Re: [PATCH -mm v6 06/13] pagemap: use walk->vma instead of calling find_vma()

[Naoya Horiguchi]

Hi Peter,

On Wed, Aug 06, 2014 at 01:30:40PM -0400, Peter Feiner wrote:
> On Fri, Aug 01, 2014 at 03:20:42PM -0400, Naoya Horiguchi wrote:
> > Page table walker has the information of the current vma in mm_walk, so
> > we don't have to call find_vma() in each pagemap_hugetlb_range() call.
> 
> You could also get rid of a bunch of code in pagemap_pte_range:

Thanks for the comment.

> ---
>  fs/proc/task_mmu.c | 33 ++++++---------------------------
>  1 file changed, 6 insertions(+), 27 deletions(-)
> 
> diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
> index 9b30bdd..e9af130 100644
> --- a/fs/proc/task_mmu.c
> +++ b/fs/proc/task_mmu.c
> @@ -1035,16 +1035,14 @@ static inline void thp_pmd_to_pagemap_entry(pagemap_entry_t *pme, struct pagemap
>  static int pagemap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
>  			     struct mm_walk *walk)
>  {
> -	struct vm_area_struct *vma;
> +	struct vm_area_struct *vma = walk->vma
>  	struct pagemapread *pm = walk->private;
>  	spinlock_t *ptl;
>  	pte_t *pte;
>  	int err = 0;
> -	pagemap_entry_t pme = make_pme(PM_NOT_PRESENT(pm->v2));
> +	pagemap_entry_t pme;
>  
> -	/* find the first VMA at or above 'addr' */
> -	vma = find_vma(walk->mm, addr);
> -	if (vma && pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
> +	if (pmd_trans_huge_lock(pmd, vma, &ptl) == 1) {
>  		int pmd_flags2;
>  
>  		if ((vma->vm_flags & VM_SOFTDIRTY) || pmd_soft_dirty(*pmd))
> @@ -1069,28 +1067,9 @@ static int pagemap_pte_range(pmd_t *pmd, unsigned long addr, unsigned long end,
>  	if (pmd_trans_unstable(pmd))
>  		return 0;
>  	for (; addr != end; addr += PAGE_SIZE) {
> -		int flags2;
> -
> -		/* check to see if we've left 'vma' behind
> -		 * and need a new, higher one */
> -		if (vma && (addr >= vma->vm_end)) {
> -			vma = find_vma(walk->mm, addr);
> -			if (vma && (vma->vm_flags & VM_SOFTDIRTY))
> -				flags2 = __PM_SOFT_DIRTY;
> -			else
> -				flags2 = 0;
> -			pme = make_pme(PM_NOT_PRESENT(pm->v2) | PM_STATUS2(pm->v2, flags2));
> -		}

Right. Soft dirty check is done in pte_to_pagemap_entry(), and vma
boundary check is done in mm/pagewalk.c, so we don't need this block
any more.

Thanks,
Naoya Horiguchi

> -
> -		/* check that 'vma' actually covers this address,
> -		 * and that it isn't a huge page vma */
> -		if (vma && (vma->vm_start <= addr) &&
> -		    !is_vm_hugetlb_page(vma)) {
> -			pte = pte_offset_map(pmd, addr);
> -			pte_to_pagemap_entry(&pme, pm, vma, addr, *pte);
> -			/* unmap before userspace copy */
> -			pte_unmap(pte);
> -		}
> +		pte = pte_offset_map(pmd, addr);
> +		pte_to_pagemap_entry(&pme, pm, vma, addr, *pte);
> +		pte_unmap(pte);
>  		err = add_to_pagemap(addr, &pme, pm);
>  		if (err)
>  			return err;
> -- 
> 2.0.0.526.g5318336
> 
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@xxxxxxxxxx  For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/