[PATCH v11 14/39] arm64/mm: Map pages for guarded control stack

[Catalin Marinas]

On Thu, Aug 22, 2024 at 02:15:17AM +0100, Mark Brown wrote:
> diff --git a/arch/arm64/include/asm/mman.h b/arch/arm64/include/asm/mman.h
> index c21849ffdd88..37dfd2882f04 100644
> --- a/arch/arm64/include/asm/mman.h
> +++ b/arch/arm64/include/asm/mman.h
> @@ -61,6 +61,15 @@ static inline bool arch_validate_flags(unsigned long vm_flags)
>  			return false;
>  	}
>  
> +	if (system_supports_gcs() && (vm_flags & VM_SHADOW_STACK)) {
> +		/* An executable GCS isn't a good idea. */
> +		if (vm_flags & VM_EXEC)
> +			return false;

Later we should look at clear VM_MAYEXEC in the core code (if the x86
folk agree).

> +
> +		/* The memory management core should prevent this */
> +		VM_WARN_ON(vm_flags & VM_SHARED);
> +	}
> +
>  	return true;
>  
>  }
> diff --git a/arch/arm64/mm/mmap.c b/arch/arm64/mm/mmap.c
> index 642bdf908b22..3ed63fc8cd0a 100644
> --- a/arch/arm64/mm/mmap.c
> +++ b/arch/arm64/mm/mmap.c
> @@ -83,9 +83,17 @@ arch_initcall(adjust_protection_map);
>  
>  pgprot_t vm_get_page_prot(unsigned long vm_flags)
>  {
> -	pteval_t prot = pgprot_val(protection_map[vm_flags &
> +	pteval_t prot;
> +
> +	/* Short circuit GCS to avoid bloating the table. */
> +	if (system_supports_gcs() && (vm_flags & VM_SHADOW_STACK)) {
> +		prot = _PAGE_GCS_RO;
> +	} else {
> +		prot = pgprot_val(protection_map[vm_flags &
>  				   (VM_READ|VM_WRITE|VM_EXEC|VM_SHARED)]);
> +	}
>  
> +	/* VM_ARM64_BTI on a GCS is rejected in arch_validate_flags() */

Not anymore.

>  	if (vm_flags & VM_ARM64_BTI)
>  		prot |= PTE_GP;

Other than the comment above, it looks good.

Reviewed-by: Catalin Marinas <catalin.marinas at arm.com>