From owner-svn-src-head@FreeBSD.ORG Tue May 27 18:18:42 2014 Return-Path: Delivered-To: svn-src-head@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [8.8.178.115]) (using TLSv1 with cipher ADH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id B4F7A610; Tue, 27 May 2014 18:18:42 +0000 (UTC) Received: from svn.freebsd.org (svn.freebsd.org [IPv6:2001:1900:2254:2068::e6a:0]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id A08BF25BB; Tue, 27 May 2014 18:18:42 +0000 (UTC) Received: from svn.freebsd.org ([127.0.1.70]) by svn.freebsd.org (8.14.8/8.14.8) with ESMTP id s4RIIgV4045231; Tue, 27 May 2014 18:18:42 GMT (envelope-from np@svn.freebsd.org) Received: (from np@localhost) by svn.freebsd.org (8.14.8/8.14.8/Submit) id s4RIIff9045225; Tue, 27 May 2014 18:18:41 GMT (envelope-from np@svn.freebsd.org) Message-Id: <201405271818.s4RIIff9045225@svn.freebsd.org> From: Navdeep Parhar Date: Tue, 27 May 2014 18:18:41 +0000 (UTC) To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: svn commit: r266757 - in head/sys: conf dev/cxgbe dev/cxgbe/common modules/cxgbe/if_cxgbe X-SVN-Group: head MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-BeenThere: svn-src-head@freebsd.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: SVN commit messages for the src tree for head/-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 27 May 2014 18:18:42 -0000 Author: np Date: Tue May 27 18:18:41 2014 New Revision: 266757 URL: http://svnweb.freebsd.org/changeset/base/266757 Log: cxgbe(4): netmap support for Terminator 5 (T5) based 10G/40G cards. Netmap gets its own hardware-assisted virtual interface and won't take over or disrupt the "normal" interface in any way. You can use both simultaneously. For kernels with DEV_NETMAP, cxgbe(4) carves out an ncxl interface (note the 'n' prefix) in the hardware to accompany each cxl interface. These two ifnet's per port share the same wire but really are separate interfaces in the hardware and software. Each gets its own L2 MAC addresses (unicast and multicast), MTU, checksum caps, etc. You should run netmap on the 'n' interfaces only, that's what they are for. With this, pkt-gen is able to transmit > 45Mpps out of a single 40G port of a T580 card. 2 port tx is at ~56Mpps total (28M + 28M) as of now. Single port receive is at 33Mpps but this is very much a work in progress. I expect it to be closer to 40Mpps once done. In any case the current effort can already saturate multiple 10G ports of a T5 card at the smallest legal packet size. T4 gear is totally untested. trantor:~# ./pkt-gen -i ncxl0 -f tx -D 00:07:43:ab:cd:ef 881.952141 main [1621] interface is ncxl0 881.952250 extract_ip_range [275] range is 10.0.0.1:0 to 10.0.0.1:0 881.952253 extract_ip_range [275] range is 10.1.0.1:0 to 10.1.0.1:0 881.962540 main [1804] mapped 334980KB at 0x801dff000 Sending on netmap:ncxl0: 4 queues, 1 threads and 1 cpus. 10.0.0.1 -> 10.1.0.1 (00:00:00:00:00:00 -> 00:07:43:ab:cd:ef) 881.962562 main [1882] Sending 512 packets every 0.000000000 s 881.962563 main [1884] Wait 2 secs for phy reset 884.088516 main [1886] Ready... 