[2/2] mm: devmap: refactor 1-based refcounting for ZONE_DEVICE pages
diff mbox series

Message ID 20191115001134.2489505-3-jhubbard@nvidia.com
State New
Headers show
Series
  • mm: devmap: page-freeing related cleanups
Related show

Commit Message

John Hubbard Nov. 15, 2019, 12:11 a.m. UTC
An upcoming patch changes and complicates the refcounting and
especially the "put page" aspects of it. In order to keep
everything clean, refactor the devmap page release routines:

* Rename put_devmap_managed_page() to page_is_devmap_managed(),
  and limit the functionality to "read only": return a bool,
  with no side effects.

* Add a new routine, put_devmap_managed_page(), to handle checking
  what kind of page it is, and what kind of refcount handling it
  requires.

* Rename __put_devmap_managed_page() to free_devmap_managed_page(),
  and limit the functionality to unconditionally freeing a devmap
  page.

This is originally based on a separate patch by Ira Weiny, which
applied to an early version of the put_user_page() experiments.
Since then, Jérôme Glisse suggested the refactoring described above.

Cc: Jan Kara <jack@suse.cz>
Cc: Jérôme Glisse <jglisse@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dan Williams <dan.j.williams@intel.com>
Suggested-by: Jérôme Glisse <jglisse@redhat.com>
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
---
 include/linux/mm.h | 27 ++++++++++++++++++++++++---
 mm/memremap.c      | 16 ++--------------
 2 files changed, 26 insertions(+), 17 deletions(-)

Comments

Jan Kara Nov. 15, 2019, 9:48 a.m. UTC | #1
On Thu 14-11-19 16:11:34, John Hubbard wrote:
> An upcoming patch changes and complicates the refcounting and
> especially the "put page" aspects of it. In order to keep
> everything clean, refactor the devmap page release routines:
> 
> * Rename put_devmap_managed_page() to page_is_devmap_managed(),
>   and limit the functionality to "read only": return a bool,
>   with no side effects.
> 
> * Add a new routine, put_devmap_managed_page(), to handle checking
>   what kind of page it is, and what kind of refcount handling it
>   requires.
> 
> * Rename __put_devmap_managed_page() to free_devmap_managed_page(),
>   and limit the functionality to unconditionally freeing a devmap
>   page.
> 
> This is originally based on a separate patch by Ira Weiny, which
> applied to an early version of the put_user_page() experiments.
> Since then, Jérôme Glisse suggested the refactoring described above.
> 
> Cc: Jan Kara <jack@suse.cz>
> Cc: Jérôme Glisse <jglisse@redhat.com>
> Cc: Christoph Hellwig <hch@lst.de>
> Cc: Dan Williams <dan.j.williams@intel.com>
> Suggested-by: Jérôme Glisse <jglisse@redhat.com>
> Signed-off-by: Ira Weiny <ira.weiny@intel.com>
> Signed-off-by: John Hubbard <jhubbard@nvidia.com>

Looks good to me. You can add:

Reviewed-by: Jan Kara <jack@suse.cz>

								Honza

> ---
>  include/linux/mm.h | 27 ++++++++++++++++++++++++---
>  mm/memremap.c      | 16 ++--------------
>  2 files changed, 26 insertions(+), 17 deletions(-)
> 
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index a2adf95b3f9c..96228376139c 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -967,9 +967,10 @@ static inline bool is_zone_device_page(const struct page *page)
>  #endif
>  
>  #ifdef CONFIG_DEV_PAGEMAP_OPS
> -void __put_devmap_managed_page(struct page *page);
> +void free_devmap_managed_page(struct page *page);
>  DECLARE_STATIC_KEY_FALSE(devmap_managed_key);
> -static inline bool put_devmap_managed_page(struct page *page)
> +
> +static inline bool page_is_devmap_managed(struct page *page)
>  {
>  	if (!static_branch_unlikely(&devmap_managed_key))
>  		return false;
> @@ -978,7 +979,6 @@ static inline bool put_devmap_managed_page(struct page *page)
>  	switch (page->pgmap->type) {
>  	case MEMORY_DEVICE_PRIVATE:
>  	case MEMORY_DEVICE_FS_DAX:
> -		__put_devmap_managed_page(page);
>  		return true;
>  	default:
>  		break;
> @@ -986,6 +986,27 @@ static inline bool put_devmap_managed_page(struct page *page)
>  	return false;
>  }
>  
> +static inline bool put_devmap_managed_page(struct page *page)
> +{
> +	bool is_devmap = page_is_devmap_managed(page);
> +
> +	if (is_devmap) {
> +		int count = page_ref_dec_return(page);
> +
> +		/*
> +		 * devmap page refcounts are 1-based, rather than 0-based: if
> +		 * refcount is 1, then the page is free and the refcount is
> +		 * stable because nobody holds a reference on the page.
> +		 */
> +		if (count == 1)
> +			free_devmap_managed_page(page);
> +		else if (!count)
> +			__put_page(page);
> +	}
> +
> +	return is_devmap;
> +}
> +
>  #else /* CONFIG_DEV_PAGEMAP_OPS */
>  static inline bool put_devmap_managed_page(struct page *page)
>  {
> diff --git a/mm/memremap.c b/mm/memremap.c
> index e899fa876a62..2ba773859031 100644
> --- a/mm/memremap.c
> +++ b/mm/memremap.c
> @@ -411,20 +411,8 @@ struct dev_pagemap *get_dev_pagemap(unsigned long pfn,
>  EXPORT_SYMBOL_GPL(get_dev_pagemap);
>  
>  #ifdef CONFIG_DEV_PAGEMAP_OPS
> -void __put_devmap_managed_page(struct page *page)
> +void free_devmap_managed_page(struct page *page)
>  {
> -	int count = page_ref_dec_return(page);
> -
> -	/* still busy */
> -	if (count > 1)
> -		return;
> -
> -	/* only triggered by the dev_pagemap shutdown path */
> -	if (count == 0) {
> -		__put_page(page);
> -		return;
> -	}
> -
>  	/* notify page idle for dax */
>  	if (!is_device_private_page(page)) {
>  		wake_up_var(&page->_refcount);
> @@ -461,5 +449,5 @@ void __put_devmap_managed_page(struct page *page)
>  	page->mapping = NULL;
>  	page->pgmap->ops->page_free(page);
>  }
> -EXPORT_SYMBOL(__put_devmap_managed_page);
> +EXPORT_SYMBOL(free_devmap_managed_page);
>  #endif /* CONFIG_DEV_PAGEMAP_OPS */
> -- 
> 2.24.0
>
Dan Williams Nov. 15, 2019, 4:41 p.m. UTC | #2
On Thu, Nov 14, 2019 at 4:11 PM John Hubbard <jhubbard@nvidia.com> wrote:
>
> An upcoming patch changes and complicates the refcounting and
> especially the "put page" aspects of it. In order to keep
> everything clean, refactor the devmap page release routines:
>
> * Rename put_devmap_managed_page() to page_is_devmap_managed(),
>   and limit the functionality to "read only": return a bool,
>   with no side effects.
>
> * Add a new routine, put_devmap_managed_page(), to handle checking
>   what kind of page it is, and what kind of refcount handling it
>   requires.
>
> * Rename __put_devmap_managed_page() to free_devmap_managed_page(),
>   and limit the functionality to unconditionally freeing a devmap
>   page.
>
> This is originally based on a separate patch by Ira Weiny, which
> applied to an early version of the put_user_page() experiments.
> Since then, Jérôme Glisse suggested the refactoring described above.
>
> Cc: Jan Kara <jack@suse.cz>
> Cc: Jérôme Glisse <jglisse@redhat.com>
> Cc: Christoph Hellwig <hch@lst.de>
> Cc: Dan Williams <dan.j.williams@intel.com>
> Suggested-by: Jérôme Glisse <jglisse@redhat.com>
> Signed-off-by: Ira Weiny <ira.weiny@intel.com>
> Signed-off-by: John Hubbard <jhubbard@nvidia.com>
> ---
>  include/linux/mm.h | 27 ++++++++++++++++++++++++---
>  mm/memremap.c      | 16 ++--------------
>  2 files changed, 26 insertions(+), 17 deletions(-)

Looks good,

Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Christoph Hellwig Nov. 18, 2019, 7:08 a.m. UTC | #3
On Thu, Nov 14, 2019 at 04:11:34PM -0800, John Hubbard wrote:
> An upcoming patch changes and complicates the refcounting and
> especially the "put page" aspects of it. In order to keep
> everything clean, refactor the devmap page release routines:
> 
> * Rename put_devmap_managed_page() to page_is_devmap_managed(),
>   and limit the functionality to "read only": return a bool,
>   with no side effects.
> 
> * Add a new routine, put_devmap_managed_page(), to handle checking
>   what kind of page it is, and what kind of refcount handling it
>   requires.
> 
> * Rename __put_devmap_managed_page() to free_devmap_managed_page(),
>   and limit the functionality to unconditionally freeing a devmap
>   page.
> 
> This is originally based on a separate patch by Ira Weiny, which
> applied to an early version of the put_user_page() experiments.
> Since then, Jérôme Glisse suggested the refactoring described above.

