Re: [PATCH v3 01/26] set_memory: add folio_{zap,restore}_direct_map helpers

[Mike Rapoport]

Hi Brendan,

On Sun, Jul 26, 2026 at 10:22:34PM +0000, Brendan Jackman wrote:
> From: Nikita Kalyazin <nikita.kalyazin@xxxxxxxxx>
> 
> Let's provide folio_{zap,restore}_direct_map helpers as preparation for
> supporting removal of the direct map for guest_memfd folios.
> In folio_zap_direct_map(), flush TLB to make sure the data is not
> accessible.  On some architectures, there may be a double TLB flush
> issued because set_direct_map_valid_noflush already performs a flush
> internally.
> 
> The new helpers need to be accessible to KVM on architectures that
> support guest_memfd (x86 and arm64).
> 
> Direct map removal gives guest_memfd the same protection that
> memfd_secret does, such as hardening against Spectre-like attacks
> through in-kernel gadgets.
> 
> Acked-by: David Hildenbrand (Arm) <david@xxxxxxxxxx>
> Signed-off-by: Nikita Kalyazin <nikita.kalyazin@xxxxxxxxx>
> [Added comment, dropped modified set_direct_map API, added highmem check]
> Signed-off-by: Brendan Jackman <jackmanb@xxxxxxxxxx>
> ---
>  include/linux/set_memory.h | 13 +++++++++++++
>  mm/memory.c                | 46 ++++++++++++++++++++++++++++++++++++++++++++++
>  2 files changed, 59 insertions(+)
> 
> diff --git a/include/linux/set_memory.h b/include/linux/set_memory.h
> index 3030d9245f5ac..1bf2a15bca118 100644
> --- a/include/linux/set_memory.h
> +++ b/include/linux/set_memory.h
> @@ -40,6 +40,15 @@ static inline int set_direct_map_valid_noflush(struct page *page,
>  	return 0;
>  }
>  
> +static inline int folio_zap_direct_map(struct folio *folio)
> +{
> +	return 0;
> +}
> +
> +static inline void folio_restore_direct_map(struct folio *folio)
> +{
> +}
> +
>  static inline bool kernel_page_present(struct page *page)
>  {
>  	return true;
> @@ -56,6 +65,10 @@ static inline bool can_set_direct_map(void)
>  }
>  #define can_set_direct_map can_set_direct_map
>  #endif
> +
> +int folio_zap_direct_map(struct folio *folio);
> +void folio_restore_direct_map(struct folio *folio);
> +
>  #endif /* CONFIG_ARCH_HAS_SET_DIRECT_MAP */
>  
>  #ifdef CONFIG_X86_64
> diff --git a/mm/memory.c b/mm/memory.c
> index a73af1fccb3d0..789c65a6d6a0e 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -78,6 +78,7 @@
>  #include <linux/sched/sysctl.h>
>  #include <linux/pgalloc.h>
>  #include <linux/uaccess.h>
> +#include <linux/set_memory.h>
>  
>  #include <trace/events/kmem.h>
>  
> @@ -7758,3 +7759,48 @@ void vma_pgtable_walk_end(struct vm_area_struct *vma)
>  	if (is_vm_hugetlb_page(vma))
>  		hugetlb_vma_unlock_read(vma);
>  }
> +
> +#ifdef CONFIG_ARCH_HAS_SET_DIRECT_MAP
> +/**
> + * folio_zap_direct_map - remove a folio from the kernel direct map
> + * @folio: folio to remove from the direct map
> + *
> + * Removes the folio from the kernel direct map and flushes the TLB.  This may
> + * require splitting huge pages in the direct map, which can fail due to memory
> + * allocation.  So far, only order-0 folios are supported; this guarantees
> + * the unmap is either a complete success or a total failure.
> + *
> + * Return: 0 on success, or a negative error code on failure.
> + */
> +int folio_zap_direct_map(struct folio *folio)
> +{
> +	struct page *page = folio_page(folio, 0);
> +	unsigned long addr = (unsigned long)page_address(page);
> +	int ret;
> +
> +	if (folio_test_large(folio) || folio_test_highmem(folio))
> +		return -EINVAL;
> +
> +	ret = set_direct_map_valid_noflush(page, 1, false);

There was a discussion about  slight differences in the semantics of
set_direct_map_valid() on x86 and on arm64 and that execmem should
apparently switch to set_direct_map_{invalid,default}.

Maybe for this series it would be better to add a patch that adds numpages
to set_direct_map_{invalid,default}_noflush and use
set_direct_map_default_noflush() here?

And maybe also pick another Nikita's patch [2] that makes set_direct_map_*
to take address?

[1] https://lore.kernel.org/all/DJ69RCVRBO0Y.3JCYSW50IC4RC@xxxxxxxxx/
[2] https://lore.kernel.org/all/20260410151746.61150-2-kalyazin@xxxxxxxxxx/

> +	flush_tlb_kernel_range(addr, addr + folio_size(folio));
> +
> +	return ret;
> +}
> +EXPORT_SYMBOL_FOR_MODULES(folio_zap_direct_map, "kvm");
> +
> +/**
> + * folio_restore_direct_map - restore the kernel direct map entry for a folio
> + * @folio: folio whose direct map entry is to be restored
> + *
> + * This may only be called after a prior successful folio_zap_direct_map() on
> + * the same folio.  Because the zap will have already split any huge pages in
> + * the direct map, restoration here only updates protection bits and cannot
> + * fail.
> + */
> +void folio_restore_direct_map(struct folio *folio)
> +{
> +	WARN_ON_ONCE(set_direct_map_valid_noflush(folio_page(folio, 0),
> +						  folio_nr_pages(folio), true));
> +}
> +EXPORT_SYMBOL_FOR_MODULES(folio_restore_direct_map, "kvm");
> +#endif /* CONFIG_ARCH_HAS_SET_DIRECT_MAP */
> 
> -- 
> 2.54.0
> 

-- 
Sincerely yours,
Mike.