From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758936AbYDRRsp (ORCPT ); Fri, 18 Apr 2008 13:48:45 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1758187AbYDRRsf (ORCPT ); Fri, 18 Apr 2008 13:48:35 -0400 Received: from rv-out-0708.google.com ([209.85.198.248]:62450 "EHLO rv-out-0506.google.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1758453AbYDRRse (ORCPT ); Fri, 18 Apr 2008 13:48:34 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:sender:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references:x-google-sender-auth; b=oNQZyaUBnk/E9l+YivDaAKmJx3uhz0J9qj0hUJeZRZZBEPb+pAXk9yfvM6bs3/lPDQpiWcRg/dA7lVmzCS/E/gaatiJwSCkhJoDNUcJldKCJKeL3iJ9zm2br8XA9eqrVLiZ6a+hGzu7H2MoCafIS6oopumiTiIZNwUeV146FLhM= Message-ID: Date: Fri, 18 Apr 2008 13:48:31 -0400 From: "Bob Copeland" To: "Szabolcs Szakacsits" Subject: Re: [PATCH 0/7] OMFS filesystem version 3 Cc: "Miklos Szeredi" , hch@infradead.org, akpm@linux-foundation.org, linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <1208041121-26787-1-git-send-email-me@bobcopeland.com> <20080412205544.5e12a7d4.akpm@linux-foundation.org> <20080413080130.GA9622@infradead.org> <20080413012001.8d7967f4.akpm@linux-foundation.org> <20080413082815.GA20108@infradead.org> <20080414004521.GA30489@hash.localnet> X-Google-Sender-Auth: 3df351095998355c Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Apr 18, 2008 at 1:35 PM, Szabolcs Szakacsits wrote: > The major reason seems to be that the FUSE version reads heavily from the > block device during pure write operations, while the kernel driver never. > There can be several non-exclusive explanations. Yep, I suspect switching to the lower-level API plus switching to mmap() and getting rid of various memcpy's in omfs.c would make a huge difference. And also drop the line count without the hash table stuff. > Moreover when you're writing to a block device from user space then the > size and position of the block should be page aligned, otherwise you end up > doing unwanted synchronous reads instead of the believed asynchronous > writes. Cool, thanks for the tips! -- Bob Copeland %% www.bobcopeland.com