Re: [PATCH v2 4/5] KVM: s390: refactor and split some gmap helpers

[Heiko Carstens <hca@linux.ibm.com>]

On Tue, May 20, 2025 at 08:26:38PM +0200, Claudio Imbrenda wrote:
> +void __gmap_helper_set_unused(struct mm_struct *mm, unsigned long vmaddr)
> +{
> +	spinlock_t *ptl;
> +	pmd_t *pmdp;
> +	pte_t *ptep;
> +
> +	mmap_assert_locked(mm);
> +
> +	if (pmd_lookup(mm, vmaddr, &pmdp))
> +		return;
> +	ptl = pmd_lock(mm, pmdp);
> +	if (!pmd_present(*pmdp) || pmd_leaf(*pmdp)) {
> +		spin_unlock(ptl);
> +		return;
> +	}
> +	spin_unlock(ptl);
> +
> +	ptep = pte_offset_map_lock(mm, pmdp, vmaddr, &ptl);
> +	if (!ptep)
> +		return;
> +	/* The last byte of a pte can be changed freely without ipte */
> +	__atomic64_or(_PAGE_UNUSED, (long *)ptep);
> +	pte_unmap_unlock(ptep, ptl);
> +}
> +EXPORT_SYMBOL_GPL(__gmap_helper_set_unused);

This is unused, as far as I can tell. I'm not sure if it is a good approach to
do all this code movements / refactorings now. Especially if you also add
(now?) dead code. I guess that e.g. this function is required for your rework
that will come later?

Imho this series causes quite a bit confusion and is not about "cleanups and
small fixes" like advertised.