From 351a4dedb34cbeb9f747f0e2309e891b6fb906cb Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Thu, 2 Jun 2016 10:23:29 +0300 Subject: qed: Utilize FW 8.10.3.0 The New QED firmware contains several fixes, including: - Wrong classification of packets in 4-port devices. - Anti-spoof interoperability with encapsulated packets. - Tx-switching of encapsulated packets. It also slightly improves Tx performance of the device. In addition, this firmware contains the necessary logic for supporting iscsi & rdma, for which we plan on pushing protocol drivers in the imminent future. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller --- include/linux/qed/common_hsi.h | 205 +++++++++++++++++++++++++++++------------ include/linux/qed/eth_common.h | 124 +++++++++++++++---------- include/linux/qed/qed_eth_if.h | 1 + 3 files changed, 222 insertions(+), 108 deletions(-) (limited to 'include/linux') diff --git a/include/linux/qed/common_hsi.h b/include/linux/qed/common_hsi.h index 3f14c7efe68f..285189a5ea6d 100644 --- a/include/linux/qed/common_hsi.h +++ b/include/linux/qed/common_hsi.h @@ -13,9 +13,19 @@ #define X_FINAL_CLEANUP_AGG_INT 1 +/* Queue Zone sizes in bytes */ +#define TSTORM_QZONE_SIZE 8 +#define MSTORM_QZONE_SIZE 0 +#define USTORM_QZONE_SIZE 8 +#define XSTORM_QZONE_SIZE 8 +#define YSTORM_QZONE_SIZE 0 +#define PSTORM_QZONE_SIZE 0 + +#define ETH_MAX_NUM_RX_QUEUES_PER_VF 16 + #define FW_MAJOR_VERSION 8 -#define FW_MINOR_VERSION 7 -#define FW_REVISION_VERSION 3 +#define FW_MINOR_VERSION 10 +#define FW_REVISION_VERSION 5 #define FW_ENGINEERING_VERSION 0 /***********************/ @@ -97,45 +107,86 @@ #define DQ_XCM_AGG_VAL_SEL_REG6 7 /* XCM agg val selection */ -#define DQ_XCM_ETH_EDPM_NUM_BDS_CMD \ - DQ_XCM_AGG_VAL_SEL_WORD2 -#define DQ_XCM_ETH_TX_BD_CONS_CMD \ - DQ_XCM_AGG_VAL_SEL_WORD3 -#define DQ_XCM_CORE_TX_BD_CONS_CMD \ - DQ_XCM_AGG_VAL_SEL_WORD3 -#define DQ_XCM_ETH_TX_BD_PROD_CMD \ - DQ_XCM_AGG_VAL_SEL_WORD4 -#define DQ_XCM_CORE_TX_BD_PROD_CMD \ - DQ_XCM_AGG_VAL_SEL_WORD4 -#define DQ_XCM_CORE_SPQ_PROD_CMD \ - DQ_XCM_AGG_VAL_SEL_WORD4 -#define DQ_XCM_ETH_GO_TO_BD_CONS_CMD DQ_XCM_AGG_VAL_SEL_WORD5 +#define DQ_XCM_CORE_TX_BD_CONS_CMD DQ_XCM_AGG_VAL_SEL_WORD3 +#define DQ_XCM_CORE_TX_BD_PROD_CMD DQ_XCM_AGG_VAL_SEL_WORD4 +#define DQ_XCM_CORE_SPQ_PROD_CMD DQ_XCM_AGG_VAL_SEL_WORD4 +#define DQ_XCM_ETH_EDPM_NUM_BDS_CMD DQ_XCM_AGG_VAL_SEL_WORD2 +#define DQ_XCM_ETH_TX_BD_CONS_CMD DQ_XCM_AGG_VAL_SEL_WORD3 +#define DQ_XCM_ETH_TX_BD_PROD_CMD DQ_XCM_AGG_VAL_SEL_WORD4 +#define DQ_XCM_ETH_GO_TO_BD_CONS_CMD DQ_XCM_AGG_VAL_SEL_WORD5 + +/* UCM agg val selection (HW) */ +#define DQ_UCM_AGG_VAL_SEL_WORD0 0 +#define DQ_UCM_AGG_VAL_SEL_WORD1 1 +#define DQ_UCM_AGG_VAL_SEL_WORD2 2 +#define DQ_UCM_AGG_VAL_SEL_WORD3 3 +#define DQ_UCM_AGG_VAL_SEL_REG0 4 +#define DQ_UCM_AGG_VAL_SEL_REG1 5 +#define DQ_UCM_AGG_VAL_SEL_REG2 6 +#define DQ_UCM_AGG_VAL_SEL_REG3 7 + +/* UCM agg val selection (FW) */ +#define DQ_UCM_ETH_PMD_TX_CONS_CMD DQ_UCM_AGG_VAL_SEL_WORD2 +#define DQ_UCM_ETH_PMD_RX_CONS_CMD DQ_UCM_AGG_VAL_SEL_WORD3 +#define DQ_UCM_ROCE_CQ_CONS_CMD DQ_UCM_AGG_VAL_SEL_REG0 +#define DQ_UCM_ROCE_CQ_PROD_CMD DQ_UCM_AGG_VAL_SEL_REG2 + +/* TCM agg val selection (HW) */ +#define DQ_TCM_AGG_VAL_SEL_WORD0 0 +#define DQ_TCM_AGG_VAL_SEL_WORD1 1 +#define DQ_TCM_AGG_VAL_SEL_WORD2 2 +#define DQ_TCM_AGG_VAL_SEL_WORD3 3 +#define DQ_TCM_AGG_VAL_SEL_REG1 4 +#define DQ_TCM_AGG_VAL_SEL_REG2 5 +#define DQ_TCM_AGG_VAL_SEL_REG6 6 +#define DQ_TCM_AGG_VAL_SEL_REG9 7 + +/* TCM agg val selection (FW) */ +#define DQ_TCM_L2B_BD_PROD_CMD \ + DQ_TCM_AGG_VAL_SEL_WORD1 +#define DQ_TCM_ROCE_RQ_PROD_CMD \ + DQ_TCM_AGG_VAL_SEL_WORD0 /* XCM agg counter flag selection */ -#define DQ_XCM_AGG_FLG_SHIFT_BIT14 0 -#define DQ_XCM_AGG_FLG_SHIFT_BIT15 1 -#define DQ_XCM_AGG_FLG_SHIFT_CF12 2 -#define DQ_XCM_AGG_FLG_SHIFT_CF13 3 -#define DQ_XCM_AGG_FLG_SHIFT_CF18 4 -#define DQ_XCM_AGG_FLG_SHIFT_CF19 5 -#define DQ_XCM_AGG_FLG_SHIFT_CF22 6 -#define DQ_XCM_AGG_FLG_SHIFT_CF23 7 +#define DQ_XCM_AGG_FLG_SHIFT_BIT14 0 +#define DQ_XCM_AGG_FLG_SHIFT_BIT15 1 +#define DQ_XCM_AGG_FLG_SHIFT_CF12 2 +#define DQ_XCM_AGG_FLG_SHIFT_CF13 3 +#define DQ_XCM_AGG_FLG_SHIFT_CF18 4 +#define DQ_XCM_AGG_FLG_SHIFT_CF19 5 +#define DQ_XCM_AGG_FLG_SHIFT_CF22 6 +#define DQ_XCM_AGG_FLG_SHIFT_CF23 7 /* XCM agg counter flag selection */ -#define DQ_XCM_ETH_DQ_CF_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF18) -#define DQ_XCM_CORE_DQ_CF_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF18) -#define DQ_XCM_ETH_TERMINATE_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF19) -#define DQ_XCM_CORE_TERMINATE_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF19) -#define DQ_XCM_ETH_SLOW_PATH_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF22) -#define DQ_XCM_CORE_SLOW_PATH_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF22) -#define DQ_XCM_ETH_TPH_EN_CMD (1 << \ - DQ_XCM_AGG_FLG_SHIFT_CF23) +#define DQ_XCM_CORE_DQ_CF_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF18) +#define DQ_XCM_CORE_TERMINATE_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF19) +#define DQ_XCM_CORE_SLOW_PATH_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF22) +#define DQ_XCM_ETH_DQ_CF_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF18) +#define DQ_XCM_ETH_TERMINATE_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF19) +#define DQ_XCM_ETH_SLOW_PATH_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF22) +#define DQ_XCM_ETH_TPH_EN_CMD (1 << DQ_XCM_AGG_FLG_SHIFT_CF23) + +/* UCM agg counter flag selection (HW) */ +#define DQ_UCM_AGG_FLG_SHIFT_CF0 0 +#define DQ_UCM_AGG_FLG_SHIFT_CF1 1 +#define DQ_UCM_AGG_FLG_SHIFT_CF3 2 +#define DQ_UCM_AGG_FLG_SHIFT_CF4 3 +#define DQ_UCM_AGG_FLG_SHIFT_CF5 4 +#define DQ_UCM_AGG_FLG_SHIFT_CF6 5 +#define DQ_UCM_AGG_FLG_SHIFT_RULE0EN 6 +#define DQ_UCM_AGG_FLG_SHIFT_RULE1EN 7 + +/* UCM agg counter flag selection (FW) */ +#define DQ_UCM_ETH_PMD_TX_ARM_CMD (1 << DQ_UCM_AGG_FLG_SHIFT_CF4) +#define DQ_UCM_ETH_PMD_RX_ARM_CMD (1 << DQ_UCM_AGG_FLG_SHIFT_CF5) + +#define DQ_REGION_SHIFT (12) + +/* DPM */ +#define DQ_DPM_WQE_BUFF_SIZE (320) + +/* Conn type ranges */ +#define DQ_CONN_TYPE_RANGE_SHIFT (4) /*****************/ /* QM CONSTANTS */ @@ -282,8 +333,6 @@ (PXP_EXTERNAL_BAR_GLOBAL_WINDOW_START + \ PXP_EXTERNAL_BAR_GLOBAL_WINDOW_LENGTH - 1) -#define PXP_ILT_PAGE_SIZE_NUM_BITS_MIN 12 -#define PXP_ILT_BLOCK_FACTOR_MULTIPLIER 1024 #define PXP_VF_BAR0_START_IGU 0 #define PXP_VF_BAR0_IGU_LENGTH 0x3000 @@ -342,6 +391,9 @@ #define PXP_VF_BAR0_GRC_WINDOW_LENGTH 32 +#define PXP_ILT_PAGE_SIZE_NUM_BITS_MIN 12 +#define PXP_ILT_BLOCK_FACTOR_MULTIPLIER 1024 + /* ILT Records */ #define PXP_NUM_ILT_RECORDS_BB 7600 #define PXP_NUM_ILT_RECORDS_K2 11000 @@ -379,6 +431,38 @@ struct async_data { u8 fw_debug_param; }; +struct coalescing_timeset { + u8 value; +#define COALESCING_TIMESET_TIMESET_MASK 0x7F +#define COALESCING_TIMESET_TIMESET_SHIFT 0 +#define COALESCING_TIMESET_VALID_MASK 0x1 +#define COALESCING_TIMESET_VALID_SHIFT 7 +}; + +struct common_prs_pf_msg_info { + __le32 value; +#define COMMON_PRS_PF_MSG_INFO_NPAR_DEFAULT_PF_MASK 0x1 +#define COMMON_PRS_PF_MSG_INFO_NPAR_DEFAULT_PF_SHIFT 0 +#define COMMON_PRS_PF_MSG_INFO_FW_DEBUG_1_MASK 0x1 +#define COMMON_PRS_PF_MSG_INFO_FW_DEBUG_1_SHIFT 1 +#define COMMON_PRS_PF_MSG_INFO_FW_DEBUG_2_MASK 0x1 +#define COMMON_PRS_PF_MSG_INFO_FW_DEBUG_2_SHIFT 2 +#define COMMON_PRS_PF_MSG_INFO_FW_DEBUG_3_MASK 0x1 +#define COMMON_PRS_PF_MSG_INFO_FW_DEBUG_3_SHIFT 3 +#define COMMON_PRS_PF_MSG_INFO_RESERVED_MASK 0xFFFFFFF +#define COMMON_PRS_PF_MSG_INFO_RESERVED_SHIFT 4 +}; + +struct common_queue_zone { + __le16 ring_drv_data_consumer; + __le16 reserved; +}; + +struct eth_rx_prod_data { + __le16 bd_prod; + __le16 cqe_prod; +}; + struct regpair { __le32 lo; __le32 hi; @@ -388,11 +472,23 @@ struct vf_pf_channel_eqe_data { struct regpair msg_addr; }; +struct malicious_vf_eqe_data { + u8 vf_id; + u8 err_id; + __le16 reserved[3]; +}; + +struct initial_cleanup_eqe_data { + u8 vf_id; + u8 reserved[7]; +}; + /* Event Data Union */ union event_ring_data { - u8 bytes[8]; - struct vf_pf_channel_eqe_data vf_pf_channel; - struct async_data async_info; + u8 bytes[8]; + struct vf_pf_channel_eqe_data vf_pf_channel; + struct malicious_vf_eqe_data malicious_vf; + struct initial_cleanup_eqe_data vf_init_cleanup; }; /* Event Ring Entry */ @@ -433,6 +529,16 @@ enum protocol_type { MAX_PROTOCOL_TYPE }; +struct ustorm_eth_queue_zone { + struct coalescing_timeset int_coalescing_timeset; + u8 reserved[3]; +}; + +struct ustorm_queue_zone { + struct ustorm_eth_queue_zone eth; + struct common_queue_zone common; +}; + /* status block structure */ struct cau_pi_entry { u32 prod; @@ -683,19 +789,4 @@ struct status_block { #define STATUS_BLOCK_ZERO_PAD3_SHIFT 24 }; -struct tunnel_parsing_flags { - u8 flags; -#define TUNNEL_PARSING_FLAGS_TYPE_MASK 0x3 -#define TUNNEL_PARSING_FLAGS_TYPE_SHIFT 0 -#define TUNNEL_PARSING_FLAGS_TENNANT_ID_EXIST_MASK 0x1 -#define TUNNEL_PARSING_FLAGS_TENNANT_ID_EXIST_SHIFT 2 -#define TUNNEL_PARSING_FLAGS_NEXT_PROTOCOL_MASK 0x3 -#define TUNNEL_PARSING_FLAGS_NEXT_PROTOCOL_SHIFT 3 -#define TUNNEL_PARSING_FLAGS_FIRSTHDRIPMATCH_MASK 0x1 -#define TUNNEL_PARSING_FLAGS_FIRSTHDRIPMATCH_SHIFT 5 -#define TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_MASK 0x1 -#define TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_SHIFT 6 -#define TUNNEL_PARSING_FLAGS_IPV4_OPTIONS_MASK 0x1 -#define TUNNEL_PARSING_FLAGS_IPV4_OPTIONS_SHIFT 7 -}; #endif /* __COMMON_HSI__ */ diff --git a/include/linux/qed/eth_common.h b/include/linux/qed/eth_common.h index 092cb0c1afcb..b5ebc697d05f 100644 --- a/include/linux/qed/eth_common.h +++ b/include/linux/qed/eth_common.h @@ -12,6 +12,8 @@ /********************/ /* ETH FW CONSTANTS */ /********************/ +#define ETH_HSI_VER_MAJOR 3 +#define ETH_HSI_VER_MINOR 0 #define ETH_CACHE_LINE_SIZE 64 #define ETH_MAX_RAMROD_PER_CON 8 @@ -57,19 +59,6 @@ #define ETH_TPA_CQE_CONT_LEN_LIST_SIZE 6 #define ETH_TPA_CQE_END_LEN_LIST_SIZE 4 -/* Queue Zone sizes */ -#define TSTORM_QZONE_SIZE 0 -#define MSTORM_QZONE_SIZE sizeof(struct mstorm_eth_queue_zone) -#define USTORM_QZONE_SIZE sizeof(struct ustorm_eth_queue_zone) -#define XSTORM_QZONE_SIZE 0 -#define YSTORM_QZONE_SIZE sizeof(struct ystorm_eth_queue_zone) -#define PSTORM_QZONE_SIZE 0 - -/* Interrupt coalescing TimeSet */ -struct coalescing_timeset { - u8 timeset; - u8 valid; -}; struct eth_tx_1st_bd_flags { u8 bitfields; @@ -97,12 +86,12 @@ struct eth_tx_data_1st_bd { u8 nbds; struct eth_tx_1st_bd_flags bd_flags; __le16 bitfields; -#define ETH_TX_DATA_1ST_BD_TUNN_CFG_OVERRIDE_MASK 0x1 -#define ETH_TX_DATA_1ST_BD_TUNN_CFG_OVERRIDE_SHIFT 0 +#define ETH_TX_DATA_1ST_BD_TUNN_FLAG_MASK 0x1 +#define ETH_TX_DATA_1ST_BD_TUNN_FLAG_SHIFT 0 #define ETH_TX_DATA_1ST_BD_RESERVED0_MASK 0x1 #define ETH_TX_DATA_1ST_BD_RESERVED0_SHIFT 1 -#define ETH_TX_DATA_1ST_BD_FW_USE_ONLY_MASK 0x3FFF -#define ETH_TX_DATA_1ST_BD_FW_USE_ONLY_SHIFT 2 +#define ETH_TX_DATA_1ST_BD_PKT_LEN_MASK 0x3FFF +#define ETH_TX_DATA_1ST_BD_PKT_LEN_SHIFT 2 }; /* The parsing information data for the second tx bd of a given packet. */ @@ -136,28 +125,51 @@ struct eth_tx_data_2nd_bd { #define ETH_TX_DATA_2ND_BD_RESERVED0_SHIFT 13 }; +struct eth_fast_path_cqe_fw_debug { + u8 reserved0; + u8 reserved1; + __le16 reserved2; +}; + +/* tunneling parsing flags */ +struct eth_tunnel_parsing_flags { + u8 flags; +#define ETH_TUNNEL_PARSING_FLAGS_TYPE_MASK 0x3 +#define ETH_TUNNEL_PARSING_FLAGS_TYPE_SHIFT 0 +#define ETH_TUNNEL_PARSING_FLAGS_TENNANT_ID_EXIST_MASK 0x1 +#define ETH_TUNNEL_PARSING_FLAGS_TENNANT_ID_EXIST_SHIFT 2 +#define ETH_TUNNEL_PARSING_FLAGS_NEXT_PROTOCOL_MASK 0x3 +#define ETH_TUNNEL_PARSING_FLAGS_NEXT_PROTOCOL_SHIFT 3 +#define ETH_TUNNEL_PARSING_FLAGS_FIRSTHDRIPMATCH_MASK 0x1 +#define ETH_TUNNEL_PARSING_FLAGS_FIRSTHDRIPMATCH_SHIFT 5 +#define ETH_TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_MASK 0x1 +#define ETH_TUNNEL_PARSING_FLAGS_IPV4_FRAGMENT_SHIFT 6 +#define ETH_TUNNEL_PARSING_FLAGS_IPV4_OPTIONS_MASK 0x1 +#define ETH_TUNNEL_PARSING_FLAGS_IPV4_OPTIONS_SHIFT 7 +}; + /* Regular ETH Rx FP CQE. */ struct eth_fast_path_rx_reg_cqe { - u8 type; - u8 bitfields; + u8 type; + u8 bitfields; #define ETH_FAST_PATH_RX_REG_CQE_RSS_HASH_TYPE_MASK 0x7 #define ETH_FAST_PATH_RX_REG_CQE_RSS_HASH_TYPE_SHIFT 0 #define ETH_FAST_PATH_RX_REG_CQE_TC_MASK 0xF #define ETH_FAST_PATH_RX_REG_CQE_TC_SHIFT 3 #define ETH_FAST_PATH_RX_REG_CQE_RESERVED0_MASK 0x1 #define ETH_FAST_PATH_RX_REG_CQE_RESERVED0_SHIFT 7 - __le16 pkt_len; - struct parsing_and_err_flags pars_flags; - __le16 vlan_tag; - __le32 rss_hash; - __le16 len_on_first_bd; - u8 placement_offset; - struct tunnel_parsing_flags tunnel_pars_flags; - u8 bd_num; - u8 reserved[7]; - u32 fw_debug; - u8 reserved1[3]; - u8 flags; + __le16 pkt_len; + struct parsing_and_err_flags pars_flags; + __le16 vlan_tag; + __le32 rss_hash; + __le16 len_on_first_bd; + u8 placement_offset; + struct eth_tunnel_parsing_flags tunnel_pars_flags; + u8 bd_num; + u8 reserved[7]; + struct eth_fast_path_cqe_fw_debug fw_debug; + u8 reserved1[3]; + u8 flags; #define ETH_FAST_PATH_RX_REG_CQE_VALID_MASK 0x1 #define ETH_FAST_PATH_RX_REG_CQE_VALID_SHIFT 0 #define ETH_FAST_PATH_RX_REG_CQE_VALID_TOGGLE_MASK 0x1 @@ -207,11 +219,11 @@ struct eth_fast_path_rx_tpa_start_cqe { __le32 rss_hash; __le16 len_on_first_bd; u8 placement_offset; - struct tunnel_parsing_flags tunnel_pars_flags; + struct eth_tunnel_parsing_flags tunnel_pars_flags; u8 tpa_agg_index; u8 header_len; __le16 ext_bd_len_list[ETH_TPA_CQE_START_LEN_LIST_SIZE]; - u32 fw_debug; + struct eth_fast_path_cqe_fw_debug fw_debug; }; /* The L4 pseudo checksum mode for Ethernet */ @@ -264,12 +276,25 @@ enum eth_rx_cqe_type { MAX_ETH_RX_CQE_TYPE }; -/* ETH Rx producers data */ -struct eth_rx_prod_data { - __le16 bd_prod; - __le16 cqe_prod; - __le16 reserved; - __le16 reserved1; +enum eth_rx_tunn_type { + ETH_RX_NO_TUNN, + ETH_RX_TUNN_GENEVE, + ETH_RX_TUNN_GRE, + ETH_RX_TUNN_VXLAN, + MAX_ETH_RX_TUNN_TYPE +}; + +/* Aggregation end reason. */ +enum eth_tpa_end_reason { + ETH_AGG_END_UNUSED, + ETH_AGG_END_SP_UPDATE, + ETH_AGG_END_MAX_LEN, + ETH_AGG_END_LAST_SEG, + ETH_AGG_END_TIMEOUT, + ETH_AGG_END_NOT_CONSISTENT, + ETH_AGG_END_OUT_OF_ORDER, + ETH_AGG_END_NON_TPA_SEG, + MAX_ETH_TPA_END_REASON }; /* The first tx bd of a given packet */ @@ -337,21 +362,18 @@ union eth_tx_bd_types { }; /* Mstorm Queue Zone */ -struct mstorm_eth_queue_zone { - struct eth_rx_prod_data rx_producers; - __le32 reserved[2]; -}; - -/* Ustorm Queue Zone */ -struct ustorm_eth_queue_zone { - struct coalescing_timeset int_coalescing_timeset; - __le16 reserved[3]; +enum eth_tx_tunn_type { + ETH_TX_TUNN_GENEVE, + ETH_TX_TUNN_TTAG, + ETH_TX_TUNN_GRE, + ETH_TX_TUNN_VXLAN, + MAX_ETH_TX_TUNN_TYPE }; /* Ystorm Queue Zone */ -struct ystorm_eth_queue_zone { - struct coalescing_timeset int_coalescing_timeset; - __le16 reserved[3]; +struct xstorm_eth_queue_zone { + struct coalescing_timeset int_coalescing_timeset; + u8 reserved[7]; }; /* ETH doorbell data */ diff --git a/include/linux/qed/qed_eth_if.h b/include/linux/qed/qed_eth_if.h index 6ae8cb4a61d3..f8ff71126d9e 100644 --- a/include/linux/qed/qed_eth_if.h +++ b/include/linux/qed/qed_eth_if.h @@ -113,6 +113,7 @@ struct qed_queue_start_common_params { u8 vport_id; u16 sb; u16 sb_idx; + u16 vf_qid; }; struct qed_tunn_params { -- cgit v1.2.3 From 3953c46c3ac7eef31a9935427371c6f54a22f1ba Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 2 Jun 2016 15:05:40 -0300 Subject: sk_buff: allow segmenting based on frag sizes This patch allows segmenting a skb based on its frags sizes instead of based on a fixed value. Signed-off-by: Marcelo Ricardo Leitner Tested-by: Xin Long Signed-off-by: David S. Miller --- include/linux/skbuff.h | 5 +++++ net/core/skbuff.c | 10 +++++++--- 2 files changed, 12 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index ee38a4127475..329a0a9ef671 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -301,6 +301,11 @@ struct sk_buff; #endif extern int sysctl_max_skb_frags; +/* Set skb_shinfo(skb)->gso_size to this in case you want skb_segment to + * segment using its current segmentation instead. + */ +#define GSO_BY_FRAGS 0xFFFF + typedef struct skb_frag_struct skb_frag_t; struct skb_frag_struct { diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 4724bcf9b0ca..97c32c75e704 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -3116,9 +3116,13 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb, int hsize; int size; - len = head_skb->len - offset; - if (len > mss) - len = mss; + if (unlikely(mss == GSO_BY_FRAGS)) { + len = list_skb->len; + } else { + len = head_skb->len - offset; + if (len > mss) + len = mss; + } hsize = skb_headlen(head_skb) - offset; if (hsize < 0) -- cgit v1.2.3 From ae7ef81ef000adeee7a87585b9135ff8a8064acc Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 2 Jun 2016 15:05:41 -0300 Subject: skbuff: introduce skb_gso_validate_mtu skb_gso_network_seglen is not enough for checking fragment sizes if skb is using GSO_BY_FRAGS as we have to check frag per frag. This patch introduces skb_gso_validate_mtu, based on the former, which will wrap the use case inside it as all calls to skb_gso_network_seglen were to validate if it fits on a given TMU, and improve the check. Signed-off-by: Marcelo Ricardo Leitner Tested-by: Xin Long Signed-off-by: David S. Miller --- include/linux/skbuff.h | 1 + net/core/skbuff.c | 31 +++++++++++++++++++++++++++++++ net/ipv4/ip_forward.c | 2 +- net/ipv4/ip_output.c | 2 +- net/ipv6/ip6_output.c | 2 +- net/mpls/af_mpls.c | 2 +- 6 files changed, 36 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 329a0a9ef671..aa3f9d7e8d5c 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -2992,6 +2992,7 @@ void skb_split(struct sk_buff *skb, struct sk_buff *skb1, const u32 len); int skb_shift(struct sk_buff *tgt, struct sk_buff *skb, int shiftlen); void skb_scrub_packet(struct sk_buff *skb, bool xnet); unsigned int skb_gso_transport_seglen(const struct sk_buff *skb); +bool skb_gso_validate_mtu(const struct sk_buff *skb, unsigned int mtu); struct sk_buff *skb_segment(struct sk_buff *skb, netdev_features_t features); struct sk_buff *skb_vlan_untag(struct sk_buff *skb); int skb_ensure_writable(struct sk_buff *skb, int write_len); diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 97c32c75e704..5ca562b56ec3 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -4392,6 +4392,37 @@ unsigned int skb_gso_transport_seglen(const struct sk_buff *skb) } EXPORT_SYMBOL_GPL(skb_gso_transport_seglen); +/** + * skb_gso_validate_mtu - Return in case such skb fits a given MTU + * + * @skb: GSO skb + * + * skb_gso_validate_mtu validates if a given skb will fit a wanted MTU + * once split. + */ +bool skb_gso_validate_mtu(const struct sk_buff *skb, unsigned int mtu) +{ + const struct skb_shared_info *shinfo = skb_shinfo(skb); + const struct sk_buff *iter; + unsigned int hlen; + + hlen = skb_gso_network_seglen(skb); + + if (shinfo->gso_size != GSO_BY_FRAGS) + return hlen <= mtu; + + /* Undo this so we can re-use header sizes */ + hlen -= GSO_BY_FRAGS; + + skb_walk_frags(skb, iter) { + if (hlen + skb_headlen(iter) > mtu) + return false; + } + + return true; +} +EXPORT_SYMBOL_GPL(skb_gso_validate_mtu); + static struct sk_buff *skb_reorder_vlan_header(struct sk_buff *skb) { if (skb_cow(skb, skb_headroom(skb)) < 0) { diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c index cbfb1808fcc4..9f0a7b96646f 100644 --- a/net/ipv4/ip_forward.c +++ b/net/ipv4/ip_forward.c @@ -54,7 +54,7 @@ static bool ip_exceeds_mtu(const struct sk_buff *skb, unsigned int mtu) if (skb->ignore_df) return false; - if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu) + if (skb_is_gso(skb) && skb_gso_validate_mtu(skb, mtu)) return false; return true; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 124bf0a66328..cbac493c913a 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -225,7 +225,7 @@ static int ip_finish_output_gso(struct net *net, struct sock *sk, /* common case: locally created skb or seglen is <= mtu */ if (((IPCB(skb)->flags & IPSKB_FORWARDED) == 0) || - skb_gso_network_seglen(skb) <= mtu) + skb_gso_validate_mtu(skb, mtu)) return ip_finish_output2(net, sk, skb); /* Slowpath - GSO segment length is exceeding the dst MTU. diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index cbf127ae7c67..6b2f60a5c1de 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -368,7 +368,7 @@ static bool ip6_pkt_too_big(const struct sk_buff *skb, unsigned int mtu) if (skb->ignore_df) return false; - if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu) + if (skb_is_gso(skb) && skb_gso_validate_mtu(skb, mtu)) return false; return true; diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c index 0b80a7140cc4..7a4aa3450dd7 100644 --- a/net/mpls/af_mpls.c +++ b/net/mpls/af_mpls.c @@ -91,7 +91,7 @@ bool mpls_pkt_too_big(const struct sk_buff *skb, unsigned int mtu) if (skb->len <= mtu) return false; - if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu) + if (skb_is_gso(skb) && skb_gso_validate_mtu(skb, mtu)) return false; return true; -- cgit v1.2.3 From 90017accff61ae89283ad9a51f9ac46ca01633fb Mon Sep 17 00:00:00 2001 From: Marcelo Ricardo Leitner Date: Thu, 2 Jun 2016 15:05:43 -0300 Subject: sctp: Add GSO support SCTP has this pecualiarity that its packets cannot be just segmented to (P)MTU. Its chunks must be contained in IP segments, padding respected. So we can't just generate a big skb, set gso_size to the fragmentation point and deliver it to IP layer. This patch takes a different approach. SCTP will now build a skb as it would be if it was received using GRO. That is, there will be a cover skb with protocol headers and children ones containing the actual segments, already segmented to a way that respects SCTP RFCs. With that, we can tell skb_segment() to just split based on frag_list, trusting its sizes are already in accordance. This way SCTP can benefit from GSO and instead of passing several packets through the stack, it can pass a single large packet. v2: - Added support for receiving GSO frames, as requested by Dave Miller. - Clear skb->cb if packet is GSO (otherwise it's not used by SCTP) - Added heuristics similar to what we have in TCP for not generating single GSO packets that fills cwnd. v3: - consider sctphdr size in skb_gso_transport_seglen() - rebased due to 5c7cdf339af5 ("gso: Remove arbitrary checks for unsupported GSO") Signed-off-by: Marcelo Ricardo Leitner Tested-by: Xin Long Signed-off-by: David S. Miller --- include/linux/netdev_features.h | 7 +- include/linux/netdevice.h | 1 + include/linux/skbuff.h | 2 + include/net/sctp/sctp.h | 4 + include/net/sctp/structs.h | 5 + net/core/ethtool.c | 1 + net/core/skbuff.c | 3 + net/sctp/Makefile | 3 +- net/sctp/input.c | 12 +- net/sctp/inqueue.c | 51 +++++- net/sctp/offload.c | 98 +++++++++++ net/sctp/output.c | 363 +++++++++++++++++++++++++++------------- net/sctp/protocol.c | 3 + net/sctp/socket.c | 2 + 14 files changed, 429 insertions(+), 126 deletions(-) create mode 100644 net/sctp/offload.c (limited to 'include/linux') diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h index aa7b2400f98c..9c6c8ef2e9e7 100644 --- a/include/linux/netdev_features.h +++ b/include/linux/netdev_features.h @@ -53,8 +53,9 @@ enum { * headers in software. */ NETIF_F_GSO_TUNNEL_REMCSUM_BIT, /* ... TUNNEL with TSO & REMCSUM */ + NETIF_F_GSO_SCTP_BIT, /* ... SCTP fragmentation */ /**/NETIF_F_GSO_LAST = /* last bit, see GSO_MASK */ - NETIF_F_GSO_TUNNEL_REMCSUM_BIT, + NETIF_F_GSO_SCTP_BIT, NETIF_F_FCOE_CRC_BIT, /* FCoE CRC32 */ NETIF_F_SCTP_CRC_BIT, /* SCTP checksum offload */ @@ -128,6 +129,7 @@ enum { #define NETIF_F_TSO_MANGLEID __NETIF_F(TSO_MANGLEID) #define NETIF_F_GSO_PARTIAL __NETIF_F(GSO_PARTIAL) #define NETIF_F_GSO_TUNNEL_REMCSUM __NETIF_F(GSO_TUNNEL_REMCSUM) +#define NETIF_F_GSO_SCTP __NETIF_F(GSO_SCTP) #define NETIF_F_HW_VLAN_STAG_FILTER __NETIF_F(HW_VLAN_STAG_FILTER) #define NETIF_F_HW_VLAN_STAG_RX __NETIF_F(HW_VLAN_STAG_RX) #define NETIF_F_HW_VLAN_STAG_TX __NETIF_F(HW_VLAN_STAG_TX) @@ -166,7 +168,8 @@ enum { NETIF_F_FSO) /* List of features with software fallbacks. */ -#define NETIF_F_GSO_SOFTWARE (NETIF_F_ALL_TSO | NETIF_F_UFO) +#define NETIF_F_GSO_SOFTWARE (NETIF_F_ALL_TSO | NETIF_F_UFO | \ + NETIF_F_GSO_SCTP) /* * If one device supports one of these features, then enable them diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index f45929ce8157..fa6df2699532 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -4012,6 +4012,7 @@ static inline bool net_gso_ok(netdev_features_t features, int gso_type) BUILD_BUG_ON(SKB_GSO_UDP_TUNNEL_CSUM != (NETIF_F_GSO_UDP_TUNNEL_CSUM >> NETIF_F_GSO_SHIFT)); BUILD_BUG_ON(SKB_GSO_PARTIAL != (NETIF_F_GSO_PARTIAL >> NETIF_F_GSO_SHIFT)); BUILD_BUG_ON(SKB_GSO_TUNNEL_REMCSUM != (NETIF_F_GSO_TUNNEL_REMCSUM >> NETIF_F_GSO_SHIFT)); + BUILD_BUG_ON(SKB_GSO_SCTP != (NETIF_F_GSO_SCTP >> NETIF_F_GSO_SHIFT)); return (features & feature) == feature; } diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index aa3f9d7e8d5c..dc0fca747c5e 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -487,6 +487,8 @@ enum { SKB_GSO_PARTIAL = 1 << 13, SKB_GSO_TUNNEL_REMCSUM = 1 << 14, + + SKB_GSO_SCTP = 1 << 15, }; #if BITS_PER_LONG > 32 diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index b392ac8382f2..632e205ca54b 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -186,6 +186,10 @@ void sctp_assocs_proc_exit(struct net *net); int sctp_remaddr_proc_init(struct net *net); void sctp_remaddr_proc_exit(struct net *net); +/* + * sctp/offload.c + */ +int sctp_offload_init(void); /* * Module global variables diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index 16b013a6191c..83c5ec58b93a 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -566,6 +566,9 @@ struct sctp_chunk { /* This points to the sk_buff containing the actual data. */ struct sk_buff *skb; + /* In case of GSO packets, this will store the head one */ + struct sk_buff *head_skb; + /* These are the SCTP headers by reverse order in a packet. * Note that some of these may happen more than once. In that * case, we point at the "current" one, whatever that means @@ -696,6 +699,8 @@ struct sctp_packet { size_t overhead; /* This is the total size of all chunks INCLUDING padding. */ size_t size; + /* This is the maximum size this packet may have */ + size_t max_size; /* The packet is destined for this transport address. * The function we finally use to pass down to the next lower diff --git a/net/core/ethtool.c b/net/core/ethtool.c index f4034817d255..977489820eb9 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -89,6 +89,7 @@ static const char netdev_features_strings[NETDEV_FEATURE_COUNT][ETH_GSTRING_LEN] [NETIF_F_GSO_UDP_TUNNEL_BIT] = "tx-udp_tnl-segmentation", [NETIF_F_GSO_UDP_TUNNEL_CSUM_BIT] = "tx-udp_tnl-csum-segmentation", [NETIF_F_GSO_PARTIAL_BIT] = "tx-gso-partial", + [NETIF_F_GSO_SCTP_BIT] = "tx-sctp-segmentation", [NETIF_F_FCOE_CRC_BIT] = "tx-checksum-fcoe-crc", [NETIF_F_SCTP_CRC_BIT] = "tx-checksum-sctp", diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 5ca562b56ec3..b6e0f95bef36 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -49,6 +49,7 @@ #include #include #include +#include #include #ifdef CONFIG_NET_CLS_ACT #include @@ -4383,6 +4384,8 @@ unsigned int skb_gso_transport_seglen(const struct sk_buff *skb) thlen += inner_tcp_hdrlen(skb); } else if (likely(shinfo->gso_type & (SKB_GSO_TCPV4 | SKB_GSO_TCPV6))) { thlen = tcp_hdrlen(skb); + } else if (unlikely(shinfo->gso_type & SKB_GSO_SCTP)) { + thlen = sizeof(struct sctphdr); } /* UFO sets gso_size to the size of the fragmentation * payload, i.e. the size of the L4 (UDP) header is already diff --git a/net/sctp/Makefile b/net/sctp/Makefile index 0fca5824ad0e..6c4f7496cec6 100644 --- a/net/sctp/Makefile +++ b/net/sctp/Makefile @@ -11,7 +11,8 @@ sctp-y := sm_statetable.o sm_statefuns.o sm_sideeffect.o \ transport.o chunk.o sm_make_chunk.o ulpevent.o \ inqueue.o outqueue.o ulpqueue.o \ tsnmap.o bind_addr.o socket.o primitive.o \ - output.o input.o debug.o ssnmap.o auth.o + output.o input.o debug.o ssnmap.o auth.o \ + offload.o sctp_probe-y := probe.o diff --git a/net/sctp/input.c b/net/sctp/input.c index 5cff2546c3dd..6f8e676d285e 100644 --- a/net/sctp/input.c +++ b/net/sctp/input.c @@ -139,7 +139,9 @@ int sctp_rcv(struct sk_buff *skb) skb->csum_valid = 0; /* Previous value not applicable */ if (skb_csum_unnecessary(skb)) __skb_decr_checksum_unnecessary(skb); - else if (!sctp_checksum_disable && sctp_rcv_checksum(net, skb) < 0) + else if (!sctp_checksum_disable && + !(skb_shinfo(skb)->gso_type & SKB_GSO_SCTP) && + sctp_rcv_checksum(net, skb) < 0) goto discard_it; skb->csum_valid = 1; @@ -1175,6 +1177,14 @@ static struct sctp_association *__sctp_rcv_lookup_harder(struct net *net, { sctp_chunkhdr_t *ch; + /* We do not allow GSO frames here as we need to linearize and + * then cannot guarantee frame boundaries. This shouldn't be an + * issue as packets hitting this are mostly INIT or INIT-ACK and + * those cannot be on GSO-style anyway. + */ + if ((skb_shinfo(skb)->gso_type & SKB_GSO_SCTP) == SKB_GSO_SCTP) + return NULL; + if (skb_linearize(skb)) return NULL; diff --git a/net/sctp/inqueue.c b/net/sctp/inqueue.c index 5ba08ceda3ab..edabbbdfca54 100644 --- a/net/sctp/inqueue.c +++ b/net/sctp/inqueue.c @@ -138,6 +138,17 @@ struct sctp_chunk *sctp_inq_pop(struct sctp_inq *queue) if (chunk->singleton || chunk->end_of_packet || chunk->pdiscard) { + if (chunk->head_skb == chunk->skb) { + chunk->skb = skb_shinfo(chunk->skb)->frag_list; + goto new_skb; + } + if (chunk->skb->next) { + chunk->skb = chunk->skb->next; + goto new_skb; + } + + if (chunk->head_skb) + chunk->skb = chunk->head_skb; sctp_chunk_free(chunk); chunk = queue->in_progress = NULL; } else { @@ -155,15 +166,15 @@ struct sctp_chunk *sctp_inq_pop(struct sctp_inq *queue) next_chunk: /* Is the queue empty? */ - if (list_empty(&queue->in_chunk_list)) + entry = sctp_list_dequeue(&queue->in_chunk_list); + if (!entry) return NULL; - entry = queue->in_chunk_list.next; chunk = list_entry(entry, struct sctp_chunk, list); - list_del_init(entry); /* Linearize if it's not GSO */ - if (skb_is_nonlinear(chunk->skb)) { + if ((skb_shinfo(chunk->skb)->gso_type & SKB_GSO_SCTP) != SKB_GSO_SCTP && + skb_is_nonlinear(chunk->skb)) { if (skb_linearize(chunk->skb)) { __SCTP_INC_STATS(dev_net(chunk->skb->dev), SCTP_MIB_IN_PKT_DISCARDS); sctp_chunk_free(chunk); @@ -174,15 +185,39 @@ next_chunk: chunk->sctp_hdr = sctp_hdr(chunk->skb); } + if ((skb_shinfo(chunk->skb)->gso_type & SKB_GSO_SCTP) == SKB_GSO_SCTP) { + /* GSO-marked skbs but without frags, handle + * them normally + */ + if (skb_shinfo(chunk->skb)->frag_list) + chunk->head_skb = chunk->skb; + + /* skbs with "cover letter" */ + if (chunk->head_skb && chunk->skb->data_len == chunk->skb->len) + chunk->skb = skb_shinfo(chunk->skb)->frag_list; + + if (WARN_ON(!chunk->skb)) { + __SCTP_INC_STATS(dev_net(chunk->skb->dev), SCTP_MIB_IN_PKT_DISCARDS); + sctp_chunk_free(chunk); + goto next_chunk; + } + } + + if (chunk->asoc) + sock_rps_save_rxhash(chunk->asoc->base.sk, chunk->skb); + queue->in_progress = chunk; +new_skb: /* This is the first chunk in the packet. */ - chunk->singleton = 1; ch = (sctp_chunkhdr_t *) chunk->skb->data; + chunk->singleton = 1; chunk->data_accepted = 0; - - if (chunk->asoc) - sock_rps_save_rxhash(chunk->asoc->base.sk, chunk->skb); + chunk->pdiscard = 0; + chunk->auth = 0; + chunk->has_asconf = 0; + chunk->end_of_packet = 0; + chunk->ecn_ce_done = 0; } chunk->chunk_hdr = ch; diff --git a/net/sctp/offload.c b/net/sctp/offload.c new file mode 100644 index 000000000000..a37887b373a7 --- /dev/null +++ b/net/sctp/offload.c @@ -0,0 +1,98 @@ +/* + * sctp_offload - GRO/GSO Offloading for SCTP + * + * Copyright (C) 2015, Marcelo Ricardo Leitner + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +static __le32 sctp_gso_make_checksum(struct sk_buff *skb) +{ + skb->ip_summed = CHECKSUM_NONE; + return sctp_compute_cksum(skb, skb_transport_offset(skb)); +} + +static struct sk_buff *sctp_gso_segment(struct sk_buff *skb, + netdev_features_t features) +{ + struct sk_buff *segs = ERR_PTR(-EINVAL); + struct sctphdr *sh; + + sh = sctp_hdr(skb); + if (!pskb_may_pull(skb, sizeof(*sh))) + goto out; + + __skb_pull(skb, sizeof(*sh)); + + if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST)) { + /* Packet is from an untrusted source, reset gso_segs. */ + struct skb_shared_info *pinfo = skb_shinfo(skb); + struct sk_buff *frag_iter; + + pinfo->gso_segs = 0; + if (skb->len != skb->data_len) { + /* Means we have chunks in here too */ + pinfo->gso_segs++; + } + + skb_walk_frags(skb, frag_iter) + pinfo->gso_segs++; + + segs = NULL; + goto out; + } + + segs = skb_segment(skb, features | NETIF_F_HW_CSUM); + if (IS_ERR(segs)) + goto out; + + /* All that is left is update SCTP CRC if necessary */ + if (!(features & NETIF_F_SCTP_CRC)) { + for (skb = segs; skb; skb = skb->next) { + if (skb->ip_summed == CHECKSUM_PARTIAL) { + sh = sctp_hdr(skb); + sh->checksum = sctp_gso_make_checksum(skb); + } + } + } + +out: + return segs; +} + +static const struct net_offload sctp_offload = { + .callbacks = { + .gso_segment = sctp_gso_segment, + }, +}; + +int __init sctp_offload_init(void) +{ + return inet_add_offload(&sctp_offload, IPPROTO_SCTP); +} diff --git a/net/sctp/output.c b/net/sctp/output.c index 9844fe573029..60499a69179d 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -84,18 +84,42 @@ static void sctp_packet_reset(struct sctp_packet *packet) struct sctp_packet *sctp_packet_config(struct sctp_packet *packet, __u32 vtag, int ecn_capable) { - struct sctp_chunk *chunk = NULL; + struct sctp_transport *tp = packet->transport; + struct sctp_association *asoc = tp->asoc; pr_debug("%s: packet:%p vtag:0x%x\n", __func__, packet, vtag); packet->vtag = vtag; + if (asoc && tp->dst) { + struct sock *sk = asoc->base.sk; + + rcu_read_lock(); + if (__sk_dst_get(sk) != tp->dst) { + dst_hold(tp->dst); + sk_setup_caps(sk, tp->dst); + } + + if (sk_can_gso(sk)) { + struct net_device *dev = tp->dst->dev; + + packet->max_size = dev->gso_max_size; + } else { + packet->max_size = asoc->pathmtu; + } + rcu_read_unlock(); + + } else { + packet->max_size = tp->pathmtu; + } + if (ecn_capable && sctp_packet_empty(packet)) { - chunk = sctp_get_ecne_prepend(packet->transport->asoc); + struct sctp_chunk *chunk; /* If there a is a prepend chunk stick it on the list before * any other chunks get appended. */ + chunk = sctp_get_ecne_prepend(asoc); if (chunk) sctp_packet_append_chunk(packet, chunk); } @@ -381,12 +405,15 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) struct sctp_transport *tp = packet->transport; struct sctp_association *asoc = tp->asoc; struct sctphdr *sh; - struct sk_buff *nskb; + struct sk_buff *nskb = NULL, *head = NULL; struct sctp_chunk *chunk, *tmp; struct sock *sk; int err = 0; int padding; /* How much padding do we need? */ + int pkt_size; __u8 has_data = 0; + int gso = 0; + int pktcount = 0; struct dst_entry *dst; unsigned char *auth = NULL; /* pointer to auth in skb data */ @@ -400,18 +427,37 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) chunk = list_entry(packet->chunk_list.next, struct sctp_chunk, list); sk = chunk->skb->sk; - /* Allocate the new skb. */ - nskb = alloc_skb(packet->size + MAX_HEADER, gfp); - if (!nskb) + /* Allocate the head skb, or main one if not in GSO */ + if (packet->size > tp->pathmtu && !packet->ipfragok) { + if (sk_can_gso(sk)) { + gso = 1; + pkt_size = packet->overhead; + } else { + /* If this happens, we trash this packet and try + * to build a new one, hopefully correct this + * time. Application may notice this error. + */ + pr_err_once("Trying to GSO but underlying device doesn't support it."); + goto nomem; + } + } else { + pkt_size = packet->size; + } + head = alloc_skb(pkt_size + MAX_HEADER, gfp); + if (!head) goto nomem; + if (gso) { + NAPI_GRO_CB(head)->last = head; + skb_shinfo(head)->gso_type = sk->sk_gso_type; + } /* Make sure the outbound skb has enough header room reserved. */ - skb_reserve(nskb, packet->overhead + MAX_HEADER); + skb_reserve(head, packet->overhead + MAX_HEADER); /* Set the owning socket so that we know where to get the * destination IP address. */ - sctp_packet_set_owner_w(nskb, sk); + sctp_packet_set_owner_w(head, sk); if (!sctp_transport_dst_check(tp)) { sctp_transport_route(tp, NULL, sctp_sk(sk)); @@ -422,11 +468,11 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) dst = dst_clone(tp->dst); if (!dst) goto no_route; - skb_dst_set(nskb, dst); + skb_dst_set(head, dst); /* Build the SCTP header. */ - sh = (struct sctphdr *)skb_push(nskb, sizeof(struct sctphdr)); - skb_reset_transport_header(nskb); + sh = (struct sctphdr *)skb_push(head, sizeof(struct sctphdr)); + skb_reset_transport_header(head); sh->source = htons(packet->source_port); sh->dest = htons(packet->destination_port); @@ -441,90 +487,133 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) sh->vtag = htonl(packet->vtag); sh->checksum = 0; - /** - * 6.10 Bundling - * - * An endpoint bundles chunks by simply including multiple - * chunks in one outbound SCTP packet. ... - */ - - /** - * 3.2 Chunk Field Descriptions - * - * The total length of a chunk (including Type, Length and - * Value fields) MUST be a multiple of 4 bytes. If the length - * of the chunk is not a multiple of 4 bytes, the sender MUST - * pad the chunk with all zero bytes and this padding is not - * included in the chunk length field. The sender should - * never pad with more than 3 bytes. - * - * [This whole comment explains WORD_ROUND() below.] - */ - pr_debug("***sctp_transmit_packet***\n"); - list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) { - list_del_init(&chunk->list); - if (sctp_chunk_is_data(chunk)) { - /* 6.3.1 C4) When data is in flight and when allowed - * by rule C5, a new RTT measurement MUST be made each - * round trip. Furthermore, new RTT measurements - * SHOULD be made no more than once per round-trip - * for a given destination transport address. - */ + do { + /* Set up convenience variables... */ + chunk = list_entry(packet->chunk_list.next, struct sctp_chunk, list); + pktcount++; - if (!chunk->resent && !tp->rto_pending) { - chunk->rtt_in_progress = 1; - tp->rto_pending = 1; + /* Calculate packet size, so it fits in PMTU. Leave + * other chunks for the next packets. + */ + if (gso) { + pkt_size = packet->overhead; + list_for_each_entry(chunk, &packet->chunk_list, list) { + int padded = WORD_ROUND(chunk->skb->len); + + if (pkt_size + padded > tp->pathmtu) + break; + pkt_size += padded; } - has_data = 1; - } + /* Allocate a new skb. */ + nskb = alloc_skb(pkt_size + MAX_HEADER, gfp); + if (!nskb) + goto nomem; - padding = WORD_ROUND(chunk->skb->len) - chunk->skb->len; - if (padding) - memset(skb_put(chunk->skb, padding), 0, padding); + /* Make sure the outbound skb has enough header + * room reserved. + */ + skb_reserve(nskb, packet->overhead + MAX_HEADER); + } else { + nskb = head; + } - /* if this is the auth chunk that we are adding, - * store pointer where it will be added and put - * the auth into the packet. + /** + * 3.2 Chunk Field Descriptions + * + * The total length of a chunk (including Type, Length and + * Value fields) MUST be a multiple of 4 bytes. If the length + * of the chunk is not a multiple of 4 bytes, the sender MUST + * pad the chunk with all zero bytes and this padding is not + * included in the chunk length field. The sender should + * never pad with more than 3 bytes. + * + * [This whole comment explains WORD_ROUND() below.] */ - if (chunk == packet->auth) - auth = skb_tail_pointer(nskb); - memcpy(skb_put(nskb, chunk->skb->len), + pkt_size -= packet->overhead; + list_for_each_entry_safe(chunk, tmp, &packet->chunk_list, list) { + list_del_init(&chunk->list); + if (sctp_chunk_is_data(chunk)) { + /* 6.3.1 C4) When data is in flight and when allowed + * by rule C5, a new RTT measurement MUST be made each + * round trip. Furthermore, new RTT measurements + * SHOULD be made no more than once per round-trip + * for a given destination transport address. + */ + + if (!chunk->resent && !tp->rto_pending) { + chunk->rtt_in_progress = 1; + tp->rto_pending = 1; + } + + has_data = 1; + } + + padding = WORD_ROUND(chunk->skb->len) - chunk->skb->len; + if (padding) + memset(skb_put(chunk->skb, padding), 0, padding); + + /* if this is the auth chunk that we are adding, + * store pointer where it will be added and put + * the auth into the packet. + */ + if (chunk == packet->auth) + auth = skb_tail_pointer(nskb); + + memcpy(skb_put(nskb, chunk->skb->len), chunk->skb->data, chunk->skb->len); - pr_debug("*** Chunk:%p[%s] %s 0x%x, length:%d, chunk->skb->len:%d, " - "rtt_in_progress:%d\n", chunk, - sctp_cname(SCTP_ST_CHUNK(chunk->chunk_hdr->type)), - chunk->has_tsn ? "TSN" : "No TSN", - chunk->has_tsn ? ntohl(chunk->subh.data_hdr->tsn) : 0, - ntohs(chunk->chunk_hdr->length), chunk->skb->len, - chunk->rtt_in_progress); - - /* - * If this is a control chunk, this is our last - * reference. Free data chunks after they've been - * acknowledged or have failed. - */ - if (!sctp_chunk_is_data(chunk)) - sctp_chunk_free(chunk); - } + pr_debug("*** Chunk:%p[%s] %s 0x%x, length:%d, chunk->skb->len:%d, rtt_in_progress:%d\n", + chunk, + sctp_cname(SCTP_ST_CHUNK(chunk->chunk_hdr->type)), + chunk->has_tsn ? "TSN" : "No TSN", + chunk->has_tsn ? ntohl(chunk->subh.data_hdr->tsn) : 0, + ntohs(chunk->chunk_hdr->length), chunk->skb->len, + chunk->rtt_in_progress); + + /* If this is a control chunk, this is our last + * reference. Free data chunks after they've been + * acknowledged or have failed. + * Re-queue auth chunks if needed. + */ + pkt_size -= WORD_ROUND(chunk->skb->len); - /* SCTP-AUTH, Section 6.2 - * The sender MUST calculate the MAC as described in RFC2104 [2] - * using the hash function H as described by the MAC Identifier and - * the shared association key K based on the endpoint pair shared key - * described by the shared key identifier. The 'data' used for the - * computation of the AUTH-chunk is given by the AUTH chunk with its - * HMAC field set to zero (as shown in Figure 6) followed by all - * chunks that are placed after the AUTH chunk in the SCTP packet. - */ - if (auth) - sctp_auth_calculate_hmac(asoc, nskb, - (struct sctp_auth_chunk *)auth, - gfp); + if (chunk == packet->auth && !list_empty(&packet->chunk_list)) + list_add(&chunk->list, &packet->chunk_list); + else if (!sctp_chunk_is_data(chunk)) + sctp_chunk_free(chunk); + + if (!pkt_size) + break; + } + + /* SCTP-AUTH, Section 6.2 + * The sender MUST calculate the MAC as described in RFC2104 [2] + * using the hash function H as described by the MAC Identifier and + * the shared association key K based on the endpoint pair shared key + * described by the shared key identifier. The 'data' used for the + * computation of the AUTH-chunk is given by the AUTH chunk with its + * HMAC field set to zero (as shown in Figure 6) followed by all + * chunks that are placed after the AUTH chunk in the SCTP packet. + */ + if (auth) + sctp_auth_calculate_hmac(asoc, nskb, + (struct sctp_auth_chunk *)auth, + gfp); + + if (!gso) + break; + + if (skb_gro_receive(&head, nskb)) + goto nomem; + nskb = NULL; + if (WARN_ON_ONCE(skb_shinfo(head)->gso_segs >= + sk->sk_gso_max_segs)) + goto nomem; + } while (!list_empty(&packet->chunk_list)); /* 2) Calculate the Adler-32 checksum of the whole packet, * including the SCTP common header and all the @@ -532,16 +621,18 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) * * Note: Adler-32 is no longer applicable, as has been replaced * by CRC32-C as described in . + * + * If it's a GSO packet, it's postponed to sctp_skb_segment. */ - if (!sctp_checksum_disable) { - if (!(dst->dev->features & NETIF_F_SCTP_CRC) || - (dst_xfrm(dst) != NULL) || packet->ipfragok) { - sh->checksum = sctp_compute_cksum(nskb, 0); + if (!sctp_checksum_disable || gso) { + if (!gso && (!(dst->dev->features & NETIF_F_SCTP_CRC) || + dst_xfrm(dst) || packet->ipfragok)) { + sh->checksum = sctp_compute_cksum(head, 0); } else { /* no need to seed pseudo checksum for SCTP */ - nskb->ip_summed = CHECKSUM_PARTIAL; - nskb->csum_start = skb_transport_header(nskb) - nskb->head; - nskb->csum_offset = offsetof(struct sctphdr, checksum); + head->ip_summed = CHECKSUM_PARTIAL; + head->csum_start = skb_transport_header(head) - head->head; + head->csum_offset = offsetof(struct sctphdr, checksum); } } @@ -557,7 +648,7 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) * Note: The works for IPv6 layer checks this bit too later * in transmission. See IP6_ECN_flow_xmit(). */ - tp->af_specific->ecn_capable(nskb->sk); + tp->af_specific->ecn_capable(sk); /* Set up the IP options. */ /* BUG: not implemented @@ -566,7 +657,7 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) /* Dump that on IP! */ if (asoc) { - asoc->stats.opackets++; + asoc->stats.opackets += pktcount; if (asoc->peer.last_sent_to != tp) /* Considering the multiple CPU scenario, this is a * "correcter" place for last_sent_to. --xguo @@ -589,16 +680,36 @@ int sctp_packet_transmit(struct sctp_packet *packet, gfp_t gfp) } } - pr_debug("***sctp_transmit_packet*** skb->len:%d\n", nskb->len); + pr_debug("***sctp_transmit_packet*** skb->len:%d\n", head->len); + + if (gso) { + /* Cleanup our debris for IP stacks */ + memset(head->cb, 0, max(sizeof(struct inet_skb_parm), + sizeof(struct inet6_skb_parm))); - nskb->ignore_df = packet->ipfragok; - tp->af_specific->sctp_xmit(nskb, tp); + skb_shinfo(head)->gso_segs = pktcount; + skb_shinfo(head)->gso_size = GSO_BY_FRAGS; + + /* We have to refresh this in case we are xmiting to + * more than one transport at a time + */ + rcu_read_lock(); + if (__sk_dst_get(sk) != tp->dst) { + dst_hold(tp->dst); + sk_setup_caps(sk, tp->dst); + } + rcu_read_unlock(); + } + head->ignore_df = packet->ipfragok; + tp->af_specific->sctp_xmit(head, tp); out: sctp_packet_reset(packet); return err; no_route: - kfree_skb(nskb); + kfree_skb(head); + if (nskb != head) + kfree_skb(nskb); if (asoc) IP_INC_STATS(sock_net(asoc->base.sk), IPSTATS_MIB_OUTNOROUTES); @@ -751,39 +862,63 @@ static sctp_xmit_t sctp_packet_will_fit(struct sctp_packet *packet, struct sctp_chunk *chunk, u16 chunk_len) { - size_t psize; - size_t pmtu; - int too_big; + size_t psize, pmtu; sctp_xmit_t retval = SCTP_XMIT_OK; psize = packet->size; - pmtu = ((packet->transport->asoc) ? - (packet->transport->asoc->pathmtu) : - (packet->transport->pathmtu)); - - too_big = (psize + chunk_len > pmtu); + if (packet->transport->asoc) + pmtu = packet->transport->asoc->pathmtu; + else + pmtu = packet->transport->pathmtu; /* Decide if we need to fragment or resubmit later. */ - if (too_big) { - /* It's OK to fragmet at IP level if any one of the following + if (psize + chunk_len > pmtu) { + /* It's OK to fragment at IP level if any one of the following * is true: - * 1. The packet is empty (meaning this chunk is greater - * the MTU) - * 2. The chunk we are adding is a control chunk - * 3. The packet doesn't have any data in it yet and data - * requires authentication. + * 1. The packet is empty (meaning this chunk is greater + * the MTU) + * 2. The packet doesn't have any data in it yet and data + * requires authentication. */ - if (sctp_packet_empty(packet) || !sctp_chunk_is_data(chunk) || + if (sctp_packet_empty(packet) || (!packet->has_data && chunk->auth)) { /* We no longer do re-fragmentation. * Just fragment at the IP layer, if we * actually hit this condition */ packet->ipfragok = 1; - } else { - retval = SCTP_XMIT_PMTU_FULL; + goto out; } + + /* It is also okay to fragment if the chunk we are + * adding is a control chunk, but only if current packet + * is not a GSO one otherwise it causes fragmentation of + * a large frame. So in this case we allow the + * fragmentation by forcing it to be in a new packet. + */ + if (!sctp_chunk_is_data(chunk) && packet->has_data) + retval = SCTP_XMIT_PMTU_FULL; + + if (psize + chunk_len > packet->max_size) + /* Hit GSO/PMTU limit, gotta flush */ + retval = SCTP_XMIT_PMTU_FULL; + + if (!packet->transport->burst_limited && + psize + chunk_len > (packet->transport->cwnd >> 1)) + /* Do not allow a single GSO packet to use more + * than half of cwnd. + */ + retval = SCTP_XMIT_PMTU_FULL; + + if (packet->transport->burst_limited && + psize + chunk_len > (packet->transport->burst_limited >> 1)) + /* Do not allow a single GSO packet to use more + * than half of original cwnd. + */ + retval = SCTP_XMIT_PMTU_FULL; + /* Otherwise it will fit in the GSO packet */ } +out: return retval; } diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c index d3d50daa248b..40022ee885d7 100644 --- a/net/sctp/protocol.c +++ b/net/sctp/protocol.c @@ -1516,6 +1516,9 @@ static __init int sctp_init(void) if (status) goto err_v6_add_protocol; + if (sctp_offload_init() < 0) + pr_crit("%s: Cannot add SCTP protocol offload\n", __func__); + out: return status; err_v6_add_protocol: diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 67154b848aa9..712fb2339baa 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -4003,6 +4003,8 @@ static int sctp_init_sock(struct sock *sk) return -ESOCKTNOSUPPORT; } + sk->sk_gso_type = SKB_GSO_SCTP; + /* Initialize default send parameters. These parameters can be * modified with the SCTP_DEFAULT_SEND_PARAM socket option. */ -- cgit v1.2.3 From a91eb52abb504a1dd3248a5d07b54e7f95d5fcf1 Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Fri, 3 Jun 2016 14:35:32 +0300 Subject: qed: Revisit chain implementation RoCE driver is going to need a 32-bit chain [current chain implementation for qed* currently supports only 16-bit producer/consumer chains]. This patch adds said support, as well as doing other slight tweaks and modifications to qed's chain API. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_dev.c | 314 +++++++++--- drivers/net/ethernet/qlogic/qed/qed_dev_api.h | 8 +- drivers/net/ethernet/qlogic/qed/qed_sp_commands.c | 5 +- drivers/net/ethernet/qlogic/qed/qed_spq.c | 40 +- drivers/net/ethernet/qlogic/qede/qede_main.c | 6 +- include/linux/qed/qed_chain.h | 553 +++++++++++++--------- include/linux/qed/qed_if.h | 3 +- 7 files changed, 623 insertions(+), 306 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index e9ce6a7bc63b..151173d4a926 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -1779,92 +1780,285 @@ void qed_hw_remove(struct qed_dev *cdev) qed_iov_free_hw_info(cdev); } -int qed_chain_alloc(struct qed_dev *cdev, - enum qed_chain_use_mode intended_use, - enum qed_chain_mode mode, - u16 num_elems, - size_t elem_size, - struct qed_chain *p_chain) +static void qed_chain_free_next_ptr(struct qed_dev *cdev, + struct qed_chain *p_chain) +{ + void *p_virt = p_chain->p_virt_addr, *p_virt_next = NULL; + dma_addr_t p_phys = p_chain->p_phys_addr, p_phys_next = 0; + struct qed_chain_next *p_next; + u32 size, i; + + if (!p_virt) + return; + + size = p_chain->elem_size * p_chain->usable_per_page; + + for (i = 0; i < p_chain->page_cnt; i++) { + if (!p_virt) + break; + + p_next = (struct qed_chain_next *)((u8 *)p_virt + size); + p_virt_next = p_next->next_virt; + p_phys_next = HILO_DMA_REGPAIR(p_next->next_phys); + + dma_free_coherent(&cdev->pdev->dev, + QED_CHAIN_PAGE_SIZE, p_virt, p_phys); + + p_virt = p_virt_next; + p_phys = p_phys_next; + } +} + +static void qed_chain_free_single(struct qed_dev *cdev, + struct qed_chain *p_chain) +{ + if (!p_chain->p_virt_addr) + return; + + dma_free_coherent(&cdev->pdev->dev, + QED_CHAIN_PAGE_SIZE, + p_chain->p_virt_addr, p_chain->p_phys_addr); +} + +static void qed_chain_free_pbl(struct qed_dev *cdev, struct qed_chain *p_chain) +{ + void **pp_virt_addr_tbl = p_chain->pbl.pp_virt_addr_tbl; + u32 page_cnt = p_chain->page_cnt, i, pbl_size; + u8 *p_pbl_virt = p_chain->pbl.p_virt_table; + + if (!pp_virt_addr_tbl) + return; + + if (!p_chain->pbl.p_virt_table) + goto out; + + for (i = 0; i < page_cnt; i++) { + if (!pp_virt_addr_tbl[i]) + break; + + dma_free_coherent(&cdev->pdev->dev, + QED_CHAIN_PAGE_SIZE, + pp_virt_addr_tbl[i], + *(dma_addr_t *)p_pbl_virt); + + p_pbl_virt += QED_CHAIN_PBL_ENTRY_SIZE; + } + + pbl_size = page_cnt * QED_CHAIN_PBL_ENTRY_SIZE; + dma_free_coherent(&cdev->pdev->dev, + pbl_size, + p_chain->pbl.p_virt_table, p_chain->pbl.p_phys_table); +out: + vfree(p_chain->pbl.pp_virt_addr_tbl); +} + +void qed_chain_free(struct qed_dev *cdev, struct qed_chain *p_chain) +{ + switch (p_chain->mode) { + case QED_CHAIN_MODE_NEXT_PTR: + qed_chain_free_next_ptr(cdev, p_chain); + break; + case QED_CHAIN_MODE_SINGLE: + qed_chain_free_single(cdev, p_chain); + break; + case QED_CHAIN_MODE_PBL: + qed_chain_free_pbl(cdev, p_chain); + break; + } +} + +static int +qed_chain_alloc_sanity_check(struct qed_dev *cdev, + enum qed_chain_cnt_type cnt_type, + size_t elem_size, u32 page_cnt) +{ + u64 chain_size = ELEMS_PER_PAGE(elem_size) * page_cnt; + + /* The actual chain size can be larger than the maximal possible value + * after rounding up the requested elements number to pages, and after + * taking into acount the unusuable elements (next-ptr elements). + * The size of a "u16" chain can be (U16_MAX + 1) since the chain + * size/capacity fields are of a u32 type. + */ + if ((cnt_type == QED_CHAIN_CNT_TYPE_U16 && + chain_size > 0x10000) || + (cnt_type == QED_CHAIN_CNT_TYPE_U32 && + chain_size > 0x100000000ULL)) { + DP_NOTICE(cdev, + "The actual chain size (0x%llx) is larger than the maximal possible value\n", + chain_size); + return -EINVAL; + } + + return 0; +} + +static int +qed_chain_alloc_next_ptr(struct qed_dev *cdev, struct qed_chain *p_chain) { - dma_addr_t p_pbl_phys = 0; - void *p_pbl_virt = NULL; + void *p_virt = NULL, *p_virt_prev = NULL; dma_addr_t p_phys = 0; - void *p_virt = NULL; - u16 page_cnt = 0; - size_t size; + u32 i; - if (mode == QED_CHAIN_MODE_SINGLE) - page_cnt = 1; - else - page_cnt = QED_CHAIN_PAGE_CNT(num_elems, elem_size, mode); + for (i = 0; i < p_chain->page_cnt; i++) { + p_virt = dma_alloc_coherent(&cdev->pdev->dev, + QED_CHAIN_PAGE_SIZE, + &p_phys, GFP_KERNEL); + if (!p_virt) { + DP_NOTICE(cdev, "Failed to allocate chain memory\n"); + return -ENOMEM; + } + + if (i == 0) { + qed_chain_init_mem(p_chain, p_virt, p_phys); + qed_chain_reset(p_chain); + } else { + qed_chain_init_next_ptr_elem(p_chain, p_virt_prev, + p_virt, p_phys); + } + + p_virt_prev = p_virt; + } + /* Last page's next element should point to the beginning of the + * chain. + */ + qed_chain_init_next_ptr_elem(p_chain, p_virt_prev, + p_chain->p_virt_addr, + p_chain->p_phys_addr); + + return 0; +} + +static int +qed_chain_alloc_single(struct qed_dev *cdev, struct qed_chain *p_chain) +{ + dma_addr_t p_phys = 0; + void *p_virt = NULL; - size = page_cnt * QED_CHAIN_PAGE_SIZE; p_virt = dma_alloc_coherent(&cdev->pdev->dev, - size, &p_phys, GFP_KERNEL); + QED_CHAIN_PAGE_SIZE, &p_phys, GFP_KERNEL); if (!p_virt) { - DP_NOTICE(cdev, "Failed to allocate chain mem\n"); - goto nomem; + DP_NOTICE(cdev, "Failed to allocate chain memory\n"); + return -ENOMEM; } - if (mode == QED_CHAIN_MODE_PBL) { - size = page_cnt * QED_CHAIN_PBL_ENTRY_SIZE; - p_pbl_virt = dma_alloc_coherent(&cdev->pdev->dev, - size, &p_pbl_phys, - GFP_KERNEL); - if (!p_pbl_virt) { - DP_NOTICE(cdev, "Failed to allocate chain pbl mem\n"); - goto nomem; - } + qed_chain_init_mem(p_chain, p_virt, p_phys); + qed_chain_reset(p_chain); - qed_chain_pbl_init(p_chain, p_virt, p_phys, page_cnt, - (u8)elem_size, intended_use, - p_pbl_phys, p_pbl_virt); - } else { - qed_chain_init(p_chain, p_virt, p_phys, page_cnt, - (u8)elem_size, intended_use, mode); + return 0; +} + +static int qed_chain_alloc_pbl(struct qed_dev *cdev, struct qed_chain *p_chain) +{ + u32 page_cnt = p_chain->page_cnt, size, i; + dma_addr_t p_phys = 0, p_pbl_phys = 0; + void **pp_virt_addr_tbl = NULL; + u8 *p_pbl_virt = NULL; + void *p_virt = NULL; + + size = page_cnt * sizeof(*pp_virt_addr_tbl); + pp_virt_addr_tbl = vmalloc(size); + if (!pp_virt_addr_tbl) { + DP_NOTICE(cdev, + "Failed to allocate memory for the chain virtual addresses table\n"); + return -ENOMEM; } + memset(pp_virt_addr_tbl, 0, size); - return 0; + /* The allocation of the PBL table is done with its full size, since it + * is expected to be successive. + * qed_chain_init_pbl_mem() is called even in a case of an allocation + * failure, since pp_virt_addr_tbl was previously allocated, and it + * should be saved to allow its freeing during the error flow. + */ + size = page_cnt * QED_CHAIN_PBL_ENTRY_SIZE; + p_pbl_virt = dma_alloc_coherent(&cdev->pdev->dev, + size, &p_pbl_phys, GFP_KERNEL); + qed_chain_init_pbl_mem(p_chain, p_pbl_virt, p_pbl_phys, + pp_virt_addr_tbl); + if (!p_pbl_virt) { + DP_NOTICE(cdev, "Failed to allocate chain pbl memory\n"); + return -ENOMEM; + } -nomem: - dma_free_coherent(&cdev->pdev->dev, - page_cnt * QED_CHAIN_PAGE_SIZE, - p_virt, p_phys); - dma_free_coherent(&cdev->pdev->dev, - page_cnt * QED_CHAIN_PBL_ENTRY_SIZE, - p_pbl_virt, p_pbl_phys); + for (i = 0; i < page_cnt; i++) { + p_virt = dma_alloc_coherent(&cdev->pdev->dev, + QED_CHAIN_PAGE_SIZE, + &p_phys, GFP_KERNEL); + if (!p_virt) { + DP_NOTICE(cdev, "Failed to allocate chain memory\n"); + return -ENOMEM; + } - return -ENOMEM; + if (i == 0) { + qed_chain_init_mem(p_chain, p_virt, p_phys); + qed_chain_reset(p_chain); + } + + /* Fill the PBL table with the physical address of the page */ + *(dma_addr_t *)p_pbl_virt = p_phys; + /* Keep the virtual address of the page */ + p_chain->pbl.pp_virt_addr_tbl[i] = p_virt; + + p_pbl_virt += QED_CHAIN_PBL_ENTRY_SIZE; + } + + return 0; } -void qed_chain_free(struct qed_dev *cdev, - struct qed_chain *p_chain) +int qed_chain_alloc(struct qed_dev *cdev, + enum qed_chain_use_mode intended_use, + enum qed_chain_mode mode, + enum qed_chain_cnt_type cnt_type, + u32 num_elems, size_t elem_size, struct qed_chain *p_chain) { - size_t size; + u32 page_cnt; + int rc = 0; - if (!p_chain->p_virt_addr) - return; + if (mode == QED_CHAIN_MODE_SINGLE) + page_cnt = 1; + else + page_cnt = QED_CHAIN_PAGE_CNT(num_elems, elem_size, mode); - if (p_chain->mode == QED_CHAIN_MODE_PBL) { - size = p_chain->page_cnt * QED_CHAIN_PBL_ENTRY_SIZE; - dma_free_coherent(&cdev->pdev->dev, size, - p_chain->pbl.p_virt_table, - p_chain->pbl.p_phys_table); + rc = qed_chain_alloc_sanity_check(cdev, cnt_type, elem_size, page_cnt); + if (rc) { + DP_NOTICE(cdev, + "Cannot allocate a chain with the given arguments:\n" + "[use_mode %d, mode %d, cnt_type %d, num_elems %d, elem_size %zu]\n", + intended_use, mode, cnt_type, num_elems, elem_size); + return rc; } - size = p_chain->page_cnt * QED_CHAIN_PAGE_SIZE; - dma_free_coherent(&cdev->pdev->dev, size, - p_chain->p_virt_addr, - p_chain->p_phys_addr); + qed_chain_init_params(p_chain, page_cnt, (u8) elem_size, intended_use, + mode, cnt_type); + + switch (mode) { + case QED_CHAIN_MODE_NEXT_PTR: + rc = qed_chain_alloc_next_ptr(cdev, p_chain); + break; + case QED_CHAIN_MODE_SINGLE: + rc = qed_chain_alloc_single(cdev, p_chain); + break; + case QED_CHAIN_MODE_PBL: + rc = qed_chain_alloc_pbl(cdev, p_chain); + break; + } + if (rc) + goto nomem; + + return 0; + +nomem: + qed_chain_free(cdev, p_chain); + return rc; } -int qed_fw_l2_queue(struct qed_hwfn *p_hwfn, - u16 src_id, u16 *dst_id) +int qed_fw_l2_queue(struct qed_hwfn *p_hwfn, u16 src_id, u16 *dst_id) { if (src_id >= RESC_NUM(p_hwfn, QED_L2_QUEUE)) { u16 min, max; - min = (u16)RESC_START(p_hwfn, QED_L2_QUEUE); + min = (u16) RESC_START(p_hwfn, QED_L2_QUEUE); max = min + RESC_NUM(p_hwfn, QED_L2_QUEUE); DP_NOTICE(p_hwfn, "l2_queue id [%d] is not valid, available indices [%d - %d]\n", diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h index dde364d6f502..f810ce45d463 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h +++ b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h @@ -245,9 +245,8 @@ int qed_chain_alloc(struct qed_dev *cdev, enum qed_chain_use_mode intended_use, enum qed_chain_mode mode, - u16 num_elems, - size_t elem_size, - struct qed_chain *p_chain); + enum qed_chain_cnt_type cnt_type, + u32 num_elems, size_t elem_size, struct qed_chain *p_chain); /** * @brief qed_chain_free - Free chain DMA memory @@ -255,8 +254,7 @@ qed_chain_alloc(struct qed_dev *cdev, * @param p_hwfn * @param p_chain */ -void qed_chain_free(struct qed_dev *cdev, - struct qed_chain *p_chain); +void qed_chain_free(struct qed_dev *cdev, struct qed_chain *p_chain); /** * @@brief qed_fw_l2_queue - Get absolute L2 queue ID diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c index 122506416478..63b93fbdc66a 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c @@ -308,6 +308,7 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn, struct qed_spq_entry *p_ent = NULL; struct qed_sp_init_data init_data; int rc = -EINVAL; + u8 page_cnt; /* update initial eq producer */ qed_eq_prod_update(p_hwfn, @@ -350,8 +351,8 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn, /* Place EQ address in RAMROD */ DMA_REGPAIR_LE(p_ramrod->event_ring_pbl_addr, p_hwfn->p_eq->chain.pbl.p_phys_table); - p_ramrod->event_ring_num_pages = (u8)p_hwfn->p_eq->chain.page_cnt; - + page_cnt = (u8)qed_chain_get_page_cnt(&p_hwfn->p_eq->chain); + p_ramrod->event_ring_num_pages = page_cnt; DMA_REGPAIR_LE(p_ramrod->consolid_q_pbl_addr, p_hwfn->p_consq->chain.pbl.p_phys_table); diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c index acac6626a1b2..ad9bf5c85c3f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_spq.c +++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c @@ -343,6 +343,7 @@ struct qed_eq *qed_eq_alloc(struct qed_hwfn *p_hwfn, if (qed_chain_alloc(p_hwfn->cdev, QED_CHAIN_USE_TO_PRODUCE, QED_CHAIN_MODE_PBL, + QED_CHAIN_CNT_TYPE_U16, num_elem, sizeof(union event_ring_element), &p_eq->chain)) { @@ -416,10 +417,10 @@ int qed_eth_cqe_completion(struct qed_hwfn *p_hwfn, ***************************************************************************/ void qed_spq_setup(struct qed_hwfn *p_hwfn) { - struct qed_spq *p_spq = p_hwfn->p_spq; - struct qed_spq_entry *p_virt = NULL; - dma_addr_t p_phys = 0; - unsigned int i = 0; + struct qed_spq *p_spq = p_hwfn->p_spq; + struct qed_spq_entry *p_virt = NULL; + dma_addr_t p_phys = 0; + u32 i, capacity; INIT_LIST_HEAD(&p_spq->pending); INIT_LIST_HEAD(&p_spq->completion_pending); @@ -431,7 +432,8 @@ void qed_spq_setup(struct qed_hwfn *p_hwfn) p_phys = p_spq->p_phys + offsetof(struct qed_spq_entry, ramrod); p_virt = p_spq->p_virt; - for (i = 0; i < p_spq->chain.capacity; i++) { + capacity = qed_chain_get_capacity(&p_spq->chain); + for (i = 0; i < capacity; i++) { DMA_REGPAIR_LE(p_virt->elem.data_ptr, p_phys); list_add_tail(&p_virt->list, &p_spq->free_pool); @@ -459,9 +461,10 @@ void qed_spq_setup(struct qed_hwfn *p_hwfn) int qed_spq_alloc(struct qed_hwfn *p_hwfn) { - struct qed_spq *p_spq = NULL; - dma_addr_t p_phys = 0; - struct qed_spq_entry *p_virt = NULL; + struct qed_spq_entry *p_virt = NULL; + struct qed_spq *p_spq = NULL; + dma_addr_t p_phys = 0; + u32 capacity; /* SPQ struct */ p_spq = @@ -475,6 +478,7 @@ int qed_spq_alloc(struct qed_hwfn *p_hwfn) if (qed_chain_alloc(p_hwfn->cdev, QED_CHAIN_USE_TO_PRODUCE, QED_CHAIN_MODE_SINGLE, + QED_CHAIN_CNT_TYPE_U16, 0, /* N/A when the mode is SINGLE */ sizeof(struct slow_path_element), &p_spq->chain)) { @@ -483,11 +487,11 @@ int qed_spq_alloc(struct qed_hwfn *p_hwfn) } /* allocate and fill the SPQ elements (incl. ramrod data list) */ + capacity = qed_chain_get_capacity(&p_spq->chain); p_virt = dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, - p_spq->chain.capacity * + capacity * sizeof(struct qed_spq_entry), - &p_phys, - GFP_KERNEL); + &p_phys, GFP_KERNEL); if (!p_virt) goto spq_allocate_fail; @@ -507,16 +511,18 @@ spq_allocate_fail: void qed_spq_free(struct qed_hwfn *p_hwfn) { struct qed_spq *p_spq = p_hwfn->p_spq; + u32 capacity; if (!p_spq) return; - if (p_spq->p_virt) + if (p_spq->p_virt) { + capacity = qed_chain_get_capacity(&p_spq->chain); dma_free_coherent(&p_hwfn->cdev->pdev->dev, - p_spq->chain.capacity * + capacity * sizeof(struct qed_spq_entry), - p_spq->p_virt, - p_spq->p_phys); + p_spq->p_virt, p_spq->p_phys); + } qed_chain_free(p_hwfn->cdev, &p_spq->chain); ; @@ -871,9 +877,9 @@ struct qed_consq *qed_consq_alloc(struct qed_hwfn *p_hwfn) if (qed_chain_alloc(p_hwfn->cdev, QED_CHAIN_USE_TO_PRODUCE, QED_CHAIN_MODE_PBL, + QED_CHAIN_CNT_TYPE_U16, QED_CHAIN_PAGE_SIZE / 0x80, - 0x80, - &p_consq->chain)) { + 0x80, &p_consq->chain)) { DP_NOTICE(p_hwfn, "Failed to allocate consq chain"); goto consq_allocate_fail; } diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 4f352476db31..edd2d908c52e 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -2817,6 +2817,7 @@ static int qede_alloc_mem_rxq(struct qede_dev *edev, rc = edev->ops->common->chain_alloc(edev->cdev, QED_CHAIN_USE_TO_CONSUME_PRODUCE, QED_CHAIN_MODE_NEXT_PTR, + QED_CHAIN_CNT_TYPE_U16, RX_RING_SIZE, sizeof(struct eth_rx_bd), &rxq->rx_bd_ring); @@ -2828,6 +2829,7 @@ static int qede_alloc_mem_rxq(struct qede_dev *edev, rc = edev->ops->common->chain_alloc(edev->cdev, QED_CHAIN_USE_TO_CONSUME, QED_CHAIN_MODE_PBL, + QED_CHAIN_CNT_TYPE_U16, RX_RING_SIZE, sizeof(union eth_rx_cqe), &rxq->rx_comp_ring); @@ -2879,9 +2881,9 @@ static int qede_alloc_mem_txq(struct qede_dev *edev, rc = edev->ops->common->chain_alloc(edev->cdev, QED_CHAIN_USE_TO_CONSUME_PRODUCE, QED_CHAIN_MODE_PBL, + QED_CHAIN_CNT_TYPE_U16, NUM_TX_BDS_MAX, - sizeof(*p_virt), - &txq->tx_pbl); + sizeof(*p_virt), &txq->tx_pbl); if (rc) goto err; diff --git a/include/linux/qed/qed_chain.h b/include/linux/qed/qed_chain.h index 5f8fcaaa6504..eceaa9ed2ae9 100644 --- a/include/linux/qed/qed_chain.h +++ b/include/linux/qed/qed_chain.h @@ -47,16 +47,56 @@ enum qed_chain_use_mode { QED_CHAIN_USE_TO_CONSUME_PRODUCE, /* Chain starts empty */ }; +enum qed_chain_cnt_type { + /* The chain's size/prod/cons are kept in 16-bit variables */ + QED_CHAIN_CNT_TYPE_U16, + + /* The chain's size/prod/cons are kept in 32-bit variables */ + QED_CHAIN_CNT_TYPE_U32, +}; + struct qed_chain_next { struct regpair next_phys; void *next_virt; }; +struct qed_chain_pbl_u16 { + u16 prod_page_idx; + u16 cons_page_idx; +}; + +struct qed_chain_pbl_u32 { + u32 prod_page_idx; + u32 cons_page_idx; +}; + struct qed_chain_pbl { + /* Base address of a pre-allocated buffer for pbl */ dma_addr_t p_phys_table; void *p_virt_table; - u16 prod_page_idx; - u16 cons_page_idx; + + /* Table for keeping the virtual addresses of the chain pages, + * respectively to the physical addresses in the pbl table. + */ + void **pp_virt_addr_tbl; + + /* Index to current used page by producer/consumer */ + union { + struct qed_chain_pbl_u16 pbl16; + struct qed_chain_pbl_u32 pbl32; + } u; +}; + +struct qed_chain_u16 { + /* Cyclic index of next element to produce/consme */ + u16 prod_idx; + u16 cons_idx; +}; + +struct qed_chain_u32 { + /* Cyclic index of next element to produce/consme */ + u32 prod_idx; + u32 cons_idx; }; struct qed_chain { @@ -64,13 +104,25 @@ struct qed_chain { dma_addr_t p_phys_addr; void *p_prod_elem; void *p_cons_elem; - u16 page_cnt; + enum qed_chain_mode mode; enum qed_chain_use_mode intended_use; /* used to produce/consume */ - u16 capacity; /*< number of _usable_ elements */ - u16 size; /* number of elements */ - u16 prod_idx; - u16 cons_idx; + enum qed_chain_cnt_type cnt_type; + + union { + struct qed_chain_u16 chain16; + struct qed_chain_u32 chain32; + } u; + + u32 page_cnt; + + /* Number of elements - capacity is for usable elements only, + * while size will contain total number of elements [for entire chain]. + */ + u32 capacity; + u32 size; + + /* Elements information for fast calculations */ u16 elem_per_page; u16 elem_per_page_mask; u16 elem_unusable; @@ -96,66 +148,69 @@ struct qed_chain { #define QED_CHAIN_PAGE_CNT(elem_cnt, elem_size, mode) \ DIV_ROUND_UP(elem_cnt, USABLE_ELEMS_PER_PAGE(elem_size, mode)) +#define is_chain_u16(p) ((p)->cnt_type == QED_CHAIN_CNT_TYPE_U16) +#define is_chain_u32(p) ((p)->cnt_type == QED_CHAIN_CNT_TYPE_U32) + /* Accessors */ static inline u16 qed_chain_get_prod_idx(struct qed_chain *p_chain) { - return p_chain->prod_idx; + return p_chain->u.chain16.prod_idx; } static inline u16 qed_chain_get_cons_idx(struct qed_chain *p_chain) { - return p_chain->cons_idx; + return p_chain->u.chain16.cons_idx; +} + +static inline u32 qed_chain_get_cons_idx_u32(struct qed_chain *p_chain) +{ + return p_chain->u.chain32.cons_idx; } static inline u16 qed_chain_get_elem_left(struct qed_chain *p_chain) { u16 used; - /* we don't need to trancate upon assignmet, as we assign u32->u16 */ - used = ((u32)0x10000u + (u32)(p_chain->prod_idx)) - - (u32)p_chain->cons_idx; + used = (u16) (((u32)0x10000 + + (u32)p_chain->u.chain16.prod_idx) - + (u32)p_chain->u.chain16.cons_idx); if (p_chain->mode == QED_CHAIN_MODE_NEXT_PTR) - used -= p_chain->prod_idx / p_chain->elem_per_page - - p_chain->cons_idx / p_chain->elem_per_page; + used -= p_chain->u.chain16.prod_idx / p_chain->elem_per_page - + p_chain->u.chain16.cons_idx / p_chain->elem_per_page; - return p_chain->capacity - used; + return (u16)(p_chain->capacity - used); } -static inline u8 qed_chain_is_full(struct qed_chain *p_chain) +static inline u32 qed_chain_get_elem_left_u32(struct qed_chain *p_chain) { - return qed_chain_get_elem_left(p_chain) == p_chain->capacity; -} + u32 used; -static inline u8 qed_chain_is_empty(struct qed_chain *p_chain) -{ - return qed_chain_get_elem_left(p_chain) == 0; -} + used = (u32) (((u64)0x100000000ULL + + (u64)p_chain->u.chain32.prod_idx) - + (u64)p_chain->u.chain32.cons_idx); + if (p_chain->mode == QED_CHAIN_MODE_NEXT_PTR) + used -= p_chain->u.chain32.prod_idx / p_chain->elem_per_page - + p_chain->u.chain32.cons_idx / p_chain->elem_per_page; -static inline u16 qed_chain_get_elem_per_page( - struct qed_chain *p_chain) -{ - return p_chain->elem_per_page; + return p_chain->capacity - used; } -static inline u16 qed_chain_get_usable_per_page( - struct qed_chain *p_chain) +static inline u16 qed_chain_get_usable_per_page(struct qed_chain *p_chain) { return p_chain->usable_per_page; } -static inline u16 qed_chain_get_unusable_per_page( - struct qed_chain *p_chain) +static inline u16 qed_chain_get_unusable_per_page(struct qed_chain *p_chain) { return p_chain->elem_unusable; } -static inline u16 qed_chain_get_size(struct qed_chain *p_chain) +static inline u32 qed_chain_get_page_cnt(struct qed_chain *p_chain) { - return p_chain->size; + return p_chain->page_cnt; } -static inline dma_addr_t -qed_chain_get_pbl_phys(struct qed_chain *p_chain) +static inline dma_addr_t qed_chain_get_pbl_phys(struct qed_chain *p_chain) { return p_chain->pbl.p_phys_table; } @@ -172,64 +227,62 @@ qed_chain_get_pbl_phys(struct qed_chain *p_chain) */ static inline void qed_chain_advance_page(struct qed_chain *p_chain, - void **p_next_elem, - u16 *idx_to_inc, - u16 *page_to_inc) + void **p_next_elem, void *idx_to_inc, void *page_to_inc) { + struct qed_chain_next *p_next = NULL; + u32 page_index = 0; switch (p_chain->mode) { case QED_CHAIN_MODE_NEXT_PTR: - { - struct qed_chain_next *p_next = *p_next_elem; + p_next = *p_next_elem; *p_next_elem = p_next->next_virt; - *idx_to_inc += p_chain->elem_unusable; + if (is_chain_u16(p_chain)) + *(u16 *)idx_to_inc += p_chain->elem_unusable; + else + *(u32 *)idx_to_inc += p_chain->elem_unusable; break; - } case QED_CHAIN_MODE_SINGLE: *p_next_elem = p_chain->p_virt_addr; break; case QED_CHAIN_MODE_PBL: - /* It is assumed pages are sequential, next element needs - * to change only when passing going back to first from last. - */ - if (++(*page_to_inc) == p_chain->page_cnt) { - *page_to_inc = 0; - *p_next_elem = p_chain->p_virt_addr; + if (is_chain_u16(p_chain)) { + if (++(*(u16 *)page_to_inc) == p_chain->page_cnt) + *(u16 *)page_to_inc = 0; + page_index = *(u16 *)page_to_inc; + } else { + if (++(*(u32 *)page_to_inc) == p_chain->page_cnt) + *(u32 *)page_to_inc = 0; + page_index = *(u32 *)page_to_inc; } + *p_next_elem = p_chain->pbl.pp_virt_addr_tbl[page_index]; } } #define is_unusable_idx(p, idx) \ - (((p)->idx & (p)->elem_per_page_mask) == (p)->usable_per_page) + (((p)->u.chain16.idx & (p)->elem_per_page_mask) == (p)->usable_per_page) + +#define is_unusable_idx_u32(p, idx) \ + (((p)->u.chain32.idx & (p)->elem_per_page_mask) == (p)->usable_per_page) +#define is_unusable_next_idx(p, idx) \ + ((((p)->u.chain16.idx + 1) & (p)->elem_per_page_mask) == \ + (p)->usable_per_page) -#define is_unusable_next_idx(p, idx) \ - ((((p)->idx + 1) & (p)->elem_per_page_mask) == (p)->usable_per_page) +#define is_unusable_next_idx_u32(p, idx) \ + ((((p)->u.chain32.idx + 1) & (p)->elem_per_page_mask) == \ + (p)->usable_per_page) -#define test_ans_skip(p, idx) \ +#define test_and_skip(p, idx) \ do { \ - if (is_unusable_idx(p, idx)) { \ - (p)->idx += (p)->elem_unusable; \ + if (is_chain_u16(p)) { \ + if (is_unusable_idx(p, idx)) \ + (p)->u.chain16.idx += (p)->elem_unusable; \ + } else { \ + if (is_unusable_idx_u32(p, idx)) \ + (p)->u.chain32.idx += (p)->elem_unusable; \ } \ } while (0) -/** - * @brief qed_chain_return_multi_produced - - * - * A chain in which the driver "Produces" elements should use this API - * to indicate previous produced elements are now consumed. - * - * @param p_chain - * @param num - */ -static inline void -qed_chain_return_multi_produced(struct qed_chain *p_chain, - u16 num) -{ - p_chain->cons_idx += num; - test_ans_skip(p_chain, cons_idx); -} - /** * @brief qed_chain_return_produced - * @@ -240,8 +293,11 @@ qed_chain_return_multi_produced(struct qed_chain *p_chain, */ static inline void qed_chain_return_produced(struct qed_chain *p_chain) { - p_chain->cons_idx++; - test_ans_skip(p_chain, cons_idx); + if (is_chain_u16(p_chain)) + p_chain->u.chain16.cons_idx++; + else + p_chain->u.chain32.cons_idx++; + test_and_skip(p_chain, cons_idx); } /** @@ -257,21 +313,33 @@ static inline void qed_chain_return_produced(struct qed_chain *p_chain) */ static inline void *qed_chain_produce(struct qed_chain *p_chain) { - void *ret = NULL; - - if ((p_chain->prod_idx & p_chain->elem_per_page_mask) == - p_chain->next_page_mask) { - qed_chain_advance_page(p_chain, &p_chain->p_prod_elem, - &p_chain->prod_idx, - &p_chain->pbl.prod_page_idx); + void *p_ret = NULL, *p_prod_idx, *p_prod_page_idx; + + if (is_chain_u16(p_chain)) { + if ((p_chain->u.chain16.prod_idx & + p_chain->elem_per_page_mask) == p_chain->next_page_mask) { + p_prod_idx = &p_chain->u.chain16.prod_idx; + p_prod_page_idx = &p_chain->pbl.u.pbl16.prod_page_idx; + qed_chain_advance_page(p_chain, &p_chain->p_prod_elem, + p_prod_idx, p_prod_page_idx); + } + p_chain->u.chain16.prod_idx++; + } else { + if ((p_chain->u.chain32.prod_idx & + p_chain->elem_per_page_mask) == p_chain->next_page_mask) { + p_prod_idx = &p_chain->u.chain32.prod_idx; + p_prod_page_idx = &p_chain->pbl.u.pbl32.prod_page_idx; + qed_chain_advance_page(p_chain, &p_chain->p_prod_elem, + p_prod_idx, p_prod_page_idx); + } + p_chain->u.chain32.prod_idx++; } - ret = p_chain->p_prod_elem; - p_chain->prod_idx++; + p_ret = p_chain->p_prod_elem; p_chain->p_prod_elem = (void *)(((u8 *)p_chain->p_prod_elem) + p_chain->elem_size); - return ret; + return p_ret; } /** @@ -282,9 +350,9 @@ static inline void *qed_chain_produce(struct qed_chain *p_chain) * @param p_chain * @param num * - * @return u16, number of unusable BDs + * @return number of unusable BDs */ -static inline u16 qed_chain_get_capacity(struct qed_chain *p_chain) +static inline u32 qed_chain_get_capacity(struct qed_chain *p_chain) { return p_chain->capacity; } @@ -297,11 +365,13 @@ static inline u16 qed_chain_get_capacity(struct qed_chain *p_chain) * * @param p_chain */ -static inline void -qed_chain_recycle_consumed(struct qed_chain *p_chain) +static inline void qed_chain_recycle_consumed(struct qed_chain *p_chain) { - test_ans_skip(p_chain, prod_idx); - p_chain->prod_idx++; + test_and_skip(p_chain, prod_idx); + if (is_chain_u16(p_chain)) + p_chain->u.chain16.prod_idx++; + else + p_chain->u.chain32.prod_idx++; } /** @@ -316,21 +386,33 @@ qed_chain_recycle_consumed(struct qed_chain *p_chain) */ static inline void *qed_chain_consume(struct qed_chain *p_chain) { - void *ret = NULL; - - if ((p_chain->cons_idx & p_chain->elem_per_page_mask) == - p_chain->next_page_mask) { + void *p_ret = NULL, *p_cons_idx, *p_cons_page_idx; + + if (is_chain_u16(p_chain)) { + if ((p_chain->u.chain16.cons_idx & + p_chain->elem_per_page_mask) == p_chain->next_page_mask) { + p_cons_idx = &p_chain->u.chain16.cons_idx; + p_cons_page_idx = &p_chain->pbl.u.pbl16.cons_page_idx; + qed_chain_advance_page(p_chain, &p_chain->p_cons_elem, + p_cons_idx, p_cons_page_idx); + } + p_chain->u.chain16.cons_idx++; + } else { + if ((p_chain->u.chain32.cons_idx & + p_chain->elem_per_page_mask) == p_chain->next_page_mask) { + p_cons_idx = &p_chain->u.chain32.cons_idx; + p_cons_page_idx = &p_chain->pbl.u.pbl32.cons_page_idx; qed_chain_advance_page(p_chain, &p_chain->p_cons_elem, - &p_chain->cons_idx, - &p_chain->pbl.cons_page_idx); + p_cons_idx, p_cons_page_idx); + } + p_chain->u.chain32.cons_idx++; } - ret = p_chain->p_cons_elem; - p_chain->cons_idx++; + p_ret = p_chain->p_cons_elem; p_chain->p_cons_elem = (void *)(((u8 *)p_chain->p_cons_elem) + p_chain->elem_size); - return ret; + return p_ret; } /** @@ -340,16 +422,33 @@ static inline void *qed_chain_consume(struct qed_chain *p_chain) */ static inline void qed_chain_reset(struct qed_chain *p_chain) { - int i; - - p_chain->prod_idx = 0; - p_chain->cons_idx = 0; - p_chain->p_cons_elem = p_chain->p_virt_addr; - p_chain->p_prod_elem = p_chain->p_virt_addr; + u32 i; + + if (is_chain_u16(p_chain)) { + p_chain->u.chain16.prod_idx = 0; + p_chain->u.chain16.cons_idx = 0; + } else { + p_chain->u.chain32.prod_idx = 0; + p_chain->u.chain32.cons_idx = 0; + } + p_chain->p_cons_elem = p_chain->p_virt_addr; + p_chain->p_prod_elem = p_chain->p_virt_addr; if (p_chain->mode == QED_CHAIN_MODE_PBL) { - p_chain->pbl.prod_page_idx = p_chain->page_cnt - 1; - p_chain->pbl.cons_page_idx = p_chain->page_cnt - 1; + /* Use (page_cnt - 1) as a reset value for the prod/cons page's + * indices, to avoid unnecessary page advancing on the first + * call to qed_chain_produce/consume. Instead, the indices + * will be advanced to page_cnt and then will be wrapped to 0. + */ + u32 reset_val = p_chain->page_cnt - 1; + + if (is_chain_u16(p_chain)) { + p_chain->pbl.u.pbl16.prod_page_idx = (u16)reset_val; + p_chain->pbl.u.pbl16.cons_page_idx = (u16)reset_val; + } else { + p_chain->pbl.u.pbl32.prod_page_idx = reset_val; + p_chain->pbl.u.pbl32.cons_page_idx = reset_val; + } } switch (p_chain->intended_use) { @@ -377,168 +476,184 @@ static inline void qed_chain_reset(struct qed_chain *p_chain) * @param intended_use * @param mode */ -static inline void qed_chain_init(struct qed_chain *p_chain, - void *p_virt_addr, - dma_addr_t p_phys_addr, - u16 page_cnt, - u8 elem_size, - enum qed_chain_use_mode intended_use, - enum qed_chain_mode mode) +static inline void qed_chain_init_params(struct qed_chain *p_chain, + u32 page_cnt, + u8 elem_size, + enum qed_chain_use_mode intended_use, + enum qed_chain_mode mode, + enum qed_chain_cnt_type cnt_type) { /* chain fixed parameters */ - p_chain->p_virt_addr = p_virt_addr; - p_chain->p_phys_addr = p_phys_addr; + p_chain->p_virt_addr = NULL; + p_chain->p_phys_addr = 0; p_chain->elem_size = elem_size; - p_chain->page_cnt = page_cnt; + p_chain->intended_use = intended_use; p_chain->mode = mode; + p_chain->cnt_type = cnt_type; - p_chain->intended_use = intended_use; p_chain->elem_per_page = ELEMS_PER_PAGE(elem_size); - p_chain->usable_per_page = - USABLE_ELEMS_PER_PAGE(elem_size, mode); - p_chain->capacity = p_chain->usable_per_page * page_cnt; - p_chain->size = p_chain->elem_per_page * page_cnt; + p_chain->usable_per_page = USABLE_ELEMS_PER_PAGE(elem_size, mode); p_chain->elem_per_page_mask = p_chain->elem_per_page - 1; - p_chain->elem_unusable = UNUSABLE_ELEMS_PER_PAGE(elem_size, mode); - p_chain->next_page_mask = (p_chain->usable_per_page & p_chain->elem_per_page_mask); - if (mode == QED_CHAIN_MODE_NEXT_PTR) { - struct qed_chain_next *p_next; - u16 i; - - for (i = 0; i < page_cnt - 1; i++) { - /* Increment mem_phy to the next page. */ - p_phys_addr += QED_CHAIN_PAGE_SIZE; - - /* Initialize the physical address of the next page. */ - p_next = (struct qed_chain_next *)((u8 *)p_virt_addr + - elem_size * - p_chain-> - usable_per_page); - - p_next->next_phys.lo = DMA_LO_LE(p_phys_addr); - p_next->next_phys.hi = DMA_HI_LE(p_phys_addr); - - /* Initialize the virtual address of the next page. */ - p_next->next_virt = (void *)((u8 *)p_virt_addr + - QED_CHAIN_PAGE_SIZE); - - /* Move to the next page. */ - p_virt_addr = p_next->next_virt; - } - - /* Last page's next should point to beginning of the chain */ - p_next = (struct qed_chain_next *)((u8 *)p_virt_addr + - elem_size * - p_chain->usable_per_page); + p_chain->page_cnt = page_cnt; + p_chain->capacity = p_chain->usable_per_page * page_cnt; + p_chain->size = p_chain->elem_per_page * page_cnt; - p_next->next_phys.lo = DMA_LO_LE(p_chain->p_phys_addr); - p_next->next_phys.hi = DMA_HI_LE(p_chain->p_phys_addr); - p_next->next_virt = p_chain->p_virt_addr; - } - qed_chain_reset(p_chain); + p_chain->pbl.p_phys_table = 0; + p_chain->pbl.p_virt_table = NULL; + p_chain->pbl.pp_virt_addr_tbl = NULL; } /** - * @brief qed_chain_pbl_init - Initalizes a basic pbl chain - * struct + * @brief qed_chain_init_mem - + * + * Initalizes a basic chain struct with its chain buffers + * * @param p_chain * @param p_virt_addr virtual address of allocated buffer's beginning * @param p_phys_addr physical address of allocated buffer's beginning - * @param page_cnt number of pages in the allocated buffer - * @param elem_size size of each element in the chain - * @param use_mode - * @param p_phys_pbl pointer to a pre-allocated side table - * which will hold physical page addresses. - * @param p_virt_pbl pointer to a pre allocated side table - * which will hold virtual page addresses. + * */ -static inline void -qed_chain_pbl_init(struct qed_chain *p_chain, - void *p_virt_addr, - dma_addr_t p_phys_addr, - u16 page_cnt, - u8 elem_size, - enum qed_chain_use_mode use_mode, - dma_addr_t p_phys_pbl, - dma_addr_t *p_virt_pbl) +static inline void qed_chain_init_mem(struct qed_chain *p_chain, + void *p_virt_addr, dma_addr_t p_phys_addr) { - dma_addr_t *p_pbl_dma = p_virt_pbl; - int i; - - qed_chain_init(p_chain, p_virt_addr, p_phys_addr, page_cnt, - elem_size, use_mode, QED_CHAIN_MODE_PBL); + p_chain->p_virt_addr = p_virt_addr; + p_chain->p_phys_addr = p_phys_addr; +} +/** + * @brief qed_chain_init_pbl_mem - + * + * Initalizes a basic chain struct with its pbl buffers + * + * @param p_chain + * @param p_virt_pbl pointer to a pre allocated side table which will hold + * virtual page addresses. + * @param p_phys_pbl pointer to a pre-allocated side table which will hold + * physical page addresses. + * @param pp_virt_addr_tbl + * pointer to a pre-allocated side table which will hold + * the virtual addresses of the chain pages. + * + */ +static inline void qed_chain_init_pbl_mem(struct qed_chain *p_chain, + void *p_virt_pbl, + dma_addr_t p_phys_pbl, + void **pp_virt_addr_tbl) +{ p_chain->pbl.p_phys_table = p_phys_pbl; p_chain->pbl.p_virt_table = p_virt_pbl; - - /* Fill the PBL with physical addresses*/ - for (i = 0; i < page_cnt; i++) { - *p_pbl_dma = p_phys_addr; - p_phys_addr += QED_CHAIN_PAGE_SIZE; - p_pbl_dma++; - } + p_chain->pbl.pp_virt_addr_tbl = pp_virt_addr_tbl; } /** - * @brief qed_chain_set_prod - sets the prod to the given - * value + * @brief qed_chain_init_next_ptr_elem - + * + * Initalizes a next pointer element + * + * @param p_chain + * @param p_virt_curr virtual address of a chain page of which the next + * pointer element is initialized + * @param p_virt_next virtual address of the next chain page + * @param p_phys_next physical address of the next chain page * - * @param prod_idx - * @param p_prod_elem */ -static inline void qed_chain_set_prod(struct qed_chain *p_chain, - u16 prod_idx, - void *p_prod_elem) +static inline void +qed_chain_init_next_ptr_elem(struct qed_chain *p_chain, + void *p_virt_curr, + void *p_virt_next, dma_addr_t p_phys_next) { - p_chain->prod_idx = prod_idx; - p_chain->p_prod_elem = p_prod_elem; + struct qed_chain_next *p_next; + u32 size; + + size = p_chain->elem_size * p_chain->usable_per_page; + p_next = (struct qed_chain_next *)((u8 *)p_virt_curr + size); + + DMA_REGPAIR_LE(p_next->next_phys, p_phys_next); + + p_next->next_virt = p_virt_next; } /** - * @brief qed_chain_get_elem - + * @brief qed_chain_get_last_elem - * - * get a pointer to an element represented by absolute idx + * Returns a pointer to the last element of the chain * * @param p_chain - * @assumption p_chain->size is a power of 2 * - * @return void*, a pointer to next element + * @return void* */ -static inline void *qed_chain_sge_get_elem(struct qed_chain *p_chain, - u16 idx) +static inline void *qed_chain_get_last_elem(struct qed_chain *p_chain) { - void *ret = NULL; - - if (idx >= p_chain->size) - return NULL; + struct qed_chain_next *p_next = NULL; + void *p_virt_addr = NULL; + u32 size, last_page_idx; - ret = (u8 *)p_chain->p_virt_addr + p_chain->elem_size * idx; + if (!p_chain->p_virt_addr) + goto out; - return ret; + switch (p_chain->mode) { + case QED_CHAIN_MODE_NEXT_PTR: + size = p_chain->elem_size * p_chain->usable_per_page; + p_virt_addr = p_chain->p_virt_addr; + p_next = (struct qed_chain_next *)((u8 *)p_virt_addr + size); + while (p_next->next_virt != p_chain->p_virt_addr) { + p_virt_addr = p_next->next_virt; + p_next = (struct qed_chain_next *)((u8 *)p_virt_addr + + size); + } + break; + case QED_CHAIN_MODE_SINGLE: + p_virt_addr = p_chain->p_virt_addr; + break; + case QED_CHAIN_MODE_PBL: + last_page_idx = p_chain->page_cnt - 1; + p_virt_addr = p_chain->pbl.pp_virt_addr_tbl[last_page_idx]; + break; + } + /* p_virt_addr points at this stage to the last page of the chain */ + size = p_chain->elem_size * (p_chain->usable_per_page - 1); + p_virt_addr = (u8 *)p_virt_addr + size; +out: + return p_virt_addr; } /** - * @brief qed_chain_sge_inc_cons_prod + * @brief qed_chain_set_prod - sets the prod to the given value * - * for sge chains, producer isn't increased serially, the ring - * is expected to be full at all times. Once elements are - * consumed, they are immediately produced. + * @param prod_idx + * @param p_prod_elem + */ +static inline void qed_chain_set_prod(struct qed_chain *p_chain, + u32 prod_idx, void *p_prod_elem) +{ + if (is_chain_u16(p_chain)) + p_chain->u.chain16.prod_idx = (u16) prod_idx; + else + p_chain->u.chain32.prod_idx = prod_idx; + p_chain->p_prod_elem = p_prod_elem; +} + +/** + * @brief qed_chain_pbl_zero_mem - set chain memory to 0 * * @param p_chain - * @param cnt - * - * @return inline void */ -static inline void -qed_chain_sge_inc_cons_prod(struct qed_chain *p_chain, - u16 cnt) +static inline void qed_chain_pbl_zero_mem(struct qed_chain *p_chain) { - p_chain->prod_idx += cnt; - p_chain->cons_idx += cnt; + u32 i, page_cnt; + + if (p_chain->mode != QED_CHAIN_MODE_PBL) + return; + + page_cnt = qed_chain_get_page_cnt(p_chain); + + for (i = 0; i < page_cnt; i++) + memset(p_chain->pbl.pp_virt_addr_tbl[i], 0, + QED_CHAIN_PAGE_SIZE); } #endif diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h index 4c29439f54bf..15efccfdc46e 100644 --- a/include/linux/qed/qed_if.h +++ b/include/linux/qed/qed_if.h @@ -325,7 +325,8 @@ struct qed_common_ops { int (*chain_alloc)(struct qed_dev *cdev, enum qed_chain_use_mode intended_use, enum qed_chain_mode mode, - u16 num_elems, + enum qed_chain_cnt_type cnt_type, + u32 num_elems, size_t elem_size, struct qed_chain *p_chain); -- cgit v1.2.3 From 7a9b6b8f6e4c52b31830d32570c2a226e27651f9 Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Fri, 3 Jun 2016 14:35:33 +0300 Subject: qed: Add common HSI for new protocols This adds the qed portion of the RoCE & iSCSI firmware HSI, as well as adding several new common HSI files which would be required by both qed and qed* protocols. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_hsi.h | 3008 +++++++++++++++++++++++++++++ drivers/net/ethernet/qlogic/qed/qed_sp.h | 26 + include/linux/qed/common_hsi.h | 190 +- include/linux/qed/iscsi_common.h | 1439 ++++++++++++++ include/linux/qed/rdma_common.h | 44 + include/linux/qed/roce_common.h | 17 + include/linux/qed/storage_common.h | 91 + include/linux/qed/tcp_common.h | 226 +++ 8 files changed, 5040 insertions(+), 1 deletion(-) create mode 100644 include/linux/qed/iscsi_common.h create mode 100644 include/linux/qed/rdma_common.h create mode 100644 include/linux/qed/roce_common.h create mode 100644 include/linux/qed/storage_common.h create mode 100644 include/linux/qed/tcp_common.h (limited to 'include/linux') diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h index d104ca790977..4cd70bfd929f 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h @@ -17,7 +17,12 @@ #include #include #include +#include +#include #include +#include +#include +#include struct qed_hwfn; struct qed_ptt; @@ -3631,6 +3636,3009 @@ struct vport_update_ramrod_data { struct eth_vport_rss_config rss_config; }; +struct mstorm_rdma_task_st_ctx { + struct regpair temp[4]; +}; + +struct rdma_close_func_ramrod_data { + u8 cnq_start_offset; + u8 num_cnqs; + u8 vf_id; + u8 vf_valid; + u8 reserved[4]; +}; + +struct rdma_cnq_params { + __le16 sb_num; + u8 sb_index; + u8 num_pbl_pages; + __le32 reserved; + struct regpair pbl_base_addr; + __le16 queue_zone_num; + u8 reserved1[6]; +}; + +struct rdma_create_cq_ramrod_data { + struct regpair cq_handle; + struct regpair pbl_addr; + __le32 max_cqes; + __le16 pbl_num_pages; + __le16 dpi; + u8 is_two_level_pbl; + u8 cnq_id; + u8 pbl_log_page_size; + u8 toggle_bit; + __le16 int_timeout; + __le16 reserved1; +}; + +struct rdma_deregister_tid_ramrod_data { + __le32 itid; + __le32 reserved; +}; + +struct rdma_destroy_cq_output_params { + __le16 cnq_num; + __le16 reserved0; + __le32 reserved1; +}; + +struct rdma_destroy_cq_ramrod_data { + struct regpair output_params_addr; +}; + +enum rdma_event_opcode { + RDMA_EVENT_UNUSED, + RDMA_EVENT_FUNC_INIT, + RDMA_EVENT_FUNC_CLOSE, + RDMA_EVENT_REGISTER_MR, + RDMA_EVENT_DEREGISTER_MR, + RDMA_EVENT_CREATE_CQ, + RDMA_EVENT_RESIZE_CQ, + RDMA_EVENT_DESTROY_CQ, + RDMA_EVENT_CREATE_SRQ, + RDMA_EVENT_MODIFY_SRQ, + RDMA_EVENT_DESTROY_SRQ, + MAX_RDMA_EVENT_OPCODE +}; + +enum rdma_fw_return_code { + RDMA_RETURN_OK = 0, + RDMA_RETURN_REGISTER_MR_BAD_STATE_ERR, + RDMA_RETURN_DEREGISTER_MR_BAD_STATE_ERR, + RDMA_RETURN_RESIZE_CQ_ERR, + RDMA_RETURN_NIG_DRAIN_REQ, + MAX_RDMA_FW_RETURN_CODE +}; + +struct rdma_init_func_hdr { + u8 cnq_start_offset; + u8 num_cnqs; + u8 cq_ring_mode; + u8 cnp_vlan_priority; + __le32 cnp_send_timeout; + u8 cnp_dscp; + u8 vf_id; + u8 vf_valid; + u8 reserved[5]; +}; + +struct rdma_init_func_ramrod_data { + struct rdma_init_func_hdr params_header; + struct rdma_cnq_params cnq_params[NUM_OF_GLOBAL_QUEUES]; +}; + +enum rdma_ramrod_cmd_id { + RDMA_RAMROD_UNUSED, + RDMA_RAMROD_FUNC_INIT, + RDMA_RAMROD_FUNC_CLOSE, + RDMA_RAMROD_REGISTER_MR, + RDMA_RAMROD_DEREGISTER_MR, + RDMA_RAMROD_CREATE_CQ, + RDMA_RAMROD_RESIZE_CQ, + RDMA_RAMROD_DESTROY_CQ, + RDMA_RAMROD_CREATE_SRQ, + RDMA_RAMROD_MODIFY_SRQ, + RDMA_RAMROD_DESTROY_SRQ, + MAX_RDMA_RAMROD_CMD_ID +}; + +struct rdma_register_tid_ramrod_data { + __le32 flags; +#define RDMA_REGISTER_TID_RAMROD_DATA_MAX_ID_MASK 0x3FFFF +#define RDMA_REGISTER_TID_RAMROD_DATA_MAX_ID_SHIFT 0 +#define RDMA_REGISTER_TID_RAMROD_DATA_PAGE_SIZE_LOG_MASK 0x1F +#define RDMA_REGISTER_TID_RAMROD_DATA_PAGE_SIZE_LOG_SHIFT 18 +#define RDMA_REGISTER_TID_RAMROD_DATA_TWO_LEVEL_PBL_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_TWO_LEVEL_PBL_SHIFT 23 +#define RDMA_REGISTER_TID_RAMROD_DATA_ZERO_BASED_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_ZERO_BASED_SHIFT 24 +#define RDMA_REGISTER_TID_RAMROD_DATA_PHY_MR_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_PHY_MR_SHIFT 25 +#define RDMA_REGISTER_TID_RAMROD_DATA_REMOTE_READ_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_REMOTE_READ_SHIFT 26 +#define RDMA_REGISTER_TID_RAMROD_DATA_REMOTE_WRITE_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_REMOTE_WRITE_SHIFT 27 +#define RDMA_REGISTER_TID_RAMROD_DATA_REMOTE_ATOMIC_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_REMOTE_ATOMIC_SHIFT 28 +#define RDMA_REGISTER_TID_RAMROD_DATA_LOCAL_WRITE_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_LOCAL_WRITE_SHIFT 29 +#define RDMA_REGISTER_TID_RAMROD_DATA_LOCAL_READ_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_LOCAL_READ_SHIFT 30 +#define RDMA_REGISTER_TID_RAMROD_DATA_ENABLE_MW_BIND_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_ENABLE_MW_BIND_SHIFT 31 + u8 flags1; +#define RDMA_REGISTER_TID_RAMROD_DATA_PBL_PAGE_SIZE_LOG_MASK 0x1F +#define RDMA_REGISTER_TID_RAMROD_DATA_PBL_PAGE_SIZE_LOG_SHIFT 0 +#define RDMA_REGISTER_TID_RAMROD_DATA_TID_TYPE_MASK 0x7 +#define RDMA_REGISTER_TID_RAMROD_DATA_TID_TYPE_SHIFT 5 + u8 flags2; +#define RDMA_REGISTER_TID_RAMROD_DATA_DMA_MR_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_DMA_MR_SHIFT 0 +#define RDMA_REGISTER_TID_RAMROD_DATA_DIF_ON_HOST_FLG_MASK 0x1 +#define RDMA_REGISTER_TID_RAMROD_DATA_DIF_ON_HOST_FLG_SHIFT 1 +#define RDMA_REGISTER_TID_RAMROD_DATA_RESERVED1_MASK 0x3F +#define RDMA_REGISTER_TID_RAMROD_DATA_RESERVED1_SHIFT 2 + u8 key; + u8 length_hi; + u8 vf_id; + u8 vf_valid; + __le16 pd; + __le32 length_lo; + __le32 itid; + __le32 reserved2; + struct regpair va; + struct regpair pbl_base; + struct regpair dif_error_addr; + struct regpair dif_runt_addr; + __le32 reserved3[2]; +}; + +struct rdma_resize_cq_output_params { + __le32 old_cq_cons; + __le32 old_cq_prod; +}; + +struct rdma_resize_cq_ramrod_data { + u8 flags; +#define RDMA_RESIZE_CQ_RAMROD_DATA_TOGGLE_BIT_MASK 0x1 +#define RDMA_RESIZE_CQ_RAMROD_DATA_TOGGLE_BIT_SHIFT 0 +#define RDMA_RESIZE_CQ_RAMROD_DATA_IS_TWO_LEVEL_PBL_MASK 0x1 +#define RDMA_RESIZE_CQ_RAMROD_DATA_IS_TWO_LEVEL_PBL_SHIFT 1 +#define RDMA_RESIZE_CQ_RAMROD_DATA_RESERVED_MASK 0x3F +#define RDMA_RESIZE_CQ_RAMROD_DATA_RESERVED_SHIFT 2 + u8 pbl_log_page_size; + __le16 pbl_num_pages; + __le32 max_cqes; + struct regpair pbl_addr; + struct regpair output_params_addr; +}; + +struct rdma_srq_context { + struct regpair temp[8]; +}; + +struct rdma_srq_create_ramrod_data { + struct regpair pbl_base_addr; + __le16 pages_in_srq_pbl; + __le16 pd_id; + struct rdma_srq_id srq_id; + __le16 page_size; + __le16 reserved1; + __le32 reserved2; + struct regpair producers_addr; +}; + +struct rdma_srq_destroy_ramrod_data { + struct rdma_srq_id srq_id; + __le32 reserved; +}; + +struct rdma_srq_modify_ramrod_data { + struct rdma_srq_id srq_id; + __le32 wqe_limit; +}; + +struct ystorm_rdma_task_st_ctx { + struct regpair temp[4]; +}; + +struct ystorm_rdma_task_ag_ctx { + u8 reserved; + u8 byte1; + __le16 msem_ctx_upd_seq; + u8 flags0; +#define YSTORM_RDMA_TASK_AG_CTX_CONNECTION_TYPE_MASK 0xF +#define YSTORM_RDMA_TASK_AG_CTX_CONNECTION_TYPE_SHIFT 0 +#define YSTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_SHIFT 4 +#define YSTORM_RDMA_TASK_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_BIT1_SHIFT 5 +#define YSTORM_RDMA_TASK_AG_CTX_VALID_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_VALID_SHIFT 6 +#define YSTORM_RDMA_TASK_AG_CTX_BIT3_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_BIT3_SHIFT 7 + u8 flags1; +#define YSTORM_RDMA_TASK_AG_CTX_CF0_MASK 0x3 +#define YSTORM_RDMA_TASK_AG_CTX_CF0_SHIFT 0 +#define YSTORM_RDMA_TASK_AG_CTX_CF1_MASK 0x3 +#define YSTORM_RDMA_TASK_AG_CTX_CF1_SHIFT 2 +#define YSTORM_RDMA_TASK_AG_CTX_CF2SPECIAL_MASK 0x3 +#define YSTORM_RDMA_TASK_AG_CTX_CF2SPECIAL_SHIFT 4 +#define YSTORM_RDMA_TASK_AG_CTX_CF0EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_CF0EN_SHIFT 6 +#define YSTORM_RDMA_TASK_AG_CTX_CF1EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_CF1EN_SHIFT 7 + u8 flags2; +#define YSTORM_RDMA_TASK_AG_CTX_BIT4_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_BIT4_SHIFT 0 +#define YSTORM_RDMA_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE0EN_SHIFT 1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE1EN_SHIFT 2 +#define YSTORM_RDMA_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE2EN_SHIFT 3 +#define YSTORM_RDMA_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE3EN_SHIFT 4 +#define YSTORM_RDMA_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE4EN_SHIFT 5 +#define YSTORM_RDMA_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE5EN_SHIFT 6 +#define YSTORM_RDMA_TASK_AG_CTX_RULE6EN_MASK 0x1 +#define YSTORM_RDMA_TASK_AG_CTX_RULE6EN_SHIFT 7 + u8 key; + __le32 mw_cnt; + u8 ref_cnt_seq; + u8 ctx_upd_seq; + __le16 dif_flags; + __le16 tx_ref_count; + __le16 last_used_ltid; + __le16 parent_mr_lo; + __le16 parent_mr_hi; + __le32 fbo_lo; + __le32 fbo_hi; +}; + +struct mstorm_rdma_task_ag_ctx { + u8 reserved; + u8 byte1; + __le16 icid; + u8 flags0; +#define MSTORM_RDMA_TASK_AG_CTX_CONNECTION_TYPE_MASK 0xF +#define MSTORM_RDMA_TASK_AG_CTX_CONNECTION_TYPE_SHIFT 0 +#define MSTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_SHIFT 4 +#define MSTORM_RDMA_TASK_AG_CTX_BIT1_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_BIT1_SHIFT 5 +#define MSTORM_RDMA_TASK_AG_CTX_BIT2_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_BIT2_SHIFT 6 +#define MSTORM_RDMA_TASK_AG_CTX_BIT3_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_BIT3_SHIFT 7 + u8 flags1; +#define MSTORM_RDMA_TASK_AG_CTX_CF0_MASK 0x3 +#define MSTORM_RDMA_TASK_AG_CTX_CF0_SHIFT 0 +#define MSTORM_RDMA_TASK_AG_CTX_CF1_MASK 0x3 +#define MSTORM_RDMA_TASK_AG_CTX_CF1_SHIFT 2 +#define MSTORM_RDMA_TASK_AG_CTX_CF2_MASK 0x3 +#define MSTORM_RDMA_TASK_AG_CTX_CF2_SHIFT 4 +#define MSTORM_RDMA_TASK_AG_CTX_CF0EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_CF0EN_SHIFT 6 +#define MSTORM_RDMA_TASK_AG_CTX_CF1EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_CF1EN_SHIFT 7 + u8 flags2; +#define MSTORM_RDMA_TASK_AG_CTX_CF2EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_CF2EN_SHIFT 0 +#define MSTORM_RDMA_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE0EN_SHIFT 1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE1EN_SHIFT 2 +#define MSTORM_RDMA_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE2EN_SHIFT 3 +#define MSTORM_RDMA_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE3EN_SHIFT 4 +#define MSTORM_RDMA_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE4EN_SHIFT 5 +#define MSTORM_RDMA_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE5EN_SHIFT 6 +#define MSTORM_RDMA_TASK_AG_CTX_RULE6EN_MASK 0x1 +#define MSTORM_RDMA_TASK_AG_CTX_RULE6EN_SHIFT 7 + u8 key; + __le32 mw_cnt; + u8 ref_cnt_seq; + u8 ctx_upd_seq; + __le16 dif_flags; + __le16 tx_ref_count; + __le16 last_used_ltid; + __le16 parent_mr_lo; + __le16 parent_mr_hi; + __le32 fbo_lo; + __le32 fbo_hi; +}; + +struct ustorm_rdma_task_st_ctx { + struct regpair temp[2]; +}; + +struct ustorm_rdma_task_ag_ctx { + u8 reserved; + u8 byte1; + __le16 icid; + u8 flags0; +#define USTORM_RDMA_TASK_AG_CTX_CONNECTION_TYPE_MASK 0xF +#define USTORM_RDMA_TASK_AG_CTX_CONNECTION_TYPE_SHIFT 0 +#define USTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_EXIST_IN_QM0_SHIFT 4 +#define USTORM_RDMA_TASK_AG_CTX_DIF_RUNT_VALID_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_DIF_RUNT_VALID_SHIFT 5 +#define USTORM_RDMA_TASK_AG_CTX_DIF_WRITE_RESULT_CF_MASK 0x3 +#define USTORM_RDMA_TASK_AG_CTX_DIF_WRITE_RESULT_CF_SHIFT 6 + u8 flags1; +#define USTORM_RDMA_TASK_AG_CTX_DIF_RESULT_TOGGLE_BIT_MASK 0x3 +#define USTORM_RDMA_TASK_AG_CTX_DIF_RESULT_TOGGLE_BIT_SHIFT 0 +#define USTORM_RDMA_TASK_AG_CTX_DIF_TX_IO_FLG_MASK 0x3 +#define USTORM_RDMA_TASK_AG_CTX_DIF_TX_IO_FLG_SHIFT 2 +#define USTORM_RDMA_TASK_AG_CTX_CF3_MASK 0x3 +#define USTORM_RDMA_TASK_AG_CTX_CF3_SHIFT 4 +#define USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_CF_MASK 0x3 +#define USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_CF_SHIFT 6 + u8 flags2; +#define USTORM_RDMA_TASK_AG_CTX_DIF_WRITE_RESULT_CF_EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_DIF_WRITE_RESULT_CF_EN_SHIFT 0 +#define USTORM_RDMA_TASK_AG_CTX_RESERVED2_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RESERVED2_SHIFT 1 +#define USTORM_RDMA_TASK_AG_CTX_RESERVED3_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RESERVED3_SHIFT 2 +#define USTORM_RDMA_TASK_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_CF_EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_CF_EN_SHIFT 4 +#define USTORM_RDMA_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE0EN_SHIFT 5 +#define USTORM_RDMA_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE1EN_SHIFT 6 +#define USTORM_RDMA_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE2EN_SHIFT 7 + u8 flags3; +#define USTORM_RDMA_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE3EN_SHIFT 0 +#define USTORM_RDMA_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE4EN_SHIFT 1 +#define USTORM_RDMA_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE5EN_SHIFT 2 +#define USTORM_RDMA_TASK_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_RDMA_TASK_AG_CTX_RULE6EN_SHIFT 3 +#define USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_TYPE_MASK 0xF +#define USTORM_RDMA_TASK_AG_CTX_DIF_ERROR_TYPE_SHIFT 4 + __le32 dif_err_intervals; + __le32 dif_error_1st_interval; + __le32 reg2; + __le32 dif_runt_value; + __le32 reg4; + __le32 reg5; +}; + +struct rdma_task_context { + struct ystorm_rdma_task_st_ctx ystorm_st_context; + struct ystorm_rdma_task_ag_ctx ystorm_ag_context; + struct tdif_task_context tdif_context; + struct mstorm_rdma_task_ag_ctx mstorm_ag_context; + struct mstorm_rdma_task_st_ctx mstorm_st_context; + struct rdif_task_context rdif_context; + struct ustorm_rdma_task_st_ctx ustorm_st_context; + struct regpair ustorm_st_padding[2]; + struct ustorm_rdma_task_ag_ctx ustorm_ag_context; +}; + +enum rdma_tid_type { + RDMA_TID_REGISTERED_MR, + RDMA_TID_FMR, + RDMA_TID_MW_TYPE1, + RDMA_TID_MW_TYPE2A, + MAX_RDMA_TID_TYPE +}; + +struct mstorm_rdma_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define MSTORM_RDMA_CONN_AG_CTX_BIT0_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_BIT0_SHIFT 0 +#define MSTORM_RDMA_CONN_AG_CTX_BIT1_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_BIT1_SHIFT 1 +#define MSTORM_RDMA_CONN_AG_CTX_CF0_MASK 0x3 +#define MSTORM_RDMA_CONN_AG_CTX_CF0_SHIFT 2 +#define MSTORM_RDMA_CONN_AG_CTX_CF1_MASK 0x3 +#define MSTORM_RDMA_CONN_AG_CTX_CF1_SHIFT 4 +#define MSTORM_RDMA_CONN_AG_CTX_CF2_MASK 0x3 +#define MSTORM_RDMA_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define MSTORM_RDMA_CONN_AG_CTX_CF0EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_CF0EN_SHIFT 0 +#define MSTORM_RDMA_CONN_AG_CTX_CF1EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_CF1EN_SHIFT 1 +#define MSTORM_RDMA_CONN_AG_CTX_CF2EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_CF2EN_SHIFT 2 +#define MSTORM_RDMA_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define MSTORM_RDMA_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define MSTORM_RDMA_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define MSTORM_RDMA_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define MSTORM_RDMA_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define MSTORM_RDMA_CONN_AG_CTX_RULE4EN_SHIFT 7 + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; +}; + +struct tstorm_rdma_conn_ag_ctx { + u8 reserved0; + u8 byte1; + u8 flags0; +#define TSTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define TSTORM_RDMA_CONN_AG_CTX_BIT1_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_BIT1_SHIFT 1 +#define TSTORM_RDMA_CONN_AG_CTX_BIT2_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_BIT2_SHIFT 2 +#define TSTORM_RDMA_CONN_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_BIT3_SHIFT 3 +#define TSTORM_RDMA_CONN_AG_CTX_BIT4_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_BIT4_SHIFT 4 +#define TSTORM_RDMA_CONN_AG_CTX_BIT5_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_BIT5_SHIFT 5 +#define TSTORM_RDMA_CONN_AG_CTX_CF0_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF0_SHIFT 6 + u8 flags1; +#define TSTORM_RDMA_CONN_AG_CTX_CF1_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF1_SHIFT 0 +#define TSTORM_RDMA_CONN_AG_CTX_CF2_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF2_SHIFT 2 +#define TSTORM_RDMA_CONN_AG_CTX_TIMER_STOP_ALL_CF_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_TIMER_STOP_ALL_CF_SHIFT 4 +#define TSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 6 + u8 flags2; +#define TSTORM_RDMA_CONN_AG_CTX_MSTORM_FLUSH_CF_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_MSTORM_FLUSH_CF_SHIFT 0 +#define TSTORM_RDMA_CONN_AG_CTX_CF6_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF6_SHIFT 2 +#define TSTORM_RDMA_CONN_AG_CTX_CF7_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF7_SHIFT 4 +#define TSTORM_RDMA_CONN_AG_CTX_CF8_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF8_SHIFT 6 + u8 flags3; +#define TSTORM_RDMA_CONN_AG_CTX_CF9_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF9_SHIFT 0 +#define TSTORM_RDMA_CONN_AG_CTX_CF10_MASK 0x3 +#define TSTORM_RDMA_CONN_AG_CTX_CF10_SHIFT 2 +#define TSTORM_RDMA_CONN_AG_CTX_CF0EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF0EN_SHIFT 4 +#define TSTORM_RDMA_CONN_AG_CTX_CF1EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF1EN_SHIFT 5 +#define TSTORM_RDMA_CONN_AG_CTX_CF2EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF2EN_SHIFT 6 +#define TSTORM_RDMA_CONN_AG_CTX_TIMER_STOP_ALL_CF_EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_TIMER_STOP_ALL_CF_EN_SHIFT 7 + u8 flags4; +#define TSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 0 +#define TSTORM_RDMA_CONN_AG_CTX_MSTORM_FLUSH_CF_EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_MSTORM_FLUSH_CF_EN_SHIFT 1 +#define TSTORM_RDMA_CONN_AG_CTX_CF6EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF6EN_SHIFT 2 +#define TSTORM_RDMA_CONN_AG_CTX_CF7EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF7EN_SHIFT 3 +#define TSTORM_RDMA_CONN_AG_CTX_CF8EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF8EN_SHIFT 4 +#define TSTORM_RDMA_CONN_AG_CTX_CF9EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF9EN_SHIFT 5 +#define TSTORM_RDMA_CONN_AG_CTX_CF10EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_CF10EN_SHIFT 6 +#define TSTORM_RDMA_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags5; +#define TSTORM_RDMA_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define TSTORM_RDMA_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define TSTORM_RDMA_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define TSTORM_RDMA_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define TSTORM_RDMA_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define TSTORM_RDMA_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define TSTORM_RDMA_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define TSTORM_RDMA_CONN_AG_CTX_RULE8EN_SHIFT 7 + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le32 reg4; + __le32 reg5; + __le32 reg6; + __le32 reg7; + __le32 reg8; + u8 byte2; + u8 byte3; + __le16 word0; + u8 byte4; + u8 byte5; + __le16 word1; + __le16 word2; + __le16 word3; + __le32 reg9; + __le32 reg10; +}; + +struct tstorm_rdma_task_ag_ctx { + u8 byte0; + u8 byte1; + __le16 word0; + u8 flags0; +#define TSTORM_RDMA_TASK_AG_CTX_NIBBLE0_MASK 0xF +#define TSTORM_RDMA_TASK_AG_CTX_NIBBLE0_SHIFT 0 +#define TSTORM_RDMA_TASK_AG_CTX_BIT0_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_BIT0_SHIFT 4 +#define TSTORM_RDMA_TASK_AG_CTX_BIT1_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_BIT1_SHIFT 5 +#define TSTORM_RDMA_TASK_AG_CTX_BIT2_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_BIT2_SHIFT 6 +#define TSTORM_RDMA_TASK_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_BIT3_SHIFT 7 + u8 flags1; +#define TSTORM_RDMA_TASK_AG_CTX_BIT4_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_BIT4_SHIFT 0 +#define TSTORM_RDMA_TASK_AG_CTX_BIT5_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_BIT5_SHIFT 1 +#define TSTORM_RDMA_TASK_AG_CTX_CF0_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF0_SHIFT 2 +#define TSTORM_RDMA_TASK_AG_CTX_CF1_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF1_SHIFT 4 +#define TSTORM_RDMA_TASK_AG_CTX_CF2_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF2_SHIFT 6 + u8 flags2; +#define TSTORM_RDMA_TASK_AG_CTX_CF3_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF3_SHIFT 0 +#define TSTORM_RDMA_TASK_AG_CTX_CF4_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF4_SHIFT 2 +#define TSTORM_RDMA_TASK_AG_CTX_CF5_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF5_SHIFT 4 +#define TSTORM_RDMA_TASK_AG_CTX_CF6_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF6_SHIFT 6 + u8 flags3; +#define TSTORM_RDMA_TASK_AG_CTX_CF7_MASK 0x3 +#define TSTORM_RDMA_TASK_AG_CTX_CF7_SHIFT 0 +#define TSTORM_RDMA_TASK_AG_CTX_CF0EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF0EN_SHIFT 2 +#define TSTORM_RDMA_TASK_AG_CTX_CF1EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF1EN_SHIFT 3 +#define TSTORM_RDMA_TASK_AG_CTX_CF2EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF2EN_SHIFT 4 +#define TSTORM_RDMA_TASK_AG_CTX_CF3EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF3EN_SHIFT 5 +#define TSTORM_RDMA_TASK_AG_CTX_CF4EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF4EN_SHIFT 6 +#define TSTORM_RDMA_TASK_AG_CTX_CF5EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF5EN_SHIFT 7 + u8 flags4; +#define TSTORM_RDMA_TASK_AG_CTX_CF6EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF6EN_SHIFT 0 +#define TSTORM_RDMA_TASK_AG_CTX_CF7EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_CF7EN_SHIFT 1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE0EN_SHIFT 2 +#define TSTORM_RDMA_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE1EN_SHIFT 3 +#define TSTORM_RDMA_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE2EN_SHIFT 4 +#define TSTORM_RDMA_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE3EN_SHIFT 5 +#define TSTORM_RDMA_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE4EN_SHIFT 6 +#define TSTORM_RDMA_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_RDMA_TASK_AG_CTX_RULE5EN_SHIFT 7 + u8 byte2; + __le16 word1; + __le32 reg0; + u8 byte3; + u8 byte4; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg1; + __le32 reg2; +}; + +struct ustorm_rdma_conn_ag_ctx { + u8 reserved; + u8 byte1; + u8 flags0; +#define USTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define USTORM_RDMA_CONN_AG_CTX_BIT1_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_BIT1_SHIFT 1 +#define USTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 2 +#define USTORM_RDMA_CONN_AG_CTX_CF1_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_CF1_SHIFT 4 +#define USTORM_RDMA_CONN_AG_CTX_CF2_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define USTORM_RDMA_CONN_AG_CTX_CF3_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_CF3_SHIFT 0 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_SE_CF_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_SE_CF_SHIFT 2 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_CF_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_CF_SHIFT 4 +#define USTORM_RDMA_CONN_AG_CTX_CF6_MASK 0x3 +#define USTORM_RDMA_CONN_AG_CTX_CF6_SHIFT 6 + u8 flags2; +#define USTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 0 +#define USTORM_RDMA_CONN_AG_CTX_CF1EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CF1EN_SHIFT 1 +#define USTORM_RDMA_CONN_AG_CTX_CF2EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CF2EN_SHIFT 2 +#define USTORM_RDMA_CONN_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_SE_CF_EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_SE_CF_EN_SHIFT 4 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_CF_EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CQ_ARM_CF_EN_SHIFT 5 +#define USTORM_RDMA_CONN_AG_CTX_CF6EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CF6EN_SHIFT 6 +#define USTORM_RDMA_CONN_AG_CTX_CQ_SE_EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CQ_SE_EN_SHIFT 7 + u8 flags3; +#define USTORM_RDMA_CONN_AG_CTX_CQ_EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_CQ_EN_SHIFT 0 +#define USTORM_RDMA_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define USTORM_RDMA_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define USTORM_RDMA_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define USTORM_RDMA_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define USTORM_RDMA_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define USTORM_RDMA_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define USTORM_RDMA_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define USTORM_RDMA_CONN_AG_CTX_RULE8EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 conn_dpi; + __le16 word1; + __le32 cq_cons; + __le32 cq_se_prod; + __le32 cq_prod; + __le32 reg3; + __le16 int_timeout; + __le16 word3; +}; + +struct xstorm_roce_conn_ag_ctx_dq_ext_ld_part { + u8 reserved0; + u8 state; + u8 flags0; +#define XSTORMROCECONNAGCTXDQEXTLDPART_EXIST_IN_QM0_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_EXIST_IN_QM0_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT1_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT1_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT2_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT2_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_EXIST_IN_QM3_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_EXIST_IN_QM3_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT4_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT4_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT5_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT5_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT6_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT6_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT7_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT7_SHIFT 7 + u8 flags1; +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT8_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT8_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT9_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT9_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT10_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT10_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT11_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT11_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT12_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT12_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT13_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT13_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT14_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT14_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_YSTORM_FLUSH_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_YSTORM_FLUSH_SHIFT 7 + u8 flags2; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF0_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF0_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF1_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF1_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF2_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF2_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF3_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF3_SHIFT 6 + u8 flags3; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF4_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF4_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF5_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF5_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF6_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF6_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_FLUSH_Q0_CF_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_FLUSH_Q0_CF_SHIFT 6 + u8 flags4; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF8_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF8_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF9_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF9_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF10_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF10_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF11_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF11_SHIFT 6 + u8 flags5; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF12_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF12_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF13_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF13_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF14_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF14_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF15_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF15_SHIFT 6 + u8 flags6; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF16_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF16_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF17_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF17_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF18_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF18_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF19_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF19_SHIFT 6 + u8 