linux-kernel - Re: [PATCH] mm: Compute mTHP order efficiently

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAGsJ_4z9Bus2=A7bdyEGx+aNake9Le-a7f8=rJZQoGT+9Gn3ZQ@mail.gmail.com>
Date: Mon, 16 Sep 2024 13:58:32 +0800
From: Barry Song <baohua@...nel.org>
To: Dev Jain <dev.jain@....com>
Cc: akpm@...ux-foundation.org, david@...hat.com, willy@...radead.org, 
	ryan.roberts@....com, anshuman.khandual@....com, hughd@...gle.com, 
	ioworker0@...il.com, wangkefeng.wang@...wei.com, 
	baolin.wang@...ux.alibaba.com, gshan@...hat.com, linux-kernel@...r.kernel.org, 
	linux-mm@...ck.org
Subject: Re: [PATCH] mm: Compute mTHP order efficiently

On Mon, Sep 16, 2024 at 1:20 PM Dev Jain <dev.jain@....com> wrote:
>
>
> On 9/16/24 10:42, Barry Song wrote:
> > On Fri, Sep 13, 2024 at 5:19 PM Dev Jain <dev.jain@....com> wrote:
> >> We use pte_range_none() to determine whether contiguous PTEs are empty
> >> for an mTHP allocation. Instead of iterating the while loop for every
> >> order, use some information, which is the first set PTE found, from the
> >> previous iteration, to eliminate some cases. The key to understanding
> >> the correctness of the patch is that the ranges we want to examine
> >> form a strictly decreasing sequence of nested intervals.
> >>
> >> Suggested-by: Ryan Roberts <ryan.roberts@....com>
> >> Signed-off-by: Dev Jain <dev.jain@....com>
> > I like this patch, but could we come up with a better subject for
> > pte_range_none()?
> > The subject is really incorrect.
>
> Are you asking me to change "Compute mTHP order efficiently" to
> something else?

Right.

Adjust the subject to more accurately reflect the specific changes
being made.

>
> >
> > Also, I'd prefer the change for alloc_anon_folio() to be separated
> > into its own patch.
> > So, one patchset with two patches, please.
>
> Fine by me.
>
> >
> >> ---
> >>   mm/memory.c | 30 +++++++++++++++++++++++-------
> >>   1 file changed, 23 insertions(+), 7 deletions(-)
> >>
> >> diff --git a/mm/memory.c b/mm/memory.c
> >> index 3c01d68065be..ffc24a48ef15 100644
> >> --- a/mm/memory.c
> >> +++ b/mm/memory.c
> >> @@ -4409,26 +4409,27 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
> >>          return ret;
> >>   }
> >>
> >> -static bool pte_range_none(pte_t *pte, int nr_pages)
> >> +static int pte_range_none(pte_t *pte, int nr_pages)
> >>   {
> >>          int i;
> >>
> >>          for (i = 0; i < nr_pages; i++) {
> >>                  if (!pte_none(ptep_get_lockless(pte + i)))
> >> -                       return false;
> >> +                       return i;
> >>          }
> >>
> >> -       return true;
> >> +       return nr_pages;
> >>   }
> >>
> >>   static struct folio *alloc_anon_folio(struct vm_fault *vmf)
> >>   {
> >>          struct vm_area_struct *vma = vmf->vma;
> >>   #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> >> +       pte_t *first_set_pte = NULL, *align_pte, *pte;
> >>          unsigned long orders;
> >>          struct folio *folio;
> >>          unsigned long addr;
> >> -       pte_t *pte;
> >> +       int max_empty;
> >>          gfp_t gfp;
> >>          int order;
> >>
> >> @@ -4463,8 +4464,23 @@ static struct folio *alloc_anon_folio(struct vm_fault *vmf)
> >>          order = highest_order(orders);
> >>          while (orders) {
> >>                  addr = ALIGN_DOWN(vmf->address, PAGE_SIZE << order);
> >> -               if (pte_range_none(pte + pte_index(addr), 1 << order))
> >> +               align_pte = pte + pte_index(addr);
> >> +
> >> +               /* Range to be scanned known to be empty */
> >> +               if (align_pte + (1 << order) <= first_set_pte)
> >>                          break;
> >> +
> >> +               /* Range to be scanned contains first_set_pte */
> >> +               if (align_pte <= first_set_pte)
> >> +                       goto repeat;
> >> +
> >> +               /* align_pte > first_set_pte, so need to check properly */
> >> +               max_empty = pte_range_none(align_pte, 1 << order);
> >> +               if (max_empty == 1 << order)
> >> +                       break;
> >> +
> >> +               first_set_pte = align_pte + max_empty;
> >> +repeat:
> >>                  order = next_order(&orders, order);
> >>          }
> >>
> >> @@ -4579,7 +4595,7 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)
> >>          if (nr_pages == 1 && vmf_pte_changed(vmf)) {
> >>                  update_mmu_tlb(vma, addr, vmf->pte);
> >>                  goto release;
> >> -       } else if (nr_pages > 1 && !pte_range_none(vmf->pte, nr_pages)) {
> >> +       } else if (nr_pages > 1 && pte_range_none(vmf->pte, nr_pages) != nr_pages) {
> >>                  update_mmu_tlb_range(vma, addr, vmf->pte, nr_pages);
> >>                  goto release;
> >>          }
> >> @@ -4915,7 +4931,7 @@ vm_fault_t finish_fault(struct vm_fault *vmf)
> >>                  update_mmu_tlb(vma, addr, vmf->pte);
> >>                  ret = VM_FAULT_NOPAGE;
> >>                  goto unlock;
> >> -       } else if (nr_pages > 1 && !pte_range_none(vmf->pte, nr_pages)) {
> >> +       } else if (nr_pages > 1 && pte_range_none(vmf->pte, nr_pages) != nr_pages) {
> >>                  update_mmu_tlb_range(vma, addr, vmf->pte, nr_pages);
> >>                  ret = VM_FAULT_NOPAGE;
> >>                  goto unlock;
> >> --
> >> 2.30.2
> >>
> > Thanks
> > Barry
>