Re: [PATCH v6 5/9] mm: thp: Extend THP to allocate anonymous large folios

[Ryan Roberts]

On 05/10/2023 13:05, Daniel Gomez wrote:
> On Fri, Sep 29, 2023 at 12:44:16PM +0100, Ryan Roberts wrote:
> 
> Hi Ryan,
>> Introduce the logic to allow THP to be configured (through the new
>> anon_orders interface we just added) to allocate large folios to back
>> anonymous memory, which are smaller than PMD-size (for example order-2,
>> order-3, order-4, etc).
>>
>> These THPs continue to be PTE-mapped, but in many cases can still
>> provide similar benefits to traditional PMD-sized THP: Page faults are
>> significantly reduced (by a factor of e.g. 4, 8, 16, etc. depending on
>> the configured order), but latency spikes are much less prominent
>> because the size of each page isn't as huge as the PMD-sized variant and
>> there is less memory to clear in each page fault. The number of per-page
>> operations (e.g. ref counting, rmap management, lru list management) are
>> also significantly reduced since those ops now become per-folio.
>>
>> Some architectures also employ TLB compression mechanisms to squeeze
>> more entries in when a set of PTEs are virtually and physically
>> contiguous and approporiately aligned. In this case, TLB misses will
>> occur less often.
>>
>> The new behaviour is disabled by default because the anon_orders
>> defaults to only enabling PMD-order, but can be enabled at runtime by
>> writing to anon_orders (see documentation in previous commit). The long
>> term aim is to default anon_orders to include suitable lower orders, but
>> there are some risks around internal fragmentation that need to be
>> better understood first.
>>
>> Signed-off-by: Ryan Roberts <ryan.roberts@xxxxxxx>
>> ---
>>  Documentation/admin-guide/mm/transhuge.rst |   9 +-
>>  include/linux/huge_mm.h                    |   6 +-
>>  mm/memory.c                                | 108 +++++++++++++++++++--
>>  3 files changed, 111 insertions(+), 12 deletions(-)
>>
>> diff --git a/Documentation/admin-guide/mm/transhuge.rst b/Documentation/admin-guide/mm/transhuge.rst
>> index 9f954e73a4ca..732c3b2f4ba8 100644
>> --- a/Documentation/admin-guide/mm/transhuge.rst
>> +++ b/Documentation/admin-guide/mm/transhuge.rst
>> @@ -353,7 +353,9 @@ anonymous transparent huge pages, it is necessary to read
>>  ``/proc/PID/smaps`` and count the AnonHugePages and AnonHugePteMap
>>  fields for each mapping. Note that in both cases, AnonHugePages refers
>>  only to PMD-mapped THPs. AnonHugePteMap refers to THPs that are mapped
>> -using PTEs.
>> +using PTEs. This includes all THPs whose order is smaller than
>> +PMD-order, as well as any PMD-order THPs that happen to be PTE-mapped
>> +for other reasons.
>>
>>  The number of file transparent huge pages mapped to userspace is available
>>  by reading ShmemPmdMapped and ShmemHugePages fields in ``/proc/meminfo``.
>> @@ -367,6 +369,11 @@ frequently will incur overhead.
>>  There are a number of counters in ``/proc/vmstat`` that may be used to
>>  monitor how successfully the system is providing huge pages for use.
>>
>> +.. note::
>> +   Currently the below counters only record events relating to
>> +   PMD-order THPs. Events relating to smaller order THPs are not
>> +   included.
>> +
