[Patch v4 1/2] mm/rmap.c: don't reuse anon_vma if we just want a copy

[Patch v4 1/2] mm/rmap.c: don't reuse anon_vma if we just want a copy

[Wei Yang]

Before commit 7a3ef208e662 ("mm: prevent endless growth of anon_vma
hierarchy"), anon_vma_clone() doesn't change dst->anon_vma. While after
this commit, anon_vma_clone() will try to reuse an exist one on forking.

But this commit go a little bit further for the case not forking.
anon_vma_clone() is called from __vma_split(), __split_vma(), copy_vma()
and anon_vma_fork(). For the first three places, the purpose here is get
a copy of src and we don't expect to touch dst->anon_vma even it is
NULL. While after that commit, it is possible to reuse an anon_vma when
dst->anon_vma is NULL. This is not we intend to have.

This patch stop reuse anon_vma for non-fork cases.

Fix commit 7a3ef208e662 ("mm: prevent endless growth of anon_vma
hierarchy")

Signed-off-by: Wei Yang <richardw.yang@linux.intel.com>

---
v4:
  * check dst->anon_vma in each iteration
v3:
  * use dst->anon_vma and src->anon_vma to get reuse state
    pointed by Konstantin Khlebnikov
---
 mm/rmap.c | 24 +++++++++++++++---------
 1 file changed, 15 insertions(+), 9 deletions(-)

diff --git a/mm/rmap.c b/mm/rmap.c
index d9a23bb773bf..c34414567474 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -250,13 +250,19 @@ static inline void unlock_anon_vma_root(struct anon_vma *root)
  * Attach the anon_vmas from src to dst.
  * Returns 0 on success, -ENOMEM on failure.
  *
- * If dst->anon_vma is NULL this function tries to find and reuse existing
- * anon_vma which has no vmas and only one child anon_vma. This prevents
- * degradation of anon_vma hierarchy to endless linear chain in case of
- * constantly forking task. On the other hand, an anon_vma with more than one
- * child isn't reused even if there was no alive vma, thus rmap walker has a
- * good chance of avoiding scanning the whole hierarchy when it searches where
- * page is mapped.
+ * anon_vma_clone() is called by __vma_split(), __split_vma(), copy_vma() and
+ * anon_vma_fork(). The first three want an exact copy of src, while the last
+ * one, anon_vma_fork(), may try to reuse an existing anon_vma to prevent
+ * endless growth of anon_vma. Since dst->anon_vma is set to NULL before call,
+ * we can identify this case by checking (!dst->anon_vma && src->anon_vma).
+ *
+ * If (!dst->anon_vma && src->anon_vma) is true, this function tries to find
+ * and reuse existing anon_vma which has no vmas and only one child anon_vma.
+ * This prevents degradation of anon_vma hierarchy to endless linear chain in
+ * case of constantly forking task. On the other hand, an anon_vma with more
+ * than one child isn't reused even if there was no alive vma, thus rmap
+ * walker has a good chance of avoiding scanning the whole hierarchy when it
+ * searches where page is mapped.
  */
 int anon_vma_clone(struct vm_area_struct *dst, struct vm_area_struct *src)
 {
@@ -286,8 +292,8 @@ int anon_vma_clone(struct vm_area_struct *dst, struct vm_area_struct *src)
 		 * will always reuse it. Root anon_vma is never reused:
 		 * it has self-parent reference and at least one child.
 		 */
-		if (!dst->anon_vma && anon_vma != src->anon_vma &&
-				anon_vma->degree < 2)
+		if (!dst->anon_vma && src->anon_vma &&
+		    anon_vma != src->anon_vma && anon_vma->degree < 2)
 			dst->anon_vma = anon_vma;
 	}
 	if (dst->anon_vma)
-- 
2.17.1

[Patch v4 2/2] mm/rmap.c: reuse mergeable anon_vma as parent when fork

[Wei Yang]

In function __anon_vma_prepare(), we will try to find anon_vma if it is
possible to reuse it. While on fork, the logic is different.

Since commit 5beb49305251 ("mm: change anon_vma linking to fix
multi-process server scalability issue"), function anon_vma_clone()
tries to allocate new anon_vma for child process. But the logic here
will allocate a new anon_vma for each vma, even in parent this vma
is mergeable and share the same anon_vma with its sibling. This may do
better for scalability issue, while it is not necessary to do so
especially after interval tree is used.

Commit 7a3ef208e662 ("mm: prevent endless growth of anon_vma hierarchy")
tries to reuse some anon_vma by counting child anon_vma and attached
vmas. While for those mergeable anon_vmas, we can just reuse it and not
necessary to go through the logic.

After this change, kernel build test reduces 20% anon_vma allocation.

Do the same kernel build test, it shows run time in sys reduced 11.6%.

Origin:

real    2m50.467s
user    17m52.002s
sys     1m51.953s

real    2m48.662s
user    17m55.464s
sys     1m50.553s

real    2m51.143s
user    17m59.687s
sys     1m53.600s

Patched:

real	2m39.933s
user	17m1.835s
sys	1m38.802s

real	2m39.321s
user	17m1.634s
sys	1m39.206s

real	2m39.575s
user	17m1.420s
sys	1m38.845s

Signed-off-by: Wei Yang <richardw.yang@linux.intel.com>
---
 mm/rmap.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/mm/rmap.c b/mm/rmap.c
index c34414567474..2c13e2bfd393 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -268,6 +268,19 @@ int anon_vma_clone(struct vm_area_struct *dst, struct vm_area_struct *src)
 {
 	struct anon_vma_chain *avc, *pavc;
 	struct anon_vma *root = NULL;
+	struct vm_area_struct *prev = dst->vm_prev, *pprev = src->vm_prev;
+
+	/*
+	 * If parent share anon_vma with its vm_prev, keep this sharing in in
+	 * child.
+	 *
+	 * 1. Parent has vm_prev, which implies we have vm_prev.
+	 * 2. Parent and its vm_prev have the same anon_vma.
+	 */
+	if (!dst->anon_vma && src->anon_vma &&
+	    pprev && pprev->anon_vma == src->anon_vma)
+		dst->anon_vma = prev->anon_vma;
+
 
 	list_for_each_entry_reverse(pavc, &src->anon_vma_chain, same_vma) {
 		struct anon_vma *anon_vma;
-- 
2.17.1

Re: [Patch v4 2/2] mm/rmap.c: reuse mergeable anon_vma as parent when fork

[Konstantin Khlebnikov]

On 11/10/2019 10.22, Wei Yang wrote:
> In function __anon_vma_prepare(), we will try to find anon_vma if it is
> possible to reuse it. While on fork, the logic is different.
> 
> Since commit 5beb49305251 ("mm: change anon_vma linking to fix
> multi-process server scalability issue"), function anon_vma_clone()
> tries to allocate new anon_vma for child process. But the logic here
> will allocate a new anon_vma for each vma, even in parent this vma
> is mergeable and share the same anon_vma with its sibling. This may do
> better for scalability issue, while it is not necessary to do so
> especially after interval tree is used.
> 
> Commit 7a3ef208e662 ("mm: prevent endless growth of anon_vma hierarchy")
> tries to reuse some anon_vma by counting child anon_vma and attached
> vmas. While for those mergeable anon_vmas, we can just reuse it and not
> necessary to go through the logic.
> 
> After this change, kernel build test reduces 20% anon_vma allocation.
> 
> Do the same kernel build test, it shows run time in sys reduced 11.6%.
> 
> Origin:
> 
> real    2m50.467s
> user    17m52.002s
> sys     1m51.953s
> 
> real    2m48.662s
> user    17m55.464s
> sys     1m50.553s
> 
> real    2m51.143s
> user    17m59.687s
> sys     1m53.600s
> 
> Patched:
> 
> real	2m39.933s
> user	17m1.835s
> sys	1m38.802s
> 
> real	2m39.321s
> user	17m1.634s
> sys	1m39.206s
> 
> real	2m39.575s
> user	17m1.420s
> sys	1m38.845s
> 
> Signed-off-by: Wei Yang <richardw.yang@linux.intel.com>

Acked-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>

> ---
>   mm/rmap.c | 13 +++++++++++++
>   1 file changed, 13 insertions(+)
> 
> diff --git a/mm/rmap.c b/mm/rmap.c
> index c34414567474..2c13e2bfd393 100644
> --- a/mm/rmap.c
> +++ b/mm/rmap.c
> @@ -268,6 +268,19 @@ int anon_vma_clone(struct vm_area_struct *dst, struct vm_area_struct *src)
>   {
>   	struct anon_vma_chain *avc, *pavc;
>   	struct anon_vma *root = NULL;
> +	struct vm_area_struct *prev = dst->vm_prev, *pprev = src->vm_prev;
> +
> +	/*
> +	 * If parent share anon_vma with its vm_prev, keep this sharing in in
> +	 * child.
> +	 *
> +	 * 1. Parent has vm_prev, which implies we have vm_prev.
> +	 * 2. Parent and its vm_prev have the same anon_vma.
> +	 */
> +	if (!dst->anon_vma && src->anon_vma &&
> +	    pprev && pprev->anon_vma == src->anon_vma)
> +		dst->anon_vma = prev->anon_vma;
> +

I believe that in present code "prev" cannot be NULL if !dst->anon_vma && src->anon_vma is true.
It would be safer to check this explicitly.

>   
>   	list_for_each_entry_reverse(pavc, &src->anon_vma_chain, same_vma) {
>   		struct anon_vma *anon_vma;
> 

Re: [Patch v4 1/2] mm/rmap.c: don't reuse anon_vma if we just want a copy

[Konstantin Khlebnikov]

On 11/10/2019 10.22, Wei Yang wrote:
> Before commit 7a3ef208e662 ("mm: prevent endless growth of anon_vma
> hierarchy"), anon_vma_clone() doesn't change dst->anon_vma. While after
> this commit, anon_vma_clone() will try to reuse an exist one on forking.
> 
> But this commit go a little bit further for the case not forking.
> anon_vma_clone() is called from __vma_split(), __split_vma(), copy_vma()
> and anon_vma_fork(). For the first three places, the purpose here is get
> a copy of src and we don't expect to touch dst->anon_vma even it is
> NULL. While after that commit, it is possible to reuse an anon_vma when
> dst->anon_vma is NULL. This is not we intend to have.
> 
> This patch stop reuse anon_vma for non-fork cases.
> 
> Fix commit 7a3ef208e662 ("mm: prevent endless growth of anon_vma
> hierarchy")
> 
> Signed-off-by: Wei Yang <richardw.yang@linux.intel.com>

Yes, reusing heuristic was designed for fork.
But this isn't strictly necessary - any vmas could share anon_vma.
For example all vmas in system could be linked with single anon_vma.

Acked-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>

> 
> ---
> v4:
>    * check dst->anon_vma in each iteration
> v3:
>    * use dst->anon_vma and src->anon_vma to get reuse state
>      pointed by Konstantin Khlebnikov
> ---
>   mm/rmap.c | 24 +++++++++++++++---------
>   1 file changed, 15 insertions(+), 9 deletions(-)
> 
> diff --git a/mm/rmap.c b/mm/rmap.c
> index d9a23bb773bf..c34414567474 100644
> --- a/mm/rmap.c
> +++ b/mm/rmap.c
> @@ -250,13 +250,19 @@ static inline void unlock_anon_vma_root(struct anon_vma *root)
>    * Attach the anon_vmas from src to dst.
>    * Returns 0 on success, -ENOMEM on failure.
>    *
> - * If dst->anon_vma is NULL this function tries to find and reuse existing
> - * anon_vma which has no vmas and only one child anon_vma. This prevents
> - * degradation of anon_vma hierarchy to endless linear chain in case of
> - * constantly forking task. On the other hand, an anon_vma with more than one
> - * child isn't reused even if there was no alive vma, thus rmap walker has a
> - * good chance of avoiding scanning the whole hierarchy when it searches where
> - * page is mapped.
> + * anon_vma_clone() is called by __vma_split(), __split_vma(), copy_vma() and
> + * anon_vma_fork(). The first three want an exact copy of src, while the last
> + * one, anon_vma_fork(), may try to reuse an existing anon_vma to prevent
> + * endless growth of anon_vma. Since dst->anon_vma is set to NULL before call,
> + * we can identify this case by checking (!dst->anon_vma && src->anon_vma).
> + *
> + * If (!dst->anon_vma && src->anon_vma) is true, this function tries to find
> + * and reuse existing anon_vma which has no vmas and only one child anon_vma.
> + * This prevents degradation of anon_vma hierarchy to endless linear chain in
> + * case of constantly forking task. On the other hand, an anon_vma with more
> + * than one child isn't reused even if there was no alive vma, thus rmap
> + * walker has a good chance of avoiding scanning the whole hierarchy when it
> + * searches where page is mapped.
>    */
>   int anon_vma_clone(struct vm_area_struct *dst, struct vm_area_struct *src)
>   {
> @@ -286,8 +292,8 @@ int anon_vma_clone(struct vm_area_struct *dst, struct vm_area_struct *src)
>   		 * will always reuse it. Root anon_vma is never reused:
>   		 * it has self-parent reference and at least one child.
>   		 */
> -		if (!dst->anon_vma && anon_vma != src->anon_vma &&
> -				anon_vma->degree < 2)
> +		if (!dst->anon_vma && src->anon_vma &&
> +		    anon_vma != src->anon_vma && anon_vma->degree < 2)
>   			dst->anon_vma = anon_vma;
>   	}
>   	if (dst->anon_vma)
> 

Re: [Patch v4 1/2] mm/rmap.c: don't reuse anon_vma if we just want a copy

[Wei Yang]

On Fri, Oct 11, 2019 at 11:04:16AM +0300, Konstantin Khlebnikov wrote:
>On 11/10/2019 10.22, Wei Yang wrote:
>> Before commit 7a3ef208e662 ("mm: prevent endless growth of anon_vma
>> hierarchy"), anon_vma_clone() doesn't change dst->anon_vma. While after
>> this commit, anon_vma_clone() will try to reuse an exist one on forking.
>> 
>> But this commit go a little bit further for the case not forking.
>> anon_vma_clone() is called from __vma_split(), __split_vma(), copy_vma()
>> and anon_vma_fork(). For the first three places, the purpose here is get
>> a copy of src and we don't expect to touch dst->anon_vma even it is
>> NULL. While after that commit, it is possible to reuse an anon_vma when
>> dst->anon_vma is NULL. This is not we intend to have.
>> 
>> This patch stop reuse anon_vma for non-fork cases.
>> 
>> Fix commit 7a3ef208e662 ("mm: prevent endless growth of anon_vma
>> hierarchy")
>> 
>> Signed-off-by: Wei Yang <richardw.yang@linux.intel.com>
>
>Yes, reusing heuristic was designed for fork.
>But this isn't strictly necessary - any vmas could share anon_vma.
>For example all vmas in system could be linked with single anon_vma.

Yes, agree with you.

Thanks for your comment :-)

>
>Acked-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
>
>> 
>> ---
>> v4:
>>    * check dst->anon_vma in each iteration
>> v3:
>>    * use dst->anon_vma and src->anon_vma to get reuse state
>>      pointed by Konstantin Khlebnikov
>> ---
>>   mm/rmap.c | 24 +++++++++++++++---------
>>   1 file changed, 15 insertions(+), 9 deletions(-)
>> 
>> diff --git a/mm/rmap.c b/mm/rmap.c
>> index d9a23bb773bf..c34414567474 100644
>> --- a/mm/rmap.c
>> +++ b/mm/rmap.c
>> @@ -250,13 +250,19 @@ static inline void unlock_anon_vma_root(struct anon_vma *root)
>>    * Attach the anon_vmas from src to dst.
>>    * Returns 0 on success, -ENOMEM on failure.
>>    *
>> - * If dst->anon_vma is NULL this function tries to find and reuse existing
>> - * anon_vma which has no vmas and only one child anon_vma. This prevents
>> - * degradation of anon_vma hierarchy to endless linear chain in case of
>> - * constantly forking task. On the other hand, an anon_vma with more than one
>> - * child isn't reused even if there was no alive vma, thus rmap walker has a
>> - * good chance of avoiding scanning the whole hierarchy when it searches where
>> - * page is mapped.
>> + * anon_vma_clone() is called by __vma_split(), __split_vma(), copy_vma() and
>> + * anon_vma_fork(). The first three want an exact copy of src, while the last
>> + * one, anon_vma_fork(), may try to reuse an existing anon_vma to prevent
>> + * endless growth of anon_vma. Since dst->anon_vma is set to NULL before call,
>> + * we can identify this case by checking (!dst->anon_vma && src->anon_vma).
>> + *
>> + * If (!dst->anon_vma && src->anon_vma) is true, this function tries to find
>> + * and reuse existing anon_vma which has no vmas and only one child anon_vma.
>> + * This prevents degradation of anon_vma hierarchy to endless linear chain in
>> + * case of constantly forking task. On the other hand, an anon_vma with more
>> + * than one child isn't reused even if there was no alive vma, thus rmap
>> + * walker has a good chance of avoiding scanning the whole hierarchy when it
>> + * searches where page is mapped.
>>    */
>>   int anon_vma_clone(struct vm_area_struct *dst, struct vm_area_struct *src)
>>   {
>> @@ -286,8 +292,8 @@ int anon_vma_clone(struct vm_area_struct *dst, struct vm_area_struct *src)
>>   		 * will always reuse it. Root anon_vma is never reused:
>>   		 * it has self-parent reference and at least one child.
>>   		 */
>> -		if (!dst->anon_vma && anon_vma != src->anon_vma &&
>> -				anon_vma->degree < 2)
>> +		if (!dst->anon_vma && src->anon_vma &&
>> +		    anon_vma != src->anon_vma && anon_vma->degree < 2)
>>   			dst->anon_vma = anon_vma;
>>   	}
>>   	if (dst->anon_vma)
>> 

-- 
Wei Yang
Help you, Help me