From owner-cvs-all@FreeBSD.ORG Thu Jun 17 18:31:36 2004 Return-Path: Delivered-To: cvs-all@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4BC0016A4CE; Thu, 17 Jun 2004 18:31:36 +0000 (GMT) Received: from freefall.freebsd.org (freefall.freebsd.org [216.136.204.21]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2B4AF43D1D; Thu, 17 Jun 2004 18:31:36 +0000 (GMT) (envelope-from bmilekic@FreeBSD.org) Received: from freefall.freebsd.org (bmilekic@localhost [127.0.0.1]) i5HIVF3g042528; Thu, 17 Jun 2004 18:31:15 GMT (envelope-from bmilekic@freefall.freebsd.org) Received: (from bmilekic@localhost) by freefall.freebsd.org (8.12.11/8.12.11/Submit) id i5HIVFGE042527; Thu, 17 Jun 2004 18:31:15 GMT (envelope-from bmilekic) Date: Thu, 17 Jun 2004 18:31:15 +0000 From: Bosko Milekic To: Alfred Perlstein Message-ID: <20040617183115.GA41663@freefall.freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4.1i cc: Max Khon cc: src-committers@FreeBSD.org cc: cvs-all@FreeBSD.org cc: cvs-src@FreeBSD.org Subject: Re: cvs commit: src/sys/sys mbuf.h src/sys/kern uipc_mbuf.c uipc_syscalls.c src/usr.bin/netstat mbuf.c src/lib/libc/sys sendfile.2 X-BeenThere: cvs-all@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: CVS commit messages for the entire tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 17 Jun 2004 18:31:36 -0000 This is not fair. This change ADDs lines to netstat -m's output, but does not/should not modify existing ones. Furthermore, the person who writes scripts to parse netstat(1)'s output should stop doing that and read the sysctls directly, doing similar to what netstat(1) itself does, instead of relying on its output, which is subject to change. Anyway, if you were going to argue against this MFC, you would have been better off pointing out that it merges to RELENG_4 something that shouldn't even be in HEAD. Namely, the sendfile(2) stats have no business in being in the mbstat struct and should have been moved out of there before this was MFC'd (you should consider fixing that, by the way, since you put them there). :-) Despite all this, since the change has already been made and it is clearly non-fatal, there is no reason for it to be backed out now, just please take caution when/if doing this sort of thing again next time. A better approach would be to: 1) Fix HEAD, move the sendfile(2) stats OUT of mbstat, and introduce new flag to netstat(1) to report sendfile(2) stats, as you suggest. 2) MFC (1), instead of backing this out first, because this change should not be a problem to anyone with enough clue to not write flaky scripts. Regards, Bosko Alfred Perlstein wrote: > >MFC'ing a change to netstat -m's output is not acceptable. > >This will break scripts. > >Please back out the netstat change. > >Why are sfbuf/sendfile stats being put under mbuf stats? > >This is bogus, please use a different flag instead of cluttering >the mbuf stats. > > >* Max Khon [040616 17:08] wrote: >> fjoe 2004-06-17 00:08:23 UTC >> >> FreeBSD src repository >> >> Modified files: (Branch: RELENG_4) >> sys/sys mbuf.h >> sys/kern uipc_mbuf.c uipc_syscalls.c >> usr.bin/netstat mbuf.c >> lib/libc/sys sendfile.2 >> Log: >> MFC: >> >> sendfile(2)-related statistics that can be viewed through netstat -m: >> - The number of times sendfile() had to do disk I/O >> - The number of times sfbuf allocation failed >> - The number of times sfbuf allocation had to wait >> and three sysctl's: kern.ipc.nsfbufs, kern.ipc.nsfbufspeak, >> kern.ipc.nsfbufsused that also can be viewed through netstat -m. >> >> Submitted by: Igor Sysoev >> >> Revision Changes Path >> 1.6.2.9 +50 -1 src/lib/libc/sys/sendfile.2 >> 1.51.2.37 +8 -0 src/sys/kern/uipc_mbuf.c >> 1.65.2.20 +6 -0 src/sys/kern/uipc_syscalls.c >> 1.44.2.26 +6 -1 src/sys/sys/mbuf.h >> 1.17.2.4 +16 -2 src/usr.bin/netstat/mbuf.c