>>  thp_fault_alloc
>>  	is incremented every time a huge page is successfully
>>  	allocated to handle a page fault.
>> diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h
>> index 2e7c338229a6..c4860476a1f5 100644
>> --- a/include/linux/huge_mm.h
>> +++ b/include/linux/huge_mm.h
>> @@ -68,9 +68,11 @@ extern struct kobj_attribute shmem_enabled_attr;
>>  #define HPAGE_PMD_NR (1<<HPAGE_PMD_ORDER)
>>
>>  /*
>> - * Mask of all large folio orders supported for anonymous THP.
>> + * Mask of all large folio orders supported for anonymous THP; all orders up to
>> + * and including PMD_ORDER, except order-0 (which is not "huge") and order-1
>> + * (which is a limitation of the THP implementation).
>>   */
>> -#define THP_ORDERS_ALL_ANON	BIT(PMD_ORDER)
>> +#define THP_ORDERS_ALL_ANON	((BIT(PMD_ORDER + 1) - 1) & ~(BIT(0) | BIT(1)))
>>
>>  /*
>>   * Mask of all large folio orders supported for file THP.
>> diff --git a/mm/memory.c b/mm/memory.c
>> index b5b82fc8e164..92ed9c782dc9 100644
>> --- a/mm/memory.c
>> +++ b/mm/memory.c
>> @@ -4059,6 +4059,87 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>>  	return ret;
>>  }
>>
>> +static bool vmf_pte_range_changed(struct vm_fault *vmf, int nr_pages)
>> +{
>> +	int i;
>> +
>> +	if (nr_pages == 1)
>> +		return vmf_pte_changed(vmf);
>> +
>> +	for (i = 0; i < nr_pages; i++) {
>> +		if (!pte_none(ptep_get_lockless(vmf->pte + i)))
>> +			return true;
>> +	}
>> +
>> +	return false;
>> +}
>> +
>> +#ifdef CONFIG_TRANSPARENT_HUGEPAGE
>> +static struct folio *alloc_anon_folio(struct vm_fault *vmf)
>> +{
>> +	gfp_t gfp;
>> +	pte_t *pte;
>> +	unsigned long addr;
>> +	struct folio *folio;
>> +	struct vm_area_struct *vma = vmf->vma;
>> +	unsigned int orders;
>> +	int order;
>> +
>> +	/*
>> +	 * If uffd is active for the vma we need per-page fault fidelity to
>> +	 * maintain the uffd semantics.
>> +	 */
>> +	if (userfaultfd_armed(vma))
>> +		goto fallback;
>> +
>> +	/*
>> +	 * Get a list of all the (large) orders below PMD_ORDER that are enabled
>> +	 * for this vma. Then filter out the orders that can't be allocated over
>> +	 * the faulting address and still be fully contained in the vma.
>> +	 */
>> +	orders = hugepage_vma_check(vma, vma->vm_flags, false, true, true,
>> +				    BIT(PMD_ORDER) - 1);
>> +	orders = transhuge_vma_suitable(vma, vmf->address, orders);
>> +
>> +	if (!orders)
>> +		goto fallback;
>> +
>> +	pte = pte_offset_map(vmf->pmd, vmf->address & PMD_MASK);
>> +	if (!pte)
>> +		return ERR_PTR(-EAGAIN);
>> +
>> +	order = first_order(orders);
>> +	while (orders) {
>> +		addr = ALIGN_DOWN(vmf->address, PAGE_SIZE << order);
>> +		vmf->pte = pte + pte_index(addr);
>> +		if (!vmf_pte_range_changed(vmf, 1 << order))
>> +			break;
>> +		order = next_order(&orders, order);
>> +	}
>> +
>> +	vmf->pte = NULL;
>> +	pte_unmap(pte);
>> +
>> +	gfp = vma_thp_gfp_mask(vma);
>> +
>> +	while (orders) {
>> +		addr = ALIGN_DOWN(vmf->address, PAGE_SIZE << order);
>> +		folio = vma_alloc_folio(gfp, order, vma, addr, true);
> 
> I was checking your series and noticed about the hugepage flag. I think
> you've changed it from v1 -> v2 from being false to true when orders >=2
> but I'm not sure about the reasoning. Is this because of your statement
> in the cover letter [1]?

