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 86D69C433F5 for ; Tue, 5 Apr 2022 01:54:35 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 8F8186B0083; Mon, 4 Apr 2022 21:49:20 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 8A7EC6B0085; Mon, 4 Apr 2022 21:49:20 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 7206B6B0087; Mon, 4 Apr 2022 21:49:20 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0243.hostedemail.com [216.40.44.243]) by kanga.kvack.org (Postfix) with ESMTP id 64FA26B0083 for ; Mon, 4 Apr 2022 21:49:20 -0400 (EDT) Received: from smtpin28.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay01.hostedemail.com (Postfix) with ESMTP id 2B30A1828AC85 for ; Tue, 5 Apr 2022 01:49:10 +0000 (UTC) X-FDA: 79321142460.28.E4537D6 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by imf11.hostedemail.com (Postfix) with ESMTP id B139D40018 for ; Tue, 5 Apr 2022 01:49:09 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1649123349; 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=95EZeCyII9xxx3gyIQ0ulWF2wAsEmWfCQ2/ndLgG/cw=; b=Mv7AcGNkNNbq1D8aJcJtFebae95Oo8GcyqjDHzyuOcKysKS2Wvi5JckLtliRxg7M3lknEz rSp13uCw9c/Eb8bwPwZOfmccsTalnswm/eK+fMpgyK711ELObSh63sbAbHJ6ogNwXCb8bN xZJJYHpWcuMzDxxW0Ao0RG5voJj/OI8= Received: from mail-il1-f200.google.com (mail-il1-f200.google.com [209.85.166.200]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-550--Lz3lX70PGO93-u_lsaClw-1; Mon, 04 Apr 2022 21:49:08 -0400 X-MC-Unique: -Lz3lX70PGO93-u_lsaClw-1 Received: by mail-il1-f200.google.com with SMTP id d13-20020a056e02214d00b002ca4d440f73so1848343ilv.15 for ; Mon, 04 Apr 2022 18:49:08 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=95EZeCyII9xxx3gyIQ0ulWF2wAsEmWfCQ2/ndLgG/cw=; b=mX6c9I4mP02Wyw8p4chhxQ6saCuffUH+7lzZDI3w6vD58SwC98II+kO16UUdVVgmNx bzAnfw7vVzJ/Aas0N5w//cmS3W7G5K14D259/JmZ6zCpzLMCQkURYbseKNI67kzgJrMR +eQLgSf3DKLD85FscespMAhdR60u3Bcv6vqdc3q3fNYZlI/DA49NvgdRNCaz0g1bpx+O ZuFbL/mQu07CnIsMnai+LZH4JyhEDLXdgtJ/iQXSE1hFApnpRL+w0mn8+R3qlVxQJSQw cpKVSR5uQTMAtThFf6w9VdgWLIvf52ErMpQbV7uk2ojwU1kGNhqIq9rdGE0riPlMiUWj /jrA== X-Gm-Message-State: AOAM532MSWLfP8jsPz/et5N6AgudP3ECzk6DcMc4bRFZZ4qY5k/rnzNT kDu2KttGaUtL1Pag9Yymb1BtkOoFm7W1ROpM1sT+3lsUL4SPSe60OppXXefWkhIYncWa6Ajez8y yvSh58+swz4g= X-Received: by 2002:a05:6602:13d5:b0:64c:9ef0:65e1 with SMTP id o21-20020a05660213d500b0064c9ef065e1mr588045iov.157.1649123346612; Mon, 04 Apr 2022 18:49:06 -0700 (PDT) X-Google-Smtp-Source: ABdhPJxvBZ1x9YNEIDP6fPb8GcdToNu45lYsI3Q8dcSYoOOQ7uq3K76BU1AVUi+jy6YnevEjPpvOKw== X-Received: by 2002:a05:6602:13d5:b0:64c:9ef0:65e1 with SMTP id o21-20020a05660213d500b0064c9ef065e1mr588020iov.157.1649123346392; Mon, 04 Apr 2022 18:49:06 -0700 (PDT) Received: from localhost.localdomain (cpec09435e3e0ee-cmc09435e3e0ec.cpe.net.cable.rogers.com. [99.241.198.116]) by smtp.gmail.com with ESMTPSA id b15-20020a05660214cf00b0064cb75d7e97sm7836568iow.53.2022.04.04.18.49.05 (version=TLS1_3 cipher=TLS_CHACHA20_POLY1305_SHA256 bits=256/256); Mon, 04 Apr 2022 18:49:06 -0700 (PDT) From: Peter Xu To: linux-kernel@vger.kernel.org, linux-mm@kvack.org Cc: Mike Kravetz , Nadav Amit , Matthew Wilcox , Mike Rapoport , David Hildenbrand , Hugh Dickins , Jerome Glisse , "Kirill A . Shutemov" , Andrea