Re: [PATCH]mmap: not merge cloned VMA

[Andi Kleen]

Shaohua Li <shaohua.li@xxxxxxxxx> writes:

> Avoid merging a VMA with another VMA which is cloned from parent process. The
> cloned VMA shares lock with parent process's VMA. If we do the merge, more vma
> area (even the new range is only for current process) uses perent process's
> anon_vma lock, so introduces scalability issues.
> find_mergeable_anon_vma already considers this.

In theory this could prevent quite some VMA merging, but I guess the 
tradeoff is worth it and that should be unlikely to hit anyways.

>  static inline int is_mergeable_anon_vma(struct anon_vma *anon_vma1,
> -					struct anon_vma *anon_vma2)
> +					struct anon_vma *anon_vma2,
> +					struct vm_area_struct *vma)
>  {
> -	return !anon_vma1 || !anon_vma2 || (anon_vma1 == anon_vma2);
> +	if ((!anon_vma1 || !anon_vma2) && (!vma ||
> +		list_is_singular(&vma->anon_vma_chain)))
> +		return 1;

I think this if () needs a comment.

-Andi
-- 
ak@xxxxxxxxxxxxxxx -- Speaking for myself only
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/