That hugepage flags is spec'ed as follows:

 * @hugepage: For hugepages try only the preferred node if possible.

The intent of passing true for orders higher than 0, is that we would prefer to
allocate a smaller order folio that is on the preferred node than a higher order
folio that is not on the preferred node. The assumption is that the on-going
cost of accessing the memory on the non-preferred node will outweigh the benefit
of allocating it as a high order folio.

Thanks,
Ryan


> 
> [1] cover letter snippet:
> 
> "to implement variable order, large folios for anonymous memory.
> (previously called ..., but now exposed as an extension to THP;
> "small-order THP")"
> 
> Thanks,
> Daniel
> 
>> +		if (folio) {
>> +			clear_huge_page(&folio->page, addr, 1 << order);
>> +			return folio;
>> +		}
>> +		order = next_order(&orders, order);
>> +	}
>> +
>> +fallback:
>> +	return vma_alloc_zeroed_movable_folio(vma, vmf->address);
>> +}
>> +#else
>> +#define alloc_anon_folio(vmf) \
>> +		vma_alloc_zeroed_movable_folio((vmf)->vma, (vmf)->address)
>> +#endif
>> +
>>  /*
>>   * We enter with non-exclusive mmap_lock (to exclude vma changes,
>>   * but allow concurrent faults), and pte mapped but not yet locked.
>> @@ -4066,6 +4147,9 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>>   */
>>  static vm_fault_t do_anonymous_page(struct vm_fault *vmf)
>>  {
>> +	int i;
>> +	int nr_pages = 1;
>> +	unsigned long addr = vmf->address;
>>  	bool uffd_wp = vmf_orig_pte_uffd_wp(vmf);
>>  	struct vm_area_struct *vma = vmf->vma;
>>  	struct folio *folio;
>> @@ -4110,10 +4194,15 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)
>>  	/* Allocate our own private page. */
>>  	if (unlikely(anon_vma_prepare(vma)))
>>  		goto oom;
>> -	folio = vma_alloc_zeroed_movable_folio(vma, vmf->address);
>> +	folio = alloc_anon_folio(vmf);
>> +	if (IS_ERR(folio))
>> +		return 0;
>>  	if (!folio)
>>  		goto oom;
>>
>> +	nr_pages = folio_nr_pages(folio);
>> +	addr = ALIGN_DOWN(vmf->address, nr_pages * PAGE_SIZE);
>> +
>>  	if (mem_cgroup_charge(folio, vma->vm_mm, GFP_KERNEL))
>>  		goto oom_free_page;
>>  	folio_throttle_swaprate(folio, GFP_KERNEL);
>> @@ -4130,12 +4219,12 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)
>>  	if (vma->vm_flags & VM_WRITE)
>>  		entry = pte_mkwrite(pte_mkdirty(entry), vma);
>>
>> -	vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd, vmf->address,
>> -			&vmf->ptl);
>> +	vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd, addr, &vmf->ptl);
>>  	if (!vmf->pte)
>>  		goto release;
>> -	if (vmf_pte_changed(vmf)) {
>> -		update_mmu_tlb(vma, vmf->address, vmf->pte);
>> +	if (vmf_pte_range_changed(vmf, nr_pages)) {
>> +		for (i = 0; i < nr_pages; i++)
>> +			update_mmu_tlb(vma, addr + PAGE_SIZE * i, vmf->pte + i);
>>  		goto release;
>>  	}
>>
>> @@ -4150,16 +4239,17 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)
>>  		return handle_userfault(vmf, VM_UFFD_MISSING);
>>  	}
>>
>> -	inc_mm_counter(vma->vm_mm, MM_ANONPAGES);
>> -	folio_add_new_anon_rmap(folio, vma, vmf->address);
>> +	folio_ref_add(folio, nr_pages - 1);
>> +	add_mm_counter(vma->vm_mm, MM_ANONPAGES, nr_pages);
>> +	folio_add_new_anon_rmap(folio, vma, addr);
>>  	folio_add_lru_vma(folio, vma);
>>  setpte:
>>  	if (uffd_wp)
>>  		entry = pte_mkuffd_wp(entry);
>> -	set_pte_at(vma->vm_mm, vmf->address, vmf->pte, entry);
>> +	set_ptes(vma->vm_mm, addr, vmf->pte, entry, nr_pages);
>>
>>  	/* No need to invalidate - it was non-present before */
>> -	update_mmu_cache_range(vmf, vma, vmf->address, vmf->pte, 1);
>> +	update_mmu_cache_range(vmf, vma, addr, vmf->pte, nr_pages);
>>  unlock:
>>  	if (vmf->pte)
>>  		pte_unmap_unlock(vmf->pte, vmf->ptl);
>> --
>> 2.25.1
>>