From patchwork Mon Sep 25 08:24:02 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Hugh Dickins X-Patchwork-Id: 13397479 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 897BECE7A8C for ; Mon, 25 Sep 2023 08:24:09 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 095BE8D0010; Mon, 25 Sep 2023 04:24:09 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 046588D0001; Mon, 25 Sep 2023 04:24:08 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id E29138D0010; Mon, 25 Sep 2023 04:24:08 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0016.hostedemail.com [216.40.44.16]) by kanga.kvack.org (Postfix) with ESMTP id CE08B8D0001 for ; Mon, 25 Sep 2023 04:24:08 -0400 (EDT) Received: from smtpin25.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay03.hostedemail.com (Postfix) with ESMTP id A3A48A0AB4 for ; Mon, 25 Sep 2023 08:24:08 +0000 (UTC) X-FDA: 81274432176.25.13FB691 Received: from mail-yb1-f179.google.com (mail-yb1-f179.google.com [209.85.219.179]) by imf27.hostedemail.com (Postfix) with ESMTP id AEBF34002C for ; Mon, 25 Sep 2023 08:24:06 +0000 (UTC) Authentication-Results: imf27.hostedemail.com; dkim=pass header.d=google.com header.s=20230601 header.b=aUyrgXeN; dmarc=pass (policy=reject) header.from=google.com; spf=pass (imf27.hostedemail.com: domain of hughd@google.com designates 209.85.219.179 as permitted sender) smtp.mailfrom=hughd@google.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1695630246; 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=aN6nR5IwDNUCMkgwxdT7Lz56gPehMPk2IXT5mlQNQ+g=; b=KAwxiqctynaDzOixMtaqa4Z4RAGkdUN0yLibwrUCayBePX84956E+Z1Lkbm89wm3B9I4wT Q7kuelhYgNKMEnR3i9vSUZzocVBexjjXOb5cin8oPTAZMidozLYysR+6YUIUkA2wceHcJD WXv/VLpZoppZYXxcrN1SRFSMa0aHyxc= ARC-Authentication-Results: i=1; imf27.hostedemail.com; dkim=pass header.d=google.com header.s=20230601 header.b=aUyrgXeN; dmarc=pass (policy=reject) header.from=google.com; spf=pass (imf27.hostedemail.com: domain of hughd@google.com designates 209.85.219.179 as permitted sender) smtp.mailfrom=hughd@google.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1695630246; a=rsa-sha256; cv=none; b=ux9CWSLo9OfP42yOlSSWTNYkWeN1eGAcciDt2NJOXlcAD7Ej0CHaVEeOIdbgHm6Ge7pbBI a/GBGqnKvpPP366AYQy2DmXp/uMQq7ocroZVxTswXqVe9JLpST96xYkIeCIJxf0lyjGX7v v1X0YR+GiKSBgzpqbyaxv8iejDMnQC4= Received: by mail-yb1-f179.google.com with SMTP id 3f1490d57ef6-d862533ea85so4897743276.0 for ; Mon, 25 Sep 2023 01:24:06 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1695630246; x=1696235046; darn=kvack.org; h=mime-version:references:message-id:in-reply-to:subject:cc:to:from :date:from:to:cc:subject:date:message-id:reply-to; bh=aN6nR5IwDNUCMkgwxdT7Lz56gPehMPk2IXT5mlQNQ+g=; b=aUyrgXeNjjQCuwzvwfF8EBzNNMHuGjFV1ckm08LmOzQyV4uOjvQKV250PyLfutEoav VMjZTSJkNCBVfLC7BY6f9IZ98lCNIiXetEEpE7MTjArYprIbzGqCUw0xz6ZpAi6j3E+k fR6HzsOoZ0tP7tCgGCp2vIn+bKw8ICCZrPxIAEie2eaiorA4M+fJ37feouNPNX7vomzc 43GRf8ZOEreU2fOiq3/hlCw2LiKA8zd2Xx5ak58JdRNo0xM569xnZOQomtgpNJaxJIGy pbCT2aKn8r9iEmg+lvBmLSr3wzGAFNT7UcZ5RZNgq5X/0GNrqi7obSWj/iYeCvvhF+Iu 4g6g== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1695630246; x=1696235046; h=mime-version:references:message-id:in-reply-to:subject:cc:to:from :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=aN6nR5IwDNUCMkgwxdT7Lz56gPehMPk2IXT5mlQNQ+g=; b=Rs98PjZmnavepNm2HQkrRMXBntluOmxz6QnQLEJO05hg4H2kXfY4lOiETnPPQ8AabX