From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Wed, 23 May 2007 16:14:20 -0500 From: Matt Mackall Subject: Re: [patch 1/3] slob: rework freelist handling Message-ID: <20070523211420.GJ11115@waste.org> References: <20070523061702.GA9449@wotan.suse.de> <20070523071200.GB9449@wotan.suse.de> <20070523183224.GD11115@waste.org> <20070523195824.GF11115@waste.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Sender: owner-linux-mm@kvack.org Return-Path: To: Christoph Lameter Cc: Nick Piggin , Andrew Morton , Linux Memory Management List List-ID: On Wed, May 23, 2007 at 01:16:06PM -0700, Christoph Lameter wrote: > Oh. And I forgot to check mm. > > See swap_prefetch.c:prefetch_suitable Doesn't look problematic to me. Or if it is, it'll die for any system that happens to use get_free_page a lot. > I do not think the updating of these counters is optional. Their use will > increase as we get more sophisticated in balancing the VM load. We cannot > have developers do an #ifdef CONFIG_SLOB around ZVC use. How does the VM deal with balancing users of get_free_page? We can probably add a couple lines to dummy up NR_SLAB_UNRECLAIMABLE, but I think the claim that SLOB is "broken" without it is completely overblown. -- Mathematics is the supreme nostalgia of our time. -- 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