I can't say I'm a big fan of this as it adds a lot more inlined
code to put_page, which has a lot of callsites.  Can't we instead
try to figure out a way to move away from the off by one refcounting?

> 
> Cc: Jan Kara <jack@suse.cz>
> Cc: Jérôme Glisse <jglisse@redhat.com>
> Cc: Christoph Hellwig <hch@lst.de>
> Cc: Dan Williams <dan.j.williams@intel.com>
> Suggested-by: Jérôme Glisse <jglisse@redhat.com>
> Signed-off-by: Ira Weiny <ira.weiny@intel.com>
> Signed-off-by: John Hubbard <jhubbard@nvidia.com>
> ---
>  include/linux/mm.h | 27 ++++++++++++++++++++++++---
>  mm/memremap.c      | 16 ++--------------
>  2 files changed, 26 insertions(+), 17 deletions(-)
> 
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index a2adf95b3f9c..96228376139c 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -967,9 +967,10 @@ static inline bool is_zone_device_page(const struct page *page)
>  #endif
>  
>  #ifdef CONFIG_DEV_PAGEMAP_OPS
> -void __put_devmap_managed_page(struct page *page);
> +void free_devmap_managed_page(struct page *page);
>  DECLARE_STATIC_KEY_FALSE(devmap_managed_key);
> -static inline bool put_devmap_managed_page(struct page *page)
> +
> +static inline bool page_is_devmap_managed(struct page *page)
>  {
>  	if (!static_branch_unlikely(&devmap_managed_key))
>  		return false;
> @@ -978,7 +979,6 @@ static inline bool put_devmap_managed_page(struct page *page)
>  	switch (page->pgmap->type) {
>  	case MEMORY_DEVICE_PRIVATE:
>  	case MEMORY_DEVICE_FS_DAX:
> -		__put_devmap_managed_page(page);
>  		return true;
>  	default:
>  		break;
> @@ -986,6 +986,27 @@ static inline bool put_devmap_managed_page(struct page *page)
>  	return false;
>  }
>  
> +static inline bool put_devmap_managed_page(struct page *page)
> +{
> +	bool is_devmap = page_is_devmap_managed(page);
> +
> +	if (is_devmap) {
> +		int count = page_ref_dec_return(page);
> +
> +		/*
> +		 * devmap page refcounts are 1-based, rather than 0-based: if
> +		 * refcount is 1, then the page is free and the refcount is
> +		 * stable because nobody holds a reference on the page.
> +		 */
> +		if (count == 1)
> +			free_devmap_managed_page(page);
> +		else if (!count)
> +			__put_page(page);
> +	}
> +
> +	return is_devmap;
> +}
> +
>  #else /* CONFIG_DEV_PAGEMAP_OPS */
>  static inline bool put_devmap_managed_page(struct page *page)
>  {
> diff --git a/mm/memremap.c b/mm/memremap.c
> index e899fa876a62..2ba773859031 100644
> --- a/mm/memremap.c
> +++ b/mm/memremap.c
> @@ -411,20 +411,8 @@ struct dev_pagemap *get_dev_pagemap(unsigned long pfn,
>  EXPORT_SYMBOL_GPL(get_dev_pagemap);
>  
>  #ifdef CONFIG_DEV_PAGEMAP_OPS
> -void __put_devmap_managed_page(struct page *page)
> +void free_devmap_managed_page(struct page *page)
>  {
> -	int count = page_ref_dec_return(page);
> -
> -	/* still busy */
> -	if (count > 1)
> -		return;
> -
> -	/* only triggered by the dev_pagemap shutdown path */
> -	if (count == 0) {
> -		__put_page(page);
> -		return;
> -	}
> -
>  	/* notify page idle for dax */
>  	if (!is_device_private_page(page)) {
>  		wake_up_var(&page->_refcount);
> @@ -461,5 +449,5 @@ void __put_devmap_managed_page(struct page *page)
>  	page->mapping = NULL;
>  	page->pgmap->ops->page_free(page);
>  }
> -EXPORT_SYMBOL(__put_devmap_managed_page);
> +EXPORT_SYMBOL(free_devmap_managed_page);
>  #endif /* CONFIG_DEV_PAGEMAP_OPS */
> -- 
> 2.24.0
> _______________________________________________
> Linux-nvdimm mailing list -- linux-nvdimm@lists.01.org
> To unsubscribe send an email to linux-nvdimm-leave@lists.01.org
---end quoted text---
Dan Williams Nov. 18, 2019, 4:57 p.m. UTC | #4
On Sun, Nov 17, 2019 at 11:09 PM Christoph Hellwig <hch@infradead.org> wrote:
>
> On Thu, Nov 14, 2019 at 04:11:34PM -0800, John Hubbard wrote:
> > An upcoming patch changes and complicates the refcounting and
> > especially the "put page" aspects of it. In order to keep
> > everything clean, refactor the devmap page release routines:
> >
> > * Rename put_devmap_managed_page() to page_is_devmap_managed(),
> >   and limit the functionality to "read only": return a bool,
> >   with no side effects.
> >
> > * Add a new routine, put_devmap_managed_page(), to handle checking
> >   what kind of page it is, and what kind of refcount handling it
> >   requires.
> >
> > * Rename __put_devmap_managed_page() to free_devmap_managed_page(),
> >   and limit the functionality to unconditionally freeing a devmap
> >   page.
> >
> > This is originally based on a separate patch by Ira Weiny, which
> > applied to an early version of the put_user_page() experiments.
> > Since then, Jérôme Glisse suggested the refactoring described above.
>
> I can't say I'm a big fan of this as it adds a lot more inlined
> code to put_page, which has a lot of callsites.  Can't we instead
> try to figure out a way to move away from the off by one refcounting?

