[PATCH 0/6] Cleanup and fixup for z3fold

[PATCH 0/6] Cleanup and fixup for z3fold

[Miaohe Lin]

Hi all,
This series contains cleanups to remove unused function, redefine macro
to improve readability and so on. Also this fixes several bugs in z3fold,
such as memory leak in z3fold_destroy_pool(). More details can be found
in the respective changelogs. Thanks!

Miaohe Lin (6):
  mm/z3fold: define macro NCHUNKS as TOTAL_CHUNKS - ZHDR_CHUNKS
  mm/z3fold: avoid possible underflow in z3fold_alloc()
  mm/z3fold: remove magic number in z3fold_create_pool()
  mm/z3fold: remove unused function handle_to_z3fold_header()
  mm/z3fold: fix potential memory leak in z3fold_destroy_pool()
  mm/z3fold: use release_z3fold_page_locked() to release locked z3fold
    page

 mm/z3fold.c | 37 ++++++++++++++-----------------------
 1 file changed, 14 insertions(+), 23 deletions(-)

-- 
2.23.0

[PATCH 1/6] mm/z3fold: define macro NCHUNKS as TOTAL_CHUNKS - ZHDR_CHUNKS

[Miaohe Lin]

To improve the code readability, we could define macro NCHUNKS as
TOTAL_CHUNKS - ZHDR_CHUNKS. No functional change intended.

Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
---
 mm/z3fold.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/z3fold.c b/mm/z3fold.c
index 56321bf37d56..04d0e493bd2e 100644
--- a/mm/z3fold.c
+++ b/mm/z3fold.c
@@ -62,7 +62,7 @@
 #define ZHDR_SIZE_ALIGNED round_up(sizeof(struct z3fold_header), CHUNK_SIZE)
 #define ZHDR_CHUNKS	(ZHDR_SIZE_ALIGNED >> CHUNK_SHIFT)
 #define TOTAL_CHUNKS	(PAGE_SIZE >> CHUNK_SHIFT)
-#define NCHUNKS		((PAGE_SIZE - ZHDR_SIZE_ALIGNED) >> CHUNK_SHIFT)
+#define NCHUNKS		(TOTAL_CHUNKS - ZHDR_CHUNKS)
 
 #define BUDDY_MASK	(0x3)
 #define BUDDY_SHIFT	2
-- 
2.23.0

[PATCH 2/6] mm/z3fold: avoid possible underflow in z3fold_alloc()

[Miaohe Lin]

It is not enough to just make sure the z3fold header is not larger than the
page size. When z3fold header is equal to PAGE_SIZE, we would underflow
when check alloc size against PAGE_SIZE - ZHDR_SIZE_ALIGNED - CHUNK_SIZE
in z3fold_alloc(). Make sure there has remaining spaces for its buddy to
fix this theoretical issue.

Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
---
This causes the below checkpatch warning:
WARNING: Comparisons should place the constant on the right side of the
test
#31: FILE: mm/z3fold.c:1812:
+       BUILD_BUG_ON(ZHDR_SIZE_ALIGNED > PAGE_SIZE - CHUNK_SIZE);

But I think the error is false positives as all members are constant.
---
 mm/z3fold.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/mm/z3fold.c b/mm/z3fold.c
index 04d0e493bd2e..e261e14b7753 100644
--- a/mm/z3fold.c
+++ b/mm/z3fold.c
@@ -1805,8 +1805,11 @@ static int __init init_z3fold(void)
 {
 	int ret;
 
-	/* Make sure the z3fold header is not larger than the page size */
-	BUILD_BUG_ON(ZHDR_SIZE_ALIGNED > PAGE_SIZE);
+	/*
+	 * Make sure the z3fold header is not larger than the page size and
+	 * there has remaining spaces for its buddy.
+	 */
+	BUILD_BUG_ON(ZHDR_SIZE_ALIGNED > PAGE_SIZE - CHUNK_SIZE);
 	ret = z3fold_mount();
 	if (ret)
 		return ret;
-- 
2.23.0

[PATCH 3/6] mm/z3fold: remove magic number in z3fold_create_pool()

[Miaohe Lin]

It's meaningless to pass a magic number 2 to __alloc_percpu() as there is
a minimum alignment size of PCPU_MIN_ALLOC_SIZE (> 2) in it. Also there is
no special alignment requirement for unbuddied. So we could replace this
magic number with nature alignment, i.e. __alignof__(struct list_head), to
improve readability.

Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
---
 mm/z3fold.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/mm/z3fold.c b/mm/z3fold.c
index e261e14b7753..988d57c143fd 100644
--- a/mm/z3fold.c
+++ b/mm/z3fold.c
@@ -1000,7 +1000,8 @@ static struct z3fold_pool *z3fold_create_pool(const char *name, gfp_t gfp,
 		goto out_c;
 	spin_lock_init(&pool->lock);
 	spin_lock_init(&pool->stale_lock);
-	pool->unbuddied = __alloc_percpu(sizeof(struct list_head)*NCHUNKS, 2);
+	pool->unbuddied = __alloc_percpu(sizeof(struct list_head) * NCHUNKS,
+					 __alignof__(struct list_head));
 	if (!pool->unbuddied)
 		goto out_pool;
 	for_each_possible_cpu(cpu) {
-- 
2.23.0

[PATCH 4/6] mm/z3fold: remove unused function handle_to_z3fold_header()

[Miaohe Lin]

handle_to_z3fold_header() is unused now. So we can remove it. As a result,
get_z3fold_header() becomes the only caller of __get_z3fold_header() and
the argument lock is always true. Therefore we could further fold the
__get_z3fold_header() into get_z3fold_header() with lock = true.

Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
---
 mm/z3fold.c | 22 ++++------------------
 1 file changed, 4 insertions(+), 18 deletions(-)

diff --git a/mm/z3fold.c b/mm/z3fold.c
index 988d57c143fd..bab08c08bf19 100644
--- a/mm/z3fold.c
+++ b/mm/z3fold.c
@@ -255,9 +255,8 @@ static inline void z3fold_page_unlock(struct z3fold_header *zhdr)
 	spin_unlock(&zhdr->page_lock);
 }
 
-
-static inline struct z3fold_header *__get_z3fold_header(unsigned long handle,
-							bool lock)
+/* return locked z3fold page if it's not headless */
+static inline struct z3fold_header *get_z3fold_header(unsigned long handle)
 {
 	struct z3fold_buddy_slots *slots;
 	struct z3fold_header *zhdr;
@@ -271,13 +270,12 @@ static inline struct z3fold_header *__get_z3fold_header(unsigned long handle,
 			read_lock(&slots->lock);
 			addr = *(unsigned long *)handle;
 			zhdr = (struct z3fold_header *)(addr & PAGE_MASK);
-			if (lock)
-				locked = z3fold_page_trylock(zhdr);
+			locked = z3fold_page_trylock(zhdr);
 			read_unlock(&slots->lock);
 			if (locked)
 				break;
 			cpu_relax();
-		} while (lock);
+		} while (true);
 	} else {
 		zhdr = (struct z3fold_header *)(handle & PAGE_MASK);
 	}
@@ -285,18 +283,6 @@ static inline struct z3fold_header *__get_z3fold_header(unsigned long handle,
 	return zhdr;
 }
 
-/* Returns the z3fold page where a given handle is stored */
-static inline struct z3fold_header *handle_to_z3fold_header(unsigned long h)
-{
-	return __get_z3fold_header(h, false);
-}
-
-/* return locked z3fold page if it's not headless */
-static inline struct z3fold_header *get_z3fold_header(unsigned long h)
-{
-	return __get_z3fold_header(h, true);
-}
-
 static inline void put_z3fold_header(struct z3fold_header *zhdr)
 {
 	struct page *page = virt_to_page(zhdr);
-- 
2.23.0

[PATCH 5/6] mm/z3fold: fix potential memory leak in z3fold_destroy_pool()

[Miaohe Lin]

There is a memoryleak in z3fold_destroy_pool() as it forgets to free_percpu
pool->unbuddied. Call free_percpu for pool->unbuddied to fix this issue.

Fixes: d30561c56f41 ("z3fold: use per-cpu unbuddied lists")
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
---
 mm/z3fold.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/mm/z3fold.c b/mm/z3fold.c
index bab08c08bf19..196d886a3436 100644
--- a/mm/z3fold.c
+++ b/mm/z3fold.c
@@ -1048,6 +1048,7 @@ static void z3fold_destroy_pool(struct z3fold_pool *pool)
 	destroy_workqueue(pool->compact_wq);
 	destroy_workqueue(pool->release_wq);
 	z3fold_unregister_migration(pool);
+	free_percpu(pool->unbuddied);
 	kfree(pool);
 }
 
-- 
2.23.0

[PATCH 6/6] mm/z3fold: use release_z3fold_page_locked() to release locked z3fold page

[Miaohe Lin]

We should use release_z3fold_page_locked() to release z3fold page when it's
locked, although it looks harmless to use release_z3fold_page() now.

Fixes: dcf5aedb24f8 ("z3fold: stricter locking and more careful reclaim")
Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
---
 mm/z3fold.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/z3fold.c b/mm/z3fold.c
index 196d886a3436..b3c0577b8095 100644
--- a/mm/z3fold.c
+++ b/mm/z3fold.c
@@ -1372,7 +1372,7 @@ static int z3fold_reclaim_page(struct z3fold_pool *pool, unsigned int retries)
 			if (zhdr->foreign_handles ||
 			    test_and_set_bit(PAGE_CLAIMED, &page->private)) {
 				if (kref_put(&zhdr->refcount,
-						release_z3fold_page))
+						release_z3fold_page_locked))
 					atomic64_dec(&pool->pages_nr);
 				else
 					z3fold_page_unlock(zhdr);
-- 
2.23.0

Re: [PATCH 6/6] mm/z3fold: use release_z3fold_page_locked() to release locked z3fold page

[Hillf Danton]

On Sat, 19 Jun 2021 17:31:51 +0800 Miaohe Lin wrote:
> We should use release_z3fold_page_locked() to release z3fold page when it's
> locked, although it looks harmless to use release_z3fold_page() now.
> 
> Fixes: dcf5aedb24f8 ("z3fold: stricter locking and more careful reclaim")
> Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
> ---
>  mm/z3fold.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/mm/z3fold.c b/mm/z3fold.c
> index 196d886a3436..b3c0577b8095 100644
> --- a/mm/z3fold.c
> +++ b/mm/z3fold.c
> @@ -1372,7 +1372,7 @@ static int z3fold_reclaim_page(struct z3fold_pool *pool, unsigned int retries)
>  			if (zhdr->foreign_handles ||
>  			    test_and_set_bit(PAGE_CLAIMED, &page->private)) {
>  				if (kref_put(&zhdr->refcount,
> -						release_z3fold_page))
> +						release_z3fold_page_locked))
>  					atomic64_dec(&pool->pages_nr);

LGTM. JFYI other issue in z3fold was reported [1] and if the fix proposed there
makes any sense to you feel free to pick it up and ask Mike for his tests.

[1] https://lore.kernel.org/linux-mm/20210316061351.1649-1-hdanton@sina.com/

Re: [PATCH 6/6] mm/z3fold: use release_z3fold_page_locked() to release locked z3fold page

[Miaohe Lin]

On 2021/6/20 8:26, Hillf Danton wrote:
> On Sat, 19 Jun 2021 17:31:51 +0800 Miaohe Lin wrote:
>> We should use release_z3fold_page_locked() to release z3fold page when it's
>> locked, although it looks harmless to use release_z3fold_page() now.
>>
>> Fixes: dcf5aedb24f8 ("z3fold: stricter locking and more careful reclaim")
>> Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
>> ---
>>  mm/z3fold.c | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/mm/z3fold.c b/mm/z3fold.c
>> index 196d886a3436..b3c0577b8095 100644
>> --- a/mm/z3fold.c
>> +++ b/mm/z3fold.c
>> @@ -1372,7 +1372,7 @@ static int z3fold_reclaim_page(struct z3fold_pool *pool, unsigned int retries)
>>  			if (zhdr->foreign_handles ||
>>  			    test_and_set_bit(PAGE_CLAIMED, &page->private)) {
>>  				if (kref_put(&zhdr->refcount,
>> -						release_z3fold_page))
>> +						release_z3fold_page_locked))
>>  					atomic64_dec(&pool->pages_nr);
> 
> LGTM. JFYI other issue in z3fold was reported [1] and if the fix proposed there
> makes any sense to you feel free to pick it up and ask Mike for his tests.
> 

Thank you for review and reply.

I browsed [1] several times but I failed to figure out what's the root cause. And I found
some bugs and possible race windows from previous code inspection. I think we can try fix
these first and see whether [1] is (hopefully) fixed. :)
Thanks again.

> [1] https://lore.kernel.org/linux-mm/20210316061351.1649-1-hdanton@sina.com/
> .
>

Re: [PATCH 0/6] Cleanup and fixup for z3fold

[Vitaly Wool]

[-- Attachment #1: Type: text/plain, Size: 966 bytes --]

On Sat, Jun 19, 2021 at 11:32 AM Miaohe Lin <linmiaohe@huawei.com> wrote:

> Hi all,
> This series contains cleanups to remove unused function, redefine macro
> to improve readability and so on. Also this fixes several bugs in z3fold,
> such as memory leak in z3fold_destroy_pool(). More details can be found
> in the respective changelogs. Thanks!
>
> Miaohe Lin (6):
>   mm/z3fold: define macro NCHUNKS as TOTAL_CHUNKS - ZHDR_CHUNKS
>   mm/z3fold: avoid possible underflow in z3fold_alloc()
>   mm/z3fold: remove magic number in z3fold_create_pool()
>   mm/z3fold: remove unused function handle_to_z3fold_header()
>   mm/z3fold: fix potential memory leak in z3fold_destroy_pool()
>   mm/z3fold: use release_z3fold_page_locked() to release locked z3fold
>     page
>
>  mm/z3fold.c | 37 ++++++++++++++-----------------------
>  1 file changed, 14 insertions(+), 23 deletions(-)
>
> Thanks for looking into this!

Reviewed-by: Vitaly Wool <vitaly.wool@konsulko.com>

[-- Attachment #2: Type: text/html, Size: 1431 bytes --]