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 Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by smtp.lore.kernel.org (Postfix) with ESMTP id 2AF05C433FE for ; Thu, 3 Nov 2022 10:51:34 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 8FD9D6B0071; Thu, 3 Nov 2022 06:51:33 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 8AE4F6B0072; Thu, 3 Nov 2022 06:51:33 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 74EC86B0073; Thu, 3 Nov 2022 06:51:33 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0013.hostedemail.com [216.40.44.13]) by kanga.kvack.org (Postfix) with ESMTP id 68DF86B0071 for ; Thu, 3 Nov 2022 06:51:33 -0400 (EDT) Received: from smtpin27.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay07.hostedemail.com (Postfix) with ESMTP id 37CB4160EC4 for ; Thu, 3 Nov 2022 10:51:33 +0000 (UTC) X-FDA: 80091814866.27.926BB18 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by imf25.hostedemail.com (Postfix) with ESMTP id AD4D0A0006 for ; Thu, 3 Nov 2022 10:51:32 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1667472692; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=FfIKIj2BjovO2d1TGoZmkvSnm5ERGLD9n+Op2feiHhQ=; b=JhPKs3+ZNroTeID2f1+lJEa8VffpuNyrCDz7K1WA06wpyOLba1qPzgdtH8F10miI4JrDOy qJjZjjo3DKZPWXH0oYDURArl6vYD+pG526CRv+84+uMnKkTG36+I6B1DHc9R5ZzOVDaDmW MyvU//VeFJnHvUls/T5j0UoLg59SBWE= Received: from mail-wr1-f72.google.com (mail-wr1-f72.google.com [209.85.221.72]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_128_GCM_SHA256) id us-mta-638-HGlFLGdwNd2G2a-DLkaAQA-1; Thu, 03 Nov 2022 06:51:29 -0400 X-MC-Unique: HGlFLGdwNd2G2a-DLkaAQA-1 Received: by mail-wr1-f72.google.com with SMTP id u13-20020adfa18d000000b00236566b5b40so362639wru.9 for ; Thu, 03 Nov 2022 03:51:28 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=content-transfer-encoding:in-reply-to:organization:references:cc:to :from:content-language:subject:user-agent:mime-version:date :message-id:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=FfIKIj2BjovO2d1TGoZmkvSnm5ERGLD9n+Op2feiHhQ=; b=Pt4xaq1VwJKuBEP4LuIJcS6aK9aksdOq37sg8L9LUSfW4up/134gjjI/wmrYw3fZdR ndH18JiOHT92B9+upkDV5pzUri56bP4EK09Jj48DyVsR6hgkOOX+3nrPlSiXCd+D77yp qGHDuoqbELHfB8dWnHdrr+7qP5y0tLU6S9n+iQfrUtI0jadiqUTJdFTjC2dmnwjhkxC/ ZciI6gKLFLJls37NGp6XTynkI5TWEdT18KgGCl0lf7hvUgZRwutj6zf+RyG6ALE9NgpW 49awNcs0rkmvdkuHBsPH4X+GRqSfLie1rTah5idZCyEDOy7B8Or3wqm1KoPTElKvgFaW K0FA== X-Gm-Message-State: ACrzQf39DYrUm+R+snaA1gk7vVCRdCZAUUZWN2zC0JWSV3nLj1t/XnK3 AB4RdIL29WDQF3IosmB0ToZh8ZUCv8rLG8NVttEE9dxwaPEXBis8wJ6RdpayZ347SFcWF6o0IEM XU8jzRzHLAT4= X-Received: by 2002:adf:ef04:0:b0:236:b58b:3166 with SMTP id e4-20020adfef04000000b00236b58b3166mr17225494wro.330.1667472687951; Thu, 03 Nov 2022 03:51:27 -0700 (PDT) X-Google-Smtp-Source: AMsMyM52izUu3gFCfTKRloSM1S8x1tX0BSlp+bZPkawxYPtXps575wscWANLRETsddgGMrTKtKfmIQ== X-Received: by 2002:adf:ef04:0:b0:236:b58b:3166 with SMTP id e4-20020adfef04000000b00236b58b3166mr17225473wro.330.1667472687582; Thu, 03 Nov 2022 03:51:27 -0700 (PDT) Received: from ?IPV6:2003:cb:c707:a400:e2d7:3ee3:8d35:ac8? (p200300cbc707a400e2d73ee38d350ac8.dip0.t-ipconnect.de. [2003:cb:c707:a400:e2d7:3ee3:8d35:ac8]) by smtp.gmail.com with ESMTPSA id q14-20020a05600c46ce00b003b4ac05a8a4sm5720194wmo.27.2022.11.03.03.51.26 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Thu, 03 Nov 2022 03:51:27 -0700 (PDT) Message-ID: Date: Thu, 3 Nov 2022 11:51:26 +0100 MIME-Version: 1.0 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.4.0 Subject: Re: [PATCH v1 4/6] mm/autonuma: use can_change_(pte|pmd)_writable() to replace savedwrite From: David Hildenbrand To: Nadav Amit Cc: kernel list , Linux-MM , linuxppc-dev , Linus Torvalds , Andrew Morton , Mel Gorman , Dave Chinner , Peter Xu , Andrea Arcangeli , Hugh Dickins , Vlastimil Babka , Michael Ellerman , Nicholas Piggin , Mike Rapoport , Anshuman Khandual References: <20221102191209.289237-1-david@redhat.com> <20221102191209.289237-5-david@redhat.com> <37192084-fc78-2d51-3bcf-1454248dcc74@redhat.com> Organization: Red Hat In-Reply-To: <37192084-fc78-2d51-3bcf-1454248dcc74@redhat.com> X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Language: en-US Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1667472692; a=rsa-sha256; cv=none; b=xjeoZEAEL0EYTngIgYW3WHusDSC3wCC++DuUY6boL+jCkdQk3wL7iHGKdy9aHcf3PwqPg0 fHoDDWrFYNy+bFLEpFE2OPwZ9iRFmNEiVhus8VqspqW8q/cp7uAgBE81Y9yUW+evQo1lbn uHqXxadwSy5gVFomCAwIxCb3rM+8Zlo= ARC-Authentication-Results: i=1; imf25.hostedemail.com; dkim=pass header.d=redhat.com header.s=mimecast20190719 header.b=JhPKs3+Z; dmarc=pass (policy=none) header.from=redhat.com; spf=pass (imf25.hostedemail.com: domain of david@redhat.com designates 170.10.133.124 as permitted sender) smtp.mailfrom=david@redhat.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1667472692; h=from:from:sender:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references:dkim-signature; bh=FfIKIj2BjovO2d1TGoZmkvSnm5ERGLD9n+Op2feiHhQ=; b=cITgnu/8rDmOnw26dVOy4eW2g9+Xut5rfSqI8HxUvTsHDsRLhfnssaJZ4aY4J2nAn+AXfh 1VWSOLZv2wkJlKCSCcGT3KBMN6fRUKw2NDRIbFPw6Fo7R9JbyD3yFEs1joSw1WMI1lRe95 15ZmB9d6MpN3dcird54m7okLqZYNjNU= X-Rspamd-Server: rspam10 X-Rspamd-Queue-Id: AD4D0A0006 X-Rspam-User: Authentication-Results: imf25.hostedemail.com; dkim=pass header.d=redhat.com header.s=mimecast20190719 header.b=JhPKs3+Z; dmarc=pass (policy=none) header.from=redhat.com; spf=pass (imf25.hostedemail.com: domain of david@redhat.com designates 170.10.133.124 as permitted sender) smtp.mailfrom=david@redhat.com X-Stat-Signature: 6xbae9na1pcmrft9duuurr4dc34qr8ao X-HE-Tag: 1667472692-632023 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: On 03.11.22 11:45, David Hildenbrand wrote: > On 02.11.22 22:22, Nadav Amit wrote: >> On Nov 2, 2022, at 12:12 PM, David Hildenbrand wrote: >> >>> !! External Email >>> >>> commit b191f9b106ea ("mm: numa: preserve PTE write permissions across a >>> NUMA hinting fault") added remembering write permissions using ordinary >>> pte_write() for PROT_NONE mapped pages to avoid write faults when >>> remapping the page !PROT_NONE on NUMA hinting faults. >>> >> >> [ snip ] >> >> Here’s a very shallow reviewed with some minor points... > > Appreciated. > >> >>> --- >>> include/linux/mm.h | 2 ++ >>> mm/huge_memory.c | 28 +++++++++++++++++----------- >>> mm/ksm.c | 9 ++++----- >>> mm/memory.c | 19 ++++++++++++++++--- >>> mm/mprotect.c | 7 ++----- >>> 5 files changed, 41 insertions(+), 24 deletions(-) >>> >>> diff --git a/include/linux/mm.h b/include/linux/mm.h >>> index 25ff9a14a777..a0deeece5e87 100644 >>> --- a/include/linux/mm.h >>> +++ b/include/linux/mm.h >>> @@ -1975,6 +1975,8 @@ extern unsigned long move_page_tables(struct vm_area_struct *vma, >>> #define MM_CP_UFFD_WP_ALL (MM_CP_UFFD_WP | \ >>> MM_CP_UFFD_WP_RESOLVE) >>> >>> +bool can_change_pte_writable(struct vm_area_struct *vma, unsigned long addr, >>> + pte_t pte); >> >> It might not be customary, but how about marking it as __pure? > > Right, there is no a single use of __pure in the mm domain. > >> >>> extern unsigned long change_protection(struct mmu_gather *tlb, >>> struct vm_area_struct *vma, unsigned long start, >>> unsigned long end, pgprot_t newprot, >>> diff --git a/mm/huge_memory.c b/mm/huge_memory.c >>> index 2ad68e91896a..45abd27d75a0 100644 >>> --- a/mm/huge_memory.c >>> +++ b/mm/huge_memory.c >>> @@ -1462,8 +1462,7 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf) >>> unsigned long haddr = vmf->address & HPAGE_PMD_MASK; >>> int page_nid = NUMA_NO_NODE; >>> int target_nid, last_cpupid = (-1 & LAST_CPUPID_MASK); >>> - bool migrated = false; >>> - bool was_writable = pmd_savedwrite(oldpmd); >>> + bool try_change_writable, migrated = false; >>> int flags = 0; >>> >>> vmf->ptl = pmd_lock(vma->vm_mm, vmf->pmd); >>> @@ -1472,13 +1471,22 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf) >>> goto out; >>> } >>> >>> + /* See mprotect_fixup(). */ >>> + if (vma->vm_flags & VM_SHARED) >>> + try_change_writable = vma_wants_writenotify(vma, vma->vm_page_prot); >>> + else >>> + try_change_writable = !!