From: Alex Sierra <alex.sierra@amd.com>
Date: Fri, 15 Jul 2022 10:05:11 -0500
Subject: mm: handling Non-LRU pages returned by vm_normal_pages
Git-commit: 3218f8712d6bba1812efd5e0d66c1e15134f2a91
Patch-mainline: v6.0-rc1
References: jsc#PED-1294
[ SLE15-SP5 backport: adjust to missing mlock rewrite ]
With DEVICE_COHERENT, we'll soon have vm_normal_pages() return
device-managed anonymous pages that are not LRU pages. Although they
behave like normal pages for purposes of mapping in CPU page, and for COW.
They do not support LRU lists, NUMA migration or THP.
Callers to follow_page() currently don't expect ZONE_DEVICE pages,
however, with DEVICE_COHERENT we might now return ZONE_DEVICE. Check for
ZONE_DEVICE pages in applicable users of follow_page() as well.
Link: https://lkml.kernel.org/r/20220715150521.18165-5-alex.sierra@amd.com
Signed-off-by: Alex Sierra <alex.sierra@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> [v2]
Reviewed-by: Alistair Popple <apopple@nvidia.com> [v6]
Cc: Christoph Hellwig <hch@lst.de>
Cc: David Hildenbrand <david@redhat.com>
Cc: Jason Gunthorpe <jgg@nvidia.com>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
---
fs/proc/task_mmu.c | 2 +-
mm/huge_memory.c | 2 +-
mm/khugepaged.c | 9 ++++++---
mm/ksm.c | 6 +++---
mm/madvise.c | 4 ++--
mm/memory.c | 10 +++++++++-
mm/mempolicy.c | 2 +-
mm/migrate.c | 4 ++--
mm/mlock.c | 5 +++--
mm/mprotect.c | 2 +-
10 files changed, 29 insertions(+), 17 deletions(-)
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -1749,7 +1749,7 @@ static struct page *can_gather_numa_stat
return NULL;
page = vm_normal_page(vma, addr, pte);
- if (!page)
+ if (!page || is_zone_device_page(page))
return NULL;
if (PageReserved(page))
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2970,7 +2970,7 @@ static int split_huge_pages_pid(int pid,
if (IS_ERR(page))
continue;
- if (!page)
+ if (!page || is_zone_device_page(page))
continue;
if (!is_transparent_hugepage(page))
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -626,7 +626,7 @@ static int __collapse_huge_page_isolate(
goto out;
}
page = vm_normal_page(vma, address, pteval);
- if (unlikely(!page)) {
+ if (unlikely(!page) || unlikely(is_zone_device_page(page))) {
result = SCAN_PAGE_NULL;
goto out;
}
@@ -1290,7 +1290,7 @@ static int khugepaged_scan_pmd(struct mm
writable = true;
page = vm_normal_page(vma, _address, pteval);
- if (unlikely(!page)) {
+ if (unlikely(!page) || unlikely(is_zone_device_page(page))) {
result = SCAN_PAGE_NULL;
goto out_unmap;
}
@@ -1482,7 +1482,8 @@ void collapse_pte_mapped_thp(struct mm_s
goto abort;
page = vm_normal_page(vma, addr, *pte);
-
+ if (WARN_ON_ONCE(page && is_zone_device_page(page)))
+ page = NULL;
/*
* Note that uprobe, debugger, or MAP_PRIVATE may change the
* page table, but the new page will not be a subpage of hpage.
