DPDK patches and discussions
 help / color / mirror / Atom feed
From: changfengnan <changfengnan@bytedance.com>
To: olivier.matz@6wind.com, andrew.rybchenko@oktetlabs.ru, dev@dpdk.org
Subject: Re: [PATCH] mempool: fix rte_mempool_avail_count may segment fault when used in multiprocess
Date: Mon, 14 Nov 2022 19:23:28 +0800	[thread overview]
Message-ID: <CAPFOzZsyTzsif=o67XRkJnzMkfNwJPGGaCDAzZBxzynmFmiyaQ@mail.gmail.com> (raw)
In-Reply-To: <20221114071439.38902-1-changfengnan@bytedance.com>

It seems that this modification ignores some certain cases, for example calling
rte_mempool_create_empty directly, maybe add a new flag bit to indicate
when to put tailq entry into rte_mempool_tailq list is a better way ?

Looking forward to any suggestions.

Thanks.
Fengnan Chang.

changfengnan <changfengnan@bytedance.com> 于2022年11月14日周一 15:14写道:
>
> rte_mempool_create put tailq entry into rte_mempool_tailq list before
> populate, and pool_data set when populate. So in multi process, if
> process A create mempool, and process B can get mempool through
> rte_mempool_lookup before pool_data set, if B call rte_mempool_lookup,
> it will cause segment fault.
> Fix this by put tailq entry into rte_mempool_tailq after populate.
>
> Signed-off-by: changfengnan <changfengnan@bytedance.com>
> ---
>  lib/mempool/rte_mempool.c | 40 +++++++++++++++++++--------------------
>  1 file changed, 19 insertions(+), 21 deletions(-)
>
> diff --git a/lib/mempool/rte_mempool.c b/lib/mempool/rte_mempool.c
> index 4c78071a34..b23d6138ff 100644
> --- a/lib/mempool/rte_mempool.c
> +++ b/lib/mempool/rte_mempool.c
> @@ -798,9 +798,7 @@ rte_mempool_create_empty(const char *name, unsigned n, unsigned elt_size,
>         int socket_id, unsigned flags)
>  {
>         char mz_name[RTE_MEMZONE_NAMESIZE];
> -       struct rte_mempool_list *mempool_list;
>         struct rte_mempool *mp = NULL;
> -       struct rte_tailq_entry *te = NULL;
>         const struct rte_memzone *mz = NULL;
>         size_t mempool_size;
>         unsigned int mz_flags = RTE_MEMZONE_1GB|RTE_MEMZONE_SIZE_HINT_ONLY;
> @@ -820,8 +818,6 @@ rte_mempool_create_empty(const char *name, unsigned n, unsigned elt_size,
>                           RTE_CACHE_LINE_MASK) != 0);
>  #endif
>
> -       mempool_list = RTE_TAILQ_CAST(rte_mempool_tailq.head, rte_mempool_list);
> -
>         /* asked for zero items */
>         if (n == 0) {
>                 rte_errno = EINVAL;
> @@ -866,14 +862,6 @@ rte_mempool_create_empty(const char *name, unsigned n, unsigned elt_size,
>         private_data_size = (private_data_size +
>                              RTE_MEMPOOL_ALIGN_MASK) & (~RTE_MEMPOOL_ALIGN_MASK);
>
> -
> -       /* try to allocate tailq entry */
> -       te = rte_zmalloc("MEMPOOL_TAILQ_ENTRY", sizeof(*te), 0);
> -       if (te == NULL) {
> -               RTE_LOG(ERR, MEMPOOL, "Cannot allocate tailq entry!\n");
> -               goto exit_unlock;
> -       }
> -
>         mempool_size = RTE_MEMPOOL_HEADER_SIZE(mp, cache_size);
>         mempool_size += private_data_size;
>         mempool_size = RTE_ALIGN_CEIL(mempool_size, RTE_MEMPOOL_ALIGN);
> @@ -908,7 +896,6 @@ rte_mempool_create_empty(const char *name, unsigned n, unsigned elt_size,
>         mp->private_data_size = private_data_size;
>         STAILQ_INIT(&mp->elt_list);
