From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-1.0 required=3.0 tests=DKIM_SIGNED,DKIM_VALID, DKIM_VALID_AU,FREEMAIL_FORGED_FROMDOMAIN,FREEMAIL_FROM, HEADER_FROM_DIFFERENT_DOMAINS,MAILING_LIST_MULTI,SPF_PASS autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 73827C04EB8 for ; Tue, 4 Dec 2018 23:34:16 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 263172082B for ; Tue, 4 Dec 2018 23:34:16 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="NYzo/NqV" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 263172082B Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=gmail.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726263AbeLDXeO (ORCPT ); Tue, 4 Dec 2018 18:34:14 -0500 Received: from mail-pg1-f193.google.com ([209.85.215.193]:42627 "EHLO mail-pg1-f193.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725875AbeLDXeN (ORCPT ); Tue, 4 Dec 2018 18:34:13 -0500 Received: by mail-pg1-f193.google.com with SMTP id d72so8087694pga.9; Tue, 04 Dec 2018 15:34:13 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=vR6rTWwdorajVFH529LIeJsZtnh4Njei7Pf4Scavbtc=; b=NYzo/NqVbKcyLv/5jyErJKWePoDYYCj4MMGh1cWhzScD9N6lyDQifSYp1SylvsSnLa 3vxC1Tle4hmFogTxGEAmxXIBnU1uQcvf9ILOilTAY++T2jchLNqYd/6U1ECxOn5EQvnn QLYx2ftgBWI7Slx5btpeMbAKhyIFxCJ73S87PcoBRD5C6OaPSa3Blm8gfSD8f63mAEVw Fv5wZo9UOKfJvVU43/4n+QZO7S3UyHy8ko9kihmz+sztB+9LolWUpevaz+z5kzKqRQWv 9OP7UJYE2Ufqqcu165HsS2tu6zhddwsKWKvhA+hGFR0mtX007ZtvcT0aMNmfVIJW5TFa Khjw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:subject:from:in-reply-to:date:cc :content-transfer-encoding:message-id:references:to; bh=vR6rTWwdorajVFH529LIeJsZtnh4Njei7Pf4Scavbtc=; b=LAgGpdrLdvqGWzsHN8Ucj9M+PemcR6/ansSSXNcVBnRJmCTUc9ukkiG3BGYUsBpcBc QtsluNQjKRB9+/ClYuRWO6EO0bnYeCp0qmk/A4dFjdoLaRu4LY1x19SF0042z3g3azWG Fie/UoCbIOogx5fAO7sm4Vd9V6jMdT0ihyj4xyyAk45RfLPZyqVhhIU89bGKtMQB9Z8K jZE834V8TcKsfK23SYIdQLq4z40nkoefk/KHvdRLkBDkQ/ZtNFkqOwpSWXFqixdi5D8/ c2eMhjtZygC0Mo1b70UK9lN6wbz2Qv4CckJ2cBdl7D1+JuDPXD0RqJtNvZ8IgDkx8sIu 7srw== X-Gm-Message-State: AA+aEWYpqPc97L9CglKGDjubP/qDEZRO4GDvS1wKKYjs3/mZ/WoGBdGa 5kIXXUeuUb0YS+8xsO2MKWY= X-Google-Smtp-Source: AFSGD/VDiEj3p1RymmTPxRDd+V5urAcTHFEoWoHFdE9yzpjtI9phUAuwNhuJnnTApLMbnUFB2Q1FEw== X-Received: by 2002:a63:de04:: with SMTP id f4mr18169415pgg.292.1543966452689; Tue, 04 Dec 2018 15:34:12 -0800 (PST) Received: from [10.2.19.70] ([208.91.2.1]) by smtp.gmail.com with ESMTPSA id x127sm26777387pfd.156.2018.12.04.15.34.10 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 04 Dec 2018 15:34:11 -0800 (PST) Content-Type: text/plain; charset=utf-8 Mime-Version: 1.0 (Mac OS X Mail 12.1 \(3445.101.1\)) Subject: Re: [PATCH 1/2] vmalloc: New flag for flush before releasing pages From: Nadav Amit In-Reply-To: Date: Tue, 4 Dec 2018 15:34:09 -0800 Cc: Andy Lutomirski , Rick Edgecombe , Andrew Morton , Will Deacon , Linux-MM , LKML , Kernel Hardening , "Naveen N . Rao" , Anil S Keshavamurthy , "David S. Miller" , Masami Hiramatsu , Steven Rostedt , Ingo Molnar , Alexei Starovoitov , Daniel Borkmann , jeyu@kernel.org, Network Development , Ard Biesheuvel , Jann Horn , Kristen Carlson Accardi , Dave Hansen , "Dock, Deneen T" , Peter Zijlstra Content-Transfer-Encoding: quoted-printable Message-Id: <1FCD958E-C875-4F26-8E59-10615D7BF22E@gmail.com> References: <20181128000754.18056-1-rick.p.edgecombe@intel.com> <20181128000754.18056-2-rick.p.edgecombe@intel.com> <4883FED1-D0EC-41B0-A90F-1A697756D41D@gmail.com> <08141F66-F3E6-4CC5-AF91-1ED5F101A54C@gmail.com> <20CC2F71-308D-42E2-8C54-F64D7CC3863F@gmail.com> To: Andy Lutomirski X-Mailer: Apple Mail (2.3445.101.1) Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org > On Dec 4, 2018, at 3:27 PM, Andy Lutomirski = wrote: >=20 >=20 >=20 >=20 > On Dec 4, 2018, at 2:48 PM, Nadav Amit wrote: >=20 >>> On Dec 4, 2018, at 11:48 AM, Andy Lutomirski = wrote: >>>=20 >>> On Tue, Dec 4, 2018 at 11:45 AM Nadav Amit = wrote: >>>>> On Dec 4, 2018, at 10:56 AM, Andy Lutomirski = wrote: >>>>>=20 >>>>> On Mon, Dec 3, 2018 at 5:43 PM Nadav Amit = wrote: >>>>>>> On Nov 27, 2018, at 4:07 PM, Rick Edgecombe = wrote: >>>>>>>=20 >>>>>>> Since vfree will lazily flush the TLB, but not lazily free the = underlying