Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753468AbYKQXW1 (ORCPT ); Mon, 17 Nov 2008 18:22:27 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751700AbYKQXWT (ORCPT ); Mon, 17 Nov 2008 18:22:19 -0500 Received: from e32.co.us.ibm.com ([32.97.110.150]:55270 "EHLO e32.co.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751630AbYKQXWS (ORCPT ); Mon, 17 Nov 2008 18:22:18 -0500 Date: Mon, 17 Nov 2008 18:22:04 -0500 From: Josh Boyer To: Benjamin Herrenschmidt Cc: Linus Torvalds , Thomas Gleixner , LKML , Steven Rostedt , linuxppc-dev@ozlabs.org, Paul Mackerras , Andrew Morton , Ingo Molnar Subject: Re: Large stack usage in fs code (especially for PPC64) Message-ID: <20081117182204.0940d3a8@zod.rchland.ibm.com> In-Reply-To: <1226963596.7178.254.camel@pasglop> References: <1226963596.7178.254.camel@pasglop> X-Mailer: Claws Mail 3.5.0 (GTK+ 2.12.8; i386-redhat-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1571 Lines: 38 On Tue, 18 Nov 2008 10:13:16 +1100 Benjamin Herrenschmidt wrote: > > > Well, it's not unacceptable on good CPU's with 4kB blocks (just an 8-entry > > array), but as you say: > > > > > On PPC64 I'm told that the page size is 64K, which makes the above equal > > > to: 64K / 512 = 128 multiply that by 8 byte words, we have 1024 bytes. > > > > Yeah. Not good. I think 64kB pages are insane. In fact, I think 32kB > > pages are insane, and 16kB pages are borderline. I've told people so. > > > > The ppc people run databases, and they don't care about sane people > > telling them the big pages suck. > > Hehe :-) > > Guess who is pushing for larger page sizes nowadays ? Embedded > people :-) In fact, we have patches submited on the list to offer the > option for ... 256K pages on some 44x embedded CPUs :-) For clarification, that workload is very precise. Namely embedded 44x CPUs used in RAID cards. I'm not entirely convinced bringing 256K pages into mainline is a good thing yet anyway. 64K pages, while seemingly insane for embedded boards that typically have less than 512 MiB of DRAM, help for a bit larger set of workloads. As a KVM host is the primary winner at the moment. But given the small number of TLB entries on these CPUs, it can pay off elsewhere as well. josh -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/