mm/huge_memory: fix lockdep complaint on 32-bit i_size_read()
commit 006d3ff27e884f80bd7d306b041afc415f63598f upstream.
Huge tmpfs testing, on 32-bit kernel with lockdep enabled, showed that
__split_huge_page() was using i_size_read() while holding the irq-safe
lru_lock and page tree lock, but the 32-bit i_size_read() uses an
irq-unsafe seqlock which should not be nested inside them.
Instead, read the i_size earlier in split_huge_page_to_list(), and pass
the end offset down to __split_huge_page(): all while holding head page
lock, which is enough to prevent truncation of that extent before the
page tree lock has been taken.
Link: http://lkml.kernel.org/r/alpine.LSU.2.11.1811261520070.2275@eggly.anvils
Fixes: baa355fd33142 ("thp: file pages support for split_huge_page()")
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Jerome Glisse <jglisse@redhat.com>
Cc: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: <stable@vger.kernel.org> [4.8+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 5beb62f..7ea8da9 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1925,12 +1925,11 @@
}
static void __split_huge_page(struct page *page, struct list_head *list,
- unsigned long flags)
+ pgoff_t end, unsigned long flags)
{
struct page *head = compound_head(page);
struct zone *zone = page_zone(head);
struct lruvec *lruvec;
- pgoff_t end = -1;
int i;
lruvec = mem_cgroup_page_lruvec(head, zone->zone_pgdat);
@@ -1938,9 +1937,6 @@
/* complete memcg works before add pages to LRU */
mem_cgroup_split_huge_fixup(head);
- if (!PageAnon(page))
- end = DIV_ROUND_UP(i_size_read(head->mapping->host), PAGE_SIZE);
-
for (i = HPAGE_PMD_NR - 1; i >= 1; i--) {
__split_huge_page_tail(head, i, lruvec, list);
/* Some pages can be beyond i_size: drop them from page cache */
@@ -2093,6 +2089,7 @@
int count, mapcount, extra_pins, ret;
bool mlocked;
unsigned long flags;
+ pgoff_t end;
VM_BUG_ON_PAGE(is_huge_zero_page(page), page);
VM_BUG_ON_PAGE(!PageLocked(page), page);
@@ -2114,6 +2111,7 @@
goto out;
}
extra_pins = 0;
+ end = -1;
mapping = NULL;
anon_vma_lock_write(anon_vma);
} else {
@@ -2129,6 +2127,15 @@
extra_pins = HPAGE_PMD_NR;
anon_vma = NULL;
i_mmap_lock_read(mapping);
+
+ /*
+ *__split_huge_page() may need to trim off pages beyond EOF:
+ * but on 32-bit, i_size_read() takes an irq-unsafe seqlock,
+ * which cannot be nested inside the page tree lock. So note
+ * end now: i_size itself may be changed at any moment, but
+ * head page lock is good enough to serialize the trimming.
+ */
+ end = DIV_ROUND_UP(i_size_read(mapping->host), PAGE_SIZE);
}
/*
@@ -2178,7 +2185,7 @@
if (mapping)
__dec_node_page_state(page, NR_SHMEM_THPS);
spin_unlock(&pgdata->split_queue_lock);
- __split_huge_page(page, list, flags);
+ __split_huge_page(page, list, end, flags);
ret = 0;
} else {
if (IS_ENABLED(CONFIG_DEBUG_VM) && mapcount) {