From mboxrd@z Thu Jan 1 00:00:00 1970 From: Matthew Wilcox Subject: Re: [PATCH 0/4] eliminate SECTOR related magic numbers and duplicated conversions Date: Tue, 5 May 2020 11:01:13 -0700 Message-ID: <20200505180113.GJ16070@bombadil.infradead.org> References: <20200505115543.1660-1-thunder.leizhen@huawei.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-block-owner@vger.kernel.org To: antlists Cc: Zhen Lei , Minchan Kim , Nitin Gupta , Sergey Senozhatsky , Jens Axboe , linux-block , Andrew Morton , linux-mm , Alasdair Kergon , Mike Snitzer , dm-devel , Song Liu , linux-raid , linux-kernel List-Id: linux-raid.ids On Tue, May 05, 2020 at 06:32:36PM +0100, antlists wrote: > On 05/05/2020 12:55, Zhen Lei wrote: > > When I studied the code of mm/swap, I found "1 << (PAGE_SHIFT - 9)" appears > > many times. So I try to clean up it. > > > > 1. Replace "1 << (PAGE_SHIFT - 9)" or similar with SECTORS_PER_PAGE > > 2. Replace "PAGE_SHIFT - 9" with SECTORS_PER_PAGE_SHIFT > > 3. Replace "9" with SECTOR_SHIFT > > 4. Replace "512" with SECTOR_SIZE > > Naive question - what is happening about 4096-byte sectors? Do we need to > forward-plan? They're fully supported already, but Linux defines a sector to be 512 bytes. So we multiply by 8 and divide by 8 a few times unnecessarily, but it's not worth making sector size be a per-device property. Good thought, though.