Received-SPF: pass (google.com: domain of linux-kernel+bounces-162134-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Precedence: bulk
MIME-Version: 1.0
References: <20240429103346.59115-1-libang.li@antgroup.com> <20240429103346.59115-6-libang.li@antgroup.com>
In-Reply-To: <20240429103346.59115-6-libang.li@antgroup.com>
From: Lance Yang <ioworker0@gmail.com>
Date: Mon, 29 Apr 2024 19:23:32 +0800
Message-ID: <CAK1f24n4usJm4=e0+jrTxGtRjqTJZvn4gBp8d_vU=p2CrU=TsA@mail.gmail.com>
Subject: Re: [PATCH v1 5/5] mm: Add update_mmu_tlb_range()
To: Bang Li <libang.li@antgroup.com>
Cc: akpm@linux-foundation.org, chenhuacai@kernel.org, 
	tsbogend@alpha.franken.de, paul.walmsley@sifive.com, palmer@dabbelt.com, 
	chris@zankel.net, jcmvbkbc@gmail.com, david@redhat.com, ryan.roberts@arm.com, 
	libang.linux@gmail.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org, 
	loongarch@lists.linux.dev, linux-mips@vger.kernel.org, 
	linux-riscv@lists.infradead.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hey Bang,

On Mon, Apr 29, 2024 at 6:36=E2=80=AFPM Bang Li <libang.li@antgroup.com> wr=
ote:
>
> After the commit 19eaf44954df ("mm: thp: support allocation of anonymous
> multi-size THP"), it may need to batch update tlb of an address range
> through the update_mmu_tlb function. We can simplify this operation by
> adding the update_mmu_tlb_range function, which may also reduce the
> execution of some unnecessary code in some architectures.
>
> Signed-off-by: Bang Li <libang.li@antgroup.com>
> ---
>  include/linux/pgtable.h | 5 +++++
>  mm/memory.c             | 4 +---
>  2 files changed, 6 insertions(+), 3 deletions(-)
>
> diff --git a/include/linux/pgtable.h b/include/linux/pgtable.h
> index 18019f037bae..73411dfebf7a 100644
> --- a/include/linux/pgtable.h
> +++ b/include/linux/pgtable.h
> @@ -734,6 +734,11 @@ static inline void update_mmu_tlb(struct vm_area_str=
uct *vma,
>                                 unsigned long address, pte_t *ptep)
>  {
>  }
> +
> +static inline void update_mmu_tlb_range(struct vm_area_struct *vma,
> +                               unsigned long address, pte_t *ptep, unsig=
ned int nr)
> +{
> +}
>  #define __HAVE_ARCH_UPDATE_MMU_TLB
>  #endif

IMO, it might be better to use a separate definition to determine whether
update_mmu_tlb_range() is overridden by a specific architecture.

Thanks,
Lance

>
> diff --git a/mm/memory.c b/mm/memory.c
> index 6647685fd3c4..1f0ca362b82a 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -4396,7 +4396,6 @@ static vm_fault_t do_anonymous_page(struct vm_fault=
 *vmf)
>         vm_fault_t ret =3D 0;
>         int nr_pages =3D 1;
>         pte_t entry;
> -       int i;
>
>         /* File mapping without ->vm_ops ? */
>         if (vma->vm_flags & VM_SHARED)
> @@ -4465,8 +4464,7 @@ static vm_fault_t do_anonymous_page(struct vm_fault=
 *vmf)
>                 update_mmu_tlb(vma, addr, vmf->pte);
>                 goto release;
>         } else if (nr_pages > 1 && !pte_range_none(vmf->pte, nr_pages)) {
> -               for (i =3D 0; i < nr_pages; i++)
> -                       update_mmu_tlb(vma, addr + PAGE_SIZE * i, vmf->pt=
e + i);
> +               update_mmu_tlb_range(vma, addr, vmf->pte, nr_pages);
>                 goto release;
>         }
>
> --
> 2.19.1.6.gb485710b
>