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=-9.8 required=3.0 tests=DKIMWL_WL_HIGH,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH, MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED, USER_AGENT_GIT autolearn=unavailable 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 A1BDEC33CB7 for ; Wed, 29 Jan 2020 03:24:56 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 5A3EA2070E for ; Wed, 29 Jan 2020 03:24:56 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=nvidia.com header.i=@nvidia.com header.b="fq26RLIw" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727124AbgA2DYs (ORCPT ); Tue, 28 Jan 2020 22:24:48 -0500 Received: from hqnvemgate25.nvidia.com ([216.228.121.64]:10761 "EHLO hqnvemgate25.nvidia.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726795AbgA2DYY (ORCPT ); Tue, 28 Jan 2020 22:24:24 -0500 Received: from hqpgpgate101.nvidia.com (Not Verified[216.228.121.13]) by hqnvemgate25.nvidia.com (using TLS: TLSv1.2, DES-CBC3-SHA) id ; Tue, 28 Jan 2020 19:24:00 -0800 Received: from hqmail.nvidia.com ([172.20.161.6]) by hqpgpgate101.nvidia.com (PGP Universal service); Tue, 28 Jan 2020 19:24:19 -0800 X-PGP-Universal: processed; by hqpgpgate101.nvidia.com on Tue, 28 Jan 2020 19:24:19 -0800 Received: from HQMAIL101.nvidia.com (172.20.187.10) by HQMAIL105.nvidia.com (172.20.187.12) with Microsoft SMTP Server (TLS) id 15.0.1473.3; Wed, 29 Jan 2020 03:24:19 +0000 Received: from hqnvemgw03.nvidia.com (10.124.88.68) by HQMAIL101.nvidia.com (172.20.187.10) with Microsoft SMTP Server (TLS) id 15.0.1473.3 via Frontend Transport; Wed, 29 Jan 2020 03:24:19 +0000 Received: from blueforge.nvidia.com (Not Verified[10.110.48.28]) by hqnvemgw03.nvidia.com with Trustwave SEG (v7,5,8,10121) id ; Tue, 28 Jan 2020 19:24:19 -0800 From: John Hubbard To: Andrew Morton CC: Al Viro , Christoph Hellwig , Dan Williams , Dave Chinner , Ira Weiny , Jan Kara , Jason Gunthorpe , Jonathan Corbet , =?UTF-8?q?J=C3=A9r=C3=B4me=20Glisse?= , "Kirill A . Shutemov" , Michal Hocko , Mike Kravetz , Shuah Khan , Vlastimil Babka , , , , , , LKML , John Hubbard , "Kirill A . Shutemov" Subject: [PATCH v2 4/8] mm/gup: track FOLL_PIN pages Date: Tue, 28 Jan 2020 19:24:13 -0800 Message-ID: <20200129032417.3085670-5-jhubbard@nvidia.com> X-Mailer: git-send-email 2.25.0 In-Reply-To: <20200129032417.3085670-1-jhubbard@nvidia.com> References: <20200129032417.3085670-1-jhubbard@nvidia.com> MIME-Version: 1.0 X-NVConfidentiality: public Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=nvidia.com; s=n1; t=1580268240; bh=ykfRWL1tp2o86eUnwxD/ZzxUI7KWe4SzkrMuGgn3euI=; h=X-PGP-Universal:From:To:CC:Subject:Date:Message-ID:X-Mailer: In-Reply-To:References:MIME-Version:X-NVConfidentiality: Content-Type:Content-Transfer-Encoding; b=fq26RLIwLRb6L0sAz0iVRWCW0eA54wM0YY+bcZ1rFLQsWz9eymlZzDEsyXKZBtzMC gTvwItEY50C1X3R3HAZvGuGuMFc9q0B60fQ3QbfcQPWtan53/cJho3yC0RfSjJcGX5 k022E0w6zBUYaACqUbeWXtznPARj/yIDyojjt4eA1ge1khxBjnoyCjMkWwOBQtsghK 6RuhQogS9cqG6A9/Rga2I901rbFZxUjY/d4dk/y1kEzHHU5BmRLPvn4MmTpm9oeXSB lLL8kCdorcbo8gBvYkLRrOVDiuoxwJSd8nZJaXmT/LCtYUnzSAJVteHh82L4B3OYeE W5KVkW5Vrol0g== Sender: linux-kselftest-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kselftest@vger.kernel.org Add tracking of pages that were pinned via FOLL_PIN. This tracking is implemented via overloading of page->_refcount: pins are added by adding GUP_PIN_COUNTING_BIAS (1024) to the refcount. This provides a fuzzy indication of pinning, and it can have false positives (and that's OK). Please see the pre-existing Documentation/core-api/pin_user_pages.rst for details. As mentioned in pin_user_pages.rst, callers who effectively set FOLL_PIN (typically via pin_user_pages*()) are required to ultimately free such pages via unpin_user_page(). Please also not the limitation, discussed in pin_user_pages.rst under the "TODO: for 1GB and larger huge pages" section. (That limitation will be removed in a following patch.) The effect of a FOLL_PIN flag is similar to that of FOLL_GET, and may be thought of as "FOLL_GET for DIO and/or RDMA use". Pages that have been pinned via FOLL_PIN are identifiable via a new function call: bool