Newsgroups: php.internals Path: news.php.net Xref: news.php.net php.internals:74280 Return-Path: Mailing-List: contact internals-help@lists.php.net; run by ezmlm Delivered-To: mailing list internals@lists.php.net Received: (qmail 31643 invoked from network); 17 May 2014 15:37:54 -0000 Received: from unknown (HELO lists.php.net) (127.0.0.1) by localhost with SMTP; 17 May 2014 15:37:54 -0000 Authentication-Results: pb1.pair.com smtp.mail=zeev@zend.com; spf=permerror; sender-id=unknown Authentication-Results: pb1.pair.com header.from=zeev@zend.com; sender-id=unknown Received-SPF: error (pb1.pair.com: domain zend.com from 209.85.220.181 cause and error) X-PHP-List-Original-Sender: zeev@zend.com X-Host-Fingerprint: 209.85.220.181 mail-vc0-f181.google.com Received: from [209.85.220.181] ([209.85.220.181:57690] helo=mail-vc0-f181.google.com) by pb1.pair.com (ecelerity 2.1.1.9-wez r(12769M)) with ESMTP id 7B/10-28602-15287735 for ; Sat, 17 May 2014 11:37:54 -0400 Received: by mail-vc0-f181.google.com with SMTP id ld13so7571622vcb.40 for ; Sat, 17 May 2014 08:37:51 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:references:from:mime-version:in-reply-to:date :message-id:subject:to:cc:content-type; bh=imBWBz0dnUWV6BzmN8kjAQBxcvI5hHxOX/vSm7il4Ag=; b=CcEi0JG8HBnOj6c+tUhnTtPK0K5YwuAy3+bPE2VSNPbCFuZagfkNdXbDgCVMeo7RhY +7SwEU1j1QqP313yrJUS+p7wbWWamjFFAMjw+xMNu7FNxyAvx1eHgbAJ9UCP0xKjffye Lh6KMd86hR8ytxaKHFDMr2+KcW0ljQ9pw9thahMX29ZlHFuNTqDxW1yR4FJOzvnIX0g4 Niywc+339ovbxIWg0RcPUH3veG5h37MzBpwf1IYALeEMx0GdP+CTxAq59xmaA3hFyBmX r4mRIm87IoxDqKnchijQb17qda0VbSk+LJ/70V8D8/wKSSj17ovOb63srJRGL/R2BKVZ yzag== X-Gm-Message-State: ALoCoQmchYO6cxKKvQSIfr7gnOG6zXy6cT7bvtoTKTOX341hU8Gmz+F0WQggrvFuR4i2LS+u1adnv7nGkvUyI2OhVhh7LsKv19v1pJZdoB1/IdamS8YdOCOZ0KGKqyAAByb3EvcfTvsh X-Received: by 10.58.126.135 with SMTP id my7mr7467552veb.27.1400341071181; Sat, 17 May 2014 08:37:51 -0700 (PDT) References: <3939936079205827190@unknownmsgid> <2422EE67-5206-4A27-9175-3C738A9128C9@ajf.me> Mime-Version: 1.0 (1.0) In-Reply-To: <2422EE67-5206-4A27-9175-3C738A9128C9@ajf.me> Date: Sat, 17 May 2014 18:37:47 +0300 Message-ID: <4081130594864738388@unknownmsgid> To: Andrea Faulds Cc: PHP internals Content-Type: multipart/alternative; boundary=047d7b675e521cd01d04f99a5016 Subject: Re: [PHP-DEV] A call for help (urgent) From: zeev@zend.com (Zeev Suraski) --047d7b675e521cd01d04f99a5016 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable On 17 =D7=91=D7=9E=D7=90=D7=99 2014, at 17:32, Andrea Faulds w= rote: Everyone, please vote yes. The No side is unlikely to lose, due to the 2/3 majority required. The Yes side does need a lot of help to win, though. Remember, every no vote requires 2 yes votes to counteract it. That has nothing to do with the fact applying this patch is morally wrong, and also may or may not be true. It negates months of hard work performed behind closed doors in secret (phpng), compared to lots of hard work done in the open (64-bit patch). Unlike phpng, this is a tactical patch that while changes a lot of code, is mostly about "dirty work" than research. Have you used PHP 3? Then maybe 4? Surely 5? All were developed in exactly the same way. First a working PoC was established, after lots of trial and error and hard work it was shared with everyone to join in. Spinning phpng's work as something evil is as ridiculous as it is wrong. Sabotaging it because of it is even worse. No way. Let=E2=80=99s use the right types for the job. =E2=80=98int=E2=80= =99 and =E2=80=98long=E2=80=99 should never be used for sizes of data structures. This is ridiculous. But I'm sure you know a lot better than Andi, Derick, Dmitry, Nikita, Rasmus, Stas, Xinchen and myself. I fail to understand who in his rift mins thinks it's both technically and morally legitimate to force this patch on this group. It=E2=80=99s not as if 64-bit will kill phpng, it=E2=80=99ll just delay it. You don't seem to realize what's going on here. It won't delay it, it'll greatly slow it down, and given phpng is all about performance - it negates its goals. Phpng's key performance gain had to do with compacting it's data structures. This goes in the opposite direction, 180 degrees. There's no way to fix it later as some suggest, it's an inherent incompatibility in directions. For the current PHP it yields an 8% memory increase. For phpng it'll be a lot more since it's data structures are more compact and therefore it'll be a lot more, relatively speaking. Zeev --047d7b675e521cd01d04f99a5016--