From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758735Ab0EYPVZ (ORCPT ); Tue, 25 May 2010 11:21:25 -0400 Received: from mail-px0-f174.google.com ([209.85.212.174]:35178 "EHLO mail-px0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757090Ab0EYPVX (ORCPT ); Tue, 25 May 2010 11:21:23 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=date:from:to:cc:subject:message-id:references:mime-version :content-type:content-disposition:in-reply-to:user-agent; b=rElvQZkdScdPvBFEfJK0XzOB6WgMKVfUTls8gpBhisYs34Tvl246V8JSooP5b38VX1 aVbGZ0tNFaVNmTArwTj1+iJ93g6oXxXLfsx1Gi8RBaNYcPMHVFKCk3y+chyQQxGnWZHi gAEYdZZhr5xg871SBhB20Fz89xUjcjKbU8KSw= Date: Wed, 26 May 2010 00:21:15 +0900 From: Minchan Kim To: Nitin Gupta Cc: Greg KH , Pekka Enberg , Andrew Morton , Hugh Dickins , Cyp , driverdev , linux-kernel Subject: Re: [PATCH 0/3] zram: generic RAM based compressed R/W block devices Message-ID: <20100525152115.GB3227@barrios-desktop> References: <1274710685-2351-1-git-send-email-ngupta@vflare.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1274710685-2351-1-git-send-email-ngupta@vflare.org> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi, Nitin. On Mon, May 24, 2010 at 07:48:02PM +0530, Nitin Gupta wrote: > Creates RAM based block devices: /dev/zramX (X = 0, 1, ...). > Pages written to these disks are compressed and stored in memory > itself. These disks allow very fast I/O and compression provides > good amounts of memory savings. > > This is enhancement over existing ramzswap driver which creates > virtual block devices (/dev/ramzswapX) which could be used only > as swap disks. > > Now, with the ability to handle any kind of I/O request, zram > devices have lot more use cases: > - /tmp storage > - various caches under /var > - swap disks > - maybe even more! :) Wow, it looks good to me. As a matter of fact, when I saw your descriptoin, I thought we can merge your feature into brd. But I changed my mind due to following as First of all, ramfs supports direct I/O. So it's hard to support compression. If we can select optionable config, it would make brd code complicated or ugly. That's because ramzswap uses own allocator, compression, extended ioctl and so on. And some users might regard write with important than read without considering space. So it would be better to keep other block device for easy use and clear code. I will review this code when I have a time. Thanks, Nitin. -- Kind regards, Minchan Kim