All of lore.kernel.org
 help / color / mirror / Atom feed
From: Chengming Zhou <chengming.zhou@linux.dev>
To: Dan Carpenter <dan.carpenter@linaro.org>
Cc: linux-mm@kvack.org
Subject: Re: [bug report] mm/zswap: use only one pool in zswap
Date: Thu, 20 Jun 2024 17:12:03 +0800	[thread overview]
Message-ID: <98c42887-c742-4fe7-9bf5-e0eb4d715c75@linux.dev> (raw)
In-Reply-To: <ca310c94-0c34-437d-811a-9c32bf8bdf6e@moroto.mountain>

On 2024/6/20 16:52, Dan Carpenter wrote:
> Hello Chengming Zhou,
> 
> Commit 6193f190fe0a ("mm/zswap: use only one pool in zswap") from Jun
> 17, 2024 (linux-next), leads to the following Smatch static checker
> warning:

Thanks much for your report and explanation!

I will fix in the next version.

> 
> 	mm/zswap.c:306 zswap_pool_create()
> 	error: potential null dereference 'pool->zpool'.  (zpool_create_pool returns null)
> 
> mm/zswap.c
>      244 static struct zswap_pool *zswap_pool_create(char *type, char *compressor)
>      245 {
>      246         struct zswap_pool *pool;
>      247         char name[38]; /* 'zswap' + 32 char (max) num + \0 */
>      248         gfp_t gfp = __GFP_NORETRY | __GFP_NOWARN | __GFP_KSWAPD_RECLAIM;
>      249         int ret;
>      250
>      251         if (!zswap_has_pool) {
>      252                 /* if either are unset, pool initialization failed, and we
>      253                  * need both params to be set correctly before trying to
>      254                  * create a pool.
>      255                  */
>      256                 if (!strcmp(type, ZSWAP_PARAM_UNSET))
>      257                         return NULL;
>      258                 if (!strcmp(compressor, ZSWAP_PARAM_UNSET))
>      259                         return NULL;
>      260         }
>      261
>      262         pool = kzalloc(sizeof(*pool), GFP_KERNEL);
>      263         if (!pool)
>      264                 return NULL;
>      265
>      266         /* unique name for each pool specifically required by zsmalloc */
>      267         snprintf(name, 38, "zswap%x", atomic_inc_return(&zswap_pools_count));
>      268         pool->zpool = zpool_create_pool(type, name, gfp);
>      269         if (!pool->zpool) {
>      270                 pr_err("%s zpool not available\n", type);
>      271                 goto error;
> 
> pool->zpool is NULL
> 
>      272         }
>      273         pr_debug("using %s zpool\n", zpool_get_type(pool->zpool));
>      274
>      275         strscpy(pool->tfm_name, compressor, sizeof(pool->tfm_name));
>      276
>      277         pool->acomp_ctx = alloc_percpu(*pool->acomp_ctx);
>      278         if (!pool->acomp_ctx) {
>      279                 pr_err("percpu alloc failed\n");
>      280                 goto error;
>      281         }
>      282
>      283         ret = cpuhp_state_add_instance(CPUHP_MM_ZSWP_POOL_PREPARE,
>      284                                        &pool->node);
>      285         if (ret)
>      286                 goto error;
>      287
>      288         /* being the current pool takes 1 ref; this func expects the
>      289          * caller to always add the new pool as the current pool
>      290          */
>      291         ret = percpu_ref_init(&pool->ref, __zswap_pool_empty,
>      292                               PERCPU_REF_ALLOW_REINIT, GFP_KERNEL);
>      293         if (ret)
>      294                 goto ref_fail;
>      295         INIT_LIST_HEAD(&pool->list);
>      296
>      297         zswap_pool_debug("created", pool);
>      298
>      299         return pool;
>      300
>      301 ref_fail:
>      302         cpuhp_state_remove_instance(CPUHP_MM_ZSWP_POOL_PREPARE, &pool->node);
>      303 error:
>      304         if (pool->acomp_ctx)
>      305                 free_percpu(pool->acomp_ctx);
> --> 306         zpool_destroy_pool(pool->zpool);
>                                     ^^^^^^^^^^^
> NULL dereference
> 
>      307         kfree(pool);
>      308         return NULL;
>      309 }
> 
> regards,
> dan carpenter


      reply	other threads:[~2024-06-20  9:12 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-20  8:52 [bug report] mm/zswap: use only one pool in zswap Dan Carpenter
2024-06-20  9:12 ` Chengming Zhou [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=98c42887-c742-4fe7-9bf5-e0eb4d715c75@linux.dev \
    --to=chengming.zhou@linux.dev \
    --cc=dan.carpenter@linaro.org \
    --cc=linux-mm@kvack.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.