From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Tue, 25 May 2004 15:01:55 -0700 (PDT) From: Linus Torvalds Subject: Re: [PATCH] ppc64: Fix possible race with set_pte on a present PTE In-Reply-To: <20040525215500.GI29378@dualathlon.random> Message-ID: References: <1085371988.15281.38.camel@gaston> <1085373839.14969.42.camel@gaston> <20040525034326.GT29378@dualathlon.random> <20040525114437.GC29154@parcelfarce.linux.theplanet.co.uk> <20040525212720.GG29378@dualathlon.random> <20040525215500.GI29378@dualathlon.random> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-mm@kvack.org Return-Path: To: Andrea Arcangeli Cc: Matthew Wilcox , Benjamin Herrenschmidt , Andrew Morton , Linux Kernel list , Ingo Molnar , Ben LaHaise , linux-mm@kvack.org, Architectures Group List-ID: On Tue, 25 May 2004, Andrea Arcangeli wrote: > > I expected the pal code to re-read the pte if the control bits asked for > page fault, like it must happen if the control bits are set to > non-present. That may or may not be true. I _think_ it wasn't true. > This latter this must be true or linux wouldn't run at all > on alpha. A "not-present" fault is a totally different fault from a "protection fault". Only the not-present fault ends up walking the page tables, if I remember correctly. The PAL-code sources are out there somewhere, so I guess this should be easy to check if I wasn't so lazy. Linus -- 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