From: Oscar Salvador <osalvador@suse.de>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org,
Peter Xu <peterx@redhat.com>, Muchun Song <muchun.song@linux.dev>,
David Hildenbrand <david@redhat.com>,
SeongJae Park <sj@kernel.org>, Miaohe Lin <linmiaohe@huawei.com>,
Michal Hocko <mhocko@suse.com>,
Matthew Wilcox <willy@infradead.org>,
Christophe Leroy <christophe.leroy@csgroup.eu>,
Oscar Salvador <osalvador@suse.de>
Subject: [PATCH 23/45] mm/mempolicy: Create queue_folios_pud to handle PUD-mapped hugetlb vmas
Date: Thu, 4 Jul 2024 06:31:10 +0200 [thread overview]
Message-ID: <20240704043132.28501-24-osalvador@suse.de> (raw)
In-Reply-To: <20240704043132.28501-1-osalvador@suse.de>
Normal THP cannot be PUD-mapped (besides devmap), but hugetlb can, so create
queue_folios_pud in order to handle PUD-mapped hugetlb vmas.
Also implement is_pud_migration_entry and pud_folio, as they will be used in this patch.
Signed-off-by: Oscar Salvador <osalvador@suse.de>
---
include/linux/pgtable.h | 1 +
include/linux/swapops.h | 12 ++++++++++++
mm/mempolicy.c | 32 ++++++++++++++++++++++++++++++++
3 files changed, 45 insertions(+)
diff --git a/include/linux/pgtable.h b/include/linux/pgtable.h
index 458e3cbc96b2..23d51fec81ac 100644
--- a/include/linux/pgtable.h
+++ b/include/linux/pgtable.h
@@ -51,6 +51,7 @@
#endif
#define pmd_folio(pmd) page_folio(pmd_page(pmd))
+#define pud_folio(pud) page_folio(pud_page(pud))
/*
* A page table page can be thought of an array like this: pXd_t[PTRS_PER_PxD]
diff --git a/include/linux/swapops.h b/include/linux/swapops.h
index 182957f0d013..a23900961d11 100644
--- a/include/linux/swapops.h
+++ b/include/linux/swapops.h
@@ -542,6 +542,18 @@ static inline bool is_pfn_swap_entry(swp_entry_t entry)
struct page_vma_mapped_walk;
+#ifdef CONFIG_HUGETLB_PAGE
+static inline int is_pud_migration_entry(pud_t pud)
+{
+ return is_swap_pud(pud) && is_migration_entry(pud_to_swp_entry(pud));
+}
+#else
+static inline int is_pud_migration_entry(pud_t pud)
+{
+ return 0;
+}
+#endif
+
#ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
extern int set_pmd_migration_entry(struct page_vma_mapped_walk *pvmw,
struct page *page);
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 5baf29da198c..93b14090d484 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -501,6 +501,37 @@ static inline bool queue_folio_required(struct folio *folio,
return node_isset(nid, *qp->nmask) == !(flags & MPOL_MF_INVERT);
}
+static int queue_folios_pud(pud_t *pud, unsigned long addr, unsigned long end,
+ struct mm_walk *walk)
+{
+ spinlock_t *ptl;
+ struct folio *folio;
+ struct vm_area_struct *vma = walk->vma;
+ struct queue_pages *qp = walk->private;
+
+ ptl = pud_huge_lock(pud, vma);
+ if (!ptl)
+ return 0;
+
+ if (unlikely(is_pud_migration_entry(*pud))) {
+ qp->nr_failed++;
+ goto out;
+ }
+ folio = pud_folio(*pud);
+ if (!queue_folio_required(folio, qp))
+ goto out;
+ if (!(qp->flags & (MPOL_MF_MOVE | MPOL_MF_MOVE_ALL)) ||
+ !vma_migratable(walk->vma) ||
+ !migrate_folio_add(folio, qp->pagelist, qp->flags, walk->vma, false))
+ qp->nr_failed++;
+
+ spin_unlock(ptl);
+out:
+ if (qp->nr_failed && strictly_unmovable(qp->flags))