That might be possible. David and I are discussing a pfn_online()
helper that might be a replacement for keeping ZONE_DEVICE pages out
of the page allocator rather than keeping their reference count
elevated.

Patch
diff mbox series

diff --git a/include/linux/mm.h b/include/linux/mm.h
index a2adf95b3f9c..96228376139c 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -967,9 +967,10 @@  static inline bool is_zone_device_page(const struct page *page)
 #endif
 
 #ifdef CONFIG_DEV_PAGEMAP_OPS
-void __put_devmap_managed_page(struct page *page);
+void free_devmap_managed_page(struct page *page);
 DECLARE_STATIC_KEY_FALSE(devmap_managed_key);
-static inline bool put_devmap_managed_page(struct page *page)
+
+static inline bool page_is_devmap_managed(struct page *page)
 {
 	if (!static_branch_unlikely(&devmap_managed_key))
 		return false;
@@ -978,7 +979,6 @@  static inline bool put_devmap_managed_page(struct page *page)
 	switch (page->pgmap->type) {
 	case MEMORY_DEVICE_PRIVATE:
 	case MEMORY_DEVICE_FS_DAX:
-		__put_devmap_managed_page(page);
 		return true;
 	default:
 		break;
@@ -986,6 +986,27 @@  static inline bool put_devmap_managed_page(struct page *page)
 	return false;
 }
 
+static inline bool put_devmap_managed_page(struct page *page)
+{
+	bool is_devmap = page_is_devmap_managed(page);
+
+	if (is_devmap) {
+		int count = page_ref_dec_return(page);
+
+		/*
+		 * devmap page refcounts are 1-based, rather than 0-based: if
+		 * refcount is 1, then the page is free and the refcount is
+		 * stable because nobody holds a reference on the page.
+		 */
+		if (count == 1)
+			free_devmap_managed_page(page);
+		else if (!count)
+			__put_page(page);
+	}
+
+	return is_devmap;
+}
+
 #else /* CONFIG_DEV_PAGEMAP_OPS */
 static inline bool put_devmap_managed_page(struct page *page)
 {
diff --git a/mm/memremap.c b/mm/memremap.c
index e899fa876a62..2ba773859031 100644
--- a/mm/memremap.c
+++ b/mm/memremap.c
@@ -411,20 +411,8 @@  struct dev_pagemap *get_dev_pagemap(unsigned long pfn,
 EXPORT_SYMBOL_GPL(get_dev_pagemap);
 
 #ifdef CONFIG_DEV_PAGEMAP_OPS
-void __put_devmap_managed_page(struct page *page)
+void free_devmap_managed_page(struct page *page)
 {
-	int count = page_ref_dec_return(page);
-
-	/* still busy */
-	if (count > 1)
-		return;
-
-	/* only triggered by the dev_pagemap shutdown path */
-	if (count == 0) {
-		__put_page(page);
-		return;
-	}
-
 	/* notify page idle for dax */
 	if (!is_device_private_page(page)) {
 		wake_up_var(&page->_refcount);
@@ -461,5 +449,5 @@  void __put_devmap_managed_page(struct page *page)
 	page->mapping = NULL;
 	page->pgmap->ops->page_free(page);
 }
-EXPORT_SYMBOL(__put_devmap_managed_page);
+EXPORT_SYMBOL(free_devmap_managed_page);
 #endif /* CONFIG_DEV_PAGEMAP_OPS */