[PATCH] dma-mapping: use page_is_ram to ensure RAM is not mapped

[Mike Rapoport]

dma_map_resource() uses pfn_valid() to ensure the range is not RAM.
However, pfn_valid() only checks for availability of the memory map for a
PFN but it does not ensure that the PFN is actually backed by RAM.

Replace pfn_valid() with page_is_ram() that does verify whether a PFN is in
RAM or not.

Signed-off-by: Mike Rapoport <rppt@xxxxxxxxxxxxx>
---
 kernel/dma/mapping.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/dma/mapping.c b/kernel/dma/mapping.c
index 7ee5284bff58..dc6064a213a2 100644
--- a/kernel/dma/mapping.c
+++ b/kernel/dma/mapping.c
@@ -296,7 +296,7 @@ dma_addr_t dma_map_resource(struct device *dev, phys_addr_t phys_addr,
 		return DMA_MAPPING_ERROR;
 
 	/* Don't allow RAM to be mapped */
-	if (WARN_ON_ONCE(pfn_valid(PHYS_PFN(phys_addr))))
+	if (WARN_ON_ONCE(page_is_ram(PHYS_PFN(phys_addr))))
 		return DMA_MAPPING_ERROR;
 
 	if (dma_map_direct(dev, ops))
-- 
2.28.0

 
> Given how later we are in the -rc cycle, I suggest we revert Anshuman's
> commit 16c9afc77660 ("arm64/mm: drop HAVE_ARCH_PFN_VALID") and try to
> assess the implications in 5.15 (the patch doesn't seem to have the
> arm64 maintainers' ack anyway ;)).
> 
> Thanks.
> 
> -- 
> Catalin

-- 
Sincerely yours,
Mike.