+ return -EIO;
+ return 0;
+}
+
static void queue_folios_pmd(pmd_t *pmd, struct mm_walk *walk)
{
struct folio *folio;
@@ -730,6 +761,7 @@ static int queue_pages_test_walk(unsigned long start, unsigned long end,
static const struct mm_walk_ops queue_pages_walk_ops = {
.hugetlb_entry = queue_folios_hugetlb,
+ .pud_entry = queue_folios_pud,
.pmd_entry = queue_folios_pte_range,
.test_walk = queue_pages_test_walk,
.walk_lock = PGWALK_RDLOCK,
--
2.26.2
next prev parent reply other threads:[~2024-07-04 4:32 UTC|newest]
Thread overview: 66+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-04 4:30 [PATCH 00/45] hugetlb pagewalk unification Oscar Salvador
2024-07-04 4:30 ` [PATCH 01/45] arch/x86: Drop own definition of pgd,p4d_leaf Oscar Salvador
2024-07-04 4:30 ` [PATCH 02/45] mm: Add {pmd,pud}_huge_lock helper Oscar Salvador
2024-07-04 15:02 ` Peter Xu
2024-07-04 4:30 ` [PATCH 03/45] mm/pagewalk: Move vma_pgtable_walk_begin and vma_pgtable_walk_end upfront Oscar Salvador
2024-07-04 4:30 ` [PATCH 04/45] mm/pagewalk: Only call pud_entry when we have a pud leaf Oscar Salvador
2024-07-04 4:30 ` [PATCH 05/45] mm/pagewalk: Enable walk_pmd_range to handle cont-pmds Oscar Salvador
2024-07-04 15:41 ` David Hildenbrand
2024-07-05 16:56 ` kernel test robot
2024-07-04 4:30 ` [PATCH 06/45] mm/pagewalk: Do not try to split non-thp pud or pmd leafs Oscar Salvador
2024-07-04 4:30 ` [PATCH 07/45] arch/s390: Enable __s390_enable_skey_pmd to handle hugetlb vmas Oscar Salvador
2024-07-04 4:30 ` [PATCH 08/45] fs/proc: Enable smaps_pmd_entry to handle PMD-mapped " Oscar Salvador
2024-07-04 4:30 ` [PATCH 09/45] mm: Implement pud-version functions for swap and vm_normal_page_pud Oscar Salvador
2024-07-04 4:30 ` [PATCH 10/45] fs/proc: Create smaps_pud_range to handle PUD-mapped hugetlb vmas Oscar Salvador
2024-07-04 4:30 ` [PATCH 11/45] fs/proc: Enable smaps_pte_entry to handle cont-pte mapped " Oscar Salvador
2024-07-04 10:30 ` David Hildenbrand
2024-07-04 4:30 ` [PATCH 12/45] fs/proc: Enable pagemap_pmd_range to handle " Oscar Salvador
2024-07-04 4:31 ` [PATCH 13/45] mm: Implement pud-version uffd functions Oscar Salvador
2024-07-05 15:48 ` kernel test robot
2024-07-05 15:48 ` kernel test robot
2024-07-04 4:31 ` [PATCH 14/45] fs/proc: Create pagemap_pud_range to handle PUD-mapped hugetlb vmas Oscar Salvador
2024-07-04 4:31 ` [PATCH 15/45] fs/proc: Adjust pte_to_pagemap_entry for " Oscar Salvador
2024-07-04 4:31 ` [PATCH 16/45] fs/proc: Enable pagemap_scan_pmd_entry to handle " Oscar Salvador
2024-07-04 4:31 ` [PATCH 17/45] mm: Implement pud-version for pud_mkinvalid and pudp_establish Oscar Salvador
2024-07-04 4:31 ` [PATCH 18/45] fs/proc: Create pagemap_scan_pud_entry to handle PUD-mapped hugetlb vmas Oscar Salvador
2024-07-04 4:31 ` [PATCH 19/45] fs/proc: Enable gather_pte_stats to handle " Oscar Salvador
2024-07-04 4:31 ` [PATCH 20/45] fs/proc: Enable gather_pte_stats to handle cont-pte mapped " Oscar Salvador
2024-07-04 4:31 ` [PATCH 21/45] fs/proc: Create gather_pud_stats to handle PUD-mapped hugetlb pages Oscar Salvador
2024-07-04 4:31 ` [PATCH 22/45] mm/mempolicy: Enable queue_folios_pmd to handle hugetlb vmas Oscar Salvador
