From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 5930 invoked by alias); 24 Sep 2017 20:14:36 -0000 Mailing-List: contact zsh-workers-help@zsh.org; run by ezmlm Precedence: bulk X-No-Archive: yes List-Id: Zsh Workers List List-Post: List-Help: X-Seq: 41757 Received: (qmail 17374 invoked by uid 1010); 24 Sep 2017 20:14:36 -0000 X-Qmail-Scanner-Diagnostics: from aok120.rev.netart.pl by f.primenet.com.au (envelope-from , uid 7791) with qmail-scanner-2.11 (clamdscan: 0.99.2/21882. spamassassin: 3.4.1. Clear:RC:0(85.128.245.120):SA:0(-1.9/5.0):. Processed in 2.740672 secs); 24 Sep 2017 20:14:36 -0000 X-Spam-Checker-Version: SpamAssassin 3.4.1 (2015-04-28) on f.primenet.com.au X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=ham autolearn_force=no version=3.4.1 X-Envelope-From: psprint@zdharma.org X-Qmail-Scanner-Mime-Attachments: | X-Qmail-Scanner-Zip-Files: | X-Virus-Scanned: by amavisd-new using ClamAV (16) Date: Sun, 24 Sep 2017 22:04:39 +0200 From: Sebastian Gniazdowski To: zsh-workers@zsh.org Message-ID: Subject: Performance gains after the commited optimizations X-Mailer: Airmail (442) MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Hello I've found out great gains coming from multiple optimizations. I'm writin= g the Doxygen-like tool in Zsh, which uses (z) flag and much processing i= n loops. Processing a 1500-lines file, gains are: - from 14 seconds to 8 seconds, thanks to patch: 2017-03-04 8522e99 40781: optimize array assignment, similar to 39995 for= string assignment =E2=80=93 from 20 seconds to 14 seconds, thanks to patch: 2016-12-28 4fb4ce1 40231: Optimise setarrvalue(). =E2=80=93 from 25 seconds to 20 seconds, thanks to patch: 2016-11-14 99acd1e 39906: More multibyte optimisations for US-ASCII. =E2=80=93 from 30 seconds to 25 seconds, thanks to patch: 2016-11-04 9c68ef0 39825: optimise mb=5Fmetastrlenend() for 7-bit charact= er. one other patch that contributed 1 sec, but no time and energy to run bis= ect I was searching for real-word confirmation of each optimization and it wa= s hard to find =C2=A0them, only once zsh-syntax-highlighting gained from = 99acd1e and 9c68ef0. Thanks for commiting those patches, they've had say = =22off-by-1=22 complexity, the possibility to break special cases, so it = wasn't easy to read them. I can now parse large file in 8 seconds, while = 5.0.8 does it in 38 seconds (in 5.1 probably Bart's heap optimizations ma= tter). -- =20 Sebastian Gniazdowski psprint /at/ zdharma.org