All of lore.kernel.org
 help / color / mirror / Atom feed
From: Dan Williams <dan.j.williams@intel.com>
To: Joao Martins <joao.m.martins@oracle.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>, Christoph Hellwig <hch@lst.de>,
	Shiyang Ruan <ruansy.fnst@fujitsu.com>,
	Matthew Wilcox <willy@infradead.org>, Jan Kara <jack@suse.cz>,
	Andrew Morton <akpm@linux-foundation.org>,
	david <david@fromorbit.com>,
	linux-fsdevel <linux-fsdevel@vger.kernel.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Linux MM <linux-mm@kvack.org>,
	linux-nvdimm <linux-nvdimm@lists.01.org>
Subject: Re: [PATCH 3/3] mm/devmap: Remove pgmap accounting in the get_user_pages_fast() path
Date: Wed, 24 Mar 2021 10:45:42 -0700	[thread overview]
Message-ID: <CAPcyv4jidaz=33oWFMB_aBPtYDLe-AA_NP-k_pfGADVt=w5Vng@mail.gmail.com> (raw)
In-Reply-To: <66514812-6a24-8e2e-7be5-c61e188fecc4@oracle.com>

On Thu, Mar 18, 2021 at 3:02 AM Joao Martins <joao.m.martins@oracle.com> wrote:
>
> On 3/18/21 4:08 AM, Dan Williams wrote:
> > Now that device-dax and filesystem-dax are guaranteed to unmap all user
> > mappings of devmap / DAX pages before tearing down the 'struct page'
> > array, get_user_pages_fast() can rely on its traditional synchronization
> > method "validate_pte(); get_page(); revalidate_pte()" to catch races with
> > device shutdown. Specifically the unmap guarantee ensures that gup-fast
> > either succeeds in taking a page reference (lock-less), or it detects a
> > need to fall back to the slow path where the device presence can be
> > revalidated with locks held.
>
> [...]
>
> > @@ -2087,21 +2078,26 @@ static int gup_pte_range(pmd_t pmd, unsigned long addr, unsigned long end,
> >  #endif /* CONFIG_ARCH_HAS_PTE_SPECIAL */
> >
> >  #if defined(CONFIG_ARCH_HAS_PTE_DEVMAP) && defined(CONFIG_TRANSPARENT_HUGEPAGE)
> > +
> >  static int __gup_device_huge(unsigned long pfn, unsigned long addr,
> >                            unsigned long end, unsigned int flags,
> >                            struct page **pages, int *nr)
> >  {
> >       int nr_start = *nr;
> > -     struct dev_pagemap *pgmap = NULL;
> >
> >       do {
> > -             struct page *page = pfn_to_page(pfn);
> > +             struct page *page;
> > +
> > +             /*
> > +              * Typically pfn_to_page() on a devmap pfn is not safe
> > +              * without holding a live reference on the hosting
> > +              * pgmap. In the gup-fast path it is safe because any
> > +              * races will be resolved by either gup-fast taking a
> > +              * reference or the shutdown path unmapping the pte to
> > +              * trigger gup-fast to fall back to the slow path.
> > +              */
> > +             page = pfn_to_page(pfn);
> >
> > -             pgmap = get_dev_pagemap(pfn, pgmap);
> > -             if (unlikely(!pgmap)) {
> > -                     undo_dev_pagemap(nr, nr_start, flags, pages);
> > -                     return 0;
> > -             }
> >               SetPageReferenced(page);
> >               pages[*nr] = page;
> >               if (unlikely(!try_grab_page(page, flags))) {
>
> So for allowing FOLL_LONGTERM[0] would it be OK if we used page->pgmap after
> try_grab_page() for checking pgmap type to see if we are in a device-dax
> longterm pin?

So, there is an effort to add a new pte bit p{m,u}d_special to disable
gup-fast for huge pages [1]. I'd like to investigate whether we could
use devmap + special as an encoding for "no longterm" and never
consult the pgmap in the gup-fast path.

