linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] fs/cramfs: Convert kmap() to kmap_local_data()
@ 2022-10-12  9:13 Fabio M. De Francesco
  2022-10-12 18:54 ` Nicolas Pitre
                   ` (2 more replies)
  0 siblings, 3 replies; 5+ messages in thread
From: Fabio M. De Francesco @ 2022-10-12  9:13 UTC (permalink / raw)
  To: Nicolas Pitre, Christian Brauner, linux-kernel, bpf
  Cc: Fabio M. De Francesco, Venkataramanan, Anirudh, Ira Weiny

The use of kmap() is being deprecated in favor of kmap_local_page().

There are two main problems with kmap(): (1) It comes with an overhead as
the mapping space is restricted and protected by a global lock for
synchronization and (2) it also requires global TLB invalidation when the
kmap’s pool wraps and it might block when the mapping space is fully
utilized until a slot becomes available.

With kmap_local_page() the mappings are per thread, CPU local, can take
page faults, and can be called from any context (including interrupts).
It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore,
the tasks can be preempted and, when they are scheduled to run again, the
kernel virtual addresses are restored and still valid.

Since its use in fs/cramfs is safe everywhere, it should be preferred.

Therefore, replace kmap() with kmap_local_page() in fs/cramfs. Instead
of open-coding kmap_local_page() + memcpy(), use memcpy_from_page().

Cc: "Venkataramanan, Anirudh" <anirudh.venkataramanan@intel.com>
Suggested-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>
---
 fs/cramfs/inode.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c
index 61ccf7722fc3..c17cbba5d45e 100644
--- a/fs/cramfs/inode.c
+++ b/fs/cramfs/inode.c
@@ -238,8 +238,7 @@ static void *cramfs_blkdev_read(struct super_block *sb, unsigned int offset,
 		struct page *page = pages[i];
 
 		if (page) {
-			memcpy(data, kmap(page), PAGE_SIZE);
-			kunmap(page);
+			memcpy_from_page(data, page, 0, PAGE_SIZE);
 			put_page(page);
 		} else
 			memset(data, 0, PAGE_SIZE);
@@ -815,7 +814,7 @@ static int cramfs_read_folio(struct file *file, struct folio *folio)
 
 	maxblock = (inode->i_size + PAGE_SIZE - 1) >> PAGE_SHIFT;
 	bytes_filled = 0;
-	pgdata = kmap(page);
+	pgdata = kmap_local_page(page);
 
 	if (page->index < maxblock) {
 		struct super_block *sb = inode->i_sb;
@@ -903,13 +902,13 @@ static int cramfs_read_folio(struct file *file, struct folio *folio)
 
 	memset(pgdata + bytes_filled, 0, PAGE_SIZE - bytes_filled);
 	flush_dcache_page(page);
-	kunmap(page);
+	kunmap_local(pgdata);
 	SetPageUptodate(page);
 	unlock_page(page);
 	return 0;
 
 err:
-	kunmap(page);
+	kunmap_local(pgdata);
 	ClearPageUptodate(page);
 	SetPageError(page);
 	unlock_page(page);
-- 
2.37.3


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH] fs/cramfs: Convert kmap() to kmap_local_data()
  2022-10-12  9:13 [PATCH] fs/cramfs: Convert kmap() to kmap_local_data() Fabio M. De Francesco
@ 2022-10-12 18:54 ` Nicolas Pitre
  2022-10-17  6:35 ` Christian Brauner
  2022-12-30 22:31 ` Ira Weiny
  2 siblings, 0 replies; 5+ messages in thread
From: Nicolas Pitre @ 2022-10-12 18:54 UTC (permalink / raw)
  To: Fabio M. De Francesco
  Cc: Christian Brauner, linux-kernel, bpf, Venkataramanan, Anirudh, Ira Weiny

[-- Attachment #1: Type: text/plain, Size: 2695 bytes --]