avo7+A9JWQ592O4R1KCbRkyNi1Ygu078Iplx3SuDeg9jPtaXw9SFfxlr8VqAndpgk1cr nRPVN9Mjxb2GW3TCfIuJNiG8u+bA09pUMDAm2uB/2+VRfTYNJBl0QQ0GstDrGcJzsVv7 BLt1WckVWQb+DeL/uKFu4NZxdTAmQ6YeTzuwOYxlGNMEQ0WOml/wFPeAcdzzBXFrMnaC AVGoCbwMTycn4WU7sdcOyjuQUFxzxCb5ODeu083aM7sjfMKPq12gT7JEwj0IFkJCBtIq ClHQ== X-Gm-Message-State: AOJu0YzXokoKUT/BUR6pwG96d4JhR0QGw9Kp0Y0Sgx7rxxpsCCZRdTuw Xq88FEb2tEE+9Ik+Hg9XEUtfug== X-Google-Smtp-Source: AGHT+IEZ5z3Rqd6K+p87ryKeqtoIFxtYPibeDUbUMGyD2lKyNvgK9u7Hoa4GbTrF1vLwXwt7ln5PhQ== X-Received: by 2002:a25:dbcc:0:b0:d0f:846c:ef7b with SMTP id g195-20020a25dbcc000000b00d0f846cef7bmr5135924ybf.17.1695630245463; Mon, 25 Sep 2023 01:24:05 -0700 (PDT) Received: from ripple.attlocal.net (172-10-233-147.lightspeed.sntcca.sbcglobal.net. [172.10.233.147]) by smtp.gmail.com with ESMTPSA id j134-20020a25238c000000b00c64533e4e20sm2118142ybj.33.2023.09.25.01.24.03 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 25 Sep 2023 01:24:04 -0700 (PDT) Date: Mon, 25 Sep 2023 01:24:02 -0700 (PDT) From: Hugh Dickins X-X-Sender: hugh@ripple.attlocal.net To: Andrew Morton cc: Andi Kleen , Christoph Lameter , Matthew Wilcox , Mike Kravetz , David Hildenbrand , Suren Baghdasaryan , Yang Shi , Sidhartha Kumar , Vishal Moola , Kefeng Wang , Greg Kroah-Hartman , Tejun Heo , Mel Gorman , Michal Hocko , linux-kernel@vger.kernel.org, linux-mm@kvack.org Subject: [PATCH 03/12] mempolicy: fix migrate_pages(2) syscall return nr_failed In-Reply-To: <2d872cef-7787-a7ca-10e-9d45a64c80b4@google.com> Message-ID: References: <2d872cef-7787-a7ca-10e-9d45a64c80b4@google.com> MIME-Version: 1.0 X-Rspamd-Queue-Id: AEBF34002C X-Rspam-User: X-Rspamd-Server: rspam04 X-Stat-Signature: qsr4ajr491wjwnnb9wjwimdonn3c615x X-HE-Tag: 1695630246-910879 X-HE-Meta: U2FsdGVkX18mtEKybVfs12ssSyqM1nv1MyftBJs0bf9fRPVaEkVb+ZoHsIctFll0iVNeSY7a/L+okHnA81kCE0BthroAL/kbijbXvBaheXq8/Yq72GgxwJVL+44ofXCBfwrlOJCwM+EvDiGTaWQsb5wh/86HNPtCePw7M9aS8t8IodQYXNxUisOAS5ZEBMS9k7w4ujYcJqj3aE5x6ALMqyxKPp8Dt0nzfh7mJjZdhoqk2hSWneabnZ+jP5YOtQvvM2XW5sx3qO5BgTQW0kt8OzlNjRe4BUnw6+9og9FaN5Sz1ZHzU5yxk/6BvNIWYi7X73YFhDpwTQnOx+RroZCsknvcbq6o2iX2KJOitXZGnuDog07EY3LqDHdRAcrIIUho3laFAcxjmh8H2cofjYxxGXpPmY4DcPeMc5j4hGNzvbrvQ4oWxT1LNvt8K2v8vuLPDdceuEAjwJI6wg6T1Qmxf14GZGjkLlOS9VCiSW8iDIv+OW8NYbeA0Hig+Tfg6WkklGpgSZc8I0q1oCf39PBKkAlS7h/x1zMcJy+w/YIvHFs512wHFqWvmRt4mJPVy7tMn+/awmAjwCGlMBBuFxKqwQTRnq3G3DswdGr6oASAmvBlAzxeKjYuaUWgIajZeooiNNcdvf4y5pIkjeBmHzNiP0wfofpWAsnguTW8jzHzfbbe7n6auDbSp5UYsn2G07vUMPhzO/FL469Tt8oFo3r8xxEg9CQRz7AdX7tMfEOgQKNS8o38+OdSVC+juwBVauOsA7qw6Ztx1SeJ4BUds006C5XmgiPNS0QCjASDZQAeYE40IF4JnxxXcueS45ipfqSz2G6vEni7oM6YNrr1+A4kZkBhYD/3WivF06Kcff1Sv7rvBGKGvUgD9Rk2Nl6GnwPH+sJQNQ2GtsD/14digVtR2j0mmkgjWDGuhTfbdijNaoDzi/a80QdW5wuDkA+DEMsLP7x8fNKJ2pHg81b8qL3 HODIp3sD u/pV5DHVxsRnFRuSjshZGJLIltklNj7xkc51JbmXFhw5aGmxdkcfOJexosHPyVd+a51XceNoYgJDe4CLlfyC/koDkR/7eKB2TEmwJHIyO1rwyiZrLQKrT0WI/HrMufnuJBD6iB1pSB1Qpez18kg8lq4RtfDVHR9tXp6iy0np8eRLEU78RDC7cRuduPknkOC/yeEGfafsdJTsBVmtrazgInFJISJTHGwUoDR7qs2NlKa1htS++sDwndwUVx2nJbTTlXO3X9s2Scc+iaBVqvZkhEGb5sahgxW4AbmgY3Mv3NeuBiDMj/RDA9ofm4dwS3PyWHHyUMGWX8AdZJE9IjZDwwtGf9LJ85x6uAQHMIaafg/W0P4IV0gKb614q84faCPkJ82sWaQ5a3yIaxPqTfSgedx6byoIxbP9/SYpT 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: "man 2 migrate_pages" says "On success migrate_pages() returns the number of pages that could not be moved". Although 