[1]: https://lore.kernel.org/linux-mm/a1fa7fa2-914b-366d-9902-e5b784e8428c@shipmail.org/
_______________________________________________
Linux-nvdimm mailing list -- linux-nvdimm@lists.01.org
To unsubscribe send an email to linux-nvdimm-leave@lists.01.org

WARNING: multiple messages have this Message-ID (diff)
From: Dan Williams <dan.j.williams@intel.com>
To: Joao Martins <joao.m.martins@oracle.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>, Christoph Hellwig <hch@lst.de>,
	Shiyang Ruan <ruansy.fnst@fujitsu.com>,
	Vishal Verma <vishal.l.verma@intel.com>,
	Dave Jiang <dave.jiang@intel.com>,
	Ira Weiny <ira.weiny@intel.com>,
	Matthew Wilcox <willy@infradead.org>, Jan Kara <jack@suse.cz>,
	Andrew Morton <akpm@linux-foundation.org>,
	david <david@fromorbit.com>,
	linux-fsdevel <linux-fsdevel@vger.kernel.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Linux MM <linux-mm@kvack.org>,
	linux-nvdimm <linux-nvdimm@lists.01.org>
Subject: Re: [PATCH 3/3] mm/devmap: Remove pgmap accounting in the get_user_pages_fast() path
Date: Wed, 24 Mar 2021 10:45:42 -0700	[thread overview]
Message-ID: <CAPcyv4jidaz=33oWFMB_aBPtYDLe-AA_NP-k_pfGADVt=w5Vng@mail.gmail.com> (raw)
In-Reply-To: <66514812-6a24-8e2e-7be5-c61e188fecc4@oracle.com>

On Thu, Mar 18, 2021 at 3:02 AM Joao Martins <joao.m.martins@oracle.com> wrote:
>
> On 3/18/21 4:08 AM, Dan Williams wrote:
> > Now that device-dax and filesystem-dax are guaranteed to unmap all user
> > mappings of devmap / DAX pages before tearing down the 'struct page'
> > array, get_user_pages_fast() can rely on its traditional synchronization
> > method "validate_pte(); get_page(); revalidate_pte()" to catch races with
> > device shutdown. Specifically the unmap guarantee ensures that gup-fast
> > either succeeds in taking a page reference (lock-less), or it detects a
> > need to fall back to the slow path where the device presence can be
> > revalidated with locks held.
>
> [...]
>
> > @@ -2087,21 +2078,26 @@ static int gup_pte_range(pmd_t pmd, unsigned long addr, unsigned long end,
> >  #endif /* CONFIG_ARCH_HAS_PTE_SPECIAL */
> >
> >  #if defined(CONFIG_ARCH_HAS_PTE_DEVMAP) && defined(CONFIG_TRANSPARENT_HUGEPAGE)
> > +
> >  static int __gup_device_huge(unsigned long pfn, unsigned long addr,
> >                            unsigned long end, unsigned int flags,
> >                            struct page **pages, int *nr)
> >  {
> >       int nr_start = *nr;
> > -     struct dev_pagemap *pgmap = NULL;
> >
> >       do {
> > -             struct page *page = pfn_to_page(pfn);
> > +             struct page *page;
> > +
> > +             /*
> > +              * Typically pfn_to_page() on a devmap pfn is not safe
> > +              * without holding a live reference on the hosting
> > +              * pgmap. In the gup-fast path it is safe because any
> > +              * races will be resolved by either gup-fast taking a
> > +              * reference or the shutdown path unmapping the pte to
> > +              * trigger gup-fast to fall back to the slow path.
> > +              */
> > +             page = pfn_to_page(pfn);
> >
> > -             pgmap = get_dev_pagemap(pfn, pgmap);
> > -             if (unlikely(!pgmap)) {
> > -                     undo_dev_pagemap(nr, nr_start, flags, pages);
> > -                     return 0;
> > -             }
> >               SetPageReferenced(page);
> >               pages[*nr] = page;
> >               if (unlikely(!try_grab_page(page, flags))) {
>
> So for allowing FOLL_LONGTERM[0] would it be OK if we used page->pgmap after
> try_grab_page() for checking pgmap type to see if we are in a device-dax
> longterm pin?

So, there is an effort to add a new pte bit p{m,u}d_special to disable
gup-fast for huge pages [1]. I'd like to investigate whether we could
use devmap + special as an encoding for "no longterm" and never
consult the pgmap in the gup-fast path.

[1]: https://lore.kernel.org/linux-mm/a1fa7fa2-914b-366d-9902-e5b784e8428c@shipmail.org/

WARNING: multiple messages have this Message-ID (diff)
From: Dan Williams <dan.j.williams@intel.com>
To: Joao Martins <joao.m.martins@oracle.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>, Christoph Hellwig <hch@lst.de>,
	Shiyang Ruan <ruansy.fnst@fujitsu.com>,
	 Vishal Verma <vishal.l.verma@intel.com>,
	Dave Jiang <dave.jiang@intel.com>,
	 Ira Weiny <ira.weiny@intel.com>,
	Matthew Wilcox <willy@infradead.org>, Jan Kara <jack@suse.cz>,
	 Andrew Morton <akpm@linux-foundation.org>,
	david <david@fromorbit.com>,
	 linux-fsdevel <linux-fsdevel@vger.kernel.org>,
	 Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Linux MM <linux-mm@kvack.org>,
	 linux-nvdimm <linux-nvdimm@lists.01.org>
Subject: Re: [PATCH 3/3] mm/devmap: Remove pgmap accounting in the get_user_pages_fast() path
Date: Wed, 24 Mar 2021 10:45:42 -0700	[thread overview]
Message-ID: <CAPcyv4jidaz=33oWFMB_aBPtYDLe-AA_NP-k_pfGADVt=w5Vng@mail.gmail.com> (raw)
In-Reply-To: <66514812-6a24-8e2e-7be5-c61e188fecc4@oracle.com>

On Thu, Mar 18, 2021 at 3:02 AM Joao Martins <joao.m.martins@oracle.com> wrote:
>
> On 3/18/21 4:08 AM, Dan Williams wrote:
> > Now that device-dax and filesystem-dax are guaranteed to unmap all user
> > mappings of devmap / DAX pages before tearing down the 'struct page'
> > array, get_user_pages_fast() can rely on its traditional synchronization
> > method "validate_pte(); get_page(); revalidate_pte()" to catch races with
> > device shutdown. Specifically the unmap guarantee ensures that gup-fast
> > either succeeds in taking a page reference (lock-less), or it detects a
> > need to fall back to the slow path where the device presence can be
> > revalidated with locks held.
>
> [...]
>
> > @@ -2087,21 +2078,26 @@ static int gup_pte_range(pmd_t pmd, unsigned long addr, unsigned long end,
> >  #endif /* CONFIG_ARCH_HAS_PTE_SPECIAL */
> >
> >  #if defined(CONFIG_ARCH_HAS_PTE_DEVMAP) && defined(CONFIG_TRANSPARENT_HUGEPAGE)
> > +
> >  static int __gup_device_huge(unsigned long pfn, unsigned long addr,
> >                            unsigned long end, unsigned int flags,
> >                            struct page **pages, int *nr)
> >  {
> >       int nr_start = *nr;
> > -     struct dev_pagemap *pgmap = NULL;
> >
> >       do {
> > -             struct page *page = pfn_to_page(pfn);
> > +             struct page *page;
> > +
> > +             /*
> > +              * Typically pfn_to_page() on a devmap pfn is not safe
> > +              * without holding a live reference on the hosting
> > +              * pgmap. In the gup-fast path it is safe because any
> > +              * races will be resolved by either gup-fast taking a
> > +              * reference or the shutdown path unmapping the pte to
> > +              * trigger gup-fast to fall back to the slow path.
> > +              */
> > +             page = pfn_to_page(pfn);
> >
> > -             pgmap = get_dev_pagemap(pfn, pgmap);
> > -             if (unlikely(!pgmap)) {
> > -                     undo_dev_pagemap(nr, nr_start, flags, pages);
> > -                     return 0;
> > -             }
> >               SetPageReferenced(page);
> >               pages[*nr] = page;
> >               if (unlikely(!try_grab_page(page, flags))) {
>
> So for allowing FOLL_LONGTERM[0] would it be OK if we used page->pgmap after
> try_grab_page() for checking pgmap type to see if we are in a device-dax
> longterm pin?

So, there is an effort to add a new pte bit p{m,u}d_special to disable
gup-fast for huge pages [1]. I'd like to investigate whether we could
use devmap + special as an encoding for "no longterm" and never
consult the pgmap in the gup-fast path.

[1]: https://lore.kernel.org/linux-mm/a1fa7fa2-914b-366d-9902-e5b784e8428c@shipmail.org/


  parent reply	other threads:[~2021-03-24 17:45 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-18  4:08 [PATCH 0/3] mm, pmem: Force unmap pmem on surprise remove Dan Williams
2021-03-18  4:08 ` Dan Williams
2021-03-18  4:08 ` [PATCH 1/3] mm/memory-failure: Prepare for mass memory_failure() Dan Williams
2021-03-18  4:08   ` Dan Williams
2021-03-18  4:08 ` [PATCH 2/3] mm, dax, pmem: Introduce dev_pagemap_failure() Dan Williams
2021-03-18  4:08   ` Dan Williams
2021-03-18  4:45   ` Darrick J. Wong
2021-03-18  4:45     ` Darrick J. Wong
2021-03-18 19:26     ` Dan Williams
2021-03-18 19:26       ` Dan Williams
2021-03-18 19:26       ` Dan Williams
2021-03-18  4:57   ` Dave Chinner
2021-03-18  4:57     ` Dave Chinner
2021-03-18 19:20     ` Dan Williams
2021-03-18 19:20       ` Dan Williams
2021-03-18 19:20       ` Dan Williams
2021-03-20  1:46       ` Dave Chinner
2021-03-20  1:46         ` Dave Chinner
2021-03-20  2:39         ` Dan Williams
2021-03-20  2:39           ` Dan Williams
2021-03-20  2:39           ` Dan Williams
2021-03-18  4:08 ` [PATCH 3/3] mm/devmap: Remove pgmap accounting in the get_user_pages_fast() path Dan Williams
2021-03-18  4:08   ` Dan Williams
2021-03-18 10:00   ` Joao Martins
2021-03-18 10:00     ` Joao Martins
2021-03-18 17:03     ` Dan Williams
2021-03-18 17:03       ` Dan Williams
2021-03-18 17:03       ` Dan Williams
2021-03-25 14:34       ` Jason Gunthorpe
2021-03-29 23:24         ` Dan Williams
2021-03-29 23:24           ` Dan Williams
2021-03-29 23:24           ` Dan Williams
2021-03-30 13:49           ` Jason Gunthorpe
2021-03-24 17:45     ` Dan Williams [this message]
2021-03-24 17:45       ` Dan Williams
2021-03-24 17:45       ` Dan Williams
2021-03-24 19:00       ` Joao Martins
2021-03-24 19:00         ` Joao Martins
2021-04-01 19:54         ` Joao Martins
2021-04-01 19:54           ` Joao Martins
2021-03-25 14:37   ` Jason Gunthorpe
2021-03-25 13:02 ` [PATCH 0/3] mm, pmem: Force unmap pmem on surprise remove David Hildenbrand
2021-03-25 13:02   ` 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='CAPcyv4jidaz=33oWFMB_aBPtYDLe-AA_NP-k_pfGADVt=w5Vng@mail.gmail.com' \
    --to=dan.j.williams@intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=david@fromorbit.com \
    --cc=hch@lst.de \
    --cc=jack@suse.cz \
    --cc=jgg@ziepe.ca \
    --cc=joao.m.martins@oracle.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-nvdimm@lists.01.org \
    --cc=ruansy.fnst@fujitsu.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.