/*
* Copyright (c) 2016-2022 Apple Inc. All rights reserved.
*
* @APPLE_OSREFERENCE_LICENSE_HEADER_START@
*
* This file contains Original Code and/or Modifications of Original Code
* as defined in and that are subject to the Apple Public Source License
* Version 2.0 (the 'License'). You may not use this file except in
* compliance with the License. The rights granted to you under the License
* may not be used to create, or enable the creation or redistribution of,
* unlawful or unlicensed copies of an Apple operating system, or to
* circumvent, violate, or enable the circumvention or violation of, any
* terms of an Apple operating system software license agreement.
*
* Please obtain a copy of the License at
* http://www.opensource.apple.com/apsl/ and read it before using this file.
*
* The Original Code and all software distributed under the License are
* distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
* EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
* INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
* Please see the License for the specific language governing rights and
* limitations under the License.
*
* @APPLE_OSREFERENCE_LICENSE_HEADER_END@
*/
#ifndef _SKYWALK_OS_PACKET_PRIVATE_H_
#define _SKYWALK_OS_PACKET_PRIVATE_H_
#if defined(PRIVATE) || defined(BSD_KERNEL_PRIVATE)
#include <skywalk/os_packet.h>
#include <skywalk/os_nexus_private.h>
#include <skywalk/os_channel_private.h>
#include <libkern/OSByteOrder.h>
#include <netinet/in.h>
#include <net/ethernet.h>
#if defined(BSD_KERNEL_PRIVATE)
/*
* Flow (currently for kernel, potentially for userland one day).
*
* XXX: When we expose this to userland, we need to be make sure to NOT
* expose kernel pointer/address values embedded within.
*
* Values in flow_{l2,l3,l4} are stored in network byte order. Pointers
* are defined using mach_vm_address_t because it's stable across user
* and kernel, and therefore keeps the structure size the same.
*
* Because this structure might be initialized on a per-packet allocation
* basis, it as well as some of its member sub-subtructures are allocated
* on a 16-bytes address boundary to allow 128-bit operations on platforms
* that support them.
*
* XXX: when adding new fields, try to leverage __pad ones first.
*
* TODO: we should consider embedding a flow_key structure here and
* use that to store the tuples. That way we can leverage that for
* flow lookups without having to copy things back-and-forth.
*/
struct __flow {
union {
/*
* The following is always zeroed out on each alloc.
*/
struct __flow_init {
/*
* Layer 3
*/
struct __flow_l3 {
union {
struct __flow_l3_ipv4_addrs {
struct in_addr _src;
struct in_addr _dst;
} _l3_ipv4;
struct __flow_l3_ipv6_addrs {
struct in6_addr _src;
struct in6_addr _dst;
} _l3_ipv6;
};
uint8_t _l3_ip_ver;
uint8_t _l3_proto;
uint8_t _l3_hlen;
unsigned _l3_is_frag : 1;
unsigned _l3_is_first_frag : 1;
unsigned _l3_reserved_flags : 6;
uint32_t _l3_frag_id;
mach_vm_address_t _l3_ptr;
} __l3;
/*
* AQM
*/
struct __flow_classq {
uint32_t _fcq_hash; /* classq-specific hash */
uint32_t _fcq_flags; /* classq-specific flags */
} __classq;
/*
* Misc.
*/
uint32_t __ulen; /* user data length */
uint8_t __ulp_encap; /* e.g. IPPROTO_QUIC */
uint8_t __pad[3];
uint64_t __pad64[2];
/*
* Flow Source.
*/
struct __flow_source {
union {
/* source identifier */
uint64_t _fsrc_id_64[2];
uint32_t _fsrc_id_32[4];
uuid_t _fsrc_id;
} __attribute__((aligned(sizeof(uint64_t))));
flowadv_idx_t _fsrc_fidx; /* flow adv. index */
uint8_t _fsrc_type; /* FLOWSRC_* mbuf.h */
uint8_t _fsrc_pad[3];
} __source;
/*
* Policy.
*/
struct __flow_policy {
uint32_t _fpc_id; /* policy id of pkt sender */
uint32_t _fpc_skip_id; /* skip policy id of pkt sender */
union {
/* process identifier */
uint64_t _fpc_euuid_64[2];
uint32_t _fpc_euuid_32[4];
uuid_t _fpc_euuid;
} __attribute__((aligned(sizeof(uint64_t))));
} __policy;
} flow_init;
uint64_t flow_init_data[16];
} __attribute((aligned(16)));
#define flow_l3 flow_init.__l3
#define flow_classq flow_init.__classq
#define flow_ulen flow_init.__ulen
#define flow_ulp_encap flow_init.__ulp_encap
#define flow_source flow_init.__source
#define flow_policy flow_init.__policy
#define flow_ipv4_addrs flow_l3._l3_ipv4
#define flow_ipv4_src flow_l3._l3_ipv4._src
#define flow_ipv4_dst flow_l3._l3_ipv4._dst
#define flow_ipv6_addrs flow_l3._l3_ipv6
#define flow_ipv6_src flow_l3._l3_ipv6._src
#define flow_ipv6_dst flow_l3._l3_ipv6._dst
#define flow_ip_ver flow_l3._l3_ip_ver
#define flow_ip_proto flow_l3._l3_proto
#define flow_ip_hlen flow_l3._l3_hlen
#define flow_ip_hdr flow_l3._l3_ptr
#define flow_ip_frag_id flow_l3._l3_frag_id
#define flow_ip_is_frag flow_l3._l3_is_frag
#define flow_ip_is_first_frag flow_l3._l3_is_first_frag
#define flow_classq_hash flow_classq._fcq_hash
#define flow_classq_flags flow_classq._fcq_flags
#define flow_src_token flow_source._fsrc_id_32[0]
#define flow_src_id flow_source._fsrc_id
#define flow_src_fidx flow_source._fsrc_fidx
#define flow_src_type flow_source._fsrc_type
#define flow_policy_id flow_policy._fpc_id
#define flow_skip_policy_id flow_policy._fpc_skip_id
#define flow_policy_euuid flow_policy._fpc_euuid
/*
* Layer 4.
*/
union {
struct __flow_l4 {
union {
struct __flow_l4_tcp {
in_port_t _src;
in_port_t _dst;
uint32_t _seq;
uint32_t _ack;
union {
struct {
#if BYTE_ORDER == LITTLE_ENDIAN
uint8_t _tcp_res:4;
uint8_t _off:4;
#else /* BYTE_ORDER == BIG_ENDIAN */
uint8_t _off:4;
uint8_t _tcp_res:4;
#endif /* BYTE_ORDER == BIG_ENDIAN */
uint8_t _flags;
uint16_t _win;
};
uint32_t _ofw;
};
} _l4_tcp;
struct __flow_l4_udp {
in_port_t _src;
in_port_t _dst;
uint32_t _ls;
} _l4_udp;
struct __flow_l4_esp {
uint32_t _spi;
} _l4_esp;
};
uint8_t _l4_hlen;
uint8_t _l4_agg_fast;
uint8_t _l4_pad[6];
mach_vm_address_t _l4_ptr;
} flow_l4;
uint64_t flow_l4_data[4];
} __attribute((aligned(sizeof(uint64_t))));
#define flow_tcp flow_l4._l4_tcp
#define flow_tcp_src flow_l4._l4_tcp._src
#define flow_tcp_dst flow_l4._l4_tcp._dst
#define flow_tcp_seq flow_l4._l4_tcp._seq
#define flow_tcp_ack flow_l4._l4_tcp._ack
#define flow_tcp_off flow_l4._l4_tcp._off
#define flow_tcp_flags flow_l4._l4_tcp._flags
#define flow_tcp_win flow_l4._l4_tcp._win
#define flow_tcp_hlen flow_l4._l4_hlen
#define flow_tcp_hdr flow_l4._l4_ptr
#define flow_tcp_agg_fast flow_l4._l4_agg_fast
#define flow_udp flow_l4._l4_udp
#define flow_udp_src flow_l4._l4_udp._src
#define flow_udp_dst flow_l4._l4_udp._dst
#define flow_udp_hlen flow_l4._l4_hlen
#define flow_udp_hdr flow_l4._l4_ptr
#define flow_esp_spi flow_l4._l4_esp._spi
} __attribute((aligned(16)));
#endif /* BSD_KERNEL_PRIVATE */
/*
* Maximum size of L2, L3 & L4 headers combined.
*/
#define PKT_MAX_PROTO_HEADER_SIZE 256
/* based on 2KB buflet size */
#define BUFLETS_MIN 1 /* Ethernet MTU (default) */
#define BUFLETS_9K_JUMBO 5 /* 9000 bytes MTU */
#define BUFLETS_GSO 46 /* 64KB GSO, Ethernet MTU */
/*
* Common buflet structure shared by {__user,__kern}_buflet.
*/
struct __buflet {
union {
/* for skmem batch alloc/free */
uint64_t __buflet_next;
/* address of next buflet in chain */
const mach_vm_address_t __nbft_addr;
};
/* buffer data address */
const mach_vm_address_t __baddr;
/* index of buflet object in the owning buflet region */
const obj_idx_t __bft_idx;
/* buffer object index in buffer region */
const obj_idx_t __bidx;
/* object index in buflet region of next buflet(for buflet chaining) */
const obj_idx_t __nbft_idx;
const uint32_t __dlim; /* maximum length */
uint32_t __doff; /* offset of data in buflet */
uint32_t __dlen; /* length of data in buflet */
const uint16_t __flag;
#define BUFLET_FLAG_EXTERNAL 0x0001
#define BUFLET_FLAG_LARGE_BUF 0x0002 /* buflet holds large buffer */
} __attribute((packed));
/*
* A buflet represents the smallest buffer fragment representing
* part of the packet. The index refers to the position of the buflet
* in the pool, and the data length represents the actual payload
* size -- not the buflet size itself as it is fixed for all objects
* in the pool.
*/
struct __user_buflet {
/*
* Common area between user and kernel variants.
*/
struct __buflet buf_com;
#define buf_addr buf_com.__baddr
#define buf_nbft_addr buf_com.__nbft_addr
#define buf_idx buf_com.__bidx
#define buf_nbft_idx buf_com.__nbft_idx
#define buf_dlim buf_com.__dlim
#define buf_dlen buf_com.__dlen
#define buf_doff buf_com.__doff
#define buf_flag buf_com.__flag
#define buf_bft_idx_reg buf_com.__bft_idx
};
#define BUFLET_HAS_LARGE_BUF(_buf) \
(((_buf)->buf_flag & BUFLET_FLAG_LARGE_BUF) != 0)
#define BUF_BADDR(_buf, _addr) \
*__DECONST(mach_vm_address_t *, &(_buf)->buf_addr) = \
(mach_vm_address_t)(_addr)
#define BUF_BIDX(_buf, _idx) \
*__DECONST(obj_idx_t *, &(_buf)->buf_idx) = (obj_idx_t)(_idx)
#define BUF_NBFT_ADDR(_buf, _addr) \
*__DECONST(mach_vm_address_t *, &(_buf)->buf_nbft_addr) = \
(mach_vm_address_t)(_addr)
#define BUF_NBFT_IDX(_buf, _idx) \
*__DECONST(obj_idx_t *, &(_buf)->buf_nbft_idx) = (obj_idx_t)(_idx)
#define BUF_BFT_IDX_REG(_buf, _idx) \
*__DECONST(obj_idx_t *, &(_buf)->buf_bft_idx_reg) = (_idx)
#define UBUF_LINK(_pubft, _ubft) do { \
ASSERT((_ubft) != NULL); \
BUF_NBFT_ADDR(_pubft, _ubft); \
BUF_NBFT_IDX(_pubft, (_ubft)->buf_bft_idx_reg); \
} while (0)
#ifdef KERNEL
#define BUF_CTOR(_buf, _baddr, _bidx, _dlim, _dlen, _doff, _nbaddr, _nbidx, _bflag) do { \
_CASSERT(sizeof ((_buf)->buf_addr) == sizeof (mach_vm_address_t)); \
_CASSERT(sizeof ((_buf)->buf_idx) == sizeof (obj_idx_t)); \
_CASSERT(sizeof ((_buf)->buf_dlim) == sizeof (uint32_t)); \
BUF_BADDR(_buf, _baddr); \
BUF_NBFT_ADDR(_buf, _nbaddr); \
BUF_BIDX(_buf, _bidx); \
BUF_NBFT_IDX(_buf, _nbidx); \
*(uint32_t *)(uintptr_t)&(_buf)->buf_dlim = (_dlim); \
(_buf)->buf_dlen = (_dlen); \
(_buf)->buf_doff = (_doff); \
*(uint16_t *)(uintptr_t)&(_buf)->buf_flag = (_bflag); \
} while (0)
#define BUF_INIT(_buf, _dlen, _doff) do { \
(_buf)->buf_dlen = (_dlen); \
(_buf)->buf_doff = (_doff); \
} while (0)
#endif /* KERNEL */
#ifdef KERNEL
#define BUF_IN_RANGE(_buf) \
((_buf)->buf_addr >= (mach_vm_address_t)(_buf)->buf_objaddr && \
((uintptr_t)(_buf)->buf_addr + (_buf)->buf_dlim) <= \
((uintptr_t)(_buf)->buf_objaddr + (_buf)->buf_objlim) && \
((_buf)->buf_doff + (_buf)->buf_dlen) <= (_buf)->buf_dlim)
#else /* !KERNEL */
#define BUF_IN_RANGE(_buf) \
(((_buf)->buf_doff + (_buf)->buf_dlen) <= (_buf)->buf_dlim)
#endif /* !KERNEL */
/*
* Metadata preamble. This structure is placed at begining of each
* __{user,kern}_{quantum,packet} object. Each user metadata object has a
* unique red zone pattern, which is an XOR of the redzone cookie and
* offset of the metadata object in the object's region. Due to the use
* of tagged pointer, we need the structure size to be multiples of 16.
* See SK_PTR_TAG() definition for details.
*/
struct __metadata_preamble {
union {
uint64_t _mdp_next; /* for batch alloc/free (K) */
uint64_t mdp_redzone; /* red zone cookie (U) */
};
const obj_idx_t mdp_idx; /* index within region (UK) */
uint16_t mdp_type; /* nexus_meta_type_t (UK) */
uint16_t mdp_subtype; /* nexus_meta_subtype_t (UK) */
};
#define METADATA_PREAMBLE_SZ (sizeof (struct __metadata_preamble))
#define METADATA_PREAMBLE(_md) \
(__unsafe_forge_single(struct __metadata_preamble *, \
((mach_vm_address_t)(_md) - METADATA_PREAMBLE_SZ)))
#define METADATA_IDX(_md) \
(METADATA_PREAMBLE(_md)->mdp_idx)
#define METADATA_TYPE(_md) \
(METADATA_PREAMBLE(_md)->mdp_type)
#define METADATA_SUBTYPE(_md) \
(METADATA_PREAMBLE(_md)->mdp_subtype)
/*
* Common packet structure shared by {__user,__kern}_quantum.
*/
struct __quantum {
union {
uuid_t __uuid; /* flow UUID */
uint8_t __val8[16];
uint16_t __val16[8];
uint32_t __val32[4];
uint64_t __val64[2];
} __flow_id_u;
#define __q_flow_id __flow_id_u.__uuid
#define __q_flow_id_val8 __flow_id_u.__val8
#define __q_flow_id_val16 __flow_id_u.__val16
#define __q_flow_id_val32 __flow_id_u.__val32
#define __q_flow_id_val64 __flow_id_u.__val64
uint32_t __q_len;
/* QoS service class, see packet_svc_class_t */
uint32_t __q_svc_class; /* PKT_SC_* values */
/*
* See notes on _QUM_{INTERNALIZE,EXTERNALIZE}() regarding
* portion of this structure above __flags that gets copied.
* Adding more user-mutable fields after __flags would also
* require adjusting those macros as well.
*/
volatile uint16_t __q_flags; /* QUMF_* flags */
uint16_t __q_pad[3];
} __attribute((aligned(sizeof(uint64_t))));
/*
* Quantum.
*
* This structure is aligned for efficient copy and accesses.
* It is the user version of the __kernel_quantum structure.
*
* XXX: Do NOT store kernel pointer/address values here.
*/
struct __user_quantum {
/*
* Common area between user and kernel variants.
*/
struct __quantum qum_com;
#define qum_flow_id qum_com.__q_flow_id
#define qum_flow_id_val8 qum_com.__q_flow_id_val8
#define qum_flow_id_val16 qum_com.__q_flow_id_val16
#define qum_flow_id_val32 qum_com.__q_flow_id_val32
#define qum_flow_id_val64 qum_com.__q_flow_id_val64
#define qum_len qum_com.__q_len
#define qum_qflags qum_com.__q_flags
#define qum_svc_class qum_com.__q_svc_class
/*
* Userland specific.
*/
struct __user_buflet qum_buf[1]; /* 1 buflet */
/*
* use count for packet.
*/
uint16_t qum_usecnt;
} __attribute((aligned(sizeof(uint64_t))));
/*
* Valid values for (16-bit) qum_qflags.
*/
#define QUM_F_FINALIZED 0x0001 /* has been finalized */
#define QUM_F_DROPPED 0x0002 /* has been dropped */
#define QUM_F_FLOW_CLASSIFIED 0x0010 /* flow has been classified */
#ifdef KERNEL
#define QUM_F_INTERNALIZED 0x1000 /* has been internalized */
#define QUM_F_KERNEL_ONLY 0x8000 /* kernel only; no user counterpart */
/* invariant flags we want to keep */
#define QUM_F_SAVE_MASK (QUM_F_KERNEL_ONLY)
/* kernel-only flags that's never externalized */
#define QUM_F_KERNEL_FLAGS (QUM_F_INTERNALIZED|QUM_F_KERNEL_ONLY)
#endif /* KERNEL */
#ifdef KERNEL
#define _KQUM_CTOR(_kqum, _flags, _len, _baddr, _bidx, _dlim, _qidx) do { \
(_kqum)->qum_flow_id_val64[0] = 0; \
(_kqum)->qum_flow_id_val64[1] = 0; \
(_kqum)->qum_qflags = (_flags); \
(_kqum)->qum_len = (_len); \
_CASSERT(sizeof(METADATA_IDX(_kqum)) == sizeof(obj_idx_t)); \
*(obj_idx_t *)(uintptr_t)&METADATA_IDX(_kqum) = (_qidx); \
BUF_CTOR(&(_kqum)->qum_buf[0], (_baddr), (_bidx), (_dlim), 0, 0, 0, \
OBJ_IDX_NONE, 0); \
} while (0)
#define _KQUM_INIT(_kqum, _flags, _len, _qidx) do { \
(_kqum)->qum_flow_id_val64[0] = 0; \
(_kqum)->qum_flow_id_val64[1] = 0; \
(_kqum)->qum_qflags = (_flags); \
(_kqum)->qum_len = (_len); \
BUF_INIT(&(_kqum)->qum_buf[0], 0, 0); \
} while (0)
#endif /* KERNEL */
/*
* Common packet structure shared by {__user,__kern}_packet.
*/
struct __packet_com {
/* Link layer (offset relevant to first buflet) */
uint16_t __link_flags; /* PKT_LINKF_* flags */
/*
* Headroom/protocol header length
*
* Since the security model of Skywalk nexus is that we doesn't trust
* packets either from above (userspace) or below (driver/firmware),
* the only metadata field that nexus makes use of from external is the
* headroom. Based on headroom, the flowswitch starts demux routine on
* l2 header, if any. The l2_len is stored in this step. Then the flow
* extraction (l3+l4 flow) begins parsing from (headroom + l2_len).
*
* __headroom is the empty buffer space before any packet data,
* it is also the equivalent to the first header offset.
*
* __l2_len is l2 (link layer) protocol header length, if any.
*/
uint8_t __headroom;
uint8_t __l2_len;
/*
* Checksum offload.
*
* Partial checksum does not require any header parsing and is
* therefore simpler to implement both in software and hardware.
*
* On transmit, PKT_CSUMF_PARTIAL indicates that a partial one's
* complement checksum to be computed on the span starting from
* pkt_csum_tx_start_off to the end of the packet, and have the
* resulted checksum value written at the location specified by
* pkt_csum_tx_stuff_off.
*
* The PKT_CSUMF_ZERO_INVERT flag is used on transmit to indicate
* that the value 0xffff (negative 0 in one's complement) must be
* substituted for the value of 0. This is set for UDP packets,
* since otherwise the receiver may not validate the checksum
* (UDP/IPv4), or drop the packet altogether (UDP/IPv6).
*
* On receive, PKT_CSUMF_PARTIAL indicates that a partial one's
* complement checksum has been computed on the span beginning at
* pkt_csum_rx_start_off to the end of the packet, and that the
* computed value is now stored in pkt_csum_rx_value.
*
* All offsets are relative to the base of the first buflet.
*/
uint32_t __csum_flags; /* PKT_CSUMF_* flags */
union {
struct {
uint16_t __csum_start_off; /* start offset */
uint16_t __csum_value; /* checksum value */
} __csum_rx;
struct {
uint16_t __csum_start_off; /* start offset */
uint16_t __csum_stuff_off; /* stuff offset */
} __csum_tx;
uint32_t __csum_data;
};
/* Compression generation count */
uint32_t __comp_gencnt;
/*
* Trace ID for each sampled packet.
* Non-zero ID indicates that the packet is being actively traced.
*/
packet_trace_id_t __trace_id;
/* Aggregation type */
uint8_t __aggr_type; /* PKT_AGGR_* values */
uint8_t __seg_cnt; /* Number of LRO-packets */
uint16_t __proto_seg_sz; /* Protocol segment size */
/*
* See notes on _PKT_{INTERNALIZE,EXTERNALIZE}() regarding portion
* of this structure above __p_flags that gets copied. Adding
* more user-mutable fields after __p_flags would also require
* adjusting those macros as well.
*/
union {
volatile uint32_t __flags32[2];
volatile uint64_t __flags; /* PKT_F_* flags */
};
} __attribute((aligned(sizeof(uint64_t))));
struct __packet {
union {
uint64_t __pkt_data[4];
struct __packet_com __pkt_com;
};
#define __p_link_flags __pkt_com.__link_flags
#define __p_headroom __pkt_com.__headroom
#define __p_l2_len __pkt_com.__l2_len
#define __p_csum_flags __pkt_com.__csum_flags
#define __p_csum_rx __pkt_com.__csum_rx
#define __p_csum_tx __pkt_com.__csum_tx
#define __p_csum_data __pkt_com.__csum_data
#define __p_comp_gencnt __pkt_com.__comp_gencnt
#define __p_aggr_type __pkt_com.__aggr_type
#define __p_seg_cnt __pkt_com.__seg_cnt
#define __p_proto_seg_sz __pkt_com.__proto_seg_sz
#define __p_trace_id __pkt_com.__trace_id
#define __p_flags32 __pkt_com.__flags32
#define __p_flags __pkt_com.__flags
};
/* optional packet token types */
#define PKT_OPT_TOKEN_TYPE_OPAQUE 1 /* token has opaque data */
#define PKT_OPT_TOKEN_TYPE_PACKET_ID 2 /* token has packet_id */
/* maximum token size */
#define PKT_OPT_MAX_TOKEN_SIZE 16
struct __packet_opt_com {
union {
uint64_t __token_data[2];
uint8_t __token[PKT_OPT_MAX_TOKEN_SIZE];
};
uint64_t __expire_ts;
uint64_t __pkt_tx_time;
uint16_t __vlan_tag;
uint16_t __token_len;
uint8_t __token_type;
uint8_t __expiry_action;
uint8_t __app_type;
uint8_t __app_metadata;
} __attribute((aligned(sizeof(uint64_t))));
struct __packet_opt {
union {
uint64_t __pkt_opt_data[5];
struct __packet_opt_com __pkt_opt_com;
};
#define __po_token_type __pkt_opt_com.__token_type
#define __po_token_len __pkt_opt_com.__token_len
#define __po_vlan_tag __pkt_opt_com.__vlan_tag
#define __po_token_data __pkt_opt_com.__token_data
#define __po_token __pkt_opt_com.__token
#define __po_expire_ts __pkt_opt_com.__expire_ts
#define __po_expiry_action __pkt_opt_com.__expiry_action
#define __po_app_type __pkt_opt_com.__app_type
#define __po_app_metadata __pkt_opt_com.__app_metadata
#define __po_pkt_tx_time __pkt_opt_com.__pkt_tx_time
};
/*
* Packet.
*
* This structure is aligned for efficient copy and accesses.
* It is the user version of the __kern_packet structure.
*
* XXX: Do NOT store kernel pointer/address values here.
*/
struct __user_packet {
struct __user_quantum pkt_qum;
/*
* pkt_flow_id is the flow identifier used by user space stack to identfy a
* flow. This identifier is passed as a metadata on all packets generated by
* the user space stack. On RX flowswitch fills in this metadata on every
* packet and can be used by user space stack for flow classification purposes.
*/
#define pkt_flow_id pkt_qum.qum_flow_id
#define pkt_flow_id_64 pkt_qum.qum_flow_id_val64
#define pkt_qum_qflags pkt_qum.qum_qflags
#define pkt_length pkt_qum.qum_len
#define pkt_qum_buf pkt_qum.qum_buf[0]
#define pkt_svc_class pkt_qum.qum_svc_class
#ifdef KERNEL
/*
* pkt_flow_token is a globally unique flow identifier generated by the
* flowswitch for each flow. Flowswitch stamps every TX packet with this
* identifier. This is the flow identifier which would be visible to the AQM
* logic and the driver.
* pkt_flow_token uses the first 4 bytes of pkt_flow_id as the storage space.
* This is not a problem as pkt_flow_id is only for flowswitch consumption
* and is not required by any other module after the flowswitch TX processing
* stage.
*/
#define pkt_flow_token pkt_qum.qum_flow_id_val32[0]
#endif /* KERNEL */
/*
* Common area between user and kernel variants.
*/
struct __packet pkt_com;
#define pkt_link_flags pkt_com.__p_link_flags
#define pkt_headroom pkt_com.__p_headroom
#define pkt_l2_len pkt_com.__p_l2_len
#define pkt_csum_flags pkt_com.__p_csum_flags
#define pkt_csum_rx_start_off pkt_com.__p_csum_rx.__csum_start_off
#define pkt_csum_rx_value pkt_com.__p_csum_rx.__csum_value
#define pkt_csum_tx_start_off pkt_com.__p_csum_tx.__csum_start_off
#define pkt_csum_tx_stuff_off pkt_com.__p_csum_tx.__csum_stuff_off
#define pkt_csum_data pkt_com.__p_csum_data
#define pkt_comp_gencnt pkt_com.__p_comp_gencnt
#define pkt_aggr_type pkt_com.__p_aggr_type
#define pkt_seg_cnt pkt_com.__p_seg_cnt
#define pkt_proto_seg_sz pkt_com.__p_proto_seg_sz
#define pkt_trace_id pkt_com.__p_trace_id
#if BYTE_ORDER == LITTLE_ENDIAN
#define pkt_pflags32 pkt_com.__p_flags32[0]
#else /* BYTE_ORDER != LITTLE_ENDIAN */
#define pkt_pflags32 pkt_com.__p_flags32[1]
#endif /* BYTE_ORDER != LITTLE_ENDIAN */
#define pkt_pflags pkt_com.__p_flags
/*
* Optional common metadata.
*/
struct __packet_opt pkt_com_opt;
/*
* Userland specific.
*/
/*
* pkt_{bufs,max} aren't part of the common area, on purpose,
* since we selectively update them on internalize/externalize.
*/
const uint16_t pkt_bufs_max; /* maximum size of buflet chain */
const uint16_t pkt_bufs_cnt; /* buflet chain size */
} __attribute((aligned(sizeof(uint64_t))));
/* the size of __user_packet structure for n total buflets */
#define _USER_PACKET_SIZE(n) sizeof(struct __user_packet)
/*
* Valid values for pkt_link_flags.
*/
#define PKT_LINKF_BCAST 0x0001 /* send/received as link-level bcast */
#define PKT_LINKF_MCAST 0x0002 /* send/received as link-level mcast */
#define PKT_LINKF_ETHFCS 0x0004 /* has Ethernet FCS */
/*
* XXX IMPORTANT - READ THIS XXX
*
* Valid values for (64-bit) pkt_pflags.
*
* The lower 32-bit values are equivalent to PKTF_* flags used by mbufs,
* hence the unused values are reserved. Do not use define any of these
* values unless they correspond to PKTF_* flags. Make sure to do the
* following when adding a value in the lower 32-bit range:
*
* a. If the flag is kernel-only, prefix it with 2 underscore characters,
* then add a PKT_F_* alias under the KERNEL block conditional. This
* will help ensure that the libsyscall code doesn't mistakenly use it.
*
* b. In pp_init(), add compile-time assertion to ensure that the PKT_F_*
* value matches the corresponding PKTF_* as defined in <sys/mbuf.h>.
*
* c. Add the new flag to PKT_F_USER_MASK depending on whether it's allowed
* to be used by userland. Flags not added to this mask will only be
* used by the kernel. We only internalize and externalize flags listed
* in PKT_F_USER_MASK.
*
* d. Add the new flag to PKT_F_COMMON_MASK.
*
* When adding an upper 32-bit value, ensure (a) and (c) above are done.
*
* Legend:
*
* (K) - Kernel-only
* (U+K) - User and kernel
* (reserved) - Only to be used for mapping with mbuf PKTF_* flags
*/
#define __PKT_F_FLOW_ID 0x0000000000000001ULL /* (K) */
#define __PKT_F_FLOW_ADV 0x0000000000000002ULL /* (K) */
/* 0x0000000000000004ULL (reserved) */
/* 0x0000000000000008ULL (reserved) */
/* 0x0000000000000010ULL (reserved) */
/* 0x0000000000000020ULL (reserved) */
/* 0x0000000000000040ULL (reserved) */
/* 0x0000000000000080ULL (reserved) */
/* 0x0000000000000100ULL (reserved) */
/* 0x0000000000000200ULL (reserved) */
#define PKT_F_WAKE_PKT 0x0000000000000400ULL /* (U+K) */
/* 0x0000000000000800ULL (reserved) */
/* 0x0000000000001000ULL (reserved) */
/* 0x0000000000002000ULL (reserved) */
/* 0x0000000000004000ULL (reserved) */
#define PKT_F_BACKGROUND 0x0000000000008000ULL /* (U+K) */
/* 0x0000000000010000ULL (reserved) */
/* 0x0000000000020000ULL (reserved) */
#define PKT_F_KEEPALIVE 0x0000000000040000ULL /* (U+K) */
#define PKT_F_REALTIME 0x0000000000080000ULL /* (U+K) */
/* 0x0000000000100000ULL (reserved) */
#define PKT_F_REXMT 0x0000000000200000ULL /* (U+K) */
/* 0x0000000000400000ULL (reserved) */
#define __PKT_F_TX_COMPL_TS_REQ 0x0000000000800000ULL /* (K) */
#define __PKT_F_TS_VALID 0x0000000001000000ULL /* (K) */
/* 0x0000000002000000ULL (reserved) */
#define __PKT_F_NEW_FLOW 0x0000000004000000ULL /* (K) */
#define __PKT_F_START_SEQ 0x0000000008000000ULL /* (K) */
#define PKT_F_LAST_PKT 0x0000000010000000ULL /* (U+K) */
/* 0x0000000020000000ULL (reserved) */
/* 0x0000000040000000ULL (reserved) */
/* 0x0000000080000000ULL (reserved) */
/* --------------------- upper 32-bit below */
#define PKT_F_OPT_GROUP_START 0x0000000100000000ULL /* (U+K) */
#define PKT_F_OPT_GROUP_END 0x0000000200000000ULL /* (U+K) */
#define PKT_F_OPT_EXPIRE_TS 0x0000000400000000ULL /* (U+K) */
#define PKT_F_OPT_TOKEN 0x0000000800000000ULL /* (U+K) */
#define __PKT_F_FLOW_DATA 0x0000001000000000ULL /* (K) */
#define __PKT_F_TX_COMPL_DATA 0x0000002000000000ULL /* (K) */
#define __PKT_F_MBUF_DATA 0x0000004000000000ULL /* (K) */
#define PKT_F_TRUNCATED 0x0000008000000000ULL /* (U+K) */
#define __PKT_F_PKT_DATA 0x0000010000000000ULL /* (K) */
#define PKT_F_PROMISC 0x0000020000000000ULL /* (U+K) */
#define PKT_F_OPT_VLTAG 0x0000040000000000ULL /* (U+K) */
#define PKT_F_OPT_VLTAG_IN_PKT 0x0000080000000000ULL /* (U+K) */
#define __PKT_F_TX_PORT_DATA 0x0000100000000000ULL /* (K) */
#define PKT_F_OPT_EXP_ACTION 0x0000200000000000ULL /* (U+K) */
#define PKT_F_OPT_APP_METADATA 0x0000400000000000ULL /* (U+K) */
#define PKT_F_L4S 0x0000800000000000ULL /* (U+K) */
#define PKT_F_OPT_TX_TIMESTAMP 0x0001000000000000ULL /* (U+K) */
/* 0x0002000000000000ULL */
/* 0x0004000000000000ULL */
/* 0x0008000000000000ULL */
/* 0x0010000000000000ULL */
/* 0x0020000000000000ULL */
/* 0x0040000000000000ULL */
/* 0x0080000000000000ULL */
#define __PKT_F_OPT_ALLOC 0x0100000000000000ULL /* (K) */
#define __PKT_F_FLOW_ALLOC 0x0200000000000000ULL /* (K) */
#define __PKT_F_TX_COMPL_ALLOC 0x0400000000000000ULL /* (K) */
/* 0x0800000000000000ULL */
/* 0x1000000000000000ULL */
/* 0x2000000000000000ULL */
/* 0x4000000000000000ULL */
/* 0x8000000000000000ULL */
/*
* Packet option flags.
*/
#define PKT_F_OPT_DATA \
(PKT_F_OPT_GROUP_START | PKT_F_OPT_GROUP_END | \
PKT_F_OPT_EXPIRE_TS | PKT_F_OPT_TOKEN | \
PKT_F_OPT_VLTAG | PKT_F_OPT_VLTAG_IN_PKT | PKT_F_OPT_EXP_ACTION | \
PKT_F_OPT_APP_METADATA | PKT_F_OPT_TX_TIMESTAMP)
#ifdef KERNEL
/*
* Flags exposed to user (and kernel). See notes above.
*/
#define PKT_F_USER_MASK \
(PKT_F_BACKGROUND | PKT_F_REALTIME | PKT_F_REXMT | \
PKT_F_LAST_PKT | PKT_F_OPT_DATA | PKT_F_PROMISC | \
PKT_F_TRUNCATED | PKT_F_WAKE_PKT | PKT_F_L4S)
/*
* Aliases for kernel-only flags. See notes above. The ones marked
* with (common) have corresponding PKTF_* definitions and are also
* included in PKT_F_COMMON_MASK below.
*/
#define PKT_F_FLOW_ID __PKT_F_FLOW_ID /* (common) */
#define PKT_F_FLOW_ADV __PKT_F_FLOW_ADV /* (common) */
#define PKT_F_TX_COMPL_TS_REQ __PKT_F_TX_COMPL_TS_REQ /* (common) */
#define PKT_F_TS_VALID __PKT_F_TS_VALID /* (common) */
#define PKT_F_NEW_FLOW __PKT_F_NEW_FLOW /* (common) */
#define PKT_F_START_SEQ __PKT_F_START_SEQ /* (common) */
#define PKT_F_FLOW_DATA __PKT_F_FLOW_DATA
#define PKT_F_TX_COMPL_DATA __PKT_F_TX_COMPL_DATA
#define PKT_F_MBUF_DATA __PKT_F_MBUF_DATA
#define PKT_F_PKT_DATA __PKT_F_PKT_DATA
#define PKT_F_OPT_ALLOC __PKT_F_OPT_ALLOC
#define PKT_F_FLOW_ALLOC __PKT_F_FLOW_ALLOC
#define PKT_F_TX_COMPL_ALLOC __PKT_F_TX_COMPL_ALLOC
#define PKT_F_TX_PORT_DATA __PKT_F_TX_PORT_DATA
/*
* Flags related to mbuf attached to the packet.
*/
#define PKT_F_MBUF_MASK (PKT_F_MBUF_DATA | PKT_F_TRUNCATED)
/*
* Flags related to packet attached to the packet.
*/
#define PKT_F_PKT_MASK (PKT_F_PKT_DATA | PKT_F_TRUNCATED)
/*
* Invariant flags kept during _PKT_COPY(). At the moment we keep
* all except those related to the attached mbuf.
*/
#define PKT_F_COPY_MASK (~(PKT_F_MBUF_MASK | PKT_F_PKT_MASK))
/*
* Lower 32-bit flags common to mbuf and __kern_packet. See notes above.
* DO NOT add flags to this mask unless they have equivalent PKTF_* flags
* defined in <sys/mbuf.h>
*/
#define PKT_F_COMMON_MASK \
(PKT_F_BACKGROUND | PKT_F_REALTIME | PKT_F_REXMT | \
PKT_F_LAST_PKT | PKT_F_FLOW_ID | PKT_F_FLOW_ADV | \
PKT_F_TX_COMPL_TS_REQ | PKT_F_TS_VALID | PKT_F_NEW_FLOW | \
PKT_F_START_SEQ | PKT_F_KEEPALIVE | PKT_F_WAKE_PKT)
/*
* Flags retained across alloc/free.
*/
#define PKT_F_INIT_MASK \
(PKT_F_OPT_ALLOC | PKT_F_FLOW_ALLOC | PKT_F_TX_COMPL_ALLOC)
#endif /* KERNEL */
/*
* 64-bit tagged pointer (limit tag to least significant byte).
* We use 2 bits to encode type, and another 2 bits for subtype.
*/
#define SK_PTR_TYPE_MASK ((uint64_t)0x3) /* 00 11 */
#define SK_PTR_SUBTYPE_MASK ((uint64_t)0xc) /* 11 00 */
#define SK_PTR_TAG_MASK ((uint64_t)0xf) /* 11 11 */
#define SK_PTR_TAG(_p) ((uint64_t)(_p) & SK_PTR_TAG_MASK)
#define SK_PTR_ADDR_MASK (~SK_PTR_TAG_MASK)
#define SK_PTR_TYPE(_p) ((uint64_t)(_p) & SK_PTR_TYPE_MASK)
#define SK_PTR_TYPE_ENC(_t) ((uint64_t)(_t) & SK_PTR_TYPE_MASK)
#define SK_PTR_SUBTYPE(_p) (((uint64_t)(_p) & SK_PTR_SUBTYPE_MASK) >> 2)
#define SK_PTR_SUBTYPE_ENC(_s) (((uint64_t)(_s) << 2) & SK_PTR_SUBTYPE_MASK)
#define SK_PTR_ADDR(_p) ((uint64_t)(_p) & SK_PTR_ADDR_MASK)
#define SK_PTR_ADDR_ENC(_p) ((uint64_t)(_p) & SK_PTR_ADDR_MASK)
#define SK_PTR_ENCODE(_p, _t, _s) \
(SK_PTR_ADDR_ENC(_p) | SK_PTR_TYPE_ENC(_t) | SK_PTR_SUBTYPE_ENC(_s))
#define SK_PTR_ADDR_UQUM(_ph) (__unsafe_forge_single(struct __user_quantum *, SK_PTR_ADDR(_ph)))
#define SK_PTR_ADDR_UPKT(_ph) (__unsafe_forge_single(struct __user_packet *, SK_PTR_ADDR(_ph)))
#ifdef KERNEL
__BEGIN_DECLS
/*
* Packets.
*/
extern struct mbuf *kern_packet_get_mbuf(const kern_packet_t);
__END_DECLS
#else /* !KERNEL */
#if defined(LIBSYSCALL_INTERFACE)
__BEGIN_DECLS
extern void pkt_subtype_assert_fail(const packet_t, uint64_t, uint64_t);
extern void pkt_type_assert_fail(const packet_t, uint64_t);
__END_DECLS
#endif /* LIBSYSCALL_INTERFACE */
#endif /* !KERNEL */
#if defined(LIBSYSCALL_INTERFACE) || defined(BSD_KERNEL_PRIVATE)
#include <skywalk/packet_common.h>
#endif /* LIBSYSCALL_INTERFACE || BSD_KERNEL_PRIVATE */
#endif /* PRIVATE || BSD_KERNEL_PRIVATE */
#endif /* !_SKYWALK_OS_PACKET_PRIVATE_H_ */