From patchwork Fri Jun 24 17:36:45 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: James Houghton X-Patchwork-Id: 12894940 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 533EDCCA473 for ; Fri, 24 Jun 2022 17:37:39 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 8734C8E024D; Fri, 24 Jun 2022 13:37:32 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 7AD0F8E0244; Fri, 24 Jun 2022 13:37:32 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 5B1548E024D; Fri, 24 Jun 2022 13:37:32 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0015.hostedemail.com [216.40.44.15]) by kanga.kvack.org (Postfix) with ESMTP id 3E2BD8E0244 for ; Fri, 24 Jun 2022 13:37:32 -0400 (EDT) Received: from smtpin13.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay06.hostedemail.com (Postfix) with ESMTP id EE36535476 for ; Fri, 24 Jun 2022 17:37:31 +0000 (UTC) X-FDA: 79613836302.13.659E5FA Received: from mail-yw1-f201.google.com (mail-yw1-f201.google.com [209.85.128.201]) by imf16.hostedemail.com (Postfix) with ESMTP id A5882180025 for ; Fri, 24 Jun 2022 17:37:31 +0000 (UTC) Received: by mail-yw1-f201.google.com with SMTP id 00721157ae682-31382419c22so26681707b3.18 for ; Fri, 24 Jun 2022 10:37:31 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; h=date:in-reply-to:message-id:mime-version:references:subject:from:to :cc; bh=LoJsmKx0RIJYFCgh3RUmS9FnkZR9nxjSRnqvv6TZGjM=; b=IZS4DM3M3K6LoUVsC8Bl04dfOKEEAWyhZp/PaBKn0d92TIDnzwMDFN7wLf3gZK8KXk Hqp1S+enLGzviCJsO6DpNpGzxKEp+T0yzrdIwA4S415ZdLqTsoPiTRgFb6aOA0kKPBAY zyheLmcN2aPMf7MJVwN4znu5j9nveQnM+HbQGBNnfXtj5p0uhIS5X82WJqumvoFeu78Z kexFoLJRgVhL59k8quEJaOagGrrlQ0qdHEHVyls+GVapUIwLCmrA+dHzKoYHQOi/e8xv vUWXfvm60zyRjMUV+v9MtvRAUv6UlhcNITygQafNxgge+q1x6oNJBsUIR4wKbgjkQNng /GzA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:date:in-reply-to:message-id:mime-version :references:subject:from:to:cc; bh=LoJsmKx0RIJYFCgh3RUmS9FnkZR9nxjSRnqvv6TZGjM=; b=mGyYPqq1aVXJCcRcts4hVwZji6fn9XkOf48zCjmvoxMfKQWaRCGapLL3dxhNTnG8F5 FmXbLMsATYodnDTA8qPHu7QWvfZ7/XOZPrQGsnsqihT8hsMM86hXTUJcXT5scBEf5uhv +Jb9QykGWgjm0ynlyc46h+q/Ye7t0lYmQ72muV1+G5U4wc5VM75WIfXh8KStkwzxOFY5 jpw1tUvNtA1vrR/JFQWf0VC2E6vYwt9wibih5e/wzcLnkitW5fBm0t05OuyyOfSr3Uw4 h8rDaPjqKyoTrhLm4x1Ui2O6zBbue02lwDLRyWUkforDfo0Rp7O3gG5AU5YaqPZenySq k9uQ== X-Gm-Message-State: AJIora84uPlFdvH3xNG1yomrNWyIiVD8LPY4etP9gV2wxqwFV+xXwjdj 5jN8LfygtxRByzLj9gZYcMBMaCx9FVavkb4Q X-Google-Smtp-Source: AGRyM1se79AgT0Cox1n1PfeD3monM06aR7cLfdIp8zaSZ2VT02A9SECnyrpKDnZIvwtzr+hFf9tsMPJX4L1dTD28 X-Received: from jthoughton.c.googlers.com ([fda3:e722:ac3:cc00:14:4d90:c0a8:2a4f]) (user=jthoughton job=sendgmr) by 2002:a25:7801:0:b0:669:b51b:10d0 with SMTP id t1-20020a257801000000b00669b51b10d0mr331321ybc.204.1656092251057; Fri, 24 Jun 2022 10:37:31 -0700 (PDT) Date: Fri, 24 Jun 2022 17:36:45 +0000 In-Reply-To: <20220624173656.2033256-1-jthoughton@google.com> Message-Id: <20220624173656.2033256-16-jthoughton@google.com> Mime-Version: 1.0 References: <20220624173656.2033256-1-jthoughton@google.com> X-Mailer: git-send-email 2.37.0.rc0.161.g10f37bed90-goog Subject: [RFC PATCH 15/26] hugetlb: make unmapping compatible with high-granularity mappings From: James Houghton To: Mike Kravetz , Muchun Song , Peter Xu Cc: David Hildenbrand , David Rientjes , Axel Rasmussen , Mina Almasry , Jue Wang , Manish Mishra , "Dr . David Alan Gilbert" , linux-mm@kvack.org, linux-kernel@vger.kernel.org, James Houghton ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1656092251; a=rsa-sha256; cv=none; b=PNUM23zcrYzkqW8Tl7YCYnij+O5IafyR8mBM5I3xtNQ9gnL9sVYFL7hUMGGNn8/IORkkFt qfkxW4WBGBm2Igt0mGtyTwY/5TJRKD73KX7iCi0lkrCu29esOV0WnuvNZImiDWzw7wvma9 Eputmd+jE2Tn66lkV5AWyHoqG6t0CjU= ARC-Authentication-Results: i=1; imf16.hostedemail.com; dkim=pass header.d=google.com header.s=20210112 header.b=IZS4DM3M; dmarc=pass (policy=reject) header.from=google.com; spf=pass (imf16.hostedemail.com: domain of 3W_a1YgoKCDwhrfmsefrmlemmejc.amkjglsv-kkitYai.mpe@flex--jthoughton.bounces.google.com designates 209.85.128.201 as permitted sender) smtp.mailfrom=3W_a1YgoKCDwhrfmsefrmlemmejc.amkjglsv-kkitYai.mpe@flex--jthoughton.bounces.google.