From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Tue, 25 May 2004 06:53:22 +0200 From: Andrea Arcangeli Subject: Re: [PATCH] ppc64: Fix possible race with set_pte on a present PTE Message-ID: <20040525045322.GX29378@dualathlon.random> References: <1085369393.15315.28.camel@gaston> <1085371988.15281.38.camel@gaston> <1085373839.14969.42.camel@gaston> <20040525034326.GT29378@dualathlon.random> <20040525042054.GU29378@dualathlon.random> <16562.52948.981913.814783@napali.hpl.hp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <16562.52948.981913.814783@napali.hpl.hp.com> Sender: owner-linux-mm@kvack.org Return-Path: To: davidm@hpl.hp.com Cc: Linus Torvalds , Benjamin Herrenschmidt , Andrew Morton , Linux Kernel list , Ingo Molnar , Ben LaHaise , linux-mm@kvack.org, Architectures Group List-ID: On Mon, May 24, 2004 at 09:43:00PM -0700, David Mosberger wrote: > >>>>> On Tue, 25 May 2004 06:20:54 +0200, Andrea Arcangeli said: > > Andrea> the only architecture that has the accessed bit in > Andrea> _hardware_ via page faults I know is ia64, but I don't know > Andrea> if it has a mode to set it without page faults > > No, it doesn't. > > Andrea> and how it is implementing the accessed bit in linux. > > If the "accessed" or "dirty" bits are zero, accessing/writing the > page will cause a fault which will be handled in a low-level > fault handler. The Linux version of these handlers simply turn > on the respective bit. See daccess_bit(), iaccess_bit(), and dirty_bit() > in arch/ia64/kernel/ivt.S. so you mean, this is being set in the arch section before ever reaching handle_mm_fault? in such case my fix should work fine for ia64 too. > Note: I'm on travel and haven't seen the context of this discussion > and don't expect to have time to think about this until I return on > Thursday. So if you don't hear from me, it's not because I'm ignoring > you... ;-) take your time ;) thanks a lot for the above hints about those ivt.S functions (though I don't speak ia64 asm very well ;) -- 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: aart@kvack.org