From mboxrd@z Thu Jan 1 00:00:00 1970 From: Matthew Wilcox Subject: Re: [RFC] Protect larger order pages from breaking up Date: Mon, 12 Feb 2018 09:36:30 -0800 Message-ID: <20180212173630.GB9396@bombadil.infradead.org> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org To: Christopher Lameter Cc: Mel Gorman , linux-mm@kvack.org, akpm@linux-foundation.org, Leon Romanovsky , linux-rdma@vger.kernel.org, Thomas Schoebel-Theuer List-Id: linux-rdma@vger.kernel.org On Fri, Feb 09, 2018 at 01:24:41PM -0600, Christopher Lameter wrote: > Control over this feature is by writing to /proc/zoneinfo. > > F.e. to ensure that 2000 16K pages stay available for jumbo > frames do > > echo "3=2000" >/proc/zoneinfo That seems ... wrong. 4k is order 0, 8k is order 1, 16k is order 2, 32k is order 3. > One can then also f.e. operate the slub allocator with > 64k pages. Specify "slub_max_order=3 slub_min_order=3" on > the kernel command line and all slab allocator allocations > will occur in 16K page sizes. This example also reads weird ;-) -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org