5.3 and 5.4 commits fixed mbind(MPOL_MF_STRICT|MPOL_MF_MOVE*) to fail with EIO when not all pages could be moved (because some could not be isolated for migration), migrate_pages(2) was left still reporting only those pages failing at the migration stage, forgetting those failing at the earlier isolation stage. Fix that by accumulating a long nr_failed count in struct queue_pages, returned by queue_pages_range() when it's not returning an error, for adding on to the nr_failed count from migrate_pages() in mm/migrate.c. A count of pages? It's more a count of folios, but changing it to pages would entail more work (also in mm/migrate.c): does not seem justified. queue_pages_range() itself should only return -EIO in the "strictly unmovable" case (STRICT without any MOVEs): in that case it's best to break out as soon as nr_failed gets set; but otherwise it should continue to isolate pages for MOVing even when nr_failed - as the mbind(2) manpage promises. This fixes mbind(MPOL_MF_STRICT|MPOL_MF_MOVE*) behavior left over from 5.3 and 5.4, and the recent syzbot need for vma_start_write() before mbind_range(): but both of those may be fixed by smaller patches. There's a case when nr_failed should be incremented when it was missed: queue_folios_pte_range() and queue_folios_hugetlb() count the transient migration entries, like queue_folios_pmd() already did. And there's a case when nr_failed should not be incremented when it would have been: in meeting later PTEs of the same large folio, which can only be isolated once: fixed by recording the current large folio in struct queue_pages. Clean up the affected functions, fixing or updating many comments. Bool migrate_folio_add(), without -EIO: true if adding, or if skipping shared (but its arguable folio_estimated_sharers() heuristic left unchanged). Use MPOL_MF_WRLOCK flag to queue_pages_range(), instead of bool lock_vma. Use explicit STRICT|MOVE* flags where queue_pages_test_walk() checks for skipping, instead of hiding them behind MPOL_MF_VALID. Signed-off-by: Hugh Dickins Reviewed-by: Matthew Wilcox (Oracle) --- mm/mempolicy.c | 322 +++++++++++++++++++++++-------------------------- 1 file changed, 149 insertions(+), 173 deletions(-) diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 42b5567e3773..937386409c28 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -111,7 +111,8 @@ /* Internal flags */ #define MPOL_MF_DISCONTIG_OK (MPOL_MF_INTERNAL << 0) /* Skip checks for continuous vmas */ -#define MPOL_MF_INVERT (MPOL_MF_INTERNAL << 1) /* Invert check for nodemask */ +#define MPOL_MF_INVERT (MPOL_MF_INTERNAL << 1) /* Invert check for nodemask */ +#define MPOL_MF_WRLOCK (MPOL_MF_INTERNAL << 2) /* Write-lock walked vmas */ static struct kmem_cache *policy_cache; static struct kmem_cache *sn_cache; @@ -416,9 +417,19 @@ static const struct mempolicy_operations mpol_ops[MPOL_MAX] = { }, }; -static int migrate_folio_add(struct folio *folio, struct list_head *foliolist, +static bool migrate_folio_add(struct folio *folio, struct list_head *foliolist, unsigned long flags); +static bool strictly_unmovable(unsigned long flags) +{ + /* + * STRICT without MOVE flags lets do_mbind() fail immediately with -EIO + * if any misplaced page is found. + */ + return (flags & (MPOL_MF_STRICT | MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) == + MPOL_MF_STRICT; +} + struct queue_pages { struct list_head *pagelist; unsigned long flags; @@ -426,6 +437,8 @@ struct queue_pages { unsigned long start; unsigned long end; struct vm_area_struct *first; + struct folio *large; /* note last large folio on pagelist */ + long nr_failed; /* could not be isolated at this time */ }; /* @@ -443,50 +456,27 @@ static inline bool queue_folio_required(struct folio *folio, return node_isset(nid, *qp->nmask) == !