On Wed, 12 Oct 2022, Fabio M. De Francesco wrote:

> The use of kmap() is being deprecated in favor of kmap_local_page().
> 
> There are two main problems with kmap(): (1) It comes with an overhead as
> the mapping space is restricted and protected by a global lock for
> synchronization and (2) it also requires global TLB invalidation when the
> kmap’s pool wraps and it might block when the mapping space is fully
> utilized until a slot becomes available.
> 
> With kmap_local_page() the mappings are per thread, CPU local, can take
> page faults, and can be called from any context (including interrupts).
> It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore,
> the tasks can be preempted and, when they are scheduled to run again, the
> kernel virtual addresses are restored and still valid.
> 
> Since its use in fs/cramfs is safe everywhere, it should be preferred.
> 
> Therefore, replace kmap() with kmap_local_page() in fs/cramfs. Instead
> of open-coding kmap_local_page() + memcpy(), use memcpy_from_page().
> 
> Cc: "Venkataramanan, Anirudh" <anirudh.venkataramanan@intel.com>
> Suggested-by: Ira Weiny <ira.weiny@intel.com>
> Signed-off-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>

Reviewed-by: Nicolas Pitre <nico@fluxnic.net>

> ---
>  fs/cramfs/inode.c | 9 ++++-----
>  1 file changed, 4 insertions(+), 5 deletions(-)
> 
> diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c
> index 61ccf7722fc3..c17cbba5d45e 100644
> --- a/fs/cramfs/inode.c
> +++ b/fs/cramfs/inode.c
> @@ -238,8 +238,7 @@ static void *cramfs_blkdev_read(struct super_block *sb, unsigned int offset,
>  		struct page *page = pages[i];
>  
>  		if (page) {
> -			memcpy(data, kmap(page), PAGE_SIZE);
> -			kunmap(page);
> +			memcpy_from_page(data, page, 0, PAGE_SIZE);
>  			put_page(page);
>  		} else
>  			memset(data, 0, PAGE_SIZE);
> @@ -815,7 +814,7 @@ static int cramfs_read_folio(struct file *file, struct folio *folio)
>  
>  	maxblock = (inode->i_size + PAGE_SIZE - 1) >> PAGE_SHIFT;
>  	bytes_filled = 0;
> -	pgdata = kmap(page);
> +	pgdata = kmap_local_page(page);
>  
>  	if (page->index < maxblock) {
>  		struct super_block *sb = inode->i_sb;
> @@ -903,13 +902,13 @@ static int cramfs_read_folio(struct file *file, struct folio *folio)
>  
>  	memset(pgdata + bytes_filled, 0, PAGE_SIZE - bytes_filled);
>  	flush_dcache_page(page);
> -	kunmap(page);
> +	kunmap_local(pgdata);
>  	SetPageUptodate(page);
>  	unlock_page(page);
>  	return 0;
>  
>  err:
> -	kunmap(page);
> +	kunmap_local(pgdata);
>  	ClearPageUptodate(page);
>  	SetPageError(page);
>  	unlock_page(page);
> -- 
> 2.37.3
> 
> 

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] fs/cramfs: Convert kmap() to kmap_local_data()
  2022-10-12  9:13 [PATCH] fs/cramfs: Convert kmap() to kmap_local_data() Fabio M. De Francesco
  2022-10-12 18:54 ` Nicolas Pitre
@ 2022-10-17  6:35 ` Christian Brauner
  2022-12-30 22:31 ` Ira Weiny
  2 siblings, 0 replies; 5+ messages in thread
From: Christian Brauner @ 2022-10-17  6:35 UTC (permalink / raw)
  To: Fabio M. De Francesco, H
  Cc: Nicolas Pitre, linux-kernel, bpf, Venkataramanan, Anirudh, Ira Weiny