(vma->vm_flags & VM_WRITE); >> >> Do you find it better to copy the code instead of extracting it to a >> separate function? > > Yeah, you're right ;) usually the issue is coming up with a suitable name. Let me try. > > vma_wants_manual_writability_change() hmm ... > >> >>> + >>> pmd = pmd_modify(oldpmd, vma->vm_page_prot); >>> page = vm_normal_page_pmd(vma, haddr, pmd); >>> if (!page) >>> goto out_map; >>> >>> /* See similar comment in do_numa_page for explanation */ >>> - if (!was_writable) >>> + if (try_change_writable && !pmd_write(pmd) && >>> + can_change_pmd_writable(vma, vmf->address, pmd)) >>> + pmd = pmd_mkwrite(pmd); >>> + if (!pmd_write(pmd)) >>> flags |= TNF_NO_GROUP; >>> >>> page_nid = page_to_nid(page); >>> @@ -1523,8 +1531,12 @@ vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf) >>> /* Restore the PMD */ >>> pmd = pmd_modify(oldpmd, vma->vm_page_prot); >>> pmd = pmd_mkyoung(pmd); >>> - if (was_writable) >>> + >>> + /* Similar to mprotect() protection updates, avoid write faults. */ >>> + if (try_change_writable && !pmd_write(pmd) && >>> + can_change_pmd_writable(vma, vmf->address, pmd)) >> >> Why do I have a deja-vu? :) >> >> There must be a way to avoid the redundant code and specifically the call to >> can_change_pmd_writable(), no? > > The issue is that as soon as we drop the page table lock, that information is stale. > Especially, after we fail migration. > > So the following should work, however, if we fail migration we wouldn't map the > page writable and would have to re-calculate: > > diff --git a/mm/memory.c b/mm/memory.c > index c5599a9279b1..a997625641e4 100644 > --- a/mm/memory.c > +++ b/mm/memory.c > @@ -4674,10 +4674,10 @@ static vm_fault_t do_numa_page(struct vm_fault *vmf) > struct vm_area_struct *vma = vmf->vma; > struct page *page = NULL; > int page_nid = NUMA_NO_NODE; > + bool writable = false; > int last_cpupid; > int target_nid; > pte_t pte, old_pte; > - bool was_writable = pte_savedwrite(vmf->orig_pte); > int flags = 0; > > /* > @@ -4696,6 +4696,17 @@ static vm_fault_t do_numa_page(struct vm_fault *vmf) > old_pte = ptep_get(vmf->pte); > pte = pte_modify(old_pte, vma->vm_page_prot); > > + /* > + * Detect now whether the PTE is or can be writable. Note that this > + * information is valid as long as we're holding the PT lock, so also on > + * the remap path below. > + */ > + writable = pte_write(pte); > + if (!writable && vma_wants_manual_writability_change(vma) && > + can_change_pte_writable(vma, vmf->address, pte); > + writable = true; > + } > + > page = vm_normal_page(vma, vmf->address, pte); > if (!page || is_zone_device_page(page)) > goto out_map; > @@ -4712,7 +4723,7 @@ static vm_fault_t do_numa_page(struct vm_fault *vmf) > * pte_dirty has unpredictable behaviour between PTE scan updates, > * background writeback, dirty balancing and application behaviour. > */ > - if (!was_writable) > + if (!writable) > flags |= TNF_NO_GROUP; > > /* > @@ -4738,6 +4749,7 @@ static vm_fault_t do_numa_page(struct vm_fault *vmf) > put_page(page); > goto out_map; > } > + writable = false; > pte_unmap_unlock(vmf->pte, vmf->ptl); > > /* Migrate to the requested node */ > @@ -4767,7 +4779,7 @@ static vm_fault_t do_numa_page(struct vm_fault *vmf) > old_pte = ptep_modify_prot_start(vma, vmf->address, vmf->pte); > pte = pte_modify(old_pte, vma->vm_page_prot); > pte = pte_mkyoung(pte); > - if (was_writable) > + if (writable) > pte = pte_mkwrite(pte); > ptep_modify_prot_commit(vma, vmf->address, vmf->pte, old_pte, pte); > update_mmu_cache(vma, vmf->address, vmf->pte); > > > To me, the less error-prone approach is to re-calculate. Hmm, thinking again, the "if (unlikely(!pte_same(*vmf->pte, vmf->orig_pte))) {" check might actually not require us to recalculate. -- Thanks, David / dhildenb