2024-07-04 4:31 ` Oscar Salvador [this message]
2024-07-04 4:31 ` [PATCH 24/45] mm/memory_failure: Enable check_hwpoisoned_pmd_entry " Oscar Salvador
2024-07-04 4:31 ` [PATCH 25/45] mm/memory-failure: Create check_hwpoisoned_pud_entry to handle PUD-mapped " Oscar Salvador
2024-07-04 4:31 ` [PATCH 26/45] mm/damon: Enable damon_young_pmd_entry to handle " Oscar Salvador
2024-07-04 4:31 ` [PATCH 27/45] mm/damon: Create damon_young_pud_entry to handle PUD-mapped " Oscar Salvador
2024-07-04 4:31 ` [PATCH 28/45] mm/damon: Enable damon_mkold_pmd_entry to handle " Oscar Salvador
2024-07-04 11:03 ` David Hildenbrand
2024-07-04 4:31 ` [PATCH 29/45] mm/damon: Create damon_mkold_pud_entry to handle PUD-mapped " Oscar Salvador
2024-07-04 4:31 ` [PATCH 30/45] mm,mincore: Enable mincore_pte_range to handle " Oscar Salvador
2024-07-04 4:31 ` [PATCH 31/45] mm/mincore: Create mincore_pud_range to handle PUD-mapped " Oscar Salvador
2024-07-04 4:31 ` [PATCH 32/45] mm/hmm: Enable hmm_vma_walk_pmd, to handle " Oscar Salvador
2024-07-04 4:31 ` [PATCH 33/45] mm/hmm: Enable hmm_vma_walk_pud to handle PUD-mapped " Oscar Salvador
2024-07-04 4:31 ` [PATCH 34/45] arch/powerpc: Skip hugetlb vmas in subpage_mark_vma_nohuge Oscar Salvador
2024-07-04 4:31 ` [PATCH 35/45] arch/s390: Skip hugetlb vmas in thp_split_mm Oscar Salvador
2024-07-04 4:31 ` [PATCH 36/45] fs/proc: Make clear_refs_test_walk skip hugetlb vmas Oscar Salvador
2024-07-04 4:31 ` [PATCH 37/45] mm/lock: Make mlock_test_walk " Oscar Salvador
2024-07-04 4:31 ` [PATCH 38/45] mm/madvise: Make swapin_test_walk " Oscar Salvador
2024-07-04 4:31 ` [PATCH 39/45] mm/madvise: Make madvise_cold_test_walk " Oscar Salvador
2024-07-04 4:31 ` [PATCH 40/45] mm/madvise: Make madvise_free_test_walk " Oscar Salvador
2024-07-04 4:31 ` [PATCH 41/45] mm/migrate_device: Make migrate_vma_test_walk " Oscar Salvador
2024-07-04 4:31 ` [PATCH 42/45] mm/memcontrol: Make mem_cgroup_move_test_walk " Oscar Salvador
2024-07-04 4:31 ` [PATCH 43/45] mm/memcontrol: Make mem_cgroup_count_test_walk " Oscar Salvador
2024-07-04 4:31 ` [PATCH 44/45] mm/hugetlb_vmemmap: Make vmemmap_test_walk " Oscar Salvador
2024-07-04 4:31 ` [PATCH 45/45] mm: Delete all hugetlb_entry entries Oscar Salvador
2024-07-04 10:13 ` [PATCH 00/45] hugetlb pagewalk unification Oscar Salvador
2024-07-04 10:44 ` David Hildenbrand
2024-07-04 14:30 ` Peter Xu
2024-07-04 15:23 ` David Hildenbrand
2024-07-04 16:43 ` Peter Xu
2024-07-08 8:18 ` Oscar Salvador
2024-07-08 14:28 ` Jason Gunthorpe
2024-07-10 3:52 ` David Hildenbrand
2024-07-10 11:26 ` Oscar Salvador
2024-07-11 0:15 ` David Hildenbrand
2024-07-11 4:48 ` Oscar Salvador
2024-07-11 4:53 ` David Hildenbrand
2024-07-08 14:35 ` Jason Gunthorpe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240704043132.28501-24-osalvador@suse.de \
--to=osalvador@suse.de \
--cc=akpm@linux-foundation.org \
--cc=christophe.leroy@csgroup.eu \
--cc=david@redhat.com \
--cc=linmiaohe@huawei.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.com \
--cc=muchun.song@linux.dev \
--cc=peterx@redhat.com \
--cc=sj@kernel.org \
--cc=willy@infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox