搞网络不知道dpdk。。。不合适。。。html
搞dpdk不知道rte_mbuf。。。不合适。。。数组
因此,搞搞搞。。。网络
上源码!!!数据结构
//关于dpdk rte_mbuf数据结构的学习 /* define a set of marker types that can be used to refer to set points in the * mbuf */ /* 定义一组可用于引用 mbuf 中的设置点的标记类型*/ __extension__ typedef void *MARKER[0]; /**< generic marker for a point in a structure */ __extension__ typedef uint8_t MARKER8[0]; /**< generic marker with 1B alignment */ __extension__ typedef uint64_t MARKER64[0]; /**< marker that allows us to overwrite 8 bytes * with a single assignment */ /** * The generic rte_mbuf, containing a packet mbuf. */ struct rte_mbuf { MARKER cacheline0; /* 柔性数组,标记开头 */ void *buf_addr; /**< Virtual address of segment buffer. */ /** * Physical address of segment buffer. * Force alignment to 8-bytes, so as to ensure we have the exact * same mbuf cacheline0 layout for 32-bit and 64-bit. This makes * working on vector drivers easier. */ RTE_STD_C11 union { rte_iova_t buf_iova; rte_iova_t buf_physaddr; /**< deprecated */ } __rte_aligned(sizeof(rte_iova_t)); /* next 8 bytes are initialised on RX descriptor rearm */ MARKER64 rearm_data; uint16_t data_off; /** * Reference counter. Its size should at least equal to the size * of port field (16 bits), to support zero-copy broadcast. * It should only be accessed using the following functions: * rte_mbuf_refcnt_update(), rte_mbuf_refcnt_read(), and * rte_mbuf_refcnt_set(). The functionality of these functions (atomic, * or non-atomic) is controlled by the CONFIG_RTE_MBUF_REFCNT_ATOMIC * config option. */ RTE_STD_C11 union { rte_atomic16_t refcnt_atomic; /**< Atomically accessed refcnt */ uint16_t refcnt; /**< Non-atomically accessed refcnt */ }; uint16_t nb_segs; /**< Number of segments. */ /** Input port (16 bits to support more than 256 virtual ports). */ uint16_t port; uint64_t ol_flags; /**< Offload features. */ /* remaining bytes are set on RX when pulling packet from descriptor */ MARKER rx_descriptor_fields1; /* * The packet type, which is the combination of outer/inner L2, L3, L4 * and tunnel types. The packet_type is about data really present in the * mbuf. Example: if vlan stripping is enabled, a received vlan packet * would have RTE_PTYPE_L2_ETHER and not RTE_PTYPE_L2_VLAN because the * vlan is stripped from the data. */ RTE_STD_C11 union { uint32_t packet_type; /**< L2/L3/L4 and tunnel information. */ struct { uint32_t l2_type:4; /**< (Outer) L2 type. */ uint32_t l3_type:4; /**< (Outer) L3 type. */ uint32_t l4_type:4; /**< (Outer) L4 type. */ uint32_t tun_type:4; /**< Tunnel type. */ RTE_STD_C11 union { uint8_t inner_esp_next_proto; /**< ESP next protocol type, valid if * RTE_PTYPE_TUNNEL_ESP tunnel type is set * on both Tx and Rx. */ __extension__ struct { uint8_t inner_l2_type:4; /**< Inner L2 type. */ uint8_t inner_l3_type:4; /**< Inner L3 type. */ }; }; uint32_t inner_l4_type:4; /**< Inner L4 type. */ }; }; uint32_t pkt_len; /**< Total pkt len: sum of all segments. */ uint16_t data_len; /**< Amount of data in segment buffer. */ /** VLAN