From: Dan Williams <dan.j.williams@intel.com>
To: Joao Martins <joao.m.martins@oracle.com>
Cc: Linux MM <linux-mm@kvack.org>,
linux-nvdimm <linux-nvdimm@lists.01.org>,
Matthew Wilcox <willy@infradead.org>,
Jason Gunthorpe <jgg@ziepe.ca>, Jane Chu <jane.chu@oracle.com>,
Muchun Song <songmuchun@bytedance.com>,
Mike Kravetz <mike.kravetz@oracle.com>,
Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [PATCH v1 04/11] mm/memremap: add ZONE_DEVICE support for compound pages
Date: Wed, 5 May 2021 11:44:29 -0700 [thread overview]
Message-ID: <CAPcyv4gs_rHL7FPqyQEb3yT4jrv8Wo_xA2ojKsppoBfmDocq8A@mail.gmail.com> (raw)
In-Reply-To: <20210325230938.30752-5-joao.m.martins@oracle.com>
On Thu, Mar 25, 2021 at 4:10 PM Joao Martins <joao.m.martins@oracle.com> wrote:
>
> Add a new align property for struct dev_pagemap which specifies that a
> pagemap is composed of a set of compound pages of size @align, instead
> of base pages. When these pages are initialised, most are initialised as
> tail pages instead of order-0 pages.
>
> For certain ZONE_DEVICE users like device-dax which have a fixed page
> size, this creates an opportunity to optimize GUP and GUP-fast walkers,
> treating it the same way as THP or hugetlb pages.
>
> Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
> ---
> include/linux/memremap.h | 13 +++++++++++++
> mm/memremap.c | 8 ++++++--
> mm/page_alloc.c | 24 +++++++++++++++++++++++-
> 3 files changed, 42 insertions(+), 3 deletions(-)
>
> diff --git a/include/linux/memremap.h b/include/linux/memremap.h
> index b46f63dcaed3..bb28d82dda5e 100644
> --- a/include/linux/memremap.h
> +++ b/include/linux/memremap.h
> @@ -114,6 +114,7 @@ struct dev_pagemap {
> struct completion done;
> enum memory_type type;
> unsigned int flags;
> + unsigned long align;
I think this wants some kernel-doc above to indicate that non-zero
means "use compound pages with tail-page dedup" and zero / PAGE_SIZE
means "use non-compound base pages". The non-zero value must be
PAGE_SIZE, PMD_PAGE_SIZE or PUD_PAGE_SIZE. Hmm, maybe it should be an
enum:
enum devmap_geometry {
DEVMAP_PTE,
DEVMAP_PMD,
DEVMAP_PUD,
}
...because it's more than just an alignment it's a structural
definition of how the memmap is laid out.
> const struct dev_pagemap_ops *ops;
> void *owner;
> int nr_range;
> @@ -130,6 +131,18 @@ static inline struct vmem_altmap *pgmap_altmap(struct dev_pagemap *pgmap)
> return NULL;
> }
>
> +static inline unsigned long pgmap_align(struct dev_pagemap *pgmap)
> +{
> + if (!pgmap || !pgmap->align)
> + return PAGE_SIZE;
> + return pgmap->align;
> +}
> +
> +static inline unsigned long pgmap_pfn_align(struct dev_pagemap *pgmap)
> +{
> + return PHYS_PFN(pgmap_align(pgmap));
> +}
> +
> #ifdef CONFIG_ZONE_DEVICE
> bool pfn_zone_device_reserved(unsigned long pfn);
> void *memremap_pages(struct dev_pagemap *pgmap, int nid);
> diff --git a/mm/memremap.c b/mm/memremap.c
> index 805d761740c4..d160853670c4 100644
> --- a/mm/memremap.c
> +++ b/mm/memremap.c
> @@ -318,8 +318,12 @@ static int pagemap_range(struct dev_pagemap *pgmap, struct mhp_params *params,
> memmap_init_zone_device(&NODE_DATA(nid)->node_zones[ZONE_DEVICE],
> PHYS_PFN(range->start),
> PHYS_PFN(range_len(range)), pgmap);
> - percpu_ref_get_many(pgmap->ref, pfn_end(pgmap, range_id)
> - - pfn_first(pgmap, range_id));
> + if (pgmap_align(pgmap) > PAGE_SIZE)
> + percpu_ref_get_many(pgmap->ref, (pfn_end(pgmap, range_id)
> + - pfn_first(pgmap, range_id)) / pgmap_pfn_align(pgmap));
> + else
> + percpu_ref_get_many(pgmap->ref, pfn_end(pgmap, range_id)
> + - pfn_first(pgmap, range_id));
> return 0;
>
> err_add_memory:
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 58974067bbd4..3a77f9e43f3a 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -6285,6 +6285,8 @@ void __ref memmap_init_zone_device(struct zone *zone,
> unsigned long pfn, end_pfn = start_pfn + nr_pages;
> struct pglist_data *pgdat = zone->zone_pgdat;
> struct vmem_altmap *altmap = pgmap_altmap(pgmap);
> + unsigned int pfn_align = pgmap_pfn_align(pgmap);
> + unsigned int order_align = order_base_2(pfn_align);
> unsigned long zone_idx = zone_idx(zone);
> unsigned long start = jiffies;
> int nid = pgdat->node_id;
> @@ -6302,10 +6304,30 @@ void __ref memmap_init_zone_device(struct zone *zone,
> nr_pages = end_pfn - start_pfn;
> }
>
> - for (pfn = start_pfn; pfn < end_pfn; pfn++) {
> + for (pfn = start_pfn; pfn < end_pfn; pfn += pfn_align) {
pfn_align is in bytes and pfn is in pages... is there a "pfn_align >>=
PAGE_SHIFT" I missed somewhere?
_______________________________________________
Linux-nvdimm mailing list -- linux-nvdimm@lists.01.org
To unsubscribe send an email to linux-nvdimm-leave@lists.01.org
next prev parent reply other threads:[~2021-05-05 18:44 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-25 23:09 [PATCH v1 00/11] mm, sparse-vmemmap: Introduce compound pagemaps Joao Martins
2021-03-25 23:09 ` [PATCH v1 01/11] memory-failure: fetch compound_head after pgmap_pfn_valid() Joao Martins
2021-04-24 0:12 ` Dan Williams
2021-04-24 19:00 ` Joao Martins
2021-03-25 23:09 ` [PATCH v1 02/11] mm/page_alloc: split prep_compound_page into head and tail subparts Joao Martins
2021-04-24 0:16 ` Dan Williams
2021-04-24 19:05 ` Joao Martins
2021-03-25 23:09 ` [PATCH v1 03/11] mm/page_alloc: refactor memmap_init_zone_device() page init Joao Martins
2021-04-24 0:18 ` Dan Williams
2021-04-24 19:05 ` Joao Martins
2021-03-25 23:09 ` [PATCH v1 04/11] mm/memremap: add ZONE_DEVICE support for compound pages Joao Martins
2021-05-05 18:44 ` Dan Williams [this message]
2021-05-05 18:58 ` Matthew Wilcox
2021-05-05 19:49 ` Joao Martins
2021-05-05 22:20 ` Dan Williams
2021-05-05 22:36 ` Joao Martins
2021-05-05 23:03 ` Dan Williams
2021-05-06 10:12 ` Joao Martins
2021-05-18 17:27 ` Joao Martins
2021-05-18 19:56 ` Jane Chu
2021-05-19 11:29 ` Joao Martins
2021-05-06 8:05 ` Aneesh Kumar K.V
2021-05-06 10:23 ` Joao Martins
2021-05-06 11:43 ` Matthew Wilcox
2021-05-06 12:15 ` Joao Martins
2021-03-25 23:09 ` [PATCH v1 05/11] mm/sparse-vmemmap: add a pgmap argument to section activation Joao Martins
2021-05-05 22:34 ` Dan Williams
2021-05-05 22:37 ` Joao Martins
2021-05-05 23:14 ` Dan Williams
2021-05-06 10:24 ` Joao Martins
2021-03-25 23:09 ` [PATCH v1 06/11] mm/sparse-vmemmap: refactor vmemmap_populate_basepages() Joao Martins
2021-05-05 22:43 ` Dan Williams
2021-05-06 10:27 ` Joao Martins
2021-05-06 18:36 ` Joao Martins
2021-03-25 23:09 ` [PATCH v1 07/11] mm/sparse-vmemmap: populate compound pagemaps Joao Martins
2021-05-06 1:18 ` Dan Williams
2021-05-06 11:01 ` Joao Martins
2021-05-10 19:19 ` Dan Williams
2021-05-13 18:45 ` Joao Martins
2021-03-25 23:09 ` [PATCH v1 08/11] mm/sparse-vmemmap: use hugepages for PUD " Joao Martins
2021-03-25 23:09 ` [PATCH v1 09/11] mm/page_alloc: reuse tail struct pages for " Joao Martins
2021-03-25 23:09 ` [PATCH v1 10/11] device-dax: compound pagemap support Joao Martins
2021-03-25 23:09 ` [PATCH v1 11/11] mm/gup: grab head page refcount once for group of subpages Joao Martins
2021-04-01 9:38 ` [PATCH v1 00/11] mm, sparse-vmemmap: Introduce compound pagemaps Joao Martins
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=CAPcyv4gs_rHL7FPqyQEb3yT4jrv8Wo_xA2ojKsppoBfmDocq8A@mail.gmail.com \
--to=dan.j.williams@intel.com \
--cc=akpm@linux-foundation.org \
--cc=jane.chu@oracle.com \
--cc=jgg@ziepe.ca \
--cc=joao.m.martins@oracle.com \
--cc=linux-mm@kvack.org \
--cc=linux-nvdimm@lists.01.org \
--cc=mike.kravetz@oracle.com \
--cc=songmuchun@bytedance.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;
as well as URLs for NNTP newsgroup(s).