Re: [PATCH] arm64: Change elfcore for_each_mte_vma() to use VMA iterator

[Catalin Marinas]

On Fri, Feb 18, 2022 at 02:37:04AM +0000, Liam Howlett wrote:
> diff --git a/arch/arm64/kernel/elfcore.c b/arch/arm64/kernel/elfcore.c
> index 3455ee4acc04..930a0bc4cac4 100644
> --- a/arch/arm64/kernel/elfcore.c
> +++ b/arch/arm64/kernel/elfcore.c
> @@ -8,9 +8,9 @@
>  #include <asm/cpufeature.h>
>  #include <asm/mte.h>
>  
> -#define for_each_mte_vma(tsk, vma)					\
> +#define for_each_mte_vma(vmi, vma)					\
>  	if (system_supports_mte())					\
> -		for (vma = tsk->mm->mmap; vma; vma = vma->vm_next)	\
> +		for_each_vma(vmi, vma)					\
>  			if (vma->vm_flags & VM_MTE)
>  
>  static unsigned long mte_vma_tag_dump_size(struct vm_area_struct *vma)
> @@ -65,8 +65,9 @@ Elf_Half elf_core_extra_phdrs(void)
>  {
>  	struct vm_area_struct *vma;
>  	int vma_count = 0;
> +	VMA_ITERATOR(vmi, current->mm, 0);
>  
> -	for_each_mte_vma(current, vma)
> +	for_each_mte_vma(vmi, vma)
>  		vma_count++;

I'm fine with the patch but it can't be applied to arm64 for-next/mte
branch as it won't build and the maple tree doesn't have the MTE
patches. Do you have a stable branch with the for_each_vma() iterator?

Anyway, if you find a way to solve the problem, for this patch:

Acked-by: Catalin Marinas <catalin.marinas@xxxxxxx>