diff mbox

[1/3] mm: New pfn_mkwrite same as page_mkwrite for VM_PFNMAP

Message ID 5512BA5D.8070609@plexistor.com (mailing list archive)
State New, archived
Headers show

Commit Message

Boaz Harrosh March 25, 2015, 1:38 p.m. UTC
From: Yigal Korman <yigal@plexistor.com>

This will allow FS that uses VM_PFNMAP | VM_MIXEDMAP (no page structs)
to get notified when access is a write to a read-only PFN.

This can happen if we mmap() a file then first mmap-read from it
to page-in a read-only PFN, than we mmap-write to the same page.

We need this functionality to fix a DAX bug, where in the scenario
above we fail to set ctime/mtime though we modified the file.
An xfstest is attached to this patchset that shows the failure
and the fix. (A DAX patch will follow)

This functionality is extra important for us, because upon
dirtying of a pmem page we also want to RDMA the page to a
remote cluster node.

We define a new pfn_mkwrite and do not reuse page_mkwrite because
  1 - The name ;-)
  2 - But mainly because it would take a very long and tedious
      audit of all page_mkwrite functions of VM_MIXEDMAP/VM_PFNMAP
      users. To make sure they do not now CRASH. For example current
      DAX code (which this is for) would crash.
      If we would want to reuse page_mkwrite, We will need to first
      patch all users, so to not-crash-on-no-page. Then enable this
      patch. But even if I did that I would not sleep so well at night.
      Adding a new vector is the safest thing to do, and is not that
      expensive. an extra pointer at a static function vector per driver.
      Also the new vector is better for performance, because else we
      Will call all current Kernel vectors, so to:
	check-ha-no-page-do-nothing and return.

No need to call it from do_shared_fault because do_wp_page is called to
change pte permissions anyway.

CC: Matthew Wilcox <matthew.r.wilcox@intel.com>
CC: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
CC: Jan Kara <jack@suse.cz>
CC: Andrew Morton <akpm@linux-foundation.org>
CC: Hugh Dickins <hughd@google.com>
CC: Mel Gorman <mgorman@suse.de>
CC: linux-mm@kvack.org

Signed-off-by: Yigal Korman <yigal@plexistor.com>
Signed-off-by: Boaz Harrosh <boaz@plexistor.com>
---
 include/linux/mm.h |  2 ++
 mm/memory.c        | 28 +++++++++++++++++++++++++++-
 2 files changed, 29 insertions(+), 1 deletion(-)

Comments

Kirill A . Shutemov March 25, 2015, 2:34 p.m. UTC | #1
On Wed, Mar 25, 2015 at 03:38:37PM +0200, Boaz Harrosh wrote:
> From: Yigal Korman <yigal@plexistor.com>
> 
> This will allow FS that uses VM_PFNMAP | VM_MIXEDMAP (no page structs)
> to get notified when access is a write to a read-only PFN.
> 
> This can happen if we mmap() a file then first mmap-read from it
> to page-in a read-only PFN, than we mmap-write to the same page.
> 
> We need this functionality to fix a DAX bug, where in the scenario
> above we fail to set ctime/mtime though we modified the file.
> An xfstest is attached to this patchset that shows the failure
> and the fix. (A DAX patch will follow)
> 
> This functionality is extra important for us, because upon
> dirtying of a pmem page we also want to RDMA the page to a
> remote cluster node.
> 
> We define a new pfn_mkwrite and do not reuse page_mkwrite because
>   1 - The name ;-)
>   2 - But mainly because it would take a very long and tedious
>       audit of all page_mkwrite functions of VM_MIXEDMAP/VM_PFNMAP
>       users. To make sure they do not now CRASH. For example current
>       DAX code (which this is for) would crash.
>       If we would want to reuse page_mkwrite, We will need to first
>       patch all users, so to not-crash-on-no-page. Then enable this
>       patch. But even if I did that I would not sleep so well at night.
>       Adding a new vector is the safest thing to do, and is not that
>       expensive. an extra pointer at a static function vector per driver.
>       Also the new vector is better for performance, because else we
>       Will call all current Kernel vectors, so to:
> 	check-ha-no-page-do-nothing and return.
> 
> No need to call it from do_shared_fault because do_wp_page is called to
> change pte permissions anyway.
> 
> CC: Matthew Wilcox <matthew.r.wilcox@intel.com>
> CC: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
> CC: Jan Kara <jack@suse.cz>
> CC: Andrew Morton <akpm@linux-foundation.org>
> CC: Hugh Dickins <hughd@google.com>
> CC: Mel Gorman <mgorman@suse.de>
> CC: linux-mm@kvack.org
> 
> Signed-off-by: Yigal Korman <yigal@plexistor.com>
> Signed-off-by: Boaz Harrosh <boaz@plexistor.com>