(flags & MPOL_MF_INVERT); } -/* - * queue_folios_pmd() has three possible return values: - * 0 - folios are placed on the right node or queued successfully, or - * special page is met, i.e. huge zero page. - * 1 - there is unmovable folio, and MPOL_MF_MOVE* & MPOL_MF_STRICT were - * specified. - * -EIO - is migration entry or only MPOL_MF_STRICT was specified and an - * existing folio was already on a node that does not follow the - * policy. - */ -static int queue_folios_pmd(pmd_t *pmd, spinlock_t *ptl, unsigned long addr, +static void queue_folios_pmd(pmd_t *pmd, spinlock_t *ptl, unsigned long addr, unsigned long end, struct mm_walk *walk) - __releases(ptl) { - int ret = 0; struct folio *folio; struct queue_pages *qp = walk->private; - unsigned long flags; if (unlikely(is_pmd_migration_entry(*pmd))) { - ret = -EIO; - goto unlock; + qp->nr_failed++; + return; } folio = pfn_folio(pmd_pfn(*pmd)); if (is_huge_zero_page(&folio->page)) { walk->action = ACTION_CONTINUE; - goto unlock; + return; } if (!queue_folio_required(folio, qp)) - goto unlock; - - flags = qp->flags; - /* go to folio migration */ - if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) { - if (!vma_migratable(walk->vma) || - migrate_folio_add(folio, qp->pagelist, flags)) { - ret = 1; - goto unlock; - } - } else - ret = -EIO; -unlock: - spin_unlock(ptl); - return ret; + return; + if (!(qp->flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) || + !vma_migratable(walk->vma) || + !migrate_folio_add(folio, qp->pagelist, qp->flags)) + qp->nr_failed++; } /* @@ -496,8 +486,6 @@ static int queue_folios_pmd(pmd_t *pmd, spinlock_t *ptl, unsigned long addr, * queue_folios_pte_range() has three possible return values: * 0 - folios are placed on the right node or queued successfully, or * special page is met, i.e. zero page. - * 1 - there is unmovable folio, and MPOL_MF_MOVE* & MPOL_MF_STRICT were - * specified. * -EIO - only MPOL_MF_STRICT was specified and an existing folio was already * on a node that does not follow the policy. */ @@ -508,14 +496,16 @@ static int queue_folios_pte_range(pmd_t *pmd, unsigned long addr, struct folio *folio; struct queue_pages *qp = walk->private; unsigned long flags = qp->flags; - bool has_unmovable = false; pte_t *pte, *mapped_pte; pte_t ptent; spinlock_t *ptl; ptl = pmd_trans_huge_lock(pmd, vma); - if (ptl) - return queue_folios_pmd(pmd, ptl, addr, end, walk); + if (ptl) { + queue_folios_pmd(pmd, ptl, addr, end, walk); + spin_unlock(ptl); + goto out; + } mapped_pte = pte = pte_offset_map_lock(walk->mm, pmd, addr, &ptl); if (!pte) { @@ -524,8 +514,13 @@ static int queue_folios_pte_range(pmd_t *pmd, unsigned long addr, } for (; addr != end; pte++, addr += PAGE_SIZE) { ptent = ptep_get(pte); - if (!pte_present(ptent)) + if (pte_none(ptent)) continue; + if (!pte_present(ptent)) { + if (is_migration_entry(pte_to_swp_entry(ptent))) + qp->nr_failed++; + continue; + } folio = vm_normal_folio(vma, addr, ptent); if (!folio || folio_is_zone_device(folio)) continue; @@ -537,97 +532,82 @@ static int queue_folios_pte_range(pmd_t *pmd, unsigned long addr, continue; if (!queue_folio_required(folio, qp)) continue; - if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) { - /* MPOL_MF_STRICT must be specified if we get here */ - if (!vma_migratable(vma)) { - has_unmovable = true; + if (!(flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) || + !vma_migratable(vma)) { + qp->nr_failed++; + if (strictly_unmovable(flags)) break; - } - + } + if (migrate_folio_add(folio, qp->pagelist, flags)) { /* - * Do not abort immediately since there may be - * temporary off LRU pages in the range. Still - * need migrate other LRU pages. + * A large folio can only be isolated from LRU once, + * but may be mapped by many PTEs (and Copy-On-Write may + * intersperse PTEs of other folios). This is a common + * case, so don't mistake it for failure (but of course + * there can be other cases of multi-mapped pages which + * this quick check does not help to filter out - and a + * search of the pagelist might grow to be prohibitive). */ - if (migrate_folio_add(folio, qp->pagelist, flags)) - has_unmovable = true; - } else - break; + if (folio_test_large(folio)) + qp->large = folio; + } else if (folio != qp->large) { + qp->nr_failed++; + if (strictly_unmovable(flags)) + break; + } } pte_unmap_unlock(mapped_pte, ptl); cond_resched(); - - if (has_unmovable) - return 1; - - return addr != end ? -EIO : 0; +out: + if (qp->nr_failed && strictly_unmovable(flags)) + return -EIO; + return 0; } static int queue_folios_hugetlb(pte_t *pte, unsigned long hmask, unsigned long addr, unsigned long end, struct mm_walk *walk) { - int ret = 0; #ifdef CONFIG_HUGETLB_PAGE struct queue_pages *qp = walk->private; - unsigned long flags = (qp->flags & MPOL_MF_VALID); + unsigned long flags = qp->flags; struct folio *folio; spinlock_t *ptl; pte_t entry; ptl = huge_pte_lock(hstate_vma(walk->vma), walk->mm, pte); entry = huge_ptep_get(pte); - if (!pte_present(entry)) + if (!pte_present(entry)) { + if (unlikely(is_hugetlb_entry_migration(entry))) + qp->nr_failed++; goto unlock; + } folio = pfn_folio(pte_pfn(entry)); if (!queue_folio_required(folio, qp)) goto unlock; - - if (flags == MPOL_MF_STRICT) { - /* - * STRICT alone means only detecting misplaced folio and no - * need to further check other vma. - */ - ret = -EIO; + if (!(flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) || + !vma_migratable(walk->vma)) { + qp->nr_failed++; goto unlock; } - - if (!vma_migratable(walk->vma)) { - /* - * Must be STRICT with MOVE*, otherwise .test_walk() have - * stopped walking current vma. - * Detecting misplaced folio but allow migrating folios which - * have been queued. - */ - ret = 1; - goto unlock; - } - /* - * With MPOL_MF_MOVE, we try to migrate only unshared folios. If it - * is shared it is likely not worth migrating. + * Unless MPOL_MF_MOVE_ALL, we try to avoid migrating a shared folio. + * Choosing not to migrate a shared folio is not counted as a failure. * * To check if the folio is shared, ideally we want to make sure * every page is mapped to the same process. Doing that is very - * expensive, so check the estimated mapcount of the folio instead. + * expensive, so check