DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Ananyev, Konstantin" <konstantin.ananyev@intel.com>
To: Olivier Matz <olivier.matz@6wind.com>, "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH v5 08/12] mbuf: fix clone support when application uses private mbuf data
Date: Tue, 21 Apr 2015 15:01:28 +0000	[thread overview]
Message-ID: <2601191342CEEE43887BDE71AB9772582142095C@irsmsx105.ger.corp.intel.com> (raw)
In-Reply-To: <1429610122-30943-9-git-send-email-olivier.matz@6wind.com>

Hi Olivier,

> -----Original Message-----
> From: Olivier Matz [mailto:olivier.matz@6wind.com]
> Sent: Tuesday, April 21, 2015 10:55 AM
> To: dev@dpdk.org
> Cc: Ananyev, Konstantin; zoltan.kiss@linaro.org; Richardson, Bruce; nhorman@tuxdriver.com; olivier.matz@6wind.com
> Subject: [PATCH v5 08/12] mbuf: fix clone support when application uses private mbuf data
> 
> Add a new private_size field in mbuf structure that should
> be initialized at mbuf pool creation. This field contains the
> size of the application private data in mbufs.
> 
> Introduce new static inline functions rte_mbuf_from_indirect()
> and rte_mbuf_to_baddr() to replace the existing macros, which
> take the private size in account when attaching and detaching
> mbufs.
> 
> Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
> Reviewed-by: Zoltan Kiss <zoltan.kiss@linaro.org>
> ---
>  examples/vhost/main.c      |  4 ++--
>  lib/librte_mbuf/rte_mbuf.c |  2 +-
>  lib/librte_mbuf/rte_mbuf.h | 59 +++++++++++++++++++++++++++++++---------------
>  3 files changed, 43 insertions(+), 22 deletions(-)
> 
> diff --git a/examples/vhost/main.c b/examples/vhost/main.c
> index 22d6a4b..195d82f 100644
> --- a/examples/vhost/main.c
> +++ b/examples/vhost/main.c
> @@ -138,7 +138,7 @@
>  /* Number of descriptors per cacheline. */
>  #define DESC_PER_CACHELINE (RTE_CACHE_LINE_SIZE / sizeof(struct vring_desc))
> 
> -#define MBUF_EXT_MEM(mb)   (RTE_MBUF_FROM_BADDR((mb)->buf_addr) != (mb))
> +#define MBUF_EXT_MEM(mb)   (rte_mbuf_from_indirect(mb) != (mb))
> 
>  /* mask of enabled ports */
>  static uint32_t enabled_port_mask = 0;
> @@ -1549,7 +1549,7 @@ attach_rxmbuf_zcp(struct virtio_net *dev)
>  static inline void pktmbuf_detach_zcp(struct rte_mbuf *m)
>  {
>  	const struct rte_mempool *mp = m->pool;
> -	void *buf = RTE_MBUF_TO_BADDR(m);
> +	void *buf = rte_mbuf_to_baddr(m);
>  	uint32_t buf_ofs;
>  	uint32_t buf_len = mp->elt_size - sizeof(*m);
>  	m->buf_physaddr = rte_mempool_virt2phy(mp, m) + sizeof(*m);
> diff --git a/lib/librte_mbuf/rte_mbuf.c b/lib/librte_mbuf/rte_mbuf.c
> index b013607..784ae8b 100644
> --- a/lib/librte_mbuf/rte_mbuf.c
> +++ b/lib/librte_mbuf/rte_mbuf.c
> @@ -129,7 +129,7 @@ rte_pktmbuf_init(struct rte_mempool *mp,
> 
>  	memset(m, 0, mp->elt_size);
> 
> -	/* start of buffer is just after mbuf structure */
> +	/* start of buffer is after mbuf structure and priv data */
>  	m->buf_addr = (char *)m + mbuf_size;
>  	m->buf_physaddr = rte_mempool_virt2phy(mp, m) + mbuf_size;
>  	m->buf_len = (uint16_t)buf_len;
> diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
> index 42db8e3..5c01c5b 100644
> --- a/lib/librte_mbuf/rte_mbuf.h
> +++ b/lib/librte_mbuf/rte_mbuf.h
> @@ -320,16 +320,40 @@ struct rte_mbuf {
>  	};
>  } __rte_cache_aligned;
> 
> +static inline uint16_t rte_pktmbuf_priv_size(struct rte_mempool *mp);
> +
>  /**
> - * Given the buf_addr returns the pointer to corresponding mbuf.
> + * Return the mbuf owning the data buffer address of an indirect mbuf.
> + *
> + * @param mi
> + *   The pointer to the indirect mbuf.
> + * @return
> + *   The address of the direct mbuf corresponding to buffer_addr.
>   */
> -#define RTE_MBUF_FROM_BADDR(ba)     (((struct rte_mbuf *)(ba)) - 1)
> +static inline struct rte_mbuf *
> +rte_mbuf_from_indirect(struct rte_mbuf *mi)
> +{
> +       struct rte_mbuf *md;
> +       md = (struct rte_mbuf *)((char *)mi->buf_addr -
> +	       sizeof(*mi) - rte_pktmbuf_priv_size(mi->pool));
> +       return md;
> +}


