On 2/1/2023 5:09 PM, David Hildenbrand wrote:
> On 01.02.23 09:17, Yin Fengwei wrote:
>> do_set_pte_range() allows to setup page table entries for a
>> specific range. It calls page_add_file_rmap_range() to take
>> advantage of batched rmap update for large folio.
>>
>> Signed-off-by: Yin Fengwei <fengwei.yin@xxxxxxxxx>
>> ---
>> include/linux/mm.h | 2 ++
>> mm/filemap.c | 1 -
>> mm/memory.c | 60 ++++++++++++++++++++++++++++++++++++----------
>> 3 files changed, 49 insertions(+), 14 deletions(-)
>>
>> diff --git a/include/linux/mm.h b/include/linux/mm.h
>> index d6f8f41514cc..96e08fcdce24 100644
>> --- a/include/linux/mm.h
>> +++ b/include/linux/mm.h
>> @@ -1162,6 +1162,8 @@ static inline pte_t maybe_mkwrite(pte_t pte, struct vm_area_struct *vma)
>> vm_fault_t do_set_pmd(struct vm_fault *vmf, struct page *page);
>> void do_set_pte(struct vm_fault *vmf, struct page *page, unsigned long addr);
>> +void do_set_pte_range(struct vm_fault *vmf, struct folio *folio,
>> + unsigned long start, unsigned long addr, unsigned int nr);
>> vm_fault_t finish_fault(struct vm_fault *vmf);
>> vm_fault_t finish_mkwrite_fault(struct vm_fault *vmf);
>> diff --git a/mm/filemap.c b/mm/filemap.c
>> index 9cc5edd8f998..95f634d11581 100644
>> --- a/mm/filemap.c
>> +++ b/mm/filemap.c
>> @@ -3386,7 +3386,6 @@ static vm_fault_t filemap_map_folio_range(struct vm_fault *vmf,
>> ref_count++;
>> do_set_pte(vmf, page, addr);
>> - update_mmu_cache(vma, addr, vmf->pte);
>> } while (vmf->pte++, page++, addr += PAGE_SIZE, ++count < nr_pages);
>> /*
>> diff --git a/mm/memory.c b/mm/memory.c
>> index 51c04bb60724..7e41142e1e4f 100644
>> --- a/mm/memory.c
>> +++ b/mm/memory.c
>> @@ -4257,7 +4257,8 @@ vm_fault_t do_set_pmd(struct vm_fault *vmf, struct page *page)
>> }
>> #endif
>> -void do_set_pte(struct vm_fault *vmf, struct page *page, unsigned long addr)
>> +static void do_set_pte_entry(struct vm_fault *vmf, struct page *page,
>> + unsigned long addr)
>> {
>> struct vm_area_struct *vma = vmf->vma;
>> bool uffd_wp = pte_marker_uffd_wp(vmf->orig_pte);
>> @@ -4277,16 +4278,52 @@ void do_set_pte(struct vm_fault *vmf, struct page *page, unsigned long addr)
>> entry = maybe_mkwrite(pte_mkdirty(entry), vma);
>> if (unlikely(uffd_wp))
>> entry = pte_mkuffd_wp(entry);
>> - /* copy-on-write page */
>> - if (write && !(vma->vm_flags & VM_SHARED)) {
>> - inc_mm_counter(vma->vm_mm, MM_ANONPAGES);
>> - page_add_new_anon_rmap(page, vma, addr);
>> - lru_cache_add_inactive_or_unevictable(page, vma);
>> - } else {
>> - inc_mm_counter(vma->vm_mm, mm_counter_file(page));
>> - page_add_file_rmap(page, vma, false);
>> - }
>> set_pte_at(vma->vm_mm, addr, vmf->pte, entry);
>> +
>> + /* no need to invalidate: a not-present page won't be cached */
>> + update_mmu_cache(vma, addr, vmf->pte);
>> +}
>> +
>> +void do_set_pte_range(struct vm_fault *vmf, struct folio *folio,
>> + unsigned long start, unsigned long addr, unsigned int nr)
>> +{
>> + unsigned int i = 0;
>> + struct page *page = folio_page(folio, start);
>> + struct vm_area_struct *vma = vmf->vma;
>> + bool cow = (vmf->flags & FAULT_FLAG_WRITE) &&
>> + !(vma->vm_flags & VM_SHARED);
>> +
>> + /*
>> + * file page: batched update rmap, mm counter.
>> + * copy-on-write page: batched update mm counter.
>> + */
>> + if (!cow) {
>> + page_add_file_rmap_range(folio, start, nr, vma, false);
>> + add_mm_counter(vma->vm_mm, mm_counter_file(page), nr);
>> + } else
>> + add_mm_counter(vma->vm_mm, MM_ANONPAGES, nr);
>> +
>> + do {
>> + if (cow) {
>> + page_add_new_anon_rmap(page, vma, addr);
>
> This doesn't work with anon pages the way you intended in this patch.
>
> Please leave anon pages out of the picture for now and make do_set_pte_range() only deal with !cow.
OK. I will move the check to do_set_pte() and only call to
do_set_pte_range() when it's !cow. Thanks.
Regards
Yin, Fengwei
>