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1656092251; 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: in-reply-to:in-reply-to:references:references:dkim-signature; bh=LoJsmKx0RIJYFCgh3RUmS9FnkZR9nxjSRnqvv6TZGjM=; b=q1wgQvKH3xe0BiuZN6uWdGclEyrfSKJSGevAXHDNaMHhV+qi6fy1SuvCCtHSRdsCzDK8E+ 1Zc3L97aXJMByFPMHGPMbsxLCD29xrB3owzO0e1XYLZnXUNTA7Yfy6J4xmspXEdLSanJC6 cAw835GSA5DxrbUR4JWKCEEj9XnFvPk= X-Rspam-User: X-Stat-Signature: 1huza4x1j8aa978b8sa9t59m1uurind4 X-Rspamd-Queue-Id: A5882180025 Authentication-Results: imf16.hostedemail.com; dkim=pass header.d=google.com header.s=20210112 header.b=IZS4DM3M; dmarc=pass (policy=reject) header.from=google.com; spf=pass (imf16.hostedemail.com: domain of 3W_a1YgoKCDwhrfmsefrmlemmejc.amkjglsv-kkitYai.mpe@flex--jthoughton.bounces.google.com designates 209.85.128.201 as permitted sender) smtp.mailfrom=3W_a1YgoKCDwhrfmsefrmlemmejc.amkjglsv-kkitYai.mpe@flex--jthoughton.bounces.google.com X-Rspamd-Server: rspam03 X-HE-Tag: 1656092251-452077 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: This enlightens __unmap_hugepage_range to deal with high-granularity mappings. This doesn't change its API; it still must be called with hugepage alignment, but it will correctly unmap hugepages that have been mapped at high granularity. Analogous to the mapcount rules introduced by hugetlb_no_page, we only drop mapcount in this case if we are unmapping an entire hugepage in one operation. This is the case when a VMA is destroyed. Eventually, functionality here can be expanded to allow users to call MADV_DONTNEED on PAGE_SIZE-aligned sections of a hugepage, but that is not done here. Signed-off-by: James Houghton --- include/asm-generic/tlb.h | 6 +-- mm/hugetlb.c | 85 ++++++++++++++++++++++++++------------- 2 files changed, 59 insertions(+), 32 deletions(-) diff --git a/include/asm-generic/tlb.h b/include/asm-generic/tlb.h index ff3e82553a76..8daa3ae460d9 100644 --- a/include/asm-generic/tlb.h +++ b/include/asm-generic/tlb.h @@ -562,9 +562,9 @@ static inline void tlb_flush_p4d_range(struct mmu_gather *tlb, __tlb_remove_tlb_entry(tlb, ptep, address); \ } while (0) -#define tlb_remove_huge_tlb_entry(h, tlb, ptep, address) \ +#define tlb_remove_huge_tlb_entry(tlb, hpte, address) \ do { \ - unsigned long _sz = huge_page_size(h); \ + unsigned long _sz = hugetlb_pte_size(&hpte); \ if (_sz >= P4D_SIZE) \ tlb_flush_p4d_range(tlb, address, _sz); \ else if (_sz >= PUD_SIZE) \ @@ -573,7 +573,7 @@ static inline void tlb_flush_p4d_range(struct mmu_gather *tlb, tlb_flush_pmd_range(tlb, address, _sz); \ else \ tlb_flush_pte_range(tlb, address, _sz); \ - __tlb_remove_tlb_entry(tlb, ptep, address); \ + __tlb_remove_tlb_entry(tlb, hpte.ptep, address);\ } while (0) /** diff --git a/mm/hugetlb.c b/mm/hugetlb.c index da30621656b8..51fc1d3f122f 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -5120,24 +5120,20 @@ static void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct { struct mm_struct *mm = vma->vm_mm; unsigned long address; - pte_t *ptep; + struct hugetlb_pte hpte; pte_t pte; spinlock_t *ptl; - struct page *page; + struct page *hpage, *subpage; struct hstate *h = hstate_vma(vma); unsigned long sz = huge_page_size(h); struct mmu_notifier_range range; bool force_flush = false; + bool hgm_enabled = hugetlb_hgm_enabled(vma); WARN_ON(!is_vm_hugetlb_page(vma)); BUG_ON(start & ~huge_page_mask(h)); BUG_ON(end & ~huge_page_mask(h)); - /* - * This