From owner-soc-status@FreeBSD.ORG Sat Jul 9 09:33:10 2011 Return-Path: Delivered-To: soc-status@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 1B2881065672 for ; Sat, 9 Jul 2011 09:33:10 +0000 (UTC) (envelope-from iputsikau@gmail.com) Received: from mail-qw0-f54.google.com (mail-qw0-f54.google.com [209.85.216.54]) by mx1.freebsd.org (Postfix) with ESMTP id D5BDA8FC08 for ; Sat, 9 Jul 2011 09:33:09 +0000 (UTC) Received: by qwc9 with SMTP id 9so1688143qwc.13 for ; Sat, 09 Jul 2011 02:33:09 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:cc:content-type; bh=apqvxIr2IPg4zlzuoJQzvbk3k787VqaOWanRxUpNYFU=; b=t1vY7Uflp+9a1TPX2OXmKQ0PJ+NT/o2cZvsHhLl6ClwzyxCgVn3EjzHkADPCMBC4mT 31GAsl/WSsMFa0PtN3aomkUD8xnCZCXn97j7ryBSx2eAYnbZLc+JUwrzaZ41wrXCmoCo RvoPLmyJk1rTaI1z+Q+GzfquV/ncSt79MpYEM= MIME-Version: 1.0 Received: by 10.229.18.77 with SMTP id v13mr2244945qca.56.1310203988874; Sat, 09 Jul 2011 02:33:08 -0700 (PDT) Received: by 10.229.32.132 with HTTP; Sat, 9 Jul 2011 02:33:08 -0700 (PDT) Date: Sat, 9 Jul 2011 12:33:08 +0300 Message-ID: From: Ilya Putsikau To: soc-status@freebsd.org Content-Type: text/plain; charset=UTF-8 Cc: Ivan Voras , Hans Petter Selasky Subject: [fuse] Status report #7 X-BeenThere: soc-status@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Summer of Code Status Reports and Discussion List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 09 Jul 2011 09:33:10 -0000 Project: Finish porting fuse Fuse can now pass fsx tests and no longer freezes with blogbench. Actually blogbench didn't freeze before, it performed a lot of reads at very big offsets. After fixing io in fuse I no longer see such behavior in blogbench. Known issues: 1. It may panic if application killed while operations in progress. It could panic before but recent changes made error easier to trigger 2. Expect much lower io performance. Additional setattr or getattr operation is now called for almost every write, io buffer size was decreased to page size, data cache is invalidated when last file closed. I'll try to improve it next week.