page_dma_pinned(struct page *page); What to do in response to encountering such a page, is left to later patchsets. There is discussion about this in [1], [2], [3], and [4]. This also changes a BUG_ON(), to a WARN_ON(), in follow_page_mask(). [1] Some slow progress on get_user_pages() (Apr 2, 2019): https://lwn.net/Articles/784574/ [2] DMA and get_user_pages() (LPC: Dec 12, 2018): https://lwn.net/Articles/774411/ [3] The trouble with get_user_pages() (Apr 30, 2018): https://lwn.net/Articles/753027/ [4] LWN kernel index: get_user_pages(): https://lwn.net/Kernel/Index/#Memory_management-get_user_pages Suggested-by: Jan Kara Suggested-by: J=C3=A9r=C3=B4me Glisse Cc: Kirill A. Shutemov Signed-off-by: John Hubbard --- include/linux/mm.h | 83 +++++++++-- include/linux/page_ref.h | 10 ++ mm/gup.c | 293 +++++++++++++++++++++++++++++++-------- mm/huge_memory.c | 29 +++- mm/hugetlb.c | 38 ++--- 5 files changed, 354 insertions(+), 99 deletions(-) diff --git a/include/linux/mm.h b/include/linux/mm.h index 080f8ac8bfb7..c5d0f4a66788 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1001,6 +1001,8 @@ static inline void get_page(struct page *page) page_ref_inc(page); } =20 +bool __must_check try_grab_page(struct page *page, unsigned int flags); + static inline __must_check bool try_get_page(struct page *page) { page =3D compound_head(page); @@ -1029,29 +1031,80 @@ static inline void put_page(struct page *page) __put_page(page); } =20 -/** - * unpin_user_page() - release a gup-pinned page - * @page: pointer to page to be released +/* + * GUP_PIN_COUNTING_BIAS, and the associated functions that use it, overlo= ad + * the page's refcount so that two separate items are tracked: the origina= l page + * reference count, and also a new count of how many pin_user_pages() call= s were + * made against the page. ("gup-pinned" is another term for the latter). + * + * With this scheme, pin_user_pages() becomes special: such pages are mark= ed as + * distinct from normal pages. As such, the unpin_user_page() call (and it= s + * variants) must be used in order to release gup-pinned pages. + * + * Choice of value: + * + * By making GUP_PIN_COUNTING_BIAS a power of two, debugging of page refer= ence + * counts with respect to pin_user_pages() and unpin_user_page() becomes + * simpler, due to the fact that adding an even power of two to the page + * refcount has the effect of using only the upper N bits, for the code th= at + * counts up using the bias value. This means that the lower bits are left= for + * the exclusive use of the original code that increments and decrements b= y one + * (or at least, by much smaller values than the bias value). * - * Pages that were pinned via pin_user_pages*() must be released via eithe= r - * unpin_user_page(), or one of the unpin_user_pages*() routines. This is = so - * that eventually such pages can be separately tracked and uniquely handl= ed. In - * particular, interactions with RDMA and filesystems need special handlin= g. + * Of course, once the lower bits overflow into the upper bits (and this i= s + * OK, because subtraction recovers the original values), then visual insp= ection + * no longer suffices to directly view the separate counts. However, for n= ormal + * applications that don't have huge page reference counts, this won't be = an + * issue. * - * unpin_user_page() and put_page() are not interchangeable, despite this = early - * implementation that makes them look the same. unpin_user_page() calls m= ust - * be perfectly matched up with pin*() calls. + * Locking: the lockless algorithm described in page_cache_get_speculative= () + * and page_cache_gup_pin_speculative() provides safe operation for + * get_user_pages and page_mkclean and other calls that race to set up pag= e + * table entries. */ -static inline void unpin_user_page(struct page *page) -{ - put_page(page); -} +#define GUP_PIN_COUNTING_BIAS (1U << 10) =20 +void unpin_user_page(struct page *page); void unpin_user_pages_dirty_lock(struct page **pages, unsigned long npages= , bool make_dirty); - void unpin_user_pages(struct page **pages, unsigned long npages); =20 +/** + * page_dma_pinned() - report if a page is pinned for DMA. + * + * This function checks if a page has been pinned via a call to + * pin_user_pages*(). + * + * For non-huge pages, the return value is partially fuzzy: false is not f= uzzy, + * because it means "definitely not pinned for DMA", but true means "proba= bly + * pinned for DMA, but possibly a false positive due to having at least + * GUP_PIN_COUNTING_BIAS worth of normal page references". + * + * False positives are OK, because: a) it's unlikely for a page to get tha= t many + * refcounts, and b) all the callers of this routine are expected to be ab= le to + * deal gracefully with a false positive. + * + * For more information, please see Documentation/vm/pin_user_pages.rst. + * + * @page: pointer to page to be queried. + * @Return: True, if it is likely that the page has been "dma-pinned". + * False, if the page is definitely not dma-pinned. + */ +static inline bool page_dma_pinned(struct page *page) +{ + /* + * page_ref_count() is signed. If that refcount overflows, then + * page_ref_count() returns a negative value, and callers will avoid + * further incrementing the refcount. + * + * Here, for that overflow case, use the signed bit to count a little + * bit higher via unsigned math, and thus still get an accurate result + * from page_dma_pinned(). + */ + return ((unsigned int)page_ref_count(compound_head(page))) >=3D + GUP_PIN_COUNTING_BIAS; +} + #if defined(CONFIG_SPARSEMEM) && !defined(CONFIG_SPARSEMEM_VMEMMAP) #define SECTION_IN_PAGE_FLAGS #endif diff --git a/include/linux/page_ref.h b/include/linux/page_ref.h index 14d14beb1f7f..b9cbe553d1e7 100644 --- a/include/linux/page_ref.h +++ b/include/linux/page_ref.h @@ -102,6 +102,16 @@ static inline void page_ref_sub(struct page *page, int= nr) __page_ref_mod(page, -nr); } =20 +static inline int page_ref_sub_return(struct page *page, int nr) +{ + int ret =3D atomic_sub_return(nr, &page->_refcount); + + if (page_ref_tracepoint_active(__tracepoint_page_ref_mod)) + __page_ref_mod(page, -nr); + + return ret; +} + static inline void page_ref_inc(struct page *page) { atomic_inc(&page->_refcount); diff --git a/mm/gup.c b/mm/gup.c index 9e117998274c..7a96490dcc54 100644 --- a/mm/gup.c +++ b/mm/gup.c @@ -44,6 +44,136 @@ static inline struct page *try_get_compound_head(struct= page *page, int refs) return head; } =20 +/* + * try_grab_compound_head() - attempt to elevate a page's refcount, by a + * flags-dependent amount. + * + * "grab" names in this file mean, "look at flags to decide whether to use + * FOLL_PIN or FOLL_GET behavior, when incrementing the page's refcount. + * + * Either FOLL_PIN or FOLL_GET (or neither) must be set, but not both at t= he + * same time. (That's true throughout the get_user_pages*() and + * pin_user_pages*() APIs.) Cases: + * + * FOLL_GET: page's refcount will be incremented by 1. + * FOLL_PIN: page's refcount will be incremented by GUP_PIN_COUNTING_BI= AS. + * + * Return: head page (with refcount appropriately incremented) for success= , or + * NULL upon failure. If neither FOLL_GET nor FOLL_PIN was set, that's + * considered failure, and furthermore, a likely bug in the caller, so a w= arning + * is also emitted. + */ +static __maybe_unused struct page *try_grab_compound_head(struct page *pag= e, + int refs, + unsigned int flags) +{ + if (flags & FOLL_GET) + return try_get_compound_head(page, refs); + else if (flags & FOLL_PIN) { + refs *=3D GUP_PIN_COUNTING_BIAS; + return try_get_compound_head(page, refs); + } + + WARN_ON_ONCE(1); + return NULL; +} + +/** + * try_grab_page() - elevate a page's refcount by a flag-dependent amount + * + * This might not do anything at all, depending on the flags argument. + * + * "grab" names in this file mean, "look at flags to decide whether to use + * FOLL_PIN or FOLL_GET behavior, when incrementing the page's refcount. + * + * @page: pointer to page to be grabbed + * @flags: gup flags: these are the FOLL_* flag values. + * + * Either FOLL_PIN or FOLL_GET (or neither) may be set, but not both at th= e same + * time. Cases: + * + * FOLL_GET: page's refcount will be incremented by 1. + * FOLL_PIN: page's refcount will be incremented by GUP_PIN_COUNTING_BI= AS. + * + * Return: true for success, or if no action was required (if neither FOLL= _PIN + * nor FOLL_GET was set, nothing is done). False for failure: FOLL_GET or + * FOLL_PIN was set, but the page could not be grabbed. + */ +bool __must_check try_grab_page(struct page *page, unsigned int flags) +{ + if (flags & FOLL_GET) + return try_get_page(page); + else if (flags & FOLL_PIN) { + page =3D compound_head(page); + + if (WARN_ON_ONCE(flags & FOLL_GET)) + return false; + + if (WARN_ON_ONCE(page_ref_count(page) <=3D 0)) + return false; + + page_ref_add(page, GUP_PIN_COUNTING_BIAS); + } + + return true; +} + +#ifdef CONFIG_DEV_PAGEMAP_OPS +static bool __unpin_devmap_managed_user_page(struct page *page) +{ + int count; + + if (!page_is_devmap_managed(page)) + return false; + + count =3D page_ref_sub_return(page, GUP_PIN_COUNTING_BIAS); + + /* + * devmap page refcounts are 1-based, rather than 0-based: if + * refcount is 1, then the page is free and the refcount is + * stable because nobody holds a reference on the page. + */ + if (count =3D=3D 1) + free_devmap_managed_page(page); + else if (!count) + __put_page(page); + + return true; +} +#else +static bool __unpin_devmap_managed_user_page(struct page *page) +{ + return false; +} +#endif /* CONFIG_DEV_PAGEMAP_OPS */ + +/** + * unpin_user_page() - release a dma-pinned page + * @page: pointer to page to be released + * + * Pages that were pinned via pin_user_pages*() must be released via eithe= r + * unpin_user_page(), or one of the unpin_user_pages*() routines. This is = so + * that such pages can be separately tracked and uniquely handled. In + * particular, interactions with RDMA and filesystems need special handlin= g. + */ +void unpin_user_page(struct page *page) +{ + page =3D compound_head(page); + + /* + * For devmap managed pages we need to catch refcount transition from + * GUP_PIN_COUNTING_BIAS to 1, when refcount reach one it means the + * page is free and we need to inform the device driver through + * callback. See include/linux/memremap.h and HMM for details. + */ + if (__unpin_devmap_managed_user_page(page)) + return; + + if (page_ref_sub_and_test(page, GUP_PIN_COUNTING_BIAS)) + __put_page(page); +} +EXPORT_SYMBOL(unpin_user_page); + /** * unpin_user_pages_dirty_lock() - release and optionally dirty gup-pinned= pages * @pages: array of pages to be maybe marked dirty, and definitely releas= ed. @@ -230,10 +360,11 @@ static struct page *follow_page_pte(struct vm_area_st= ruct *vma, } =20 page =3D vm_normal_page(vma, address, pte); - if (!page && pte_devmap(pte) && (flags & FOLL_GET)) { + if (!page && pte_devmap(pte) && (flags & (FOLL_GET | FOLL_PIN))) { /* - * Only return device mapping pages in the FOLL_GET case since - * they are only valid while holding the pgmap reference. + * Only return device mapping pages in the FOLL_GET or FOLL_PIN + * case since they are only valid while holding the pgmap + * reference. */ *pgmap =3D get_dev_pagemap(pte_pfn(pte), *pgmap); if (*pgmap) @@ -271,11 +402,10 @@ static struct page *follow_page_pte(struct vm_area_st= ruct *vma, goto retry; } =20 - if (flags & FOLL_GET) { - if (unlikely(!try_get_page(page))) { - page =3D ERR_PTR(-ENOMEM); - goto out; - } + /* try_grab_page() does nothing unless FOLL_GET or FOLL_PIN is set. */ + if (unlikely(!try_grab_page(page, flags))) { + page =3D ERR_PTR(-ENOMEM); + goto out; } if (flags & FOLL_TOUCH) { if ((flags & FOLL_WRITE) && @@ -537,7 +667,7 @@ static struct page *follow_page_mask(struct vm_area_str= uct *vma, /* make this handle hugepd */ page =3D follow_huge_addr(mm, address, flags & FOLL_WRITE); if (!IS_ERR(page)) { - BUG_ON(flags & FOLL_GET); + WARN_ON_ONCE(flags & (FOLL_GET | FOLL_PIN)); return page; } =20 @@ -1675,6 +1805,15 @@ long get_user_pages_remote(struct task_struct *tsk, = struct mm_struct *mm, { return 0; } + +static long __get_user_pages_remote(struct task_struct *tsk, + struct mm_struct *mm, + unsigned long start, unsigned long nr_pages, + unsigned int gup_flags, struct page **pages, + struct vm_area_struct **vmas, int *locked) +{ + return 0; +} #endif /* !CONFIG_MMU */ =20 /* @@ -1870,13 +2009,17 @@ static inline pte_t gup_get_pte(pte_t *ptep) #endif /* CONFIG_GUP_GET_PTE_LOW_HIGH */ =20 static void __maybe_unused undo_dev_pagemap(int *nr, int nr_start, + unsigned int flags, struct page **pages) { while ((*nr) - nr_start) { struct page *page =3D pages[--(*nr)]; =20 ClearPageReferenced(page); - put_page(page); + if (flags & FOLL_PIN) + unpin_user_page(page); + else + put_page(page); } } =20 @@ -1909,7 +2052,7 @@ static int gup_pte_range(pmd_t pmd, unsigned long add= r, unsigned long end, =20 pgmap =3D get_dev_pagemap(pte_pfn(pte), pgmap); if (unlikely(!pgmap)) { - undo_dev_pagemap(nr, nr_start, pages); + undo_dev_pagemap(nr, nr_start, flags, pages); goto pte_unmap; } } else if (pte_special(pte)) @@ -1918,7 +2061,7 @@ static int gup_pte_range(pmd_t pmd, unsigned long add= r, unsigned long end, VM_BUG_ON(!pfn_valid(pte_pfn(pte))); page =3D pte_page(pte); =20 - head =3D try_get_compound_head(page, 1); + head =3D try_grab_compound_head(page, 1, flags); if (!head) goto pte_unmap; =20 @@ -1974,12 +2117,15 @@ static int __gup_device_huge(unsigned long pfn, uns= igned long addr, =20 pgmap =3D get_dev_pagemap(pfn, pgmap); if (unlikely(!pgmap)) { - undo_dev_pagemap(nr, nr_start, pages); + undo_dev_pagemap(nr, nr_start, flags, pages); return 0; } SetPageReferenced(page); pages[*nr] =3D page; - get_page(page); + if (unlikely(!try_grab_page(page, flags))) { + undo_dev_pagemap(nr, nr_start, flags, pages); + return 0; + } (*nr)++; pfn++; } while (addr +=3D PAGE_SIZE, addr !=3D end); @@ -2001,7 +2147,7 @@ static int __gup_device_huge_pmd(pmd_t orig, pmd_t *p= mdp, unsigned long addr, return 0; =20 if (unlikely(pmd_val(orig) !=3D pmd_val(*pmdp))) { - undo_dev_pagemap(nr, nr_start, pages); + undo_dev_pagemap(nr, nr_start, flags, pages); return 0; } return 1; @@ -2019,7 +2165,7 @@ static int __gup_device_huge_pud(pud_t orig, pud_t *p= udp, unsigned long addr, return 0; =20 if (unlikely(pud_val(orig) !=3D pud_val(*pudp))) { - undo_dev_pagemap(nr, nr_start, pages); + undo_dev_pagemap(nr, nr_start, flags, pages); return 0; } return 1; @@ -2053,8 +2199,11 @@ static int record_subpages(struct page *page, unsign= ed long addr, return nr; } =20 -static void put_compound_head(struct page *page, int refs) +static void put_compound_head(struct page *page, int refs, unsigned int fl= ags) { + if (flags & FOLL_PIN) + refs *=3D GUP_PIN_COUNTING_BIAS; + VM_BUG_ON_PAGE(page_ref_count(page) < refs, page); /* * Calling put_page() for each ref is unnecessarily slow. Only the last @@ -2098,12 +2247,12 @@ static int gup_hugepte(pte_t *ptep, unsigned long s= z, unsigned long addr, page =3D head + ((addr & (sz-1)) >> PAGE_SHIFT); refs =3D record_subpages(page, addr, end, pages + *nr); =20 - head =3D try_get_compound_head(head, refs); + head =3D try_grab_compound_head(head, refs, flags); if (!head) return 0; =20 if (unlikely(pte_val(pte) !=3D pte_val(*ptep))) { - put_compound_head(head, refs); + put_compound_head(head, refs, flags); return 0; } =20 @@ -2158,12 +2307,12 @@ static int gup_huge_pmd(pmd_t orig, pmd_t *pmdp, un= signed long addr, page =3D pmd_page(orig) + ((addr & ~PMD_MASK) >> PAGE_SHIFT); refs =3D record_subpages(page, addr, end, pages + *nr); =20 - head =3D try_get_compound_head(pmd_page(orig), refs); + head =3D try_grab_compound_head(pmd_page(orig), refs, flags); if (!head) return 0; =20 if (unlikely(pmd_val(orig) !=3D pmd_val(*pmdp))) { - put_compound_head(head, refs); + put_compound_head(head, refs, flags); return 0; } =20 @@ -2192,12 +2341,12 @@ static int gup_huge_pud(pud_t orig, pud_t *pudp, un= signed long addr, page =3D pud_page(orig) + ((addr & ~PUD_MASK) >> PAGE_SHIFT); refs =3D record_subpages(page, addr, end, pages + *nr); =20 - head =3D try_get_compound_head(pud_page(orig), refs); + head =3D try_grab_compound_head(pud_page(orig), refs, flags); if (!head) return 0; =20 if (unlikely(pud_val(orig) !=3D pud_val(*pudp))) { - put_compound_head(head, refs); + put_compound_head(head, refs, flags); return 0; } =20 @@ -2221,12 +2370,12 @@ static int gup_huge_pgd(pgd_t orig, pgd_t *pgdp, un= signed long addr, page =3D pgd_page(orig) + ((addr & ~PGDIR_MASK) >> PAGE_SHIFT); refs =3D record_subpages(page, addr, end, pages + *nr); =20 - head =3D try_get_compound_head(pgd_page(orig), refs); + head =3D try_grab_compound_head(pgd_page(orig), refs, flags); if (!head) return 0; =20 if (unlikely(pgd_val(orig) !=3D pgd_val(*pgdp))) { - put_compound_head(head, refs); + put_compound_head(head, refs, flags); return 0; } =20 @@ -2389,6 +2538,14 @@ int __get_user_pages_fast(unsigned long start, int n= r_pages, int write, unsigned long len, end; unsigned long flags; int nr =3D 0; + /* + * Internally (within mm/gup.c), gup fast variants must set FOLL_GET, + * because gup fast is always a "pin with a +1 page refcount" request. + */ + unsigned int gup_flags =3D FOLL_GET; + + if (write) + gup_flags |=3D FOLL_WRITE; =20 start =3D untagged_addr(start) & PAGE_MASK; len =3D (unsigned long) nr_pages << PAGE_SHIFT; @@ -2414,7 +2571,7 @@ int __get_user_pages_fast(unsigned long start, int nr= _pages, int write, if (IS_ENABLED(CONFIG_HAVE_FAST_GUP) && gup_fast_permitted(start, end)) { local_irq_save(flags); - gup_pgd_range(start, end, write ? FOLL_WRITE : 0, pages, &nr); + gup_pgd_range(start, end, gup_flags, pages, &nr); local_irq_restore(flags); } =20 @@ -2453,7 +2610,7 @@ static int internal_get_user_pages_fast(unsigned long= start, int nr_pages, int nr =3D 0, ret =3D 0; =20 if (WARN_ON_ONCE(gup_flags & ~(FOLL_WRITE | FOLL_LONGTERM | - FOLL_FORCE | FOLL_PIN))) + FOLL_FORCE | FOLL_PIN | FOLL_GET))) return -EINVAL; =20 start =3D untagged_addr(start) & PAGE_MASK; @@ -2496,11 +2653,11 @@ static int internal_get_user_pages_fast(unsigned lo= ng start, int nr_pages, =20 /** * get_user_pages_fast() - pin user pages in memory - * @start: starting user address - * @nr_pages: number of pages from start to pin - * @gup_flags: flags modifying pin behaviour - * @pages: array that receives pointers to the pages pinned. - * Should be at least nr_pages long. + * @start: starting user address + * @nr_pages: number of pages from start to pin + * @gup_flags: flags modifying pin behaviour + * @pages: array that receives pointers to the pages pinned. + * Should be at least nr_pages long. * * Attempt to pin user pages in memory without taking mm->mmap_sem. * If not successful, it will fall back to taking the lock and @@ -2520,6 +2677,13 @@ int get_user_pages_fast(unsigned long start, int nr_= pages, if (WARN_ON_ONCE(gup_flags & FOLL_PIN)) return -EINVAL; =20 + /* + * The caller may or may not have explicitly set FOLL_GET; either way is + * OK. However, internally (within mm/gup.c), gup fast variants must set + * FOLL_GET, because gup fast is always a "pin with a +1 page refcount" + * request. + */ + gup_flags |=3D FOLL_GET; return internal_get_user_pages_fast(start, nr_pages, gup_flags, pages); } EXPORT_SYMBOL_GPL(get_user_pages_fast); @@ -2527,9 +2691,12 @@ EXPORT_SYMBOL_GPL(get_user_pages_fast); /** * pin_user_pages_fast() - pin user pages in memory without taking locks * - * For now, this is a placeholder function, until various call sites are - * converted to use the correct get_user_pages*() or pin_user_pages*() API= . So, - * this is identical to get_user_pages_fast(). + * Nearly the same as get_user_pages_fast(), except that FOLL_PIN is set. = See + * get_user_pages_fast() for documentation on the function arguments, beca= use + * the arguments here are identical. + * + * FOLL_PIN means that the pages must be released via unpin_user_page(). P= lease + * see Documentation/vm/pin_user_pages.rst for further details. * * This is intended for Case 1 (DIO) in Documentation/vm/pin_user_pages.rs= t. It * is NOT intended for Case 2 (RDMA: long-term pins). @@ -2537,21 +2704,24 @@ EXPORT_SYMBOL_GPL(get_user_pages_fast); int pin_user_pages_fast(unsigned long start, int nr_pages, unsigned int gup_flags, struct page **pages) { - /* - * This is a placeholder, until the pin functionality is activated. - * Until then, just behave like the corresponding get_user_pages*() - * routine. - */ - return get_user_pages_fast(start, nr_pages, gup_flags, pages); + /* FOLL_GET and FOLL_PIN are mutually exclusive. */ + if (WARN_ON_ONCE(gup_flags & FOLL_GET)) + return -EINVAL; + + gup_flags |=3D FOLL_PIN; + return internal_get_user_pages_fast(start, nr_pages, gup_flags, pages); } EXPORT_SYMBOL_GPL(pin_user_pages_fast); =20 /** * pin_user_pages_remote() - pin pages of a remote process (task !