From mboxrd@z Thu Jan 1 00:00:00 1970 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <16898.46622.108835.631425@cargo.ozlabs.ibm.com> Date: Fri, 4 Feb 2005 10:39:10 +1100 From: Paul Mackerras Subject: Re: A scrub daemon (prezeroing) In-Reply-To: References: <1106828124.19262.45.camel@hades.cambridge.redhat.com> <20050202153256.GA19615@logos.cnet> <20050202163110.GB23132@logos.cnet> Sender: owner-linux-mm@kvack.org Return-Path: To: Rik van Riel Cc: Marcelo Tosatti , Christoph Lameter , David Woodhouse , linux-mm@kvack.org, linux-kernel@vger.kernel.org, akpm@osdl.org List-ID: Rik van Riel writes: > I'm not convinced. Zeroing a page takes 2000-4000 CPU > cycles, while faulting the page from RAM into cache takes > 200-400 CPU cycles per cache line, or 6000-12000 CPU > cycles. On my G5 it takes ~200 cycles to zero a whole page. In other words it takes about the same time to zero a page as to bring in a single cache line from memory. (PPC has an instruction to establish a whole cache line of zeroes in modified state without reading anything from memory.) Thus I can't see how prezeroing can ever be a win on ppc64. Regards, Paul. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: aart@kvack.org