Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1259624

Re: [PATCH 1/8] mm: support madvise(MADV_FREE)

From Shaohua Li <shli@kernel.org>
Newsgroups linux.kernel
Subject Re: [PATCH 1/8] mm: support madvise(MADV_FREE)
Date 2015-10-30 17:50 +0100
Message-ID <qpl2j-8b8-29@gated-at.bofh.it> (permalink)
References <qpbYZ-2NI-5@gated-at.bofh.it> <qpbZ0-2NI-23@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


On Fri, Oct 30, 2015 at 04:01:37PM +0900, Minchan Kim wrote:
> +static int madvise_free_pte_range(pmd_t *pmd, unsigned long addr,
> +				unsigned long end, struct mm_walk *walk)
> +
> +{
> +	struct mmu_gather *tlb = walk->private;
> +	struct mm_struct *mm = tlb->mm;
> +	struct vm_area_struct *vma = walk->vma;
> +	spinlock_t *ptl;
> +	pte_t *pte, ptent;
> +	struct page *page;
> +
> +	split_huge_page_pmd(vma, addr, pmd);
> +	if (pmd_trans_unstable(pmd))
> +		return 0;
> +
> +	pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
> +	arch_enter_lazy_mmu_mode();
> +	for (; addr != end; pte++, addr += PAGE_SIZE) {
> +		ptent = *pte;
> +
> +		if (!pte_present(ptent))
> +			continue;
> +
> +		page = vm_normal_page(vma, addr, ptent);
> +		if (!page)
> +			continue;
> +
> +		if (PageSwapCache(page)) {
> +			if (!trylock_page(page))
> +				continue;
> +
> +			if (!try_to_free_swap(page)) {
> +				unlock_page(page);
> +				continue;
> +			}
> +
> +			ClearPageDirty(page);
> +			unlock_page(page);
> +		}
> +
> +		/*
> +		 * Some of architecture(ex, PPC) don't update TLB
> +		 * with set_pte_at and tlb_remove_tlb_entry so for
> +		 * the portability, remap the pte with old|clean
> +		 * after pte clearing.
> +		 */
> +		ptent = ptep_get_and_clear_full(mm, addr, pte,
> +						tlb->fullmm);
> +		ptent = pte_mkold(ptent);
> +		ptent = pte_mkclean(ptent);
> +		set_pte_at(mm, addr, pte, ptent);
> +		tlb_remove_tlb_entry(tlb, pte, addr);

The orginal ptent might not be dirty. In that case, the tlb_remove_tlb_entry
is unnecessary, so please add a check. In practice, I saw more TLB flush with
FREE compared to DONTNEED because of this issue.

Thanks,
Shaohua
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH 0/8] MADV_FREE support Minchan Kim <minchan@kernel.org> - 2015-10-30 08:10 +0100
  [PATCH 3/8] arch: uapi: asm: mman.h: Let MADV_FREE have same value for all architectures Minchan Kim <minchan@kernel.org> - 2015-10-30 08:10 +0100
    Re: [PATCH 3/8] arch: uapi: asm: mman.h: Let MADV_FREE have same  value for all architectures Hugh Dickins <hughd@google.com> - 2015-11-02 01:10 +0100
      Re: [PATCH 3/8] arch: uapi: asm: mman.h: Let MADV_FREE have same  value for all architectures Minchan Kim <minchan@kernel.org> - 2015-11-03 03:40 +0100
        Re: [PATCH 3/8] arch: uapi: asm: mman.h: Let MADV_FREE have same  value for all architectures David Miller <davem@davemloft.net> - 2015-11-03 04:40 +0100
          Re: [PATCH 3/8] arch: uapi: asm: mman.h: Let MADV_FREE have same  value for all architectures Minchan Kim <minchan@kernel.org> - 2015-11-03 05:40 +0100
      Re: [PATCH 3/8] arch: uapi: asm: mman.h: Let MADV_FREE have same  value for all architectures Minchan Kim <minchan@kernel.org> - 2015-11-03 03:40 +0100
  [PATCH 1/8] mm: support madvise(MADV_FREE) Minchan Kim <minchan@kernel.org> - 2015-10-30 08:10 +0100
    Re: [PATCH 1/8] mm: support madvise(MADV_FREE) Shaohua Li <shli@kernel.org> - 2015-10-30 17:50 +0100
      Re: [PATCH 1/8] mm: support madvise(MADV_FREE) Minchan Kim <minchan@kernel.org> - 2015-11-03 01:20 +0100
  [PATCH 6/8] mm: lru_deactivate_fn should clear PG_referenced Minchan Kim <minchan@kernel.org> - 2015-10-30 08:10 +0100
    Re: [PATCH 6/8] mm: lru_deactivate_fn should clear PG_referenced Michal Hocko <mhocko@kernel.org> - 2015-10-30 13:50 +0100
      Re: [PATCH 6/8] mm: lru_deactivate_fn should clear PG_referenced Minchan Kim <minchan@kernel.org> - 2015-11-03 02:20 +0100
        Re: [PATCH 6/8] mm: lru_deactivate_fn should clear PG_referenced Michal Hocko <mhocko@kernel.org> - 2015-11-04 09:30 +0100
  [PATCH 4/8] mm: free swp_entry in madvise_free Minchan Kim <minchan@kernel.org> - 2015-10-30 08:10 +0100
    Re: [PATCH 4/8] mm: free swp_entry in madvise_free Michal Hocko <mhocko@kernel.org> - 2015-10-30 13:30 +0100
      Re: [PATCH 4/8] mm: free swp_entry in madvise_free Minchan Kim <minchan@kernel.org> - 2015-11-03 02:00 +0100
  Re: [PATCH 0/8] MADV_FREE support David Rientjes <rientjes@google.com> - 2015-11-01 06:00 +0100
    Re: [PATCH 0/8] MADV_FREE support Daniel Micay <danielmicay@gmail.com> - 2015-11-01 07:40 +0100
      Re: [PATCH 0/8] MADV_FREE support Minchan Kim <minchan@kernel.org> - 2015-11-03 03:30 +0100
      Re: [PATCH 0/8] MADV_FREE support David Rientjes <rientjes@google.com> - 2015-11-04 21:20 +0100

csiph-web