Re: [RFC 1/7] mm: introduce MADV_COOL

[Hillf Danton <hdanton@sina.com>]

On Tue, 28 May 2019 18:58:15 +0800 Minchan Kim wrote:
> On Tue, May 28, 2019 at 04:53:01PM +0800, Hillf Danton wrote:
> >
> > On Mon, 20 May 2019 12:52:48 +0900 Minchan Kim wrote:
> > > +static int madvise_cool_pte_range(pmd_t *pmd, unsigned long addr,
> > > +				unsigned long end, struct mm_walk *walk)
> > > +{
> > > +	pte_t *orig_pte, *pte, ptent;
> > > +	spinlock_t *ptl;
> > > +	struct page *page;
> > > +	struct vm_area_struct *vma = walk->vma;
> > > +	unsigned long next;
> > > +
> > > +	next = pmd_addr_end(addr, end);
> > > +	if (pmd_trans_huge(*pmd)) {
> > > +		spinlock_t *ptl;
> >
> > Seems not needed with another ptl declared above.
>
> Will remove it.
>
> > > +
> > > +		ptl = pmd_trans_huge_lock(pmd, vma);
> > > +		if (!ptl)
> > > +			return 0;
> > > +
> > > +		if (is_huge_zero_pmd(*pmd))
> > > +			goto huge_unlock;
> > > +
> > > +		page = pmd_page(*pmd);
> > > +		if (page_mapcount(page) > 1)
> > > +			goto huge_unlock;
> > > +
> > > +		if (next - addr != HPAGE_PMD_SIZE) {
> > > +			int err;
> >
> > Alternately, we deactivate thp only if the address range from userspace
> > is sane enough, in order to avoid complex works we have to do here.
>
> Not sure it's a good idea. That's the way we have done in MADV_FREE
> so want to be consistent.
>
Fair.

> > > +
> > > +			get_page(page);
> > > +			spin_unlock(ptl);
> > > +			lock_page(page);
> > > +			err = split_huge_page(page);
> > > +			unlock_page(page);
> > > +			put_page(page);
> > > +			if (!err)
> > > +				goto regular_page;
> > > +			return 0;
> > > +		}
> > > +
> > > +		pmdp_test_and_clear_young(vma, addr, pmd);
> > > +		deactivate_page(page);
> > > +huge_unlock:
> > > +		spin_unlock(ptl);
> > > +		return 0;
> > > +	}
> > > +
> > > +	if (pmd_trans_unstable(pmd))
> > > +		return 0;
> > > +
> > > +regular_page:
> >
> > Take a look at pending signal?
>
> Do you have any reason to see pending signal here? I want to know what's
> your requirement so that what's the better place to handle it.
>
We could bail out without work done IMO if there is a fatal siganl pending.
And we can do that, if it makes sense to you, before the hard work.

> >
> > > +	orig_pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
> > > +	for (pte = orig_pte; addr < end; pte++, addr += PAGE_SIZE) {
> >
> > s/end/next/ ?
>
> Why do you think it should be next?
>
Simply based on the following line, and afraid that next != end
	> > > +	next = pmd_addr_end(addr, end);

> > > +		ptent = *pte;
> > > +
> > > +		if (pte_none(ptent))
> > > +			continue;
> > > +
> > > +		if (!pte_present(ptent))
> > > +			continue;
> > > +
> > > +		page = vm_normal_page(vma, addr, ptent);
> > > +		if (!page)
> > > +			continue;
> > > +
> > > +		if (page_mapcount(page) > 1)
> > > +			continue;
> > > +
> > > +		ptep_test_and_clear_young(vma, addr, pte);
> > > +		deactivate_page(page);
> > > +	}
> > > +
> > > +	pte_unmap_unlock(orig_pte, ptl);
> > > +	cond_resched();
> > > +
> > > +	return 0;
> > > +}
> > > +
> > > +static long madvise_cool(struct vm_area_struct *vma,
> > > +			unsigned long start_addr, unsigned long end_addr)
> > > +{
> > > +	struct mm_struct *mm = vma->vm_mm;
> > > +	struct mmu_gather tlb;
> > > +
> > > +	if (vma->vm_flags & (VM_LOCKED|VM_HUGETLB|VM_PFNMAP))
> > > +		return -EINVAL;
> >
> > No service in case of VM_IO?
>
> I don't know VM_IO would have regular LRU pages but just follow normal
> convention for DONTNEED and FREE.
> Do you have anything in your mind?
>
I want to skip a mapping set up for DMA.

BR
Hillf