From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 2A83A7691C for ; Thu, 21 Dec 2023 22:29:33 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linux-foundation.org header.i=@linux-foundation.org header.b="P8mjjMpe" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 9C895C433C7; Thu, 21 Dec 2023 22:29:33 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linux-foundation.org; s=korg; t=1703197773; bh=prrIFgbdpfQuQa+nUy2cJ8iXxozctYzw5Bn2BaIPTIQ=; h=Date:To:From:Subject:From; b=P8mjjMpeF1p/ValL9fzOIDXIYcjWPyHismtD2DTF2Ibr9726gasXTUbP8+u+TiDVW 9OmD2L4hUEnXJqwV9Q+/bX0XM1lWcJPHEets7ICjv5Vp7EsuJzsCJa2Tbwyxqz7uxI 1pmh3Qy5VglbmJlAOJRI0EQN1g7/WLb8b1hMBjZk= Date: Thu, 21 Dec 2023 14:29:33 -0800 To: mm-commits@vger.kernel.org,willy@infradead.org,songmuchun@bytedance.com,ryan.roberts@arm.com,peterx@redhat.com,muchun.song@linux.dev,hughd@google.com,fengwei.yin@intel.com,david@redhat.com,akpm@linux-foundation.org From: Andrew Morton Subject: + mm-rmap-introduce-and-use-hugetlb_try_share_anon_rmap.patch added to mm-unstable branch Message-Id: <20231221222933.9C895C433C7@smtp.kernel.org> Precedence: bulk X-Mailing-List: mm-commits@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: The patch titled Subject: mm/rmap: introduce and use hugetlb_try_share_anon_rmap() has been added to the -mm mm-unstable branch. Its filename is mm-rmap-introduce-and-use-hugetlb_try_share_anon_rmap.patch This patch will shortly appear at https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-rmap-introduce-and-use-hugetlb_try_share_anon_rmap.patch This patch will later appear in the mm-unstable branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/process/submit-checklist.rst when testing your code *** The -mm tree is included into linux-next via the mm-everything branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm and is updated there every 2-3 working days ------------------------------------------------------ From: David Hildenbrand Subject: mm/rmap: introduce and use hugetlb_try_share_anon_rmap() Date: Wed, 20 Dec 2023 23:44:29 +0100 hugetlb rmap handling differs quite a lot from "ordinary" rmap code. For example, hugetlb currently only supports entire mappings, and treats any mapping as mapped using a single "logical PTE". Let's move it out of the way so we can overhaul our "ordinary" rmap. implementation/interface. So let's introduce and use hugetlb_try_dup_anon_rmap() to make all hugetlb handling use dedicated hugetlb_* rmap functions. Add sanity checks that we end up with the right folios in the right functions. Note that try_to_unmap_one() does not need care. Easy to spot because among all that nasty hugetlb special-casing in that function, we're not using set_huge_pte_at() on the anon path -- well, and that code assumes that we would want to swapout. Link: https://lkml.kernel.org/r/20231220224504.646757-6-david@redhat.com Signed-off-by: David Hildenbrand Reviewed-by: Yin Fengwei Reviewed-by: Ryan Roberts Cc: Hugh Dickins Cc: Matthew Wilcox (Oracle) Cc: Muchun Song Cc: Muchun Song Cc: Peter Xu Signed-off-by: Andrew Morton --- include/linux/rmap.h | 25 +++++++++++++++++++++++++ mm/rmap.c | 15 ++++++++++----- 2 files changed, 35 insertions(+), 5 deletions(-) --- a/include/linux/rmap.h~mm-rmap-introduce-and-use-hugetlb_try_share_anon_rmap +++ a/include/linux/rmap.h @@ -227,6 +227,30 @@ static inline int hugetlb_try_dup_anon_r return 0; } +/* See page_try_share_anon_rmap() */ +static inline int hugetlb_try_share_anon_rmap(struct folio *folio) +{ + VM_WARN_ON_FOLIO(!folio_test_hugetlb(folio), folio); + VM_WARN_ON_FOLIO(!folio_test_anon(folio), folio); + VM_WARN_ON_FOLIO(!PageAnonExclusive(&folio->page), folio); + + /* Paired with the memory barrier in try_grab_folio(). */ + if (IS_ENABLED(CONFIG_HAVE_FAST_GUP)) + smp_mb(); + + if (unlikely(folio_maybe_dma_pinned(folio))) + return -EBUSY; + ClearPageAnonExclusive(&folio->page); + + /* + * This is conceptually a smp_wmb() paired with the smp_rmb() in + * gup_must_unshare(). + */ + if (IS_ENABLED(CONFIG_HAVE_FAST_GUP)) + smp_mb__after_atomic(); + return 0; +} + static inline void hugetlb_add_file_rmap(struct folio *folio) { VM_WARN_ON_FOLIO(!folio_test_hugetlb(folio), folio); @@ -331,6 +355,7 @@ dup: */ static inline int page_try_share_anon_rmap(struct page *page) { + VM_WARN_ON(folio_test_hugetlb(page_folio(page))); VM_BUG_ON_PAGE(!PageAnon(page) || !PageAnonExclusive(page), page); /* device private pages cannot get pinned via GUP. */ --- a/mm/rmap.c~mm-rmap-introduce-and-use-hugetlb_try_share_anon_rmap +++ a/mm/rmap.c @@ -2149,13 +2149,18 @@ static bool try_to_migrate_one(struct fo !anon_exclusive, subpage); /* See page_try_share_anon_rmap(): clear PTE first. */ - if (anon_exclusive && - page_try_share_anon_rmap(subpage)) { - if (folio_test_hugetlb(folio)) + if (folio_test_hugetlb(folio)) { + if (anon_exclusive && + hugetlb_try_share_anon_rmap(folio)) { set_huge_pte_at(mm, address, pvmw.pte, pteval, hsz); - else - set_pte_at(mm, address, pvmw.pte, pteval); + ret = false; + page_vma_mapped_walk_done(&pvmw); + break; + } + } else if (anon_exclusive && + page_try_share_anon_rmap(subpage)) { + set_pte_at(mm, address, pvmw.pte, pteval); ret = false; page_vma_mapped_walk_done(&pvmw); break; _ Patches currently in -mm which might be from david@redhat.com are mm-rmap-rename-hugepage_add-to-hugetlb_add.patch mm-rmap-introduce-and-use-hugetlb_remove_rmap.patch mm-rmap-introduce-and-use-hugetlb_add_file_rmap.patch mm-rmap-introduce-and-use-hugetlb_try_dup_anon_rmap.patch mm-rmap-introduce-and-use-hugetlb_try_share_anon_rmap.patch mm-rmap-add-hugetlb-sanity-checks-for-anon-rmap-handling.patch mm-rmap-convert-folio_add_file_rmap_range-into-folio_add_file_rmap_.patch mm-memory-page_add_file_rmap-folio_add_file_rmap_.patch mm-huge_memory-page_add_file_rmap-folio_add_file_rmap_pmd.patch mm-migrate-page_add_file_rmap-folio_add_file_rmap_pte.patch mm-userfaultfd-page_add_file_rmap-folio_add_file_rmap_pte.patch mm-rmap-remove-page_add_file_rmap.patch mm-rmap-factor-out-adding-folio-mappings-into-__folio_add_rmap.patch mm-rmap-introduce-folio_add_anon_rmap_.patch mm-huge_memory-batch-rmap-operations-in-__split_huge_pmd_locked.patch mm-huge_memory-page_add_anon_rmap-folio_add_anon_rmap_pmd.patch mm-migrate-page_add_anon_rmap-folio_add_anon_rmap_pte.patch mm-ksm-page_add_anon_rmap-folio_add_anon_rmap_pte.patch mm-swapfile-page_add_anon_rmap-folio_add_anon_rmap_pte.patch mm-memory-page_add_anon_rmap-folio_add_anon_rmap_pte.patch mm-rmap-remove-page_add_anon_rmap.patch mm-rmap-remove-rmap_compound.patch mm-rmap-introduce-folio_remove_rmap_.patch kernel-events-uprobes-page_remove_rmap-folio_remove_rmap_pte.patch mm-huge_memory-page_remove_rmap-folio_remove_rmap_pmd.patch mm-khugepaged-page_remove_rmap-folio_remove_rmap_pte.patch mm-ksm-page_remove_rmap-folio_remove_rmap_pte.patch mm-memory-page_remove_rmap-folio_remove_rmap_pte.patch mm-migrate_device-page_remove_rmap-folio_remove_rmap_pte.patch mm-rmap-page_remove_rmap-folio_remove_rmap_pte.patch documentation-stop-referring-to-page_remove_rmap.patch mm-rmap-remove-page_remove_rmap.patch mm-rmap-convert-page_dup_file_rmap-to-folio_dup_file_rmap_.patch mm-rmap-introduce-folio_try_dup_anon_rmap_.patch mm-huge_memory-page_try_dup_anon_rmap-folio_try_dup_anon_rmap_pmd.patch mm-memory-page_try_dup_anon_rmap-folio_try_dup_anon_rmap_pte.patch mm-rmap-remove-page_try_dup_anon_rmap.patch mm-convert-page_try_share_anon_rmap-to-folio_try_share_anon_rmap_.patch mm-rmap-rename-compound_mapped-to-entirely_mapped.patch mm-remove-one-last-reference-to-page_add__rmap.patch