TCI (CPU order), valid if PKT_RX_VLAN_STRIPPED is set. */ uint16_t vlan_tci; union { uint32_t rss; /**< RSS hash result if RSS enabled */ struct { RTE_STD_C11 union { struct { uint16_t hash; uint16_t id; }; uint32_t lo; /**< Second 4 flexible bytes */ }; uint32_t hi; /**< First 4 flexible bytes or FD ID, dependent on PKT_RX_FDIR_* flag in ol_flags. */ } fdir; /**< Filter identifier if FDIR enabled */ struct { uint32_t lo; uint32_t hi; } sched; /**< Hierarchical scheduler */ uint32_t usr; /**< User defined tags. See rte_distributor_process() */ } hash; /**< hash information */ /** Outer VLAN TCI (CPU order), valid if PKT_RX_QINQ_STRIPPED is set. */ uint16_t vlan_tci_outer; uint16_t buf_len; /**< Length of segment buffer. */ /** Valid if PKT_RX_TIMESTAMP is set. The unit and time reference * are not normalized but are always the same for a given port. */ uint64_t timestamp; /* second cache line - fields only used in slow path or on TX */ MARKER cacheline1 __rte_cache_min_aligned; RTE_STD_C11 union { void *userdata; /**< Can be used for external metadata */ uint64_t udata64; /**< Allow 8-byte userdata on 32-bit */ }; struct rte_mempool *pool; /**< Pool from which mbuf was allocated. */ struct rte_mbuf *next; /**< Next segment of scattered packet. */ /* fields to support TX offloads */ RTE_STD_C11 union { uint64_t tx_offload; /**< combined for easy fetch */ __extension__ struct { uint64_t l2_len:7; /**< L2 (MAC) Header Length for non-tunneling pkt. * Outer_L4_len + ... + Inner_L2_len for tunneling pkt. */ uint64_t l3_len:9; /**< L3 (IP) Header Length. */ uint64_t l4_len:8; /**< L4 (TCP/UDP) Header Length. */ uint64_t tso_segsz:16; /**< TCP TSO segment size */ /* fields for TX offloading of tunnels */ uint64_t outer_l3_len:9; /**< Outer L3 (IP) Hdr Length. */ uint64_t outer_l2_len:7; /**< Outer L2 (MAC) Hdr Length. */ /* uint64_t unused:8; */ }; }; /** Size of the application private data. In case of an indirect * mbuf, it stores the direct mbuf private data size. */ uint16_t priv_size; /** Timesync flags for use with IEEE1588. */ uint16_t timesync; /** Sequence number. See also rte_reorder_insert(). */ uint32_t seqn; }
好家伙,果真mbuf,大名鼎鼎。下面分别对每一个字段进行学习解释。app
下面按照出现顺序对每一个字段进行解释。ide
MARKER cacheline0; typedef void *MARKER[0]; /**< generic marker for a point in a structure */
查看typedef,发现这是一个柔性数组。长度为0,因此这里在编译时是不占用内存滴。只是一个标记喽。MARKER嘛。函数
void *buf_addr; /**< Virtual address of segment buffer. */
有图就容易解释了,一些指针、成员或函数结果的内容在下表中列出,mbuf指针简写为m性能
m | 首部,即mbuf结构体 |
m->buf_addr | headroom起始地址 |
m->data_off | data起始地址相对于buf_addr的偏移 |
m->buf_len | mbuf和priv以后内存的长度,包含headroom |
m->pkt_len | 整个mbuf链的data总长度 |
m->data_len | 实际data的长度 |
m->buf_addr+m->data_off | 实际data的起始地址 |
rte_pktmbuf_mtod(m) | 同上 |
rte_pktmbuf_data_len(m) | 同m->data_len |
rte_pktmbuf_pkt_len | 同m->pkt_len |
rte_pktmbuf_data_room_size | 同m->buf_len |
rte_pktmbuf_headroom | headroom长度 |
rte_pktmbuf_tailroom | 尾部剩余空间长度 |
综合图片解释以及上述表格的备注。这里buf_addr就是rte_mbuf结构体尾部,headroom起始地址。学习