This is not going to apply to -mm. do_wp_page() is reworked there.
BTW, shouldn't we rename it to do_wp_fault() or something?

> ---
>  include/linux/mm.h |  2 ++
>  mm/memory.c        | 28 +++++++++++++++++++++++++++-

Documentation/filesystems/Locking ?

>  2 files changed, 29 insertions(+), 1 deletion(-)
> 
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index 47a9392..1cd820c 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -250,6 +250,8 @@ struct vm_operations_struct {
>  	/* notification that a previously read-only page is about to become
>  	 * writable, if an error is returned it will cause a SIGBUS */
>  	int (*page_mkwrite)(struct vm_area_struct *vma, struct vm_fault *vmf);
> +	/* same as page_mkwrite when using VM_PFNMAP|VM_MIXEDMAP */

New line before the comment?

> +	int (*pfn_mkwrite)(struct vm_area_struct *vma, struct vm_fault *vmf);
>  
>  	/* called by access_process_vm when get_user_pages() fails, typically
>  	 * for use by special VMAs that can switch between memory and hardware
> diff --git a/mm/memory.c b/mm/memory.c
> index 8068893..8d640d1 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -1982,6 +1982,23 @@ static int do_page_mkwrite(struct vm_area_struct *vma, struct page *page,
>  	return ret;
>  }
>  
> +static int do_pfn_mkwrite(struct vm_area_struct *vma, unsigned long address)
> +{
> +	if (vma->vm_ops && vma->vm_ops->pfn_mkwrite) {
> +		struct vm_fault vmf = {
> +			.page = 0,

.page = NULL,

> +			.pgoff = (((address & PAGE_MASK) - vma->vm_start)
> +						>> PAGE_SHIFT) + vma->vm_pgoff,

.pgoff = linear_page_index(vma, address),

> +			.virtual_address = (void __user *)(address & PAGE_MASK),
> +			.flags = FAULT_FLAG_WRITE | FAULT_FLAG_MKWRITE,
> +		};
> +
> +		return vma->vm_ops->pfn_mkwrite(vma, &vmf);
> +	}
> +
> +	return 0;
> +}
> +
>  /*
>   * This routine handles present pages, when users try to write
>   * to a shared page. It is done by copying the page to a new address
> @@ -2025,8 +2042,17 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
>  		 * accounting on raw pfn maps.
>  		 */
>  		if ((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
> -				     (VM_WRITE|VM_SHARED))
> +				     (VM_WRITE|VM_SHARED)) {
> +			pte_unmap_unlock(page_table, ptl);

It would be nice to avoid ptl drop if ->pfn_mkwrite is not defined for the
vma.