Arcangeli , Andrew Morton , Axel Rasmussen , Alistair Popple , peterx@redhat.com Subject: [PATCH v8 13/23] mm/hugetlb: Take care of UFFDIO_COPY_MODE_WP Date: Mon, 4 Apr 2022 21:49:04 -0400 Message-Id: <20220405014904.14643-1-peterx@redhat.com> X-Mailer: git-send-email 2.32.0 In-Reply-To: <20220405014646.13522-1-peterx@redhat.com> References: <20220405014646.13522-1-peterx@redhat.com> MIME-Version: 1.0 X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Type: text/plain; charset="US-ASCII"; x-default=true X-Stat-Signature: wtq5fj7ywzjzpurbd3poo4pitjme3mx1 Authentication-Results: imf11.hostedemail.com; dkim=pass header.d=redhat.com header.s=mimecast20190719 header.b=Mv7AcGNk; spf=none (imf11.hostedemail.com: domain of peterx@redhat.com has no SPF policy when checking 170.10.133.124) smtp.mailfrom=peterx@redhat.com; dmarc=pass (policy=none) header.from=redhat.com X-Rspam-User: X-Rspamd-Server: rspam02 X-Rspamd-Queue-Id: B139D40018 X-HE-Tag: 1649123349-81044 Content-Transfer-Encoding: quoted-printable 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: Pass the wp_copy variable into hugetlb_mcopy_atomic_pte() thoughout the s= tack. Apply the UFFD_WP bit if UFFDIO_COPY_MODE_WP is with UFFDIO_COPY. Hugetlb pages are only managed by hugetlbfs, so we're safe even without s= etting dirty bit in the huge pte if the page is installed as read-only. However= we'd better still keep the dirty bit set for a read-only UFFDIO_COPY pte (when UFFDIO_COPY_MODE_WP bit is set), not only to match what we do with shmem,= but also because the page does contain dirty data that the kernel just copied= from the userspace. Signed-off-by: Peter Xu --- include/linux/hugetlb.h | 6 ++++-- mm/hugetlb.c | 29 +++++++++++++++++++++++------ mm/userfaultfd.c | 14 +++++++++----- 3 files changed, 36 insertions(+), 13 deletions(-) diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 53c1b6082a4c..6347298778b6 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -160,7 +160,8 @@ int hugetlb_mcopy_atomic_pte(struct mm_struct *dst_mm= , pte_t *dst_pte, unsigned long dst_addr, unsigned long src_addr, enum mcopy_atomic_mode mode, - struct page **pagep); + struct page **pagep, + bool wp_copy); #endif /* CONFIG_USERFAULTFD */ bool hugetlb_reserve_pages(struct inode *inode, long from, long to, struct vm_area_struct *vma, @@ -355,7 +356,8 @@ static inline int hugetlb_mcopy_atomic_pte(struct mm_= struct *dst_mm, unsigned long dst_addr, unsigned long src_addr, enum mcopy_atomic_mode mode, - struct page **pagep) + struct page **pagep, + bool wp_copy) { BUG(); return 0; diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 82df0fcfedf9..c94deead22b2 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -5795,7 +5795,8 @@ int hugetlb_mcopy_atomic_pte(struct mm_struct *dst_= mm, unsigned long dst_addr, unsigned long src_addr, enum mcopy_atomic_mode mode, - struct page **pagep) + struct page **pagep, + bool wp_copy) { bool is_continue =3D (mode =3D=3D MCOPY_ATOMIC_CONTINUE); struct hstate *h =3D hstate_vma(dst_vma); @@ -5925,7 +5926,12 @@ int hugetlb_mcopy_atomic_pte(struct mm_struct *dst= _mm, goto out_release_unlock; =20 ret =3D -EEXIST; - if (!huge_pte_none(huge_ptep_get(dst_pte))) + /* + * We allow to overwrite