Return-Path: Delivered-To: new-httpd-archive@hyperreal.org Received: (qmail 12034 invoked by uid 6000); 6 Mar 1998 21:53:15 -0000 Received: (qmail 12010 invoked from network); 6 Mar 1998 21:53:10 -0000 Received: from twinlark.arctic.org (204.62.130.91) by taz.hyperreal.org with SMTP; 6 Mar 1998 21:53:10 -0000 Received: (qmail 17382 invoked by uid 500); 6 Mar 1998 21:53:34 -0000 Date: Fri, 6 Mar 1998 13:53:33 -0800 (PST) From: Dean Gaudet To: new-httpd@apache.org Subject: Re: mmap() versus read() (fwd) In-Reply-To: <350065C0.9F00627C@prairie.nodak.edu> Message-ID: X-Comment: Visit http://www.arctic.org/~dgaudet/legal for information regarding copyright and disclaimer. MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: new-httpd-owner@apache.org Precedence: bulk Reply-To: new-httpd@apache.org On Fri, 6 Mar 1998, Igor Tatarinov wrote: > How come then that "The Linux Kernel" by David Rusling > (http://sunsite.unc.edu/LDP/LDP/tlk/tlk.html, chapter 4.7) > says that Linux always prefetches the following page on a page fault? > (as does Solaris) I wouldn't be surprised if the book is wrong. > If it does prefetch another page then it reads 8K at once. Apache's > IOBUFSIZE is also 8K so there should be no difference !? > Unless, the prefetch is issued as a separate I/O ? The difference is that read() readahead is like 16k or 32k. Dean