is a hugetlb vma, all the pte entries should point - * to huge page. - */ - tlb_change_page_size(tlb, sz); tlb_start_vma(tlb, vma); /* @@ -5148,25 +5144,43 @@ static void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct adjust_range_if_pmd_sharing_possible(vma, &range.start, &range.end); mmu_notifier_invalidate_range_start(&range); address = start; - for (; address < end; address += sz) { - ptep = huge_pte_offset(mm, address, sz); - if (!ptep) + + while (address < end) { + pte_t *ptep = huge_pte_offset(mm, address, sz); + + if (!ptep) { + address += sz; continue; + } + hugetlb_pte_populate(&hpte, ptep, huge_page_shift(h)); + if (hgm_enabled) { + int ret = huge_pte_alloc_high_granularity( + &hpte, mm, vma, address, PAGE_SHIFT, + HUGETLB_SPLIT_NEVER, + /*write_locked=*/true); + /* + * We will never split anything, so this should always + * succeed. + */ + BUG_ON(ret); + } - ptl = huge_pte_lock(h, mm, ptep); - if (huge_pmd_unshare(mm, vma, &address, ptep)) { + ptl = hugetlb_pte_lock(mm, &hpte); + if (!hgm_enabled && huge_pmd_unshare( + mm, vma, &address, hpte.ptep)) { spin_unlock(ptl); tlb_flush_pmd_range(tlb, address & PUD_MASK, PUD_SIZE); force_flush = true; - continue; + goto next_hpte; } - pte = huge_ptep_get(ptep); - if (huge_pte_none(pte)) { + if (hugetlb_pte_none(&hpte)) { spin_unlock(ptl); - continue; + goto next_hpte; } + pte = hugetlb_ptep_get(&hpte); + /* * Migrating hugepage or HWPoisoned hugepage is already * unmapped and its refcount is dropped, so just clear pte here. @@ -5180,24 +5194,27 @@ static void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct */ if (pte_swp_uffd_wp_any(pte) && !(zap_flags & ZAP_FLAG_DROP_MARKER)) - set_huge_pte_at(mm, address, ptep, + set_huge_pte_at(mm, address, hpte.ptep, make_pte_marker(PTE_MARKER_UFFD_WP)); else - huge_pte_clear(mm, address, ptep, sz); + huge_pte_clear(mm, address, hpte.ptep, + hugetlb_pte_size(&hpte)); spin_unlock(ptl); - continue; + goto next_hpte; } - page = pte_page(pte); + subpage = pte_page(pte); + BUG_ON(!subpage); + hpage = compound_head(subpage); /* * If a reference page is supplied, it is because a specific * page is being unmapped, not a range. Ensure the page we * are about to unmap is the actual page of interest. */ if (ref_page) { - if (page != ref_page) { + if (hpage != ref_page) { spin_unlock(ptl); - continue; + goto next_hpte; } /* * Mark the VMA as having unmapped its page so that @@ -5207,25 +5224,35 @@ static void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct set_vma_resv_flags(vma, HPAGE_RESV_UNMAPPED); } - pte = huge_ptep_get_and_clear(mm, address, ptep); - tlb_remove_huge_tlb_entry(h, tlb, ptep, address); + pte = huge_ptep_get_and_clear(mm, address, hpte.ptep); + tlb_change_page_size(tlb, hugetlb_pte_size(&hpte)); + tlb_remove_huge_tlb_entry(tlb, hpte, address); if (huge_pte_dirty(pte)) - set_page_dirty(page); + set_page_dirty(hpage); /* Leave a uffd-wp pte marker if needed */ if (huge_pte_uffd_wp(pte) && !(zap_flags & ZAP_FLAG_DROP_MARKER)) - set_huge_pte_at(mm, address, ptep, + set_huge_pte_at(mm, address, hpte.ptep, make_pte_marker(PTE_MARKER_UFFD_WP)); - hugetlb_count_sub(pages_per_huge_page(h), mm); - page_remove_rmap(page, vma, true); + + hugetlb_count_sub(hugetlb_pte_size(&hpte)/PAGE_SIZE, mm); + + /* + * If we are unmapping the entire page, remove it from the + * rmap. + */ + if (IS_ALIGNED(address, sz) && address + sz <= end) + page_remove_rmap(hpage, vma, true); spin_unlock(ptl); - tlb_remove_page_size(tlb, page, huge_page_size(h)); + tlb_remove_page_size(tlb, subpage, hugetlb_pte_size(&hpte)); /* * Bail out after unmapping reference page if supplied */ if (ref_page) break; +next_hpte: + address += hugetlb_pte_size(&hpte); } mmu_notifier_invalidate_range_end(&range); tlb_end_vma(tlb, vma);