that can be done, I think. Thanks, Mathieu From debbugs-submit-bounces@debbugs.gnu.org Mon Jul 13 09:34:25 2020 Received: (at 40626) by debbugs.gnu.org; 13 Jul 2020 13:34:25 +0000 Received: from localhost ([127.0.0.1]:47781 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1juybA-0007eN-Nf for submit@debbugs.gnu.org; Mon, 13 Jul 2020 09:34:25 -0400 Received: from eggs.gnu.org ([209.51.188.92]:38646) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1juyb8-0007eA-QM for 40626@debbugs.gnu.org; Mon, 13 Jul 2020 09:34:23 -0400 Received: from fencepost.gnu.org ([2001:470:142:3::e]:35894) by eggs.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1juyb3-0001K4-8c; Mon, 13 Jul 2020 09:34:17 -0400 Received: from [2a01:e0a:1d:7270:af76:b9b:ca24:c465] (port=34724 helo=ribbon) by fencepost.gnu.org with esmtpsa (TLS1.2:RSA_AES_256_CBC_SHA1:256) (Exim 4.82) (envelope-from ) id 1juyb2-00020l-Pz; Mon, 13 Jul 2020 09:34:17 -0400 From: =?utf-8?Q?Ludovic_Court=C3=A8s?= To: Denis 'GNUtoo' Carikli Subject: Re: bug#40626: Poor performance on low-end ARMv7 devices References: <87a73exj5v.fsf@inria.fr> <20200703015725.11a6515c@primarylaptop.localdomain> X-URL: http://www.fdn.fr/~lcourtes/ X-Revolutionary-Date: 26 Messidor an 228 de la =?utf-8?Q?R=C3=A9volution?= X-PGP-Key-ID: 0x090B11993D9AEBB5 X-PGP-Key: http://www.fdn.fr/~lcourtes/ludovic.asc X-PGP-Fingerprint: 3CE4 6455 8A84 FDC6 9DB4 0CFB 090B 1199 3D9A EBB5 X-OS: x86_64-pc-linux-gnu Date: Mon, 13 Jul 2020 15:34:15 +0200 In-Reply-To: <20200703015725.11a6515c@primarylaptop.localdomain> (Denis Carikli's message of "Fri, 3 Jul 2020 01:57:25 +0200") Message-ID: <87ft9vr12g.fsf@gnu.org> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/26.3 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-Spam-Score: -2.3 (--) X-Debbugs-Envelope-To: 40626 Cc: 40626@debbugs.gnu.org X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: debbugs-submit-bounces@debbugs.gnu.org Sender: "Debbugs-submit" X-Spam-Score: -3.3 (---) Hi, Denis 'GNUtoo' Carikli skribis: > Many ARM Single Board Computers are commonly used with microSD for > storage, and some microSD cards are extremely slow (and sometimes > unreliable when they are old). > > Could the performance issues be related to storage device I/Os? It could be the reason; it=E2=80=99s definitely the case on the board I was using here. Still I wonder what could be done on our side to improve on this. Ludo=E2=80=99. From debbugs-submit-bounces@debbugs.gnu.org Tue Jul 14 00:49:08 2020 Received: (at 40626) by debbugs.gnu.org; 14 Jul 2020 04:49:08 +0000 Received: from localhost ([127.0.0.1]:49930 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jvCsO-0006mL-5c for submit@debbugs.gnu.org; Tue, 14 Jul 2020 00:49:08 -0400 Received: from cyberdimension.org ([80.67.179.20]:42120 helo=gnutoo.cyberdimension.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jvCsL-0006mC-Qz for 40626@debbugs.gnu.org; Tue, 14 Jul 2020 00:49:06 -0400 Received: from gnutoo.cyberdimension.org (localhost [127.0.0.1]) by cyberdimension.org (OpenSMTPD) with ESMTP id c0da11dd; Tue, 14 Jul 2020 04:47:14 +0000 (UTC) Received: from primarylaptop.localdomain (localhost.localdomain [::1]) by gnutoo.cyberdimension.org (OpenSMTPD) with ESMTP id 698a7363; Tue, 14 Jul 2020 04:47:14 +0000 (UTC) Date: Tue, 14 Jul 2020 06:39:59 +0200 From: Denis 'GNUtoo' Carikli To: Ludovic =?UTF-8?B?Q291cnTDqHM=?