From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Subject: Re: [PATCH 04/11] fs: add support for allowing applications to pass in write life time hints To: Christoph Hellwig Cc: "Darrick J. Wong" , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, adilger@dilger.ca, martin.petersen@oracle.com References: <1497498312-17704-1-git-send-email-axboe@kernel.dk> <1497498312-17704-5-git-send-email-axboe@kernel.dk> <20170615041503.GB4521@birch.djwong.org> <20170615081914.GD10599@infradead.org> <7f8e2e3c-fa15-6864-7afd-b82e5afb87b5@kernel.dk> <20170616073031.GB8682@infradead.org> From: Jens Axboe Message-ID: Date: Fri, 16 Jun 2017 08:35:49 -0600 MIME-Version: 1.0 In-Reply-To: <20170616073031.GB8682@infradead.org> Content-Type: text/plain; charset=utf-8 List-ID: On 06/16/2017 01:30 AM, Christoph Hellwig wrote: > On Thu, Jun 15, 2017 at 09:23:02AM -0600, Jens Axboe wrote: >> And then I remembered why fadvise _sucks_. It returns the error value >> directly. So 0 is success > 0 is some error. That does not work well >> for adding a set/get interface. Additionally, with fadvise, we have >> to overload either 'offset' or 'length' for the write hint for the >> set operation. Not super pretty. >> >> Any objections to making the auxiliary interface fcntl(2) based? >> That would be a cleaner fit, imho. > > fcntl sounds fine to me. v5 implemented that, much better fit than fadvise. -- Jens Axboe