Re: [PATCH v3 1/2] mm: Abstract THP allocation

[David Hildenbrand]

On 11.09.24 11:27, David Hildenbrand wrote:
> On 11.09.24 08:55, Dev Jain wrote:
> > In preparation for the second patch, abstract away the THP allocation
> > logic present in the create_huge_pmd() path, which corresponds to the
> > faulting case when no page is present.
> >
> > There should be no functional change as a result of applying
> > this patch.
>
> Hi,
>
> > Signed-off-by: Dev Jain <dev.jain@xxxxxxx>
> > ---
> >    mm/huge_memory.c | 110 +++++++++++++++++++++++++++++------------------
> >    1 file changed, 67 insertions(+), 43 deletions(-)
> >
> > diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> > index 67c86a5d64a6..b96a1ff2bf40 100644
> > --- a/mm/huge_memory.c
> > +++ b/mm/huge_memory.c
> > @@ -943,47 +943,88 @@ unsigned long thp_get_unmapped_area(struct file *filp, unsigned long addr,
> >    }
> >    EXPORT_SYMBOL_GPL(thp_get_unmapped_area);
> >    
> > -static vm_fault_t __do_huge_pmd_anonymous_page(struct vm_fault *vmf,
> > -			struct page *page, gfp_t gfp)
> > +static struct folio *pmd_thp_fault_alloc(gfp_t gfp, struct vm_area_struct *vma,
> > +					 unsigned long haddr, unsigned long addr)
>
> I suggest calling this something like "vma_alloc_anon_folio_pmd()"? Then
> it's more consistent with vma_alloc_folio().
>
> Also, likely we should just only pass in "addr" and calculate "haddr"
> ourselves, it's cheap and reduces the number of function parameters.
>
> >    {
> > -	struct vm_area_struct *vma = vmf->vma;
> > -	struct folio *folio = page_folio(page);
> > -	pgtable_t pgtable;
> > -	unsigned long haddr = vmf->address & HPAGE_PMD_MASK;
> > -	vm_fault_t ret = 0;
> > +	const int order = HPAGE_PMD_ORDER;
> > +	struct folio *folio = vma_alloc_folio(gfp, order, vma, haddr, true);
> >    
> > -	VM_BUG_ON_FOLIO(!folio_test_large(folio), folio);
> > +	if (unlikely(!folio)) {
> > +		count_vm_event(THP_FAULT_FALLBACK);
> > +		count_mthp_stat(order, MTHP_STAT_ANON_FAULT_FALLBACK);
> > +		goto out;
> > +	}
> >    
> > +	VM_BUG_ON_FOLIO(!folio_test_large(folio), folio);
> >    	if (mem_cgroup_charge(folio, vma->vm_mm, gfp)) {
> >    		folio_put(folio);
> >    		count_vm_event(THP_FAULT_FALLBACK);
> >    		count_vm_event(THP_FAULT_FALLBACK_CHARGE);
> > -		count_mthp_stat(HPAGE_PMD_ORDER, MTHP_STAT_ANON_FAULT_FALLBACK);
> > -		count_mthp_stat(HPAGE_PMD_ORDER, MTHP_STAT_ANON_FAULT_FALLBACK_CHARGE);
> > -		return VM_FAULT_FALLBACK;
> > +		count_mthp_stat(order, MTHP_STAT_ANON_FAULT_FALLBACK);
> > +		count_mthp_stat(order, MTHP_STAT_ANON_FAULT_FALLBACK_CHARGE);
> > +		goto out;
> >    	}
> >    	folio_throttle_swaprate(folio, gfp);
> >    
> > -	pgtable = pte_alloc_one(vma->vm_mm);
> > -	if (unlikely(!pgtable)) {
> > -		ret = VM_FAULT_OOM;
> > -		goto release;
> > -	}
> > -
> > -	folio_zero_user(folio, vmf->address);
> > +	folio_zero_user(folio, addr);
> >    	/*
> >    	 * The memory barrier inside __folio_mark_uptodate makes sure that
> >    	 * folio_zero_user writes become visible before the set_pmd_at()
> >    	 * write.
> >    	 */
> >    	__folio_mark_uptodate(folio);
> > +out:
> > +	return folio;
> > +}
> > +
> > +static void __pmd_thp_fault_success_stats(struct vm_area_struct *vma)
> > +{
> > +	count_vm_event(THP_FAULT_ALLOC);
> > +	count_mthp_stat(HPAGE_PMD_ORDER, MTHP_STAT_ANON_FAULT_ALLOC);
> > +	count_memcg_event_mm(vma->vm_mm, THP_FAULT_ALLOC);
> > +}
>
> Why isn't that moved into map_pmd_thp()
>
> Note that in this patch you do:
>
> map_pmd_thp(folio, vmf, vma, haddr);
> spin_unlock(vmf->ptl);
> __pmd_thp_fault_success_stats(vma);
>
> But in patch #2
>
> map_pmd_thp(folio, vmf, vma, haddr);
> __pmd_thp_fault_success_stats(vma);
> goto unlock;
> release:
> 	folio_put(folio);
> unlock:
> 	spin_unlock(vmf->ptl);
>
> Please make that consistent, meaning:
>
> 1) Inline __pmd_thp_fault_success_stats() into map_pmd_thp(). No need to
> have the separated out.
>
> 2) Either do the PTL unlocking in __pmd_thp_fault_success_stats() or in
>      the caller. In the caller is likely easiest. Adjusting the counters
>      should be cheap, if not we could revisit this later with real data.
>
> > +
> > +static void map_pmd_thp(struct folio *folio, struct vm_fault *vmf,
> > +			struct vm_area_struct *vma, unsigned long haddr)
> > +{
> > +	pmd_t entry;
> > +
> > +	entry = mk_huge_pmd(&folio->page, vma->vm_page_prot);
> > +	entry = maybe_pmd_mkwrite(pmd_mkdirty(entry), vma);
> > +	folio_add_new_anon_rmap(folio, vma, haddr, RMAP_EXCLUSIVE);
> > +	folio_add_lru_vma(folio, vma);
> > +	set_pmd_at(vma->vm_mm, haddr, vmf->pmd, entry);
> > +	update_mmu_cache_pmd(vma, vmf->address, vmf->pmd);
>
> It's quite weird to see a mixture of haddr and vmf->address, and likely
> this mixture is wrong or not not required.
>
> Looking at arc's update_mmu_cache_pmd() implementation, I cannot see how
> passing in the unaligned address would do the right thing. But maybe arc
> also doesn't trigger that code path ... who knows :)
>
>
> Staring at some other update_mmu_cache_pmd() users, it's quite
> inconsistent. Primarily only do_huge_pmd_numa_page() and
> __do_huge_pmd_anonymous_page() use the unaligned address. The others
> seem to use the aligned address ... as one would expect when modifying a
> PMD.
>
>
> I suggest to change this function to *not* pass in the vmf, and rename
> it to something like:
>
> static void folio_map_anon_pmd(struct folio *folio, struct
> vm_area_struct *vma, pmd_t *pmd, unsigned long haddr)

... or better "map_anon_folio_pmd" so it better matches 
vma_alloc_folio_pmd() suggested above.

Could also do

vma_map_anon_folio_pmd() :)

--
Cheers,

David / dhildenb