a pte marker: consider when both MISSING|WP + * registered, we firstly wr-protect a none pte which has no page cache + * page backing it, then access the page. + */ + if (!huge_pte_none_mostly(huge_ptep_get(dst_pte))) goto out_release_unlock; =20 if (vm_shared) { @@ -5935,17 +5941,28 @@ int hugetlb_mcopy_atomic_pte(struct mm_struct *ds= t_mm, hugepage_add_new_anon_rmap(page, dst_vma, dst_addr); } =20 - /* For CONTINUE on a non-shared VMA, don't set VM_WRITE for CoW. */ - if (is_continue && !vm_shared) + /* + * For either: (1) CONTINUE on a non-shared VMA, or (2) UFFDIO_COPY + * with wp flag set, don't set pte write bit. + */ + if (wp_copy || (is_continue && !vm_shared)) writable =3D 0; else writable =3D dst_vma->vm_flags & VM_WRITE; =20 _dst_pte =3D make_huge_pte(dst_vma, page, writable); - if (writable) - _dst_pte =3D huge_pte_mkdirty(_dst_pte); + /* + * Always mark UFFDIO_COPY page dirty; note that this may not be + * extremely important for hugetlbfs for now since swapping is not + * supported, but we should still be clear in that this page cannot be + * thrown away at will, even if write bit not set. + */ + _dst_pte =3D huge_pte_mkdirty(_dst_pte); _dst_pte =3D pte_mkyoung(_dst_pte); =20 + if (wp_copy) + _dst_pte =3D huge_pte_mkuffd_wp(_dst_pte); + set_huge_pte_at(dst_mm, dst_addr, dst_pte, _dst_pte); =20 (void)huge_ptep_set_access_flags(dst_vma, dst_addr, dst_pte, _dst_pte, diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c index b1c875b77fbb..da0b3ed2a6b5 100644 --- a/mm/userfaultfd.c +++ b/mm/userfaultfd.c @@ -304,7 +304,8 @@ static __always_inline ssize_t __mcopy_atomic_hugetlb= (struct mm_struct *dst_mm, unsigned long dst_start, unsigned long src_start, unsigned long len, - enum mcopy_atomic_mode mode) + enum mcopy_atomic_mode mode, + bool wp_copy) { int vm_shared =3D dst_vma->vm_flags & VM_SHARED; ssize_t err; @@ -392,7 +393,7 @@ static __always_inline ssize_t __mcopy_atomic_hugetlb= (struct mm_struct *dst_mm, } =20 if (mode !=3D MCOPY_ATOMIC_CONTINUE && - !huge_pte_none(huge_ptep_get(dst_pte))) { + !huge_pte_none_mostly(huge_ptep_get(dst_pte))) { err =3D -EEXIST; mutex_unlock(&hugetlb_fault_mutex_table[hash]); i_mmap_unlock_read(mapping); @@ -400,7 +401,8 @@ static __always_inline ssize_t __mcopy_atomic_hugetlb= (struct mm_struct *dst_mm, } =20 err =3D hugetlb_mcopy_atomic_pte(dst_mm, dst_pte, dst_vma, - dst_addr, src_addr, mode, &page); + dst_addr, src_addr, mode, &page, + wp_copy); =20 mutex_unlock(&hugetlb_fault_mutex_table[hash]); i_mmap_unlock_read(mapping); @@ -455,7 +457,8 @@ extern ssize_t __mcopy_atomic_hugetlb(struct mm_struc= t *dst_mm, unsigned long dst_start, unsigned long src_start, unsigned long len, - enum mcopy_atomic_mode mode); + enum mcopy_atomic_mode mode, + bool wp_copy); #endif /* CONFIG_HUGETLB_PAGE */ =20 static __always_inline ssize_t mfill_atomic_pte(struct mm_struct *dst_mm= , @@ -575,7 +578,8 @@ static __always_inline ssize_t __mcopy_atomic(struct = mm_struct *dst_mm, */ if (is_vm_hugetlb_page(dst_vma)) return __mcopy_atomic_hugetlb(dst_mm, dst_vma, dst_start, - src_start, len, mcopy_mode); + src_start, len, mcopy_mode, + wp_copy); =20 if (!vma_is_anonymous(dst_vma) && !vma_is_shmem(dst_vma)) goto out_unlock; --=20 2.32.0