On Wed, Oct 12, 2022 at 11:13:56AM +0200, Fabio M. De Francesco wrote:
> The use of kmap() is being deprecated in favor of kmap_local_page().
> 
> There are two main problems with kmap(): (1) It comes with an overhead as
> the mapping space is restricted and protected by a global lock for
> synchronization and (2) it also requires global TLB invalidation when the
> kmap’s pool wraps and it might block when the mapping space is fully
> utilized until a slot becomes available.
> 
> With kmap_local_page() the mappings are per thread, CPU local, can take
> page faults, and can be called from any context (including interrupts).
> It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore,
> the tasks can be preempted and, when they are scheduled to run again, the
> kernel virtual addresses are restored and still valid.
> 
> Since its use in fs/cramfs is safe everywhere, it should be preferred.
> 
> Therefore, replace kmap() with kmap_local_page() in fs/cramfs. Instead
> of open-coding kmap_local_page() + memcpy(), use memcpy_from_page().
> 
> Cc: "Venkataramanan, Anirudh" <anirudh.venkataramanan@intel.com>
> Suggested-by: Ira Weiny <ira.weiny@intel.com>
> Signed-off-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>
> ---

Looks good to me,
Reviewed-by: Christian Brauner (Microsoft) <brauner@kernel.org>

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] fs/cramfs: Convert kmap() to kmap_local_data()
  2022-10-12  9:13 [PATCH] fs/cramfs: Convert kmap() to kmap_local_data() Fabio M. De Francesco
  2022-10-12 18:54 ` Nicolas Pitre
  2022-10-17  6:35 ` Christian Brauner
@ 2022-12-30 22:31 ` Ira Weiny
  2022-12-31 20:55   ` Nicolas Pitre
  2 siblings, 1 reply; 5+ messages in thread
From: Ira Weiny @ 2022-12-30 22:31 UTC (permalink / raw)
  To: Fabio M. De Francesco
  Cc: Nicolas Pitre, Christian Brauner, linux-kernel, bpf,
	Venkataramanan, Anirudh

On Wed, Oct 12, 2022 at 11:13:56AM +0200, Fabio M. De Francesco wrote:
> The use of kmap() is being deprecated in favor of kmap_local_page().
> 
> There are two main problems with kmap(): (1) It comes with an overhead as
> the mapping space is restricted and protected by a global lock for
> synchronization and (2) it also requires global TLB invalidation when the
> kmap’s pool wraps and it might block when the mapping space is fully
> utilized until a slot becomes available.
> 
> With kmap_local_page() the mappings are per thread, CPU local, can take
> page faults, and can be called from any context (including interrupts).
> It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore,
> the tasks can be preempted and, when they are scheduled to run again, the
> kernel virtual addresses are restored and still valid.
> 
> Since its use in fs/cramfs is safe everywhere, it should be preferred.
> 
> Therefore, replace kmap() with kmap_local_page() in fs/cramfs. Instead
> of open-coding kmap_local_page() + memcpy(), use memcpy_from_page().
> 
> Cc: "Venkataramanan, Anirudh" <anirudh.venkataramanan@intel.com>
> Suggested-by: Ira Weiny <ira.weiny@intel.com>

Nicolas, I was auditing the kmap conversion status and I see you reviewed this
but did not merge it.  Are you still the maintainer for fs/cramfs?  Should
Fabio send this to someone else?