=3D curr= ent) * - * For now, this is a placeholder function, until various call sites are - * converted to use the correct get_user_pages*() or pin_user_pages*() API= . So, - * this is identical to get_user_pages_remote(). + * Nearly the same as get_user_pages_remote(), except that FOLL_PIN is set= . See + * get_user_pages_remote() for documentation on the function arguments, be= cause + * the arguments here are identical. + * + * FOLL_PIN means that the pages must be released via unpin_user_page(). P= lease + * see Documentation/vm/pin_user_pages.rst for details. * * This is intended for Case 1 (DIO) in Documentation/vm/pin_user_pages.rs= t. It * is NOT intended for Case 2 (RDMA: long-term pins). @@ -2561,22 +2731,24 @@ long pin_user_pages_remote(struct task_struct *tsk,= struct mm_struct *mm, unsigned int gup_flags, struct page **pages, struct vm_area_struct **vmas, int *locked) { - /* - * This is a placeholder, until the pin functionality is activated. - * Until then, just behave like the corresponding get_user_pages*() - * routine. - */ - return get_user_pages_remote(tsk, mm, start, nr_pages, gup_flags, pages, - vmas, locked); + /* FOLL_GET and FOLL_PIN are mutually exclusive. */ + if (WARN_ON_ONCE(gup_flags & FOLL_GET)) + return -EINVAL; + + gup_flags |=3D FOLL_PIN; + return __get_user_pages_remote(tsk, mm, start, nr_pages, gup_flags, + pages, vmas, locked); } EXPORT_SYMBOL(pin_user_pages_remote); =20 /** * pin_user_pages() - pin user pages in memory for use by other devices * - * For now, this is a placeholder function, until various call sites are - * converted to use the correct get_user_pages*() or pin_user_pages*() API= . So, - * this is identical to get_user_pages(). + * Nearly the same as get_user_pages(), except that FOLL_TOUCH is not set,= and + * FOLL_PIN is set. + * + * FOLL_PIN means that the pages must be released via unpin_user_page(). P= lease + * see Documentation/vm/pin_user_pages.rst for details. * * This is intended for Case 1 (DIO) in Documentation/vm/pin_user_pages.rs= t. It * is NOT intended for Case 2 (RDMA: long-term pins). @@ -2585,11 +2757,12 @@ long pin_user_pages(unsigned long start, unsigned l= ong nr_pages, unsigned int gup_flags, struct page **pages, struct vm_area_struct **vmas) { - /* - * This is a placeholder, until the pin functionality is activated. - * Until then, just behave like the corresponding get_user_pages*() - * routine. - */ - return get_user_pages(start, nr_pages, gup_flags, pages, vmas); + /* FOLL_GET and FOLL_PIN are mutually exclusive. */ + if (WARN_ON_ONCE(gup_flags & FOLL_GET)) + return -EINVAL; + + gup_flags |=3D FOLL_PIN; + return __gup_longterm_locked(current, current->mm, start, nr_pages, + pages, vmas, gup_flags); } EXPORT_SYMBOL(pin_user_pages); diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 0a55dec68925..b1079aaa6f24 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -958,6 +958,11 @@ struct page *follow_devmap_pmd(struct vm_area_struct *= vma, unsigned long addr, */ WARN_ONCE(flags & FOLL_COW, "mm: In follow_devmap_pmd with FOLL_COW set")= ; =20 + /* FOLL_GET and FOLL_PIN are mutually exclusive. */ + if (WARN_ON_ONCE((flags & (FOLL_PIN | FOLL_GET)) =3D=3D + (FOLL_PIN | FOLL_GET))) + return NULL; + if (flags & FOLL_WRITE && !pmd_write(*pmd)) return NULL; =20 @@ -973,7 +978,7 @@ struct page *follow_devmap_pmd(struct vm_area_struct *v= ma, unsigned long addr, * device mapped pages can only be returned if the * caller will manage the page reference count. */ - if (!(flags & FOLL_GET)) + if (!(flags & (FOLL_GET | FOLL_PIN))) return ERR_PTR(-EEXIST); =20 pfn +=3D (addr & ~PMD_MASK) >> PAGE_SHIFT; @@ -981,7 +986,8 @@ struct page *follow_devmap_pmd(struct vm_area_struct *v= ma, unsigned long addr, if (!