Really puzzled, why do you remove prv_size from mbuf?
Now if let say we attach mbuf (mi) with priv_size == X to mbuf (md) with priv_size == Y,
then rte_mbuf_from_indirect(mi) would return: mi->buf_addr - 128 - X, while it should be: mi->buf_addr - 128 - Y.

Another thing, rte_mbuf_from_indirect() and rte_mbuf_to_baddr() would probably become a bit slower -
as another level of indirection is added.

Konstantin

> 
>  /**
> - * Given the pointer to mbuf returns an address where it's  buf_addr
> - * should point to.
> + * Return the buffer address embedded in the given mbuf.
> + *
> + * @param md
> + *   The pointer to the mbuf.
> + * @return
> + *   The address of the data buffer owned by the mbuf.
>   */
> -#define RTE_MBUF_TO_BADDR(mb)       (((struct rte_mbuf *)(mb)) + 1)
> +static inline char *
> +rte_mbuf_to_baddr(struct rte_mbuf *md)
> +{
> +       char *buffer_addr;
> +       buffer_addr = (char *)md + sizeof(*md) + rte_pktmbuf_priv_size(md->pool);
> +       return buffer_addr;
> +}
> 
>  /**
>   * Returns TRUE if given mbuf is indirect, or FALSE otherwise.
> @@ -771,6 +795,7 @@ static inline struct rte_mbuf *rte_pktmbuf_alloc(struct rte_mempool *mp)
> 
>  /**
>   * Attach packet mbuf to another packet mbuf.
> + *
>   * After attachment we refer the mbuf we attached as 'indirect',
>   * while mbuf we attached to as 'direct'.
>   * Right now, not supported:
> @@ -784,7 +809,6 @@ static inline struct rte_mbuf *rte_pktmbuf_alloc(struct rte_mempool *mp)
>   * @param md
>   *   The direct packet mbuf.
>   */
> -
>  static inline void rte_pktmbuf_attach(struct rte_mbuf *mi, struct rte_mbuf *md)
>  {
>  	RTE_MBUF_ASSERT(RTE_MBUF_DIRECT(md) &&
> @@ -815,7 +839,8 @@ static inline void rte_pktmbuf_attach(struct rte_mbuf *mi, struct rte_mbuf *md)
>  }
> 
>  /**
> - * Detach an indirect packet mbuf -
> + * Detach an indirect packet mbuf.
> + *
>   *  - restore original mbuf address and length values.
>   *  - reset pktmbuf data and data_len to their default values.
>   *  All other fields of the given packet mbuf will be left intact.
> @@ -823,22 +848,18 @@ static inline void rte_pktmbuf_attach(struct rte_mbuf *mi, struct rte_mbuf *md)
>   * @param m
>   *   The indirect attached packet mbuf.
>   */
> -
>  static inline void rte_pktmbuf_detach(struct rte_mbuf *m)
>  {
> -	const struct rte_mempool *mp = m->pool;
> -	void *buf = RTE_MBUF_TO_BADDR(m);
> -	uint32_t buf_len = mp->elt_size - sizeof(*m);
> -	m->buf_physaddr = rte_mempool_virt2phy(mp, m) + sizeof (*m);
> +	struct rte_mempool *mp = m->pool;
> +	uint32_t mbuf_size, buf_len;
> 
> -	m->buf_addr = buf;
> +	mbuf_size = sizeof(struct rte_mbuf) + rte_pktmbuf_priv_size(mp);
> +	buf_len = rte_pktmbuf_data_room_size(mp);
> +	m->buf_addr = rte_mbuf_to_baddr(m);
> +	m->buf_physaddr = rte_mempool_virt2phy(mp, m) + mbuf_size;
>  	m->buf_len = (uint16_t)buf_len;
> -
> -	m->data_off = (RTE_PKTMBUF_HEADROOM <= m->buf_len) ?
> -			RTE_PKTMBUF_HEADROOM : m->buf_len;
> -
> +	m->data_off = RTE_MIN(RTE_PKTMBUF_HEADROOM, (uint16_t)m->buf_len);
>  	m->data_len = 0;
> -
>  	m->ol_flags = 0;
>  }
> 
> @@ -867,7 +888,7 @@ __rte_pktmbuf_prefree_seg(struct rte_mbuf *m)
>  		 *  - free attached mbuf segment
>  		 */
>  		if (RTE_MBUF_INDIRECT(m)) {
> -			struct rte_mbuf *md = RTE_MBUF_FROM_BADDR(m->buf_addr);
> +			struct rte_mbuf *md = rte_mbuf_from_indirect(m);
>  			rte_pktmbuf_detach(m);
>  			if (rte_mbuf_refcnt_update(md, -1) == 0)
>  				__rte_mbuf_raw_free(md);
> --
> 2.1.4

  reply	other threads:[~2015-04-21 15:01 UTC|newest]

Thread overview: 101+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-03-25 17:00 [dpdk-dev] [PATCH 0/5] mbuf: enhancements of mbuf clones Olivier Matz
2015-03-25 17:00 ` [dpdk-dev] [PATCH 1/5] mbuf: fix clone support when application uses private mbuf data Olivier Matz
2015-03-26 13:35   ` Bruce Richardson
2015-03-26 15:30     ` Olivier MATZ
2015-03-25 17:00 ` [dpdk-dev] [PATCH 2/5] mbuf: allow to clone an indirect mbuf Olivier Matz
2015-03-25 17:00 ` [dpdk-dev] [PATCH 3/5] test/mbuf: rename mc variable in m Olivier Matz
2015-03-25 17:00 ` [dpdk-dev] [PATCH 4/5] test/mbuf: enhance mbuf refcnt test Olivier Matz
2015-03-25 17:00 ` [dpdk-dev] [PATCH 5/5] test/mbuf: verify that cloning a clone works properly Olivier Matz
2015-03-26  8:48   ` Olivier MATZ
2015-03-26 15:59 ` [dpdk-dev] [PATCH v2 0/5] mbuf: enhancements of mbuf clones Olivier Matz
2015-03-26 15:59   ` [dpdk-dev] [PATCH v2 1/5] mbuf: fix clone support when application uses private mbuf data Olivier Matz
2015-03-26 17:13     ` Zoltan Kiss
2015-03-27  0:24     ` Ananyev, Konstantin
2015-03-27  9:07       ` Olivier MATZ
2015-03-27 13:56         ` Olivier MATZ
2015-03-27 14:25           ` Ananyev, Konstantin
2015-03-27 15:17             ` Olivier MATZ
2015-03-27 18:11               ` Zoltan Kiss
2015-03-28 21:19                 ` Olivier MATZ
2015-03-30 12:34               ` Ananyev, Konstantin
2015-03-30 19:55                 ` Olivier MATZ
2015-03-30 23:17                   ` Ananyev, Konstantin
2015-03-31 19:01                     ` Olivier MATZ
2015-04-01 13:48                       ` Ananyev, Konstantin
2015-04-01 15:18                         ` Olivier MATZ
2015-03-26 15:59   ` [dpdk-dev] [PATCH v2 2/5] mbuf: allow to clone an indirect mbuf Olivier Matz
2015-03-26 15:59   ` [dpdk-dev] [PATCH v2 3/5] test/mbuf: rename mc variable in m Olivier Matz
2015-03-26 15:59   ` [dpdk-dev] [PATCH v2 4/5] test/mbuf: enhance mbuf refcnt test Olivier Matz
2015-03-26 15:59   ` [dpdk-dev] [PATCH v2 5/5] test/mbuf: verify that cloning a clone works properly Olivier Matz
2015-03-31 19:22   ` [dpdk-dev] [PATCH v3 0/5] mbuf: enhancements of mbuf clones Olivier Matz
2015-03-31 19:23     ` [dpdk-dev] [PATCH v3 1/5] mbuf: fix clone support when application uses private mbuf data Olivier Matz
2015-04-02 14:32       ` Zoltan Kiss
2015-04-02 17:21       ` Ananyev, Konstantin
2015-04-06 21:49         ` Olivier MATZ
2015-04-07 12:40           ` Ananyev, Konstantin
2015-04-07 15:45             ` Olivier MATZ
2015-04-07 17:17               ` Ananyev, Konstantin
2015-04-08  9:44                 ` Olivier MATZ
2015-04-08 13:45                   ` Ananyev, Konstantin
2015-04-09 13:06                     ` Olivier MATZ
2015-04-20 15:41       ` [dpdk-dev] [PATCH v4 00/12] mbuf: enhancements of mbuf clones Olivier Matz
2015-04-20 15:41         ` [dpdk-dev] [PATCH v4 01/12] mbuf: fix mbuf data room size calculation rte_pktmbuf_pool_init Olivier Matz
2015-04-20 15:41         ` [dpdk-dev] [PATCH v4 02/12] examples: always initialize mbuf_pool private area Olivier Matz
2015-04-20 15:41         ` [dpdk-dev] [PATCH v4 03/12] mbuf: add accessors to get data room size and private size Olivier Matz
2015-04-20 15:41         ` [dpdk-dev] [PATCH v4 04/12] mbuf: fix rte_pktmbuf_init when mbuf private size is not zero Olivier Matz
2015-04-20 15:41         ` [dpdk-dev] [PATCH v4 05/12] testpmd: use standard functions to initialize mbufs and mbuf pool Olivier Matz
2015-04-20 15:41         ` [dpdk-dev] [PATCH v4 06/12] mbuf: introduce a new helper to create a " Olivier Matz
2015-04-20 15:41         ` [dpdk-dev] [PATCH v4 07/12] apps: use rte_pktmbuf_pool_create to create mbuf pools Olivier Matz
2015-04-20 15:41         ` [dpdk-dev] [PATCH v4 08/12] mbuf: fix clone support when application uses private mbuf data Olivier Matz
2015-04-20 15:41         ` [dpdk-dev] [PATCH v4 09/12] mbuf: allow to clone an indirect mbuf Olivier Matz
2015-04-20 15:41         ` [dpdk-dev] [PATCH v4 10/12] test/mbuf: rename mc variable in m Olivier Matz
2015-04-20 15:41         ` [dpdk-dev] [PATCH v4 11/12] test/mbuf: enhance mbuf refcnt test Olivier Matz
2015-04-20 15:41         ` [dpdk-dev] [PATCH v4 12/12] test/mbuf: verify that cloning a clone works properly Olivier Matz
2015-04-20 16:53         ` [dpdk-dev] [PATCH v4 00/12] mbuf: enhancements of mbuf clones Neil Horman
2015-04-20 17:07           ` Olivier MATZ
2015-04-20 17:21             ` Neil Horman
2015-04-20 18:24               ` Olivier MATZ
2015-04-21  9:55         ` [dpdk-dev] [PATCH v5 " Olivier Matz
2015-04-21  9:55           ` [dpdk-dev] [PATCH v5 01/12] mbuf: fix mbuf data room size calculation rte_pktmbuf_pool_init Olivier Matz
2015-04-21  9:55           ` [dpdk-dev] [PATCH v5 02/12] examples: always initialize mbuf_pool private area Olivier Matz
2015-04-21  9:55           ` [dpdk-dev] [PATCH v5 03/12] mbuf: add accessors to get data room size and private size Olivier Matz
2015-04-21  9:55           ` [dpdk-dev] [PATCH v5 04/12] mbuf: fix rte_pktmbuf_init when mbuf private size is not zero Olivier Matz
2015-04-21 15:07             ` Ananyev, Konstantin
2015-04-21  9:55           ` [dpdk-dev] [PATCH v5 05/12] testpmd: use standard functions to initialize mbufs and mbuf pool Olivier Matz
2015-04-21  9:55           ` [dpdk-dev] [PATCH v5 06/12] mbuf: introduce a new helper to create a " Olivier Matz
2015-04-21  9:55           ` [dpdk-dev] [PATCH v5 07/12] apps: use rte_pktmbuf_pool_create to create mbuf pools Olivier Matz
2015-04-21  9:55           ` [dpdk-dev] [PATCH v5 08/12] mbuf: fix clone support when application uses private mbuf data Olivier Matz
2015-04-21 15:01             ` Ananyev, Konstantin [this message]
2015-04-21 15:26               ` Olivier MATZ
2015-04-21  9:55           ` [dpdk-dev] [PATCH v5 09/12] mbuf: allow to clone an indirect mbuf Olivier Matz
2015-04-21  9:55           ` [dpdk-dev] [PATCH v5 10/12] test/mbuf: rename mc variable in m Olivier Matz
2015-04-21  9:55           ` [dpdk-dev] [PATCH v5 11/12] test/mbuf: enhance mbuf refcnt test Olivier Matz
2015-04-21  9:55           ` [dpdk-dev] [PATCH v5 12/12] test/mbuf: verify that cloning a clone works properly Olivier Matz
2015-04-21 11:50           ` [dpdk-dev] [PATCH v5 00/12] mbuf: enhancements of mbuf clones Neil Horman
2015-04-22  9:57           ` [dpdk-dev] [PATCH v6 00/13] " Olivier Matz
2015-04-22  9:57             ` [dpdk-dev] [PATCH v6 01/13] mbuf: fix mbuf data room size calculation rte_pktmbuf_pool_init Olivier Matz
2015-04-22  9:57             ` [dpdk-dev] [PATCH v6 02/13] examples: always initialize mbuf_pool private area Olivier Matz
2015-04-22  9:57             ` [dpdk-dev] [PATCH v6 03/13] mbuf: add accessors to get data room size and private size Olivier Matz
2015-04-28  9:15               ` Thomas Monjalon
2015-04-22  9:57             ` [dpdk-dev] [PATCH v6 04/13] mbuf: fix rte_pktmbuf_init when mbuf private size is not zero Olivier Matz
2015-04-22  9:57             ` [dpdk-dev] [PATCH v6 05/13] testpmd: use standard functions to initialize mbufs and mbuf pool Olivier Matz
2015-04-22  9:57             ` [dpdk-dev] [PATCH v6 06/13] mbuf: introduce a new helper to create a " Olivier Matz
2015-04-22  9:57             ` [dpdk-dev] [PATCH v6 07/13] apps: use rte_pktmbuf_pool_create to create mbuf pools Olivier Matz
2015-04-22  9:57             ` [dpdk-dev] [PATCH v6 08/13] mbuf: fix clone support when application uses private mbuf data Olivier Matz
2015-04-22  9:57             ` [dpdk-dev] [PATCH v6 09/13] mbuf: allow to clone an indirect mbuf Olivier Matz
2015-04-22  9:57             ` [dpdk-dev] [PATCH v6 10/13] test/mbuf: rename mc variable in m Olivier Matz
2015-04-22  9:57             ` [dpdk-dev] [PATCH v6 11/13] test/mbuf: enhance mbuf refcnt test Olivier Matz
2015-04-22  9:57             ` [dpdk-dev] [PATCH v6 12/13] test/mbuf: verify that cloning a clone works properly Olivier Matz
2015-04-22  9:57             ` [dpdk-dev] [PATCH v6 13/13] test/mbuf: add a test case for clone with different priv size Olivier Matz
2015-04-22 11:59             ` [dpdk-dev] [PATCH v6 00/13] mbuf: enhancements of mbuf clones Ananyev, Konstantin
2015-04-24 10:38               ` Zoltan Kiss
2015-04-27 17:38                 ` Thomas Monjalon
2015-04-28 11:15                   ` Zoltan Kiss
2015-05-07  1:57                 ` Xu, HuilongX
2015-05-07  7:32                   ` Olivier MATZ
2015-05-07  9:39                     ` Ananyev, Konstantin
2015-04-28  9:50               ` Thomas Monjalon
2015-03-31 19:23     ` [dpdk-dev] [PATCH v3 2/5] mbuf: allow to clone an indirect mbuf Olivier Matz
2015-03-31 19:23     ` [dpdk-dev] [PATCH v3 3/5] test/mbuf: rename mc variable in m Olivier Matz
2015-03-31 19:23     ` [dpdk-dev] [PATCH v3 4/5] test/mbuf: enhance mbuf refcnt test Olivier Matz
2015-03-31 19:23     ` [dpdk-dev] [PATCH v3 5/5] test/mbuf: verify that cloning a clone works properly Olivier Matz

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=2601191342CEEE43887BDE71AB9772582142095C@irsmsx105.ger.corp.intel.com \
    --to=konstantin.ananyev@intel.com \
    --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).