| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <cevxzukpt4363kdlb5ofre5tquuvm6jyphstp6nlxjxqibj4wx@2yg4xwlcxwg5>
Date: Wed, 21 Jan 2026 23:58:45 +0000
From: Yosry Ahmed <yosry.ahmed@...ux.dev>
To: Sergey Senozhatsky <senozhatsky@...omium.org>
Cc: Nhat Pham <nphamcs@...il.com>,
Andrew Morton <akpm@...ux-foundation.org>, Minchan Kim <minchan@...nel.org>,
Johannes Weiner <hannes@...xchg.org>, Brian Geffon <bgeffon@...gle.com>, linux-kernel@...r.kernel.org,
linux-mm@...ck.org
Subject: Re: [RFC PATCH] zsmalloc: make common caches global
On Wed, Jan 21, 2026 at 12:41:39PM +0900, Sergey Senozhatsky wrote:
> On (26/01/19 13:44), Nhat Pham wrote:
> > On Thu, Jan 15, 2026 at 9:53 PM Sergey Senozhatsky
> > <senozhatsky@...omium.org> wrote:
> > >
> > > On (26/01/16 13:48), Sergey Senozhatsky wrote:
> > > > Currently, zsmalloc creates kmem_cache of handles and zspages
> > > > for each pool, which may be suboptimal from the memory usage
> > > > point of view (extra internal fragmentation per pool). Systems
> > > > that create multiple zsmalloc pools may benefit from shared
> > > > common zsmalloc caches.
> > >
> > > This is step 1.
> > >
> > > Step 2 is to look into possibility of sharing zsmalloc pools.
> > > E.g. if there are N zram devices in the system, do we really need
> > > N zsmalloc pools? Can we just share a single pool between them?
> >
> > Ditto for zswap (although here, we almost always only have a single zswap pool).
>
> COMPLETELY UNTESTED (current linux-next doesn't boot for me, hitting
> an "Oops: stack guard page: 0000" early during boot).
>
> So I'm thinking of something like below. Basically have a Kconfig
> option to turn zsmalloc into a singleton pool mode, transparently
> for zsmalloc users.
Why do we need a config option? Is the main concern with a single pool
lock contention? If yes, we can probably measure it by spawning many
zram devices and stressing them at the same time.
>
> ---
> mm/Kconfig | 11 ++++++++
> mm/zsmalloc.c | 73 ++++++++++++++++++++++++++++++++++++++++++---------
> 2 files changed, 72 insertions(+), 12 deletions(-)
>
> diff --git a/mm/Kconfig b/mm/Kconfig
> index 4fc1a171dffa..ff6855e74c3d 100644
> --- a/mm/Kconfig
> +++ b/mm/Kconfig
> @@ -132,6 +132,17 @@ menu "Zsmalloc allocator options"
>
> comment "Zsmalloc is a common backend allocator for zswap & zram"
>
> +config ZSMALLOC_SINGLETON_POOL
> + bool "Use a singleton zsmalloc pool"
> + default n
> + help
> + This option enables the use of a single global zsmalloc pool
> + instance for all users of zsmalloc (e.g., zswap, zram). This
> + reduces memory overhead and fragmentation by sharing size class
> + configurations and memory between different users.
> +
> + If unsure, say N.
> +
> config ZSMALLOC_STAT
> bool "Export zsmalloc statistics"
> select DEBUG_FS
> diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
> index 8df45aa1b5c8..acd14b001342 100644
> --- a/mm/zsmalloc.c
> +++ b/mm/zsmalloc.c
> @@ -224,6 +224,10 @@ struct zs_pool {
> atomic_t compaction_in_progress;
> };
>
> +#ifdef CONFIG_ZSMALLOC_SINGLETON_POOL
> +static struct zs_pool *zs_singleton_pool;
> +#endif
> +
> static inline void zpdesc_set_first(struct zpdesc *zpdesc)
> {
> SetPagePrivate(zpdesc_page(zpdesc));
> @@ -2051,17 +2055,7 @@ static int calculate_zspage_chain_size(int class_size)
> return chain_size;
> }
>
> -/**
> - * zs_create_pool - Creates an allocation pool to work from.
> - * @name: pool name to be created
> - *
> - * This function must be called before anything when using
> - * the zsmalloc allocator.
> - *
> - * On success, a pointer to the newly created pool is returned,
> - * otherwise NULL.
> - */
> -struct zs_pool *zs_create_pool(const char *name)
> +static struct zs_pool *__zs_create_pool(const char *name)
> {
> int i;
> struct zs_pool *pool;
> @@ -2170,9 +2164,29 @@ struct zs_pool *zs_create_pool(const char *name)
> zs_destroy_pool(pool);
> return NULL;
> }
> +
> +/**
> + * zs_create_pool - Creates an allocation pool to work from.
> + * @name: pool name to be created
> + *
> + * This function must be called before anything when using
> + * the zsmalloc allocator.
> + *
> + * On success, a pointer to the newly created pool is returned,
> + * otherwise NULL.
> + */
> +struct zs_pool *zs_create_pool(const char *name)
> +{
> +#ifdef CONFIG_ZSMALLOC_SINGLETON_POOL
> + return zs_singleton_pool;
> +#else
> + return __zs_create_pool(name);
> +#endif
> +
> +}
> EXPORT_SYMBOL_GPL(zs_create_pool);
>
> -void zs_destroy_pool(struct zs_pool *pool)
> +static void __zs_destroy_pool(struct zs_pool *pool)
> {
> int i;
>
> @@ -2203,8 +2217,35 @@ void zs_destroy_pool(struct zs_pool *pool)
> kfree(pool->name);
> kfree(pool);
> }
> +
> +void zs_destroy_pool(struct zs_pool *pool __maybe_unused)
> +{
> +#ifndef CONFIG_ZSMALLOC_SINGLETON_POOL
> + __zs_destroy_pool(pool);
> +#endif
> +}
> EXPORT_SYMBOL_GPL(zs_destroy_pool);
>
> +static void zs_destroy_singleton_pool(void)
> +{
> +#ifdef CONFIG_ZSMALLOC_SINGLETON_POOL
> + if (zs_singleton_pool) {
> + __zs_destroy_pool(zs_singleton_pool);
> + zs_singleton_pool = NULL;
> + }
> +#endif
> +}
> +
> +static int zs_create_singleton_pool(void)
> +{
> +#ifdef CONFIG_ZSMALLOC_SINGLETON_POOL
> + zs_singleton_pool = __zs_create_pool("zsmalloc");
> + if (!zs_singleton_pool)
> + return -ENOMEM;
> +#endif
> + return 0;
> +}
> +
> static void zs_destroy_caches(void)
> {
> kmem_cache_destroy(handle_cachep);
> @@ -2235,9 +2276,16 @@ static int __init zs_init(void)
> if (rc)
> return rc;
>
> + rc = zs_create_singleton_pool();
> + if (rc) {
> + zs_destroy_caches();
> + return rc;
> + }
> +
> #ifdef CONFIG_COMPACTION
> rc = set_movable_ops(&zsmalloc_mops, PGTY_zsmalloc);
> if (rc) {
> + zs_destroy_singleton_pool();
> zs_destroy_caches();
> return rc;
> }
> @@ -2252,6 +2300,7 @@ static void __exit zs_exit(void)
> set_movable_ops(NULL, PGTY_zsmalloc);
> #endif
> zs_stat_exit();
> + zs_destroy_singleton_pool();
> zs_destroy_caches();
> }
>
> --
> 2.52.0.457.g6b5491de43-goog
>
Powered by blists - more mailing lists