flags7; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF20_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF20_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF21_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF21_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_SLOW_PATH_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_SLOW_PATH_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF0EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF0EN_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF1EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF1EN_SHIFT 7 + u8 flags8; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF2EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF2EN_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF3EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF3EN_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF4EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF4EN_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF5EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF5EN_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF6EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF6EN_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_FLUSH_Q0_CF_EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_FLUSH_Q0_CF_EN_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF8EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF8EN_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF9EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF9EN_SHIFT 7 + u8 flags9; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF10EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF10EN_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF11EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF11EN_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF12EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF12EN_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF13EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF13EN_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF14EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF14EN_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF15EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF15EN_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF16EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF16EN_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF17EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF17EN_SHIFT 7 + u8 flags10; +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF18EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF18EN_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF19EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF19EN_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF20EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF20EN_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF21EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF21EN_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_SLOW_PATH_EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_SLOW_PATH_EN_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF23EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF23EN_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE0EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE0EN_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE1EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE1EN_SHIFT 7 + u8 flags11; +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE2EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE2EN_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE3EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE3EN_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE4EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE4EN_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE5EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE5EN_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE6EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE6EN_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE7EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE7EN_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED1_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED1_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE9EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE9EN_SHIFT 7 + u8 flags12; +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE10EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE10EN_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE11EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE11EN_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED2_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED2_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED3_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED3_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE14EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE14EN_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE15EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE15EN_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE16EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE16EN_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE17EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE17EN_SHIFT 7 + u8 flags13; +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE18EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE18EN_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE19EN_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RULE19EN_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED4_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED4_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED5_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED5_SHIFT 3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED6_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED6_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED7_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED7_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED8_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED8_SHIFT 6 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED9_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_A0_RESERVED9_SHIFT 7 + u8 flags14; +#define XSTORMROCECONNAGCTXDQEXTLDPART_MIGRATION_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_MIGRATION_SHIFT 0 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT17_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_BIT17_SHIFT 1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_DPM_PORT_NUM_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_DPM_PORT_NUM_SHIFT 2 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RESERVED_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_RESERVED_SHIFT 4 +#define XSTORMROCECONNAGCTXDQEXTLDPART_ROCE_EDPM_ENABLE_MASK 0x1 +#define XSTORMROCECONNAGCTXDQEXTLDPART_ROCE_EDPM_ENABLE_SHIFT 5 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF23_MASK 0x3 +#define XSTORMROCECONNAGCTXDQEXTLDPART_CF23_SHIFT 6 + u8 byte2; + __le16 physical_q0; + __le16 word1; + __le16 word2; + __le16 word3; + __le16 word4; + __le16 word5; + __le16 conn_dpi; + u8 byte3; + u8 byte4; + u8 byte5; + u8 byte6; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 snd_nxt_psn; + __le32 reg4; +}; + +struct xstorm_rdma_conn_ag_ctx { + u8 reserved0; + u8 state; + u8 flags0; +#define XSTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_BIT1_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT1_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT2_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT2_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM3_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_EXIST_IN_QM3_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_BIT4_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT4_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_BIT5_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT5_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_BIT6_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT6_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_BIT7_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT7_SHIFT 7 + u8 flags1; +#define XSTORM_RDMA_CONN_AG_CTX_BIT8_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT8_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_BIT9_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT9_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT10_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT10_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_BIT11_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT11_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_BIT12_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT12_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_BIT13_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT13_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_BIT14_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT14_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_YSTORM_FLUSH_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_YSTORM_FLUSH_SHIFT 7 + u8 flags2; +#define XSTORM_RDMA_CONN_AG_CTX_CF0_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF0_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF1_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF1_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF2_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF2_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF3_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF3_SHIFT 6 + u8 flags3; +#define XSTORM_RDMA_CONN_AG_CTX_CF4_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF4_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF5_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF5_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF6_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF6_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 6 + u8 flags4; +#define XSTORM_RDMA_CONN_AG_CTX_CF8_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF8_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF9_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF9_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF10_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF10_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF11_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF11_SHIFT 6 + u8 flags5; +#define XSTORM_RDMA_CONN_AG_CTX_CF12_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF12_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF13_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF13_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF14_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF14_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF15_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF15_SHIFT 6 + u8 flags6; +#define XSTORM_RDMA_CONN_AG_CTX_CF16_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF16_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF17_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF17_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF18_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF18_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF19_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF19_SHIFT 6 + u8 flags7; +#define XSTORM_RDMA_CONN_AG_CTX_CF20_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF20_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF21_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF21_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_SLOW_PATH_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_SLOW_PATH_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF0EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF0EN_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_CF1EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF1EN_SHIFT 7 + u8 flags8; +#define XSTORM_RDMA_CONN_AG_CTX_CF2EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF2EN_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF3EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF3EN_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_CF4EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF4EN_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF5EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF5EN_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_CF6EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF6EN_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_CF8EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF8EN_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_CF9EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF9EN_SHIFT 7 + u8 flags9; +#define XSTORM_RDMA_CONN_AG_CTX_CF10EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF10EN_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF11EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF11EN_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_CF12EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF12EN_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF13EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF13EN_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_CF14EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF14EN_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF15EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF15EN_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_CF16EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF16EN_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_CF17EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF17EN_SHIFT 7 + u8 flags10; +#define XSTORM_RDMA_CONN_AG_CTX_CF18EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF18EN_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_CF19EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF19EN_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_CF20EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF20EN_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_CF21EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF21EN_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_SLOW_PATH_EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_SLOW_PATH_EN_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_CF23EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_CF23EN_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE0EN_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE1EN_SHIFT 7 + u8 flags11; +#define XSTORM_RDMA_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE2EN_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE3EN_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE4EN_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE5EN_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE6EN_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE7EN_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED1_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED1_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_RULE9EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE9EN_SHIFT 7 + u8 flags12; +#define XSTORM_RDMA_CONN_AG_CTX_RULE10EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE10EN_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_RULE11EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE11EN_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED2_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED2_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED3_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED3_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_RULE14EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE14EN_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_RULE15EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE15EN_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_RULE16EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE16EN_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_RULE17EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE17EN_SHIFT 7 + u8 flags13; +#define XSTORM_RDMA_CONN_AG_CTX_RULE18EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE18EN_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_RULE19EN_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RULE19EN_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED4_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED4_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED5_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED5_SHIFT 3 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED6_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED6_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED7_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED7_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED8_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED8_SHIFT 6 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED9_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_A0_RESERVED9_SHIFT 7 + u8 flags14; +#define XSTORM_RDMA_CONN_AG_CTX_MIGRATION_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_MIGRATION_SHIFT 0 +#define XSTORM_RDMA_CONN_AG_CTX_BIT17_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_BIT17_SHIFT 1 +#define XSTORM_RDMA_CONN_AG_CTX_DPM_PORT_NUM_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_DPM_PORT_NUM_SHIFT 2 +#define XSTORM_RDMA_CONN_AG_CTX_RESERVED_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_RESERVED_SHIFT 4 +#define XSTORM_RDMA_CONN_AG_CTX_ROCE_EDPM_ENABLE_MASK 0x1 +#define XSTORM_RDMA_CONN_AG_CTX_ROCE_EDPM_ENABLE_SHIFT 5 +#define XSTORM_RDMA_CONN_AG_CTX_CF23_MASK 0x3 +#define XSTORM_RDMA_CONN_AG_CTX_CF23_SHIFT 6 + u8 byte2; + __le16 physical_q0; + __le16 word1; + __le16 word2; + __le16 word3; + __le16 word4; + __le16 word5; + __le16 conn_dpi; + u8 byte3; + u8 byte4; + u8 byte5; + u8 byte6; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 snd_nxt_psn; + __le32 reg4; + __le32 reg5; + __le32 reg6; +}; + +struct ystorm_rdma_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define YSTORM_RDMA_CONN_AG_CTX_BIT0_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_BIT0_SHIFT 0 +#define YSTORM_RDMA_CONN_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_BIT1_SHIFT 1 +#define YSTORM_RDMA_CONN_AG_CTX_CF0_MASK 0x3 +#define YSTORM_RDMA_CONN_AG_CTX_CF0_SHIFT 2 +#define YSTORM_RDMA_CONN_AG_CTX_CF1_MASK 0x3 +#define YSTORM_RDMA_CONN_AG_CTX_CF1_SHIFT 4 +#define YSTORM_RDMA_CONN_AG_CTX_CF2_MASK 0x3 +#define YSTORM_RDMA_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define YSTORM_RDMA_CONN_AG_CTX_CF0EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_CF0EN_SHIFT 0 +#define YSTORM_RDMA_CONN_AG_CTX_CF1EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_CF1EN_SHIFT 1 +#define YSTORM_RDMA_CONN_AG_CTX_CF2EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_CF2EN_SHIFT 2 +#define YSTORM_RDMA_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define YSTORM_RDMA_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define YSTORM_RDMA_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define YSTORM_RDMA_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define YSTORM_RDMA_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_RDMA_CONN_AG_CTX_RULE4EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le32 reg0; + __le32 reg1; + __le16 word1; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg2; + __le32 reg3; +}; + +struct mstorm_roce_conn_st_ctx { + struct regpair temp[6]; +}; + +struct pstorm_roce_conn_st_ctx { + struct regpair temp[16]; +}; + +struct ystorm_roce_conn_st_ctx { + struct regpair temp[2]; +}; + +struct xstorm_roce_conn_st_ctx { + struct regpair temp[22]; +}; + +struct tstorm_roce_conn_st_ctx { + struct regpair temp[30]; +}; + +struct ustorm_roce_conn_st_ctx { + struct regpair temp[12]; +}; + +struct roce_conn_context { + struct ystorm_roce_conn_st_ctx ystorm_st_context; + struct regpair ystorm_st_padding[2]; + struct pstorm_roce_conn_st_ctx pstorm_st_context; + struct xstorm_roce_conn_st_ctx xstorm_st_context; + struct regpair xstorm_st_padding[2]; + struct xstorm_rdma_conn_ag_ctx xstorm_ag_context; + struct tstorm_rdma_conn_ag_ctx tstorm_ag_context; + struct timers_context timer_context; + struct ustorm_rdma_conn_ag_ctx ustorm_ag_context; + struct tstorm_roce_conn_st_ctx tstorm_st_context; + struct mstorm_roce_conn_st_ctx mstorm_st_context; + struct ustorm_roce_conn_st_ctx ustorm_st_context; + struct regpair ustorm_st_padding[2]; +}; + +struct roce_create_qp_req_ramrod_data { + __le16 flags; +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_ROCE_FLAVOR_MASK 0x3 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_ROCE_FLAVOR_SHIFT 0 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_FMR_AND_RESERVED_EN_MASK 0x1 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_FMR_AND_RESERVED_EN_SHIFT 2 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_SIGNALED_COMP_MASK 0x1 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_SIGNALED_COMP_SHIFT 3 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_PRI_MASK 0x7 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_PRI_SHIFT 4 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_RESERVED_MASK 0x1 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_RESERVED_SHIFT 7 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_ERR_RETRY_CNT_MASK 0xF +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_ERR_RETRY_CNT_SHIFT 8 +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_RNR_NAK_CNT_MASK 0xF +#define ROCE_CREATE_QP_REQ_RAMROD_DATA_RNR_NAK_CNT_SHIFT 12 + u8 max_ord; + u8 traffic_class; + u8 hop_limit; + u8 orq_num_pages; + __le16 p_key; + __le32 flow_label; + __le32 dst_qp_id; + __le32 ack_timeout_val; + __le32 initial_psn; + __le16 mtu; + __le16 pd; + __le16 sq_num_pages; + __le16 reseved2; + struct regpair sq_pbl_addr; + struct regpair orq_pbl_addr; + __le16 local_mac_addr[3]; + __le16 remote_mac_addr[3]; + __le16 vlan_id; + __le16 udp_src_port; + __le32 src_gid[4]; + __le32 dst_gid[4]; + struct regpair qp_handle_for_cqe; + struct regpair qp_handle_for_async; + u8 stats_counter_id; + u8 reserved3[7]; + __le32 cq_cid; + __le16 physical_queue0; + __le16 dpi; +}; + +struct roce_create_qp_resp_ramrod_data { + __le16 flags; +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_ROCE_FLAVOR_MASK 0x3 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_ROCE_FLAVOR_SHIFT 0 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RDMA_RD_EN_MASK 0x1 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RDMA_RD_EN_SHIFT 2 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RDMA_WR_EN_MASK 0x1 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RDMA_WR_EN_SHIFT 3 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_ATOMIC_EN_MASK 0x1 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_ATOMIC_EN_SHIFT 4 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_SRQ_FLG_MASK 0x1 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_SRQ_FLG_SHIFT 5 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_E2E_FLOW_CONTROL_EN_MASK 0x1 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_E2E_FLOW_CONTROL_EN_SHIFT 6 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RESERVED0_MASK 0x1 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_RESERVED0_SHIFT 7 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_PRI_MASK 0x7 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_PRI_SHIFT 8 +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_MASK 0x1F +#define ROCE_CREATE_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_SHIFT 11 + u8 max_ird; + u8 traffic_class; + u8 hop_limit; + u8 irq_num_pages; + __le16 p_key; + __le32 flow_label; + __le32 dst_qp_id; + u8 stats_counter_id; + u8 reserved1; + __le16 mtu; + __le32 initial_psn; + __le16 pd; + __le16 rq_num_pages; + struct rdma_srq_id srq_id; + struct regpair rq_pbl_addr; + struct regpair irq_pbl_addr; + __le16 local_mac_addr[3]; + __le16 remote_mac_addr[3]; + __le16 vlan_id; + __le16 udp_src_port; + __le32 src_gid[4]; + __le32 dst_gid[4]; + struct regpair qp_handle_for_cqe; + struct regpair qp_handle_for_async; + __le32 reserved2[2]; + __le32 cq_cid; + __le16 physical_queue0; + __le16 dpi; +}; + +struct roce_destroy_qp_req_output_params { + __le32 num_bound_mw; + __le32 reserved; +}; + +struct roce_destroy_qp_req_ramrod_data { + struct regpair output_params_addr; +}; + +struct roce_destroy_qp_resp_output_params { + __le32 num_invalidated_mw; + __le32 reserved; +}; + +struct roce_destroy_qp_resp_ramrod_data { + struct regpair output_params_addr; +}; + +enum roce_event_opcode { + ROCE_EVENT_CREATE_QP = 11, + ROCE_EVENT_MODIFY_QP, + ROCE_EVENT_QUERY_QP, + ROCE_EVENT_DESTROY_QP, + MAX_ROCE_EVENT_OPCODE +}; + +struct roce_modify_qp_req_ramrod_data { + __le16 flags; +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_MOVE_TO_ERR_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_MOVE_TO_ERR_FLG_SHIFT 0 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_MOVE_TO_SQD_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_MOVE_TO_SQD_FLG_SHIFT 1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_EN_SQD_ASYNC_NOTIFY_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_EN_SQD_ASYNC_NOTIFY_SHIFT 2 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_P_KEY_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_P_KEY_FLG_SHIFT 3 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ADDRESS_VECTOR_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ADDRESS_VECTOR_FLG_SHIFT 4 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_MAX_ORD_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_MAX_ORD_FLG_SHIFT 5 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RNR_NAK_CNT_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RNR_NAK_CNT_FLG_SHIFT 6 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ERR_RETRY_CNT_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ERR_RETRY_CNT_FLG_SHIFT 7 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ACK_TIMEOUT_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ACK_TIMEOUT_FLG_SHIFT 8 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PRI_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PRI_FLG_SHIFT 9 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PRI_MASK 0x7 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_PRI_SHIFT 10 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RESERVED1_MASK 0x7 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RESERVED1_SHIFT 13 + u8 fields; +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ERR_RETRY_CNT_MASK 0xF +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_ERR_RETRY_CNT_SHIFT 0 +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RNR_NAK_CNT_MASK 0xF +#define ROCE_MODIFY_QP_REQ_RAMROD_DATA_RNR_NAK_CNT_SHIFT 4 + u8 max_ord; + u8 traffic_class; + u8 hop_limit; + __le16 p_key; + __le32 flow_label; + __le32 ack_timeout_val; + __le16 mtu; + __le16 reserved2; + __le32 reserved3[3]; + __le32 src_gid[4]; + __le32 dst_gid[4]; +}; + +struct roce_modify_qp_resp_ramrod_data { + __le16 flags; +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MOVE_TO_ERR_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MOVE_TO_ERR_FLG_SHIFT 0 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_RD_EN_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_RD_EN_SHIFT 1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_WR_EN_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_WR_EN_SHIFT 2 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_ATOMIC_EN_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_ATOMIC_EN_SHIFT 3 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_P_KEY_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_P_KEY_FLG_SHIFT 4 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_ADDRESS_VECTOR_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_ADDRESS_VECTOR_FLG_SHIFT 5 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MAX_IRD_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MAX_IRD_FLG_SHIFT 6 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_PRI_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_PRI_FLG_SHIFT 7 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_FLG_SHIFT 8 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_OPS_EN_FLG_MASK 0x1 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RDMA_OPS_EN_FLG_SHIFT 9 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RESERVED1_MASK 0x3F +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_RESERVED1_SHIFT 10 + u8 fields; +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_PRI_MASK 0x7 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_PRI_SHIFT 0 +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_MASK 0x1F +#define ROCE_MODIFY_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER_SHIFT 3 + u8 max_ird; + u8 traffic_class; + u8 hop_limit; + __le16 p_key; + __le32 flow_label; + __le16 mtu; + __le16 reserved2; + __le32 src_gid[4]; + __le32 dst_gid[4]; +}; + +struct roce_query_qp_req_output_params { + __le32 psn; + __le32 flags; +#define ROCE_QUERY_QP_REQ_OUTPUT_PARAMS_ERR_FLG_MASK 0x1 +#define ROCE_QUERY_QP_REQ_OUTPUT_PARAMS_ERR_FLG_SHIFT 0 +#define ROCE_QUERY_QP_REQ_OUTPUT_PARAMS_SQ_DRAINING_FLG_MASK 0x1 +#define ROCE_QUERY_QP_REQ_OUTPUT_PARAMS_SQ_DRAINING_FLG_SHIFT 1 +#define ROCE_QUERY_QP_REQ_OUTPUT_PARAMS_RESERVED0_MASK 0x3FFFFFFF +#define ROCE_QUERY_QP_REQ_OUTPUT_PARAMS_RESERVED0_SHIFT 2 +}; + +struct roce_query_qp_req_ramrod_data { + struct regpair output_params_addr; +}; + +struct roce_query_qp_resp_output_params { + __le32 psn; + __le32 err_flag; +#define ROCE_QUERY_QP_RESP_OUTPUT_PARAMS_ERROR_FLG_MASK 0x1 +#define ROCE_QUERY_QP_RESP_OUTPUT_PARAMS_ERROR_FLG_SHIFT 0 +#define ROCE_QUERY_QP_RESP_OUTPUT_PARAMS_RESERVED0_MASK 0x7FFFFFFF +#define ROCE_QUERY_QP_RESP_OUTPUT_PARAMS_RESERVED0_SHIFT 1 +}; + +struct roce_query_qp_resp_ramrod_data { + struct regpair output_params_addr; +}; + +enum roce_ramrod_cmd_id { + ROCE_RAMROD_CREATE_QP = 11, + ROCE_RAMROD_MODIFY_QP, + ROCE_RAMROD_QUERY_QP, + ROCE_RAMROD_DESTROY_QP, + MAX_ROCE_RAMROD_CMD_ID +}; + +struct mstorm_roce_req_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define MSTORM_ROCE_REQ_CONN_AG_CTX_BIT0_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_BIT0_SHIFT 0 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_BIT1_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_BIT1_SHIFT 1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF0_MASK 0x3 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF0_SHIFT 2 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF1_MASK 0x3 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF1_SHIFT 4 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF2_MASK 0x3 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_SHIFT 0 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_SHIFT 1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_SHIFT 2 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define MSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_SHIFT 7 + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; +}; + +struct mstorm_roce_resp_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define MSTORM_ROCE_RESP_CONN_AG_CTX_BIT0_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_BIT0_SHIFT 0 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_BIT1_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_BIT1_SHIFT 1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF0_MASK 0x3 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF0_SHIFT 2 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF1_MASK 0x3 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF1_SHIFT 4 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF2_MASK 0x3 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_SHIFT 0 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_SHIFT 1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_SHIFT 2 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define MSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_SHIFT 7 + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; +}; + +enum roce_flavor { + PLAIN_ROCE /* RoCE v1 */ , + RROCE_IPV4 /* RoCE v2 (Routable RoCE) over ipv4 */ , + RROCE_IPV6 /* RoCE v2 (Routable RoCE) over ipv6 */ , + MAX_ROCE_FLAVOR +}; + +struct tstorm_roce_req_conn_ag_ctx { + u8 reserved0; + u8 state; + u8 flags0; +#define TSTORM_ROCE_REQ_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RX_ERROR_OCCURED_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RX_ERROR_OCCURED_SHIFT 1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TX_CQE_ERROR_OCCURED_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TX_CQE_ERROR_OCCURED_SHIFT 2 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_BIT3_SHIFT 3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_MSTORM_FLUSH_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_MSTORM_FLUSH_SHIFT 4 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_CACHED_ORQ_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_CACHED_ORQ_SHIFT 5 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_CF_SHIFT 6 + u8 flags1; +#define TSTORM_ROCE_REQ_CONN_AG_CTX_CF1_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_CF1_SHIFT 0 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_SQ_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_SQ_CF_SHIFT 2 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_STOP_ALL_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_STOP_ALL_CF_SHIFT 4 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 6 + u8 flags2; +#define TSTORM_ROCE_REQ_CONN_AG_CTX_MSTORM_FLUSH_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_MSTORM_FLUSH_CF_SHIFT 0 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SET_TIMER_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SET_TIMER_CF_SHIFT 2 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TX_ASYNC_ERROR_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TX_ASYNC_ERROR_CF_SHIFT 4 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RXMIT_DONE_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RXMIT_DONE_CF_SHIFT 6 + u8 flags3; +#define TSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_SCAN_COMPLETED_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_SCAN_COMPLETED_CF_SHIFT 0 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SQ_DRAIN_COMPLETED_CF_MASK 0x3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SQ_DRAIN_COMPLETED_CF_SHIFT 2 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_CF_EN_SHIFT 4 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_SHIFT 5 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_SQ_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_SQ_CF_EN_SHIFT 6 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_STOP_ALL_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TIMER_STOP_ALL_CF_EN_SHIFT 7 + u8 flags4; +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 0 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_MSTORM_FLUSH_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_MSTORM_FLUSH_CF_EN_SHIFT 1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SET_TIMER_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SET_TIMER_CF_EN_SHIFT 2 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TX_ASYNC_ERROR_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_TX_ASYNC_ERROR_CF_EN_SHIFT 3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RXMIT_DONE_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RXMIT_DONE_CF_EN_SHIFT 4 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_SCAN_COMPLETED_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_SCAN_COMPLETED_CF_EN_SHIFT 5 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SQ_DRAIN_COMPLETED_CF_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SQ_DRAIN_COMPLETED_CF_EN_SHIFT 6 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags5; +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SND_SQ_CONS_EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_SND_SQ_CONS_EN_SHIFT 5 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define TSTORM_ROCE_REQ_CONN_AG_CTX_RULE8EN_SHIFT 7 + __le32 reg0; + __le32 snd_nxt_psn; + __le32 snd_max_psn; + __le32 orq_prod; + __le32 reg4; + __le32 reg5; + __le32 reg6; + __le32 reg7; + __le32 reg8; + u8 tx_cqe_error_type; + u8 orq_cache_idx; + __le16 snd_sq_cons_th; + u8 byte4; + u8 byte5; + __le16 snd_sq_cons; + __le16 word2; + __le16 word3; + __le32 reg9; + __le32 reg10; +}; + +struct tstorm_roce_resp_conn_ag_ctx { + u8 byte0; + u8 state; + u8 flags0; +#define TSTORM_ROCE_RESP_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT1_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT1_SHIFT 1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT2_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT2_SHIFT 2 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT3_SHIFT 3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_MSTORM_FLUSH_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_MSTORM_FLUSH_SHIFT 4 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT5_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_BIT5_SHIFT 5 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF0_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF0_SHIFT 6 + u8 flags1; +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_SHIFT 0 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_TX_ERROR_CF_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_TX_ERROR_CF_SHIFT 2 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF3_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF3_SHIFT 4 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 6 + u8 flags2; +#define TSTORM_ROCE_RESP_CONN_AG_CTX_MSTORM_FLUSH_CF_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_MSTORM_FLUSH_CF_SHIFT 0 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF6_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF6_SHIFT 2 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF7_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF7_SHIFT 4 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF8_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF8_SHIFT 6 + u8 flags3; +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF9_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF9_SHIFT 0 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF10_MASK 0x3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF10_SHIFT 2 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_SHIFT 4 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_EN_SHIFT 5 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_TX_ERROR_CF_EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_TX_ERROR_CF_EN_SHIFT 6 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF3EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF3EN_SHIFT 7 + u8 flags4; +#define TSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 0 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_MSTORM_FLUSH_CF_EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_MSTORM_FLUSH_CF_EN_SHIFT 1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF6EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF6EN_SHIFT 2 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF7EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF7EN_SHIFT 3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF8EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF8EN_SHIFT 4 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF9EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF9EN_SHIFT 5 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF10EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_CF10EN_SHIFT 6 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags5; +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RQ_RULE_EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RQ_RULE_EN_SHIFT 5 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define TSTORM_ROCE_RESP_CONN_AG_CTX_RULE8EN_SHIFT 7 + __le32 psn_and_rxmit_id_echo; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le32 reg4; + __le32 reg5; + __le32 reg6; + __le32 reg7; + __le32 reg8; + u8 tx_async_error_type; + u8 byte3; + __le16 rq_cons; + u8 byte4; + u8 byte5; + __le16 rq_prod; + __le16 conn_dpi; + __le16 irq_cons; + __le32 num_invlidated_mw; + __le32 reg10; +}; + +struct ustorm_roce_req_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define USTORM_ROCE_REQ_CONN_AG_CTX_BIT0_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_BIT0_SHIFT 0 +#define USTORM_ROCE_REQ_CONN_AG_CTX_BIT1_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_BIT1_SHIFT 1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF0_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF0_SHIFT 2 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF1_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF1_SHIFT 4 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF2_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF3_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF3_SHIFT 0 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF4_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF4_SHIFT 2 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF5_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF5_SHIFT 4 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF6_MASK 0x3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF6_SHIFT 6 + u8 flags2; +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_SHIFT 0 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_SHIFT 1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_SHIFT 2 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF4EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF4EN_SHIFT 4 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF5EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF5EN_SHIFT 5 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF6EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_CF6EN_SHIFT 6 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags3; +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define USTORM_ROCE_REQ_CONN_AG_CTX_RULE8EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le16 word2; + __le16 word3; +}; + +struct ustorm_roce_resp_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define USTORM_ROCE_RESP_CONN_AG_CTX_BIT0_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_BIT0_SHIFT 0 +#define USTORM_ROCE_RESP_CONN_AG_CTX_BIT1_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_BIT1_SHIFT 1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF0_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF0_SHIFT 2 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF1_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF1_SHIFT 4 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF2_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF3_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF3_SHIFT 0 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF4_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF4_SHIFT 2 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF5_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF5_SHIFT 4 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF6_MASK 0x3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF6_SHIFT 6 + u8 flags2; +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_SHIFT 0 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_SHIFT 1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_SHIFT 2 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF4EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF4EN_SHIFT 4 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF5EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF5EN_SHIFT 5 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF6EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_CF6EN_SHIFT 6 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags3; +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define USTORM_ROCE_RESP_CONN_AG_CTX_RULE8EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le16 word2; + __le16 word3; +}; + +struct xstorm_roce_req_conn_ag_ctx { + u8 reserved0; + u8 state; + u8 flags0; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED1_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED1_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED2_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED2_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_EXIST_IN_QM3_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_EXIST_IN_QM3_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED3_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED3_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED4_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED4_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED5_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED5_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED6_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED6_SHIFT 7 + u8 flags1; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED7_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED7_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED8_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED8_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT10_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT10_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT11_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT11_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT12_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT12_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT13_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT13_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_STATE_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_ERROR_STATE_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_YSTORM_FLUSH_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_YSTORM_FLUSH_SHIFT 7 + u8 flags2; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF0_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF0_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF1_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF1_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF2_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF2_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF3_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF3_SHIFT 6 + u8 flags3; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SQ_FLUSH_CF_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SQ_FLUSH_CF_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RX_ERROR_CF_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RX_ERROR_CF_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SND_RXMIT_CF_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SND_RXMIT_CF_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 6 + u8 flags4; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF8_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF8_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF9_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF9_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF10_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF10_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF11_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF11_SHIFT 6 + u8 flags5; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF12_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF12_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF13_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF13_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FMR_ENDED_CF_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FMR_ENDED_CF_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF15_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF15_SHIFT 6 + u8 flags6; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF16_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF16_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF17_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF17_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF18_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF18_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF19_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF19_SHIFT 6 + u8 flags7; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF20_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF20_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF21_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF21_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SLOW_PATH_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SLOW_PATH_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_SHIFT 7 + u8 flags8; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF3EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF3EN_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SQ_FLUSH_CF_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SQ_FLUSH_CF_EN_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RX_ERROR_CF_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RX_ERROR_CF_EN_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SND_RXMIT_CF_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SND_RXMIT_CF_EN_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF8EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF8EN_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF9EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF9EN_SHIFT 7 + u8 flags9; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF10EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF10EN_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF11EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF11EN_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF12EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF12EN_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF13EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF13EN_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FME_ENDED_CF_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_FME_ENDED_CF_EN_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF15EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF15EN_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF16EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF16EN_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF17EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF17EN_SHIFT 7 + u8 flags10; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF18EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF18EN_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF19EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF19EN_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF20EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF20EN_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF21EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF21EN_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SLOW_PATH_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SLOW_PATH_EN_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF23EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF23EN_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_SHIFT 7 + u8 flags11; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE5EN_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE6EN_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_E2E_CREDIT_RULE_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_E2E_CREDIT_RULE_EN_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED1_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED1_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE9EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE9EN_SHIFT 7 + u8 flags12; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SQ_PROD_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_SQ_PROD_EN_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE11EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE11EN_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED2_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED2_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED3_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED3_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_INV_FENCE_RULE_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_INV_FENCE_RULE_EN_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE15EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE15EN_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_ORQ_FENCE_RULE_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_ORQ_FENCE_RULE_EN_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_MAX_ORD_RULE_EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_MAX_ORD_RULE_EN_SHIFT 7 + u8 flags13; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE18EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE18EN_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE19EN_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RULE19EN_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED4_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED4_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED5_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED5_SHIFT 3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED6_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED6_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED7_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED7_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED8_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED8_SHIFT 6 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED9_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_A0_RESERVED9_SHIFT 7 + u8 flags14; +#define XSTORM_ROCE_REQ_CONN_AG_CTX_MIGRATION_FLAG_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_MIGRATION_FLAG_SHIFT 0 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT17_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_BIT17_SHIFT 1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_DPM_PORT_NUM_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_DPM_PORT_NUM_SHIFT 2 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_RESERVED_SHIFT 4 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_ROCE_EDPM_ENABLE_MASK 0x1 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_ROCE_EDPM_ENABLE_SHIFT 5 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF23_MASK 0x3 +#define XSTORM_ROCE_REQ_CONN_AG_CTX_CF23_SHIFT 6 + u8 byte2; + __le16 physical_q0; + __le16 word1; + __le16 sq_cmp_cons; + __le16 sq_cons; + __le16 sq_prod; + __le16 word5; + __le16 conn_dpi; + u8 byte3; + u8 byte4; + u8 byte5; + u8 byte6; + __le32 lsn; + __le32 ssn; + __le32 snd_una_psn; + __le32 snd_nxt_psn; + __le32 reg4; + __le32 orq_cons_th; + __le32 orq_cons; +}; + +struct xstorm_roce_resp_conn_ag_ctx { + u8 reserved0; + u8 state; + u8 flags0; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED1_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED1_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED2_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED2_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_EXIST_IN_QM3_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_EXIST_IN_QM3_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED3_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED3_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED4_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED4_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED5_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED5_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED6_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED6_SHIFT 7 + u8 flags1; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED7_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED7_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED8_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RESERVED8_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT10_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT10_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT11_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT11_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT12_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT12_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT13_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT13_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_ERROR_STATE_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_ERROR_STATE_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_YSTORM_FLUSH_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_YSTORM_FLUSH_SHIFT 7 + u8 flags2; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF0_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF0_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF1_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF1_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF2_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF2_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF3_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF3_SHIFT 6 + u8 flags3; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RXMIT_CF_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RXMIT_CF_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FORCE_ACK_CF_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FORCE_ACK_CF_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_SHIFT 6 + u8 flags4; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF8_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF8_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF9_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF9_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF10_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF10_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF11_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF11_SHIFT 6 + u8 flags5; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF12_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF12_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF13_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF13_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF14_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF14_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF15_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF15_SHIFT 6 + u8 flags6; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF16_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF16_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF17_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF17_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF18_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF18_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF19_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF19_SHIFT 6 + u8 flags7; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF20_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF20_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF21_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF21_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_SLOW_PATH_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_SLOW_PATH_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_SHIFT 7 + u8 flags8; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF3EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF3EN_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RXMIT_CF_EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RXMIT_CF_EN_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RX_ERROR_CF_EN_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FORCE_ACK_CF_EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FORCE_ACK_CF_EN_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_FLUSH_Q0_CF_EN_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF8EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF8EN_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF9EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF9EN_SHIFT 7 + u8 flags9; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF10EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF10EN_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF11EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF11EN_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF12EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF12EN_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF13EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF13EN_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF14EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF14EN_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF15EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF15EN_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF16EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF16EN_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF17EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF17EN_SHIFT 7 + u8 flags10; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF18EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF18EN_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF19EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF19EN_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF20EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF20EN_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF21EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF21EN_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_SLOW_PATH_EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_SLOW_PATH_EN_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF23EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF23EN_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_SHIFT 7 + u8 flags11; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE5EN_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE6EN_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE7EN_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED1_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED1_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE9EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE9EN_SHIFT 7 + u8 flags12; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE10EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE10EN_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_IRQ_PROD_RULE_EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_IRQ_PROD_RULE_EN_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED2_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED2_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED3_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED3_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE14EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE14EN_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE15EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE15EN_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE16EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE16EN_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE17EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE17EN_SHIFT 7 + u8 flags13; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE18EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE18EN_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE19EN_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_RULE19EN_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED4_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED4_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED5_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED5_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED6_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED6_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED7_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED7_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED8_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED8_SHIFT 6 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED9_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_A0_RESERVED9_SHIFT 7 + u8 flags14; +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT16_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT16_SHIFT 0 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT17_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT17_SHIFT 1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT18_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT18_SHIFT 2 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT19_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT19_SHIFT 3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT20_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT20_SHIFT 4 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT21_MASK 0x1 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_BIT21_SHIFT 5 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF23_MASK 0x3 +#define XSTORM_ROCE_RESP_CONN_AG_CTX_CF23_SHIFT 6 + u8 byte2; + __le16 physical_q0; + __le16 word1; + __le16 irq_prod; + __le16 word3; + __le16 word4; + __le16 word5; + __le16 irq_cons; + u8 rxmit_opcode; + u8 byte4; + u8 byte5; + u8 byte6; + __le32 rxmit_psn_and_id; + __le32 rxmit_bytes_length; + __le32 psn; + __le32 reg3; + __le32 reg4; + __le32 reg5; + __le32 msn_and_syndrome; +}; + +struct ystorm_roce_req_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define YSTORM_ROCE_REQ_CONN_AG_CTX_BIT0_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_BIT0_SHIFT 0 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_BIT1_SHIFT 1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF0_MASK 0x3 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF0_SHIFT 2 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF1_MASK 0x3 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF1_SHIFT 4 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF2_MASK 0x3 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF0EN_SHIFT 0 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF1EN_SHIFT 1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_CF2EN_SHIFT 2 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_ROCE_REQ_CONN_AG_CTX_RULE4EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le32 reg0; + __le32 reg1; + __le16 word1; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg2; + __le32 reg3; +}; + +struct ystorm_roce_resp_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define YSTORM_ROCE_RESP_CONN_AG_CTX_BIT0_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_BIT0_SHIFT 0 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_BIT1_SHIFT 1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF0_MASK 0x3 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF0_SHIFT 2 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF1_MASK 0x3 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF1_SHIFT 4 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF2_MASK 0x3 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF0EN_SHIFT 0 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF1EN_SHIFT 1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_CF2EN_SHIFT 2 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_ROCE_RESP_CONN_AG_CTX_RULE4EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le32 reg0; + __le32 reg1; + __le16 word1; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg2; + __le32 reg3; +}; + +struct ystorm_iscsi_conn_st_ctx { + __le32 reserved[4]; +}; + +struct pstorm_iscsi_tcp_conn_st_ctx { + __le32 tcp[32]; + __le32 iscsi[4]; +}; + +struct xstorm_iscsi_tcp_conn_st_ctx { + __le32 reserved_iscsi[40]; + __le32 reserved_tcp[4]; +}; + +struct xstorm_iscsi_conn_ag_ctx { + u8 cdu_validation; + u8 state; + u8 flags0; +#define XSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM1_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM1_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_RESERVED1_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RESERVED1_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM3_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM3_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT4_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT4_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_RESERVED2_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RESERVED2_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT6_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT6_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT7_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT7_SHIFT 7 + u8 flags1; +#define XSTORM_ISCSI_CONN_AG_CTX_BIT8_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT8_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT9_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT9_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT10_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT10_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT11_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT11_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT12_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT12_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT13_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT13_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT14_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT14_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_TX_TRUNCATE_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_TX_TRUNCATE_SHIFT 7 + u8 flags2; +#define XSTORM_ISCSI_CONN_AG_CTX_CF0_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF0_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_CF1_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF1_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF2_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF2_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_SHIFT 6 + u8 flags3; +#define XSTORM_ISCSI_CONN_AG_CTX_CF4_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF4_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_CF5_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF5_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF6_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF6_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_CF7_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF7_SHIFT 6 + u8 flags4; +#define XSTORM_ISCSI_CONN_AG_CTX_CF8_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF8_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_CF9_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF9_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF10_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF10_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_CF11_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF11_SHIFT 6 + u8 flags5; +#define XSTORM_ISCSI_CONN_AG_CTX_CF12_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF12_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_CF13_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF13_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF14_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF14_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_UPDATE_STATE_TO_BASE_CF_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_UPDATE_STATE_TO_BASE_CF_SHIFT 6 + u8 flags6; +#define XSTORM_ISCSI_CONN_AG_CTX_CF16_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF16_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_CF17_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF17_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF18_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF18_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_DQ_FLUSH_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_DQ_FLUSH_SHIFT 6 + u8 flags7; +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q1_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q1_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_SLOW_PATH_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_SLOW_PATH_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_CF0EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF0EN_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_CF1EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF1EN_SHIFT 7 + u8 flags8; +#define XSTORM_ISCSI_CONN_AG_CTX_CF2EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF2EN_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_EN_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF4EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF4EN_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF5EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF5EN_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF6EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF6EN_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_CF7EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF7EN_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_CF8EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF8EN_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_CF9EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF9EN_SHIFT 7 + u8 flags9; +#define XSTORM_ISCSI_CONN_AG_CTX_CF10EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF10EN_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_CF11EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF11EN_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF12EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF12EN_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_CF13EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF13EN_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_CF14EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF14EN_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_UPDATE_STATE_TO_BASE_CF_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_UPDATE_STATE_TO_BASE_CF_EN_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_CF16EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF16EN_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_CF17EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF17EN_SHIFT 7 + u8 flags10; +#define XSTORM_ISCSI_CONN_AG_CTX_CF18EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_CF18EN_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_DQ_FLUSH_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_DQ_FLUSH_EN_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_EN_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q1_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q1_EN_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_SLOW_PATH_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_SLOW_PATH_EN_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_PROC_ONLY_CLEANUP_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_PROC_ONLY_CLEANUP_EN_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE0EN_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_MORE_TO_SEND_DEC_RULE_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_MORE_TO_SEND_DEC_RULE_EN_SHIFT 7 + u8 flags11; +#define XSTORM_ISCSI_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE2EN_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE3EN_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_RESERVED3_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RESERVED3_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE5EN_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE6EN_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE7EN_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED1_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED1_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE9EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE9EN_SHIFT 7 + u8 flags12; +#define XSTORM_ISCSI_CONN_AG_CTX_SQ_DEC_RULE_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_SQ_DEC_RULE_EN_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE11EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE11EN_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED2_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED2_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED3_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED3_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE14EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE14EN_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE15EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE15EN_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE16EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE16EN_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE17EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_RULE17EN_SHIFT 7 + u8 flags13; +#define XSTORM_ISCSI_CONN_AG_CTX_R2TQ_DEC_RULE_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_R2TQ_DEC_RULE_EN_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_HQ_DEC_RULE_EN_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_HQ_DEC_RULE_EN_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED4_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED4_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED5_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED5_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED6_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED6_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED7_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED7_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED8_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED8_SHIFT 6 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED9_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_A0_RESERVED9_SHIFT 7 + u8 flags14; +#define XSTORM_ISCSI_CONN_AG_CTX_BIT16_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT16_SHIFT 0 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT17_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT17_SHIFT 1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT18_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT18_SHIFT 2 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT19_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT19_SHIFT 3 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT20_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_BIT20_SHIFT 4 +#define XSTORM_ISCSI_CONN_AG_CTX_DUMMY_READ_DONE_MASK 0x1 +#define XSTORM_ISCSI_CONN_AG_CTX_DUMMY_READ_DONE_SHIFT 5 +#define XSTORM_ISCSI_CONN_AG_CTX_PROC_ONLY_CLEANUP_MASK 0x3 +#define XSTORM_ISCSI_CONN_AG_CTX_PROC_ONLY_CLEANUP_SHIFT 6 + u8 byte2; + __le16 physical_q0; + __le16 physical_q1; + __le16 dummy_dorq_var; + __le16 sq_cons; + __le16 sq_prod; + __le16 word5; + __le16 slow_io_total_data_tx_update; + u8 byte3; + u8 byte4; + u8 byte5; + u8 byte6; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 more_to_send_seq; + __le32 reg4; + __le32 reg5; + __le32 hq_scan_next_relevant_ack; + __le16 r2tq_prod; + __le16 r2tq_cons; + __le16 hq_prod; + __le16 hq_cons; + __le32 remain_seq; + __le32 bytes_to_next_pdu; + __le32 hq_tcp_seq; + u8 byte7; + u8 byte8; + u8 byte9; + u8 byte10; + u8 byte11; + u8 byte12; + u8 byte13; + u8 byte14; + u8 byte15; + u8 byte16; + __le16 word11; + __le32 reg10; + __le32 reg11; + __le32 exp_stat_sn; + __le32 reg13; + __le32 reg14; + __le32 reg15; + __le32 reg16; + __le32 reg17; +}; + +struct tstorm_iscsi_conn_ag_ctx { + u8 reserved0; + u8 state; + u8 flags0; +#define TSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_EXIST_IN_QM0_SHIFT 0 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT1_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT1_SHIFT 1 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT2_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT2_SHIFT 2 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT3_SHIFT 3 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT4_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT4_SHIFT 4 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT5_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_BIT5_SHIFT 5 +#define TSTORM_ISCSI_CONN_AG_CTX_CF0_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF0_SHIFT 6 + u8 flags1; +#define TSTORM_ISCSI_CONN_AG_CTX_CF1_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF1_SHIFT 0 +#define TSTORM_ISCSI_CONN_AG_CTX_CF2_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF2_SHIFT 2 +#define TSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_SHIFT 4 +#define TSTORM_ISCSI_CONN_AG_CTX_CF4_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF4_SHIFT 6 + u8 flags2; +#define TSTORM_ISCSI_CONN_AG_CTX_CF5_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF5_SHIFT 0 +#define TSTORM_ISCSI_CONN_AG_CTX_CF6_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF6_SHIFT 2 +#define TSTORM_ISCSI_CONN_AG_CTX_CF7_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF7_SHIFT 4 +#define TSTORM_ISCSI_CONN_AG_CTX_CF8_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF8_SHIFT 6 + u8 flags3; +#define TSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_SHIFT 0 +#define TSTORM_ISCSI_CONN_AG_CTX_CF10_MASK 0x3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF10_SHIFT 2 +#define TSTORM_ISCSI_CONN_AG_CTX_CF0EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF0EN_SHIFT 4 +#define TSTORM_ISCSI_CONN_AG_CTX_CF1EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF1EN_SHIFT 5 +#define TSTORM_ISCSI_CONN_AG_CTX_CF2EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF2EN_SHIFT 6 +#define TSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_TIMER_STOP_ALL_EN_SHIFT 7 + u8 flags4; +#define TSTORM_ISCSI_CONN_AG_CTX_CF4EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF4EN_SHIFT 0 +#define TSTORM_ISCSI_CONN_AG_CTX_CF5EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF5EN_SHIFT 1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF6EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF6EN_SHIFT 2 +#define TSTORM_ISCSI_CONN_AG_CTX_CF7EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF7EN_SHIFT 3 +#define TSTORM_ISCSI_CONN_AG_CTX_CF8EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF8EN_SHIFT 4 +#define TSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_FLUSH_Q0_EN_SHIFT 5 +#define TSTORM_ISCSI_CONN_AG_CTX_CF10EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_CF10EN_SHIFT 6 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags5; +#define TSTORM_ISCSI_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define TSTORM_ISCSI_CONN_AG_CTX_RULE8EN_SHIFT 7 + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le32 reg4; + __le32 reg5; + __le32 reg6; + __le32 reg7; + __le32 reg8; + u8 byte2; + u8 byte3; + __le16 word0; +}; + +struct ustorm_iscsi_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define USTORM_ISCSI_CONN_AG_CTX_BIT0_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_BIT0_SHIFT 0 +#define USTORM_ISCSI_CONN_AG_CTX_BIT1_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_BIT1_SHIFT 1 +#define USTORM_ISCSI_CONN_AG_CTX_CF0_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF0_SHIFT 2 +#define USTORM_ISCSI_CONN_AG_CTX_CF1_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF1_SHIFT 4 +#define USTORM_ISCSI_CONN_AG_CTX_CF2_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define USTORM_ISCSI_CONN_AG_CTX_CF3_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF3_SHIFT 0 +#define USTORM_ISCSI_CONN_AG_CTX_CF4_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF4_SHIFT 2 +#define USTORM_ISCSI_CONN_AG_CTX_CF5_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF5_SHIFT 4 +#define USTORM_ISCSI_CONN_AG_CTX_CF6_MASK 0x3 +#define USTORM_ISCSI_CONN_AG_CTX_CF6_SHIFT 6 + u8 flags2; +#define USTORM_ISCSI_CONN_AG_CTX_CF0EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF0EN_SHIFT 0 +#define USTORM_ISCSI_CONN_AG_CTX_CF1EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF1EN_SHIFT 1 +#define USTORM_ISCSI_CONN_AG_CTX_CF2EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF2EN_SHIFT 2 +#define USTORM_ISCSI_CONN_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_ISCSI_CONN_AG_CTX_CF4EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF4EN_SHIFT 4 +#define USTORM_ISCSI_CONN_AG_CTX_CF5EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF5EN_SHIFT 5 +#define USTORM_ISCSI_CONN_AG_CTX_CF6EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_CF6EN_SHIFT 6 +#define USTORM_ISCSI_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE0EN_SHIFT 7 + u8 flags3; +#define USTORM_ISCSI_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE1EN_SHIFT 0 +#define USTORM_ISCSI_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE2EN_SHIFT 1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE3EN_SHIFT 2 +#define USTORM_ISCSI_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE4EN_SHIFT 3 +#define USTORM_ISCSI_CONN_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE5EN_SHIFT 4 +#define USTORM_ISCSI_CONN_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE6EN_SHIFT 5 +#define USTORM_ISCSI_CONN_AG_CTX_RULE7EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE7EN_SHIFT 6 +#define USTORM_ISCSI_CONN_AG_CTX_RULE8EN_MASK 0x1 +#define USTORM_ISCSI_CONN_AG_CTX_RULE8EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; + __le32 reg2; + __le32 reg3; + __le16 word2; + __le16 word3; +}; + +struct tstorm_iscsi_conn_st_ctx { + __le32 reserved[40]; +}; + +struct mstorm_iscsi_conn_ag_ctx { + u8 reserved; + u8 state; + u8 flags0; +#define MSTORM_ISCSI_CONN_AG_CTX_BIT0_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_BIT0_SHIFT 0 +#define MSTORM_ISCSI_CONN_AG_CTX_BIT1_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_BIT1_SHIFT 1 +#define MSTORM_ISCSI_CONN_AG_CTX_CF0_MASK 0x3 +#define MSTORM_ISCSI_CONN_AG_CTX_CF0_SHIFT 2 +#define MSTORM_ISCSI_CONN_AG_CTX_CF1_MASK 0x3 +#define MSTORM_ISCSI_CONN_AG_CTX_CF1_SHIFT 4 +#define MSTORM_ISCSI_CONN_AG_CTX_CF2_MASK 0x3 +#define MSTORM_ISCSI_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define MSTORM_ISCSI_CONN_AG_CTX_CF0EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_CF0EN_SHIFT 0 +#define MSTORM_ISCSI_CONN_AG_CTX_CF1EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_CF1EN_SHIFT 1 +#define MSTORM_ISCSI_CONN_AG_CTX_CF2EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_CF2EN_SHIFT 2 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define MSTORM_ISCSI_CONN_AG_CTX_RULE4EN_SHIFT 7 + __le16 word0; + __le16 word1; + __le32 reg0; + __le32 reg1; +}; + +struct mstorm_iscsi_tcp_conn_st_ctx { + __le32 reserved_tcp[20]; + __le32 reserved_iscsi[8]; +}; + +struct ustorm_iscsi_conn_st_ctx { + __le32 reserved[52]; +}; + +struct iscsi_conn_context { + struct ystorm_iscsi_conn_st_ctx ystorm_st_context; + struct regpair ystorm_st_padding[2]; + struct pstorm_iscsi_tcp_conn_st_ctx pstorm_st_context; + struct regpair pstorm_st_padding[2]; + struct pb_context xpb2_context; + struct xstorm_iscsi_tcp_conn_st_ctx xstorm_st_context; + struct regpair xstorm_st_padding[2]; + struct xstorm_iscsi_conn_ag_ctx xstorm_ag_context; + struct tstorm_iscsi_conn_ag_ctx tstorm_ag_context; + struct regpair tstorm_ag_padding[2]; + struct timers_context timer_context; + struct ustorm_iscsi_conn_ag_ctx ustorm_ag_context; + struct pb_context upb_context; + struct tstorm_iscsi_conn_st_ctx tstorm_st_context; + struct regpair tstorm_st_padding[2]; + struct mstorm_iscsi_conn_ag_ctx mstorm_ag_context; + struct mstorm_iscsi_tcp_conn_st_ctx mstorm_st_context; + struct ustorm_iscsi_conn_st_ctx ustorm_st_context; +}; + +struct iscsi_init_ramrod_params { + struct iscsi_spe_func_init iscsi_init_spe; + struct tcp_init_params tcp_init; +}; + +struct ystorm_iscsi_conn_ag_ctx { + u8 byte0; + u8 byte1; + u8 flags0; +#define YSTORM_ISCSI_CONN_AG_CTX_BIT0_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_BIT0_SHIFT 0 +#define YSTORM_ISCSI_CONN_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_BIT1_SHIFT 1 +#define YSTORM_ISCSI_CONN_AG_CTX_CF0_MASK 0x3 +#define YSTORM_ISCSI_CONN_AG_CTX_CF0_SHIFT 2 +#define YSTORM_ISCSI_CONN_AG_CTX_CF1_MASK 0x3 +#define YSTORM_ISCSI_CONN_AG_CTX_CF1_SHIFT 4 +#define YSTORM_ISCSI_CONN_AG_CTX_CF2_MASK 0x3 +#define YSTORM_ISCSI_CONN_AG_CTX_CF2_SHIFT 6 + u8 flags1; +#define YSTORM_ISCSI_CONN_AG_CTX_CF0EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_CF0EN_SHIFT 0 +#define YSTORM_ISCSI_CONN_AG_CTX_CF1EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_CF1EN_SHIFT 1 +#define YSTORM_ISCSI_CONN_AG_CTX_CF2EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_CF2EN_SHIFT 2 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE0EN_SHIFT 3 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE1EN_SHIFT 4 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE2EN_SHIFT 5 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE3EN_SHIFT 6 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_ISCSI_CONN_AG_CTX_RULE4EN_SHIFT 7 + u8 byte2; + u8 byte3; + __le16 word0; + __le32 reg0; + __le32 reg1; + __le16 word1; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg2; + __le32 reg3; +}; #define VF_MAX_STATIC 192 #define MCP_GLOB_PATH_MAX 2 diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp.h b/drivers/net/ethernet/qlogic/qed/qed_sp.h index ea4e9ce53e0a..a548504c3420 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sp.h +++ b/drivers/net/ethernet/qlogic/qed/qed_sp.h @@ -63,6 +63,32 @@ union ramrod_data { struct vport_update_ramrod_data vport_update; struct vport_filter_update_ramrod_data vport_filter_update; + struct rdma_init_func_ramrod_data rdma_init_func; + struct rdma_close_func_ramrod_data rdma_close_func; + struct rdma_register_tid_ramrod_data rdma_register_tid; + struct rdma_deregister_tid_ramrod_data rdma_deregister_tid; + struct roce_create_qp_resp_ramrod_data roce_create_qp_resp; + struct roce_create_qp_req_ramrod_data roce_create_qp_req; + struct roce_modify_qp_resp_ramrod_data roce_modify_qp_resp; + struct roce_modify_qp_req_ramrod_data roce_modify_qp_req; + struct roce_query_qp_resp_ramrod_data roce_query_qp_resp; + struct roce_query_qp_req_ramrod_data roce_query_qp_req; + struct roce_destroy_qp_resp_ramrod_data roce_destroy_qp_resp; + struct roce_destroy_qp_req_ramrod_data roce_destroy_qp_req; + struct rdma_create_cq_ramrod_data rdma_create_cq; + struct rdma_resize_cq_ramrod_data rdma_resize_cq; + struct rdma_destroy_cq_ramrod_data rdma_destroy_cq; + struct rdma_srq_create_ramrod_data rdma_create_srq; + struct rdma_srq_destroy_ramrod_data rdma_destroy_srq; + struct rdma_srq_modify_ramrod_data rdma_modify_srq; + + struct iscsi_slow_path_hdr iscsi_empty; + struct iscsi_init_ramrod_params iscsi_init; + struct iscsi_spe_func_dstry iscsi_destroy; + struct iscsi_spe_conn_offload iscsi_conn_offload; + struct iscsi_conn_update_ramrod_params iscsi_conn_update; + struct iscsi_spe_conn_termination iscsi_conn_terminate; + struct vf_start_ramrod_data vf_start; struct vf_stop_ramrod_data vf_stop; }; diff --git a/include/linux/qed/common_hsi.h b/include/linux/qed/common_hsi.h index 285189a5ea6d..077a3b6cc80f 100644 --- a/include/linux/qed/common_hsi.h +++ b/include/linux/qed/common_hsi.h @@ -12,6 +12,7 @@ #define CORE_SPQE_PAGE_SIZE_BYTES 4096 #define X_FINAL_CLEANUP_AGG_INT 1 +#define NUM_OF_GLOBAL_QUEUES 128 /* Queue Zone sizes in bytes */ #define TSTORM_QZONE_SIZE 8 @@ -694,7 +695,10 @@ struct parsing_and_err_flags { #define PARSING_AND_ERR_FLAGS_TUNNELL4CHKSMERROR_SHIFT 15 }; -/* Concrete Function ID. */ +struct pb_context { + __le32 crc[4]; +}; + struct pxp_concrete_fid { __le16 fid; #define PXP_CONCRETE_FID_PFID_MASK 0xF @@ -761,6 +765,72 @@ struct pxp_ptt_entry { }; /* RSS hash type */ +struct rdif_task_context { + __le32 initial_ref_tag; + __le16 app_tag_value; + __le16 app_tag_mask; + u8 flags0; +#define RDIF_TASK_CONTEXT_IGNOREAPPTAG_MASK 0x1 +#define RDIF_TASK_CONTEXT_IGNOREAPPTAG_SHIFT 0 +#define RDIF_TASK_CONTEXT_INITIALREFTAGVALID_MASK 0x1 +#define RDIF_TASK_CONTEXT_INITIALREFTAGVALID_SHIFT 1 +#define RDIF_TASK_CONTEXT_HOSTGUARDTYPE_MASK 0x1 +#define RDIF_TASK_CONTEXT_HOSTGUARDTYPE_SHIFT 2 +#define RDIF_TASK_CONTEXT_SETERRORWITHEOP_MASK 0x1 +#define RDIF_TASK_CONTEXT_SETERRORWITHEOP_SHIFT 3 +#define RDIF_TASK_CONTEXT_PROTECTIONTYPE_MASK 0x3 +#define RDIF_TASK_CONTEXT_PROTECTIONTYPE_SHIFT 4 +#define RDIF_TASK_CONTEXT_CRC_SEED_MASK 0x1 +#define RDIF_TASK_CONTEXT_CRC_SEED_SHIFT 6 +#define RDIF_TASK_CONTEXT_KEEPREFTAGCONST_MASK 0x1 +#define RDIF_TASK_CONTEXT_KEEPREFTAGCONST_SHIFT 7 + u8 partial_dif_data[7]; + __le16 partial_crc_value; + __le16 partial_checksum_value; + __le32 offset_in_io; + __le16 flags1; +#define RDIF_TASK_CONTEXT_VALIDATEGUARD_MASK 0x1 +#define RDIF_TASK_CONTEXT_VALIDATEGUARD_SHIFT 0 +#define RDIF_TASK_CONTEXT_VALIDATEAPPTAG_MASK 0x1 +#define RDIF_TASK_CONTEXT_VALIDATEAPPTAG_SHIFT 1 +#define RDIF_TASK_CONTEXT_VALIDATEREFTAG_MASK 0x1 +#define RDIF_TASK_CONTEXT_VALIDATEREFTAG_SHIFT 2 +#define RDIF_TASK_CONTEXT_FORWARDGUARD_MASK 0x1 +#define RDIF_TASK_CONTEXT_FORWARDGUARD_SHIFT 3 +#define RDIF_TASK_CONTEXT_FORWARDAPPTAG_MASK 0x1 +#define RDIF_TASK_CONTEXT_FORWARDAPPTAG_SHIFT 4 +#define RDIF_TASK_CONTEXT_FORWARDREFTAG_MASK 0x1 +#define RDIF_TASK_CONTEXT_FORWARDREFTAG_SHIFT 5 +#define RDIF_TASK_CONTEXT_INTERVALSIZE_MASK 0x7 +#define RDIF_TASK_CONTEXT_INTERVALSIZE_SHIFT 6 +#define RDIF_TASK_CONTEXT_HOSTINTERFACE_MASK 0x3 +#define RDIF_TASK_CONTEXT_HOSTINTERFACE_SHIFT 9 +#define RDIF_TASK_CONTEXT_DIFBEFOREDATA_MASK 0x1 +#define RDIF_TASK_CONTEXT_DIFBEFOREDATA_SHIFT 11 +#define RDIF_TASK_CONTEXT_RESERVED0_MASK 0x1 +#define RDIF_TASK_CONTEXT_RESERVED0_SHIFT 12 +#define RDIF_TASK_CONTEXT_NETWORKINTERFACE_MASK 0x1 +#define RDIF_TASK_CONTEXT_NETWORKINTERFACE_SHIFT 13 +#define RDIF_TASK_CONTEXT_FORWARDAPPTAGWITHMASK_MASK 0x1 +#define RDIF_TASK_CONTEXT_FORWARDAPPTAGWITHMASK_SHIFT 14 +#define RDIF_TASK_CONTEXT_FORWARDREFTAGWITHMASK_MASK 0x1 +#define RDIF_TASK_CONTEXT_FORWARDREFTAGWITHMASK_SHIFT 15 + __le16 state; +#define RDIF_TASK_CONTEXT_RECEIVEDDIFBYTESLEFT_MASK 0xF +#define RDIF_TASK_CONTEXT_RECEIVEDDIFBYTESLEFT_SHIFT 0 +#define RDIF_TASK_CONTEXT_TRANSMITEDDIFBYTESLEFT_MASK 0xF +#define RDIF_TASK_CONTEXT_TRANSMITEDDIFBYTESLEFT_SHIFT 4 +#define RDIF_TASK_CONTEXT_ERRORINIO_MASK 0x1 +#define RDIF_TASK_CONTEXT_ERRORINIO_SHIFT 8 +#define RDIF_TASK_CONTEXT_CHECKSUMOVERFLOW_MASK 0x1 +#define RDIF_TASK_CONTEXT_CHECKSUMOVERFLOW_SHIFT 9 +#define RDIF_TASK_CONTEXT_REFTAGMASK_MASK 0xF +#define RDIF_TASK_CONTEXT_REFTAGMASK_SHIFT 10 +#define RDIF_TASK_CONTEXT_RESERVED1_MASK 0x3 +#define RDIF_TASK_CONTEXT_RESERVED1_SHIFT 14 + __le32 reserved2; +}; + enum rss_hash_type { RSS_HASH_TYPE_DEFAULT = 0, RSS_HASH_TYPE_IPV4 = 1, @@ -789,4 +859,122 @@ struct status_block { #define STATUS_BLOCK_ZERO_PAD3_SHIFT 24 }; +struct tdif_task_context { + __le32 initial_ref_tag; + __le16 app_tag_value; + __le16 app_tag_mask; + __le16 partial_crc_valueB; + __le16 partial_checksum_valueB; + __le16 stateB; +#define TDIF_TASK_CONTEXT_RECEIVEDDIFBYTESLEFTB_MASK 0xF +#define TDIF_TASK_CONTEXT_RECEIVEDDIFBYTESLEFTB_SHIFT 0 +#define TDIF_TASK_CONTEXT_TRANSMITEDDIFBYTESLEFTB_MASK 0xF +#define TDIF_TASK_CONTEXT_TRANSMITEDDIFBYTESLEFTB_SHIFT 4 +#define TDIF_TASK_CONTEXT_ERRORINIOB_MASK 0x1 +#define TDIF_TASK_CONTEXT_ERRORINIOB_SHIFT 8 +#define TDIF_TASK_CONTEXT_CHECKSUMOVERFLOW_MASK 0x1 +#define TDIF_TASK_CONTEXT_CHECKSUMOVERFLOW_SHIFT 9 +#define TDIF_TASK_CONTEXT_RESERVED0_MASK 0x3F +#define TDIF_TASK_CONTEXT_RESERVED0_SHIFT 10 + u8 reserved1; + u8 flags0; +#define TDIF_TASK_CONTEXT_IGNOREAPPTAG_MASK 0x1 +#define TDIF_TASK_CONTEXT_IGNOREAPPTAG_SHIFT 0 +#define TDIF_TASK_CONTEXT_INITIALREFTAGVALID_MASK 0x1 +#define TDIF_TASK_CONTEXT_INITIALREFTAGVALID_SHIFT 1 +#define TDIF_TASK_CONTEXT_HOSTGUARDTYPE_MASK 0x1 +#define TDIF_TASK_CONTEXT_HOSTGUARDTYPE_SHIFT 2 +#define TDIF_TASK_CONTEXT_SETERRORWITHEOP_MASK 0x1 +#define TDIF_TASK_CONTEXT_SETERRORWITHEOP_SHIFT 3 +#define TDIF_TASK_CONTEXT_PROTECTIONTYPE_MASK 0x3 +#define TDIF_TASK_CONTEXT_PROTECTIONTYPE_SHIFT 4 +#define TDIF_TASK_CONTEXT_CRC_SEED_MASK 0x1 +#define TDIF_TASK_CONTEXT_CRC_SEED_SHIFT 6 +#define TDIF_TASK_CONTEXT_RESERVED2_MASK 0x1 +#define TDIF_TASK_CONTEXT_RESERVED2_SHIFT 7 + __le32 flags1; +#define TDIF_TASK_CONTEXT_VALIDATEGUARD_MASK 0x1 +#define TDIF_TASK_CONTEXT_VALIDATEGUARD_SHIFT 0 +#define TDIF_TASK_CONTEXT_VALIDATEAPPTAG_MASK 0x1 +#define TDIF_TASK_CONTEXT_VALIDATEAPPTAG_SHIFT 1 +#define TDIF_TASK_CONTEXT_VALIDATEREFTAG_MASK 0x1 +#define TDIF_TASK_CONTEXT_VALIDATEREFTAG_SHIFT 2 +#define TDIF_TASK_CONTEXT_FORWARDGUARD_MASK 0x1 +#define TDIF_TASK_CONTEXT_FORWARDGUARD_SHIFT 3 +#define TDIF_TASK_CONTEXT_FORWARDAPPTAG_MASK 0x1 +#define TDIF_TASK_CONTEXT_FORWARDAPPTAG_SHIFT 4 +#define TDIF_TASK_CONTEXT_FORWARDREFTAG_MASK 0x1 +#define TDIF_TASK_CONTEXT_FORWARDREFTAG_SHIFT 5 +#define TDIF_TASK_CONTEXT_INTERVALSIZE_MASK 0x7 +#define TDIF_TASK_CONTEXT_INTERVALSIZE_SHIFT 6 +#define TDIF_TASK_CONTEXT_HOSTINTERFACE_MASK 0x3 +#define TDIF_TASK_CONTEXT_HOSTINTERFACE_SHIFT 9 +#define TDIF_TASK_CONTEXT_DIFBEFOREDATA_MASK 0x1 +#define TDIF_TASK_CONTEXT_DIFBEFOREDATA_SHIFT 11 +#define TDIF_TASK_CONTEXT_RESERVED3_MASK 0x1 +#define TDIF_TASK_CONTEXT_RESERVED3_SHIFT 12 +#define TDIF_TASK_CONTEXT_NETWORKINTERFACE_MASK 0x1 +#define TDIF_TASK_CONTEXT_NETWORKINTERFACE_SHIFT 13 +#define TDIF_TASK_CONTEXT_RECEIVEDDIFBYTESLEFTA_MASK 0xF +#define TDIF_TASK_CONTEXT_RECEIVEDDIFBYTESLEFTA_SHIFT 14 +#define TDIF_TASK_CONTEXT_TRANSMITEDDIFBYTESLEFTA_MASK 0xF +#define TDIF_TASK_CONTEXT_TRANSMITEDDIFBYTESLEFTA_SHIFT 18 +#define TDIF_TASK_CONTEXT_ERRORINIOA_MASK 0x1 +#define TDIF_TASK_CONTEXT_ERRORINIOA_SHIFT 22 +#define TDIF_TASK_CONTEXT_CHECKSUMOVERFLOWA_MASK 0x1 +#define TDIF_TASK_CONTEXT_CHECKSUMOVERFLOWA_SHIFT 23 +#define TDIF_TASK_CONTEXT_REFTAGMASK_MASK 0xF +#define TDIF_TASK_CONTEXT_REFTAGMASK_SHIFT 24 +#define TDIF_TASK_CONTEXT_FORWARDAPPTAGWITHMASK_MASK 0x1 +#define TDIF_TASK_CONTEXT_FORWARDAPPTAGWITHMASK_SHIFT 28 +#define TDIF_TASK_CONTEXT_FORWARDREFTAGWITHMASK_MASK 0x1 +#define TDIF_TASK_CONTEXT_FORWARDREFTAGWITHMASK_SHIFT 29 +#define TDIF_TASK_CONTEXT_KEEPREFTAGCONST_MASK 0x1 +#define TDIF_TASK_CONTEXT_KEEPREFTAGCONST_SHIFT 30 +#define TDIF_TASK_CONTEXT_RESERVED4_MASK 0x1 +#define TDIF_TASK_CONTEXT_RESERVED4_SHIFT 31 + __le32 offset_in_iob; + __le16 partial_crc_value_a; + __le16 partial_checksum_valuea_; + __le32 offset_in_ioa; + u8 partial_dif_data_a[8]; + u8 partial_dif_data_b[8]; +}; + +struct timers_context { + __le32 logical_client0; +#define TIMERS_CONTEXT_EXPIRATIONTIMELC0_MASK 0xFFFFFFF +#define TIMERS_CONTEXT_EXPIRATIONTIMELC0_SHIFT 0 +#define TIMERS_CONTEXT_VALIDLC0_MASK 0x1 +#define TIMERS_CONTEXT_VALIDLC0_SHIFT 28 +#define TIMERS_CONTEXT_ACTIVELC0_MASK 0x1 +#define TIMERS_CONTEXT_ACTIVELC0_SHIFT 29 +#define TIMERS_CONTEXT_RESERVED0_MASK 0x3 +#define TIMERS_CONTEXT_RESERVED0_SHIFT 30 + __le32 logical_client1; +#define TIMERS_CONTEXT_EXPIRATIONTIMELC1_MASK 0xFFFFFFF +#define TIMERS_CONTEXT_EXPIRATIONTIMELC1_SHIFT 0 +#define TIMERS_CONTEXT_VALIDLC1_MASK 0x1 +#define TIMERS_CONTEXT_VALIDLC1_SHIFT 28 +#define TIMERS_CONTEXT_ACTIVELC1_MASK 0x1 +#define TIMERS_CONTEXT_ACTIVELC1_SHIFT 29 +#define TIMERS_CONTEXT_RESERVED1_MASK 0x3 +#define TIMERS_CONTEXT_RESERVED1_SHIFT 30 + __le32 logical_client2; +#define TIMERS_CONTEXT_EXPIRATIONTIMELC2_MASK 0xFFFFFFF +#define TIMERS_CONTEXT_EXPIRATIONTIMELC2_SHIFT 0 +#define TIMERS_CONTEXT_VALIDLC2_MASK 0x1 +#define TIMERS_CONTEXT_VALIDLC2_SHIFT 28 +#define TIMERS_CONTEXT_ACTIVELC2_MASK 0x1 +#define TIMERS_CONTEXT_ACTIVELC2_SHIFT 29 +#define TIMERS_CONTEXT_RESERVED2_MASK 0x3 +#define TIMERS_CONTEXT_RESERVED2_SHIFT 30 + __le32 host_expiration_fields; +#define TIMERS_CONTEXT_HOSTEXPRIRATIONVALUE_MASK 0xFFFFFFF +#define TIMERS_CONTEXT_HOSTEXPRIRATIONVALUE_SHIFT 0 +#define TIMERS_CONTEXT_HOSTEXPRIRATIONVALID_MASK 0x1 +#define TIMERS_CONTEXT_HOSTEXPRIRATIONVALID_SHIFT 28 +#define TIMERS_CONTEXT_RESERVED3_MASK 0x7 +#define TIMERS_CONTEXT_RESERVED3_SHIFT 29 +}; #endif /* __COMMON_HSI__ */ diff --git a/include/linux/qed/iscsi_common.h b/include/linux/qed/iscsi_common.h new file mode 100644 index 000000000000..b3c0feb15ae9 --- /dev/null +++ b/include/linux/qed/iscsi_common.h @@ -0,0 +1,1439 @@ +/* QLogic qed NIC Driver + * Copyright (c) 2015 QLogic Corporation + * + * This software is available under the terms of the GNU General Public License + * (GPL) Version 2, available from the file COPYING in the main directory of + * this source tree. + */ + +#ifndef __ISCSI_COMMON__ +#define __ISCSI_COMMON__ +/**********************/ +/* ISCSI FW CONSTANTS */ +/**********************/ + +/* iSCSI HSI constants */ +#define ISCSI_DEFAULT_MTU (1500) + +/* Current iSCSI HSI version number composed of two fields (16 bit) */ +#define ISCSI_HSI_MAJOR_VERSION (0) +#define ISCSI_HSI_MINOR_VERSION (0) + +/* KWQ (kernel work queue) layer codes */ +#define ISCSI_SLOW_PATH_LAYER_CODE (6) + +/* CQE completion status */ +#define ISCSI_EQE_COMPLETION_SUCCESS (0x0) +#define ISCSI_EQE_RST_CONN_RCVD (0x1) + +/* iSCSI parameter defaults */ +#define ISCSI_DEFAULT_HEADER_DIGEST (0) +#define ISCSI_DEFAULT_DATA_DIGEST (0) +#define ISCSI_DEFAULT_INITIAL_R2T (1) +#define ISCSI_DEFAULT_IMMEDIATE_DATA (1) +#define ISCSI_DEFAULT_MAX_PDU_LENGTH (0x2000) +#define ISCSI_DEFAULT_FIRST_BURST_LENGTH (0x10000) +#define ISCSI_DEFAULT_MAX_BURST_LENGTH (0x40000) +#define ISCSI_DEFAULT_MAX_OUTSTANDING_R2T (1) + +/* iSCSI parameter limits */ +#define ISCSI_MIN_VAL_MAX_PDU_LENGTH (0x200) +#define ISCSI_MAX_VAL_MAX_PDU_LENGTH (0xffffff) +#define ISCSI_MIN_VAL_BURST_LENGTH (0x200) +#define ISCSI_MAX_VAL_BURST_LENGTH (0xffffff) +#define ISCSI_MIN_VAL_MAX_OUTSTANDING_R2T (1) +#define ISCSI_MAX_VAL_MAX_OUTSTANDING_R2T (0xff) + +/* iSCSI reserved params */ +#define ISCSI_ITT_ALL_ONES (0xffffffff) +#define ISCSI_TTT_ALL_ONES (0xffffffff) + +#define ISCSI_OPTION_1_OFF_CHIP_TCP 1 +#define ISCSI_OPTION_2_ON_CHIP_TCP 2 + +#define ISCSI_INITIATOR_MODE 0 +#define ISCSI_TARGET_MODE 1 + +/* iSCSI request op codes */ +#define ISCSI_OPCODE_NOP_OUT_NO_IMM (0) +#define ISCSI_OPCODE_NOP_OUT ( \ + ISCSI_OPCODE_NOP_OUT_NO_IMM | 0x40) +#define ISCSI_OPCODE_SCSI_CMD_NO_IMM (1) +#define ISCSI_OPCODE_SCSI_CMD ( \ + ISCSI_OPCODE_SCSI_CMD_NO_IMM | 0x40) +#define ISCSI_OPCODE_TMF_REQUEST_NO_IMM (2) +#define ISCSI_OPCODE_TMF_REQUEST ( \ + ISCSI_OPCODE_TMF_REQUEST_NO_IMM | 0x40) +#define ISCSI_OPCODE_LOGIN_REQUEST_NO_IMM (3) +#define ISCSI_OPCODE_LOGIN_REQUEST ( \ + ISCSI_OPCODE_LOGIN_REQUEST_NO_IMM | 0x40) +#define ISCSI_OPCODE_TEXT_REQUEST_NO_IMM (4) +#define ISCSI_OPCODE_TEXT_REQUEST ( \ + ISCSI_OPCODE_TEXT_REQUEST_NO_IMM | 0x40) +#define ISCSI_OPCODE_DATA_OUT (5) +#define ISCSI_OPCODE_LOGOUT_REQUEST_NO_IMM (6) +#define ISCSI_OPCODE_LOGOUT_REQUEST ( \ + ISCSI_OPCODE_LOGOUT_REQUEST_NO_IMM | 0x40) + +/* iSCSI response/messages op codes */ +#define ISCSI_OPCODE_NOP_IN (0x20) +#define ISCSI_OPCODE_SCSI_RESPONSE (0x21) +#define ISCSI_OPCODE_TMF_RESPONSE (0x22) +#define ISCSI_OPCODE_LOGIN_RESPONSE (0x23) +#define ISCSI_OPCODE_TEXT_RESPONSE (0x24) +#define ISCSI_OPCODE_DATA_IN (0x25) +#define ISCSI_OPCODE_LOGOUT_RESPONSE (0x26) +#define ISCSI_OPCODE_R2T (0x31) +#define ISCSI_OPCODE_ASYNC_MSG (0x32) +#define ISCSI_OPCODE_REJECT (0x3f) + +/* iSCSI stages */ +#define ISCSI_STAGE_SECURITY_NEGOTIATION (0) +#define ISCSI_STAGE_LOGIN_OPERATIONAL_NEGOTIATION (1) +#define ISCSI_STAGE_FULL_FEATURE_PHASE (3) + +/* iSCSI CQE errors */ +#define CQE_ERROR_BITMAP_DATA_DIGEST (0x08) +#define CQE_ERROR_BITMAP_RCV_ON_INVALID_CONN (0x10) +#define CQE_ERROR_BITMAP_DATA_TRUNCATED (0x20) + +struct cqe_error_bitmap { + u8 cqe_error_status_bits; +#define CQE_ERROR_BITMAP_DIF_ERR_BITS_MASK 0x7 +#define CQE_ERROR_BITMAP_DIF_ERR_BITS_SHIFT 0 +#define CQE_ERROR_BITMAP_DATA_DIGEST_ERR_MASK 0x1 +#define CQE_ERROR_BITMAP_DATA_DIGEST_ERR_SHIFT 3 +#define CQE_ERROR_BITMAP_RCV_ON_INVALID_CONN_MASK 0x1 +#define CQE_ERROR_BITMAP_RCV_ON_INVALID_CONN_SHIFT 4 +#define CQE_ERROR_BITMAP_DATA_TRUNCATED_ERR_MASK 0x1 +#define CQE_ERROR_BITMAP_DATA_TRUNCATED_ERR_SHIFT 5 +#define CQE_ERROR_BITMAP_UNDER_RUN_ERR_MASK 0x1 +#define CQE_ERROR_BITMAP_UNDER_RUN_ERR_SHIFT 6 +#define CQE_ERROR_BITMAP_RESERVED2_MASK 0x1 +#define CQE_ERROR_BITMAP_RESERVED2_SHIFT 7 +}; + +union cqe_error_status { + u8 error_status; + struct cqe_error_bitmap error_bits; +}; + +struct data_hdr { + __le32 data[12]; +}; + +struct iscsi_async_msg_hdr { + __le16 reserved0; + u8 flags_attr; +#define ISCSI_ASYNC_MSG_HDR_RSRV_MASK 0x7F +#define ISCSI_ASYNC_MSG_HDR_RSRV_SHIFT 0 +#define ISCSI_ASYNC_MSG_HDR_CONST1_MASK 0x1 +#define ISCSI_ASYNC_MSG_HDR_CONST1_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_ASYNC_MSG_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_ASYNC_MSG_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_ASYNC_MSG_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_ASYNC_MSG_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 all_ones; + __le32 reserved1; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le16 param1_rsrv; + u8 async_vcode; + u8 async_event; + __le16 param3_rsrv; + __le16 param2_rsrv; + __le32 reserved7; +}; + +struct iscsi_sge { + struct regpair sge_addr; + __le16 sge_len; + __le16 reserved0; + __le32 reserved1; +}; + +struct iscsi_cached_sge_ctx { + struct iscsi_sge sge; + struct regpair reserved; + __le32 dsgl_curr_offset[2]; +}; + +struct iscsi_cmd_hdr { + __le16 reserved1; + u8 flags_attr; +#define ISCSI_CMD_HDR_ATTR_MASK 0x7 +#define ISCSI_CMD_HDR_ATTR_SHIFT 0 +#define ISCSI_CMD_HDR_RSRV_MASK 0x3 +#define ISCSI_CMD_HDR_RSRV_SHIFT 3 +#define ISCSI_CMD_HDR_WRITE_MASK 0x1 +#define ISCSI_CMD_HDR_WRITE_SHIFT 5 +#define ISCSI_CMD_HDR_READ_MASK 0x1 +#define ISCSI_CMD_HDR_READ_SHIFT 6 +#define ISCSI_CMD_HDR_FINAL_MASK 0x1 +#define ISCSI_CMD_HDR_FINAL_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_CMD_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_CMD_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_CMD_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_CMD_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 expected_transfer_length; + __le32 cmd_sn; + __le32 exp_stat_sn; + __le32 cdb[4]; +}; + +struct iscsi_common_hdr { + u8 hdr_status; + u8 hdr_response; + u8 hdr_flags; + u8 hdr_first_byte; +#define ISCSI_COMMON_HDR_OPCODE_MASK 0x3F +#define ISCSI_COMMON_HDR_OPCODE_SHIFT 0 +#define ISCSI_COMMON_HDR_IMM_MASK 0x1 +#define ISCSI_COMMON_HDR_IMM_SHIFT 6 +#define ISCSI_COMMON_HDR_RSRV_MASK 0x1 +#define ISCSI_COMMON_HDR_RSRV_SHIFT 7 + __le32 hdr_second_dword; +#define ISCSI_COMMON_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_COMMON_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_COMMON_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_COMMON_HDR_TOTAL_AHS_LEN_SHIFT 24 + __le32 lun_reserved[4]; + __le32 data[6]; +}; + +struct iscsi_conn_offload_params { + struct regpair sq_pbl_addr; + struct regpair r2tq_pbl_addr; + struct regpair xhq_pbl_addr; + struct regpair uhq_pbl_addr; + __le32 initial_ack; + __le16 physical_q0; + __le16 physical_q1; + u8 flags; +#define ISCSI_CONN_OFFLOAD_PARAMS_TCP_ON_CHIP_1B_MASK 0x1 +#define ISCSI_CONN_OFFLOAD_PARAMS_TCP_ON_CHIP_1B_SHIFT 0 +#define ISCSI_CONN_OFFLOAD_PARAMS_TARGET_MODE_MASK 0x1 +#define ISCSI_CONN_OFFLOAD_PARAMS_TARGET_MODE_SHIFT 1 +#define ISCSI_CONN_OFFLOAD_PARAMS_RESERVED1_MASK 0x3F +#define ISCSI_CONN_OFFLOAD_PARAMS_RESERVED1_SHIFT 2 + u8 pbl_page_size_log; + u8 pbe_page_size_log; + u8 default_cq; + __le32 stat_sn; +}; + +struct iscsi_slow_path_hdr { + u8 op_code; + u8 flags; +#define ISCSI_SLOW_PATH_HDR_RESERVED0_MASK 0xF +#define ISCSI_SLOW_PATH_HDR_RESERVED0_SHIFT 0 +#define ISCSI_SLOW_PATH_HDR_LAYER_CODE_MASK 0x7 +#define ISCSI_SLOW_PATH_HDR_LAYER_CODE_SHIFT 4 +#define ISCSI_SLOW_PATH_HDR_RESERVED1_MASK 0x1 +#define ISCSI_SLOW_PATH_HDR_RESERVED1_SHIFT 7 +}; + +struct iscsi_conn_update_ramrod_params { + struct iscsi_slow_path_hdr hdr; + __le16 conn_id; + __le32 fw_cid; + u8 flags; +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_HD_EN_MASK 0x1 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_HD_EN_SHIFT 0 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_DD_EN_MASK 0x1 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_DD_EN_SHIFT 1 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_INITIAL_R2T_MASK 0x1 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_INITIAL_R2T_SHIFT 2 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_IMMEDIATE_DATA_MASK 0x1 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_IMMEDIATE_DATA_SHIFT 3 +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_RESERVED1_MASK 0xF +#define ISCSI_CONN_UPDATE_RAMROD_PARAMS_RESERVED1_SHIFT 4 + u8 reserved0[3]; + __le32 max_seq_size; + __le32 max_send_pdu_length; + __le32 max_recv_pdu_length; + __le32 first_seq_length; + __le32 exp_stat_sn; +}; + +struct iscsi_ext_cdb_cmd_hdr { + __le16 reserved1; + u8 flags_attr; +#define ISCSI_EXT_CDB_CMD_HDR_ATTR_MASK 0x7 +#define ISCSI_EXT_CDB_CMD_HDR_ATTR_SHIFT 0 +#define ISCSI_EXT_CDB_CMD_HDR_RSRV_MASK 0x3 +#define ISCSI_EXT_CDB_CMD_HDR_RSRV_SHIFT 3 +#define ISCSI_EXT_CDB_CMD_HDR_WRITE_MASK 0x1 +#define ISCSI_EXT_CDB_CMD_HDR_WRITE_SHIFT 5 +#define ISCSI_EXT_CDB_CMD_HDR_READ_MASK 0x1 +#define ISCSI_EXT_CDB_CMD_HDR_READ_SHIFT 6 +#define ISCSI_EXT_CDB_CMD_HDR_FINAL_MASK 0x1 +#define ISCSI_EXT_CDB_CMD_HDR_FINAL_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_EXT_CDB_CMD_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_EXT_CDB_CMD_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_EXT_CDB_CMD_HDR_CDB_SIZE_MASK 0xFF +#define ISCSI_EXT_CDB_CMD_HDR_CDB_SIZE_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 expected_transfer_length; + __le32 cmd_sn; + __le32 exp_stat_sn; + struct iscsi_sge cdb_sge; +}; + +struct iscsi_login_req_hdr { + u8 version_min; + u8 version_max; + u8 flags_attr; +#define ISCSI_LOGIN_REQ_HDR_NSG_MASK 0x3 +#define ISCSI_LOGIN_REQ_HDR_NSG_SHIFT 0 +#define ISCSI_LOGIN_REQ_HDR_CSG_MASK 0x3 +#define ISCSI_LOGIN_REQ_HDR_CSG_SHIFT 2 +#define ISCSI_LOGIN_REQ_HDR_RSRV_MASK 0x3 +#define ISCSI_LOGIN_REQ_HDR_RSRV_SHIFT 4 +#define ISCSI_LOGIN_REQ_HDR_C_MASK 0x1 +#define ISCSI_LOGIN_REQ_HDR_C_SHIFT 6 +#define ISCSI_LOGIN_REQ_HDR_T_MASK 0x1 +#define ISCSI_LOGIN_REQ_HDR_T_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_LOGIN_REQ_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_LOGIN_REQ_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_LOGIN_REQ_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_LOGIN_REQ_HDR_TOTAL_AHS_LEN_SHIFT 24 + __le32 isid_TABC; + __le16 tsih; + __le16 isid_d; + __le32 itt; + __le16 reserved1; + __le16 cid; + __le32 cmd_sn; + __le32 exp_stat_sn; + __le32 reserved2[4]; +}; + +struct iscsi_logout_req_hdr { + __le16 reserved0; + u8 reason_code; + u8 opcode; + __le32 reserved1; + __le32 reserved2[2]; + __le32 itt; + __le16 reserved3; + __le16 cid; + __le32 cmd_sn; + __le32 exp_stat_sn; + __le32 reserved4[4]; +}; + +struct iscsi_data_out_hdr { + __le16 reserved1; + u8 flags_attr; +#define ISCSI_DATA_OUT_HDR_RSRV_MASK 0x7F +#define ISCSI_DATA_OUT_HDR_RSRV_SHIFT 0 +#define ISCSI_DATA_OUT_HDR_FINAL_MASK 0x1 +#define ISCSI_DATA_OUT_HDR_FINAL_SHIFT 7 + u8 opcode; + __le32 reserved2; + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 reserved3; + __le32 exp_stat_sn; + __le32 reserved4; + __le32 data_sn; + __le32 buffer_offset; + __le32 reserved5; +}; + +struct iscsi_data_in_hdr { + u8 status_rsvd; + u8 reserved1; + u8 flags; +#define ISCSI_DATA_IN_HDR_STATUS_MASK 0x1 +#define ISCSI_DATA_IN_HDR_STATUS_SHIFT 0 +#define ISCSI_DATA_IN_HDR_UNDERFLOW_MASK 0x1 +#define ISCSI_DATA_IN_HDR_UNDERFLOW_SHIFT 1 +#define ISCSI_DATA_IN_HDR_OVERFLOW_MASK 0x1 +#define ISCSI_DATA_IN_HDR_OVERFLOW_SHIFT 2 +#define ISCSI_DATA_IN_HDR_RSRV_MASK 0x7 +#define ISCSI_DATA_IN_HDR_RSRV_SHIFT 3 +#define ISCSI_DATA_IN_HDR_ACK_MASK 0x1 +#define ISCSI_DATA_IN_HDR_ACK_SHIFT 6 +#define ISCSI_DATA_IN_HDR_FINAL_MASK 0x1 +#define ISCSI_DATA_IN_HDR_FINAL_SHIFT 7 + u8 opcode; + __le32 reserved2; + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 data_sn; + __le32 buffer_offset; + __le32 residual_count; +}; + +struct iscsi_r2t_hdr { + u8 reserved0[3]; + u8 opcode; + __le32 reserved2; + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 r2t_sn; + __le32 buffer_offset; + __le32 desired_data_trns_len; +}; + +struct iscsi_nop_out_hdr { + __le16 reserved1; + u8 flags_attr; +#define ISCSI_NOP_OUT_HDR_RSRV_MASK 0x7F +#define ISCSI_NOP_OUT_HDR_RSRV_SHIFT 0 +#define ISCSI_NOP_OUT_HDR_CONST1_MASK 0x1 +#define ISCSI_NOP_OUT_HDR_CONST1_SHIFT 7 + u8 opcode; + __le32 reserved2; + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 cmd_sn; + __le32 exp_stat_sn; + __le32 reserved3; + __le32 reserved4; + __le32 reserved5; + __le32 reserved6; +}; + +struct iscsi_nop_in_hdr { + __le16 reserved0; + u8 flags_attr; +#define ISCSI_NOP_IN_HDR_RSRV_MASK 0x7F +#define ISCSI_NOP_IN_HDR_RSRV_SHIFT 0 +#define ISCSI_NOP_IN_HDR_CONST1_MASK 0x1 +#define ISCSI_NOP_IN_HDR_CONST1_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_NOP_IN_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_NOP_IN_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_NOP_IN_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_NOP_IN_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 reserved5; + __le32 reserved6; + __le32 reserved7; +}; + +struct iscsi_login_response_hdr { + u8 version_active; + u8 version_max; + u8 flags_attr; +#define ISCSI_LOGIN_RESPONSE_HDR_NSG_MASK 0x3 +#define ISCSI_LOGIN_RESPONSE_HDR_NSG_SHIFT 0 +#define ISCSI_LOGIN_RESPONSE_HDR_CSG_MASK 0x3 +#define ISCSI_LOGIN_RESPONSE_HDR_CSG_SHIFT 2 +#define ISCSI_LOGIN_RESPONSE_HDR_RSRV_MASK 0x3 +#define ISCSI_LOGIN_RESPONSE_HDR_RSRV_SHIFT 4 +#define ISCSI_LOGIN_RESPONSE_HDR_C_MASK 0x1 +#define ISCSI_LOGIN_RESPONSE_HDR_C_SHIFT 6 +#define ISCSI_LOGIN_RESPONSE_HDR_T_MASK 0x1 +#define ISCSI_LOGIN_RESPONSE_HDR_T_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_LOGIN_RESPONSE_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_LOGIN_RESPONSE_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_LOGIN_RESPONSE_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_LOGIN_RESPONSE_HDR_TOTAL_AHS_LEN_SHIFT 24 + __le32 isid_TABC; + __le16 tsih; + __le16 isid_d; + __le32 itt; + __le32 reserved1; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le16 reserved2; + u8 status_detail; + u8 status_class; + __le32 reserved4[2]; +}; + +struct iscsi_logout_response_hdr { + u8 reserved1; + u8 response; + u8 flags; + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_LOGOUT_RESPONSE_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_LOGOUT_RESPONSE_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_LOGOUT_RESPONSE_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_LOGOUT_RESPONSE_HDR_TOTAL_AHS_LEN_SHIFT 24 + __le32 reserved2[2]; + __le32 itt; + __le32 reserved3; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 reserved4; + __le16 time2retain; + __le16 time2wait; + __le32 reserved5[1]; +}; + +struct iscsi_text_request_hdr { + __le16 reserved0; + u8 flags_attr; +#define ISCSI_TEXT_REQUEST_HDR_RSRV_MASK 0x3F +#define ISCSI_TEXT_REQUEST_HDR_RSRV_SHIFT 0 +#define ISCSI_TEXT_REQUEST_HDR_C_MASK 0x1 +#define ISCSI_TEXT_REQUEST_HDR_C_SHIFT 6 +#define ISCSI_TEXT_REQUEST_HDR_F_MASK 0x1 +#define ISCSI_TEXT_REQUEST_HDR_F_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_TEXT_REQUEST_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_TEXT_REQUEST_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_TEXT_REQUEST_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_TEXT_REQUEST_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 cmd_sn; + __le32 exp_stat_sn; + __le32 reserved4[4]; +}; + +struct iscsi_text_response_hdr { + __le16 reserved1; + u8 flags; +#define ISCSI_TEXT_RESPONSE_HDR_RSRV_MASK 0x3F +#define ISCSI_TEXT_RESPONSE_HDR_RSRV_SHIFT 0 +#define ISCSI_TEXT_RESPONSE_HDR_C_MASK 0x1 +#define ISCSI_TEXT_RESPONSE_HDR_C_SHIFT 6 +#define ISCSI_TEXT_RESPONSE_HDR_F_MASK 0x1 +#define ISCSI_TEXT_RESPONSE_HDR_F_SHIFT 7 + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_TEXT_RESPONSE_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_TEXT_RESPONSE_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_TEXT_RESPONSE_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_TEXT_RESPONSE_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 ttt; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 reserved4[3]; +}; + +struct iscsi_tmf_request_hdr { + __le16 reserved0; + u8 function; + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_TMF_REQUEST_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_TMF_REQUEST_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_TMF_REQUEST_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_TMF_REQUEST_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 rtt; + __le32 cmd_sn; + __le32 exp_stat_sn; + __le32 ref_cmd_sn; + __le32 exp_data_sn; + __le32 reserved4[2]; +}; + +struct iscsi_tmf_response_hdr { + u8 reserved2; + u8 hdr_response; + u8 hdr_flags; + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_TMF_RESPONSE_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_TMF_RESPONSE_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_TMF_RESPONSE_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_TMF_RESPONSE_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair reserved0; + __le32 itt; + __le32 rtt; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 reserved4[3]; +}; + +struct iscsi_response_hdr { + u8 hdr_status; + u8 hdr_response; + u8 hdr_flags; + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_RESPONSE_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_RESPONSE_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_RESPONSE_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_RESPONSE_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair lun; + __le32 itt; + __le32 snack_tag; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 exp_data_sn; + __le32 bi_residual_count; + __le32 residual_count; +}; + +struct iscsi_reject_hdr { + u8 reserved4; + u8 hdr_reason; + u8 hdr_flags; + u8 opcode; + __le32 hdr_second_dword; +#define ISCSI_REJECT_HDR_DATA_SEG_LEN_MASK 0xFFFFFF +#define ISCSI_REJECT_HDR_DATA_SEG_LEN_SHIFT 0 +#define ISCSI_REJECT_HDR_TOTAL_AHS_LEN_MASK 0xFF +#define ISCSI_REJECT_HDR_TOTAL_AHS_LEN_SHIFT 24 + struct regpair reserved0; + __le32 reserved1; + __le32 reserved2; + __le32 stat_sn; + __le32 exp_cmd_sn; + __le32 max_cmd_sn; + __le32 data_sn; + __le32 reserved3[2]; +}; + +union iscsi_task_hdr { + struct iscsi_common_hdr common; + struct data_hdr data; + struct iscsi_cmd_hdr cmd; + struct iscsi_ext_cdb_cmd_hdr ext_cdb_cmd; + struct iscsi_login_req_hdr login_req; + struct iscsi_logout_req_hdr logout_req; + struct iscsi_data_out_hdr data_out; + struct iscsi_data_in_hdr data_in; + struct iscsi_r2t_hdr r2t; + struct iscsi_nop_out_hdr nop_out; + struct iscsi_nop_in_hdr nop_in; + struct iscsi_login_response_hdr login_response; + struct iscsi_logout_response_hdr logout_response; + struct iscsi_text_request_hdr text_request; + struct iscsi_text_response_hdr text_response; + struct iscsi_tmf_request_hdr tmf_request; + struct iscsi_tmf_response_hdr tmf_response; + struct iscsi_response_hdr response; + struct iscsi_reject_hdr reject; + struct iscsi_async_msg_hdr async_msg; +}; + +struct iscsi_cqe_common { + __le16 conn_id; + u8 cqe_type; + union cqe_error_status error_bitmap; + __le32 reserved[3]; + union iscsi_task_hdr iscsi_hdr; +}; + +struct iscsi_cqe_solicited { + __le16 conn_id; + u8 cqe_type; + union cqe_error_status error_bitmap; + __le16 itid; + u8 task_type; + u8 fw_dbg_field; + __le32 reserved1[2]; + union iscsi_task_hdr iscsi_hdr; +}; + +struct iscsi_cqe_unsolicited { + __le16 conn_id; + u8 cqe_type; + union cqe_error_status error_bitmap; + __le16 reserved0; + u8 reserved1; + u8 unsol_cqe_type; + struct regpair rqe_opaque; + union iscsi_task_hdr iscsi_hdr; +}; + +union iscsi_cqe { + struct iscsi_cqe_common cqe_common; + struct iscsi_cqe_solicited cqe_solicited; + struct iscsi_cqe_unsolicited cqe_unsolicited; +}; + +enum iscsi_cqes_type { + ISCSI_CQE_TYPE_SOLICITED = 1, + ISCSI_CQE_TYPE_UNSOLICITED, + ISCSI_CQE_TYPE_SOLICITED_WITH_SENSE + , + ISCSI_CQE_TYPE_TASK_CLEANUP, + ISCSI_CQE_TYPE_DUMMY, + MAX_ISCSI_CQES_TYPE +}; + +enum iscsi_cqe_unsolicited_type { + ISCSI_CQE_UNSOLICITED_NONE, + ISCSI_CQE_UNSOLICITED_SINGLE, + ISCSI_CQE_UNSOLICITED_FIRST, + ISCSI_CQE_UNSOLICITED_MIDDLE, + ISCSI_CQE_UNSOLICITED_LAST, + MAX_ISCSI_CQE_UNSOLICITED_TYPE +}; + +struct iscsi_virt_sgl_ctx { + struct regpair sgl_base; + struct regpair dsgl_base; + __le32 sgl_initial_offset; + __le32 dsgl_initial_offset; + __le32 dsgl_curr_offset[2]; +}; + +struct iscsi_sgl_var_params { + u8 sgl_ptr; + u8 dsgl_ptr; + __le16 sge_offset; + __le16 dsge_offset; +}; + +struct iscsi_phys_sgl_ctx { + struct regpair sgl_base; + struct regpair dsgl_base; + u8 sgl_size; + u8 dsgl_size; + __le16 reserved; + struct iscsi_sgl_var_params var_params[2]; +}; + +union iscsi_data_desc_ctx { + struct iscsi_virt_sgl_ctx virt_sgl; + struct iscsi_phys_sgl_ctx phys_sgl; + struct iscsi_cached_sge_ctx cached_sge; +}; + +struct iscsi_debug_modes { + u8 flags; +#define ISCSI_DEBUG_MODES_ASSERT_IF_RX_CONN_ERROR_MASK 0x1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RX_CONN_ERROR_SHIFT 0 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_RESET_MASK 0x1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_RESET_SHIFT 1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_FIN_MASK 0x1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_FIN_SHIFT 2 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_CLEANUP_MASK 0x1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_CLEANUP_SHIFT 3 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_REJECT_OR_ASYNC_MASK 0x1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_REJECT_OR_ASYNC_SHIFT 4 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_NOP_MASK 0x1 +#define ISCSI_DEBUG_MODES_ASSERT_IF_RECV_NOP_SHIFT 5 +#define ISCSI_DEBUG_MODES_RESERVED0_MASK 0x3 +#define ISCSI_DEBUG_MODES_RESERVED0_SHIFT 6 +}; + +struct iscsi_dif_flags { + u8 flags; +#define ISCSI_DIF_FLAGS_PROT_INTERVAL_SIZE_LOG_MASK 0xF +#define ISCSI_DIF_FLAGS_PROT_INTERVAL_SIZE_LOG_SHIFT 0 +#define ISCSI_DIF_FLAGS_DIF_TO_PEER_MASK 0x1 +#define ISCSI_DIF_FLAGS_DIF_TO_PEER_SHIFT 4 +#define ISCSI_DIF_FLAGS_HOST_INTERFACE_MASK 0x7 +#define ISCSI_DIF_FLAGS_HOST_INTERFACE_SHIFT 5 +}; + +enum iscsi_eqe_opcode { + ISCSI_EVENT_TYPE_INIT_FUNC = 0, + ISCSI_EVENT_TYPE_DESTROY_FUNC, + ISCSI_EVENT_TYPE_OFFLOAD_CONN, + ISCSI_EVENT_TYPE_UPDATE_CONN, + ISCSI_EVENT_TYPE_CLEAR_SQ, + ISCSI_EVENT_TYPE_TERMINATE_CONN, + ISCSI_EVENT_TYPE_ASYN_CONNECT_COMPLETE, + ISCSI_EVENT_TYPE_ASYN_TERMINATE_DONE, + RESERVED8, + RESERVED9, + ISCSI_EVENT_TYPE_START_OF_ERROR_TYPES = 10, + ISCSI_EVENT_TYPE_ASYN_ABORT_RCVD, + ISCSI_EVENT_TYPE_ASYN_CLOSE_RCVD, + ISCSI_EVENT_TYPE_ASYN_SYN_RCVD, + ISCSI_EVENT_TYPE_ASYN_MAX_RT_TIME, + ISCSI_EVENT_TYPE_ASYN_MAX_RT_CNT, + ISCSI_EVENT_TYPE_ASYN_MAX_KA_PROBES_CNT, + ISCSI_EVENT_TYPE_ASYN_FIN_WAIT2, + ISCSI_EVENT_TYPE_ISCSI_CONN_ERROR, + ISCSI_EVENT_TYPE_TCP_CONN_ERROR, + ISCSI_EVENT_TYPE_ASYN_DELETE_OOO_ISLES, + MAX_ISCSI_EQE_OPCODE +}; + +enum iscsi_error_types { + ISCSI_STATUS_NONE = 0, + ISCSI_CQE_ERROR_UNSOLICITED_RCV_ON_INVALID_CONN = 1, + ISCSI_CONN_ERROR_TASK_CID_MISMATCH, + ISCSI_CONN_ERROR_TASK_NOT_VALID, + ISCSI_CONN_ERROR_RQ_RING_IS_FULL, + ISCSI_CONN_ERROR_CMDQ_RING_IS_FULL, + ISCSI_CONN_ERROR_HQE_CACHING_FAILED, + ISCSI_CONN_ERROR_HEADER_DIGEST_ERROR, + ISCSI_CONN_ERROR_LOCAL_COMPLETION_ERROR, + ISCSI_CONN_ERROR_DATA_OVERRUN, + ISCSI_CONN_ERROR_OUT_OF_SGES_ERROR, + ISCSI_CONN_ERROR_TCP_SEG_PROC_URG_ERROR, + ISCSI_CONN_ERROR_TCP_SEG_PROC_IP_OPTIONS_ERROR, + ISCSI_CONN_ERROR_TCP_SEG_PROC_CONNECT_INVALID_WS_OPTION, + ISCSI_CONN_ERROR_TCP_IP_FRAGMENT_ERROR, + ISCSI_CONN_ERROR_PROTOCOL_ERR_AHS_LEN, + ISCSI_CONN_ERROR_PROTOCOL_ERR_AHS_TYPE, + ISCSI_CONN_ERROR_PROTOCOL_ERR_ITT_OUT_OF_RANGE, + ISCSI_CONN_ERROR_PROTOCOL_ERR_TTT_OUT_OF_RANGE, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DATA_SEG_LEN_EXCEEDS_PDU_SIZE, + ISCSI_CONN_ERROR_PROTOCOL_ERR_INVALID_OPCODE, + ISCSI_CONN_ERROR_PROTOCOL_ERR_INVALID_OPCODE_BEFORE_UPDATE, + ISCSI_CONN_ERROR_UNVALID_NOPIN_DSL, + ISCSI_CONN_ERROR_PROTOCOL_ERR_R2T_CARRIES_NO_DATA, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DATA_SN, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DATA_IN_TTT, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DATA_OUT_ITT, + ISCSI_CONN_ERROR_PROTOCOL_ERR_R2T_TTT, + ISCSI_CONN_ERROR_PROTOCOL_ERR_R2T_BUFFER_OFFSET, + ISCSI_CONN_ERROR_PROTOCOL_ERR_BUFFER_OFFSET_OOO, + ISCSI_CONN_ERROR_PROTOCOL_ERR_R2T_SN, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DESIRED_DATA_TRNS_LEN_0, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DESIRED_DATA_TRNS_LEN_1, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DESIRED_DATA_TRNS_LEN_2, + ISCSI_CONN_ERROR_PROTOCOL_ERR_LUN, + ISCSI_CONN_ERROR_PROTOCOL_ERR_F_BIT_ZERO, + ISCSI_CONN_ERROR_PROTOCOL_ERR_F_BIT_ZERO_S_BIT_ONE, + ISCSI_CONN_ERROR_PROTOCOL_ERR_EXP_STAT_SN, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DSL_NOT_ZERO, + ISCSI_CONN_ERROR_PROTOCOL_ERR_INVALID_DSL, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DATA_SEG_LEN_TOO_BIG, + ISCSI_CONN_ERROR_PROTOCOL_ERR_OUTSTANDING_R2T_COUNT, + ISCSI_CONN_ERROR_PROTOCOL_ERR_DIF_TX, + ISCSI_CONN_ERROR_SENSE_DATA_LENGTH, + ISCSI_CONN_ERROR_DATA_PLACEMENT_ERROR, + ISCSI_ERROR_UNKNOWN, + MAX_ISCSI_ERROR_TYPES +}; + +struct iscsi_mflags { + u8 mflags; +#define ISCSI_MFLAGS_SLOW_IO_MASK 0x1 +#define ISCSI_MFLAGS_SLOW_IO_SHIFT 0 +#define ISCSI_MFLAGS_SINGLE_SGE_MASK 0x1 +#define ISCSI_MFLAGS_SINGLE_SGE_SHIFT 1 +#define ISCSI_MFLAGS_RESERVED_MASK 0x3F +#define ISCSI_MFLAGS_RESERVED_SHIFT 2 +}; + +struct iscsi_sgl { + struct regpair sgl_addr; + __le16 updated_sge_size; + __le16 updated_sge_offset; + __le32 byte_offset; +}; + +union iscsi_mstorm_sgl { + struct iscsi_sgl sgl_struct; + struct iscsi_sge single_sge; +}; + +enum iscsi_ramrod_cmd_id { + ISCSI_RAMROD_CMD_ID_UNUSED = 0, + ISCSI_RAMROD_CMD_ID_INIT_FUNC = 1, + ISCSI_RAMROD_CMD_ID_DESTROY_FUNC = 2, + ISCSI_RAMROD_CMD_ID_OFFLOAD_CONN = 3, + ISCSI_RAMROD_CMD_ID_UPDATE_CONN = 4, + ISCSI_RAMROD_CMD_ID_TERMINATION_CONN = 5, + ISCSI_RAMROD_CMD_ID_CLEAR_SQ = 6, + MAX_ISCSI_RAMROD_CMD_ID +}; + +struct iscsi_reg1 { + __le32 reg1_map; +#define ISCSI_REG1_NUM_FAST_SGES_MASK 0x7 +#define ISCSI_REG1_NUM_FAST_SGES_SHIFT 0 +#define ISCSI_REG1_RESERVED1_MASK 0x1FFFFFFF +#define ISCSI_REG1_RESERVED1_SHIFT 3 +}; + +union iscsi_seq_num { + __le16 data_sn; + __le16 r2t_sn; +}; + +struct iscsi_spe_conn_offload { + struct iscsi_slow_path_hdr hdr; + __le16 conn_id; + __le32 fw_cid; + struct iscsi_conn_offload_params iscsi; + struct tcp_offload_params tcp; +}; + +struct iscsi_spe_conn_offload_option2 { + struct iscsi_slow_path_hdr hdr; + __le16 conn_id; + __le32 fw_cid; + struct iscsi_conn_offload_params iscsi; + struct tcp_offload_params_opt2 tcp; +}; + +struct iscsi_spe_conn_termination { + struct iscsi_slow_path_hdr hdr; + __le16 conn_id; + __le32 fw_cid; + u8 abortive; + u8 reserved0[7]; + struct regpair queue_cnts_addr; + struct regpair query_params_addr; +}; + +struct iscsi_spe_func_dstry { + struct iscsi_slow_path_hdr hdr; + __le16 reserved0; + __le32 reserved1; +}; + +struct iscsi_spe_func_init { + struct iscsi_slow_path_hdr hdr; + __le16 half_way_close_timeout; + u8 num_sq_pages_in_ring; + u8 num_r2tq_pages_in_ring; + u8 num_uhq_pages_in_ring; + u8 ll2_rx_queue_id; + u8 ooo_enable; + struct iscsi_debug_modes debug_mode; + __le16 reserved1; + __le32 reserved2; + __le32 reserved3; + __le32 reserved4; + struct scsi_init_func_params func_params; + struct scsi_init_func_queues q_params; +}; + +struct ystorm_iscsi_task_state { + union iscsi_data_desc_ctx sgl_ctx_union; + __le32 buffer_offset[2]; + __le16 bytes_nxt_dif; + __le16 rxmit_bytes_nxt_dif; + union iscsi_seq_num seq_num_union; + u8 dif_bytes_leftover; + u8 rxmit_dif_bytes_leftover; + __le16 reuse_count; + struct iscsi_dif_flags dif_flags; + u8 local_comp; + __le32 exp_r2t_sn; + __le32 sgl_offset[2]; +}; + +struct ystorm_iscsi_task_st_ctx { + struct ystorm_iscsi_task_state state; + union iscsi_task_hdr pdu_hdr; +}; + +struct ystorm_iscsi_task_ag_ctx { + u8 reserved; + u8 byte1; + __le16 word0; + u8 flags0; +#define YSTORM_ISCSI_TASK_AG_CTX_NIBBLE0_MASK 0xF +#define YSTORM_ISCSI_TASK_AG_CTX_NIBBLE0_SHIFT 0 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT0_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT0_SHIFT 4 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT1_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT1_SHIFT 5 +#define YSTORM_ISCSI_TASK_AG_CTX_VALID_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_VALID_SHIFT 6 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT3_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT3_SHIFT 7 + u8 flags1; +#define YSTORM_ISCSI_TASK_AG_CTX_CF0_MASK 0x3 +#define YSTORM_ISCSI_TASK_AG_CTX_CF0_SHIFT 0 +#define YSTORM_ISCSI_TASK_AG_CTX_CF1_MASK 0x3 +#define YSTORM_ISCSI_TASK_AG_CTX_CF1_SHIFT 2 +#define YSTORM_ISCSI_TASK_AG_CTX_CF2SPECIAL_MASK 0x3 +#define YSTORM_ISCSI_TASK_AG_CTX_CF2SPECIAL_SHIFT 4 +#define YSTORM_ISCSI_TASK_AG_CTX_CF0EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_CF0EN_SHIFT 6 +#define YSTORM_ISCSI_TASK_AG_CTX_CF1EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_CF1EN_SHIFT 7 + u8 flags2; +#define YSTORM_ISCSI_TASK_AG_CTX_BIT4_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_BIT4_SHIFT 0 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE0EN_SHIFT 1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE1EN_SHIFT 2 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE2EN_SHIFT 3 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE3EN_SHIFT 4 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE4EN_SHIFT 5 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE5EN_SHIFT 6 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE6EN_MASK 0x1 +#define YSTORM_ISCSI_TASK_AG_CTX_RULE6EN_SHIFT 7 + u8 byte2; + __le32 TTT; + u8 byte3; + u8 byte4; + __le16 word1; +}; + +struct mstorm_iscsi_task_ag_ctx { + u8 cdu_validation; + u8 byte1; + __le16 task_cid; + u8 flags0; +#define MSTORM_ISCSI_TASK_AG_CTX_CONNECTION_TYPE_MASK 0xF +#define MSTORM_ISCSI_TASK_AG_CTX_CONNECTION_TYPE_SHIFT 0 +#define MSTORM_ISCSI_TASK_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_EXIST_IN_QM0_SHIFT 4 +#define MSTORM_ISCSI_TASK_AG_CTX_BIT1_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_BIT1_SHIFT 5 +#define MSTORM_ISCSI_TASK_AG_CTX_VALID_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_VALID_SHIFT 6 +#define MSTORM_ISCSI_TASK_AG_CTX_TASK_CLEANUP_FLAG_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_TASK_CLEANUP_FLAG_SHIFT 7 + u8 flags1; +#define MSTORM_ISCSI_TASK_AG_CTX_TASK_CLEANUP_CF_MASK 0x3 +#define MSTORM_ISCSI_TASK_AG_CTX_TASK_CLEANUP_CF_SHIFT 0 +#define MSTORM_ISCSI_TASK_AG_CTX_CF1_MASK 0x3 +#define MSTORM_ISCSI_TASK_AG_CTX_CF1_SHIFT 2 +#define MSTORM_ISCSI_TASK_AG_CTX_CF2_MASK 0x3 +#define MSTORM_ISCSI_TASK_AG_CTX_CF2_SHIFT 4 +#define MSTORM_ISCSI_TASK_AG_CTX_TASK_CLEANUP_CF_EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_TASK_CLEANUP_CF_EN_SHIFT 6 +#define MSTORM_ISCSI_TASK_AG_CTX_CF1EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_CF1EN_SHIFT 7 + u8 flags2; +#define MSTORM_ISCSI_TASK_AG_CTX_CF2EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_CF2EN_SHIFT 0 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE0EN_SHIFT 1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE1EN_SHIFT 2 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE2EN_SHIFT 3 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE3EN_SHIFT 4 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE4EN_SHIFT 5 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE5EN_SHIFT 6 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE6EN_MASK 0x1 +#define MSTORM_ISCSI_TASK_AG_CTX_RULE6EN_SHIFT 7 + u8 byte2; + __le32 reg0; + u8 byte3; + u8 byte4; + __le16 word1; +}; + +struct ustorm_iscsi_task_ag_ctx { + u8 reserved; + u8 state; + __le16 icid; + u8 flags0; +#define USTORM_ISCSI_TASK_AG_CTX_CONNECTION_TYPE_MASK 0xF +#define USTORM_ISCSI_TASK_AG_CTX_CONNECTION_TYPE_SHIFT 0 +#define USTORM_ISCSI_TASK_AG_CTX_EXIST_IN_QM0_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_EXIST_IN_QM0_SHIFT 4 +#define USTORM_ISCSI_TASK_AG_CTX_BIT1_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_BIT1_SHIFT 5 +#define USTORM_ISCSI_TASK_AG_CTX_HQ_SCANNED_CF_MASK 0x3 +#define USTORM_ISCSI_TASK_AG_CTX_HQ_SCANNED_CF_SHIFT 6 + u8 flags1; +#define USTORM_ISCSI_TASK_AG_CTX_RESERVED1_MASK 0x3 +#define USTORM_ISCSI_TASK_AG_CTX_RESERVED1_SHIFT 0 +#define USTORM_ISCSI_TASK_AG_CTX_R2T2RECV_MASK 0x3 +#define USTORM_ISCSI_TASK_AG_CTX_R2T2RECV_SHIFT 2 +#define USTORM_ISCSI_TASK_AG_CTX_CF3_MASK 0x3 +#define USTORM_ISCSI_TASK_AG_CTX_CF3_SHIFT 4 +#define USTORM_ISCSI_TASK_AG_CTX_DIF_ERROR_CF_MASK 0x3 +#define USTORM_ISCSI_TASK_AG_CTX_DIF_ERROR_CF_SHIFT 6 + u8 flags2; +#define USTORM_ISCSI_TASK_AG_CTX_HQ_SCANNED_CF_EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_HQ_SCANNED_CF_EN_SHIFT 0 +#define USTORM_ISCSI_TASK_AG_CTX_DISABLE_DATA_ACKED_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_DISABLE_DATA_ACKED_SHIFT 1 +#define USTORM_ISCSI_TASK_AG_CTX_R2T2RECV_EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_R2T2RECV_EN_SHIFT 2 +#define USTORM_ISCSI_TASK_AG_CTX_CF3EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_CF3EN_SHIFT 3 +#define USTORM_ISCSI_TASK_AG_CTX_DIF_ERROR_CF_EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_DIF_ERROR_CF_EN_SHIFT 4 +#define USTORM_ISCSI_TASK_AG_CTX_CMP_DATA_TOTAL_EXP_EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_CMP_DATA_TOTAL_EXP_EN_SHIFT 5 +#define USTORM_ISCSI_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_RULE1EN_SHIFT 6 +#define USTORM_ISCSI_TASK_AG_CTX_CMP_CONT_RCV_EXP_EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_CMP_CONT_RCV_EXP_EN_SHIFT 7 + u8 flags3; +#define USTORM_ISCSI_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_RULE3EN_SHIFT 0 +#define USTORM_ISCSI_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_RULE4EN_SHIFT 1 +#define USTORM_ISCSI_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_RULE5EN_SHIFT 2 +#define USTORM_ISCSI_TASK_AG_CTX_RULE6EN_MASK 0x1 +#define USTORM_ISCSI_TASK_AG_CTX_RULE6EN_SHIFT 3 +#define USTORM_ISCSI_TASK_AG_CTX_DIF_ERROR_TYPE_MASK 0xF +#define USTORM_ISCSI_TASK_AG_CTX_DIF_ERROR_TYPE_SHIFT 4 + __le32 dif_err_intervals; + __le32 dif_error_1st_interval; + __le32 rcv_cont_len; + __le32 exp_cont_len; + __le32 total_data_acked; + __le32 exp_data_acked; + u8 next_tid_valid; + u8 byte3; + __le16 word1; + __le16 next_tid; + __le16 word3; + __le32 hdr_residual_count; + __le32 exp_r2t_sn; +}; + +struct mstorm_iscsi_task_st_ctx { + union iscsi_mstorm_sgl sgl_union; + struct iscsi_dif_flags dif_flags; + struct iscsi_mflags flags; + u8 sgl_size; + u8 host_sge_index; + __le16 dix_cur_sge_offset; + __le16 dix_cur_sge_size; + __le32 data_offset_rtid; + u8 dif_offset; + u8 dix_sgl_size; + u8 dix_sge_index; + u8 task_type; + struct regpair sense_db; + struct regpair dix_sgl_cur_sge; + __le32 rem_task_size; + __le16 reuse_count; + __le16 dif_data_residue; + u8 reserved0[4]; + __le32 reserved1[1]; +}; + +struct ustorm_iscsi_task_st_ctx { + __le32 rem_rcv_len; + __le32 exp_data_transfer_len; + __le32 exp_data_sn; + struct regpair lun; + struct iscsi_reg1 reg1; + u8 flags2; +#define USTORM_ISCSI_TASK_ST_CTX_AHS_EXIST_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_AHS_EXIST_SHIFT 0 +#define USTORM_ISCSI_TASK_ST_CTX_RESERVED1_MASK 0x7F +#define USTORM_ISCSI_TASK_ST_CTX_RESERVED1_SHIFT 1 + u8 reserved2; + __le16 reserved3; + __le32 reserved4; + __le32 reserved5; + __le32 reserved6; + __le32 reserved7; + u8 task_type; + u8 error_flags; +#define USTORM_ISCSI_TASK_ST_CTX_DATA_DIGEST_ERROR_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_DATA_DIGEST_ERROR_SHIFT 0 +#define USTORM_ISCSI_TASK_ST_CTX_DATA_TRUNCATED_ERROR_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_DATA_TRUNCATED_ERROR_SHIFT 1 +#define USTORM_ISCSI_TASK_ST_CTX_UNDER_RUN_ERROR_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_UNDER_RUN_ERROR_SHIFT 2 +#define USTORM_ISCSI_TASK_ST_CTX_RESERVED8_MASK 0x1F +#define USTORM_ISCSI_TASK_ST_CTX_RESERVED8_SHIFT 3 + u8 flags; +#define USTORM_ISCSI_TASK_ST_CTX_CQE_WRITE_MASK 0x3 +#define USTORM_ISCSI_TASK_ST_CTX_CQE_WRITE_SHIFT 0 +#define USTORM_ISCSI_TASK_ST_CTX_LOCAL_COMP_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_LOCAL_COMP_SHIFT 2 +#define USTORM_ISCSI_TASK_ST_CTX_Q0_R2TQE_WRITE_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_Q0_R2TQE_WRITE_SHIFT 3 +#define USTORM_ISCSI_TASK_ST_CTX_TOTALDATAACKED_DONE_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_TOTALDATAACKED_DONE_SHIFT 4 +#define USTORM_ISCSI_TASK_ST_CTX_HQSCANNED_DONE_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_HQSCANNED_DONE_SHIFT 5 +#define USTORM_ISCSI_TASK_ST_CTX_R2T2RECV_DONE_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_R2T2RECV_DONE_SHIFT 6 +#define USTORM_ISCSI_TASK_ST_CTX_RESERVED0_MASK 0x1 +#define USTORM_ISCSI_TASK_ST_CTX_RESERVED0_SHIFT 7 + u8 cq_rss_number; +}; + +struct iscsi_task_context { + struct ystorm_iscsi_task_st_ctx ystorm_st_context; + struct regpair ystorm_st_padding[2]; + struct ystorm_iscsi_task_ag_ctx ystorm_ag_context; + struct regpair ystorm_ag_padding[2]; + struct tdif_task_context tdif_context; + struct mstorm_iscsi_task_ag_ctx mstorm_ag_context; + struct regpair mstorm_ag_padding[2]; + struct ustorm_iscsi_task_ag_ctx ustorm_ag_context; + struct mstorm_iscsi_task_st_ctx mstorm_st_context; + struct ustorm_iscsi_task_st_ctx ustorm_st_context; + struct rdif_task_context rdif_context; +}; + +enum iscsi_task_type { + ISCSI_TASK_TYPE_INITIATOR_WRITE, + ISCSI_TASK_TYPE_INITIATOR_READ, + ISCSI_TASK_TYPE_MIDPATH, + ISCSI_TASK_TYPE_UNSOLIC, + ISCSI_TASK_TYPE_EXCHCLEANUP, + ISCSI_TASK_TYPE_IRRELEVANT, + ISCSI_TASK_TYPE_TARGET_WRITE, + ISCSI_TASK_TYPE_TARGET_READ, + ISCSI_TASK_TYPE_TARGET_RESPONSE, + ISCSI_TASK_TYPE_LOGIN_RESPONSE, + MAX_ISCSI_TASK_TYPE +}; + +union iscsi_ttt_txlen_union { + __le32 desired_tx_len; + __le32 ttt; +}; + +struct iscsi_uhqe { + __le32 reg1; +#define ISCSI_UHQE_PDU_PAYLOAD_LEN_MASK 0xFFFFF +#define ISCSI_UHQE_PDU_PAYLOAD_LEN_SHIFT 0 +#define ISCSI_UHQE_LOCAL_COMP_MASK 0x1 +#define ISCSI_UHQE_LOCAL_COMP_SHIFT 20 +#define ISCSI_UHQE_TOGGLE_BIT_MASK 0x1 +#define ISCSI_UHQE_TOGGLE_BIT_SHIFT 21 +#define ISCSI_UHQE_PURE_PAYLOAD_MASK 0x1 +#define ISCSI_UHQE_PURE_PAYLOAD_SHIFT 22 +#define ISCSI_UHQE_LOGIN_RESPONSE_PDU_MASK 0x1 +#define ISCSI_UHQE_LOGIN_RESPONSE_PDU_SHIFT 23 +#define ISCSI_UHQE_TASK_ID_HI_MASK 0xFF +#define ISCSI_UHQE_TASK_ID_HI_SHIFT 24 + __le32 reg2; +#define ISCSI_UHQE_BUFFER_OFFSET_MASK 0xFFFFFF +#define ISCSI_UHQE_BUFFER_OFFSET_SHIFT 0 +#define ISCSI_UHQE_TASK_ID_LO_MASK 0xFF +#define ISCSI_UHQE_TASK_ID_LO_SHIFT 24 +}; + +struct iscsi_wqe_field { + __le32 contlen_cdbsize_field; +#define ISCSI_WQE_FIELD_CONT_LEN_MASK 0xFFFFFF +#define ISCSI_WQE_FIELD_CONT_LEN_SHIFT 0 +#define ISCSI_WQE_FIELD_CDB_SIZE_MASK 0xFF +#define ISCSI_WQE_FIELD_CDB_SIZE_SHIFT 24 +}; + +union iscsi_wqe_field_union { + struct iscsi_wqe_field cont_field; + __le32 prev_tid; +}; + +struct iscsi_wqe { + __le16 task_id; + u8 flags; +#define ISCSI_WQE_WQE_TYPE_MASK 0x7 +#define ISCSI_WQE_WQE_TYPE_SHIFT 0 +#define ISCSI_WQE_NUM_FAST_SGES_MASK 0x7 +#define ISCSI_WQE_NUM_FAST_SGES_SHIFT 3 +#define ISCSI_WQE_PTU_INVALIDATE_MASK 0x1 +#define ISCSI_WQE_PTU_INVALIDATE_SHIFT 6 +#define ISCSI_WQE_RESPONSE_MASK 0x1 +#define ISCSI_WQE_RESPONSE_SHIFT 7 + struct iscsi_dif_flags prot_flags; + union iscsi_wqe_field_union cont_prevtid_union; +}; + +enum iscsi_wqe_type { + ISCSI_WQE_TYPE_NORMAL, + ISCSI_WQE_TYPE_TASK_CLEANUP, + ISCSI_WQE_TYPE_MIDDLE_PATH, + ISCSI_WQE_TYPE_LOGIN, + ISCSI_WQE_TYPE_FIRST_R2T_CONT, + ISCSI_WQE_TYPE_NONFIRST_R2T_CONT, + ISCSI_WQE_TYPE_RESPONSE, + MAX_ISCSI_WQE_TYPE +}; + +struct iscsi_xhqe { + union iscsi_ttt_txlen_union ttt_or_txlen; + __le32 exp_stat_sn; + struct iscsi_dif_flags prot_flags; + u8 total_ahs_length; + u8 opcode; + u8 flags; +#define ISCSI_XHQE_NUM_FAST_SGES_MASK 0x7 +#define ISCSI_XHQE_NUM_FAST_SGES_SHIFT 0 +#define ISCSI_XHQE_FINAL_MASK 0x1 +#define ISCSI_XHQE_FINAL_SHIFT 3 +#define ISCSI_XHQE_SUPER_IO_MASK 0x1 +#define ISCSI_XHQE_SUPER_IO_SHIFT 4 +#define ISCSI_XHQE_STATUS_BIT_MASK 0x1 +#define ISCSI_XHQE_STATUS_BIT_SHIFT 5 +#define ISCSI_XHQE_RESERVED_MASK 0x3 +#define ISCSI_XHQE_RESERVED_SHIFT 6 + union iscsi_seq_num seq_num_union; + __le16 reserved1; +}; + +struct mstorm_iscsi_stats_drv { + struct regpair iscsi_rx_dropped_pdus_task_not_valid; +}; + +struct ooo_opaque { + __le32 cid; + u8 drop_isle; + u8 drop_size; + u8 ooo_opcode; + u8 ooo_isle; +}; + +struct pstorm_iscsi_stats_drv { + struct regpair iscsi_tx_bytes_cnt; + struct regpair iscsi_tx_packet_cnt; +}; + +struct tstorm_iscsi_stats_drv { + struct regpair iscsi_rx_bytes_cnt; + struct regpair iscsi_rx_packet_cnt; + struct regpair iscsi_rx_new_ooo_isle_events_cnt; + __le32 iscsi_cmdq_threshold_cnt; + __le32 iscsi_rq_threshold_cnt; + __le32 iscsi_immq_threshold_cnt; +}; + +struct ustorm_iscsi_stats_drv { + struct regpair iscsi_rx_data_pdu_cnt; + struct regpair iscsi_rx_r2t_pdu_cnt; + struct regpair iscsi_rx_total_pdu_cnt; +}; + +struct xstorm_iscsi_stats_drv { + struct regpair iscsi_tx_go_to_slow_start_event_cnt; + struct regpair iscsi_tx_fast_retransmit_event_cnt; +}; + +struct ystorm_iscsi_stats_drv { + struct regpair iscsi_tx_data_pdu_cnt; + struct regpair iscsi_tx_r2t_pdu_cnt; + struct regpair iscsi_tx_total_pdu_cnt; +}; + +struct iscsi_db_data { + u8 params; +#define ISCSI_DB_DATA_DEST_MASK 0x3 +#define ISCSI_DB_DATA_DEST_SHIFT 0 +#define ISCSI_DB_DATA_AGG_CMD_MASK 0x3 +#define ISCSI_DB_DATA_AGG_CMD_SHIFT 2 +#define ISCSI_DB_DATA_BYPASS_EN_MASK 0x1 +#define ISCSI_DB_DATA_BYPASS_EN_SHIFT 4 +#define ISCSI_DB_DATA_RESERVED_MASK 0x1 +#define ISCSI_DB_DATA_RESERVED_SHIFT 5 +#define ISCSI_DB_DATA_AGG_VAL_SEL_MASK 0x3 +#define ISCSI_DB_DATA_AGG_VAL_SEL_SHIFT 6 + u8 agg_flags; + __le16 sq_prod; +}; + +struct tstorm_iscsi_task_ag_ctx { + u8 byte0; + u8 byte1; + __le16 word0; + u8 flags0; +#define TSTORM_ISCSI_TASK_AG_CTX_NIBBLE0_MASK 0xF +#define TSTORM_ISCSI_TASK_AG_CTX_NIBBLE0_SHIFT 0 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT0_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT0_SHIFT 4 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT1_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT1_SHIFT 5 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT2_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT2_SHIFT 6 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT3_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT3_SHIFT 7 + u8 flags1; +#define TSTORM_ISCSI_TASK_AG_CTX_BIT4_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT4_SHIFT 0 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT5_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_BIT5_SHIFT 1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF0_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF0_SHIFT 2 +#define TSTORM_ISCSI_TASK_AG_CTX_CF1_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF1_SHIFT 4 +#define TSTORM_ISCSI_TASK_AG_CTX_CF2_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF2_SHIFT 6 + u8 flags2; +#define TSTORM_ISCSI_TASK_AG_CTX_CF3_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF3_SHIFT 0 +#define TSTORM_ISCSI_TASK_AG_CTX_CF4_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF4_SHIFT 2 +#define TSTORM_ISCSI_TASK_AG_CTX_CF5_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF5_SHIFT 4 +#define TSTORM_ISCSI_TASK_AG_CTX_CF6_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF6_SHIFT 6 + u8 flags3; +#define TSTORM_ISCSI_TASK_AG_CTX_CF7_MASK 0x3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF7_SHIFT 0 +#define TSTORM_ISCSI_TASK_AG_CTX_CF0EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF0EN_SHIFT 2 +#define TSTORM_ISCSI_TASK_AG_CTX_CF1EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF1EN_SHIFT 3 +#define TSTORM_ISCSI_TASK_AG_CTX_CF2EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF2EN_SHIFT 4 +#define TSTORM_ISCSI_TASK_AG_CTX_CF3EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF3EN_SHIFT 5 +#define TSTORM_ISCSI_TASK_AG_CTX_CF4EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF4EN_SHIFT 6 +#define TSTORM_ISCSI_TASK_AG_CTX_CF5EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF5EN_SHIFT 7 + u8 flags4; +#define TSTORM_ISCSI_TASK_AG_CTX_CF6EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF6EN_SHIFT 0 +#define TSTORM_ISCSI_TASK_AG_CTX_CF7EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_CF7EN_SHIFT 1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE0EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE0EN_SHIFT 2 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE1EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE1EN_SHIFT 3 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE2EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE2EN_SHIFT 4 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE3EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE3EN_SHIFT 5 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE4EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE4EN_SHIFT 6 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE5EN_MASK 0x1 +#define TSTORM_ISCSI_TASK_AG_CTX_RULE5EN_SHIFT 7 + u8 byte2; + __le16 word1; + __le32 reg0; + u8 byte3; + u8 byte4; + __le16 word2; + __le16 word3; + __le16 word4; + __le32 reg1; + __le32 reg2; +}; + +#endif /* __ISCSI_COMMON__ */ diff --git a/include/linux/qed/rdma_common.h b/include/linux/qed/rdma_common.h new file mode 100644 index 000000000000..187991c1f439 --- /dev/null +++ b/include/linux/qed/rdma_common.h @@ -0,0 +1,44 @@ +/* QLogic qed NIC Driver + * Copyright (c) 2015 QLogic Corporation + * + * This software is available under the terms of the GNU General Public License + * (GPL) Version 2, available from the file COPYING in the main directory of + * this source tree. + */ + +#ifndef __RDMA_COMMON__ +#define __RDMA_COMMON__ +/************************/ +/* RDMA FW CONSTANTS */ +/************************/ + +#define RDMA_RESERVED_LKEY (0) +#define RDMA_RING_PAGE_SIZE (0x1000) + +#define RDMA_MAX_SGE_PER_SQ_WQE (4) +#define RDMA_MAX_SGE_PER_RQ_WQE (4) + +#define RDMA_MAX_DATA_SIZE_IN_WQE (0x7FFFFFFF) + +#define RDMA_REQ_RD_ATOMIC_ELM_SIZE (0x50) +#define RDMA_RESP_RD_ATOMIC_ELM_SIZE (0x20) + +#define RDMA_MAX_CQS (64 * 1024) +#define RDMA_MAX_TIDS (128 * 1024 - 1) +#define RDMA_MAX_PDS (64 * 1024) + +#define RDMA_NUM_STATISTIC_COUNTERS MAX_NUM_VPORTS + +#define RDMA_TASK_TYPE (PROTOCOLID_ROCE) + +struct rdma_srq_id { + __le16 srq_idx; + __le16 opaque_fid; +}; + +struct rdma_srq_producers { + __le32 sge_prod; + __le32 wqe_prod; +}; + +#endif /* __RDMA_COMMON__ */ diff --git a/include/linux/qed/roce_common.h b/include/linux/qed/roce_common.h new file mode 100644 index 000000000000..2eeaf3dc6646 --- /dev/null +++ b/include/linux/qed/roce_common.h @@ -0,0 +1,17 @@ +/* QLogic qed NIC Driver + * Copyright (c) 2015 QLogic Corporation + * + * This software is available under the terms of the GNU General Public License + * (GPL) Version 2, available from the file COPYING in the main directory of + * this source tree. + */ + +#ifndef __ROCE_COMMON__ +#define __ROCE_COMMON__ + +#define ROCE_REQ_MAX_INLINE_DATA_SIZE (256) +#define ROCE_REQ_MAX_SINGLE_SQ_WQE_SIZE (288) + +#define ROCE_MAX_QPS (32 * 1024) + +#endif /* __ROCE_COMMON__ */ diff --git a/include/linux/qed/storage_common.h b/include/linux/qed/storage_common.h new file mode 100644 index 000000000000..3b8e1efd9bc2 --- /dev/null +++ b/include/linux/qed/storage_common.h @@ -0,0 +1,91 @@ +/* QLogic qed NIC Driver + * Copyright (c) 2015 QLogic Corporation + * + * This software is available under the terms of the GNU General Public License + * (GPL) Version 2, available from the file COPYING in the main directory of + * this source tree. + */ + +#ifndef __STORAGE_COMMON__ +#define __STORAGE_COMMON__ + +#define NUM_OF_CMDQS_CQS (NUM_OF_GLOBAL_QUEUES / 2) +#define BDQ_NUM_RESOURCES (4) + +#define BDQ_ID_RQ (0) +#define BDQ_ID_IMM_DATA (1) +#define BDQ_NUM_IDS (2) + +#define BDQ_MAX_EXTERNAL_RING_SIZE (1 << 15) + +struct scsi_bd { + struct regpair address; + struct regpair opaque; +}; + +struct scsi_bdq_ram_drv_data { + __le16 external_producer; + __le16 reserved0[3]; +}; + +struct scsi_drv_cmdq { + __le16 cmdq_cons; + __le16 reserved0; + __le32 reserved1; +}; + +struct scsi_init_func_params { + __le16 num_tasks; + u8 log_page_size; + u8 debug_mode; + u8 reserved2[12]; +}; + +struct scsi_init_func_queues { + struct regpair glbl_q_params_addr; + __le16 rq_buffer_size; + __le16 cq_num_entries; + __le16 cmdq_num_entries; + u8 bdq_resource_id; + u8 q_validity; +#define SCSI_INIT_FUNC_QUEUES_RQ_VALID_MASK 0x1 +#define SCSI_INIT_FUNC_QUEUES_RQ_VALID_SHIFT 0 +#define SCSI_INIT_FUNC_QUEUES_IMM_DATA_VALID_MASK 0x1 +#define SCSI_INIT_FUNC_QUEUES_IMM_DATA_VALID_SHIFT 1 +#define SCSI_INIT_FUNC_QUEUES_CMD_VALID_MASK 0x1 +#define SCSI_INIT_FUNC_QUEUES_CMD_VALID_SHIFT 2 +#define SCSI_INIT_FUNC_QUEUES_RESERVED_VALID_MASK 0x1F +#define SCSI_INIT_FUNC_QUEUES_RESERVED_VALID_SHIFT 3 + u8 num_queues; + u8 queue_relative_offset; + u8 cq_sb_pi; + u8 cmdq_sb_pi; + __le16 cq_cmdq_sb_num_arr[NUM_OF_CMDQS_CQS]; + __le16 reserved0; + u8 bdq_pbl_num_entries[BDQ_NUM_IDS]; + struct regpair bdq_pbl_base_address[BDQ_NUM_IDS]; + __le16 bdq_xoff_threshold[BDQ_NUM_IDS]; + __le16 bdq_xon_threshold[BDQ_NUM_IDS]; + __le16 cmdq_xoff_threshold; + __le16 cmdq_xon_threshold; + __le32 reserved1; +}; + +struct scsi_ram_per_bdq_resource_drv_data { + struct scsi_bdq_ram_drv_data drv_data_per_bdq_id[BDQ_NUM_IDS]; +}; + +struct scsi_sge { + struct regpair sge_addr; + __le16 sge_len; + __le16 reserved0; + __le32 reserved1; +}; + +struct scsi_terminate_extra_params { + __le16 unsolicited_cq_count; + __le16 cmdq_count; + u8 reserved[4]; +}; + +#endif /* __STORAGE_COMMON__ */ diff --git a/include/linux/qed/tcp_common.h b/include/linux/qed/tcp_common.h new file mode 100644 index 000000000000..accba0e6b704 --- /dev/null +++ b/include/linux/qed/tcp_common.h @@ -0,0 +1,226 @@ +/* QLogic qed NIC Driver + * Copyright (c) 2015 QLogic Corporation + * + * This software is available under the terms of the GNU General Public License + * (GPL) Version 2, available from the file COPYING in the main directory of + * this source tree. + */ + +#ifndef __TCP_COMMON__ +#define __TCP_COMMON__ + +#define TCP_INVALID_TIMEOUT_VAL -1 + +enum tcp_connect_mode { + TCP_CONNECT_ACTIVE, + TCP_CONNECT_PASSIVE, + MAX_TCP_CONNECT_MODE +}; + +struct tcp_init_params { + __le32 max_cwnd; + __le16 dup_ack_threshold; + __le16 tx_sws_timer; + __le16 min_rto; + __le16 min_rto_rt; + __le16 max_rto; + u8 maxfinrt; + u8 reserved[1]; +}; + +enum tcp_ip_version { + TCP_IPV4, + TCP_IPV6, + MAX_TCP_IP_VERSION +}; + +struct tcp_offload_params { + __le16 local_mac_addr_lo; + __le16 local_mac_addr_mid; + __le16 local_mac_addr_hi; + __le16 remote_mac_addr_lo; + __le16 remote_mac_addr_mid; + __le16 remote_mac_addr_hi; + __le16 vlan_id; + u8 flags; +#define TCP_OFFLOAD_PARAMS_TS_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_TS_EN_SHIFT 0 +#define TCP_OFFLOAD_PARAMS_DA_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_DA_EN_SHIFT 1 +#define TCP_OFFLOAD_PARAMS_KA_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_KA_EN_SHIFT 2 +#define TCP_OFFLOAD_PARAMS_NAGLE_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_NAGLE_EN_SHIFT 3 +#define TCP_OFFLOAD_PARAMS_DA_CNT_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_DA_CNT_EN_SHIFT 4 +#define TCP_OFFLOAD_PARAMS_FIN_SENT_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_FIN_SENT_SHIFT 5 +#define TCP_OFFLOAD_PARAMS_FIN_RECEIVED_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_FIN_RECEIVED_SHIFT 6 +#define TCP_OFFLOAD_PARAMS_RESERVED0_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_RESERVED0_SHIFT 7 + u8 ip_version; + __le32 remote_ip[4]; + __le32 local_ip[4]; + __le32 flow_label; + u8 ttl; + u8 tos_or_tc; + __le16 remote_port; + __le16 local_port; + __le16 mss; + u8 rcv_wnd_scale; + u8 connect_mode; + __le16 srtt; + __le32 cwnd; + __le32 ss_thresh; + __le16 reserved1; + u8 ka_max_probe_cnt; + u8 dup_ack_theshold; + __le32 rcv_next; + __le32 snd_una; + __le32 snd_next; + __le32 snd_max; + __le32 snd_wnd; + __le32 rcv_wnd; + __le32 snd_wl1; + __le32 ts_time; + __le32 ts_recent; + __le32 ts_recent_age; + __le32 total_rt; + __le32 ka_timeout_delta; + __le32 rt_timeout_delta; + u8 dup_ack_cnt; + u8 snd_wnd_probe_cnt; + u8 ka_probe_cnt; + u8 rt_cnt; + __le16 rtt_var; + __le16 reserved2; + __le32 ka_timeout; + __le32 ka_interval; + __le32 max_rt_time; + __le32 initial_rcv_wnd; + u8 snd_wnd_scale; + u8 ack_frequency; + __le16 da_timeout_value; + __le32 ts_ticks_per_second; +}; + +struct tcp_offload_params_opt2 { + __le16 local_mac_addr_lo; + __le16 local_mac_addr_mid; + __le16 local_mac_addr_hi; + __le16 remote_mac_addr_lo; + __le16 remote_mac_addr_mid; + __le16 remote_mac_addr_hi; + __le16 vlan_id; + u8 flags; +#define TCP_OFFLOAD_PARAMS_OPT2_TS_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_OPT2_TS_EN_SHIFT 0 +#define TCP_OFFLOAD_PARAMS_OPT2_DA_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_OPT2_DA_EN_SHIFT 1 +#define TCP_OFFLOAD_PARAMS_OPT2_KA_EN_MASK 0x1 +#define TCP_OFFLOAD_PARAMS_OPT2_KA_EN_SHIFT 2 +#define TCP_OFFLOAD_PARAMS_OPT2_RESERVED0_MASK 0x1F +#define TCP_OFFLOAD_PARAMS_OPT2_RESERVED0_SHIFT 3 + u8 ip_version; + __le32 remote_ip[4]; + __le32 local_ip[4]; + __le32 flow_label; + u8 ttl; + u8 tos_or_tc; + __le16 remote_port; + __le16 local_port; + __le16 mss; + u8 rcv_wnd_scale; + u8 connect_mode; + __le16 syn_ip_payload_length; + __le32 syn_phy_addr_lo; + __le32 syn_phy_addr_hi; + __le32 reserved1[22]; +}; + +enum tcp_seg_placement_event { + TCP_EVENT_ADD_PEN, + TCP_EVENT_ADD_NEW_ISLE, + TCP_EVENT_ADD_ISLE_RIGHT, + TCP_EVENT_ADD_ISLE_LEFT, + TCP_EVENT_JOIN, + TCP_EVENT_NOP, + MAX_TCP_SEG_PLACEMENT_EVENT +}; + +struct tcp_update_params { + __le16 flags; +#define TCP_UPDATE_PARAMS_REMOTE_MAC_ADDR_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_REMOTE_MAC_ADDR_CHANGED_SHIFT 0 +#define TCP_UPDATE_PARAMS_MSS_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_MSS_CHANGED_SHIFT 1 +#define TCP_UPDATE_PARAMS_TTL_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_TTL_CHANGED_SHIFT 2 +#define TCP_UPDATE_PARAMS_TOS_OR_TC_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_TOS_OR_TC_CHANGED_SHIFT 3 +#define TCP_UPDATE_PARAMS_KA_TIMEOUT_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_KA_TIMEOUT_CHANGED_SHIFT 4 +#define TCP_UPDATE_PARAMS_KA_INTERVAL_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_KA_INTERVAL_CHANGED_SHIFT 5 +#define TCP_UPDATE_PARAMS_MAX_RT_TIME_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_MAX_RT_TIME_CHANGED_SHIFT 6 +#define TCP_UPDATE_PARAMS_FLOW_LABEL_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_FLOW_LABEL_CHANGED_SHIFT 7 +#define TCP_UPDATE_PARAMS_INITIAL_RCV_WND_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_INITIAL_RCV_WND_CHANGED_SHIFT 8 +#define TCP_UPDATE_PARAMS_KA_MAX_PROBE_CNT_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_KA_MAX_PROBE_CNT_CHANGED_SHIFT 9 +#define TCP_UPDATE_PARAMS_KA_EN_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_KA_EN_CHANGED_SHIFT 10 +#define TCP_UPDATE_PARAMS_NAGLE_EN_CHANGED_MASK 0x1 +#define TCP_UPDATE_PARAMS_NAGLE_EN_CHANGED_SHIFT 11 +#define TCP_UPDATE_PARAMS_KA_EN_MASK 0x1 +#define TCP_UPDATE_PARAMS_KA_EN_SHIFT 12 +#define TCP_UPDATE_PARAMS_NAGLE_EN_MASK 0x1 +#define TCP_UPDATE_PARAMS_NAGLE_EN_SHIFT 13 +#define TCP_UPDATE_PARAMS_KA_RESTART_MASK 0x1 +#define TCP_UPDATE_PARAMS_KA_RESTART_SHIFT 14 +#define TCP_UPDATE_PARAMS_RETRANSMIT_RESTART_MASK 0x1 +#define TCP_UPDATE_PARAMS_RETRANSMIT_RESTART_SHIFT 15 + __le16 remote_mac_addr_lo; + __le16 remote_mac_addr_mid; + __le16 remote_mac_addr_hi; + __le16 mss; + u8 ttl; + u8 tos_or_tc; + __le32 ka_timeout; + __le32 ka_interval; + __le32 max_rt_time; + __le32 flow_label; + __le32 initial_rcv_wnd; + u8 ka_max_probe_cnt; + u8 reserved1[7]; +}; + +struct tcp_upload_params { + __le32 rcv_next; + __le32 snd_una; + __le32 snd_next; + __le32 snd_max; + __le32 snd_wnd; + __le32 rcv_wnd; + __le32 snd_wl1; + __le32 cwnd; + __le32 ss_thresh; + __le16 srtt; + __le16 rtt_var; + __le32 ts_time; + __le32 ts_recent; + __le32 ts_recent_age; + __le32 total_rt; + __le32 ka_timeout_delta; + __le32 rt_timeout_delta; + u8 dup_ack_cnt; + u8 snd_wnd_probe_cnt; + u8 ka_probe_cnt; + u8 rt_cnt; + __le32 reserved; +}; + +#endif /* __TCP_COMMON__ */ -- cgit v1.2.3 From c5ac93191d7e6977c5c3465ac94c73ebb8a8ecba Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Fri, 3 Jun 2016 14:35:34 +0300 Subject: qed: Add iscsi/rdma personalities This patch adds in the ecore 2 new personalities in addition to QED_PCI_ETH - QED_PCI_ISCSI and QED_PCI_ETH_ROCE. Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed.h | 4 ++ drivers/net/ethernet/qlogic/qed/qed_dev.c | 6 +++ drivers/net/ethernet/qlogic/qed/qed_hsi.h | 6 ++- drivers/net/ethernet/qlogic/qed/qed_main.c | 5 +- drivers/net/ethernet/qlogic/qed/qed_mcp.c | 13 ++++- drivers/net/ethernet/qlogic/qed/qed_sp_commands.c | 17 +++++- include/linux/qed/common_hsi.h | 4 +- include/linux/qed/qed_if.h | 65 +++++++++++++++++++++++ 8 files changed, 114 insertions(+), 6 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h index 1042f2af854a..4417f127f766 100644 --- a/drivers/net/ethernet/qlogic/qed/qed.h +++ b/drivers/net/ethernet/qlogic/qed/qed.h @@ -127,6 +127,8 @@ struct qed_tunn_update_params { */ enum qed_pci_personality { QED_PCI_ETH, + QED_PCI_ISCSI, + QED_PCI_ETH_ROCE, QED_PCI_DEFAULT /* default in shmem */ }; @@ -170,6 +172,8 @@ enum QED_PORT_MODE { enum qed_dev_cap { QED_DEV_CAP_ETH, + QED_DEV_CAP_ISCSI, + QED_DEV_CAP_ROCE, }; struct qed_hw_info { diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index 151173d4a926..30c6d2eb6e4c 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -1472,6 +1472,12 @@ static int qed_hw_get_nvm_info(struct qed_hwfn *p_hwfn, if (device_capabilities & NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ETHERNET) __set_bit(QED_DEV_CAP_ETH, &p_hwfn->hw_info.device_capabilities); + if (device_capabilities & NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ISCSI) + __set_bit(QED_DEV_CAP_ISCSI, + &p_hwfn->hw_info.device_capabilities); + if (device_capabilities & NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ROCE) + __set_bit(QED_DEV_CAP_ROCE, + &p_hwfn->hw_info.device_capabilities); return qed_mcp_fill_shmem_func_info(p_hwfn, p_ptt); } diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h index 4cd70bfd929f..72e0ad6d16c6 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h @@ -650,7 +650,7 @@ struct mstorm_vf_zone { /* personality per PF */ enum personality_type { BAD_PERSONALITY_TYP, - PERSONALITY_RESERVED, + PERSONALITY_ISCSI, PERSONALITY_RESERVED2, PERSONALITY_RDMA_AND_ETH, PERSONALITY_RESERVED3, @@ -7072,6 +7072,8 @@ struct public_func { #define FUNC_MF_CFG_PROTOCOL_MASK 0x000000f0 #define FUNC_MF_CFG_PROTOCOL_SHIFT 4 #define FUNC_MF_CFG_PROTOCOL_ETHERNET 0x00000000 +#define FUNC_MF_CFG_PROTOCOL_ISCSI 0x00000010 +#define FUNC_MF_CFG_PROTOCOL_ROCE 0x00000030 #define FUNC_MF_CFG_PROTOCOL_MAX 0x00000030 #define FUNC_MF_CFG_MIN_BW_MASK 0x0000ff00 @@ -7405,6 +7407,8 @@ struct nvm_cfg1_glob { u32 misc_sig; u32 device_capabilities; #define NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ETHERNET 0x1 +#define NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ISCSI 0x4 +#define NVM_CFG1_GLOB_DEVICE_CAPABILITIES_ROCE 0x8 u32 power_dissipated; u32 power_consumed; u32 efi_version; diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index c807f67a02c5..f41f78d6a3f9 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -207,6 +207,8 @@ int qed_fill_dev_info(struct qed_dev *cdev, dev_info->pci_mem_start = cdev->pci_params.mem_start; dev_info->pci_mem_end = cdev->pci_params.mem_end; dev_info->pci_irq = cdev->pci_params.irq; + dev_info->rdma_supported = + (cdev->hwfns[0].hw_info.personality == QED_PCI_ETH_ROCE); dev_info->is_mf_default = IS_MF_DEFAULT(&cdev->hwfns[0]); ether_addr_copy(dev_info->hw_mac, cdev->hwfns[0].hw_info.hw_mac_addr); @@ -901,7 +903,8 @@ static int qed_slowpath_stop(struct qed_dev *cdev) if (IS_PF(cdev)) { qed_free_stream_mem(cdev); - qed_sriov_disable(cdev, true); + if (IS_QED_ETH_IF(cdev)) + qed_sriov_disable(cdev, true); qed_nic_stop(cdev); qed_slowpath_irq_free(cdev); diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c index 2c143b33c002..a240f26344a4 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c @@ -977,7 +977,18 @@ qed_mcp_get_shmem_proto(struct qed_hwfn *p_hwfn, switch (p_info->config & FUNC_MF_CFG_PROTOCOL_MASK) { case FUNC_MF_CFG_PROTOCOL_ETHERNET: - *p_proto = QED_PCI_ETH; + if (test_bit(QED_DEV_CAP_ROCE, + &p_hwfn->hw_info.device_capabilities)) + *p_proto = QED_PCI_ETH_ROCE; + else + *p_proto = QED_PCI_ETH; + break; + case FUNC_MF_CFG_PROTOCOL_ISCSI: + *p_proto = QED_PCI_ISCSI; + break; + case FUNC_MF_CFG_PROTOCOL_ROCE: + DP_NOTICE(p_hwfn, "RoCE personality is not a valid value!\n"); + rc = -EINVAL; break; default: rc = -EINVAL; diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c index 63b93fbdc66a..a52f3fc051f5 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c @@ -358,11 +358,26 @@ int qed_sp_pf_start(struct qed_hwfn *p_hwfn, qed_tunn_set_pf_start_params(p_hwfn, p_tunn, &p_ramrod->tunnel_config); - p_hwfn->hw_info.personality = PERSONALITY_ETH; if (IS_MF_SI(p_hwfn)) p_ramrod->allow_npar_tx_switching = allow_npar_tx_switch; + switch (p_hwfn->hw_info.personality) { + case QED_PCI_ETH: + p_ramrod->personality = PERSONALITY_ETH; + break; + case QED_PCI_ISCSI: + p_ramrod->personality = PERSONALITY_ISCSI; + break; + case QED_PCI_ETH_ROCE: + p_ramrod->personality = PERSONALITY_RDMA_AND_ETH; + break; + default: + DP_NOTICE(p_hwfn, "Unkown personality %d\n", + p_hwfn->hw_info.personality); + p_ramrod->personality = PERSONALITY_ETH; + } + if (p_hwfn->cdev->p_iov_info) { struct qed_hw_sriov_info *p_iov = p_hwfn->cdev->p_iov_info; diff --git a/include/linux/qed/common_hsi.h b/include/linux/qed/common_hsi.h index 077a3b6cc80f..40c0ada01806 100644 --- a/include/linux/qed/common_hsi.h +++ b/include/linux/qed/common_hsi.h @@ -517,9 +517,9 @@ enum mf_mode { /* Per-protocol connection types */ enum protocol_type { - PROTOCOLID_RESERVED1, + PROTOCOLID_ISCSI, PROTOCOLID_RESERVED2, - PROTOCOLID_RESERVED3, + PROTOCOLID_ROCE, PROTOCOLID_CORE, PROTOCOLID_ETH, PROTOCOLID_RESERVED4, diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h index 15efccfdc46e..e8cc49f9688a 100644 --- a/include/linux/qed/qed_if.h +++ b/include/linux/qed/qed_if.h @@ -58,8 +58,70 @@ struct qed_eth_pf_params { u16 num_cons; }; +/* Most of the the parameters below are described in the FW iSCSI / TCP HSI */ +struct qed_iscsi_pf_params { + u64 glbl_q_params_addr; + u64 bdq_pbl_base_addr[2]; + u32 max_cwnd; + u16 cq_num_entries; + u16 cmdq_num_entries; + u16 dup_ack_threshold; + u16 tx_sws_timer; + u16 min_rto; + u16 min_rto_rt; + u16 max_rto; + + /* The following parameters are used during HW-init + * and these parameters need to be passed as arguments + * to update_pf_params routine invoked before slowpath start + */ + u16 num_cons; + u16 num_tasks; + + /* The following parameters are used during protocol-init */ + u16 half_way_close_timeout; + u16 bdq_xoff_threshold[2]; + u16 bdq_xon_threshold[2]; + u16 cmdq_xoff_threshold; + u16 cmdq_xon_threshold; + u16 rq_buffer_size; + + u8 num_sq_pages_in_ring; + u8 num_r2tq_pages_in_ring; + u8 num_uhq_pages_in_ring; + u8 num_queues; + u8 log_page_size; + u8 rqe_log_size; + u8 max_fin_rt; + u8 gl_rq_pi; + u8 gl_cmd_pi; + u8 debug_mode; + u8 ll2_ooo_queue_id; + u8 ooo_enable; + + u8 is_target; + u8 bdq_pbl_num_entries[2]; +}; + +struct qed_rdma_pf_params { + /* Supplied to QED during resource allocation (may affect the ILT and + * the doorbell BAR). + */ + u32 min_dpis; /* number of requested DPIs */ + u32 num_mrs; /* number of requested memory regions */ + u32 num_qps; /* number of requested Queue Pairs */ + u32 num_srqs; /* number of requested SRQ */ + u8 roce_edpm_mode; /* see QED_ROCE_EDPM_MODE_ENABLE */ + u8 gl_pi; /* protocol index */ + + /* Will allocate rate limiters to be used with QPs */ + u8 enable_dcqcn; +}; + struct qed_pf_params { struct qed_eth_pf_params eth_pf_params; + struct qed_iscsi_pf_params iscsi_pf_params; + struct qed_rdma_pf_params rdma_pf_params; }; enum qed_int_mode { @@ -100,6 +162,8 @@ struct qed_dev_info { /* MFW version */ u32 mfw_rev; + bool rdma_supported; + u32 flash_size; u8 mf_mode; bool tx_switching; @@ -111,6 +175,7 @@ enum qed_sb_type { enum qed_protocol { QED_PROTOCOL_ETH, + QED_PROTOCOL_ISCSI, }; struct qed_link_params { -- cgit v1.2.3 From 81b1251d3b761b303955408dd6a49618b5683c2b Mon Sep 17 00:00:00 2001 From: Yuval Mintz Date: Sat, 4 Jun 2016 08:20:16 +0300 Subject: qed: Fix next-ptr chains for BE / 32-bit Commit a91eb52abb50 ("qed: Revisit chain implementation") contains an incorrect implementation for BE platforms, as device's regpairs containing addresses are LE and they're not converted correctly when read back. In addition, it raises a compilation warning for 32-bit platforms where dma_addr_t is a 32-bit variable. Reported-by: kbuild test robot Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller --- include/linux/qed/qed_chain.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/qed/qed_chain.h b/include/linux/qed/qed_chain.h index eceaa9ed2ae9..7e441bdeabdc 100644 --- a/include/linux/qed/qed_chain.h +++ b/include/linux/qed/qed_chain.h @@ -25,10 +25,9 @@ } while (0) #define HILO_GEN(hi, lo, type) ((((type)(hi)) << 32) + (lo)) -#define HILO_DMA(hi, lo) HILO_GEN(hi, lo, dma_addr_t) #define HILO_64(hi, lo) HILO_GEN((le32_to_cpu(hi)), (le32_to_cpu(lo)), u64) -#define HILO_DMA_REGPAIR(regpair) (HILO_DMA(regpair.hi, regpair.lo)) #define HILO_64_REGPAIR(regpair) (HILO_64(regpair.hi, regpair.lo)) +#define HILO_DMA_REGPAIR(regpair) ((dma_addr_t)HILO_64_REGPAIR(regpair)) enum qed_chain_mode { /* Each Page contains a next pointer at its end */ -- cgit v1.2.3 From 6eb17e0df74c036eba7548915b37f009403fe09e Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:09 +0800 Subject: ACPI: bus: add stub acpi_dev_found() to linux/acpi.h acpi_dev_found() will be used to detect if a given ACPI device is in the system. It will be compiled in non-ACPI case, but the function is in acpi_bus.h and acpi_bus.h can only be used in ACPI case, so this patch add the stub function to linux/acpi.h to make compiled successfully in non-ACPI cases. Cc: Rafael J. Wysocki Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller --- include/linux/acpi.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/linux') diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 288fac5294f5..3025d1930f6e 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -543,6 +543,11 @@ struct platform_device *acpi_create_platform_device(struct acpi_device *); struct fwnode_handle; +static inline bool acpi_dev_found(const char *hid) +{ + return false; +} + static inline bool is_acpi_node(struct fwnode_handle *fwnode) { return false; -- cgit v1.2.3 From 4ae399241adba66ad72e5973a1004f37ffbe67cd Mon Sep 17 00:00:00 2001 From: Kejian Yan Date: Fri, 3 Jun 2016 10:55:10 +0800 Subject: ACPI: bus: add stub acpi_evaluate_dsm() to linux/acpi.h acpi_evaluate_dsm() will be used to handle the _DSM method in ACPI case. It will be compiled in non-ACPI case, but the function is in acpi_bus.h and acpi_bus.h can only be used in ACPI case, so this patch add the stub function to linux/acpi.h to make compiled successfully in non-ACPI cases. Cc: Rafael J. Wysocki Signed-off-by: Kejian Yan Signed-off-by: Yisen Zhuang Signed-off-by: David S. Miller --- include/linux/acpi.h | 8 ++++++++ 1 file changed, 8 insertions(+) (limited to 'include/linux') diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 3025d1930f6e..4d4bb4955682 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -659,6 +659,14 @@ static inline bool acpi_driver_match_device(struct device *dev, return false; } +static inline union acpi_object *acpi_evaluate_dsm(acpi_handle handle, + const u8 *uuid, + int rev, int func, + union acpi_object *argv4) +{ + return NULL; +} + static inline int acpi_device_uevent_modalias(struct device *dev, struct kobj_uevent_env *env) { -- cgit v1.2.3 From f9eb8aea2a1e12fc2f584d1627deeb957435a801 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 6 Jun 2016 09:37:15 -0700 Subject: net_sched: transform qdisc running bit into a seqcount Instead of using a single bit (__QDISC___STATE_RUNNING) in sch->__state, use a seqcount. This adds lockdep support, but more importantly it will allow us to sample qdisc/class statistics without having to grab qdisc root lock. Signed-off-by: Eric Dumazet Cc: Cong Wang Cc: Jamal Hadi Salim Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 2 ++ drivers/net/ppp/ppp_generic.c | 3 +++ drivers/net/team/team.c | 2 ++ include/linux/netdevice.h | 1 + include/net/sch_generic.h | 15 ++++----------- net/bluetooth/6lowpan.c | 2 ++ net/core/dev.c | 2 +- net/ieee802154/6lowpan/core.c | 3 +++ net/l2tp/l2tp_eth.c | 4 ++++ net/sched/sch_generic.c | 14 ++++++++++---- 10 files changed, 32 insertions(+), 16 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 941ec99cd3b6..681af31a60ed 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -4610,6 +4610,7 @@ static int bond_check_params(struct bond_params *params) static struct lock_class_key bonding_netdev_xmit_lock_key; static struct lock_class_key bonding_netdev_addr_lock_key; static struct lock_class_key bonding_tx_busylock_key; +static struct lock_class_key bonding_qdisc_running_key; static void bond_set_lockdep_class_one(struct net_device *dev, struct netdev_queue *txq, @@ -4625,6 +4626,7 @@ static void bond_set_lockdep_class(struct net_device *dev) &bonding_netdev_addr_lock_key); netdev_for_each_tx_queue(dev, bond_set_lockdep_class_one, NULL); dev->qdisc_tx_busylock = &bonding_tx_busylock_key; + dev->qdisc_running_key = &bonding_qdisc_running_key; } /* Called from registration process */ diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c index 8dedafa1a95d..aeabaa42317f 100644 --- a/drivers/net/ppp/ppp_generic.c +++ b/drivers/net/ppp/ppp_generic.c @@ -1313,9 +1313,12 @@ ppp_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats64) } static struct lock_class_key ppp_tx_busylock; +static struct lock_class_key ppp_qdisc_running_key; + static int ppp_dev_init(struct net_device *dev) { dev->qdisc_tx_busylock = &ppp_tx_busylock; + dev->qdisc_running_key = &ppp_qdisc_running_key; return 0; } diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index 2ace126533cd..00eb38956a2c 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -1577,6 +1577,7 @@ static const struct team_option team_options[] = { static struct lock_class_key team_netdev_xmit_lock_key; static struct lock_class_key team_netdev_addr_lock_key; static struct lock_class_key team_tx_busylock_key; +static struct lock_class_key team_qdisc_running_key; static void team_set_lockdep_class_one(struct net_device *dev, struct netdev_queue *txq, @@ -1590,6 +1591,7 @@ static void team_set_lockdep_class(struct net_device *dev) lockdep_set_class(&dev->addr_list_lock, &team_netdev_addr_lock_key); netdev_for_each_tx_queue(dev, team_set_lockdep_class_one, NULL); dev->qdisc_tx_busylock = &team_tx_busylock_key; + dev->qdisc_running_key = &team_qdisc_running_key; } static int team_init(struct net_device *dev) diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index fa6df2699532..59d7e06d88d5 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1862,6 +1862,7 @@ struct net_device { #endif struct phy_device *phydev; struct lock_class_key *qdisc_tx_busylock; + struct lock_class_key *qdisc_running_key; bool proto_down; }; #define to_net_dev(d) container_of(d, struct net_device, dev) diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index a1fd76c22a59..bff8d895ef8a 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -29,13 +29,6 @@ enum qdisc_state_t { __QDISC_STATE_THROTTLED, }; -/* - * following bits are only changed while qdisc lock is held - */ -enum qdisc___state_t { - __QDISC___STATE_RUNNING = 1, -}; - struct qdisc_size_table { struct rcu_head rcu; struct list_head list; @@ -93,7 +86,7 @@ struct Qdisc { unsigned long state; struct sk_buff_head q; struct gnet_stats_basic_packed bstats; - unsigned int __state; + seqcount_t running; struct gnet_stats_queue qstats; struct rcu_head rcu_head; int padded; @@ -104,20 +97,20 @@ struct Qdisc { static inline bool qdisc_is_running(const struct Qdisc *qdisc) { - return (qdisc->__state & __QDISC___STATE_RUNNING) ? true : false; + return (raw_read_seqcount(&qdisc->running) & 1) ? true : false; } static inline bool qdisc_run_begin(struct Qdisc *qdisc) { if (qdisc_is_running(qdisc)) return false; - qdisc->__state |= __QDISC___STATE_RUNNING; + write_seqcount_begin(&qdisc->running); return true; } static inline void qdisc_run_end(struct Qdisc *qdisc) { - qdisc->__state &= ~__QDISC___STATE_RUNNING; + write_seqcount_end(&qdisc->running); } static inline bool qdisc_may_bulk(const struct Qdisc *qdisc) diff --git a/net/bluetooth/6lowpan.c b/net/bluetooth/6lowpan.c index 780089d75915..977a11e418d0 100644 --- a/net/bluetooth/6lowpan.c +++ b/net/bluetooth/6lowpan.c @@ -629,6 +629,7 @@ static netdev_tx_t bt_xmit(struct sk_buff *skb, struct net_device *netdev) static struct lock_class_key bt_tx_busylock; static struct lock_class_key bt_netdev_xmit_lock_key; +static struct lock_class_key bt_qdisc_running_key; static void bt_set_lockdep_class_one(struct net_device *dev, struct netdev_queue *txq, @@ -641,6 +642,7 @@ static int bt_dev_init(struct net_device *dev) { netdev_for_each_tx_queue(dev, bt_set_lockdep_class_one, NULL); dev->qdisc_tx_busylock = &bt_tx_busylock; + dev->qdisc_running_key = &bt_qdisc_running_key; return 0; } diff --git a/net/core/dev.c b/net/core/dev.c index 896b686d1966..e0bcc39f4a7d 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3075,7 +3075,7 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q, /* * Heuristic to force contended enqueues to serialize on a * separate lock before trying to get qdisc main lock. - * This permits __QDISC___STATE_RUNNING owner to get the lock more + * This permits qdisc->running owner to get the lock more * often and dequeue packets faster. */ contended = qdisc_is_running(q); diff --git a/net/ieee802154/6lowpan/core.c b/net/ieee802154/6lowpan/core.c index dd085db8580e..14aa5effd29a 100644 --- a/net/ieee802154/6lowpan/core.c +++ b/net/ieee802154/6lowpan/core.c @@ -60,6 +60,7 @@ static struct header_ops lowpan_header_ops = { static struct lock_class_key lowpan_tx_busylock; static struct lock_class_key lowpan_netdev_xmit_lock_key; +static struct lock_class_key lowpan_qdisc_running_key; static void lowpan_set_lockdep_class_one(struct net_device *ldev, struct netdev_queue *txq, @@ -73,6 +74,8 @@ static int lowpan_dev_init(struct net_device *ldev) { netdev_for_each_tx_queue(ldev, lowpan_set_lockdep_class_one, NULL); ldev->qdisc_tx_busylock = &lowpan_tx_busylock; + ldev->qdisc_running_key = &lowpan_qdisc_running_key; + return 0; } diff --git a/net/l2tp/l2tp_eth.c b/net/l2tp/l2tp_eth.c index e253c26f31ac..c00d72d182fa 100644 --- a/net/l2tp/l2tp_eth.c +++ b/net/l2tp/l2tp_eth.c @@ -68,6 +68,8 @@ static inline struct l2tp_eth_net *l2tp_eth_pernet(struct net *net) } static struct lock_class_key l2tp_eth_tx_busylock; +static struct lock_class_key l2tp_qdisc_running_key; + static int l2tp_eth_dev_init(struct net_device *dev) { struct l2tp_eth *priv = netdev_priv(dev); @@ -76,6 +78,8 @@ static int l2tp_eth_dev_init(struct net_device *dev) eth_hw_addr_random(dev); eth_broadcast_addr(dev->broadcast); dev->qdisc_tx_busylock = &l2tp_eth_tx_busylock; + dev->qdisc_running_key = &l2tp_qdisc_running_key; + return 0; } diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 269dd71b3828..cebea73e70ac 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -110,7 +110,7 @@ static struct sk_buff *dequeue_skb(struct Qdisc *q, bool *validate, /* * Transmit possibly several skbs, and handle the return status as - * required. Holding the __QDISC___STATE_RUNNING bit guarantees that + * required. Owning running seqcount bit guarantees that * only one CPU can execute this function. * * Returns to the caller: @@ -137,10 +137,10 @@ int sch_direct_xmit(struct sk_buff *skb, struct Qdisc *q, HARD_TX_UNLOCK(dev, txq); } else { - spin_lock(root_lock); + spin_lock_nested(root_lock, SINGLE_DEPTH_NESTING); return qdisc_qlen(q); } - spin_lock(root_lock); + spin_lock_nested(root_lock, SINGLE_DEPTH_NESTING); if (dev_xmit_complete(ret)) { /* Driver sent out skb successfully or skb was consumed */ @@ -163,7 +163,7 @@ int sch_direct_xmit(struct sk_buff *skb, struct Qdisc *q, /* * NOTE: Called under qdisc_lock(q) with locally disabled BH. * - * __QDISC___STATE_RUNNING guarantees only one CPU can process + * running seqcount guarantees only one CPU can process * this qdisc at a time. qdisc_lock(q) serializes queue accesses for * this queue. * @@ -379,6 +379,7 @@ struct Qdisc noop_qdisc = { .list = LIST_HEAD_INIT(noop_qdisc.list), .q.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock), .dev_queue = &noop_netdev_queue, + .running = SEQCNT_ZERO(noop_qdisc.running), .busylock = __SPIN_LOCK_UNLOCKED(noop_qdisc.busylock), }; EXPORT_SYMBOL(noop_qdisc); @@ -537,6 +538,7 @@ struct Qdisc_ops pfifo_fast_ops __read_mostly = { EXPORT_SYMBOL(pfifo_fast_ops); static struct lock_class_key qdisc_tx_busylock; +static struct lock_class_key qdisc_running_key; struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue, const struct Qdisc_ops *ops) @@ -570,6 +572,10 @@ struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue, lockdep_set_class(&sch->busylock, dev->qdisc_tx_busylock ?: &qdisc_tx_busylock); + seqcount_init(&sch->running); + lockdep_set_class(&sch->running, + dev->qdisc_running_key ?: &qdisc_running_key); + sch->ops = ops; sch->enqueue = ops->enqueue; sch->dequeue = ops->dequeue; -- cgit v1.2.3 From 6ad8c632ee48ae099aa13704ef18a641220fe211 Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Wed, 8 Jun 2016 06:22:10 -0400 Subject: qed: Add support for query/config dcbx. Query API reads the dcbx data from the device shared memory and return it to the caller. The config API configures the user provided dcbx values on the device, and initiates the dcbx negotiation with the peer. Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_dcbx.c | 533 ++++++++++++++++++++++++++++- drivers/net/ethernet/qlogic/qed/qed_dcbx.h | 28 ++ drivers/net/ethernet/qlogic/qed/qed_hsi.h | 3 + include/linux/qed/qed_if.h | 90 +++++ 4 files changed, 653 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c index 21ec1c2df2c7..e782484d2d89 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c @@ -252,7 +252,7 @@ qed_dcbx_process_tlv(struct qed_hwfn *p_hwfn, if (p_data->arr[type].update) continue; - enable = (type == DCBX_PROTOCOL_ETH) ? false : dcbx_enabled; + enable = !(type == DCBX_PROTOCOL_ETH); qed_dcbx_update_app_info(p_data, p_hwfn, enable, true, priority, tc, type); } @@ -351,6 +351,293 @@ qed_dcbx_copy_mib(struct qed_hwfn *p_hwfn, return rc; } +#ifdef CONFIG_DCB +static void +qed_dcbx_get_priority_info(struct qed_hwfn *p_hwfn, + struct qed_dcbx_app_prio *p_prio, + struct qed_dcbx_results *p_results) +{ + u8 val; + + p_prio->roce = QED_DCBX_INVALID_PRIORITY; + p_prio->roce_v2 = QED_DCBX_INVALID_PRIORITY; + p_prio->iscsi = QED_DCBX_INVALID_PRIORITY; + p_prio->fcoe = QED_DCBX_INVALID_PRIORITY; + + if (p_results->arr[DCBX_PROTOCOL_ROCE].update && + p_results->arr[DCBX_PROTOCOL_ROCE].enable) + p_prio->roce = p_results->arr[DCBX_PROTOCOL_ROCE].priority; + + if (p_results->arr[DCBX_PROTOCOL_ROCE_V2].update && + p_results->arr[DCBX_PROTOCOL_ROCE_V2].enable) { + val = p_results->arr[DCBX_PROTOCOL_ROCE_V2].priority; + p_prio->roce_v2 = val; + } + + if (p_results->arr[DCBX_PROTOCOL_ISCSI].update && + p_results->arr[DCBX_PROTOCOL_ISCSI].enable) + p_prio->iscsi = p_results->arr[DCBX_PROTOCOL_ISCSI].priority; + + if (p_results->arr[DCBX_PROTOCOL_FCOE].update && + p_results->arr[DCBX_PROTOCOL_FCOE].enable) + p_prio->fcoe = p_results->arr[DCBX_PROTOCOL_FCOE].priority; + + if (p_results->arr[DCBX_PROTOCOL_ETH].update && + p_results->arr[DCBX_PROTOCOL_ETH].enable) + p_prio->eth = p_results->arr[DCBX_PROTOCOL_ETH].priority; + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "Priorities: iscsi %d, roce %d, roce v2 %d, fcoe %d, eth %d\n", + p_prio->iscsi, p_prio->roce, p_prio->roce_v2, p_prio->fcoe, + p_prio->eth); +} + +static void +qed_dcbx_get_app_data(struct qed_hwfn *p_hwfn, + struct dcbx_app_priority_feature *p_app, + struct dcbx_app_priority_entry *p_tbl, + struct qed_dcbx_params *p_params) +{ + struct qed_app_entry *entry; + u8 pri_map; + int i; + + p_params->app_willing = QED_MFW_GET_FIELD(p_app->flags, + DCBX_APP_WILLING); + p_params->app_valid = QED_MFW_GET_FIELD(p_app->flags, DCBX_APP_ENABLED); + p_params->app_error = QED_MFW_GET_FIELD(p_app->flags, DCBX_APP_ERROR); + p_params->num_app_entries = QED_MFW_GET_FIELD(p_app->flags, + DCBX_APP_NUM_ENTRIES); + for (i = 0; i < DCBX_MAX_APP_PROTOCOL; i++) { + entry = &p_params->app_entry[i]; + entry->ethtype = !(QED_MFW_GET_FIELD(p_tbl[i].entry, + DCBX_APP_SF)); + pri_map = QED_MFW_GET_FIELD(p_tbl[i].entry, DCBX_APP_PRI_MAP); + entry->prio = ffs(pri_map) - 1; + entry->proto_id = QED_MFW_GET_FIELD(p_tbl[i].entry, + DCBX_APP_PROTOCOL_ID); + qed_dcbx_get_app_protocol_type(p_hwfn, p_tbl[i].entry, + entry->proto_id, + &entry->proto_type); + } + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "APP params: willing %d, valid %d error = %d\n", + p_params->app_willing, p_params->app_valid, + p_params->app_error); +} + +static void +qed_dcbx_get_pfc_data(struct qed_hwfn *p_hwfn, + u32 pfc, struct qed_dcbx_params *p_params) +{ + u8 pfc_map; + + p_params->pfc.willing = QED_MFW_GET_FIELD(pfc, DCBX_PFC_WILLING); + p_params->pfc.max_tc = QED_MFW_GET_FIELD(pfc, DCBX_PFC_CAPS); + p_params->pfc.enabled = QED_MFW_GET_FIELD(pfc, DCBX_PFC_ENABLED); + pfc_map = QED_MFW_GET_FIELD(pfc, DCBX_PFC_PRI_EN_BITMAP); + p_params->pfc.prio[0] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_0); + p_params->pfc.prio[1] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_1); + p_params->pfc.prio[2] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_2); + p_params->pfc.prio[3] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_3); + p_params->pfc.prio[4] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_4); + p_params->pfc.prio[5] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_5); + p_params->pfc.prio[6] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_6); + p_params->pfc.prio[7] = !!(pfc_map & DCBX_PFC_PRI_EN_BITMAP_PRI_7); + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "PFC params: willing %d, pfc_bitmap %d\n", + p_params->pfc.willing, pfc_map); +} + +static void +qed_dcbx_get_ets_data(struct qed_hwfn *p_hwfn, + struct dcbx_ets_feature *p_ets, + struct qed_dcbx_params *p_params) +{ + u32 bw_map[2], tsa_map[2], pri_map; + int i; + + p_params->ets_willing = QED_MFW_GET_FIELD(p_ets->flags, + DCBX_ETS_WILLING); + p_params->ets_enabled = QED_MFW_GET_FIELD(p_ets->flags, + DCBX_ETS_ENABLED); + p_params->ets_cbs = QED_MFW_GET_FIELD(p_ets->flags, DCBX_ETS_CBS); + p_params->max_ets_tc = QED_MFW_GET_FIELD(p_ets->flags, + DCBX_ETS_MAX_TCS); + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "ETS params: willing %d, ets_cbs %d pri_tc_tbl_0 %x max_ets_tc %d\n", + p_params->ets_willing, + p_params->ets_cbs, + p_ets->pri_tc_tbl[0], p_params->max_ets_tc); + + /* 8 bit tsa and bw data corresponding to each of the 8 TC's are + * encoded in a type u32 array of size 2. + */ + bw_map[0] = be32_to_cpu(p_ets->tc_bw_tbl[0]); + bw_map[1] = be32_to_cpu(p_ets->tc_bw_tbl[1]); + tsa_map[0] = be32_to_cpu(p_ets->tc_tsa_tbl[0]); + tsa_map[1] = be32_to_cpu(p_ets->tc_tsa_tbl[1]); + pri_map = be32_to_cpu(p_ets->pri_tc_tbl[0]); + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) { + p_params->ets_tc_bw_tbl[i] = ((u8 *)bw_map)[i]; + p_params->ets_tc_tsa_tbl[i] = ((u8 *)tsa_map)[i]; + p_params->ets_pri_tc_tbl[i] = QED_DCBX_PRIO2TC(pri_map, i); + DP_VERBOSE(p_hwfn, QED_MSG_DCB, + "elem %d bw_tbl %x tsa_tbl %x\n", + i, p_params->ets_tc_bw_tbl[i], + p_params->ets_tc_tsa_tbl[i]); + } +} + +static void +qed_dcbx_get_common_params(struct qed_hwfn *p_hwfn, + struct dcbx_app_priority_feature *p_app, + struct dcbx_app_priority_entry *p_tbl, + struct dcbx_ets_feature *p_ets, + u32 pfc, struct qed_dcbx_params *p_params) +{ + qed_dcbx_get_app_data(p_hwfn, p_app, p_tbl, p_params); + qed_dcbx_get_ets_data(p_hwfn, p_ets, p_params); + qed_dcbx_get_pfc_data(p_hwfn, pfc, p_params); +} + +static void +qed_dcbx_get_local_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, struct qed_dcbx_get *params) +{ + struct dcbx_features *p_feat; + + p_feat = &p_hwfn->p_dcbx_info->local_admin.features; + qed_dcbx_get_common_params(p_hwfn, &p_feat->app, + p_feat->app.app_pri_tbl, &p_feat->ets, + p_feat->pfc, ¶ms->local.params); + params->local.valid = true; +} + +static void +qed_dcbx_get_remote_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, struct qed_dcbx_get *params) +{ + struct dcbx_features *p_feat; + + p_feat = &p_hwfn->p_dcbx_info->remote.features; + qed_dcbx_get_common_params(p_hwfn, &p_feat->app, + p_feat->app.app_pri_tbl, &p_feat->ets, + p_feat->pfc, ¶ms->remote.params); + params->remote.valid = true; +} + +static void +qed_dcbx_get_operational_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_dcbx_get *params) +{ + struct qed_dcbx_operational_params *p_operational; + struct qed_dcbx_results *p_results; + struct dcbx_features *p_feat; + bool enabled, err; + u32 flags; + bool val; + + flags = p_hwfn->p_dcbx_info->operational.flags; + + /* If DCBx version is non zero, then negotiation + * was successfuly performed + */ + p_operational = ¶ms->operational; + enabled = !!(QED_MFW_GET_FIELD(flags, DCBX_CONFIG_VERSION) != + DCBX_CONFIG_VERSION_DISABLED); + if (!enabled) { + p_operational->enabled = enabled; + p_operational->valid = false; + return; + } + + p_feat = &p_hwfn->p_dcbx_info->operational.features; + p_results = &p_hwfn->p_dcbx_info->results; + + val = !!(QED_MFW_GET_FIELD(flags, DCBX_CONFIG_VERSION) == + DCBX_CONFIG_VERSION_IEEE); + p_operational->ieee = val; + val = !!(QED_MFW_GET_FIELD(flags, DCBX_CONFIG_VERSION) == + DCBX_CONFIG_VERSION_CEE); + p_operational->cee = val; + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, "Version support: ieee %d, cee %d\n", + p_operational->ieee, p_operational->cee); + + qed_dcbx_get_common_params(p_hwfn, &p_feat->app, + p_feat->app.app_pri_tbl, &p_feat->ets, + p_feat->pfc, ¶ms->operational.params); + qed_dcbx_get_priority_info(p_hwfn, &p_operational->app_prio, p_results); + err = QED_MFW_GET_FIELD(p_feat->app.flags, DCBX_APP_ERROR); + p_operational->err = err; + p_operational->enabled = enabled; + p_operational->valid = true; +} + +static void +qed_dcbx_get_local_lldp_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_dcbx_get *params) +{ + struct lldp_config_params_s *p_local; + + p_local = &p_hwfn->p_dcbx_info->lldp_local[LLDP_NEAREST_BRIDGE]; + + memcpy(params->lldp_local.local_chassis_id, p_local->local_chassis_id, + ARRAY_SIZE(p_local->local_chassis_id)); + memcpy(params->lldp_local.local_port_id, p_local->local_port_id, + ARRAY_SIZE(p_local->local_port_id)); +} + +static void +qed_dcbx_get_remote_lldp_params(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_dcbx_get *params) +{ + struct lldp_status_params_s *p_remote; + + p_remote = &p_hwfn->p_dcbx_info->lldp_remote[LLDP_NEAREST_BRIDGE]; + + memcpy(params->lldp_remote.peer_chassis_id, p_remote->peer_chassis_id, + ARRAY_SIZE(p_remote->peer_chassis_id)); + memcpy(params->lldp_remote.peer_port_id, p_remote->peer_port_id, + ARRAY_SIZE(p_remote->peer_port_id)); +} + +static int +qed_dcbx_get_params(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + struct qed_dcbx_get *p_params, + enum qed_mib_read_type type) +{ + switch (type) { + case QED_DCBX_REMOTE_MIB: + qed_dcbx_get_remote_params(p_hwfn, p_ptt, p_params); + break; + case QED_DCBX_LOCAL_MIB: + qed_dcbx_get_local_params(p_hwfn, p_ptt, p_params); + break; + case QED_DCBX_OPERATIONAL_MIB: + qed_dcbx_get_operational_params(p_hwfn, p_ptt, p_params); + break; + case QED_DCBX_REMOTE_LLDP_MIB: + qed_dcbx_get_remote_lldp_params(p_hwfn, p_ptt, p_params); + break; + case QED_DCBX_LOCAL_LLDP_MIB: + qed_dcbx_get_local_lldp_params(p_hwfn, p_ptt, p_params); + break; + default: + DP_ERR(p_hwfn, "MIB read err, unknown mib type %d\n", type); + return -EINVAL; + } + + return 0; +} +#endif + static int qed_dcbx_read_local_lldp_mib(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { @@ -561,3 +848,247 @@ void qed_dcbx_set_pf_update_params(struct qed_dcbx_results *p_src, p_dcb_data = &p_dest->eth_dcb_data; qed_dcbx_update_protocol_data(p_dcb_data, p_src, DCBX_PROTOCOL_ETH); } + +#ifdef CONFIG_DCB +static int qed_dcbx_query_params(struct qed_hwfn *p_hwfn, + struct qed_dcbx_get *p_get, + enum qed_mib_read_type type) +{ + struct qed_ptt *p_ptt; + int rc; + + p_ptt = qed_ptt_acquire(p_hwfn); + if (!p_ptt) + return -EBUSY; + + rc = qed_dcbx_read_mib(p_hwfn, p_ptt, type); + if (rc) + goto out; + + rc = qed_dcbx_get_params(p_hwfn, p_ptt, p_get, type); + +out: + qed_ptt_release(p_hwfn, p_ptt); + return rc; +} + +static void +qed_dcbx_set_pfc_data(struct qed_hwfn *p_hwfn, + u32 *pfc, struct qed_dcbx_params *p_params) +{ + u8 pfc_map = 0; + int i; + + if (p_params->pfc.willing) + *pfc |= DCBX_PFC_WILLING_MASK; + else + *pfc &= ~DCBX_PFC_WILLING_MASK; + + if (p_params->pfc.enabled) + *pfc |= DCBX_PFC_ENABLED_MASK; + else + *pfc &= ~DCBX_PFC_ENABLED_MASK; + + *pfc &= ~DCBX_PFC_CAPS_MASK; + *pfc |= (u32)p_params->pfc.max_tc << DCBX_PFC_CAPS_SHIFT; + + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) + if (p_params->pfc.prio[i]) + pfc_map |= BIT(i); + + *pfc |= (pfc_map << DCBX_PFC_PRI_EN_BITMAP_SHIFT); + + DP_VERBOSE(p_hwfn, QED_MSG_DCB, "pfc = 0x%x\n", *pfc); +} + +static void +qed_dcbx_set_ets_data(struct qed_hwfn *p_hwfn, + struct dcbx_ets_feature *p_ets, + struct qed_dcbx_params *p_params) +{ + u8 *bw_map, *tsa_map; + u32 val; + int i; + + if (p_params->ets_willing) + p_ets->flags |= DCBX_ETS_WILLING_MASK; + else + p_ets->flags &= ~DCBX_ETS_WILLING_MASK; + + if (p_params->ets_cbs) + p_ets->flags |= DCBX_ETS_CBS_MASK; + else + p_ets->flags &= ~DCBX_ETS_CBS_MASK; + + if (p_params->ets_enabled) + p_ets->flags |= DCBX_ETS_ENABLED_MASK; + else + p_ets->flags &= ~DCBX_ETS_ENABLED_MASK; + + p_ets->flags &= ~DCBX_ETS_MAX_TCS_MASK; + p_ets->flags |= (u32)p_params->max_ets_tc << DCBX_ETS_MAX_TCS_SHIFT; + + bw_map = (u8 *)&p_ets->tc_bw_tbl[0]; + tsa_map = (u8 *)&p_ets->tc_tsa_tbl[0]; + p_ets->pri_tc_tbl[0] = 0; + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) { + bw_map[i] = p_params->ets_tc_bw_tbl[i]; + tsa_map[i] = p_params->ets_tc_tsa_tbl[i]; + /* Copy the priority value to the corresponding 4 bits in the + * traffic class table. + */ + val = (((u32)p_params->ets_pri_tc_tbl[i]) << ((7 - i) * 4)); + p_ets->pri_tc_tbl[0] |= val; + } + p_ets->pri_tc_tbl[0] = cpu_to_be32(p_ets->pri_tc_tbl[0]); + for (i = 0; i < 2; i++) { + p_ets->tc_bw_tbl[i] = cpu_to_be32(p_ets->tc_bw_tbl[i]); + p_ets->tc_tsa_tbl[i] = cpu_to_be32(p_ets->tc_tsa_tbl[i]); + } +} + +static void +qed_dcbx_set_app_data(struct qed_hwfn *p_hwfn, + struct dcbx_app_priority_feature *p_app, + struct qed_dcbx_params *p_params) +{ + u32 *entry; + int i; + + if (p_params->app_willing) + p_app->flags |= DCBX_APP_WILLING_MASK; + else + p_app->flags &= ~DCBX_APP_WILLING_MASK; + + if (p_params->app_valid) + p_app->flags |= DCBX_APP_ENABLED_MASK; + else + p_app->flags &= ~DCBX_APP_ENABLED_MASK; + + p_app->flags &= ~DCBX_APP_NUM_ENTRIES_MASK; + p_app->flags |= (u32)p_params->num_app_entries << + DCBX_APP_NUM_ENTRIES_SHIFT; + + for (i = 0; i < DCBX_MAX_APP_PROTOCOL; i++) { + entry = &p_app->app_pri_tbl[i].entry; + *entry &= ~DCBX_APP_SF_MASK; + if (p_params->app_entry[i].ethtype) + *entry |= ((u32)DCBX_APP_SF_ETHTYPE << + DCBX_APP_SF_SHIFT); + else + *entry |= ((u32)DCBX_APP_SF_PORT << DCBX_APP_SF_SHIFT); + *entry &= ~DCBX_APP_PROTOCOL_ID_MASK; + *entry |= ((u32)p_params->app_entry[i].proto_id << + DCBX_APP_PROTOCOL_ID_SHIFT); + *entry &= ~DCBX_APP_PRI_MAP_MASK; + *entry |= ((u32)(p_params->app_entry[i].prio) << + DCBX_APP_PRI_MAP_SHIFT); + } +} + +static void +qed_dcbx_set_local_params(struct qed_hwfn *p_hwfn, + struct dcbx_local_params *local_admin, + struct qed_dcbx_set *params) +{ + local_admin->flags = 0; + memcpy(&local_admin->features, + &p_hwfn->p_dcbx_info->operational.features, + sizeof(local_admin->features)); + + if (params->enabled) + local_admin->config = params->ver_num; + else + local_admin->config = DCBX_CONFIG_VERSION_DISABLED; + + if (params->override_flags & QED_DCBX_OVERRIDE_PFC_CFG) + qed_dcbx_set_pfc_data(p_hwfn, &local_admin->features.pfc, + ¶ms->config.params); + + if (params->override_flags & QED_DCBX_OVERRIDE_ETS_CFG) + qed_dcbx_set_ets_data(p_hwfn, &local_admin->features.ets, + ¶ms->config.params); + + if (params->override_flags & QED_DCBX_OVERRIDE_APP_CFG) + qed_dcbx_set_app_data(p_hwfn, &local_admin->features.app, + ¶ms->config.params); +} + +int qed_dcbx_config_params(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + struct qed_dcbx_set *params, bool hw_commit) +{ + struct dcbx_local_params local_admin; + struct qed_dcbx_mib_meta_data data; + u32 resp = 0, param = 0; + int rc = 0; + + if (!hw_commit) { + memcpy(&p_hwfn->p_dcbx_info->set, params, + sizeof(struct qed_dcbx_set)); + return 0; + } + + /* clear set-parmas cache */ + memset(&p_hwfn->p_dcbx_info->set, 0, sizeof(p_hwfn->p_dcbx_info->set)); + + memset(&local_admin, 0, sizeof(local_admin)); + qed_dcbx_set_local_params(p_hwfn, &local_admin, params); + + data.addr = p_hwfn->mcp_info->port_addr + + offsetof(struct public_port, local_admin_dcbx_mib); + data.local_admin = &local_admin; + data.size = sizeof(struct dcbx_local_params); + qed_memcpy_to(p_hwfn, p_ptt, data.addr, data.local_admin, data.size); + + rc = qed_mcp_cmd(p_hwfn, p_ptt, DRV_MSG_CODE_SET_DCBX, + 1 << DRV_MB_PARAM_LLDP_SEND_SHIFT, &resp, ¶m); + if (rc) + DP_NOTICE(p_hwfn, "Failed to send DCBX update request\n"); + + return rc; +} + +int qed_dcbx_get_config_params(struct qed_hwfn *p_hwfn, + struct qed_dcbx_set *params) +{ + struct qed_dcbx_get *dcbx_info; + int rc; + + if (p_hwfn->p_dcbx_info->set.config.valid) { + memcpy(params, &p_hwfn->p_dcbx_info->set, + sizeof(struct qed_dcbx_set)); + return 0; + } + + dcbx_info = kmalloc(sizeof(*dcbx_info), GFP_KERNEL); + if (!dcbx_info) { + DP_ERR(p_hwfn, "Failed to allocate struct qed_dcbx_info\n"); + return -ENOMEM; + } + + rc = qed_dcbx_query_params(p_hwfn, dcbx_info, QED_DCBX_OPERATIONAL_MIB); + if (rc) { + kfree(dcbx_info); + return rc; + } + + p_hwfn->p_dcbx_info->set.override_flags = 0; + p_hwfn->p_dcbx_info->set.ver_num = DCBX_CONFIG_VERSION_DISABLED; + if (dcbx_info->operational.cee) + p_hwfn->p_dcbx_info->set.ver_num |= DCBX_CONFIG_VERSION_CEE; + if (dcbx_info->operational.ieee) + p_hwfn->p_dcbx_info->set.ver_num |= DCBX_CONFIG_VERSION_IEEE; + + p_hwfn->p_dcbx_info->set.enabled = dcbx_info->operational.enabled; + memcpy(&p_hwfn->p_dcbx_info->set.config.params, + &dcbx_info->operational.params, + sizeof(struct qed_dcbx_admin_params)); + p_hwfn->p_dcbx_info->set.config.valid = true; + + memcpy(params, &p_hwfn->p_dcbx_info->set, sizeof(struct qed_dcbx_set)); + + kfree(dcbx_info); + + return 0; +} +#endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.h b/drivers/net/ethernet/qlogic/qed/qed_dcbx.h index e7f834dbda2d..9ba681643d05 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.h +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.h @@ -33,6 +33,24 @@ struct qed_dcbx_app_data { u8 tc; /* Traffic Class */ }; +#ifdef CONFIG_DCB +#define QED_DCBX_VERSION_DISABLED 0 +#define QED_DCBX_VERSION_IEEE 1 +#define QED_DCBX_VERSION_CEE 2 + +struct qed_dcbx_set { +#define QED_DCBX_OVERRIDE_STATE BIT(0) +#define QED_DCBX_OVERRIDE_PFC_CFG BIT(1) +#define QED_DCBX_OVERRIDE_ETS_CFG BIT(2) +#define QED_DCBX_OVERRIDE_APP_CFG BIT(3) +#define QED_DCBX_OVERRIDE_DSCP_CFG BIT(4) + u32 override_flags; + bool enabled; + struct qed_dcbx_admin_params config; + u32 ver_num; +}; +#endif + struct qed_dcbx_results { bool dcbx_enabled; u8 pf_id; @@ -55,6 +73,9 @@ struct qed_dcbx_info { struct qed_dcbx_results results; struct dcbx_mib operational; struct dcbx_mib remote; +#ifdef CONFIG_DCB + struct qed_dcbx_set set; +#endif u8 dcbx_cap; }; @@ -67,6 +88,13 @@ struct qed_dcbx_mib_meta_data { u32 addr; }; +#ifdef CONFIG_DCB +int qed_dcbx_get_config_params(struct qed_hwfn *, struct qed_dcbx_set *); + +int qed_dcbx_config_params(struct qed_hwfn *, + struct qed_ptt *, struct qed_dcbx_set *, bool); +#endif + /* QED local interface routines */ int qed_dcbx_mib_update_event(struct qed_hwfn *, diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h index 72e0ad6d16c6..592784019994 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h @@ -7244,6 +7244,9 @@ struct public_drv_mb { #define DRV_MB_PARAM_CFG_VF_MSIX_VF_ID_MASK 0x000000FF #define DRV_MB_PARAM_CFG_VF_MSIX_SB_NUM_SHIFT 8 #define DRV_MB_PARAM_CFG_VF_MSIX_SB_NUM_MASK 0x0000FF00 +#define DRV_MB_PARAM_LLDP_SEND_MASK 0x00000001 +#define DRV_MB_PARAM_LLDP_SEND_SHIFT 0 + #define DRV_MB_PARAM_SET_LED_MODE_OPER 0x0 #define DRV_MB_PARAM_SET_LED_MODE_ON 0x1 diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h index e8cc49f9688a..e1d5122e8a96 100644 --- a/include/linux/qed/qed_if.h +++ b/include/linux/qed/qed_if.h @@ -34,6 +34,96 @@ enum dcbx_protocol_type { DCBX_MAX_PROTOCOL_TYPE }; +#ifdef CONFIG_DCB +#define QED_LLDP_CHASSIS_ID_STAT_LEN 4 +#define QED_LLDP_PORT_ID_STAT_LEN 4 +#define QED_DCBX_MAX_APP_PROTOCOL 32 +#define QED_MAX_PFC_PRIORITIES 8 +#define QED_DCBX_DSCP_SIZE 64 + +struct qed_dcbx_lldp_remote { + u32 peer_chassis_id[QED_LLDP_CHASSIS_ID_STAT_LEN]; + u32 peer_port_id[QED_LLDP_PORT_ID_STAT_LEN]; + bool enable_rx; + bool enable_tx; + u32 tx_interval; + u32 max_credit; +}; + +struct qed_dcbx_lldp_local { + u32 local_chassis_id[QED_LLDP_CHASSIS_ID_STAT_LEN]; + u32 local_port_id[QED_LLDP_PORT_ID_STAT_LEN]; +}; + +struct qed_dcbx_app_prio { + u8 roce; + u8 roce_v2; + u8 fcoe; + u8 iscsi; + u8 eth; +}; + +struct qed_dbcx_pfc_params { + bool willing; + bool enabled; + u8 prio[QED_MAX_PFC_PRIORITIES]; + u8 max_tc; +}; + +struct qed_app_entry { + bool ethtype; + bool enabled; + u8 prio; + u16 proto_id; + enum dcbx_protocol_type proto_type; +}; + +struct qed_dcbx_params { + struct qed_app_entry app_entry[QED_DCBX_MAX_APP_PROTOCOL]; + u16 num_app_entries; + bool app_willing; + bool app_valid; + bool app_error; + bool ets_willing; + bool ets_enabled; + bool ets_cbs; + bool valid; + u8 ets_pri_tc_tbl[QED_MAX_PFC_PRIORITIES]; + u8 ets_tc_bw_tbl[QED_MAX_PFC_PRIORITIES]; + u8 ets_tc_tsa_tbl[QED_MAX_PFC_PRIORITIES]; + struct qed_dbcx_pfc_params pfc; + u8 max_ets_tc; +}; + +struct qed_dcbx_admin_params { + struct qed_dcbx_params params; + bool valid; +}; + +struct qed_dcbx_remote_params { + struct qed_dcbx_params params; + bool valid; +}; + +struct qed_dcbx_operational_params { + struct qed_dcbx_app_prio app_prio; + struct qed_dcbx_params params; + bool valid; + bool enabled; + bool ieee; + bool cee; + u32 err; +}; + +struct qed_dcbx_get { + struct qed_dcbx_operational_params operational; + struct qed_dcbx_lldp_remote lldp_remote; + struct qed_dcbx_lldp_local lldp_local; + struct qed_dcbx_remote_params remote; + struct qed_dcbx_admin_params local; +}; +#endif + enum qed_led_mode { QED_LED_MODE_OFF, QED_LED_MODE_ON, -- cgit v1.2.3 From a1d8d8a51e8317269dd127d94b9de14f67d9563f Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Wed, 8 Jun 2016 06:22:11 -0400 Subject: qed: Add dcbnl support. This patch adds the implementation for both cee/ieee dcbnl callbacks by using the qed query/config APIs. Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_dcbx.c | 1090 ++++++++++++++++++++++++++++ drivers/net/ethernet/qlogic/qed/qed_l2.c | 7 + include/linux/qed/qed_eth_if.h | 62 ++ 3 files changed, 1159 insertions(+) (limited to 'include/linux') diff --git a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c index e782484d2d89..d0dc28f93c0e 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dcbx.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dcbx.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include @@ -18,6 +19,9 @@ #include "qed_dcbx.h" #include "qed_hsi.h" #include "qed_sp.h" +#ifdef CONFIG_DCB +#include +#endif #define QED_DCBX_MAX_MIB_READ_TRY (100) #define QED_ETH_TYPE_DEFAULT (0) @@ -1091,4 +1095,1090 @@ int qed_dcbx_get_config_params(struct qed_hwfn *p_hwfn, return 0; } + +static struct qed_dcbx_get *qed_dcbnl_get_dcbx(struct qed_hwfn *hwfn, + enum qed_mib_read_type type) +{ + struct qed_dcbx_get *dcbx_info; + + dcbx_info = kmalloc(sizeof(*dcbx_info), GFP_KERNEL); + if (!dcbx_info) { + DP_ERR(hwfn->cdev, "Failed to allocate memory for dcbx_info\n"); + return NULL; + } + + if (qed_dcbx_query_params(hwfn, dcbx_info, type)) { + kfree(dcbx_info); + return NULL; + } + + if ((type == QED_DCBX_OPERATIONAL_MIB) && + !dcbx_info->operational.enabled) { + DP_INFO(hwfn, "DCBX is not enabled/operational\n"); + kfree(dcbx_info); + return NULL; + } + + return dcbx_info; +} + +static u8 qed_dcbnl_getstate(struct qed_dev *cdev) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + bool enabled; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 0; + + enabled = dcbx_info->operational.enabled; + DP_VERBOSE(hwfn, QED_MSG_DCB, "DCB state = %d\n", enabled); + kfree(dcbx_info); + + return enabled; +} + +static u8 qed_dcbnl_setstate(struct qed_dev *cdev, u8 state) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "DCB state = %d\n", state); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + dcbx_set.enabled = !!state; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return 1; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc ? 1 : 0; +} + +static void qed_dcbnl_getpgtccfgtx(struct qed_dev *cdev, int tc, u8 *prio_type, + u8 *pgid, u8 *bw_pct, u8 *up_map) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "tc = %d\n", tc); + *prio_type = *pgid = *bw_pct = *up_map = 0; + if (tc < 0 || tc >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid tc %d\n", tc); + return; + } + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return; + + *pgid = dcbx_info->operational.params.ets_pri_tc_tbl[tc]; + kfree(dcbx_info); +} + +static void qed_dcbnl_getpgbwgcfgtx(struct qed_dev *cdev, int pgid, u8 *bw_pct) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + *bw_pct = 0; + DP_VERBOSE(hwfn, QED_MSG_DCB, "pgid = %d\n", pgid); + if (pgid < 0 || pgid >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid pgid %d\n", pgid); + return; + } + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return; + + *bw_pct = dcbx_info->operational.params.ets_tc_bw_tbl[pgid]; + DP_VERBOSE(hwfn, QED_MSG_DCB, "bw_pct = %d\n", *bw_pct); + kfree(dcbx_info); +} + +static void qed_dcbnl_getpgtccfgrx(struct qed_dev *cdev, int tc, u8 *prio, + u8 *bwg_id, u8 *bw_pct, u8 *up_map) +{ + DP_INFO(QED_LEADING_HWFN(cdev), "Rx ETS is not supported\n"); + *prio = *bwg_id = *bw_pct = *up_map = 0; +} + +static void qed_dcbnl_getpgbwgcfgrx(struct qed_dev *cdev, + int bwg_id, u8 *bw_pct) +{ + DP_INFO(QED_LEADING_HWFN(cdev), "Rx ETS is not supported\n"); + *bw_pct = 0; +} + +static void qed_dcbnl_getpfccfg(struct qed_dev *cdev, + int priority, u8 *setting) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "priority = %d\n", priority); + if (priority < 0 || priority >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid priority %d\n", priority); + return; + } + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return; + + *setting = dcbx_info->operational.params.pfc.prio[priority]; + DP_VERBOSE(hwfn, QED_MSG_DCB, "setting = %d\n", *setting); + kfree(dcbx_info); +} + +static void qed_dcbnl_setpfccfg(struct qed_dev *cdev, int priority, u8 setting) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "priority = %d setting = %d\n", + priority, setting); + if (priority < 0 || priority >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid priority %d\n", priority); + return; + } + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + dcbx_set.config.params.pfc.prio[priority] = !!setting; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); +} + +static u8 qed_dcbnl_getcap(struct qed_dev *cdev, int capid, u8 *cap) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int rc = 0; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "capid = %d\n", capid); + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 1; + + switch (capid) { + case DCB_CAP_ATTR_PG: + case DCB_CAP_ATTR_PFC: + case DCB_CAP_ATTR_UP2TC: + case DCB_CAP_ATTR_GSP: + *cap = true; + break; + case DCB_CAP_ATTR_PG_TCS: + case DCB_CAP_ATTR_PFC_TCS: + *cap = 0x80; + break; + case DCB_CAP_ATTR_DCBX: + *cap = (DCB_CAP_DCBX_LLD_MANAGED | DCB_CAP_DCBX_VER_CEE | + DCB_CAP_DCBX_VER_IEEE); + break; + default: + *cap = false; + rc = 1; + } + + DP_VERBOSE(hwfn, QED_MSG_DCB, "id = %d caps = %d\n", capid, *cap); + kfree(dcbx_info); + + return rc; +} + +static int qed_dcbnl_getnumtcs(struct qed_dev *cdev, int tcid, u8 *num) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int rc = 0; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "tcid = %d\n", tcid); + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + switch (tcid) { + case DCB_NUMTCS_ATTR_PG: + *num = dcbx_info->operational.params.max_ets_tc; + break; + case DCB_NUMTCS_ATTR_PFC: + *num = dcbx_info->operational.params.pfc.max_tc; + break; + default: + rc = -EINVAL; + } + + kfree(dcbx_info); + DP_VERBOSE(hwfn, QED_MSG_DCB, "numtcs = %d\n", *num); + + return rc; +} + +static u8 qed_dcbnl_getpfcstate(struct qed_dev *cdev) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + bool enabled; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 0; + + enabled = dcbx_info->operational.params.pfc.enabled; + DP_VERBOSE(hwfn, QED_MSG_DCB, "pfc state = %d\n", enabled); + kfree(dcbx_info); + + return enabled; +} + +static u8 qed_dcbnl_getdcbx(struct qed_dev *cdev) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + u8 mode = 0; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 0; + + if (dcbx_info->operational.enabled) + mode |= DCB_CAP_DCBX_LLD_MANAGED; + if (dcbx_info->operational.ieee) + mode |= DCB_CAP_DCBX_VER_IEEE; + if (dcbx_info->operational.cee) + mode |= DCB_CAP_DCBX_VER_CEE; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "dcb mode = %d\n", mode); + kfree(dcbx_info); + + return mode; +} + +static void qed_dcbnl_setpgtccfgtx(struct qed_dev *cdev, + int tc, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, + "tc = %d pri_type = %d pgid = %d bw_pct = %d up_map = %d\n", + tc, pri_type, pgid, bw_pct, up_map); + + if (tc < 0 || tc >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid tc %d\n", tc); + return; + } + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.ets_pri_tc_tbl[tc] = pgid; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); +} + +static void qed_dcbnl_setpgtccfgrx(struct qed_dev *cdev, int prio, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map) +{ + DP_INFO(QED_LEADING_HWFN(cdev), "Rx ETS is not supported\n"); +} + +static void qed_dcbnl_setpgbwgcfgtx(struct qed_dev *cdev, int pgid, u8 bw_pct) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "pgid = %d bw_pct = %d\n", pgid, bw_pct); + if (pgid < 0 || pgid >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid pgid %d\n", pgid); + return; + } + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.ets_tc_bw_tbl[pgid] = bw_pct; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); +} + +static void qed_dcbnl_setpgbwgcfgrx(struct qed_dev *cdev, int pgid, u8 bw_pct) +{ + DP_INFO(QED_LEADING_HWFN(cdev), "Rx ETS is not supported\n"); +} + +static u8 qed_dcbnl_setall(struct qed_dev *cdev) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return 1; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 1); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +static int qed_dcbnl_setnumtcs(struct qed_dev *cdev, int tcid, u8 num) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "tcid = %d num = %d\n", tcid, num); + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + switch (tcid) { + case DCB_NUMTCS_ATTR_PG: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.max_ets_tc = num; + break; + case DCB_NUMTCS_ATTR_PFC: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + dcbx_set.config.params.pfc.max_tc = num; + break; + default: + DP_INFO(hwfn, "Invalid tcid %d\n", tcid); + return -EINVAL; + } + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EINVAL; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return 0; +} + +static void qed_dcbnl_setpfcstate(struct qed_dev *cdev, u8 state) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "new state = %d\n", state); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + dcbx_set.config.params.pfc.enabled = !!state; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); +} + +static int qed_dcbnl_getapp(struct qed_dev *cdev, u8 idtype, u16 idval) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_app_entry *entry; + bool ethtype; + u8 prio = 0; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + ethtype = !!(idtype == DCB_APP_IDTYPE_ETHTYPE); + for (i = 0; i < QED_DCBX_MAX_APP_PROTOCOL; i++) { + entry = &dcbx_info->operational.params.app_entry[i]; + if ((entry->ethtype == ethtype) && (entry->proto_id == idval)) { + prio = entry->prio; + break; + } + } + + if (i == QED_DCBX_MAX_APP_PROTOCOL) { + DP_ERR(cdev, "App entry (%d, %d) not found\n", idtype, idval); + kfree(dcbx_info); + return -EINVAL; + } + + kfree(dcbx_info); + + return prio; +} + +static int qed_dcbnl_setapp(struct qed_dev *cdev, + u8 idtype, u16 idval, u8 pri_map) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_app_entry *entry; + struct qed_ptt *ptt; + bool ethtype; + int rc, i; + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return -EINVAL; + + ethtype = !!(idtype == DCB_APP_IDTYPE_ETHTYPE); + for (i = 0; i < QED_DCBX_MAX_APP_PROTOCOL; i++) { + entry = &dcbx_set.config.params.app_entry[i]; + if ((entry->ethtype == ethtype) && (entry->proto_id == idval)) + break; + /* First empty slot */ + if (!entry->proto_id) + break; + } + + if (i == QED_DCBX_MAX_APP_PROTOCOL) { + DP_ERR(cdev, "App table is full\n"); + return -EBUSY; + } + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_APP_CFG; + dcbx_set.config.params.app_entry[i].ethtype = ethtype; + dcbx_set.config.params.app_entry[i].proto_id = idval; + dcbx_set.config.params.app_entry[i].prio = pri_map; + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EBUSY; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +static u8 qed_dcbnl_setdcbx(struct qed_dev *cdev, u8 mode) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "new mode = %x\n", mode); + + if (!(mode & DCB_CAP_DCBX_VER_IEEE) && !(mode & DCB_CAP_DCBX_VER_CEE)) { + DP_INFO(hwfn, "Allowed mode is cee, ieee or both\n"); + return 1; + } + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + dcbx_set.ver_num = 0; + if (mode & DCB_CAP_DCBX_VER_CEE) { + dcbx_set.ver_num |= DCBX_CONFIG_VERSION_CEE; + dcbx_set.enabled = true; + } + + if (mode & DCB_CAP_DCBX_VER_IEEE) { + dcbx_set.ver_num |= DCBX_CONFIG_VERSION_IEEE; + dcbx_set.enabled = true; + } + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return 1; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return 0; +} + +static u8 qed_dcbnl_getfeatcfg(struct qed_dev *cdev, int featid, u8 *flags) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "Feature id = %d\n", featid); + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return 1; + + *flags = 0; + switch (featid) { + case DCB_FEATCFG_ATTR_PG: + if (dcbx_info->operational.params.ets_enabled) + *flags = DCB_FEATCFG_ENABLE; + else + *flags = DCB_FEATCFG_ERROR; + break; + case DCB_FEATCFG_ATTR_PFC: + if (dcbx_info->operational.params.pfc.enabled) + *flags = DCB_FEATCFG_ENABLE; + else + *flags = DCB_FEATCFG_ERROR; + break; + case DCB_FEATCFG_ATTR_APP: + if (dcbx_info->operational.params.app_valid) + *flags = DCB_FEATCFG_ENABLE; + else + *flags = DCB_FEATCFG_ERROR; + break; + default: + DP_INFO(hwfn, "Invalid feature-ID %d\n", featid); + kfree(dcbx_info); + return 1; + } + + DP_VERBOSE(hwfn, QED_MSG_DCB, "flags = %d\n", *flags); + kfree(dcbx_info); + + return 0; +} + +static u8 qed_dcbnl_setfeatcfg(struct qed_dev *cdev, int featid, u8 flags) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_set dcbx_set; + bool enabled, willing; + struct qed_ptt *ptt; + int rc; + + DP_VERBOSE(hwfn, QED_MSG_DCB, "featid = %d flags = %d\n", + featid, flags); + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return 1; + + enabled = !!(flags & DCB_FEATCFG_ENABLE); + willing = !!(flags & DCB_FEATCFG_WILLING); + switch (featid) { + case DCB_FEATCFG_ATTR_PG: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.ets_enabled = enabled; + dcbx_set.config.params.ets_willing = willing; + break; + case DCB_FEATCFG_ATTR_PFC: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + dcbx_set.config.params.pfc.enabled = enabled; + dcbx_set.config.params.pfc.willing = willing; + break; + case DCB_FEATCFG_ATTR_APP: + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_APP_CFG; + dcbx_set.config.params.app_willing = willing; + break; + default: + DP_INFO(hwfn, "Invalid feature-ID %d\n", featid); + return 1; + } + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return 1; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return 0; +} + +static int qed_dcbnl_peer_getappinfo(struct qed_dev *cdev, + struct dcb_peer_app_info *info, + u16 *app_count) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_REMOTE_MIB); + if (!dcbx_info) + return -EINVAL; + + info->willing = dcbx_info->remote.params.app_willing; + info->error = dcbx_info->remote.params.app_error; + *app_count = dcbx_info->remote.params.num_app_entries; + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_peer_getapptable(struct qed_dev *cdev, + struct dcb_app *table) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_REMOTE_MIB); + if (!dcbx_info) + return -EINVAL; + + for (i = 0; i < dcbx_info->remote.params.num_app_entries; i++) { + if (dcbx_info->remote.params.app_entry[i].ethtype) + table[i].selector = DCB_APP_IDTYPE_ETHTYPE; + else + table[i].selector = DCB_APP_IDTYPE_PORTNUM; + table[i].priority = dcbx_info->remote.params.app_entry[i].prio; + table[i].protocol = + dcbx_info->remote.params.app_entry[i].proto_id; + } + + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_cee_peer_getpfc(struct qed_dev *cdev, struct cee_pfc *pfc) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_REMOTE_MIB); + if (!dcbx_info) + return -EINVAL; + + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) + if (dcbx_info->remote.params.pfc.prio[i]) + pfc->pfc_en |= BIT(i); + + pfc->tcs_supported = dcbx_info->remote.params.pfc.max_tc; + DP_VERBOSE(hwfn, QED_MSG_DCB, "pfc state = %d tcs_supported = %d\n", + pfc->pfc_en, pfc->tcs_supported); + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_cee_peer_getpg(struct qed_dev *cdev, struct cee_pg *pg) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_REMOTE_MIB); + if (!dcbx_info) + return -EINVAL; + + pg->willing = dcbx_info->remote.params.ets_willing; + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) { + pg->pg_bw[i] = dcbx_info->remote.params.ets_tc_bw_tbl[i]; + pg->prio_pg[i] = dcbx_info->remote.params.ets_pri_tc_tbl[i]; + } + + DP_VERBOSE(hwfn, QED_MSG_DCB, "willing = %d", pg->willing); + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_get_ieee_pfc(struct qed_dev *cdev, + struct ieee_pfc *pfc, bool remote) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_params *params; + struct qed_dcbx_get *dcbx_info; + int rc, i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + return -EINVAL; + } + + if (remote) { + memset(dcbx_info, 0, sizeof(*dcbx_info)); + rc = qed_dcbx_query_params(hwfn, dcbx_info, + QED_DCBX_REMOTE_MIB); + if (rc) { + kfree(dcbx_info); + return -EINVAL; + } + + params = &dcbx_info->remote.params; + } else { + params = &dcbx_info->operational.params; + } + + pfc->pfc_cap = params->pfc.max_tc; + pfc->pfc_en = 0; + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) + if (params->pfc.prio[i]) + pfc->pfc_en |= BIT(i); + + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_ieee_getpfc(struct qed_dev *cdev, struct ieee_pfc *pfc) +{ + return qed_dcbnl_get_ieee_pfc(cdev, pfc, false); +} + +static int qed_dcbnl_ieee_setpfc(struct qed_dev *cdev, struct ieee_pfc *pfc) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc, i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + kfree(dcbx_info); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return -EINVAL; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_PFC_CFG; + for (i = 0; i < QED_MAX_PFC_PRIORITIES; i++) + dcbx_set.config.params.pfc.prio[i] = !!(pfc->pfc_en & BIT(i)); + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EINVAL; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +static int qed_dcbnl_get_ieee_ets(struct qed_dev *cdev, + struct ieee_ets *ets, bool remote) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_dcbx_params *params; + int rc; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + if (remote) { + memset(dcbx_info, 0, sizeof(*dcbx_info)); + rc = qed_dcbx_query_params(hwfn, dcbx_info, + QED_DCBX_REMOTE_MIB); + if (rc) { + kfree(dcbx_info); + return -EINVAL; + } + + params = &dcbx_info->remote.params; + } else { + params = &dcbx_info->operational.params; + } + + ets->ets_cap = params->max_ets_tc; + ets->willing = params->ets_willing; + ets->cbs = params->ets_cbs; + memcpy(ets->tc_tx_bw, params->ets_tc_bw_tbl, sizeof(ets->tc_tx_bw)); + memcpy(ets->tc_tsa, params->ets_tc_tsa_tbl, sizeof(ets->tc_tsa)); + memcpy(ets->prio_tc, params->ets_pri_tc_tbl, sizeof(ets->prio_tc)); + kfree(dcbx_info); + + return 0; +} + +static int qed_dcbnl_ieee_getets(struct qed_dev *cdev, struct ieee_ets *ets) +{ + return qed_dcbnl_get_ieee_ets(cdev, ets, false); +} + +static int qed_dcbnl_ieee_setets(struct qed_dev *cdev, struct ieee_ets *ets) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_dcbx_set dcbx_set; + struct qed_ptt *ptt; + int rc; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + kfree(dcbx_info); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return -EINVAL; + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_ETS_CFG; + dcbx_set.config.params.max_ets_tc = ets->ets_cap; + dcbx_set.config.params.ets_willing = ets->willing; + dcbx_set.config.params.ets_cbs = ets->cbs; + memcpy(dcbx_set.config.params.ets_tc_bw_tbl, ets->tc_tx_bw, + sizeof(ets->tc_tx_bw)); + memcpy(dcbx_set.config.params.ets_tc_tsa_tbl, ets->tc_tsa, + sizeof(ets->tc_tsa)); + memcpy(dcbx_set.config.params.ets_pri_tc_tbl, ets->prio_tc, + sizeof(ets->prio_tc)); + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EINVAL; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +int qed_dcbnl_ieee_peer_getets(struct qed_dev *cdev, struct ieee_ets *ets) +{ + return qed_dcbnl_get_ieee_ets(cdev, ets, true); +} + +int qed_dcbnl_ieee_peer_getpfc(struct qed_dev *cdev, struct ieee_pfc *pfc) +{ + return qed_dcbnl_get_ieee_pfc(cdev, pfc, true); +} + +int qed_dcbnl_ieee_getapp(struct qed_dev *cdev, struct dcb_app *app) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_app_entry *entry; + bool ethtype; + u8 prio = 0; + int i; + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + /* ieee defines the selector field value for ethertype to be 1 */ + ethtype = !!((app->selector - 1) == DCB_APP_IDTYPE_ETHTYPE); + for (i = 0; i < QED_DCBX_MAX_APP_PROTOCOL; i++) { + entry = &dcbx_info->operational.params.app_entry[i]; + if ((entry->ethtype == ethtype) && + (entry->proto_id == app->protocol)) { + prio = entry->prio; + break; + } + } + + if (i == QED_DCBX_MAX_APP_PROTOCOL) { + DP_ERR(cdev, "App entry (%d, %d) not found\n", app->selector, + app->protocol); + kfree(dcbx_info); + return -EINVAL; + } + + app->priority = ffs(prio) - 1; + + kfree(dcbx_info); + + return 0; +} + +int qed_dcbnl_ieee_setapp(struct qed_dev *cdev, struct dcb_app *app) +{ + struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); + struct qed_dcbx_get *dcbx_info; + struct qed_dcbx_set dcbx_set; + struct qed_app_entry *entry; + struct qed_ptt *ptt; + bool ethtype; + int rc, i; + + if (app->priority < 0 || app->priority >= QED_MAX_PFC_PRIORITIES) { + DP_INFO(hwfn, "Invalid priority %d\n", app->priority); + return -EINVAL; + } + + dcbx_info = qed_dcbnl_get_dcbx(hwfn, QED_DCBX_OPERATIONAL_MIB); + if (!dcbx_info) + return -EINVAL; + + if (!dcbx_info->operational.ieee) { + DP_INFO(hwfn, "DCBX is not enabled/operational in IEEE mode\n"); + kfree(dcbx_info); + return -EINVAL; + } + + kfree(dcbx_info); + + memset(&dcbx_set, 0, sizeof(dcbx_set)); + rc = qed_dcbx_get_config_params(hwfn, &dcbx_set); + if (rc) + return -EINVAL; + + /* ieee defines the selector field value for ethertype to be 1 */ + ethtype = !!((app->selector - 1) == DCB_APP_IDTYPE_ETHTYPE); + for (i = 0; i < QED_DCBX_MAX_APP_PROTOCOL; i++) { + entry = &dcbx_set.config.params.app_entry[i]; + if ((entry->ethtype == ethtype) && + (entry->proto_id == app->protocol)) + break; + /* First empty slot */ + if (!entry->proto_id) + break; + } + + if (i == QED_DCBX_MAX_APP_PROTOCOL) { + DP_ERR(cdev, "App table is full\n"); + return -EBUSY; + } + + dcbx_set.override_flags |= QED_DCBX_OVERRIDE_APP_CFG; + dcbx_set.config.params.app_entry[i].ethtype = ethtype; + dcbx_set.config.params.app_entry[i].proto_id = app->protocol; + dcbx_set.config.params.app_entry[i].prio = BIT(app->priority); + + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EBUSY; + + rc = qed_dcbx_config_params(hwfn, ptt, &dcbx_set, 0); + + qed_ptt_release(hwfn, ptt); + + return rc; +} + +const struct qed_eth_dcbnl_ops qed_dcbnl_ops_pass = { + .getstate = qed_dcbnl_getstate, + .setstate = qed_dcbnl_setstate, + .getpgtccfgtx = qed_dcbnl_getpgtccfgtx, + .getpgbwgcfgtx = qed_dcbnl_getpgbwgcfgtx, + .getpgtccfgrx = qed_dcbnl_getpgtccfgrx, + .getpgbwgcfgrx = qed_dcbnl_getpgbwgcfgrx, + .getpfccfg = qed_dcbnl_getpfccfg, + .setpfccfg = qed_dcbnl_setpfccfg, + .getcap = qed_dcbnl_getcap, + .getnumtcs = qed_dcbnl_getnumtcs, + .getpfcstate = qed_dcbnl_getpfcstate, + .getdcbx = qed_dcbnl_getdcbx, + .setpgtccfgtx = qed_dcbnl_setpgtccfgtx, + .setpgtccfgrx = qed_dcbnl_setpgtccfgrx, + .setpgbwgcfgtx = qed_dcbnl_setpgbwgcfgtx, + .setpgbwgcfgrx = qed_dcbnl_setpgbwgcfgrx, + .setall = qed_dcbnl_setall, + .setnumtcs = qed_dcbnl_setnumtcs, + .setpfcstate = qed_dcbnl_setpfcstate, + .setapp = qed_dcbnl_setapp, + .setdcbx = qed_dcbnl_setdcbx, + .setfeatcfg = qed_dcbnl_setfeatcfg, + .getfeatcfg = qed_dcbnl_getfeatcfg, + .getapp = qed_dcbnl_getapp, + .peer_getappinfo = qed_dcbnl_peer_getappinfo, + .peer_getapptable = qed_dcbnl_peer_getapptable, + .cee_peer_getpfc = qed_dcbnl_cee_peer_getpfc, + .cee_peer_getpg = qed_dcbnl_cee_peer_getpg, + .ieee_getpfc = qed_dcbnl_ieee_getpfc, + .ieee_setpfc = qed_dcbnl_ieee_setpfc, + .ieee_getets = qed_dcbnl_ieee_getets, + .ieee_setets = qed_dcbnl_ieee_setets, + .ieee_peer_getpfc = qed_dcbnl_ieee_peer_getpfc, + .ieee_peer_getets = qed_dcbnl_ieee_peer_getets, + .ieee_getapp = qed_dcbnl_ieee_getapp, + .ieee_setapp = qed_dcbnl_ieee_setapp, +}; + #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c index 2ee496e26df6..d121a8bf6b20 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_l2.c +++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c @@ -2166,10 +2166,17 @@ static int qed_fp_cqe_completion(struct qed_dev *dev, extern const struct qed_iov_hv_ops qed_iov_ops_pass; #endif +#ifdef CONFIG_DCB +extern const struct qed_eth_dcbnl_ops qed_dcbnl_ops_pass; +#endif + static const struct qed_eth_ops qed_eth_ops_pass = { .common = &qed_common_ops_pass, #ifdef CONFIG_QED_SRIOV .iov = &qed_iov_ops_pass, +#endif +#ifdef CONFIG_DCB + .dcb = &qed_dcbnl_ops_pass, #endif .fill_dev_info = &qed_fill_eth_dev_info, .register_ops = &qed_register_eth_ops, diff --git a/include/linux/qed/qed_eth_if.h b/include/linux/qed/qed_eth_if.h index f8ff71126d9e..71d523b4bc54 100644 --- a/include/linux/qed/qed_eth_if.h +++ b/include/linux/qed/qed_eth_if.h @@ -128,11 +128,73 @@ struct qed_eth_cb_ops { void (*force_mac) (void *dev, u8 *mac); }; +#ifdef CONFIG_DCB +/* Prototype declaration of qed_eth_dcbnl_ops should match with the declaration + * of dcbnl_rtnl_ops structure. + */ +struct qed_eth_dcbnl_ops { + /* IEEE 802.1Qaz std */ + int (*ieee_getpfc)(struct qed_dev *cdev, struct ieee_pfc *pfc); + int (*ieee_setpfc)(struct qed_dev *cdev, struct ieee_pfc *pfc); + int (*ieee_getets)(struct qed_dev *cdev, struct ieee_ets *ets); + int (*ieee_setets)(struct qed_dev *cdev, struct ieee_ets *ets); + int (*ieee_peer_getets)(struct qed_dev *cdev, struct ieee_ets *ets); + int (*ieee_peer_getpfc)(struct qed_dev *cdev, struct ieee_pfc *pfc); + int (*ieee_getapp)(struct qed_dev *cdev, struct dcb_app *app); + int (*ieee_setapp)(struct qed_dev *cdev, struct dcb_app *app); + + /* CEE std */ + u8 (*getstate)(struct qed_dev *cdev); + u8 (*setstate)(struct qed_dev *cdev, u8 state); + void (*getpgtccfgtx)(struct qed_dev *cdev, int prio, u8 *prio_type, + u8 *pgid, u8 *bw_pct, u8 *up_map); + void (*getpgbwgcfgtx)(struct qed_dev *cdev, int pgid, u8 *bw_pct); + void (*getpgtccfgrx)(struct qed_dev *cdev, int prio, u8 *prio_type, + u8 *pgid, u8 *bw_pct, u8 *up_map); + void (*getpgbwgcfgrx)(struct qed_dev *cdev, int pgid, u8 *bw_pct); + void (*getpfccfg)(struct qed_dev *cdev, int prio, u8 *setting); + void (*setpfccfg)(struct qed_dev *cdev, int prio, u8 setting); + u8 (*getcap)(struct qed_dev *cdev, int capid, u8 *cap); + int (*getnumtcs)(struct qed_dev *cdev, int tcid, u8 *num); + u8 (*getpfcstate)(struct qed_dev *cdev); + int (*getapp)(struct qed_dev *cdev, u8 idtype, u16 id); + u8 (*getfeatcfg)(struct qed_dev *cdev, int featid, u8 *flags); + + /* DCBX configuration */ + u8 (*getdcbx)(struct qed_dev *cdev); + void (*setpgtccfgtx)(struct qed_dev *cdev, int prio, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map); + void (*setpgtccfgrx)(struct qed_dev *cdev, int prio, + u8 pri_type, u8 pgid, u8 bw_pct, u8 up_map); + void (*setpgbwgcfgtx)(struct qed_dev *cdev, int pgid, u8 bw_pct); + void (*setpgbwgcfgrx)(struct qed_dev *cdev, int pgid, u8 bw_pct); + u8 (*setall)(struct qed_dev *cdev); + int (*setnumtcs)(struct qed_dev *cdev, int tcid, u8 num); + void (*setpfcstate)(struct qed_dev *cdev, u8 state); + int (*setapp)(struct qed_dev *cdev, u8 idtype, u16 idval, u8 up); + u8 (*setdcbx)(struct qed_dev *cdev, u8 state); + u8 (*setfeatcfg)(struct qed_dev *cdev, int featid, u8 flags); + + /* Peer apps */ + int (*peer_getappinfo)(struct qed_dev *cdev, + struct dcb_peer_app_info *info, + u16 *app_count); + int (*peer_getapptable)(struct qed_dev *cdev, struct dcb_app *table); + + /* CEE peer */ + int (*cee_peer_getpfc)(struct qed_dev *cdev, struct cee_pfc *pfc); + int (*cee_peer_getpg)(struct qed_dev *cdev, struct cee_pg *pg); +}; +#endif + struct qed_eth_ops { const struct qed_common_ops *common; #ifdef CONFIG_QED_SRIOV const struct qed_iov_hv_ops *iov; #endif +#ifdef CONFIG_DCB + const struct qed_eth_dcbnl_ops *dcb; +#endif int (*fill_dev_info)(struct qed_dev *cdev, struct qed_dev_eth_info *info); -- cgit v1.2.3 From 123b36526592f009bf8eccb7c8833aeda296d9cf Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 8 Jun 2016 07:22:49 -0700 Subject: net: sched: fix missing doc annotations "make htmldocs" complains otherwise: .//net/core/gen_stats.c:168: warning: No description found for parameter 'running' .//include/linux/netdevice.h:1867: warning: No description found for parameter 'qdisc_running_key' Fixes: f9eb8aea2a1e ("net_sched: transform qdisc running bit into a seqcount") Fixes: edb09eb17ed8 ("net: sched: do not acquire qdisc spinlock in qdisc/class stats dump") Signed-off-by: Eric Dumazet Reported-by: kbuild test robot Signed-off-by: David S. Miller --- include/linux/netdevice.h | 3 ++- net/core/gen_stats.c | 1 + 2 files changed, 3 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 59d7e06d88d5..541562333ba5 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1594,7 +1594,8 @@ enum netdev_priv_flags { * @phydev: Physical device may attach itself * for hardware timestamping * - * @qdisc_tx_busylock: XXX: need comments on this one + * @qdisc_tx_busylock: lockdep class annotating Qdisc->busylock spinlock + * @qdisc_running_key: lockdep class annotating Qdisc->running seqcount * * @proto_down: protocol port state information can be sent to the * switch driver and used to set the phys state of the diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c index d9c210caff32..32207e6a942c 100644 --- a/net/core/gen_stats.c +++ b/net/core/gen_stats.c @@ -150,6 +150,7 @@ EXPORT_SYMBOL(__gnet_stats_copy_basic); /** * gnet_stats_copy_basic - copy basic statistics into statistic TLV + * @running: seqcount_t pointer * @d: dumping handle * @cpu: copy statistic per cpu * @b: basic statistics -- cgit v1.2.3 From d3fff6c443fe8f8a5ef2bdcea45e2ff39db948c7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 9 Jun 2016 07:45:12 -0700 Subject: net: add netdev_lockdep_set_classes() helper It is time to add netdev_lockdep_set_classes() helper so that lockdep annotations per device type are easier to manage. This removes a lot of copies and missing annotations. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 24 +----------------------- drivers/net/ppp/ppp_generic.c | 6 +----- drivers/net/team/team.c | 21 +-------------------- include/linux/netdevice.h | 17 +++++++++++++++++ net/bluetooth/6lowpan.c | 15 +-------------- net/ieee802154/6lowpan/core.c | 16 +--------------- net/l2tp/l2tp_eth.c | 6 +----- 7 files changed, 23 insertions(+), 82 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 681af31a60ed..90157e20357e 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -4607,28 +4607,6 @@ static int bond_check_params(struct bond_params *params) return 0; } -static struct lock_class_key bonding_netdev_xmit_lock_key; -static struct lock_class_key bonding_netdev_addr_lock_key; -static struct lock_class_key bonding_tx_busylock_key; -static struct lock_class_key bonding_qdisc_running_key; - -static void bond_set_lockdep_class_one(struct net_device *dev, - struct netdev_queue *txq, - void *_unused) -{ - lockdep_set_class(&txq->_xmit_lock, - &bonding_netdev_xmit_lock_key); -} - -static void bond_set_lockdep_class(struct net_device *dev) -{ - lockdep_set_class(&dev->addr_list_lock, - &bonding_netdev_addr_lock_key); - netdev_for_each_tx_queue(dev, bond_set_lockdep_class_one, NULL); - dev->qdisc_tx_busylock = &bonding_tx_busylock_key; - dev->qdisc_running_key = &bonding_qdisc_running_key; -} - /* Called from registration process */ static int bond_init(struct net_device *bond_dev) { @@ -4641,7 +4619,7 @@ static int bond_init(struct net_device *bond_dev) if (!bond->wq) return -ENOMEM; - bond_set_lockdep_class(bond_dev); + netdev_lockdep_set_classes(bond_dev); list_add_tail(&bond->bond_list, &bn->dev_list); diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c index aeabaa42317f..17953ab15000 100644 --- a/drivers/net/ppp/ppp_generic.c +++ b/drivers/net/ppp/ppp_generic.c @@ -1312,13 +1312,9 @@ ppp_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats64) return stats64; } -static struct lock_class_key ppp_tx_busylock; -static struct lock_class_key ppp_qdisc_running_key; - static int ppp_dev_init(struct net_device *dev) { - dev->qdisc_tx_busylock = &ppp_tx_busylock; - dev->qdisc_running_key = &ppp_qdisc_running_key; + netdev_lockdep_set_classes(dev); return 0; } diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index 00eb38956a2c..0a1bb8387d96 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -1574,25 +1574,6 @@ static const struct team_option team_options[] = { }, }; -static struct lock_class_key team_netdev_xmit_lock_key; -static struct lock_class_key team_netdev_addr_lock_key; -static struct lock_class_key team_tx_busylock_key; -static struct lock_class_key team_qdisc_running_key; - -static void team_set_lockdep_class_one(struct net_device *dev, - struct netdev_queue *txq, - void *unused) -{ - lockdep_set_class(&txq->_xmit_lock, &team_netdev_xmit_lock_key); -} - -static void team_set_lockdep_class(struct net_device *dev) -{ - lockdep_set_class(&dev->addr_list_lock, &team_netdev_addr_lock_key); - netdev_for_each_tx_queue(dev, team_set_lockdep_class_one, NULL); - dev->qdisc_tx_busylock = &team_tx_busylock_key; - dev->qdisc_running_key = &team_qdisc_running_key; -} static int team_init(struct net_device *dev) { @@ -1628,7 +1609,7 @@ static int team_init(struct net_device *dev) goto err_options_register; netif_carrier_off(dev); - team_set_lockdep_class(dev); + netdev_lockdep_set_classes(dev); return 0; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 541562333ba5..4f234b102892 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1946,6 +1946,23 @@ static inline void netdev_for_each_tx_queue(struct net_device *dev, f(dev, &dev->_tx[i], arg); } +#define netdev_lockdep_set_classes(dev) \ +{ \ + static struct lock_class_key qdisc_tx_busylock_key; \ + static struct lock_class_key qdisc_running_key; \ + static struct lock_class_key qdisc_xmit_lock_key; \ + static struct lock_class_key dev_addr_list_lock_key; \ + unsigned int i; \ + \ + (dev)->qdisc_tx_busylock = &qdisc_tx_busylock_key; \ + (dev)->qdisc_running_key = &qdisc_running_key; \ + lockdep_set_class(&(dev)->addr_list_lock, \ + &dev_addr_list_lock_key); \ + for (i = 0; i < (dev)->num_tx_queues; i++) \ + lockdep_set_class(&(dev)->_tx[i]._xmit_lock, \ + &qdisc_xmit_lock_key); \ +} + struct netdev_queue *netdev_pick_tx(struct net_device *dev, struct sk_buff *skb, void *accel_priv); diff --git a/net/bluetooth/6lowpan.c b/net/bluetooth/6lowpan.c index 977a11e418d0..d020299baba4 100644 --- a/net/bluetooth/6lowpan.c +++ b/net/bluetooth/6lowpan.c @@ -627,22 +627,9 @@ static netdev_tx_t bt_xmit(struct sk_buff *skb, struct net_device *netdev) return err < 0 ? NET_XMIT_DROP : err; } -static struct lock_class_key bt_tx_busylock; -static struct lock_class_key bt_netdev_xmit_lock_key; -static struct lock_class_key bt_qdisc_running_key; - -static void bt_set_lockdep_class_one(struct net_device *dev, - struct netdev_queue *txq, - void *_unused) -{ - lockdep_set_class(&txq->_xmit_lock, &bt_netdev_xmit_lock_key); -} - static int bt_dev_init(struct net_device *dev) { - netdev_for_each_tx_queue(dev, bt_set_lockdep_class_one, NULL); - dev->qdisc_tx_busylock = &bt_tx_busylock; - dev->qdisc_running_key = &bt_qdisc_running_key; + netdev_lockdep_set_classes(dev); return 0; } diff --git a/net/ieee802154/6lowpan/core.c b/net/ieee802154/6lowpan/core.c index 14aa5effd29a..4e2b30894224 100644 --- a/net/ieee802154/6lowpan/core.c +++ b/net/ieee802154/6lowpan/core.c @@ -58,23 +58,9 @@ static struct header_ops lowpan_header_ops = { .create = lowpan_header_create, }; -static struct lock_class_key lowpan_tx_busylock; -static struct lock_class_key lowpan_netdev_xmit_lock_key; -static struct lock_class_key lowpan_qdisc_running_key; - -static void lowpan_set_lockdep_class_one(struct net_device *ldev, - struct netdev_queue *txq, - void *_unused) -{ - lockdep_set_class(&txq->_xmit_lock, - &lowpan_netdev_xmit_lock_key); -} - static int lowpan_dev_init(struct net_device *ldev) { - netdev_for_each_tx_queue(ldev, lowpan_set_lockdep_class_one, NULL); - ldev->qdisc_tx_busylock = &lowpan_tx_busylock; - ldev->qdisc_running_key = &lowpan_qdisc_running_key; + netdev_lockdep_set_classes(ldev); return 0; } diff --git a/net/l2tp/l2tp_eth.c b/net/l2tp/l2tp_eth.c index c00d72d182fa..57fc5a46ce06 100644 --- a/net/l2tp/l2tp_eth.c +++ b/net/l2tp/l2tp_eth.c @@ -67,9 +67,6 @@ static inline struct l2tp_eth_net *l2tp_eth_pernet(struct net *net) return net_generic(net, l2tp_eth_net_id); } -static struct lock_class_key l2tp_eth_tx_busylock; -static struct lock_class_key l2tp_qdisc_running_key; - static int l2tp_eth_dev_init(struct net_device *dev) { struct l2tp_eth *priv = netdev_priv(dev); @@ -77,8 +74,7 @@ static int l2tp_eth_dev_init(struct net_device *dev) priv->dev = dev; eth_hw_addr_random(dev); eth_broadcast_addr(dev->broadcast); - dev->qdisc_tx_busylock = &l2tp_eth_tx_busylock; - dev->qdisc_running_key = &l2tp_qdisc_running_key; + netdev_lockdep_set_classes(dev); return 0; } -- cgit v1.2.3 From 967dd82ffc52e9d8ea0defde094f9a39a3f4eeed Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Thu, 9 Jun 2016 18:23:53 -0700 Subject: net: dsa: b53: Add support for Broadcom RoboSwitch This patch adds support for Broadcom's BCM53xx switch family, also known as RoboSwitch. Some of these switches are ubiquituous, found in home routers, Wi-Fi routers, DSL and cable modem gateways and other networking related products. This drivers adds the library driver (b53_common.c) as well as a few bus glue drivers for MDIO, SPI, Switch Register Access Block (SRAB) and memory-mapped I/O into a SoC's address space (Broadcom BCM63xx/33xx). Basic operations are supported to bring the Layer 1/2 up and running, but not much more at this point, subsequent patches add the remaining features. Signed-off-by: Florian Fainelli Signed-off-by: David S. Miller --- Documentation/devicetree/bindings/net/dsa/b53.txt | 88 ++ MAINTAINERS | 8 + drivers/net/dsa/Kconfig | 2 + drivers/net/dsa/Makefile | 2 + drivers/net/dsa/b53/Kconfig | 33 + drivers/net/dsa/b53/Makefile | 6 + drivers/net/dsa/b53/b53_common.c | 1158 +++++++++++++++++++++ drivers/net/dsa/b53/b53_mdio.c | 381 +++++++ drivers/net/dsa/b53/b53_mmap.c | 260 +++++ drivers/net/dsa/b53/b53_priv.h | 322 ++++++ drivers/net/dsa/b53/b53_regs.h | 358 +++++++ drivers/net/dsa/b53/b53_spi.c | 331 ++++++ drivers/net/dsa/b53/b53_srab.c | 415 ++++++++ include/linux/platform_data/b53.h | 33 + 14 files changed, 3397 insertions(+) create mode 100644 Documentation/devicetree/bindings/net/dsa/b53.txt create mode 100644 drivers/net/dsa/b53/Kconfig create mode 100644 drivers/net/dsa/b53/Makefile create mode 100644 drivers/net/dsa/b53/b53_common.c create mode 100644 drivers/net/dsa/b53/b53_mdio.c create mode 100644 drivers/net/dsa/b53/b53_mmap.c create mode 100644 drivers/net/dsa/b53/b53_priv.h create mode 100644 drivers/net/dsa/b53/b53_regs.h create mode 100644 drivers/net/dsa/b53/b53_spi.c create mode 100644 drivers/net/dsa/b53/b53_srab.c create mode 100644 include/linux/platform_data/b53.h (limited to 'include/linux') diff --git a/Documentation/devicetree/bindings/net/dsa/b53.txt b/Documentation/devicetree/bindings/net/dsa/b53.txt new file mode 100644 index 000000000000..ca752db14dff --- /dev/null +++ b/Documentation/devicetree/bindings/net/dsa/b53.txt @@ -0,0 +1,88 @@ +Broadcom BCM53xx Ethernet switches +================================== + +Required properties: + +- compatible: For external switch chips, compatible string must be exactly one + of: "brcm,bcm5325" + "brcm,bcm53115" + "brcm,bcm53125" + "brcm,bcm53128" + "brcm,bcm5365" + "brcm,bcm5395" + "brcm,bcm5397" + "brcm,bcm5398" + + For the BCM5310x SoCs with an integrated switch, must be one of: + "brcm,bcm53010-srab" + "brcm,bcm53011-srab" + "brcm,bcm53012-srab" + "brcm,bcm53018-srab" + "brcm,bcm53019-srab" and the mandatory "brcm,bcm5301x-srab" string + + For the BCM63xx/33xx SoCs with an integrated switch, must be one of: + "brcm,bcm3384-switch" + "brcm,bcm6328-switch" + "brcm,bcm6368-switch" and the mandatory "brcm,bcm63xx-switch" + +See Documentation/devicetree/bindings/dsa/dsa.txt for a list of additional +required and optional properties. + +Examples: + +Ethernet switch connected via MDIO to the host, CPU port wired to eth0: + + eth0: ethernet@10001000 { + compatible = "brcm,unimac"; + reg = <0x10001000 0x1000>; + + fixed-link { + speed = <1000>; + duplex-full; + }; + }; + + mdio0: mdio@10000000 { + compatible = "brcm,unimac-mdio"; + #address-cells = <1>; + #size-cells = <0>; + + switch0: ethernet-switch@30 { + compatible = "brcm,bcm53125"; + #address-cells = <1>; + #size-cells = <0>; + + ports { + port0@0 { + reg = <0>; + label = "lan1"; + }; + + port1@1 { + reg = <1>; + label = "lan2"; + }; + + port5@5 { + reg = <5>; + label = "cable-modem"; + fixed-link { + speed = <1000>; + duplex-full; + }; + phy-mode = "rgmii-txid"; + }; + + port8@8 { + reg = <8>; + label = "cpu"; + fixed-link { + speed = <1000>; + duplex-full; + }; + phy-mode = "rgmii-txid"; + ethernet = <ð0>; + }; + }; + }; + }; diff --git a/MAINTAINERS b/MAINTAINERS index 16e1500d7c52..b29a08884eed 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2454,6 +2454,14 @@ L: netdev@vger.kernel.org S: Supported F: drivers/net/ethernet/broadcom/b44.* +BROADCOM B53 ETHERNET SWITCH DRIVER +M: Florian Fainelli +L: netdev@vger.kernel.org +L: openwrt-devel@lists.openwrt.org (subscribers-only) +S: Supported +F: drivers/net/dsa/b53/* +F: include/linux/platform_data/b53.h + BROADCOM GENET ETHERNET DRIVER M: Florian Fainelli L: netdev@vger.kernel.org diff --git a/drivers/net/dsa/Kconfig b/drivers/net/dsa/Kconfig index 200663c43ce9..be481e15ec9b 100644 --- a/drivers/net/dsa/Kconfig +++ b/drivers/net/dsa/Kconfig @@ -28,4 +28,6 @@ config NET_DSA_BCM_SF2 This enables support for the Broadcom Starfighter 2 Ethernet switch chips. +source "drivers/net/dsa/b53/Kconfig" + endmenu diff --git a/drivers/net/dsa/Makefile b/drivers/net/dsa/Makefile index 76b751dd9efd..97bc70a7f3c7 100644 --- a/drivers/net/dsa/Makefile +++ b/drivers/net/dsa/Makefile @@ -1,3 +1,5 @@ obj-$(CONFIG_NET_DSA_MV88E6060) += mv88e6060.o obj-$(CONFIG_NET_DSA_MV88E6XXX) += mv88e6xxx.o obj-$(CONFIG_NET_DSA_BCM_SF2) += bcm_sf2.o + +obj-y += b53/ diff --git a/drivers/net/dsa/b53/Kconfig b/drivers/net/dsa/b53/Kconfig new file mode 100644 index 000000000000..27f32a50df57 --- /dev/null +++ b/drivers/net/dsa/b53/Kconfig @@ -0,0 +1,33 @@ +menuconfig B53 + tristate "Broadcom BCM53xx managed switch support" + depends on NET_DSA + help + This driver adds support for Broadcom managed switch chips. It supports + BCM5325E, BCM5365, BCM539x, BCM53115 and BCM53125 as well as BCM63XX + integrated switches. + +config B53_SPI_DRIVER + tristate "B53 SPI connected switch driver" + depends on B53 && SPI + help + Select to enable support for registering switches configured through SPI. + +config B53_MDIO_DRIVER + tristate "B53 MDIO connected switch driver" + depends on B53 + help + Select to enable support for registering switches configured through MDIO. + +config B53_MMAP_DRIVER + tristate "B53 MMAP connected switch driver" + depends on B53 && HAS_IOMEM + help + Select to enable support for memory-mapped switches like the BCM63XX + integrated switches. + +config B53_SRAB_DRIVER + tristate "B53 SRAB connected switch driver" + depends on B53 && HAS_IOMEM + help + Select to enable support for memory-mapped Switch Register Access + Bridge Registers (SRAB) like it is found on the BCM53010 diff --git a/drivers/net/dsa/b53/Makefile b/drivers/net/dsa/b53/Makefile new file mode 100644 index 000000000000..7e6f9a8bfd75 --- /dev/null +++ b/drivers/net/dsa/b53/Makefile @@ -0,0 +1,6 @@ +obj-$(CONFIG_B53) += b53_common.o + +obj-$(CONFIG_B53_SPI_DRIVER) += b53_spi.o +obj-$(CONFIG_B53_MDIO_DRIVER) += b53_mdio.o +obj-$(CONFIG_B53_MMAP_DRIVER) += b53_mmap.o +obj-$(CONFIG_B53_SRAB_DRIVER) += b53_srab.o diff --git a/drivers/net/dsa/b53/b53_common.c b/drivers/net/dsa/b53/b53_common.c new file mode 100644 index 000000000000..6f0337d6dfa4 --- /dev/null +++ b/drivers/net/dsa/b53/b53_common.c @@ -0,0 +1,1158 @@ +/* + * B53 switch driver main logic + * + * Copyright (C) 2011-2013 Jonas Gorski + * Copyright (C) 2016 Florian Fainelli + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include +#include +#include +#include +#include +#include +#include +#include + +#include "b53_regs.h" +#include "b53_priv.h" + +struct b53_mib_desc { + u8 size; + u8 offset; + const char *name; +}; + +/* BCM5365 MIB counters */ +static const struct b53_mib_desc b53_mibs_65[] = { + { 8, 0x00, "TxOctets" }, + { 4, 0x08, "TxDropPkts" }, + { 4, 0x10, "TxBroadcastPkts" }, + { 4, 0x14, "TxMulticastPkts" }, + { 4, 0x18, "TxUnicastPkts" }, + { 4, 0x1c, "TxCollisions" }, + { 4, 0x20, "TxSingleCollision" }, + { 4, 0x24, "TxMultipleCollision" }, + { 4, 0x28, "TxDeferredTransmit" }, + { 4, 0x2c, "TxLateCollision" }, + { 4, 0x30, "TxExcessiveCollision" }, + { 4, 0x38, "TxPausePkts" }, + { 8, 0x44, "RxOctets" }, + { 4, 0x4c, "RxUndersizePkts" }, + { 4, 0x50, "RxPausePkts" }, + { 4, 0x54, "Pkts64Octets" }, + { 4, 0x58, "Pkts65to127Octets" }, + { 4, 0x5c, "Pkts128to255Octets" }, + { 4, 0x60, "Pkts256to511Octets" }, + { 4, 0x64, "Pkts512to1023Octets" }, + { 4, 0x68, "Pkts1024to1522Octets" }, + { 4, 0x6c, "RxOversizePkts" }, + { 4, 0x70, "RxJabbers" }, + { 4, 0x74, "RxAlignmentErrors" }, + { 4, 0x78, "RxFCSErrors" }, + { 8, 0x7c, "RxGoodOctets" }, + { 4, 0x84, "RxDropPkts" }, + { 4, 0x88, "RxUnicastPkts" }, + { 4, 0x8c, "RxMulticastPkts" }, + { 4, 0x90, "RxBroadcastPkts" }, + { 4, 0x94, "RxSAChanges" }, + { 4, 0x98, "RxFragments" }, +}; + +#define B53_MIBS_65_SIZE ARRAY_SIZE(b53_mibs_65) + +/* BCM63xx MIB counters */ +static const struct b53_mib_desc b53_mibs_63xx[] = { + { 8, 0x00, "TxOctets" }, + { 4, 0x08, "TxDropPkts" }, + { 4, 0x0c, "TxQoSPkts" }, + { 4, 0x10, "TxBroadcastPkts" }, + { 4, 0x14, "TxMulticastPkts" }, + { 4, 0x18, "TxUnicastPkts" }, + { 4, 0x1c, "TxCollisions" }, + { 4, 0x20, "TxSingleCollision" }, + { 4, 0x24, "TxMultipleCollision" }, + { 4, 0x28, "TxDeferredTransmit" }, + { 4, 0x2c, "TxLateCollision" }, + { 4, 0x30, "TxExcessiveCollision" }, + { 4, 0x38, "TxPausePkts" }, + { 8, 0x3c, "TxQoSOctets" }, + { 8, 0x44, "RxOctets" }, + { 4, 0x4c, "RxUndersizePkts" }, + { 4, 0x50, "RxPausePkts" }, + { 4, 0x54, "Pkts64Octets" }, + { 4, 0x58, "Pkts65to127Octets" }, + { 4, 0x5c, "Pkts128to255Octets" }, + { 4, 0x60, "Pkts256to511Octets" }, + { 4, 0x64, "Pkts512to1023Octets" }, + { 4, 0x68, "Pkts1024to1522Octets" }, + { 4, 0x6c, "RxOversizePkts" }, + { 4, 0x70, "RxJabbers" }, + { 4, 0x74, "RxAlignmentErrors" }, + { 4, 0x78, "RxFCSErrors" }, + { 8, 0x7c, "RxGoodOctets" }, + { 4, 0x84, "RxDropPkts" }, + { 4, 0x88, "RxUnicastPkts" }, + { 4, 0x8c, "RxMulticastPkts" }, + { 4, 0x90, "RxBroadcastPkts" }, + { 4, 0x94, "RxSAChanges" }, + { 4, 0x98, "RxFragments" }, + { 4, 0xa0, "RxSymbolErrors" }, + { 4, 0xa4, "RxQoSPkts" }, + { 8, 0xa8, "RxQoSOctets" }, + { 4, 0xb0, "Pkts1523to2047Octets" }, + { 4, 0xb4, "Pkts2048to4095Octets" }, + { 4, 0xb8, "Pkts4096to8191Octets" }, + { 4, 0xbc, "Pkts8192to9728Octets" }, + { 4, 0xc0, "RxDiscarded" }, +}; + +#define B53_MIBS_63XX_SIZE ARRAY_SIZE(b53_mibs_63xx) + +/* MIB counters */ +static const struct b53_mib_desc b53_mibs[] = { + { 8, 0x00, "TxOctets" }, + { 4, 0x08, "TxDropPkts" }, + { 4, 0x10, "TxBroadcastPkts" }, + { 4, 0x14, "TxMulticastPkts" }, + { 4, 0x18, "TxUnicastPkts" }, + { 4, 0x1c, "TxCollisions" }, + { 4, 0x20, "TxSingleCollision" }, + { 4, 0x24, "TxMultipleCollision" }, + { 4, 0x28, "TxDeferredTransmit" }, + { 4, 0x2c, "TxLateCollision" }, + { 4, 0x30, "TxExcessiveCollision" }, + { 4, 0x38, "TxPausePkts" }, + { 8, 0x50, "RxOctets" }, + { 4, 0x58, "RxUndersizePkts" }, + { 4, 0x5c, "RxPausePkts" }, + { 4, 0x60, "Pkts64Octets" }, + { 4, 0x64, "Pkts65to127Octets" }, + { 4, 0x68, "Pkts128to255Octets" }, + { 4, 0x6c, "Pkts256to511Octets" }, + { 4, 0x70, "Pkts512to1023Octets" }, + { 4, 0x74, "Pkts1024to1522Octets" }, + { 4, 0x78, "RxOversizePkts" }, + { 4, 0x7c, "RxJabbers" }, + { 4, 0x80, "RxAlignmentErrors" }, + { 4, 0x84, "RxFCSErrors" }, + { 8, 0x88, "RxGoodOctets" }, + { 4, 0x90, "RxDropPkts" }, + { 4, 0x94, "RxUnicastPkts" }, + { 4, 0x98, "RxMulticastPkts" }, + { 4, 0x9c, "RxBroadcastPkts" }, + { 4, 0xa0, "RxSAChanges" }, + { 4, 0xa4, "RxFragments" }, + { 4, 0xa8, "RxJumboPkts" }, + { 4, 0xac, "RxSymbolErrors" }, + { 4, 0xc0, "RxDiscarded" }, +}; + +#define B53_MIBS_SIZE ARRAY_SIZE(b53_mibs) + +static int b53_do_vlan_op(struct b53_device *dev, u8 op) +{ + unsigned int i; + + b53_write8(dev, B53_ARLIO_PAGE, dev->vta_regs[0], VTA_START_CMD | op); + + for (i = 0; i < 10; i++) { + u8 vta; + + b53_read8(dev, B53_ARLIO_PAGE, dev->vta_regs[0], &vta); + if (!(vta & VTA_START_CMD)) + return 0; + + usleep_range(100, 200); + } + + return -EIO; +} + +static void b53_set_vlan_entry(struct b53_device *dev, u16 vid, u16 members, + u16 untag) +{ + if (is5325(dev)) { + u32 entry = 0; + + if (members) { + entry = ((untag & VA_UNTAG_MASK_25) << VA_UNTAG_S_25) | + members; + if (dev->core_rev >= 3) + entry |= VA_VALID_25_R4 | vid << VA_VID_HIGH_S; + else + entry |= VA_VALID_25; + } + + b53_write32(dev, B53_VLAN_PAGE, B53_VLAN_WRITE_25, entry); + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_TABLE_ACCESS_25, vid | + VTA_RW_STATE_WR | VTA_RW_OP_EN); + } else if (is5365(dev)) { + u16 entry = 0; + + if (members) + entry = ((untag & VA_UNTAG_MASK_65) << VA_UNTAG_S_65) | + members | VA_VALID_65; + + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_WRITE_65, entry); + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_TABLE_ACCESS_65, vid | + VTA_RW_STATE_WR | VTA_RW_OP_EN); + } else { + b53_write16(dev, B53_ARLIO_PAGE, dev->vta_regs[1], vid); + b53_write32(dev, B53_ARLIO_PAGE, dev->vta_regs[2], + (untag << VTE_UNTAG_S) | members); + + b53_do_vlan_op(dev, VTA_CMD_WRITE); + } +} + +void b53_set_forwarding(struct b53_device *dev, int enable) +{ + u8 mgmt; + + b53_read8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, &mgmt); + + if (enable) + mgmt |= SM_SW_FWD_EN; + else + mgmt &= ~SM_SW_FWD_EN; + + b53_write8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, mgmt); +} + +static void b53_enable_vlan(struct b53_device *dev, int enable) +{ + u8 mgmt, vc0, vc1, vc4 = 0, vc5; + + b53_read8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, &mgmt); + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL0, &vc0); + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL1, &vc1); + + if (is5325(dev) || is5365(dev)) { + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4_25, &vc4); + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5_25, &vc5); + } else if (is63xx(dev)) { + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4_63XX, &vc4); + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5_63XX, &vc5); + } else { + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4, &vc4); + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5, &vc5); + } + + mgmt &= ~SM_SW_FWD_MODE; + + if (enable) { + vc0 |= VC0_VLAN_EN | VC0_VID_CHK_EN | VC0_VID_HASH_VID; + vc1 |= VC1_RX_MCST_UNTAG_EN | VC1_RX_MCST_FWD_EN; + vc4 &= ~VC4_ING_VID_CHECK_MASK; + vc4 |= VC4_ING_VID_VIO_DROP << VC4_ING_VID_CHECK_S; + vc5 |= VC5_DROP_VTABLE_MISS; + + if (is5325(dev)) + vc0 &= ~VC0_RESERVED_1; + + if (is5325(dev) || is5365(dev)) + vc1 |= VC1_RX_MCST_TAG_EN; + + if (!is5325(dev) && !is5365(dev)) { + if (dev->allow_vid_4095) + vc5 |= VC5_VID_FFF_EN; + else + vc5 &= ~VC5_VID_FFF_EN; + } + } else { + vc0 &= ~(VC0_VLAN_EN | VC0_VID_CHK_EN | VC0_VID_HASH_VID); + vc1 &= ~(VC1_RX_MCST_UNTAG_EN | VC1_RX_MCST_FWD_EN); + vc4 &= ~VC4_ING_VID_CHECK_MASK; + vc5 &= ~VC5_DROP_VTABLE_MISS; + + if (is5325(dev) || is5365(dev)) + vc4 |= VC4_ING_VID_VIO_FWD << VC4_ING_VID_CHECK_S; + else + vc4 |= VC4_ING_VID_VIO_TO_IMP << VC4_ING_VID_CHECK_S; + + if (is5325(dev) || is5365(dev)) + vc1 &= ~VC1_RX_MCST_TAG_EN; + + if (!is5325(dev) && !is5365(dev)) + vc5 &= ~VC5_VID_FFF_EN; + } + + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL0, vc0); + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL1, vc1); + + if (is5325(dev) || is5365(dev)) { + /* enable the high 8 bit vid check on 5325 */ + if (is5325(dev) && enable) + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL3, + VC3_HIGH_8BIT_EN); + else + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL3, 0); + + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4_25, vc4); + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5_25, vc5); + } else if (is63xx(dev)) { + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_CTRL3_63XX, 0); + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4_63XX, vc4); + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5_63XX, vc5); + } else { + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_CTRL3, 0); + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4, vc4); + b53_write8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL5, vc5); + } + + b53_write8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, mgmt); +} + +static int b53_set_jumbo(struct b53_device *dev, bool enable, bool allow_10_100) +{ + u32 port_mask = 0; + u16 max_size = JMS_MIN_SIZE; + + if (is5325(dev) || is5365(dev)) + return -EINVAL; + + if (enable) { + port_mask = dev->enabled_ports; + max_size = JMS_MAX_SIZE; + if (allow_10_100) + port_mask |= JPM_10_100_JUMBO_EN; + } + + b53_write32(dev, B53_JUMBO_PAGE, dev->jumbo_pm_reg, port_mask); + return b53_write16(dev, B53_JUMBO_PAGE, dev->jumbo_size_reg, max_size); +} + +static int b53_flush_arl(struct b53_device *dev) +{ + unsigned int i; + + b53_write8(dev, B53_CTRL_PAGE, B53_FAST_AGE_CTRL, + FAST_AGE_DONE | FAST_AGE_DYNAMIC | FAST_AGE_STATIC); + + for (i = 0; i < 10; i++) { + u8 fast_age_ctrl; + + b53_read8(dev, B53_CTRL_PAGE, B53_FAST_AGE_CTRL, + &fast_age_ctrl); + + if (!(fast_age_ctrl & FAST_AGE_DONE)) + goto out; + + msleep(1); + } + + return -ETIMEDOUT; +out: + /* Only age dynamic entries (default behavior) */ + b53_write8(dev, B53_CTRL_PAGE, B53_FAST_AGE_CTRL, FAST_AGE_DYNAMIC); + return 0; +} + +static int b53_enable_port(struct dsa_switch *ds, int port, + struct phy_device *phy) +{ + struct b53_device *dev = ds_to_priv(ds); + + /* Clear the Rx and Tx disable bits and set to no spanning tree */ + b53_write8(dev, B53_CTRL_PAGE, B53_PORT_CTRL(port), 0); + + return 0; +} + +static void b53_disable_port(struct dsa_switch *ds, int port, + struct phy_device *phy) +{ + struct b53_device *dev = ds_to_priv(ds); + u8 reg; + + /* Disable Tx/Rx for the port */ + b53_read8(dev, B53_CTRL_PAGE, B53_PORT_CTRL(port), ®); + reg |= PORT_CTRL_RX_DISABLE | PORT_CTRL_TX_DISABLE; + b53_write8(dev, B53_CTRL_PAGE, B53_PORT_CTRL(port), reg); +} + +static void b53_enable_cpu_port(struct b53_device *dev) +{ + unsigned int cpu_port = dev->cpu_port; + u8 port_ctrl; + + /* BCM5325 CPU port is at 8 */ + if ((is5325(dev) || is5365(dev)) && cpu_port == B53_CPU_PORT_25) + cpu_port = B53_CPU_PORT; + + port_ctrl = PORT_CTRL_RX_BCST_EN | + PORT_CTRL_RX_MCST_EN | + PORT_CTRL_RX_UCST_EN; + b53_write8(dev, B53_CTRL_PAGE, B53_PORT_CTRL(cpu_port), port_ctrl); +} + +static void b53_enable_mib(struct b53_device *dev) +{ + u8 gc; + + b53_read8(dev, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, &gc); + gc &= ~(GC_RESET_MIB | GC_MIB_AC_EN); + b53_write8(dev, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, gc); +} + +static int b53_configure_vlan(struct b53_device *dev) +{ + int i; + + /* clear all vlan entries */ + if (is5325(dev) || is5365(dev)) { + for (i = 1; i < dev->num_vlans; i++) + b53_set_vlan_entry(dev, i, 0, 0); + } else { + b53_do_vlan_op(dev, VTA_CMD_CLEAR); + } + + b53_enable_vlan(dev, false); + + b53_for_each_port(dev, i) + b53_write16(dev, B53_VLAN_PAGE, + B53_VLAN_PORT_DEF_TAG(i), 1); + + if (!is5325(dev) && !is5365(dev)) + b53_set_jumbo(dev, dev->enable_jumbo, false); + + return 0; +} + +static void b53_switch_reset_gpio(struct b53_device *dev) +{ + int gpio = dev->reset_gpio; + + if (gpio < 0) + return; + + /* Reset sequence: RESET low(50ms)->high(20ms) + */ + gpio_set_value(gpio, 0); + mdelay(50); + + gpio_set_value(gpio, 1); + mdelay(20); + + dev->current_page = 0xff; +} + +static int b53_switch_reset(struct b53_device *dev) +{ + u8 mgmt; + + b53_switch_reset_gpio(dev); + + if (is539x(dev)) { + b53_write8(dev, B53_CTRL_PAGE, B53_SOFTRESET, 0x83); + b53_write8(dev, B53_CTRL_PAGE, B53_SOFTRESET, 0x00); + } + + b53_read8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, &mgmt); + + if (!(mgmt & SM_SW_FWD_EN)) { + mgmt &= ~SM_SW_FWD_MODE; + mgmt |= SM_SW_FWD_EN; + + b53_write8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, mgmt); + b53_read8(dev, B53_CTRL_PAGE, B53_SWITCH_MODE, &mgmt); + + if (!(mgmt & SM_SW_FWD_EN)) { + dev_err(dev->dev, "Failed to enable switch!\n"); + return -EINVAL; + } + } + + b53_enable_mib(dev); + + return b53_flush_arl(dev); +} + +static int b53_phy_read16(struct dsa_switch *ds, int addr, int reg) +{ + struct b53_device *priv = ds_to_priv(ds); + u16 value = 0; + int ret; + + if (priv->ops->phy_read16) + ret = priv->ops->phy_read16(priv, addr, reg, &value); + else + ret = b53_read16(priv, B53_PORT_MII_PAGE(addr), + reg * 2, &value); + + return ret ? ret : value; +} + +static int b53_phy_write16(struct dsa_switch *ds, int addr, int reg, u16 val) +{ + struct b53_device *priv = ds_to_priv(ds); + + if (priv->ops->phy_write16) + return priv->ops->phy_write16(priv, addr, reg, val); + + return b53_write16(priv, B53_PORT_MII_PAGE(addr), reg * 2, val); +} + +static int b53_reset_switch(struct b53_device *priv) +{ + /* reset vlans */ + priv->enable_jumbo = false; + + memset(priv->ports, 0, sizeof(*priv->ports) * priv->num_ports); + + return b53_switch_reset(priv); +} + +static int b53_apply_config(struct b53_device *priv) +{ + /* disable switching */ + b53_set_forwarding(priv, 0); + + b53_configure_vlan(priv); + + /* enable switching */ + b53_set_forwarding(priv, 1); + + return 0; +} + +static void b53_reset_mib(struct b53_device *priv) +{ + u8 gc; + + b53_read8(priv, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, &gc); + + b53_write8(priv, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, gc | GC_RESET_MIB); + msleep(1); + b53_write8(priv, B53_MGMT_PAGE, B53_GLOBAL_CONFIG, gc & ~GC_RESET_MIB); + msleep(1); +} + +static const struct b53_mib_desc *b53_get_mib(struct b53_device *dev) +{ + if (is5365(dev)) + return b53_mibs_65; + else if (is63xx(dev)) + return b53_mibs_63xx; + else + return b53_mibs; +} + +static unsigned int b53_get_mib_size(struct b53_device *dev) +{ + if (is5365(dev)) + return B53_MIBS_65_SIZE; + else if (is63xx(dev)) + return B53_MIBS_63XX_SIZE; + else + return B53_MIBS_SIZE; +} + +static void b53_get_strings(struct dsa_switch *ds, int port, uint8_t *data) +{ + struct b53_device *dev = ds_to_priv(ds); + const struct b53_mib_desc *mibs = b53_get_mib(dev); + unsigned int mib_size = b53_get_mib_size(dev); + unsigned int i; + + for (i = 0; i < mib_size; i++) + memcpy(data + i * ETH_GSTRING_LEN, + mibs[i].name, ETH_GSTRING_LEN); +} + +static void b53_get_ethtool_stats(struct dsa_switch *ds, int port, + uint64_t *data) +{ + struct b53_device *dev = ds_to_priv(ds); + const struct b53_mib_desc *mibs = b53_get_mib(dev); + unsigned int mib_size = b53_get_mib_size(dev); + const struct b53_mib_desc *s; + unsigned int i; + u64 val = 0; + + if (is5365(dev) && port == 5) + port = 8; + + mutex_lock(&dev->stats_mutex); + + for (i = 0; i < mib_size; i++) { + s = &mibs[i]; + + if (mibs->size == 8) { + b53_read64(dev, B53_MIB_PAGE(port), s->offset, &val); + } else { + u32 val32; + + b53_read32(dev, B53_MIB_PAGE(port), s->offset, + &val32); + val = val32; + } + data[i] = (u64)val; + } + + mutex_unlock(&dev->stats_mutex); +} + +static int b53_get_sset_count(struct dsa_switch *ds) +{ + struct b53_device *dev = ds_to_priv(ds); + + return b53_get_mib_size(dev); +} + +static int b53_set_addr(struct dsa_switch *ds, u8 *addr) +{ + return 0; +} + +static int b53_setup(struct dsa_switch *ds) +{ + struct b53_device *dev = ds_to_priv(ds); + unsigned int port; + int ret; + + ret = b53_reset_switch(dev); + if (ret) { + dev_err(ds->dev, "failed to reset switch\n"); + return ret; + } + + b53_reset_mib(dev); + + ret = b53_apply_config(dev); + if (ret) + dev_err(ds->dev, "failed to apply configuration\n"); + + for (port = 0; port < dev->num_ports; port++) { + if (BIT(port) & ds->enabled_port_mask) + b53_enable_port(ds, port, NULL); + else if (dsa_is_cpu_port(ds, port)) + b53_enable_cpu_port(dev); + else + b53_disable_port(ds, port, NULL); + } + + return ret; +} + +static void b53_adjust_link(struct dsa_switch *ds, int port, + struct phy_device *phydev) +{ + struct b53_device *dev = ds_to_priv(ds); + u8 rgmii_ctrl = 0, reg = 0, off; + + if (!phy_is_pseudo_fixed_link(phydev)) + return; + + /* Override the port settings */ + if (port == dev->cpu_port) { + off = B53_PORT_OVERRIDE_CTRL; + reg = PORT_OVERRIDE_EN; + } else { + off = B53_GMII_PORT_OVERRIDE_CTRL(port); + reg = GMII_PO_EN; + } + + /* Set the link UP */ + if (phydev->link) + reg |= PORT_OVERRIDE_LINK; + + if (phydev->duplex == DUPLEX_FULL) + reg |= PORT_OVERRIDE_FULL_DUPLEX; + + switch (phydev->speed) { + case 2000: + reg |= PORT_OVERRIDE_SPEED_2000M; + /* fallthrough */ + case SPEED_1000: + reg |= PORT_OVERRIDE_SPEED_1000M; + break; + case SPEED_100: + reg |= PORT_OVERRIDE_SPEED_100M; + break; + case SPEED_10: + reg |= PORT_OVERRIDE_SPEED_10M; + break; + default: + dev_err(ds->dev, "unknown speed: %d\n", phydev->speed); + return; + } + + /* Enable flow control on BCM5301x's CPU port */ + if (is5301x(dev) && port == dev->cpu_port) + reg |= PORT_OVERRIDE_RX_FLOW | PORT_OVERRIDE_TX_FLOW; + + if (phydev->pause) { + if (phydev->asym_pause) + reg |= PORT_OVERRIDE_TX_FLOW; + reg |= PORT_OVERRIDE_RX_FLOW; + } + + b53_write8(dev, B53_CTRL_PAGE, off, reg); + + if (is531x5(dev) && phy_interface_is_rgmii(phydev)) { + if (port == 8) + off = B53_RGMII_CTRL_IMP; + else + off = B53_RGMII_CTRL_P(port); + + /* Configure the port RGMII clock delay by DLL disabled and + * tx_clk aligned timing (restoring to reset defaults) + */ + b53_read8(dev, B53_CTRL_PAGE, off, &rgmii_ctrl); + rgmii_ctrl &= ~(RGMII_CTRL_DLL_RXC | RGMII_CTRL_DLL_TXC | + RGMII_CTRL_TIMING_SEL); + + /* PHY_INTERFACE_MODE_RGMII_TXID means TX internal delay, make + * sure that we enable the port TX clock internal delay to + * account for this internal delay that is inserted, otherwise + * the switch won't be able to receive correctly. + * + * PHY_INTERFACE_MODE_RGMII means that we are not introducing + * any delay neither on transmission nor reception, so the + * BCM53125 must also be configured accordingly to account for + * the lack of delay and introduce + * + * The BCM53125 switch has its RX clock and TX clock control + * swapped, hence the reason why we modify the TX clock path in + * the "RGMII" case + */ + if (phydev->interface == PHY_INTERFACE_MODE_RGMII_TXID) + rgmii_ctrl |= RGMII_CTRL_DLL_TXC; + if (phydev->interface == PHY_INTERFACE_MODE_RGMII) + rgmii_ctrl |= RGMII_CTRL_DLL_TXC | RGMII_CTRL_DLL_RXC; + rgmii_ctrl |= RGMII_CTRL_TIMING_SEL; + b53_write8(dev, B53_CTRL_PAGE, off, rgmii_ctrl); + + dev_info(ds->dev, "Configured port %d for %s\n", port, + phy_modes(phydev->interface)); + } + + /* configure MII port if necessary */ + if (is5325(dev)) { + b53_read8(dev, B53_CTRL_PAGE, B53_PORT_OVERRIDE_CTRL, + ®); + + /* reverse mii needs to be enabled */ + if (!(reg & PORT_OVERRIDE_RV_MII_25)) { + b53_write8(dev, B53_CTRL_PAGE, B53_PORT_OVERRIDE_CTRL, + reg | PORT_OVERRIDE_RV_MII_25); + b53_read8(dev, B53_CTRL_PAGE, B53_PORT_OVERRIDE_CTRL, + ®); + + if (!(reg & PORT_OVERRIDE_RV_MII_25)) { + dev_err(ds->dev, + "Failed to enable reverse MII mode\n"); + return; + } + } + } else if (is5301x(dev)) { + if (port != dev->cpu_port) { + u8 po_reg = B53_GMII_PORT_OVERRIDE_CTRL(dev->cpu_port); + u8 gmii_po; + + b53_read8(dev, B53_CTRL_PAGE, po_reg, &gmii_po); + gmii_po |= GMII_PO_LINK | + GMII_PO_RX_FLOW | + GMII_PO_TX_FLOW | + GMII_PO_EN | + GMII_PO_SPEED_2000M; + b53_write8(dev, B53_CTRL_PAGE, po_reg, gmii_po); + } + } +} + +static struct dsa_switch_driver b53_switch_ops = { + .tag_protocol = DSA_TAG_PROTO_NONE, + .setup = b53_setup, + .set_addr = b53_set_addr, + .get_strings = b53_get_strings, + .get_ethtool_stats = b53_get_ethtool_stats, + .get_sset_count = b53_get_sset_count, + .phy_read = b53_phy_read16, + .phy_write = b53_phy_write16, + .adjust_link = b53_adjust_link, + .port_enable = b53_enable_port, + .port_disable = b53_disable_port, +}; + +struct b53_chip_data { + u32 chip_id; + const char *dev_name; + u16 vlans; + u16 enabled_ports; + u8 cpu_port; + u8 vta_regs[3]; + u8 duplex_reg; + u8 jumbo_pm_reg; + u8 jumbo_size_reg; +}; + +#define B53_VTA_REGS \ + { B53_VT_ACCESS, B53_VT_INDEX, B53_VT_ENTRY } +#define B53_VTA_REGS_9798 \ + { B53_VT_ACCESS_9798, B53_VT_INDEX_9798, B53_VT_ENTRY_9798 } +#define B53_VTA_REGS_63XX \ + { B53_VT_ACCESS_63XX, B53_VT_INDEX_63XX, B53_VT_ENTRY_63XX } + +static const struct b53_chip_data b53_switch_chips[] = { + { + .chip_id = BCM5325_DEVICE_ID, + .dev_name = "BCM5325", + .vlans = 16, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT_25, + .duplex_reg = B53_DUPLEX_STAT_FE, + }, + { + .chip_id = BCM5365_DEVICE_ID, + .dev_name = "BCM5365", + .vlans = 256, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT_25, + .duplex_reg = B53_DUPLEX_STAT_FE, + }, + { + .chip_id = BCM5395_DEVICE_ID, + .dev_name = "BCM5395", + .vlans = 4096, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT, + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM5397_DEVICE_ID, + .dev_name = "BCM5397", + .vlans = 4096, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT, + .vta_regs = B53_VTA_REGS_9798, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM5398_DEVICE_ID, + .dev_name = "BCM5398", + .vlans = 4096, + .enabled_ports = 0x7f, + .cpu_port = B53_CPU_PORT, + .vta_regs = B53_VTA_REGS_9798, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53115_DEVICE_ID, + .dev_name = "BCM53115", + .vlans = 4096, + .enabled_ports = 0x1f, + .vta_regs = B53_VTA_REGS, + .cpu_port = B53_CPU_PORT, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53125_DEVICE_ID, + .dev_name = "BCM53125", + .vlans = 4096, + .enabled_ports = 0xff, + .cpu_port = B53_CPU_PORT, + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53128_DEVICE_ID, + .dev_name = "BCM53128", + .vlans = 4096, + .enabled_ports = 0x1ff, + .cpu_port = B53_CPU_PORT, + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM63XX_DEVICE_ID, + .dev_name = "BCM63xx", + .vlans = 4096, + .enabled_ports = 0, /* pdata must provide them */ + .cpu_port = B53_CPU_PORT, + .vta_regs = B53_VTA_REGS_63XX, + .duplex_reg = B53_DUPLEX_STAT_63XX, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK_63XX, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE_63XX, + }, + { + .chip_id = BCM53010_DEVICE_ID, + .dev_name = "BCM53010", + .vlans = 4096, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53011_DEVICE_ID, + .dev_name = "BCM53011", + .vlans = 4096, + .enabled_ports = 0x1bf, + .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53012_DEVICE_ID, + .dev_name = "BCM53012", + .vlans = 4096, + .enabled_ports = 0x1bf, + .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53018_DEVICE_ID, + .dev_name = "BCM53018", + .vlans = 4096, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, + { + .chip_id = BCM53019_DEVICE_ID, + .dev_name = "BCM53019", + .vlans = 4096, + .enabled_ports = 0x1f, + .cpu_port = B53_CPU_PORT_25, /* TODO: auto detect */ + .vta_regs = B53_VTA_REGS, + .duplex_reg = B53_DUPLEX_STAT_GE, + .jumbo_pm_reg = B53_JUMBO_PORT_MASK, + .jumbo_size_reg = B53_JUMBO_MAX_SIZE, + }, +}; + +static int b53_switch_init(struct b53_device *dev) +{ + struct dsa_switch *ds = dev->ds; + unsigned int i; + int ret; + + for (i = 0; i < ARRAY_SIZE(b53_switch_chips); i++) { + const struct b53_chip_data *chip = &b53_switch_chips[i]; + + if (chip->chip_id == dev->chip_id) { + if (!dev->enabled_ports) + dev->enabled_ports = chip->enabled_ports; + dev->name = chip->dev_name; + dev->duplex_reg = chip->duplex_reg; + dev->vta_regs[0] = chip->vta_regs[0]; + dev->vta_regs[1] = chip->vta_regs[1]; + dev->vta_regs[2] = chip->vta_regs[2]; + dev->jumbo_pm_reg = chip->jumbo_pm_reg; + ds->drv = &b53_switch_ops; + dev->cpu_port = chip->cpu_port; + dev->num_vlans = chip->vlans; + break; + } + } + + /* check which BCM5325x version we have */ + if (is5325(dev)) { + u8 vc4; + + b53_read8(dev, B53_VLAN_PAGE, B53_VLAN_CTRL4_25, &vc4); + + /* check reserved bits */ + switch (vc4 & 3) { + case 1: + /* BCM5325E */ + break; + case 3: + /* BCM5325F - do not use port 4 */ + dev->enabled_ports &= ~BIT(4); + break; + default: +/* On the BCM47XX SoCs this is the supported internal switch.*/ +#ifndef CONFIG_BCM47XX + /* BCM5325M */ + return -EINVAL; +#else + break; +#endif + } + } else if (dev->chip_id == BCM53115_DEVICE_ID) { + u64 strap_value; + + b53_read48(dev, B53_STAT_PAGE, B53_STRAP_VALUE, &strap_value); + /* use second IMP port if GMII is enabled */ + if (strap_value & SV_GMII_CTRL_115) + dev->cpu_port = 5; + } + + /* cpu port is always last */ + dev->num_ports = dev->cpu_port + 1; + dev->enabled_ports |= BIT(dev->cpu_port); + + dev->ports = devm_kzalloc(dev->dev, + sizeof(struct b53_port) * dev->num_ports, + GFP_KERNEL); + if (!dev->ports) + return -ENOMEM; + + dev->reset_gpio = b53_switch_get_reset_gpio(dev); + if (dev->reset_gpio >= 0) { + ret = devm_gpio_request_one(dev->dev, dev->reset_gpio, + GPIOF_OUT_INIT_HIGH, "robo_reset"); + if (ret) + return ret; + } + + return 0; +} + +struct b53_device *b53_switch_alloc(struct device *base, struct b53_io_ops *ops, + void *priv) +{ + struct dsa_switch *ds; + struct b53_device *dev; + + ds = devm_kzalloc(base, sizeof(*ds) + sizeof(*dev), GFP_KERNEL); + if (!ds) + return NULL; + + dev = (struct b53_device *)(ds + 1); + + ds->priv = dev; + ds->dev = base; + dev->dev = base; + + dev->ds = ds; + dev->priv = priv; + dev->ops = ops; + mutex_init(&dev->reg_mutex); + mutex_init(&dev->stats_mutex); + + return dev; +} +EXPORT_SYMBOL(b53_switch_alloc); + +int b53_switch_detect(struct b53_device *dev) +{ + u32 id32; + u16 tmp; + u8 id8; + int ret; + + ret = b53_read8(dev, B53_MGMT_PAGE, B53_DEVICE_ID, &id8); + if (ret) + return ret; + + switch (id8) { + case 0: + /* BCM5325 and BCM5365 do not have this register so reads + * return 0. But the read operation did succeed, so assume this + * is one of them. + * + * Next check if we can write to the 5325's VTA register; for + * 5365 it is read only. + */ + b53_write16(dev, B53_VLAN_PAGE, B53_VLAN_TABLE_ACCESS_25, 0xf); + b53_read16(dev, B53_VLAN_PAGE, B53_VLAN_TABLE_ACCESS_25, &tmp); + + if (tmp == 0xf) + dev->chip_id = BCM5325_DEVICE_ID; + else + dev->chip_id = BCM5365_DEVICE_ID; + break; + case BCM5395_DEVICE_ID: + case BCM5397_DEVICE_ID: + case BCM5398_DEVICE_ID: + dev->chip_id = id8; + break; + default: + ret = b53_read32(dev, B53_MGMT_PAGE, B53_DEVICE_ID, &id32); + if (ret) + return ret; + + switch (id32) { + case BCM53115_DEVICE_ID: + case BCM53125_DEVICE_ID: + case BCM53128_DEVICE_ID: + case BCM53010_DEVICE_ID: + case BCM53011_DEVICE_ID: + case BCM53012_DEVICE_ID: + case BCM53018_DEVICE_ID: + case BCM53019_DEVICE_ID: + dev->chip_id = id32; + break; + default: + pr_err("unsupported switch detected (BCM53%02x/BCM%x)\n", + id8, id32); + return -ENODEV; + } + } + + if (dev->chip_id == BCM5325_DEVICE_ID) + return b53_read8(dev, B53_STAT_PAGE, B53_REV_ID_25, + &dev->core_rev); + else + return b53_read8(dev, B53_MGMT_PAGE, B53_REV_ID, + &dev->core_rev); +} +EXPORT_SYMBOL(b53_switch_detect); + +int b53_switch_register(struct b53_device *dev) +{ + int ret; + + if (dev->pdata) { + dev->chip_id = dev->pdata->chip_id; + dev->enabled_ports = dev->pdata->enabled_ports; + } + + if (!dev->chip_id && b53_switch_detect(dev)) + return -EINVAL; + + ret = b53_switch_init(dev); + if (ret) + return ret; + + pr_info("found switch: %s, rev %i\n", dev->name, dev->core_rev); + + return dsa_register_switch(dev->ds, dev->ds->dev->of_node); +} +EXPORT_SYMBOL(b53_switch_register); + +MODULE_AUTHOR("Jonas Gorski "); +MODULE_DESCRIPTION("B53 switch library"); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/net/dsa/b53/b53_mdio.c b/drivers/net/dsa/b53/b53_mdio.c new file mode 100644 index 000000000000..c6cf7cf57cc3 --- /dev/null +++ b/drivers/net/dsa/b53/b53_mdio.c @@ -0,0 +1,381 @@ +/* + * B53 register access through MII registers + * + * Copyright (C) 2011-2013 Jonas Gorski + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include +#include + +#include "b53_priv.h" + +/* MII registers */ +#define REG_MII_PAGE 0x10 /* MII Page register */ +#define REG_MII_ADDR 0x11 /* MII Address register */ +#define REG_MII_DATA0 0x18 /* MII Data register 0 */ +#define REG_MII_DATA1 0x19 /* MII Data register 1 */ +#define REG_MII_DATA2 0x1a /* MII Data register 2 */ +#define REG_MII_DATA3 0x1b /* MII Data register 3 */ + +#define REG_MII_PAGE_ENABLE BIT(0) +#define REG_MII_ADDR_WRITE BIT(0) +#define REG_MII_ADDR_READ BIT(1) + +static int b53_mdio_op(struct b53_device *dev, u8 page, u8 reg, u16 op) +{ + int i; + u16 v; + int ret; + struct mii_bus *bus = dev->priv; + + if (dev->current_page != page) { + /* set page number */ + v = (page << 8) | REG_MII_PAGE_ENABLE; + ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_PAGE, v); + if (ret) + return ret; + dev->current_page = page; + } + + /* set register address */ + v = (reg << 8) | op; + ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, REG_MII_ADDR, v); + if (ret) + return ret; + + /* check if operation completed */ + for (i = 0; i < 5; ++i) { + v = mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_ADDR); + if (!(v & (REG_MII_ADDR_WRITE | REG_MII_ADDR_READ))) + break; + usleep_range(10, 100); + } + + if (WARN_ON(i == 5)) + return -EIO; + + return 0; +} + +static int b53_mdio_read8(struct b53_device *dev, u8 page, u8 reg, u8 *val) +{ + struct mii_bus *bus = dev->priv; + int ret; + + ret = b53_mdio_op(dev, page, reg, REG_MII_ADDR_READ); + if (ret) + return ret; + + *val = mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0) & 0xff; + + return 0; +} + +static int b53_mdio_read16(struct b53_device *dev, u8 page, u8 reg, u16 *val) +{ + struct mii_bus *bus = dev->priv; + int ret; + + ret = b53_mdio_op(dev, page, reg, REG_MII_ADDR_READ); + if (ret) + return ret; + + *val = mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, REG_MII_DATA0); + + return 0; +} + +static int b53_mdio_read32(struct b53_device *dev, u8 page, u8 reg, u32 *val) +{ + struct mii_bus *bus = dev->priv; + int ret; + + ret = b53_mdio_op(dev, page, reg, REG_MII_ADDR_READ); + if (ret) + return ret; + + *val = mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, REG_MII_DATA0); + *val |= mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA1) << 16; + + return 0; +} + +static int b53_mdio_read48(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + struct mii_bus *bus = dev->priv; + u64 temp = 0; + int i; + int ret; + + ret = b53_mdio_op(dev, page, reg, REG_MII_ADDR_READ); + if (ret) + return ret; + + for (i = 2; i >= 0; i--) { + temp <<= 16; + temp |= mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0 + i); + } + + *val = temp; + + return 0; +} + +static int b53_mdio_read64(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + struct mii_bus *bus = dev->priv; + u64 temp = 0; + int i; + int ret; + + ret = b53_mdio_op(dev, page, reg, REG_MII_ADDR_READ); + if (ret) + return ret; + + for (i = 3; i >= 0; i--) { + temp <<= 16; + temp |= mdiobus_read_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0 + i); + } + + *val = temp; + + return 0; +} + +static int b53_mdio_write8(struct b53_device *dev, u8 page, u8 reg, u8 value) +{ + struct mii_bus *bus = dev->priv; + int ret; + + ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0, value); + if (ret) + return ret; + + return b53_mdio_op(dev, page, reg, REG_MII_ADDR_WRITE); +} + +static int b53_mdio_write16(struct b53_device *dev, u8 page, u8 reg, + u16 value) +{ + struct mii_bus *bus = dev->priv; + int ret; + + ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0, value); + if (ret) + return ret; + + return b53_mdio_op(dev, page, reg, REG_MII_ADDR_WRITE); +} + +static int b53_mdio_write32(struct b53_device *dev, u8 page, u8 reg, + u32 value) +{ + struct mii_bus *bus = dev->priv; + unsigned int i; + u32 temp = value; + + for (i = 0; i < 2; i++) { + int ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0 + i, + temp & 0xffff); + if (ret) + return ret; + temp >>= 16; + } + + return b53_mdio_op(dev, page, reg, REG_MII_ADDR_WRITE); +} + +static int b53_mdio_write48(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + struct mii_bus *bus = dev->priv; + unsigned int i; + u64 temp = value; + + for (i = 0; i < 3; i++) { + int ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0 + i, + temp & 0xffff); + if (ret) + return ret; + temp >>= 16; + } + + return b53_mdio_op(dev, page, reg, REG_MII_ADDR_WRITE); +} + +static int b53_mdio_write64(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + struct mii_bus *bus = dev->priv; + unsigned int i; + u64 temp = value; + + for (i = 0; i < 4; i++) { + int ret = mdiobus_write_nested(bus, BRCM_PSEUDO_PHY_ADDR, + REG_MII_DATA0 + i, + temp & 0xffff); + if (ret) + return ret; + temp >>= 16; + } + + return b53_mdio_op(dev, page, reg, REG_MII_ADDR_WRITE); +} + +static int b53_mdio_phy_read16(struct b53_device *dev, int addr, int reg, + u16 *value) +{ + struct mii_bus *bus = dev->priv; + + *value = mdiobus_read_nested(bus, addr, reg); + + return 0; +} + +static int b53_mdio_phy_write16(struct b53_device *dev, int addr, int reg, + u16 value) +{ + struct mii_bus *bus = dev->bus; + + return mdiobus_write_nested(bus, addr, reg, value); +} + +static struct b53_io_ops b53_mdio_ops = { + .read8 = b53_mdio_read8, + .read16 = b53_mdio_read16, + .read32 = b53_mdio_read32, + .read48 = b53_mdio_read48, + .read64 = b53_mdio_read64, + .write8 = b53_mdio_write8, + .write16 = b53_mdio_write16, + .write32 = b53_mdio_write32, + .write48 = b53_mdio_write48, + .write64 = b53_mdio_write64, + .phy_read16 = b53_mdio_phy_read16, + .phy_write16 = b53_mdio_phy_write16, +}; + +#define B53_BRCM_OUI_1 0x0143bc00 +#define B53_BRCM_OUI_2 0x03625c00 +#define B53_BRCM_OUI_3 0x00406000 + +static int b53_mdio_probe(struct mdio_device *mdiodev) +{ + struct b53_device *dev; + u32 phy_id; + int ret; + + /* allow the generic PHY driver to take over the non-management MDIO + * addresses + */ + if (mdiodev->addr != BRCM_PSEUDO_PHY_ADDR && mdiodev->addr != 0) { + dev_err(&mdiodev->dev, "leaving address %d to PHY\n", + mdiodev->addr); + return -ENODEV; + } + + /* read the first port's id */ + phy_id = mdiobus_read(mdiodev->bus, 0, 2) << 16; + phy_id |= mdiobus_read(mdiodev->bus, 0, 3); + + /* BCM5325, BCM539x (OUI_1) + * BCM53125, BCM53128 (OUI_2) + * BCM5365 (OUI_3) + */ + if ((phy_id & 0xfffffc00) != B53_BRCM_OUI_1 && + (phy_id & 0xfffffc00) != B53_BRCM_OUI_2 && + (phy_id & 0xfffffc00) != B53_BRCM_OUI_3) { + dev_err(&mdiodev->dev, "Unsupported device: 0x%08x\n", phy_id); + return -ENODEV; + } + + dev = b53_switch_alloc(&mdiodev->dev, &b53_mdio_ops, mdiodev->bus); + if (!dev) + return -ENOMEM; + + /* we don't use page 0xff, so force a page set */ + dev->current_page = 0xff; + dev->bus = mdiodev->bus; + + dev_set_drvdata(&mdiodev->dev, dev); + + ret = b53_switch_register(dev); + if (ret) { + dev_err(&mdiodev->dev, "failed to register switch: %i\n", ret); + return ret; + } + + return ret; +} + +static void b53_mdio_remove(struct mdio_device *mdiodev) +{ + struct b53_device *dev = dev_get_drvdata(&mdiodev->dev); + struct dsa_switch *ds = dev->ds; + + dsa_unregister_switch(ds); +} + +static const struct of_device_id b53_of_match[] = { + { .compatible = "brcm,bcm5325" }, + { .compatible = "brcm,bcm53115" }, + { .compatible = "brcm,bcm53125" }, + { .compatible = "brcm,bcm53128" }, + { .compatible = "brcm,bcm5365" }, + { .compatible = "brcm,bcm5395" }, + { .compatible = "brcm,bcm5397" }, + { .compatible = "brcm,bcm5398" }, + { /* sentinel */ }, +}; +MODULE_DEVICE_TABLE(of, b53_of_match); + +static struct mdio_driver b53_mdio_driver = { + .probe = b53_mdio_probe, + .remove = b53_mdio_remove, + .mdiodrv.driver = { + .name = "bcm53xx", + .of_match_table = b53_of_match, + }, +}; + +static int __init b53_mdio_driver_register(void) +{ + return mdio_driver_register(&b53_mdio_driver); +} +module_init(b53_mdio_driver_register); + +static void __exit b53_mdio_driver_unregister(void) +{ + mdio_driver_unregister(&b53_mdio_driver); +} +module_exit(b53_mdio_driver_unregister); + +MODULE_DESCRIPTION("B53 MDIO access driver"); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/net/dsa/b53/b53_mmap.c b/drivers/net/dsa/b53/b53_mmap.c new file mode 100644 index 000000000000..f115ee25c0e8 --- /dev/null +++ b/drivers/net/dsa/b53/b53_mmap.c @@ -0,0 +1,260 @@ +/* + * B53 register access through memory mapped registers + * + * Copyright (C) 2012-2013 Jonas Gorski + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include + +#include "b53_priv.h" + +struct b53_mmap_priv { + void __iomem *regs; +}; + +static int b53_mmap_read8(struct b53_device *dev, u8 page, u8 reg, u8 *val) +{ + u8 __iomem *regs = dev->priv; + + *val = readb(regs + (page << 8) + reg); + + return 0; +} + +static int b53_mmap_read16(struct b53_device *dev, u8 page, u8 reg, u16 *val) +{ + u8 __iomem *regs = dev->priv; + + if (WARN_ON(reg % 2)) + return -EINVAL; + + if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN) && dev->pdata && + dev->pdata->big_endian) + *val = __raw_readw(regs + (page << 8) + reg); + else + *val = readw(regs + (page << 8) + reg); + + return 0; +} + +static int b53_mmap_read32(struct b53_device *dev, u8 page, u8 reg, u32 *val) +{ + u8 __iomem *regs = dev->priv; + + if (WARN_ON(reg % 4)) + return -EINVAL; + + if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN) && dev->pdata && + dev->pdata->big_endian) + *val = __raw_readl(regs + (page << 8) + reg); + else + *val = readl(regs + (page << 8) + reg); + + return 0; +} + +static int b53_mmap_read48(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + if (WARN_ON(reg % 2)) + return -EINVAL; + + if (reg % 4) { + u16 lo; + u32 hi; + + b53_mmap_read16(dev, page, reg, &lo); + b53_mmap_read32(dev, page, reg + 2, &hi); + + *val = ((u64)hi << 16) | lo; + } else { + u32 lo; + u16 hi; + + b53_mmap_read32(dev, page, reg, &lo); + b53_mmap_read16(dev, page, reg + 4, &hi); + + *val = ((u64)hi << 32) | lo; + } + + return 0; +} + +static int b53_mmap_read64(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + u32 hi, lo; + + if (WARN_ON(reg % 4)) + return -EINVAL; + + b53_mmap_read32(dev, page, reg, &lo); + b53_mmap_read32(dev, page, reg + 4, &hi); + + *val = ((u64)hi << 32) | lo; + + return 0; +} + +static int b53_mmap_write8(struct b53_device *dev, u8 page, u8 reg, u8 value) +{ + u8 __iomem *regs = dev->priv; + + writeb(value, regs + (page << 8) + reg); + + return 0; +} + +static int b53_mmap_write16(struct b53_device *dev, u8 page, u8 reg, + u16 value) +{ + u8 __iomem *regs = dev->priv; + + if (WARN_ON(reg % 2)) + return -EINVAL; + + if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN) && dev->pdata && + dev->pdata->big_endian) + __raw_writew(value, regs + (page << 8) + reg); + else + writew(value, regs + (page << 8) + reg); + + return 0; +} + +static int b53_mmap_write32(struct b53_device *dev, u8 page, u8 reg, + u32 value) +{ + u8 __iomem *regs = dev->priv; + + if (WARN_ON(reg % 4)) + return -EINVAL; + + if (IS_ENABLED(CONFIG_CPU_BIG_ENDIAN) && dev->pdata && + dev->pdata->big_endian) + __raw_writel(value, regs + (page << 8) + reg); + else + writel(value, regs + (page << 8) + reg); + + return 0; +} + +static int b53_mmap_write48(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + if (WARN_ON(reg % 2)) + return -EINVAL; + + if (reg % 4) { + u32 hi = (u32)(value >> 16); + u16 lo = (u16)value; + + b53_mmap_write16(dev, page, reg, lo); + b53_mmap_write32(dev, page, reg + 2, hi); + } else { + u16 hi = (u16)(value >> 32); + u32 lo = (u32)value; + + b53_mmap_write32(dev, page, reg, lo); + b53_mmap_write16(dev, page, reg + 4, hi); + } + + return 0; +} + +static int b53_mmap_write64(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + u32 hi, lo; + + hi = upper_32_bits(value); + lo = lower_32_bits(value); + + if (WARN_ON(reg % 4)) + return -EINVAL; + + b53_mmap_write32(dev, page, reg, lo); + b53_mmap_write32(dev, page, reg + 4, hi); + + return 0; +} + +static struct b53_io_ops b53_mmap_ops = { + .read8 = b53_mmap_read8, + .read16 = b53_mmap_read16, + .read32 = b53_mmap_read32, + .read48 = b53_mmap_read48, + .read64 = b53_mmap_read64, + .write8 = b53_mmap_write8, + .write16 = b53_mmap_write16, + .write32 = b53_mmap_write32, + .write48 = b53_mmap_write48, + .write64 = b53_mmap_write64, +}; + +static int b53_mmap_probe(struct platform_device *pdev) +{ + struct b53_platform_data *pdata = pdev->dev.platform_data; + struct b53_device *dev; + + if (!pdata) + return -EINVAL; + + dev = b53_switch_alloc(&pdev->dev, &b53_mmap_ops, pdata->regs); + if (!dev) + return -ENOMEM; + + if (pdata) + dev->pdata = pdata; + + platform_set_drvdata(pdev, dev); + + return b53_switch_register(dev); +} + +static int b53_mmap_remove(struct platform_device *pdev) +{ + struct b53_device *dev = platform_get_drvdata(pdev); + + if (dev) + b53_switch_remove(dev); + + return 0; +} + +static const struct of_device_id b53_mmap_of_table[] = { + { .compatible = "brcm,bcm3384-switch" }, + { .compatible = "brcm,bcm6328-switch" }, + { .compatible = "brcm,bcm6368-switch" }, + { .compatible = "brcm,bcm63xx-switch" }, + { /* sentinel */ }, +}; + +static struct platform_driver b53_mmap_driver = { + .probe = b53_mmap_probe, + .remove = b53_mmap_remove, + .driver = { + .name = "b53-switch", + .of_match_table = b53_mmap_of_table, + }, +}; + +module_platform_driver(b53_mmap_driver); +MODULE_AUTHOR("Jonas Gorski "); +MODULE_DESCRIPTION("B53 MMAP access driver"); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/net/dsa/b53/b53_priv.h b/drivers/net/dsa/b53/b53_priv.h new file mode 100644 index 000000000000..c4844155546b --- /dev/null +++ b/drivers/net/dsa/b53/b53_priv.h @@ -0,0 +1,322 @@ +/* + * B53 common definitions + * + * Copyright (C) 2011-2013 Jonas Gorski + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#ifndef __B53_PRIV_H +#define __B53_PRIV_H + +#include +#include +#include +#include + +struct b53_device; + +struct b53_io_ops { + int (*read8)(struct b53_device *dev, u8 page, u8 reg, u8 *value); + int (*read16)(struct b53_device *dev, u8 page, u8 reg, u16 *value); + int (*read32)(struct b53_device *dev, u8 page, u8 reg, u32 *value); + int (*read48)(struct b53_device *dev, u8 page, u8 reg, u64 *value); + int (*read64)(struct b53_device *dev, u8 page, u8 reg, u64 *value); + int (*write8)(struct b53_device *dev, u8 page, u8 reg, u8 value); + int (*write16)(struct b53_device *dev, u8 page, u8 reg, u16 value); + int (*write32)(struct b53_device *dev, u8 page, u8 reg, u32 value); + int (*write48)(struct b53_device *dev, u8 page, u8 reg, u64 value); + int (*write64)(struct b53_device *dev, u8 page, u8 reg, u64 value); + int (*phy_read16)(struct b53_device *dev, int addr, int reg, u16 *value); + int (*phy_write16)(struct b53_device *dev, int addr, int reg, u16 value); +}; + +enum { + BCM5325_DEVICE_ID = 0x25, + BCM5365_DEVICE_ID = 0x65, + BCM5395_DEVICE_ID = 0x95, + BCM5397_DEVICE_ID = 0x97, + BCM5398_DEVICE_ID = 0x98, + BCM53115_DEVICE_ID = 0x53115, + BCM53125_DEVICE_ID = 0x53125, + BCM53128_DEVICE_ID = 0x53128, + BCM63XX_DEVICE_ID = 0x6300, + BCM53010_DEVICE_ID = 0x53010, + BCM53011_DEVICE_ID = 0x53011, + BCM53012_DEVICE_ID = 0x53012, + BCM53018_DEVICE_ID = 0x53018, + BCM53019_DEVICE_ID = 0x53019, +}; + +#define B53_N_PORTS 9 +#define B53_N_PORTS_25 6 + +struct b53_port { +}; + +struct b53_device { + struct dsa_switch *ds; + struct b53_platform_data *pdata; + const char *name; + + struct mutex reg_mutex; + struct mutex stats_mutex; + const struct b53_io_ops *ops; + + /* chip specific data */ + u32 chip_id; + u8 core_rev; + u8 vta_regs[3]; + u8 duplex_reg; + u8 jumbo_pm_reg; + u8 jumbo_size_reg; + int reset_gpio; + + /* used ports mask */ + u16 enabled_ports; + unsigned int cpu_port; + + /* connect specific data */ + u8 current_page; + struct device *dev; + + /* Master MDIO bus we got probed from */ + struct mii_bus *bus; + + /* Slave MDIO bus we created */ + struct mii_bus *slave_bus; + void *priv; + + /* run time configuration */ + unsigned enable_jumbo:1; + unsigned allow_vid_4095:1; + unsigned int num_vlans; + unsigned int num_ports; + struct b53_port *ports; +}; + +#define b53_for_each_port(dev, i) \ + for (i = 0; i < B53_N_PORTS; i++) \ + if (dev->enabled_ports & BIT(i)) + + +static inline int is5325(struct b53_device *dev) +{ + return dev->chip_id == BCM5325_DEVICE_ID; +} + +static inline int is5365(struct b53_device *dev) +{ +#ifdef CONFIG_BCM47XX + return dev->chip_id == BCM5365_DEVICE_ID; +#else + return 0; +#endif +} + +static inline int is5397_98(struct b53_device *dev) +{ + return dev->chip_id == BCM5397_DEVICE_ID || + dev->chip_id == BCM5398_DEVICE_ID; +} + +static inline int is539x(struct b53_device *dev) +{ + return dev->chip_id == BCM5395_DEVICE_ID || + dev->chip_id == BCM5397_DEVICE_ID || + dev->chip_id == BCM5398_DEVICE_ID; +} + +static inline int is531x5(struct b53_device *dev) +{ + return dev->chip_id == BCM53115_DEVICE_ID || + dev->chip_id == BCM53125_DEVICE_ID || + dev->chip_id == BCM53128_DEVICE_ID; +} + +static inline int is63xx(struct b53_device *dev) +{ +#ifdef CONFIG_BCM63XX + return dev->chip_id == BCM63XX_DEVICE_ID; +#else + return 0; +#endif +} + +static inline int is5301x(struct b53_device *dev) +{ + return dev->chip_id == BCM53010_DEVICE_ID || + dev->chip_id == BCM53011_DEVICE_ID || + dev->chip_id == BCM53012_DEVICE_ID || + dev->chip_id == BCM53018_DEVICE_ID || + dev->chip_id == BCM53019_DEVICE_ID; +} + +#define B53_CPU_PORT_25 5 +#define B53_CPU_PORT 8 + +static inline int is_cpu_port(struct b53_device *dev, int port) +{ + return dev->cpu_port; +} + +struct b53_device *b53_switch_alloc(struct device *base, struct b53_io_ops *ops, + void *priv); + +int b53_switch_detect(struct b53_device *dev); + +int b53_switch_register(struct b53_device *dev); + +static inline void b53_switch_remove(struct b53_device *dev) +{ + dsa_unregister_switch(dev->ds); +} + +static inline int b53_read8(struct b53_device *dev, u8 page, u8 reg, u8 *val) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->read8(dev, page, reg, val); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_read16(struct b53_device *dev, u8 page, u8 reg, u16 *val) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->read16(dev, page, reg, val); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_read32(struct b53_device *dev, u8 page, u8 reg, u32 *val) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->read32(dev, page, reg, val); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_read48(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->read48(dev, page, reg, val); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_read64(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->read64(dev, page, reg, val); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_write8(struct b53_device *dev, u8 page, u8 reg, u8 value) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->write8(dev, page, reg, value); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_write16(struct b53_device *dev, u8 page, u8 reg, + u16 value) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->write16(dev, page, reg, value); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_write32(struct b53_device *dev, u8 page, u8 reg, + u32 value) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->write32(dev, page, reg, value); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_write48(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->write48(dev, page, reg, value); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +static inline int b53_write64(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + int ret; + + mutex_lock(&dev->reg_mutex); + ret = dev->ops->write64(dev, page, reg, value); + mutex_unlock(&dev->reg_mutex); + + return ret; +} + +#ifdef CONFIG_BCM47XX + +#include +#include +#include +static inline int b53_switch_get_reset_gpio(struct b53_device *dev) +{ + enum bcm47xx_board board = bcm47xx_board_get(); + + switch (board) { + case BCM47XX_BOARD_LINKSYS_WRT300NV11: + case BCM47XX_BOARD_LINKSYS_WRT310NV1: + return 8; + default: + return bcm47xx_nvram_gpio_pin("robo_reset"); + } +} +#else +static inline int b53_switch_get_reset_gpio(struct b53_device *dev) +{ + return -ENOENT; +} +#endif +#endif diff --git a/drivers/net/dsa/b53/b53_regs.h b/drivers/net/dsa/b53/b53_regs.h new file mode 100644 index 000000000000..ccf8af7717d3 --- /dev/null +++ b/drivers/net/dsa/b53/b53_regs.h @@ -0,0 +1,358 @@ +/* + * B53 register definitions + * + * Copyright (C) 2004 Broadcom Corporation + * Copyright (C) 2011-2013 Jonas Gorski + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#ifndef __B53_REGS_H +#define __B53_REGS_H + +/* Management Port (SMP) Page offsets */ +#define B53_CTRL_PAGE 0x00 /* Control */ +#define B53_STAT_PAGE 0x01 /* Status */ +#define B53_MGMT_PAGE 0x02 /* Management Mode */ +#define B53_MIB_AC_PAGE 0x03 /* MIB Autocast */ +#define B53_ARLCTRL_PAGE 0x04 /* ARL Control */ +#define B53_ARLIO_PAGE 0x05 /* ARL Access */ +#define B53_FRAMEBUF_PAGE 0x06 /* Management frame access */ +#define B53_MEM_ACCESS_PAGE 0x08 /* Memory access */ + +/* PHY Registers */ +#define B53_PORT_MII_PAGE(i) (0x10 + (i)) /* Port i MII Registers */ +#define B53_IM_PORT_PAGE 0x18 /* Inverse MII Port (to EMAC) */ +#define B53_ALL_PORT_PAGE 0x19 /* All ports MII (broadcast) */ + +/* MIB registers */ +#define B53_MIB_PAGE(i) (0x20 + (i)) + +/* Quality of Service (QoS) Registers */ +#define B53_QOS_PAGE 0x30 + +/* Port VLAN Page */ +#define B53_PVLAN_PAGE 0x31 + +/* VLAN Registers */ +#define B53_VLAN_PAGE 0x34 + +/* Jumbo Frame Registers */ +#define B53_JUMBO_PAGE 0x40 + +/* CFP Configuration Registers Page */ +#define B53_CFP_PAGE 0xa1 + +/************************************************************************* + * Control Page registers + *************************************************************************/ + +/* Port Control Register (8 bit) */ +#define B53_PORT_CTRL(i) (0x00 + (i)) +#define PORT_CTRL_RX_DISABLE BIT(0) +#define PORT_CTRL_TX_DISABLE BIT(1) +#define PORT_CTRL_RX_BCST_EN BIT(2) /* Broadcast RX (P8 only) */ +#define PORT_CTRL_RX_MCST_EN BIT(3) /* Multicast RX (P8 only) */ +#define PORT_CTRL_RX_UCST_EN BIT(4) /* Unicast RX (P8 only) */ +#define PORT_CTRL_STP_STATE_S 5 +#define PORT_CTRL_STP_STATE_MASK (0x7 << PORT_CTRL_STP_STATE_S) + +/* SMP Control Register (8 bit) */ +#define B53_SMP_CTRL 0x0a + +/* Switch Mode Control Register (8 bit) */ +#define B53_SWITCH_MODE 0x0b +#define SM_SW_FWD_MODE BIT(0) /* 1 = Managed Mode */ +#define SM_SW_FWD_EN BIT(1) /* Forwarding Enable */ + +/* IMP Port state override register (8 bit) */ +#define B53_PORT_OVERRIDE_CTRL 0x0e +#define PORT_OVERRIDE_LINK BIT(0) +#define PORT_OVERRIDE_FULL_DUPLEX BIT(1) /* 0 = Half Duplex */ +#define PORT_OVERRIDE_SPEED_S 2 +#define PORT_OVERRIDE_SPEED_10M (0 << PORT_OVERRIDE_SPEED_S) +#define PORT_OVERRIDE_SPEED_100M (1 << PORT_OVERRIDE_SPEED_S) +#define PORT_OVERRIDE_SPEED_1000M (2 << PORT_OVERRIDE_SPEED_S) +#define PORT_OVERRIDE_RV_MII_25 BIT(4) /* BCM5325 only */ +#define PORT_OVERRIDE_RX_FLOW BIT(4) +#define PORT_OVERRIDE_TX_FLOW BIT(5) +#define PORT_OVERRIDE_SPEED_2000M BIT(6) /* BCM5301X only, requires setting 1000M */ +#define PORT_OVERRIDE_EN BIT(7) /* Use the register contents */ + +/* Power-down mode control */ +#define B53_PD_MODE_CTRL_25 0x0f + +/* IP Multicast control (8 bit) */ +#define B53_IP_MULTICAST_CTRL 0x21 +#define B53_IPMC_FWD_EN BIT(1) +#define B53_UC_FWD_EN BIT(6) +#define B53_MC_FWD_EN BIT(7) + +/* (16 bit) */ +#define B53_UC_FLOOD_MASK 0x32 +#define B53_MC_FLOOD_MASK 0x34 +#define B53_IPMC_FLOOD_MASK 0x36 + +/* + * Override Ports 0-7 State on devices with xMII interfaces (8 bit) + * + * For port 8 still use B53_PORT_OVERRIDE_CTRL + * Please note that not all ports are available on every hardware, e.g. BCM5301X + * don't include overriding port 6, BCM63xx also have some limitations. + */ +#define B53_GMII_PORT_OVERRIDE_CTRL(i) (0x58 + (i)) +#define GMII_PO_LINK BIT(0) +#define GMII_PO_FULL_DUPLEX BIT(1) /* 0 = Half Duplex */ +#define GMII_PO_SPEED_S 2 +#define GMII_PO_SPEED_10M (0 << GMII_PO_SPEED_S) +#define GMII_PO_SPEED_100M (1 << GMII_PO_SPEED_S) +#define GMII_PO_SPEED_1000M (2 << GMII_PO_SPEED_S) +#define GMII_PO_RX_FLOW BIT(4) +#define GMII_PO_TX_FLOW BIT(5) +#define GMII_PO_EN BIT(6) /* Use the register contents */ +#define GMII_PO_SPEED_2000M BIT(7) /* BCM5301X only, requires setting 1000M */ + +#define B53_RGMII_CTRL_IMP 0x60 +#define RGMII_CTRL_ENABLE_GMII BIT(7) +#define RGMII_CTRL_TIMING_SEL BIT(2) +#define RGMII_CTRL_DLL_RXC BIT(1) +#define RGMII_CTRL_DLL_TXC BIT(0) + +#define B53_RGMII_CTRL_P(i) (B53_RGMII_CTRL_IMP + (i)) + +/* Software reset register (8 bit) */ +#define B53_SOFTRESET 0x79 +#define SW_RST BIT(7) +#define EN_SW_RST BIT(4) + +/* Fast Aging Control register (8 bit) */ +#define B53_FAST_AGE_CTRL 0x88 +#define FAST_AGE_STATIC BIT(0) +#define FAST_AGE_DYNAMIC BIT(1) +#define FAST_AGE_PORT BIT(2) +#define FAST_AGE_VLAN BIT(3) +#define FAST_AGE_STP BIT(4) +#define FAST_AGE_MC BIT(5) +#define FAST_AGE_DONE BIT(7) + +/************************************************************************* + * Status Page registers + *************************************************************************/ + +/* Link Status Summary Register (16bit) */ +#define B53_LINK_STAT 0x00 + +/* Link Status Change Register (16 bit) */ +#define B53_LINK_STAT_CHANGE 0x02 + +/* Port Speed Summary Register (16 bit for FE, 32 bit for GE) */ +#define B53_SPEED_STAT 0x04 +#define SPEED_PORT_FE(reg, port) (((reg) >> (port)) & 1) +#define SPEED_PORT_GE(reg, port) (((reg) >> 2 * (port)) & 3) +#define SPEED_STAT_10M 0 +#define SPEED_STAT_100M 1 +#define SPEED_STAT_1000M 2 + +/* Duplex Status Summary (16 bit) */ +#define B53_DUPLEX_STAT_FE 0x06 +#define B53_DUPLEX_STAT_GE 0x08 +#define B53_DUPLEX_STAT_63XX 0x0c + +/* Revision ID register for BCM5325 */ +#define B53_REV_ID_25 0x50 + +/* Strap Value (48 bit) */ +#define B53_STRAP_VALUE 0x70 +#define SV_GMII_CTRL_115 BIT(27) + +/************************************************************************* + * Management Mode Page Registers + *************************************************************************/ + +/* Global Management Config Register (8 bit) */ +#define B53_GLOBAL_CONFIG 0x00 +#define GC_RESET_MIB 0x01 +#define GC_RX_BPDU_EN 0x02 +#define GC_MIB_AC_HDR_EN 0x10 +#define GC_MIB_AC_EN 0x20 +#define GC_FRM_MGMT_PORT_M 0xC0 +#define GC_FRM_MGMT_PORT_04 0x00 +#define GC_FRM_MGMT_PORT_MII 0x80 + +/* Broadcom Header control register (8 bit) */ +#define B53_BRCM_HDR 0x03 +#define BRCM_HDR_P8_EN BIT(0) /* Enable tagging on port 8 */ +#define BRCM_HDR_P5_EN BIT(1) /* Enable tagging on port 5 */ + +/* Device ID register (8 or 32 bit) */ +#define B53_DEVICE_ID 0x30 + +/* Revision ID register (8 bit) */ +#define B53_REV_ID 0x40 + +/************************************************************************* + * ARL Access Page Registers + *************************************************************************/ + +/* VLAN Table Access Register (8 bit) */ +#define B53_VT_ACCESS 0x80 +#define B53_VT_ACCESS_9798 0x60 /* for BCM5397/BCM5398 */ +#define B53_VT_ACCESS_63XX 0x60 /* for BCM6328/62/68 */ +#define VTA_CMD_WRITE 0 +#define VTA_CMD_READ 1 +#define VTA_CMD_CLEAR 2 +#define VTA_START_CMD BIT(7) + +/* VLAN Table Index Register (16 bit) */ +#define B53_VT_INDEX 0x81 +#define B53_VT_INDEX_9798 0x61 +#define B53_VT_INDEX_63XX 0x62 + +/* VLAN Table Entry Register (32 bit) */ +#define B53_VT_ENTRY 0x83 +#define B53_VT_ENTRY_9798 0x63 +#define B53_VT_ENTRY_63XX 0x64 +#define VTE_MEMBERS 0x1ff +#define VTE_UNTAG_S 9 +#define VTE_UNTAG (0x1ff << 9) + +/************************************************************************* + * Port VLAN Registers + *************************************************************************/ + +/* Port VLAN mask (16 bit) IMP port is always 8, also on 5325 & co */ +#define B53_PVLAN_PORT_MASK(i) ((i) * 2) + +/************************************************************************* + * 802.1Q Page Registers + *************************************************************************/ + +/* Global QoS Control (8 bit) */ +#define B53_QOS_GLOBAL_CTL 0x00 + +/* Enable 802.1Q for individual Ports (16 bit) */ +#define B53_802_1P_EN 0x04 + +/************************************************************************* + * VLAN Page Registers + *************************************************************************/ + +/* VLAN Control 0 (8 bit) */ +#define B53_VLAN_CTRL0 0x00 +#define VC0_8021PF_CTRL_MASK 0x3 +#define VC0_8021PF_CTRL_NONE 0x0 +#define VC0_8021PF_CTRL_CHANGE_PRI 0x1 +#define VC0_8021PF_CTRL_CHANGE_VID 0x2 +#define VC0_8021PF_CTRL_CHANGE_BOTH 0x3 +#define VC0_8021QF_CTRL_MASK 0xc +#define VC0_8021QF_CTRL_CHANGE_PRI 0x1 +#define VC0_8021QF_CTRL_CHANGE_VID 0x2 +#define VC0_8021QF_CTRL_CHANGE_BOTH 0x3 +#define VC0_RESERVED_1 BIT(1) +#define VC0_DROP_VID_MISS BIT(4) +#define VC0_VID_HASH_VID BIT(5) +#define VC0_VID_CHK_EN BIT(6) /* Use VID,DA or VID,SA */ +#define VC0_VLAN_EN BIT(7) /* 802.1Q VLAN Enabled */ + +/* VLAN Control 1 (8 bit) */ +#define B53_VLAN_CTRL1 0x01 +#define VC1_RX_MCST_TAG_EN BIT(1) +#define VC1_RX_MCST_FWD_EN BIT(2) +#define VC1_RX_MCST_UNTAG_EN BIT(3) + +/* VLAN Control 2 (8 bit) */ +#define B53_VLAN_CTRL2 0x02 + +/* VLAN Control 3 (8 bit when BCM5325, 16 bit else) */ +#define B53_VLAN_CTRL3 0x03 +#define B53_VLAN_CTRL3_63XX 0x04 +#define VC3_MAXSIZE_1532 BIT(6) /* 5325 only */ +#define VC3_HIGH_8BIT_EN BIT(7) /* 5325 only */ + +/* VLAN Control 4 (8 bit) */ +#define B53_VLAN_CTRL4 0x05 +#define B53_VLAN_CTRL4_25 0x04 +#define B53_VLAN_CTRL4_63XX 0x06 +#define VC4_ING_VID_CHECK_S 6 +#define VC4_ING_VID_CHECK_MASK (0x3 << VC4_ING_VID_CHECK_S) +#define VC4_ING_VID_VIO_FWD 0 /* forward, but do not learn */ +#define VC4_ING_VID_VIO_DROP 1 /* drop VID violations */ +#define VC4_NO_ING_VID_CHK 2 /* do not check */ +#define VC4_ING_VID_VIO_TO_IMP 3 /* redirect to MII port */ + +/* VLAN Control 5 (8 bit) */ +#define B53_VLAN_CTRL5 0x06 +#define B53_VLAN_CTRL5_25 0x05 +#define B53_VLAN_CTRL5_63XX 0x07 +#define VC5_VID_FFF_EN BIT(2) +#define VC5_DROP_VTABLE_MISS BIT(3) + +/* VLAN Control 6 (8 bit) */ +#define B53_VLAN_CTRL6 0x07 +#define B53_VLAN_CTRL6_63XX 0x08 + +/* VLAN Table Access Register (16 bit) */ +#define B53_VLAN_TABLE_ACCESS_25 0x06 /* BCM5325E/5350 */ +#define B53_VLAN_TABLE_ACCESS_65 0x08 /* BCM5365 */ +#define VTA_VID_LOW_MASK_25 0xf +#define VTA_VID_LOW_MASK_65 0xff +#define VTA_VID_HIGH_S_25 4 +#define VTA_VID_HIGH_S_65 8 +#define VTA_VID_HIGH_MASK_25 (0xff << VTA_VID_HIGH_S_25E) +#define VTA_VID_HIGH_MASK_65 (0xf << VTA_VID_HIGH_S_65) +#define VTA_RW_STATE BIT(12) +#define VTA_RW_STATE_RD 0 +#define VTA_RW_STATE_WR BIT(12) +#define VTA_RW_OP_EN BIT(13) + +/* VLAN Read/Write Registers for (16/32 bit) */ +#define B53_VLAN_WRITE_25 0x08 +#define B53_VLAN_WRITE_65 0x0a +#define B53_VLAN_READ 0x0c +#define VA_MEMBER_MASK 0x3f +#define VA_UNTAG_S_25 6 +#define VA_UNTAG_MASK_25 0x3f +#define VA_UNTAG_S_65 7 +#define VA_UNTAG_MASK_65 0x1f +#define VA_VID_HIGH_S 12 +#define VA_VID_HIGH_MASK (0xffff << VA_VID_HIGH_S) +#define VA_VALID_25 BIT(20) +#define VA_VALID_25_R4 BIT(24) +#define VA_VALID_65 BIT(14) + +/* VLAN Port Default Tag (16 bit) */ +#define B53_VLAN_PORT_DEF_TAG(i) (0x10 + 2 * (i)) + +/************************************************************************* + * Jumbo Frame Page Registers + *************************************************************************/ + +/* Jumbo Enable Port Mask (bit i == port i enabled) (32 bit) */ +#define B53_JUMBO_PORT_MASK 0x01 +#define B53_JUMBO_PORT_MASK_63XX 0x04 +#define JPM_10_100_JUMBO_EN BIT(24) /* GigE always enabled */ + +/* Good Frame Max Size without 802.1Q TAG (16 bit) */ +#define B53_JUMBO_MAX_SIZE 0x05 +#define B53_JUMBO_MAX_SIZE_63XX 0x08 +#define JMS_MIN_SIZE 1518 +#define JMS_MAX_SIZE 9724 + +/************************************************************************* + * CFP Configuration Page Registers + *************************************************************************/ + +/* CFP Control Register with ports map (8 bit) */ +#define B53_CFP_CTRL 0x00 + +#endif /* !__B53_REGS_H */ diff --git a/drivers/net/dsa/b53/b53_spi.c b/drivers/net/dsa/b53/b53_spi.c new file mode 100644 index 000000000000..2bda0b5f1578 --- /dev/null +++ b/drivers/net/dsa/b53/b53_spi.c @@ -0,0 +1,331 @@ +/* + * B53 register access through SPI + * + * Copyright (C) 2011-2013 Jonas Gorski + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#include + +#include +#include +#include +#include +#include + +#include "b53_priv.h" + +#define B53_SPI_DATA 0xf0 + +#define B53_SPI_STATUS 0xfe +#define B53_SPI_CMD_SPIF BIT(7) +#define B53_SPI_CMD_RACK BIT(5) + +#define B53_SPI_CMD_READ 0x00 +#define B53_SPI_CMD_WRITE 0x01 +#define B53_SPI_CMD_NORMAL 0x60 +#define B53_SPI_CMD_FAST 0x10 + +#define B53_SPI_PAGE_SELECT 0xff + +static inline int b53_spi_read_reg(struct spi_device *spi, u8 reg, u8 *val, + unsigned int len) +{ + u8 txbuf[2]; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_READ; + txbuf[1] = reg; + + return spi_write_then_read(spi, txbuf, 2, val, len); +} + +static inline int b53_spi_clear_status(struct spi_device *spi) +{ + unsigned int i; + u8 rxbuf; + int ret; + + for (i = 0; i < 10; i++) { + ret = b53_spi_read_reg(spi, B53_SPI_STATUS, &rxbuf, 1); + if (ret) + return ret; + + if (!(rxbuf & B53_SPI_CMD_SPIF)) + break; + + mdelay(1); + } + + if (i == 10) + return -EIO; + + return 0; +} + +static inline int b53_spi_set_page(struct spi_device *spi, u8 page) +{ + u8 txbuf[3]; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_WRITE; + txbuf[1] = B53_SPI_PAGE_SELECT; + txbuf[2] = page; + + return spi_write(spi, txbuf, sizeof(txbuf)); +} + +static inline int b53_prepare_reg_access(struct spi_device *spi, u8 page) +{ + int ret = b53_spi_clear_status(spi); + + if (ret) + return ret; + + return b53_spi_set_page(spi, page); +} + +static int b53_spi_prepare_reg_read(struct spi_device *spi, u8 reg) +{ + u8 rxbuf; + int retry_count; + int ret; + + ret = b53_spi_read_reg(spi, reg, &rxbuf, 1); + if (ret) + return ret; + + for (retry_count = 0; retry_count < 10; retry_count++) { + ret = b53_spi_read_reg(spi, B53_SPI_STATUS, &rxbuf, 1); + if (ret) + return ret; + + if (rxbuf & B53_SPI_CMD_RACK) + break; + + mdelay(1); + } + + if (retry_count == 10) + return -EIO; + + return 0; +} + +static int b53_spi_read(struct b53_device *dev, u8 page, u8 reg, u8 *data, + unsigned int len) +{ + struct spi_device *spi = dev->priv; + int ret; + + ret = b53_prepare_reg_access(spi, page); + if (ret) + return ret; + + ret = b53_spi_prepare_reg_read(spi, reg); + if (ret) + return ret; + + return b53_spi_read_reg(spi, B53_SPI_DATA, data, len); +} + +static int b53_spi_read8(struct b53_device *dev, u8 page, u8 reg, u8 *val) +{ + return b53_spi_read(dev, page, reg, val, 1); +} + +static int b53_spi_read16(struct b53_device *dev, u8 page, u8 reg, u16 *val) +{ + int ret = b53_spi_read(dev, page, reg, (u8 *)val, 2); + + if (!ret) + *val = le16_to_cpu(*val); + + return ret; +} + +static int b53_spi_read32(struct b53_device *dev, u8 page, u8 reg, u32 *val) +{ + int ret = b53_spi_read(dev, page, reg, (u8 *)val, 4); + + if (!ret) + *val = le32_to_cpu(*val); + + return ret; +} + +static int b53_spi_read48(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + int ret; + + *val = 0; + ret = b53_spi_read(dev, page, reg, (u8 *)val, 6); + if (!ret) + *val = le64_to_cpu(*val); + + return ret; +} + +static int b53_spi_read64(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + int ret = b53_spi_read(dev, page, reg, (u8 *)val, 8); + + if (!ret) + *val = le64_to_cpu(*val); + + return ret; +} + +static int b53_spi_write8(struct b53_device *dev, u8 page, u8 reg, u8 value) +{ + struct spi_device *spi = dev->priv; + int ret; + u8 txbuf[3]; + + ret = b53_prepare_reg_access(spi, page); + if (ret) + return ret; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_WRITE; + txbuf[1] = reg; + txbuf[2] = value; + + return spi_write(spi, txbuf, sizeof(txbuf)); +} + +static int b53_spi_write16(struct b53_device *dev, u8 page, u8 reg, u16 value) +{ + struct spi_device *spi = dev->priv; + int ret; + u8 txbuf[4]; + + ret = b53_prepare_reg_access(spi, page); + if (ret) + return ret; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_WRITE; + txbuf[1] = reg; + put_unaligned_le16(value, &txbuf[2]); + + return spi_write(spi, txbuf, sizeof(txbuf)); +} + +static int b53_spi_write32(struct b53_device *dev, u8 page, u8 reg, u32 value) +{ + struct spi_device *spi = dev->priv; + int ret; + u8 txbuf[6]; + + ret = b53_prepare_reg_access(spi, page); + if (ret) + return ret; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_WRITE; + txbuf[1] = reg; + put_unaligned_le32(value, &txbuf[2]); + + return spi_write(spi, txbuf, sizeof(txbuf)); +} + +static int b53_spi_write48(struct b53_device *dev, u8 page, u8 reg, u64 value) +{ + struct spi_device *spi = dev->priv; + int ret; + u8 txbuf[10]; + + ret = b53_prepare_reg_access(spi, page); + if (ret) + return ret; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_WRITE; + txbuf[1] = reg; + put_unaligned_le64(value, &txbuf[2]); + + return spi_write(spi, txbuf, sizeof(txbuf) - 2); +} + +static int b53_spi_write64(struct b53_device *dev, u8 page, u8 reg, u64 value) +{ + struct spi_device *spi = dev->priv; + int ret; + u8 txbuf[10]; + + ret = b53_prepare_reg_access(spi, page); + if (ret) + return ret; + + txbuf[0] = B53_SPI_CMD_NORMAL | B53_SPI_CMD_WRITE; + txbuf[1] = reg; + put_unaligned_le64(value, &txbuf[2]); + + return spi_write(spi, txbuf, sizeof(txbuf)); +} + +static struct b53_io_ops b53_spi_ops = { + .read8 = b53_spi_read8, + .read16 = b53_spi_read16, + .read32 = b53_spi_read32, + .read48 = b53_spi_read48, + .read64 = b53_spi_read64, + .write8 = b53_spi_write8, + .write16 = b53_spi_write16, + .write32 = b53_spi_write32, + .write48 = b53_spi_write48, + .write64 = b53_spi_write64, +}; + +static int b53_spi_probe(struct spi_device *spi) +{ + struct b53_device *dev; + int ret; + + dev = b53_switch_alloc(&spi->dev, &b53_spi_ops, spi); + if (!dev) + return -ENOMEM; + + if (spi->dev.platform_data) + dev->pdata = spi->dev.platform_data; + + ret = b53_switch_register(dev); + if (ret) + return ret; + + spi_set_drvdata(spi, dev); + + return 0; +} + +static int b53_spi_remove(struct spi_device *spi) +{ + struct b53_device *dev = spi_get_drvdata(spi); + + if (dev) + b53_switch_remove(dev); + + return 0; +} + +static struct spi_driver b53_spi_driver = { + .driver = { + .name = "b53-switch", + .bus = &spi_bus_type, + .owner = THIS_MODULE, + }, + .probe = b53_spi_probe, + .remove = b53_spi_remove, +}; + +module_spi_driver(b53_spi_driver); + +MODULE_AUTHOR("Jonas Gorski "); +MODULE_DESCRIPTION("B53 SPI access driver"); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/net/dsa/b53/b53_srab.c b/drivers/net/dsa/b53/b53_srab.c new file mode 100644 index 000000000000..70fd47284535 --- /dev/null +++ b/drivers/net/dsa/b53/b53_srab.c @@ -0,0 +1,415 @@ +/* + * B53 register access through Switch Register Access Bridge Registers + * + * Copyright (C) 2013 Hauke Mehrtens + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#include +#include +#include +#include +#include + +#include "b53_priv.h" + +/* command and status register of the SRAB */ +#define B53_SRAB_CMDSTAT 0x2c +#define B53_SRAB_CMDSTAT_RST BIT(2) +#define B53_SRAB_CMDSTAT_WRITE BIT(1) +#define B53_SRAB_CMDSTAT_GORDYN BIT(0) +#define B53_SRAB_CMDSTAT_PAGE 24 +#define B53_SRAB_CMDSTAT_REG 16 + +/* high order word of write data to switch registe */ +#define B53_SRAB_WD_H 0x30 + +/* low order word of write data to switch registe */ +#define B53_SRAB_WD_L 0x34 + +/* high order word of read data from switch register */ +#define B53_SRAB_RD_H 0x38 + +/* low order word of read data from switch register */ +#define B53_SRAB_RD_L 0x3c + +/* command and status register of the SRAB */ +#define B53_SRAB_CTRLS 0x40 +#define B53_SRAB_CTRLS_RCAREQ BIT(3) +#define B53_SRAB_CTRLS_RCAGNT BIT(4) +#define B53_SRAB_CTRLS_SW_INIT_DONE BIT(6) + +/* the register captures interrupt pulses from the switch */ +#define B53_SRAB_INTR 0x44 +#define B53_SRAB_INTR_P(x) BIT(x) +#define B53_SRAB_SWITCH_PHY BIT(8) +#define B53_SRAB_1588_SYNC BIT(9) +#define B53_SRAB_IMP1_SLEEP_TIMER BIT(10) +#define B53_SRAB_P7_SLEEP_TIMER BIT(11) +#define B53_SRAB_IMP0_SLEEP_TIMER BIT(12) + +struct b53_srab_priv { + void __iomem *regs; +}; + +static int b53_srab_request_grant(struct b53_device *dev) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + u32 ctrls; + int i; + + ctrls = readl(regs + B53_SRAB_CTRLS); + ctrls |= B53_SRAB_CTRLS_RCAREQ; + writel(ctrls, regs + B53_SRAB_CTRLS); + + for (i = 0; i < 20; i++) { + ctrls = readl(regs + B53_SRAB_CTRLS); + if (ctrls & B53_SRAB_CTRLS_RCAGNT) + break; + usleep_range(10, 100); + } + if (WARN_ON(i == 5)) + return -EIO; + + return 0; +} + +static void b53_srab_release_grant(struct b53_device *dev) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + u32 ctrls; + + ctrls = readl(regs + B53_SRAB_CTRLS); + ctrls &= ~B53_SRAB_CTRLS_RCAREQ; + writel(ctrls, regs + B53_SRAB_CTRLS); +} + +static int b53_srab_op(struct b53_device *dev, u8 page, u8 reg, u32 op) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int i; + u32 cmdstat; + + /* set register address */ + cmdstat = (page << B53_SRAB_CMDSTAT_PAGE) | + (reg << B53_SRAB_CMDSTAT_REG) | + B53_SRAB_CMDSTAT_GORDYN | + op; + writel(cmdstat, regs + B53_SRAB_CMDSTAT); + + /* check if operation completed */ + for (i = 0; i < 5; ++i) { + cmdstat = readl(regs + B53_SRAB_CMDSTAT); + if (!(cmdstat & B53_SRAB_CMDSTAT_GORDYN)) + break; + usleep_range(10, 100); + } + + if (WARN_ON(i == 5)) + return -EIO; + + return 0; +} + +static int b53_srab_read8(struct b53_device *dev, u8 page, u8 reg, u8 *val) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + ret = b53_srab_op(dev, page, reg, 0); + if (ret) + goto err; + + *val = readl(regs + B53_SRAB_RD_L) & 0xff; + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_read16(struct b53_device *dev, u8 page, u8 reg, u16 *val) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + ret = b53_srab_op(dev, page, reg, 0); + if (ret) + goto err; + + *val = readl(regs + B53_SRAB_RD_L) & 0xffff; + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_read32(struct b53_device *dev, u8 page, u8 reg, u32 *val) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + ret = b53_srab_op(dev, page, reg, 0); + if (ret) + goto err; + + *val = readl(regs + B53_SRAB_RD_L); + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_read48(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + ret = b53_srab_op(dev, page, reg, 0); + if (ret) + goto err; + + *val = readl(regs + B53_SRAB_RD_L); + *val += ((u64)readl(regs + B53_SRAB_RD_H) & 0xffff) << 32; + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_read64(struct b53_device *dev, u8 page, u8 reg, u64 *val) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + ret = b53_srab_op(dev, page, reg, 0); + if (ret) + goto err; + + *val = readl(regs + B53_SRAB_RD_L); + *val += (u64)readl(regs + B53_SRAB_RD_H) << 32; + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_write8(struct b53_device *dev, u8 page, u8 reg, u8 value) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + writel(value, regs + B53_SRAB_WD_L); + + ret = b53_srab_op(dev, page, reg, B53_SRAB_CMDSTAT_WRITE); + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_write16(struct b53_device *dev, u8 page, u8 reg, + u16 value) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + writel(value, regs + B53_SRAB_WD_L); + + ret = b53_srab_op(dev, page, reg, B53_SRAB_CMDSTAT_WRITE); + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_write32(struct b53_device *dev, u8 page, u8 reg, + u32 value) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + writel(value, regs + B53_SRAB_WD_L); + + ret = b53_srab_op(dev, page, reg, B53_SRAB_CMDSTAT_WRITE); + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_write48(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + writel((u32)value, regs + B53_SRAB_WD_L); + writel((u16)(value >> 32), regs + B53_SRAB_WD_H); + + ret = b53_srab_op(dev, page, reg, B53_SRAB_CMDSTAT_WRITE); + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static int b53_srab_write64(struct b53_device *dev, u8 page, u8 reg, + u64 value) +{ + struct b53_srab_priv *priv = dev->priv; + u8 __iomem *regs = priv->regs; + int ret = 0; + + ret = b53_srab_request_grant(dev); + if (ret) + goto err; + + writel((u32)value, regs + B53_SRAB_WD_L); + writel((u32)(value >> 32), regs + B53_SRAB_WD_H); + + ret = b53_srab_op(dev, page, reg, B53_SRAB_CMDSTAT_WRITE); + +err: + b53_srab_release_grant(dev); + + return ret; +} + +static struct b53_io_ops b53_srab_ops = { + .read8 = b53_srab_read8, + .read16 = b53_srab_read16, + .read32 = b53_srab_read32, + .read48 = b53_srab_read48, + .read64 = b53_srab_read64, + .write8 = b53_srab_write8, + .write16 = b53_srab_write16, + .write32 = b53_srab_write32, + .write48 = b53_srab_write48, + .write64 = b53_srab_write64, +}; + +static int b53_srab_probe(struct platform_device *pdev) +{ + struct b53_srab_priv *priv; + struct b53_device *dev; + struct resource *r; + + priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + r = platform_get_resource(pdev, IORESOURCE_MEM, 0); + priv->regs = devm_ioremap_resource(&pdev->dev, r); + if (IS_ERR(priv->regs)) + return -ENOMEM; + + dev = b53_switch_alloc(&pdev->dev, &b53_srab_ops, priv); + if (!dev) + return -ENOMEM; + + platform_set_drvdata(pdev, dev); + + return b53_switch_register(dev); +} + +static int b53_srab_remove(struct platform_device *pdev) +{ + struct b53_device *dev = platform_get_drvdata(pdev); + + if (dev) + b53_switch_remove(dev); + + return 0; +} + +static const struct of_device_id b53_srab_of_match[] = { + { .compatible = "brcm,bcm53010-srab" }, + { .compatible = "brcm,bcm53011-srab" }, + { .compatible = "brcm,bcm53012-srab" }, + { .compatible = "brcm,bcm53018-srab" }, + { .compatible = "brcm,bcm53019-srab" }, + { .compatible = "brcm,bcm5301x-srab" }, + { /* sentinel */ }, +}; + +static struct platform_driver b53_srab_driver = { + .probe = b53_srab_probe, + .remove = b53_srab_remove, + .driver = { + .name = "b53-srab-switch", + .of_match_table = b53_srab_of_match, + }, +}; + +module_platform_driver(b53_srab_driver); +MODULE_AUTHOR("Hauke Mehrtens "); +MODULE_DESCRIPTION("B53 Switch Register Access Bridge Registers (SRAB) access driver"); +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/include/linux/platform_data/b53.h b/include/linux/platform_data/b53.h new file mode 100644 index 000000000000..69d279c0da96 --- /dev/null +++ b/include/linux/platform_data/b53.h @@ -0,0 +1,33 @@ +/* + * B53 platform data + * + * Copyright (C) 2013 Jonas Gorski + * + * Permission to use, copy, modify, and/or distribute this software for any + * purpose with or without fee is hereby granted, provided that the above + * copyright notice and this permission notice appear in all copies. + * + * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES + * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES + * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN + * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF + * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. + */ + +#ifndef __B53_H +#define __B53_H + +#include + +struct b53_platform_data { + u32 chip_id; + u16 enabled_ports; + + /* only used by MMAP'd driver */ + unsigned big_endian:1; + void __iomem *regs; +}; + +#endif -- cgit v1.2.3 From 2341e0775747864b684abe8627f3d45b167f2940 Mon Sep 17 00:00:00 2001 From: David Howells Date: Thu, 9 Jun 2016 23:02:51 +0100 Subject: rxrpc: Simplify connect() implementation and simplify sendmsg() op Simplify the RxRPC connect() implementation. It will just note the destination address it is given, and if a sendmsg() comes along with no address, this will be assigned as the address. No transport struct will be held internally, which will allow us to remove this later. Simplify sendmsg() also. Whilst a call is active, userspace refers to it by a private unique user ID specified in a control message. When sendmsg() sees a user ID that doesn't map to an extant call, it creates a new call for that user ID and attempts to add it. If, when we try to add it, the user ID is now registered, we now reject the message with -EEXIST. We should never see this situation unless two threads are racing, trying to create a call with the same ID - which would be an error. It also isn't required to provide sendmsg() with an address - provided the control message data holds a user ID that maps to a currently active call. Signed-off-by: David Howells Signed-off-by: David S. Miller --- include/linux/rxrpc.h | 18 +++-- net/rxrpc/af_rxrpc.c | 177 ++++++++++++++++--------------------------- net/rxrpc/ar-call.c | 158 ++++++++++++++++----------------------- net/rxrpc/ar-connection.c | 17 ----- net/rxrpc/ar-internal.h | 22 +++--- net/rxrpc/ar-output.c | 186 ++++++++++++++++++++++------------------------ 6 files changed, 234 insertions(+), 344 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rxrpc.h b/include/linux/rxrpc.h index a53915cd5581..1e8f216e2cf1 100644 --- a/include/linux/rxrpc.h +++ b/include/linux/rxrpc.h @@ -40,16 +40,18 @@ struct sockaddr_rxrpc { /* * RxRPC control messages + * - If neither abort or accept are specified, the message is a data message. * - terminal messages mean that a user call ID tag can be recycled + * - s/r/- indicate whether these are applicable to sendmsg() and/or recvmsg() */ -#define RXRPC_USER_CALL_ID 1 /* user call ID specifier */ -#define RXRPC_ABORT 2 /* abort request / notification [terminal] */ -#define RXRPC_ACK 3 /* [Server] RPC op final ACK received [terminal] */ -#define RXRPC_NET_ERROR 5 /* network error received [terminal] */ -#define RXRPC_BUSY 6 /* server busy received [terminal] */ -#define RXRPC_LOCAL_ERROR 7 /* local error generated [terminal] */ -#define RXRPC_NEW_CALL 8 /* [Server] new incoming call notification */ -#define RXRPC_ACCEPT 9 /* [Server] accept request */ +#define RXRPC_USER_CALL_ID 1 /* sr: user call ID specifier */ +#define RXRPC_ABORT 2 /* sr: abort request / notification [terminal] */ +#define RXRPC_ACK 3 /* -r: [Service] RPC op final ACK received [terminal] */ +#define RXRPC_NET_ERROR 5 /* -r: network error received [terminal] */ +#define RXRPC_BUSY 6 /* -r: server busy received [terminal] */ +#define RXRPC_LOCAL_ERROR 7 /* -r: local error generated [terminal] */ +#define RXRPC_NEW_CALL 8 /* -r: [Service] new incoming call notification */ +#define RXRPC_ACCEPT 9 /* s-: [Service] accept request */ /* * RxRPC security levels diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 7840b8e7da80..38512a200db6 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -139,33 +139,33 @@ static int rxrpc_bind(struct socket *sock, struct sockaddr *saddr, int len) lock_sock(&rx->sk); - if (rx->sk.sk_state != RXRPC_UNCONNECTED) { + if (rx->sk.sk_state != RXRPC_UNBOUND) { ret = -EINVAL; goto error_unlock; } memcpy(&rx->srx, srx, sizeof(rx->srx)); - /* Find or create a local transport endpoint to use */ local = rxrpc_lookup_local(&rx->srx); if (IS_ERR(local)) { ret = PTR_ERR(local); goto error_unlock; } - rx->local = local; - if (srx->srx_service) { + if (rx->srx.srx_service) { write_lock_bh(&local->services_lock); list_for_each_entry(prx, &local->services, listen_link) { - if (prx->srx.srx_service == srx->srx_service) + if (prx->srx.srx_service == rx->srx.srx_service) goto service_in_use; } + rx->local = local; list_add_tail(&rx->listen_link, &local->services); write_unlock_bh(&local->services_lock); rx->sk.sk_state = RXRPC_SERVER_BOUND; } else { + rx->local = local; rx->sk.sk_state = RXRPC_CLIENT_BOUND; } @@ -174,8 +174,9 @@ static int rxrpc_bind(struct socket *sock, struct sockaddr *saddr, int len) return 0; service_in_use: - ret = -EADDRINUSE; write_unlock_bh(&local->services_lock); + rxrpc_put_local(local); + ret = -EADDRINUSE; error_unlock: release_sock(&rx->sk); error: @@ -197,11 +198,11 @@ static int rxrpc_listen(struct socket *sock, int backlog) lock_sock(&rx->sk); switch (rx->sk.sk_state) { - case RXRPC_UNCONNECTED: + case RXRPC_UNBOUND: ret = -EADDRNOTAVAIL; break; + case RXRPC_CLIENT_UNBOUND: case RXRPC_CLIENT_BOUND: - case RXRPC_CLIENT_CONNECTED: default: ret = -EBUSY; break; @@ -221,20 +222,18 @@ static int rxrpc_listen(struct socket *sock, int backlog) /* * find a transport by address */ -static struct rxrpc_transport *rxrpc_name_to_transport(struct socket *sock, - struct sockaddr *addr, - int addr_len, int flags, - gfp_t gfp) +struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_sock *rx, + struct sockaddr *addr, + int addr_len, int flags, + gfp_t gfp) { struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *) addr; struct rxrpc_transport *trans; - struct rxrpc_sock *rx = rxrpc_sk(sock->sk); struct rxrpc_peer *peer; _enter("%p,%p,%d,%d", rx, addr, addr_len, flags); ASSERT(rx->local != NULL); - ASSERT(rx->sk.sk_state > RXRPC_UNCONNECTED); if (rx->srx.transport_type != srx->transport_type) return ERR_PTR(-ESOCKTNOSUPPORT); @@ -256,7 +255,7 @@ static struct rxrpc_transport *rxrpc_name_to_transport(struct socket *sock, /** * rxrpc_kernel_begin_call - Allow a kernel service to begin a call * @sock: The socket on which to make the call - * @srx: The address of the peer to contact (defaults to socket setting) + * @srx: The address of the peer to contact * @key: The security context to use (defaults to socket setting) * @user_call_ID: The ID to use * @@ -282,25 +281,14 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, lock_sock(&rx->sk); - if (srx) { - trans = rxrpc_name_to_transport(sock, (struct sockaddr *) srx, - sizeof(*srx), 0, gfp); - if (IS_ERR(trans)) { - call = ERR_CAST(trans); - trans = NULL; - goto out_notrans; - } - } else { - trans = rx->trans; - if (!trans) { - call = ERR_PTR(-ENOTCONN); - goto out_notrans; - } - atomic_inc(&trans->usage); + trans = rxrpc_name_to_transport(rx, (struct sockaddr *)srx, + sizeof(*srx), 0, gfp); + if (IS_ERR(trans)) { + call = ERR_CAST(trans); + trans = NULL; + goto out_notrans; } - if (!srx) - srx = &rx->srx; if (!key) key = rx->key; if (key && !key->payload.data[0]) @@ -312,8 +300,7 @@ struct rxrpc_call *rxrpc_kernel_begin_call(struct socket *sock, goto out; } - call = rxrpc_get_client_call(rx, trans, bundle, user_call_ID, true, - gfp); + call = rxrpc_new_client_call(rx, trans, bundle, user_call_ID, gfp); rxrpc_put_bundle(trans, bundle); out: rxrpc_put_transport(trans); @@ -369,11 +356,8 @@ EXPORT_SYMBOL(rxrpc_kernel_intercept_rx_messages); static int rxrpc_connect(struct socket *sock, struct sockaddr *addr, int addr_len, int flags) { - struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *) addr; - struct sock *sk = sock->sk; - struct rxrpc_transport *trans; - struct rxrpc_local *local; - struct rxrpc_sock *rx = rxrpc_sk(sk); + struct sockaddr_rxrpc *srx = (struct sockaddr_rxrpc *)addr; + struct rxrpc_sock *rx = rxrpc_sk(sock->sk); int ret; _enter("%p,%p,%d,%d", rx, addr, addr_len, flags); @@ -386,45 +370,28 @@ static int rxrpc_connect(struct socket *sock, struct sockaddr *addr, lock_sock(&rx->sk); + ret = -EISCONN; + if (test_bit(RXRPC_SOCK_CONNECTED, &rx->flags)) + goto error; + switch (rx->sk.sk_state) { - case RXRPC_UNCONNECTED: - /* find a local transport endpoint if we don't have one already */ - ASSERTCMP(rx->local, ==, NULL); - rx->srx.srx_family = AF_RXRPC; - rx->srx.srx_service = 0; - rx->srx.transport_type = srx->transport_type; - rx->srx.transport_len = sizeof(sa_family_t); - rx->srx.transport.family = srx->transport.family; - local = rxrpc_lookup_local(&rx->srx); - if (IS_ERR(local)) { - release_sock(&rx->sk); - return PTR_ERR(local); - } - rx->local = local; - rx->sk.sk_state = RXRPC_CLIENT_BOUND; + case RXRPC_UNBOUND: + rx->sk.sk_state = RXRPC_CLIENT_UNBOUND; + case RXRPC_CLIENT_UNBOUND: case RXRPC_CLIENT_BOUND: break; - case RXRPC_CLIENT_CONNECTED: - release_sock(&rx->sk); - return -EISCONN; default: - release_sock(&rx->sk); - return -EBUSY; /* server sockets can't connect as well */ - } - - trans = rxrpc_name_to_transport(sock, addr, addr_len, flags, - GFP_KERNEL); - if (IS_ERR(trans)) { - release_sock(&rx->sk); - _leave(" = %ld", PTR_ERR(trans)); - return PTR_ERR(trans); + ret = -EBUSY; + goto error; } - rx->trans = trans; - rx->sk.sk_state = RXRPC_CLIENT_CONNECTED; + rx->connect_srx = *srx; + set_bit(RXRPC_SOCK_CONNECTED, &rx->flags); + ret = 0; +error: release_sock(&rx->sk); - return 0; + return ret; } /* @@ -438,7 +405,7 @@ static int rxrpc_connect(struct socket *sock, struct sockaddr *addr, */ static int rxrpc_sendmsg(struct socket *sock, struct msghdr *m, size_t len) { - struct rxrpc_transport *trans; + struct rxrpc_local *local; struct rxrpc_sock *rx = rxrpc_sk(sock->sk); int ret; @@ -455,48 +422,38 @@ static int rxrpc_sendmsg(struct socket *sock, struct msghdr *m, size_t len) } } - trans = NULL; lock_sock(&rx->sk); - if (m->msg_name) { - ret = -EISCONN; - trans = rxrpc_name_to_transport(sock, m->msg_name, - m->msg_namelen, 0, GFP_KERNEL); - if (IS_ERR(trans)) { - ret = PTR_ERR(trans); - trans = NULL; - goto out; - } - } else { - trans = rx->trans; - if (trans) - atomic_inc(&trans->usage); - } - switch (rx->sk.sk_state) { - case RXRPC_SERVER_LISTENING: - if (!m->msg_name) { - ret = rxrpc_server_sendmsg(rx, m, len); - break; + case RXRPC_UNBOUND: + local = rxrpc_lookup_local(&rx->srx); + if (IS_ERR(local)) { + ret = PTR_ERR(local); + goto error_unlock; } - case RXRPC_SERVER_BOUND: + + rx->local = local; + rx->sk.sk_state = RXRPC_CLIENT_UNBOUND; + /* Fall through */ + + case RXRPC_CLIENT_UNBOUND: case RXRPC_CLIENT_BOUND: - if (!m->msg_name) { - ret = -ENOTCONN; - break; + if (!m->msg_name && + test_bit(RXRPC_SOCK_CONNECTED, &rx->flags)) { + m->msg_name = &rx->connect_srx; + m->msg_namelen = sizeof(rx->connect_srx); } - case RXRPC_CLIENT_CONNECTED: - ret = rxrpc_client_sendmsg(rx, trans, m, len); + case RXRPC_SERVER_BOUND: + case RXRPC_SERVER_LISTENING: + ret = rxrpc_do_sendmsg(rx, m, len); break; default: - ret = -ENOTCONN; + ret = -EINVAL; break; } -out: +error_unlock: release_sock(&rx->sk); - if (trans) - rxrpc_put_transport(trans); _leave(" = %d", ret); return ret; } @@ -523,7 +480,7 @@ static int rxrpc_setsockopt(struct socket *sock, int level, int optname, if (optlen != 0) goto error; ret = -EISCONN; - if (rx->sk.sk_state != RXRPC_UNCONNECTED) + if (rx->sk.sk_state != RXRPC_UNBOUND) goto error; set_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags); goto success; @@ -533,7 +490,7 @@ static int rxrpc_setsockopt(struct socket *sock, int level, int optname, if (rx->key) goto error; ret = -EISCONN; - if (rx->sk.sk_state != RXRPC_UNCONNECTED) + if (rx->sk.sk_state != RXRPC_UNBOUND) goto error; ret = rxrpc_request_key(rx, optval, optlen); goto error; @@ -543,7 +500,7 @@ static int rxrpc_setsockopt(struct socket *sock, int level, int optname, if (rx->key) goto error; ret = -EISCONN; - if (rx->sk.sk_state != RXRPC_UNCONNECTED) + if (rx->sk.sk_state != RXRPC_UNBOUND) goto error; ret = rxrpc_server_keyring(rx, optval, optlen); goto error; @@ -553,7 +510,7 @@ static int rxrpc_setsockopt(struct socket *sock, int level, int optname, if (optlen != sizeof(unsigned int)) goto error; ret = -EISCONN; - if (rx->sk.sk_state != RXRPC_UNCONNECTED) + if (rx->sk.sk_state != RXRPC_UNBOUND) goto error; ret = get_user(min_sec_level, (unsigned int __user *) optval); @@ -632,7 +589,7 @@ static int rxrpc_create(struct net *net, struct socket *sock, int protocol, return -ENOMEM; sock_init_data(sock, sk); - sk->sk_state = RXRPC_UNCONNECTED; + sk->sk_state = RXRPC_UNBOUND; sk->sk_write_space = rxrpc_write_space; sk->sk_max_ack_backlog = sysctl_rxrpc_max_qlen; sk->sk_destruct = rxrpc_sock_destructor; @@ -705,14 +662,6 @@ static int rxrpc_release_sock(struct sock *sk) rx->conn = NULL; } - if (rx->bundle) { - rxrpc_put_bundle(rx->trans, rx->bundle); - rx->bundle = NULL; - } - if (rx->trans) { - rxrpc_put_transport(rx->trans); - rx->trans = NULL; - } if (rx->local) { rxrpc_put_local(rx->local); rx->local = NULL; diff --git a/net/rxrpc/ar-call.c b/net/rxrpc/ar-call.c index 1fbaae1cba5f..68125dc4cb7c 100644 --- a/net/rxrpc/ar-call.c +++ b/net/rxrpc/ar-call.c @@ -195,6 +195,43 @@ struct rxrpc_call *rxrpc_find_call_hash( return ret; } +/* + * find an extant server call + * - called in process context with IRQs enabled + */ +struct rxrpc_call *rxrpc_find_call_by_user_ID(struct rxrpc_sock *rx, + unsigned long user_call_ID) +{ + struct rxrpc_call *call; + struct rb_node *p; + + _enter("%p,%lx", rx, user_call_ID); + + read_lock(&rx->call_lock); + + p = rx->calls.rb_node; + while (p) { + call = rb_entry(p, struct rxrpc_call, sock_node); + + if (user_call_ID < call->user_call_ID) + p = p->rb_left; + else if (user_call_ID > call->user_call_ID) + p = p->rb_right; + else + goto found_extant_call; + } + + read_unlock(&rx->call_lock); + _leave(" = NULL"); + return NULL; + +found_extant_call: + rxrpc_get_call(call); + read_unlock(&rx->call_lock); + _leave(" = %p [%d]", call, atomic_read(&call->usage)); + return call; +} + /* * allocate a new call */ @@ -311,51 +348,27 @@ static struct rxrpc_call *rxrpc_alloc_client_call( * set up a call for the given data * - called in process context with IRQs enabled */ -struct rxrpc_call *rxrpc_get_client_call(struct rxrpc_sock *rx, +struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *rx, struct rxrpc_transport *trans, struct rxrpc_conn_bundle *bundle, unsigned long user_call_ID, - int create, gfp_t gfp) { - struct rxrpc_call *call, *candidate; - struct rb_node *p, *parent, **pp; + struct rxrpc_call *call, *xcall; + struct rb_node *parent, **pp; - _enter("%p,%d,%d,%lx,%d", - rx, trans ? trans->debug_id : -1, bundle ? bundle->debug_id : -1, - user_call_ID, create); + _enter("%p,%d,%d,%lx", + rx, trans->debug_id, bundle ? bundle->debug_id : -1, + user_call_ID); - /* search the extant calls first for one that matches the specified - * user ID */ - read_lock(&rx->call_lock); - - p = rx->calls.rb_node; - while (p) { - call = rb_entry(p, struct rxrpc_call, sock_node); - - if (user_call_ID < call->user_call_ID) - p = p->rb_left; - else if (user_call_ID > call->user_call_ID) - p = p->rb_right; - else - goto found_extant_call; + call = rxrpc_alloc_client_call(rx, trans, bundle, gfp); + if (IS_ERR(call)) { + _leave(" = %ld", PTR_ERR(call)); + return call; } - read_unlock(&rx->call_lock); - - if (!create || !trans) - return ERR_PTR(-EBADSLT); - - /* not yet present - create a candidate for a new record and then - * redo the search */ - candidate = rxrpc_alloc_client_call(rx, trans, bundle, gfp); - if (IS_ERR(candidate)) { - _leave(" = %ld", PTR_ERR(candidate)); - return candidate; - } - - candidate->user_call_ID = user_call_ID; - __set_bit(RXRPC_CALL_HAS_USERID, &candidate->flags); + call->user_call_ID = user_call_ID; + __set_bit(RXRPC_CALL_HAS_USERID, &call->flags); write_lock(&rx->call_lock); @@ -363,19 +376,16 @@ struct rxrpc_call *rxrpc_get_client_call(struct rxrpc_sock *rx, parent = NULL; while (*pp) { parent = *pp; - call = rb_entry(parent, struct rxrpc_call, sock_node); + xcall = rb_entry(parent, struct rxrpc_call, sock_node); - if (user_call_ID < call->user_call_ID) + if (user_call_ID < xcall->user_call_ID) pp = &(*pp)->rb_left; - else if (user_call_ID > call->user_call_ID) + else if (user_call_ID > xcall->user_call_ID) pp = &(*pp)->rb_right; else - goto found_extant_second; + goto found_user_ID_now_present; } - /* second search also failed; add the new call */ - call = candidate; - candidate = NULL; rxrpc_get_call(call); rb_link_node(&call->sock_node, parent, pp); @@ -391,20 +401,16 @@ struct rxrpc_call *rxrpc_get_client_call(struct rxrpc_sock *rx, _leave(" = %p [new]", call); return call; - /* we found the call in the list immediately */ -found_extant_call: - rxrpc_get_call(call); - read_unlock(&rx->call_lock); - _leave(" = %p [extant %d]", call, atomic_read(&call->usage)); - return call; - - /* we found the call on the second time through the list */ -found_extant_second: - rxrpc_get_call(call); + /* We unexpectedly found the user ID in the list after taking + * the call_lock. This shouldn't happen unless the user races + * with itself and tries to add the same user ID twice at the + * same time in different threads. + */ +found_user_ID_now_present: write_unlock(&rx->call_lock); - rxrpc_put_call(candidate); - _leave(" = %p [second %d]", call, atomic_read(&call->usage)); - return call; + rxrpc_put_call(call); + _leave(" = -EEXIST [%p]", call); + return ERR_PTR(-EEXIST); } /* @@ -565,46 +571,6 @@ old_call: return ERR_PTR(-ECONNRESET); } -/* - * find an extant server call - * - called in process context with IRQs enabled - */ -struct rxrpc_call *rxrpc_find_server_call(struct rxrpc_sock *rx, - unsigned long user_call_ID) -{ - struct rxrpc_call *call; - struct rb_node *p; - - _enter("%p,%lx", rx, user_call_ID); - - /* search the extant calls for one that matches the specified user - * ID */ - read_lock(&rx->call_lock); - - p = rx->calls.rb_node; - while (p) { - call = rb_entry(p, struct rxrpc_call, sock_node); - - if (user_call_ID < call->user_call_ID) - p = p->rb_left; - else if (user_call_ID > call->user_call_ID) - p = p->rb_right; - else - goto found_extant_call; - } - - read_unlock(&rx->call_lock); - _leave(" = NULL"); - return NULL; - - /* we found the call in the list immediately */ -found_extant_call: - rxrpc_get_call(call); - read_unlock(&rx->call_lock); - _leave(" = %p [%d]", call, atomic_read(&call->usage)); - return call; -} - /* * detach a call from a socket and set up for release */ diff --git a/net/rxrpc/ar-connection.c b/net/rxrpc/ar-connection.c index d67b1f1b5001..8ecde4b77b55 100644 --- a/net/rxrpc/ar-connection.c +++ b/net/rxrpc/ar-connection.c @@ -80,11 +80,6 @@ struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *rx, _enter("%p{%x},%x,%hx,", rx, key_serial(key), trans->debug_id, service_id); - if (rx->trans == trans && rx->bundle) { - atomic_inc(&rx->bundle->usage); - return rx->bundle; - } - /* search the extant bundles first for one that matches the specified * user ID */ spin_lock(&trans->client_lock); @@ -138,10 +133,6 @@ struct rxrpc_conn_bundle *rxrpc_get_bundle(struct rxrpc_sock *rx, rb_insert_color(&bundle->node, &trans->bundles); spin_unlock(&trans->client_lock); _net("BUNDLE new on trans %d", trans->debug_id); - if (!rx->bundle && rx->sk.sk_state == RXRPC_CLIENT_CONNECTED) { - atomic_inc(&bundle->usage); - rx->bundle = bundle; - } _leave(" = %p [new]", bundle); return bundle; @@ -150,10 +141,6 @@ found_extant_bundle: atomic_inc(&bundle->usage); spin_unlock(&trans->client_lock); _net("BUNDLE old on trans %d", trans->debug_id); - if (!rx->bundle && rx->sk.sk_state == RXRPC_CLIENT_CONNECTED) { - atomic_inc(&bundle->usage); - rx->bundle = bundle; - } _leave(" = %p [extant %d]", bundle, atomic_read(&bundle->usage)); return bundle; @@ -163,10 +150,6 @@ found_extant_second: spin_unlock(&trans->client_lock); kfree(candidate); _net("BUNDLE old2 on trans %d", trans->debug_id); - if (!rx->bundle && rx->sk.sk_state == RXRPC_CLIENT_CONNECTED) { - atomic_inc(&bundle->usage); - rx->bundle = bundle; - } _leave(" = %p [second %d]", bundle, atomic_read(&bundle->usage)); return bundle; } diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index 18ab5c50ba87..b89dcdcbc65a 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -39,9 +39,9 @@ struct rxrpc_crypt { * sk_state for RxRPC sockets */ enum { - RXRPC_UNCONNECTED = 0, + RXRPC_UNBOUND = 0, + RXRPC_CLIENT_UNBOUND, /* Unbound socket used as client */ RXRPC_CLIENT_BOUND, /* client local address bound */ - RXRPC_CLIENT_CONNECTED, /* client is connected */ RXRPC_SERVER_BOUND, /* server local address bound */ RXRPC_SERVER_LISTENING, /* server listening for connections */ RXRPC_CLOSE, /* socket is being closed */ @@ -55,8 +55,6 @@ struct rxrpc_sock { struct sock sk; rxrpc_interceptor_t interceptor; /* kernel service Rx interceptor function */ struct rxrpc_local *local; /* local endpoint */ - struct rxrpc_transport *trans; /* transport handler */ - struct rxrpc_conn_bundle *bundle; /* virtual connection bundle */ struct rxrpc_connection *conn; /* exclusive virtual connection */ struct list_head listen_link; /* link in the local endpoint's listen list */ struct list_head secureq; /* calls awaiting connection security clearance */ @@ -65,11 +63,13 @@ struct rxrpc_sock { struct key *securities; /* list of server security descriptors */ struct rb_root calls; /* outstanding calls on this socket */ unsigned long flags; +#define RXRPC_SOCK_CONNECTED 0 /* connect_srx is set */ #define RXRPC_SOCK_EXCLUSIVE_CONN 1 /* exclusive connection for a client socket */ rwlock_t call_lock; /* lock for calls */ u32 min_sec_level; /* minimum security level */ #define RXRPC_SECURITY_MAX RXRPC_SECURITY_ENCRYPT struct sockaddr_rxrpc srx; /* local address */ + struct sockaddr_rxrpc connect_srx; /* Default client address from connect() */ sa_family_t proto; /* protocol created with */ }; @@ -477,6 +477,10 @@ extern u32 rxrpc_epoch; extern atomic_t rxrpc_debug_id; extern struct workqueue_struct *rxrpc_workqueue; +extern struct rxrpc_transport *rxrpc_name_to_transport(struct rxrpc_sock *, + struct sockaddr *, + int, int, gfp_t); + /* * ar-accept.c */ @@ -502,14 +506,14 @@ extern rwlock_t rxrpc_call_lock; struct rxrpc_call *rxrpc_find_call_hash(struct rxrpc_host_header *, void *, sa_family_t, const void *); -struct rxrpc_call *rxrpc_get_client_call(struct rxrpc_sock *, +struct rxrpc_call *rxrpc_find_call_by_user_ID(struct rxrpc_sock *, unsigned long); +struct rxrpc_call *rxrpc_new_client_call(struct rxrpc_sock *, struct rxrpc_transport *, struct rxrpc_conn_bundle *, - unsigned long, int, gfp_t); + unsigned long, gfp_t); struct rxrpc_call *rxrpc_incoming_call(struct rxrpc_sock *, struct rxrpc_connection *, struct rxrpc_host_header *); -struct rxrpc_call *rxrpc_find_server_call(struct rxrpc_sock *, unsigned long); void rxrpc_release_call(struct rxrpc_call *); void rxrpc_release_calls_on_socket(struct rxrpc_sock *); void __rxrpc_put_call(struct rxrpc_call *); @@ -581,9 +585,7 @@ int rxrpc_get_server_data_key(struct rxrpc_connection *, const void *, time_t, extern unsigned int rxrpc_resend_timeout; int rxrpc_send_packet(struct rxrpc_transport *, struct sk_buff *); -int rxrpc_client_sendmsg(struct rxrpc_sock *, struct rxrpc_transport *, - struct msghdr *, size_t); -int rxrpc_server_sendmsg(struct rxrpc_sock *, struct msghdr *, size_t); +int rxrpc_do_sendmsg(struct rxrpc_sock *, struct msghdr *, size_t); /* * ar-peer.c diff --git a/net/rxrpc/ar-output.c b/net/rxrpc/ar-output.c index ea619535f0ed..2e3c4064e29c 100644 --- a/net/rxrpc/ar-output.c +++ b/net/rxrpc/ar-output.c @@ -32,13 +32,13 @@ static int rxrpc_send_data(struct rxrpc_sock *rx, /* * extract control messages from the sendmsg() control buffer */ -static int rxrpc_sendmsg_cmsg(struct rxrpc_sock *rx, struct msghdr *msg, +static int rxrpc_sendmsg_cmsg(struct msghdr *msg, unsigned long *user_call_ID, enum rxrpc_command *command, - u32 *abort_code, - bool server) + u32 *abort_code) { struct cmsghdr *cmsg; + bool got_user_ID = false; int len; *command = RXRPC_CMD_SEND_DATA; @@ -70,6 +70,7 @@ static int rxrpc_sendmsg_cmsg(struct rxrpc_sock *rx, struct msghdr *msg, CMSG_DATA(cmsg); } _debug("User Call ID %lx", *user_call_ID); + got_user_ID = true; break; case RXRPC_ABORT: @@ -90,8 +91,6 @@ static int rxrpc_sendmsg_cmsg(struct rxrpc_sock *rx, struct msghdr *msg, *command = RXRPC_CMD_ACCEPT; if (len != 0) return -EINVAL; - if (!server) - return -EISCONN; break; default: @@ -99,6 +98,8 @@ static int rxrpc_sendmsg_cmsg(struct rxrpc_sock *rx, struct msghdr *msg, } } + if (!got_user_ID) + return -EINVAL; _leave(" = 0"); return 0; } @@ -125,56 +126,97 @@ static void rxrpc_send_abort(struct rxrpc_call *call, u32 abort_code) write_unlock_bh(&call->state_lock); } +/* + * Create a new client call for sendmsg(). + */ +static struct rxrpc_call * +rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, + unsigned long user_call_ID) +{ + struct rxrpc_conn_bundle *bundle; + struct rxrpc_transport *trans; + struct rxrpc_call *call; + struct key *key; + long ret; + + DECLARE_SOCKADDR(struct sockaddr_rxrpc *, srx, msg->msg_name); + + _enter(""); + + if (!msg->msg_name) + return ERR_PTR(-EDESTADDRREQ); + + trans = rxrpc_name_to_transport(rx, msg->msg_name, msg->msg_namelen, 0, + GFP_KERNEL); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + goto out; + } + + key = rx->key; + if (key && !rx->key->payload.data[0]) + key = NULL; + bundle = rxrpc_get_bundle(rx, trans, key, srx->srx_service, GFP_KERNEL); + if (IS_ERR(bundle)) { + ret = PTR_ERR(bundle); + goto out_trans; + } + + call = rxrpc_new_client_call(rx, trans, bundle, user_call_ID, + GFP_KERNEL); + rxrpc_put_bundle(trans, bundle); + rxrpc_put_transport(trans); + if (IS_ERR(call)) { + ret = PTR_ERR(call); + goto out_trans; + } + + _leave(" = %p\n", call); + return call; + +out_trans: + rxrpc_put_transport(trans); +out: + _leave(" = %ld", ret); + return ERR_PTR(ret); +} + /* * send a message forming part of a client call through an RxRPC socket * - caller holds the socket locked * - the socket may be either a client socket or a server socket */ -int rxrpc_client_sendmsg(struct rxrpc_sock *rx, struct rxrpc_transport *trans, - struct msghdr *msg, size_t len) +int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) { - struct rxrpc_conn_bundle *bundle; enum rxrpc_command cmd; struct rxrpc_call *call; unsigned long user_call_ID = 0; - struct key *key; - u16 service_id; u32 abort_code = 0; int ret; _enter(""); - ASSERT(trans != NULL); - - ret = rxrpc_sendmsg_cmsg(rx, msg, &user_call_ID, &cmd, &abort_code, - false); + ret = rxrpc_sendmsg_cmsg(msg, &user_call_ID, &cmd, &abort_code); if (ret < 0) return ret; - bundle = NULL; - if (trans) { - service_id = rx->srx.srx_service; - if (msg->msg_name) { - DECLARE_SOCKADDR(struct sockaddr_rxrpc *, srx, - msg->msg_name); - service_id = srx->srx_service; - } - key = rx->key; - if (key && !rx->key->payload.data[0]) - key = NULL; - bundle = rxrpc_get_bundle(rx, trans, key, service_id, - GFP_KERNEL); - if (IS_ERR(bundle)) - return PTR_ERR(bundle); + if (cmd == RXRPC_CMD_ACCEPT) { + if (rx->sk.sk_state != RXRPC_SERVER_LISTENING) + return -EINVAL; + call = rxrpc_accept_call(rx, user_call_ID); + if (IS_ERR(call)) + return PTR_ERR(call); + rxrpc_put_call(call); + return 0; } - call = rxrpc_get_client_call(rx, trans, bundle, user_call_ID, - abort_code == 0, GFP_KERNEL); - if (trans) - rxrpc_put_bundle(trans, bundle); - if (IS_ERR(call)) { - _leave(" = %ld", PTR_ERR(call)); - return PTR_ERR(call); + call = rxrpc_find_call_by_user_ID(rx, user_call_ID); + if (!call) { + if (cmd != RXRPC_CMD_SEND_DATA) + return -EBADSLT; + call = rxrpc_new_client_call_for_sendmsg(rx, msg, user_call_ID); + if (IS_ERR(call)) + return PTR_ERR(call); } _debug("CALL %d USR %lx ST %d on CONN %p", @@ -182,14 +224,21 @@ int rxrpc_client_sendmsg(struct rxrpc_sock *rx, struct rxrpc_transport *trans, if (call->state >= RXRPC_CALL_COMPLETE) { /* it's too late for this call */ - ret = -ESHUTDOWN; + ret = -ECONNRESET; } else if (cmd == RXRPC_CMD_SEND_ABORT) { rxrpc_send_abort(call, abort_code); + ret = 0; } else if (cmd != RXRPC_CMD_SEND_DATA) { ret = -EINVAL; - } else if (call->state != RXRPC_CALL_CLIENT_SEND_REQUEST) { + } else if (!call->in_clientflag && + call->state != RXRPC_CALL_CLIENT_SEND_REQUEST) { /* request phase complete for this client call */ ret = -EPROTO; + } else if (call->in_clientflag && + call->state != RXRPC_CALL_SERVER_ACK_REQUEST && + call->state != RXRPC_CALL_SERVER_SEND_REPLY) { + /* Reply phase not begun or not complete for service call. */ + ret = -EPROTO; } else { ret = rxrpc_send_data(rx, call, msg, len); } @@ -267,67 +316,6 @@ void rxrpc_kernel_abort_call(struct rxrpc_call *call, u32 abort_code) EXPORT_SYMBOL(rxrpc_kernel_abort_call); -/* - * send a message through a server socket - * - caller holds the socket locked - */ -int rxrpc_server_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) -{ - enum rxrpc_command cmd; - struct rxrpc_call *call; - unsigned long user_call_ID = 0; - u32 abort_code = 0; - int ret; - - _enter(""); - - ret = rxrpc_sendmsg_cmsg(rx, msg, &user_call_ID, &cmd, &abort_code, - true); - if (ret < 0) - return ret; - - if (cmd == RXRPC_CMD_ACCEPT) { - call = rxrpc_accept_call(rx, user_call_ID); - if (IS_ERR(call)) - return PTR_ERR(call); - rxrpc_put_call(call); - return 0; - } - - call = rxrpc_find_server_call(rx, user_call_ID); - if (!call) - return -EBADSLT; - if (call->state >= RXRPC_CALL_COMPLETE) { - ret = -ESHUTDOWN; - goto out; - } - - switch (cmd) { - case RXRPC_CMD_SEND_DATA: - if (call->state != RXRPC_CALL_CLIENT_SEND_REQUEST && - call->state != RXRPC_CALL_SERVER_ACK_REQUEST && - call->state != RXRPC_CALL_SERVER_SEND_REPLY) { - /* Tx phase not yet begun for this call */ - ret = -EPROTO; - break; - } - - ret = rxrpc_send_data(rx, call, msg, len); - break; - - case RXRPC_CMD_SEND_ABORT: - rxrpc_send_abort(call, abort_code); - break; - default: - BUG(); - } - - out: - rxrpc_put_call(call); - _leave(" = %d", ret); - return ret; -} - /* * send a packet through the transport endpoint */ -- cgit v1.2.3 From e434863718d4b99dd0d6e0cefd3c5e79e4fa2083 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Thu, 9 Jun 2016 10:21:00 -0700 Subject: net: vrf: Fix crash when IPv6 is disabled at boot time Frank Kellermann reported a kernel crash with 4.5.0 when IPv6 is disabled at boot using the kernel option ipv6.disable=1. Using current net-next with the boot option: $ ip link add red type vrf table 1001 Generates: [12210.919584] BUG: unable to handle kernel NULL pointer dereference at 0000000000000748 [12210.921341] IP: [] fib6_get_table+0x2c/0x5a [12210.922537] PGD b79e3067 PUD bb32b067 PMD 0 [12210.923479] Oops: 0000 [#1] SMP [12210.924001] Modules linked in: ipvlan 8021q garp mrp stp llc [12210.925130] CPU: 3 PID: 1177 Comm: ip Not tainted 4.7.0-rc1+ #235 [12210.926168] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.7.5-20140531_083030-gandalf 04/01/2014 [12210.928065] task: ffff8800b9ac4640 ti: ffff8800bacac000 task.ti: ffff8800bacac000 [12210.929328] RIP: 0010:[] [] fib6_get_table+0x2c/0x5a [12210.930697] RSP: 0018:ffff8800bacaf888 EFLAGS: 00010202 [12210.931563] RAX: 0000000000000748 RBX: ffffffff81a9e280 RCX: ffff8800b9ac4e28 [12210.932688] RDX: 00000000000000e9 RSI: 0000000000000002 RDI: 0000000000000286 [12210.933820] RBP: ffff8800bacaf898 R08: ffff8800b9ac4df0 R09: 000000000052001b [12210.934941] R10: 00000000657c0000 R11: 000000000000c649 R12: 00000000000003e9 [12210.936032] R13: 00000000000003e9 R14: ffff8800bace7800 R15: ffff8800bb3ec000 [12210.937103] FS: 00007faa1766c700(0000) GS:ffff88013ac00000(0000) knlGS:0000000000000000 [12210.938321] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [12210.939166] CR2: 0000000000000748 CR3: 00000000b79d6000 CR4: 00000000000406e0 [12210.940278] Stack: [12210.940603] ffff8800bb3ec000 ffffffff81a9e280 ffff8800bacaf8c8 ffffffff814b3135 [12210.941818] ffff8800bb3ec000 ffffffff81a9e280 ffffffff81a9e280 ffff8800bace7800 [12210.943040] ffff8800bacaf8f0 ffffffff81397c88 ffff8800bb3ec000 ffffffff81a9e280 [12210.944288] Call Trace: [12210.944688] [] fib6_new_table+0x24/0x8a [12210.945516] [] vrf_dev_init+0xd4/0x162 [12210.946328] [] register_netdevice+0x100/0x396 [12210.947209] [] vrf_newlink+0x40/0xb3 [12210.948001] [] rtnl_newlink+0x5d3/0x6d5 ... The problem above is due to the fact that the fib hash table is not allocated when IPv6 is disabled at boot. As for the VRF driver it should not do any IPv6 initializations if IPv6 is disabled, so it needs to know if IPv6 is disabled at boot. The disable parameter is private to the IPv6 module, so provide an accessor for modules to determine if IPv6 was disabled at boot time. Fixes: 35402e3136634 ("net: Add IPv6 support to VRF device") Signed-off-by: David Ahern Signed-off-by: David S. Miller --- drivers/net/vrf.c | 7 +++++++ include/linux/ipv6.h | 7 +++++++ net/ipv6/af_inet6.c | 6 ++++++ 3 files changed, 20 insertions(+) (limited to 'include/linux') diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c index b82e3527924e..b4d746943bc5 100644 --- a/drivers/net/vrf.c +++ b/drivers/net/vrf.c @@ -407,6 +407,10 @@ static int vrf_rt6_create(struct net_device *dev) struct rt6_info *rt6, *rt6_local; int rc = -ENOMEM; + /* IPv6 can be CONFIG enabled and then disabled runtime */ + if (!ipv6_mod_enabled()) + return 0; + rt6i_table = fib6_new_table(net, vrf->tb_id); if (!rt6i_table) goto out; @@ -919,6 +923,9 @@ static int vrf_fib_rule(const struct net_device *dev, __u8 family, bool add_it) struct sk_buff *skb; int err; + if (family == AF_INET6 && !ipv6_mod_enabled()) + return 0; + skb = nlmsg_new(vrf_fib_rule_nl_size(), GFP_KERNEL); if (!skb) return -ENOMEM; diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h index 5c91b0b055d4..c6dbcd84a2c7 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -283,6 +283,8 @@ struct tcp6_timewait_sock { }; #if IS_ENABLED(CONFIG_IPV6) +bool ipv6_mod_enabled(void); + static inline struct ipv6_pinfo *inet6_sk(const struct sock *__sk) { return sk_fullsock(__sk) ? inet_sk(__sk)->pinet6 : NULL; @@ -326,6 +328,11 @@ static inline int inet_v6_ipv6only(const struct sock *sk) #define ipv6_only_sock(sk) 0 #define ipv6_sk_rxinfo(sk) 0 +static inline bool ipv6_mod_enabled(void) +{ + return false; +} + static inline struct ipv6_pinfo * inet6_sk(const struct sock *__sk) { return NULL; diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index bfa86f040c16..2076c21107d0 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -92,6 +92,12 @@ MODULE_PARM_DESC(disable_ipv6, "Disable IPv6 on all interfaces"); module_param_named(autoconf, ipv6_defaults.autoconf, int, 0444); MODULE_PARM_DESC(autoconf, "Enable IPv6 address autoconfiguration on all interfaces"); +bool ipv6_mod_enabled(void) +{ + return disable_ipv6_mod == 0; +} +EXPORT_SYMBOL_GPL(ipv6_mod_enabled); + static __inline__ struct ipv6_pinfo *inet6_sk_generic(struct sock *sk) { const int offset = sk->sk_prot->obj_size - sizeof(struct ipv6_pinfo); -- cgit v1.2.3 From 7486216b3a0bd26375b17b2cc168a311106cea70 Mon Sep 17 00:00:00 2001 From: Saeed Mahameed Date: Thu, 9 Jun 2016 15:11:34 +0300 Subject: {net,IB}/mlx5: mlx5_ifc updates Introducing mlx5_ifc updates for upcoming ConnectX-4 features. Needed bits and hardware structures for mlx5e netdev: - MLX5_CQ_PERIOD_NUM_MODES for adaptive moderation support - QoS rate limiting - SQ context rate limiting - Auto negotiation fields in PTYS register - Source SQN field in flow table entry match structure - DCBX parameters Needed bits and hardware structures for IB: - New XRQ opcodes, commands and capabilities layout - Extend q counters definition to support IB. Signed-off-by: Saeed Mahameed Signed-off-by: Leon Romanovsky Signed-off-by: Leon Romanovsky --- include/linux/mlx5/mlx5_ifc.h | 275 ++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 263 insertions(+), 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 9a05cd7e5890..209add93159f 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -123,6 +123,10 @@ enum { MLX5_CMD_OP_DRAIN_DCT = 0x712, MLX5_CMD_OP_QUERY_DCT = 0x713, MLX5_CMD_OP_ARM_DCT_FOR_KEY_VIOLATION = 0x714, + MLX5_CMD_OP_CREATE_XRQ = 0x717, + MLX5_CMD_OP_DESTROY_XRQ = 0x718, + MLX5_CMD_OP_QUERY_XRQ = 0x719, + MLX5_CMD_OP_ARM_XRQ = 0x71a, MLX5_CMD_OP_QUERY_VPORT_STATE = 0x750, MLX5_CMD_OP_MODIFY_VPORT_STATE = 0x751, MLX5_CMD_OP_QUERY_ESW_VPORT_CONTEXT = 0x752, @@ -139,6 +143,8 @@ enum { MLX5_CMD_OP_ALLOC_Q_COUNTER = 0x771, MLX5_CMD_OP_DEALLOC_Q_COUNTER = 0x772, MLX5_CMD_OP_QUERY_Q_COUNTER = 0x773, + MLX5_CMD_OP_SET_RATE_LIMIT = 0x780, + MLX5_CMD_OP_QUERY_RATE_LIMIT = 0x781, MLX5_CMD_OP_ALLOC_PD = 0x800, MLX5_CMD_OP_DEALLOC_PD = 0x801, MLX5_CMD_OP_ALLOC_UAR = 0x802, @@ -361,7 +367,8 @@ struct mlx5_ifc_fte_match_set_lyr_2_4_bits { }; struct mlx5_ifc_fte_match_set_misc_bits { - u8 reserved_at_0[0x20]; + u8 reserved_at_0[0x8]; + u8 source_sqn[0x18]; u8 reserved_at_20[0x10]; u8 source_port[0x10]; @@ -505,6 +512,17 @@ struct mlx5_ifc_e_switch_cap_bits { u8 reserved_at_20[0x7e0]; }; +struct mlx5_ifc_qos_cap_bits { + u8 packet_pacing[0x1]; + u8 reserved_0[0x1f]; + u8 reserved_1[0x20]; + u8 packet_pacing_max_rate[0x20]; + u8 packet_pacing_min_rate[0x20]; + u8 reserved_2[0x10]; + u8 packet_pacing_rate_table_size[0x10]; + u8 reserved_3[0x760]; +}; + struct mlx5_ifc_per_protocol_networking_offload_caps_bits { u8 csum_cap[0x1]; u8 vlan_cap[0x1]; @@ -744,7 +762,8 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 out_of_seq_cnt[0x1]; u8 vport_counters[0x1]; - u8 reserved_at_182[0x4]; + u8 retransmission_q_counters[0x1]; + u8 reserved_at_183[0x3]; u8 max_qp_cnt[0xa]; u8 pkey_table_size[0x10]; @@ -771,7 +790,9 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 log_max_msg[0x5]; u8 reserved_at_1c8[0x4]; u8 max_tc[0x4]; - u8 reserved_at_1d0[0x6]; + u8 reserved_at_1d0[0x1]; + u8 dcbx[0x1]; + u8 reserved_at_1d2[0x4]; u8 rol_s[0x1]; u8 rol_g[0x1]; u8 reserved_at_1d8[0x1]; @@ -803,7 +824,7 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 tph[0x1]; u8 rf[0x1]; u8 dct[0x1]; - u8 reserved_at_21b[0x1]; + u8 qos[0x1]; u8 eth_net_offloads[0x1]; u8 roce[0x1]; u8 atomic[0x1]; @@ -929,7 +950,15 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 cqe_compression_timeout[0x10]; u8 cqe_compression_max_num[0x10]; - u8 reserved_at_5e0[0x220]; + u8 reserved_at_5e0[0x10]; + u8 tag_matching[0x1]; + u8 rndv_offload_rc[0x1]; + u8 rndv_offload_dc[0x1]; + u8 log_tag_matching_list_sz[0x5]; + u8 reserved_at_5e8[0x3]; + u8 log_max_xrq[0x5]; + + u8 reserved_at_5f0[0x200]; }; enum mlx5_flow_destination_type { @@ -1967,7 +1996,7 @@ struct mlx5_ifc_qpc_bits { u8 reserved_at_560[0x5]; u8 rq_type[0x3]; - u8 srqn_rmpn[0x18]; + u8 srqn_rmpn_xrqn[0x18]; u8 reserved_at_580[0x8]; u8 rmsn[0x18]; @@ -2018,6 +2047,7 @@ union mlx5_ifc_hca_cap_union_bits { struct mlx5_ifc_flow_table_eswitch_cap_bits flow_table_eswitch_cap; struct mlx5_ifc_e_switch_cap_bits e_switch_cap; struct mlx5_ifc_vector_calc_cap_bits vector_calc_cap; + struct mlx5_ifc_qos_cap_bits qos_cap; u8 reserved_at_0[0x8000]; }; @@ -2244,8 +2274,9 @@ struct mlx5_ifc_sqc_bits { u8 reserved_at_40[0x8]; u8 cqn[0x18]; - u8 reserved_at_60[0xa0]; + u8 reserved_at_60[0x90]; + u8 packet_pacing_rate_limit_index[0x10]; u8 tis_lst_sz[0x10]; u8 reserved_at_110[0x10]; @@ -2593,7 +2624,7 @@ struct mlx5_ifc_dctc_bits { u8 reserved_at_98[0x8]; u8 reserved_at_a0[0x8]; - u8 srqn[0x18]; + u8 srqn_xrqn[0x18]; u8 reserved_at_c0[0x8]; u8 pd[0x18]; @@ -2645,6 +2676,7 @@ enum { enum { MLX5_CQ_PERIOD_MODE_START_FROM_EQE = 0x0, MLX5_CQ_PERIOD_MODE_START_FROM_CQE = 0x1, + MLX5_CQ_PERIOD_NUM_MODES }; struct mlx5_ifc_cqc_bits { @@ -2722,6 +2754,54 @@ struct mlx5_ifc_query_adapter_param_block_bits { u8 vsd_contd_psid[16][0x8]; }; +enum { + MLX5_XRQC_STATE_GOOD = 0x0, + MLX5_XRQC_STATE_ERROR = 0x1, +}; + +enum { + MLX5_XRQC_TOPOLOGY_NO_SPECIAL_TOPOLOGY = 0x0, + MLX5_XRQC_TOPOLOGY_TAG_MATCHING = 0x1, +}; + +enum { + MLX5_XRQC_OFFLOAD_RNDV = 0x1, +}; + +struct mlx5_ifc_tag_matching_topology_context_bits { + u8 log_matching_list_sz[0x4]; + u8 reserved_at_4[0xc]; + u8 append_next_index[0x10]; + + u8 sw_phase_cnt[0x10]; + u8 hw_phase_cnt[0x10]; + + u8 reserved_at_40[0x40]; +}; + +struct mlx5_ifc_xrqc_bits { + u8 state[0x4]; + u8 rlkey[0x1]; + u8 reserved_at_5[0xf]; + u8 topology[0x4]; + u8 reserved_at_18[0x4]; + u8 offload[0x4]; + + u8 reserved_at_20[0x8]; + u8 user_index[0x18]; + + u8 reserved_at_40[0x8]; + u8 cqn[0x18]; + + u8 reserved_at_60[0xa0]; + + struct mlx5_ifc_tag_matching_topology_context_bits tag_matching_topology_context; + + u8 reserved_at_180[0x180]; + + struct mlx5_ifc_wq_bits wq; +}; + union mlx5_ifc_modify_field_select_resize_field_select_auto_bits { struct mlx5_ifc_modify_field_select_bits modify_field_select; struct mlx5_ifc_resize_field_select_bits resize_field_select; @@ -3144,6 +3224,30 @@ struct mlx5_ifc_rst2init_qp_in_bits { u8 reserved_at_800[0x80]; }; +struct mlx5_ifc_query_xrq_out_bits { + u8 status[0x8]; + u8 reserved_at_8[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_at_40[0x40]; + + struct mlx5_ifc_xrqc_bits xrq_context; +}; + +struct mlx5_ifc_query_xrq_in_bits { + u8 opcode[0x10]; + u8 reserved_at_10[0x10]; + + u8 reserved_at_20[0x10]; + u8 op_mod[0x10]; + + u8 reserved_at_40[0x8]; + u8 xrqn[0x18]; + + u8 reserved_at_60[0x20]; +}; + struct mlx5_ifc_query_xrc_srq_out_bits { u8 status[0x8]; u8 reserved_at_8[0x18]; @@ -3547,7 +3651,27 @@ struct mlx5_ifc_query_q_counter_out_bits { u8 out_of_sequence[0x20]; - u8 reserved_at_1e0[0x620]; + u8 reserved_at_1e0[0x20]; + + u8 duplicate_request[0x20]; + + u8 reserved_at_220[0x20]; + + u8 rnr_nak_retry_err[0x20]; + + u8 reserved_at_260[0x20]; + + u8 packet_seq_err[0x20]; + + u8 reserved_at_2a0[0x20]; + + u8 implied_nak_seq_err[0x20]; + + u8 reserved_at_2e0[0x20]; + + u8 local_ack_timeout_err[0x20]; + + u8 reserved_at_320[0x4e0]; }; struct mlx5_ifc_query_q_counter_in_bits { @@ -4998,6 +5122,28 @@ struct mlx5_ifc_detach_from_mcg_in_bits { u8 multicast_gid[16][0x8]; }; +struct mlx5_ifc_destroy_xrq_out_bits { + u8 status[0x8]; + u8 reserved_at_8[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_at_40[0x40]; +}; + +struct mlx5_ifc_destroy_xrq_in_bits { + u8 opcode[0x10]; + u8 reserved_at_10[0x10]; + + u8 reserved_at_20[0x10]; + u8 op_mod[0x10]; + + u8 reserved_at_40[0x8]; + u8 xrqn[0x18]; + + u8 reserved_at_60[0x20]; +}; + struct mlx5_ifc_destroy_xrc_srq_out_bits { u8 status[0x8]; u8 reserved_at_8[0x18]; @@ -5583,6 +5729,30 @@ struct mlx5_ifc_dealloc_flow_counter_in_bits { u8 reserved_at_60[0x20]; }; +struct mlx5_ifc_create_xrq_out_bits { + u8 status[0x8]; + u8 reserved_at_8[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_at_40[0x8]; + u8 xrqn[0x18]; + + u8 reserved_at_60[0x20]; +}; + +struct mlx5_ifc_create_xrq_in_bits { + u8 opcode[0x10]; + u8 reserved_at_10[0x10]; + + u8 reserved_at_20[0x10]; + u8 op_mod[0x10]; + + u8 reserved_at_40[0x40]; + + struct mlx5_ifc_xrqc_bits xrq_context; +}; + struct mlx5_ifc_create_xrc_srq_out_bits { u8 status[0x8]; u8 reserved_at_8[0x18]; @@ -6124,6 +6294,29 @@ struct mlx5_ifc_attach_to_mcg_in_bits { u8 multicast_gid[16][0x8]; }; +struct mlx5_ifc_arm_xrq_out_bits { + u8 status[0x8]; + u8 reserved_at_8[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_at_40[0x40]; +}; + +struct mlx5_ifc_arm_xrq_in_bits { + u8 opcode[0x10]; + u8 reserved_at_10[0x10]; + + u8 reserved_at_20[0x10]; + u8 op_mod[0x10]; + + u8 reserved_at_40[0x8]; + u8 xrqn[0x18]; + + u8 reserved_at_60[0x10]; + u8 lwm[0x10]; +}; + struct mlx5_ifc_arm_xrc_srq_out_bits { u8 status[0x8]; u8 reserved_at_8[0x18]; @@ -6161,7 +6354,8 @@ struct mlx5_ifc_arm_rq_out_bits { }; enum { - MLX5_ARM_RQ_IN_OP_MOD_SRQ_ = 0x1, + MLX5_ARM_RQ_IN_OP_MOD_SRQ = 0x1, + MLX5_ARM_RQ_IN_OP_MOD_XRQ = 0x2, }; struct mlx5_ifc_arm_rq_in_bits { @@ -6354,6 +6548,30 @@ struct mlx5_ifc_add_vxlan_udp_dport_in_bits { u8 vxlan_udp_port[0x10]; }; +struct mlx5_ifc_set_rate_limit_out_bits { + u8 status[0x8]; + u8 reserved_at_8[0x18]; + + u8 syndrome[0x20]; + + u8 reserved_at_40[0x40]; +}; + +struct mlx5_ifc_set_rate_limit_in_bits { + u8 opcode[0x10]; + u8 reserved_at_10[0x10]; + + u8 reserved_at_20[0x10]; + u8 op_mod[0x10]; + + u8 reserved_at_40[0x10]; + u8 rate_limit_index[0x10]; + + u8 reserved_at_60[0x20]; + + u8 rate_limit[0x20]; +}; + struct mlx5_ifc_access_register_out_bits { u8 status[0x8]; u8 reserved_at_8[0x18]; @@ -6478,12 +6696,15 @@ struct mlx5_ifc_pude_reg_bits { }; struct mlx5_ifc_ptys_reg_bits { - u8 reserved_at_0[0x8]; + u8 an_disable_cap[0x1]; + u8 an_disable_admin[0x1]; + u8 reserved_at_2[0x6]; u8 local_port[0x8]; u8 reserved_at_10[0xd]; u8 proto_mask[0x3]; - u8 reserved_at_20[0x40]; + u8 an_status[0x4]; + u8 reserved_at_24[0x3c]; u8 eth_proto_capability[0x20]; @@ -7444,4 +7665,34 @@ struct mlx5_ifc_mcia_reg_bits { u8 dword_11[0x20]; }; +struct mlx5_ifc_dcbx_param_bits { + u8 dcbx_cee_cap[0x1]; + u8 dcbx_ieee_cap[0x1]; + u8 dcbx_standby_cap[0x1]; + u8 reserved_at_0[0x5]; + u8 port_number[0x8]; + u8 reserved_at_10[0xa]; + u8 max_application_table_size[6]; + u8 reserved_at_20[0x15]; + u8 version_oper[0x3]; + u8 reserved_at_38[5]; + u8 version_admin[0x3]; + u8 willing_admin[0x1]; + u8 reserved_at_41[0x3]; + u8 pfc_cap_oper[0x4]; + u8 reserved_at_48[0x4]; + u8 pfc_cap_admin[0x4]; + u8 reserved_at_50[0x4]; + u8 num_of_tc_oper[0x4]; + u8 reserved_at_58[0x4]; + u8 num_of_tc_admin[0x4]; + u8 remote_willing[0x1]; + u8 reserved_at_61[3]; + u8 remote_pfc_cap[4]; + u8 reserved_at_68[0x14]; + u8 remote_num_of_tc[0x4]; + u8 reserved_at_80[0x18]; + u8 error[0x8]; + u8 reserved_at_a0[0x160]; +}; #endif /* MLX5_IFC_H */ -- cgit v1.2.3 From a70b506efe899dc8d650eafcc0b11fc9ee746627 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 10 Jun 2016 21:19:06 +0200 Subject: bpf: enforce recursion limit on redirects Respect the stack's xmit_recursion limit for calls into dev_queue_xmit(). Currently, they are not handeled by the limiter when attached to clsact's egress parent, for example, and a buggy program redirecting it to the same device again could run into stack overflow eventually. It would be good if we could notify an admin to give him a chance to react. We reuse xmit_recursion instead of having one private to eBPF, so that the stack's current recursion depth will be taken into account as well. Follow-up to commit 3896d655f4d4 ("bpf: introduce bpf_clone_redirect() helper") and 27b29f63058d ("bpf: add bpf_redirect() helper"). Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 ++ net/core/dev.c | 6 ++---- net/core/filter.c | 55 +++++++++++++++++++++++++++++------------------ 3 files changed, 38 insertions(+), 25 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 4f234b102892..94eef356a65f 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2389,6 +2389,8 @@ void synchronize_net(void); int init_dummy_netdev(struct net_device *dev); DECLARE_PER_CPU(int, xmit_recursion); +#define XMIT_RECURSION_LIMIT 10 + static inline int dev_recursion_level(void) { return this_cpu_read(xmit_recursion); diff --git a/net/core/dev.c b/net/core/dev.c index c43c9d2a88cf..b14835757141 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3144,8 +3144,6 @@ static void skb_update_prio(struct sk_buff *skb) DEFINE_PER_CPU(int, xmit_recursion); EXPORT_SYMBOL(xmit_recursion); -#define RECURSION_LIMIT 10 - /** * dev_loopback_xmit - loop back @skb * @net: network namespace this loopback is happening in @@ -3388,8 +3386,8 @@ static int __dev_queue_xmit(struct sk_buff *skb, void *accel_priv) int cpu = smp_processor_id(); /* ok because BHs are off */ if (txq->xmit_lock_owner != cpu) { - - if (__this_cpu_read(xmit_recursion) > RECURSION_LIMIT) + if (unlikely(__this_cpu_read(xmit_recursion) > + XMIT_RECURSION_LIMIT)) goto recursion_alert; skb = validate_xmit_skb(skb, dev); diff --git a/net/core/filter.c b/net/core/filter.c index 68adb5f52110..d11744d10e00 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1603,9 +1603,36 @@ static const struct bpf_func_proto bpf_csum_diff_proto = { .arg5_type = ARG_ANYTHING, }; +static inline int __bpf_rx_skb(struct net_device *dev, struct sk_buff *skb) +{ + if (skb_at_tc_ingress(skb)) + skb_postpush_rcsum(skb, skb_mac_header(skb), skb->mac_len); + + return dev_forward_skb(dev, skb); +} + +static inline int __bpf_tx_skb(struct net_device *dev, struct sk_buff *skb) +{ + int ret; + + if (unlikely(__this_cpu_read(xmit_recursion) > XMIT_RECURSION_LIMIT)) { + net_crit_ratelimited("bpf: recursion limit reached on datapath, buggy bpf program?\n"); + kfree_skb(skb); + return -ENETDOWN; + } + + skb->dev = dev; + + __this_cpu_inc(xmit_recursion); + ret = dev_queue_xmit(skb); + __this_cpu_dec(xmit_recursion); + + return ret; +} + static u64 bpf_clone_redirect(u64 r1, u64 ifindex, u64 flags, u64 r4, u64 r5) { - struct sk_buff *skb = (struct sk_buff *) (long) r1, *skb2; + struct sk_buff *skb = (struct sk_buff *) (long) r1; struct net_device *dev; if (unlikely(flags & ~(BPF_F_INGRESS))) @@ -1615,19 +1642,12 @@ static u64 bpf_clone_redirect(u64 r1, u64 ifindex, u64 flags, u64 r4, u64 r5) if (unlikely(!dev)) return -EINVAL; - skb2 = skb_clone(skb, GFP_ATOMIC); - if (unlikely(!skb2)) + skb = skb_clone(skb, GFP_ATOMIC); + if (unlikely(!skb)) return -ENOMEM; - if (flags & BPF_F_INGRESS) { - if (skb_at_tc_ingress(skb2)) - skb_postpush_rcsum(skb2, skb_mac_header(skb2), - skb2->mac_len); - return dev_forward_skb(dev, skb2); - } - - skb2->dev = dev; - return dev_queue_xmit(skb2); + return flags & BPF_F_INGRESS ? + __bpf_rx_skb(dev, skb) : __bpf_tx_skb(dev, skb); } static const struct bpf_func_proto bpf_clone_redirect_proto = { @@ -1671,15 +1691,8 @@ int skb_do_redirect(struct sk_buff *skb) return -EINVAL; } - if (ri->flags & BPF_F_INGRESS) { - if (skb_at_tc_ingress(skb)) - skb_postpush_rcsum(skb, skb_mac_header(skb), - skb->mac_len); - return dev_forward_skb(dev, skb); - } - - skb->dev = dev; - return dev_queue_xmit(skb); + return ri->flags & BPF_F_INGRESS ? + __bpf_rx_skb(dev, skb) : __bpf_tx_skb(dev, skb); } static const struct bpf_func_proto bpf_redirect_proto = { -- cgit v1.2.3 From fd2a0437dc33b6425cabf74cc7fc7fdba6d5903b Mon Sep 17 00:00:00 2001 From: Mike Rapoport Date: Wed, 8 Jun 2016 16:09:18 +0300 Subject: virtio_net: introduce virtio_net_hdr_{from,to}_skb The code for conversion between virtio_net_hdr and skb GSO info is duplicated at several places. Let's put it to a common place to allow reuse. Signed-off-by: Mike Rapoport Signed-off-by: David S. Miller --- include/linux/virtio_net.h | 101 +++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 101 insertions(+) create mode 100644 include/linux/virtio_net.h (limited to 'include/linux') diff --git a/include/linux/virtio_net.h b/include/linux/virtio_net.h new file mode 100644 index 000000000000..1c912f85e041 --- /dev/null +++ b/include/linux/virtio_net.h @@ -0,0 +1,101 @@ +#ifndef _LINUX_VIRTIO_NET_H +#define _LINUX_VIRTIO_NET_H + +#include +#include + +static inline int virtio_net_hdr_to_skb(struct sk_buff *skb, + const struct virtio_net_hdr *hdr, + bool little_endian) +{ + unsigned short gso_type = 0; + + if (hdr->gso_type != VIRTIO_NET_HDR_GSO_NONE) { + switch (hdr->gso_type & ~VIRTIO_NET_HDR_GSO_ECN) { + case VIRTIO_NET_HDR_GSO_TCPV4: + gso_type = SKB_GSO_TCPV4; + break; + case VIRTIO_NET_HDR_GSO_TCPV6: + gso_type = SKB_GSO_TCPV6; + break; + case VIRTIO_NET_HDR_GSO_UDP: + gso_type = SKB_GSO_UDP; + break; + default: + return -EINVAL; + } + + if (hdr->gso_type & VIRTIO_NET_HDR_GSO_ECN) + gso_type |= SKB_GSO_TCP_ECN; + + if (hdr->gso_size == 0) + return -EINVAL; + } + + if (hdr->flags & VIRTIO_NET_HDR_F_NEEDS_CSUM) { + u16 start = __virtio16_to_cpu(little_endian, hdr->csum_start); + u16 off = __virtio16_to_cpu(little_endian, hdr->csum_offset); + + if (!skb_partial_csum_set(skb, start, off)) + return -EINVAL; + } + + if (hdr->gso_type != VIRTIO_NET_HDR_GSO_NONE) { + u16 gso_size = __virtio16_to_cpu(little_endian, hdr->gso_size); + + skb_shinfo(skb)->gso_size = gso_size; + skb_shinfo(skb)->gso_type = gso_type; + + /* Header must be checked, and gso_segs computed. */ + skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; + skb_shinfo(skb)->gso_segs = 0; + } + + return 0; +} + +static inline int virtio_net_hdr_from_skb(const struct sk_buff *skb, + struct virtio_net_hdr *hdr, + bool little_endian) +{ + memset(hdr, 0, sizeof(*hdr)); + + if (skb_is_gso(skb)) { + struct skb_shared_info *sinfo = skb_shinfo(skb); + + /* This is a hint as to how much should be linear. */ + hdr->hdr_len = __cpu_to_virtio16(little_endian, + skb_headlen(skb)); + hdr->gso_size = __cpu_to_virtio16(little_endian, + sinfo->gso_size); + if (sinfo->gso_type & SKB_GSO_TCPV4) + hdr->gso_type = VIRTIO_NET_HDR_GSO_TCPV4; + else if (sinfo->gso_type & SKB_GSO_TCPV6) + hdr->gso_type = VIRTIO_NET_HDR_GSO_TCPV6; + else if (sinfo->gso_type & SKB_GSO_UDP) + hdr->gso_type = VIRTIO_NET_HDR_GSO_UDP; + else + return -EINVAL; + if (sinfo->gso_type & SKB_GSO_TCP_ECN) + hdr->gso_type |= VIRTIO_NET_HDR_GSO_ECN; + } else + hdr->gso_type = VIRTIO_NET_HDR_GSO_NONE; + + if (skb->ip_summed == CHECKSUM_PARTIAL) { + hdr->flags = VIRTIO_NET_HDR_F_NEEDS_CSUM; + if (skb_vlan_tag_present(skb)) + hdr->csum_start = __cpu_to_virtio16(little_endian, + skb_checksum_start_offset(skb) + VLAN_HLEN); + else + hdr->csum_start = __cpu_to_virtio16(little_endian, + skb_checksum_start_offset(skb)); + hdr->csum_offset = __cpu_to_virtio16(little_endian, + skb->csum_offset); + } else if (skb->ip_summed == CHECKSUM_UNNECESSARY) { + hdr->flags = VIRTIO_NET_HDR_F_DATA_VALID; + } /* else everything is zero */ + + return 0; +} + +#endif /* _LINUX_VIRTIO_BYTEORDER */ -- cgit v1.2.3 From 002245cc6407c8ff65b8024554080eb6de1a8e2c Mon Sep 17 00:00:00 2001 From: Zi Shen Lim Date: Wed, 8 Jun 2016 21:18:47 -0700 Subject: bpf: fix missing header inclusion Commit 0fc174dea545 ("ebpf: make internal bpf API independent of CONFIG_BPF_SYSCALL ifdefs") introduced usage of ERR_PTR() in bpf_prog_get(), however did not include linux/err.h. Without this patch, when compiling arm64 BPF without CONFIG_BPF_SYSCALL: ... In file included from arch/arm64/net/bpf_jit_comp.c:21:0: include/linux/bpf.h: In function 'bpf_prog_get': include/linux/bpf.h:235:9: error: implicit declaration of function 'ERR_PTR' [-Werror=implicit-function-declaration] return ERR_PTR(-EOPNOTSUPP); ^ include/linux/bpf.h:235:9: warning: return makes pointer from integer without a cast [-Wint-conversion] In file included from include/linux/rwsem.h:17:0, from include/linux/mm_types.h:10, from include/linux/sched.h:27, from arch/arm64/include/asm/compat.h:25, from arch/arm64/include/asm/stat.h:23, from include/linux/stat.h:5, from include/linux/compat.h:12, from include/linux/filter.h:10, from arch/arm64/net/bpf_jit_comp.c:22: include/linux/err.h: At top level: include/linux/err.h:23:35: error: conflicting types for 'ERR_PTR' static inline void * __must_check ERR_PTR(long error) ^ In file included from arch/arm64/net/bpf_jit_comp.c:21:0: include/linux/bpf.h:235:9: note: previous implicit declaration of 'ERR_PTR' was here return ERR_PTR(-EOPNOTSUPP); ^ ... Fixes: 0fc174dea545 ("ebpf: make internal bpf API independent of CONFIG_BPF_SYSCALL ifdefs") Suggested-by: Daniel Borkmann Signed-off-by: Zi Shen Lim Acked-by: Daniel Borkmann Signed-off-by: David S. Miller --- include/linux/bpf.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 8ee27b8afe81..1bcae82c6cb1 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -11,6 +11,7 @@ #include #include #include +#include struct bpf_map; -- cgit v1.2.3 From f20e6657a8758fe8d074889a6f1883674f01c7f2 Mon Sep 17 00:00:00 2001 From: Pramod Kumar Date: Fri, 10 Jun 2016 11:03:45 +0530 Subject: mdio: mux: Enhanced MDIO mux framework for integrated multiplexers An integrated multiplexer uses same address space for "muxed bus selection" and "generation of mdio transaction" hence its good to register parent bus from mux driver. Hence added a mechanism where mux driver could register a parent bus and pass it down to framework via mdio_mux_init api. Signed-off-by: Pramod Kumar Reviewed-by: Andrew Lunn Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/net/phy/mdio-mux-gpio.c | 2 +- drivers/net/phy/mdio-mux-mmioreg.c | 2 +- drivers/net/phy/mdio-mux.c | 28 ++++++++++++++++++---------- include/linux/mdio-mux.h | 4 +++- 4 files changed, 23 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/phy/mdio-mux-gpio.c b/drivers/net/phy/mdio-mux-gpio.c index 7ddb1ab70891..919949960a10 100644 --- a/drivers/net/phy/mdio-mux-gpio.c +++ b/drivers/net/phy/mdio-mux-gpio.c @@ -55,7 +55,7 @@ static int mdio_mux_gpio_probe(struct platform_device *pdev) return PTR_ERR(s->gpios); r = mdio_mux_init(&pdev->dev, - mdio_mux_gpio_switch_fn, &s->mux_handle, s); + mdio_mux_gpio_switch_fn, &s->mux_handle, s, NULL); if (r != 0) { gpiod_put_array(s->gpios); diff --git a/drivers/net/phy/mdio-mux-mmioreg.c b/drivers/net/phy/mdio-mux-mmioreg.c index 7fde454fbc4f..d0bed52c8d16 100644 --- a/drivers/net/phy/mdio-mux-mmioreg.c +++ b/drivers/net/phy/mdio-mux-mmioreg.c @@ -126,7 +126,7 @@ static int mdio_mux_mmioreg_probe(struct platform_device *pdev) } ret = mdio_mux_init(&pdev->dev, mdio_mux_mmioreg_switch_fn, - &s->mux_handle, s); + &s->mux_handle, s, NULL); if (ret) { dev_err(&pdev->dev, "failed to register mdio-mux bus %s\n", np->full_name); diff --git a/drivers/net/phy/mdio-mux.c b/drivers/net/phy/mdio-mux.c index 5c81d6faf304..dbd4ecc205dc 100644 --- a/drivers/net/phy/mdio-mux.c +++ b/drivers/net/phy/mdio-mux.c @@ -89,7 +89,8 @@ static int parent_count; int mdio_mux_init(struct device *dev, int (*switch_fn)(int cur, int desired, void *data), void **mux_handle, - void *data) + void *data, + struct mii_bus *mux_bus) { struct device_node *parent_bus_node; struct device_node *child_bus_node; @@ -101,10 +102,21 @@ int mdio_mux_init(struct device *dev, if (!dev->of_node) return -ENODEV; - parent_bus_node = of_parse_phandle(dev->of_node, "mdio-parent-bus", 0); + if (!mux_bus) { + parent_bus_node = of_parse_phandle(dev->of_node, + "mdio-parent-bus", 0); - if (!parent_bus_node) - return -ENODEV; + if (!parent_bus_node) + return -ENODEV; + + parent_bus = of_mdio_find_bus(parent_bus_node); + if (!parent_bus) { + ret_val = -EPROBE_DEFER; + goto err_parent_bus; + } + } else { + parent_bus = mux_bus; + } pb = devm_kzalloc(dev, sizeof(*pb), GFP_KERNEL); if (pb == NULL) { @@ -112,11 +124,6 @@ int mdio_mux_init(struct device *dev, goto err_parent_bus; } - parent_bus = of_mdio_find_bus(parent_bus_node); - if (parent_bus == NULL) { - ret_val = -EPROBE_DEFER; - goto err_parent_bus; - } pb->switch_data = data; pb->switch_fn = switch_fn; @@ -177,7 +184,8 @@ int mdio_mux_init(struct device *dev, put_device(&pb->mii_bus->dev); err_parent_bus: - of_node_put(parent_bus_node); + if (!mux_bus) + of_node_put(parent_bus_node); return ret_val; } EXPORT_SYMBOL_GPL(mdio_mux_init); diff --git a/include/linux/mdio-mux.h b/include/linux/mdio-mux.h index a243dbba8659..61f5b21b31c7 100644 --- a/include/linux/mdio-mux.h +++ b/include/linux/mdio-mux.h @@ -10,11 +10,13 @@ #ifndef __LINUX_MDIO_MUX_H #define __LINUX_MDIO_MUX_H #include +#include int mdio_mux_init(struct device *dev, int (*switch_fn) (int cur, int desired, void *data), void **mux_handle, - void *data); + void *data, + struct mii_bus *mux_bus); void mdio_mux_uninit(void *mux_handle); -- cgit v1.2.3 From 99860208bc62d8ebd5c57495b84856506fe075bc Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sat, 11 Jun 2016 12:46:04 +0200 Subject: sched: remove NET_XMIT_POLICED sch_atm returns this when TC_ACT_SHOT classification occurs. But all other schedulers that use tc_classify (htb, hfsc, drr, fq_codel ...) return NET_XMIT_SUCCESS | __BYPASS in this case so just do that in atm. BATMAN uses it as an intermediate return value to signal forwarding vs. buffering, but it did not return POLICED to callers outside of BATMAN. Reviewed-by: Sven Eckelmann Signed-off-by: Florian Westphal Signed-off-by: David S. Miller --- include/linux/netdevice.h | 1 - net/batman-adv/routing.c | 2 +- net/batman-adv/send.c | 4 ++-- net/core/pktgen.c | 1 - net/sched/sch_api.c | 2 -- net/sched/sch_atm.c | 2 +- 6 files changed, 4 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 94eef356a65f..d101e4d904ba 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -90,7 +90,6 @@ void netdev_set_default_ethtool_ops(struct net_device *dev, #define NET_XMIT_SUCCESS 0x00 #define NET_XMIT_DROP 0x01 /* skb dropped */ #define NET_XMIT_CN 0x02 /* congestion notification */ -#define NET_XMIT_POLICED 0x03 /* skb is shot by police */ #define NET_XMIT_MASK 0x0f /* qdisc flags in net/sch_generic.h */ /* NET_XMIT_CN is special. It does not guarantee that this packet is lost. It diff --git a/net/batman-adv/routing.c b/net/batman-adv/routing.c index e3857ed4057f..f75091c983ee 100644 --- a/net/batman-adv/routing.c +++ b/net/batman-adv/routing.c @@ -653,7 +653,7 @@ static int batadv_route_unicast_packet(struct sk_buff *skb, len + ETH_HLEN); ret = NET_RX_SUCCESS; - } else if (res == NET_XMIT_POLICED) { + } else if (res == -EINPROGRESS) { /* skb was buffered and consumed */ ret = NET_RX_SUCCESS; } diff --git a/net/batman-adv/send.c b/net/batman-adv/send.c index f2f125684ed9..b1a4e8a811c8 100644 --- a/net/batman-adv/send.c +++ b/net/batman-adv/send.c @@ -156,7 +156,7 @@ int batadv_send_unicast_skb(struct sk_buff *skb, * attempted. * * Return: NET_XMIT_SUCCESS on success, NET_XMIT_DROP on failure, or - * NET_XMIT_POLICED if the skb is buffered for later transmit. + * -EINPROGRESS if the skb is buffered for later transmit. */ int batadv_send_skb_to_orig(struct sk_buff *skb, struct batadv_orig_node *orig_node, @@ -188,7 +188,7 @@ int batadv_send_skb_to_orig(struct sk_buff *skb, * network coding fails, then send the packet as usual. */ if (recv_if && batadv_nc_skb_forward(skb, neigh_node)) { - ret = NET_XMIT_POLICED; + ret = -EINPROGRESS; } else { batadv_send_unicast_skb(skb, neigh_node); ret = NET_XMIT_SUCCESS; diff --git a/net/core/pktgen.c b/net/core/pktgen.c index 8b02df0d354d..f74ab9c3b38f 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -3463,7 +3463,6 @@ xmit_more: break; case NET_XMIT_DROP: case NET_XMIT_CN: - case NET_XMIT_POLICED: /* skb has been consumed */ pkt_dev->errors++; break; diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index 401eda6de682..12ebde845523 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -95,8 +95,6 @@ static int tclass_notify(struct net *net, struct sk_buff *oskb, Expected action: do not backoff, but wait until queue will clear. NET_XMIT_CN - probably this packet enqueued, but another one dropped. Expected action: backoff or ignore - NET_XMIT_POLICED - dropped by police. - Expected action: backoff or error to real-time apps. Auxiliary routines: diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c index 7e6c12dfc66a..0785b239ddf9 100644 --- a/net/sched/sch_atm.c +++ b/net/sched/sch_atm.c @@ -363,7 +363,7 @@ static int atm_tc_enqueue(struct sk_buff *skb, struct Qdisc *sch) struct atm_flow_data *flow; struct tcf_result res; int result; - int ret = NET_XMIT_POLICED; + int ret = NET_XMIT_SUCCESS | __NET_XMIT_BYPASS; pr_debug("atm_tc_enqueue(skb %p,sch %p,[qdisc %p])\n", skb, sch, p); result = TC_POLICE_OK; /* be nice to gcc */ -- cgit v1.2.3 From 2e0ab8ca83c122f275b21ea917d52fee506910bf Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 13 Jun 2016 23:54:31 +0300 Subject: ptr_ring: array based FIFO for pointers A simple array based FIFO of pointers. Intended for net stack which commonly has a single consumer/producer. Signed-off-by: Michael S. Tsirkin Acked-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller --- include/linux/ptr_ring.h | 264 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 264 insertions(+) create mode 100644 include/linux/ptr_ring.h (limited to 'include/linux') diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h new file mode 100644 index 000000000000..633406f9af8e --- /dev/null +++ b/include/linux/ptr_ring.h @@ -0,0 +1,264 @@ +/* + * Definitions for the 'struct ptr_ring' datastructure. + * + * Author: + * Michael S. Tsirkin + * + * Copyright (C) 2016 Red Hat, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * This is a limited-size FIFO maintaining pointers in FIFO order, with + * one CPU producing entries and another consuming entries from a FIFO. + * + * This implementation tries to minimize cache-contention when there is a + * single producer and a single consumer CPU. + */ + +#ifndef _LINUX_PTR_RING_H +#define _LINUX_PTR_RING_H 1 + +#ifdef __KERNEL__ +#include +#include +#include +#include +#include +#include +#include +#endif + +struct ptr_ring { + int producer ____cacheline_aligned_in_smp; + spinlock_t producer_lock; + int consumer ____cacheline_aligned_in_smp; + spinlock_t consumer_lock; + /* Shared consumer/producer data */ + /* Read-only by both the producer and the consumer */ + int size ____cacheline_aligned_in_smp; /* max entries in queue */ + void **queue; +}; + +/* Note: callers invoking this in a loop must use a compiler barrier, + * for example cpu_relax(). + * Callers don't need to take producer lock - if they don't + * the next call to __ptr_ring_produce may fail. + */ +static inline bool __ptr_ring_full(struct ptr_ring *r) +{ + return r->queue[r->producer]; +} + +static inline bool ptr_ring_full(struct ptr_ring *r) +{ + barrier(); + return __ptr_ring_full(r); +} + +/* Note: callers invoking this in a loop must use a compiler barrier, + * for example cpu_relax(). + */ +static inline int __ptr_ring_produce(struct ptr_ring *r, void *ptr) +{ + if (__ptr_ring_full(r)) + return -ENOSPC; + + r->queue[r->producer++] = ptr; + if (unlikely(r->producer >= r->size)) + r->producer = 0; + return 0; +} + +static inline int ptr_ring_produce(struct ptr_ring *r, void *ptr) +{ + int ret; + + spin_lock(&r->producer_lock); + ret = __ptr_ring_produce(r, ptr); + spin_unlock(&r->producer_lock); + + return ret; +} + +static inline int ptr_ring_produce_irq(struct ptr_ring *r, void *ptr) +{ + int ret; + + spin_lock_irq(&r->producer_lock); + ret = __ptr_ring_produce(r, ptr); + spin_unlock_irq(&r->producer_lock); + + return ret; +} + +static inline int ptr_ring_produce_any(struct ptr_ring *r, void *ptr) +{ + unsigned long flags; + int ret; + + spin_lock_irqsave(&r->producer_lock, flags); + ret = __ptr_ring_produce(r, ptr); + spin_unlock_irqrestore(&r->producer_lock, flags); + + return ret; +} + +static inline int ptr_ring_produce_bh(struct ptr_ring *r, void *ptr) +{ + int ret; + + spin_lock_bh(&r->producer_lock); + ret = __ptr_ring_produce(r, ptr); + spin_unlock_bh(&r->producer_lock); + + return ret; +} + +/* Note: callers invoking this in a loop must use a compiler barrier, + * for example cpu_relax(). Callers must take consumer_lock + * if they dereference the pointer - see e.g. PTR_RING_PEEK_CALL. + * There's no need for a lock if pointer is merely tested - see e.g. + * ptr_ring_empty. + */ +static inline void *__ptr_ring_peek(struct ptr_ring *r) +{ + return r->queue[r->consumer]; +} + +static inline bool ptr_ring_empty(struct ptr_ring *r) +{ + barrier(); + return !__ptr_ring_peek(r); +} + +/* Must only be called after __ptr_ring_peek returned !NULL */ +static inline void __ptr_ring_discard_one(struct ptr_ring *r) +{ + r->queue[r->consumer++] = NULL; + if (unlikely(r->consumer >= r->size)) + r->consumer = 0; +} + +static inline void *__ptr_ring_consume(struct ptr_ring *r) +{ + void *ptr; + + ptr = __ptr_ring_peek(r); + if (ptr) + __ptr_ring_discard_one(r); + + return ptr; +} + +static inline void *ptr_ring_consume(struct ptr_ring *r) +{ + void *ptr; + + spin_lock(&r->consumer_lock); + ptr = __ptr_ring_consume(r); + spin_unlock(&r->consumer_lock); + + return ptr; +} + +static inline void *ptr_ring_consume_irq(struct ptr_ring *r) +{ + void *ptr; + + spin_lock_irq(&r->consumer_lock); + ptr = __ptr_ring_consume(r); + spin_unlock_irq(&r->consumer_lock); + + return ptr; +} + +static inline void *ptr_ring_consume_any(struct ptr_ring *r) +{ + unsigned long flags; + void *ptr; + + spin_lock_irqsave(&r->consumer_lock, flags); + ptr = __ptr_ring_consume(r); + spin_unlock_irqrestore(&r->consumer_lock, flags); + + return ptr; +} + +static inline void *ptr_ring_consume_bh(struct ptr_ring *r) +{ + void *ptr; + + spin_lock_bh(&r->consumer_lock); + ptr = __ptr_ring_consume(r); + spin_unlock_bh(&r->consumer_lock); + + return ptr; +} + +/* Cast to structure type and call a function without discarding from FIFO. + * Function must return a value. + * Callers must take consumer_lock. + */ +#define __PTR_RING_PEEK_CALL(r, f) ((f)(__ptr_ring_peek(r))) + +#define PTR_RING_PEEK_CALL(r, f) ({ \ + typeof((f)(NULL)) __PTR_RING_PEEK_CALL_v; \ + \ + spin_lock(&(r)->consumer_lock); \ + __PTR_RING_PEEK_CALL_v = __PTR_RING_PEEK_CALL(r, f); \ + spin_unlock(&(r)->consumer_lock); \ + __PTR_RING_PEEK_CALL_v; \ +}) + +#define PTR_RING_PEEK_CALL_IRQ(r, f) ({ \ + typeof((f)(NULL)) __PTR_RING_PEEK_CALL_v; \ + \ + spin_lock_irq(&(r)->consumer_lock); \ + __PTR_RING_PEEK_CALL_v = __PTR_RING_PEEK_CALL(r, f); \ + spin_unlock_irq(&(r)->consumer_lock); \ + __PTR_RING_PEEK_CALL_v; \ +}) + +#define PTR_RING_PEEK_CALL_BH(r, f) ({ \ + typeof((f)(NULL)) __PTR_RING_PEEK_CALL_v; \ + \ + spin_lock_bh(&(r)->consumer_lock); \ + __PTR_RING_PEEK_CALL_v = __PTR_RING_PEEK_CALL(r, f); \ + spin_unlock_bh(&(r)->consumer_lock); \ + __PTR_RING_PEEK_CALL_v; \ +}) + +#define PTR_RING_PEEK_CALL_ANY(r, f) ({ \ + typeof((f)(NULL)) __PTR_RING_PEEK_CALL_v; \ + unsigned long __PTR_RING_PEEK_CALL_f;\ + \ + spin_lock_irqsave(&(r)->consumer_lock, __PTR_RING_PEEK_CALL_f); \ + __PTR_RING_PEEK_CALL_v = __PTR_RING_PEEK_CALL(r, f); \ + spin_unlock_irqrestore(&(r)->consumer_lock, __PTR_RING_PEEK_CALL_f); \ + __PTR_RING_PEEK_CALL_v; \ +}) + +static inline int ptr_ring_init(struct ptr_ring *r, int size, gfp_t gfp) +{ + r->queue = kzalloc(ALIGN(size * sizeof *(r->queue), SMP_CACHE_BYTES), + gfp); + if (!r->queue) + return -ENOMEM; + + r->size = size; + r->producer = r->consumer = 0; + spin_lock_init(&r->producer_lock); + spin_lock_init(&r->consumer_lock); + + return 0; +} + +static inline void ptr_ring_cleanup(struct ptr_ring *r) +{ + kfree(r->queue); +} + +#endif /* _LINUX_PTR_RING_H */ -- cgit v1.2.3 From ad69f35d1dc0a79f86627ca56e01f86512602a49 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 13 Jun 2016 23:54:41 +0300 Subject: skb_array: array based FIFO for skbs A simple array based FIFO of pointers. Intended for net stack so uses skbs for type safety. Implemented as a set of wrappers around ptr_ring. Signed-off-by: Michael S. Tsirkin Acked-by: Jesper Dangaard Brouer Tested-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller --- include/linux/skb_array.h | 144 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 144 insertions(+) create mode 100644 include/linux/skb_array.h (limited to 'include/linux') diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h new file mode 100644 index 000000000000..c4c090223333 --- /dev/null +++ b/include/linux/skb_array.h @@ -0,0 +1,144 @@ +/* + * Definitions for the 'struct skb_array' datastructure. + * + * Author: + * Michael S. Tsirkin + * + * Copyright (C) 2016 Red Hat, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * Limited-size FIFO of skbs. Can be used more or less whenever + * sk_buff_head can be used, except you need to know the queue size in + * advance. + * Implemented as a type-safe wrapper around ptr_ring. + */ + +#ifndef _LINUX_SKB_ARRAY_H +#define _LINUX_SKB_ARRAY_H 1 + +#ifdef __KERNEL__ +#include +#include +#include +#endif + +struct skb_array { + struct ptr_ring ring; +}; + +/* Might be slightly faster than skb_array_full below, but callers invoking + * this in a loop must use a compiler barrier, for example cpu_relax(). + */ +static inline bool __skb_array_full(struct skb_array *a) +{ + return __ptr_ring_full(&a->ring); +} + +static inline bool skb_array_full(struct skb_array *a) +{ + return ptr_ring_full(&a->ring); +} + +static inline int skb_array_produce(struct skb_array *a, struct sk_buff *skb) +{ + return ptr_ring_produce(&a->ring, skb); +} + +static inline int skb_array_produce_irq(struct skb_array *a, struct sk_buff *skb) +{ + return ptr_ring_produce_irq(&a->ring, skb); +} + +static inline int skb_array_produce_bh(struct skb_array *a, struct sk_buff *skb) +{ + return ptr_ring_produce_bh(&a->ring, skb); +} + +static inline int skb_array_produce_any(struct skb_array *a, struct sk_buff *skb) +{ + return ptr_ring_produce_any(&a->ring, skb); +} + +/* Might be slightly faster than skb_array_empty below, but callers invoking + * this in a loop must take care to use a compiler barrier, for example + * cpu_relax(). + */ +static inline bool __skb_array_empty(struct skb_array *a) +{ + return !__ptr_ring_peek(&a->ring); +} + +static inline bool skb_array_empty(struct skb_array *a) +{ + return ptr_ring_empty(&a->ring); +} + +static inline struct sk_buff *skb_array_consume(struct skb_array *a) +{ + return ptr_ring_consume(&a->ring); +} + +static inline struct sk_buff *skb_array_consume_irq(struct skb_array *a) +{ + return ptr_ring_consume_irq(&a->ring); +} + +static inline struct sk_buff *skb_array_consume_any(struct skb_array *a) +{ + return ptr_ring_consume_any(&a->ring); +} + +static inline struct sk_buff *skb_array_consume_bh(struct skb_array *a) +{ + return ptr_ring_consume_bh(&a->ring); +} + +static inline int __skb_array_len_with_tag(struct sk_buff *skb) +{ + if (likely(skb)) { + int len = skb->len; + + if (skb_vlan_tag_present(skb)) + len += VLAN_HLEN; + + return len; + } else { + return 0; + } +} + +static inline int skb_array_peek_len(struct skb_array *a) +{ + return PTR_RING_PEEK_CALL(&a->ring, __skb_array_len_with_tag); +} + +static inline int skb_array_peek_len_irq(struct skb_array *a) +{ + return PTR_RING_PEEK_CALL_IRQ(&a->ring, __skb_array_len_with_tag); +} + +static inline int skb_array_peek_len_bh(struct skb_array *a) +{ + return PTR_RING_PEEK_CALL_BH(&a->ring, __skb_array_len_with_tag); +} + +static inline int skb_array_peek_len_any(struct skb_array *a) +{ + return PTR_RING_PEEK_CALL_ANY(&a->ring, __skb_array_len_with_tag); +} + +static inline int skb_array_init(struct skb_array *a, int size, gfp_t gfp) +{ + return ptr_ring_init(&a->ring, size, gfp); +} + +static inline void skb_array_cleanup(struct skb_array *a) +{ + ptr_ring_cleanup(&a->ring); +} + +#endif /* _LINUX_SKB_ARRAY_H */ -- cgit v1.2.3 From 5d49de532002f02755decd1758aac53063a68625 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 13 Jun 2016 23:54:45 +0300 Subject: ptr_ring: resize support This adds ring resize support. Seems to be necessary as users such as tun allow userspace control over queue size. If resize is used, this costs us ability to peek at queue without consumer lock - should not be a big deal as peek and consumer are usually run on the same CPU. If ring is made bigger, ring contents is preserved. If ring is made smaller, extra pointers are passed to an optional destructor callback. Cleanup function also gains destructor callback such that all pointers in queue can be cleaned up. This changes some APIs but we don't have any users yet, so it won't break bisect. Signed-off-by: Michael S. Tsirkin Acked-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller --- include/linux/ptr_ring.h | 157 ++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 143 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h index 633406f9af8e..562a65e8bcc0 100644 --- a/include/linux/ptr_ring.h +++ b/include/linux/ptr_ring.h @@ -43,9 +43,9 @@ struct ptr_ring { }; /* Note: callers invoking this in a loop must use a compiler barrier, - * for example cpu_relax(). - * Callers don't need to take producer lock - if they don't - * the next call to __ptr_ring_produce may fail. + * for example cpu_relax(). If ring is ever resized, callers must hold + * producer_lock - see e.g. ptr_ring_full. Otherwise, if callers don't hold + * producer_lock, the next call to __ptr_ring_produce may fail. */ static inline bool __ptr_ring_full(struct ptr_ring *r) { @@ -54,16 +54,55 @@ static inline bool __ptr_ring_full(struct ptr_ring *r) static inline bool ptr_ring_full(struct ptr_ring *r) { - barrier(); - return __ptr_ring_full(r); + bool ret; + + spin_lock(&r->producer_lock); + ret = __ptr_ring_full(r); + spin_unlock(&r->producer_lock); + + return ret; +} + +static inline bool ptr_ring_full_irq(struct ptr_ring *r) +{ + bool ret; + + spin_lock_irq(&r->producer_lock); + ret = __ptr_ring_full(r); + spin_unlock_irq(&r->producer_lock); + + return ret; +} + +static inline bool ptr_ring_full_any(struct ptr_ring *r) +{ + unsigned long flags; + bool ret; + + spin_lock_irqsave(&r->producer_lock, flags); + ret = __ptr_ring_full(r); + spin_unlock_irqrestore(&r->producer_lock, flags); + + return ret; +} + +static inline bool ptr_ring_full_bh(struct ptr_ring *r) +{ + bool ret; + + spin_lock_bh(&r->producer_lock); + ret = __ptr_ring_full(r); + spin_unlock_bh(&r->producer_lock); + + return ret; } /* Note: callers invoking this in a loop must use a compiler barrier, - * for example cpu_relax(). + * for example cpu_relax(). Callers must hold producer_lock. */ static inline int __ptr_ring_produce(struct ptr_ring *r, void *ptr) { - if (__ptr_ring_full(r)) + if (r->queue[r->producer]) return -ENOSPC; r->queue[r->producer++] = ptr; @@ -120,20 +159,68 @@ static inline int ptr_ring_produce_bh(struct ptr_ring *r, void *ptr) /* Note: callers invoking this in a loop must use a compiler barrier, * for example cpu_relax(). Callers must take consumer_lock * if they dereference the pointer - see e.g. PTR_RING_PEEK_CALL. - * There's no need for a lock if pointer is merely tested - see e.g. - * ptr_ring_empty. + * If ring is never resized, and if the pointer is merely + * tested, there's no need to take the lock - see e.g. __ptr_ring_empty. */ static inline void *__ptr_ring_peek(struct ptr_ring *r) { return r->queue[r->consumer]; } -static inline bool ptr_ring_empty(struct ptr_ring *r) +/* Note: callers invoking this in a loop must use a compiler barrier, + * for example cpu_relax(). Callers must take consumer_lock + * if the ring is ever resized - see e.g. ptr_ring_empty. + */ +static inline bool __ptr_ring_empty(struct ptr_ring *r) { - barrier(); return !__ptr_ring_peek(r); } +static inline bool ptr_ring_empty(struct ptr_ring *r) +{ + bool ret; + + spin_lock(&r->consumer_lock); + ret = __ptr_ring_empty(r); + spin_unlock(&r->consumer_lock); + + return ret; +} + +static inline bool ptr_ring_empty_irq(struct ptr_ring *r) +{ + bool ret; + + spin_lock_irq(&r->consumer_lock); + ret = __ptr_ring_empty(r); + spin_unlock_irq(&r->consumer_lock); + + return ret; +} + +static inline bool ptr_ring_empty_any(struct ptr_ring *r) +{ + unsigned long flags; + bool ret; + + spin_lock_irqsave(&r->consumer_lock, flags); + ret = __ptr_ring_empty(r); + spin_unlock_irqrestore(&r->consumer_lock, flags); + + return ret; +} + +static inline bool ptr_ring_empty_bh(struct ptr_ring *r) +{ + bool ret; + + spin_lock_bh(&r->consumer_lock); + ret = __ptr_ring_empty(r); + spin_unlock_bh(&r->consumer_lock); + + return ret; +} + /* Must only be called after __ptr_ring_peek returned !NULL */ static inline void __ptr_ring_discard_one(struct ptr_ring *r) { @@ -241,10 +328,14 @@ static inline void *ptr_ring_consume_bh(struct ptr_ring *r) __PTR_RING_PEEK_CALL_v; \ }) +static inline void **__ptr_ring_init_queue_alloc(int size, gfp_t gfp) +{ + return kzalloc(ALIGN(size * sizeof(void *), SMP_CACHE_BYTES), gfp); +} + static inline int ptr_ring_init(struct ptr_ring *r, int size, gfp_t gfp) { - r->queue = kzalloc(ALIGN(size * sizeof *(r->queue), SMP_CACHE_BYTES), - gfp); + r->queue = __ptr_ring_init_queue_alloc(size, gfp); if (!r->queue) return -ENOMEM; @@ -256,8 +347,46 @@ static inline int ptr_ring_init(struct ptr_ring *r, int size, gfp_t gfp) return 0; } -static inline void ptr_ring_cleanup(struct ptr_ring *r) +static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, + void (*destroy)(void *)) +{ + unsigned long flags; + int producer = 0; + void **queue = __ptr_ring_init_queue_alloc(size, gfp); + void **old; + void *ptr; + + if (!queue) + return -ENOMEM; + + spin_lock_irqsave(&(r)->producer_lock, flags); + + while ((ptr = ptr_ring_consume(r))) + if (producer < size) + queue[producer++] = ptr; + else if (destroy) + destroy(ptr); + + r->size = size; + r->producer = producer; + r->consumer = 0; + old = r->queue; + r->queue = queue; + + spin_unlock_irqrestore(&(r)->producer_lock, flags); + + kfree(old); + + return 0; +} + +static inline void ptr_ring_cleanup(struct ptr_ring *r, void (*destroy)(void *)) { + void *ptr; + + if (destroy) + while ((ptr = ptr_ring_consume(r))) + destroy(ptr); kfree(r->queue); } -- cgit v1.2.3 From 7d7072e3bad5569f636d3c54a36da40976bfd505 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Mon, 13 Jun 2016 23:54:50 +0300 Subject: skb_array: resize support Update skb_array after ptr_ring API changes. Signed-off-by: Michael S. Tsirkin Acked-by: Jesper Dangaard Brouer Tested-by: Jesper Dangaard Brouer Signed-off-by: David S. Miller --- include/linux/skb_array.h | 33 +++++++++++++++++++++++++++++---- 1 file changed, 29 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h index c4c090223333..678bfbf78ac4 100644 --- a/include/linux/skb_array.h +++ b/include/linux/skb_array.h @@ -63,9 +63,9 @@ static inline int skb_array_produce_any(struct skb_array *a, struct sk_buff *skb return ptr_ring_produce_any(&a->ring, skb); } -/* Might be slightly faster than skb_array_empty below, but callers invoking - * this in a loop must take care to use a compiler barrier, for example - * cpu_relax(). +/* Might be slightly faster than skb_array_empty below, but only safe if the + * array is never resized. Also, callers invoking this in a loop must take care + * to use a compiler barrier, for example cpu_relax(). */ static inline bool __skb_array_empty(struct skb_array *a) { @@ -77,6 +77,21 @@ static inline bool skb_array_empty(struct skb_array *a) return ptr_ring_empty(&a->ring); } +static inline bool skb_array_empty_bh(struct skb_array *a) +{ + return ptr_ring_empty_bh(&a->ring); +} + +static inline bool skb_array_empty_irq(struct skb_array *a) +{ + return ptr_ring_empty_irq(&a->ring); +} + +static inline bool skb_array_empty_any(struct skb_array *a) +{ + return ptr_ring_empty_any(&a->ring); +} + static inline struct sk_buff *skb_array_consume(struct skb_array *a) { return ptr_ring_consume(&a->ring); @@ -136,9 +151,19 @@ static inline int skb_array_init(struct skb_array *a, int size, gfp_t gfp) return ptr_ring_init(&a->ring, size, gfp); } +void __skb_array_destroy_skb(void *ptr) +{ + kfree_skb(ptr); +} + +int skb_array_resize(struct skb_array *a, int size, gfp_t gfp) +{ + return ptr_ring_resize(&a->ring, size, gfp, __skb_array_destroy_skb); +} + static inline void skb_array_cleanup(struct skb_array *a) { - ptr_ring_cleanup(&a->ring); + ptr_ring_cleanup(&a->ring, __skb_array_destroy_skb); } #endif /* _LINUX_SKB_ARRAY_H */ -- cgit v1.2.3 From 1b5c5493e3e68181be344cb51bf9df192d05ffc2 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 13 Jun 2016 20:21:50 -0700 Subject: net_sched: add the ability to defer skb freeing qdisc are changed under RTNL protection and often while blocking BH and root qdisc spinlock. When lots of skbs need to be dropped, we free them under these locks causing TX/RX freezes, and more generally latency spikes. This commit adds rtnl_kfree_skbs(), used to queue skbs for deferred freeing. Actual freeing happens right after RTNL is released, with appropriate scheduling points. rtnl_qdisc_drop() can also be used in place of disc_drop() when RTNL is held. qdisc_reset_queue() and __qdisc_reset_queue() get the new behavior, so standard qdiscs like pfifo, pfifo_fast... have their ->reset() method automatically handled. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/rtnetlink.h | 5 +++-- include/net/sch_generic.h | 16 ++++++++++++---- net/core/rtnetlink.c | 22 ++++++++++++++++++++++ net/sched/sch_generic.c | 2 +- 4 files changed, 38 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h index c006cc900c44..2daece8979f7 100644 --- a/include/linux/rtnetlink.h +++ b/include/linux/rtnetlink.h @@ -89,8 +89,9 @@ void net_inc_egress_queue(void); void net_dec_egress_queue(void); #endif -extern void rtnetlink_init(void); -extern void __rtnl_unlock(void); +void rtnetlink_init(void); +void __rtnl_unlock(void); +void rtnl_kfree_skbs(struct sk_buff *head, struct sk_buff *tail); #define ASSERT_RTNL() do { \ if (unlikely(!rtnl_is_locked())) { \ diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h index 9a0d177884c6..4f7cee8344c4 100644 --- a/include/net/sch_generic.h +++ b/include/net/sch_generic.h @@ -683,19 +683,21 @@ static inline struct sk_buff *qdisc_dequeue_peeked(struct Qdisc *sch) return skb; } -static inline void __qdisc_reset_queue(struct Qdisc *sch, - struct sk_buff_head *list) +static inline void __qdisc_reset_queue(struct sk_buff_head *list) { /* * We do not know the backlog in bytes of this list, it * is up to the caller to correct it */ - __skb_queue_purge(list); + if (!skb_queue_empty(list)) { + rtnl_kfree_skbs(list->next, list->prev); + __skb_queue_head_init(list); + } } static inline void qdisc_reset_queue(struct Qdisc *sch) { - __qdisc_reset_queue(sch, &sch->q); + __qdisc_reset_queue(&sch->q); sch->qstats.backlog = 0; } @@ -716,6 +718,12 @@ static inline struct Qdisc *qdisc_replace(struct Qdisc *sch, struct Qdisc *new, return old; } +static inline void rtnl_qdisc_drop(struct sk_buff *skb, struct Qdisc *sch) +{ + rtnl_kfree_skbs(skb, skb); + qdisc_qstats_drop(sch); +} + static inline int qdisc_drop(struct sk_buff *skb, struct Qdisc *sch) { kfree_skb(skb); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index d69c4644f8f2..eb49ca24274a 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -71,9 +71,31 @@ void rtnl_lock(void) } EXPORT_SYMBOL(rtnl_lock); +static struct sk_buff *defer_kfree_skb_list; +void rtnl_kfree_skbs(struct sk_buff *head, struct sk_buff *tail) +{ + if (head && tail) { + tail->next = defer_kfree_skb_list; + defer_kfree_skb_list = head; + } +} +EXPORT_SYMBOL(rtnl_kfree_skbs); + void __rtnl_unlock(void) { + struct sk_buff *head = defer_kfree_skb_list; + + defer_kfree_skb_list = NULL; + mutex_unlock(&rtnl_mutex); + + while (head) { + struct sk_buff *next = head->next; + + kfree_skb(head); + cond_resched(); + head = next; + } } void rtnl_unlock(void) diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 0c9cb516f2e3..773b632e1e33 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -493,7 +493,7 @@ static void pfifo_fast_reset(struct Qdisc *qdisc) struct pfifo_fast_priv *priv = qdisc_priv(qdisc); for (prio = 0; prio < PFIFO_FAST_BANDS; prio++) - __qdisc_reset_queue(qdisc, band2list(priv, prio)); + __qdisc_reset_queue(band2list(priv, prio)); priv->bitmap = 0; qdisc->qstats.backlog = 0; -- cgit v1.2.3 From 8626a0c83b0d471d859bcd908d016874df951fc3 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:16 +0200 Subject: 6lowpan: add private neighbour data This patch will introduce a 6lowpan neighbour private data. Like the interface private data we handle private data for generic 6lowpan and for link-layer specific 6lowpan. The current first use case if to save the short address for a 802.15.4 6lowpan neighbour. Cc: David S. Miller Reviewed-by: Stefan Schmidt Acked-by: YOSHIFUJI Hideaki Signed-off-by: Alexander Aring Signed-off-by: David S. Miller --- include/linux/netdevice.h | 3 +-- include/net/6lowpan.h | 10 ++++++++++ net/ieee802154/6lowpan/core.c | 12 ++++++++++++ 3 files changed, 23 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index d101e4d904ba..36e43bd422f8 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1483,8 +1483,7 @@ enum netdev_priv_flags { * @perm_addr: Permanent hw address * @addr_assign_type: Hw address assignment type * @addr_len: Hardware address length - * @neigh_priv_len; Used in neigh_alloc(), - * initialized only in atm/clip.c + * @neigh_priv_len: Used in neigh_alloc() * @dev_id: Used to differentiate devices that share * the same link layer address * @dev_port: Used to differentiate devices that share diff --git a/include/net/6lowpan.h b/include/net/6lowpan.h index da84cf920b78..2d9b9d39221e 100644 --- a/include/net/6lowpan.h +++ b/include/net/6lowpan.h @@ -141,6 +141,16 @@ struct lowpan_dev { u8 priv[0] __aligned(sizeof(void *)); }; +struct lowpan_802154_neigh { + __le16 short_addr; +}; + +static inline +struct lowpan_802154_neigh *lowpan_802154_neigh(void *neigh_priv) +{ + return neigh_priv; +} + static inline struct lowpan_dev *lowpan_dev(const struct net_device *dev) { diff --git a/net/ieee802154/6lowpan/core.c b/net/ieee802154/6lowpan/core.c index 4e2b30894224..8c004a0c8d64 100644 --- a/net/ieee802154/6lowpan/core.c +++ b/net/ieee802154/6lowpan/core.c @@ -81,11 +81,21 @@ static int lowpan_stop(struct net_device *dev) return 0; } +static int lowpan_neigh_construct(struct neighbour *n) +{ + struct lowpan_802154_neigh *neigh = lowpan_802154_neigh(neighbour_priv(n)); + + /* default no short_addr is available for a neighbour */ + neigh->short_addr = cpu_to_le16(IEEE802154_ADDR_SHORT_UNSPEC); + return 0; +} + static const struct net_device_ops lowpan_netdev_ops = { .ndo_init = lowpan_dev_init, .ndo_start_xmit = lowpan_xmit, .ndo_open = lowpan_open, .ndo_stop = lowpan_stop, + .ndo_neigh_construct = lowpan_neigh_construct, }; static void lowpan_setup(struct net_device *ldev) @@ -150,6 +160,8 @@ static int lowpan_newlink(struct net *src_net, struct net_device *ldev, wdev->needed_headroom; ldev->needed_tailroom = wdev->needed_tailroom; + ldev->neigh_priv_len = sizeof(struct lowpan_802154_neigh); + ret = lowpan_register_netdevice(ldev, LOWPAN_LLTYPE_IEEE802154); if (ret < 0) { dev_put(wdev); -- cgit v1.2.3 From f997c55c1dc8841b3ee4df0493d0ac7966d42165 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 15 Jun 2016 21:20:23 +0200 Subject: ipv6: introduce neighbour discovery ops This patch introduces neighbour discovery ops callback structure. The idea is to separate the handling for 6LoWPAN into the 6lowpan module. These callback offers 6lowpan different handling, such as 802.15.4 short address handling or RFC6775 (Neighbor Discovery Optimization for IPv6 over 6LoWPANs). Cc: David S. Miller Cc: Alexey Kuznetsov Cc: James Morris Cc: Hideaki YOSHIFUJI Cc: Patrick McHardy Acked-by: YOSHIFUJI Hideaki Signed-off-by: Alexander Aring Signed-off-by: David S. Miller --- include/linux/netdevice.h | 5 ++ include/net/ndisc.h | 197 +++++++++++++++++++++++++++++++++++++++++++++- net/ipv6/addrconf.c | 13 ++- net/ipv6/ndisc.c | 101 ++++++++++++++++-------- net/ipv6/route.c | 8 +- 5 files changed, 284 insertions(+), 40 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 36e43bd422f8..890158e99159 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1456,6 +1456,8 @@ enum netdev_priv_flags { * @netdev_ops: Includes several pointers to callbacks, * if one wants to override the ndo_*() functions * @ethtool_ops: Management operations + * @ndisc_ops: Includes callbacks for different IPv6 neighbour + * discovery handling. Necessary for e.g. 6LoWPAN. * @header_ops: Includes callbacks for creating,parsing,caching,etc * of Layer 2 headers. * @@ -1672,6 +1674,9 @@ struct net_device { #ifdef CONFIG_NET_L3_MASTER_DEV const struct l3mdev_ops *l3mdev_ops; #endif +#if IS_ENABLED(CONFIG_IPV6) + const struct ndisc_ops *ndisc_ops; +#endif const struct header_ops *header_ops; diff --git a/include/net/ndisc.h b/include/net/ndisc.h index c8962adf24d0..a5e276703cb3 100644 --- a/include/net/ndisc.h +++ b/include/net/ndisc.h @@ -58,6 +58,7 @@ struct inet6_dev; struct net_device; struct net_proto_family; struct sk_buff; +struct prefix_info; extern struct neigh_table nd_tbl; @@ -110,9 +111,182 @@ struct ndisc_options { #define NDISC_OPT_SPACE(len) (((len)+2+7)&~7) -struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len, +struct ndisc_options *ndisc_parse_options(const struct net_device *dev, + u8 *opt, int opt_len, struct ndisc_options *ndopts); +#define NDISC_OPS_REDIRECT_DATA_SPACE 2 + +/* + * This structure defines the hooks for IPv6 neighbour discovery. + * The following hooks can be defined; unless noted otherwise, they are + * optional and can be filled with a null pointer. + * + * int (*is_useropt)(u8 nd_opt_type): + * This function is called when IPv6 decide RA userspace options. if + * this function returns 1 then the option given by nd_opt_type will + * be handled as userspace option additional to the IPv6 options. + * + * int (*parse_options)(const struct net_device *dev, + * struct nd_opt_hdr *nd_opt, + * struct ndisc_options *ndopts): + * This function is called while parsing ndisc ops and put each position + * as pointer into ndopts. If this function return unequal 0, then this + * function took care about the ndisc option, if 0 then the IPv6 ndisc + * option parser will take care about that option. + * + * void (*update)(const struct net_device *dev, struct neighbour *n, + * u32 flags, u8 icmp6_type, + * const struct ndisc_options *ndopts): + * This function is called when IPv6 ndisc updates the neighbour cache + * entry. Additional options which can be updated may be previously + * parsed by parse_opts callback and accessible over ndopts parameter. + * + * int (*opt_addr_space)(const struct net_device *dev, u8 icmp6_type, + * struct neighbour *neigh, u8 *ha_buf, + * u8 **ha): + * This function is called when the necessary option space will be + * calculated before allocating a skb. The parameters neigh, ha_buf + * abd ha are available on NDISC_REDIRECT messages only. + * + * void (*fill_addr_option)(const struct net_device *dev, + * struct sk_buff *skb, u8 icmp6_type, + * const u8 *ha): + * This function is called when the skb will finally fill the option + * fields inside skb. NOTE: this callback should fill the option + * fields to the skb which are previously indicated by opt_space + * parameter. That means the decision to add such option should + * not lost between these two callbacks, e.g. protected by interface + * up state. + * + * void (*prefix_rcv_add_addr)(struct net *net, struct net_device *dev, + * const struct prefix_info *pinfo, + * struct inet6_dev *in6_dev, + * struct in6_addr *addr, + * int addr_type, u32 addr_flags, + * bool sllao, bool tokenized, + * __u32 valid_lft, u32 prefered_lft, + * bool dev_addr_generated): + * This function is called when a RA messages is received with valid + * PIO option fields and an IPv6 address will be added to the interface + * for autoconfiguration. The parameter dev_addr_generated reports about + * if the address was based on dev->dev_addr or not. This can be used + * to add a second address if link-layer operates with two link layer + * addresses. E.g. 802.15.4 6LoWPAN. + */ +struct ndisc_ops { + int (*is_useropt)(u8 nd_opt_type); + int (*parse_options)(const struct net_device *dev, + struct nd_opt_hdr *nd_opt, + struct ndisc_options *ndopts); + void (*update)(const struct net_device *dev, struct neighbour *n, + u32 flags, u8 icmp6_type, + const struct ndisc_options *ndopts); + int (*opt_addr_space)(const struct net_device *dev, u8 icmp6_type, + struct neighbour *neigh, u8 *ha_buf, + u8 **ha); + void (*fill_addr_option)(const struct net_device *dev, + struct sk_buff *skb, u8 icmp6_type, + const u8 *ha); + void (*prefix_rcv_add_addr)(struct net *net, struct net_device *dev, + const struct prefix_info *pinfo, + struct inet6_dev *in6_dev, + struct in6_addr *addr, + int addr_type, u32 addr_flags, + bool sllao, bool tokenized, + __u32 valid_lft, u32 prefered_lft, + bool dev_addr_generated); +}; + +#if IS_ENABLED(CONFIG_IPV6) +static inline int ndisc_ops_is_useropt(const struct net_device *dev, + u8 nd_opt_type) +{ + if (dev->ndisc_ops && dev->ndisc_ops->is_useropt) + return dev->ndisc_ops->is_useropt(nd_opt_type); + else + return 0; +} + +static inline int ndisc_ops_parse_options(const struct net_device *dev, + struct nd_opt_hdr *nd_opt, + struct ndisc_options *ndopts) +{ + if (dev->ndisc_ops && dev->ndisc_ops->parse_options) + return dev->ndisc_ops->parse_options(dev, nd_opt, ndopts); + else + return 0; +} + +static inline void ndisc_ops_update(const struct net_device *dev, + struct neighbour *n, u32 flags, + u8 icmp6_type, + const struct ndisc_options *ndopts) +{ + if (dev->ndisc_ops && dev->ndisc_ops->update) + dev->ndisc_ops->update(dev, n, flags, icmp6_type, ndopts); +} + +static inline int ndisc_ops_opt_addr_space(const struct net_device *dev, + u8 icmp6_type) +{ + if (dev->ndisc_ops && dev->ndisc_ops->opt_addr_space && + icmp6_type != NDISC_REDIRECT) + return dev->ndisc_ops->opt_addr_space(dev, icmp6_type, NULL, + NULL, NULL); + else + return 0; +} + +static inline int ndisc_ops_redirect_opt_addr_space(const struct net_device *dev, + struct neighbour *neigh, + u8 *ha_buf, u8 **ha) +{ + if (dev->ndisc_ops && dev->ndisc_ops->opt_addr_space) + return dev->ndisc_ops->opt_addr_space(dev, NDISC_REDIRECT, + neigh, ha_buf, ha); + else + return 0; +} + +static inline void ndisc_ops_fill_addr_option(const struct net_device *dev, + struct sk_buff *skb, + u8 icmp6_type) +{ + if (dev->ndisc_ops && dev->ndisc_ops->fill_addr_option && + icmp6_type != NDISC_REDIRECT) + dev->ndisc_ops->fill_addr_option(dev, skb, icmp6_type, NULL); +} + +static inline void ndisc_ops_fill_redirect_addr_option(const struct net_device *dev, + struct sk_buff *skb, + const u8 *ha) +{ + if (dev->ndisc_ops && dev->ndisc_ops->fill_addr_option) + dev->ndisc_ops->fill_addr_option(dev, skb, NDISC_REDIRECT, ha); +} + +static inline void ndisc_ops_prefix_rcv_add_addr(struct net *net, + struct net_device *dev, + const struct prefix_info *pinfo, + struct inet6_dev *in6_dev, + struct in6_addr *addr, + int addr_type, u32 addr_flags, + bool sllao, bool tokenized, + __u32 valid_lft, + u32 prefered_lft, + bool dev_addr_generated) +{ + if (dev->ndisc_ops && dev->ndisc_ops->prefix_rcv_add_addr) + dev->ndisc_ops->prefix_rcv_add_addr(net, dev, pinfo, in6_dev, + addr, addr_type, + addr_flags, sllao, + tokenized, valid_lft, + prefered_lft, + dev_addr_generated); +} +#endif + /* * Return the padding between the option length and the start of the * link addr. Currently only IP-over-InfiniBand needs this, although @@ -132,11 +306,25 @@ static inline int __ndisc_opt_addr_space(unsigned char addr_len, int pad) return NDISC_OPT_SPACE(addr_len + pad); } -static inline int ndisc_opt_addr_space(struct net_device *dev) +#if IS_ENABLED(CONFIG_IPV6) +static inline int ndisc_opt_addr_space(struct net_device *dev, u8 icmp6_type) +{ + return __ndisc_opt_addr_space(dev->addr_len, + ndisc_addr_option_pad(dev->type)) + + ndisc_ops_opt_addr_space(dev, icmp6_type); +} + +static inline int ndisc_redirect_opt_addr_space(struct net_device *dev, + struct neighbour *neigh, + u8 *ops_data_buf, + u8 **ops_data) { return __ndisc_opt_addr_space(dev->addr_len, - ndisc_addr_option_pad(dev->type)); + ndisc_addr_option_pad(dev->type)) + + ndisc_ops_redirect_opt_addr_space(dev, neigh, ops_data_buf, + ops_data); } +#endif static inline u8 *__ndisc_opt_addr_data(struct nd_opt_hdr *p, unsigned char addr_len, int prepad) @@ -205,6 +393,9 @@ void ndisc_send_redirect(struct sk_buff *skb, const struct in6_addr *target); int ndisc_mc_map(const struct in6_addr *addr, char *buf, struct net_device *dev, int dir); +void ndisc_update(const struct net_device *dev, struct neighbour *neigh, + const u8 *lladdr, u8 new, u32 flags, u8 icmp6_type, + struct ndisc_options *ndopts); /* * IGMP diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 44c1cefdb299..b6e9bdc610f2 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2531,7 +2531,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) if (pinfo->autoconf && in6_dev->cnf.autoconf) { struct in6_addr addr; - bool tokenized = false; + bool tokenized = false, dev_addr_generated = false; if (pinfo->prefix_len == 64) { memcpy(&addr, &pinfo->prefix, 8); @@ -2550,6 +2550,8 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) } else if (ipv6_generate_eui64(addr.s6_addr + 8, dev) && ipv6_inherit_eui64(addr.s6_addr + 8, in6_dev)) { goto put; + } else { + dev_addr_generated = true; } goto ok; } @@ -2565,6 +2567,15 @@ ok: prefered_lft); if (err) goto put; + + /* Ignore error case here because previous prefix add addr was + * successful which will be notified. + */ + ndisc_ops_prefix_rcv_add_addr(net, dev, pinfo, in6_dev, &addr, + addr_type, addr_flags, sllao, + tokenized, valid_lft, + prefered_lft, + dev_addr_generated); } inet6_prefix_notify(RTM_NEWPREFIX, in6_dev, pinfo); put: diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index a7b9468e684a..2f4afd17954b 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -172,10 +172,19 @@ static void __ndisc_fill_addr_option(struct sk_buff *skb, int type, void *data, } static inline void ndisc_fill_addr_option(struct sk_buff *skb, int type, - void *data) + void *data, u8 icmp6_type) { __ndisc_fill_addr_option(skb, type, data, skb->dev->addr_len, ndisc_addr_option_pad(skb->dev->type)); + ndisc_ops_fill_addr_option(skb->dev, skb, icmp6_type); +} + +static inline void ndisc_fill_redirect_addr_option(struct sk_buff *skb, + void *ha, + const u8 *ops_data) +{ + ndisc_fill_addr_option(skb, ND_OPT_TARGET_LL_ADDR, ha, NDISC_REDIRECT); + ndisc_ops_fill_redirect_addr_option(skb->dev, skb, ops_data); } static struct nd_opt_hdr *ndisc_next_option(struct nd_opt_hdr *cur, @@ -191,24 +200,28 @@ static struct nd_opt_hdr *ndisc_next_option(struct nd_opt_hdr *cur, return cur <= end && cur->nd_opt_type == type ? cur : NULL; } -static inline int ndisc_is_useropt(struct nd_opt_hdr *opt) +static inline int ndisc_is_useropt(const struct net_device *dev, + struct nd_opt_hdr *opt) { return opt->nd_opt_type == ND_OPT_RDNSS || - opt->nd_opt_type == ND_OPT_DNSSL; + opt->nd_opt_type == ND_OPT_DNSSL || + ndisc_ops_is_useropt(dev, opt->nd_opt_type); } -static struct nd_opt_hdr *ndisc_next_useropt(struct nd_opt_hdr *cur, +static struct nd_opt_hdr *ndisc_next_useropt(const struct net_device *dev, + struct nd_opt_hdr *cur, struct nd_opt_hdr *end) { if (!cur || !end || cur >= end) return NULL; do { cur = ((void *)cur) + (cur->nd_opt_len << 3); - } while (cur < end && !ndisc_is_useropt(cur)); - return cur <= end && ndisc_is_useropt(cur) ? cur : NULL; + } while (cur < end && !ndisc_is_useropt(dev, cur)); + return cur <= end && ndisc_is_useropt(dev, cur) ? cur : NULL; } -struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len, +struct ndisc_options *ndisc_parse_options(const struct net_device *dev, + u8 *opt, int opt_len, struct ndisc_options *ndopts) { struct nd_opt_hdr *nd_opt = (struct nd_opt_hdr *)opt; @@ -223,6 +236,8 @@ struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len, l = nd_opt->nd_opt_len << 3; if (opt_len < l || l == 0) return NULL; + if (ndisc_ops_parse_options(dev, nd_opt, ndopts)) + goto next_opt; switch (nd_opt->nd_opt_type) { case ND_OPT_SOURCE_LL_ADDR: case ND_OPT_TARGET_LL_ADDR: @@ -249,7 +264,7 @@ struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len, break; #endif default: - if (ndisc_is_useropt(nd_opt)) { + if (ndisc_is_useropt(dev, nd_opt)) { ndopts->nd_useropts_end = nd_opt; if (!ndopts->nd_useropts) ndopts->nd_useropts = nd_opt; @@ -266,6 +281,7 @@ struct ndisc_options *ndisc_parse_options(u8 *opt, int opt_len, nd_opt->nd_opt_len); } } +next_opt: opt_len -= l; nd_opt = ((void *)nd_opt) + l; } @@ -515,7 +531,8 @@ void ndisc_send_na(struct net_device *dev, const struct in6_addr *daddr, if (!dev->addr_len) inc_opt = 0; if (inc_opt) - optlen += ndisc_opt_addr_space(dev); + optlen += ndisc_opt_addr_space(dev, + NDISC_NEIGHBOUR_ADVERTISEMENT); skb = ndisc_alloc_skb(dev, sizeof(*msg) + optlen); if (!skb) @@ -534,8 +551,8 @@ void ndisc_send_na(struct net_device *dev, const struct in6_addr *daddr, if (inc_opt) ndisc_fill_addr_option(skb, ND_OPT_TARGET_LL_ADDR, - dev->dev_addr); - + dev->dev_addr, + NDISC_NEIGHBOUR_ADVERTISEMENT); ndisc_send_skb(skb, daddr, src_addr); } @@ -580,7 +597,8 @@ void ndisc_send_ns(struct net_device *dev, const struct in6_addr *solicit, if (ipv6_addr_any(saddr)) inc_opt = false; if (inc_opt) - optlen += ndisc_opt_addr_space(dev); + optlen += ndisc_opt_addr_space(dev, + NDISC_NEIGHBOUR_SOLICITATION); skb = ndisc_alloc_skb(dev, sizeof(*msg) + optlen); if (!skb) @@ -596,7 +614,8 @@ void ndisc_send_ns(struct net_device *dev, const struct in6_addr *solicit, if (inc_opt) ndisc_fill_addr_option(skb, ND_OPT_SOURCE_LL_ADDR, - dev->dev_addr); + dev->dev_addr, + NDISC_NEIGHBOUR_SOLICITATION); ndisc_send_skb(skb, daddr, saddr); } @@ -632,7 +651,7 @@ void ndisc_send_rs(struct net_device *dev, const struct in6_addr *saddr, } #endif if (send_sllao) - optlen += ndisc_opt_addr_space(dev); + optlen += ndisc_opt_addr_space(dev, NDISC_ROUTER_SOLICITATION); skb = ndisc_alloc_skb(dev, sizeof(*msg) + optlen); if (!skb) @@ -647,7 +666,8 @@ void ndisc_send_rs(struct net_device *dev, const struct in6_addr *saddr, if (send_sllao) ndisc_fill_addr_option(skb, ND_OPT_SOURCE_LL_ADDR, - dev->dev_addr); + dev->dev_addr, + NDISC_ROUTER_SOLICITATION); ndisc_send_skb(skb, daddr, saddr); } @@ -708,6 +728,15 @@ static int pndisc_is_router(const void *pkey, return ret; } +void ndisc_update(const struct net_device *dev, struct neighbour *neigh, + const u8 *lladdr, u8 new, u32 flags, u8 icmp6_type, + struct ndisc_options *ndopts) +{ + neigh_update(neigh, lladdr, new, flags); + /* report ndisc ops about neighbour update */ + ndisc_ops_update(dev, neigh, flags, icmp6_type, ndopts); +} + static void ndisc_recv_ns(struct sk_buff *skb) { struct nd_msg *msg = (struct nd_msg *)skb_transport_header(skb); @@ -744,7 +773,7 @@ static void ndisc_recv_ns(struct sk_buff *skb) return; } - if (!ndisc_parse_options(msg->opt, ndoptlen, &ndopts)) { + if (!ndisc_parse_options(dev, msg->opt, ndoptlen, &ndopts)) { ND_PRINTK(2, warn, "NS: invalid ND options\n"); return; } @@ -862,9 +891,10 @@ have_ifp: neigh = __neigh_lookup(&nd_tbl, saddr, dev, !inc || lladdr || !dev->addr_len); if (neigh) - neigh_update(neigh, lladdr, NUD_STALE, + ndisc_update(dev, neigh, lladdr, NUD_STALE, NEIGH_UPDATE_F_WEAK_OVERRIDE| - NEIGH_UPDATE_F_OVERRIDE); + NEIGH_UPDATE_F_OVERRIDE, + NDISC_NEIGHBOUR_SOLICITATION, &ndopts); if (neigh || !dev->header_ops) { ndisc_send_na(dev, saddr, &msg->target, !!is_router, true, (ifp != NULL && inc), inc); @@ -917,7 +947,7 @@ static void ndisc_recv_na(struct sk_buff *skb) idev->cnf.drop_unsolicited_na) return; - if (!ndisc_parse_options(msg->opt, ndoptlen, &ndopts)) { + if (!ndisc_parse_options(dev, msg->opt, ndoptlen, &ndopts)) { ND_PRINTK(2, warn, "NS: invalid ND option\n"); return; } @@ -973,12 +1003,13 @@ static void ndisc_recv_na(struct sk_buff *skb) goto out; } - neigh_update(neigh, lladdr, + ndisc_update(dev, neigh, lladdr, msg->icmph.icmp6_solicited ? NUD_REACHABLE : NUD_STALE, NEIGH_UPDATE_F_WEAK_OVERRIDE| (msg->icmph.icmp6_override ? NEIGH_UPDATE_F_OVERRIDE : 0)| NEIGH_UPDATE_F_OVERRIDE_ISROUTER| - (msg->icmph.icmp6_router ? NEIGH_UPDATE_F_ISROUTER : 0)); + (msg->icmph.icmp6_router ? NEIGH_UPDATE_F_ISROUTER : 0), + NDISC_NEIGHBOUR_ADVERTISEMENT, &ndopts); if ((old_flags & ~neigh->flags) & NTF_ROUTER) { /* @@ -1023,7 +1054,7 @@ static void ndisc_recv_rs(struct sk_buff *skb) goto out; /* Parse ND options */ - if (!ndisc_parse_options(rs_msg->opt, ndoptlen, &ndopts)) { + if (!ndisc_parse_options(skb->dev, rs_msg->opt, ndoptlen, &ndopts)) { ND_PRINTK(2, notice, "NS: invalid ND option, ignored\n"); goto out; } @@ -1037,10 +1068,11 @@ static void ndisc_recv_rs(struct sk_buff *skb) neigh = __neigh_lookup(&nd_tbl, saddr, skb->dev, 1); if (neigh) { - neigh_update(neigh, lladdr, NUD_STALE, + ndisc_update(skb->dev, neigh, lladdr, NUD_STALE, NEIGH_UPDATE_F_WEAK_OVERRIDE| NEIGH_UPDATE_F_OVERRIDE| - NEIGH_UPDATE_F_OVERRIDE_ISROUTER); + NEIGH_UPDATE_F_OVERRIDE_ISROUTER, + NDISC_ROUTER_SOLICITATION, &ndopts); neigh_release(neigh); } out: @@ -1141,7 +1173,7 @@ static void ndisc_router_discovery(struct sk_buff *skb) return; } - if (!ndisc_parse_options(opt, optlen, &ndopts)) { + if (!ndisc_parse_options(skb->dev, opt, optlen, &ndopts)) { ND_PRINTK(2, warn, "RA: invalid ND options\n"); return; } @@ -1335,11 +1367,12 @@ skip_linkparms: goto out; } } - neigh_update(neigh, lladdr, NUD_STALE, + ndisc_update(skb->dev, neigh, lladdr, NUD_STALE, NEIGH_UPDATE_F_WEAK_OVERRIDE| NEIGH_UPDATE_F_OVERRIDE| NEIGH_UPDATE_F_OVERRIDE_ISROUTER| - NEIGH_UPDATE_F_ISROUTER); + NEIGH_UPDATE_F_ISROUTER, + NDISC_ROUTER_ADVERTISEMENT, &ndopts); } if (!ipv6_accept_ra(in6_dev)) { @@ -1427,7 +1460,8 @@ skip_routeinfo: struct nd_opt_hdr *p; for (p = ndopts.nd_useropts; p; - p = ndisc_next_useropt(p, ndopts.nd_useropts_end)) { + p = ndisc_next_useropt(skb->dev, p, + ndopts.nd_useropts_end)) { ndisc_ra_useropt(skb, p); } } @@ -1465,7 +1499,7 @@ static void ndisc_redirect_rcv(struct sk_buff *skb) return; } - if (!ndisc_parse_options(msg->opt, ndoptlen, &ndopts)) + if (!ndisc_parse_options(skb->dev, msg->opt, ndoptlen, &ndopts)) return; if (!ndopts.nd_opts_rh) { @@ -1510,7 +1544,8 @@ void ndisc_send_redirect(struct sk_buff *skb, const struct in6_addr *target) struct dst_entry *dst; struct flowi6 fl6; int rd_len; - u8 ha_buf[MAX_ADDR_LEN], *ha = NULL; + u8 ha_buf[MAX_ADDR_LEN], *ha = NULL, + ops_data_buf[NDISC_OPS_REDIRECT_DATA_SPACE], *ops_data = NULL; int oif = l3mdev_fib_oif(dev); bool ret; @@ -1569,7 +1604,9 @@ void ndisc_send_redirect(struct sk_buff *skb, const struct in6_addr *target) memcpy(ha_buf, neigh->ha, dev->addr_len); read_unlock_bh(&neigh->lock); ha = ha_buf; - optlen += ndisc_opt_addr_space(dev); + optlen += ndisc_redirect_opt_addr_space(dev, neigh, + ops_data_buf, + &ops_data); } else read_unlock_bh(&neigh->lock); @@ -1600,7 +1637,7 @@ void ndisc_send_redirect(struct sk_buff *skb, const struct in6_addr *target) */ if (ha) - ndisc_fill_addr_option(buff, ND_OPT_TARGET_LL_ADDR, ha); + ndisc_fill_redirect_addr_option(buff, ha, ops_data); /* * build redirect option and copy skb over to the new packet. diff --git a/net/ipv6/route.c b/net/ipv6/route.c index d51a1a48b839..9e1516785dac 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -2201,7 +2201,7 @@ static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_bu * first-hop router for the specified ICMP Destination Address. */ - if (!ndisc_parse_options(msg->opt, optlen, &ndopts)) { + if (!ndisc_parse_options(skb->dev, msg->opt, optlen, &ndopts)) { net_dbg_ratelimited("rt6_redirect: invalid ND options\n"); return; } @@ -2236,12 +2236,12 @@ static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_bu * We have finally decided to accept it. */ - neigh_update(neigh, lladdr, NUD_STALE, + ndisc_update(skb->dev, neigh, lladdr, NUD_STALE, NEIGH_UPDATE_F_WEAK_OVERRIDE| NEIGH_UPDATE_F_OVERRIDE| (on_link ? 0 : (NEIGH_UPDATE_F_OVERRIDE_ISROUTER| - NEIGH_UPDATE_F_ISROUTER)) - ); + NEIGH_UPDATE_F_ISROUTER)), + NDISC_REDIRECT, &ndopts); nrt = ip6_rt_cache_alloc(rt, &msg->dest, NULL); if (!nrt) -- cgit v1.2.3 From 61d1b6a42fec61c5065f54cc62cef02b483c69fb Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 15 Jun 2016 22:47:12 +0200 Subject: bpf, maps: add release callback Add a release callback for maps that is invoked when the last reference to its struct file is gone and the struct file about to be released by vfs. The handler will be used by fd array maps. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/bpf.h | 3 ++- kernel/bpf/syscall.c | 7 ++++++- 2 files changed, 8 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 1bcae82c6cb1..29b5a1ae22cb 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -19,7 +19,8 @@ struct bpf_map; struct bpf_map_ops { /* funcs callable from userspace (via syscall) */ struct bpf_map *(*map_alloc)(union bpf_attr *attr); - void (*map_free)(struct bpf_map *); + void (*map_release)(struct bpf_map *map, struct file *map_file); + void (*map_free)(struct bpf_map *map); int (*map_get_next_key)(struct bpf_map *map, void *key, void *next_key); /* funcs callable from userspace and from eBPF programs */ diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 46ecce4b79ed..fc3adcd064b1 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -124,7 +124,12 @@ void bpf_map_put_with_uref(struct bpf_map *map) static int bpf_map_release(struct inode *inode, struct file *filp) { - bpf_map_put_with_uref(filp->private_data); + struct bpf_map *map = filp->private_data; + + if (map->ops->map_release) + map->ops->map_release(map, filp); + + bpf_map_put_with_uref(map); return 0; } -- cgit v1.2.3 From d056a788765e67773124f520159185bc89f5d1ad Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 15 Jun 2016 22:47:13 +0200 Subject: bpf, maps: extend map_fd_get_ptr arguments This patch extends map_fd_get_ptr() callback that is used by fd array maps, so that struct file pointer from the related map can be passed in. It's safe to remove map_update_elem() callback for the two maps since this is only allowed from syscall side, but not from eBPF programs for these two map types. Like in per-cpu map case, bpf_fd_array_map_update_elem() needs to be called directly here due to the extra argument. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/bpf.h | 12 +++++++++--- kernel/bpf/arraymap.c | 16 +++++++++------- kernel/bpf/syscall.c | 6 ++++++ 3 files changed, 24 insertions(+), 10 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 29b5a1ae22cb..d7b43e73fe87 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -29,8 +29,9 @@ struct bpf_map_ops { int (*map_delete_elem)(struct bpf_map *map, void *key); /* funcs called by prog_array and perf_event_array map */ - void *(*map_fd_get_ptr) (struct bpf_map *map, int fd); - void (*map_fd_put_ptr) (void *ptr); + void *(*map_fd_get_ptr)(struct bpf_map *map, struct file *map_file, + int fd); + void (*map_fd_put_ptr)(void *ptr); }; struct bpf_map { @@ -169,7 +170,7 @@ struct bpf_array { u64 bpf_tail_call(u64 ctx, u64 r2, u64 index, u64 r4, u64 r5); u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5); -void bpf_fd_array_map_clear(struct bpf_map *map); + bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp); const struct bpf_func_proto *bpf_get_trace_printk_proto(void); @@ -207,8 +208,13 @@ int bpf_percpu_hash_update(struct bpf_map *map, void *key, void *value, u64 flags); int bpf_percpu_array_update(struct bpf_map *map, void *key, void *value, u64 flags); + int bpf_stackmap_copy(struct bpf_map *map, void *key, void *value); +int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file, + void *key, void *value, u64 map_flags); +void bpf_fd_array_map_clear(struct bpf_map *map); + /* memcpy that is used with 8-byte aligned pointers, power-of-8 size and * forced to use 'long' read/writes to try to atomically copy long counters. * Best-effort only. No barriers here, since it _will_ race with concurrent diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 76d5a794e426..bfedcbdb4d84 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -328,8 +328,8 @@ static void *fd_array_map_lookup_elem(struct bpf_map *map, void *key) } /* only called from syscall */ -static int fd_array_map_update_elem(struct bpf_map *map, void *key, - void *value, u64 map_flags) +int bpf_fd_array_map_update_elem(struct bpf_map *map, struct file *map_file, + void *key, void *value, u64 map_flags) { struct bpf_array *array = container_of(map, struct bpf_array, map); void *new_ptr, *old_ptr; @@ -342,7 +342,7 @@ static int fd_array_map_update_elem(struct bpf_map *map, void *key, return -E2BIG; ufd = *(u32 *)value; - new_ptr = map->ops->map_fd_get_ptr(map, ufd); + new_ptr = map->ops->map_fd_get_ptr(map, map_file, ufd); if (IS_ERR(new_ptr)) return PTR_ERR(new_ptr); @@ -371,10 +371,12 @@ static int fd_array_map_delete_elem(struct bpf_map *map, void *key) } } -static void *prog_fd_array_get_ptr(struct bpf_map *map, int fd) +static void *prog_fd_array_get_ptr(struct bpf_map *map, + struct file *map_file, int fd) { struct bpf_array *array = container_of(map, struct bpf_array, map); struct bpf_prog *prog = bpf_prog_get(fd); + if (IS_ERR(prog)) return prog; @@ -382,6 +384,7 @@ static void *prog_fd_array_get_ptr(struct bpf_map *map, int fd) bpf_prog_put(prog); return ERR_PTR(-EINVAL); } + return prog; } @@ -407,7 +410,6 @@ static const struct bpf_map_ops prog_array_ops = { .map_free = fd_array_map_free, .map_get_next_key = array_map_get_next_key, .map_lookup_elem = fd_array_map_lookup_elem, - .map_update_elem = fd_array_map_update_elem, .map_delete_elem = fd_array_map_delete_elem, .map_fd_get_ptr = prog_fd_array_get_ptr, .map_fd_put_ptr = prog_fd_array_put_ptr, @@ -431,7 +433,8 @@ static void perf_event_array_map_free(struct bpf_map *map) fd_array_map_free(map); } -static void *perf_event_fd_array_get_ptr(struct bpf_map *map, int fd) +static void *perf_event_fd_array_get_ptr(struct bpf_map *map, + struct file *map_file, int fd) { struct perf_event *event; const struct perf_event_attr *attr; @@ -474,7 +477,6 @@ static const struct bpf_map_ops perf_event_array_ops = { .map_free = perf_event_array_map_free, .map_get_next_key = array_map_get_next_key, .map_lookup_elem = fd_array_map_lookup_elem, - .map_update_elem = fd_array_map_update_elem, .map_delete_elem = fd_array_map_delete_elem, .map_fd_get_ptr = perf_event_fd_array_get_ptr, .map_fd_put_ptr = perf_event_fd_array_put_ptr, diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index fc3adcd064b1..c23a4e9311b3 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -392,6 +392,12 @@ static int map_update_elem(union bpf_attr *attr) err = bpf_percpu_hash_update(map, key, value, attr->flags); } else if (map->map_type == BPF_MAP_TYPE_PERCPU_ARRAY) { err = bpf_percpu_array_update(map, key, value, attr->flags); + } else if (map->map_type == BPF_MAP_TYPE_PERF_EVENT_ARRAY || + map->map_type == BPF_MAP_TYPE_PROG_ARRAY) { + rcu_read_lock(); + err = bpf_fd_array_map_update_elem(map, f.file, key, value, + attr->flags); + rcu_read_unlock(); } else { rcu_read_lock(); err = map->ops->map_update_elem(map, key, value, attr->flags); -- cgit v1.2.3 From 3b1efb196eee45b2f0c4994e0c43edb5e367f620 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Wed, 15 Jun 2016 22:47:14 +0200 Subject: bpf, maps: flush own entries on perf map release The behavior of perf event arrays are quite different from all others as they are tightly coupled to perf event fds, f.e. shown recently by commit e03e7ee34fdd ("perf/bpf: Convert perf_event_array to use struct file") to make refcounting on perf event more robust. A remaining issue that the current code still has is that since additions to the perf event array take a reference on the struct file via perf_event_get() and are only released via fput() (that cleans up the perf event eventually via perf_event_release_kernel()) when the element is either manually removed from the map from user space or automatically when the last reference on the perf event map is dropped. However, this leads us to dangling struct file's when the map gets pinned after the application owning the perf event descriptor exits, and since the struct file reference will in such case only be manually dropped or via pinned file removal, it leads to the perf event living longer than necessary, consuming needlessly resources for that time. Relations between perf event fds and bpf perf event map fds can be rather complex. F.e. maps can act as demuxers among different perf event fds that can possibly be owned by different threads and based on the index selection from the program, events get dispatched to one of the per-cpu fd endpoints. One perf event fd (or, rather a per-cpu set of them) can also live in multiple perf event maps at the same time, listening for events. Also, another requirement is that perf event fds can get closed from application side after they have been attached to the perf event map, so that on exit perf event map will take care of dropping their references eventually. Likewise, when such maps are pinned, the intended behavior is that a user application does bpf_obj_get(), puts its fds in there and on exit when fd is released, they are dropped from the map again, so the map acts rather as connector endpoint. This also makes perf event maps inherently different from program arrays as described in more detail in commit c9da161c6517 ("bpf: fix clearing on persistent program array maps"). To tackle this, map entries are marked by the map struct file that added the element to the map. And when the last reference to that map struct file is released from user space, then the tracked entries are purged from the map. This is okay, because new map struct files instances resp. frontends to the anon inode are provided via bpf_map_new_fd() that is called when we invoke bpf_obj_get_user() for retrieving a pinned map, but also when an initial instance is created via map_create(). The rest is resolved by the vfs layer automatically for us by keeping reference count on the map's struct file. Any concurrent updates on the map slot are fine as well, it just means that perf_event_fd_array_release() needs to delete less of its own entires. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/bpf.h | 9 +++++ kernel/bpf/arraymap.c | 102 ++++++++++++++++++++++++++++++++++------------- kernel/trace/bpf_trace.c | 18 ++++----- 3 files changed, 91 insertions(+), 38 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index d7b43e73fe87..9adfef694a25 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -13,6 +13,7 @@ #include #include +struct perf_event; struct bpf_map; /* map is generic key/value storage optionally accesible by eBPF programs */ @@ -166,8 +167,16 @@ struct bpf_array { void __percpu *pptrs[0] __aligned(8); }; }; + #define MAX_TAIL_CALL_CNT 32 +struct bpf_event_entry { + struct perf_event *event; + struct file *perf_file; + struct file *map_file; + struct rcu_head rcu; +}; + u64 bpf_tail_call(u64 ctx, u64 r2, u64 index, u64 r4, u64 r5); u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5); diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index bfedcbdb4d84..5af30732697b 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -427,59 +427,105 @@ static int __init register_prog_array_map(void) } late_initcall(register_prog_array_map); -static void perf_event_array_map_free(struct bpf_map *map) +static struct bpf_event_entry *bpf_event_entry_gen(struct file *perf_file, + struct file *map_file) { - bpf_fd_array_map_clear(map); - fd_array_map_free(map); + struct bpf_event_entry *ee; + + ee = kzalloc(sizeof(*ee), GFP_KERNEL); + if (ee) { + ee->event = perf_file->private_data; + ee->perf_file = perf_file; + ee->map_file = map_file; + } + + return ee; +} + +static void __bpf_event_entry_free(struct rcu_head *rcu) +{ + struct bpf_event_entry *ee; + + ee = container_of(rcu, struct bpf_event_entry, rcu); + fput(ee->perf_file); + kfree(ee); +} + +static void bpf_event_entry_free_rcu(struct bpf_event_entry *ee) +{ + call_rcu(&ee->rcu, __bpf_event_entry_free); } static void *perf_event_fd_array_get_ptr(struct bpf_map *map, struct file *map_file, int fd) { - struct perf_event *event; const struct perf_event_attr *attr; - struct file *file; + struct bpf_event_entry *ee; + struct perf_event *event; + struct file *perf_file; - file = perf_event_get(fd); - if (IS_ERR(file)) - return file; + perf_file = perf_event_get(fd); + if (IS_ERR(perf_file)) + return perf_file; - event = file->private_data; + event = perf_file->private_data; + ee = ERR_PTR(-EINVAL); attr = perf_event_attrs(event); - if (IS_ERR(attr)) - goto err; - - if (attr->inherit) - goto err; - - if (attr->type == PERF_TYPE_RAW) - return file; - - if (attr->type == PERF_TYPE_HARDWARE) - return file; + if (IS_ERR(attr) || attr->inherit) + goto err_out; + + switch (attr->type) { + case PERF_TYPE_SOFTWARE: + if (attr->config != PERF_COUNT_SW_BPF_OUTPUT) + goto err_out; + /* fall-through */ + case PERF_TYPE_RAW: + case PERF_TYPE_HARDWARE: + ee = bpf_event_entry_gen(perf_file, map_file); + if (ee) + return ee; + ee = ERR_PTR(-ENOMEM); + /* fall-through */ + default: + break; + } - if (attr->type == PERF_TYPE_SOFTWARE && - attr->config == PERF_COUNT_SW_BPF_OUTPUT) - return file; -err: - fput(file); - return ERR_PTR(-EINVAL); +err_out: + fput(perf_file); + return ee; } static void perf_event_fd_array_put_ptr(void *ptr) { - fput((struct file *)ptr); + bpf_event_entry_free_rcu(ptr); +} + +static void perf_event_fd_array_release(struct bpf_map *map, + struct file *map_file) +{ + struct bpf_array *array = container_of(map, struct bpf_array, map); + struct bpf_event_entry *ee; + int i; + + rcu_read_lock(); + for (i = 0; i < array->map.max_entries; i++) { + ee = READ_ONCE(array->ptrs[i]); + if (ee && ee->map_file == map_file) + fd_array_map_delete_elem(map, &i); + } + rcu_read_unlock(); } static const struct bpf_map_ops perf_event_array_ops = { .map_alloc = fd_array_map_alloc, - .map_free = perf_event_array_map_free, + .map_free = fd_array_map_free, .map_get_next_key = array_map_get_next_key, .map_lookup_elem = fd_array_map_lookup_elem, .map_delete_elem = fd_array_map_delete_elem, .map_fd_get_ptr = perf_event_fd_array_get_ptr, .map_fd_put_ptr = perf_event_fd_array_put_ptr, + .map_release = perf_event_fd_array_release, }; static struct bpf_map_type_list perf_event_array_type __read_mostly = { diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 720b7bb01d43..037ea6ea3cb2 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -192,18 +192,17 @@ static u64 bpf_perf_event_read(u64 r1, u64 index, u64 r3, u64 r4, u64 r5) { struct bpf_map *map = (struct bpf_map *) (unsigned long) r1; struct bpf_array *array = container_of(map, struct bpf_array, map); + struct bpf_event_entry *ee; struct perf_event *event; - struct file *file; if (unlikely(index >= array->map.max_entries)) return -E2BIG; - file = READ_ONCE(array->ptrs[index]); - if (unlikely(!file)) + ee = READ_ONCE(array->ptrs[index]); + if (unlikely(!ee)) return -ENOENT; - event = file->private_data; - + event = ee->event; /* make sure event is local and doesn't have pmu::count */ if (event->oncpu != smp_processor_id() || event->pmu->count) @@ -233,8 +232,8 @@ static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) u64 index = flags & BPF_F_INDEX_MASK; void *data = (void *) (long) r4; struct perf_sample_data sample_data; + struct bpf_event_entry *ee; struct perf_event *event; - struct file *file; struct perf_raw_record raw = { .size = size, .data = data, @@ -247,12 +246,11 @@ static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) if (unlikely(index >= array->map.max_entries)) return -E2BIG; - file = READ_ONCE(array->ptrs[index]); - if (unlikely(!file)) + ee = READ_ONCE(array->ptrs[index]); + if (unlikely(!ee)) return -ENOENT; - event = file->private_data; - + event = ee->event; if (unlikely(event->attr.type != PERF_TYPE_SOFTWARE || event->attr.config != PERF_COUNT_SW_BPF_OUTPUT)) return -EINVAL; -- cgit v1.2.3 From cecbc5563a02289164fa6379130243cbe08b2dd6 Mon Sep 17 00:00:00 2001 From: Vincent Palatin Date: Wed, 15 Jun 2016 11:32:21 -0700 Subject: net: stmmac: allow to split suspend/resume from init/exit callbacks Let the stmmac platform drivers provide dedicated suspend and resume callbacks rather than always re-using the init and exits callbacks. If the driver does not provide the suspend or resume callback, we fall back to the old behavior trying to use exit or init. This allows a specific platform to perform only a partial power-down on suspend if Wake-on-Lan is enabled but always perform the full shutdown sequence if the module is unloaded. Signed-off-by: Vincent Palatin Signed-off-by: David S. Miller --- drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 8 ++++++-- include/linux/stmmac.h | 2 ++ 2 files changed, 8 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 409db913b117..a96714d34560 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -411,7 +411,9 @@ static int stmmac_pltfr_suspend(struct device *dev) struct platform_device *pdev = to_platform_device(dev); ret = stmmac_suspend(dev); - if (priv->plat->exit) + if (priv->plat->suspend) + priv->plat->suspend(pdev, priv->plat->bsp_priv); + else if (priv->plat->exit) priv->plat->exit(pdev, priv->plat->bsp_priv); return ret; @@ -430,7 +432,9 @@ static int stmmac_pltfr_resume(struct device *dev) struct stmmac_priv *priv = netdev_priv(ndev); struct platform_device *pdev = to_platform_device(dev); - if (priv->plat->init) + if (priv->plat->resume) + priv->plat->resume(pdev, priv->plat->bsp_priv); + else if (priv->plat->init) priv->plat->init(pdev, priv->plat->bsp_priv); return stmmac_resume(dev); diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h index ffdaca9c01af..0507dbfbf63c 100644 --- a/include/linux/stmmac.h +++ b/include/linux/stmmac.h @@ -135,6 +135,8 @@ struct plat_stmmacenet_data { void (*bus_setup)(void __iomem *ioaddr); int (*init)(struct platform_device *pdev, void *priv); void (*exit)(struct platform_device *pdev, void *priv); + void (*suspend)(struct platform_device *pdev, void *priv); + void (*resume)(struct platform_device *pdev, void *priv); void *bsp_priv; struct stmmac_axi *axi; int has_gmac4; -- cgit v1.2.3 From 7c46a640de6fcc4f35d0702710356a024eadf68f Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:21:00 -0700 Subject: net: Merge VXLAN and GENEVE push notifiers into a single notifier This patch merges the notifiers for VXLAN and GENEVE into a single UDP tunnel notifier. The idea is that we will want to only have to make one notifier call to receive the list of ports for VXLAN and GENEVE tunnels that need to be offloaded. In addition we add a new set of ndo functions named ndo_udp_tunnel_add and ndo_udp_tunnel_del that are meant to allow us to track the tunnel meta-data such as port and address family as tunnels are added and removed. The tunnel meta-data is now transported in a structure named udp_tunnel_info which for now carries the type, address family, and port number. In the future this could be updated so that we can include a tuple of values including things such as the destination IP address and other fields. I also ended up going with a naming scheme that consisted of using the prefix udp_tunnel on function names. I applied this to the notifier and ndo ops as well so that it hopefully points to the fact that these are primarily used in the udp_tunnel functions. Signed-off-by: Alexander Duyck Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller --- drivers/net/geneve.c | 2 +- drivers/net/vxlan.c | 2 +- include/linux/netdevice.h | 22 ++++++++++++++++++++-- include/net/geneve.h | 3 +-- include/net/udp_tunnel.h | 6 ++++++ include/net/vxlan.h | 4 ++-- net/ipv4/udp_tunnel.c | 10 ++++++++++ 7 files changed, 41 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index af6f676ca444..aa61708bea69 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -1497,7 +1497,7 @@ static int geneve_netdevice_event(struct notifier_block *unused, { struct net_device *dev = netdev_notifier_info_to_dev(ptr); - if (event == NETDEV_OFFLOAD_PUSH_GENEVE) + if (event == NETDEV_UDP_TUNNEL_PUSH_INFO) geneve_push_rx_ports(dev); return NOTIFY_DONE; diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c index adbd979da22d..31aeec967175 100644 --- a/drivers/net/vxlan.c +++ b/drivers/net/vxlan.c @@ -3239,7 +3239,7 @@ static int vxlan_netdevice_event(struct notifier_block *unused, if (event == NETDEV_UNREGISTER) vxlan_handle_lowerdev_unregister(vn, dev); - else if (event == NETDEV_OFFLOAD_PUSH_VXLAN) + else if (event == NETDEV_UDP_TUNNEL_PUSH_INFO) vxlan_push_rx_ports(dev); return NOTIFY_DONE; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 890158e99159..577d2a1814b1 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -61,6 +61,8 @@ struct wireless_dev; /* 802.15.4 specific */ struct wpan_dev; struct mpls_dev; +/* UDP Tunnel offloads */ +struct udp_tunnel_info; void netdev_set_default_ethtool_ops(struct net_device *dev, const struct ethtool_ops *ops); @@ -1050,6 +1052,19 @@ struct tc_to_netdev { * address family that vxlan is not listening to anymore. The operation * is protected by the vxlan_net->sock_lock. * + * void (*ndo_udp_tunnel_add)(struct net_device *dev, + * struct udp_tunnel_info *ti); + * Called by UDP tunnel to notify a driver about the UDP port and socket + * address family that a UDP tunnel is listnening to. It is called only + * when a new port starts listening. The operation is protected by the + * RTNL. + * + * void (*ndo_udp_tunnel_del)(struct net_device *dev, + * struct udp_tunnel_info *ti); + * Called by UDP tunnel to notify the driver about a UDP port and socket + * address family that the UDP tunnel is not listening to anymore. The + * operation is protected by the RTNL. + * * void* (*ndo_dfwd_add_station)(struct net_device *pdev, * struct net_device *dev) * Called by upper layer devices to accelerate switching or other @@ -1269,6 +1284,10 @@ struct net_device_ops { void (*ndo_del_geneve_port)(struct net_device *dev, sa_family_t sa_family, __be16 port); + void (*ndo_udp_tunnel_add)(struct net_device *dev, + struct udp_tunnel_info *ti); + void (*ndo_udp_tunnel_del)(struct net_device *dev, + struct udp_tunnel_info *ti); void* (*ndo_dfwd_add_station)(struct net_device *pdev, struct net_device *dev); void (*ndo_dfwd_del_station)(struct net_device *pdev, @@ -2255,8 +2274,7 @@ struct netdev_lag_lower_state_info { #define NETDEV_BONDING_INFO 0x0019 #define NETDEV_PRECHANGEUPPER 0x001A #define NETDEV_CHANGELOWERSTATE 0x001B -#define NETDEV_OFFLOAD_PUSH_VXLAN 0x001C -#define NETDEV_OFFLOAD_PUSH_GENEVE 0x001D +#define NETDEV_UDP_TUNNEL_PUSH_INFO 0x001C int register_netdevice_notifier(struct notifier_block *nb); int unregister_netdevice_notifier(struct notifier_block *nb); diff --git a/include/net/geneve.h b/include/net/geneve.h index f8aff18d6702..3410c4b5a382 100644 --- a/include/net/geneve.h +++ b/include/net/geneve.h @@ -61,8 +61,7 @@ struct genevehdr { static inline void geneve_get_rx_port(struct net_device *netdev) { - ASSERT_RTNL(); - call_netdevice_notifiers(NETDEV_OFFLOAD_PUSH_GENEVE, netdev); + udp_tunnel_get_rx_info(netdev); } #ifdef CONFIG_INET diff --git a/include/net/udp_tunnel.h b/include/net/udp_tunnel.h index 71afbea873a0..1c9408a04213 100644 --- a/include/net/udp_tunnel.h +++ b/include/net/udp_tunnel.h @@ -117,6 +117,12 @@ void udp_tunnel_push_rx_port(struct net_device *dev, struct socket *sock, void udp_tunnel_notify_add_rx_port(struct socket *sock, unsigned short type); void udp_tunnel_notify_del_rx_port(struct socket *sock, unsigned short type); +static inline void udp_tunnel_get_rx_info(struct net_device *dev) +{ + ASSERT_RTNL(); + call_netdevice_notifiers(NETDEV_UDP_TUNNEL_PUSH_INFO, dev); +} + /* Transmit the skb using UDP encapsulation. */ void udp_tunnel_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb, __be32 src, __be32 dst, __u8 tos, __u8 ttl, diff --git a/include/net/vxlan.h b/include/net/vxlan.h index 7d944941f32f..c62e2ed1c3af 100644 --- a/include/net/vxlan.h +++ b/include/net/vxlan.h @@ -4,6 +4,7 @@ #include #include #include +#include /* VXLAN protocol (RFC 7348) header: * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ @@ -390,8 +391,7 @@ static inline __be32 vxlan_compute_rco(unsigned int start, unsigned int offset) static inline void vxlan_get_rx_port(struct net_device *netdev) { - ASSERT_RTNL(); - call_netdevice_notifiers(NETDEV_OFFLOAD_PUSH_VXLAN, netdev); + udp_tunnel_get_rx_info(netdev); } static inline unsigned short vxlan_get_sk_family(struct vxlan_sock *vs) diff --git a/net/ipv4/udp_tunnel.c b/net/ipv4/udp_tunnel.c index 8174753e6494..683e494d9000 100644 --- a/net/ipv4/udp_tunnel.c +++ b/net/ipv4/udp_tunnel.c @@ -79,6 +79,11 @@ EXPORT_SYMBOL_GPL(setup_udp_tunnel_sock); static void __udp_tunnel_push_rx_port(struct net_device *dev, struct udp_tunnel_info *ti) { + if (dev->netdev_ops->ndo_udp_tunnel_add) { + dev->netdev_ops->ndo_udp_tunnel_add(dev, ti); + return; + } + switch (ti->type) { case UDP_TUNNEL_TYPE_VXLAN: if (!dev->netdev_ops->ndo_add_vxlan_port) @@ -137,6 +142,11 @@ EXPORT_SYMBOL_GPL(udp_tunnel_notify_add_rx_port); static void __udp_tunnel_pull_rx_port(struct net_device *dev, struct udp_tunnel_info *ti) { + if (dev->netdev_ops->ndo_udp_tunnel_del) { + dev->netdev_ops->ndo_udp_tunnel_del(dev, ti); + return; + } + switch (ti->type) { case UDP_TUNNEL_TYPE_VXLAN: if (!dev->netdev_ops->ndo_del_vxlan_port) -- cgit v1.2.3 From 1938ee1fd3de74d761a60806b048df652666afec Mon Sep 17 00:00:00 2001 From: Alexander Duyck Date: Thu, 16 Jun 2016 12:23:12 -0700 Subject: net: Remove deprecated tunnel specific UDP offload functions Now that we have all the drivers using udp_tunnel_get_rx_ports, ndo_add_udp_enc_rx_port, and ndo_del_udp_enc_rx_port we can drop the function calls that were specific to VXLAN and GENEVE. Signed-off-by: Alexander Duyck Acked-by: Hannes Frederic Sowa Signed-off-by: David S. Miller --- include/linux/netdevice.h | 38 ----------------------- include/net/geneve.h | 5 --- include/net/vxlan.h | 5 --- net/ipv4/udp_tunnel.c | 79 +++++++++-------------------------------------- 4 files changed, 14 insertions(+), 113 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 577d2a1814b1..e84d9d23c2d5 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1026,32 +1026,6 @@ struct tc_to_netdev { * not implement this, it is assumed that the hw is not able to have * multiple net devices on single physical port. * - * void (*ndo_add_vxlan_port)(struct net_device *dev, - * sa_family_t sa_family, __be16 port); - * Called by vxlan to notify a driver about the UDP port and socket - * address family that vxlan is listening to. It is called only when - * a new port starts listening. The operation is protected by the - * vxlan_net->sock_lock. - * - * void (*ndo_add_geneve_port)(struct net_device *dev, - * sa_family_t sa_family, __be16 port); - * Called by geneve to notify a driver about the UDP port and socket - * address family that geneve is listnening to. It is called only when - * a new port starts listening. The operation is protected by the - * geneve_net->sock_lock. - * - * void (*ndo_del_geneve_port)(struct net_device *dev, - * sa_family_t sa_family, __be16 port); - * Called by geneve to notify the driver about a UDP port and socket - * address family that geneve is not listening to anymore. The operation - * is protected by the geneve_net->sock_lock. - * - * void (*ndo_del_vxlan_port)(struct net_device *dev, - * sa_family_t sa_family, __be16 port); - * Called by vxlan to notify the driver about a UDP port and socket - * address family that vxlan is not listening to anymore. The operation - * is protected by the vxlan_net->sock_lock. - * * void (*ndo_udp_tunnel_add)(struct net_device *dev, * struct udp_tunnel_info *ti); * Called by UDP tunnel to notify a driver about the UDP port and socket @@ -1272,18 +1246,6 @@ struct net_device_ops { struct netdev_phys_item_id *ppid); int (*ndo_get_phys_port_name)(struct net_device *dev, char *name, size_t len); - void (*ndo_add_vxlan_port)(struct net_device *dev, - sa_family_t sa_family, - __be16 port); - void (*ndo_del_vxlan_port)(struct net_device *dev, - sa_family_t sa_family, - __be16 port); - void (*ndo_add_geneve_port)(struct net_device *dev, - sa_family_t sa_family, - __be16 port); - void (*ndo_del_geneve_port)(struct net_device *dev, - sa_family_t sa_family, - __be16 port); void (*ndo_udp_tunnel_add)(struct net_device *dev, struct udp_tunnel_info *ti); void (*ndo_udp_tunnel_del)(struct net_device *dev, diff --git a/include/net/geneve.h b/include/net/geneve.h index 3410c4b5a382..ec0327d4331b 100644 --- a/include/net/geneve.h +++ b/include/net/geneve.h @@ -59,11 +59,6 @@ struct genevehdr { struct geneve_opt options[]; }; -static inline void geneve_get_rx_port(struct net_device *netdev) -{ - udp_tunnel_get_rx_info(netdev); -} - #ifdef CONFIG_INET struct net_device *geneve_dev_create_fb(struct net *net, const char *name, u8 name_assign_type, u16 dst_port); diff --git a/include/net/vxlan.h b/include/net/vxlan.h index c62e2ed1c3af..b96d0360c095 100644 --- a/include/net/vxlan.h +++ b/include/net/vxlan.h @@ -389,11 +389,6 @@ static inline __be32 vxlan_compute_rco(unsigned int start, unsigned int offset) return vni_field; } -static inline void vxlan_get_rx_port(struct net_device *netdev) -{ - udp_tunnel_get_rx_info(netdev); -} - static inline unsigned short vxlan_get_sk_family(struct vxlan_sock *vs) { return vs->sock->sk->sk_family; diff --git a/net/ipv4/udp_tunnel.c b/net/ipv4/udp_tunnel.c index 683e494d9000..58bd39fb14b4 100644 --- a/net/ipv4/udp_tunnel.c +++ b/net/ipv4/udp_tunnel.c @@ -76,47 +76,20 @@ void setup_udp_tunnel_sock(struct net *net, struct socket *sock, } EXPORT_SYMBOL_GPL(setup_udp_tunnel_sock); -static void __udp_tunnel_push_rx_port(struct net_device *dev, - struct udp_tunnel_info *ti) -{ - if (dev->netdev_ops->ndo_udp_tunnel_add) { - dev->netdev_ops->ndo_udp_tunnel_add(dev, ti); - return; - } - - switch (ti->type) { - case UDP_TUNNEL_TYPE_VXLAN: - if (!dev->netdev_ops->ndo_add_vxlan_port) - break; - - dev->netdev_ops->ndo_add_vxlan_port(dev, - ti->sa_family, - ti->port); - break; - case UDP_TUNNEL_TYPE_GENEVE: - if (!dev->netdev_ops->ndo_add_geneve_port) - break; - - dev->netdev_ops->ndo_add_geneve_port(dev, - ti->sa_family, - ti->port); - break; - default: - break; - } -} - void udp_tunnel_push_rx_port(struct net_device *dev, struct socket *sock, unsigned short type) { struct sock *sk = sock->sk; struct udp_tunnel_info ti; + if (!dev->netdev_ops->ndo_udp_tunnel_add) + return; + ti.type = type; ti.sa_family = sk->sk_family; ti.port = inet_sk(sk)->inet_sport; - __udp_tunnel_push_rx_port(dev, &ti); + dev->netdev_ops->ndo_udp_tunnel_add(dev, &ti); } EXPORT_SYMBOL_GPL(udp_tunnel_push_rx_port); @@ -133,42 +106,15 @@ void udp_tunnel_notify_add_rx_port(struct socket *sock, unsigned short type) ti.port = inet_sk(sk)->inet_sport; rcu_read_lock(); - for_each_netdev_rcu(net, dev) - __udp_tunnel_push_rx_port(dev, &ti); + for_each_netdev_rcu(net, dev) { + if (!dev->netdev_ops->ndo_udp_tunnel_add) + continue; + dev->netdev_ops->ndo_udp_tunnel_add(dev, &ti); + } rcu_read_unlock(); } EXPORT_SYMBOL_GPL(udp_tunnel_notify_add_rx_port); -static void __udp_tunnel_pull_rx_port(struct net_device *dev, - struct udp_tunnel_info *ti) -{ - if (dev->netdev_ops->ndo_udp_tunnel_del) { - dev->netdev_ops->ndo_udp_tunnel_del(dev, ti); - return; - } - - switch (ti->type) { - case UDP_TUNNEL_TYPE_VXLAN: - if (!dev->netdev_ops->ndo_del_vxlan_port) - break; - - dev->netdev_ops->ndo_del_vxlan_port(dev, - ti->sa_family, - ti->port); - break; - case UDP_TUNNEL_TYPE_GENEVE: - if (!dev->netdev_ops->ndo_del_geneve_port) - break; - - dev->netdev_ops->ndo_del_geneve_port(dev, - ti->sa_family, - ti->port); - break; - default: - break; - } -} - /* Notify netdevs that UDP port is no more listening */ void udp_tunnel_notify_del_rx_port(struct socket *sock, unsigned short type) { @@ -182,8 +128,11 @@ void udp_tunnel_notify_del_rx_port(struct socket *sock, unsigned short type) ti.port = inet_sk(sk)->inet_sport; rcu_read_lock(); - for_each_netdev_rcu(net, dev) - __udp_tunnel_pull_rx_port(dev, &ti); + for_each_netdev_rcu(net, dev) { + if (!dev->netdev_ops->ndo_udp_tunnel_del) + continue; + dev->netdev_ops->ndo_udp_tunnel_del(dev, &ti); + } rcu_read_unlock(); } EXPORT_SYMBOL_GPL(udp_tunnel_notify_del_rx_port); -- cgit v1.2.3 From b1cadc1a0949c82ff7fcb15603e3caf2d32ff9f6 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 18 Jun 2016 21:52:02 -0700 Subject: ipv6: icmp: add a force_saddr param to icmp6_send() SIT or GRE tunnels might want to translate an IPV4 address into a v4mapped one when translating ICMP to ICMPv6. This patch adds the parameter to icmp6_send() but does not change icmpv6_send() signature. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/icmpv6.h | 3 ++- net/ipv6/icmp.c | 7 +++++-- net/ipv6/ip6_icmp.c | 2 +- 3 files changed, 8 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index 630f45335c73..432611a297fb 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -14,7 +14,8 @@ static inline struct icmp6hdr *icmp6_hdr(const struct sk_buff *skb) #if IS_ENABLED(CONFIG_IPV6) extern void icmpv6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info); -typedef void ip6_icmp_send_t(struct sk_buff *skb, u8 type, u8 code, __u32 info); +typedef void ip6_icmp_send_t(struct sk_buff *skb, u8 type, u8 code, __u32 info, + const struct in6_addr *force_saddr); extern int inet6_register_icmp_sender(ip6_icmp_send_t *fn); extern int inet6_unregister_icmp_sender(ip6_icmp_send_t *fn); diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index e32a72fb9982..6c57e6e90301 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -388,7 +388,8 @@ relookup_failed: /* * Send an ICMP message in response to a packet in error */ -static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) +static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info, + const struct in6_addr *force_saddr) { struct net *net = dev_net(skb->dev); struct inet6_dev *idev = NULL; @@ -475,6 +476,8 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) memset(&fl6, 0, sizeof(fl6)); fl6.flowi6_proto = IPPROTO_ICMPV6; fl6.daddr = hdr->saddr; + if (force_saddr) + saddr = force_saddr; if (saddr) fl6.saddr = *saddr; fl6.flowi6_mark = mark; @@ -551,7 +554,7 @@ out: */ void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos) { - icmp6_send(skb, ICMPV6_PARAMPROB, code, pos); + icmp6_send(skb, ICMPV6_PARAMPROB, code, pos, NULL); kfree_skb(skb); } diff --git a/net/ipv6/ip6_icmp.c b/net/ipv6/ip6_icmp.c index 14dacc544c3e..713676f14a0e 100644 --- a/net/ipv6/ip6_icmp.c +++ b/net/ipv6/ip6_icmp.c @@ -39,7 +39,7 @@ void icmpv6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) if (!send) goto out; - send(skb, type, code, info); + send(skb, type, code, info, NULL); out: rcu_read_unlock(); } -- cgit v1.2.3 From 5fbba8ac9358f1e796c8aedcccc3487364643723 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 18 Jun 2016 21:52:03 -0700 Subject: ip6: move ipip6_err_gen_icmpv6_unreach() We want to use this helper from GRE as well, so this is the time to move it in net/ipv6/icmp.c Also add a @nhs parameter, since SIT and GRE have different values for the header(s) to skip. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/icmpv6.h | 1 + net/ipv6/icmp.c | 39 +++++++++++++++++++++++++++++++++++++++ net/ipv6/sit.c | 38 +------------------------------------- 3 files changed, 41 insertions(+), 37 deletions(-) (limited to 'include/linux') diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index 432611a297fb..9796481edbdb 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -18,6 +18,7 @@ typedef void ip6_icmp_send_t(struct sk_buff *skb, u8 type, u8 code, __u32 info, const struct in6_addr *force_saddr); extern int inet6_register_icmp_sender(ip6_icmp_send_t *fn); extern int inet6_unregister_icmp_sender(ip6_icmp_send_t *fn); +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs); #else diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 6c57e6e90301..07bc63c23712 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -558,6 +558,45 @@ void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos) kfree_skb(skb); } +/* Generate icmpv6 with type/code ICMPV6_DEST_UNREACH/ICMPV6_ADDR_UNREACH + * if sufficient data bytes are available + * @nhs is the size of the tunnel header(s) : + * Either an IPv4 header for SIT encap + * an IPv4 header + GRE header for GRE encap + */ +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs) +{ + struct rt6_info *rt; + struct sk_buff *skb2; + + if (!pskb_may_pull(skb, nhs + sizeof(struct ipv6hdr) + 8)) + return 1; + + skb2 = skb_clone(skb, GFP_ATOMIC); + + if (!skb2) + return 1; + + skb_dst_drop(skb2); + skb_pull(skb2, nhs); + skb_reset_network_header(skb2); + + rt = rt6_lookup(dev_net(skb->dev), &ipv6_hdr(skb2)->saddr, NULL, 0, 0); + + if (rt && rt->dst.dev) + skb2->dev = rt->dst.dev; + + icmpv6_send(skb2, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0); + + if (rt) + ip6_rt_put(rt); + + kfree_skb(skb2); + + return 0; +} +EXPORT_SYMBOL(ip6_err_gen_icmpv6_unreach); + static void icmpv6_echo_reply(struct sk_buff *skb) { struct net *net = dev_net(skb->dev); diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index d9f2bd6ef72d..78e84d6793ee 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -479,42 +479,6 @@ static void ipip6_tunnel_uninit(struct net_device *dev) dev_put(dev); } -/* Generate icmpv6 with type/code ICMPV6_DEST_UNREACH/ICMPV6_ADDR_UNREACH - * if sufficient data bytes are available - */ -static int ipip6_err_gen_icmpv6_unreach(struct sk_buff *skb) -{ - int ihl = ((const struct iphdr *)skb->data)->ihl*4; - struct rt6_info *rt; - struct sk_buff *skb2; - - if (!pskb_may_pull(skb, ihl + sizeof(struct ipv6hdr) + 8)) - return 1; - - skb2 = skb_clone(skb, GFP_ATOMIC); - - if (!skb2) - return 1; - - skb_dst_drop(skb2); - skb_pull(skb2, ihl); - skb_reset_network_header(skb2); - - rt = rt6_lookup(dev_net(skb->dev), &ipv6_hdr(skb2)->saddr, NULL, 0, 0); - - if (rt && rt->dst.dev) - skb2->dev = rt->dst.dev; - - icmpv6_send(skb2, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0); - - if (rt) - ip6_rt_put(rt); - - kfree_skb(skb2); - - return 0; -} - static int ipip6_err(struct sk_buff *skb, u32 info) { const struct iphdr *iph = (const struct iphdr *)skb->data; @@ -575,7 +539,7 @@ static int ipip6_err(struct sk_buff *skb, u32 info) goto out; err = 0; - if (!ipip6_err_gen_icmpv6_unreach(skb)) + if (!ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4)) goto out; if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED) -- cgit v1.2.3 From 2d7a3b276be2d032a6c1a48ced87a474327ee3d3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 18 Jun 2016 21:52:04 -0700 Subject: ipv6: translate ICMP_TIME_EXCEEDED to ICMPV6_TIME_EXCEED For better traceroute/mtr support for SIT and GRE tunnels, we translate IPV4 ICMP ICMP_TIME_EXCEEDED to ICMPV6_TIME_EXCEED We also have to translate the IPv4 source IP address of ICMP message to IPv6 v4mapped. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/linux/icmpv6.h | 2 +- net/ipv6/icmp.c | 12 +++++++++--- net/ipv6/sit.c | 6 +++--- 3 files changed, 13 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index 9796481edbdb..97ae98071a03 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -18,7 +18,7 @@ typedef void ip6_icmp_send_t(struct sk_buff *skb, u8 type, u8 code, __u32 info, const struct in6_addr *force_saddr); extern int inet6_register_icmp_sender(ip6_icmp_send_t *fn); extern int inet6_unregister_icmp_sender(ip6_icmp_send_t *fn); -int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs); +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type); #else diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 07bc63c23712..867aebc34248 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -564,8 +564,9 @@ void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos) * Either an IPv4 header for SIT encap * an IPv4 header + GRE header for GRE encap */ -int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs) +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type) { + struct in6_addr temp_saddr; struct rt6_info *rt; struct sk_buff *skb2; @@ -586,8 +587,13 @@ int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs) if (rt && rt->dst.dev) skb2->dev = rt->dst.dev; - icmpv6_send(skb2, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, 0); - + ipv6_addr_set_v4mapped(ip_hdr(skb)->saddr, &temp_saddr); + if (type == ICMP_TIME_EXCEEDED) + icmp6_send(skb2, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, + 0, &temp_saddr); + else + icmp6_send(skb2, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, + 0, &temp_saddr); if (rt) ip6_rt_put(rt); diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 78e84d6793ee..d7a36114eb50 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -535,11 +535,11 @@ static int ipip6_err(struct sk_buff *skb, u32 info) goto out; } - if (t->parms.iph.daddr == 0) + err = 0; + if (!ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4, type)) goto out; - err = 0; - if (!ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4)) + if (t->parms.iph.daddr == 0) goto out; if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED) -- cgit v1.2.3 From 20e1954fe238dbe5f8d3a979e593fe352bd703cf Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 18 Jun 2016 21:52:06 -0700 Subject: ipv6: RFC 4884 partial support for SIT/GRE tunnels When receiving an ICMPv4 message containing extensions as defined in RFC 4884, and translating it to ICMPv6 at SIT or GRE tunnel, we need some extra manipulation in order to properly forward the extensions. This patch only takes care of Time Exceeded messages as they are the ones that typically carry information from various routers in a fabric during a traceroute session. It also avoids complex skb logic if the data_len is not a multiple of 8. RFC states : The "original datagram" field MUST contain at least 128 octets. If the original datagram did not contain 128 octets, the "original datagram" field MUST be zero padded to 128 octets. In practice routers use 128 bytes of original datagram, not more. Initial translation was added in commit ca15a078bd90 ("sit: generate icmpv6 error when receiving icmpv4 error") Signed-off-by: Eric Dumazet Cc: Oussama Ghorbel Signed-off-by: David S. Miller --- include/linux/icmpv6.h | 3 ++- include/uapi/linux/icmp.h | 1 + net/ipv4/ip_gre.c | 5 ++++- net/ipv6/icmp.c | 28 ++++++++++++++++++++++++---- net/ipv6/sit.c | 4 +++- 5 files changed, 34 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/include/linux/icmpv6.h b/include/linux/icmpv6.h index 97ae98071a03..57086e9fc64c 100644 --- a/include/linux/icmpv6.h +++ b/include/linux/icmpv6.h @@ -18,7 +18,8 @@ typedef void ip6_icmp_send_t(struct sk_buff *skb, u8 type, u8 code, __u32 info, const struct in6_addr *force_saddr); extern int inet6_register_icmp_sender(ip6_icmp_send_t *fn); extern int inet6_unregister_icmp_sender(ip6_icmp_send_t *fn); -int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type); +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type, + unsigned int data_len); #else diff --git a/include/uapi/linux/icmp.h b/include/uapi/linux/icmp.h index 16fff055f734..fddd9d736284 100644 --- a/include/uapi/linux/icmp.h +++ b/include/uapi/linux/icmp.h @@ -79,6 +79,7 @@ struct icmphdr { __be16 __unused; __be16 mtu; } frag; + __u8 reserved[4]; } un; }; diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index ab4cff8e563d..8eec78f53f9e 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -144,6 +144,7 @@ static void ipgre_err(struct sk_buff *skb, u32 info, const struct iphdr *iph; const int type = icmp_hdr(skb)->type; const int code = icmp_hdr(skb)->code; + unsigned int data_len = 0; struct ip_tunnel *t; switch (type) { @@ -169,6 +170,7 @@ static void ipgre_err(struct sk_buff *skb, u32 info, case ICMP_TIME_EXCEEDED: if (code != ICMP_EXC_TTL) return; + data_len = icmp_hdr(skb)->un.reserved[1] * 4; /* RFC 4884 4.1 */ break; case ICMP_REDIRECT: @@ -189,7 +191,8 @@ static void ipgre_err(struct sk_buff *skb, u32 info, #if IS_ENABLED(CONFIG_IPV6) if (tpi->proto == htons(ETH_P_IPV6) && - !ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4 + tpi->hdr_len, type)) + !ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4 + tpi->hdr_len, + type, data_len)) return; #endif diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 867aebc34248..fd11f5856ce8 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -564,16 +564,22 @@ void icmpv6_param_prob(struct sk_buff *skb, u8 code, int pos) * Either an IPv4 header for SIT encap * an IPv4 header + GRE header for GRE encap */ -int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type) +int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type, + unsigned int data_len) { struct in6_addr temp_saddr; struct rt6_info *rt; struct sk_buff *skb2; + u32 info = 0; if (!pskb_may_pull(skb, nhs + sizeof(struct ipv6hdr) + 8)) return 1; - skb2 = skb_clone(skb, GFP_ATOMIC); + /* RFC 4884 (partial) support for ICMP extensions */ + if (data_len < 128 || (data_len & 7) || skb->len < data_len) + data_len = 0; + + skb2 = data_len ? skb_copy(skb, GFP_ATOMIC) : skb_clone(skb, GFP_ATOMIC); if (!skb2) return 1; @@ -588,12 +594,26 @@ int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type) skb2->dev = rt->dst.dev; ipv6_addr_set_v4mapped(ip_hdr(skb)->saddr, &temp_saddr); + + if (data_len) { + /* RFC 4884 (partial) support : + * insert 0 padding at the end, before the extensions + */ + __skb_push(skb2, nhs); + skb_reset_network_header(skb2); + memmove(skb2->data, skb2->data + nhs, data_len - nhs); + memset(skb2->data + data_len - nhs, 0, nhs); + /* RFC 4884 4.5 : Length is measured in 64-bit words, + * and stored in reserved[0] + */ + info = (data_len/8) << 24; + } if (type == ICMP_TIME_EXCEEDED) icmp6_send(skb2, ICMPV6_TIME_EXCEED, ICMPV6_EXC_HOPLIMIT, - 0, &temp_saddr); + info, &temp_saddr); else icmp6_send(skb2, ICMPV6_DEST_UNREACH, ICMPV6_ADDR_UNREACH, - 0, &temp_saddr); + info, &temp_saddr); if (rt) ip6_rt_put(rt); diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index d7a36114eb50..cdd714690f95 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -484,6 +484,7 @@ static int ipip6_err(struct sk_buff *skb, u32 info) const struct iphdr *iph = (const struct iphdr *)skb->data; const int type = icmp_hdr(skb)->type; const int code = icmp_hdr(skb)->code; + unsigned int data_len = 0; struct ip_tunnel *t; int err; @@ -508,6 +509,7 @@ static int ipip6_err(struct sk_buff *skb, u32 info) case ICMP_TIME_EXCEEDED: if (code != ICMP_EXC_TTL) return 0; + data_len = icmp_hdr(skb)->un.reserved[1] * 4; /* RFC 4884 4.1 */ break; case ICMP_REDIRECT: break; @@ -536,7 +538,7 @@ static int ipip6_err(struct sk_buff *skb, u32 info) } err = 0; - if (!ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4, type)) + if (!ip6_err_gen_icmpv6_unreach(skb, iph->ihl * 4, type, data_len)) goto out; if (t->parms.iph.daddr == 0) -- cgit v1.2.3 From cc8feb8edd92d854be552fe4f5e0eeabca40b9ee Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 4 Apr 2016 14:00:37 +0100 Subject: rxrpc: Fix exclusive connection handling "Exclusive connections" are meant to be used for a single client call and then scrapped. The idea is to limit the use of the negotiated security context. The current code, however, isn't doing this: it is instead restricting the socket to a single virtual connection and doing all the calls over that. This is changed such that the socket no longer maintains a special virtual connection over which it will do all the calls, but rather gets a new one each time a new exclusive call is made. Further, using a socket option for this is a poor choice. It should be done on sendmsg with a control message marker instead so that calls can be marked exclusive individually. To that end, add RXRPC_EXCLUSIVE_CALL which, if passed to sendmsg() as a control message element, will cause the call to be done on an single-use connection. The socket option (RXRPC_EXCLUSIVE_CONNECTION) still exists and, if set, will override any lack of RXRPC_EXCLUSIVE_CALL being specified so that programs using the setsockopt() will appear to work the same. Signed-off-by: David Howells --- include/linux/rxrpc.h | 3 +- net/rxrpc/af_rxrpc.c | 7 +--- net/rxrpc/ar-internal.h | 7 ++-- net/rxrpc/conn_object.c | 97 ++++++++++++++++++++----------------------------- net/rxrpc/output.c | 19 +++++++--- 5 files changed, 60 insertions(+), 73 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rxrpc.h b/include/linux/rxrpc.h index 1e8f216e2cf1..c68307bc306f 100644 --- a/include/linux/rxrpc.h +++ b/include/linux/rxrpc.h @@ -35,7 +35,7 @@ struct sockaddr_rxrpc { */ #define RXRPC_SECURITY_KEY 1 /* [clnt] set client security key */ #define RXRPC_SECURITY_KEYRING 2 /* [srvr] set ring of server security keys */ -#define RXRPC_EXCLUSIVE_CONNECTION 3 /* [clnt] use exclusive RxRPC connection */ +#define RXRPC_EXCLUSIVE_CONNECTION 3 /* Deprecated; use RXRPC_EXCLUSIVE_CALL instead */ #define RXRPC_MIN_SECURITY_LEVEL 4 /* minimum security level */ /* @@ -52,6 +52,7 @@ struct sockaddr_rxrpc { #define RXRPC_LOCAL_ERROR 7 /* -r: local error generated [terminal] */ #define RXRPC_NEW_CALL 8 /* -r: [Service] new incoming call notification */ #define RXRPC_ACCEPT 9 /* s-: [Service] accept request */ +#define RXRPC_EXCLUSIVE_CALL 10 /* s-: Call should be on exclusive connection */ /* * RxRPC security levels diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c index 48b45a0280c0..73f5c553eef4 100644 --- a/net/rxrpc/af_rxrpc.c +++ b/net/rxrpc/af_rxrpc.c @@ -494,7 +494,7 @@ static int rxrpc_setsockopt(struct socket *sock, int level, int optname, ret = -EISCONN; if (rx->sk.sk_state != RXRPC_UNBOUND) goto error; - set_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags); + rx->exclusive = true; goto success; case RXRPC_SECURITY_KEY: @@ -669,11 +669,6 @@ static int rxrpc_release_sock(struct sock *sk) flush_workqueue(rxrpc_workqueue); rxrpc_purge_queue(&sk->sk_receive_queue); - if (rx->conn) { - rxrpc_put_connection(rx->conn); - rx->conn = NULL; - } - if (rx->local) { rxrpc_put_local(rx->local); rx->local = NULL; diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h index efe6673deb28..4ca99445e0b7 100644 --- a/net/rxrpc/ar-internal.h +++ b/net/rxrpc/ar-internal.h @@ -37,6 +37,8 @@ struct rxrpc_crypt { #define rxrpc_queue_call(CALL) rxrpc_queue_work(&(CALL)->processor) #define rxrpc_queue_conn(CONN) rxrpc_queue_work(&(CONN)->processor) +struct rxrpc_connection; + /* * sk_state for RxRPC sockets */ @@ -57,7 +59,6 @@ struct rxrpc_sock { struct sock sk; rxrpc_interceptor_t interceptor; /* kernel service Rx interceptor function */ struct rxrpc_local *local; /* local endpoint */ - struct rxrpc_connection *conn; /* exclusive virtual connection */ struct list_head listen_link; /* link in the local endpoint's listen list */ struct list_head secureq; /* calls awaiting connection security clearance */ struct list_head acceptq; /* calls awaiting acceptance */ @@ -66,13 +67,13 @@ struct rxrpc_sock { struct rb_root calls; /* outstanding calls on this socket */ unsigned long flags; #define RXRPC_SOCK_CONNECTED 0 /* connect_srx is set */ -#define RXRPC_SOCK_EXCLUSIVE_CONN 1 /* exclusive connection for a client socket */ rwlock_t call_lock; /* lock for calls */ u32 min_sec_level; /* minimum security level */ #define RXRPC_SECURITY_MAX RXRPC_SECURITY_ENCRYPT + bool exclusive; /* Exclusive connection for a client socket */ + sa_family_t family; /* Protocol family created with */ struct sockaddr_rxrpc srx; /* local address */ struct sockaddr_rxrpc connect_srx; /* Default client address from connect() */ - sa_family_t family; /* protocol family created with */ }; #define rxrpc_sk(__sk) container_of((__sk), struct rxrpc_sock, sk) diff --git a/net/rxrpc/conn_object.c b/net/rxrpc/conn_object.c index c6787b6f459f..6164373d6ce3 100644 --- a/net/rxrpc/conn_object.c +++ b/net/rxrpc/conn_object.c @@ -328,71 +328,57 @@ static int rxrpc_connect_exclusive(struct rxrpc_sock *rx, _enter(""); - conn = rx->conn; + conn = rxrpc_alloc_connection(gfp); if (!conn) { - /* not yet present - create a candidate for a new connection - * and then redo the check */ - conn = rxrpc_alloc_connection(gfp); - if (!conn) { - _leave(" = -ENOMEM"); - return -ENOMEM; - } + _leave(" = -ENOMEM"); + return -ENOMEM; + } - conn->trans = trans; - conn->bundle = NULL; - conn->params = *cp; - conn->proto.local = cp->local; - conn->proto.epoch = rxrpc_epoch; - conn->proto.cid = 0; - conn->proto.in_clientflag = 0; - conn->proto.family = cp->peer->srx.transport.family; - conn->out_clientflag = RXRPC_CLIENT_INITIATED; - conn->state = RXRPC_CONN_CLIENT; - conn->avail_calls = RXRPC_MAXCALLS - 1; - - key_get(conn->params.key); - - ret = rxrpc_init_client_conn_security(conn); - if (ret < 0) { - key_put(conn->params.key); - kfree(conn); - _leave(" = %d [key]", ret); - return ret; - } + conn->trans = trans; + conn->bundle = NULL; + conn->params = *cp; + conn->proto.local = cp->local; + conn->proto.epoch = rxrpc_epoch; + conn->proto.cid = 0; + conn->proto.in_clientflag = 0; + conn->proto.family = cp->peer->srx.transport.family; + conn->out_clientflag = RXRPC_CLIENT_INITIATED; + conn->state = RXRPC_CONN_CLIENT; + conn->avail_calls = RXRPC_MAXCALLS - 1; + + key_get(conn->params.key); + + ret = rxrpc_init_client_conn_security(conn); + if (ret < 0) { + key_put(conn->params.key); + kfree(conn); + _leave(" = %d [key]", ret); + return ret; + } - write_lock_bh(&rxrpc_connection_lock); - list_add_tail(&conn->link, &rxrpc_connections); - write_unlock_bh(&rxrpc_connection_lock); + write_lock_bh(&rxrpc_connection_lock); + list_add_tail(&conn->link, &rxrpc_connections); + write_unlock_bh(&rxrpc_connection_lock); - spin_lock(&trans->client_lock); - atomic_inc(&trans->usage); + spin_lock(&trans->client_lock); + atomic_inc(&trans->usage); - _net("CONNECT EXCL new %d on TRANS %d", - conn->debug_id, conn->trans->debug_id); + _net("CONNECT EXCL new %d on TRANS %d", + conn->debug_id, conn->trans->debug_id); - rxrpc_assign_connection_id(conn); - rx->conn = conn; - } else { - spin_lock(&trans->client_lock); - } + rxrpc_assign_connection_id(conn); - /* we've got a connection with a free channel and we can now attach the - * call to it - * - we're holding the transport's client lock - * - we're holding a reference on the connection + /* Since no one else can use the connection, we just use the first + * channel. */ - for (chan = 0; chan < RXRPC_MAXCALLS; chan++) - if (!conn->channels[chan]) - goto found_channel; - goto no_free_channels; - -found_channel: + chan = 0; atomic_inc(&conn->usage); conn->channels[chan] = call; + conn->call_counter = 1; call->conn = conn; call->channel = chan; call->cid = conn->proto.cid | chan; - call->call_id = ++conn->call_counter; + call->call_id = 1; _net("CONNECT client on conn %d chan %d as call %x", conn->debug_id, chan, call->call_id); @@ -402,11 +388,6 @@ found_channel: rxrpc_add_call_ID_to_conn(conn, call); _leave(" = 0"); return 0; - -no_free_channels: - spin_unlock(&trans->client_lock); - _leave(" = -ENOSR"); - return -ENOSR; } /* @@ -427,7 +408,7 @@ int rxrpc_connect_call(struct rxrpc_sock *rx, _enter("%p,%lx,", rx, call->user_call_ID); - if (test_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags)) + if (cp->exclusive) return rxrpc_connect_exclusive(rx, cp, trans, call, gfp); spin_lock(&trans->client_lock); diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c index becbaa7c0a7c..6f8ab0ef839f 100644 --- a/net/rxrpc/output.c +++ b/net/rxrpc/output.c @@ -35,7 +35,8 @@ static int rxrpc_send_data(struct rxrpc_sock *rx, static int rxrpc_sendmsg_cmsg(struct msghdr *msg, unsigned long *user_call_ID, enum rxrpc_command *command, - u32 *abort_code) + u32 *abort_code, + bool *_exclusive) { struct cmsghdr *cmsg; bool got_user_ID = false; @@ -93,6 +94,11 @@ static int rxrpc_sendmsg_cmsg(struct msghdr *msg, return -EINVAL; break; + case RXRPC_EXCLUSIVE_CALL: + *_exclusive = true; + if (len != 0) + return -EINVAL; + break; default: return -EINVAL; } @@ -131,7 +137,7 @@ static void rxrpc_send_abort(struct rxrpc_call *call, u32 abort_code) */ static struct rxrpc_call * rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, - unsigned long user_call_ID) + unsigned long user_call_ID, bool exclusive) { struct rxrpc_conn_parameters cp; struct rxrpc_conn_bundle *bundle; @@ -155,7 +161,7 @@ rxrpc_new_client_call_for_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, cp.local = rx->local; cp.key = rx->key; cp.security_level = rx->min_sec_level; - cp.exclusive = test_bit(RXRPC_SOCK_EXCLUSIVE_CONN, &rx->flags); + cp.exclusive = rx->exclusive | exclusive; cp.service_id = srx->srx_service; trans = rxrpc_name_to_transport(&cp, msg->msg_name, msg->msg_namelen, GFP_KERNEL); @@ -201,12 +207,14 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) enum rxrpc_command cmd; struct rxrpc_call *call; unsigned long user_call_ID = 0; + bool exclusive = false; u32 abort_code = 0; int ret; _enter(""); - ret = rxrpc_sendmsg_cmsg(msg, &user_call_ID, &cmd, &abort_code); + ret = rxrpc_sendmsg_cmsg(msg, &user_call_ID, &cmd, &abort_code, + &exclusive); if (ret < 0) return ret; @@ -224,7 +232,8 @@ int rxrpc_do_sendmsg(struct rxrpc_sock *rx, struct msghdr *msg, size_t len) if (!call) { if (cmd != RXRPC_CMD_SEND_DATA) return -EBADSLT; - call = rxrpc_new_client_call_for_sendmsg(rx, msg, user_call_ID); + call = rxrpc_new_client_call_for_sendmsg(rx, msg, user_call_ID, + exclusive); if (IS_ERR(call)) return PTR_ERR(call); } -- cgit v1.2.3 From d1bd330a229fc8a69f0e7532138dfd42b4542fd4 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Tue, 21 Jun 2016 01:17:17 +0100 Subject: of_mdio: Enable fixed PHY support if driver is a module The fixed_phy driver doesn't have to be built-in, and it's important that of_mdio supports it even if it's a module. Signed-off-by: Ben Hutchings Acked-by: Florian Fainelli Signed-off-by: David S. Miller --- drivers/of/of_mdio.c | 2 +- include/linux/of_mdio.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index e051e1b57609..de68707a99c7 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -361,7 +361,7 @@ struct phy_device *of_phy_attach(struct net_device *dev, } EXPORT_SYMBOL(of_phy_attach); -#if defined(CONFIG_FIXED_PHY) +#if IS_ENABLED(CONFIG_FIXED_PHY) /* * of_phy_is_fixed_link() and of_phy_register_fixed_link() must * support two DT bindings: diff --git a/include/linux/of_mdio.h b/include/linux/of_mdio.h index 8f2237eb3485..6c8cb9aa4c00 100644 --- a/include/linux/of_mdio.h +++ b/include/linux/of_mdio.h @@ -69,7 +69,7 @@ static inline int of_mdio_parse_addr(struct device *dev, } #endif /* CONFIG_OF */ -#if defined(CONFIG_OF) && defined(CONFIG_FIXED_PHY) +#if defined(CONFIG_OF) && IS_ENABLED(CONFIG_FIXED_PHY) extern int of_phy_register_fixed_link(struct device_node *np); extern bool of_phy_is_fixed_link(struct device_node *np); #else -- cgit v1.2.3 From af7d5185263133f859dd4f35d45594deef9db854 Mon Sep 17 00:00:00 2001 From: Rana Shahout Date: Tue, 21 Jun 2016 12:43:59 +0300 Subject: net/mlx4_en: Add DCB PFC support through CEE netlink commands This patch adds support for reading and updating priority flow control (PFC) attributes in the driver via netlink. Signed-off-by: Rana Shahout Signed-off-by: Eran Ben Elisha Signed-off-by: Eugenia Emantayev Signed-off-by: Tariq Toukan Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c | 277 +++++++++++++++++++++++-- drivers/net/ethernet/mellanox/mlx4/en_netdev.c | 25 +++ drivers/net/ethernet/mellanox/mlx4/fw.c | 1 + drivers/net/ethernet/mellanox/mlx4/fw.h | 1 + drivers/net/ethernet/mellanox/mlx4/main.c | 1 + drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 26 +++ drivers/net/ethernet/mellanox/mlx4/port.c | 12 ++ include/linux/mlx4/device.h | 2 + 8 files changed, 331 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c index f01918c63f28..99c6bbdff501 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_dcb_nl.c @@ -37,6 +37,11 @@ #include "mlx4_en.h" #include "fw_qos.h" +enum { + MLX4_CEE_STATE_DOWN = 0, + MLX4_CEE_STATE_UP = 1, +}; + /* Definitions for QCN */ @@ -80,13 +85,202 @@ struct mlx4_congestion_control_mb_prio_802_1_qau_statistics { __be32 reserved3[4]; }; +static u8 mlx4_en_dcbnl_getcap(struct net_device *dev, int capid, u8 *cap) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + + switch (capid) { + case DCB_CAP_ATTR_PFC: + *cap = true; + break; + case DCB_CAP_ATTR_DCBX: + *cap = priv->cee_params.dcbx_cap; + break; + case DCB_CAP_ATTR_PFC_TCS: + *cap = 1 << mlx4_max_tc(priv->mdev->dev); + break; + default: + *cap = false; + break; + } + + return 0; +} + +static u8 mlx4_en_dcbnl_getpfcstate(struct net_device *netdev) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + return priv->cee_params.dcb_cfg.pfc_state; +} + +static void mlx4_en_dcbnl_setpfcstate(struct net_device *netdev, u8 state) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + priv->cee_params.dcb_cfg.pfc_state = state; +} + +static void mlx4_en_dcbnl_get_pfc_cfg(struct net_device *netdev, int priority, + u8 *setting) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + *setting = priv->cee_params.dcb_cfg.tc_config[priority].dcb_pfc; +} + +static void mlx4_en_dcbnl_set_pfc_cfg(struct net_device *netdev, int priority, + u8 setting) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + priv->cee_params.dcb_cfg.tc_config[priority].dcb_pfc = setting; + priv->cee_params.dcb_cfg.pfc_state = true; +} + +static int mlx4_en_dcbnl_getnumtcs(struct net_device *netdev, int tcid, u8 *num) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + + if (!(priv->flags & MLX4_EN_FLAG_DCB_ENABLED)) + return -EINVAL; + + if (tcid == DCB_NUMTCS_ATTR_PFC) + *num = mlx4_max_tc(priv->mdev->dev); + else + *num = 0; + + return 0; +} + +static u8 mlx4_en_dcbnl_set_all(struct net_device *netdev) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + struct mlx4_en_dev *mdev = priv->mdev; + struct mlx4_en_cee_config *dcb_cfg = &priv->cee_params.dcb_cfg; + int err = 0; + + if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + return -EINVAL; + + if (dcb_cfg->pfc_state) { + int tc; + + priv->prof->rx_pause = 0; + priv->prof->tx_pause = 0; + for (tc = 0; tc < CEE_DCBX_MAX_PRIO; tc++) { + u8 tc_mask = 1 << tc; + + switch (dcb_cfg->tc_config[tc].dcb_pfc) { + case pfc_disabled: + priv->prof->tx_ppp &= ~tc_mask; + priv->prof->rx_ppp &= ~tc_mask; + break; + case pfc_enabled_full: + priv->prof->tx_ppp |= tc_mask; + priv->prof->rx_ppp |= tc_mask; + break; + case pfc_enabled_tx: + priv->prof->tx_ppp |= tc_mask; + priv->prof->rx_ppp &= ~tc_mask; + break; + case pfc_enabled_rx: + priv->prof->tx_ppp &= ~tc_mask; + priv->prof->rx_ppp |= tc_mask; + break; + default: + break; + } + } + en_dbg(DRV, priv, "Set pfc on\n"); + } else { + priv->prof->rx_pause = 1; + priv->prof->tx_pause = 1; + en_dbg(DRV, priv, "Set pfc off\n"); + } + + err = mlx4_SET_PORT_general(mdev->dev, priv->port, + priv->rx_skb_size + ETH_FCS_LEN, + priv->prof->tx_pause, + priv->prof->tx_ppp, + priv->prof->rx_pause, + priv->prof->rx_ppp); + if (err) + en_err(priv, "Failed setting pause params\n"); + return err; +} + +static u8 mlx4_en_dcbnl_get_state(struct net_device *dev) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + + if (priv->flags & MLX4_EN_FLAG_DCB_ENABLED) + return MLX4_CEE_STATE_UP; + + return MLX4_CEE_STATE_DOWN; +} + +static u8 mlx4_en_dcbnl_set_state(struct net_device *dev, u8 state) +{ + struct mlx4_en_priv *priv = netdev_priv(dev); + int num_tcs = 0; + + if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + return 1; + + if (!!(state) == !!(priv->flags & MLX4_EN_FLAG_DCB_ENABLED)) + return 0; + + if (state) { + priv->flags |= MLX4_EN_FLAG_DCB_ENABLED; + num_tcs = IEEE_8021QAZ_MAX_TCS; + } else { + priv->flags &= ~MLX4_EN_FLAG_DCB_ENABLED; + } + + return mlx4_en_setup_tc(dev, num_tcs); +} + +/* On success returns a non-zero 802.1p user priority bitmap + * otherwise returns 0 as the invalid user priority bitmap to + * indicate an error. + */ +static int mlx4_en_dcbnl_getapp(struct net_device *netdev, u8 idtype, u16 id) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + struct dcb_app app = { + .selector = idtype, + .protocol = id, + }; + if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + return 0; + + return dcb_getapp(netdev, &app); +} + +static int mlx4_en_dcbnl_setapp(struct net_device *netdev, u8 idtype, + u16 id, u8 up) +{ + struct mlx4_en_priv *priv = netdev_priv(netdev); + struct dcb_app app; + + if (!(priv->cee_params.dcbx_cap & DCB_CAP_DCBX_VER_CEE)) + return -EINVAL; + + memset(&app, 0, sizeof(struct dcb_app)); + app.selector = idtype; + app.protocol = id; + app.priority = up; + + return dcb_setapp(netdev, &app); +} + static int mlx4_en_dcbnl_ieee_getets(struct net_device *dev, struct ieee_ets *ets) { struct mlx4_en_priv *priv = netdev_priv(dev); struct ieee_ets *my_ets = &priv->ets; - /* No IEEE PFC settings available */ if (!my_ets) return -EINVAL; @@ -237,18 +431,51 @@ static int mlx4_en_dcbnl_ieee_setpfc(struct net_device *dev, static u8 mlx4_en_dcbnl_getdcbx(struct net_device *dev) { - return DCB_CAP_DCBX_HOST | DCB_CAP_DCBX_VER_IEEE; + struct mlx4_en_priv *priv = netdev_priv(dev); + + return priv->cee_params.dcbx_cap; } static u8 mlx4_en_dcbnl_setdcbx(struct net_device *dev, u8 mode) { + struct mlx4_en_priv *priv = netdev_priv(dev); + struct ieee_ets ets = {0}; + struct ieee_pfc pfc = {0}; + + if (mode == priv->cee_params.dcbx_cap) + return 0; + if ((mode & DCB_CAP_DCBX_LLD_MANAGED) || - (mode & DCB_CAP_DCBX_VER_CEE) || - !(mode & DCB_CAP_DCBX_VER_IEEE) || + ((mode & DCB_CAP_DCBX_VER_IEEE) && + (mode & DCB_CAP_DCBX_VER_CEE)) || !(mode & DCB_CAP_DCBX_HOST)) - return 1; + goto err; + + priv->cee_params.dcbx_cap = mode; + + ets.ets_cap = IEEE_8021QAZ_MAX_TCS; + pfc.pfc_cap = IEEE_8021QAZ_MAX_TCS; + + if (mode & DCB_CAP_DCBX_VER_IEEE) { + if (mlx4_en_dcbnl_ieee_setets(dev, &ets)) + goto err; + if (mlx4_en_dcbnl_ieee_setpfc(dev, &pfc)) + goto err; + } else if (mode & DCB_CAP_DCBX_VER_CEE) { + if (mlx4_en_dcbnl_set_all(dev)) + goto err; + } else { + if (mlx4_en_dcbnl_ieee_setets(dev, &ets)) + goto err; + if (mlx4_en_dcbnl_ieee_setpfc(dev, &pfc)) + goto err; + if (mlx4_en_setup_tc(dev, 0)) + goto err; + } return 0; +err: + return 1; } #define MLX4_RATELIMIT_UNITS_IN_KB 100000 /* rate-limit HW unit in Kbps */ @@ -463,24 +690,46 @@ static int mlx4_en_dcbnl_ieee_getqcnstats(struct net_device *dev, } const struct dcbnl_rtnl_ops mlx4_en_dcbnl_ops = { - .ieee_getets = mlx4_en_dcbnl_ieee_getets, - .ieee_setets = mlx4_en_dcbnl_ieee_setets, - .ieee_getmaxrate = mlx4_en_dcbnl_ieee_getmaxrate, - .ieee_setmaxrate = mlx4_en_dcbnl_ieee_setmaxrate, - .ieee_getpfc = mlx4_en_dcbnl_ieee_getpfc, - .ieee_setpfc = mlx4_en_dcbnl_ieee_setpfc, + .ieee_getets = mlx4_en_dcbnl_ieee_getets, + .ieee_setets = mlx4_en_dcbnl_ieee_setets, + .ieee_getmaxrate = mlx4_en_dcbnl_ieee_getmaxrate, + .ieee_setmaxrate = mlx4_en_dcbnl_ieee_setmaxrate, + .ieee_getqcn = mlx4_en_dcbnl_ieee_getqcn, + .ieee_setqcn = mlx4_en_dcbnl_ieee_setqcn, + .ieee_getqcnstats = mlx4_en_dcbnl_ieee_getqcnstats, + .ieee_getpfc = mlx4_en_dcbnl_ieee_getpfc, + .ieee_setpfc = mlx4_en_dcbnl_ieee_setpfc, + + .getstate = mlx4_en_dcbnl_get_state, + .setstate = mlx4_en_dcbnl_set_state, + .getpfccfg = mlx4_en_dcbnl_get_pfc_cfg, + .setpfccfg = mlx4_en_dcbnl_set_pfc_cfg, + .setall = mlx4_en_dcbnl_set_all, + .getcap = mlx4_en_dcbnl_getcap, + .getnumtcs = mlx4_en_dcbnl_getnumtcs, + .getpfcstate = mlx4_en_dcbnl_getpfcstate, + .setpfcstate = mlx4_en_dcbnl_setpfcstate, + .getapp = mlx4_en_dcbnl_getapp, + .setapp = mlx4_en_dcbnl_setapp, .getdcbx = mlx4_en_dcbnl_getdcbx, .setdcbx = mlx4_en_dcbnl_setdcbx, - .ieee_getqcn = mlx4_en_dcbnl_ieee_getqcn, - .ieee_setqcn = mlx4_en_dcbnl_ieee_setqcn, - .ieee_getqcnstats = mlx4_en_dcbnl_ieee_getqcnstats, }; const struct dcbnl_rtnl_ops mlx4_en_dcbnl_pfc_ops = { .ieee_getpfc = mlx4_en_dcbnl_ieee_getpfc, .ieee_setpfc = mlx4_en_dcbnl_ieee_setpfc, + .setstate = mlx4_en_dcbnl_set_state, + .getpfccfg = mlx4_en_dcbnl_get_pfc_cfg, + .setpfccfg = mlx4_en_dcbnl_set_pfc_cfg, + .setall = mlx4_en_dcbnl_set_all, + .getnumtcs = mlx4_en_dcbnl_getnumtcs, + .getpfcstate = mlx4_en_dcbnl_getpfcstate, + .setpfcstate = mlx4_en_dcbnl_setpfcstate, + .getapp = mlx4_en_dcbnl_getapp, + .setapp = mlx4_en_dcbnl_setapp, + .getdcbx = mlx4_en_dcbnl_getdcbx, .setdcbx = mlx4_en_dcbnl_setdcbx, }; diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 8e318d21321f..d42083a8a104 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -67,6 +67,17 @@ int mlx4_en_setup_tc(struct net_device *dev, u8 up) offset += priv->num_tx_rings_p_up; } +#ifdef CONFIG_MLX4_EN_DCB + if (!mlx4_is_slave(priv->mdev->dev)) { + if (up) { + priv->flags |= MLX4_EN_FLAG_DCB_ENABLED; + } else { + priv->flags &= ~MLX4_EN_FLAG_DCB_ENABLED; + priv->cee_params.dcb_cfg.pfc_state = false; + } + } +#endif /* CONFIG_MLX4_EN_DCB */ + return 0; } @@ -2815,6 +2826,9 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, struct mlx4_en_priv *priv; int i; int err; +#ifdef CONFIG_MLX4_EN_DCB + struct tc_configuration *tc; +#endif dev = alloc_etherdev_mqs(sizeof(struct mlx4_en_priv), MAX_TX_RINGS, MAX_RX_RINGS); @@ -2881,6 +2895,17 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port, priv->msg_enable = MLX4_EN_MSG_LEVEL; #ifdef CONFIG_MLX4_EN_DCB if (!mlx4_is_slave(priv->mdev->dev)) { + priv->cee_params.dcbx_cap = DCB_CAP_DCBX_VER_CEE | + DCB_CAP_DCBX_HOST | + DCB_CAP_DCBX_VER_IEEE; + priv->flags |= MLX4_EN_DCB_ENABLED; + priv->cee_params.dcb_cfg.pfc_state = false; + + for (i = 0; i < MLX4_EN_NUM_UP; i++) { + tc = &priv->cee_params.dcb_cfg.tc_config[i]; + tc->dcb_pfc = pfc_disabled; + } + if (mdev->dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_ETS_CFG) { dev->dcbnl_ops = &mlx4_en_dcbnl_ops; } else { diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c index e97094598b2d..f4497cf4d06d 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.c +++ b/drivers/net/ethernet/mellanox/mlx4/fw.c @@ -1128,6 +1128,7 @@ int mlx4_QUERY_PORT(struct mlx4_dev *dev, int port, struct mlx4_port_cap *port_c port_cap->max_pkeys = 1 << (field & 0xf); MLX4_GET(field, outbox, QUERY_PORT_MAX_VL_OFFSET); port_cap->max_vl = field & 0xf; + port_cap->max_tc_eth = field >> 4; MLX4_GET(field, outbox, QUERY_PORT_MAX_MACVLAN_OFFSET); port_cap->log_max_macs = field & 0xf; port_cap->log_max_vlans = field >> 4; diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.h b/drivers/net/ethernet/mellanox/mlx4/fw.h index 7ea258af636a..cdbd76f10ced 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.h +++ b/drivers/net/ethernet/mellanox/mlx4/fw.h @@ -53,6 +53,7 @@ struct mlx4_port_cap { int ib_mtu; int max_port_width; int max_vl; + int max_tc_eth; int max_gids; int max_pkeys; u64 def_mac; diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c index 12c77a70abdb..3564aad778a3 100644 --- a/drivers/net/ethernet/mellanox/mlx4/main.c +++ b/drivers/net/ethernet/mellanox/mlx4/main.c @@ -292,6 +292,7 @@ static int _mlx4_dev_port(struct mlx4_dev *dev, int port, dev->caps.pkey_table_len[port] = port_cap->max_pkeys; dev->caps.port_width_cap[port] = port_cap->max_port_width; dev->caps.eth_mtu_cap[port] = port_cap->eth_mtu; + dev->caps.max_tc_eth = port_cap->max_tc_eth; dev->caps.def_mac[port] = port_cap->def_mac; dev->caps.supported_type[port] = port_cap->supported_port_types; dev->caps.suggested_type[port] = port_cap->suggested_type; diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index 6b3b0fefabad..d39bf594abe4 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -448,6 +448,27 @@ struct mlx4_en_frag_info { #define MLX4_EN_TC_ETS 7 +enum dcb_pfc_type { + pfc_disabled = 0, + pfc_enabled_full, + pfc_enabled_tx, + pfc_enabled_rx +}; + +struct tc_configuration { + enum dcb_pfc_type dcb_pfc; +}; + +struct mlx4_en_cee_config { + bool pfc_state; + struct tc_configuration tc_config[MLX4_EN_NUM_UP]; +}; + +struct mlx4_en_cee_params { + u8 dcbx_cap; + struct mlx4_en_cee_config dcb_cfg; +}; + #endif struct ethtool_flow_id { @@ -467,6 +488,9 @@ enum { MLX4_EN_FLAG_RX_FILTER_NEEDED = (1 << 3), MLX4_EN_FLAG_FORCE_PROMISC = (1 << 4), MLX4_EN_FLAG_RX_CSUM_NON_TCP_UDP = (1 << 5), +#ifdef CONFIG_MLX4_EN_DCB + MLX4_EN_FLAG_DCB_ENABLED = (1 << 6), +#endif }; #define PORT_BEACON_MAX_LIMIT (65535) @@ -568,9 +592,11 @@ struct mlx4_en_priv { u32 counter_index; #ifdef CONFIG_MLX4_EN_DCB +#define MLX4_EN_DCB_ENABLED 0x3 struct ieee_ets ets; u16 maxrate[IEEE_8021QAZ_MAX_TCS]; enum dcbnl_cndd_states cndd_state[IEEE_8021QAZ_MAX_TCS]; + struct mlx4_en_cee_params cee_params; #endif #ifdef CONFIG_RFS_ACCEL spinlock_t filters_lock; diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c index 087b23b320cb..3d2095e5c61c 100644 --- a/drivers/net/ethernet/mellanox/mlx4/port.c +++ b/drivers/net/ethernet/mellanox/mlx4/port.c @@ -52,6 +52,7 @@ #define MLX4_FLAG_V_IGNORE_FCS_MASK 0x2 #define MLX4_IGNORE_FCS_MASK 0x1 +#define MLNX4_TX_MAX_NUMBER 8 void mlx4_init_mac_table(struct mlx4_dev *dev, struct mlx4_mac_table *table) { @@ -2015,3 +2016,14 @@ out: return ret; } EXPORT_SYMBOL(mlx4_get_module_info); + +int mlx4_max_tc(struct mlx4_dev *dev) +{ + u8 num_tc = dev->caps.max_tc_eth; + + if (!num_tc) + num_tc = MLNX4_TX_MAX_NUMBER; + + return num_tc; +} +EXPORT_SYMBOL(mlx4_max_tc); diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h index 80dec87a94f8..4dbc1450bbe0 100644 --- a/include/linux/mlx4/device.h +++ b/include/linux/mlx4/device.h @@ -535,6 +535,7 @@ struct mlx4_caps { int max_rq_desc_sz; int max_qp_init_rdma; int max_qp_dest_rdma; + int max_tc_eth; u32 *qp0_qkey; u32 *qp0_proxy; u32 *qp1_proxy; @@ -1494,6 +1495,7 @@ int mlx4_mr_rereg_mem_write(struct mlx4_dev *dev, struct mlx4_mr *mr, int mlx4_get_module_info(struct mlx4_dev *dev, u8 port, u16 offset, u16 size, u8 *data); +int mlx4_max_tc(struct mlx4_dev *dev); /* Returns true if running in low memory profile (kdump kernel) */ static inline bool mlx4_low_memory_profile(void) -- cgit v1.2.3 From 722003ac40c2c397bd5bc2b714125bc82ab27043 Mon Sep 17 00:00:00 2001 From: Sudarsana Reddy Kalluru Date: Tue, 21 Jun 2016 09:36:21 -0400 Subject: qed: Add support for coalescing config read/update. This patch adds support for configuring the device tx/rx coalescing timeout values in the order of micro seconds. It also adds APIs for upper layer drivers for reading/updating the coalescing values. Signed-off-by: Sudarsana Reddy Kalluru Signed-off-by: Yuval Mintz Signed-off-by: David S. Miller --- drivers/net/ethernet/qlogic/qed/qed_dev.c | 104 +++++++++++++++++++++++++ drivers/net/ethernet/qlogic/qed/qed_dev_api.h | 47 +++++++++++ drivers/net/ethernet/qlogic/qed/qed_hw.c | 23 ++++++ drivers/net/ethernet/qlogic/qed/qed_int.c | 75 ++++++++++++++++-- drivers/net/ethernet/qlogic/qed/qed_int.h | 3 + drivers/net/ethernet/qlogic/qed/qed_main.c | 34 ++++++++ drivers/net/ethernet/qlogic/qed/qed_reg_addr.h | 2 + include/linux/qed/qed_if.h | 24 ++++++ 8 files changed, 307 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index e45cff4df280..b26fe267a150 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -2222,6 +2222,110 @@ int qed_fw_rss_eng(struct qed_hwfn *p_hwfn, return 0; } +static int qed_set_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u32 hw_addr, void *p_eth_qzone, + size_t eth_qzone_size, u8 timeset) +{ + struct coalescing_timeset *p_coal_timeset; + + if (p_hwfn->cdev->int_coalescing_mode != QED_COAL_MODE_ENABLE) { + DP_NOTICE(p_hwfn, "Coalescing configuration not enabled\n"); + return -EINVAL; + } + + p_coal_timeset = p_eth_qzone; + memset(p_coal_timeset, 0, eth_qzone_size); + SET_FIELD(p_coal_timeset->value, COALESCING_TIMESET_TIMESET, timeset); + SET_FIELD(p_coal_timeset->value, COALESCING_TIMESET_VALID, 1); + qed_memcpy_to(p_hwfn, p_ptt, hw_addr, p_eth_qzone, eth_qzone_size); + + return 0; +} + +int qed_set_rxq_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u16 coalesce, u8 qid, u16 sb_id) +{ + struct ustorm_eth_queue_zone eth_qzone; + u8 timeset, timer_res; + u16 fw_qid = 0; + u32 address; + int rc; + + /* Coalesce = (timeset << timer-resolution), timeset is 7bit wide */ + if (coalesce <= 0x7F) { + timer_res = 0; + } else if (coalesce <= 0xFF) { + timer_res = 1; + } else if (coalesce <= 0x1FF) { + timer_res = 2; + } else { + DP_ERR(p_hwfn, "Invalid coalesce value - %d\n", coalesce); + return -EINVAL; + } + timeset = (u8)(coalesce >> timer_res); + + rc = qed_fw_l2_queue(p_hwfn, (u16)qid, &fw_qid); + if (rc) + return rc; + + rc = qed_int_set_timer_res(p_hwfn, p_ptt, timer_res, sb_id, false); + if (rc) + goto out; + + address = BAR0_MAP_REG_USDM_RAM + USTORM_ETH_QUEUE_ZONE_OFFSET(fw_qid); + + rc = qed_set_coalesce(p_hwfn, p_ptt, address, ð_qzone, + sizeof(struct ustorm_eth_queue_zone), timeset); + if (rc) + goto out; + + p_hwfn->cdev->rx_coalesce_usecs = coalesce; +out: + return rc; +} + +int qed_set_txq_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u16 coalesce, u8 qid, u16 sb_id) +{ + struct xstorm_eth_queue_zone eth_qzone; + u8 timeset, timer_res; + u16 fw_qid = 0; + u32 address; + int rc; + + /* Coalesce = (timeset << timer-resolution), timeset is 7bit wide */ + if (coalesce <= 0x7F) { + timer_res = 0; + } else if (coalesce <= 0xFF) { + timer_res = 1; + } else if (coalesce <= 0x1FF) { + timer_res = 2; + } else { + DP_ERR(p_hwfn, "Invalid coalesce value - %d\n", coalesce); + return -EINVAL; + } + timeset = (u8)(coalesce >> timer_res); + + rc = qed_fw_l2_queue(p_hwfn, (u16)qid, &fw_qid); + if (rc) + return rc; + + rc = qed_int_set_timer_res(p_hwfn, p_ptt, timer_res, sb_id, true); + if (rc) + goto out; + + address = BAR0_MAP_REG_XSDM_RAM + XSTORM_ETH_QUEUE_ZONE_OFFSET(fw_qid); + + rc = qed_set_coalesce(p_hwfn, p_ptt, address, ð_qzone, + sizeof(struct xstorm_eth_queue_zone), timeset); + if (rc) + goto out; + + p_hwfn->cdev->tx_coalesce_usecs = coalesce; +out: + return rc; +} + /* Calculate final WFQ values for all vports and configure them. * After this configuration each vport will have * approx min rate = min_pf_rate * (vport_wfq / QED_WFQ_UNIT) diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h index f810ce45d463..343bb0344f62 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev_api.h +++ b/drivers/net/ethernet/qlogic/qed/qed_dev_api.h @@ -212,6 +212,20 @@ qed_dmae_host2grc(struct qed_hwfn *p_hwfn, u32 size_in_dwords, u32 flags); + /** + * @brief qed_dmae_grc2host - Read data from dmae data offset + * to source address using the given ptt + * + * @param p_ptt + * @param grc_addr (dmae_data_offset) + * @param dest_addr + * @param size_in_dwords + * @param flags - one of the flags defined above + */ +int qed_dmae_grc2host(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u32 grc_addr, dma_addr_t dest_addr, u32 size_in_dwords, + u32 flags); + /** * @brief qed_dmae_host2host - copy data from to source address * to a destination adress (for SRIOV) using the given ptt @@ -308,4 +322,37 @@ int qed_fw_rss_eng(struct qed_hwfn *p_hwfn, int qed_final_cleanup(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u16 id, bool is_vf); +/** + * @brief qed_set_rxq_coalesce - Configure coalesce parameters for an Rx queue + * The fact that we can configure coalescing to up to 511, but on varying + * accuracy [the bigger the value the less accurate] up to a mistake of 3usec + * for the highest values. + * + * @param p_hwfn + * @param p_ptt + * @param coalesce - Coalesce value in micro seconds. + * @param qid - Queue index. + * @param qid - SB Id + * + * @return int + */ +int qed_set_rxq_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u16 coalesce, u8 qid, u16 sb_id); + +/** + * @brief qed_set_txq_coalesce - Configure coalesce parameters for a Tx queue + * While the API allows setting coalescing per-qid, all tx queues sharing a + * SB should be in same range [i.e., either 0-0x7f, 0x80-0xff or 0x100-0x1ff] + * otherwise configuration would break. + * + * @param p_hwfn + * @param p_ptt + * @param coalesce - Coalesce value in micro seconds. + * @param qid - Queue index. + * @param qid - SB Id + * + * @return int + */ +int qed_set_txq_coalesce(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u16 coalesce, u8 qid, u16 sb_id); #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.c b/drivers/net/ethernet/qlogic/qed/qed_hw.c index 2693c30981eb..e17885321faf 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hw.c +++ b/drivers/net/ethernet/qlogic/qed/qed_hw.c @@ -768,6 +768,29 @@ int qed_dmae_host2grc(struct qed_hwfn *p_hwfn, return rc; } +int +qed_dmae_grc2host(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u32 grc_addr, + dma_addr_t dest_addr, u32 size_in_dwords, u32 flags) +{ + u32 grc_addr_in_dw = grc_addr / sizeof(u32); + struct qed_dmae_params params; + int rc; + + memset(¶ms, 0, sizeof(struct qed_dmae_params)); + params.flags = flags; + + mutex_lock(&p_hwfn->dmae_info.mutex); + + rc = qed_dmae_execute_command(p_hwfn, p_ptt, grc_addr_in_dw, + dest_addr, QED_DMAE_ADDRESS_GRC, + QED_DMAE_ADDRESS_HOST_VIRT, + size_in_dwords, ¶ms); + + mutex_unlock(&p_hwfn->dmae_info.mutex); + + return rc; +} + int qed_dmae_host2host(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c index 09a6ad3d22dd..8fa50fa23c8d 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.c +++ b/drivers/net/ethernet/qlogic/qed/qed_int.c @@ -2418,6 +2418,7 @@ void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn, { struct qed_dev *cdev = p_hwfn->cdev; u32 cau_state; + u8 timer_res; memset(p_sb_entry, 0, sizeof(*p_sb_entry)); @@ -2443,6 +2444,23 @@ void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn, cdev->tx_coalesce_usecs = QED_CAU_DEF_TX_USECS; } + /* Coalesce = (timeset << timer-res), timeset is 7bit wide */ + if (cdev->rx_coalesce_usecs <= 0x7F) + timer_res = 0; + else if (cdev->rx_coalesce_usecs <= 0xFF) + timer_res = 1; + else + timer_res = 2; + SET_FIELD(p_sb_entry->params, CAU_SB_ENTRY_TIMER_RES0, timer_res); + + if (cdev->tx_coalesce_usecs <= 0x7F) + timer_res = 0; + else if (cdev->tx_coalesce_usecs <= 0xFF) + timer_res = 1; + else + timer_res = 2; + SET_FIELD(p_sb_entry->params, CAU_SB_ENTRY_TIMER_RES1, timer_res); + SET_FIELD(p_sb_entry->data, CAU_SB_ENTRY_STATE0, cau_state); SET_FIELD(p_sb_entry->data, CAU_SB_ENTRY_STATE1, cau_state); } @@ -2484,17 +2502,28 @@ void qed_int_cau_conf_sb(struct qed_hwfn *p_hwfn, /* Configure pi coalescing if set */ if (p_hwfn->cdev->int_coalescing_mode == QED_COAL_MODE_ENABLE) { - u8 timeset = p_hwfn->cdev->rx_coalesce_usecs >> - (QED_CAU_DEF_RX_TIMER_RES + 1); + u8 timeset, timer_res; u8 num_tc = 1, i; + /* timeset = (coalesce >> timer-res), timeset is 7bit wide */ + if (p_hwfn->cdev->rx_coalesce_usecs <= 0x7F) + timer_res = 0; + else if (p_hwfn->cdev->rx_coalesce_usecs <= 0xFF) + timer_res = 1; + else + timer_res = 2; + timeset = (u8)(p_hwfn->cdev->rx_coalesce_usecs >> timer_res); qed_int_cau_conf_pi(p_hwfn, p_ptt, igu_sb_id, RX_PI, QED_COAL_RX_STATE_MACHINE, timeset); - timeset = p_hwfn->cdev->tx_coalesce_usecs >> - (QED_CAU_DEF_TX_TIMER_RES + 1); - + if (p_hwfn->cdev->tx_coalesce_usecs <= 0x7F) + timer_res = 0; + else if (p_hwfn->cdev->tx_coalesce_usecs <= 0xFF) + timer_res = 1; + else + timer_res = 2; + timeset = (u8)(p_hwfn->cdev->tx_coalesce_usecs >> timer_res); for (i = 0; i < num_tc; i++) { qed_int_cau_conf_pi(p_hwfn, p_ptt, igu_sb_id, TX_PI(i), @@ -3199,3 +3228,39 @@ void qed_int_disable_post_isr_release(struct qed_dev *cdev) for_each_hwfn(cdev, i) cdev->hwfns[i].b_int_requested = false; } + +int qed_int_set_timer_res(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u8 timer_res, u16 sb_id, bool tx) +{ + struct cau_sb_entry sb_entry; + int rc; + + if (!p_hwfn->hw_init_done) { + DP_ERR(p_hwfn, "hardware not initialized yet\n"); + return -EINVAL; + } + + rc = qed_dmae_grc2host(p_hwfn, p_ptt, CAU_REG_SB_VAR_MEMORY + + sb_id * sizeof(u64), + (u64)(uintptr_t)&sb_entry, 2, 0); + if (rc) { + DP_ERR(p_hwfn, "dmae_grc2host failed %d\n", rc); + return rc; + } + + if (tx) + SET_FIELD(sb_entry.params, CAU_SB_ENTRY_TIMER_RES1, timer_res); + else + SET_FIELD(sb_entry.params, CAU_SB_ENTRY_TIMER_RES0, timer_res); + + rc = qed_dmae_host2grc(p_hwfn, p_ptt, + (u64)(uintptr_t)&sb_entry, + CAU_REG_SB_VAR_MEMORY + + sb_id * sizeof(u64), 2, 0); + if (rc) { + DP_ERR(p_hwfn, "dmae_host2grc failed %d\n", rc); + return rc; + } + + return rc; +} diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.h b/drivers/net/ethernet/qlogic/qed/qed_int.h index 20b468637504..0948be64dc78 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.h +++ b/drivers/net/ethernet/qlogic/qed/qed_int.h @@ -389,6 +389,9 @@ void qed_init_cau_sb_entry(struct qed_hwfn *p_hwfn, u16 vf_number, u8 vf_valid); +int qed_int_set_timer_res(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, + u8 timer_res, u16 sb_id, bool tx); + #define QED_MAPPING_MEMORY_SIZE(dev) (NUM_OF_SBS(dev)) #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index 6c4606b44b7e..e32ee57cdfee 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -1303,6 +1303,38 @@ static int qed_drain(struct qed_dev *cdev) return 0; } +static void qed_get_coalesce(struct qed_dev *cdev, u16 *rx_coal, u16 *tx_coal) +{ + *rx_coal = cdev->rx_coalesce_usecs; + *tx_coal = cdev->tx_coalesce_usecs; +} + +static int qed_set_coalesce(struct qed_dev *cdev, u16 rx_coal, u16 tx_coal, + u8 qid, u16 sb_id) +{ + struct qed_hwfn *hwfn; + struct qed_ptt *ptt; + int hwfn_index; + int status = 0; + + hwfn_index = qid % cdev->num_hwfns; + hwfn = &cdev->hwfns[hwfn_index]; + ptt = qed_ptt_acquire(hwfn); + if (!ptt) + return -EAGAIN; + + status = qed_set_rxq_coalesce(hwfn, ptt, rx_coal, + qid / cdev->num_hwfns, sb_id); + if (status) + goto out; + status = qed_set_txq_coalesce(hwfn, ptt, tx_coal, + qid / cdev->num_hwfns, sb_id); +out: + qed_ptt_release(hwfn, ptt); + + return status; +} + static int qed_set_led(struct qed_dev *cdev, enum qed_led_mode mode) { struct qed_hwfn *hwfn = QED_LEADING_HWFN(cdev); @@ -1349,5 +1381,7 @@ const struct qed_common_ops qed_common_ops_pass = { .update_msglvl = &qed_init_dp, .chain_alloc = &qed_chain_alloc, .chain_free = &qed_chain_free, + .get_coalesce = &qed_get_coalesce, + .set_coalesce = &qed_set_coalesce, .set_led = &qed_set_led, }; diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h index aa08ddbd95df..f6b86ca1ff79 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h +++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h @@ -80,6 +80,8 @@ 0x1f00000UL #define BAR0_MAP_REG_TSDM_RAM \ 0x1c80000UL +#define BAR0_MAP_REG_XSDM_RAM \ + 0x1e00000UL #define NIG_REG_RX_LLH_BRB_GATE_DNTFWD_PERPF \ 0x5011f4UL #define PRS_REG_SEARCH_TCP \ diff --git a/include/linux/qed/qed_if.h b/include/linux/qed/qed_if.h index e1d5122e8a96..b1e3c57c7117 100644 --- a/include/linux/qed/qed_if.h +++ b/include/linux/qed/qed_if.h @@ -488,6 +488,30 @@ struct qed_common_ops { void (*chain_free)(struct qed_dev *cdev, struct qed_chain *p_chain); +/** + * @brief get_coalesce - Get coalesce parameters in usec + * + * @param cdev + * @param rx_coal - Rx coalesce value in usec + * @param tx_coal - Tx coalesce value in usec + * + */ + void (*get_coalesce)(struct qed_dev *cdev, u16 *rx_coal, u16 *tx_coal); + +/** + * @brief set_coalesce - Configure Rx coalesce value in usec + * + * @param cdev + * @param rx_coal - Rx coalesce value in usec + * @param tx_coal - Tx coalesce value in usec + * @param qid - Queue index + * @param sb_id - Status Block Id + * + * @return 0 on success, error otherwise. + */ + int (*set_coalesce)(struct qed_dev *cdev, u16 rx_coal, u16 tx_coal, + u8 qid, u16 sb_id); + /** * @brief set_led - Configure LED mode * -- cgit v1.2.3 From 1466cc5b23d18e7b6b8f1a45443d595393dbcae7 Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Thu, 23 Jun 2016 17:02:37 +0300 Subject: net/mlx5: Rate limit tables support Configuring and managing HW rate limit tables. The HW holds a table of rate limits, each rate is associated with an index in that table. Later a Send Queue uses this index to set the rate limit. Multiple Send Queues can have the same rate limit, which is represented by a single entry in this table. Even though a rate can be shared, each queue is being rate limited independently of others. The SW shadow of this table holds the rate itself, the index in the HW table and the refcount (number of queues) working with this rate. The exported functions are mlx5_rl_add_rate and mlx5_rl_remove_rate. Number of different rates and their values are derived from HW capabilities. Signed-off-by: Yevgeny Petrilin Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/Makefile | 5 +- drivers/net/ethernet/mellanox/mlx5/core/fw.c | 6 + drivers/net/ethernet/mellanox/mlx5/core/main.c | 10 ++ drivers/net/ethernet/mellanox/mlx5/core/rl.c | 209 +++++++++++++++++++++++ include/linux/mlx5/device.h | 4 + include/linux/mlx5/driver.h | 27 +++ 6 files changed, 259 insertions(+), 2 deletions(-) create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/rl.c (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index 9ea7b583096a..0c8a7dcea483 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -1,8 +1,9 @@ obj-$(CONFIG_MLX5_CORE) += mlx5_core.o mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ - health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \ - mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o fs_counters.o + health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o \ + mad.o transobj.o vport.o sriov.o fs_cmd.o fs_core.o \ + fs_counters.o rl.o mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o \ en_main.o en_fs.o en_ethtool.o en_tx.o en_rx.o \ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c index 75c7ae6a5cc4..77fc1aa26114 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c @@ -151,6 +151,12 @@ int mlx5_query_hca_caps(struct mlx5_core_dev *dev) return err; } + if (MLX5_CAP_GEN(dev, qos)) { + err = mlx5_core_get_caps(dev, MLX5_CAP_QOS); + if (err) + return err; + } + return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index a19b59348dd6..08cae3485960 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -1144,6 +1144,13 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) dev_err(&pdev->dev, "Failed to init flow steering\n"); goto err_fs; } + + err = mlx5_init_rl_table(dev); + if (err) { + dev_err(&pdev->dev, "Failed to init rate limiting\n"); + goto err_rl; + } + #ifdef CONFIG_MLX5_CORE_EN err = mlx5_eswitch_init(dev); if (err) { @@ -1183,6 +1190,8 @@ err_sriov: mlx5_eswitch_cleanup(dev->priv.eswitch); #endif err_reg_dev: + mlx5_cleanup_rl_table(dev); +err_rl: mlx5_cleanup_fs(dev); err_fs: mlx5_cleanup_mkey_table(dev); @@ -1253,6 +1262,7 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv) mlx5_eswitch_cleanup(dev->priv.eswitch); #endif + mlx5_cleanup_rl_table(dev); mlx5_cleanup_fs(dev); mlx5_cleanup_mkey_table(dev); mlx5_cleanup_srq_table(dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/rl.c b/drivers/net/ethernet/mellanox/mlx5/core/rl.c new file mode 100644 index 000000000000..c07c28bd3d55 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/rl.c @@ -0,0 +1,209 @@ +/* + * Copyright (c) 2013-2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include +#include +#include +#include "mlx5_core.h" + +/* Finds an entry where we can register the given rate + * If the rate already exists, return the entry where it is registered, + * otherwise return the first available entry. + * If the table is full, return NULL + */ +static struct mlx5_rl_entry *find_rl_entry(struct mlx5_rl_table *table, + u32 rate) +{ + struct mlx5_rl_entry *ret_entry = NULL; + bool empty_found = false; + int i; + + for (i = 0; i < table->max_size; i++) { + if (table->rl_entry[i].rate == rate) + return &table->rl_entry[i]; + if (!empty_found && !table->rl_entry[i].rate) { + empty_found = true; + ret_entry = &table->rl_entry[i]; + } + } + + return ret_entry; +} + +static int mlx5_set_rate_limit_cmd(struct mlx5_core_dev *dev, + u32 rate, u16 index) +{ + u32 in[MLX5_ST_SZ_DW(set_rate_limit_in)]; + u32 out[MLX5_ST_SZ_DW(set_rate_limit_out)]; + + memset(in, 0, sizeof(in)); + memset(out, 0, sizeof(out)); + + MLX5_SET(set_rate_limit_in, in, opcode, + MLX5_CMD_OP_SET_RATE_LIMIT); + MLX5_SET(set_rate_limit_in, in, rate_limit_index, index); + MLX5_SET(set_rate_limit_in, in, rate_limit, rate); + + return mlx5_cmd_exec_check_status(dev, in, sizeof(in), + out, sizeof(out)); +} + +bool mlx5_rl_is_in_range(struct mlx5_core_dev *dev, u32 rate) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + + return (rate <= table->max_rate && rate >= table->min_rate); +} +EXPORT_SYMBOL(mlx5_rl_is_in_range); + +int mlx5_rl_add_rate(struct mlx5_core_dev *dev, u32 rate, u16 *index) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + struct mlx5_rl_entry *entry; + int err = 0; + + mutex_lock(&table->rl_lock); + + if (!rate || !mlx5_rl_is_in_range(dev, rate)) { + mlx5_core_err(dev, "Invalid rate: %u, should be %u to %u\n", + rate, table->min_rate, table->max_rate); + err = -EINVAL; + goto out; + } + + entry = find_rl_entry(table, rate); + if (!entry) { + mlx5_core_err(dev, "Max number of %u rates reached\n", + table->max_size); + err = -ENOSPC; + goto out; + } + if (entry->refcount) { + /* rate already configured */ + entry->refcount++; + } else { + /* new rate limit */ + err = mlx5_set_rate_limit_cmd(dev, rate, entry->index); + if (err) { + mlx5_core_err(dev, "Failed configuring rate: %u (%d)\n", + rate, err); + goto out; + } + entry->rate = rate; + entry->refcount = 1; + } + *index = entry->index; + +out: + mutex_unlock(&table->rl_lock); + return err; +} +EXPORT_SYMBOL(mlx5_rl_add_rate); + +void mlx5_rl_remove_rate(struct mlx5_core_dev *dev, u32 rate) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + struct mlx5_rl_entry *entry = NULL; + + /* 0 is a reserved value for unlimited rate */ + if (rate == 0) + return; + + mutex_lock(&table->rl_lock); + entry = find_rl_entry(table, rate); + if (!entry || !entry->refcount) { + mlx5_core_warn(dev, "Rate %u is not configured\n", rate); + goto out; + } + + entry->refcount--; + if (!entry->refcount) { + /* need to remove rate */ + mlx5_set_rate_limit_cmd(dev, 0, entry->index); + entry->rate = 0; + } + +out: + mutex_unlock(&table->rl_lock); +} +EXPORT_SYMBOL(mlx5_rl_remove_rate); + +int mlx5_init_rl_table(struct mlx5_core_dev *dev) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + int i; + + mutex_init(&table->rl_lock); + if (!MLX5_CAP_GEN(dev, qos) || !MLX5_CAP_QOS(dev, packet_pacing)) { + table->max_size = 0; + return 0; + } + + /* First entry is reserved for unlimited rate */ + table->max_size = MLX5_CAP_QOS(dev, packet_pacing_rate_table_size) - 1; + table->max_rate = MLX5_CAP_QOS(dev, packet_pacing_max_rate); + table->min_rate = MLX5_CAP_QOS(dev, packet_pacing_min_rate); + + table->rl_entry = kcalloc(table->max_size, sizeof(struct mlx5_rl_entry), + GFP_KERNEL); + if (!table->rl_entry) + return -ENOMEM; + + /* The index represents the index in HW rate limit table + * Index 0 is reserved for unlimited rate + */ + for (i = 0; i < table->max_size; i++) + table->rl_entry[i].index = i + 1; + + /* Index 0 is reserved */ + mlx5_core_info(dev, "Rate limit: %u rates are supported, range: %uMbps to %uMbps\n", + table->max_size, + table->min_rate >> 10, + table->max_rate >> 10); + + return 0; +} + +void mlx5_cleanup_rl_table(struct mlx5_core_dev *dev) +{ + struct mlx5_rl_table *table = &dev->priv.rl_table; + int i; + + /* Clear all configured rates */ + for (i = 0; i < table->max_size; i++) + if (table->rl_entry[i].rate) + mlx5_set_rate_limit_cmd(dev, 0, + table->rl_entry[i].index); + + kfree(dev->priv.rl_table.rl_entry); +} diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index 73a48479892d..e0a3ed758287 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -1330,6 +1330,7 @@ enum mlx5_cap_type { MLX5_CAP_ESWITCH, MLX5_CAP_RESERVED, MLX5_CAP_VECTOR_CALC, + MLX5_CAP_QOS, /* NUM OF CAP Types */ MLX5_CAP_NUM }; @@ -1414,6 +1415,9 @@ enum mlx5_cap_type { MLX5_GET(vector_calc_cap, \ mdev->hca_caps_cur[MLX5_CAP_VECTOR_CALC], cap) +#define MLX5_CAP_QOS(mdev, cap)\ + MLX5_GET(qos_cap, mdev->hca_caps_cur[MLX5_CAP_QOS], cap) + enum { MLX5_CMD_STAT_OK = 0x0, MLX5_CMD_STAT_INT_ERR = 0x1, diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 80776d0c52dc..46260fdc5305 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -481,6 +481,21 @@ struct mlx5_fc_stats { struct mlx5_eswitch; +struct mlx5_rl_entry { + u32 rate; + u16 index; + u16 refcount; +}; + +struct mlx5_rl_table { + /* protect rate limit table */ + struct mutex rl_lock; + u16 max_size; + u32 max_rate; + u32 min_rate; + struct mlx5_rl_entry *rl_entry; +}; + struct mlx5_priv { char name[MLX5_MAX_NAME_LEN]; struct mlx5_eq_table eq_table; @@ -544,6 +559,7 @@ struct mlx5_priv { struct mlx5_flow_root_namespace *esw_ingress_root_ns; struct mlx5_fc_stats fc_stats; + struct mlx5_rl_table rl_table; }; enum mlx5_device_state { @@ -861,6 +877,12 @@ int mlx5_query_odp_caps(struct mlx5_core_dev *dev, int mlx5_core_query_ib_ppcnt(struct mlx5_core_dev *dev, u8 port_num, void *out, size_t sz); +int mlx5_init_rl_table(struct mlx5_core_dev *dev); +void mlx5_cleanup_rl_table(struct mlx5_core_dev *dev); +int mlx5_rl_add_rate(struct mlx5_core_dev *dev, u32 rate, u16 *index); +void mlx5_rl_remove_rate(struct mlx5_core_dev *dev, u32 rate); +bool mlx5_rl_is_in_range(struct mlx5_core_dev *dev, u32 rate); + static inline int fw_initializing(struct mlx5_core_dev *dev) { return ioread32be(&dev->iseg->initializing) >> 31; @@ -938,6 +960,11 @@ static inline int mlx5_get_gid_table_len(u16 param) return 8 * (1 << param); } +static inline bool mlx5_rl_is_supported(struct mlx5_core_dev *dev) +{ + return !!(dev->priv.rl_table.max_size); +} + enum { MLX5_TRIGGERED_CMD_COMP = (u64)1 << 32, }; -- cgit v1.2.3 From 667daedaecd15b89d0ded7af49519f28d6ea2cf4 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Thu, 23 Jun 2016 17:02:42 +0300 Subject: net/mlx5e: Toggle link only after modifying port parameters Add a dedicated function to toggle port link. It should be called only after setting a port register. Toggle will set port link to down and bring it back up in case that it's admin status was up. Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c | 9 +-------- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 7 +------ drivers/net/ethernet/mellanox/mlx5/core/port.c | 12 ++++++++++++ include/linux/mlx5/port.h | 1 + 4 files changed, 15 insertions(+), 14 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index b2db180ae2a5..e6883132b555 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -191,7 +191,6 @@ static int mlx5e_dcbnl_ieee_setpfc(struct net_device *dev, { struct mlx5e_priv *priv = netdev_priv(dev); struct mlx5_core_dev *mdev = priv->mdev; - enum mlx5_port_status ps; u8 curr_pfc_en; int ret; @@ -200,14 +199,8 @@ static int mlx5e_dcbnl_ieee_setpfc(struct net_device *dev, if (pfc->pfc_en == curr_pfc_en) return 0; - mlx5_query_port_admin_status(mdev, &ps); - if (ps == MLX5_PORT_UP) - mlx5_set_port_admin_status(mdev, MLX5_PORT_DOWN); - ret = mlx5_set_port_pfc(mdev, pfc->pfc_en, pfc->pfc_en); - - if (ps == MLX5_PORT_UP) - mlx5_set_port_admin_status(mdev, MLX5_PORT_UP); + mlx5_toggle_port_link(mdev); return ret; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index c4be3941957e..d0d3dcfb4de9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -795,7 +795,6 @@ static int mlx5e_set_settings(struct net_device *netdev, u32 link_modes; u32 speed; u32 eth_proto_cap, eth_proto_admin; - enum mlx5_port_status ps; int err; speed = ethtool_cmd_speed(cmd); @@ -829,12 +828,8 @@ static int mlx5e_set_settings(struct net_device *netdev, if (link_modes == eth_proto_admin) goto out; - mlx5_query_port_admin_status(mdev, &ps); - if (ps == MLX5_PORT_UP) - mlx5_set_port_admin_status(mdev, MLX5_PORT_DOWN); mlx5_set_port_proto(mdev, link_modes, MLX5_PTYS_EN); - if (ps == MLX5_PORT_UP) - mlx5_set_port_admin_status(mdev, MLX5_PORT_UP); + mlx5_toggle_port_link(mdev); out: return err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index 3e35611b19c3..1562e7310f5b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -222,6 +222,18 @@ int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, } EXPORT_SYMBOL_GPL(mlx5_set_port_proto); +/* This function should be used after setting a port register only */ +void mlx5_toggle_port_link(struct mlx5_core_dev *dev) +{ + enum mlx5_port_status ps; + + mlx5_query_port_admin_status(dev, &ps); + mlx5_set_port_admin_status(dev, MLX5_PORT_DOWN); + if (ps == MLX5_PORT_UP) + mlx5_set_port_admin_status(dev, MLX5_PORT_UP); +} +EXPORT_SYMBOL_GPL(mlx5_toggle_port_link); + int mlx5_set_port_admin_status(struct mlx5_core_dev *dev, enum mlx5_port_status status) { diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h index 9851862c0ec5..4adfac15f0e9 100644 --- a/include/linux/mlx5/port.h +++ b/include/linux/mlx5/port.h @@ -67,6 +67,7 @@ int mlx5_query_port_proto_oper(struct mlx5_core_dev *dev, u8 local_port); int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, int proto_mask); +void mlx5_toggle_port_link(struct mlx5_core_dev *dev); int mlx5_set_port_admin_status(struct mlx5_core_dev *dev, enum mlx5_port_status status); int mlx5_query_port_admin_status(struct mlx5_core_dev *dev, -- cgit v1.2.3 From 52244d960755936fa9c8ce54d583d0ed46f24fb6 Mon Sep 17 00:00:00 2001 From: Gal Pressman Date: Thu, 23 Jun 2016 17:02:46 +0300 Subject: net/mlx5e: Report correct auto negotiation and allow toggling Previous to this patch auto negotiation was reported off although it was on by default in hardware. This patch reports the correct information to ethtool and allows the user to toggle it on/off. Added another parameter to set port proto function in order to pass the auto negotiation field to the hardware. Signed-off-by: Gal Pressman Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- .../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 42 ++++++++++++++++++---- drivers/net/ethernet/mellanox/mlx5/core/port.c | 36 ++++++++++++++++--- include/linux/mlx5/port.h | 15 ++++++-- 3 files changed, 80 insertions(+), 13 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 4c560e003d56..39a4d961a58e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -702,6 +702,8 @@ static int mlx5e_get_link_ksettings(struct net_device *netdev, u32 eth_proto_admin; u32 eth_proto_lp; u32 eth_proto_oper; + u8 an_disable_admin; + u8 an_status; int err; err = mlx5_query_port_ptys(mdev, out, sizeof(out), MLX5_PTYS_EN, 1); @@ -712,10 +714,12 @@ static int mlx5e_get_link_ksettings(struct net_device *netdev, goto err_query_ptys; } - eth_proto_cap = MLX5_GET(ptys_reg, out, eth_proto_capability); - eth_proto_admin = MLX5_GET(ptys_reg, out, eth_proto_admin); - eth_proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper); - eth_proto_lp = MLX5_GET(ptys_reg, out, eth_proto_lp_advertise); + eth_proto_cap = MLX5_GET(ptys_reg, out, eth_proto_capability); + eth_proto_admin = MLX5_GET(ptys_reg, out, eth_proto_admin); + eth_proto_oper = MLX5_GET(ptys_reg, out, eth_proto_oper); + eth_proto_lp = MLX5_GET(ptys_reg, out, eth_proto_lp_advertise); + an_disable_admin = MLX5_GET(ptys_reg, out, an_disable_admin); + an_status = MLX5_GET(ptys_reg, out, an_status); ethtool_link_ksettings_zero_link_mode(link_ksettings, supported); ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising); @@ -729,6 +733,18 @@ static int mlx5e_get_link_ksettings(struct net_device *netdev, link_ksettings->base.port = get_connector_port(eth_proto_oper); get_lp_advertising(eth_proto_lp, link_ksettings); + if (an_status == MLX5_AN_COMPLETE) + ethtool_link_ksettings_add_link_mode(link_ksettings, + lp_advertising, Autoneg); + + link_ksettings->base.autoneg = an_disable_admin ? AUTONEG_DISABLE : + AUTONEG_ENABLE; + ethtool_link_ksettings_add_link_mode(link_ksettings, supported, + Autoneg); + if (!an_disable_admin) + ethtool_link_ksettings_add_link_mode(link_ksettings, + advertising, Autoneg); + err_query_ptys: return err; } @@ -764,9 +780,14 @@ static int mlx5e_set_link_ksettings(struct net_device *netdev, { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; + u32 eth_proto_cap, eth_proto_admin; + bool an_changes = false; + u8 an_disable_admin; + u8 an_disable_cap; + bool an_disable; u32 link_modes; + u8 an_status; u32 speed; - u32 eth_proto_cap, eth_proto_admin; int err; speed = link_ksettings->base.speed; @@ -797,10 +818,17 @@ static int mlx5e_set_link_ksettings(struct net_device *netdev, goto out; } - if (link_modes == eth_proto_admin) + mlx5_query_port_autoneg(mdev, MLX5_PTYS_EN, &an_status, + &an_disable_cap, &an_disable_admin); + + an_disable = link_ksettings->base.autoneg == AUTONEG_DISABLE; + an_changes = ((!an_disable && an_disable_admin) || + (an_disable && !an_disable_admin)); + + if (!an_changes && link_modes == eth_proto_admin) goto out; - mlx5_set_port_proto(mdev, link_modes, MLX5_PTYS_EN); + mlx5_set_port_ptys(mdev, an_disable, link_modes, MLX5_PTYS_EN); mlx5_toggle_port_link(mdev); out: diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index 1562e7310f5b..752c08127138 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -202,15 +202,24 @@ int mlx5_query_port_proto_oper(struct mlx5_core_dev *dev, } EXPORT_SYMBOL_GPL(mlx5_query_port_proto_oper); -int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, - int proto_mask) +int mlx5_set_port_ptys(struct mlx5_core_dev *dev, bool an_disable, + u32 proto_admin, int proto_mask) { - u32 in[MLX5_ST_SZ_DW(ptys_reg)]; u32 out[MLX5_ST_SZ_DW(ptys_reg)]; + u32 in[MLX5_ST_SZ_DW(ptys_reg)]; + u8 an_disable_admin; + u8 an_disable_cap; + u8 an_status; + + mlx5_query_port_autoneg(dev, proto_mask, &an_status, + &an_disable_cap, &an_disable_admin); + if (!an_disable_cap && an_disable) + return -EPERM; memset(in, 0, sizeof(in)); MLX5_SET(ptys_reg, in, local_port, 1); + MLX5_SET(ptys_reg, in, an_disable_admin, an_disable); MLX5_SET(ptys_reg, in, proto_mask, proto_mask); if (proto_mask == MLX5_PTYS_EN) MLX5_SET(ptys_reg, in, eth_proto_admin, proto_admin); @@ -220,7 +229,7 @@ int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, return mlx5_core_access_reg(dev, in, sizeof(in), out, sizeof(out), MLX5_REG_PTYS, 0, 1); } -EXPORT_SYMBOL_GPL(mlx5_set_port_proto); +EXPORT_SYMBOL_GPL(mlx5_set_port_ptys); /* This function should be used after setting a port register only */ void mlx5_toggle_port_link(struct mlx5_core_dev *dev) @@ -530,6 +539,25 @@ int mlx5_query_port_pfc(struct mlx5_core_dev *dev, u8 *pfc_en_tx, u8 *pfc_en_rx) } EXPORT_SYMBOL_GPL(mlx5_query_port_pfc); +void mlx5_query_port_autoneg(struct mlx5_core_dev *dev, int proto_mask, + u8 *an_status, + u8 *an_disable_cap, u8 *an_disable_admin) +{ + u32 out[MLX5_ST_SZ_DW(ptys_reg)]; + + *an_status = 0; + *an_disable_cap = 0; + *an_disable_admin = 0; + + if (mlx5_query_port_ptys(dev, out, sizeof(out), proto_mask, 1)) + return; + + *an_status = MLX5_GET(ptys_reg, out, an_status); + *an_disable_cap = MLX5_GET(ptys_reg, out, an_disable_cap); + *an_disable_admin = MLX5_GET(ptys_reg, out, an_disable_admin); +} +EXPORT_SYMBOL_GPL(mlx5_query_port_autoneg); + int mlx5_max_tc(struct mlx5_core_dev *mdev) { u8 num_tc = MLX5_CAP_GEN(mdev, max_tc) ? : 8; diff --git a/include/linux/mlx5/port.h b/include/linux/mlx5/port.h index 4adfac15f0e9..e3012cc64b8a 100644 --- a/include/linux/mlx5/port.h +++ b/include/linux/mlx5/port.h @@ -47,6 +47,14 @@ enum mlx5_module_id { MLX5_MODULE_ID_QSFP28 = 0x11, }; +enum mlx5_an_status { + MLX5_AN_UNAVAILABLE = 0, + MLX5_AN_COMPLETE = 1, + MLX5_AN_FAILED = 2, + MLX5_AN_LINK_UP = 3, + MLX5_AN_LINK_DOWN = 4, +}; + #define MLX5_EEPROM_MAX_BYTES 32 #define MLX5_EEPROM_IDENTIFIER_BYTE_MASK 0x000000ff #define MLX5_I2C_ADDR_LOW 0x50 @@ -65,14 +73,17 @@ int mlx5_query_port_link_width_oper(struct mlx5_core_dev *dev, int mlx5_query_port_proto_oper(struct mlx5_core_dev *dev, u8 *proto_oper, int proto_mask, u8 local_port); -int mlx5_set_port_proto(struct mlx5_core_dev *dev, u32 proto_admin, - int proto_mask); +int mlx5_set_port_ptys(struct mlx5_core_dev *dev, bool an_disable, + u32 proto_admin, int proto_mask); void mlx5_toggle_port_link(struct mlx5_core_dev *dev); int mlx5_set_port_admin_status(struct mlx5_core_dev *dev, enum mlx5_port_status status); int mlx5_query_port_admin_status(struct mlx5_core_dev *dev, enum mlx5_port_status *status); int mlx5_set_port_beacon(struct mlx5_core_dev *dev, u16 beacon_duration); +void mlx5_query_port_autoneg(struct mlx5_core_dev *dev, int proto_mask, + u8 *an_status, + u8 *an_disable_cap, u8 *an_disable_admin); int mlx5_set_port_mtu(struct mlx5_core_dev *dev, u16 mtu, u8 port); void mlx5_query_port_max_mtu(struct mlx5_core_dev *dev, u16 *max_mtu, u8 port); -- cgit v1.2.3 From a5e4bd991362223346e1d3561e61d7a25797fe25 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 24 Jun 2016 11:24:08 +0200 Subject: of_mdio: select fixed phy support unconditionally Calling the fixed-phy functions when CONFIG_FIXED_PHY=m as a previous change tried cannot work if the caller is in built-in code: drivers/of/built-in.o: In function `of_phy_register_fixed_link': of_reserved_mem.c:(.text+0x85e0): undefined reference to `fixed_phy_register' Making of_mdio depend on 'FIXED_PHY || !FIXED_PHY' would solve this dependency by enforcing that OF_MDIO itself becomes a loadable module when FIXED_PHY=y, but that creates a different dependency as it breaks any built-in ethernet driver that uses of_mdio. Making FIXED_PHY a bool option also cannot work, since it depends on PHYLIB, which again is tristate. This version now uses 'select FIXED_PHY' to ensure that the fixed-phy portion of of_mdio is not optional. The main downside of this is a small increase in code size for cases that do not need fixed phy support, but it should avoid all of the link-time problems. Signed-off-by: Arnd Bergmann Fixes: d1bd330a229f ("of_mdio: Enable fixed PHY support if driver is a module") Acked-by: Randy Dunlap Signed-off-by: David S. Miller --- drivers/of/Kconfig | 1 + drivers/of/of_mdio.c | 2 -- include/linux/of_mdio.h | 8 ++------ 3 files changed, 3 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/drivers/of/Kconfig b/drivers/of/Kconfig index b3bec3aaa45d..bc07ad30c9bf 100644 --- a/drivers/of/Kconfig +++ b/drivers/of/Kconfig @@ -74,6 +74,7 @@ config OF_NET config OF_MDIO def_tristate PHYLIB depends on PHYLIB + select FIXED_PHY help OpenFirmware MDIO bus (Ethernet PHY) accessors diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index de68707a99c7..e2b50bc12f23 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -361,7 +361,6 @@ struct phy_device *of_phy_attach(struct net_device *dev, } EXPORT_SYMBOL(of_phy_attach); -#if IS_ENABLED(CONFIG_FIXED_PHY) /* * of_phy_is_fixed_link() and of_phy_register_fixed_link() must * support two DT bindings: @@ -451,4 +450,3 @@ int of_phy_register_fixed_link(struct device_node *np) return -ENODEV; } EXPORT_SYMBOL(of_phy_register_fixed_link); -#endif diff --git a/include/linux/of_mdio.h b/include/linux/of_mdio.h index 6c8cb9aa4c00..4b04587d0441 100644 --- a/include/linux/of_mdio.h +++ b/include/linux/of_mdio.h @@ -25,6 +25,8 @@ struct phy_device *of_phy_attach(struct net_device *dev, extern struct mii_bus *of_mdio_find_bus(struct device_node *mdio_np); extern int of_mdio_parse_addr(struct device *dev, const struct device_node *np); +extern int of_phy_register_fixed_link(struct device_node *np); +extern bool of_phy_is_fixed_link(struct device_node *np); #else /* CONFIG_OF */ static inline int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np) @@ -67,12 +69,6 @@ static inline int of_mdio_parse_addr(struct device *dev, { return -ENOSYS; } -#endif /* CONFIG_OF */ - -#if defined(CONFIG_OF) && IS_ENABLED(CONFIG_FIXED_PHY) -extern int of_phy_register_fixed_link(struct device_node *np); -extern bool of_phy_is_fixed_link(struct device_node *np); -#else static inline int of_phy_register_fixed_link(struct device_node *np) { return -ENOSYS; -- cgit v1.2.3 From 02e57b9d7c8ce9e403f15f48fb91dd6549aaf465 Mon Sep 17 00:00:00 2001 From: Giuseppe CAVALLARO Date: Fri, 24 Jun 2016 15:16:26 +0200 Subject: drivers: net: stmmac: add port selection programming In case of SGMII more, for example when a MAC2MAC connection is needed, the port selection bits (inside the MAC configuration registers) have to be programmed according to the link selected. So the patch adds a new DT parameter to pass the port selection and to programmed related PCS and CORE to use it. Signed-off-by: Giuseppe Cavallaro Signed-off-by: David S. Miller --- Documentation/devicetree/bindings/net/stmmac.txt | 3 +++ drivers/net/ethernet/stmicro/stmmac/common.h | 1 + drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c | 15 +++++++++++++++ drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c | 15 +++++++++++++++ drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 3 ++- drivers/net/ethernet/stmicro/stmmac/stmmac_main.c | 15 ++++++++++++++- drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c | 2 ++ include/linux/stmmac.h | 1 + 8 files changed, 53 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/Documentation/devicetree/bindings/net/stmmac.txt b/Documentation/devicetree/bindings/net/stmmac.txt index 95816c5fc589..41b49e6075f5 100644 --- a/Documentation/devicetree/bindings/net/stmmac.txt +++ b/Documentation/devicetree/bindings/net/stmmac.txt @@ -47,6 +47,9 @@ Optional properties: supported by this device instance - snps,perfect-filter-entries: Number of perfect filter entries supported by this device instance +- snps,ps-speed: port selection speed that can be passed to the core when + PCS is supported. For example, this is used in case of SGMII + and MAC2MAC connection. - AXI BUS Mode parameters: below the list of all the parameters to program the AXI register inside the DMA module: - snps,lpi_en: enable Low Power Interface diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 51077a8dc1f5..2533b91f1421 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -531,6 +531,7 @@ struct mac_device_info { unsigned int rx_csum; unsigned int pcs; unsigned int pmt; + unsigned int ps; }; struct mac_device_info *dwmac1000_setup(void __iomem *ioaddr, int mcbins, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c index 0d31f2f80d07..cbefe9e2207c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac1000_core.c @@ -46,6 +46,21 @@ static void dwmac1000_core_init(struct mac_device_info *hw, int mtu) if (mtu > 2000) value |= GMAC_CONTROL_JE; + if (hw->ps) { + value |= GMAC_CONTROL_TE; + + if (hw->ps == SPEED_1000) { + value &= ~GMAC_CONTROL_PS; + } else { + value |= GMAC_CONTROL_PS; + + if (hw->ps == SPEED_10) + value &= ~GMAC_CONTROL_FES; + else + value |= GMAC_CONTROL_FES; + } + } + writel(value, ioaddr + GMAC_CONTROL); /* Mask GMAC interrupts */ diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index 747f3cf1e20b..df5580dcdfed 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -32,6 +32,21 @@ static void dwmac4_core_init(struct mac_device_info *hw, int mtu) if (mtu > 2000) value |= GMAC_CONFIG_JE; + if (hw->ps) { + value |= GMAC_CONFIG_TE; + + if (hw->ps == SPEED_1000) { + value &= ~GMAC_CONFIG_PS; + } else { + value |= GMAC_CONFIG_PS; + + if (hw->ps == SPEED_10) + value &= ~GMAC_CONFIG_FES; + else + value |= GMAC_CONFIG_FES; + } + } + writel(value, ioaddr + GMAC_CONFIG); /* Mask GMAC interrupts */ diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c index da2d9b5b740a..1e06173fc9d7 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c @@ -380,7 +380,8 @@ static int stmmac_ethtool_setsettings(struct net_device *dev, spin_lock(&priv->lock); if (priv->hw->mac->pcs_ctrl_ane) - priv->hw->mac->pcs_ctrl_ane(priv->ioaddr, 1, 0, 0); + priv->hw->mac->pcs_ctrl_ane(priv->ioaddr, 1, + priv->hw->ps, 0); spin_unlock(&priv->lock); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 6142fce9ffac..aab777c1ba33 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -1666,6 +1666,19 @@ static int stmmac_hw_setup(struct net_device *dev, bool init_ptp) if (priv->plat->bus_setup) priv->plat->bus_setup(priv->ioaddr); + /* PS and related bits will be programmed according to the speed */ + if (priv->hw->pcs) { + int speed = priv->plat->mac_port_sel_speed; + + if ((speed == SPEED_10) || (speed == SPEED_100) || + (speed == SPEED_1000)) { + priv->hw->ps = speed; + } else { + dev_warn(priv->device, "invalid port speed\n"); + priv->hw->ps = 0; + } + } + /* Initialize the MAC Core */ priv->hw->mac->core_init(priv->hw, dev->mtu); @@ -1716,7 +1729,7 @@ static int stmmac_hw_setup(struct net_device *dev, bool init_ptp) } if (priv->hw->pcs && priv->hw->mac->pcs_ctrl_ane) - priv->hw->mac->pcs_ctrl_ane(priv->hw, 1, 0, 0); + priv->hw->mac->pcs_ctrl_ane(priv->hw, 1, priv->hw->ps, 0); /* set TX ring length */ if (priv->hw->dma->set_tx_ring_len) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index a96714d34560..f7dfc0ae8e9c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -319,6 +319,8 @@ stmmac_probe_config_dt(struct platform_device *pdev, const char **mac) pr_warn("force_sf_dma_mode is ignored if force_thresh_dma_mode is set."); } + of_property_read_u32(np, "snps,ps-speed", &plat->mac_port_sel_speed); + plat->axi = stmmac_axi_setup(pdev); return plat; diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h index 0507dbfbf63c..705840e0438f 100644 --- a/include/linux/stmmac.h +++ b/include/linux/stmmac.h @@ -141,5 +141,6 @@ struct plat_stmmacenet_data { struct stmmac_axi *axi; int has_gmac4; bool tso_en; + int mac_port_sel_speed; }; #endif -- cgit v1.2.3 From e98e915e11ad1efb11147122bd4932ec6b3425da Mon Sep 17 00:00:00 2001 From: Masashi Honma Date: Wed, 22 Jun 2016 20:23:03 +0900 Subject: wireless: Use macro instead of number Use IEEE80211_MIN_ACTION_SIZE macro for robust management frame check. Signed-off-by: Masashi Honma Signed-off-by: Johannes Berg --- include/linux/ieee80211.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index b118744d3382..1daebb307e6e 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -2464,7 +2464,7 @@ static inline bool _ieee80211_is_robust_mgmt_frame(struct ieee80211_hdr *hdr) */ static inline bool ieee80211_is_robust_mgmt_frame(struct sk_buff *skb) { - if (skb->len < 25) + if (skb->len < IEEE80211_MIN_ACTION_SIZE) return false; return _ieee80211_is_robust_mgmt_frame((void *)skb->data); } -- cgit v1.2.3 From 46f6b06050b736dab4d41494dae27b883cddc365 Mon Sep 17 00:00:00 2001 From: Masashi Honma Date: Wed, 22 Jun 2016 19:55:20 +0900 Subject: mac80211: Encrypt "Group addressed privacy" action frames Previously, the action frames to group address was not encrypted. But [1] "Table 8-38 Category values" indicates "Mesh" and "Multihop" category action frames should be encrypted (Group addressed privacy == yes). And the encyption key should be MGTK ([1] 10.13 Group addressed robust management frame procedures). So this patch modifies the code to make it suitable for spec. [1] IEEE Std 802.11-2012 Signed-off-by: Masashi Honma Signed-off-by: Johannes Berg --- include/linux/ieee80211.h | 30 ++++++++++++++++++++++++++++++ net/mac80211/rx.c | 7 ++++++- net/mac80211/tx.c | 6 +++++- 3 files changed, 41 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ieee80211.h b/include/linux/ieee80211.h index 1daebb307e6e..a80516fd65c8 100644 --- a/include/linux/ieee80211.h +++ b/include/linux/ieee80211.h @@ -19,6 +19,7 @@ #include #include +#include #include #include @@ -2486,6 +2487,35 @@ static inline bool ieee80211_is_public_action(struct ieee80211_hdr *hdr, return mgmt->u.action.category == WLAN_CATEGORY_PUBLIC; } +/** + * _ieee80211_is_group_privacy_action - check if frame is a group addressed + * privacy action frame + * @hdr: the frame + */ +static inline bool _ieee80211_is_group_privacy_action(struct ieee80211_hdr *hdr) +{ + struct ieee80211_mgmt *mgmt = (void *)hdr; + + if (!ieee80211_is_action(hdr->frame_control) || + !is_multicast_ether_addr(hdr->addr1)) + return false; + + return mgmt->u.action.category == WLAN_CATEGORY_MESH_ACTION || + mgmt->u.action.category == WLAN_CATEGORY_MULTIHOP_ACTION; +} + +/** + * ieee80211_is_group_privacy_action - check if frame is a group addressed + * privacy action frame + * @skb: the skb containing the frame, length will be checked + */ +static inline bool ieee80211_is_group_privacy_action(struct sk_buff *skb) +{ + if (skb->len < IEEE80211_MIN_ACTION_SIZE) + return false; + return _ieee80211_is_group_privacy_action((void *)skb->data); +} + /** * ieee80211_tu_to_usec - convert time units (TU) to microseconds * @tu: the TUs diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c index 9a1eb70cb120..2e8a9024625a 100644 --- a/net/mac80211/rx.c +++ b/net/mac80211/rx.c @@ -1624,8 +1624,13 @@ ieee80211_rx_h_decrypt(struct ieee80211_rx_data *rx) if (mmie_keyidx < NUM_DEFAULT_KEYS || mmie_keyidx >= NUM_DEFAULT_KEYS + NUM_DEFAULT_MGMT_KEYS) return RX_DROP_MONITOR; /* unexpected BIP keyidx */ - if (rx->sta) + if (rx->sta) { + if (ieee80211_is_group_privacy_action(skb) && + test_sta_flag(rx->sta, WLAN_STA_MFP)) + return RX_DROP_MONITOR; + rx->key = rcu_dereference(rx->sta->gtk[mmie_keyidx]); + } if (!rx->key) rx->key = rcu_dereference(rx->sdata->keys[mmie_keyidx]); } else if (!ieee80211_has_protected(fc)) { diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index 44ec605a5682..fa8d38eb9236 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -593,6 +593,9 @@ ieee80211_tx_h_select_key(struct ieee80211_tx_data *tx) else if (tx->sta && (key = rcu_dereference(tx->sta->ptk[tx->sta->ptk_idx]))) tx->key = key; + else if (ieee80211_is_group_privacy_action(tx->skb) && + (key = rcu_dereference(tx->sdata->default_multicast_key))) + tx->key = key; else if (ieee80211_is_mgmt(hdr->frame_control) && is_multicast_ether_addr(hdr->addr1) && ieee80211_is_robust_mgmt_frame(tx->skb) && @@ -625,7 +628,8 @@ ieee80211_tx_h_select_key(struct ieee80211_tx_data *tx) case WLAN_CIPHER_SUITE_GCMP_256: if (!ieee80211_is_data_present(hdr->frame_control) && !ieee80211_use_mfp(hdr->frame_control, tx->sta, - tx->skb)) + tx->skb) && + !ieee80211_is_group_privacy_action(tx->skb)) tx->key = NULL; else skip_hw = (tx->key->conf.flags & -- cgit v1.2.3 From 982fb490c298896d15e9323a882f34a57c11ff56 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Thu, 30 Jun 2016 14:45:31 +0800 Subject: ptr_ring: support zero length ring Sometimes, we need zero length ring. But current code will crash since we don't do any check before accessing the ring. This patch fixes this. Signed-off-by: Jason Wang Signed-off-by: David S. Miller --- include/linux/ptr_ring.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h index 562a65e8bcc0..d78b8b89c707 100644 --- a/include/linux/ptr_ring.h +++ b/include/linux/ptr_ring.h @@ -102,7 +102,7 @@ static inline bool ptr_ring_full_bh(struct ptr_ring *r) */ static inline int __ptr_ring_produce(struct ptr_ring *r, void *ptr) { - if (r->queue[r->producer]) + if (unlikely(!r->size) || r->queue[r->producer]) return -ENOSPC; r->queue[r->producer++] = ptr; @@ -164,7 +164,9 @@ static inline int ptr_ring_produce_bh(struct ptr_ring *r, void *ptr) */ static inline void *__ptr_ring_peek(struct ptr_ring *r) { - return r->queue[r->consumer]; + if (likely(r->size)) + return r->queue[r->consumer]; + return NULL; } /* Note: callers invoking this in a loop must use a compiler barrier, -- cgit v1.2.3 From fd68adec9de3104c236ffbcb3bd829d3e635a444 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Thu, 30 Jun 2016 14:45:32 +0800 Subject: skb_array: minor tweak Signed-off-by: Michael S. Tsirkin Signed-off-by: Jason Wang Signed-off-by: David S. Miller --- include/linux/skb_array.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h index 678bfbf78ac4..2dd0d1e4ee7e 100644 --- a/include/linux/skb_array.h +++ b/include/linux/skb_array.h @@ -151,12 +151,12 @@ static inline int skb_array_init(struct skb_array *a, int size, gfp_t gfp) return ptr_ring_init(&a->ring, size, gfp); } -void __skb_array_destroy_skb(void *ptr) +static void __skb_array_destroy_skb(void *ptr) { kfree_skb(ptr); } -int skb_array_resize(struct skb_array *a, int size, gfp_t gfp) +static inline int skb_array_resize(struct skb_array *a, int size, gfp_t gfp) { return ptr_ring_resize(&a->ring, size, gfp, __skb_array_destroy_skb); } -- cgit v1.2.3 From 59e6ae53248a72d83cec77dd704b6990b2394479 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Thu, 30 Jun 2016 14:45:33 +0800 Subject: ptr_ring: support resizing multiple queues Sometimes, we need support resizing multiple queues at once. This is because it was not easy to recover to recover from a partial failure of multiple queues resizing. Signed-off-by: Michael S. Tsirkin Signed-off-by: Jason Wang Signed-off-by: David S. Miller --- include/linux/ptr_ring.h | 71 +++++++++++++++++++++++++++++++++++----- tools/virtio/ringtest/ptr_ring.c | 5 +++ 2 files changed, 67 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/ptr_ring.h b/include/linux/ptr_ring.h index d78b8b89c707..2052011bf9fb 100644 --- a/include/linux/ptr_ring.h +++ b/include/linux/ptr_ring.h @@ -349,20 +349,14 @@ static inline int ptr_ring_init(struct ptr_ring *r, int size, gfp_t gfp) return 0; } -static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, - void (*destroy)(void *)) +static inline void **__ptr_ring_swap_queue(struct ptr_ring *r, void **queue, + int size, gfp_t gfp, + void (*destroy)(void *)) { - unsigned long flags; int producer = 0; - void **queue = __ptr_ring_init_queue_alloc(size, gfp); void **old; void *ptr; - if (!queue) - return -ENOMEM; - - spin_lock_irqsave(&(r)->producer_lock, flags); - while ((ptr = ptr_ring_consume(r))) if (producer < size) queue[producer++] = ptr; @@ -375,6 +369,23 @@ static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, old = r->queue; r->queue = queue; + return old; +} + +static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, + void (*destroy)(void *)) +{ + unsigned long flags; + void **queue = __ptr_ring_init_queue_alloc(size, gfp); + void **old; + + if (!queue) + return -ENOMEM; + + spin_lock_irqsave(&(r)->producer_lock, flags); + + old = __ptr_ring_swap_queue(r, queue, size, gfp, destroy); + spin_unlock_irqrestore(&(r)->producer_lock, flags); kfree(old); @@ -382,6 +393,48 @@ static inline int ptr_ring_resize(struct ptr_ring *r, int size, gfp_t gfp, return 0; } +static inline int ptr_ring_resize_multiple(struct ptr_ring **rings, int nrings, + int size, + gfp_t gfp, void (*destroy)(void *)) +{ + unsigned long flags; + void ***queues; + int i; + + queues = kmalloc(nrings * sizeof *queues, gfp); + if (!queues) + goto noqueues; + + for (i = 0; i < nrings; ++i) { + queues[i] = __ptr_ring_init_queue_alloc(size, gfp); + if (!queues[i]) + goto nomem; + } + + for (i = 0; i < nrings; ++i) { + spin_lock_irqsave(&(rings[i])->producer_lock, flags); + queues[i] = __ptr_ring_swap_queue(rings[i], queues[i], + size, gfp, destroy); + spin_unlock_irqrestore(&(rings[i])->producer_lock, flags); + } + + for (i = 0; i < nrings; ++i) + kfree(queues[i]); + + kfree(queues); + + return 0; + +nomem: + while (--i >= 0) + kfree(queues[i]); + + kfree(queues); + +noqueues: + return -ENOMEM; +} + static inline void ptr_ring_cleanup(struct ptr_ring *r, void (*destroy)(void *)) { void *ptr; diff --git a/tools/virtio/ringtest/ptr_ring.c b/tools/virtio/ringtest/ptr_ring.c index 74abd746ae91..68e4f9f0da3a 100644 --- a/tools/virtio/ringtest/ptr_ring.c +++ b/tools/virtio/ringtest/ptr_ring.c @@ -17,6 +17,11 @@ typedef pthread_spinlock_t spinlock_t; typedef int gfp_t; +static void *kmalloc(unsigned size, gfp_t gfp) +{ + return memalign(64, size); +} + static void *kzalloc(unsigned size, gfp_t gfp) { void *p = memalign(64, size); -- cgit v1.2.3 From bf900b3dbefee49855c17aa09fb4245346a78fb3 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Thu, 30 Jun 2016 14:45:34 +0800 Subject: skb_array: add wrappers for resizing Signed-off-by: Michael S. Tsirkin Signed-off-by: Jason Wang Signed-off-by: David S. Miller --- include/linux/skb_array.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux') diff --git a/include/linux/skb_array.h b/include/linux/skb_array.h index 2dd0d1e4ee7e..f4dfade428f0 100644 --- a/include/linux/skb_array.h +++ b/include/linux/skb_array.h @@ -161,6 +161,15 @@ static inline int skb_array_resize(struct skb_array *a, int size, gfp_t gfp) return ptr_ring_resize(&a->ring, size, gfp, __skb_array_destroy_skb); } +static inline int skb_array_resize_multiple(struct skb_array **rings, + int nrings, int size, gfp_t gfp) +{ + BUILD_BUG_ON(offsetof(struct skb_array, ring)); + return ptr_ring_resize_multiple((struct ptr_ring **)rings, + nrings, size, gfp, + __skb_array_destroy_skb); +} + static inline void skb_array_cleanup(struct skb_array *a) { ptr_ring_cleanup(&a->ring, __skb_array_destroy_skb); -- cgit v1.2.3 From 08294a26e15d7baf1e14ee569e9f2bc82a7ae768 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Thu, 30 Jun 2016 14:45:35 +0800 Subject: net: introduce NETDEV_CHANGE_TX_QUEUE_LEN This patch introduces a new event - NETDEV_CHANGE_TX_QUEUE_LEN, this will be triggered when tx_queue_len. It could be used by net device who want to do some processing at that time. An example is tun who may want to resize tx array when tx_queue_len is changed. Cc: John Fastabend Signed-off-by: Jason Wang Acked-by: John Fastabend Signed-off-by: David S. Miller --- include/linux/netdevice.h | 1 + net/core/net-sysfs.c | 15 ++++++++++++++- net/core/rtnetlink.c | 16 ++++++++++++---- 3 files changed, 27 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index e84d9d23c2d5..7dc2ec74122a 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -2237,6 +2237,7 @@ struct netdev_lag_lower_state_info { #define NETDEV_PRECHANGEUPPER 0x001A #define NETDEV_CHANGELOWERSTATE 0x001B #define NETDEV_UDP_TUNNEL_PUSH_INFO 0x001C +#define NETDEV_CHANGE_TX_QUEUE_LEN 0x001E int register_netdevice_notifier(struct notifier_block *nb); int unregister_netdevice_notifier(struct notifier_block *nb); diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 7a0b616557ab..6e4f34721080 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -322,7 +322,20 @@ NETDEVICE_SHOW_RW(flags, fmt_hex); static int change_tx_queue_len(struct net_device *dev, unsigned long new_len) { - dev->tx_queue_len = new_len; + int res, orig_len = dev->tx_queue_len; + + if (new_len != orig_len) { + dev->tx_queue_len = new_len; + res = call_netdevice_notifiers(NETDEV_CHANGE_TX_QUEUE_LEN, dev); + res = notifier_to_errno(res); + if (res) { + netdev_err(dev, + "refused to change device tx_queue_len\n"); + dev->tx_queue_len = orig_len; + return -EFAULT; + } + } + return 0; } diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index cfed7bc14ee6..a9e3805af739 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1927,11 +1927,19 @@ static int do_setlink(const struct sk_buff *skb, if (tb[IFLA_TXQLEN]) { unsigned long value = nla_get_u32(tb[IFLA_TXQLEN]); - - if (dev->tx_queue_len ^ value) + unsigned long orig_len = dev->tx_queue_len; + + if (dev->tx_queue_len ^ value) { + dev->tx_queue_len = value; + err = call_netdevice_notifiers( + NETDEV_CHANGE_TX_QUEUE_LEN, dev); + err = notifier_to_errno(err); + if (err) { + dev->tx_queue_len = orig_len; + goto errout; + } status |= DO_SETLINK_NOTIFY; - - dev->tx_queue_len = value; + } } if (tb[IFLA_OPERSTATE]) -- cgit v1.2.3 From 1576d98605998fb59d121a39581129e134217182 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Thu, 30 Jun 2016 14:45:36 +0800 Subject: tun: switch to use skb array for tx We used to queue tx packets in sk_receive_queue, this is less efficient since it requires spinlocks to synchronize between producer and consumer. This patch tries to address this by: - switch from sk_receive_queue to a skb_array, and resize it when tx_queue_len was changed. - introduce a new proto_ops peek_len which was used for peeking the skb length. - implement a tun version of peek_len for vhost_net to use and convert vhost_net to use peek_len if possible. Pktgen test shows about 15.3% improvement on guest receiving pps for small buffers: Before: ~1300000pps After : ~1500000pps Signed-off-by: Jason Wang Signed-off-by: David S. Miller --- drivers/net/tun.c | 138 +++++++++++++++++++++++++++++++++++++++++++++++++--- drivers/vhost/net.c | 16 +++++- include/linux/net.h | 1 + 3 files changed, 146 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/tun.c b/drivers/net/tun.c index 4884802e0af1..74752159ec34 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -71,6 +71,7 @@ #include #include #include +#include #include @@ -167,6 +168,7 @@ struct tun_file { }; struct list_head next; struct tun_struct *detached; + struct skb_array tx_array; }; struct tun_flow_entry { @@ -515,7 +517,11 @@ static struct tun_struct *tun_enable_queue(struct tun_file *tfile) static void tun_queue_purge(struct tun_file *tfile) { - skb_queue_purge(&tfile->sk.sk_receive_queue); + struct sk_buff *skb; + + while ((skb = skb_array_consume(&tfile->tx_array)) != NULL) + kfree_skb(skb); + skb_queue_purge(&tfile->sk.sk_error_queue); } @@ -560,6 +566,8 @@ static void __tun_detach(struct tun_file *tfile, bool clean) tun->dev->reg_state == NETREG_REGISTERED) unregister_netdevice(tun->dev); } + if (tun) + skb_array_cleanup(&tfile->tx_array); sock_put(&tfile->sk); } } @@ -613,6 +621,7 @@ static void tun_detach_all(struct net_device *dev) static int tun_attach(struct tun_struct *tun, struct file *file, bool skip_filter) { struct tun_file *tfile = file->private_data; + struct net_device *dev = tun->dev; int err; err = security_tun_dev_attach(tfile->socket.sk, tun->security); @@ -642,6 +651,13 @@ static int tun_attach(struct tun_struct *tun, struct file *file, bool skip_filte if (!err) goto out; } + + if (!tfile->detached && + skb_array_init(&tfile->tx_array, dev->tx_queue_len, GFP_KERNEL)) { + err = -ENOMEM; + goto out; + } + tfile->queue_index = tun->numqueues; tfile->socket.sk->sk_shutdown &= ~RCV_SHUTDOWN; rcu_assign_pointer(tfile->tun, tun); @@ -891,8 +907,8 @@ static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev) nf_reset(skb); - /* Enqueue packet */ - skb_queue_tail(&tfile->socket.sk->sk_receive_queue, skb); + if (skb_array_produce(&tfile->tx_array, skb)) + goto drop; /* Notify and wake up reader process */ if (tfile->flags & TUN_FASYNC) @@ -1107,7 +1123,7 @@ static unsigned int tun_chr_poll(struct file *file, poll_table *wait) poll_wait(file, sk_sleep(sk), wait); - if (!skb_queue_empty(&sk->sk_receive_queue)) + if (!skb_array_empty(&tfile->tx_array)) mask |= POLLIN | POLLRDNORM; if (sock_writeable(sk) || @@ -1426,22 +1442,61 @@ done: return total; } +static struct sk_buff *tun_ring_recv(struct tun_file *tfile, int noblock, + int *err) +{ + DECLARE_WAITQUEUE(wait, current); + struct sk_buff *skb = NULL; + + skb = skb_array_consume(&tfile->tx_array); + if (skb) + goto out; + if (noblock) { + *err = -EAGAIN; + goto out; + } + + add_wait_queue(&tfile->wq.wait, &wait); + current->state = TASK_INTERRUPTIBLE; + + while (1) { + skb = skb_array_consume(&tfile->tx_array); + if (skb) + break; + if (signal_pending(current)) { + *err = -ERESTARTSYS; + break; + } + if (tfile->socket.sk->sk_shutdown & RCV_SHUTDOWN) { + *err = -EFAULT; + break; + } + + schedule(); + } + + current->state = TASK_RUNNING; + remove_wait_queue(&tfile->wq.wait, &wait); + +out: + return skb; +} + static ssize_t tun_do_read(struct tun_struct *tun, struct tun_file *tfile, struct iov_iter *to, int noblock) { struct sk_buff *skb; ssize_t ret; - int peeked, err, off = 0; + int err; tun_debug(KERN_INFO, tun, "tun_do_read\n"); if (!iov_iter_count(to)) return 0; - /* Read frames from queue */ - skb = __skb_recv_datagram(tfile->socket.sk, noblock ? MSG_DONTWAIT : 0, - &peeked, &off, &err); + /* Read frames from ring */ + skb = tun_ring_recv(tfile, noblock, &err); if (!skb) return err; @@ -1574,8 +1629,25 @@ out: return ret; } +static int tun_peek_len(struct socket *sock) +{ + struct tun_file *tfile = container_of(sock, struct tun_file, socket); + struct tun_struct *tun; + int ret = 0; + + tun = __tun_get(tfile); + if (!tun) + return 0; + + ret = skb_array_peek_len(&tfile->tx_array); + tun_put(tun); + + return ret; +} + /* Ops structure to mimic raw sockets with tun */ static const struct proto_ops tun_socket_ops = { + .peek_len = tun_peek_len, .sendmsg = tun_sendmsg, .recvmsg = tun_recvmsg, }; @@ -2397,6 +2469,53 @@ static const struct ethtool_ops tun_ethtool_ops = { .get_ts_info = ethtool_op_get_ts_info, }; +static int tun_queue_resize(struct tun_struct *tun) +{ + struct net_device *dev = tun->dev; + struct tun_file *tfile; + struct skb_array **arrays; + int n = tun->numqueues + tun->numdisabled; + int ret, i; + + arrays = kmalloc(sizeof *arrays * n, GFP_KERNEL); + if (!arrays) + return -ENOMEM; + + for (i = 0; i < tun->numqueues; i++) { + tfile = rtnl_dereference(tun->tfiles[i]); + arrays[i] = &tfile->tx_array; + } + list_for_each_entry(tfile, &tun->disabled, next) + arrays[i++] = &tfile->tx_array; + + ret = skb_array_resize_multiple(arrays, n, + dev->tx_queue_len, GFP_KERNEL); + + kfree(arrays); + return ret; +} + +static int tun_device_event(struct notifier_block *unused, + unsigned long event, void *ptr) +{ + struct net_device *dev = netdev_notifier_info_to_dev(ptr); + struct tun_struct *tun = netdev_priv(dev); + + switch (event) { + case NETDEV_CHANGE_TX_QUEUE_LEN: + if (tun_queue_resize(tun)) + return NOTIFY_BAD; + break; + default: + break; + } + + return NOTIFY_DONE; +} + +static struct notifier_block tun_notifier_block __read_mostly = { + .notifier_call = tun_device_event, +}; static int __init tun_init(void) { @@ -2416,6 +2535,8 @@ static int __init tun_init(void) pr_err("Can't register misc device %d\n", TUN_MINOR); goto err_misc; } + + register_netdevice_notifier(&tun_notifier_block); return 0; err_misc: rtnl_link_unregister(&tun_link_ops); @@ -2427,6 +2548,7 @@ static void tun_cleanup(void) { misc_deregister(&tun_miscdev); rtnl_link_unregister(&tun_link_ops); + unregister_netdevice_notifier(&tun_notifier_block); } /* Get an underlying socket object from tun file. Returns error unless file is diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c index 1d3e45f84549..e032ca397371 100644 --- a/drivers/vhost/net.c +++ b/drivers/vhost/net.c @@ -481,10 +481,14 @@ out: static int peek_head_len(struct sock *sk) { + struct socket *sock = sk->sk_socket; struct sk_buff *head; int len = 0; unsigned long flags; + if (sock->ops->peek_len) + return sock->ops->peek_len(sock); + spin_lock_irqsave(&sk->sk_receive_queue.lock, flags); head = skb_peek(&sk->sk_receive_queue); if (likely(head)) { @@ -497,6 +501,16 @@ static int peek_head_len(struct sock *sk) return len; } +static int sk_has_rx_data(struct sock *sk) +{ + struct socket *sock = sk->sk_socket; + + if (sock->ops->peek_len) + return sock->ops->peek_len(sock); + + return skb_queue_empty(&sk->sk_receive_queue); +} + static int vhost_net_rx_peek_head_len(struct vhost_net *net, struct sock *sk) { struct vhost_net_virtqueue *nvq = &net->vqs[VHOST_NET_VQ_TX]; @@ -513,7 +527,7 @@ static int vhost_net_rx_peek_head_len(struct vhost_net *net, struct sock *sk) endtime = busy_clock() + vq->busyloop_timeout; while (vhost_can_busy_poll(&net->dev, endtime) && - skb_queue_empty(&sk->sk_receive_queue) && + !sk_has_rx_data(sk) && vhost_vq_avail_empty(&net->dev, vq)) cpu_relax_lowlatency(); diff --git a/include/linux/net.h b/include/linux/net.h index 25aa03b51c4e..b9f0ff4d489c 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -185,6 +185,7 @@ struct proto_ops { ssize_t (*splice_read)(struct socket *sock, loff_t *ppos, struct pipe_inode_info *pipe, size_t len, unsigned int flags); int (*set_peek_off)(struct sock *sk, int val); + int (*peek_len)(struct socket *sock); }; #define DECLARE_SOCKADDR(type, dst, src) \ -- cgit v1.2.3 From 4ae89ad92477219b504a49966ee010fe8dcb85af Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Fri, 24 Jun 2016 11:32:26 -0700 Subject: etherdevice.h & bridge: netfilter: Add and use ether_addr_equal_masked There are code duplications of a masked ethernet address comparison here so make it a separate function instead. Miscellanea: o Neaten alignment of FWINV macro uses to make it clearer for the reader Signed-off-by: Joe Perches Acked-by: David S. Miller Signed-off-by: Pablo Neira Ayuso --- include/linux/etherdevice.h | 23 +++++++++++++++++++ net/bridge/netfilter/ebt_arp.c | 17 +++++--------- net/bridge/netfilter/ebt_stp.c | 49 ++++++++++++++++++----------------------- net/bridge/netfilter/ebtables.c | 17 +++++--------- 4 files changed, 57 insertions(+), 49 deletions(-) (limited to 'include/linux') diff --git a/include/linux/etherdevice.h b/include/linux/etherdevice.h index 37ff4a6faa9a..6fec9e81bd70 100644 --- a/include/linux/etherdevice.h +++ b/include/linux/etherdevice.h @@ -373,6 +373,29 @@ static inline bool ether_addr_equal_unaligned(const u8 *addr1, const u8 *addr2) #endif } +/** + * ether_addr_equal_masked - Compare two Ethernet addresses with a mask + * @addr1: Pointer to a six-byte array containing the 1st Ethernet address + * @addr2: Pointer to a six-byte array containing the 2nd Ethernet address + * @mask: Pointer to a six-byte array containing the Ethernet address bitmask + * + * Compare two Ethernet addresses with a mask, returns true if for every bit + * set in the bitmask the equivalent bits in the ethernet addresses are equal. + * Using a mask with all bits set is a slower ether_addr_equal. + */ +static inline bool ether_addr_equal_masked(const u8 *addr1, const u8 *addr2, + const u8 *mask) +{ + int i; + + for (i = 0; i < ETH_ALEN; i++) { + if ((addr1[i] ^ addr2[i]) & mask[i]) + return false; + } + + return true; +} + /** * is_etherdev_addr - Tell if given Ethernet address belongs to the device. * @dev: Pointer to a device structure diff --git a/net/bridge/netfilter/ebt_arp.c b/net/bridge/netfilter/ebt_arp.c index cd457b891b27..cca0a899ee15 100644 --- a/net/bridge/netfilter/ebt_arp.c +++ b/net/bridge/netfilter/ebt_arp.c @@ -65,7 +65,6 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) if (info->bitmask & (EBT_ARP_SRC_MAC | EBT_ARP_DST_MAC)) { const unsigned char *mp; unsigned char _mac[ETH_ALEN]; - uint8_t verdict, i; if (ah->ar_hln != ETH_ALEN || ah->ar_hrd != htons(ARPHRD_ETHER)) return false; @@ -74,11 +73,9 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) sizeof(_mac), &_mac); if (mp == NULL) return false; - verdict = 0; - for (i = 0; i < 6; i++) - verdict |= (mp[i] ^ info->smaddr[i]) & - info->smmsk[i]; - if (FWINV(verdict != 0, EBT_ARP_SRC_MAC)) + if (FWINV(!ether_addr_equal_masked(mp, info->smaddr, + info->smmsk), + EBT_ARP_SRC_MAC)) return false; } @@ -88,11 +85,9 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) sizeof(_mac), &_mac); if (mp == NULL) return false; - verdict = 0; - for (i = 0; i < 6; i++) - verdict |= (mp[i] ^ info->dmaddr[i]) & - info->dmmsk[i]; - if (FWINV(verdict != 0, EBT_ARP_DST_MAC)) + if (FWINV(!ether_addr_equal_masked(mp, info->dmaddr, + info->dmmsk), + EBT_ARP_DST_MAC)) return false; } } diff --git a/net/bridge/netfilter/ebt_stp.c b/net/bridge/netfilter/ebt_stp.c index e77f90bf8db3..45f73d55422f 100644 --- a/net/bridge/netfilter/ebt_stp.c +++ b/net/bridge/netfilter/ebt_stp.c @@ -46,7 +46,6 @@ static bool ebt_filter_config(const struct ebt_stp_info *info, const struct ebt_stp_config_info *c; u16 v16; u32 v32; - int verdict, i; c = &info->config; if ((info->bitmask & EBT_STP_FLAGS) && @@ -54,66 +53,62 @@ static bool ebt_filter_config(const struct ebt_stp_info *info, return false; if (info->bitmask & EBT_STP_ROOTPRIO) { v16 = NR16(stpc->root); - if (FWINV(v16 < c->root_priol || - v16 > c->root_priou, EBT_STP_ROOTPRIO)) + if (FWINV(v16 < c->root_priol || v16 > c->root_priou, + EBT_STP_ROOTPRIO)) return false; } if (info->bitmask & EBT_STP_ROOTADDR) { - verdict = 0; - for (i = 0; i < 6; i++) - verdict |= (stpc->root[2+i] ^ c->root_addr[i]) & - c->root_addrmsk[i]; - if (FWINV(verdict != 0, EBT_STP_ROOTADDR)) + if (FWINV(!ether_addr_equal_masked(&stpc->root[2], c->root_addr, + c->root_addrmsk), + EBT_STP_ROOTADDR)) return false; } if (info->bitmask & EBT_STP_ROOTCOST) { v32 = NR32(stpc->root_cost); - if (FWINV(v32 < c->root_costl || - v32 > c->root_costu, EBT_STP_ROOTCOST)) + if (FWINV(v32 < c->root_costl || v32 > c->root_costu, + EBT_STP_ROOTCOST)) return false; } if (info->bitmask & EBT_STP_SENDERPRIO) { v16 = NR16(stpc->sender); - if (FWINV(v16 < c->sender_priol || - v16 > c->sender_priou, EBT_STP_SENDERPRIO)) + if (FWINV(v16 < c->sender_priol || v16 > c->sender_priou, + EBT_STP_SENDERPRIO)) return false; } if (info->bitmask & EBT_STP_SENDERADDR) { - verdict = 0; - for (i = 0; i < 6; i++) - verdict |= (stpc->sender[2+i] ^ c->sender_addr[i]) & - c->sender_addrmsk[i]; - if (FWINV(verdict != 0, EBT_STP_SENDERADDR)) + if (FWINV(!ether_addr_equal_masked(&stpc->sender[2], + c->sender_addr, + c->sender_addrmsk), + EBT_STP_SENDERADDR)) return false; } if (info->bitmask & EBT_STP_PORT) { v16 = NR16(stpc->port); - if (FWINV(v16 < c->portl || - v16 > c->portu, EBT_STP_PORT)) + if (FWINV(v16 < c->portl || v16 > c->portu, EBT_STP_PORT)) return false; } if (info->bitmask & EBT_STP_MSGAGE) { v16 = NR16(stpc->msg_age); - if (FWINV(v16 < c->msg_agel || - v16 > c->msg_ageu, EBT_STP_MSGAGE)) + if (FWINV(v16 < c->msg_agel || v16 > c->msg_ageu, + EBT_STP_MSGAGE)) return false; } if (info->bitmask & EBT_STP_MAXAGE) { v16 = NR16(stpc->max_age); - if (FWINV(v16 < c->max_agel || - v16 > c->max_ageu, EBT_STP_MAXAGE)) + if (FWINV(v16 < c->max_agel || v16 > c->max_ageu, + EBT_STP_MAXAGE)) return false; } if (info->bitmask & EBT_STP_HELLOTIME) { v16 = NR16(stpc->hello_time); - if (FWINV(v16 < c->hello_timel || - v16 > c->hello_timeu, EBT_STP_HELLOTIME)) + if (FWINV(v16 < c->hello_timel || v16 > c->hello_timeu, + EBT_STP_HELLOTIME)) return false; } if (info->bitmask & EBT_STP_FWDD) { v16 = NR16(stpc->forward_delay); - if (FWINV(v16 < c->forward_delayl || - v16 > c->forward_delayu, EBT_STP_FWDD)) + if (FWINV(v16 < c->forward_delayl || v16 > c->forward_delayu, + EBT_STP_FWDD)) return false; } return true; diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index 5a61f35412a0..5721a25be860 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -130,7 +130,6 @@ ebt_basic_match(const struct ebt_entry *e, const struct sk_buff *skb, const struct ethhdr *h = eth_hdr(skb); const struct net_bridge_port *p; __be16 ethproto; - int verdict, i; if (skb_vlan_tag_present(skb)) ethproto = htons(ETH_P_8021Q); @@ -157,19 +156,15 @@ ebt_basic_match(const struct ebt_entry *e, const struct sk_buff *skb, return 1; if (e->bitmask & EBT_SOURCEMAC) { - verdict = 0; - for (i = 0; i < 6; i++) - verdict |= (h->h_source[i] ^ e->sourcemac[i]) & - e->sourcemsk[i]; - if (FWINV2(verdict != 0, EBT_ISOURCE)) + if (FWINV2(!ether_addr_equal_masked(h->h_source, + e->sourcemac, e->sourcemsk), + EBT_ISOURCE)) return 1; } if (e->bitmask & EBT_DESTMAC) { - verdict = 0; - for (i = 0; i < 6; i++) - verdict |= (h->h_dest[i] ^ e->destmac[i]) & - e->destmsk[i]; - if (FWINV2(verdict != 0, EBT_IDEST)) + if (FWINV2(!ether_addr_equal_masked(h->h_dest, + e->destmac, e->destmsk), + EBT_IDEST)) return 1; } return 0; -- cgit v1.2.3 From 1aacde3d22c42281236155c1ef6d7a5aa32a826b Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 30 Jun 2016 17:24:43 +0200 Subject: bpf: generally move prog destruction to RCU deferral Jann Horn reported following analysis that could potentially result in a very hard to trigger (if not impossible) UAF race, to quote his event timeline: - Set up a process with threads T1, T2 and T3 - Let T1 set up a socket filter F1 that invokes another filter F2 through a BPF map [tail call] - Let T1 trigger the socket filter via a unix domain socket write, don't wait for completion - Let T2 call PERF_EVENT_IOC_SET_BPF with F2, don't wait for completion - Now T2 should be behind bpf_prog_get(), but before bpf_prog_put() - Let T3 close the file descriptor for F2, dropping the reference count of F2 to 2 - At this point, T1 should have looked up F2 from the map, but not finished executing it - Let T3 remove F2 from the BPF map, dropping the reference count of F2 to 1 - Now T2 should call bpf_prog_put() (wrong BPF program type), dropping the reference count of F2 to 0 and scheduling bpf_prog_free_deferred() via schedule_work() - At this point, the BPF program could be freed - BPF execution is still running in a freed BPF program While at PERF_EVENT_IOC_SET_BPF time it's only guaranteed that the perf event fd we're doing the syscall on doesn't disappear from underneath us for whole syscall time, it may not be the case for the bpf fd used as an argument only after we did the put. It needs to be a valid fd pointing to a BPF program at the time of the call to make the bpf_prog_get() and while T2 gets preempted, F2 must have dropped reference to 1 on the other CPU. The fput() from the close() in T3 should also add additionally delay to the reference drop via exit_task_work() when bpf_prog_release() gets called as well as scheduling bpf_prog_free_deferred(). That said, it makes nevertheless sense to move the BPF prog destruction generally after RCU grace period to guarantee that such scenario above, but also others as recently fixed in ceb56070359b ("bpf, perf: delay release of BPF prog after grace period") with regards to tail calls won't happen. Integrating bpf_prog_free_deferred() directly into the RCU callback is not allowed since the invocation might happen from either softirq or process context, so we're not permitted to block. Reviewing all bpf_prog_put() invocations from eBPF side (note, cBPF -> eBPF progs don't use this for their destruction) with call_rcu() look good to me. Since we don't know whether at the time of attaching the program, we're already part of a tail call map, we need to use RCU variant. However, due to this, there won't be severely more stress on the RCU callback queue: situations with above bpf_prog_get() and bpf_prog_put() combo in practice normally won't lead to releases, but even if they would, enough effort/ cycles have to be put into loading a BPF program into the kernel already. Reported-by: Jann Horn Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/bpf.h | 5 ----- kernel/bpf/arraymap.c | 4 +--- kernel/bpf/syscall.c | 13 +++---------- kernel/events/core.c | 2 +- 4 files changed, 5 insertions(+), 19 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 8411032ac90d..749549888b86 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -220,7 +220,6 @@ void bpf_register_map_type(struct bpf_map_type_list *tl); struct bpf_prog *bpf_prog_get(u32 ufd); struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog); void bpf_prog_put(struct bpf_prog *prog); -void bpf_prog_put_rcu(struct bpf_prog *prog); struct bpf_map *bpf_map_get_with_uref(u32 ufd); struct bpf_map *__bpf_map_get(struct fd f); @@ -281,10 +280,6 @@ static inline struct bpf_prog *bpf_prog_get(u32 ufd) static inline void bpf_prog_put(struct bpf_prog *prog) { } - -static inline void bpf_prog_put_rcu(struct bpf_prog *prog) -{ -} #endif /* CONFIG_BPF_SYSCALL */ /* verifier prototypes for helper functions called from eBPF programs */ diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index 5af30732697b..4ec57a649b1f 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -390,9 +390,7 @@ static void *prog_fd_array_get_ptr(struct bpf_map *map, static void prog_fd_array_put_ptr(void *ptr) { - struct bpf_prog *prog = ptr; - - bpf_prog_put_rcu(prog); + bpf_prog_put(ptr); } /* decrement refcnt of all bpf_progs that are stored in this map */ diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index c23a4e9311b3..f6806a1d7ed9 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -623,7 +623,7 @@ static void bpf_prog_uncharge_memlock(struct bpf_prog *prog) free_uid(user); } -static void __prog_put_common(struct rcu_head *rcu) +static void __bpf_prog_put_rcu(struct rcu_head *rcu) { struct bpf_prog_aux *aux = container_of(rcu, struct bpf_prog_aux, rcu); @@ -632,17 +632,10 @@ static void __prog_put_common(struct rcu_head *rcu) bpf_prog_free(aux->prog); } -/* version of bpf_prog_put() that is called after a grace period */ -void bpf_prog_put_rcu(struct bpf_prog *prog) -{ - if (atomic_dec_and_test(&prog->aux->refcnt)) - call_rcu(&prog->aux->rcu, __prog_put_common); -} - void bpf_prog_put(struct bpf_prog *prog) { if (atomic_dec_and_test(&prog->aux->refcnt)) - __prog_put_common(&prog->aux->rcu); + call_rcu(&prog->aux->rcu, __bpf_prog_put_rcu); } EXPORT_SYMBOL_GPL(bpf_prog_put); @@ -650,7 +643,7 @@ static int bpf_prog_release(struct inode *inode, struct file *filp) { struct bpf_prog *prog = filp->private_data; - bpf_prog_put_rcu(prog); + bpf_prog_put(prog); return 0; } diff --git a/kernel/events/core.c b/kernel/events/core.c index 85cd41878a74..9c51ec3f0f44 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -7529,7 +7529,7 @@ static void perf_event_free_bpf_prog(struct perf_event *event) prog = event->tp_event->prog; if (prog) { event->tp_event->prog = NULL; - bpf_prog_put_rcu(prog); + bpf_prog_put(prog); } } -- cgit v1.2.3 From 113214be7f6c98dd6d0435e4765aea8dea91662c Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 30 Jun 2016 17:24:44 +0200 Subject: bpf: refactor bpf_prog_get and type check into helper Since bpf_prog_get() and program type check is used in a couple of places, refactor this into a small helper function that we can make use of. Since the non RO prog->aux part is not used in performance critical paths and a program destruction via RCU is rather very unlikley when doing the put, we shouldn't have an issue just doing the bpf_prog_get() + prog->type != type check, but actually not taking the ref at all (due to being in fdget() / fdput() section of the bpf fd) is even cleaner and makes the diff smaller as well, so just go for that. Callsites are changed to make use of the new helper where possible. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/bpf.h | 7 +++++++ kernel/bpf/syscall.c | 27 +++++++++++++++++++-------- net/core/filter.c | 13 +------------ net/kcm/kcmsock.c | 8 +------- net/packet/af_packet.c | 6 +----- net/sched/act_bpf.c | 7 +------ net/sched/cls_bpf.c | 7 +------ 7 files changed, 31 insertions(+), 44 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 749549888b86..b3336b4f5d04 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -218,6 +218,7 @@ void bpf_register_prog_type(struct bpf_prog_type_list *tl); void bpf_register_map_type(struct bpf_map_type_list *tl); struct bpf_prog *bpf_prog_get(u32 ufd); +struct bpf_prog *bpf_prog_get_type(u32 ufd, enum bpf_prog_type type); struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog); void bpf_prog_put(struct bpf_prog *prog); @@ -277,6 +278,12 @@ static inline struct bpf_prog *bpf_prog_get(u32 ufd) return ERR_PTR(-EOPNOTSUPP); } +static inline struct bpf_prog *bpf_prog_get_type(u32 ufd, + enum bpf_prog_type type) +{ + return ERR_PTR(-EOPNOTSUPP); +} + static inline void bpf_prog_put(struct bpf_prog *prog) { } diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index f6806a1d7ed9..22863d9872b1 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -657,7 +657,7 @@ int bpf_prog_new_fd(struct bpf_prog *prog) O_RDWR | O_CLOEXEC); } -static struct bpf_prog *__bpf_prog_get(struct fd f) +static struct bpf_prog *____bpf_prog_get(struct fd f) { if (!f.file) return ERR_PTR(-EBADF); @@ -678,24 +678,35 @@ struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog) return prog; } -/* called by sockets/tracing/seccomp before attaching program to an event - * pairs with bpf_prog_put() - */ -struct bpf_prog *bpf_prog_get(u32 ufd) +static struct bpf_prog *__bpf_prog_get(u32 ufd, enum bpf_prog_type *type) { struct fd f = fdget(ufd); struct bpf_prog *prog; - prog = __bpf_prog_get(f); + prog = ____bpf_prog_get(f); if (IS_ERR(prog)) return prog; + if (type && prog->type != *type) { + prog = ERR_PTR(-EINVAL); + goto out; + } prog = bpf_prog_inc(prog); +out: fdput(f); - return prog; } -EXPORT_SYMBOL_GPL(bpf_prog_get); + +struct bpf_prog *bpf_prog_get(u32 ufd) +{ + return __bpf_prog_get(ufd, NULL); +} + +struct bpf_prog *bpf_prog_get_type(u32 ufd, enum bpf_prog_type type) +{ + return __bpf_prog_get(ufd, &type); +} +EXPORT_SYMBOL_GPL(bpf_prog_get_type); /* last field in 'union bpf_attr' used by this command */ #define BPF_PROG_LOAD_LAST_FIELD kern_version diff --git a/net/core/filter.c b/net/core/filter.c index 76f9a4938be4..76fee35da244 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -1301,21 +1301,10 @@ int sk_reuseport_attach_filter(struct sock_fprog *fprog, struct sock *sk) static struct bpf_prog *__get_bpf(u32 ufd, struct sock *sk) { - struct bpf_prog *prog; - if (sock_flag(sk, SOCK_FILTER_LOCKED)) return ERR_PTR(-EPERM); - prog = bpf_prog_get(ufd); - if (IS_ERR(prog)) - return prog; - - if (prog->type != BPF_PROG_TYPE_SOCKET_FILTER) { - bpf_prog_put(prog); - return ERR_PTR(-EINVAL); - } - - return prog; + return bpf_prog_get_type(ufd, BPF_PROG_TYPE_SOCKET_FILTER); } int sk_attach_bpf(u32 ufd, struct sock *sk) diff --git a/net/kcm/kcmsock.c b/net/kcm/kcmsock.c index 0b68ba730a06..cb39e05b166c 100644 --- a/net/kcm/kcmsock.c +++ b/net/kcm/kcmsock.c @@ -1765,18 +1765,12 @@ static int kcm_attach_ioctl(struct socket *sock, struct kcm_attach *info) if (!csock) return -ENOENT; - prog = bpf_prog_get(info->bpf_fd); + prog = bpf_prog_get_type(info->bpf_fd, BPF_PROG_TYPE_SOCKET_FILTER); if (IS_ERR(prog)) { err = PTR_ERR(prog); goto out; } - if (prog->type != BPF_PROG_TYPE_SOCKET_FILTER) { - bpf_prog_put(prog); - err = -EINVAL; - goto out; - } - err = kcm_attach(sock, csock, prog); if (err) { bpf_prog_put(prog); diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index d1f3b9e977e5..48b58957adf4 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -1588,13 +1588,9 @@ static int fanout_set_data_ebpf(struct packet_sock *po, char __user *data, if (copy_from_user(&fd, data, len)) return -EFAULT; - new = bpf_prog_get(fd); + new = bpf_prog_get_type(fd, BPF_PROG_TYPE_SOCKET_FILTER); if (IS_ERR(new)) return PTR_ERR(new); - if (new->type != BPF_PROG_TYPE_SOCKET_FILTER) { - bpf_prog_put(new); - return -EINVAL; - } __fanout_set_data_bpf(po->fanout, new); return 0; diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c index f7b6cf49ea6f..ef74bffa6101 100644 --- a/net/sched/act_bpf.c +++ b/net/sched/act_bpf.c @@ -223,15 +223,10 @@ static int tcf_bpf_init_from_efd(struct nlattr **tb, struct tcf_bpf_cfg *cfg) bpf_fd = nla_get_u32(tb[TCA_ACT_BPF_FD]); - fp = bpf_prog_get(bpf_fd); + fp = bpf_prog_get_type(bpf_fd, BPF_PROG_TYPE_SCHED_ACT); if (IS_ERR(fp)) return PTR_ERR(fp); - if (fp->type != BPF_PROG_TYPE_SCHED_ACT) { - bpf_prog_put(fp); - return -EINVAL; - } - if (tb[TCA_ACT_BPF_NAME]) { name = kmemdup(nla_data(tb[TCA_ACT_BPF_NAME]), nla_len(tb[TCA_ACT_BPF_NAME]), diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c index 7b342c779da7..c3002c2c68bb 100644 --- a/net/sched/cls_bpf.c +++ b/net/sched/cls_bpf.c @@ -272,15 +272,10 @@ static int cls_bpf_prog_from_efd(struct nlattr **tb, struct cls_bpf_prog *prog, bpf_fd = nla_get_u32(tb[TCA_BPF_FD]); - fp = bpf_prog_get(bpf_fd); + fp = bpf_prog_get_type(bpf_fd, BPF_PROG_TYPE_SCHED_CLS); if (IS_ERR(fp)) return PTR_ERR(fp); - if (fp->type != BPF_PROG_TYPE_SCHED_CLS) { - bpf_prog_put(fp); - return -EINVAL; - } - if (tb[TCA_BPF_NAME]) { name = kmemdup(nla_data(tb[TCA_BPF_NAME]), nla_len(tb[TCA_BPF_NAME]), -- cgit v1.2.3 From 1f3fe7ebf6136c341012db9f554d4caa566fcbaa Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Thu, 30 Jun 2016 10:28:42 -0700 Subject: cgroup: Add cgroup_get_from_fd Add a helper function to get a cgroup2 from a fd. It will be stored in a bpf array (BPF_MAP_TYPE_CGROUP_ARRAY) which will be introduced in the later patch. Signed-off-by: Martin KaFai Lau Cc: Alexei Starovoitov Cc: Daniel Borkmann Cc: Tejun Heo Acked-by: Tejun Heo Signed-off-by: David S. Miller --- include/linux/cgroup.h | 1 + kernel/cgroup.c | 35 +++++++++++++++++++++++++++++++++++ 2 files changed, 36 insertions(+) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index a20320c666fd..984f73b719a9 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -87,6 +87,7 @@ struct cgroup_subsys_state *css_tryget_online_from_dir(struct dentry *dentry, struct cgroup_subsys *ss); struct cgroup *cgroup_get_from_path(const char *path); +struct cgroup *cgroup_get_from_fd(int fd); int cgroup_attach_task_all(struct task_struct *from, struct task_struct *); int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from); diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 75c0ff00aca6..50787cd61da2 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -62,6 +62,7 @@ #include #include #include +#include #include /* @@ -6209,6 +6210,40 @@ struct cgroup *cgroup_get_from_path(const char *path) } EXPORT_SYMBOL_GPL(cgroup_get_from_path); +/** + * cgroup_get_from_fd - get a cgroup pointer from a fd + * @fd: fd obtained by open(cgroup2_dir) + * + * Find the cgroup from a fd which should be obtained + * by opening a cgroup directory. Returns a pointer to the + * cgroup on success. ERR_PTR is returned if the cgroup + * cannot be found. + */ +struct cgroup *cgroup_get_from_fd(int fd) +{ + struct cgroup_subsys_state *css; + struct cgroup *cgrp; + struct file *f; + + f = fget_raw(fd); + if (!f) + return ERR_PTR(-EBADF); + + css = css_tryget_online_from_dir(f->f_path.dentry, NULL); + fput(f); + if (IS_ERR(css)) + return ERR_CAST(css); + + cgrp = css->cgroup; + if (!cgroup_on_dfl(cgrp)) { + cgroup_put(cgrp); + return ERR_PTR(-EBADF); + } + + return cgrp; +} +EXPORT_SYMBOL_GPL(cgroup_get_from_fd); + /* * sock->sk_cgrp_data handling. For more info, see sock_cgroup_data * definition in cgroup-defs.h. -- cgit v1.2.3 From acbc2004d7129a1ecf02414c1da8808bdc06d5a2 Mon Sep 17 00:00:00 2001 From: Or Gerlitz Date: Fri, 1 Jul 2016 14:50:58 +0300 Subject: net/mlx5: Introduce offloads steering namespace Add a new namespace (MLX5_FLOW_NAMESPACE_OFFLOADS) to be populated with flow steering rules that deal with rules that have have to be executed before the EN NIC steering rules are matched. The namespace is located after the bypass name-space and before the kernel name-space. Therefore, it precedes the HW processing done for rules set for the kernel NIC name-space. Under SRIOV, it would allow us to match on e-switch missed packet and forward them to the relevant VF representor TIR. Signed-off-by: Or Gerlitz Signed-off-by: Amir Vadai Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 11 ++++++++++- include/linux/mlx5/fs.h | 1 + 2 files changed, 11 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index e912a3d2505e..b0401104afb9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -83,6 +83,11 @@ #define ANCHOR_NUM_LEVELS 1 #define ANCHOR_NUM_PRIOS 1 #define ANCHOR_MIN_LEVEL (BY_PASS_MIN_LEVEL + 1) + +#define OFFLOADS_MAX_FT 1 +#define OFFLOADS_NUM_PRIOS 1 +#define OFFLOADS_MIN_LEVEL (ANCHOR_MIN_LEVEL + 1) + struct node_caps { size_t arr_sz; long *caps; @@ -98,7 +103,7 @@ static struct init_tree_node { int num_levels; } root_fs = { .type = FS_TYPE_NAMESPACE, - .ar_size = 4, + .ar_size = 5, .children = (struct init_tree_node[]) { ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0, FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en), @@ -107,6 +112,9 @@ static struct init_tree_node { FS_CAP(flow_table_properties_nic_receive.flow_table_modify)), ADD_NS(ADD_MULTIPLE_PRIO(MLX5_BY_PASS_NUM_PRIOS, BY_PASS_PRIO_NUM_LEVELS))), + ADD_PRIO(0, OFFLOADS_MIN_LEVEL, 0, {}, + ADD_NS(ADD_MULTIPLE_PRIO(OFFLOADS_NUM_PRIOS, OFFLOADS_MAX_FT))), + ADD_PRIO(0, KERNEL_MIN_LEVEL, 0, {}, ADD_NS(ADD_MULTIPLE_PRIO(1, 1), ADD_MULTIPLE_PRIO(KERNEL_NIC_NUM_PRIOS, @@ -1369,6 +1377,7 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, switch (type) { case MLX5_FLOW_NAMESPACE_BYPASS: + case MLX5_FLOW_NAMESPACE_OFFLOADS: case MLX5_FLOW_NAMESPACE_KERNEL: case MLX5_FLOW_NAMESPACE_LEFTOVERS: case MLX5_FLOW_NAMESPACE_ANCHOR: diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index 4b7a107d9c19..6ad111938709 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -54,6 +54,7 @@ static inline void build_leftovers_ft_param(int *priority, enum mlx5_flow_namespace_type { MLX5_FLOW_NAMESPACE_BYPASS, + MLX5_FLOW_NAMESPACE_OFFLOADS, MLX5_FLOW_NAMESPACE_KERNEL, MLX5_FLOW_NAMESPACE_LEFTOVERS, MLX5_FLOW_NAMESPACE_ANCHOR, -- cgit v1.2.3 From b50d292b4399f4eb11e82d0430aacf62dd5d5365 Mon Sep 17 00:00:00 2001 From: Hadar Hen Zion Date: Fri, 1 Jul 2016 14:51:04 +0300 Subject: net/mlx5e: Create NIC global resources only once To allow creating more than one netdev over the same PCI function, we change the driver such that global NIC resources are created once and later be shared amongst all the mlx5e netdevs running over that port. Move the CQ UAR, PD (pdn), Transport Domain (tdn), MKey resources from being kept in the mlx5e priv part to a new resources structure (mlx5e_resources) placed under the mlx5_core device. This patch doesn't add any new functionality. Signed-off-by: Hadar Hen Zion Reviewed-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/Makefile | 6 +- drivers/net/ethernet/mellanox/mlx5/core/en.h | 6 +- .../net/ethernet/mellanox/mlx5/core/en_common.c | 112 +++++++++++++++++++ drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 124 +++++++-------------- include/linux/mlx5/driver.h | 13 +++ 5 files changed, 171 insertions(+), 90 deletions(-) create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_common.c (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index 96f18264a0ef..9b14dadd9309 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -6,8 +6,8 @@ mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ fs_counters.o rl.o mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o eswitch_offloads.o \ - en_main.o en_fs.o en_ethtool.o en_tx.o en_rx.o \ - en_rx_am.o en_txrx.o en_clock.o vxlan.o en_tc.o \ - en_arfs.o + en_main.o en_common.o en_fs.o en_ethtool.o en_tx.o \ + en_rx.o en_rx_am.o en_txrx.o en_clock.o vxlan.o \ + en_tc.o en_arfs.o mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) += en_dcbnl.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index b97511bf4c7b..3226b92a344d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -570,10 +570,6 @@ struct mlx5e_priv { unsigned long state; struct mutex state_lock; /* Protects Interface state */ - struct mlx5_uar cq_uar; - u32 pdn; - u32 tdn; - struct mlx5_core_mkey mkey; struct mlx5_core_mkey umr_mkey; struct mlx5e_rq drop_rq; @@ -788,5 +784,7 @@ int mlx5e_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb, #endif u16 mlx5e_get_max_inline_cap(struct mlx5_core_dev *mdev); +int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev); +void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev); #endif /* __MLX5_EN_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c new file mode 100644 index 000000000000..33b373216594 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c @@ -0,0 +1,112 @@ +/* + * Copyright (c) 2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include "en.h" + +/* mlx5e global resources should be placed in this file. + * Global resources are common to all the netdevices crated on the same nic. + */ + +static int mlx5e_create_mkey(struct mlx5_core_dev *mdev, u32 pdn, + struct mlx5_core_mkey *mkey) +{ + struct mlx5_create_mkey_mbox_in *in; + int err; + + in = mlx5_vzalloc(sizeof(*in)); + if (!in) + return -ENOMEM; + + in->seg.flags = MLX5_PERM_LOCAL_WRITE | + MLX5_PERM_LOCAL_READ | + MLX5_ACCESS_MODE_PA; + in->seg.flags_pd = cpu_to_be32(pdn | MLX5_MKEY_LEN64); + in->seg.qpn_mkey7_0 = cpu_to_be32(0xffffff << 8); + + err = mlx5_core_create_mkey(mdev, mkey, in, sizeof(*in), NULL, NULL, + NULL); + + kvfree(in); + + return err; +} + +int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev) +{ + struct mlx5e_resources *res = &mdev->mlx5e_res; + int err; + + err = mlx5_alloc_map_uar(mdev, &res->cq_uar, false); + if (err) { + mlx5_core_err(mdev, "alloc_map uar failed, %d\n", err); + return err; + } + + err = mlx5_core_alloc_pd(mdev, &res->pdn); + if (err) { + mlx5_core_err(mdev, "alloc pd failed, %d\n", err); + goto err_unmap_free_uar; + } + + err = mlx5_core_alloc_transport_domain(mdev, &res->td.tdn); + if (err) { + mlx5_core_err(mdev, "alloc td failed, %d\n", err); + goto err_dealloc_pd; + } + + err = mlx5e_create_mkey(mdev, res->pdn, &res->mkey); + if (err) { + mlx5_core_err(mdev, "create mkey failed, %d\n", err); + goto err_dealloc_transport_domain; + } + + return 0; + +err_dealloc_transport_domain: + mlx5_core_dealloc_transport_domain(mdev, res->td.tdn); +err_dealloc_pd: + mlx5_core_dealloc_pd(mdev, res->pdn); +err_unmap_free_uar: + mlx5_unmap_free_uar(mdev, &res->cq_uar); + + return err; +} + +void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev) +{ + struct mlx5e_resources *res = &mdev->mlx5e_res; + + mlx5_core_destroy_mkey(mdev, &res->mkey); + mlx5_core_dealloc_transport_domain(mdev, res->td.tdn); + mlx5_core_dealloc_pd(mdev, res->pdn); + mlx5_unmap_free_uar(mdev, &res->cq_uar); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index a64ce5df5810..9b2e2b211cee 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -858,7 +858,7 @@ static int mlx5e_create_cq(struct mlx5e_channel *c, mcq->comp = mlx5e_completion_event; mcq->event = mlx5e_cq_error_event; mcq->irqn = irqn; - mcq->uar = &priv->cq_uar; + mcq->uar = &mdev->mlx5e_res.cq_uar; for (i = 0; i < mlx5_cqwq_get_size(&cq->wq); i++) { struct mlx5_cqe64 *cqe = mlx5_cqwq_get_wqe(&cq->wq, i); @@ -1136,7 +1136,7 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, c->cpu = cpu; c->pdev = &priv->mdev->pdev->dev; c->netdev = priv->netdev; - c->mkey_be = cpu_to_be32(priv->mkey.key); + c->mkey_be = cpu_to_be32(priv->mdev->mlx5e_res.mkey.key); c->num_tc = priv->params.num_tc; if (priv->params.rx_am_enabled) @@ -1252,7 +1252,7 @@ static void mlx5e_build_rq_param(struct mlx5e_priv *priv, MLX5_SET(wq, wq, end_padding_mode, MLX5_WQ_END_PAD_MODE_ALIGN); MLX5_SET(wq, wq, log_wq_stride, ilog2(sizeof(struct mlx5e_rx_wqe))); MLX5_SET(wq, wq, log_wq_sz, priv->params.log_rq_size); - MLX5_SET(wq, wq, pd, priv->pdn); + MLX5_SET(wq, wq, pd, priv->mdev->mlx5e_res.pdn); MLX5_SET(rqc, rqc, counter_set_id, priv->q_counter); param->wq.buf_numa_node = dev_to_node(&priv->mdev->pdev->dev); @@ -1277,7 +1277,7 @@ static void mlx5e_build_sq_param_common(struct mlx5e_priv *priv, void *wq = MLX5_ADDR_OF(sqc, sqc, wq); MLX5_SET(wq, wq, log_wq_stride, ilog2(MLX5_SEND_WQE_BB)); - MLX5_SET(wq, wq, pd, priv->pdn); + MLX5_SET(wq, wq, pd, priv->mdev->mlx5e_res.pdn); param->wq.buf_numa_node = dev_to_node(&priv->mdev->pdev->dev); } @@ -1299,7 +1299,7 @@ static void mlx5e_build_common_cq_param(struct mlx5e_priv *priv, { void *cqc = param->cqc; - MLX5_SET(cqc, cqc, uar_page, priv->cq_uar.index); + MLX5_SET(cqc, cqc, uar_page, priv->mdev->mlx5e_res.cq_uar.index); } static void mlx5e_build_rx_cq_param(struct mlx5e_priv *priv, @@ -1920,7 +1920,7 @@ static int mlx5e_create_drop_cq(struct mlx5e_priv *priv, mcq->comp = mlx5e_completion_event; mcq->event = mlx5e_cq_error_event; mcq->irqn = irqn; - mcq->uar = &priv->cq_uar; + mcq->uar = &mdev->mlx5e_res.cq_uar; cq->priv = priv; @@ -1986,7 +1986,7 @@ static int mlx5e_create_tis(struct mlx5e_priv *priv, int tc) memset(in, 0, sizeof(in)); MLX5_SET(tisc, tisc, prio, tc << 1); - MLX5_SET(tisc, tisc, transport_domain, priv->tdn); + MLX5_SET(tisc, tisc, transport_domain, mdev->mlx5e_res.td.tdn); return mlx5_core_create_tis(mdev, in, sizeof(in), &priv->tisn[tc]); } @@ -2029,7 +2029,7 @@ static void mlx5e_build_indir_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, { void *hfso = MLX5_ADDR_OF(tirc, tirc, rx_hash_field_selector_outer); - MLX5_SET(tirc, tirc, transport_domain, priv->tdn); + MLX5_SET(tirc, tirc, transport_domain, priv->mdev->mlx5e_res.td.tdn); #define MLX5_HASH_IP (MLX5_HASH_FIELD_SEL_SRC_IP |\ MLX5_HASH_FIELD_SEL_DST_IP) @@ -2136,7 +2136,7 @@ static void mlx5e_build_indir_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, static void mlx5e_build_direct_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, u32 rqtn) { - MLX5_SET(tirc, tirc, transport_domain, priv->tdn); + MLX5_SET(tirc, tirc, transport_domain, priv->mdev->mlx5e_res.td.tdn); mlx5e_build_tir_ctx_lro(tirc, priv); @@ -3082,31 +3082,6 @@ static void mlx5e_build_netdev(struct net_device *netdev) mlx5e_set_netdev_dev_addr(netdev); } -static int mlx5e_create_mkey(struct mlx5e_priv *priv, u32 pdn, - struct mlx5_core_mkey *mkey) -{ - struct mlx5_core_dev *mdev = priv->mdev; - struct mlx5_create_mkey_mbox_in *in; - int err; - - in = mlx5_vzalloc(sizeof(*in)); - if (!in) - return -ENOMEM; - - in->seg.flags = MLX5_PERM_LOCAL_WRITE | - MLX5_PERM_LOCAL_READ | - MLX5_ACCESS_MODE_PA; - in->seg.flags_pd = cpu_to_be32(pdn | MLX5_MKEY_LEN64); - in->seg.qpn_mkey7_0 = cpu_to_be32(0xffffff << 8); - - err = mlx5_core_create_mkey(mdev, mkey, in, sizeof(*in), NULL, NULL, - NULL); - - kvfree(in); - - return err; -} - static void mlx5e_create_q_counter(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; @@ -3149,7 +3124,7 @@ static int mlx5e_create_umr_mkey(struct mlx5e_priv *priv) MLX5_ACCESS_MODE_MTT; mkc->qpn_mkey7_0 = cpu_to_be32(0xffffff << 8); - mkc->flags_pd = cpu_to_be32(priv->pdn); + mkc->flags_pd = cpu_to_be32(mdev->mlx5e_res.pdn); mkc->len = cpu_to_be64(npages << PAGE_SHIFT); mkc->xlt_oct_size = cpu_to_be32(mlx5e_get_mtt_octw(npages)); mkc->log2_page_size = PAGE_SHIFT; @@ -3169,9 +3144,6 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev) int nch = mlx5e_get_max_num_channels(mdev); int err; - if (mlx5e_check_required_hca_cap(mdev)) - return NULL; - netdev = alloc_etherdev_mqs(sizeof(struct mlx5e_priv), nch * MLX5E_MAX_NUM_TC, nch); @@ -3191,34 +3163,10 @@ static void *mlx5e_create_netdev(struct mlx5_core_dev *mdev) if (!priv->wq) goto err_free_netdev; - err = mlx5_alloc_map_uar(mdev, &priv->cq_uar, false); - if (err) { - mlx5_core_err(mdev, "alloc_map uar failed, %d\n", err); - goto err_destroy_wq; - } - - err = mlx5_core_alloc_pd(mdev, &priv->pdn); - if (err) { - mlx5_core_err(mdev, "alloc pd failed, %d\n", err); - goto err_unmap_free_uar; - } - - err = mlx5_core_alloc_transport_domain(mdev, &priv->tdn); - if (err) { - mlx5_core_err(mdev, "alloc td failed, %d\n", err); - goto err_dealloc_pd; - } - - err = mlx5e_create_mkey(priv, priv->pdn, &priv->mkey); - if (err) { - mlx5_core_err(mdev, "create mkey failed, %d\n", err); - goto err_dealloc_transport_domain; - } - err = mlx5e_create_umr_mkey(priv); if (err) { mlx5_core_err(mdev, "create umr mkey failed, %d\n", err); - goto err_destroy_mkey; + goto err_destroy_wq; } err = mlx5e_create_tises(priv); @@ -3304,18 +3252,6 @@ err_destroy_tises: err_destroy_umr_mkey: mlx5_core_destroy_mkey(mdev, &priv->umr_mkey); -err_destroy_mkey: - mlx5_core_destroy_mkey(mdev, &priv->mkey); - -err_dealloc_transport_domain: - mlx5_core_dealloc_transport_domain(mdev, priv->tdn); - -err_dealloc_pd: - mlx5_core_dealloc_pd(mdev, priv->pdn); - -err_unmap_free_uar: - mlx5_unmap_free_uar(mdev, &priv->cq_uar); - err_destroy_wq: destroy_workqueue(priv->wq); @@ -3325,9 +3261,27 @@ err_free_netdev: return NULL; } -static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, void *vpriv) +static void *mlx5e_add(struct mlx5_core_dev *mdev) +{ + void *ret; + + if (mlx5e_check_required_hca_cap(mdev)) + return NULL; + + if (mlx5e_create_mdev_resources(mdev)) + return NULL; + + ret = mlx5e_create_netdev(mdev); + if (!ret) { + mlx5e_destroy_mdev_resources(mdev); + return NULL; + } + return ret; +} + +static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, + struct mlx5e_priv *priv) { - struct mlx5e_priv *priv = vpriv; struct net_device *netdev = priv->netdev; set_bit(MLX5E_STATE_DESTROYING, &priv->state); @@ -3351,10 +3305,6 @@ static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, void *vpriv) mlx5e_close_drop_rq(priv); mlx5e_destroy_tises(priv); mlx5_core_destroy_mkey(priv->mdev, &priv->umr_mkey); - mlx5_core_destroy_mkey(priv->mdev, &priv->mkey); - mlx5_core_dealloc_transport_domain(priv->mdev, priv->tdn); - mlx5_core_dealloc_pd(priv->mdev, priv->pdn); - mlx5_unmap_free_uar(priv->mdev, &priv->cq_uar); cancel_delayed_work_sync(&priv->update_stats_work); destroy_workqueue(priv->wq); @@ -3362,6 +3312,14 @@ static void mlx5e_destroy_netdev(struct mlx5_core_dev *mdev, void *vpriv) free_netdev(netdev); } +static void mlx5e_remove(struct mlx5_core_dev *mdev, void *vpriv) +{ + struct mlx5e_priv *priv = vpriv; + + mlx5e_destroy_netdev(mdev, priv); + mlx5e_destroy_mdev_resources(mdev); +} + static void *mlx5e_get_netdev(void *vpriv) { struct mlx5e_priv *priv = vpriv; @@ -3370,8 +3328,8 @@ static void *mlx5e_get_netdev(void *vpriv) } static struct mlx5_interface mlx5e_interface = { - .add = mlx5e_create_netdev, - .remove = mlx5e_destroy_netdev, + .add = mlx5e_add, + .remove = mlx5e_remove, .event = mlx5e_async_event, .protocol = MLX5_INTERFACE_PROTOCOL_ETH, .get_dev = mlx5e_get_netdev, diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 46260fdc5305..e22b3456b2ee 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -578,6 +578,18 @@ enum mlx5_pci_status { MLX5_PCI_STATUS_ENABLED, }; +struct mlx5_td { + struct list_head tirs_list; + u32 tdn; +}; + +struct mlx5e_resources { + struct mlx5_uar cq_uar; + u32 pdn; + struct mlx5_td td; + struct mlx5_core_mkey mkey; +}; + struct mlx5_core_dev { struct pci_dev *pdev; /* sync pci state */ @@ -602,6 +614,7 @@ struct mlx5_core_dev { struct mlx5_profile *profile; atomic_t num_qps; u32 issi; + struct mlx5e_resources mlx5e_res; #ifdef CONFIG_RFS_ACCEL struct cpu_rmap *rmap; #endif -- cgit v1.2.3 From c37a2dfa67f7920b14ea77dc9f9f9660f7a1f6dd Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Fri, 24 Jun 2016 13:25:22 -0700 Subject: netfilter: Convert FWINV<[foo]> macros and uses to NF_INVF netfilter uses multiple FWINV #defines with identical form that hide a specific structure variable and dereference it with a invflags member. $ git grep "#define FWINV" include/linux/netfilter_bridge/ebtables.h:#define FWINV(bool,invflg) ((bool) ^ !!(info->invflags & invflg)) net/bridge/netfilter/ebtables.c:#define FWINV2(bool, invflg) ((bool) ^ !!(e->invflags & invflg)) net/ipv4/netfilter/arp_tables.c:#define FWINV(bool, invflg) ((bool) ^ !!(arpinfo->invflags & (invflg))) net/ipv4/netfilter/ip_tables.c:#define FWINV(bool, invflg) ((bool) ^ !!(ipinfo->invflags & (invflg))) net/ipv6/netfilter/ip6_tables.c:#define FWINV(bool, invflg) ((bool) ^ !!(ip6info->invflags & (invflg))) net/netfilter/xt_tcpudp.c:#define FWINVTCP(bool, invflg) ((bool) ^ !!(tcpinfo->invflags & (invflg))) Consolidate these macros into a single NF_INVF macro. Miscellanea: o Neaten the alignment around these uses o A few lines are > 80 columns for intelligibility Signed-off-by: Joe Perches Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/x_tables.h | 4 +++ include/linux/netfilter_bridge/ebtables.h | 2 -- net/bridge/netfilter/ebt_802_3.c | 6 ++-- net/bridge/netfilter/ebt_arp.c | 38 +++++++++++----------- net/bridge/netfilter/ebt_ip.c | 28 ++++++++--------- net/bridge/netfilter/ebt_ip6.c | 41 +++++++++++++----------- net/bridge/netfilter/ebt_stp.c | 52 ++++++++++++++++--------------- net/bridge/netfilter/ebtables.c | 27 ++++++++-------- net/ipv4/netfilter/arp_tables.c | 41 ++++++++++++------------ net/ipv4/netfilter/ip_tables.c | 20 ++++++------ net/ipv6/netfilter/ip6_tables.c | 16 +++++----- net/netfilter/xt_tcpudp.c | 7 ++--- 12 files changed, 144 insertions(+), 138 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index dc4f58a3cdcc..e94e81ab2b58 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -6,6 +6,10 @@ #include #include +/* Test a struct->invflags and a boolean for inequality */ +#define NF_INVF(ptr, flag, boolean) \ + ((boolean) ^ !!((ptr)->invflags & (flag))) + /** * struct xt_action_param - parameters for matches/targets * diff --git a/include/linux/netfilter_bridge/ebtables.h b/include/linux/netfilter_bridge/ebtables.h index 2ea517c7c6b9..984b2112c77b 100644 --- a/include/linux/netfilter_bridge/ebtables.h +++ b/include/linux/netfilter_bridge/ebtables.h @@ -115,8 +115,6 @@ extern unsigned int ebt_do_table(struct sk_buff *skb, const struct nf_hook_state *state, struct ebt_table *table); -/* Used in the kernel match() functions */ -#define FWINV(bool,invflg) ((bool) ^ !!(info->invflags & invflg)) /* True if the hook mask denotes that the rule is in a base chain, * used in the check() functions */ #define BASE_CHAIN (par->hook_mask & (1 << NF_BR_NUMHOOKS)) diff --git a/net/bridge/netfilter/ebt_802_3.c b/net/bridge/netfilter/ebt_802_3.c index 2a449b7ab8fa..5fc4affd9fdb 100644 --- a/net/bridge/netfilter/ebt_802_3.c +++ b/net/bridge/netfilter/ebt_802_3.c @@ -20,16 +20,16 @@ ebt_802_3_mt(const struct sk_buff *skb, struct xt_action_param *par) __be16 type = hdr->llc.ui.ctrl & IS_UI ? hdr->llc.ui.type : hdr->llc.ni.type; if (info->bitmask & EBT_802_3_SAP) { - if (FWINV(info->sap != hdr->llc.ui.ssap, EBT_802_3_SAP)) + if (NF_INVF(info, EBT_802_3_SAP, info->sap != hdr->llc.ui.ssap)) return false; - if (FWINV(info->sap != hdr->llc.ui.dsap, EBT_802_3_SAP)) + if (NF_INVF(info, EBT_802_3_SAP, info->sap != hdr->llc.ui.dsap)) return false; } if (info->bitmask & EBT_802_3_TYPE) { if (!(hdr->llc.ui.dsap == CHECK_TYPE && hdr->llc.ui.ssap == CHECK_TYPE)) return false; - if (FWINV(info->type != type, EBT_802_3_TYPE)) + if (NF_INVF(info, EBT_802_3_TYPE, info->type != type)) return false; } diff --git a/net/bridge/netfilter/ebt_arp.c b/net/bridge/netfilter/ebt_arp.c index cca0a899ee15..227142282b45 100644 --- a/net/bridge/netfilter/ebt_arp.c +++ b/net/bridge/netfilter/ebt_arp.c @@ -25,14 +25,14 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) ah = skb_header_pointer(skb, 0, sizeof(_arph), &_arph); if (ah == NULL) return false; - if (info->bitmask & EBT_ARP_OPCODE && FWINV(info->opcode != - ah->ar_op, EBT_ARP_OPCODE)) + if ((info->bitmask & EBT_ARP_OPCODE) && + NF_INVF(info, EBT_ARP_OPCODE, info->opcode != ah->ar_op)) return false; - if (info->bitmask & EBT_ARP_HTYPE && FWINV(info->htype != - ah->ar_hrd, EBT_ARP_HTYPE)) + if ((info->bitmask & EBT_ARP_HTYPE) && + NF_INVF(info, EBT_ARP_HTYPE, info->htype != ah->ar_hrd)) return false; - if (info->bitmask & EBT_ARP_PTYPE && FWINV(info->ptype != - ah->ar_pro, EBT_ARP_PTYPE)) + if ((info->bitmask & EBT_ARP_PTYPE) && + NF_INVF(info, EBT_ARP_PTYPE, info->ptype != ah->ar_pro)) return false; if (info->bitmask & (EBT_ARP_SRC_IP | EBT_ARP_DST_IP | EBT_ARP_GRAT)) { @@ -51,14 +51,16 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) sizeof(daddr), &daddr); if (dap == NULL) return false; - if (info->bitmask & EBT_ARP_SRC_IP && - FWINV(info->saddr != (*sap & info->smsk), EBT_ARP_SRC_IP)) + if ((info->bitmask & EBT_ARP_SRC_IP) && + NF_INVF(info, EBT_ARP_SRC_IP, + info->saddr != (*sap & info->smsk))) return false; - if (info->bitmask & EBT_ARP_DST_IP && - FWINV(info->daddr != (*dap & info->dmsk), EBT_ARP_DST_IP)) + if ((info->bitmask & EBT_ARP_DST_IP) && + NF_INVF(info, EBT_ARP_DST_IP, + info->daddr != (*dap & info->dmsk))) return false; - if (info->bitmask & EBT_ARP_GRAT && - FWINV(*dap != *sap, EBT_ARP_GRAT)) + if ((info->bitmask & EBT_ARP_GRAT) && + NF_INVF(info, EBT_ARP_GRAT, *dap != *sap)) return false; } @@ -73,9 +75,9 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) sizeof(_mac), &_mac); if (mp == NULL) return false; - if (FWINV(!ether_addr_equal_masked(mp, info->smaddr, - info->smmsk), - EBT_ARP_SRC_MAC)) + if (NF_INVF(info, EBT_ARP_SRC_MAC, + !ether_addr_equal_masked(mp, info->smaddr, + info->smmsk))) return false; } @@ -85,9 +87,9 @@ ebt_arp_mt(const struct sk_buff *skb, struct xt_action_param *par) sizeof(_mac), &_mac); if (mp == NULL) return false; - if (FWINV(!ether_addr_equal_masked(mp, info->dmaddr, - info->dmmsk), - EBT_ARP_DST_MAC)) + if (NF_INVF(info, EBT_ARP_DST_MAC, + !ether_addr_equal_masked(mp, info->dmaddr, + info->dmmsk))) return false; } } diff --git a/net/bridge/netfilter/ebt_ip.c b/net/bridge/netfilter/ebt_ip.c index 23bca62d58d2..d06968bdf5ec 100644 --- a/net/bridge/netfilter/ebt_ip.c +++ b/net/bridge/netfilter/ebt_ip.c @@ -36,19 +36,19 @@ ebt_ip_mt(const struct sk_buff *skb, struct xt_action_param *par) ih = skb_header_pointer(skb, 0, sizeof(_iph), &_iph); if (ih == NULL) return false; - if (info->bitmask & EBT_IP_TOS && - FWINV(info->tos != ih->tos, EBT_IP_TOS)) + if ((info->bitmask & EBT_IP_TOS) && + NF_INVF(info, EBT_IP_TOS, info->tos != ih->tos)) return false; - if (info->bitmask & EBT_IP_SOURCE && - FWINV((ih->saddr & info->smsk) != - info->saddr, EBT_IP_SOURCE)) + if ((info->bitmask & EBT_IP_SOURCE) && + NF_INVF(info, EBT_IP_SOURCE, + (ih->saddr & info->smsk) != info->saddr)) return false; if ((info->bitmask & EBT_IP_DEST) && - FWINV((ih->daddr & info->dmsk) != - info->daddr, EBT_IP_DEST)) + NF_INVF(info, EBT_IP_DEST, + (ih->daddr & info->dmsk) != info->daddr)) return false; if (info->bitmask & EBT_IP_PROTO) { - if (FWINV(info->protocol != ih->protocol, EBT_IP_PROTO)) + if (NF_INVF(info, EBT_IP_PROTO, info->protocol != ih->protocol)) return false; if (!(info->bitmask & EBT_IP_DPORT) && !(info->bitmask & EBT_IP_SPORT)) @@ -61,16 +61,16 @@ ebt_ip_mt(const struct sk_buff *skb, struct xt_action_param *par) return false; if (info->bitmask & EBT_IP_DPORT) { u32 dst = ntohs(pptr->dst); - if (FWINV(dst < info->dport[0] || - dst > info->dport[1], - EBT_IP_DPORT)) + if (NF_INVF(info, EBT_IP_DPORT, + dst < info->dport[0] || + dst > info->dport[1])) return false; } if (info->bitmask & EBT_IP_SPORT) { u32 src = ntohs(pptr->src); - if (FWINV(src < info->sport[0] || - src > info->sport[1], - EBT_IP_SPORT)) + if (NF_INVF(info, EBT_IP_SPORT, + src < info->sport[0] || + src > info->sport[1])) return false; } } diff --git a/net/bridge/netfilter/ebt_ip6.c b/net/bridge/netfilter/ebt_ip6.c index 98de6e7fd86d..4617491be41e 100644 --- a/net/bridge/netfilter/ebt_ip6.c +++ b/net/bridge/netfilter/ebt_ip6.c @@ -45,15 +45,18 @@ ebt_ip6_mt(const struct sk_buff *skb, struct xt_action_param *par) ih6 = skb_header_pointer(skb, 0, sizeof(_ip6h), &_ip6h); if (ih6 == NULL) return false; - if (info->bitmask & EBT_IP6_TCLASS && - FWINV(info->tclass != ipv6_get_dsfield(ih6), EBT_IP6_TCLASS)) + if ((info->bitmask & EBT_IP6_TCLASS) && + NF_INVF(info, EBT_IP6_TCLASS, + info->tclass != ipv6_get_dsfield(ih6))) return false; - if ((info->bitmask & EBT_IP6_SOURCE && - FWINV(ipv6_masked_addr_cmp(&ih6->saddr, &info->smsk, - &info->saddr), EBT_IP6_SOURCE)) || - (info->bitmask & EBT_IP6_DEST && - FWINV(ipv6_masked_addr_cmp(&ih6->daddr, &info->dmsk, - &info->daddr), EBT_IP6_DEST))) + if (((info->bitmask & EBT_IP6_SOURCE) && + NF_INVF(info, EBT_IP6_SOURCE, + ipv6_masked_addr_cmp(&ih6->saddr, &info->smsk, + &info->saddr))) || + ((info->bitmask & EBT_IP6_DEST) && + NF_INVF(info, EBT_IP6_DEST, + ipv6_masked_addr_cmp(&ih6->daddr, &info->dmsk, + &info->daddr)))) return false; if (info->bitmask & EBT_IP6_PROTO) { uint8_t nexthdr = ih6->nexthdr; @@ -63,7 +66,7 @@ ebt_ip6_mt(const struct sk_buff *skb, struct xt_action_param *par) offset_ph = ipv6_skip_exthdr(skb, sizeof(_ip6h), &nexthdr, &frag_off); if (offset_ph == -1) return false; - if (FWINV(info->protocol != nexthdr, EBT_IP6_PROTO)) + if (NF_INVF(info, EBT_IP6_PROTO, info->protocol != nexthdr)) return false; if (!(info->bitmask & (EBT_IP6_DPORT | EBT_IP6_SPORT | EBT_IP6_ICMP6))) @@ -76,22 +79,24 @@ ebt_ip6_mt(const struct sk_buff *skb, struct xt_action_param *par) return false; if (info->bitmask & EBT_IP6_DPORT) { u16 dst = ntohs(pptr->tcpudphdr.dst); - if (FWINV(dst < info->dport[0] || - dst > info->dport[1], EBT_IP6_DPORT)) + if (NF_INVF(info, EBT_IP6_DPORT, + dst < info->dport[0] || + dst > info->dport[1])) return false; } if (info->bitmask & EBT_IP6_SPORT) { u16 src = ntohs(pptr->tcpudphdr.src); - if (FWINV(src < info->sport[0] || - src > info->sport[1], EBT_IP6_SPORT)) + if (NF_INVF(info, EBT_IP6_SPORT, + src < info->sport[0] || + src > info->sport[1])) return false; } if ((info->bitmask & EBT_IP6_ICMP6) && - FWINV(pptr->icmphdr.type < info->icmpv6_type[0] || - pptr->icmphdr.type > info->icmpv6_type[1] || - pptr->icmphdr.code < info->icmpv6_code[0] || - pptr->icmphdr.code > info->icmpv6_code[1], - EBT_IP6_ICMP6)) + NF_INVF(info, EBT_IP6_ICMP6, + pptr->icmphdr.type < info->icmpv6_type[0] || + pptr->icmphdr.type > info->icmpv6_type[1] || + pptr->icmphdr.code < info->icmpv6_code[0] || + pptr->icmphdr.code > info->icmpv6_code[1])) return false; } return true; diff --git a/net/bridge/netfilter/ebt_stp.c b/net/bridge/netfilter/ebt_stp.c index 45f73d55422f..3140eb912d7e 100644 --- a/net/bridge/netfilter/ebt_stp.c +++ b/net/bridge/netfilter/ebt_stp.c @@ -49,66 +49,68 @@ static bool ebt_filter_config(const struct ebt_stp_info *info, c = &info->config; if ((info->bitmask & EBT_STP_FLAGS) && - FWINV(c->flags != stpc->flags, EBT_STP_FLAGS)) + NF_INVF(info, EBT_STP_FLAGS, c->flags != stpc->flags)) return false; if (info->bitmask & EBT_STP_ROOTPRIO) { v16 = NR16(stpc->root); - if (FWINV(v16 < c->root_priol || v16 > c->root_priou, - EBT_STP_ROOTPRIO)) + if (NF_INVF(info, EBT_STP_ROOTPRIO, + v16 < c->root_priol || v16 > c->root_priou)) return false; } if (info->bitmask & EBT_STP_ROOTADDR) { - if (FWINV(!ether_addr_equal_masked(&stpc->root[2], c->root_addr, - c->root_addrmsk), - EBT_STP_ROOTADDR)) + if (NF_INVF(info, EBT_STP_ROOTADDR, + !ether_addr_equal_masked(&stpc->root[2], + c->root_addr, + c->root_addrmsk))) return false; } if (info->bitmask & EBT_STP_ROOTCOST) { v32 = NR32(stpc->root_cost); - if (FWINV(v32 < c->root_costl || v32 > c->root_costu, - EBT_STP_ROOTCOST)) + if (NF_INVF(info, EBT_STP_ROOTCOST, + v32 < c->root_costl || v32 > c->root_costu)) return false; } if (info->bitmask & EBT_STP_SENDERPRIO) { v16 = NR16(stpc->sender); - if (FWINV(v16 < c->sender_priol || v16 > c->sender_priou, - EBT_STP_SENDERPRIO)) + if (NF_INVF(info, EBT_STP_SENDERPRIO, + v16 < c->sender_priol || v16 > c->sender_priou)) return false; } if (info->bitmask & EBT_STP_SENDERADDR) { - if (FWINV(!ether_addr_equal_masked(&stpc->sender[2], - c->sender_addr, - c->sender_addrmsk), - EBT_STP_SENDERADDR)) + if (NF_INVF(info, EBT_STP_SENDERADDR, + !ether_addr_equal_masked(&stpc->sender[2], + c->sender_addr, + c->sender_addrmsk))) return false; } if (info->bitmask & EBT_STP_PORT) { v16 = NR16(stpc->port); - if (FWINV(v16 < c->portl || v16 > c->portu, EBT_STP_PORT)) + if (NF_INVF(info, EBT_STP_PORT, + v16 < c->portl || v16 > c->portu)) return false; } if (info->bitmask & EBT_STP_MSGAGE) { v16 = NR16(stpc->msg_age); - if (FWINV(v16 < c->msg_agel || v16 > c->msg_ageu, - EBT_STP_MSGAGE)) + if (NF_INVF(info, EBT_STP_MSGAGE, + v16 < c->msg_agel || v16 > c->msg_ageu)) return false; } if (info->bitmask & EBT_STP_MAXAGE) { v16 = NR16(stpc->max_age); - if (FWINV(v16 < c->max_agel || v16 > c->max_ageu, - EBT_STP_MAXAGE)) + if (NF_INVF(info, EBT_STP_MAXAGE, + v16 < c->max_agel || v16 > c->max_ageu)) return false; } if (info->bitmask & EBT_STP_HELLOTIME) { v16 = NR16(stpc->hello_time); - if (FWINV(v16 < c->hello_timel || v16 > c->hello_timeu, - EBT_STP_HELLOTIME)) + if (NF_INVF(info, EBT_STP_HELLOTIME, + v16 < c->hello_timel || v16 > c->hello_timeu)) return false; } if (info->bitmask & EBT_STP_FWDD) { v16 = NR16(stpc->forward_delay); - if (FWINV(v16 < c->forward_delayl || v16 > c->forward_delayu, - EBT_STP_FWDD)) + if (NF_INVF(info, EBT_STP_FWDD, + v16 < c->forward_delayl || v16 > c->forward_delayu)) return false; } return true; @@ -130,8 +132,8 @@ ebt_stp_mt(const struct sk_buff *skb, struct xt_action_param *par) if (memcmp(sp, header, sizeof(header))) return false; - if (info->bitmask & EBT_STP_TYPE && - FWINV(info->type != sp->type, EBT_STP_TYPE)) + if ((info->bitmask & EBT_STP_TYPE) && + NF_INVF(info, EBT_STP_TYPE, info->type != sp->type)) return false; if (sp->type == BPDU_TYPE_CONFIG && diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index 5721a25be860..cceac5bb658f 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -121,7 +121,6 @@ ebt_dev_check(const char *entry, const struct net_device *device) return devname[i] != entry[i] && entry[i] != 1; } -#define FWINV2(bool, invflg) ((bool) ^ !!(e->invflags & invflg)) /* process standard matches */ static inline int ebt_basic_match(const struct ebt_entry *e, const struct sk_buff *skb, @@ -137,34 +136,36 @@ ebt_basic_match(const struct ebt_entry *e, const struct sk_buff *skb, ethproto = h->h_proto; if (e->bitmask & EBT_802_3) { - if (FWINV2(eth_proto_is_802_3(ethproto), EBT_IPROTO)) + if (NF_INVF(e, EBT_IPROTO, eth_proto_is_802_3(ethproto))) return 1; } else if (!(e->bitmask & EBT_NOPROTO) && - FWINV2(e->ethproto != ethproto, EBT_IPROTO)) + NF_INVF(e, EBT_IPROTO, e->ethproto != ethproto)) return 1; - if (FWINV2(ebt_dev_check(e->in, in), EBT_IIN)) + if (NF_INVF(e, EBT_IIN, ebt_dev_check(e->in, in))) return 1; - if (FWINV2(ebt_dev_check(e->out, out), EBT_IOUT)) + if (NF_INVF(e, EBT_IOUT, ebt_dev_check(e->out, out))) return 1; /* rcu_read_lock()ed by nf_hook_slow */ if (in && (p = br_port_get_rcu(in)) != NULL && - FWINV2(ebt_dev_check(e->logical_in, p->br->dev), EBT_ILOGICALIN)) + NF_INVF(e, EBT_ILOGICALIN, + ebt_dev_check(e->logical_in, p->br->dev))) return 1; if (out && (p = br_port_get_rcu(out)) != NULL && - FWINV2(ebt_dev_check(e->logical_out, p->br->dev), EBT_ILOGICALOUT)) + NF_INVF(e, EBT_ILOGICALOUT, + ebt_dev_check(e->logical_out, p->br->dev))) return 1; if (e->bitmask & EBT_SOURCEMAC) { - if (FWINV2(!ether_addr_equal_masked(h->h_source, - e->sourcemac, e->sourcemsk), - EBT_ISOURCE)) + if (NF_INVF(e, EBT_ISOURCE, + !ether_addr_equal_masked(h->h_source, e->sourcemac, + e->sourcemsk))) return 1; } if (e->bitmask & EBT_DESTMAC) { - if (FWINV2(!ether_addr_equal_masked(h->h_dest, - e->destmac, e->destmsk), - EBT_IDEST)) + if (NF_INVF(e, EBT_IDEST, + !ether_addr_equal_masked(h->h_dest, e->destmac, + e->destmsk))) return 1; } return 0; diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index 2033f929aa66..c8dd9e26b185 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -89,22 +89,20 @@ static inline int arp_packet_match(const struct arphdr *arphdr, __be32 src_ipaddr, tgt_ipaddr; long ret; -#define FWINV(bool, invflg) ((bool) ^ !!(arpinfo->invflags & (invflg))) - - if (FWINV((arphdr->ar_op & arpinfo->arpop_mask) != arpinfo->arpop, - ARPT_INV_ARPOP)) + if (NF_INVF(arpinfo, ARPT_INV_ARPOP, + (arphdr->ar_op & arpinfo->arpop_mask) != arpinfo->arpop)) return 0; - if (FWINV((arphdr->ar_hrd & arpinfo->arhrd_mask) != arpinfo->arhrd, - ARPT_INV_ARPHRD)) + if (NF_INVF(arpinfo, ARPT_INV_ARPHRD, + (arphdr->ar_hrd & arpinfo->arhrd_mask) != arpinfo->arhrd)) return 0; - if (FWINV((arphdr->ar_pro & arpinfo->arpro_mask) != arpinfo->arpro, - ARPT_INV_ARPPRO)) + if (NF_INVF(arpinfo, ARPT_INV_ARPPRO, + (arphdr->ar_pro & arpinfo->arpro_mask) != arpinfo->arpro)) return 0; - if (FWINV((arphdr->ar_hln & arpinfo->arhln_mask) != arpinfo->arhln, - ARPT_INV_ARPHLN)) + if (NF_INVF(arpinfo, ARPT_INV_ARPHLN, + (arphdr->ar_hln & arpinfo->arhln_mask) != arpinfo->arhln)) return 0; src_devaddr = arpptr; @@ -115,31 +113,32 @@ static inline int arp_packet_match(const struct arphdr *arphdr, arpptr += dev->addr_len; memcpy(&tgt_ipaddr, arpptr, sizeof(u32)); - if (FWINV(arp_devaddr_compare(&arpinfo->src_devaddr, src_devaddr, dev->addr_len), - ARPT_INV_SRCDEVADDR) || - FWINV(arp_devaddr_compare(&arpinfo->tgt_devaddr, tgt_devaddr, dev->addr_len), - ARPT_INV_TGTDEVADDR)) + if (NF_INVF(arpinfo, ARPT_INV_SRCDEVADDR, + arp_devaddr_compare(&arpinfo->src_devaddr, src_devaddr, + dev->addr_len)) || + NF_INVF(arpinfo, ARPT_INV_TGTDEVADDR, + arp_devaddr_compare(&arpinfo->tgt_devaddr, tgt_devaddr, + dev->addr_len))) return 0; - if (FWINV((src_ipaddr & arpinfo->smsk.s_addr) != arpinfo->src.s_addr, - ARPT_INV_SRCIP) || - FWINV(((tgt_ipaddr & arpinfo->tmsk.s_addr) != arpinfo->tgt.s_addr), - ARPT_INV_TGTIP)) + if (NF_INVF(arpinfo, ARPT_INV_SRCIP, + (src_ipaddr & arpinfo->smsk.s_addr) != arpinfo->src.s_addr) || + NF_INVF(arpinfo, ARPT_INV_TGTIP, + (tgt_ipaddr & arpinfo->tmsk.s_addr) != arpinfo->tgt.s_addr)) return 0; /* Look for ifname matches. */ ret = ifname_compare(indev, arpinfo->iniface, arpinfo->iniface_mask); - if (FWINV(ret != 0, ARPT_INV_VIA_IN)) + if (NF_INVF(arpinfo, ARPT_INV_VIA_IN, ret != 0)) return 0; ret = ifname_compare(outdev, arpinfo->outiface, arpinfo->outiface_mask); - if (FWINV(ret != 0, ARPT_INV_VIA_OUT)) + if (NF_INVF(arpinfo, ARPT_INV_VIA_OUT, ret != 0)) return 0; return 1; -#undef FWINV } static inline int arp_checkentry(const struct arpt_arp *arp) diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index 54906e0e8e0c..f0df66f54ce6 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -58,32 +58,31 @@ ip_packet_match(const struct iphdr *ip, { unsigned long ret; -#define FWINV(bool, invflg) ((bool) ^ !!(ipinfo->invflags & (invflg))) - - if (FWINV((ip->saddr&ipinfo->smsk.s_addr) != ipinfo->src.s_addr, - IPT_INV_SRCIP) || - FWINV((ip->daddr&ipinfo->dmsk.s_addr) != ipinfo->dst.s_addr, - IPT_INV_DSTIP)) + if (NF_INVF(ipinfo, IPT_INV_SRCIP, + (ip->saddr & ipinfo->smsk.s_addr) != ipinfo->src.s_addr) || + NF_INVF(ipinfo, IPT_INV_DSTIP, + (ip->daddr & ipinfo->dmsk.s_addr) != ipinfo->dst.s_addr)) return false; ret = ifname_compare_aligned(indev, ipinfo->iniface, ipinfo->iniface_mask); - if (FWINV(ret != 0, IPT_INV_VIA_IN)) + if (NF_INVF(ipinfo, IPT_INV_VIA_IN, ret != 0)) return false; ret = ifname_compare_aligned(outdev, ipinfo->outiface, ipinfo->outiface_mask); - if (FWINV(ret != 0, IPT_INV_VIA_OUT)) + if (NF_INVF(ipinfo, IPT_INV_VIA_OUT, ret != 0)) return false; /* Check specific protocol */ if (ipinfo->proto && - FWINV(ip->protocol != ipinfo->proto, IPT_INV_PROTO)) + NF_INVF(ipinfo, IPT_INV_PROTO, ip->protocol != ipinfo->proto)) return false; /* If we have a fragment rule but the packet is not a fragment * then we return zero */ - if (FWINV((ipinfo->flags&IPT_F_FRAG) && !isfrag, IPT_INV_FRAG)) + if (NF_INVF(ipinfo, IPT_INV_FRAG, + (ipinfo->flags & IPT_F_FRAG) && !isfrag)) return false; return true; @@ -122,7 +121,6 @@ static inline bool unconditional(const struct ipt_entry *e) return e->target_offset == sizeof(struct ipt_entry) && memcmp(&e->ip, &uncond, sizeof(uncond)) == 0; -#undef FWINV } /* for const-correctness */ diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 63e06c3dd319..61ed95054efa 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -73,22 +73,22 @@ ip6_packet_match(const struct sk_buff *skb, unsigned long ret; const struct ipv6hdr *ipv6 = ipv6_hdr(skb); -#define FWINV(bool, invflg) ((bool) ^ !!(ip6info->invflags & (invflg))) - - if (FWINV(ipv6_masked_addr_cmp(&ipv6->saddr, &ip6info->smsk, - &ip6info->src), IP6T_INV_SRCIP) || - FWINV(ipv6_masked_addr_cmp(&ipv6->daddr, &ip6info->dmsk, - &ip6info->dst), IP6T_INV_DSTIP)) + if (NF_INVF(ip6info, IP6T_INV_SRCIP, + ipv6_masked_addr_cmp(&ipv6->saddr, &ip6info->smsk, + &ip6info->src)) || + NF_INVF(ip6info, IP6T_INV_DSTIP, + ipv6_masked_addr_cmp(&ipv6->daddr, &ip6info->dmsk, + &ip6info->dst))) return false; ret = ifname_compare_aligned(indev, ip6info->iniface, ip6info->iniface_mask); - if (FWINV(ret != 0, IP6T_INV_VIA_IN)) + if (NF_INVF(ip6info, IP6T_INV_VIA_IN, ret != 0)) return false; ret = ifname_compare_aligned(outdev, ip6info->outiface, ip6info->outiface_mask); - if (FWINV(ret != 0, IP6T_INV_VIA_OUT)) + if (NF_INVF(ip6info, IP6T_INV_VIA_OUT, ret != 0)) return false; /* ... might want to do something with class and flowlabel here ... */ diff --git a/net/netfilter/xt_tcpudp.c b/net/netfilter/xt_tcpudp.c index c14d4645daa3..ade024c90f4f 100644 --- a/net/netfilter/xt_tcpudp.c +++ b/net/netfilter/xt_tcpudp.c @@ -83,8 +83,6 @@ static bool tcp_mt(const struct sk_buff *skb, struct xt_action_param *par) return false; } -#define FWINVTCP(bool, invflg) ((bool) ^ !!(tcpinfo->invflags & (invflg))) - th = skb_header_pointer(skb, par->thoff, sizeof(_tcph), &_tcph); if (th == NULL) { /* We've been asked to examine this packet, and we @@ -102,9 +100,8 @@ static bool tcp_mt(const struct sk_buff *skb, struct xt_action_param *par) ntohs(th->dest), !!(tcpinfo->invflags & XT_TCP_INV_DSTPT))) return false; - if (!FWINVTCP((((unsigned char *)th)[13] & tcpinfo->flg_mask) - == tcpinfo->flg_cmp, - XT_TCP_INV_FLAGS)) + if (!NF_INVF(tcpinfo, XT_TCP_INV_FLAGS, + (((unsigned char *)th)[13] & tcpinfo->flg_mask) == tcpinfo->flg_cmp)) return false; if (tcpinfo->option) { if (th->doff * 4 < sizeof(_tcph)) { -- cgit v1.2.3 From 8b10cab64c134ffbffac96edd1899d303d3afcac Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sat, 2 Jul 2016 06:43:14 -0400 Subject: net: simplify and make pkt_type_ok() available for other users Suggested-by: Daniel Borkmann Signed-off-by: Jamal Hadi Salim Acked-by: Daniel Borkmann Signed-off-by: David S. Miller --- include/linux/skbuff.h | 10 ++++++++++ net/netfilter/nft_meta.c | 9 +-------- 2 files changed, 11 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index dc0fca747c5e..638b0e004310 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -37,6 +37,7 @@ #include #include #include +#include #include /* The interface for checksum offload between the stack and networking drivers @@ -881,6 +882,15 @@ static inline struct rtable *skb_rtable(const struct sk_buff *skb) return (struct rtable *)skb_dst(skb); } +/* For mangling skb->pkt_type from user space side from applications + * such as nft, tc, etc, we only allow a conservative subset of + * possible pkt_types to be set. +*/ +static inline bool skb_pkt_type_ok(u32 ptype) +{ + return ptype <= PACKET_OTHERHOST; +} + void kfree_skb(struct sk_buff *skb); void kfree_skb_list(struct sk_buff *segs); void skb_tx_error(struct sk_buff *skb); diff --git a/net/netfilter/nft_meta.c b/net/netfilter/nft_meta.c index 16c50b0dd426..03e5e33b5c39 100644 --- a/net/netfilter/nft_meta.c +++ b/net/netfilter/nft_meta.c @@ -199,13 +199,6 @@ err: } EXPORT_SYMBOL_GPL(nft_meta_get_eval); -/* don't change or set _LOOPBACK, _USER, etc. */ -static bool pkt_type_ok(u32 p) -{ - return p == PACKET_HOST || p == PACKET_BROADCAST || - p == PACKET_MULTICAST || p == PACKET_OTHERHOST; -} - void nft_meta_set_eval(const struct nft_expr *expr, struct nft_regs *regs, const struct nft_pktinfo *pkt) @@ -223,7 +216,7 @@ void nft_meta_set_eval(const struct nft_expr *expr, break; case NFT_META_PKTTYPE: if (skb->pkt_type != value && - pkt_type_ok(value) && pkt_type_ok(skb->pkt_type)) + skb_pkt_type_ok(value) && skb_pkt_type_ok(skb->pkt_type)) skb->pkt_type = value; break; case NFT_META_NFTRACE: -- cgit v1.2.3 From 7ce856aaaf13a5dc969ac5f998e5daaf1abe4cd2 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Mon, 4 Jul 2016 08:23:12 +0200 Subject: mlxsw: spectrum: Add couple of lower device helper functions Add functions that iterate over lower devices and find port device. As a dependency add netdev_for_each_all_lower_dev and netdev_for_each_all_lower_dev_rcu macro with netdev_all_lower_get_next and netdev_all_lower_get_next_rcu shelpers. Also, add functions to return mlxsw struct according to lower device found and mlxsw_port struct with a reference to lower device. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 65 ++++++++++++++++++++++++-- drivers/net/ethernet/mellanox/mlxsw/spectrum.h | 3 ++ include/linux/netdevice.h | 18 +++++++ net/core/dev.c | 46 ++++++++++++++++++ 4 files changed, 127 insertions(+), 5 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index f0799898817d..f54fd6ab6ec9 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -2567,6 +2567,66 @@ static struct mlxsw_driver mlxsw_sp_driver = { .profile = &mlxsw_sp_config_profile, }; +static bool mlxsw_sp_port_dev_check(const struct net_device *dev) +{ + return dev->netdev_ops == &mlxsw_sp_port_netdev_ops; +} + +static struct mlxsw_sp_port *mlxsw_sp_port_dev_lower_find(struct net_device *dev) +{ + struct net_device *lower_dev; + struct list_head *iter; + + if (mlxsw_sp_port_dev_check(dev)) + return netdev_priv(dev); + + netdev_for_each_all_lower_dev(dev, lower_dev, iter) { + if (mlxsw_sp_port_dev_check(lower_dev)) + return netdev_priv(lower_dev); + } + return NULL; +} + +static struct mlxsw_sp *mlxsw_sp_lower_get(struct net_device *dev) +{ + struct mlxsw_sp_port *mlxsw_sp_port; + + mlxsw_sp_port = mlxsw_sp_port_dev_lower_find(dev); + return mlxsw_sp_port ? mlxsw_sp_port->mlxsw_sp : NULL; +} + +static struct mlxsw_sp_port *mlxsw_sp_port_dev_lower_find_rcu(struct net_device *dev) +{ + struct net_device *lower_dev; + struct list_head *iter; + + if (mlxsw_sp_port_dev_check(dev)) + return netdev_priv(dev); + + netdev_for_each_all_lower_dev_rcu(dev, lower_dev, iter) { + if (mlxsw_sp_port_dev_check(lower_dev)) + return netdev_priv(lower_dev); + } + return NULL; +} + +struct mlxsw_sp_port *mlxsw_sp_port_lower_dev_hold(struct net_device *dev) +{ + struct mlxsw_sp_port *mlxsw_sp_port; + + rcu_read_lock(); + mlxsw_sp_port = mlxsw_sp_port_dev_lower_find_rcu(dev); + if (mlxsw_sp_port) + dev_hold(mlxsw_sp_port->dev); + rcu_read_unlock(); + return mlxsw_sp_port; +} + +void mlxsw_sp_port_dev_put(struct mlxsw_sp_port *mlxsw_sp_port) +{ + dev_put(mlxsw_sp_port->dev); +} + static bool mlxsw_sp_lag_port_fid_member(struct mlxsw_sp_port *lag_port, u16 fid) { @@ -2647,11 +2707,6 @@ int mlxsw_sp_port_fdb_flush(struct mlxsw_sp_port *mlxsw_sp_port, u16 fid) return mlxsw_sp_port_fdb_flush_by_port_fid(mlxsw_sp_port, fid); } -static bool mlxsw_sp_port_dev_check(const struct net_device *dev) -{ - return dev->netdev_ops == &mlxsw_sp_port_netdev_ops; -} - static bool mlxsw_sp_master_bridge_check(struct mlxsw_sp *mlxsw_sp, struct net_device *br_dev) { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index 877a8793fd97..fefff255b30d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -292,6 +292,9 @@ struct mlxsw_sp_port { struct list_head vports_list; }; +struct mlxsw_sp_port *mlxsw_sp_port_lower_dev_hold(struct net_device *dev); +void mlxsw_sp_port_dev_put(struct mlxsw_sp_port *mlxsw_sp_port); + static inline bool mlxsw_sp_port_is_pause_en(const struct mlxsw_sp_port *mlxsw_sp_port) { diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 7dc2ec74122a..0c6ee2c5099f 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3804,12 +3804,30 @@ void *netdev_lower_get_next_private_rcu(struct net_device *dev, void *netdev_lower_get_next(struct net_device *dev, struct list_head **iter); + #define netdev_for_each_lower_dev(dev, ldev, iter) \ for (iter = (dev)->adj_list.lower.next, \ ldev = netdev_lower_get_next(dev, &(iter)); \ ldev; \ ldev = netdev_lower_get_next(dev, &(iter))) +struct net_device *netdev_all_lower_get_next(struct net_device *dev, + struct list_head **iter); +struct net_device *netdev_all_lower_get_next_rcu(struct net_device *dev, + struct list_head **iter); + +#define netdev_for_each_all_lower_dev(dev, ldev, iter) \ + for (iter = (dev)->all_adj_list.lower.next, \ + ldev = netdev_all_lower_get_next(dev, &(iter)); \ + ldev; \ + ldev = netdev_all_lower_get_next(dev, &(iter))) + +#define netdev_for_each_all_lower_dev_rcu(dev, ldev, iter) \ + for (iter = (dev)->all_adj_list.lower.next, \ + ldev = netdev_all_lower_get_next_rcu(dev, &(iter)); \ + ldev; \ + ldev = netdev_all_lower_get_next_rcu(dev, &(iter))) + void *netdev_adjacent_get_private(struct list_head *adj_list); void *netdev_lower_get_first_private_rcu(struct net_device *dev); struct net_device *netdev_master_upper_dev_get(struct net_device *dev); diff --git a/net/core/dev.c b/net/core/dev.c index aba10d2a8bc3..a4f3b0a9aeaf 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -5444,6 +5444,52 @@ void *netdev_lower_get_next(struct net_device *dev, struct list_head **iter) } EXPORT_SYMBOL(netdev_lower_get_next); +/** + * netdev_all_lower_get_next - Get the next device from all lower neighbour list + * @dev: device + * @iter: list_head ** of the current position + * + * Gets the next netdev_adjacent from the dev's all lower neighbour + * list, starting from iter position. The caller must hold RTNL lock or + * its own locking that guarantees that the neighbour all lower + * list will remain unchanged. + */ +struct net_device *netdev_all_lower_get_next(struct net_device *dev, struct list_head **iter) +{ + struct netdev_adjacent *lower; + + lower = list_entry(*iter, struct netdev_adjacent, list); + + if (&lower->list == &dev->all_adj_list.lower) + return NULL; + + *iter = lower->list.next; + + return lower->dev; +} +EXPORT_SYMBOL(netdev_all_lower_get_next); + +/** + * netdev_all_lower_get_next_rcu - Get the next device from all + * lower neighbour list, RCU variant + * @dev: device + * @iter: list_head ** of the current position + * + * Gets the next netdev_adjacent from the dev's all lower neighbour + * list, starting from iter position. The caller must hold RCU read lock. + */ +struct net_device *netdev_all_lower_get_next_rcu(struct net_device *dev, + struct list_head **iter) +{ + struct netdev_adjacent *lower; + + lower = list_first_or_null_rcu(&dev->all_adj_list.lower, + struct netdev_adjacent, list); + + return lower ? lower->dev : NULL; +} +EXPORT_SYMBOL(netdev_all_lower_get_next_rcu); + /** * netdev_lower_get_first_private_rcu - Get the first ->private from the * lower neighbour list, RCU -- cgit v1.2.3 From c5bb17302e734967822be559cf661704b707b4ed Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 4 Jul 2016 17:23:05 +0300 Subject: net/mlx5: Refactor mlx5_add_flow_rule Reduce the set of arguments passed to mlx5_add_flow_rule by introducing flow_spec structure. Signed-off-by: Maor Gottlieb Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/infiniband/hw/mlx5/main.c | 21 ++--- drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c | 68 ++++++-------- drivers/net/ethernet/mellanox/mlx5/core/en_fs.c | 96 ++++++++------------ drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 31 +++---- drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 100 ++++++++------------- .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 55 ++++++------ drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 22 ++--- include/linux/mlx5/fs.h | 10 ++- 8 files changed, 171 insertions(+), 232 deletions(-) (limited to 'include/linux') diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index b48ad85315dc..dad63f038bb8 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -1528,21 +1528,18 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, { struct mlx5_flow_table *ft = ft_prio->flow_table; struct mlx5_ib_flow_handler *handler; + struct mlx5_flow_spec *spec; void *ib_flow = flow_attr + 1; - u8 match_criteria_enable = 0; unsigned int spec_index; - u32 *match_c; - u32 *match_v; u32 action; int err = 0; if (!is_valid_attr(flow_attr)) return ERR_PTR(-EINVAL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); + spec = mlx5_vzalloc(sizeof(*spec)); handler = kzalloc(sizeof(*handler), GFP_KERNEL); - if (!handler || !match_c || !match_v) { + if (!handler || !spec) { err = -ENOMEM; goto free; } @@ -1550,7 +1547,8 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, INIT_LIST_HEAD(&handler->list); for (spec_index = 0; spec_index < flow_attr->num_of_specs; spec_index++) { - err = parse_flow_attr(match_c, match_v, ib_flow); + err = parse_flow_attr(spec->match_criteria, + spec->match_value, ib_flow); if (err < 0) goto free; @@ -1558,11 +1556,11 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, } /* Outer header support only */ - match_criteria_enable = (!outer_header_zero(match_c)) << 0; + spec->match_criteria_enable = (!outer_header_zero(spec->match_criteria)) + << 0; action = dst ? MLX5_FLOW_CONTEXT_ACTION_FWD_DEST : MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO; - handler->rule = mlx5_add_flow_rule(ft, match_criteria_enable, - match_c, match_v, + handler->rule = mlx5_add_flow_rule(ft, spec, action, MLX5_FS_DEFAULT_FLOW_TAG, dst); @@ -1578,8 +1576,7 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, free: if (err) kfree(handler); - kfree(match_c); - kfree(match_v); + kvfree(spec); return err ? ERR_PTR(err) : handler; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c index 10f18d46b8ab..a8cb38789774 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_arfs.c @@ -175,15 +175,12 @@ static int arfs_add_default_rule(struct mlx5e_priv *priv, { struct arfs_table *arfs_t = &priv->fs.arfs.arfs_tables[type]; struct mlx5_flow_destination dest; - u8 match_criteria_enable = 0; struct mlx5e_tir *tir = priv->indir_tir; - u32 *match_criteria; - u32 *match_value; + struct mlx5_flow_spec *spec; int err = 0; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); err = -ENOMEM; goto out; @@ -208,8 +205,7 @@ static int arfs_add_default_rule(struct mlx5e_priv *priv, goto out; } - arfs_t->default_rule = mlx5_add_flow_rule(arfs_t->ft.t, match_criteria_enable, - match_criteria, match_value, + arfs_t->default_rule = mlx5_add_flow_rule(arfs_t->ft.t, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, &dest); @@ -220,8 +216,7 @@ static int arfs_add_default_rule(struct mlx5e_priv *priv, __func__, type); } out: - kvfree(match_criteria); - kvfree(match_value); + kvfree(spec); return err; } @@ -475,23 +470,20 @@ static struct mlx5_flow_rule *arfs_add_rule(struct mlx5e_priv *priv, struct mlx5_flow_rule *rule = NULL; struct mlx5_flow_destination dest; struct arfs_table *arfs_table; - u8 match_criteria_enable = 0; + struct mlx5_flow_spec *spec; struct mlx5_flow_table *ft; - u32 *match_criteria; - u32 *match_value; int err = 0; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); err = -ENOMEM; goto out; } - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ethertype); - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.ethertype, ntohs(tuple->etype)); arfs_table = arfs_get_table(arfs, tuple->ip_proto, tuple->etype); if (!arfs_table) { @@ -501,59 +493,58 @@ static struct mlx5_flow_rule *arfs_add_rule(struct mlx5e_priv *priv, ft = arfs_table->ft.t; if (tuple->ip_proto == IPPROTO_TCP) { - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.tcp_dport); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.tcp_sport); - MLX5_SET(fte_match_param, match_value, outer_headers.tcp_dport, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.tcp_dport, ntohs(tuple->dst_port)); - MLX5_SET(fte_match_param, match_value, outer_headers.tcp_sport, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.tcp_sport, ntohs(tuple->src_port)); } else { - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.udp_dport); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.udp_sport); - MLX5_SET(fte_match_param, match_value, outer_headers.udp_dport, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.udp_dport, ntohs(tuple->dst_port)); - MLX5_SET(fte_match_param, match_value, outer_headers.udp_sport, + MLX5_SET(fte_match_param, spec->match_value, outer_headers.udp_sport, ntohs(tuple->src_port)); } if (tuple->etype == htons(ETH_P_IP)) { - memcpy(MLX5_ADDR_OF(fte_match_param, match_value, + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4), &tuple->src_ipv4, 4); - memcpy(MLX5_ADDR_OF(fte_match_param, match_value, + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4), &tuple->dst_ipv4, 4); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.src_ipv4_src_ipv6.ipv4_layout.ipv4); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4); } else { - memcpy(MLX5_ADDR_OF(fte_match_param, match_value, + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), &tuple->src_ipv6, 16); - memcpy(MLX5_ADDR_OF(fte_match_param, match_value, + memcpy(MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), &tuple->dst_ipv6, 16); - memset(MLX5_ADDR_OF(fte_match_param, match_criteria, + memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), 0xff, 16); - memset(MLX5_ADDR_OF(fte_match_param, match_criteria, + memset(MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), 0xff, 16); } dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; dest.tir_num = priv->direct_tir[arfs_rule->rxq].tirn; - rule = mlx5_add_flow_rule(ft, match_criteria_enable, match_criteria, - match_value, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + rule = mlx5_add_flow_rule(ft, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, &dest); if (IS_ERR(rule)) { @@ -563,8 +554,7 @@ static struct mlx5_flow_rule *arfs_add_rule(struct mlx5e_priv *priv, } out: - kvfree(match_criteria); - kvfree(match_value); + kvfree(spec); return err ? ERR_PTR(err) : rule; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c index 606e69b4babc..2e1e86316fe7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c @@ -156,19 +156,18 @@ enum mlx5e_vlan_rule_type { static int __mlx5e_add_vlan_rule(struct mlx5e_priv *priv, enum mlx5e_vlan_rule_type rule_type, - u16 vid, u32 *mc, u32 *mv) + u16 vid, struct mlx5_flow_spec *spec) { struct mlx5_flow_table *ft = priv->fs.vlan.ft.t; struct mlx5_flow_destination dest; - u8 match_criteria_enable = 0; struct mlx5_flow_rule **rule_p; int err = 0; dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; dest.ft = priv->fs.l2.ft.t; - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.vlan_tag); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.vlan_tag); switch (rule_type) { case MLX5E_VLAN_RULE_TYPE_UNTAGGED: @@ -176,17 +175,19 @@ static int __mlx5e_add_vlan_rule(struct mlx5e_priv *priv, break; case MLX5E_VLAN_RULE_TYPE_ANY_VID: rule_p = &priv->fs.vlan.any_vlan_rule; - MLX5_SET(fte_match_param, mv, outer_headers.vlan_tag, 1); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.vlan_tag, 1); break; default: /* MLX5E_VLAN_RULE_TYPE_MATCH_VID */ rule_p = &priv->fs.vlan.active_vlans_rule[vid]; - MLX5_SET(fte_match_param, mv, outer_headers.vlan_tag, 1); - MLX5_SET_TO_ONES(fte_match_param, mc, outer_headers.first_vid); - MLX5_SET(fte_match_param, mv, outer_headers.first_vid, vid); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.vlan_tag, 1); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, + outer_headers.first_vid); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.first_vid, + vid); break; } - *rule_p = mlx5_add_flow_rule(ft, match_criteria_enable, mc, mv, + *rule_p = mlx5_add_flow_rule(ft, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, &dest); @@ -203,27 +204,21 @@ static int __mlx5e_add_vlan_rule(struct mlx5e_priv *priv, static int mlx5e_add_vlan_rule(struct mlx5e_priv *priv, enum mlx5e_vlan_rule_type rule_type, u16 vid) { - u32 *match_criteria; - u32 *match_value; + struct mlx5_flow_spec *spec; int err = 0; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); - err = -ENOMEM; - goto add_vlan_rule_out; + return -ENOMEM; } if (rule_type == MLX5E_VLAN_RULE_TYPE_MATCH_VID) mlx5e_vport_context_update_vlans(priv); - err = __mlx5e_add_vlan_rule(priv, rule_type, vid, match_criteria, - match_value); + err = __mlx5e_add_vlan_rule(priv, rule_type, vid, spec); -add_vlan_rule_out: - kvfree(match_criteria); - kvfree(match_value); + kvfree(spec); return err; } @@ -598,32 +593,27 @@ static struct mlx5_flow_rule *mlx5e_generate_ttc_rule(struct mlx5e_priv *priv, u8 proto) { struct mlx5_flow_rule *rule; - u8 match_criteria_enable = 0; - u32 *match_criteria; - u32 *match_value; + struct mlx5_flow_spec *spec; int err = 0; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); - err = -ENOMEM; - goto out; + return ERR_PTR(-ENOMEM); } if (proto) { - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.ip_protocol); - MLX5_SET(fte_match_param, match_value, outer_headers.ip_protocol, proto); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ip_protocol); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.ip_protocol, proto); } if (etype) { - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.ethertype); - MLX5_SET(fte_match_param, match_value, outer_headers.ethertype, etype); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ethertype); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.ethertype, etype); } - rule = mlx5_add_flow_rule(ft, match_criteria_enable, - match_criteria, match_value, + rule = mlx5_add_flow_rule(ft, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, dest); @@ -631,9 +621,8 @@ static struct mlx5_flow_rule *mlx5e_generate_ttc_rule(struct mlx5e_priv *priv, err = PTR_ERR(rule); netdev_err(priv->netdev, "%s: add rule failed\n", __func__); } -out: - kvfree(match_criteria); - kvfree(match_value); + + kvfree(spec); return err ? ERR_PTR(err) : rule; } @@ -792,24 +781,20 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_priv *priv, { struct mlx5_flow_table *ft = priv->fs.l2.ft.t; struct mlx5_flow_destination dest; - u8 match_criteria_enable = 0; - u32 *match_criteria; - u32 *match_value; + struct mlx5_flow_spec *spec; int err = 0; u8 *mc_dmac; u8 *mv_dmac; - match_value = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - match_criteria = mlx5_vzalloc(MLX5_ST_SZ_BYTES(fte_match_param)); - if (!match_value || !match_criteria) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { netdev_err(priv->netdev, "%s: alloc failed\n", __func__); - err = -ENOMEM; - goto add_l2_rule_out; + return -ENOMEM; } - mc_dmac = MLX5_ADDR_OF(fte_match_param, match_criteria, + mc_dmac = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers.dmac_47_16); - mv_dmac = MLX5_ADDR_OF(fte_match_param, match_value, + mv_dmac = MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.dmac_47_16); dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; @@ -817,13 +802,13 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_priv *priv, switch (type) { case MLX5E_FULLMATCH: - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; eth_broadcast_addr(mc_dmac); ether_addr_copy(mv_dmac, ai->addr); break; case MLX5E_ALLMULTI: - match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; mc_dmac[0] = 0x01; mv_dmac[0] = 0x01; break; @@ -832,8 +817,7 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_priv *priv, break; } - ai->rule = mlx5_add_flow_rule(ft, match_criteria_enable, match_criteria, - match_value, + ai->rule = mlx5_add_flow_rule(ft, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, MLX5_FS_DEFAULT_FLOW_TAG, &dest); if (IS_ERR(ai->rule)) { @@ -843,9 +827,7 @@ static int mlx5e_add_l2_flow_rule(struct mlx5e_priv *priv, ai->rule = NULL; } -add_l2_rule_out: - kvfree(match_criteria); - kvfree(match_value); + kvfree(spec); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 704c3d30493e..3261e8b1286e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -50,7 +50,7 @@ struct mlx5e_tc_flow { #define MLX5E_TC_TABLE_NUM_GROUPS 4 static struct mlx5_flow_rule *mlx5e_tc_add_flow(struct mlx5e_priv *priv, - u32 *match_c, u32 *match_v, + struct mlx5_flow_spec *spec, u32 action, u32 flow_tag) { struct mlx5_core_dev *dev = priv->mdev; @@ -88,8 +88,8 @@ static struct mlx5_flow_rule *mlx5e_tc_add_flow(struct mlx5e_priv *priv, table_created = true; } - rule = mlx5_add_flow_rule(priv->fs.tc.t, MLX5_MATCH_OUTER_HEADERS, - match_c, match_v, + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + rule = mlx5_add_flow_rule(priv->fs.tc.t, spec, action, flow_tag, &dest); @@ -126,12 +126,13 @@ static void mlx5e_tc_del_flow(struct mlx5e_priv *priv, } } -static int parse_cls_flower(struct mlx5e_priv *priv, - u32 *match_c, u32 *match_v, +static int parse_cls_flower(struct mlx5e_priv *priv, struct mlx5_flow_spec *spec, struct tc_cls_flower_offload *f) { - void *headers_c = MLX5_ADDR_OF(fte_match_param, match_c, outer_headers); - void *headers_v = MLX5_ADDR_OF(fte_match_param, match_v, outer_headers); + void *headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + outer_headers); + void *headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, + outer_headers); u16 addr_type = 0; u8 ip_proto = 0; @@ -342,12 +343,11 @@ int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol, struct tc_cls_flower_offload *f) { struct mlx5e_tc_table *tc = &priv->fs.tc; - u32 *match_c; - u32 *match_v; int err = 0; u32 flow_tag; u32 action; struct mlx5e_tc_flow *flow; + struct mlx5_flow_spec *spec; struct mlx5_flow_rule *old = NULL; flow = rhashtable_lookup_fast(&tc->ht, &f->cookie, @@ -357,16 +357,15 @@ int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol, else flow = kzalloc(sizeof(*flow), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_c || !match_v || !flow) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec || !flow) { err = -ENOMEM; goto err_free; } flow->cookie = f->cookie; - err = parse_cls_flower(priv, match_c, match_v, f); + err = parse_cls_flower(priv, spec, f); if (err < 0) goto err_free; @@ -379,8 +378,7 @@ int mlx5e_configure_flower(struct mlx5e_priv *priv, __be16 protocol, if (err) goto err_free; - flow->rule = mlx5e_tc_add_flow(priv, match_c, match_v, action, - flow_tag); + flow->rule = mlx5e_tc_add_flow(priv, spec, action, flow_tag); if (IS_ERR(flow->rule)) { err = PTR_ERR(flow->rule); goto err_hash_del; @@ -398,8 +396,7 @@ err_free: if (!old) kfree(flow); out: - kfree(match_c); - kfree(match_v); + kvfree(spec); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index f0a973557f7b..f6d667797ee1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -329,25 +329,23 @@ __esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule, MLX5_MATCH_OUTER_HEADERS); struct mlx5_flow_rule *flow_rule = NULL; struct mlx5_flow_destination dest; + struct mlx5_flow_spec *spec; void *mv_misc = NULL; void *mc_misc = NULL; u8 *dmac_v = NULL; u8 *dmac_c = NULL; - u32 *match_v; - u32 *match_c; if (rx_rule) match_header |= MLX5_MATCH_MISC_PARAMETERS; - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { pr_warn("FDB: Failed to alloc match parameters\n"); - goto out; + return NULL; } - - dmac_v = MLX5_ADDR_OF(fte_match_param, match_v, + dmac_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, outer_headers.dmac_47_16); - dmac_c = MLX5_ADDR_OF(fte_match_param, match_c, + dmac_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, outer_headers.dmac_47_16); if (match_header & MLX5_MATCH_OUTER_HEADERS) { @@ -356,8 +354,10 @@ __esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule, } if (match_header & MLX5_MATCH_MISC_PARAMETERS) { - mv_misc = MLX5_ADDR_OF(fte_match_param, match_v, misc_parameters); - mc_misc = MLX5_ADDR_OF(fte_match_param, match_c, misc_parameters); + mv_misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, + misc_parameters); + mc_misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + misc_parameters); MLX5_SET(fte_match_set_misc, mv_misc, source_port, UPLINK_VPORT); MLX5_SET_TO_ONES(fte_match_set_misc, mc_misc, source_port); } @@ -368,11 +368,9 @@ __esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule, esw_debug(esw->dev, "\tFDB add rule dmac_v(%pM) dmac_c(%pM) -> vport(%d)\n", dmac_v, dmac_c, vport); + spec->match_criteria_enable = match_header; flow_rule = - mlx5_add_flow_rule(esw->fdb_table.fdb, - match_header, - match_c, - match_v, + mlx5_add_flow_rule(esw->fdb_table.fdb, spec, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, 0, &dest); if (IS_ERR(flow_rule)) { @@ -381,9 +379,8 @@ __esw_fdb_set_vport_rule(struct mlx5_eswitch *esw, u32 vport, bool rx_rule, dmac_v, dmac_c, vport, PTR_ERR(flow_rule)); flow_rule = NULL; } -out: - kfree(match_v); - kfree(match_c); + + kvfree(spec); return flow_rule; } @@ -1293,9 +1290,8 @@ static void esw_vport_disable_ingress_acl(struct mlx5_eswitch *esw, static int esw_vport_ingress_config(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { + struct mlx5_flow_spec *spec; u8 smac[ETH_ALEN]; - u32 *match_v; - u32 *match_c; int err = 0; u8 *smac_v; @@ -1329,9 +1325,8 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, "vport[%d] configure ingress rules, vlan(%d) qos(%d)\n", vport->vport, vport->vlan, vport->qos); - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { err = -ENOMEM; esw_warn(esw->dev, "vport[%d] configure ingress rules failed, err(%d)\n", vport->vport, err); @@ -1339,22 +1334,20 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, } if (vport->vlan || vport->qos) - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.vlan_tag); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.vlan_tag); if (vport->spoofchk) { - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.smac_47_16); - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.smac_15_0); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.smac_47_16); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.smac_15_0); smac_v = MLX5_ADDR_OF(fte_match_param, - match_v, + spec->match_value, outer_headers.smac_47_16); ether_addr_copy(smac_v, smac); } + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; vport->ingress.allow_rule = - mlx5_add_flow_rule(vport->ingress.acl, - MLX5_MATCH_OUTER_HEADERS, - match_c, - match_v, + mlx5_add_flow_rule(vport->ingress.acl, spec, MLX5_FLOW_CONTEXT_ACTION_ALLOW, 0, NULL); if (IS_ERR(vport->ingress.allow_rule)) { @@ -1365,13 +1358,9 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, goto out; } - memset(match_c, 0, MLX5_ST_SZ_BYTES(fte_match_param)); - memset(match_v, 0, MLX5_ST_SZ_BYTES(fte_match_param)); + memset(spec, 0, sizeof(*spec)); vport->ingress.drop_rule = - mlx5_add_flow_rule(vport->ingress.acl, - 0, - match_c, - match_v, + mlx5_add_flow_rule(vport->ingress.acl, spec, MLX5_FLOW_CONTEXT_ACTION_DROP, 0, NULL); if (IS_ERR(vport->ingress.drop_rule)) { @@ -1385,17 +1374,14 @@ static int esw_vport_ingress_config(struct mlx5_eswitch *esw, out: if (err) esw_vport_cleanup_ingress_rules(esw, vport); - - kfree(match_v); - kfree(match_c); + kvfree(spec); return err; } static int esw_vport_egress_config(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { - u32 *match_v; - u32 *match_c; + struct mlx5_flow_spec *spec; int err = 0; esw_vport_cleanup_egress_rules(esw, vport); @@ -1411,9 +1397,8 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, "vport[%d] configure egress rules, vlan(%d) qos(%d)\n", vport->vport, vport->vlan, vport->qos); - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { err = -ENOMEM; esw_warn(esw->dev, "vport[%d] configure egress rules failed, err(%d)\n", vport->vport, err); @@ -1421,16 +1406,14 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, } /* Allowed vlan rule */ - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.vlan_tag); - MLX5_SET_TO_ONES(fte_match_param, match_v, outer_headers.vlan_tag); - MLX5_SET_TO_ONES(fte_match_param, match_c, outer_headers.first_vid); - MLX5_SET(fte_match_param, match_v, outer_headers.first_vid, vport->vlan); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.vlan_tag); + MLX5_SET_TO_ONES(fte_match_param, spec->match_value, outer_headers.vlan_tag); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.first_vid); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.first_vid, vport->vlan); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; vport->egress.allowed_vlan = - mlx5_add_flow_rule(vport->egress.acl, - MLX5_MATCH_OUTER_HEADERS, - match_c, - match_v, + mlx5_add_flow_rule(vport->egress.acl, spec, MLX5_FLOW_CONTEXT_ACTION_ALLOW, 0, NULL); if (IS_ERR(vport->egress.allowed_vlan)) { @@ -1442,13 +1425,9 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, } /* Drop others rule (star rule) */ - memset(match_c, 0, MLX5_ST_SZ_BYTES(fte_match_param)); - memset(match_v, 0, MLX5_ST_SZ_BYTES(fte_match_param)); + memset(spec, 0, sizeof(*spec)); vport->egress.drop_rule = - mlx5_add_flow_rule(vport->egress.acl, - 0, - match_c, - match_v, + mlx5_add_flow_rule(vport->egress.acl, spec, MLX5_FLOW_CONTEXT_ACTION_DROP, 0, NULL); if (IS_ERR(vport->egress.drop_rule)) { @@ -1458,8 +1437,7 @@ static int esw_vport_egress_config(struct mlx5_eswitch *esw, vport->egress.drop_rule = NULL; } out: - kfree(match_v); - kfree(match_c); + kvfree(spec); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index ed8ad988f07a..1842dfb4636b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -43,37 +43,35 @@ mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, u32 sqn { struct mlx5_flow_destination dest; struct mlx5_flow_rule *flow_rule; - int match_header = MLX5_MATCH_MISC_PARAMETERS; - u32 *match_v, *match_c; + struct mlx5_flow_spec *spec; void *misc; - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { esw_warn(esw->dev, "FDB: Failed to alloc match parameters\n"); flow_rule = ERR_PTR(-ENOMEM); goto out; } - misc = MLX5_ADDR_OF(fte_match_param, match_v, misc_parameters); + misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters); MLX5_SET(fte_match_set_misc, misc, source_sqn, sqn); MLX5_SET(fte_match_set_misc, misc, source_port, 0x0); /* source vport is 0 */ - misc = MLX5_ADDR_OF(fte_match_param, match_c, misc_parameters); + misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters); MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_sqn); MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port); + spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS; dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; dest.vport_num = vport; - flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, match_header, match_c, - match_v, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, spec, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, 0, &dest); if (IS_ERR(flow_rule)) esw_warn(esw->dev, "FDB: Failed to add send to vport rule err %ld\n", PTR_ERR(flow_rule)); out: - kfree(match_v); - kfree(match_c); + kvfree(spec); return flow_rule; } @@ -138,12 +136,11 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) { struct mlx5_flow_destination dest; struct mlx5_flow_rule *flow_rule = NULL; - u32 *match_v, *match_c; + struct mlx5_flow_spec *spec; int err = 0; - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { esw_warn(esw->dev, "FDB: Failed to alloc match parameters\n"); err = -ENOMEM; goto out; @@ -152,8 +149,9 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; dest.vport_num = 0; - flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, 0, match_c, match_v, - MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, 0, &dest); + flow_rule = mlx5_add_flow_rule(esw->fdb_table.fdb, spec, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + 0, &dest); if (IS_ERR(flow_rule)) { err = PTR_ERR(flow_rule); esw_warn(esw->dev, "FDB: Failed to add miss flow rule err %d\n", err); @@ -162,8 +160,7 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) esw->fdb_table.offloads.miss_rule = flow_rule; out: - kfree(match_v); - kfree(match_c); + kvfree(spec); return err; } @@ -351,29 +348,28 @@ mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, int vport, u32 tirn) { struct mlx5_flow_destination dest; struct mlx5_flow_rule *flow_rule; - int match_header = MLX5_MATCH_MISC_PARAMETERS; - u32 *match_v, *match_c; + struct mlx5_flow_spec *spec; void *misc; - match_v = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - match_c = kzalloc(MLX5_ST_SZ_BYTES(fte_match_param), GFP_KERNEL); - if (!match_v || !match_c) { + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) { esw_warn(esw->dev, "Failed to alloc match parameters\n"); flow_rule = ERR_PTR(-ENOMEM); goto out; } - misc = MLX5_ADDR_OF(fte_match_param, match_v, misc_parameters); + misc = MLX5_ADDR_OF(fte_match_param, spec->match_value, misc_parameters); MLX5_SET(fte_match_set_misc, misc, source_port, vport); - misc = MLX5_ADDR_OF(fte_match_param, match_c, misc_parameters); + misc = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, misc_parameters); MLX5_SET_TO_ONES(fte_match_set_misc, misc, source_port); + spec->match_criteria_enable = MLX5_MATCH_MISC_PARAMETERS; dest.type = MLX5_FLOW_DESTINATION_TYPE_TIR; dest.tir_num = tirn; - flow_rule = mlx5_add_flow_rule(esw->offloads.ft_offloads, match_header, match_c, - match_v, MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, + flow_rule = mlx5_add_flow_rule(esw->offloads.ft_offloads, spec, + MLX5_FLOW_CONTEXT_ACTION_FWD_DEST, 0, &dest); if (IS_ERR(flow_rule)) { esw_warn(esw->dev, "fs offloads: Failed to add vport rx rule err %ld\n", PTR_ERR(flow_rule)); @@ -381,8 +377,7 @@ mlx5_eswitch_create_vport_rx_rule(struct mlx5_eswitch *esw, int vport, u32 tirn) } out: - kfree(match_v); - kfree(match_c); + kvfree(spec); return flow_rule; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index b0401104afb9..7fcdae1cf053 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -1160,9 +1160,7 @@ static bool dest_is_valid(struct mlx5_flow_destination *dest, static struct mlx5_flow_rule * _mlx5_add_flow_rule(struct mlx5_flow_table *ft, - u8 match_criteria_enable, - u32 *match_criteria, - u32 *match_value, + struct mlx5_flow_spec *spec, u32 action, u32 flow_tag, struct mlx5_flow_destination *dest) @@ -1176,22 +1174,23 @@ _mlx5_add_flow_rule(struct mlx5_flow_table *ft, nested_lock_ref_node(&ft->node, FS_MUTEX_GRANDPARENT); fs_for_each_fg(g, ft) if (compare_match_criteria(g->mask.match_criteria_enable, - match_criteria_enable, + spec->match_criteria_enable, g->mask.match_criteria, - match_criteria)) { - rule = add_rule_fg(g, match_value, + spec->match_criteria)) { + rule = add_rule_fg(g, spec->match_value, action, flow_tag, dest); if (!IS_ERR(rule) || PTR_ERR(rule) != -ENOSPC) goto unlock; } - g = create_autogroup(ft, match_criteria_enable, match_criteria); + g = create_autogroup(ft, spec->match_criteria_enable, + spec->match_criteria); if (IS_ERR(g)) { rule = (void *)g; goto unlock; } - rule = add_rule_fg(g, match_value, + rule = add_rule_fg(g, spec->match_value, action, flow_tag, dest); if (IS_ERR(rule)) { /* Remove assumes refcount > 0 and autogroup creates a group @@ -1215,9 +1214,7 @@ static bool fwd_next_prio_supported(struct mlx5_flow_table *ft) struct mlx5_flow_rule * mlx5_add_flow_rule(struct mlx5_flow_table *ft, - u8 match_criteria_enable, - u32 *match_criteria, - u32 *match_value, + struct mlx5_flow_spec *spec, u32 action, u32 flow_tag, struct mlx5_flow_destination *dest) @@ -1248,8 +1245,7 @@ mlx5_add_flow_rule(struct mlx5_flow_table *ft, } } - rule = _mlx5_add_flow_rule(ft, match_criteria_enable, match_criteria, - match_value, action, flow_tag, dest); + rule = _mlx5_add_flow_rule(ft, spec, action, flow_tag, dest); if (sw_action == MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO) { if (!IS_ERR_OR_NULL(rule) && diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index 6ad111938709..d22fe7e5a39a 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -68,6 +68,12 @@ struct mlx5_flow_group; struct mlx5_flow_rule; struct mlx5_flow_namespace; +struct mlx5_flow_spec { + u8 match_criteria_enable; + u32 match_criteria[MLX5_ST_SZ_DW(fte_match_param)]; + u32 match_value[MLX5_ST_SZ_DW(fte_match_param)]; +}; + struct mlx5_flow_destination { enum mlx5_flow_destination_type type; union { @@ -116,9 +122,7 @@ void mlx5_destroy_flow_group(struct mlx5_flow_group *fg); */ struct mlx5_flow_rule * mlx5_add_flow_rule(struct mlx5_flow_table *ft, - u8 match_criteria_enable, - u32 *match_criteria, - u32 *match_value, + struct mlx5_flow_spec *spec, u32 action, u32 flow_tag, struct mlx5_flow_destination *dest); -- cgit v1.2.3 From fba53f7b571925b8a0d59d460ad6de1fda928a3e Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 4 Jul 2016 17:23:06 +0300 Subject: net/mlx5: Introduce mlx5_flow_steering structure Instead of having all steering private name spaces and steering module fields flat in mlx5_core_priv, we wrap them in mlx5_flow_steering for better modularity and API exposure. Signed-off-by: Maor Gottlieb Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 134 ++++++++++++---------- drivers/net/ethernet/mellanox/mlx5/core/fs_core.h | 8 ++ include/linux/mlx5/driver.h | 6 +- 3 files changed, 84 insertions(+), 64 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 7fcdae1cf053..0f969cb2291a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -1363,12 +1363,13 @@ void mlx5_destroy_flow_group(struct mlx5_flow_group *fg) struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, enum mlx5_flow_namespace_type type) { - struct mlx5_flow_root_namespace *root_ns = dev->priv.root_ns; + struct mlx5_flow_steering *steering = dev->priv.steering; + struct mlx5_flow_root_namespace *root_ns; int prio; struct fs_prio *fs_prio; struct mlx5_flow_namespace *ns; - if (!root_ns) + if (!steering) return NULL; switch (type) { @@ -1380,24 +1381,28 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, prio = type; break; case MLX5_FLOW_NAMESPACE_FDB: - if (dev->priv.fdb_root_ns) - return &dev->priv.fdb_root_ns->ns; + if (steering->fdb_root_ns) + return &steering->fdb_root_ns->ns; else return NULL; case MLX5_FLOW_NAMESPACE_ESW_EGRESS: - if (dev->priv.esw_egress_root_ns) - return &dev->priv.esw_egress_root_ns->ns; + if (steering->esw_egress_root_ns) + return &steering->esw_egress_root_ns->ns; else return NULL; case MLX5_FLOW_NAMESPACE_ESW_INGRESS: - if (dev->priv.esw_ingress_root_ns) - return &dev->priv.esw_ingress_root_ns->ns; + if (steering->esw_ingress_root_ns) + return &steering->esw_ingress_root_ns->ns; else return NULL; default: return NULL; } + root_ns = steering->root_ns; + if (!root_ns) + return NULL; + fs_prio = find_prio(&root_ns->ns, prio); if (!fs_prio) return NULL; @@ -1483,13 +1488,13 @@ static bool has_required_caps(struct mlx5_core_dev *dev, struct node_caps *caps) return true; } -static int init_root_tree_recursive(struct mlx5_core_dev *dev, +static int init_root_tree_recursive(struct mlx5_flow_steering *steering, struct init_tree_node *init_node, struct fs_node *fs_parent_node, struct init_tree_node *init_parent_node, int prio) { - int max_ft_level = MLX5_CAP_FLOWTABLE(dev, + int max_ft_level = MLX5_CAP_FLOWTABLE(steering->dev, flow_table_properties_nic_receive. max_ft_level); struct mlx5_flow_namespace *fs_ns; @@ -1500,7 +1505,7 @@ static int init_root_tree_recursive(struct mlx5_core_dev *dev, if (init_node->type == FS_TYPE_PRIO) { if ((init_node->min_ft_level > max_ft_level) || - !has_required_caps(dev, &init_node->caps)) + !has_required_caps(steering->dev, &init_node->caps)) return 0; fs_get_obj(fs_ns, fs_parent_node); @@ -1521,7 +1526,7 @@ static int init_root_tree_recursive(struct mlx5_core_dev *dev, } prio = 0; for (i = 0; i < init_node->ar_size; i++) { - err = init_root_tree_recursive(dev, &init_node->children[i], + err = init_root_tree_recursive(steering, &init_node->children[i], base, init_node, prio); if (err) return err; @@ -1534,7 +1539,7 @@ static int init_root_tree_recursive(struct mlx5_core_dev *dev, return 0; } -static int init_root_tree(struct mlx5_core_dev *dev, +static int init_root_tree(struct mlx5_flow_steering *steering, struct init_tree_node *init_node, struct fs_node *fs_parent_node) { @@ -1544,7 +1549,7 @@ static int init_root_tree(struct mlx5_core_dev *dev, fs_get_obj(fs_ns, fs_parent_node); for (i = 0; i < init_node->ar_size; i++) { - err = init_root_tree_recursive(dev, &init_node->children[i], + err = init_root_tree_recursive(steering, &init_node->children[i], &fs_ns->node, init_node, i); if (err) @@ -1553,7 +1558,7 @@ static int init_root_tree(struct mlx5_core_dev *dev, return 0; } -static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_core_dev *dev, +static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_flow_steering *steering, enum fs_flow_table_type table_type) { @@ -1565,7 +1570,7 @@ static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_core_dev *dev if (!root_ns) return NULL; - root_ns->dev = dev; + root_ns->dev = steering->dev; root_ns->table_type = table_type; ns = &root_ns->ns; @@ -1620,46 +1625,45 @@ static void set_prio_attrs(struct mlx5_flow_root_namespace *root_ns) #define ANCHOR_PRIO 0 #define ANCHOR_SIZE 1 #define ANCHOR_LEVEL 0 -static int create_anchor_flow_table(struct mlx5_core_dev - *dev) +static int create_anchor_flow_table(struct mlx5_flow_steering *steering) { struct mlx5_flow_namespace *ns = NULL; struct mlx5_flow_table *ft; - ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_ANCHOR); + ns = mlx5_get_flow_namespace(steering->dev, MLX5_FLOW_NAMESPACE_ANCHOR); if (!ns) return -EINVAL; ft = mlx5_create_flow_table(ns, ANCHOR_PRIO, ANCHOR_SIZE, ANCHOR_LEVEL); if (IS_ERR(ft)) { - mlx5_core_err(dev, "Failed to create last anchor flow table"); + mlx5_core_err(steering->dev, "Failed to create last anchor flow table"); return PTR_ERR(ft); } return 0; } -static int init_root_ns(struct mlx5_core_dev *dev) +static int init_root_ns(struct mlx5_flow_steering *steering) { - dev->priv.root_ns = create_root_ns(dev, FS_FT_NIC_RX); - if (IS_ERR_OR_NULL(dev->priv.root_ns)) + steering->root_ns = create_root_ns(steering, FS_FT_NIC_RX); + if (IS_ERR_OR_NULL(steering->root_ns)) goto cleanup; - if (init_root_tree(dev, &root_fs, &dev->priv.root_ns->ns.node)) + if (init_root_tree(steering, &root_fs, &steering->root_ns->ns.node)) goto cleanup; - set_prio_attrs(dev->priv.root_ns); + set_prio_attrs(steering->root_ns); - if (create_anchor_flow_table(dev)) + if (create_anchor_flow_table(steering)) goto cleanup; return 0; cleanup: - mlx5_cleanup_fs(dev); + mlx5_cleanup_fs(steering->dev); return -ENOMEM; } -static void cleanup_single_prio_root_ns(struct mlx5_core_dev *dev, +static void cleanup_single_prio_root_ns(struct mlx5_flow_steering *steering, struct mlx5_flow_root_namespace *root_ns) { struct fs_node *prio; @@ -1672,11 +1676,11 @@ static void cleanup_single_prio_root_ns(struct mlx5_core_dev *dev, struct fs_node, list); if (tree_remove_node(prio)) - mlx5_core_warn(dev, + mlx5_core_warn(steering->dev, "Flow steering priority wasn't destroyed, refcount > 1\n"); } if (tree_remove_node(&root_ns->ns.node)) - mlx5_core_warn(dev, + mlx5_core_warn(steering->dev, "Flow steering namespace wasn't destroyed, refcount > 1\n"); root_ns = NULL; } @@ -1690,12 +1694,12 @@ static void destroy_flow_tables(struct fs_prio *prio) mlx5_destroy_flow_table(iter); } -static void cleanup_root_ns(struct mlx5_core_dev *dev) +static void cleanup_root_ns(struct mlx5_flow_steering *steering) { - struct mlx5_flow_root_namespace *root_ns = dev->priv.root_ns; + struct mlx5_flow_root_namespace *root_ns = steering->root_ns; struct fs_prio *iter_prio; - if (!MLX5_CAP_GEN(dev, nic_flow_table)) + if (!MLX5_CAP_GEN(steering->dev, nic_flow_table)) return; if (!root_ns) @@ -1720,7 +1724,7 @@ static void cleanup_root_ns(struct mlx5_core_dev *dev) fs_get_obj(obj_iter_prio2, iter_prio2); destroy_flow_tables(obj_iter_prio2); if (tree_remove_node(iter_prio2)) { - mlx5_core_warn(dev, + mlx5_core_warn(steering->dev, "Priority %d wasn't destroyed, refcount > 1\n", obj_iter_prio2->prio); return; @@ -1737,7 +1741,7 @@ static void cleanup_root_ns(struct mlx5_core_dev *dev) struct fs_node, list); if (tree_remove_node(iter_ns)) { - mlx5_core_warn(dev, + mlx5_core_warn(steering->dev, "Namespace wasn't destroyed, refcount > 1\n"); return; } @@ -1754,7 +1758,7 @@ static void cleanup_root_ns(struct mlx5_core_dev *dev) fs_get_obj(obj_prio_node, prio_node); if (tree_remove_node(prio_node)) { - mlx5_core_warn(dev, + mlx5_core_warn(steering->dev, "Priority %d wasn't destroyed, refcount > 1\n", obj_prio_node->prio); return; @@ -1762,70 +1766,75 @@ static void cleanup_root_ns(struct mlx5_core_dev *dev) } if (tree_remove_node(&root_ns->ns.node)) { - mlx5_core_warn(dev, + mlx5_core_warn(steering->dev, "root namespace wasn't destroyed, refcount > 1\n"); return; } - dev->priv.root_ns = NULL; + steering->root_ns = NULL; } void mlx5_cleanup_fs(struct mlx5_core_dev *dev) { + struct mlx5_flow_steering *steering = dev->priv.steering; + if (MLX5_CAP_GEN(dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) return; - cleanup_root_ns(dev); - cleanup_single_prio_root_ns(dev, dev->priv.fdb_root_ns); - cleanup_single_prio_root_ns(dev, dev->priv.esw_egress_root_ns); - cleanup_single_prio_root_ns(dev, dev->priv.esw_ingress_root_ns); + cleanup_root_ns(steering); + cleanup_single_prio_root_ns(steering, steering->esw_egress_root_ns); + cleanup_single_prio_root_ns(steering, steering->esw_ingress_root_ns); + cleanup_single_prio_root_ns(steering, steering->fdb_root_ns); mlx5_cleanup_fc_stats(dev); + kfree(steering); } -static int init_fdb_root_ns(struct mlx5_core_dev *dev) +static int init_fdb_root_ns(struct mlx5_flow_steering *steering) { struct fs_prio *prio; - dev->priv.fdb_root_ns = create_root_ns(dev, FS_FT_FDB); - if (!dev->priv.fdb_root_ns) + steering->fdb_root_ns = create_root_ns(steering, FS_FT_FDB); + if (!steering->fdb_root_ns) return -ENOMEM; /* Create single prio */ - prio = fs_create_prio(&dev->priv.fdb_root_ns->ns, 0, 1); + prio = fs_create_prio(&steering->fdb_root_ns->ns, 0, 1); if (IS_ERR(prio)) { - cleanup_single_prio_root_ns(dev, dev->priv.fdb_root_ns); + cleanup_single_prio_root_ns(steering, steering->fdb_root_ns); return PTR_ERR(prio); } else { return 0; } } -static int init_egress_acl_root_ns(struct mlx5_core_dev *dev) +static int init_ingress_acl_root_ns(struct mlx5_flow_steering *steering) { struct fs_prio *prio; - dev->priv.esw_egress_root_ns = create_root_ns(dev, FS_FT_ESW_EGRESS_ACL); - if (!dev->priv.esw_egress_root_ns) + steering->esw_egress_root_ns = create_root_ns(steering, FS_FT_ESW_EGRESS_ACL); + if (!steering->esw_egress_root_ns) return -ENOMEM; /* create 1 prio*/ - prio = fs_create_prio(&dev->priv.esw_egress_root_ns->ns, 0, MLX5_TOTAL_VPORTS(dev)); + prio = fs_create_prio(&steering->esw_egress_root_ns->ns, 0, + MLX5_TOTAL_VPORTS(steering->dev)); if (IS_ERR(prio)) return PTR_ERR(prio); else return 0; } -static int init_ingress_acl_root_ns(struct mlx5_core_dev *dev) +static int init_egress_acl_root_ns(struct mlx5_flow_steering *steering) { struct fs_prio *prio; - dev->priv.esw_ingress_root_ns = create_root_ns(dev, FS_FT_ESW_INGRESS_ACL); - if (!dev->priv.esw_ingress_root_ns) + steering->esw_ingress_root_ns = create_root_ns(steering, FS_FT_ESW_INGRESS_ACL); + if (!steering->esw_ingress_root_ns) return -ENOMEM; /* create 1 prio*/ - prio = fs_create_prio(&dev->priv.esw_ingress_root_ns->ns, 0, MLX5_TOTAL_VPORTS(dev)); + prio = fs_create_prio(&steering->esw_ingress_root_ns->ns, 0, + MLX5_TOTAL_VPORTS(steering->dev)); if (IS_ERR(prio)) return PTR_ERR(prio); else @@ -1834,6 +1843,7 @@ static int init_ingress_acl_root_ns(struct mlx5_core_dev *dev) int mlx5_init_fs(struct mlx5_core_dev *dev) { + struct mlx5_flow_steering *steering; int err = 0; if (MLX5_CAP_GEN(dev, port_type) != MLX5_CAP_PORT_TYPE_ETH) @@ -1843,26 +1853,32 @@ int mlx5_init_fs(struct mlx5_core_dev *dev) if (err) return err; + steering = kzalloc(sizeof(*steering), GFP_KERNEL); + if (!steering) + return -ENOMEM; + steering->dev = dev; + dev->priv.steering = steering; + if (MLX5_CAP_GEN(dev, nic_flow_table) && MLX5_CAP_FLOWTABLE_NIC_RX(dev, ft_support)) { - err = init_root_ns(dev); + err = init_root_ns(steering); if (err) goto err; } if (MLX5_CAP_GEN(dev, eswitch_flow_table)) { if (MLX5_CAP_ESW_FLOWTABLE_FDB(dev, ft_support)) { - err = init_fdb_root_ns(dev); + err = init_fdb_root_ns(steering); if (err) goto err; } if (MLX5_CAP_ESW_EGRESS_ACL(dev, ft_support)) { - err = init_egress_acl_root_ns(dev); + err = init_egress_acl_root_ns(steering); if (err) goto err; } if (MLX5_CAP_ESW_INGRESS_ACL(dev, ft_support)) { - err = init_ingress_acl_root_ns(dev); + err = init_ingress_acl_root_ns(steering); if (err) goto err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h index aa41a7314691..d7ba91a1eea2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h @@ -55,6 +55,14 @@ enum fs_fte_status { FS_FTE_STATUS_EXISTING = 1UL << 0, }; +struct mlx5_flow_steering { + struct mlx5_core_dev *dev; + struct mlx5_flow_root_namespace *root_ns; + struct mlx5_flow_root_namespace *fdb_root_ns; + struct mlx5_flow_root_namespace *esw_egress_root_ns; + struct mlx5_flow_root_namespace *esw_ingress_root_ns; +}; + struct fs_node { struct list_head list; struct list_head children; diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index e22b3456b2ee..f21c45941887 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -550,14 +550,10 @@ struct mlx5_priv { struct list_head ctx_list; spinlock_t ctx_lock; + struct mlx5_flow_steering *steering; struct mlx5_eswitch *eswitch; struct mlx5_core_sriov sriov; unsigned long pci_dev_data; - struct mlx5_flow_root_namespace *root_ns; - struct mlx5_flow_root_namespace *fdb_root_ns; - struct mlx5_flow_root_namespace *esw_egress_root_ns; - struct mlx5_flow_root_namespace *esw_ingress_root_ns; - struct mlx5_fc_stats fc_stats; struct mlx5_rl_table rl_table; }; -- cgit v1.2.3 From 6dc6071cfcde6cf687f8d288c9cef9ee6ee24dc7 Mon Sep 17 00:00:00 2001 From: Maor Gottlieb Date: Mon, 4 Jul 2016 17:23:08 +0300 Subject: net/mlx5e: Add ethtool flow steering support Implement etrhtool set_rxnfc callback to support ethtool flow spec direct steering. This patch adds only the support of ether flow type spec. L3/L4 flow specs support will be added in downstream patches. Signed-off-by: Maor Gottlieb Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/Makefile | 2 +- drivers/net/ethernet/mellanox/mlx5/core/en.h | 20 ++ .../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 21 ++ drivers/net/ethernet/mellanox/mlx5/core/en_fs.c | 3 + .../ethernet/mellanox/mlx5/core/en_fs_ethtool.c | 393 +++++++++++++++++++++ drivers/net/ethernet/mellanox/mlx5/core/fs_core.c | 28 +- include/linux/mlx5/fs.h | 1 + 7 files changed, 456 insertions(+), 12 deletions(-) create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index a574deabdda8..05cc1effc13c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -8,6 +8,6 @@ mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \ mlx5_core-$(CONFIG_MLX5_CORE_EN) += wq.o eswitch.o eswitch_offloads.o \ en_main.o en_common.o en_fs.o en_ethtool.o en_tx.o \ en_rx.o en_rx_am.o en_txrx.o en_clock.o vxlan.o \ - en_tc.o en_arfs.o en_rep.o + en_tc.o en_arfs.o en_rep.o en_fs_ethtool.o mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) += en_dcbnl.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 00643a116492..357320e79de2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -544,8 +544,22 @@ enum { MLX5E_ARFS_FT_LEVEL }; +struct mlx5e_ethtool_table { + struct mlx5_flow_table *ft; + int num_rules; +}; + +#define ETHTOOL_NUM_L2_FTS 4 + +struct mlx5e_ethtool_steering { + struct mlx5e_ethtool_table l2_ft[ETHTOOL_NUM_L2_FTS]; + struct list_head rules; + int tot_num_rules; +}; + struct mlx5e_flow_steering { struct mlx5_flow_namespace *ns; + struct mlx5e_ethtool_steering ethtool; struct mlx5e_tc_table tc; struct mlx5e_vlan_table vlan; struct mlx5e_l2_table l2; @@ -701,6 +715,12 @@ int mlx5e_create_flow_steering(struct mlx5e_priv *priv); void mlx5e_destroy_flow_steering(struct mlx5e_priv *priv); void mlx5e_init_l2_addr(struct mlx5e_priv *priv); void mlx5e_destroy_flow_table(struct mlx5e_flow_table *ft); +int mlx5e_ethtool_flow_replace(struct mlx5e_priv *priv, + struct ethtool_rx_flow_spec *fs); +int mlx5e_ethtool_flow_remove(struct mlx5e_priv *priv, + int location); +void mlx5e_ethtool_init_steering(struct mlx5e_priv *priv); +void mlx5e_ethtool_cleanup_steering(struct mlx5e_priv *priv); void mlx5e_set_rx_mode_work(struct work_struct *work); void mlx5e_fill_hwstamp(struct mlx5e_tstamp *clock, u64 timestamp, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 7e61ffa96732..edbb665f4d22 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -1368,6 +1368,26 @@ static u32 mlx5e_get_priv_flags(struct net_device *netdev) return priv->pflags; } +static int mlx5e_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd) +{ + int err = 0; + struct mlx5e_priv *priv = netdev_priv(dev); + + switch (cmd->cmd) { + case ETHTOOL_SRXCLSRLINS: + err = mlx5e_ethtool_flow_replace(priv, &cmd->fs); + break; + case ETHTOOL_SRXCLSRLDEL: + err = mlx5e_ethtool_flow_remove(priv, cmd->fs.location); + break; + default: + err = -EOPNOTSUPP; + break; + } + + return err; +} + const struct ethtool_ops mlx5e_ethtool_ops = { .get_drvinfo = mlx5e_get_drvinfo, .get_link = ethtool_op_get_link, @@ -1387,6 +1407,7 @@ const struct ethtool_ops mlx5e_ethtool_ops = { .get_rxfh = mlx5e_get_rxfh, .set_rxfh = mlx5e_set_rxfh, .get_rxnfc = mlx5e_get_rxnfc, + .set_rxnfc = mlx5e_set_rxnfc, .get_tunable = mlx5e_get_tunable, .set_tunable = mlx5e_set_tunable, .get_pauseparam = mlx5e_get_pauseparam, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c index 2e1e86316fe7..1587a9fd5724 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs.c @@ -1084,6 +1084,8 @@ int mlx5e_create_flow_steering(struct mlx5e_priv *priv) goto err_destroy_l2_table; } + mlx5e_ethtool_init_steering(priv); + return 0; err_destroy_l2_table: @@ -1103,4 +1105,5 @@ void mlx5e_destroy_flow_steering(struct mlx5e_priv *priv) mlx5e_destroy_l2_table(priv); mlx5e_destroy_ttc_table(priv); mlx5e_arfs_destroy_tables(priv); + mlx5e_ethtool_cleanup_steering(priv); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c new file mode 100644 index 000000000000..ee28a9fc0b9d --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_fs_ethtool.c @@ -0,0 +1,393 @@ +/* + * Copyright (c) 2016, Mellanox Technologies. All rights reserved. + * + * This software is available to you under a choice of one of two + * licenses. You may choose to be licensed under the terms of the GNU + * General Public License (GPL) Version 2, available from the file + * COPYING in the main directory of this source tree, or the + * OpenIB.org BSD license below: + * + * Redistribution and use in source and binary forms, with or + * without modification, are permitted provided that the following + * conditions are met: + * + * - Redistributions of source code must retain the above + * copyright notice, this list of conditions and the following + * disclaimer. + * + * - Redistributions in binary form must reproduce the above + * copyright notice, this list of conditions and the following + * disclaimer in the documentation and/or other materials + * provided with the distribution. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, + * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND + * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS + * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN + * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + * SOFTWARE. + */ + +#include +#include "en.h" + +struct mlx5e_ethtool_rule { + struct list_head list; + struct ethtool_rx_flow_spec flow_spec; + struct mlx5_flow_rule *rule; + struct mlx5e_ethtool_table *eth_ft; +}; + +static void put_flow_table(struct mlx5e_ethtool_table *eth_ft) +{ + if (!--eth_ft->num_rules) { + mlx5_destroy_flow_table(eth_ft->ft); + eth_ft->ft = NULL; + } +} + +#define MLX5E_ETHTOOL_L2_PRIO 0 +#define MLX5E_ETHTOOL_NUM_ENTRIES 64000 +#define MLX5E_ETHTOOL_NUM_GROUPS 10 +static struct mlx5e_ethtool_table *get_flow_table(struct mlx5e_priv *priv, + struct ethtool_rx_flow_spec *fs, + int num_tuples) +{ + struct mlx5e_ethtool_table *eth_ft; + struct mlx5_flow_namespace *ns; + struct mlx5_flow_table *ft; + int max_tuples; + int table_size; + int prio; + + switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) { + case ETHER_FLOW: + max_tuples = ETHTOOL_NUM_L2_FTS; + prio = max_tuples - num_tuples; + eth_ft = &priv->fs.ethtool.l2_ft[prio]; + prio += MLX5E_ETHTOOL_L2_PRIO; + break; + default: + return ERR_PTR(-EINVAL); + } + + eth_ft->num_rules++; + if (eth_ft->ft) + return eth_ft; + + ns = mlx5_get_flow_namespace(priv->mdev, + MLX5_FLOW_NAMESPACE_ETHTOOL); + if (!ns) + return ERR_PTR(-ENOTSUPP); + + table_size = min_t(u32, BIT(MLX5_CAP_FLOWTABLE(priv->mdev, + flow_table_properties_nic_receive.log_max_ft_size)), + MLX5E_ETHTOOL_NUM_ENTRIES); + ft = mlx5_create_auto_grouped_flow_table(ns, prio, + table_size, + MLX5E_ETHTOOL_NUM_GROUPS, 0); + if (IS_ERR(ft)) + return (void *)ft; + + eth_ft->ft = ft; + return eth_ft; +} + +static void mask_spec(u8 *mask, u8 *val, size_t size) +{ + unsigned int i; + + for (i = 0; i < size; i++, mask++, val++) + *((u8 *)val) = *((u8 *)mask) & *((u8 *)val); +} + +static int set_flow_attrs(u32 *match_c, u32 *match_v, + struct ethtool_rx_flow_spec *fs) +{ + void *outer_headers_c = MLX5_ADDR_OF(fte_match_param, match_c, + outer_headers); + void *outer_headers_v = MLX5_ADDR_OF(fte_match_param, match_v, + outer_headers); + u32 flow_type = fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT); + struct ethhdr *eth_val; + struct ethhdr *eth_mask; + + switch (flow_type) { + case ETHER_FLOW: + eth_mask = &fs->m_u.ether_spec; + eth_val = &fs->h_u.ether_spec; + + mask_spec((u8 *)eth_mask, (u8 *)eth_val, sizeof(*eth_mask)); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_c, smac_47_16), + eth_mask->h_source); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_v, smac_47_16), + eth_val->h_source); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_c, dmac_47_16), + eth_mask->h_dest); + ether_addr_copy(MLX5_ADDR_OF(fte_match_set_lyr_2_4, + outer_headers_v, dmac_47_16), + eth_val->h_dest); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, ethertype, + ntohs(eth_mask->h_proto)); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, ethertype, + ntohs(eth_val->h_proto)); + break; + default: + return -EINVAL; + } + + if ((fs->flow_type & FLOW_EXT) && + (fs->m_ext.vlan_tci & cpu_to_be16(VLAN_VID_MASK))) { + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + vlan_tag, 1); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + vlan_tag, 1); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_c, + first_vid, 0xfff); + MLX5_SET(fte_match_set_lyr_2_4, outer_headers_v, + first_vid, ntohs(fs->h_ext.vlan_tci)); + } + + return 0; +} + +static void add_rule_to_list(struct mlx5e_priv *priv, + struct mlx5e_ethtool_rule *rule) +{ + struct mlx5e_ethtool_rule *iter; + struct list_head *head = &priv->fs.ethtool.rules; + + list_for_each_entry(iter, &priv->fs.ethtool.rules, list) { + if (iter->flow_spec.location > rule->flow_spec.location) + break; + head = &iter->list; + } + priv->fs.ethtool.tot_num_rules++; + list_add(&rule->list, head); +} + +static bool outer_header_zero(u32 *match_criteria) +{ + int size = MLX5_ST_SZ_BYTES(fte_match_param); + char *outer_headers_c = MLX5_ADDR_OF(fte_match_param, match_criteria, + outer_headers); + + return outer_headers_c[0] == 0 && !memcmp(outer_headers_c, + outer_headers_c + 1, + size - 1); +} + +static struct mlx5_flow_rule *add_ethtool_flow_rule(struct mlx5e_priv *priv, + struct mlx5_flow_table *ft, + struct ethtool_rx_flow_spec *fs) +{ + struct mlx5_flow_destination *dst = NULL; + struct mlx5_flow_spec *spec; + struct mlx5_flow_rule *rule; + int err = 0; + u32 action; + + spec = mlx5_vzalloc(sizeof(*spec)); + if (!spec) + return ERR_PTR(-ENOMEM); + err = set_flow_attrs(spec->match_criteria, spec->match_value, + fs); + if (err) + goto free; + + if (fs->ring_cookie == RX_CLS_FLOW_DISC) { + action = MLX5_FLOW_CONTEXT_ACTION_DROP; + } else { + dst = kzalloc(sizeof(*dst), GFP_KERNEL); + if (!dst) { + err = -ENOMEM; + goto free; + } + + dst->type = MLX5_FLOW_DESTINATION_TYPE_TIR; + dst->tir_num = priv->direct_tir[fs->ring_cookie].tirn; + action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + } + + spec->match_criteria_enable = (!outer_header_zero(spec->match_criteria)); + rule = mlx5_add_flow_rule(ft, spec, action, + MLX5_FS_DEFAULT_FLOW_TAG, dst); + if (IS_ERR(rule)) { + err = PTR_ERR(rule); + netdev_err(priv->netdev, "%s: failed to add ethtool steering rule: %d\n", + __func__, err); + goto free; + } +free: + kvfree(spec); + kfree(dst); + return err ? ERR_PTR(err) : rule; +} + +static void del_ethtool_rule(struct mlx5e_priv *priv, + struct mlx5e_ethtool_rule *eth_rule) +{ + if (eth_rule->rule) + mlx5_del_flow_rule(eth_rule->rule); + list_del(ð_rule->list); + priv->fs.ethtool.tot_num_rules--; + put_flow_table(eth_rule->eth_ft); + kfree(eth_rule); +} + +static struct mlx5e_ethtool_rule *find_ethtool_rule(struct mlx5e_priv *priv, + int location) +{ + struct mlx5e_ethtool_rule *iter; + + list_for_each_entry(iter, &priv->fs.ethtool.rules, list) { + if (iter->flow_spec.location == location) + return iter; + } + return NULL; +} + +static struct mlx5e_ethtool_rule *get_ethtool_rule(struct mlx5e_priv *priv, + int location) +{ + struct mlx5e_ethtool_rule *eth_rule; + + eth_rule = find_ethtool_rule(priv, location); + if (eth_rule) + del_ethtool_rule(priv, eth_rule); + + eth_rule = kzalloc(sizeof(*eth_rule), GFP_KERNEL); + if (!eth_rule) + return ERR_PTR(-ENOMEM); + + add_rule_to_list(priv, eth_rule); + return eth_rule; +} + +#define MAX_NUM_OF_ETHTOOL_RULES BIT(10) +static int validate_flow(struct mlx5e_priv *priv, + struct ethtool_rx_flow_spec *fs) +{ + struct ethhdr *eth_mask; + int num_tuples = 0; + + if (fs->location >= MAX_NUM_OF_ETHTOOL_RULES) + return -EINVAL; + + if (fs->ring_cookie >= priv->params.num_channels && + fs->ring_cookie != RX_CLS_FLOW_DISC) + return -EINVAL; + + switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) { + case ETHER_FLOW: + eth_mask = &fs->m_u.ether_spec; + if (!is_zero_ether_addr(eth_mask->h_dest)) + num_tuples++; + if (!is_zero_ether_addr(eth_mask->h_source)) + num_tuples++; + if (eth_mask->h_proto) + num_tuples++; + break; + default: + return -EINVAL; + } + if ((fs->flow_type & FLOW_EXT)) { + if (fs->m_ext.vlan_etype || + (fs->m_ext.vlan_tci != cpu_to_be16(VLAN_VID_MASK))) + return -EINVAL; + + if (fs->m_ext.vlan_tci) { + if (be16_to_cpu(fs->h_ext.vlan_tci) >= VLAN_N_VID) + return -EINVAL; + } + num_tuples++; + } + + return num_tuples; +} + +int mlx5e_ethtool_flow_replace(struct mlx5e_priv *priv, + struct ethtool_rx_flow_spec *fs) +{ + struct mlx5e_ethtool_table *eth_ft; + struct mlx5e_ethtool_rule *eth_rule; + struct mlx5_flow_rule *rule; + int num_tuples; + int err; + + num_tuples = validate_flow(priv, fs); + if (num_tuples <= 0) { + netdev_warn(priv->netdev, "%s: flow is not valid\n", __func__); + return -EINVAL; + } + + eth_ft = get_flow_table(priv, fs, num_tuples); + if (IS_ERR(eth_ft)) + return PTR_ERR(eth_ft); + + eth_rule = get_ethtool_rule(priv, fs->location); + if (IS_ERR(eth_rule)) { + put_flow_table(eth_ft); + return PTR_ERR(eth_rule); + } + + eth_rule->flow_spec = *fs; + eth_rule->eth_ft = eth_ft; + if (!eth_ft->ft) { + err = -EINVAL; + goto del_ethtool_rule; + } + rule = add_ethtool_flow_rule(priv, eth_ft->ft, fs); + if (IS_ERR(rule)) { + err = PTR_ERR(rule); + goto del_ethtool_rule; + } + + eth_rule->rule = rule; + + return 0; + +del_ethtool_rule: + del_ethtool_rule(priv, eth_rule); + + return err; +} + +int mlx5e_ethtool_flow_remove(struct mlx5e_priv *priv, + int location) +{ + struct mlx5e_ethtool_rule *eth_rule; + int err = 0; + + if (location >= MAX_NUM_OF_ETHTOOL_RULES) + return -ENOSPC; + + eth_rule = find_ethtool_rule(priv, location); + if (!eth_rule) { + err = -ENOENT; + goto out; + } + + del_ethtool_rule(priv, eth_rule); +out: + return err; +} + +void mlx5e_ethtool_cleanup_steering(struct mlx5e_priv *priv) +{ + struct mlx5e_ethtool_rule *iter; + struct mlx5e_ethtool_rule *temp; + + list_for_each_entry_safe(iter, temp, &priv->fs.ethtool.rules, list) + del_ethtool_rule(priv, iter); +} + +void mlx5e_ethtool_init_steering(struct mlx5e_priv *priv) +{ + INIT_LIST_HEAD(&priv->fs.ethtool.rules); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 3e9577537b52..83fa98fde795 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -67,13 +67,21 @@ #define FS_REQUIRED_CAPS(...) {.arr_sz = INIT_CAPS_ARRAY_SIZE(__VA_ARGS__), \ .caps = (long[]) {__VA_ARGS__} } +#define FS_CHAINING_CAPS FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en), \ + FS_CAP(flow_table_properties_nic_receive.modify_root), \ + FS_CAP(flow_table_properties_nic_receive.identified_miss_table_mode), \ + FS_CAP(flow_table_properties_nic_receive.flow_table_modify)) + #define LEFTOVERS_NUM_LEVELS 1 #define LEFTOVERS_NUM_PRIOS 1 #define BY_PASS_PRIO_NUM_LEVELS 1 -#define BY_PASS_MIN_LEVEL (KERNEL_MIN_LEVEL + MLX5_BY_PASS_NUM_PRIOS +\ +#define BY_PASS_MIN_LEVEL (ETHTOOL_MIN_LEVEL + MLX5_BY_PASS_NUM_PRIOS +\ LEFTOVERS_NUM_PRIOS) +#define ETHTOOL_PRIO_NUM_LEVELS 1 +#define ETHTOOL_NUM_PRIOS 4 +#define ETHTOOL_MIN_LEVEL (KERNEL_MIN_LEVEL + ETHTOOL_NUM_PRIOS) /* Vlan, mac, ttc, aRFS */ #define KERNEL_NIC_PRIO_NUM_LEVELS 4 #define KERNEL_NIC_NUM_PRIOS 1 @@ -103,27 +111,24 @@ static struct init_tree_node { int num_levels; } root_fs = { .type = FS_TYPE_NAMESPACE, - .ar_size = 5, + .ar_size = 6, .children = (struct init_tree_node[]) { ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0, - FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en), - FS_CAP(flow_table_properties_nic_receive.modify_root), - FS_CAP(flow_table_properties_nic_receive.identified_miss_table_mode), - FS_CAP(flow_table_properties_nic_receive.flow_table_modify)), + FS_CHAINING_CAPS, ADD_NS(ADD_MULTIPLE_PRIO(MLX5_BY_PASS_NUM_PRIOS, BY_PASS_PRIO_NUM_LEVELS))), ADD_PRIO(0, OFFLOADS_MIN_LEVEL, 0, {}, ADD_NS(ADD_MULTIPLE_PRIO(OFFLOADS_NUM_PRIOS, OFFLOADS_MAX_FT))), - + ADD_PRIO(0, ETHTOOL_MIN_LEVEL, 0, + FS_CHAINING_CAPS, + ADD_NS(ADD_MULTIPLE_PRIO(ETHTOOL_NUM_PRIOS, + ETHTOOL_PRIO_NUM_LEVELS))), ADD_PRIO(0, KERNEL_MIN_LEVEL, 0, {}, ADD_NS(ADD_MULTIPLE_PRIO(1, 1), ADD_MULTIPLE_PRIO(KERNEL_NIC_NUM_PRIOS, KERNEL_NIC_PRIO_NUM_LEVELS))), ADD_PRIO(0, BY_PASS_MIN_LEVEL, 0, - FS_REQUIRED_CAPS(FS_CAP(flow_table_properties_nic_receive.flow_modify_en), - FS_CAP(flow_table_properties_nic_receive.modify_root), - FS_CAP(flow_table_properties_nic_receive.identified_miss_table_mode), - FS_CAP(flow_table_properties_nic_receive.flow_table_modify)), + FS_CHAINING_CAPS, ADD_NS(ADD_MULTIPLE_PRIO(LEFTOVERS_NUM_PRIOS, LEFTOVERS_NUM_LEVELS))), ADD_PRIO(0, ANCHOR_MIN_LEVEL, 0, {}, ADD_NS(ADD_MULTIPLE_PRIO(ANCHOR_NUM_PRIOS, ANCHOR_NUM_LEVELS))), @@ -1375,6 +1380,7 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, switch (type) { case MLX5_FLOW_NAMESPACE_BYPASS: case MLX5_FLOW_NAMESPACE_OFFLOADS: + case MLX5_FLOW_NAMESPACE_ETHTOOL: case MLX5_FLOW_NAMESPACE_KERNEL: case MLX5_FLOW_NAMESPACE_LEFTOVERS: case MLX5_FLOW_NAMESPACE_ANCHOR: diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h index d22fe7e5a39a..e036d6030867 100644 --- a/include/linux/mlx5/fs.h +++ b/include/linux/mlx5/fs.h @@ -55,6 +55,7 @@ static inline void build_leftovers_ft_param(int *priority, enum mlx5_flow_namespace_type { MLX5_FLOW_NAMESPACE_BYPASS, MLX5_FLOW_NAMESPACE_OFFLOADS, + MLX5_FLOW_NAMESPACE_ETHTOOL, MLX5_FLOW_NAMESPACE_KERNEL, MLX5_FLOW_NAMESPACE_LEFTOVERS, MLX5_FLOW_NAMESPACE_ANCHOR, -- cgit v1.2.3 From 503eebc265dcf5c512454fd5a6b6673ea4f1d7f2 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 5 Jul 2016 11:27:37 +0200 Subject: net: add dev arg to ndo_neigh_construct/destroy As the following patch will allow upper devices to follow the call down lower devices, we need to add dev here and not rely on n->dev. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller --- drivers/net/ethernet/rocker/rocker_main.c | 3 ++- include/linux/netdevice.h | 6 ++++-- net/atm/clip.c | 2 +- net/core/neighbour.c | 4 ++-- net/ieee802154/6lowpan/core.c | 2 +- 5 files changed, 10 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c index 28b775e5a9ad..f0b09b05ed3f 100644 --- a/drivers/net/ethernet/rocker/rocker_main.c +++ b/drivers/net/ethernet/rocker/rocker_main.c @@ -1996,7 +1996,8 @@ static int rocker_port_change_proto_down(struct net_device *dev, return 0; } -static void rocker_port_neigh_destroy(struct neighbour *n) +static void rocker_port_neigh_destroy(struct net_device *dev, + struct neighbour *n) { struct rocker_port *rocker_port = netdev_priv(n->dev); int err; diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 0c6ee2c5099f..91af73c9dd51 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1209,8 +1209,10 @@ struct net_device_ops { netdev_features_t features); int (*ndo_set_features)(struct net_device *dev, netdev_features_t features); - int (*ndo_neigh_construct)(struct neighbour *n); - void (*ndo_neigh_destroy)(struct neighbour *n); + int (*ndo_neigh_construct)(struct net_device *dev, + struct neighbour *n); + void (*ndo_neigh_destroy)(struct net_device *dev, + struct neighbour *n); int (*ndo_fdb_add)(struct ndmsg *ndm, struct nlattr *tb[], diff --git a/net/atm/clip.c b/net/atm/clip.c index e07f551a863c..53b4ac09e7b7 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -286,7 +286,7 @@ static const struct neigh_ops clip_neigh_ops = { .connected_output = neigh_direct_output, }; -static int clip_constructor(struct neighbour *neigh) +static int clip_constructor(struct net_device *dev, struct neighbour *neigh) { struct atmarp_entry *entry = neighbour_priv(neigh); diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 510cd62fcb99..952aabb5aa56 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -473,7 +473,7 @@ struct neighbour *__neigh_create(struct neigh_table *tbl, const void *pkey, } if (dev->netdev_ops->ndo_neigh_construct) { - error = dev->netdev_ops->ndo_neigh_construct(n); + error = dev->netdev_ops->ndo_neigh_construct(dev, n); if (error < 0) { rc = ERR_PTR(error); goto out_neigh_release; @@ -701,7 +701,7 @@ void neigh_destroy(struct neighbour *neigh) neigh->arp_queue_len_bytes = 0; if (dev->netdev_ops->ndo_neigh_destroy) - dev->netdev_ops->ndo_neigh_destroy(neigh); + dev->netdev_ops->ndo_neigh_destroy(dev, neigh); dev_put(dev); neigh_parms_put(neigh->parms); diff --git a/net/ieee802154/6lowpan/core.c b/net/ieee802154/6lowpan/core.c index 8c004a0c8d64..935ab932e841 100644 --- a/net/ieee802154/6lowpan/core.c +++ b/net/ieee802154/6lowpan/core.c @@ -81,7 +81,7 @@ static int lowpan_stop(struct net_device *dev) return 0; } -static int lowpan_neigh_construct(struct neighbour *n) +static int lowpan_neigh_construct(struct net_device *dev, struct neighbour *n) { struct lowpan_802154_neigh *neigh = lowpan_802154_neigh(neighbour_priv(n)); -- cgit v1.2.3 From 18bfb924f0005a728caadd90ba755b2a660bf441 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Tue, 5 Jul 2016 11:27:38 +0200 Subject: net: introduce default neigh_construct/destroy ndo calls for L2 upper devices L2 upper device needs to propagate neigh_construct/destroy calls down to lower devices. Do this by defining default ndo functions and use them in team, bond, bridge and vlan. Signed-off-by: Jiri Pirko Reviewed-by: Ido Schimmel Signed-off-by: David S. Miller --- drivers/net/bonding/bond_main.c | 2 ++ drivers/net/team/team.c | 2 ++ include/linux/netdevice.h | 4 ++++ net/8021q/vlan_dev.c | 2 ++ net/bridge/br_device.c | 2 ++ net/core/dev.c | 44 +++++++++++++++++++++++++++++++++++++++++ 6 files changed, 56 insertions(+) (limited to 'include/linux') diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 90157e20357e..480d73ac7d1b 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -4137,6 +4137,8 @@ static const struct net_device_ops bond_netdev_ops = { .ndo_add_slave = bond_enslave, .ndo_del_slave = bond_release, .ndo_fix_features = bond_fix_features, + .ndo_neigh_construct = netdev_default_l2upper_neigh_construct, + .ndo_neigh_destroy = netdev_default_l2upper_neigh_destroy, .ndo_bridge_setlink = switchdev_port_bridge_setlink, .ndo_bridge_getlink = switchdev_port_bridge_getlink, .ndo_bridge_dellink = switchdev_port_bridge_dellink, diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c index f9eebea83516..a380649bf6b5 100644 --- a/drivers/net/team/team.c +++ b/drivers/net/team/team.c @@ -2002,6 +2002,8 @@ static const struct net_device_ops team_netdev_ops = { .ndo_add_slave = team_add_slave, .ndo_del_slave = team_del_slave, .ndo_fix_features = team_fix_features, + .ndo_neigh_construct = netdev_default_l2upper_neigh_construct, + .ndo_neigh_destroy = netdev_default_l2upper_neigh_destroy, .ndo_change_carrier = team_change_carrier, .ndo_bridge_setlink = switchdev_port_bridge_setlink, .ndo_bridge_getlink = switchdev_port_bridge_getlink, diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 91af73c9dd51..49736a31acaa 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3845,6 +3845,10 @@ void *netdev_lower_dev_get_private(struct net_device *dev, struct net_device *lower_dev); void netdev_lower_state_changed(struct net_device *lower_dev, void *lower_state_info); +int netdev_default_l2upper_neigh_construct(struct net_device *dev, + struct neighbour *n); +void netdev_default_l2upper_neigh_destroy(struct net_device *dev, + struct neighbour *n); /* RSS keys are 40 or 52 bytes long */ #define NETDEV_RSS_KEY_LEN 52 diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 86ae75b77390..c8f422c90856 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -790,6 +790,8 @@ static const struct net_device_ops vlan_netdev_ops = { .ndo_netpoll_cleanup = vlan_dev_netpoll_cleanup, #endif .ndo_fix_features = vlan_dev_fix_features, + .ndo_neigh_construct = netdev_default_l2upper_neigh_construct, + .ndo_neigh_destroy = netdev_default_l2upper_neigh_destroy, .ndo_fdb_add = switchdev_port_fdb_add, .ndo_fdb_del = switchdev_port_fdb_del, .ndo_fdb_dump = switchdev_port_fdb_dump, diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 0c39e0f6da09..8eecd0ec22f2 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -349,6 +349,8 @@ static const struct net_device_ops br_netdev_ops = { .ndo_add_slave = br_add_slave, .ndo_del_slave = br_del_slave, .ndo_fix_features = br_fix_features, + .ndo_neigh_construct = netdev_default_l2upper_neigh_construct, + .ndo_neigh_destroy = netdev_default_l2upper_neigh_destroy, .ndo_fdb_add = br_fdb_add, .ndo_fdb_del = br_fdb_delete, .ndo_fdb_dump = br_fdb_dump, diff --git a/net/core/dev.c b/net/core/dev.c index a4f3b0a9aeaf..b92d63bfde7a 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -6087,6 +6087,50 @@ void netdev_lower_state_changed(struct net_device *lower_dev, } EXPORT_SYMBOL(netdev_lower_state_changed); +int netdev_default_l2upper_neigh_construct(struct net_device *dev, + struct neighbour *n) +{ + struct net_device *lower_dev, *stop_dev; + struct list_head *iter; + int err; + + netdev_for_each_lower_dev(dev, lower_dev, iter) { + if (!lower_dev->netdev_ops->ndo_neigh_construct) + continue; + err = lower_dev->netdev_ops->ndo_neigh_construct(lower_dev, n); + if (err) { + stop_dev = lower_dev; + goto rollback; + } + } + return 0; + +rollback: + netdev_for_each_lower_dev(dev, lower_dev, iter) { + if (lower_dev == stop_dev) + break; + if (!lower_dev->netdev_ops->ndo_neigh_destroy) + continue; + lower_dev->netdev_ops->ndo_neigh_destroy(lower_dev, n); + } + return err; +} +EXPORT_SYMBOL_GPL(netdev_default_l2upper_neigh_construct); + +void netdev_default_l2upper_neigh_destroy(struct net_device *dev, + struct neighbour *n) +{ + struct net_device *lower_dev; + struct list_head *iter; + + netdev_for_each_lower_dev(dev, lower_dev, iter) { + if (!lower_dev->netdev_ops->ndo_neigh_destroy) + continue; + lower_dev->netdev_ops->ndo_neigh_destroy(lower_dev, n); + } +} +EXPORT_SYMBOL_GPL(netdev_default_l2upper_neigh_destroy); + static void dev_change_rx_flags(struct net_device *dev, int flags) { const struct net_device_ops *ops = dev->netdev_ops; -- cgit v1.2.3 From c1adf20052d80f776849fa2c1acb472cdeb7786c Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 1 Jul 2016 07:53:51 +0100 Subject: Introduce rb_replace_node_rcu() Implement an RCU-safe variant of rb_replace_node() and rearrange rb_replace_node() to do things in the same order. Signed-off-by: David Howells Acked-by: Peter Zijlstra (Intel) --- include/linux/rbtree.h | 2 ++ include/linux/rbtree_augmented.h | 13 +++++++++++++ lib/rbtree.c | 26 ++++++++++++++++++++++++-- 3 files changed, 39 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rbtree.h b/include/linux/rbtree.h index b6900099ea81..e585018498d5 100644 --- a/include/linux/rbtree.h +++ b/include/linux/rbtree.h @@ -76,6 +76,8 @@ extern struct rb_node *rb_next_postorder(const struct rb_node *); /* Fast replacement of a single node without remove/rebalance/add/rebalance */ extern void rb_replace_node(struct rb_node *victim, struct rb_node *new, struct rb_root *root); +extern void rb_replace_node_rcu(struct rb_node *victim, struct rb_node *new, + struct rb_root *root); static inline void rb_link_node(struct rb_node *node, struct rb_node *parent, struct rb_node **rb_link) diff --git a/include/linux/rbtree_augmented.h b/include/linux/rbtree_augmented.h index 14d7b831b63a..d076183e49be 100644 --- a/include/linux/rbtree_augmented.h +++ b/include/linux/rbtree_augmented.h @@ -130,6 +130,19 @@ __rb_change_child(struct rb_node *old, struct rb_node *new, WRITE_ONCE(root->rb_node, new); } +static inline void +__rb_change_child_rcu(struct rb_node *old, struct rb_node *new, + struct rb_node *parent, struct rb_root *root) +{ + if (parent) { + if (parent->rb_left == old) + rcu_assign_pointer(parent->rb_left, new); + else + rcu_assign_pointer(parent->rb_right, new); + } else + rcu_assign_pointer(root->rb_node, new); +} + extern void __rb_erase_color(struct rb_node *parent, struct rb_root *root, void (*augment_rotate)(struct rb_node *old, struct rb_node *new)); diff --git a/lib/rbtree.c b/lib/rbtree.c index 1356454e36de..eb8a19fee110 100644 --- a/lib/rbtree.c +++ b/lib/rbtree.c @@ -539,17 +539,39 @@ void rb_replace_node(struct rb_node *victim, struct rb_node *new, { struct rb_node *parent = rb_parent(victim); + /* Copy the pointers/colour from the victim to the replacement */ + *new = *victim; + /* Set the surrounding nodes to point to the replacement */ - __rb_change_child(victim, new, parent, root); if (victim->rb_left) rb_set_parent(victim->rb_left, new); if (victim->rb_right) rb_set_parent(victim->rb_right, new); + __rb_change_child(victim, new, parent, root); +} +EXPORT_SYMBOL(rb_replace_node); + +void rb_replace_node_rcu(struct rb_node *victim, struct rb_node *new, + struct rb_root *root) +{ + struct rb_node *parent = rb_parent(victim); /* Copy the pointers/colour from the victim to the replacement */ *new = *victim; + + /* Set the surrounding nodes to point to the replacement */ + if (victim->rb_left) + rb_set_parent(victim->rb_left, new); + if (victim->rb_right) + rb_set_parent(victim->rb_right, new); + + /* Set the parent's pointer to the new node last after an RCU barrier + * so that the pointers onwards are seen to be set correctly when doing + * an RCU walk over the tree. + */ + __rb_change_child_rcu(victim, new, parent, root); } -EXPORT_SYMBOL(rb_replace_node); +EXPORT_SYMBOL(rb_replace_node_rcu); static struct rb_node *rb_left_deepest_node(const struct rb_node *node) { -- cgit v1.2.3 From 995f1405610bd8446c5be37d2ffc031a7729e406 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 1 Jul 2016 13:44:53 -0700 Subject: rcu: Suppress sparse warnings for rcu_dereference_raw() Data structures that are used both with and without RCU protection are difficult to write in a sparse-clean manner. If you mark the relevant pointers with __rcu, sparse will complain about all non-RCU uses, but if you don't mark those pointers, sparse will complain about all RCU uses. This commit therefore suppresses sparse warnings for rcu_dereference_raw(), allowing mixed-protection data structures to avoid these warnings. Reported-by: David Howells Signed-off-by: Paul E. McKenney Signed-off-by: David Howells --- include/linux/rcupdate.h | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 5f1533e3d032..85830e6c797b 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -611,6 +611,12 @@ static inline void rcu_preempt_sleep_check(void) rcu_dereference_sparse(p, space); \ ((typeof(*p) __force __kernel *)(p)); \ }) +#define rcu_dereference_raw(p) \ +({ \ + /* Dependency order vs. p above. */ \ + typeof(p) ________p1 = lockless_dereference(p); \ + ((typeof(*p) __force __kernel *)(________p1)); \ +}) /** * RCU_INITIALIZER() - statically initialize an RCU-protected global variable @@ -729,8 +735,6 @@ static inline void rcu_preempt_sleep_check(void) __rcu_dereference_check((p), (c) || rcu_read_lock_sched_held(), \ __rcu) -#define rcu_dereference_raw(p) rcu_dereference_check(p, 1) /*@@@ needed? @@@*/ - /* * The tracing infrastructure traces RCU (we want that), but unfortunately * some of the RCU checks causes tracing to lock up the system. -- cgit v1.2.3 From 9cc577dd25b9762df7f353658426bb2e048c480a Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 6 Jul 2016 23:32:24 +0200 Subject: ieee802154: add ieee802154_skb_dst_pan helper This patch adds ieee802154_skb_dst_pan function to get the pointer address of the destination pan id at skb mac pointer. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann --- include/linux/ieee802154.h | 16 ++++++++++++++++ include/net/mac802154.h | 29 +++++++++++++++++++++++++++++ 2 files changed, 45 insertions(+) (limited to 'include/linux') diff --git a/include/linux/ieee802154.h b/include/linux/ieee802154.h index acedbb68a5a3..91f4665fea63 100644 --- a/include/linux/ieee802154.h +++ b/include/linux/ieee802154.h @@ -31,6 +31,8 @@ #define IEEE802154_MIN_PSDU_LEN 9 #define IEEE802154_FCS_LEN 2 #define IEEE802154_MAX_AUTH_TAG_LEN 16 +#define IEEE802154_FC_LEN 2 +#define IEEE802154_SEQ_LEN 1 /* General MAC frame format: * 2 bytes: Frame Control @@ -221,9 +223,14 @@ enum { #define IEEE802154_FCTL_ACKREQ 0x0020 #define IEEE802154_FCTL_SECEN 0x0004 #define IEEE802154_FCTL_INTRA_PAN 0x0040 +#define IEEE802154_FCTL_DADDR 0x0c00 #define IEEE802154_FTYPE_DATA 0x0001 +#define IEEE802154_FCTL_ADDR_NONE 0x0000 +#define IEEE802154_FCTL_DADDR_SHORT 0x0800 +#define IEEE802154_FCTL_DADDR_EXTENDED 0x0c00 + /* * ieee802154_is_data - check if type is IEEE802154_FTYPE_DATA * @fc: frame control bytes in little-endian byteorder @@ -261,6 +268,15 @@ static inline bool ieee802154_is_intra_pan(__le16 fc) return fc & cpu_to_le16(IEEE802154_FCTL_INTRA_PAN); } +/* + * ieee802154_daddr_mode - get daddr mode from fc + * @fc: frame control bytes in little-endian byteorder + */ +static inline __le16 ieee802154_daddr_mode(__le16 fc) +{ + return fc & cpu_to_le16(IEEE802154_FCTL_DADDR); +} + /** * ieee802154_is_valid_psdu_len - check if psdu len is valid * available lengths: diff --git a/include/net/mac802154.h b/include/net/mac802154.h index e465c8551ac3..b3f7cd868fe9 100644 --- a/include/net/mac802154.h +++ b/include/net/mac802154.h @@ -257,6 +257,35 @@ static inline __le16 ieee802154_get_fc_from_skb(const struct sk_buff *skb) return get_unaligned_le16(skb_mac_header(skb)); } +/** + * ieee802154_skb_dst_pan - get the pointer to destination pan field + * @fc: mac header frame control field + * @skb: skb where the destination pan pointer will be get from + */ +static inline unsigned char *ieee802154_skb_dst_pan(__le16 fc, + const struct sk_buff *skb) +{ + unsigned char *dst_pan; + + switch (ieee802154_daddr_mode(fc)) { + case cpu_to_le16(IEEE802154_FCTL_ADDR_NONE): + dst_pan = NULL; + break; + case cpu_to_le16(IEEE802154_FCTL_DADDR_SHORT): + case cpu_to_le16(IEEE802154_FCTL_DADDR_EXTENDED): + dst_pan = skb_mac_header(skb) + + IEEE802154_FC_LEN + + IEEE802154_SEQ_LEN; + break; + default: + WARN_ONCE(1, "invalid addr mode detected"); + dst_pan = NULL; + break; + } + + return dst_pan; +} + /** * ieee802154_be64_to_le64 - copies and convert be64 to le64 * @le64_dst: le64 destination pointer -- cgit v1.2.3 From 19580cc1ed299c736b56b45c7576b477f185f8f5 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Wed, 6 Jul 2016 23:32:25 +0200 Subject: ieee802154: add ieee802154_skb_src_pan helper This patch adds ieee802154_skb_src_pan function to get the pointer address of the source pan id at skb mac pointer. Signed-off-by: Alexander Aring Signed-off-by: Marcel Holtmann --- include/linux/ieee802154.h | 13 ++++++++++ include/net/mac802154.h | 59 ++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 72 insertions(+) (limited to 'include/linux') diff --git a/include/linux/ieee802154.h b/include/linux/ieee802154.h index 91f4665fea63..ddb890174a0e 100644 --- a/include/linux/ieee802154.h +++ b/include/linux/ieee802154.h @@ -50,6 +50,7 @@ #define IEEE802154_EXTENDED_ADDR_LEN 8 #define IEEE802154_SHORT_ADDR_LEN 2 +#define IEEE802154_PAN_ID_LEN 2 #define IEEE802154_LIFS_PERIOD 40 #define IEEE802154_SIFS_PERIOD 12 @@ -224,12 +225,15 @@ enum { #define IEEE802154_FCTL_SECEN 0x0004 #define IEEE802154_FCTL_INTRA_PAN 0x0040 #define IEEE802154_FCTL_DADDR 0x0c00 +#define IEEE802154_FCTL_SADDR 0xc000 #define IEEE802154_FTYPE_DATA 0x0001 #define IEEE802154_FCTL_ADDR_NONE 0x0000 #define IEEE802154_FCTL_DADDR_SHORT 0x0800 #define IEEE802154_FCTL_DADDR_EXTENDED 0x0c00 +#define IEEE802154_FCTL_SADDR_SHORT 0x8000 +#define IEEE802154_FCTL_SADDR_EXTENDED 0xc000 /* * ieee802154_is_data - check if type is IEEE802154_FTYPE_DATA @@ -277,6 +281,15 @@ static inline __le16 ieee802154_daddr_mode(__le16 fc) return fc & cpu_to_le16(IEEE802154_FCTL_DADDR); } +/* + * ieee802154_saddr_mode - get saddr mode from fc + * @fc: frame control bytes in little-endian byteorder + */ +static inline __le16 ieee802154_saddr_mode(__le16 fc) +{ + return fc & cpu_to_le16(IEEE802154_FCTL_SADDR); +} + /** * ieee802154_is_valid_psdu_len - check if psdu len is valid * available lengths: diff --git a/include/net/mac802154.h b/include/net/mac802154.h index b3f7cd868fe9..ec01b35bc969 100644 --- a/include/net/mac802154.h +++ b/include/net/mac802154.h @@ -286,6 +286,65 @@ static inline unsigned char *ieee802154_skb_dst_pan(__le16 fc, return dst_pan; } +/** + * ieee802154_skb_src_pan - get the pointer to source pan field + * @fc: mac header frame control field + * @skb: skb where the source pan pointer will be get from + */ +static inline unsigned char *ieee802154_skb_src_pan(__le16 fc, + const struct sk_buff *skb) +{ + unsigned char *src_pan; + + switch (ieee802154_saddr_mode(fc)) { + case cpu_to_le16(IEEE802154_FCTL_ADDR_NONE): + src_pan = NULL; + break; + case cpu_to_le16(IEEE802154_FCTL_SADDR_SHORT): + case cpu_to_le16(IEEE802154_FCTL_SADDR_EXTENDED): + /* if intra-pan and source addr mode is non none, + * then source pan id is equal destination pan id. + */ + if (ieee802154_is_intra_pan(fc)) { + src_pan = ieee802154_skb_dst_pan(fc, skb); + break; + } + + switch (ieee802154_daddr_mode(fc)) { + case cpu_to_le16(IEEE802154_FCTL_ADDR_NONE): + src_pan = skb_mac_header(skb) + + IEEE802154_FC_LEN + + IEEE802154_SEQ_LEN; + break; + case cpu_to_le16(IEEE802154_FCTL_DADDR_SHORT): + src_pan = skb_mac_header(skb) + + IEEE802154_FC_LEN + + IEEE802154_SEQ_LEN + + IEEE802154_PAN_ID_LEN + + IEEE802154_SHORT_ADDR_LEN; + break; + case cpu_to_le16(IEEE802154_FCTL_DADDR_EXTENDED): + src_pan = skb_mac_header(skb) + + IEEE802154_FC_LEN + + IEEE802154_SEQ_LEN + + IEEE802154_PAN_ID_LEN + + IEEE802154_EXTENDED_ADDR_LEN; + break; + default: + WARN_ONCE(1, "invalid addr mode detected"); + src_pan = NULL; + break; + } + break; + default: + WARN_ONCE(1, "invalid addr mode detected"); + src_pan = NULL; + break; + } + + return src_pan; +} + /** * ieee802154_be64_to_le64 - copies and convert be64 to le64 * @le64_dst: le64 destination pointer -- cgit v1.2.3 From 3467f0d433016c45d1851f3587d32816b7b2ffb0 Mon Sep 17 00:00:00 2001 From: Martin Blumenstingl Date: Thu, 23 Jun 2016 16:57:09 +0200 Subject: ath9k: Allow configuration of LED polarity in platform data. Some devices running OpenWrt need this and it makes sense to add this to ath9k_platform_data as the next patches will add a devicetree (boolean) property for it as well. Suggested-by: Vittorio Gambaletta Signed-off-by: Martin Blumenstingl Signed-off-by: Kalle Valo --- drivers/net/wireless/ath/ath9k/init.c | 3 +++ include/linux/ath9k_platform.h | 1 + 2 files changed, 4 insertions(+) (limited to 'include/linux') diff --git a/drivers/net/wireless/ath/ath9k/init.c b/drivers/net/wireless/ath/ath9k/init.c index 2ee8624755f7..384929d71696 100644 --- a/drivers/net/wireless/ath/ath9k/init.c +++ b/drivers/net/wireless/ath/ath9k/init.c @@ -527,6 +527,9 @@ static int ath9k_init_soc_platform(struct ath_softc *sc) return ret; } + if (pdata->led_active_high) + ah->config.led_active_high = true; + if (pdata->tx_gain_buffalo) ah->config.tx_gain_buffalo = true; diff --git a/include/linux/ath9k_platform.h b/include/linux/ath9k_platform.h index e66153d60bd5..76860a461ed2 100644 --- a/include/linux/ath9k_platform.h +++ b/include/linux/ath9k_platform.h @@ -40,6 +40,7 @@ struct ath9k_platform_data { bool tx_gain_buffalo; bool disable_2ghz; bool disable_5ghz; + bool led_active_high; int (*get_mac_revision)(void); int (*external_reset)(void); -- cgit v1.2.3 From 29cc6679076a00a6ce193004dcf2d14ae7c428a5 Mon Sep 17 00:00:00 2001 From: Amir Vadai Date: Thu, 14 Jul 2016 10:32:37 +0300 Subject: net/mlx5: Store counters in rbtree instead of list In order to use bulk counters, we need to have counters sorted by id. Signed-off-by: Amir Vadai Reviewed-by: Or Gerlitz Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/fs_core.h | 1 + .../net/ethernet/mellanox/mlx5/core/fs_counters.c | 64 ++++++++++++++++++---- include/linux/mlx5/driver.h | 2 +- 3 files changed, 56 insertions(+), 11 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h index d7ba91a1eea2..9cffb6aeb4e9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h @@ -111,6 +111,7 @@ struct mlx5_fc_cache { }; struct mlx5_fc { + struct rb_node node; struct list_head list; /* last{packets,bytes} members are used when calculating the delta since diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c index 164dc37fda72..aaf8fd149cc6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c @@ -32,6 +32,7 @@ #include #include +#include #include "mlx5_core.h" #include "fs_core.h" #include "fs_cmd.h" @@ -68,6 +69,27 @@ * elapsed, the thread will actually query the hardware. */ +static void mlx5_fc_stats_insert(struct rb_root *root, struct mlx5_fc *counter) +{ + struct rb_node **new = &root->rb_node; + struct rb_node *parent = NULL; + + while (*new) { + struct mlx5_fc *this = container_of(*new, struct mlx5_fc, node); + int result = counter->id - this->id; + + parent = *new; + if (result < 0) + new = &((*new)->rb_left); + else + new = &((*new)->rb_right); + } + + /* Add new node and rebalance tree. */ + rb_link_node(&counter->node, parent, new); + rb_insert_color(&counter->node, root); +} + static void mlx5_fc_stats_work(struct work_struct *work) { struct mlx5_core_dev *dev = container_of(work, struct mlx5_core_dev, @@ -75,25 +97,35 @@ static void mlx5_fc_stats_work(struct work_struct *work) struct mlx5_fc_stats *fc_stats = &dev->priv.fc_stats; unsigned long now = jiffies; struct mlx5_fc *counter; - struct mlx5_fc *tmp; + struct rb_node *node; + LIST_HEAD(tmplist); int err = 0; spin_lock(&fc_stats->addlist_lock); - list_splice_tail_init(&fc_stats->addlist, &fc_stats->list); + list_splice_tail_init(&fc_stats->addlist, &tmplist); - if (!list_empty(&fc_stats->list)) + if (!list_empty(&tmplist) || !RB_EMPTY_ROOT(&fc_stats->counters)) queue_delayed_work(fc_stats->wq, &fc_stats->work, MLX5_FC_STATS_PERIOD); spin_unlock(&fc_stats->addlist_lock); - list_for_each_entry_safe(counter, tmp, &fc_stats->list, list) { - struct mlx5_fc_cache *c = &counter->cache; + list_for_each_entry(counter, &tmplist, list) + mlx5_fc_stats_insert(&fc_stats->counters, counter); + + node = rb_first(&fc_stats->counters); + while (node) { + struct mlx5_fc_cache *c; u64 packets; u64 bytes; + counter = rb_entry(node, struct mlx5_fc, node); + c = &counter->cache; + + node = rb_next(node); + if (counter->deleted) { - list_del(&counter->list); + rb_erase(&counter->node, &fc_stats->counters); mlx5_cmd_fc_free(dev, counter->id); @@ -176,7 +208,7 @@ int mlx5_init_fc_stats(struct mlx5_core_dev *dev) { struct mlx5_fc_stats *fc_stats = &dev->priv.fc_stats; - INIT_LIST_HEAD(&fc_stats->list); + fc_stats->counters = RB_ROOT; INIT_LIST_HEAD(&fc_stats->addlist); spin_lock_init(&fc_stats->addlist_lock); @@ -194,20 +226,32 @@ void mlx5_cleanup_fc_stats(struct mlx5_core_dev *dev) struct mlx5_fc_stats *fc_stats = &dev->priv.fc_stats; struct mlx5_fc *counter; struct mlx5_fc *tmp; + struct rb_node *node; cancel_delayed_work_sync(&dev->priv.fc_stats.work); destroy_workqueue(dev->priv.fc_stats.wq); dev->priv.fc_stats.wq = NULL; - list_splice_tail_init(&fc_stats->addlist, &fc_stats->list); - - list_for_each_entry_safe(counter, tmp, &fc_stats->list, list) { + list_for_each_entry_safe(counter, tmp, &fc_stats->addlist, list) { list_del(&counter->list); mlx5_cmd_fc_free(dev, counter->id); kfree(counter); } + + node = rb_first(&fc_stats->counters); + while (node) { + counter = rb_entry(node, struct mlx5_fc, node); + + node = rb_next(node); + + rb_erase(&counter->node, &fc_stats->counters); + + mlx5_cmd_fc_free(dev, counter->id); + + kfree(counter); + } } void mlx5_fc_query_cached(struct mlx5_fc *counter, diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 81e8396574f4..a041b99fceac 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -469,7 +469,7 @@ struct mlx5_irq_info { }; struct mlx5_fc_stats { - struct list_head list; + struct rb_root counters; struct list_head addlist; /* protect addlist add/splice operations */ spinlock_t addlist_lock; -- cgit v1.2.3 From a351a1b03bf169f77891060be30036ef71cbe618 Mon Sep 17 00:00:00 2001 From: Amir Vadai Date: Thu, 14 Jul 2016 10:32:38 +0300 Subject: net/mlx5: Introduce bulk reading of flow counters This commit utilize the ability of ConnectX-4 to bulk read flow counters. Few bulk counter queries could be done instead of issuing thousands of firmware commands per second to get statistics of all flows set to HW, such as those programmed when we offload tc filters. Counters are stored sorted by hardware id, and queried in blocks (id + number of counters). Due to hardware requirement, start of block and number of counters in a block must be four aligned. Reviewed-by: Or Gerlitz Signed-off-by: Amir Vadai Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c | 67 +++++++++++++++++ drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h | 12 ++++ .../net/ethernet/mellanox/mlx5/core/fs_counters.c | 83 ++++++++++++++++------ include/linux/mlx5/mlx5_ifc.h | 8 ++- 4 files changed, 146 insertions(+), 24 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c index a5bb6b695242..9134010e2921 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c @@ -413,3 +413,70 @@ int mlx5_cmd_fc_query(struct mlx5_core_dev *dev, u16 id, return 0; } + +struct mlx5_cmd_fc_bulk { + u16 id; + int num; + int outlen; + u32 out[0]; +}; + +struct mlx5_cmd_fc_bulk * +mlx5_cmd_fc_bulk_alloc(struct mlx5_core_dev *dev, u16 id, int num) +{ + struct mlx5_cmd_fc_bulk *b; + int outlen = sizeof(*b) + + MLX5_ST_SZ_BYTES(query_flow_counter_out) + + MLX5_ST_SZ_BYTES(traffic_counter) * num; + + b = kzalloc(outlen, GFP_KERNEL); + if (!b) + return NULL; + + b->id = id; + b->num = num; + b->outlen = outlen; + + return b; +} + +void mlx5_cmd_fc_bulk_free(struct mlx5_cmd_fc_bulk *b) +{ + kfree(b); +} + +int +mlx5_cmd_fc_bulk_query(struct mlx5_core_dev *dev, struct mlx5_cmd_fc_bulk *b) +{ + u32 in[MLX5_ST_SZ_DW(query_flow_counter_in)]; + + memset(in, 0, sizeof(in)); + + MLX5_SET(query_flow_counter_in, in, opcode, + MLX5_CMD_OP_QUERY_FLOW_COUNTER); + MLX5_SET(query_flow_counter_in, in, op_mod, 0); + MLX5_SET(query_flow_counter_in, in, flow_counter_id, b->id); + MLX5_SET(query_flow_counter_in, in, num_of_counters, b->num); + + return mlx5_cmd_exec_check_status(dev, in, sizeof(in), + b->out, b->outlen); +} + +void mlx5_cmd_fc_bulk_get(struct mlx5_core_dev *dev, + struct mlx5_cmd_fc_bulk *b, u16 id, + u64 *packets, u64 *bytes) +{ + int index = id - b->id; + void *stats; + + if (index < 0 || index >= b->num) { + mlx5_core_warn(dev, "Flow counter id (0x%x) out of range (0x%x..0x%x). Counter ignored.\n", + id, b->id, b->id + b->num - 1); + return; + } + + stats = MLX5_ADDR_OF(query_flow_counter_out, b->out, + flow_statistics[index]); + *packets = MLX5_GET64(traffic_counter, stats, packets); + *bytes = MLX5_GET64(traffic_counter, stats, octets); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h index fc4f7b83fe0a..158844cef82b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h @@ -76,4 +76,16 @@ int mlx5_cmd_fc_alloc(struct mlx5_core_dev *dev, u16 *id); int mlx5_cmd_fc_free(struct mlx5_core_dev *dev, u16 id); int mlx5_cmd_fc_query(struct mlx5_core_dev *dev, u16 id, u64 *packets, u64 *bytes); + +struct mlx5_cmd_fc_bulk; + +struct mlx5_cmd_fc_bulk * +mlx5_cmd_fc_bulk_alloc(struct mlx5_core_dev *dev, u16 id, int num); +void mlx5_cmd_fc_bulk_free(struct mlx5_cmd_fc_bulk *b); +int +mlx5_cmd_fc_bulk_query(struct mlx5_core_dev *dev, struct mlx5_cmd_fc_bulk *b); +void mlx5_cmd_fc_bulk_get(struct mlx5_core_dev *dev, + struct mlx5_cmd_fc_bulk *b, u16 id, + u64 *packets, u64 *bytes); + #endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c index aaf8fd149cc6..c2877e9de8a1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_counters.c @@ -90,16 +90,66 @@ static void mlx5_fc_stats_insert(struct rb_root *root, struct mlx5_fc *counter) rb_insert_color(&counter->node, root); } +static struct rb_node *mlx5_fc_stats_query(struct mlx5_core_dev *dev, + struct mlx5_fc *first, + u16 last_id) +{ + struct mlx5_cmd_fc_bulk *b; + struct rb_node *node = NULL; + u16 afirst_id; + int num; + int err; + int max_bulk = 1 << MLX5_CAP_GEN(dev, log_max_flow_counter_bulk); + + /* first id must be aligned to 4 when using bulk query */ + afirst_id = first->id & ~0x3; + + /* number of counters to query inc. the last counter */ + num = ALIGN(last_id - afirst_id + 1, 4); + if (num > max_bulk) { + num = max_bulk; + last_id = afirst_id + num - 1; + } + + b = mlx5_cmd_fc_bulk_alloc(dev, afirst_id, num); + if (!b) { + mlx5_core_err(dev, "Error allocating resources for bulk query\n"); + return NULL; + } + + err = mlx5_cmd_fc_bulk_query(dev, b); + if (err) { + mlx5_core_err(dev, "Error doing bulk query: %d\n", err); + goto out; + } + + for (node = &first->node; node; node = rb_next(node)) { + struct mlx5_fc *counter = rb_entry(node, struct mlx5_fc, node); + struct mlx5_fc_cache *c = &counter->cache; + + if (counter->id > last_id) + break; + + mlx5_cmd_fc_bulk_get(dev, b, + counter->id, &c->packets, &c->bytes); + } + +out: + mlx5_cmd_fc_bulk_free(b); + + return node; +} + static void mlx5_fc_stats_work(struct work_struct *work) { struct mlx5_core_dev *dev = container_of(work, struct mlx5_core_dev, priv.fc_stats.work.work); struct mlx5_fc_stats *fc_stats = &dev->priv.fc_stats; unsigned long now = jiffies; - struct mlx5_fc *counter; + struct mlx5_fc *counter = NULL; + struct mlx5_fc *last = NULL; struct rb_node *node; LIST_HEAD(tmplist); - int err = 0; spin_lock(&fc_stats->addlist_lock); @@ -115,12 +165,7 @@ static void mlx5_fc_stats_work(struct work_struct *work) node = rb_first(&fc_stats->counters); while (node) { - struct mlx5_fc_cache *c; - u64 packets; - u64 bytes; - counter = rb_entry(node, struct mlx5_fc, node); - c = &counter->cache; node = rb_next(node); @@ -133,26 +178,20 @@ static void mlx5_fc_stats_work(struct work_struct *work) continue; } - if (time_before(now, fc_stats->next_query)) - continue; + last = counter; + } - err = mlx5_cmd_fc_query(dev, counter->id, &packets, &bytes); - if (err) { - pr_err("Error querying stats for counter id %d\n", - counter->id); - continue; - } + if (time_before(now, fc_stats->next_query) || !last) + return; - if (packets == c->packets) - continue; + node = rb_first(&fc_stats->counters); + while (node) { + counter = rb_entry(node, struct mlx5_fc, node); - c->lastuse = jiffies; - c->packets = packets; - c->bytes = bytes; + node = mlx5_fc_stats_query(dev, counter, last->id); } - if (time_after_eq(now, fc_stats->next_query)) - fc_stats->next_query = now + MLX5_FC_STATS_PERIOD; + fc_stats->next_query = now + MLX5_FC_STATS_PERIOD; } struct mlx5_fc *mlx5_fc_create(struct mlx5_core_dev *dev, bool aging) diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 152421cc6f44..d671e4e8e7db 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -893,7 +893,10 @@ struct mlx5_ifc_cmd_hca_cap_bits { u8 reserved_at_330[0xb]; u8 log_max_xrcd[0x5]; - u8 reserved_at_340[0x20]; + u8 reserved_at_340[0x8]; + u8 log_max_flow_counter_bulk[0x8]; + u8 max_flow_counter[0x10]; + u8 reserved_at_360[0x3]; u8 log_max_rq[0x5]; @@ -980,7 +983,8 @@ struct mlx5_ifc_dest_format_struct_bits { }; struct mlx5_ifc_flow_counter_list_bits { - u8 reserved_at_0[0x10]; + u8 clear[0x1]; + u8 num_of_counters[0xf]; u8 flow_counter_id[0x10]; u8 reserved_at_20[0x20]; -- cgit v1.2.3 From 77501a79cec40eac65c59ee7af3f786c703ead9c Mon Sep 17 00:00:00 2001 From: Philipp Zabel Date: Thu, 14 Jul 2016 16:29:43 +0200 Subject: net: phy: micrel: Add KSZ8041FTL fiber mode support We can't detect the FXEN (fiber mode) bootstrap pin, so configure it via a boolean device tree property "micrel,fiber-mode". If it is enabled, auto-negotiation is not supported. The only available modes are 100base-fx (full duplex and half duplex). Signed-off-by: Philipp Zabel Signed-off-by: David S. Miller --- Documentation/devicetree/bindings/net/micrel.txt | 10 +++++++ drivers/net/phy/micrel.c | 34 ++++++++++++++++++++++-- include/linux/micrel_phy.h | 1 + 3 files changed, 43 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/Documentation/devicetree/bindings/net/micrel.txt b/Documentation/devicetree/bindings/net/micrel.txt index 87496a8c64ab..8d157f0295a5 100644 --- a/Documentation/devicetree/bindings/net/micrel.txt +++ b/Documentation/devicetree/bindings/net/micrel.txt @@ -35,3 +35,13 @@ Optional properties: supported clocks: - KSZ8021, KSZ8031, KSZ8081, KSZ8091: "rmii-ref": The RMII reference input clock. Used to determine the XI input clock. + + - micrel,fiber-mode: If present the PHY is configured to operate in fiber mode + + Some PHYs, such as the KSZ8041FTL variant, support fiber mode, enabled + by the FXEN boot strapping pin. It can't be determined from the PHY + registers whether the PHY is in fiber mode, so this boolean device tree + property can be used to describe it. + + In fiber mode, auto-negotiation is disabled and the PHY can only work in + 100base-fx (full and half duplex) modes. diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c index 5a8fefc25157..059f13b60fe0 100644 --- a/drivers/net/phy/micrel.c +++ b/drivers/net/phy/micrel.c @@ -311,6 +311,36 @@ static int kszphy_config_init(struct phy_device *phydev) return 0; } +static int ksz8041_config_init(struct phy_device *phydev) +{ + struct device_node *of_node = phydev->mdio.dev.of_node; + + /* Limit supported and advertised modes in fiber mode */ + if (of_property_read_bool(of_node, "micrel,fiber-mode")) { + phydev->dev_flags |= MICREL_PHY_FXEN; + phydev->supported &= SUPPORTED_FIBRE | + SUPPORTED_100baseT_Full | + SUPPORTED_100baseT_Half; + phydev->advertising &= ADVERTISED_FIBRE | + ADVERTISED_100baseT_Full | + ADVERTISED_100baseT_Half; + phydev->autoneg = AUTONEG_DISABLE; + } + + return kszphy_config_init(phydev); +} + +static int ksz8041_config_aneg(struct phy_device *phydev) +{ + /* Skip auto-negotiation in fiber mode */ + if (phydev->dev_flags & MICREL_PHY_FXEN) { + phydev->speed = SPEED_100; + return 0; + } + + return genphy_config_aneg(phydev); +} + static int ksz9021_load_values_from_of(struct phy_device *phydev, const struct device_node *of_node, u16 reg, @@ -788,8 +818,8 @@ static struct phy_driver ksphy_driver[] = { .flags = PHY_HAS_MAGICANEG | PHY_HAS_INTERRUPT, .driver_data = &ksz8041_type, .probe = kszphy_probe, - .config_init = kszphy_config_init, - .config_aneg = genphy_config_aneg, + .config_init = ksz8041_config_init, + .config_aneg = ksz8041_config_aneg, .read_status = genphy_read_status, .ack_interrupt = kszphy_ack_interrupt, .config_intr = kszphy_config_intr, diff --git a/include/linux/micrel_phy.h b/include/linux/micrel_phy.h index 2e5b194b9b19..257173e0095e 100644 --- a/include/linux/micrel_phy.h +++ b/include/linux/micrel_phy.h @@ -37,6 +37,7 @@ /* struct phy_device dev_flags definitions */ #define MICREL_PHY_50MHZ_CLK 0x00000001 +#define MICREL_PHY_FXEN 0x00000002 #define MICREL_KSZ9021_EXTREG_CTRL 0xB #define MICREL_KSZ9021_EXTREG_DATA_WRITE 0xC -- cgit v1.2.3 From 7e3f977edd0bd9ea6104156feba95bb5ae9bdd38 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 14 Jul 2016 18:08:03 +0200 Subject: perf, events: add non-linear data support for raw records This patch adds support for non-linear data on raw records. It extends raw records to have one or multiple fragments that will be written linearly into the ring slot, where each fragment can optionally have a custom callback handler to walk and extract complex, possibly non-linear data. If a callback handler is provided for a fragment, then the new __output_custom() will be used instead of __output_copy() for the perf_output_sample() part. perf_prepare_sample() does all the size calculation only once, so perf_output_sample() doesn't need to redo the same work anymore, meaning real_size and padding will be cached in the raw record. The raw record becomes 32 bytes in size without holes; to not increase it further and to avoid doing unnecessary recalculations in fast-path, we can reuse next pointer of the last fragment, idea here is borrowed from ZERO_OR_NULL_PTR(), which should keep the perf_output_sample() path for PERF_SAMPLE_RAW minimal. This facility is needed for BPF's event output helper as a first user that will, in a follow-up, add an additional perf_raw_frag to its perf_raw_record in order to be able to more efficiently dump skb context after a linear head meta data related to it. skbs can be non-linear and thus need a custom output function to dump buffers. Currently, the skb data needs to be copied twice; with the help of __output_custom() this work only needs to be done once. Future users could be things like XDP/BPF programs that work on different context though and would thus also have a different callback function. The few users of raw records are adapted to initialize their frag data from the raw record itself, no change in behavior for them. The code is based upon a PoC diff provided by Peter Zijlstra [1]. [1] http://thread.gmane.org/gmane.linux.network/421294 Suggested-by: Peter Zijlstra Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- arch/s390/kernel/perf_cpum_sf.c | 9 ++++-- arch/x86/events/amd/ibs.c | 8 +++-- include/linux/perf_event.h | 20 ++++++++++++- kernel/events/core.c | 66 ++++++++++++++++++++++++++++------------- kernel/events/internal.h | 16 +++++++--- kernel/trace/bpf_trace.c | 6 ++-- 6 files changed, 93 insertions(+), 32 deletions(-) (limited to 'include/linux') diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c index a8e832166417..92619cce57ed 100644 --- a/arch/s390/kernel/perf_cpum_sf.c +++ b/arch/s390/kernel/perf_cpum_sf.c @@ -979,12 +979,15 @@ static int perf_push_sample(struct perf_event *event, struct sf_raw_sample *sfr) struct pt_regs regs; struct perf_sf_sde_regs *sde_regs; struct perf_sample_data data; - struct perf_raw_record raw; + struct perf_raw_record raw = { + .frag = { + .size = sfr->size, + .data = sfr, + }, + }; /* Setup perf sample */ perf_sample_data_init(&data, 0, event->hw.last_period); - raw.size = sfr->size; - raw.data = sfr; data.raw = &raw; /* Setup pt_regs to look like an CPU-measurement external interrupt diff --git a/arch/x86/events/amd/ibs.c b/arch/x86/events/amd/ibs.c index feb90f6730e8..72dea2f40fc4 100644 --- a/arch/x86/events/amd/ibs.c +++ b/arch/x86/events/amd/ibs.c @@ -655,8 +655,12 @@ fail: } if (event->attr.sample_type & PERF_SAMPLE_RAW) { - raw.size = sizeof(u32) + ibs_data.size; - raw.data = ibs_data.data; + raw = (struct perf_raw_record){ + .frag = { + .size = sizeof(u32) + ibs_data.size, + .data = ibs_data.data, + }, + }; data.raw = &raw; } diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 1a827cecd62f..e79e6c6fed89 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -69,9 +69,22 @@ struct perf_callchain_entry_ctx { bool contexts_maxed; }; +typedef unsigned long (*perf_copy_f)(void *dst, const void *src, + unsigned long len); + +struct perf_raw_frag { + union { + struct perf_raw_frag *next; + unsigned long pad; + }; + perf_copy_f copy; + void *data; + u32 size; +} __packed; + struct perf_raw_record { + struct perf_raw_frag frag; u32 size; - void *data; }; /* @@ -1283,6 +1296,11 @@ extern void perf_restore_debug_store(void); static inline void perf_restore_debug_store(void) { } #endif +static __always_inline bool perf_raw_frag_last(const struct perf_raw_frag *frag) +{ + return frag->pad < sizeof(u64); +} + #define perf_output_put(handle, x) perf_output_copy((handle), &(x), sizeof(x)) /* diff --git a/kernel/events/core.c b/kernel/events/core.c index 9c51ec3f0f44..b1891b6b5c1f 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -5553,16 +5553,26 @@ void perf_output_sample(struct perf_output_handle *handle, } if (sample_type & PERF_SAMPLE_RAW) { - if (data->raw) { - u32 raw_size = data->raw->size; - u32 real_size = round_up(raw_size + sizeof(u32), - sizeof(u64)) - sizeof(u32); - u64 zero = 0; - - perf_output_put(handle, real_size); - __output_copy(handle, data->raw->data, raw_size); - if (real_size - raw_size) - __output_copy(handle, &zero, real_size - raw_size); + struct perf_raw_record *raw = data->raw; + + if (raw) { + struct perf_raw_frag *frag = &raw->frag; + + perf_output_put(handle, raw->size); + do { + if (frag->copy) { + __output_custom(handle, frag->copy, + frag->data, frag->size); + } else { + __output_copy(handle, frag->data, + frag->size); + } + if (perf_raw_frag_last(frag)) + break; + frag = frag->next; + } while (1); + if (frag->pad) + __output_skip(handle, NULL, frag->pad); } else { struct { u32 size; @@ -5687,14 +5697,28 @@ void perf_prepare_sample(struct perf_event_header *header, } if (sample_type & PERF_SAMPLE_RAW) { - int size = sizeof(u32); - - if (data->raw) - size += data->raw->size; - else - size += sizeof(u32); + struct perf_raw_record *raw = data->raw; + int size; + + if (raw) { + struct perf_raw_frag *frag = &raw->frag; + u32 sum = 0; + + do { + sum += frag->size; + if (perf_raw_frag_last(frag)) + break; + frag = frag->next; + } while (1); + + size = round_up(sum + sizeof(u32), sizeof(u64)); + raw->size = size - sizeof(u32); + frag->pad = raw->size - sum; + } else { + size = sizeof(u64); + } - header->size += round_up(size, sizeof(u64)); + header->size += size; } if (sample_type & PERF_SAMPLE_BRANCH_STACK) { @@ -7331,7 +7355,7 @@ static struct pmu perf_swevent = { static int perf_tp_filter_match(struct perf_event *event, struct perf_sample_data *data) { - void *record = data->raw->data; + void *record = data->raw->frag.data; /* only top level events have filters set */ if (event->parent) @@ -7387,8 +7411,10 @@ void perf_tp_event(u16 event_type, u64 count, void *record, int entry_size, struct perf_event *event; struct perf_raw_record raw = { - .size = entry_size, - .data = record, + .frag = { + .size = entry_size, + .data = record, + }, }; perf_sample_data_init(&data, 0, 0); diff --git a/kernel/events/internal.h b/kernel/events/internal.h index 05f9f6d626df..2417eb5512cd 100644 --- a/kernel/events/internal.h +++ b/kernel/events/internal.h @@ -123,10 +123,7 @@ static inline unsigned long perf_aux_size(struct ring_buffer *rb) return rb->aux_nr_pages << PAGE_SHIFT; } -#define DEFINE_OUTPUT_COPY(func_name, memcpy_func) \ -static inline unsigned long \ -func_name(struct perf_output_handle *handle, \ - const void *buf, unsigned long len) \ +#define __DEFINE_OUTPUT_COPY_BODY(memcpy_func) \ { \ unsigned long size, written; \ \ @@ -152,6 +149,17 @@ func_name(struct perf_output_handle *handle, \ return len; \ } +#define DEFINE_OUTPUT_COPY(func_name, memcpy_func) \ +static inline unsigned long \ +func_name(struct perf_output_handle *handle, \ + const void *buf, unsigned long len) \ +__DEFINE_OUTPUT_COPY_BODY(memcpy_func) + +static inline unsigned long +__output_custom(struct perf_output_handle *handle, perf_copy_f copy_func, + const void *buf, unsigned long len) +__DEFINE_OUTPUT_COPY_BODY(copy_func) + static inline unsigned long memcpy_common(void *dst, const void *src, unsigned long n) { diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index 094c716154ed..35ab1b2b041b 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -245,8 +245,10 @@ static u64 bpf_perf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) struct bpf_event_entry *ee; struct perf_event *event; struct perf_raw_record raw = { - .size = size, - .data = data, + .frag = { + .size = size, + .data = data, + }, }; if (unlikely(flags & ~(BPF_F_INDEX_MASK))) -- cgit v1.2.3 From 555c8a8623a3a87b3c990ba30b7fd2e5914e41d2 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 14 Jul 2016 18:08:05 +0200 Subject: bpf: avoid stack copy and use skb ctx for event output This work addresses a couple of issues bpf_skb_event_output() helper currently has: i) We need two copies instead of just a single one for the skb data when it should be part of a sample. The data can be non-linear and thus needs to be extracted via bpf_skb_load_bytes() helper first, and then copied once again into the ring buffer slot. ii) Since bpf_skb_load_bytes() currently needs to be used first, the helper needs to see a constant size on the passed stack buffer to make sure BPF verifier can do sanity checks on it during verification time. Thus, just passing skb->len (or any other non-constant value) wouldn't work, but changing bpf_skb_load_bytes() is also not the proper solution, since the two copies are generally still needed. iii) bpf_skb_load_bytes() is just for rather small buffers like headers, since they need to sit on the limited BPF stack anyway. Instead of working around in bpf_skb_load_bytes(), this work improves the bpf_skb_event_output() helper to address all 3 at once. We can make use of the passed in skb context that we have in the helper anyway, and use some of the reserved flag bits as a length argument. The helper will use the new __output_custom() facility from perf side with bpf_skb_copy() as callback helper to walk and extract the data. It will pass the data for setup to bpf_event_output(), which generates and pushes the raw record with an additional frag part. The linear data used in the first frag of the record serves as programmatically defined meta data passed along with the appended sample. Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/bpf.h | 7 ++++++- include/uapi/linux/bpf.h | 2 ++ kernel/bpf/core.c | 6 ++++-- kernel/trace/bpf_trace.c | 33 +++++++++++++++------------------ net/core/filter.c | 43 ++++++++++++++++++++++++++++++++++++++++++- 5 files changed, 69 insertions(+), 22 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index b3336b4f5d04..c13e92b00bf5 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -209,7 +209,12 @@ u64 bpf_get_stackid(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5); bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *fp); const struct bpf_func_proto *bpf_get_trace_printk_proto(void); -const struct bpf_func_proto *bpf_get_event_output_proto(void); + +typedef unsigned long (*bpf_ctx_copy_t)(void *dst, const void *src, + unsigned long len); + +u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, + void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy); #ifdef CONFIG_BPF_SYSCALL DECLARE_PER_CPU(int, bpf_prog_active); diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index 262a7e883b19..c4d922439d20 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -401,6 +401,8 @@ enum bpf_func_id { /* BPF_FUNC_perf_event_output and BPF_FUNC_perf_event_read flags. */ #define BPF_F_INDEX_MASK 0xffffffffULL #define BPF_F_CURRENT_CPU BPF_F_INDEX_MASK +/* BPF_FUNC_perf_event_output for sk_buff input context. */ +#define BPF_F_CTXLEN_MASK (0xfffffULL << 32) /* user accessible mirror of in-kernel sk_buff. * new fields can only be added to the end of this structure diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index d638062f66d6..03fd23d4d587 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -1054,9 +1054,11 @@ const struct bpf_func_proto * __weak bpf_get_trace_printk_proto(void) return NULL; } -const struct bpf_func_proto * __weak bpf_get_event_output_proto(void) +u64 __weak +bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, + void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy) { - return NULL; + return -ENOTSUPP; } /* Always built-in helper functions. */ diff --git a/kernel/trace/bpf_trace.c b/kernel/trace/bpf_trace.c index c35883a9bc11..ebfbb7dd7033 100644 --- a/kernel/trace/bpf_trace.c +++ b/kernel/trace/bpf_trace.c @@ -298,29 +298,26 @@ static const struct bpf_func_proto bpf_perf_event_output_proto = { static DEFINE_PER_CPU(struct pt_regs, bpf_pt_regs); -static u64 bpf_event_output(u64 r1, u64 r2, u64 flags, u64 r4, u64 size) +u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, + void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy) { struct pt_regs *regs = this_cpu_ptr(&bpf_pt_regs); + struct perf_raw_frag frag = { + .copy = ctx_copy, + .size = ctx_size, + .data = ctx, + }; + struct perf_raw_record raw = { + .frag = { + .next = ctx_size ? &frag : NULL, + .size = meta_size, + .data = meta, + }, + }; perf_fetch_caller_regs(regs); - return bpf_perf_event_output((long)regs, r2, flags, r4, size); -} - -static const struct bpf_func_proto bpf_event_output_proto = { - .func = bpf_event_output, - .gpl_only = true, - .ret_type = RET_INTEGER, - .arg1_type = ARG_PTR_TO_CTX, - .arg2_type = ARG_CONST_MAP_PTR, - .arg3_type = ARG_ANYTHING, - .arg4_type = ARG_PTR_TO_STACK, - .arg5_type = ARG_CONST_STACK_SIZE, -}; - -const struct bpf_func_proto *bpf_get_event_output_proto(void) -{ - return &bpf_event_output_proto; + return __bpf_perf_event_output(regs, map, flags, &raw); } static u64 bpf_get_current_task(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5) diff --git a/net/core/filter.c b/net/core/filter.c index 10c4a2f9e8bb..22e3992c8b48 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2025,6 +2025,47 @@ bool bpf_helper_changes_skb_data(void *func) return false; } +static unsigned long bpf_skb_copy(void *dst_buff, const void *skb, + unsigned long len) +{ + void *ptr = skb_header_pointer(skb, 0, len, dst_buff); + + if (unlikely(!ptr)) + return len; + if (ptr != dst_buff) + memcpy(dst_buff, ptr, len); + + return 0; +} + +static u64 bpf_skb_event_output(u64 r1, u64 r2, u64 flags, u64 r4, + u64 meta_size) +{ + struct sk_buff *skb = (struct sk_buff *)(long) r1; + struct bpf_map *map = (struct bpf_map *)(long) r2; + u64 skb_size = (flags & BPF_F_CTXLEN_MASK) >> 32; + void *meta = (void *)(long) r4; + + if (unlikely(flags & ~(BPF_F_CTXLEN_MASK | BPF_F_INDEX_MASK))) + return -EINVAL; + if (unlikely(skb_size > skb->len)) + return -EFAULT; + + return bpf_event_output(map, flags, meta, meta_size, skb, skb_size, + bpf_skb_copy); +} + +static const struct bpf_func_proto bpf_skb_event_output_proto = { + .func = bpf_skb_event_output, + .gpl_only = true, + .ret_type = RET_INTEGER, + .arg1_type = ARG_PTR_TO_CTX, + .arg2_type = ARG_CONST_MAP_PTR, + .arg3_type = ARG_ANYTHING, + .arg4_type = ARG_PTR_TO_STACK, + .arg5_type = ARG_CONST_STACK_SIZE, +}; + static unsigned short bpf_tunnel_key_af(u64 flags) { return flags & BPF_F_TUNINFO_IPV6 ? AF_INET6 : AF_INET; @@ -2357,7 +2398,7 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) case BPF_FUNC_get_hash_recalc: return &bpf_get_hash_recalc_proto; case BPF_FUNC_perf_event_output: - return bpf_get_event_output_proto(); + return &bpf_skb_event_output_proto; case BPF_FUNC_get_smp_processor_id: return &bpf_get_smp_processor_id_proto; #ifdef CONFIG_SOCK_CGROUP_DATA -- cgit v1.2.3 From 43b9e127406079d187794a5140a2411fbc6df2df Mon Sep 17 00:00:00 2001 From: Nikolay Aleksandrov Date: Thu, 14 Jul 2016 19:28:27 +0300 Subject: net: ipmr/ip6mr: add support for keeping an entry age In preparation for hardware offloading of ipmr/ip6mr we need an interface that allows to check (and later update) the age of entries. Relying on stats alone can show activity but not actual age of the entry, furthermore when there're tens of thousands of entries a lot of the hardware implementations only support "hit" bits which are cleared on read to denote that the entry was active and shouldn't be aged out, these can then be naturally translated into age timestamp and will be compatible with the software forwarding age. Using a lastuse entry doesn't affect performance because the members in that cache line are written to along with the age. Since all new users are encouraged to use ipmr via netlink, this is exported via the RTA_EXPIRES attribute. Also do a minor local variable declaration style adjustment - arrange them longest to shortest. Signed-off-by: Nikolay Aleksandrov CC: Roopa Prabhu CC: Shrijeet Mukherjee CC: Satish Ashok CC: Donald Sharp CC: David S. Miller CC: Alexey Kuznetsov CC: James Morris CC: Hideaki YOSHIFUJI CC: Patrick McHardy Signed-off-by: David S. Miller --- include/linux/mroute.h | 1 + include/linux/mroute6.h | 1 + net/ipv4/ipmr.c | 13 +++++++++---- net/ipv6/ip6mr.c | 13 +++++++++---- 4 files changed, 20 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/mroute.h b/include/linux/mroute.h index bf9b322cb0b0..d351fd3e1049 100644 --- a/include/linux/mroute.h +++ b/include/linux/mroute.h @@ -104,6 +104,7 @@ struct mfc_cache { unsigned long bytes; unsigned long pkt; unsigned long wrong_if; + unsigned long lastuse; unsigned char ttls[MAXVIFS]; /* TTL thresholds */ } res; } mfc_un; diff --git a/include/linux/mroute6.h b/include/linux/mroute6.h index 66982e764051..3987b64040c5 100644 --- a/include/linux/mroute6.h +++ b/include/linux/mroute6.h @@ -92,6 +92,7 @@ struct mfc6_cache { unsigned long bytes; unsigned long pkt; unsigned long wrong_if; + unsigned long lastuse; unsigned char ttls[MAXMIFS]; /* TTL thresholds */ } res; } mfc_un; diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index 5ad48ec77710..e0d76f5f0113 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1150,6 +1150,7 @@ static int ipmr_mfc_add(struct net *net, struct mr_table *mrt, c->mfc_origin = mfc->mfcc_origin.s_addr; c->mfc_mcastgrp = mfc->mfcc_mcastgrp.s_addr; c->mfc_parent = mfc->mfcc_parent; + c->mfc_un.res.lastuse = jiffies; ipmr_update_thresholds(mrt, c, mfc->mfcc_ttls); if (!mrtsock) c->mfc_flags |= MFC_STATIC; @@ -1792,6 +1793,7 @@ static void ip_mr_forward(struct net *net, struct mr_table *mrt, vif = cache->mfc_parent; cache->mfc_un.res.pkt++; cache->mfc_un.res.bytes += skb->len; + cache->mfc_un.res.lastuse = jiffies; if (cache->mfc_origin == htonl(INADDR_ANY) && true_vifi >= 0) { struct mfc_cache *cache_proxy; @@ -2071,10 +2073,10 @@ drop: static int __ipmr_fill_mroute(struct mr_table *mrt, struct sk_buff *skb, struct mfc_cache *c, struct rtmsg *rtm) { - int ct; - struct rtnexthop *nhp; - struct nlattr *mp_attr; struct rta_mfc_stats mfcs; + struct nlattr *mp_attr; + struct rtnexthop *nhp; + int ct; /* If cache is unresolved, don't try to parse IIF and OIF */ if (c->mfc_parent >= MAXVIFS) @@ -2106,7 +2108,10 @@ static int __ipmr_fill_mroute(struct mr_table *mrt, struct sk_buff *skb, mfcs.mfcs_packets = c->mfc_un.res.pkt; mfcs.mfcs_bytes = c->mfc_un.res.bytes; mfcs.mfcs_wrong_if = c->mfc_un.res.wrong_if; - if (nla_put_64bit(skb, RTA_MFC_STATS, sizeof(mfcs), &mfcs, RTA_PAD) < 0) + if (nla_put_64bit(skb, RTA_MFC_STATS, sizeof(mfcs), &mfcs, RTA_PAD) || + nla_put_u64_64bit(skb, RTA_EXPIRES, + jiffies_to_clock_t(c->mfc_un.res.lastuse), + RTA_PAD)) return -EMSGSIZE; rtm->rtm_type = RTN_MULTICAST; diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index c7ca0f5d1a3b..7adce139d92a 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -1500,6 +1500,7 @@ static int ip6mr_mfc_add(struct net *net, struct mr6_table *mrt, c->mf6c_origin = mfc->mf6cc_origin.sin6_addr; c->mf6c_mcastgrp = mfc->mf6cc_mcastgrp.sin6_addr; c->mf6c_parent = mfc->mf6cc_parent; + c->mfc_un.res.lastuse = jiffies; ip6mr_update_thresholds(mrt, c, ttls); if (!mrtsock) c->mfc_flags |= MFC_STATIC; @@ -2092,6 +2093,7 @@ static void ip6_mr_forward(struct net *net, struct mr6_table *mrt, vif = cache->mf6c_parent; cache->mfc_un.res.pkt++; cache->mfc_un.res.bytes += skb->len; + cache->mfc_un.res.lastuse = jiffies; if (ipv6_addr_any(&cache->mf6c_origin) && true_vifi >= 0) { struct mfc6_cache *cache_proxy; @@ -2234,10 +2236,10 @@ int ip6_mr_input(struct sk_buff *skb) static int __ip6mr_fill_mroute(struct mr6_table *mrt, struct sk_buff *skb, struct mfc6_cache *c, struct rtmsg *rtm) { - int ct; - struct rtnexthop *nhp; - struct nlattr *mp_attr; struct rta_mfc_stats mfcs; + struct nlattr *mp_attr; + struct rtnexthop *nhp; + int ct; /* If cache is unresolved, don't try to parse IIF and OIF */ if (c->mf6c_parent >= MAXMIFS) @@ -2270,7 +2272,10 @@ static int __ip6mr_fill_mroute(struct mr6_table *mrt, struct sk_buff *skb, mfcs.mfcs_packets = c->mfc_un.res.pkt; mfcs.mfcs_bytes = c->mfc_un.res.bytes; mfcs.mfcs_wrong_if = c->mfc_un.res.wrong_if; - if (nla_put_64bit(skb, RTA_MFC_STATS, sizeof(mfcs), &mfcs, RTA_PAD) < 0) + if (nla_put_64bit(skb, RTA_MFC_STATS, sizeof(mfcs), &mfcs, RTA_PAD) || + nla_put_u64_64bit(skb, RTA_EXPIRES, + jiffies_to_clock_t(c->mfc_un.res.lastuse), + RTA_PAD)) return -EMSGSIZE; rtm->rtm_type = RTN_MULTICAST; -- cgit v1.2.3 From b786241253041c13d94309ca4dace301833f63d1 Mon Sep 17 00:00:00 2001 From: Dongpo Li Date: Fri, 15 Jul 2016 16:26:34 +0800 Subject: of_mdio: Abstract a general interface for phy connect Abstract a general interface "of_phy_get_and_connect" for PHY connect. User will have no bother with getting "phy-mode" and "phy-handle" any more. Suggested-by: Arnd Bergmann Signed-off-by: Dongpo Li Reviewed-by: Jiancheng Xue Signed-off-by: David S. Miller --- drivers/of/of_mdio.c | 36 ++++++++++++++++++++++++++++++++++++ include/linux/of_mdio.h | 10 ++++++++++ 2 files changed, 46 insertions(+) (limited to 'include/linux') diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c index e2b50bc12f23..b470f7e3521d 100644 --- a/drivers/of/of_mdio.c +++ b/drivers/of/of_mdio.c @@ -19,6 +19,7 @@ #include #include #include +#include #include MODULE_AUTHOR("Grant Likely "); @@ -331,6 +332,41 @@ struct phy_device *of_phy_connect(struct net_device *dev, } EXPORT_SYMBOL(of_phy_connect); +/** + * of_phy_get_and_connect + * - Get phy node and connect to the phy described in the device tree + * @dev: pointer to net_device claiming the phy + * @np: Pointer to device tree node for the net_device claiming the phy + * @hndlr: Link state callback for the network device + * + * If successful, returns a pointer to the phy_device with the embedded + * struct device refcount incremented by one, or NULL on failure. The + * refcount must be dropped by calling phy_disconnect() or phy_detach(). + */ +struct phy_device *of_phy_get_and_connect(struct net_device *dev, + struct device_node *np, + void (*hndlr)(struct net_device *)) +{ + phy_interface_t iface; + struct device_node *phy_np; + struct phy_device *phy; + + iface = of_get_phy_mode(np); + if (iface < 0) + return NULL; + + phy_np = of_parse_phandle(np, "phy-handle", 0); + if (!phy_np) + return NULL; + + phy = of_phy_connect(dev, phy_np, hndlr, 0, iface); + + of_node_put(phy_np); + + return phy; +} +EXPORT_SYMBOL(of_phy_get_and_connect); + /** * of_phy_attach - Attach to a PHY without starting the state machine * @dev: pointer to net_device claiming the phy diff --git a/include/linux/of_mdio.h b/include/linux/of_mdio.h index 4b04587d0441..2ab233661ae5 100644 --- a/include/linux/of_mdio.h +++ b/include/linux/of_mdio.h @@ -19,6 +19,9 @@ extern struct phy_device *of_phy_connect(struct net_device *dev, struct device_node *phy_np, void (*hndlr)(struct net_device *), u32 flags, phy_interface_t iface); +extern struct phy_device * +of_phy_get_and_connect(struct net_device *dev, struct device_node *np, + void (*hndlr)(struct net_device *)); struct phy_device *of_phy_attach(struct net_device *dev, struct device_node *phy_np, u32 flags, phy_interface_t iface); @@ -52,6 +55,13 @@ static inline struct phy_device *of_phy_connect(struct net_device *dev, return NULL; } +static inline struct phy_device * +of_phy_get_and_connect(struct net_device *dev, struct device_node *np, + void (*hndlr)(struct net_device *)) +{ + return NULL; +} + static inline struct phy_device *of_phy_attach(struct net_device *dev, struct device_node *phy_np, u32 flags, phy_interface_t iface) -- cgit v1.2.3 From 4360fa22ad5b48a1d1e10e31ffb383ed8c977435 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 6 Jun 2016 11:02:03 +0200 Subject: drivers: misc: ti-st: Use int instead of fuzzy char for callback status On mips and parisc: drivers/bluetooth/btwilink.c: In function 'ti_st_open': drivers/bluetooth/btwilink.c:174:21: warning: overflow in implicit constant conversion [-Woverflow] hst->reg_status = -EINPROGRESS; drivers/nfc/nfcwilink.c: In function 'nfcwilink_open': drivers/nfc/nfcwilink.c:396:31: warning: overflow in implicit constant conversion [-Woverflow] drv->st_register_cb_status = -EINPROGRESS; There are actually two issues: 1. Whether "char" is signed or unsigned depends on the architecture. As the completion callback data is used to pass a (negative) error code, it should always be signed. 2. EINPROGRESS is 150 on mips, 245 on parisc. Hence -EINPROGRESS doesn't fit in a signed 8-bit number. Change the callback status from "char" to "int" to fix these. Signed-off-by: Geert Uytterhoeven Acked-by: Mauro Carvalho Chehab Acked-by: Samuel Ortiz Signed-off-by: Marcel Holtmann --- drivers/bluetooth/btwilink.c | 4 ++-- drivers/media/radio/wl128x/fmdrv_common.c | 2 +- drivers/misc/ti-st/st_core.c | 2 +- drivers/nfc/nfcwilink.c | 4 ++-- include/linux/ti_wilink_st.h | 2 +- 5 files changed, 7 insertions(+), 7 deletions(-) (limited to 'include/linux') diff --git a/drivers/bluetooth/btwilink.c b/drivers/bluetooth/btwilink.c index 24a652f9252b..485281b3f167 100644 --- a/drivers/bluetooth/btwilink.c +++ b/drivers/bluetooth/btwilink.c @@ -51,7 +51,7 @@ */ struct ti_st { struct hci_dev *hdev; - char reg_status; + int reg_status; long (*st_write) (struct sk_buff *); struct completion wait_reg_completion; }; @@ -83,7 +83,7 @@ static inline void ti_st_tx_complete(struct ti_st *hst, int pkt_type) * status.ti_st_open() function will wait for signal from this * API when st_register() function returns ST_PENDING. */ -static void st_reg_completion_cb(void *priv_data, char data) +static void st_reg_completion_cb(void *priv_data, int data) { struct ti_st *lhst = priv_data; diff --git a/drivers/media/radio/wl128x/fmdrv_common.c b/drivers/media/radio/wl128x/fmdrv_common.c index 3f9e6df7d837..642b89c66bcb 100644 --- a/drivers/media/radio/wl128x/fmdrv_common.c +++ b/drivers/media/radio/wl128x/fmdrv_common.c @@ -1472,7 +1472,7 @@ static long fm_st_receive(void *arg, struct sk_buff *skb) * Called by ST layer to indicate protocol registration completion * status. */ -static void fm_st_reg_comp_cb(void *arg, char data) +static void fm_st_reg_comp_cb(void *arg, int data) { struct fmdev *fmdev; diff --git a/drivers/misc/ti-st/st_core.c b/drivers/misc/ti-st/st_core.c index dcdbd58672cc..00051590e00f 100644 --- a/drivers/misc/ti-st/st_core.c +++ b/drivers/misc/ti-st/st_core.c @@ -141,7 +141,7 @@ static void st_send_frame(unsigned char chnl_id, struct st_data_s *st_gdata) * This function is being called with spin lock held, protocol drivers are * only expected to complete their waits and do nothing more than that. */ -static void st_reg_complete(struct st_data_s *st_gdata, char err) +static void st_reg_complete(struct st_data_s *st_gdata, int err) { unsigned char i = 0; pr_info(" %s ", __func__); diff --git a/drivers/nfc/nfcwilink.c b/drivers/nfc/nfcwilink.c index f81e500e7650..3fbd18b8e473 100644 --- a/drivers/nfc/nfcwilink.c +++ b/drivers/nfc/nfcwilink.c @@ -94,7 +94,7 @@ struct nfcwilink { struct nci_dev *ndev; unsigned long flags; - char st_register_cb_status; + int st_register_cb_status; long (*st_write) (struct sk_buff *); struct completion completed; @@ -320,7 +320,7 @@ exit: } /* Called by ST when registration is complete */ -static void nfcwilink_register_complete(void *priv_data, char data) +static void nfcwilink_register_complete(void *priv_data, int data) { struct nfcwilink *drv = priv_data; diff --git a/include/linux/ti_wilink_st.h b/include/linux/ti_wilink_st.h index 0a0d56834c8e..f2293028ab9d 100644 --- a/include/linux/ti_wilink_st.h +++ b/include/linux/ti_wilink_st.h @@ -71,7 +71,7 @@ struct st_proto_s { enum proto_type type; long (*recv) (void *, struct sk_buff *); unsigned char (*match_packet) (const unsigned char *data); - void (*reg_complete_cb) (void *, char data); + void (*reg_complete_cb) (void *, int data); long (*write) (struct sk_buff *skb); void *priv_data; -- cgit v1.2.3 From f4dc77713f8016d2e8a3295e1c9c53a21f296def Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 14 Jul 2016 17:51:26 +0200 Subject: netfilter: x_tables: speed up jump target validation The dummy ruleset I used to test the original validation change was broken, most rules were unreachable and were not tested by mark_source_chains(). In some cases rulesets that used to load in a few seconds now require several minutes. sample ruleset that shows the behaviour: echo "*filter" for i in $(seq 0 100000);do printf ":chain_%06x - [0:0]\n" $i done for i in $(seq 0 100000);do printf -- "-A INPUT -j chain_%06x\n" $i printf -- "-A INPUT -j chain_%06x\n" $i printf -- "-A INPUT -j chain_%06x\n" $i done echo COMMIT [ pipe result into iptables-restore ] This ruleset will be about 74mbyte in size, with ~500k searches though all 500k[1] rule entries. iptables-restore will take forever (gave up after 10 minutes) Instead of always searching the entire blob for a match, fill an array with the start offsets of every single ipt_entry struct, then do a binary search to check if the jump target is present or not. After this change ruleset restore times get again close to what one gets when reverting 36472341017529e (~3 seconds on my workstation). [1] every user-defined rule gets an implicit RETURN, so we get 300k jumps + 100k userchains + 100k returns -> 500k rule entries Fixes: 36472341017529e ("netfilter: x_tables: validate targets of jumps") Reported-by: Jeff Wu Tested-by: Jeff Wu Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/linux/netfilter/x_tables.h | 4 +++ net/ipv4/netfilter/arp_tables.c | 47 ++++++++++++++++++----------------- net/ipv4/netfilter/ip_tables.c | 45 ++++++++++++++++++---------------- net/ipv6/netfilter/ip6_tables.c | 45 ++++++++++++++++++---------------- net/netfilter/x_tables.c | 50 ++++++++++++++++++++++++++++++++++++++ 5 files changed, 127 insertions(+), 64 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h index e94e81ab2b58..2ad1a2b289b5 100644 --- a/include/linux/netfilter/x_tables.h +++ b/include/linux/netfilter/x_tables.h @@ -250,6 +250,10 @@ int xt_check_entry_offsets(const void *base, const char *elems, unsigned int target_offset, unsigned int next_offset); +unsigned int *xt_alloc_entry_offsets(unsigned int size); +bool xt_find_jump_offset(const unsigned int *offsets, + unsigned int target, unsigned int size); + int xt_check_match(struct xt_mtchk_param *, unsigned int size, u_int8_t proto, bool inv_proto); int xt_check_target(struct xt_tgchk_param *, unsigned int size, u_int8_t proto, diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index c8dd9e26b185..b31df597fd37 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -299,23 +299,12 @@ static inline bool unconditional(const struct arpt_entry *e) memcmp(&e->arp, &uncond, sizeof(uncond)) == 0; } -static bool find_jump_target(const struct xt_table_info *t, - const struct arpt_entry *target) -{ - struct arpt_entry *iter; - - xt_entry_foreach(iter, t->entries, t->size) { - if (iter == target) - return true; - } - return false; -} - /* Figures out from what hook each rule can be called: returns 0 if * there are loops. Puts hook bitmask in comefrom. */ static int mark_source_chains(const struct xt_table_info *newinfo, - unsigned int valid_hooks, void *entry0) + unsigned int valid_hooks, void *entry0, + unsigned int *offsets) { unsigned int hook; @@ -388,10 +377,11 @@ static int mark_source_chains(const struct xt_table_info *newinfo, XT_STANDARD_TARGET) == 0 && newpos >= 0) { /* This a jump; chase it. */ + if (!xt_find_jump_offset(offsets, newpos, + newinfo->number)) + return 0; e = (struct arpt_entry *) (entry0 + newpos); - if (!find_jump_target(newinfo, e)) - return 0; } else { /* ... this is a fallthru */ newpos = pos + e->next_offset; @@ -543,6 +533,7 @@ static int translate_table(struct xt_table_info *newinfo, void *entry0, const struct arpt_replace *repl) { struct arpt_entry *iter; + unsigned int *offsets; unsigned int i; int ret = 0; @@ -555,6 +546,9 @@ static int translate_table(struct xt_table_info *newinfo, void *entry0, newinfo->underflow[i] = 0xFFFFFFFF; } + offsets = xt_alloc_entry_offsets(newinfo->number); + if (!offsets) + return -ENOMEM; i = 0; /* Walk through entries, checking offsets. */ @@ -565,17 +559,20 @@ static int translate_table(struct xt_table_info *newinfo, void *entry0, repl->underflow, repl->valid_hooks); if (ret != 0) - break; + goto out_free; + if (i < repl->num_entries) + offsets[i] = (void *)iter - entry0; ++i; if (strcmp(arpt_get_target(iter)->u.user.name, XT_ERROR_TARGET) == 0) ++newinfo->stacksize; } if (ret != 0) - return ret; + goto out_free; + ret = -EINVAL; if (i != repl->num_entries) - return -EINVAL; + goto out_free; /* Check hooks all assigned */ for (i = 0; i < NF_ARP_NUMHOOKS; i++) { @@ -583,13 +580,16 @@ static int translate_table(struct xt_table_info *newinfo, void *entry0, if (!(repl->valid_hooks & (1 << i))) continue; if (newinfo->hook_entry[i] == 0xFFFFFFFF) - return -EINVAL; + goto out_free; if (newinfo->underflow[i] == 0xFFFFFFFF) - return -EINVAL; + goto out_free; } - if (!mark_source_chains(newinfo, repl->valid_hooks, entry0)) - return -ELOOP; + if (!mark_source_chains(newinfo, repl->valid_hooks, entry0, offsets)) { + ret = -ELOOP; + goto out_free; + } + kvfree(offsets); /* Finally, each sanity check must pass */ i = 0; @@ -609,6 +609,9 @@ static int translate_table(struct xt_table_info *newinfo, void *entry0, return ret; } + return ret; + out_free: + kvfree(offsets); return ret; } diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index f0df66f54ce6..f993545a3373 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -373,23 +373,12 @@ ipt_do_table(struct sk_buff *skb, else return verdict; } -static bool find_jump_target(const struct xt_table_info *t, - const struct ipt_entry *target) -{ - struct ipt_entry *iter; - - xt_entry_foreach(iter, t->entries, t->size) { - if (iter == target) - return true; - } - return false; -} - /* Figures out from what hook each rule can be called: returns 0 if there are loops. Puts hook bitmask in comefrom. */ static int mark_source_chains(const struct xt_table_info *newinfo, - unsigned int valid_hooks, void *entry0) + unsigned int valid_hooks, void *entry0, + unsigned int *offsets) { unsigned int hook; @@ -458,10 +447,11 @@ mark_source_chains(const struct xt_table_info *newinfo, XT_STANDARD_TARGET) == 0 && newpos >= 0) { /* This a jump; chase it. */ + if (!xt_find_jump_offset(offsets, newpos, + newinfo->number)) + return 0; e = (struct ipt_entry *) (entry0 + newpos); - if (!find_jump_target(newinfo, e)) - return 0; } else { /* ... this is a fallthru */ newpos = pos + e->next_offset; @@ -694,6 +684,7 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, const struct ipt_replace *repl) { struct ipt_entry *iter; + unsigned int *offsets; unsigned int i; int ret = 0; @@ -706,6 +697,9 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, newinfo->underflow[i] = 0xFFFFFFFF; } + offsets = xt_alloc_entry_offsets(newinfo->number); + if (!offsets) + return -ENOMEM; i = 0; /* Walk through entries, checking offsets. */ xt_entry_foreach(iter, entry0, newinfo->size) { @@ -715,15 +709,18 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, repl->underflow, repl->valid_hooks); if (ret != 0) - return ret; + goto out_free; + if (i < repl->num_entries) + offsets[i] = (void *)iter - entry0; ++i; if (strcmp(ipt_get_target(iter)->u.user.name, XT_ERROR_TARGET) == 0) ++newinfo->stacksize; } + ret = -EINVAL; if (i != repl->num_entries) - return -EINVAL; + goto out_free; /* Check hooks all assigned */ for (i = 0; i < NF_INET_NUMHOOKS; i++) { @@ -731,13 +728,16 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, if (!(repl->valid_hooks & (1 << i))) continue; if (newinfo->hook_entry[i] == 0xFFFFFFFF) - return -EINVAL; + goto out_free; if (newinfo->underflow[i] == 0xFFFFFFFF) - return -EINVAL; + goto out_free; } - if (!mark_source_chains(newinfo, repl->valid_hooks, entry0)) - return -ELOOP; + if (!mark_source_chains(newinfo, repl->valid_hooks, entry0, offsets)) { + ret = -ELOOP; + goto out_free; + } + kvfree(offsets); /* Finally, each sanity check must pass */ i = 0; @@ -757,6 +757,9 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, return ret; } + return ret; + out_free: + kvfree(offsets); return ret; } diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c index 61ed95054efa..552fac2f390a 100644 --- a/net/ipv6/netfilter/ip6_tables.c +++ b/net/ipv6/netfilter/ip6_tables.c @@ -402,23 +402,12 @@ ip6t_do_table(struct sk_buff *skb, else return verdict; } -static bool find_jump_target(const struct xt_table_info *t, - const struct ip6t_entry *target) -{ - struct ip6t_entry *iter; - - xt_entry_foreach(iter, t->entries, t->size) { - if (iter == target) - return true; - } - return false; -} - /* Figures out from what hook each rule can be called: returns 0 if there are loops. Puts hook bitmask in comefrom. */ static int mark_source_chains(const struct xt_table_info *newinfo, - unsigned int valid_hooks, void *entry0) + unsigned int valid_hooks, void *entry0, + unsigned int *offsets) { unsigned int hook; @@ -487,10 +476,11 @@ mark_source_chains(const struct xt_table_info *newinfo, XT_STANDARD_TARGET) == 0 && newpos >= 0) { /* This a jump; chase it. */ + if (!xt_find_jump_offset(offsets, newpos, + newinfo->number)) + return 0; e = (struct ip6t_entry *) (entry0 + newpos); - if (!find_jump_target(newinfo, e)) - return 0; } else { /* ... this is a fallthru */ newpos = pos + e->next_offset; @@ -724,6 +714,7 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, const struct ip6t_replace *repl) { struct ip6t_entry *iter; + unsigned int *offsets; unsigned int i; int ret = 0; @@ -736,6 +727,9 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, newinfo->underflow[i] = 0xFFFFFFFF; } + offsets = xt_alloc_entry_offsets(newinfo->number); + if (!offsets) + return -ENOMEM; i = 0; /* Walk through entries, checking offsets. */ xt_entry_foreach(iter, entry0, newinfo->size) { @@ -745,15 +739,18 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, repl->underflow, repl->valid_hooks); if (ret != 0) - return ret; + goto out_free; + if (i < repl->num_entries) + offsets[i] = (void *)iter - entry0; ++i; if (strcmp(ip6t_get_target(iter)->u.user.name, XT_ERROR_TARGET) == 0) ++newinfo->stacksize; } + ret = -EINVAL; if (i != repl->num_entries) - return -EINVAL; + goto out_free; /* Check hooks all assigned */ for (i = 0; i < NF_INET_NUMHOOKS; i++) { @@ -761,13 +758,16 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, if (!(repl->valid_hooks & (1 << i))) continue; if (newinfo->hook_entry[i] == 0xFFFFFFFF) - return -EINVAL; + goto out_free; if (newinfo->underflow[i] == 0xFFFFFFFF) - return -EINVAL; + goto out_free; } - if (!mark_source_chains(newinfo, repl->valid_hooks, entry0)) - return -ELOOP; + if (!mark_source_chains(newinfo, repl->valid_hooks, entry0, offsets)) { + ret = -ELOOP; + goto out_free; + } + kvfree(offsets); /* Finally, each sanity check must pass */ i = 0; @@ -787,6 +787,9 @@ translate_table(struct net *net, struct xt_table_info *newinfo, void *entry0, return ret; } + return ret; + out_free: + kvfree(offsets); return ret; } diff --git a/net/netfilter/x_tables.c b/net/netfilter/x_tables.c index fe0e2db632c7..e0aa7c1d0224 100644 --- a/net/netfilter/x_tables.c +++ b/net/netfilter/x_tables.c @@ -702,6 +702,56 @@ int xt_check_entry_offsets(const void *base, } EXPORT_SYMBOL(xt_check_entry_offsets); +/** + * xt_alloc_entry_offsets - allocate array to store rule head offsets + * + * @size: number of entries + * + * Return: NULL or kmalloc'd or vmalloc'd array + */ +unsigned int *xt_alloc_entry_offsets(unsigned int size) +{ + unsigned int *off; + + off = kcalloc(size, sizeof(unsigned int), GFP_KERNEL | __GFP_NOWARN); + + if (off) + return off; + + if (size < (SIZE_MAX / sizeof(unsigned int))) + off = vmalloc(size * sizeof(unsigned int)); + + return off; +} +EXPORT_SYMBOL(xt_alloc_entry_offsets); + +/** + * xt_find_jump_offset - check if target is a valid jump offset + * + * @offsets: array containing all valid rule start offsets of a rule blob + * @target: the jump target to search for + * @size: entries in @offset + */ +bool xt_find_jump_offset(const unsigned int *offsets, + unsigned int target, unsigned int size) +{ + int m, low = 0, hi = size; + + while (hi > low) { + m = (low + hi) / 2u; + + if (offsets[m] > target) + hi = m; + else if (offsets[m] < target) + low = m + 1; + else + return true; + } + + return false; +} +EXPORT_SYMBOL(xt_find_jump_offset); + int xt_check_target(struct xt_tgchk_param *par, unsigned int size, u_int8_t proto, bool inv_proto) { -- cgit v1.2.3 From cc2d1de06f0572a51437d1f31633d81afea5eb47 Mon Sep 17 00:00:00 2001 From: Rafał Miłecki Date: Fri, 8 Jul 2016 17:14:18 +0200 Subject: bcma: define ChipCommon B MII registers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We don't have access to datasheets to document all the bits but we can name these registers at least. Signed-off-by: Rafał Miłecki Signed-off-by: Kalle Valo --- drivers/bcma/driver_chipcommon_b.c | 9 +++++---- include/linux/bcma/bcma_driver_chipcommon.h | 3 +++ 2 files changed, 8 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/bcma/driver_chipcommon_b.c b/drivers/bcma/driver_chipcommon_b.c index c20b5f4ff290..57f10b58b47c 100644 --- a/drivers/bcma/driver_chipcommon_b.c +++ b/drivers/bcma/driver_chipcommon_b.c @@ -33,11 +33,12 @@ static bool bcma_wait_reg(struct bcma_bus *bus, void __iomem *addr, u32 mask, void bcma_chipco_b_mii_write(struct bcma_drv_cc_b *ccb, u32 offset, u32 value) { struct bcma_bus *bus = ccb->core->bus; + void __iomem *mii = ccb->mii; - writel(offset, ccb->mii + 0x00); - bcma_wait_reg(bus, ccb->mii + 0x00, 0x0100, 0x0000, 100); - writel(value, ccb->mii + 0x04); - bcma_wait_reg(bus, ccb->mii + 0x00, 0x0100, 0x0000, 100); + writel(offset, mii + BCMA_CCB_MII_MNG_CTL); + bcma_wait_reg(bus, mii + BCMA_CCB_MII_MNG_CTL, 0x0100, 0x0000, 100); + writel(value, mii + BCMA_CCB_MII_MNG_CMD_DATA); + bcma_wait_reg(bus, mii + BCMA_CCB_MII_MNG_CTL, 0x0100, 0x0000, 100); } EXPORT_SYMBOL_GPL(bcma_chipco_b_mii_write); diff --git a/include/linux/bcma/bcma_driver_chipcommon.h b/include/linux/bcma/bcma_driver_chipcommon.h index a5ac2cad5cb7..b20e3d56253f 100644 --- a/include/linux/bcma/bcma_driver_chipcommon.h +++ b/include/linux/bcma/bcma_driver_chipcommon.h @@ -504,6 +504,9 @@ #define BCMA_CC_PMU1_PLL0_PC2_NDIV_INT_MASK 0x1ff00000 #define BCMA_CC_PMU1_PLL0_PC2_NDIV_INT_SHIFT 20 +#define BCMA_CCB_MII_MNG_CTL 0x0000 +#define BCMA_CCB_MII_MNG_CMD_DATA 0x0004 + /* BCM4331 ChipControl numbers. */ #define BCMA_CHIPCTL_4331_BT_COEXIST BIT(0) /* 0 disable */ #define BCMA_CHIPCTL_4331_SECI BIT(1) /* 0 SECI is disabled (JATG functional) */ -- cgit v1.2.3 From 59d3656d5bf504f771fc44fdbc7a9a8590795f22 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:46 -0700 Subject: bpf: add bpf_prog_add api for bulk prog refcnt A subsystem may need to store many copies of a bpf program, each deserving its own reference. Rather than requiring the caller to loop one by one (with possible mid-loop failure), add a bulk bpf_prog_add api. Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/bpf.h | 1 + kernel/bpf/syscall.c | 12 +++++++++--- 2 files changed, 10 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index c13e92b00bf5..75a5ae6bee07 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -224,6 +224,7 @@ void bpf_register_map_type(struct bpf_map_type_list *tl); struct bpf_prog *bpf_prog_get(u32 ufd); struct bpf_prog *bpf_prog_get_type(u32 ufd, enum bpf_prog_type type); +struct bpf_prog *bpf_prog_add(struct bpf_prog *prog, int i); struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog); void bpf_prog_put(struct bpf_prog *prog); diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index 96d938a22050..228f962447a5 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -670,14 +670,20 @@ static struct bpf_prog *____bpf_prog_get(struct fd f) return f.file->private_data; } -struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog) +struct bpf_prog *bpf_prog_add(struct bpf_prog *prog, int i) { - if (atomic_inc_return(&prog->aux->refcnt) > BPF_MAX_REFCNT) { - atomic_dec(&prog->aux->refcnt); + if (atomic_add_return(i, &prog->aux->refcnt) > BPF_MAX_REFCNT) { + atomic_sub(i, &prog->aux->refcnt); return ERR_PTR(-EBUSY); } return prog; } +EXPORT_SYMBOL_GPL(bpf_prog_add); + +struct bpf_prog *bpf_prog_inc(struct bpf_prog *prog) +{ + return bpf_prog_add(prog, 1); +} static struct bpf_prog *__bpf_prog_get(u32 ufd, enum bpf_prog_type *type) { -- cgit v1.2.3 From 6a773a15a1e8874e5eccd2f29190c31085912c95 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:47 -0700 Subject: bpf: add XDP prog type for early driver filter Add a new bpf prog type that is intended to run in early stages of the packet rx path. Only minimal packet metadata will be available, hence a new context type, struct xdp_md, is exposed to userspace. So far only expose the packet start and end pointers, and only in read mode. An XDP program must return one of the well known enum values, all other return codes are reserved for future use. Unfortunately, this restriction is hard to enforce at verification time, so take the approach of warning at runtime when such programs are encountered. Out of bounds return codes should alias to XDP_ABORTED. Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/filter.h | 18 +++++++++++ include/uapi/linux/bpf.h | 20 ++++++++++++ kernel/bpf/verifier.c | 1 + net/core/filter.c | 79 ++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 118 insertions(+) (limited to 'include/linux') diff --git a/include/linux/filter.h b/include/linux/filter.h index 6fc31ef1da2d..15d816a8b755 100644 --- a/include/linux/filter.h +++ b/include/linux/filter.h @@ -368,6 +368,11 @@ struct bpf_skb_data_end { void *data_end; }; +struct xdp_buff { + void *data; + void *data_end; +}; + /* compute the linear packet data range [data, data_end) which * will be accessed by cls_bpf and act_bpf programs */ @@ -429,6 +434,18 @@ static inline u32 bpf_prog_run_clear_cb(const struct bpf_prog *prog, return BPF_PROG_RUN(prog, skb); } +static inline u32 bpf_prog_run_xdp(const struct bpf_prog *prog, + struct xdp_buff *xdp) +{ + u32 ret; + + rcu_read_lock(); + ret = BPF_PROG_RUN(prog, (void *)xdp); + rcu_read_unlock(); + + return ret; +} + static inline unsigned int bpf_prog_size(unsigned int proglen) { return max(sizeof(struct bpf_prog), @@ -509,6 +526,7 @@ bool bpf_helper_changes_skb_data(void *func); struct bpf_prog *bpf_patch_insn_single(struct bpf_prog *prog, u32 off, const struct bpf_insn *patch, u32 len); +void bpf_warn_invalid_xdp_action(u32 act); #ifdef CONFIG_BPF_JIT extern int bpf_jit_enable; diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h index c4d922439d20..a51786566c2f 100644 --- a/include/uapi/linux/bpf.h +++ b/include/uapi/linux/bpf.h @@ -94,6 +94,7 @@ enum bpf_prog_type { BPF_PROG_TYPE_SCHED_CLS, BPF_PROG_TYPE_SCHED_ACT, BPF_PROG_TYPE_TRACEPOINT, + BPF_PROG_TYPE_XDP, }; #define BPF_PSEUDO_MAP_FD 1 @@ -439,4 +440,23 @@ struct bpf_tunnel_key { __u32 tunnel_label; }; +/* User return codes for XDP prog type. + * A valid XDP program must return one of these defined values. All other + * return codes are reserved for future use. Unknown return codes will result + * in packet drop. + */ +enum xdp_action { + XDP_ABORTED = 0, + XDP_DROP, + XDP_PASS, +}; + +/* user accessible metadata for XDP packet hook + * new fields must be added to the end of this structure + */ +struct xdp_md { + __u32 data; + __u32 data_end; +}; + #endif /* _UAPI__LINUX_BPF_H__ */ diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index e206c2181412..a8d67d097b0d 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -713,6 +713,7 @@ static int check_ptr_alignment(struct verifier_env *env, struct reg_state *reg, switch (env->prog->type) { case BPF_PROG_TYPE_SCHED_CLS: case BPF_PROG_TYPE_SCHED_ACT: + case BPF_PROG_TYPE_XDP: break; default: verbose("verifier is misconfigured\n"); diff --git a/net/core/filter.c b/net/core/filter.c index 22e3992c8b48..6c627bc4be6e 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2410,6 +2410,12 @@ tc_cls_act_func_proto(enum bpf_func_id func_id) } } +static const struct bpf_func_proto * +xdp_func_proto(enum bpf_func_id func_id) +{ + return sk_filter_func_proto(func_id); +} + static bool __is_valid_access(int off, int size, enum bpf_access_type type) { if (off < 0 || off >= sizeof(struct __sk_buff)) @@ -2477,6 +2483,44 @@ static bool tc_cls_act_is_valid_access(int off, int size, return __is_valid_access(off, size, type); } +static bool __is_valid_xdp_access(int off, int size, + enum bpf_access_type type) +{ + if (off < 0 || off >= sizeof(struct xdp_md)) + return false; + if (off % size != 0) + return false; + if (size != 4) + return false; + + return true; +} + +static bool xdp_is_valid_access(int off, int size, + enum bpf_access_type type, + enum bpf_reg_type *reg_type) +{ + if (type == BPF_WRITE) + return false; + + switch (off) { + case offsetof(struct xdp_md, data): + *reg_type = PTR_TO_PACKET; + break; + case offsetof(struct xdp_md, data_end): + *reg_type = PTR_TO_PACKET_END; + break; + } + + return __is_valid_xdp_access(off, size, type); +} + +void bpf_warn_invalid_xdp_action(u32 act) +{ + WARN_ONCE(1, "Illegal XDP return value %u, expect packet loss\n", act); +} +EXPORT_SYMBOL_GPL(bpf_warn_invalid_xdp_action); + static u32 bpf_net_convert_ctx_access(enum bpf_access_type type, int dst_reg, int src_reg, int ctx_off, struct bpf_insn *insn_buf, @@ -2628,6 +2672,29 @@ static u32 bpf_net_convert_ctx_access(enum bpf_access_type type, int dst_reg, return insn - insn_buf; } +static u32 xdp_convert_ctx_access(enum bpf_access_type type, int dst_reg, + int src_reg, int ctx_off, + struct bpf_insn *insn_buf, + struct bpf_prog *prog) +{ + struct bpf_insn *insn = insn_buf; + + switch (ctx_off) { + case offsetof(struct xdp_md, data): + *insn++ = BPF_LDX_MEM(bytes_to_bpf_size(FIELD_SIZEOF(struct xdp_buff, data)), + dst_reg, src_reg, + offsetof(struct xdp_buff, data)); + break; + case offsetof(struct xdp_md, data_end): + *insn++ = BPF_LDX_MEM(bytes_to_bpf_size(FIELD_SIZEOF(struct xdp_buff, data_end)), + dst_reg, src_reg, + offsetof(struct xdp_buff, data_end)); + break; + } + + return insn - insn_buf; +} + static const struct bpf_verifier_ops sk_filter_ops = { .get_func_proto = sk_filter_func_proto, .is_valid_access = sk_filter_is_valid_access, @@ -2640,6 +2707,12 @@ static const struct bpf_verifier_ops tc_cls_act_ops = { .convert_ctx_access = bpf_net_convert_ctx_access, }; +static const struct bpf_verifier_ops xdp_ops = { + .get_func_proto = xdp_func_proto, + .is_valid_access = xdp_is_valid_access, + .convert_ctx_access = xdp_convert_ctx_access, +}; + static struct bpf_prog_type_list sk_filter_type __read_mostly = { .ops = &sk_filter_ops, .type = BPF_PROG_TYPE_SOCKET_FILTER, @@ -2655,11 +2728,17 @@ static struct bpf_prog_type_list sched_act_type __read_mostly = { .type = BPF_PROG_TYPE_SCHED_ACT, }; +static struct bpf_prog_type_list xdp_type __read_mostly = { + .ops = &xdp_ops, + .type = BPF_PROG_TYPE_XDP, +}; + static int __init register_sk_filter_ops(void) { bpf_register_prog_type(&sk_filter_type); bpf_register_prog_type(&sched_cls_type); bpf_register_prog_type(&sched_act_type); + bpf_register_prog_type(&xdp_type); return 0; } -- cgit v1.2.3 From a7862b45849fe2f8610a2bec89235580f55d337f Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:48 -0700 Subject: net: add ndo to setup/query xdp prog in adapter rx Add one new netdev op for drivers implementing the BPF_PROG_TYPE_XDP filter. The single op is used for both setup/query of the xdp program, modelled after ndo_setup_tc. Signed-off-by: Brenden Blanco Signed-off-by: David S. Miller --- include/linux/netdevice.h | 34 ++++++++++++++++++++++++++++++++++ net/core/dev.c | 33 +++++++++++++++++++++++++++++++++ 2 files changed, 67 insertions(+) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 49736a31acaa..fab9a1c2a2ac 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -63,6 +63,7 @@ struct wpan_dev; struct mpls_dev; /* UDP Tunnel offloads */ struct udp_tunnel_info; +struct bpf_prog; void netdev_set_default_ethtool_ops(struct net_device *dev, const struct ethtool_ops *ops); @@ -799,6 +800,33 @@ struct tc_to_netdev { }; }; +/* These structures hold the attributes of xdp state that are being passed + * to the netdevice through the xdp op. + */ +enum xdp_netdev_command { + /* Set or clear a bpf program used in the earliest stages of packet + * rx. The prog will have been loaded as BPF_PROG_TYPE_XDP. The callee + * is responsible for calling bpf_prog_put on any old progs that are + * stored. In case of error, the callee need not release the new prog + * reference, but on success it takes ownership and must bpf_prog_put + * when it is no longer used. + */ + XDP_SETUP_PROG, + /* Check if a bpf program is set on the device. The callee should + * return true if a program is currently attached and running. + */ + XDP_QUERY_PROG, +}; + +struct netdev_xdp { + enum xdp_netdev_command command; + union { + /* XDP_SETUP_PROG */ + struct bpf_prog *prog; + /* XDP_QUERY_PROG */ + bool prog_attached; + }; +}; /* * This structure defines the management hooks for network devices. @@ -1087,6 +1115,9 @@ struct tc_to_netdev { * appropriate rx headroom value allows avoiding skb head copy on * forward. Setting a negative value resets the rx headroom to the * default value. + * int (*ndo_xdp)(struct net_device *dev, struct netdev_xdp *xdp); + * This function is used to set or query state related to XDP on the + * netdevice. See definition of enum xdp_netdev_command for details. * */ struct net_device_ops { @@ -1271,6 +1302,8 @@ struct net_device_ops { struct sk_buff *skb); void (*ndo_set_rx_headroom)(struct net_device *dev, int needed_headroom); + int (*ndo_xdp)(struct net_device *dev, + struct netdev_xdp *xdp); }; /** @@ -3257,6 +3290,7 @@ int dev_get_phys_port_id(struct net_device *dev, int dev_get_phys_port_name(struct net_device *dev, char *name, size_t len); int dev_change_proto_down(struct net_device *dev, bool proto_down); +int dev_change_xdp_fd(struct net_device *dev, int fd); struct sk_buff *validate_xmit_skb_list(struct sk_buff *skb, struct net_device *dev); struct sk_buff *dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev, struct netdev_queue *txq, int *ret); diff --git a/net/core/dev.c b/net/core/dev.c index 7894e406c806..2a9c39f8824e 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -94,6 +94,7 @@ #include #include #include +#include #include #include #include @@ -6614,6 +6615,38 @@ int dev_change_proto_down(struct net_device *dev, bool proto_down) } EXPORT_SYMBOL(dev_change_proto_down); +/** + * dev_change_xdp_fd - set or clear a bpf program for a device rx path + * @dev: device + * @fd: new program fd or negative value to clear + * + * Set or clear a bpf program for a device + */ +int dev_change_xdp_fd(struct net_device *dev, int fd) +{ + const struct net_device_ops *ops = dev->netdev_ops; + struct bpf_prog *prog = NULL; + struct netdev_xdp xdp = {}; + int err; + + if (!ops->ndo_xdp) + return -EOPNOTSUPP; + if (fd >= 0) { + prog = bpf_prog_get_type(fd, BPF_PROG_TYPE_XDP); + if (IS_ERR(prog)) + return PTR_ERR(prog); + } + + xdp.command = XDP_SETUP_PROG; + xdp.prog = prog; + err = ops->ndo_xdp(dev, &xdp); + if (err < 0 && prog) + bpf_prog_put(prog); + + return err; +} +EXPORT_SYMBOL(dev_change_xdp_fd); + /** * dev_new_index - allocate an ifindex * @net: the applicable net namespace -- cgit v1.2.3 From 224e92e02a769b8028ca2450443586af8b4f1715 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Tue, 19 Jul 2016 12:16:54 -0700 Subject: net/mlx4_en: break out tx_desc write into separate function In preparation for writing the tx descriptor from multiple functions, create a helper for both normal and blueflame access. Signed-off-by: Brenden Blanco Signed-off-by: David S. Miller --- drivers/infiniband/hw/mlx4/qp.c | 11 +-- drivers/net/ethernet/mellanox/mlx4/en_tx.c | 134 ++++++++++++++++------------- include/linux/mlx4/qp.h | 18 ++-- 3 files changed, 92 insertions(+), 71 deletions(-) (limited to 'include/linux') diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index 8db8405c1e99..768085f59566 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -232,7 +232,7 @@ static void stamp_send_wqe(struct mlx4_ib_qp *qp, int n, int size) } } else { ctrl = buf = get_send_wqe(qp, n & (qp->sq.wqe_cnt - 1)); - s = (ctrl->fence_size & 0x3f) << 4; + s = (ctrl->qpn_vlan.fence_size & 0x3f) << 4; for (i = 64; i < s; i += 64) { wqe = buf + i; *wqe = cpu_to_be32(0xffffffff); @@ -264,7 +264,7 @@ static void post_nop_wqe(struct mlx4_ib_qp *qp, int n, int size) inl->byte_count = cpu_to_be32(1 << 31 | (size - s - sizeof *inl)); } ctrl->srcrb_flags = 0; - ctrl->fence_size = size / 16; + ctrl->qpn_vlan.fence_size = size / 16; /* * Make sure descriptor is fully written before setting ownership bit * (because HW can start executing as soon as we do). @@ -1992,7 +1992,8 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp, ctrl = get_send_wqe(qp, i); ctrl->owner_opcode = cpu_to_be32(1 << 31); if (qp->sq_max_wqes_per_wr == 1) - ctrl->fence_size = 1 << (qp->sq.wqe_shift - 4); + ctrl->qpn_vlan.fence_size = + 1 << (qp->sq.wqe_shift - 4); stamp_send_wqe(qp, i, 1 << qp->sq.wqe_shift); } @@ -3169,8 +3170,8 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, wmb(); *lso_wqe = lso_hdr_sz; - ctrl->fence_size = (wr->send_flags & IB_SEND_FENCE ? - MLX4_WQE_CTRL_FENCE : 0) | size; + ctrl->qpn_vlan.fence_size = (wr->send_flags & IB_SEND_FENCE ? + MLX4_WQE_CTRL_FENCE : 0) | size; /* * Make sure descriptor is fully written before diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c index 76aa4d27183c..2f56018ddae9 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c @@ -631,8 +631,7 @@ static int get_real_size(const struct sk_buff *skb, static void build_inline_wqe(struct mlx4_en_tx_desc *tx_desc, const struct sk_buff *skb, const struct skb_shared_info *shinfo, - int real_size, u16 *vlan_tag, - int tx_ind, void *fragptr) + void *fragptr) { struct mlx4_wqe_inline_seg *inl = &tx_desc->inl; int spc = MLX4_INLINE_ALIGN - CTRL_SIZE - sizeof *inl; @@ -700,10 +699,66 @@ static void mlx4_bf_copy(void __iomem *dst, const void *src, __iowrite64_copy(dst, src, bytecnt / 8); } +void mlx4_en_xmit_doorbell(struct mlx4_en_tx_ring *ring) +{ + wmb(); + /* Since there is no iowrite*_native() that writes the + * value as is, without byteswapping - using the one + * the doesn't do byteswapping in the relevant arch + * endianness. + */ +#if defined(__LITTLE_ENDIAN) + iowrite32( +#else + iowrite32be( +#endif + ring->doorbell_qpn, + ring->bf.uar->map + MLX4_SEND_DOORBELL); +} + +static void mlx4_en_tx_write_desc(struct mlx4_en_tx_ring *ring, + struct mlx4_en_tx_desc *tx_desc, + union mlx4_wqe_qpn_vlan qpn_vlan, + int desc_size, int bf_index, + __be32 op_own, bool bf_ok, + bool send_doorbell) +{ + tx_desc->ctrl.qpn_vlan = qpn_vlan; + + if (bf_ok) { + op_own |= htonl((bf_index & 0xffff) << 8); + /* Ensure new descriptor hits memory + * before setting ownership of this descriptor to HW + */ + dma_wmb(); + tx_desc->ctrl.owner_opcode = op_own; + + wmb(); + + mlx4_bf_copy(ring->bf.reg + ring->bf.offset, &tx_desc->ctrl, + desc_size); + + wmb(); + + ring->bf.offset ^= ring->bf.buf_size; + } else { + /* Ensure new descriptor hits memory + * before setting ownership of this descriptor to HW + */ + dma_wmb(); + tx_desc->ctrl.owner_opcode = op_own; + if (send_doorbell) + mlx4_en_xmit_doorbell(ring); + else + ring->xmit_more++; + } +} + netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) { struct skb_shared_info *shinfo = skb_shinfo(skb); struct mlx4_en_priv *priv = netdev_priv(dev); + union mlx4_wqe_qpn_vlan qpn_vlan = {}; struct device *ddev = priv->ddev; struct mlx4_en_tx_ring *ring; struct mlx4_en_tx_desc *tx_desc; @@ -715,7 +770,6 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) int real_size; u32 index, bf_index; __be32 op_own; - u16 vlan_tag = 0; u16 vlan_proto = 0; int i_frag; int lso_header_size; @@ -725,6 +779,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) bool stop_queue; bool inline_ok; u32 ring_cons; + bool bf_ok; tx_ind = skb_get_queue_mapping(skb); ring = priv->tx_ring[tx_ind]; @@ -749,9 +804,17 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) goto tx_drop; } + bf_ok = ring->bf_enabled; if (skb_vlan_tag_present(skb)) { - vlan_tag = skb_vlan_tag_get(skb); + qpn_vlan.vlan_tag = cpu_to_be16(skb_vlan_tag_get(skb)); vlan_proto = be16_to_cpu(skb->vlan_proto); + if (vlan_proto == ETH_P_8021AD) + qpn_vlan.ins_vlan = MLX4_WQE_CTRL_INS_SVLAN; + else if (vlan_proto == ETH_P_8021Q) + qpn_vlan.ins_vlan = MLX4_WQE_CTRL_INS_CVLAN; + else + qpn_vlan.ins_vlan = 0; + bf_ok = false; } netdev_txq_bql_enqueue_prefetchw(ring->tx_queue); @@ -771,6 +834,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) else { tx_desc = (struct mlx4_en_tx_desc *) ring->bounce_buf; bounce = true; + bf_ok = false; } /* Save skb in tx_info ring */ @@ -907,8 +971,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) AVG_PERF_COUNTER(priv->pstats.tx_pktsz_avg, skb->len); if (tx_info->inl) - build_inline_wqe(tx_desc, skb, shinfo, real_size, &vlan_tag, - tx_ind, fragptr); + build_inline_wqe(tx_desc, skb, shinfo, fragptr); if (skb->encapsulation) { union { @@ -946,60 +1009,15 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) real_size = (real_size / 16) & 0x3f; - if (ring->bf_enabled && desc_size <= MAX_BF && !bounce && - !skb_vlan_tag_present(skb) && send_doorbell) { - tx_desc->ctrl.bf_qpn = ring->doorbell_qpn | - cpu_to_be32(real_size); - - op_own |= htonl((bf_index & 0xffff) << 8); - /* Ensure new descriptor hits memory - * before setting ownership of this descriptor to HW - */ - dma_wmb(); - tx_desc->ctrl.owner_opcode = op_own; - - wmb(); - - mlx4_bf_copy(ring->bf.reg + ring->bf.offset, &tx_desc->ctrl, - desc_size); - - wmb(); - - ring->bf.offset ^= ring->bf.buf_size; - } else { - tx_desc->ctrl.vlan_tag = cpu_to_be16(vlan_tag); - if (vlan_proto == ETH_P_8021AD) - tx_desc->ctrl.ins_vlan = MLX4_WQE_CTRL_INS_SVLAN; - else if (vlan_proto == ETH_P_8021Q) - tx_desc->ctrl.ins_vlan = MLX4_WQE_CTRL_INS_CVLAN; - else - tx_desc->ctrl.ins_vlan = 0; + bf_ok &= desc_size <= MAX_BF && send_doorbell; - tx_desc->ctrl.fence_size = real_size; + if (bf_ok) + qpn_vlan.bf_qpn = ring->doorbell_qpn | cpu_to_be32(real_size); + else + qpn_vlan.fence_size = real_size; - /* Ensure new descriptor hits memory - * before setting ownership of this descriptor to HW - */ - dma_wmb(); - tx_desc->ctrl.owner_opcode = op_own; - if (send_doorbell) { - wmb(); - /* Since there is no iowrite*_native() that writes the - * value as is, without byteswapping - using the one - * the doesn't do byteswapping in the relevant arch - * endianness. - */ -#if defined(__LITTLE_ENDIAN) - iowrite32( -#else - iowrite32be( -#endif - ring->doorbell_qpn, - ring->bf.uar->map + MLX4_SEND_DOORBELL); - } else { - ring->xmit_more++; - } - } + mlx4_en_tx_write_desc(ring, tx_desc, qpn_vlan, desc_size, bf_index, + op_own, bf_ok, send_doorbell); if (unlikely(stop_queue)) { /* If queue was emptied after the if (stop_queue) , and before diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h index 587cdf943b52..deaa2217214d 100644 --- a/include/linux/mlx4/qp.h +++ b/include/linux/mlx4/qp.h @@ -291,16 +291,18 @@ enum { MLX4_WQE_CTRL_FORCE_LOOPBACK = 1 << 0, }; +union mlx4_wqe_qpn_vlan { + struct { + __be16 vlan_tag; + u8 ins_vlan; + u8 fence_size; + }; + __be32 bf_qpn; +}; + struct mlx4_wqe_ctrl_seg { __be32 owner_opcode; - union { - struct { - __be16 vlan_tag; - u8 ins_vlan; - u8 fence_size; - }; - __be32 bf_qpn; - }; + union mlx4_wqe_qpn_vlan qpn_vlan; /* * High 24 bits are SRC remote buffer; low 8 bits are flags: * [7] SO (strong ordering) -- cgit v1.2.3 From cc2e0b3fbcdd9667d7b7ecdf36d7b4d3647681d6 Mon Sep 17 00:00:00 2001 From: Brenden Blanco Date: Wed, 20 Jul 2016 07:55:52 -0700 Subject: bpf: fix implicit declaration of bpf_prog_add For the ifndef case of CONFIG_BPF_SYSCALL, an inline version of bpf_prog_add needs to exist otherwise the build breaks on some configs. drivers/net/ethernet/mellanox/mlx4/en_netdev.c:2544:10: error: implicit declaration of function 'bpf_prog_add' prog = bpf_prog_add(prog, priv->rx_ring_num - 1); The function is introduced in 59d3656d5bf50 ("bpf: add bpf_prog_add api for bulk prog refcnt") and first used in 47f1afdba2b87 ("net/mlx4_en: add support for fast rx drop bpf program"). Fixes: 47f1afdba2b87 ("net/mlx4_en: add support for fast rx drop bpf program") Reported-by: kbuild test robot Reported-by: Tariq Toukan Signed-off-by: Brenden Blanco Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/bpf.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 75a5ae6bee07..36da0749205a 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -289,6 +289,10 @@ static inline struct bpf_prog *bpf_prog_get_type(u32 ufd, { return ERR_PTR(-EOPNOTSUPP); } +static inline struct bpf_prog *bpf_prog_add(struct bpf_prog *prog, int i) +{ + return ERR_PTR(-EOPNOTSUPP); +} static inline void bpf_prog_put(struct bpf_prog *prog) { -- cgit v1.2.3 From b87f7936a93246804cf70e7e2e0568799c948bb1 Mon Sep 17 00:00:00 2001 From: Yotam Gigi Date: Thu, 21 Jul 2016 12:03:12 +0200 Subject: net/sched: Add match-all classifier hw offloading. Following the work that have been done on offloading classifiers like u32 and flower, now the match-all classifier hw offloading is possible. if the interface supports tc offloading. To control the offloading, two tc flags have been introduced: skip_sw and skip_hw. Typical usage: tc filter add dev eth25 parent ffff: \ matchall skip_sw \ action mirred egress mirror \ dev eth27 Signed-off-by: Yotam Gigi Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- include/linux/netdevice.h | 2 ++ include/net/pkt_cls.h | 11 +++++++ include/uapi/linux/pkt_cls.h | 1 + net/sched/cls_matchall.c | 76 ++++++++++++++++++++++++++++++++++++++++++-- 4 files changed, 87 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 43c749b1b619..076df5360ba5 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -787,6 +787,7 @@ enum { TC_SETUP_MQPRIO, TC_SETUP_CLSU32, TC_SETUP_CLSFLOWER, + TC_SETUP_MATCHALL, }; struct tc_cls_u32_offload; @@ -797,6 +798,7 @@ struct tc_to_netdev { u8 tc; struct tc_cls_u32_offload *cls_u32; struct tc_cls_flower_offload *cls_flower; + struct tc_cls_matchall_offload *cls_mall; }; }; diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h index 3722dda0199d..6f8d65342d3a 100644 --- a/include/net/pkt_cls.h +++ b/include/net/pkt_cls.h @@ -442,4 +442,15 @@ struct tc_cls_flower_offload { struct tcf_exts *exts; }; +enum tc_matchall_command { + TC_CLSMATCHALL_REPLACE, + TC_CLSMATCHALL_DESTROY, +}; + +struct tc_cls_matchall_offload { + enum tc_matchall_command command; + struct tcf_exts *exts; + unsigned long cookie; +}; + #endif diff --git a/include/uapi/linux/pkt_cls.h b/include/uapi/linux/pkt_cls.h index a32494887e01..d1c1ccaba787 100644 --- a/include/uapi/linux/pkt_cls.h +++ b/include/uapi/linux/pkt_cls.h @@ -439,6 +439,7 @@ enum { TCA_MATCHALL_UNSPEC, TCA_MATCHALL_CLASSID, TCA_MATCHALL_ACT, + TCA_MATCHALL_FLAGS, __TCA_MATCHALL_MAX, }; diff --git a/net/sched/cls_matchall.c b/net/sched/cls_matchall.c index 8a6b4de7a99a..25927b6c4436 100644 --- a/net/sched/cls_matchall.c +++ b/net/sched/cls_matchall.c @@ -21,6 +21,7 @@ struct cls_mall_filter { struct tcf_result res; u32 handle; struct rcu_head rcu; + u32 flags; }; struct cls_mall_head { @@ -34,6 +35,9 @@ static int mall_classify(struct sk_buff *skb, const struct tcf_proto *tp, struct cls_mall_head *head = rcu_dereference_bh(tp->root); struct cls_mall_filter *f = head->filter; + if (tc_skip_sw(f->flags)) + return -1; + return tcf_exts_exec(skb, &f->exts, res); } @@ -55,18 +59,61 @@ static void mall_destroy_filter(struct rcu_head *head) struct cls_mall_filter *f = container_of(head, struct cls_mall_filter, rcu); tcf_exts_destroy(&f->exts); + kfree(f); } +static int mall_replace_hw_filter(struct tcf_proto *tp, + struct cls_mall_filter *f, + unsigned long cookie) +{ + struct net_device *dev = tp->q->dev_queue->dev; + struct tc_to_netdev offload; + struct tc_cls_matchall_offload mall_offload = {0}; + + offload.type = TC_SETUP_MATCHALL; + offload.cls_mall = &mall_offload; + offload.cls_mall->command = TC_CLSMATCHALL_REPLACE; + offload.cls_mall->exts = &f->exts; + offload.cls_mall->cookie = cookie; + + return dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->protocol, + &offload); +} + +static void mall_destroy_hw_filter(struct tcf_proto *tp, + struct cls_mall_filter *f, + unsigned long cookie) +{ + struct net_device *dev = tp->q->dev_queue->dev; + struct tc_to_netdev offload; + struct tc_cls_matchall_offload mall_offload = {0}; + + offload.type = TC_SETUP_MATCHALL; + offload.cls_mall = &mall_offload; + offload.cls_mall->command = TC_CLSMATCHALL_DESTROY; + offload.cls_mall->exts = NULL; + offload.cls_mall->cookie = cookie; + + dev->netdev_ops->ndo_setup_tc(dev, tp->q->handle, tp->protocol, + &offload); +} + static bool mall_destroy(struct tcf_proto *tp, bool force) { struct cls_mall_head *head = rtnl_dereference(tp->root); + struct net_device *dev = tp->q->dev_queue->dev; + struct cls_mall_filter *f = head->filter; - if (!force && head->filter) + if (!force && f) return false; - if (head->filter) - call_rcu(&head->filter->rcu, mall_destroy_filter); + if (f) { + if (tc_should_offload(dev, tp, f->flags)) + mall_destroy_hw_filter(tp, f, (unsigned long) f); + + call_rcu(&f->rcu, mall_destroy_filter); + } RCU_INIT_POINTER(tp->root, NULL); kfree_rcu(head, rcu); return true; @@ -117,8 +164,10 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, { struct cls_mall_head *head = rtnl_dereference(tp->root); struct cls_mall_filter *fold = (struct cls_mall_filter *) *arg; + struct net_device *dev = tp->q->dev_queue->dev; struct cls_mall_filter *f; struct nlattr *tb[TCA_MATCHALL_MAX + 1]; + u32 flags = 0; int err; if (!tca[TCA_OPTIONS]) @@ -135,6 +184,12 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, if (err < 0) return err; + if (tb[TCA_MATCHALL_FLAGS]) { + flags = nla_get_u32(tb[TCA_MATCHALL_FLAGS]); + if (!tc_flags_valid(flags)) + return -EINVAL; + } + f = kzalloc(sizeof(*f), GFP_KERNEL); if (!f) return -ENOBUFS; @@ -144,11 +199,22 @@ static int mall_change(struct net *net, struct sk_buff *in_skb, if (!handle) handle = 1; f->handle = handle; + f->flags = flags; err = mall_set_parms(net, tp, f, base, tb, tca[TCA_RATE], ovr); if (err) goto errout; + if (tc_should_offload(dev, tp, flags)) { + err = mall_replace_hw_filter(tp, f, (unsigned long) f); + if (err) { + if (tc_skip_sw(flags)) + goto errout; + else + err = 0; + } + } + *arg = (unsigned long) f; rcu_assign_pointer(head->filter, f); @@ -163,6 +229,10 @@ static int mall_delete(struct tcf_proto *tp, unsigned long arg) { struct cls_mall_head *head = rtnl_dereference(tp->root); struct cls_mall_filter *f = (struct cls_mall_filter *) arg; + struct net_device *dev = tp->q->dev_queue->dev; + + if (tc_should_offload(dev, tp, f->flags)) + mall_destroy_hw_filter(tp, f, (unsigned long) f); RCU_INIT_POINTER(head->filter, NULL); tcf_unbind_filter(tp, &f->res); -- cgit v1.2.3 From aa7145c16d6bf086538ad7eb20c807513bfa5efc Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Fri, 22 Jul 2016 01:19:42 +0200 Subject: bpf, events: fix offset in skb copy handler This patch fixes the __output_custom() routine we currently use with bpf_skb_copy(). I missed that when len is larger than the size of the current handle, we can issue multiple invocations of copy_func, and __output_custom() advances destination but also source buffer by the written amount of bytes. When we have __output_custom(), this is actually wrong since in that case the source buffer points to a non-linear object, in our case an skb, which the copy_func helper is supposed to walk. Therefore, since this is non-linear we thus need to pass the offset into the helper, so that copy_func can use it for extracting the data from the source object. Therefore, adjust the callback signatures properly and pass offset into the skb_header_pointer() invoked from bpf_skb_copy() callback. The __DEFINE_OUTPUT_COPY_BODY() is adjusted to accommodate for two things: i) to pass in whether we should advance source buffer or not; this is a compile-time constant condition, ii) to pass in the offset for __output_custom(), which we do with help of __VA_ARGS__, so everything can stay inlined as is currently. Both changes allow for adapting the __output_* fast-path helpers w/o extra overhead. Fixes: 555c8a8623a3 ("bpf: avoid stack copy and use skb ctx for event output") Fixes: 7e3f977edd0b ("perf, events: add non-linear data support for raw records") Signed-off-by: Daniel Borkmann Acked-by: Alexei Starovoitov Signed-off-by: David S. Miller --- include/linux/bpf.h | 2 +- include/linux/perf_event.h | 2 +- kernel/events/internal.h | 15 ++++++++++----- net/core/filter.c | 4 ++-- 4 files changed, 14 insertions(+), 9 deletions(-) (limited to 'include/linux') diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 36da0749205a..11134238417d 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -211,7 +211,7 @@ bool bpf_prog_array_compatible(struct bpf_array *array, const struct bpf_prog *f const struct bpf_func_proto *bpf_get_trace_printk_proto(void); typedef unsigned long (*bpf_ctx_copy_t)(void *dst, const void *src, - unsigned long len); + unsigned long off, unsigned long len); u64 bpf_event_output(struct bpf_map *map, u64 flags, void *meta, u64 meta_size, void *ctx, u64 ctx_size, bpf_ctx_copy_t ctx_copy); diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index e79e6c6fed89..15e55b7ee096 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -70,7 +70,7 @@ struct perf_callchain_entry_ctx { }; typedef unsigned long (*perf_copy_f)(void *dst, const void *src, - unsigned long len); + unsigned long off, unsigned long len); struct perf_raw_frag { union { diff --git a/kernel/events/internal.h b/kernel/events/internal.h index 2417eb5512cd..486fd78eb8d5 100644 --- a/kernel/events/internal.h +++ b/kernel/events/internal.h @@ -123,18 +123,19 @@ static inline unsigned long perf_aux_size(struct ring_buffer *rb) return rb->aux_nr_pages << PAGE_SHIFT; } -#define __DEFINE_OUTPUT_COPY_BODY(memcpy_func) \ +#define __DEFINE_OUTPUT_COPY_BODY(advance_buf, memcpy_func, ...) \ { \ unsigned long size, written; \ \ do { \ size = min(handle->size, len); \ - written = memcpy_func(handle->addr, buf, size); \ + written = memcpy_func(__VA_ARGS__); \ written = size - written; \ \ len -= written; \ handle->addr += written; \ - buf += written; \ + if (advance_buf) \ + buf += written; \ handle->size -= written; \ if (!handle->size) { \ struct ring_buffer *rb = handle->rb; \ @@ -153,12 +154,16 @@ static inline unsigned long perf_aux_size(struct ring_buffer *rb) static inline unsigned long \ func_name(struct perf_output_handle *handle, \ const void *buf, unsigned long len) \ -__DEFINE_OUTPUT_COPY_BODY(memcpy_func) +__DEFINE_OUTPUT_COPY_BODY(true, memcpy_func, handle->addr, buf, size) static inline unsigned long __output_custom(struct perf_output_handle *handle, perf_copy_f copy_func, const void *buf, unsigned long len) -__DEFINE_OUTPUT_COPY_BODY(copy_func) +{ + unsigned long orig_len = len; + __DEFINE_OUTPUT_COPY_BODY(false, copy_func, handle->addr, buf, + orig_len - len, size) +} static inline unsigned long memcpy_common(void *dst, const void *src, unsigned long n) diff --git a/net/core/filter.c b/net/core/filter.c index 0b521353008d..5708999f8a79 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2026,9 +2026,9 @@ bool bpf_helper_changes_skb_data(void *func) } static unsigned long bpf_skb_copy(void *dst_buff, const void *skb, - unsigned long len) + unsigned long off, unsigned long len) { - void *ptr = skb_header_pointer(skb, 0, len, dst_buff); + void *ptr = skb_header_pointer(skb, off, len, dst_buff); if (unlikely(!ptr)) return len; -- cgit v1.2.3 From ae76715d153e33c249b6850361e4d8d775388b5a Mon Sep 17 00:00:00 2001 From: Hadar Hen Zion Date: Sun, 24 Jul 2016 16:12:39 +0300 Subject: net/mlx5e: Check the minimum inline header mode before xmit Each send queue (SQ) has inline mode that defines the minimal required inline headers in the SQ WQE. Before sending each packet check that the minimum required headers on the WQE are copied. Signed-off-by: Hadar Hen Zion Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 1 + drivers/net/ethernet/mellanox/mlx5/core/en_tx.c | 49 +++++++++++++++++++++++-- include/linux/mlx5/device.h | 7 ++++ 3 files changed, 53 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 4cbd452fec25..2c20c7b14043 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -398,6 +398,7 @@ struct mlx5e_sq { u32 sqn; u16 bf_buf_size; u16 max_inline; + u8 min_inline_mode; u16 edge; struct device *pdev; struct mlx5e_tstamp *tstamp; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index 5740b465ef84..e073bf59890d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -128,6 +128,50 @@ u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb, return priv->channeltc_to_txq_map[channel_ix][up]; } +static inline int mlx5e_skb_l2_header_offset(struct sk_buff *skb) +{ +#define MLX5E_MIN_INLINE (ETH_HLEN + VLAN_HLEN) + + return max(skb_network_offset(skb), MLX5E_MIN_INLINE); +} + +static inline int mlx5e_skb_l3_header_offset(struct sk_buff *skb) +{ + struct flow_keys keys; + + if (skb_transport_header_was_set(skb)) + return skb_transport_offset(skb); + else if (skb_flow_dissect_flow_keys(skb, &keys, 0)) + return keys.control.thoff; + else + return mlx5e_skb_l2_header_offset(skb); +} + +static inline unsigned int mlx5e_calc_min_inline(enum mlx5_inline_modes mode, + struct sk_buff *skb) +{ + int hlen; + + switch (mode) { + case MLX5_INLINE_MODE_TCP_UDP: + hlen = eth_get_headlen(skb->data, skb_headlen(skb)); + if (hlen == ETH_HLEN && !skb_vlan_tag_present(skb)) + hlen += VLAN_HLEN; + return hlen; + case MLX5_INLINE_MODE_IP: + /* When transport header is set to zero, it means no transport + * header. When transport header is set to 0xff's, it means + * transport header wasn't set. + */ + if (skb_transport_offset(skb)) + return mlx5e_skb_l3_header_offset(skb); + /* fall through */ + case MLX5_INLINE_MODE_L2: + default: + return mlx5e_skb_l2_header_offset(skb); + } +} + static inline u16 mlx5e_get_inline_hdr_size(struct mlx5e_sq *sq, struct sk_buff *skb, bool bf) { @@ -135,8 +179,6 @@ static inline u16 mlx5e_get_inline_hdr_size(struct mlx5e_sq *sq, * headers and occur before the data gather. * Therefore these headers must be copied into the WQE */ -#define MLX5E_MIN_INLINE (ETH_HLEN + VLAN_HLEN) - if (bf) { u16 ihs = skb_headlen(skb); @@ -146,8 +188,7 @@ static inline u16 mlx5e_get_inline_hdr_size(struct mlx5e_sq *sq, if (ihs <= sq->max_inline) return skb_headlen(skb); } - - return max(skb_network_offset(skb), MLX5E_MIN_INLINE); + return mlx5e_calc_min_inline(sq->min_inline_mode, skb); } static inline void mlx5e_tx_skb_pull_inline(unsigned char **skb_data, diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index e0a3ed758287..0b6d15cddb2f 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -129,6 +129,13 @@ __mlx5_mask(typ, fld)) tmp; \ }) +enum mlx5_inline_modes { + MLX5_INLINE_MODE_NONE, + MLX5_INLINE_MODE_L2, + MLX5_INLINE_MODE_IP, + MLX5_INLINE_MODE_TCP_UDP, +}; + enum { MLX5_MAX_COMMANDS = 32, MLX5_CMD_DATA_BLOCK_SIZE = 512, -- cgit v1.2.3 From cff92d7c7ebd7ceddd4def6b39e0302585b1eb14 Mon Sep 17 00:00:00 2001 From: Hadar Hen Zion Date: Sun, 24 Jul 2016 16:12:40 +0300 Subject: net/mlx5e: Query minimum required header copy during xmit Add support for query the minimum inline mode from the Firmware. It is required for correct TX steering according to L3/L4 packet headers. Each send queue (SQ) has inline mode that defines the minimal required headers that needs to be copied into the SQ WQE. The driver asks the Firmware for the wqe_inline_mode device capability value. In case the device capability defined as "vport context" the driver must check the reported min inline mode from the vport context before creating its SQs. Signed-off-by: Hadar Hen Zion Signed-off-by: Saeed Mahameed Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx5/core/en.h | 7 +++++++ drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 24 +++++++++++++++++++++++ drivers/net/ethernet/mellanox/mlx5/core/vport.c | 12 ++++++++++++ include/linux/mlx5/mlx5_ifc.h | 10 +++++++--- include/linux/mlx5/vport.h | 2 ++ 5 files changed, 52 insertions(+), 3 deletions(-) (limited to 'include/linux') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 2c20c7b14043..1b495efa7490 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -129,6 +129,12 @@ static inline int mlx5_max_log_rq_size(int wq_type) } } +enum { + MLX5E_INLINE_MODE_L2, + MLX5E_INLINE_MODE_VPORT_CONTEXT, + MLX5_INLINE_MODE_NOT_REQUIRED, +}; + struct mlx5e_tx_wqe { struct mlx5_wqe_ctrl_seg ctrl; struct mlx5_wqe_eth_seg eth; @@ -188,6 +194,7 @@ struct mlx5e_params { bool lro_en; u32 lro_wqe_sz; u16 tx_max_inline; + u8 tx_min_inline_mode; u8 rss_hfunc; u8 toeplitz_hash_key[40]; u32 indirection_rqt[MLX5E_INDIR_RQT_SIZE]; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index bdcb699fe63e..870bea37c57c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -56,6 +56,7 @@ struct mlx5e_sq_param { u32 sqc[MLX5_ST_SZ_DW(sqc)]; struct mlx5_wq_param wq; u16 max_inline; + u8 min_inline_mode; bool icosq; }; @@ -649,6 +650,9 @@ static int mlx5e_create_sq(struct mlx5e_channel *c, } sq->bf_buf_size = (1 << MLX5_CAP_GEN(mdev, log_bf_reg_size)) / 2; sq->max_inline = param->max_inline; + sq->min_inline_mode = + MLX5_CAP_ETH(mdev, wqe_inline_mode) == MLX5E_INLINE_MODE_VPORT_CONTEXT ? + param->min_inline_mode : 0; err = mlx5e_alloc_sq_db(sq, cpu_to_node(c->cpu)); if (err) @@ -731,6 +735,7 @@ static int mlx5e_enable_sq(struct mlx5e_sq *sq, struct mlx5e_sq_param *param) MLX5_SET(sqc, sqc, tis_num_0, param->icosq ? 0 : priv->tisn[sq->tc]); MLX5_SET(sqc, sqc, cqn, sq->cq.mcq.cqn); + MLX5_SET(sqc, sqc, min_wqe_inline_mode, sq->min_inline_mode); MLX5_SET(sqc, sqc, state, MLX5_SQC_STATE_RST); MLX5_SET(sqc, sqc, tis_lst_sz, param->icosq ? 0 : 1); MLX5_SET(sqc, sqc, flush_in_error_en, 1); @@ -1343,6 +1348,7 @@ static void mlx5e_build_sq_param(struct mlx5e_priv *priv, MLX5_SET(wq, wq, log_wq_sz, priv->params.log_sq_size); param->max_inline = priv->params.tx_max_inline; + param->min_inline_mode = priv->params.tx_min_inline_mode; } static void mlx5e_build_common_cq_param(struct mlx5e_priv *priv, @@ -2978,6 +2984,23 @@ void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) MLX5E_PARAMS_DEFAULT_RX_CQ_MODERATION_USEC_FROM_CQE; } +static void mlx5e_query_min_inline(struct mlx5_core_dev *mdev, + u8 *min_inline_mode) +{ + switch (MLX5_CAP_ETH(mdev, wqe_inline_mode)) { + case MLX5E_INLINE_MODE_L2: + *min_inline_mode = MLX5_INLINE_MODE_L2; + break; + case MLX5E_INLINE_MODE_VPORT_CONTEXT: + mlx5_query_nic_vport_min_inline(mdev, + min_inline_mode); + break; + case MLX5_INLINE_MODE_NOT_REQUIRED: + *min_inline_mode = MLX5_INLINE_MODE_NONE; + break; + } +} + static void mlx5e_build_nic_netdev_priv(struct mlx5_core_dev *mdev, struct net_device *netdev, const struct mlx5e_profile *profile, @@ -3043,6 +3066,7 @@ static void mlx5e_build_nic_netdev_priv(struct mlx5_core_dev *mdev, priv->params.tx_cq_moderation.pkts = MLX5E_PARAMS_DEFAULT_TX_CQ_MODERATION_PKTS; priv->params.tx_max_inline = mlx5e_get_max_inline_cap(mdev); + mlx5e_query_min_inline(mdev, &priv->params.tx_min_inline_mode); priv->params.num_tc = 1; priv->params.rss_hfunc = ETH_RSS_HASH_XOR; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index 91846dfcbe9c..21365d06982b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -135,6 +135,18 @@ static int mlx5_modify_nic_vport_context(struct mlx5_core_dev *mdev, void *in, return mlx5_cmd_exec_check_status(mdev, in, inlen, out, sizeof(out)); } +void mlx5_query_nic_vport_min_inline(struct mlx5_core_dev *mdev, + u8 *min_inline_mode) +{ + u32 out[MLX5_ST_SZ_DW(query_nic_vport_context_out)] = {0}; + + mlx5_query_nic_vport_context(mdev, 0, out, sizeof(out)); + + *min_inline_mode = MLX5_GET(query_nic_vport_context_out, out, + nic_vport_context.min_wqe_inline_mode); +} +EXPORT_SYMBOL_GPL(mlx5_query_nic_vport_min_inline); + int mlx5_query_nic_vport_mac_address(struct mlx5_core_dev *mdev, u16 vport, u8 *addr) { diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index d671e4e8e7db..21bc4557b67a 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -536,7 +536,8 @@ struct mlx5_ifc_per_protocol_networking_offload_caps_bits { u8 self_lb_en_modifiable[0x1]; u8 reserved_at_9[0x2]; u8 max_lso_cap[0x5]; - u8 reserved_at_10[0x4]; + u8 reserved_at_10[0x2]; + u8 wqe_inline_mode[0x2]; u8 rss_ind_tbl_cap[0x4]; u8 reg_umr_sq[0x1]; u8 scatter_fcs[0x1]; @@ -2270,7 +2271,8 @@ struct mlx5_ifc_sqc_bits { u8 cd_master[0x1]; u8 fre[0x1]; u8 flush_in_error_en[0x1]; - u8 reserved_at_4[0x4]; + u8 reserved_at_4[0x1]; + u8 min_wqe_inline_mode[0x3]; u8 state[0x4]; u8 reg_umr[0x1]; u8 reserved_at_d[0x13]; @@ -2367,7 +2369,9 @@ struct mlx5_ifc_rmpc_bits { }; struct mlx5_ifc_nic_vport_context_bits { - u8 reserved_at_0[0x1f]; + u8 reserved_at_0[0x5]; + u8 min_wqe_inline_mode[0x3]; + u8 reserved_at_8[0x17]; u8 roce_en[0x1]; u8 arm_change_event[0x1]; diff --git a/include/linux/mlx5/vport.h b/include/linux/mlx5/vport.h index 6c16c198f680..e087b7d047ac 100644 --- a/include/linux/mlx5/vport.h +++ b/include/linux/mlx5/vport.h @@ -43,6 +43,8 @@ int mlx5_modify_vport_admin_state(struct mlx5_core_dev *mdev, u8 opmod, u16 vport, u8 state); int mlx5_query_nic_vport_mac_address(struct mlx5_core_dev *mdev, u16 vport, u8 *addr); +void mlx5_query_nic_vport_min_inline(struct mlx5_core_dev *mdev, + u8 *min_inline); int mlx5_modify_nic_vport_mac_address(struct mlx5_core_dev *dev, u16 vport, u8 *addr); int mlx5_query_nic_vport_mtu(struct mlx5_core_dev *mdev, u16 *mtu); -- cgit v1.2.3