From: Andrey Konovalov <andreyknvl@xxxxxxxxxx> Since commit 7a3b83537188 ("kasan: use separate (un)poison implementation for integrated init"), when all init, kasan_has_integrated_init(), and skip_kasan_poison are true, free_pages_prepare() doesn't initialize the page. This is wrong. Fix it by remembering whether kasan_poison_pages() performed initialization, and call kernel_init_free_pages() if it didn't. Fixes: 7a3b83537188 ("kasan: use separate (un)poison implementation for integrated init") Signed-off-by: Andrey Konovalov <andreyknvl@xxxxxxxxxx> --- mm/page_alloc.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 0673db27dd12..2ada09a58e4b 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1360,9 +1360,14 @@ static __always_inline bool free_pages_prepare(struct page *page, * With hardware tag-based KASAN, memory tags must be set before the * page becomes unavailable via debug_pagealloc or arch_free_page. */ - if (!skip_kasan_poison) + if (!skip_kasan_poison) { kasan_poison_pages(page, order, init); - if (init && !kasan_has_integrated_init()) + + /* Memory is already initialized if KASAN did it internally. */ + if (kasan_has_integrated_init()) + init = false; + } + if (init) kernel_init_free_pages(page, 1 << order); /* -- 2.25.1