> +			ret = do_pfn_mkwrite(vma, address);
> +			if (ret & VM_FAULT_ERROR)
> +				return ret;
> +			page_table = pte_offset_map_lock(mm, pmd, address,
> +							 &ptl);
> +			if (!pte_same(*page_table, orig_pte))
> +				goto unlock;
>  			goto reuse;
> +		}
>  		goto gotten;
>  	}
>
Dave Hansen March 25, 2015, 3:08 p.m. UTC | #2
On 03/25/2015 06:38 AM, Boaz Harrosh wrote:
>  /*
>   * This routine handles present pages, when users try to write
>   * to a shared page. It is done by copying the page to a new address
> @@ -2025,8 +2042,17 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
>  		 * accounting on raw pfn maps.
>  		 */
>  		if ((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
> -				     (VM_WRITE|VM_SHARED))
> +				     (VM_WRITE|VM_SHARED)) {
> +			pte_unmap_unlock(page_table, ptl);
> +			ret = do_pfn_mkwrite(vma, address);
> +			if (ret & VM_FAULT_ERROR)
> +				return ret;
> +			page_table = pte_offset_map_lock(mm, pmd, address,
> +							 &ptl);
> +			if (!pte_same(*page_table, orig_pte))
> +				goto unlock;
>  			goto reuse;
> +		}
>  		goto gotten;
>  	}

This adds a lock release/reacquire in a place where the lock was
previously just held.  Could you explain a bit why this is safe?

Also, that pte_same() check looks a bit fragile.  It seems like it would
fail if the hardware, for instance, set the accessed bit in here
somewhere.  Is that what we want?
Kirill A . Shutemov March 25, 2015, 3:13 p.m. UTC | #3
On Wed, Mar 25, 2015 at 08:08:24AM -0700, Dave Hansen wrote:
> On 03/25/2015 06:38 AM, Boaz Harrosh wrote:
> >  /*
> >   * This routine handles present pages, when users try to write
> >   * to a shared page. It is done by copying the page to a new address
> > @@ -2025,8 +2042,17 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
> >  		 * accounting on raw pfn maps.
> >  		 */
> >  		if ((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
> > -				     (VM_WRITE|VM_SHARED))
> > +				     (VM_WRITE|VM_SHARED)) {
> > +			pte_unmap_unlock(page_table, ptl);
> > +			ret = do_pfn_mkwrite(vma, address);
> > +			if (ret & VM_FAULT_ERROR)
> > +				return ret;
> > +			page_table = pte_offset_map_lock(mm, pmd, address,
> > +							 &ptl);
> > +			if (!pte_same(*page_table, orig_pte))
> > +				goto unlock;
> >  			goto reuse;
> > +		}
> >  		goto gotten;
> >  	}
> 
> This adds a lock release/reacquire in a place where the lock was
> previously just held.  Could you explain a bit why this is safe?

It's common practice in page fault codepath. See code around
->page_mkwrite for example.
> 
> Also, that pte_same() check looks a bit fragile.  It seems like it would
> fail if the hardware, for instance, set the accessed bit in here
> somewhere.  Is that what we want?

In this case we will cancel this fault handling and fault again. No
problems here.
Boaz Harrosh March 26, 2015, 7:49 a.m. UTC | #4
On 03/25/2015 04:34 PM, Kirill A. Shutemov wrote:
> On Wed, Mar 25, 2015 at 03:38:37PM +0200, Boaz Harrosh wrote:
>> From: Yigal Korman <yigal@plexistor.com>
>>
>> This will allow FS that uses VM_PFNMAP | VM_MIXEDMAP (no page structs)
>> to get notified when access is a write to a read-only PFN.
>>
>> This can happen if we mmap() a file then first mmap-read from it
>> to page-in a read-only PFN, than we mmap-write to the same page.
>>
>> We need this functionality to fix a DAX bug, where in the scenario
>> above we fail to set ctime/mtime though we modified the file.
>> An xfstest is attached to this patchset that shows the failure
>> and the fix. (A DAX patch will follow)
>>
>> This functionality is extra important for us, because upon
>> dirtying of a pmem page we also want to RDMA the page to a
>> remote cluster node.
>>
>> We define a new pfn_mkwrite and do not reuse page_mkwrite because
>>   1 - The name ;-)
>>   2 - But mainly because it would take a very long and tedious
>>       audit of all page_mkwrite functions of VM_MIXEDMAP/VM_PFNMAP
>>       users. To make sure they do not now CRASH. For example current
>>       DAX code (which this is for) would crash.
>>       If we would want to reuse page_mkwrite, We will need to first
>>       patch all users, so to not-crash-on-no-page. Then enable this
>>       patch. But even if I did that I would not sleep so well at night.
>>       Adding a new vector is the safest thing to do, and is not that
>>       expensive. an extra pointer at a static function vector per driver.
>>       Also the new vector is better for performance, because else we
>>       Will call all current Kernel vectors, so to:
>> 	check-ha-no-page-do-nothing and return.
>>
>> No need to call it from do_shared_fault because do_wp_page is called to
>> change pte permissions anyway.
>>
>> CC: Matthew Wilcox <matthew.r.wilcox@intel.com>
>> CC: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
>> CC: Jan Kara <jack@suse.cz>
>> CC: Andrew Morton <akpm@linux-foundation.org>
>> CC: Hugh Dickins <hughd@google.com>
>> CC: Mel Gorman <mgorman@suse.de>
>> CC: linux-mm@kvack.org
>>
>> Signed-off-by: Yigal Korman <yigal@plexistor.com>
>> Signed-off-by: Boaz Harrosh <boaz@plexistor.com>
> 
> This is not going to apply to -mm. do_wp_page() is reworked there.
> BTW, shouldn't we rename it to do_wp_fault() or something?
> 

Wowhoo you were not kidding ;-)

I'll redo this patch based on linux-next/akpm branch. I will
need an hard up testing. Current patch I had for 6 month and
I'm confident about it. I'll need to stare at this real hard.

>> ---
>>  include/linux/mm.h |  2 ++
>>  mm/memory.c        | 28 +++++++++++++++++++++++++++-
> 
> Documentation/filesystems/Locking ?
> 
>>  2 files changed, 29 insertions(+), 1 deletion(-)
>>
>> diff --git a/include/linux/mm.h b/include/linux/mm.h
>> index 47a9392..1cd820c 100644
>> --- a/include/linux/mm.h
>> +++ b/include/linux/mm.h
>> @@ -250,6 +250,8 @@ struct vm_operations_struct {
>>  	/* notification that a previously read-only page is about to become
>>  	 * writable, if an error is returned it will cause a SIGBUS */
>>  	int (*page_mkwrite)(struct vm_area_struct *vma, struct vm_fault *vmf);
>> +	/* same as page_mkwrite when using VM_PFNMAP|VM_MIXEDMAP */
> 
> New line before the comment?
> 
>> +	int (*pfn_mkwrite)(struct vm_area_struct *vma, struct vm_fault *vmf);
>>  
>>  	/* called by access_process_vm when get_user_pages() fails, typically
>>  	 * for use by special VMAs that can switch between memory and hardware
>> diff --git a/mm/memory.c b/mm/memory.c
>> index 8068893..8d640d1 100644
>> --- a/mm/memory.c
>> +++ b/mm/memory.c
>> @@ -1982,6 +1982,23 @@ static int do_page_mkwrite(struct vm_area_struct *vma, struct page *page,
>>  	return ret;
>>  }
>>  
>> +static int do_pfn_mkwrite(struct vm_area_struct *vma, unsigned long address)
>> +{
>> +	if (vma->vm_ops && vma->vm_ops->pfn_mkwrite) {
>> +		struct vm_fault vmf = {
>> +			.page = 0,
> 
> .page = NULL,
> 
>> +			.pgoff = (((address & PAGE_MASK) - vma->vm_start)
>> +						>> PAGE_SHIFT) + vma->vm_pgoff,
> 
> .pgoff = linear_page_index(vma, address),
> 
>> +			.virtual_address = (void __user *)(address & PAGE_MASK),
>> +			.flags = FAULT_FLAG_WRITE | FAULT_FLAG_MKWRITE,
>> +		};
>> +
>> +		return vma->vm_ops->pfn_mkwrite(vma, &vmf);
>> +	}
>> +
>> +	return 0;
>> +}
>> +
>>  /*
>>   * This routine handles present pages, when users try to write
>>   * to a shared page. It is done by copying the page to a new address
>> @@ -2025,8 +2042,17 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
>>  		 * accounting on raw pfn maps.
>>  		 */
>>  		if ((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
>> -				     (VM_WRITE|VM_SHARED))
>> +				     (VM_WRITE|VM_SHARED)) {
>> +			pte_unmap_unlock(page_table, ptl);
> 
> It would be nice to avoid ptl drop if ->pfn_mkwrite is not defined for the
> vma.

OK Yes, I will move the if (vma->vm_ops && vma->vm_ops->pfn_mkwrite)
to out here surrounding the unlock/lock

> 
>> +			ret = do_pfn_mkwrite(vma, address);
>> +			if (ret & VM_FAULT_ERROR)
>> +				return ret;
>> +			page_table = pte_offset_map_lock(mm, pmd, address,
>> +							 &ptl);
>> +			if (!pte_same(*page_table, orig_pte))
>> +				goto unlock;
>>  			goto reuse;
>> +		}
>>  		goto gotten;
>>  	}
>>  

Thank you Kirill, very much. I was hopping you'll have a look at this
see all the fine implications.

I will fix and send, after some hard testing.

Boaz
diff mbox

Patch

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 47a9392..1cd820c 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -250,6 +250,8 @@  struct vm_operations_struct {
 	/* notification that a previously read-only page is about to become
 	 * writable, if an error is returned it will cause a SIGBUS */
 	int (*page_mkwrite)(struct vm_area_struct *vma, struct vm_fault *vmf);
+	/* same as page_mkwrite when using VM_PFNMAP|VM_MIXEDMAP */
+	int (*pfn_mkwrite)(struct vm_area_struct *vma, struct vm_fault *vmf);
 
 	/* called by access_process_vm when get_user_pages() fails, typically
 	 * for use by special VMAs that can switch between memory and hardware
diff --git a/mm/memory.c b/mm/memory.c
index 8068893..8d640d1 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1982,6 +1982,23 @@  static int do_page_mkwrite(struct vm_area_struct *vma, struct page *page,
 	return ret;
 }
 
+static int do_pfn_mkwrite(struct vm_area_struct *vma, unsigned long address)
+{
+	if (vma->vm_ops && vma->vm_ops->pfn_mkwrite) {
+		struct vm_fault vmf = {
+			.page = 0,
+			.pgoff = (((address & PAGE_MASK) - vma->vm_start)
+						>> PAGE_SHIFT) + vma->vm_pgoff,
+			.virtual_address = (void __user *)(address & PAGE_MASK),
+			.flags = FAULT_FLAG_WRITE | FAULT_FLAG_MKWRITE,
+		};
+
+		return vma->vm_ops->pfn_mkwrite(vma, &vmf);
+	}
+
+	return 0;
+}
+
 /*
  * This routine handles present pages, when users try to write
  * to a shared page. It is done by copying the page to a new address
@@ -2025,8 +2042,17 @@  static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		 * accounting on raw pfn maps.
 		 */
 		if ((vma->vm_flags & (VM_WRITE|VM_SHARED)) ==
-				     (VM_WRITE|VM_SHARED))
+				     (VM_WRITE|VM_SHARED)) {
+			pte_unmap_unlock(page_table, ptl);
+			ret = do_pfn_mkwrite(vma, address);
+			if (ret & VM_FAULT_ERROR)
+				return ret;
+			page_table = pte_offset_map_lock(mm, pmd, address,
+							 &ptl);
+			if (!pte_same(*page_table, orig_pte))
+				goto unlock;
 			goto reuse;
+		}
 		goto gotten;
 	}