From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from bombadil.infradead.org ([65.50.211.133]:33110 "EHLO bombadil.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751646AbdJSH2F (ORCPT ); Thu, 19 Oct 2017 03:28:05 -0400 Date: Thu, 19 Oct 2017 00:28:04 -0700 From: Christoph Hellwig To: Jaegeuk Kim Cc: linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, linux-f2fs-devel@lists.sourceforge.net Subject: Re: [PATCH] f2fs: limit # of inmemory pages Message-ID: <20171019072804.GA16321@infradead.org> References: <20171019021516.65627-1-jaegeuk@kernel.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20171019021516.65627-1-jaegeuk@kernel.org> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Wed, Oct 18, 2017 at 07:15:16PM -0700, Jaegeuk Kim wrote: > If some abnormal users try lots of atomic write operations, f2fs is able to > produce pinned pages in the main memory which affects system performance. > This patch limits that as 20% over total memory size, and if f2fs reaches > to the limit, it will drop all the inmemory pages. Can you please document your "atomic" writes and how they pin memory in a mail to linux-abi? And how they compare to my O_ATOMIC implement as well as the papers it is based on? I tried to take a quick look at the code and it looks very bogus to me in addition to being sneaked in through fs-specific ioctls without any review whatsoever.