[PATCHv3] zsmalloc: allow only one active pool compaction context

[PATCHv3] zsmalloc: allow only one active pool compaction context

[Sergey Senozhatsky]

zsmalloc pool can be compacted concurrently by many contexts,
e.g.

 cc1 handle_mm_fault()
      do_anonymous_page()
       __alloc_pages_slowpath()
        try_to_free_pages()
         do_try_to_free_pages(
          lru_gen_shrink_node()
           shrink_slab()
            do_shrink_slab()
             zs_shrinker_scan()
              zs_compact()

Pool compaction is currently (basically) single-threaded as
it is performed under pool->lock. Having multiple compaction
threads results in unnecessary contention, as each thread
competes for pool->lock. This, in turn, affects all zsmalloc
operations such as zs_malloc(), zs_map_object(), zs_free(), etc.

Introduce the pool->compaction_in_progress atomic variable,
which ensures that only one compaction context can run at a
time. This reduces overall pool->lock contention in (corner)
cases when many contexts attempt to shrink zspool simultaneously.

Fixes: c0547d0b6a4b ("zsmalloc: consolidate zs_pool's migrate_lock and size_class's locks")
Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org>
---
 mm/zsmalloc.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
index cc81dfba05a0..44ddaf5d601e 100644
--- a/mm/zsmalloc.c
+++ b/mm/zsmalloc.c
@@ -264,6 +264,7 @@ struct zs_pool {
 	struct work_struct free_work;
 #endif
 	spinlock_t lock;
+	atomic_t compaction_in_progress;
 };
 
 struct zspage {
@@ -2274,6 +2275,15 @@ unsigned long zs_compact(struct zs_pool *pool)
 	struct size_class *class;
 	unsigned long pages_freed = 0;
 
+	/*
+	 * Pool compaction is performed under pool->lock so it is basically
+	 * single-threaded. Having more than one thread in __zs_compact()
+	 * will increase pool->lock contention, which will impact other
+	 * zsmalloc operations that need pool->lock.
+	 */
+	if (atomic_xchg(&pool->compaction_in_progress, 1))
+		return 0;
+
 	for (i = ZS_SIZE_CLASSES - 1; i >= 0; i--) {
 		class = pool->size_class[i];
 		if (class->index != i)
@@ -2281,6 +2291,7 @@ unsigned long zs_compact(struct zs_pool *pool)
 		pages_freed += __zs_compact(pool, class);
 	}
 	atomic_long_add(pages_freed, &pool->stats.pages_compacted);
+	atomic_set(&pool->compaction_in_progress, 0);
 
 	return pages_freed;
 }
@@ -2388,6 +2399,7 @@ struct zs_pool *zs_create_pool(const char *name)
 
 	init_deferred_free(pool);
 	spin_lock_init(&pool->lock);
+	atomic_set(&pool->compaction_in_progress, 0);
 
 	pool->name = kstrdup(name, GFP_KERNEL);
 	if (!pool->name)
-- 
2.40.0.634.g4ca3ef3211-goog

Re: [PATCHv3] zsmalloc: allow only one active pool compaction context

[Yosry Ahmed]

On Tue, Apr 18, 2023 at 12:46 AM Sergey Senozhatsky
<senozhatsky@chromium.org> wrote:
>
> zsmalloc pool can be compacted concurrently by many contexts,
> e.g.
>
>  cc1 handle_mm_fault()
>       do_anonymous_page()
>        __alloc_pages_slowpath()
>         try_to_free_pages()
>          do_try_to_free_pages(
>           lru_gen_shrink_node()
>            shrink_slab()
>             do_shrink_slab()
>              zs_shrinker_scan()
>               zs_compact()
>
> Pool compaction is currently (basically) single-threaded as
> it is performed under pool->lock. Having multiple compaction
> threads results in unnecessary contention, as each thread
> competes for pool->lock. This, in turn, affects all zsmalloc
> operations such as zs_malloc(), zs_map_object(), zs_free(), etc.
>
> Introduce the pool->compaction_in_progress atomic variable,
> which ensures that only one compaction context can run at a
> time. This reduces overall pool->lock contention in (corner)
> cases when many contexts attempt to shrink zspool simultaneously.
>
> Fixes: c0547d0b6a4b ("zsmalloc: consolidate zs_pool's migrate_lock and size_class's locks")
> Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org>


Reviewed-by: Yosry Ahmed <yosryahmed@google.com>

> ---
>  mm/zsmalloc.c | 12 ++++++++++++
>  1 file changed, 12 insertions(+)
>
> diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
> index cc81dfba05a0..44ddaf5d601e 100644
> --- a/mm/zsmalloc.c
> +++ b/mm/zsmalloc.c
> @@ -264,6 +264,7 @@ struct zs_pool {
>         struct work_struct free_work;
>  #endif
>         spinlock_t lock;
> +       atomic_t compaction_in_progress;
>  };
>
>  struct zspage {
> @@ -2274,6 +2275,15 @@ unsigned long zs_compact(struct zs_pool *pool)
>         struct size_class *class;
>         unsigned long pages_freed = 0;
>
> +       /*
> +        * Pool compaction is performed under pool->lock so it is basically
> +        * single-threaded. Having more than one thread in __zs_compact()
> +        * will increase pool->lock contention, which will impact other
> +        * zsmalloc operations that need pool->lock.
> +        */
> +       if (atomic_xchg(&pool->compaction_in_progress, 1))
> +               return 0;
> +
>         for (i = ZS_SIZE_CLASSES - 1; i >= 0; i--) {
>                 class = pool->size_class[i];
>                 if (class->index != i)
> @@ -2281,6 +2291,7 @@ unsigned long zs_compact(struct zs_pool *pool)
>                 pages_freed += __zs_compact(pool, class);
>         }
>         atomic_long_add(pages_freed, &pool->stats.pages_compacted);
> +       atomic_set(&pool->compaction_in_progress, 0);
>
>         return pages_freed;
>  }
> @@ -2388,6 +2399,7 @@ struct zs_pool *zs_create_pool(const char *name)
>
>         init_deferred_free(pool);
>         spin_lock_init(&pool->lock);
> +       atomic_set(&pool->compaction_in_progress, 0);
>
>         pool->name = kstrdup(name, GFP_KERNEL);
>         if (!pool->name)
> --
> 2.40.0.634.g4ca3ef3211-goog
>