From owner-freebsd-current@FreeBSD.ORG Mon Dec 22 20:58:14 2008 Return-Path: Delivered-To: freebsd-current@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 7BAC41065670; Mon, 22 Dec 2008 20:58:14 +0000 (UTC) (envelope-from sobomax@FreeBSD.org) Received: from sippysoft.com (gk1.360sip.com [72.236.70.240]) by mx1.freebsd.org (Postfix) with ESMTP id 0E2988FC08; Mon, 22 Dec 2008 20:58:13 +0000 (UTC) (envelope-from sobomax@FreeBSD.org) Received: from [192.168.1.38] (S0106001372fd1e07.vs.shawcable.net [70.71.171.106]) (authenticated bits=0) by sippysoft.com (8.13.8/8.13.8) with ESMTP id mBMKvsa1063552 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Mon, 22 Dec 2008 12:57:55 -0800 (PST) (envelope-from sobomax@FreeBSD.org) Message-ID: <494FFF42.7090102@FreeBSD.org> Date: Mon, 22 Dec 2008 12:57:38 -0800 From: Maxim Sobolev Organization: Sippy Software, Inc. User-Agent: Thunderbird 2.0.0.18 (Windows/20081105) MIME-Version: 1.0 To: Norikatsu Shigemura References: <20081221012752.cdc5cbfc.nork@FreeBSD.org> <20081221211949.GS1176@hoeg.nl> <20081222091203.GA28920@freebsd.org> <494F740E.3040502@FreeBSD.org> <20081223002901.9b71e60d.nork@FreeBSD.org> In-Reply-To: <20081223002901.9b71e60d.nork@FreeBSD.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Ed Schouten , Roman Divacky , FreeBSD Current Subject: Re: Change select(2) to kevent(2) on script(1)... X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 22 Dec 2008 20:58:14 -0000 Norikatsu Shigemura wrote: > Hi sobomax! > > On Mon, 22 Dec 2008 03:03:42 -0800 > Maxim Sobolev wrote: >> Roman Divacky wrote: >>> I believe other OSes should be able to do the porting effort when they >>> want to use FreeBSD software. >>> if kqueue-in-script makes any improvement go for it. >> I really doubt there would be any. It's just 2 descriptors, and if >> select(2) can't handle 2 descriptors efficiently them perhaps it is >> broken and has to be fixed instead. > > I think that performance improvement is significant(I don't > think performance improved by my patch). But my patch is the And can you explain where that "significant improvement" comes from? Are you saying that tty layer / pseudo-terminal driver is somehow much more efficient with kqeue(2) compared to select(2)/poll(2)? There is something broken about it if so. In any case without any numbers this discussion is pretty much pointless. -Maxim