= Subject: Re: bug#40626: Poor performance on low-end ARMv7 devices Message-ID: <20200714063959.19196a53@primarylaptop.localdomain> In-Reply-To: <87ft9vr12g.fsf@gnu.org> References: <87a73exj5v.fsf@inria.fr> <20200703015725.11a6515c@primarylaptop.localdomain> <87ft9vr12g.fsf@gnu.org> X-Mailer: Claws Mail 3.17.5 (GTK+ 2.24.32; i686-pc-linux-gnu) MIME-Version: 1.0 Content-Type: multipart/signed; boundary="Sig_/zD7benGhrppGhhjBGIlqW3s"; protocol="application/pgp-signature"; micalg=pgp-sha256 X-Spam-Score: -0.0 (/) X-Debbugs-Envelope-To: 40626 Cc: 40626@debbugs.gnu.org X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: debbugs-submit-bounces@debbugs.gnu.org Sender: "Debbugs-submit" X-Spam-Score: -1.0 (-) --Sig_/zD7benGhrppGhhjBGIlqW3s Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On Mon, 13 Jul 2020 15:34:15 +0200 Ludovic Court=C3=A8s wrote: > Hi, >=20 > Denis 'GNUtoo' Carikli skribis: >=20 > > Many ARM Single Board Computers are commonly used with microSD for > > storage, and some microSD cards are extremely slow (and sometimes > > unreliable when they are old). > > > > Could the performance issues be related to storage device I/Os? >=20 > It could be the reason; it=E2=80=99s definitely the case on the board I w= as > using here. >=20 > Still I wonder what could be done on our side to improve on this. I guessed that was the intent. A way to deal with that could be to validate if it's actually the case, for instance by installing an SSD, and doing performance comparison with the time command. Then if it helps a lot, we probably need to trace the filesystem access and optimize it somehow. Maybe that can be done with BPF or gprof, but I never looked into obtimizing I/O performances, so I'm unsure if that's the right approach. Guile may have profiling tools too. As for benchmarking the CPU, I've patched phoronix-test-suite in a very quick and dirty way in Parabola, so we could at least have some benchmarks. For now we only have "compilation" benchmarks with source code that should be FSDG compliant. With it I've found that the compilation performances can vary a lot between different ARM and x86 boards: - An I.MX6 Quad with 2G of RAM is about 4.5x faster than the AM335x of a Beaglebone Green with 512M of RAM. - A Lime2 A20 is about 1.7x faster than the Beaglebone Green and 2.5x slower than the board with the I.MX6 Quad. - My smartphone (Galaxy SIII GT-I9300) is about 2x faster than my server (PC Engines APU1) for compilation. Denis. --Sig_/zD7benGhrppGhhjBGIlqW3s Content-Type: application/pgp-signature Content-Description: OpenPGP digital signature -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEEeC+d2+Nrp/PU3kkGX138wUF34mMFAl8NNyAACgkQX138wUF3 4mPm7hAAlyGuNNBHykJhMYi8Zs4PUV4lNWRrvMfc0aO4JufCTJJgqfMooZppiFoe U5bbMB5I4bkdp5DfxUdkTQU+mq2zy4QoX3OhjyBUrgQJUfIkIsdMNOih96rjnI+J St3tLK7rnTLzNCOYCOUpXCFQD9lESkTZwgkF/nsuoCU3L1YbQpub/JZHANef9FFA iR9/UzpVIvLCr06U8vDKVt/RV8Xw5ZBlxHtAQhWENOc/o4ljTGCLbhsGhx0Ee9Az c27U0YUUcg/8X67n1u2Wr/idSmOHzqgnUuO9fRQlwRAI02o4lqe10ALrScWmD0U5 lBPUMCseL//HQWCM+raUeh/Ya/fRiBhm6WkxNv8xeACS5w+82zPR9PUc5OzJ8LOz 0hogEPmvqRpyfqHB+et5CiL+Y16gyof42wnNVIeGsbXrMMUTuPofBqI4ric6EUKL Uj/H+PrthFJF+v2qpuOGHFv0Sb2OkFMdJfNj0ZfTVdM4olmGRXPBzV+F9f+cxVOC