From mboxrd@z Thu Jan 1 00:00:00 1970 Content-Type: text/plain; charset="iso-8859-1" From: Daniel Phillips Subject: Re: [RFC] Page table sharing Date: Wed, 20 Feb 2002 15:57:37 +0100 References: In-Reply-To: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Message-Id: Sender: owner-linux-mm@kvack.org Return-Path: To: Hugh Dickins Cc: Linus Torvalds , Rik van Riel , dmccr@us.ibm.com, Kernel Mailing List , linux-mm@kvack.org, Robert Love , mingo@redhat.com, Andrew Morton , manfred@colorfullife.com, wli@holomorphy.com List-ID: On February 20, 2002 03:38 pm, Hugh Dickins wrote: > On Wed, 20 Feb 2002, Daniel Phillips wrote: > > > > Looking at the current try_to_swap_out code I see only a local invalidate, > > flush_tlb_page(vma, address), why is that? How do we know that this mm could > > not be in context on another cpu? > > I made the same mistake a few months ago: not noticing #ifndef CONFIG_SMP > in the header. arch/i386/kernel/smp.c has the real i386 flush_tlb_page(). OK, well if I'm making the same mistakes then I'm likely on the right track ;) So it seems that what we need for tlb invalidate of shared page tables is not worse than what we already have, though there's some extra bookkeeping to handle. Why would we run into your page dirty propagation problem with shared page tables and not with the current code? -- Daniel -- 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/