pages, >>>>>>> it often leaves stale TLB entries to freed pages that could get = re-used. This is >>>>>>> undesirable for cases where the memory being freed has special = permissions such >>>>>>> as executable. >>>>>>=20 >>>>>> So I am trying to finish my patch-set for preventing transient = W+X mappings >>>>>> from taking space, by handling kprobes & ftrace that I missed = (thanks again for >>>>>> pointing it out). >>>>>>=20 >>>>>> But all of the sudden, I don=E2=80=99t understand why we have the = problem that this >>>>>> (your) patch-set deals with at all. We already change the = mappings to make >>>>>> the memory writable before freeing the memory, so why can=E2=80=99t= we make it >>>>>> non-executable at the same time? Actually, why do we make the = module memory, >>>>>> including its data executable before freeing it??? >>>>>=20 >>>>> All the code you're looking at is IMO a very awkward and possibly >>>>> incorrect of doing what's actually necessary: putting the direct = map >>>>> the way it wants to be. >>>>>=20 >>>>> Can't we shove this entirely mess into vunmap? Have a flag (as = part >>>>> of vmalloc like in Rick's patch or as a flag passed to a vfree = variant >>>>> directly) that makes the vunmap code that frees the underlying = pages >>>>> also reset their permissions? >>>>>=20 >>>>> Right now, we muck with set_memory_rw() and set_memory_nx(), which >>>>> both have very awkward (and inconsistent with each other!) = semantics >>>>> when called on vmalloc memory. And they have their own flushes, = which >>>>> is inefficient. Maybe the right solution is for vunmap to remove = the >>>>> vmap area PTEs, call into a function like set_memory_rw() that = resets >>>>> the direct maps to their default permissions *without* flushing, = and >>>>> then to do a single flush for everything. Or, even better, to = cause >>>>> the change_page_attr code to do the flush and also to flush the = vmap >>>>> area all at once so that very small free operations can flush = single >>>>> pages instead of flushing globally. >>>>=20 >>>> Thanks for the explanation. I read it just after I realized that = indeed the >>>> whole purpose of this code is to get cpa_process_alias() >>>> update the corresponding direct mapping. >>>>=20 >>>> This thing (pageattr.c) indeed seems over-engineered and very = unintuitive. >>>> Right now I have a list of patch-sets that I owe, so I don=E2=80=99t = have the time >>>> to deal with it. >>>>=20 >>>> But, I still think that disable_ro_nx() should not call = set_memory_x(). >>>> IIUC, this breaks W+X of the direct-mapping which correspond with = the module >>>> memory. Does it ever stop being W+X?? I=E2=80=99ll have another = look. >>>=20 >>> Dunno. I did once chase down a bug where some memory got freed = while >>> it was still read-only, and the results were hilarious and hard to >>> debug, since the explosion happened long after the buggy code >>> finished. >>=20 >> This piece of code causes me pain and misery. >>=20 >> So, it turns out that the direct map is *not* changed if you just = change >> the NX-bit. See change_page_attr_set_clr(): >>=20 >> /* No alias checking for _NX bit modifications */ >> checkalias =3D (pgprot_val(mask_set) | pgprot_val(mask_clr)) !=3D= _PAGE_NX; >>=20 >> How many levels of abstraction are broken in the way? What would = happen >> if somebody tries to change the NX-bit and some other bit in the PTE? >> Luckily, I don=E2=80=99t think someone does=E2=80=A6 at least for = now. >>=20 >> So, again, I think the change I proposed makes sense. nios2 does not = have >> set_memory_x() and it will not be affected. >>=20 >> [ I can add a comment, although I don=E2=80=99t have know if nios2 = has an NX bit, >> and I don=E2=80=99t find any code that defines PTEs. Actually where = is pte_present() >> of nios2 being defined? Whatever. ] >=20 > At least rename the function, then. The last thing we need is for > disable_ro_nx to *enable* NX. The code is so horrible right now (IMHO), that it will not make it much worse. But, yes, I will of course change the name. I just want to finish this text_poke() patch-set and W+X mappings keep popping up. Thanks (as usual) for your help.