4.4-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Shaokun Zhang <zhangshao...@hisilicon.com>

commit 20c27a4270c775d7ed661491af8ac03264d60fc6 upstream.

__sync_icache_dcache unconditionally skips the cache maintenance for
anonymous pages, under the assumption that flushing is only required in
the presence of D-side aliases [see 7249b79f6b4cc ("arm64: Do not flush
the D-cache for anonymous pages")].

Unfortunately, this breaks migration of anonymous pages holding
self-modifying code, where userspace cannot be reasonably expected to
reissue maintenance instructions in response to a migration.

This patch fixes the problem by removing the broken page_mapping(page)
check from the cache syncing code, otherwise we may end up fetching and
executing stale instructions from the PoU.

Cc: Catalin Marinas <catalin.mari...@arm.com>
Cc: Will Deacon <will.dea...@arm.com>
Cc: Mark Rutland <mark.rutl...@arm.com>
Cc: <sta...@vger.kernel.org>
Reviewed-by: Catalin Marinas <catalin.mari...@arm.com>
Signed-off-by: Shaokun Zhang <zhangshao...@hisilicon.com>
Signed-off-by: Will Deacon <will.dea...@arm.com>
Signed-off-by: Sasha Levin <sasha.le...@oracle.com>
Cc: Amanieu d'Antras <aman...@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gre...@linuxfoundation.org>

---
 arch/arm64/mm/flush.c |    4 ----
 1 file changed, 4 deletions(-)

--- a/arch/arm64/mm/flush.c
+++ b/arch/arm64/mm/flush.c
@@ -70,10 +70,6 @@ void __sync_icache_dcache(pte_t pte, uns
 {
        struct page *page = pte_page(pte);
 
-       /* no flushing needed for anonymous pages */
-       if (!page_mapping(page))
-               return;
-
        if (!test_and_set_bit(PG_dcache_clean, &page->flags)) {
                __flush_dcache_area(page_address(page),
                                PAGE_SIZE << compound_order(page));


Reply via email to