[PATCH v1 3/4] arm64: decouple check whether pfn is in linear map from pfn_valid()

[David Hildenbrand]

On 20.04.21 11:09, Mike Rapoport wrote:
> From: Mike Rapoport <rppt at linux.ibm.com>
> 
> The intended semantics of pfn_valid() is to verify whether there is a
> struct page for the pfn in question and nothing else.
> 
> Yet, on arm64 it is used to distinguish memory areas that are mapped in the
> linear map vs those that require ioremap() to access them.
> 
> Introduce a dedicated pfn_is_map_memory() wrapper for
> memblock_is_map_memory() to perform such check and use it where
> appropriate.
> 
> Using a wrapper allows to avoid cyclic include dependencies.
> 
> Signed-off-by: Mike Rapoport <rppt at linux.ibm.com>
> ---
>   arch/arm64/include/asm/memory.h | 2 +-
>   arch/arm64/include/asm/page.h   | 1 +
>   arch/arm64/kvm/mmu.c            | 2 +-
>   arch/arm64/mm/init.c            | 6 ++++++
>   arch/arm64/mm/ioremap.c         | 4 ++--
>   arch/arm64/mm/mmu.c             | 2 +-
>   6 files changed, 12 insertions(+), 5 deletions(-)
> 
> diff --git a/arch/arm64/include/asm/memory.h b/arch/arm64/include/asm/memory.h
> index 0aabc3be9a75..194f9f993d30 100644
> --- a/arch/arm64/include/asm/memory.h
> +++ b/arch/arm64/include/asm/memory.h
> @@ -351,7 +351,7 @@ static inline void *phys_to_virt(phys_addr_t x)
>   
>   #define virt_addr_valid(addr)	({					\
>   	__typeof__(addr) __addr = __tag_reset(addr);			\
> -	__is_lm_address(__addr) && pfn_valid(virt_to_pfn(__addr));	\
> +	__is_lm_address(__addr) && pfn_is_map_memory(virt_to_pfn(__addr));	\
>   })
>   
>   void dump_mem_limit(void);
> diff --git a/arch/arm64/include/asm/page.h b/arch/arm64/include/asm/page.h
> index 012cffc574e8..99a6da91f870 100644
> --- a/arch/arm64/include/asm/page.h
> +++ b/arch/arm64/include/asm/page.h
> @@ -38,6 +38,7 @@ void copy_highpage(struct page *to, struct page *from);
>   typedef struct page *pgtable_t;
>   
>   extern int pfn_valid(unsigned long);
> +extern int pfn_is_map_memory(unsigned long);
>   
>   #include <asm/memory.h>
>   
> diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c
> index 8711894db8c2..23dd99e29b23 100644
> --- a/arch/arm64/kvm/mmu.c
> +++ b/arch/arm64/kvm/mmu.c
> @@ -85,7 +85,7 @@ void kvm_flush_remote_tlbs(struct kvm *kvm)
>   
>   static bool kvm_is_device_pfn(unsigned long pfn)
>   {
> -	return !pfn_valid(pfn);
> +	return !pfn_is_map_memory(pfn);
>   }
>   
>   /*
> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
> index 3685e12aba9b..c54e329aca15 100644
> --- a/arch/arm64/mm/init.c
> +++ b/arch/arm64/mm/init.c
> @@ -258,6 +258,12 @@ int pfn_valid(unsigned long pfn)
>   }
>   EXPORT_SYMBOL(pfn_valid);
>   
> +int pfn_is_map_memory(unsigned long pfn)
> +{

I think you might have to add (see pfn_valid())

if (PHYS_PFN(PFN_PHYS(pfn)) != pfn)
	return 0;

to catch false positives.

> +	return memblock_is_map_memory(PFN_PHYS(pfn));
> +}
> +EXPORT_SYMBOL(pfn_is_map_memory);
> +
>   static phys_addr_t memory_limit = PHYS_ADDR_MAX;
>   
>   /*
> diff --git a/arch/arm64/mm/ioremap.c b/arch/arm64/mm/ioremap.c
> index b5e83c46b23e..b7c81dacabf0 100644
> --- a/arch/arm64/mm/ioremap.c
> +++ b/arch/arm64/mm/ioremap.c
> @@ -43,7 +43,7 @@ static void __iomem *__ioremap_caller(phys_addr_t phys_addr, size_t size,
>   	/*
>   	 * Don't allow RAM to be mapped.
>   	 */
> -	if (WARN_ON(pfn_valid(__phys_to_pfn(phys_addr))))
> +	if (WARN_ON(pfn_is_map_memory(__phys_to_pfn(phys_addr))))
>   		return NULL;
>   
>   	area = get_vm_area_caller(size, VM_IOREMAP, caller);
> @@ -84,7 +84,7 @@ EXPORT_SYMBOL(iounmap);
>   void __iomem *ioremap_cache(phys_addr_t phys_addr, size_t size)
>   {
>   	/* For normal memory we already have a cacheable mapping. */
> -	if (pfn_valid(__phys_to_pfn(phys_addr)))
> +	if (pfn_is_map_memory(__phys_to_pfn(phys_addr)))
>   		return (void __iomem *)__phys_to_virt(phys_addr);
>   
>   	return __ioremap_caller(phys_addr, size, __pgprot(PROT_NORMAL),
> diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c
> index 5d9550fdb9cf..26045e9adbd7 100644
> --- a/arch/arm64/mm/mmu.c
> +++ b/arch/arm64/mm/mmu.c
> @@ -81,7 +81,7 @@ void set_swapper_pgd(pgd_t *pgdp, pgd_t pgd)
>   pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
>   			      unsigned long size, pgprot_t vma_prot)
>   {
> -	if (!pfn_valid(pfn))
> +	if (!pfn_is_map_memory(pfn))
>   		return pgprot_noncached(vma_prot);
>   	else if (file->f_flags & O_SYNC)
>   		return pgprot_writecombine(vma_prot);
> 

As discussed, in the future it would be nice if we could just rely on 
the memmap state. There are cases where pfn_is_map_memory() will now be 
slower than pfn_valid() -- e.g., we don't check for valid_section() in 
case of CONFIG_SPARSEMEM. This would apply where pfn_valid() would have 
returned "0".

As we're not creating the direct map, kern_addr_valid() shouldn't need 
love. It'd be some kind of ugly if some generic code used by arm64 would 
be relying in case of arm64 on pfn_valid() to return the expected 
result; I doubt it.

Acked-by: David Hildenbrand <david at redhat.com>

-- 
Thanks,

David / dhildenb