Re: [PATCH 1/6] arm64: pgtable: Fix pte_accessible()

[Yu Zhao]

On Fri, Nov 20, 2020 at 02:35:52PM +0000, Will Deacon wrote:
> pte_accessible() is used by ptep_clear_flush() to figure out whether TLB
> invalidation is necessary when unmapping pages for reclaim. Although our
> implementation is correct according to the architecture, returning true
> only for valid, young ptes in the absence of racing page-table
> modifications, this is in fact flawed due to lazy invalidation of old
> ptes in ptep_clear_flush_young() where we elide the expensive DSB
> instruction for completing the TLB invalidation.
> 
> Rather than penalise the aging path, adjust pte_accessible() to return
> true for any valid pte, even if the access flag is cleared.

The chance of a system hitting reclaim is proportional to how long
it's been running, and that of having mapped but yet to be accessed
PTEs is reciprocal, so to speak. I don't reboot my devices everyday,
and therefore:

Acked-by: Yu Zhao <yuzhao@xxxxxxxxxx>

> Cc: <stable@xxxxxxxxxxxxxxx>
> Fixes: 76c714be0e5e ("arm64: pgtable: implement pte_accessible()")
> Reported-by: Yu Zhao <yuzhao@xxxxxxxxxx>
> Signed-off-by: Will Deacon <will@xxxxxxxxxx>
> ---
>  arch/arm64/include/asm/pgtable.h | 4 +---
>  1 file changed, 1 insertion(+), 3 deletions(-)
> 
> diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h
> index 4ff12a7adcfd..1bdf51f01e73 100644
> --- a/arch/arm64/include/asm/pgtable.h
> +++ b/arch/arm64/include/asm/pgtable.h
> @@ -115,8 +115,6 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
>  #define pte_valid(pte)		(!!(pte_val(pte) & PTE_VALID))
>  #define pte_valid_not_user(pte) \
>  	((pte_val(pte) & (PTE_VALID | PTE_USER)) == PTE_VALID)
> -#define pte_valid_young(pte) \
> -	((pte_val(pte) & (PTE_VALID | PTE_AF)) == (PTE_VALID | PTE_AF))
>  #define pte_valid_user(pte) \
>  	((pte_val(pte) & (PTE_VALID | PTE_USER)) == (PTE_VALID | PTE_USER))
>  
> @@ -126,7 +124,7 @@ extern unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)];
>   * remapped as PROT_NONE but are yet to be flushed from the TLB.
>   */
>  #define pte_accessible(mm, pte)	\
> -	(mm_tlb_flush_pending(mm) ? pte_present(pte) : pte_valid_young(pte))
> +	(mm_tlb_flush_pending(mm) ? pte_present(pte) : pte_valid(pte))
>  
>  /*
>   * p??_access_permitted() is true for valid user mappings (subject to the
> -- 
> 2.29.2.454.gaff20da3a2-goog
>