From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mars.matrox.com (mars.matrox.com [192.168.1.29]) by itchy.matrox.com (Postfix) with ESMTP id B9D4F19E13 for ; Wed, 14 Jul 2004 18:41:40 -0400 (EDT) Received: (from root@localhost) by mars.matrox.com (8.11.6/8.11.6) id i6EMh9128140 for ; Wed, 14 Jul 2004 18:43:09 -0400 (EDT) Received: from dyn-152-170.matrox.com (dyn-152-170.matrox.com [192.168.152.170]) by pluton.matrox.com (8.12.9/8.12.9) with ESMTP id i6EMh6hC027846 for ; Wed, 14 Jul 2004 18:43:06 -0400 (EDT) Subject: [Fwd: remap_page_range() vs nopage()] From: Michel Hubert Content-Type: text/plain Message-Id: <1089844986.15840.144.camel@blackcomb> Mime-Version: 1.0 Date: 14 Jul 2004 18:43:06 -0400 Content-Transfer-Encoding: 7bit Sender: owner-linux-mm@kvack.org Return-Path: To: linux-mm@kvack.org List-ID: Hello, I previously posted this question to kernelnewbies.org but without getting any answer. I hope it is not too basic for this mailing list... It's written in Linux Device Driver 2nd edition that remap_page_range (which maps an entire range at once) should be used for device IO whereas nopage (which maps a single page at a time) should be used for real physical memory. However, I noticed that mmap_mem() in drivers/char/mem.c uses exclusively remap_page_range. How could this work when dealing with non contiguous physical memory ? Thank you, Michel -- 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