FWIW LGTM

Reviewed-by: Ira Weiny <ira.weiny@intel.com>

> Signed-off-by: Fabio M. De Francesco <fmdefrancesco@gmail.com>
> ---
>  fs/cramfs/inode.c | 9 ++++-----
>  1 file changed, 4 insertions(+), 5 deletions(-)
> 
> diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c
> index 61ccf7722fc3..c17cbba5d45e 100644
> --- a/fs/cramfs/inode.c
> +++ b/fs/cramfs/inode.c
> @@ -238,8 +238,7 @@ static void *cramfs_blkdev_read(struct super_block *sb, unsigned int offset,
>  		struct page *page = pages[i];
>  
>  		if (page) {
> -			memcpy(data, kmap(page), PAGE_SIZE);
> -			kunmap(page);
> +			memcpy_from_page(data, page, 0, PAGE_SIZE);
>  			put_page(page);
>  		} else
>  			memset(data, 0, PAGE_SIZE);
> @@ -815,7 +814,7 @@ static int cramfs_read_folio(struct file *file, struct folio *folio)
>  
>  	maxblock = (inode->i_size + PAGE_SIZE - 1) >> PAGE_SHIFT;
>  	bytes_filled = 0;
> -	pgdata = kmap(page);
> +	pgdata = kmap_local_page(page);
>  
>  	if (page->index < maxblock) {
>  		struct super_block *sb = inode->i_sb;
> @@ -903,13 +902,13 @@ static int cramfs_read_folio(struct file *file, struct folio *folio)
>  
>  	memset(pgdata + bytes_filled, 0, PAGE_SIZE - bytes_filled);
>  	flush_dcache_page(page);
> -	kunmap(page);
> +	kunmap_local(pgdata);
>  	SetPageUptodate(page);
>  	unlock_page(page);
>  	return 0;
>  
>  err:
> -	kunmap(page);
> +	kunmap_local(pgdata);
>  	ClearPageUptodate(page);
>  	SetPageError(page);
>  	unlock_page(page);
> -- 
> 2.37.3
> 

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] fs/cramfs: Convert kmap() to kmap_local_data()
  2022-12-30 22:31 ` Ira Weiny
@ 2022-12-31 20:55   ` Nicolas Pitre
  0 siblings, 0 replies; 5+ messages in thread
From: Nicolas Pitre @ 2022-12-31 20:55 UTC (permalink / raw)
  To: Ira Weiny
  Cc: Fabio M. De Francesco, Christian Brauner, linux-kernel, bpf,
	Venkataramanan, Anirudh

[-- Attachment #1: Type: text/plain, Size: 1548 bytes --]

On Fri, 30 Dec 2022, Ira Weiny wrote:

> On Wed, Oct 12, 2022 at 11:13:56AM +0200, Fabio M. De Francesco wrote:
> > The use of kmap() is being deprecated in favor of kmap_local_page().
> > 
> > There are two main problems with kmap(): (1) It comes with an overhead as
> > the mapping space is restricted and protected by a global lock for
> > synchronization and (2) it also requires global TLB invalidation when the
> > kmap’s pool wraps and it might block when the mapping space is fully
> > utilized until a slot becomes available.
> > 
> > With kmap_local_page() the mappings are per thread, CPU local, can take
> > page faults, and can be called from any context (including interrupts).
> > It is faster than kmap() in kernels with HIGHMEM enabled. Furthermore,
> > the tasks can be preempted and, when they are scheduled to run again, the
> > kernel virtual addresses are restored and still valid.
> > 
> > Since its use in fs/cramfs is safe everywhere, it should be preferred.
> > 
> > Therefore, replace kmap() with kmap_local_page() in fs/cramfs. Instead
> > of open-coding kmap_local_page() + memcpy(), use memcpy_from_page().
> > 
> > Cc: "Venkataramanan, Anirudh" <anirudh.venkataramanan@intel.com>
> > Suggested-by: Ira Weiny <ira.weiny@intel.com>
> 
> Nicolas, I was auditing the kmap conversion status and I see you reviewed this
> but did not merge it.  Are you still the maintainer for fs/cramfs?  Should
> Fabio send this to someone else?

Please send it to Al Viro with my reviewed-by.


Nicolas

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2022-12-31 20:59 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-10-12  9:13 [PATCH] fs/cramfs: Convert kmap() to kmap_local_data() Fabio M. De Francesco
2022-10-12 18:54 ` Nicolas Pitre
2022-10-17  6:35 ` Christian Brauner
2022-12-30 22:31 ` Ira Weiny
2022-12-31 20:55   ` Nicolas Pitre

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).