>         STAILQ_INIT(&mp->mem_list);
> -
>         /*
>          * local_cache pointer is set even if cache_size is zero.
>          * The local_cache points to just past the elt_pa[] array.
> @@ -922,12 +909,6 @@ rte_mempool_create_empty(const char *name, unsigned n, unsigned elt_size,
>                         mempool_cache_init(&mp->local_cache[lcore_id],
>                                            cache_size);
>         }
> -
> -       te->data = mp;
> -
> -       rte_mcfg_tailq_write_lock();
> -       TAILQ_INSERT_TAIL(mempool_list, te, next);
> -       rte_mcfg_tailq_write_unlock();
>         rte_mcfg_mempool_write_unlock();
>
>         rte_mempool_trace_create_empty(name, n, elt_size, cache_size,
> @@ -936,7 +917,6 @@ rte_mempool_create_empty(const char *name, unsigned n, unsigned elt_size,
>
>  exit_unlock:
>         rte_mcfg_mempool_write_unlock();
> -       rte_free(te);
>         rte_mempool_free(mp);
>         return NULL;
>  }
> @@ -951,11 +931,22 @@ rte_mempool_create(const char *name, unsigned n, unsigned elt_size,
>  {
>         int ret;
>         struct rte_mempool *mp;
> +       struct rte_mempool_list *mempool_list;
> +       struct rte_tailq_entry *te = NULL;
> +
> +       /* try to allocate tailq entry */
> +       te = rte_zmalloc("MEMPOOL_TAILQ_ENTRY", sizeof(*te), 0);
> +       if (te == NULL) {
> +               RTE_LOG(ERR, MEMPOOL, "Cannot allocate tailq entry!\n");
> +               return NULL;
> +       }
>
>         mp = rte_mempool_create_empty(name, n, elt_size, cache_size,
>                 private_data_size, socket_id, flags);
> -       if (mp == NULL)
> +       if (mp == NULL) {
> +               rte_free(te);
>                 return NULL;
> +       }
>
>         /*
>          * Since we have 4 combinations of the SP/SC/MP/MC examine the flags to
> @@ -984,12 +975,19 @@ rte_mempool_create(const char *name, unsigned n, unsigned elt_size,
>         if (obj_init)
>                 rte_mempool_obj_iter(mp, obj_init, obj_init_arg);
>
> +       te->data = mp;
> +       mempool_list = RTE_TAILQ_CAST(rte_mempool_tailq.head, rte_mempool_list);
> +       rte_mcfg_tailq_write_lock();
> +       TAILQ_INSERT_TAIL(mempool_list, te, next);
> +       rte_mcfg_tailq_write_unlock();
> +
>         rte_mempool_trace_create(name, n, elt_size, cache_size,
>                 private_data_size, mp_init, mp_init_arg, obj_init,
>                 obj_init_arg, flags, mp);
>         return mp;
>
>   fail:
> +       rte_free(te);
>         rte_mempool_free(mp);
>         return NULL;
>  }
> --
> 2.37.0 (Apple Git-136)
>

  parent reply	other threads:[~2022-11-14 22:13 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-14  7:14 changfengnan
2022-11-14  9:55 ` Morten Brørup
2022-11-14 11:23 ` changfengnan [this message]
2022-11-14 20:43 ` David Marchand
2022-11-15  1:51   ` [External] " Fengnan Chang
2022-11-15  7:23     ` Morten Brørup
2022-11-15  7:47     ` David Marchand
2022-11-15  8:29       ` Olivier Matz
2022-11-15 11:30         ` Fengnan Chang

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='CAPFOzZsyTzsif=o67XRkJnzMkfNwJPGGaCDAzZBxzynmFmiyaQ@mail.gmail.com' \
    --to=changfengnan@bytedance.com \
    --cc=andrew.rybchenko@oktetlabs.ru \
    --cc=dev@dpdk.org \
    --cc=olivier.matz@6wind.com \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).