From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from host-212-158-219-180.bulldogdsl.com ([212.158.219.180] helo=aeryn.fluff.org.uk) by canuck.infradead.org with esmtp (Exim 4.43 #1 (Red Hat Linux)) id 1D39uT-0006yb-9u for linux-mtd@lists.infradead.org; Mon, 21 Feb 2005 04:31:22 -0500 Date: Mon, 21 Feb 2005 09:31:00 +0000 From: Ben Dooks To: Steven Scholz Message-ID: <20050221093100.GF28744@home.fluff.org> References: <421998DA.1060503@imc-berlin.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <421998DA.1060503@imc-berlin.de> Cc: MTD List Subject: Re: ioremap_cached in drivers/mtd/maps ? List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Mon, Feb 21, 2005 at 09:16:26AM +0100, Steven Scholz wrote: > Hi there, > > it looks like only lubbock-flash.c is using ioremap_cached(). > > Can I use ioremap_cached() with normal NOR flashes? > Is it only for DMA transfer from/to flash? you could use it for such, but it also allows an cached area for flash read, which makes it faster to execute code from. > Does it improve something when using JFFS2? > What do I have to consider? .inval_cache function? It allows the mapping driver to have a seperate read-only cached area, the .inval_cache function is needed to flush this area when writes happen to the un-cached area. -- Ben (ben@fluff.org, http://www.fluff.org/) 'a smiley only costs 4 bytes'