*pgmap) return ERR_PTR(-EFAULT); page =3D pfn_to_page(pfn); - get_page(page); + if (!try_grab_page(page, flags)) + page =3D ERR_PTR(-ENOMEM); =20 return page; } @@ -1101,6 +1107,11 @@ struct page *follow_devmap_pud(struct vm_area_struct= *vma, unsigned long addr, if (flags & FOLL_WRITE && !pud_write(*pud)) return NULL; =20 + /* FOLL_GET and FOLL_PIN are mutually exclusive. */ + if (WARN_ON_ONCE((flags & (FOLL_PIN | FOLL_GET)) =3D=3D + (FOLL_PIN | FOLL_GET))) + return NULL; + if (pud_present(*pud) && pud_devmap(*pud)) /* pass */; else @@ -1112,8 +1123,10 @@ struct page *follow_devmap_pud(struct vm_area_struct= *vma, unsigned long addr, /* * device mapped pages can only be returned if the * caller will manage the page reference count. + * + * At least one of FOLL_GET | FOLL_PIN must be set, so assert that here: */ - if (!(flags & FOLL_GET)) + if (!(flags & (FOLL_GET | FOLL_PIN))) return ERR_PTR(-EEXIST); =20 pfn +=3D (addr & ~PUD_MASK) >> PAGE_SHIFT; @@ -1121,7 +1134,8 @@ struct page *follow_devmap_pud(struct vm_area_struct = *vma, unsigned long addr, if (!*pgmap) return ERR_PTR(-EFAULT); page =3D pfn_to_page(pfn); - get_page(page); + if (!try_grab_page(page, flags)) + page =3D ERR_PTR(-ENOMEM); =20 return page; } @@ -1497,8 +1511,13 @@ struct page *follow_trans_huge_pmd(struct vm_area_st= ruct *vma, =20 page =3D pmd_page(*pmd); VM_BUG_ON_PAGE(!PageHead(page) && !is_zone_device_page(page), page); + + if (!try_grab_page(page, flags)) + return ERR_PTR(-ENOMEM); + if (flags & FOLL_TOUCH) touch_pmd(vma, addr, pmd, flags); + if ((flags & FOLL_MLOCK) && (vma->vm_flags & VM_LOCKED)) { /* * We don't mlock() pte-mapped THPs. This way we can avoid @@ -1535,8 +1554,6 @@ struct page *follow_trans_huge_pmd(struct vm_area_str= uct *vma, skip_mlock: page +=3D (addr & ~HPAGE_PMD_MASK) >> PAGE_SHIFT; VM_BUG_ON_PAGE(!PageCompound(page) && !is_zone_device_page(page), page); - if (flags & FOLL_GET) - get_page(page); =20 out: return page; diff --git a/mm/hugetlb.c b/mm/hugetlb.c index dd8737a94bec..487e998fd38e 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -4375,19 +4375,6 @@ long follow_hugetlb_page(struct mm_struct *mm, struc= t vm_area_struct *vma, pfn_offset =3D (vaddr & ~huge_page_mask(h)) >> PAGE_SHIFT; page =3D pte_page(huge_ptep_get(pte)); =20 - /* - * Instead of doing 'try_get_page()' below in the same_page - * loop, just check the count once here. - */ - if (unlikely(page_count(page) <=3D 0)) { - if (pages) { - spin_unlock(ptl); - remainder =3D 0; - err =3D -ENOMEM; - break; - } - } - /* * If subpage information not requested, update counters * and skip the same_page loop below. @@ -4405,7 +4392,13 @@ long follow_hugetlb_page(struct mm_struct *mm, struc= t vm_area_struct *vma, same_page: if (pages) { pages[i] =3D mem_map_offset(page, pfn_offset); - get_page(pages[i]); + if (!try_grab_page(pages[i], flags)) { + spin_unlock(ptl); + remainder =3D 0; + err =3D -ENOMEM; + WARN_ON_ONCE(1); + break; + } } =20 if (vmas) @@ -4965,6 +4958,12 @@ follow_huge_pmd(struct mm_struct *mm, unsigned long = address, struct page *page =3D NULL; spinlock_t *ptl; pte_t pte; + + /* FOLL_GET and FOLL_PIN are mutually exclusive. */ + if (WARN_ON_ONCE((flags & (FOLL_PIN | FOLL_GET)) =3D=3D + (FOLL_PIN | FOLL_GET))) + return NULL; + retry: ptl =3D pmd_lockptr(mm, pmd); spin_lock(ptl); @@ -4977,8 +4976,11 @@ follow_huge_pmd(struct mm_struct *mm, unsigned long = address, pte =3D huge_ptep_get((pte_t *)pmd); if (pte_present(pte)) { page =3D pmd_page(*pmd) + ((address & ~PMD_MASK) >> PAGE_SHIFT); - if (flags & FOLL_GET) - get_page(page); + if (unlikely(!try_grab_page(page, flags))) { + WARN_ON_ONCE(1); + page =3D NULL; + goto out; + } } else { if (is_hugetlb_entry_migration(pte)) { spin_unlock(ptl); @@ -4999,7 +5001,7 @@ struct page * __weak follow_huge_pud(struct mm_struct *mm, unsigned long address, pud_t *pud, int flags) { - if (flags & FOLL_GET) + if (flags & (FOLL_GET | FOLL_PIN)) return NULL; =20 return pte_page(*(pte_t *)pud) + ((address & ~PUD_MASK) >> PAGE_SHIFT); @@ -5008,7 +5010,7 @@ follow_huge_pud(struct mm_struct *mm, unsigned long a= ddress, struct page * __weak follow_huge_pgd(struct mm_struct *mm, unsigned long address, pgd_t *pgd, i= nt flags) { - if (flags & FOLL_GET) + if (flags & (FOLL_GET | FOLL_PIN)) return NULL; =20 return pte_page(*(pte_t *)pgd) + ((address & ~PGDIR_MASK) >> PAGE_SHIFT); --=20 2.25.0