From owner-freebsd-net Tue Oct 23 12: 0:36 2001 Delivered-To: freebsd-net@freebsd.org Received: from elvis.mu.org (elvis.mu.org [216.33.66.196]) by hub.freebsd.org (Postfix) with ESMTP id 3C9A737B403 for ; Tue, 23 Oct 2001 12:00:34 -0700 (PDT) Received: by elvis.mu.org (Postfix, from userid 1192) id 4511781D01; Tue, 23 Oct 2001 14:00:34 -0500 (CDT) Date: Tue, 23 Oct 2001 14:00:34 -0500 From: Alfred Perlstein To: Luigi Rizzo Cc: net@FreeBSD.ORG Subject: Re: performance issues with M_PREPEND on clusters Message-ID: <20011023140034.M15052@elvis.mu.org> References: <20011023110307.A34494@iguana.aciri.org> <20011023132813.I15052@elvis.mu.org> <20011023114650.C34494@iguana.aciri.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20011023114650.C34494@iguana.aciri.org>; from rizzo@aciri.org on Tue, Oct 23, 2001 at 11:46:50AM -0700 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org * Luigi Rizzo [011023 13:50] wrote: > On Tue, Oct 23, 2001 at 01:28:13PM -0500, Alfred Perlstein wrote: > ... > > > In the mbuf code, M_LEADINGSPACE always returns 0 when M_EXT is > > > set, instead of calling the second part of M_WRITABLE to check > > > whether there is a chance of writing into the cluster. This means > ... > > I've seen this brought up before, this is not what you want to > > do otherwise you risk corrupting EXT data. The right thing to > > do is have the people allocating writable EXT bufs mark them > > as such. Other EXT type buffers such as sendfile bufs can not > > be written to no matter what the sharecount is. > > Actually, the BSD code works the other way around: > > * standard EXT bufs as returned by MCLGET _are_ writable if > m_ext.ext_free == NULL and their refcnt is 1 (they _must_ be, > otherwise we would have errors when m_free() decides to > dispose of a cluster, which is a form of writing), > > and > > * people allocating "other EXT type" implicitly mark the buffer > as not-writable by setting m_ext.ext_free to the proper routine. > This is certainly what the sendfile bufs that you mentioned do. > > M_WRITABLE (used in the patch i proposed) does exactly the checks > that you do when you m_free() a cluster. If this is wrong, then > m_free() is broken! > > Am i missing something else ? Yes, you're right, I was mistaken in my paranioa, however you're missing the fact that one may want to allocate an EXT buf and still have it writeable. I like your patch, it's a good idea, go for it. -Alfred To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message