/** * Physical address of segment buffer. * Force alignment to 8-bytes, so as to ensure we have the exact * same mbuf cacheline0 layout for 32-bit and 64-bit. This makes * working on vector drivers easier. */ RTE_STD_C11 union { rte_iova_t buf_iova; rte_iova_t buf_physaddr; /**< deprecated */ } __rte_aligned(sizeof(rte_iova_t));
段缓冲区的物理地址。 强制8字节对齐,保证在32位和64位有相同的cacheline0。这块暂时无需关注。fetch
/* next 8 bytes are initialised on RX descriptor rearm */ MARKER64 rearm_data;
接下来的 8 个字节在 RX 描述符重装时初始化 。
uint16_t data_off;
data起始地址相对于buf_addr的偏移。要获取data的位置,m->buf_addr + m->data_off ,就是对应的data的实际指针。通常中间间隔是一个headroom的大小。
/** * Reference counter. Its size should at least equal to the size * of port field (16 bits), to support zero-copy broadcast. * It should only be accessed using the following functions: * rte_mbuf_refcnt_update(), rte_mbuf_refcnt_read(), and * rte_mbuf_refcnt_set(). The functionality of these functions (atomic, * or non-atomic) is controlled by the CONFIG_RTE_MBUF_REFCNT_ATOMIC * config option. */ RTE_STD_C11 union { rte_atomic16_t refcnt_atomic; /**< Atomically accessed refcnt */ uint16_t refcnt; /**< Non-atomically accessed refcnt */ };
引用计数。这里用union实现了原子访问和非原子访问2种。计数的规格至少等于端口字段的大小16bits,(用来支持零拷贝广播?不明白)。
uint16_t nb_segs; /**< Number of segments. */
分片数。
/** Input port (16 bits to support more than 256 virtual ports). */ uint16_t port;
入接口id号。
uint64_t ol_flags; /**< Offload features. */
offload特性标记。
offload特性,主要是指将本来在协议栈中进行的IP分片、TCP分段、重组、checksum校验等操做,转移到网卡硬件中进行,下降系统CPU的消耗,提升处理性能。
/* remaining bytes are set on RX when pulling packet from descriptor */ MARKER rx_descriptor_fields1;
从描述符中提取数据包时,剩余字节设置在 RX 上。标记使用,MARKER。。。
/* * The packet type, which is the combination of outer/inner L2, L3, L4 * and tunnel types. The packet_type is about data really present in the * mbuf. Example: if vlan stripping is enabled, a received vlan packet * would have RTE_PTYPE_L2_ETHER and not RTE_PTYPE_L2_VLAN because the * vlan is stripped from the data. */ /* 数据包类型,它是外部/内部 L二、L三、L4 和隧道类型的组合。 * packet_type 是关于 mbuf 中真正存在的数据。 * 若是启用了 vlan 剥离,则接收到的 vlan 数据包将具备 RTE_PTYPE_L2_ETHER * 而不是 RTE_PTYPE_L2_VLAN,由于 vlan 已从数据中剥离。 */ RTE_STD_C11 union { uint32_t packet_type; /**< L2/L3/L4 and tunnel information. */ struct { uint32_t l2_type:4; /**< (Outer) L2 type. */ uint32_t l3_type:4; /**< (Outer) L3 type. */ uint32_t l4_type:4; /**< (Outer) L4 type. */ uint32_t tun_type:4; /**< Tunnel type. */ RTE_STD_C11 union { uint8_t inner_esp_next_proto; /**< ESP next protocol type, valid if * RTE_PTYPE_TUNNEL_ESP tunnel type is set * on both Tx and Rx. */ __extension__ struct { uint8_t inner_l2_type:4; /**< Inner L2 type. */ uint8_t inner_l3_type:4; /**< Inner L3 type. */ }; }; uint32_t inner_l4_type:4; /**< Inner L4 type. */ }; };
此数据结构比较清晰,无需多余解释。有一个疑问,这里的inner && outer具体是什么呢?
uint32_t pkt_len; /**< Total pkt len: sum of all segments. */ uint16_t data_len; /**< Amount of data in segment buffer. */
pkt_len,包括全部分片的长度。
data_len,当前的数据长度。若是没有分片,pkt_len与data_len数值应该是相同的。也就是pkt_len >= data_len.
/** VLAN TCI (CPU order), valid if PKT_RX_VLAN_STRIPPED is set. */ uint16_t vlan_tci;
只有开启了PKT_RX_VLAN_STRIPPED标记,此字段才是有效的。vlan时使用,学习vlan时,须要关注此字段。
union { uint32_t rss; /**< RSS hash result if RSS enabled */ struct { RTE_STD_C11 union { struct { uint16_t hash; uint16_t id; }; uint32_t lo; /**< Second 4 flexible bytes */ }; uint32_t hi; /**< First 4 flexible bytes or FD ID, dependent on PKT_RX_FDIR_* flag in ol_flags. */ } fdir; /**< Filter identifier if FDIR enabled */ struct { uint32_t lo; uint32_t hi; } sched; /**< Hierarchical scheduler */ uint32_t usr; /**< User defined tags. See rte_distributor_process() */ } hash; /**< hash information */
哈希数据。这里是一个union。当RSS开启时,对应rss字段是哈希结果。学习RSS时,关注一下。
/** Outer VLAN TCI (CPU order), valid if PKT_RX_QINQ_STRIPPED is set. */ uint16_t vlan_tci_outer;
只有开启了QINQ剥离时,此字段有效。外部vlan相关。
uint16_t buf_len; /**< Length of segment buffer. */
mbuf和priv以后内存的长度,包含headroom。
/** Valid if PKT_RX_TIMESTAMP is set. The unit and time reference * are not normalized but are always the same for a given port. */ uint64_t timestamp;
时间戳。PKT_RX_TIMESAMP开启时,此字段有效。单位和时间参考未标准化,但对于给定端口始终相同。
/* second cache line - fields only used in slow path or on TX */ MARKER cacheline1 __rte_cache_min_aligned;
第二个cacheline,这部份内容仅用在慢路或者发包流程中。
RTE_STD_C11 union { void *userdata; /**< Can be used for external metadata */ uint64_t udata64; /**< Allow 8-byte userdata on 32-bit */ }; //#define RTE_STD_C11 __extension__
__extension__字段用于消除编译告警。
这里是一个union,
在userdata指针总能够用来存放额外的元数据。
udata64,能够存放8字节的用户数据。
struct rte_mempool *pool; /**< Pool from which mbuf was allocated. */
标识本mbuf是从哪一个rte_mempool池子中申请到的。也就是该mbuf是哪一个rte_mempool池子的。
struct rte_mbuf *next; /**< Next segment of scattered packet. */
在分片报文中,标记下一个报文的位置。
/* fields to support TX offloads */ /* 用于支持发包硬件卸载的字段 */ RTE_STD_C11 union { uint64_t tx_offload; /**< combined for easy fetch */ /* tx_offload 组合起来,方便取用 */ __extension__ struct { uint64_t l2_len:7; /**< L2 (MAC) Header Length for non-tunneling pkt. * Outer_L4_len + ... + Inner_L2_len for tunneling pkt. */ uint64_t l3_len:9; /**< L3 (IP) Header Length. */ uint64_t l4_len:8; /**< L4 (TCP/UDP) Header Length. */ uint64_t tso_segsz:16; /**< TCP TSO segment size */ /* TSO(TCP Segment Offload)是一种利用网卡的少许处理能力, 下降CPU发送数据包负载的技术,须要网卡硬件及驱动的支持。 */ /* fields for TX offloading of tunnels */ uint64_t outer_l3_len:9; /**< Outer L3 (IP) Hdr Length. */ uint64_t outer_l2_len:7; /**< Outer L2 (MAC) Hdr Length. */ /* uint64_t unused:8; */ }; };
支持硬件发包卸载的字段内容。内部为一个union。其中tx_offload字段是为了容易获取搞出来的。
/** Size of the application private data. In case of an indirect * mbuf, it stores the direct mbuf private data size. */ uint16_t priv_size;
应用程序私有数据的大小。
在indirect mbuf 的状况下,它存储direct mbuf 私有数据大小。 关于direct mbuf与indirect mbuf的区别,参考连接
10. Mbuf Library — Data Plane Development Kit 21.08.0-rc1 documentation (dpdk.org)
/** Timesync flags for use with IEEE1588. */ /* IEEE1588 协议,又称 PTP( precise time protocol,精确时间协议), * 能够达到亚微秒级别时间同步精度,于 2002 年发布 version 1, * 2008 年发布 version 2。 */ uint16_t timesync;
时间同步。参考IEEE1588。
/** Sequence number. See also rte_reorder_insert(). */ uint32_t seqn;
序列号。这个是哪里用到呢?
rte_mbuf的数据结构学习完毕。有一些遗留的问题,后续来完善。