linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Ryan Roberts <ryan.roberts@arm.com>
To: David Hildenbrand <david@redhat.com>, linux-kernel@vger.kernel.org
Cc: linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>,
	"Matthew Wilcox (Oracle)" <willy@infradead.org>,
	Hugh Dickins <hughd@google.com>,
	Yin Fengwei <fengwei.yin@intel.com>,
	Mike Kravetz <mike.kravetz@oracle.com>,
	Muchun Song <muchun.song@linux.dev>, Peter Xu <peterx@redhat.com>
Subject: Re: [PATCH v1 13/39] mm/rmap: factor out adding folio mappings into __folio_add_rmap()
Date: Mon, 18 Dec 2023 16:07:27 +0000	[thread overview]
Message-ID: <e5b69854-2bb3-4df4-92cc-3101e9691615@arm.com> (raw)
In-Reply-To: <20231211155652.131054-14-david@redhat.com>

On 11/12/2023 15:56, David Hildenbrand wrote:
> Let's factor it out to prepare for reuse as we convert
> page_add_anon_rmap() to folio_add_anon_rmap_[pte|ptes|pmd]().
> 
> Make the compiler always special-case on the granularity by using
> __always_inline.
> 
> Reviewed-by: Yin Fengwei <fengwei.yin@intel.com>
> Signed-off-by: David Hildenbrand <david@redhat.com>
> ---
>  mm/rmap.c | 81 ++++++++++++++++++++++++++++++-------------------------
>  1 file changed, 45 insertions(+), 36 deletions(-)
> 
> diff --git a/mm/rmap.c b/mm/rmap.c
> index 2ff2f11275e5..c5761986a411 100644
> --- a/mm/rmap.c
> +++ b/mm/rmap.c
> @@ -1157,6 +1157,49 @@ int folio_total_mapcount(struct folio *folio)
>  	return mapcount;
>  }
>  
> +static __always_inline unsigned int __folio_add_rmap(struct folio *folio,
> +		struct page *page, int nr_pages, enum rmap_mode mode,
> +		unsigned int *nr_pmdmapped)
> +{
> +	atomic_t *mapped = &folio->_nr_pages_mapped;
> +	int first, nr = 0;
> +
> +	__folio_rmap_sanity_checks(folio, page, nr_pages, mode);
> +
> +	/* Is page being mapped by PTE? Is this its first map to be added? */

I suspect this comment is left over from the old version? It sounds a bit odd in
its new context.

> +	switch (mode) {
> +	case RMAP_MODE_PTE:
> +		do {
> +			first = atomic_inc_and_test(&page->_mapcount);
> +			if (first && folio_test_large(folio)) {
> +				first = atomic_inc_return_relaxed(mapped);
> +				first = (first < COMPOUND_MAPPED);
> +			}
> +
> +			if (first)
> +				nr++;
> +		} while (page++, --nr_pages > 0);
> +		break;
> +	case RMAP_MODE_PMD:
> +		first = atomic_inc_and_test(&folio->_entire_mapcount);
> +		if (first) {
> +			nr = atomic_add_return_relaxed(COMPOUND_MAPPED, mapped);
> +			if (likely(nr < COMPOUND_MAPPED + COMPOUND_MAPPED)) {
> +				*nr_pmdmapped = folio_nr_pages(folio);
> +				nr = *nr_pmdmapped - (nr & FOLIO_PAGES_MAPPED);
> +				/* Raced ahead of a remove and another add? */
> +				if (unlikely(nr < 0))
> +					nr = 0;
> +			} else {
> +				/* Raced ahead of a remove of COMPOUND_MAPPED */
> +				nr = 0;
> +			}
> +		}
> +		break;
> +	}
> +	return nr;
> +}
> +
>  /**
>   * folio_move_anon_rmap - move a folio to our anon_vma
>   * @folio:	The folio to move to our anon_vma
> @@ -1380,45 +1423,11 @@ static __always_inline void __folio_add_file_rmap(struct folio *folio,
>  		struct page *page, int nr_pages, struct vm_area_struct *vma,
>  		enum rmap_mode mode)
>  {
> -	atomic_t *mapped = &folio->_nr_pages_mapped;
> -	unsigned int nr_pmdmapped = 0, first;
> -	int nr = 0;
> +	unsigned int nr, nr_pmdmapped = 0;

You're still being inconsistent with signed/unsigned here. Is there a reason
these can't be signed like nr_pages in the interface?