@@ -1500,6 +1501,8 @@ void collapse_pte_mapped_thp(struct mm_s
if (pte_none(*pte))
continue;
page = vm_normal_page(vma, addr, *pte);
+ if (WARN_ON_ONCE(page && is_zone_device_page(page)))
+ goto abort;
page_remove_rmap(page, false);
}
--- a/mm/ksm.c
+++ b/mm/ksm.c
@@ -474,7 +474,7 @@ static int break_ksm(struct vm_area_stru
cond_resched();
page = follow_page(vma, addr,
FOLL_GET | FOLL_MIGRATION | FOLL_REMOTE);
- if (IS_ERR_OR_NULL(page))
+ if (IS_ERR_OR_NULL(page) || is_zone_device_page(page))
break;
if (PageKsm(page))
ret = handle_mm_fault(vma, addr,
@@ -559,7 +559,7 @@ static struct page *get_mergeable_page(s
goto out;
page = follow_page(vma, addr, FOLL_GET);
- if (IS_ERR_OR_NULL(page))
+ if (IS_ERR_OR_NULL(page) || is_zone_device_page(page))
goto out;
if (PageAnon(page)) {
flush_anon_page(vma, page, addr);
@@ -2298,7 +2298,7 @@ next_mm:
if (ksm_test_exit(mm))
break;
*page = follow_page(vma, ksm_scan.address, FOLL_GET);
- if (IS_ERR_OR_NULL(*page)) {
+ if (IS_ERR_OR_NULL(*page) || is_zone_device_page(*page)) {
ksm_scan.address += PAGE_SIZE;
cond_resched();
continue;
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -406,7 +406,7 @@ regular_page:
continue;
page = vm_normal_page(vma, addr, ptent);
- if (!page)
+ if (!page || is_zone_device_page(page))
continue;
/*
@@ -616,7 +616,7 @@ static int madvise_free_pte_range(pmd_t
}
page = vm_normal_page(vma, addr, ptent);
- if (!page)
+ if (!page || is_zone_device_page(page))
continue;
/*
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -620,6 +620,14 @@ struct page *vm_normal_page(struct vm_ar
if (is_zero_pfn(pfn))
return NULL;
if (pte_devmap(pte))
+ /*
+ * NOTE: New users of ZONE_DEVICE will not set pte_devmap()
+ * and will have refcounts incremented on their struct pages
+ * when they are inserted into PTEs, thus they are safe to
+ * return here. Legacy ZONE_DEVICE pages that set pte_devmap()
+ * do not have refcounts. Example of legacy ZONE_DEVICE is
+ * MEMORY_DEVICE_FS_DAX type in pmem or virtio_fs drivers.
+ */
return NULL;
print_bad_pte(vma, addr, pte, NULL);
@@ -4343,7 +4351,7 @@ static vm_fault_t do_numa_page(struct vm
pte = pte_modify(old_pte, vma->vm_page_prot);
page = vm_normal_page(vma, vmf->address, pte);
- if (!page)
+ if (!page || is_zone_device_page(page))
goto out_map;
/* TODO: handle PTE-mapped THP */
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -526,7 +526,7 @@ static int queue_pages_pte_range(pmd_t *
if (!pte_present(*pte))
continue;
page = vm_normal_page(vma, addr, *pte);
- if (!page)
+ if (!page || is_zone_device_page(page))
continue;
/*
* vm_normal_page() filters out zero pages, but there might
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -1601,7 +1601,7 @@ static int add_page_for_migration(struct
goto out;
err = -ENOENT;
- if (!page)
+ if (!page || is_zone_device_page(page))
goto out;
err = 0;
@@ -1784,7 +1784,7 @@ static void do_pages_stat_array(struct m
if (IS_ERR(page))
goto set_status;
- if (page) {
+ if (page && !is_zone_device_page(page)) {
err = page_to_nid(page);
put_page(page);
} else {
--- a/mm/mlock.c
+++ b/mm/mlock.c
@@ -377,7 +377,8 @@ static unsigned long __munlock_pagevec_f
* Break if page could not be obtained or the page's node+zone does not
* match
*/
- if (!page || page_zone(page) != zone)
+ if (!page || is_zone_device_page(page) ||
+ page_zone(page) != zone)
break;
/*
@@ -440,7 +441,7 @@ void munlock_vma_pages_range(struct vm_a
*/
page = follow_page(vma, start, FOLL_GET | FOLL_DUMP);
- if (page && !IS_ERR(page)) {
+ if (page && !IS_ERR(page) && !is_zone_device_page(page)) {
if (PageTransTail(page)) {
VM_BUG_ON_PAGE(PageMlocked(page), page);
put_page(page); /* follow_page_mask() */
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -89,7 +89,7 @@ static unsigned long change_pte_range(st
continue;
page = vm_normal_page(vma, addr, oldpte);
- if (!page || PageKsm(page))
+ if (!page || is_zone_device_page(page) || PageKsm(page))
continue;
/* Also skip shared copy-on-write pages */