the estimated sharers of the folio instead. */ - if (flags & (MPOL_MF_MOVE_ALL) || - (flags & MPOL_MF_MOVE && folio_estimated_sharers(folio) == 1 && - !hugetlb_pmd_shared(pte))) { - if (!isolate_hugetlb(folio, qp->pagelist) && - (flags & MPOL_MF_STRICT)) - /* - * Failed to isolate folio but allow migrating pages - * which have been queued. - */ - ret = 1; - } + if ((flags & MPOL_MF_MOVE_ALL) || + (folio_estimated_sharers(folio) == 1 && !hugetlb_pmd_shared(pte))) + if (!isolate_hugetlb(folio, qp->pagelist)) + qp->nr_failed++; unlock: spin_unlock(ptl); -#else - BUG(); + if (qp->nr_failed && strictly_unmovable(flags)) + return -EIO; #endif - return ret; + return 0; } #ifdef CONFIG_NUMA_BALANCING @@ -708,8 +688,11 @@ static int queue_pages_test_walk(unsigned long start, unsigned long end, return 1; } - /* queue pages from current vma */ - if (flags & MPOL_MF_VALID) + /* + * Check page nodes, and queue pages to move, in the current vma. + * But if no moving, and no strict checking, the scan can be skipped. + */ + if (flags & (MPOL_MF_STRICT | MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) return 0; return 1; } @@ -731,22 +714,22 @@ static const struct mm_walk_ops queue_pages_lock_vma_walk_ops = { /* * Walk through page tables and collect pages to be migrated. * - * If pages found in a given range are on a set of nodes (determined by - * @nodes and @flags,) it's isolated and queued to the pagelist which is - * passed via @private. + * If pages found in a given range are not on the required set of @nodes, + * and migration is allowed, they are isolated and queued to the pagelist + * which is passed via @private. * - * queue_pages_range() has three possible return values: - * 1 - there is unmovable page, but MPOL_MF_MOVE* & MPOL_MF_STRICT were - * specified. - * 0 - queue pages successfully or no misplaced page. - * errno - i.e. misplaced pages with MPOL_MF_STRICT specified (-EIO) or - * memory range specified by nodemask and maxnode points outside - * your accessible address space (-EFAULT) + * queue_pages_range() may return: + * 0 - all pages already on the right node, or successfully queued for moving + * (or neither strict checking nor moving requested: only range checking). + * >0 - this number of misplaced folios could not be queued for moving + * (a hugetlbfs page or a transparent huge page being counted as 1). + * -EIO - a misplaced page found, when MPOL_MF_STRICT specified without MOVEs. + * -EFAULT - a hole in the memory range, when MPOL_MF_DISCONTIG_OK unspecified. */ -static int +static long queue_pages_range(struct mm_struct *mm, unsigned long start, unsigned long end, nodemask_t *nodes, unsigned long flags, - struct list_head *pagelist, bool lock_vma) + struct list_head *pagelist) { int err; struct queue_pages qp = { @@ -757,7 +740,7 @@ queue_pages_range(struct mm_struct *mm, unsigned long start, unsigned long end, .end = end, .first = NULL, }; - const struct mm_walk_ops *ops = lock_vma ? + const struct mm_walk_ops *ops = (flags & MPOL_MF_WRLOCK) ? &queue_pages_lock_vma_walk_ops : &queue_pages_walk_ops; err = walk_page_range(mm, start, end, ops, &qp); @@ -766,7 +749,7 @@ queue_pages_range(struct mm_struct *mm, unsigned long start, unsigned long end, /* whole range in hole */ err = -EFAULT; - return err; + return err ? : qp.nr_failed; } /* @@ -1029,16 +1012,16 @@ static long do_get_mempolicy(int *policy, nodemask_t *nmask, } #ifdef CONFIG_MIGRATION -static int migrate_folio_add(struct folio *folio, struct list_head *foliolist, +static bool migrate_folio_add(struct folio *folio, struct list_head *foliolist, unsigned long flags) { /* - * We try to migrate only unshared folios. If it is shared it - * is likely not worth migrating. + * Unless MPOL_MF_MOVE_ALL, we try to avoid migrating a shared folio. + * Choosing not to migrate a shared folio is not counted as a failure. * * To check if the folio is shared, ideally we want to make sure * every page is mapped to the same process. Doing that is very - * expensive, so check the estimated mapcount of the folio instead. + * expensive, so check the estimated sharers of the folio instead. */ if ((flags & MPOL_MF_MOVE_ALL) || folio_estimated_sharers(folio) == 1) { if (folio_isolate_lru(folio)) { @@ -1046,32 +1029,31 @@ static int migrate_folio_add(struct folio *folio, struct list_head *foliolist, node_stat_mod_folio(folio, NR_ISOLATED_ANON + folio_is_file_lru(folio), folio_nr_pages(folio)); - } else if (flags & MPOL_MF_STRICT) { + } else { /* * Non-movable folio may reach here. And, there may be * temporary off LRU folios or non-LRU movable folios. * Treat them as unmovable folios since they can't be - * isolated, so they can't be moved at the moment. It - * should return -EIO for this case too. + * isolated, so they can't be moved at the moment. */ - return -EIO; + return false; } } - - return 0; + return true; } /* * Migrate pages from one node to a target node. * Returns error or the number of pages not migrated. */ -static int migrate_to_node(struct mm_struct *mm, int source, int dest, - int flags) +static long migrate_to_node(struct mm_struct *mm, int source, int dest, + int flags) { nodemask_t nmask; struct vm_area_struct *vma; LIST_HEAD(pagelist); - int err = 0; + long nr_failed; + long err = 0; struct migration_target_control mtc = { .nid = dest, .gfp_mask = GFP_HIGHUSER_MOVABLE | __GFP_THISNODE, @@ -1080,23 +1062,27 @@ static int migrate_to_node(struct mm_struct *mm, int source, int dest, nodes_clear(nmask); node_set(source, nmask); - /* - * This does not "check" the range but isolates all pages that - * need migration. Between passing in the full user address - * space range and MPOL_MF_DISCONTIG_OK, this call can not fail. - */ - vma = find_vma(mm, 0); VM_BUG_ON(!(flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL))); - queue_pages_range(mm, vma->vm_start, mm->task_size, &nmask, - flags | MPOL_MF_DISCONTIG_OK, &pagelist, false); + vma = find_vma(mm, 0); + + /* + * This does not migrate the range, but isolates all pages that + * need migration. Between passing in the full user address + * space range and MPOL_MF_DISCONTIG_OK, this call cannot fail, + * but passes back the count of pages which could not be isolated. + */ + nr_failed = queue_pages_range(mm, vma->vm_start, mm->task_size, &nmask, + flags | MPOL_MF_DISCONTIG_OK, &pagelist); if (!list_empty(&pagelist)) { err = migrate_pages(&pagelist, alloc_migration_target, NULL, - (unsigned long)&mtc, MIGRATE_SYNC, MR_SYSCALL, NULL); + (unsigned long)&mtc, MIGRATE_SYNC, MR_SYSCALL, NULL); if (err) putback_movable_pages(&pagelist); } + if (err >= 0) + err += nr_failed; return err; } @@ -1109,8 +1095,8 @@ static int migrate_to_node(struct mm_struct *mm, int source, int dest, int do_migrate_pages(struct mm_struct *mm, const nodemask_t *from, const nodemask_t *to, int flags) { - int busy = 0; - int err = 0; + long nr_failed = 0; + long err = 0; nodemask_t tmp; lru_cache_disable(); @@ -1192,7 +1178,7 @@ int do_migrate_pages(struct mm_struct *mm, const nodemask_t *from, node_clear(source, tmp); err = migrate_to_node(mm, source, dest, flags); if (err > 0) - busy += err; + nr_failed += err; if (err < 0) break; } @@ -1201,8 +1187,7 @@ int do_migrate_pages(struct mm_struct *mm, const nodemask_t *from, lru_cache_enable(); if (err < 0) return err; - return busy; - + return (nr_failed < INT_MAX) ? nr_failed : INT_MAX; } /* @@ -1241,10 +1226,10 @@ static struct folio *new_folio(struct folio *src, unsigned long start) } #else -static int migrate_folio_add(struct folio *folio, struct list_head *foliolist, +static bool migrate_folio_add(struct folio *folio, struct list_head *foliolist, unsigned long flags) { - return -EIO; + return false; } int do_migrate_pages(struct mm_struct *mm, const nodemask_t *from, @@ -1268,8 +1253,8 @@ static long do_mbind(unsigned long start, unsigned long len, struct vma_iterator vmi; struct mempolicy *new; unsigned long end; - int err; - int ret; + long err; + long nr_failed; LIST_HEAD(pagelist); if (flags & ~(unsigned long)MPOL_MF_VALID) @@ -1309,10 +1294,8 @@ static long do_mbind(unsigned long start, unsigned long len, start, start + len, mode, mode_flags, nmask ? nodes_addr(*nmask)[0] : NUMA_NO_NODE); - if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) { - + if (flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) lru_cache_disable(); - } { NODEMASK_SCRATCH(scratch); if (scratch) { @@ -1328,44 +1311,37 @@ static long do_mbind(unsigned long start, unsigned long len, goto mpol_out; /* - * Lock the VMAs before scanning for pages to migrate, to ensure we don't - * miss a concurrently inserted page. + * Lock the VMAs before scanning for pages to migrate, + * to ensure we don't miss a concurrently inserted page. */ - ret = queue_pages_range(mm, start, end, nmask, - flags | MPOL_MF_INVERT, &pagelist, true); + nr_failed = queue_pages_range(mm, start, end, nmask, + flags | MPOL_MF_INVERT | MPOL_MF_WRLOCK, &pagelist); - if (ret < 0) { - err = ret; - goto up_out; - } - - vma_iter_init(&vmi, mm, start); - prev = vma_prev(&vmi); - for_each_vma_range(vmi, vma, end) { - err = mbind_range(&vmi, vma, &prev, start, end, new); - if (err) - break; + if (nr_failed < 0) { + err = nr_failed; + } else { + vma_iter_init(&vmi, mm, start); + prev = vma_prev(&vmi); + for_each_vma_range(vmi, vma, end) { + err = mbind_range(&vmi, vma, &prev, start, end, new); + if (err) + break; + } } if (!err) { - int nr_failed = 0; - if (!list_empty(&pagelist)) { WARN_ON_ONCE(flags & MPOL_MF_LAZY); - nr_failed = migrate_pages(&pagelist, new_folio, NULL, + nr_failed |= migrate_pages(&pagelist, new_folio, NULL, start, MIGRATE_SYNC, MR_MEMPOLICY_MBIND, NULL); - if (nr_failed) - putback_movable_pages(&pagelist); } - - if ((ret > 0) || (nr_failed && (flags & MPOL_MF_STRICT))) + if (nr_failed && (flags & MPOL_MF_STRICT)) err = -EIO; - } else { -up_out: - if (!list_empty(&pagelist)) - putback_movable_pages(&pagelist); } + if (!list_empty(&pagelist)) + putback_movable_pages(&pagelist); + mmap_write_unlock(mm); mpol_out: mpol_put(new);