>  
>  	VM_WARN_ON_FOLIO(folio_test_anon(folio), folio);
> -	__folio_rmap_sanity_checks(folio, page, nr_pages, mode);
> -
> -	/* Is page being mapped by PTE? Is this its first map to be added? */
> -	switch (mode) {
> -	case RMAP_MODE_PTE:
> -		do {
> -			first = atomic_inc_and_test(&page->_mapcount);
> -			if (first && folio_test_large(folio)) {
> -				first = atomic_inc_return_relaxed(mapped);
> -				first = (first < COMPOUND_MAPPED);
> -			}
> -
> -			if (first)
> -				nr++;
> -		} while (page++, --nr_pages > 0);
> -		break;
> -	case RMAP_MODE_PMD:
> -		first = atomic_inc_and_test(&folio->_entire_mapcount);
> -		if (first) {
> -			nr = atomic_add_return_relaxed(COMPOUND_MAPPED, mapped);
> -			if (likely(nr < COMPOUND_MAPPED + COMPOUND_MAPPED)) {
> -				nr_pmdmapped = folio_nr_pages(folio);
> -				nr = nr_pmdmapped - (nr & FOLIO_PAGES_MAPPED);
> -				/* Raced ahead of a remove and another add? */
> -				if (unlikely(nr < 0))
> -					nr = 0;
> -			} else {
> -				/* Raced ahead of a remove of COMPOUND_MAPPED */
> -				nr = 0;
> -			}
> -		}
> -		break;
> -	}
>  
> +	nr = __folio_add_rmap(folio, page, nr_pages, mode, &nr_pmdmapped);
>  	if (nr_pmdmapped)
>  		__lruvec_stat_mod_folio(folio, folio_test_swapbacked(folio) ?
>  			NR_SHMEM_PMDMAPPED : NR_FILE_PMDMAPPED, nr_pmdmapped);



  reply	other threads:[~2023-12-18 16:07 UTC|newest]

Thread overview: 70+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-11 15:56 [PATCH v1 00/39] mm/rmap: interface overhaul David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 01/39] mm/rmap: rename hugepage_add* to hugetlb_add* David Hildenbrand
2023-12-11 16:14   ` Ryan Roberts
2023-12-11 16:24   ` Matthew Wilcox
2023-12-11 15:56 ` [PATCH v1 02/39] mm/rmap: introduce and use hugetlb_remove_rmap() David Hildenbrand
2023-12-11 16:15   ` Ryan Roberts
2023-12-11 16:33   ` Matthew Wilcox
2023-12-11 16:35     ` David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 03/39] mm/rmap: introduce and use hugetlb_add_file_rmap() David Hildenbrand
2023-12-11 16:17   ` Ryan Roberts
2023-12-11 15:56 ` [PATCH v1 04/39] mm/rmap: introduce and use hugetlb_try_dup_anon_rmap() David Hildenbrand
2023-12-11 16:25   ` Ryan Roberts
2023-12-11 15:56 ` [PATCH v1 05/39] mm/rmap: introduce and use hugetlb_try_share_anon_rmap() David Hildenbrand
2023-12-11 16:29   ` Ryan Roberts
2023-12-11 15:56 ` [PATCH v1 06/39] mm/rmap: add hugetlb sanity checks David Hildenbrand
2023-12-11 16:29   ` Ryan Roberts
2023-12-13  9:03   ` David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 07/39] mm/rmap: convert folio_add_file_rmap_range() into folio_add_file_rmap_[pte|ptes|pmd]() David Hildenbrand
2023-12-13  5:33   ` Yin Fengwei
2023-12-13  8:47     ` David Hildenbrand
2023-12-18 15:48   ` Ryan Roberts
2023-12-11 15:56 ` [PATCH v1 08/39] mm/memory: page_add_file_rmap() -> folio_add_file_rmap_[pte|pmd]() David Hildenbrand
2023-12-18 15:56   ` Ryan Roberts
2023-12-11 15:56 ` [PATCH v1 09/39] mm/huge_memory: page_add_file_rmap() -> folio_add_file_rmap_pmd() David Hildenbrand
2023-12-18 15:58   ` Ryan Roberts
2023-12-11 15:56 ` [PATCH v1 10/39] mm/migrate: page_add_file_rmap() -> folio_add_file_rmap_pte() David Hildenbrand
2023-12-18 15:58   ` Ryan Roberts
2023-12-11 15:56 ` [PATCH v1 11/39] mm/userfaultfd: " David Hildenbrand
2023-12-18 15:59   ` Ryan Roberts
2023-12-11 15:56 ` [PATCH v1 12/39] mm/rmap: remove page_add_file_rmap() David Hildenbrand
2023-12-18 16:00   ` Ryan Roberts
2023-12-11 15:56 ` [PATCH v1 13/39] mm/rmap: factor out adding folio mappings into __folio_add_rmap() David Hildenbrand
2023-12-18 16:07   ` Ryan Roberts [this message]
2023-12-18 17:06     ` David Hildenbrand
2023-12-19  8:40       ` Ryan Roberts
2023-12-11 15:56 ` [PATCH v1 14/39] mm/rmap: introduce folio_add_anon_rmap_[pte|ptes|pmd]() David Hildenbrand
2023-12-15  2:26   ` Yin, Fengwei
2023-12-15 15:16     ` David Hildenbrand
2023-12-18 16:26   ` Ryan Roberts
2023-12-18 17:02     ` David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 15/39] mm/huge_memory: batch rmap operations in __split_huge_pmd_locked() David Hildenbrand
2023-12-15  2:27   ` Yin, Fengwei
2023-12-15  2:39   ` Yin, Fengwei
2023-12-18 16:22   ` Ryan Roberts
2023-12-18 17:03     ` David Hildenbrand
2023-12-19  8:42       ` Ryan Roberts
2023-12-11 15:56 ` [PATCH v1 16/39] mm/huge_memory: page_add_anon_rmap() -> folio_add_anon_rmap_pmd() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 17/39] mm/migrate: page_add_anon_rmap() -> folio_add_anon_rmap_pte() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 18/39] mm/ksm: " David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 19/39] mm/swapfile: " David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 20/39] mm/memory: " David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 21/39] mm/rmap: remove page_add_anon_rmap() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 22/39] mm/rmap: remove RMAP_COMPOUND David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 23/39] mm/rmap: introduce folio_remove_rmap_[pte|ptes|pmd]() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 24/39] kernel/events/uprobes: page_remove_rmap() -> folio_remove_rmap_pte() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 25/39] mm/huge_memory: page_remove_rmap() -> folio_remove_rmap_pmd() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 26/39] mm/khugepaged: page_remove_rmap() -> folio_remove_rmap_pte() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 27/39] mm/ksm: " David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 28/39] mm/memory: " David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 29/39] mm/migrate_device: " David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 30/39] mm/rmap: " David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 31/39] Documentation: stop referring to page_remove_rmap() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 32/39] mm/rmap: remove page_remove_rmap() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 33/39] mm/rmap: convert page_dup_file_rmap() to folio_dup_file_rmap_[pte|ptes|pmd]() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 34/39] mm/rmap: introduce folio_try_dup_anon_rmap_[pte|ptes|pmd]() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 35/39] mm/huge_memory: page_try_dup_anon_rmap() -> folio_try_dup_anon_rmap_pmd() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 36/39] mm/memory: page_try_dup_anon_rmap() -> folio_try_dup_anon_rmap_pte() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 37/39] mm/rmap: remove page_try_dup_anon_rmap() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 38/39] mm: convert page_try_share_anon_rmap() to folio_try_share_anon_rmap_[pte|pmd]() David Hildenbrand
2023-12-11 15:56 ` [PATCH v1 39/39] mm/rmap: rename COMPOUND_MAPPED to ENTIRELY_MAPPED David Hildenbrand

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=e5b69854-2bb3-4df4-92cc-3101e9691615@arm.com \
    --to=ryan.roberts@arm.com \
    --cc=akpm@linux-foundation.org \
    --cc=david@redhat.com \
    --cc=fengwei.yin@intel.com \
    --cc=hughd@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mike.kravetz@oracle.com \
    --cc=muchun.song@linux.dev \
    --cc=peterx@redhat.com \
    --cc=willy@infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox