Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1365755 > unrolled thread

[RFC PATCH 1/2] mm/hugetlbfs: Attempt PUD_SIZE mapping alignment if PMD sharing enabled

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  [RFC PATCH 1/2] mm/hugetlbfs: Attempt PUD_SIZE mapping alignment if PMD sharing enabled Mike Kravetz <mike.kravetz@oracle.com> - 2016-03-29 03:20 +0200
    Re: [RFC PATCH 1/2] mm/hugetlbfs: Attempt PUD_SIZE mapping alignment if PMD sharing enabled "Hillf Danton" <hillf.zj@alibaba-inc.com> - 2016-03-29 06:00 +0200
      Re: [RFC PATCH 1/2] mm/hugetlbfs: Attempt PUD_SIZE mapping alignment  if PMD sharing enabled Mike Kravetz <mike.kravetz@oracle.com> - 2016-03-29 18:40 +0200
    Re: [RFC PATCH 1/2] mm/hugetlbfs: Attempt PUD_SIZE mapping  alignment if PMD sharing enabled Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> - 2016-03-31 04:30 +0200
      Re: [RFC PATCH 1/2] mm/hugetlbfs: Attempt PUD_SIZE mapping alignment  if PMD sharing enabled Mike Kravetz <mike.kravetz@oracle.com> - 2016-03-31 18:50 +0200

#1365755 — [RFC PATCH 1/2] mm/hugetlbfs: Attempt PUD_SIZE mapping alignment if PMD sharing enabled

When creating a hugetlb mapping, attempt PUD_SIZE alignment if the
following conditions are met:
- Address passed to mmap or shmat is NULL
- The mapping is flaged as shared
- The mapping is at least PUD_SIZE in length
If a PUD_SIZE aligned mapping can not be created, then fall back to a
huge page size mapping.

Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
---
 fs/hugetlbfs/inode.c | 29 +++++++++++++++++++++++++++--
 1 file changed, 27 insertions(+), 2 deletions(-)

diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 540ddc9..22b2e38 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -175,6 +175,17 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 	struct vm_area_struct *vma;
 	struct hstate *h = hstate_file(file);
 	struct vm_unmapped_area_info info;
+	bool pud_size_align = false;
+	unsigned long ret_addr;
+
+	/*
+	 * If PMD sharing is enabled, align to PUD_SIZE to facilitate
+	 * sharing.  Only attempt alignment if no address was passed in,
+	 * flags indicate sharing and size is big enough.
+	 */
+	if (IS_ENABLED(CONFIG_ARCH_WANT_HUGE_PMD_SHARE) &&
+	    !addr && flags & MAP_SHARED && len >= PUD_SIZE)
+		pud_size_align = true;
 
 	if (len & ~huge_page_mask(h))
 		return -EINVAL;
@@ -199,9 +210,23 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 	info.length = len;
 	info.low_limit = TASK_UNMAPPED_BASE;
 	info.high_limit = TASK_SIZE;
-	info.align_mask = PAGE_MASK & ~huge_page_mask(h);
+	if (pud_size_align)
+		info.align_mask = PAGE_MASK & (PUD_SIZE - 1);
+	else
+		info.align_mask = PAGE_MASK & ~huge_page_mask(h);
 	info.align_offset = 0;
-	return vm_unmapped_area(&info);
+	ret_addr = vm_unmapped_area(&info);
+
+	/*
+	 * If failed with PUD_SIZE alignment, try again with huge page
+	 * size alignment.
+	 */
+	if ((ret_addr & ~PAGE_MASK) && pud_size_align) {
+		info.align_mask = PAGE_MASK & ~huge_page_mask(h);
+		ret_addr = vm_unmapped_area(&info);
+	}
+
+	return ret_addr;
 }
 #endif
 
-- 
2.4.3

[toc] | [next] | [standalone]

#1365831

> 
> When creating a hugetlb mapping, attempt PUD_SIZE alignment if the
> following conditions are met:
> - Address passed to mmap or shmat is NULL
> - The mapping is flaged as shared
> - The mapping is at least PUD_SIZE in length
> If a PUD_SIZE aligned mapping can not be created, then fall back to a
> huge page size mapping.
> 
> Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
> ---
>  fs/hugetlbfs/inode.c | 29 +++++++++++++++++++++++++++--
>  1 file changed, 27 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
> index 540ddc9..22b2e38 100644
> --- a/fs/hugetlbfs/inode.c
> +++ b/fs/hugetlbfs/inode.c
> @@ -175,6 +175,17 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
>  	struct vm_area_struct *vma;
>  	struct hstate *h = hstate_file(file);
>  	struct vm_unmapped_area_info info;
> +	bool pud_size_align = false;
> +	unsigned long ret_addr;
> +
> +	/*
> +	 * If PMD sharing is enabled, align to PUD_SIZE to facilitate
> +	 * sharing.  Only attempt alignment if no address was passed in,
> +	 * flags indicate sharing and size is big enough.
> +	 */
> +	if (IS_ENABLED(CONFIG_ARCH_WANT_HUGE_PMD_SHARE) &&
> +	    !addr && flags & MAP_SHARED && len >= PUD_SIZE)
> +		pud_size_align = true;
> 
>  	if (len & ~huge_page_mask(h))
>  		return -EINVAL;
> @@ -199,9 +210,23 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
>  	info.length = len;
>  	info.low_limit = TASK_UNMAPPED_BASE;
>  	info.high_limit = TASK_SIZE;
> -	info.align_mask = PAGE_MASK & ~huge_page_mask(h);
> +	if (pud_size_align)
> +		info.align_mask = PAGE_MASK & (PUD_SIZE - 1);
> +	else
> +		info.align_mask = PAGE_MASK & ~huge_page_mask(h);
>  	info.align_offset = 0;
> -	return vm_unmapped_area(&info);
> +	ret_addr = vm_unmapped_area(&info);
> +
> +	/*
> +	 * If failed with PUD_SIZE alignment, try again with huge page
> +	 * size alignment.
> +	 */

Can we avoid going another round as long as it is a file with
the PUD page size?

Hillf
> +	if ((ret_addr & ~PAGE_MASK) && pud_size_align) {
> +		info.align_mask = PAGE_MASK & ~huge_page_mask(h);
> +		ret_addr = vm_unmapped_area(&info);
> +	}
> +
> +	return ret_addr;
>  }
>  #endif
> 
> --
> 2.4.3

[toc] | [prev] | [next] | [standalone]

#1366449 — Re: [RFC PATCH 1/2] mm/hugetlbfs: Attempt PUD_SIZE mapping alignment if PMD sharing enabled

On 03/28/2016 08:50 PM, Hillf Danton wrote:
>>
>> When creating a hugetlb mapping, attempt PUD_SIZE alignment if the
>> following conditions are met:
>> - Address passed to mmap or shmat is NULL
>> - The mapping is flaged as shared
>> - The mapping is at least PUD_SIZE in length
>> If a PUD_SIZE aligned mapping can not be created, then fall back to a
>> huge page size mapping.
>>
>> Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
>> ---
>>  fs/hugetlbfs/inode.c | 29 +++++++++++++++++++++++++++--
>>  1 file changed, 27 insertions(+), 2 deletions(-)
>>
>> diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
>> index 540ddc9..22b2e38 100644
>> --- a/fs/hugetlbfs/inode.c
>> +++ b/fs/hugetlbfs/inode.c
>> @@ -175,6 +175,17 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
>>  	struct vm_area_struct *vma;
>>  	struct hstate *h = hstate_file(file);
>>  	struct vm_unmapped_area_info info;
>> +	bool pud_size_align = false;
>> +	unsigned long ret_addr;
>> +
>> +	/*
>> +	 * If PMD sharing is enabled, align to PUD_SIZE to facilitate
>> +	 * sharing.  Only attempt alignment if no address was passed in,
>> +	 * flags indicate sharing and size is big enough.
>> +	 */
>> +	if (IS_ENABLED(CONFIG_ARCH_WANT_HUGE_PMD_SHARE) &&
>> +	    !addr && flags & MAP_SHARED && len >= PUD_SIZE)
>> +		pud_size_align = true;
>>
>>  	if (len & ~huge_page_mask(h))
>>  		return -EINVAL;
>> @@ -199,9 +210,23 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
>>  	info.length = len;
>>  	info.low_limit = TASK_UNMAPPED_BASE;
>>  	info.high_limit = TASK_SIZE;
>> -	info.align_mask = PAGE_MASK & ~huge_page_mask(h);
>> +	if (pud_size_align)
>> +		info.align_mask = PAGE_MASK & (PUD_SIZE - 1);
>> +	else
>> +		info.align_mask = PAGE_MASK & ~huge_page_mask(h);
>>  	info.align_offset = 0;
>> -	return vm_unmapped_area(&info);
>> +	ret_addr = vm_unmapped_area(&info);
>> +
>> +	/*
>> +	 * If failed with PUD_SIZE alignment, try again with huge page
>> +	 * size alignment.
>> +	 */
> 
> Can we avoid going another round as long as it is a file with
> the PUD page size?

Yes, that brings up a good point.

Since we only do PMD sharing with PMD_SIZE huge pages, that should be
part of the check as to whether we try PUD_SIZE alignment.  The initial
check should be expanded as follows:

if (IS_ENABLED(CONFIG_ARCH_WANT_HUGE_PMD_SHARE) && !addr &&
    flags & MAP_SHARED && huge_page_size(h) == PMD_SIZE && len >= PUD_SIZE)
	pud_size_align = true;

In that case, pud_size_align remains false and we do not retry.

-- 
Mike Kravetz

> 
> Hillf
>> +	if ((ret_addr & ~PAGE_MASK) && pud_size_align) {
>> +		info.align_mask = PAGE_MASK & ~huge_page_mask(h);
>> +		ret_addr = vm_unmapped_area(&info);
>> +	}
>> +
>> +	return ret_addr;
>>  }
>>  #endif
>>
>> --
>> 2.4.3
> 

[toc] | [prev] | [next] | [standalone]

#1367755 — Re: [RFC PATCH 1/2] mm/hugetlbfs: Attempt PUD_SIZE mapping alignment if PMD sharing enabled

On Mon, Mar 28, 2016 at 06:12:49PM -0700, Mike Kravetz wrote:
> When creating a hugetlb mapping, attempt PUD_SIZE alignment if the
> following conditions are met:
> - Address passed to mmap or shmat is NULL
> - The mapping is flaged as shared
> - The mapping is at least PUD_SIZE in length
> If a PUD_SIZE aligned mapping can not be created, then fall back to a
> huge page size mapping.

It would be kinder if the patch description includes why this change.
Simply "to facilitate pmd sharing" is helpful for someone who read
"git log".

> 
> Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
> ---
>  fs/hugetlbfs/inode.c | 29 +++++++++++++++++++++++++++--
>  1 file changed, 27 insertions(+), 2 deletions(-)
> 
> diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
> index 540ddc9..22b2e38 100644
> --- a/fs/hugetlbfs/inode.c
> +++ b/fs/hugetlbfs/inode.c
> @@ -175,6 +175,17 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
>  	struct vm_area_struct *vma;
>  	struct hstate *h = hstate_file(file);
>  	struct vm_unmapped_area_info info;
> +	bool pud_size_align = false;
> +	unsigned long ret_addr;
> +
> +	/*
> +	 * If PMD sharing is enabled, align to PUD_SIZE to facilitate
> +	 * sharing.  Only attempt alignment if no address was passed in,
> +	 * flags indicate sharing and size is big enough.
> +	 */
> +	if (IS_ENABLED(CONFIG_ARCH_WANT_HUGE_PMD_SHARE) &&
> +	    !addr && flags & MAP_SHARED && len >= PUD_SIZE)
> +		pud_size_align = true;

This code will have duplicates in the next patch, so how about checking
this in a separate check routine?

Thanks,
Naoya Horiguchi

[toc] | [prev] | [next] | [standalone]

#1368439 — Re: [RFC PATCH 1/2] mm/hugetlbfs: Attempt PUD_SIZE mapping alignment if PMD sharing enabled

On 03/30/2016 07:18 PM, Naoya Horiguchi wrote:
> On Mon, Mar 28, 2016 at 06:12:49PM -0700, Mike Kravetz wrote:
>> When creating a hugetlb mapping, attempt PUD_SIZE alignment if the
>> following conditions are met:
>> - Address passed to mmap or shmat is NULL
>> - The mapping is flaged as shared
>> - The mapping is at least PUD_SIZE in length
>> If a PUD_SIZE aligned mapping can not be created, then fall back to a
>> huge page size mapping.
> 
> It would be kinder if the patch description includes why this change.
> Simply "to facilitate pmd sharing" is helpful for someone who read
> "git log".

Ok, will do.

> 
>>
>> Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
>> ---
>>  fs/hugetlbfs/inode.c | 29 +++++++++++++++++++++++++++--
>>  1 file changed, 27 insertions(+), 2 deletions(-)
>>
>> diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
>> index 540ddc9..22b2e38 100644
>> --- a/fs/hugetlbfs/inode.c
>> +++ b/fs/hugetlbfs/inode.c
>> @@ -175,6 +175,17 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
>>  	struct vm_area_struct *vma;
>>  	struct hstate *h = hstate_file(file);
>>  	struct vm_unmapped_area_info info;
>> +	bool pud_size_align = false;
>> +	unsigned long ret_addr;
>> +
>> +	/*
>> +	 * If PMD sharing is enabled, align to PUD_SIZE to facilitate
>> +	 * sharing.  Only attempt alignment if no address was passed in,
>> +	 * flags indicate sharing and size is big enough.
>> +	 */
>> +	if (IS_ENABLED(CONFIG_ARCH_WANT_HUGE_PMD_SHARE) &&
>> +	    !addr && flags & MAP_SHARED && len >= PUD_SIZE)
>> +		pud_size_align = true;
> 
> This code will have duplicates in the next patch, so how about checking
> this in a separate check routine?

Good suggestion,  thanks
-- 
Mike Kravetz

> 
> Thanks,
> Naoya Horiguchi
> 

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web