From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pekka Enberg Subject: Re: [PATCH v6] add MAP_UNLOCKED mmap flag Date: Tue, 19 Jan 2010 10:44:23 +0200 Message-ID: <84144f021001190044s397c6665qb00af48235d2d818@mail.gmail.com> References: <20100118141938.GI30698@redhat.com> <20100118170816.GA22111@redhat.com> <84144f021001181009m52f7eaebp2bd746f92de08da9@mail.gmail.com> <20100118181942.GD22111@redhat.com> <20100118191031.0088f49a@lxorguk.ukuu.org.uk> <20100119071734.GG14345@redhat.com> <84144f021001182337o274c8ed3q8ce60581094bc2b9@mail.gmail.com> <20100119075205.GI14345@redhat.com> <84144f021001190007q54a334dfwed64189e6cf0b7c4@mail.gmail.com> <20100119082638.GK14345@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Return-path: In-Reply-To: <20100119082638.GK14345-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> Sender: linux-api-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Gleb Natapov Cc: Alan Cox , linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org, kosaki.motohiro-+CUm20s59erQFUHtdCDX3A@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org, andrew.c.morrow-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org, "Paul E. McKenney" , Rik van Riel , Johannes Weiner , minchan.kim-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org List-Id: linux-api@vger.kernel.org Hi Gleb, On Tue, Jan 19, 2010 at 10:26 AM, Gleb Natapov wrote: >> design would still be broken, no? Did you try using (or extending) >> posix_madvise(MADV_DONTNEED) for the guest address space? It seems to > After mlockall() I can't even allocate guest address space. Or do you mean > instead of mlockall()? Then how MADV_DONTNEED will help? It just drops > page table for the address range (which is not what I need) and does not > have any long time effect. Oh right, MADV_DONTNEED is no good. On Tue, Jan 19, 2010 at 10:26 AM, Gleb Natapov wrote: >> me that you're trying to use a big hammer (mlock) when a polite hint >> for the VM would probably be sufficient for it do its job. >> > I what to tell to VM "swap this, don't swap that" and as far as I see > there is no other way to do it currently. Yeah, which is why I was suggesting that maybe posix_madvise() needs to be extended to have a MADV_NEED_BUT_LESS_IMPORTANT flag that can be used as a hint by mm/vmscan.c to first swap the guest address spaces. Pekka