ReTmE1WbfNXOpekrSQ+ez/v9JJfVBvd3G6CGmRbJJMRn9qHfsgQndzN/hb8wyE/t 4K+32vItRvqC7/I/pRzP4GrVtOhRJCczoFNG04eyCVpW4VtymLPTPpxjmbY5Rmse CV/9ClF1JdhPdCJODtx1/O7797uZxFDMi4pVjU+HpUncKbYbTD0= =vYYm -----END PGP SIGNATURE----- --Sig_/zD7benGhrppGhhjBGIlqW3s-- From debbugs-submit-bounces@debbugs.gnu.org Tue Jul 14 09:05:52 2020 Received: (at 40626) by debbugs.gnu.org; 14 Jul 2020 13:05:52 +0000 Received: from localhost ([127.0.0.1]:50409 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jvKd6-0006GT-FX for submit@debbugs.gnu.org; Tue, 14 Jul 2020 09:05:52 -0400 Received: from eggs.gnu.org ([209.51.188.92]:36356) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1jvKd3-0006G8-1i for 40626@debbugs.gnu.org; Tue, 14 Jul 2020 09:05:51 -0400 Received: from fencepost.gnu.org ([2001:470:142:3::e]:58568) by eggs.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1jvKcx-0004kW-FN; Tue, 14 Jul 2020 09:05:43 -0400 Received: from [2a01:e0a:1d:7270:af76:b9b:ca24:c465] (port=37176 helo=ribbon) by fencepost.gnu.org with esmtpsa (TLS1.2:RSA_AES_256_CBC_SHA1:256) (Exim 4.82) (envelope-from ) id 1jvKcw-0008So-O6; Tue, 14 Jul 2020 09:05:43 -0400 From: =?utf-8?Q?Ludovic_Court=C3=A8s?= To: Denis 'GNUtoo' Carikli Subject: Re: bug#40626: Poor performance on low-end ARMv7 devices References: <87a73exj5v.fsf@inria.fr> <20200703015725.11a6515c@primarylaptop.localdomain> <87ft9vr12g.fsf@gnu.org> <20200714063959.19196a53@primarylaptop.localdomain> Date: Tue, 14 Jul 2020 15:05:41 +0200 In-Reply-To: <20200714063959.19196a53@primarylaptop.localdomain> (Denis Carikli's message of "Tue, 14 Jul 2020 06:39:59 +0200") Message-ID: <87ft9umel6.fsf@gnu.org> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/26.3 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-Spam-Score: -2.3 (--) X-Debbugs-Envelope-To: 40626 Cc: 40626@debbugs.gnu.org X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: debbugs-submit-bounces@debbugs.gnu.org Sender: "Debbugs-submit" X-Spam-Score: -3.3 (---) Hi, Denis 'GNUtoo' Carikli skribis: > A way to deal with that could be to validate if it's actually the case, > for instance by installing an SSD, and doing performance comparison > with the time command. > > Then if it helps a lot, we probably need to trace the filesystem access > and optimize it somehow. Maybe that can be done with BPF or gprof, but > I never looked into obtimizing I/O performances, so I'm unsure if > that's the right approach. Guile may have profiling tools too. Yeah, the profiles at suggest a couple of things to investigate in Guile. > As for benchmarking the CPU, I've patched phoronix-test-suite in a very > quick and dirty way in Parabola, so we could at least have some > benchmarks. For now we only have "compilation" benchmarks with source > code that should be FSDG compliant. > > With it I've found that the compilation performances can vary a lot > between different ARM and x86 boards: > - An I.MX6 Quad with 2G of RAM is about 4.5x faster than the AM335x of a > Beaglebone Green with 512M of RAM. > - A Lime2 A20 is about 1.7x faster than the Beaglebone Green and 2.5x > slower than the board with the I.MX6 Quad. > - My smartphone (Galaxy SIII GT-I9300) is about 2x faster than my server > (PC Engines APU1) for compilation. Interesting, thanks! Ludo=E2=80=99.