884.088535 nm_open [457] overriding ifname ncxl0 ringid 0x0 flags 0x1 884.088607 sender_body [996] start 884.093246 sender_body [1064] drop copy 885.090435 main_thread [1418] 45206353 pps (45289533 pkts in 1001840 usec) 886.091600 main_thread [1418] 45322792 pps (45375593 pkts in 1001165 usec) 887.092435 main_thread [1418] 45313992 pps (45351784 pkts in 1000834 usec) 888.094434 main_thread [1418] 45315765 pps (45406397 pkts in 1002000 usec) 889.095434 main_thread [1418] 45333218 pps (45378551 pkts in 1001000 usec) 890.097434 main_thread [1418] 45315247 pps (45405877 pkts in 1002000 usec) 891.099434 main_thread [1418] 45326515 pps (45417168 pkts in 1002000 usec) 892.101434 main_thread [1418] 45333039 pps (45423705 pkts in 1002000 usec) 893.103434 main_thread [1418] 45324105 pps (45414708 pkts in 1001999 usec) 894.105434 main_thread [1418] 45318042 pps (45408723 pkts in 1002001 usec) 895.106434 main_thread [1418] 45332430 pps (45377762 pkts in 1001000 usec) 896.107434 main_thread [1418] 45338072 pps (45383410 pkts in 1001000 usec) ... Relnotes: Yes Sponsored by: Chelsio Communications. Added: head/sys/dev/cxgbe/t4_netmap.c (contents, props changed) Modified: head/sys/conf/files head/sys/dev/cxgbe/adapter.h head/sys/dev/cxgbe/common/common.h head/sys/dev/cxgbe/common/t4_hw.c head/sys/dev/cxgbe/t4_main.c head/sys/dev/cxgbe/t4_sge.c head/sys/modules/cxgbe/if_cxgbe/Makefile Modified: head/sys/conf/files ============================================================================== --- head/sys/conf/files Tue May 27 16:30:54 2014 (r266756) +++ head/sys/conf/files Tue May 27 18:18:41 2014 (r266757) @@ -1127,6 +1127,8 @@ dev/cxgb/cxgb_t3fw.c optional cxgb cxgb compile-with "${NORMAL_C} -I$S/dev/cxgb" dev/cxgbe/t4_main.c optional cxgbe pci \ compile-with "${NORMAL_C} -I$S/dev/cxgbe" +dev/cxgbe/t4_netmap.c optional cxgbe pci \ + compile-with "${NORMAL_C} -I$S/dev/cxgbe" dev/cxgbe/t4_sge.c optional cxgbe pci \ compile-with "${NORMAL_C} -I$S/dev/cxgbe" dev/cxgbe/t4_l2t.c optional cxgbe pci \ Modified: head/sys/dev/cxgbe/adapter.h ============================================================================== --- head/sys/dev/cxgbe/adapter.h Tue May 27 16:30:54 2014 (r266756) +++ head/sys/dev/cxgbe/adapter.h Tue May 27 18:18:41 2014 (r266757) @@ -54,6 +54,7 @@ #include #include "offload.h" +#include "common/t4_msg.h" #include "firmware/t4fw_interface.h" MALLOC_DECLARE(M_CXGBE); @@ -131,6 +132,7 @@ enum { RX_IQ_ESIZE = 64, /* At least 64 so CPL_RX_PKT will fit */ EQ_ESIZE = 64, /* All egress queues use this entry size */ + SGE_MAX_WR_NDESC = SGE_MAX_WR_LEN / EQ_ESIZE, /* max WR size in desc */ RX_FL_ESIZE = EQ_ESIZE, /* 8 64bit addresses */ #if MJUMPAGESIZE != MCLBYTES @@ -155,6 +157,17 @@ enum { }; enum { + XGMAC_MTU = (1 << 0), + XGMAC_PROMISC = (1 << 1), + XGMAC_ALLMULTI = (1 << 2), + XGMAC_VLANEX = (1 << 3), + XGMAC_UCADDR = (1 << 4), + XGMAC_MCADDRS = (1 << 5), + + XGMAC_ALL = 0xffff +}; + +enum { /* flags understood by begin_synchronized_op */ HOLD_LOCK = (1 << 0), SLEEP_OK = (1 << 1), @@ -168,7 +181,7 @@ enum { /* adapter flags */ FULL_INIT_DONE = (1 << 0), FW_OK = (1 << 1), - INTR_DIRECT = (1 << 2), /* direct interrupts for everything */ + /* INTR_DIRECT = (1 << 2), No longer used. */ MASTER_PF = (1 << 3), ADAP_SYSCTL_CTX = (1 << 4), TOM_INIT_DONE = (1 << 5), @@ -181,6 +194,10 @@ enum { PORT_INIT_DONE = (1 << 1), PORT_SYSCTL_CTX = (1 << 2), HAS_TRACEQ = (1 << 3), + INTR_RXQ = (1 << 4), /* All NIC rxq's take interrupts */ + INTR_OFLD_RXQ = (1 << 5), /* All TOE rxq's take interrupts */ + INTR_NM_RXQ = (1 << 6), /* All netmap rxq's take interrupts */ + INTR_ALL = (INTR_RXQ | INTR_OFLD_RXQ | INTR_NM_RXQ), }; #define IS_DOOMED(pi) ((pi)->flags & DOOMED) @@ -225,6 +242,19 @@ struct port_info { int nofldrxq; /* # of offload rx queues */ int first_ofld_rxq; /* index of first offload rx queue */ #endif +#ifdef DEV_NETMAP + int nnmtxq; /* # of netmap tx queues */ + int first_nm_txq; /* index of first netmap tx queue */ + int nnmrxq; /* # of netmap rx queues */ + int first_nm_rxq; /* index of first netmap rx queue */ + + struct ifnet *nm_ifp; + struct ifmedia nm_media; + int nmif_flags; + uint16_t nm_viid; + int16_t nm_xact_addr_filt; + uint16_t nm_rss_size; /* size of netmap VI's RSS table slice */ +#endif int tmr_idx; int pktc_idx; int qsize_rxq; @@ -362,7 +392,7 @@ struct sge_eq { struct tx_desc *desc; /* KVA of descriptor ring */ bus_addr_t ba; /* bus address of descriptor ring */ struct sge_qstat *spg; /* status page, for convenience */ - int doorbells; + uint16_t doorbells; volatile uint32_t *udb; /* KVA of doorbell (lies within BAR2) */ u_int udb_qid; /* relative qid within the doorbell page */ uint16_t cap; /* max # of desc, for convenience */ @@ -538,6 +568,77 @@ struct sge_wrq { uint32_t no_desc; /* out of hardware descriptors */ } __aligned(CACHE_LINE_SIZE); + +#ifdef DEV_NETMAP +#define CPL_PAD (RX_IQ_ESIZE - sizeof(struct rsp_ctrl) - \ + sizeof(struct rss_header)) +struct nm_iq_desc { + struct rss_header rss; + union { + uint8_t cpl[CPL_PAD]; + struct cpl_fw6_msg fw6_msg; + struct cpl_rx_pkt rx_pkt; + } u; + struct rsp_ctrl rsp; +}; +CTASSERT(sizeof(struct nm_iq_desc) == RX_IQ_ESIZE); + +struct sge_nm_rxq { + struct port_info *pi; + + struct nm_iq_desc *iq_desc; + uint16_t iq_abs_id; + uint16_t iq_cntxt_id; + uint16_t iq_cidx; + uint16_t iq_sidx; + uint8_t iq_gen; + + __be64 *fl_desc; + uint16_t fl_cntxt_id; + uint32_t fl_cidx; + uint32_t fl_pidx; + uint32_t fl_sidx; + uint32_t fl_db_val; + u_int fl_hwidx:4; + + u_int nid; /* netmap ring # for this queue */ + + /* infrequently used items after this */ + + bus_dma_tag_t iq_desc_tag; + bus_dmamap_t iq_desc_map; + bus_addr_t iq_ba; + int intr_idx; + + bus_dma_tag_t fl_desc_tag; + bus_dmamap_t fl_desc_map; + bus_addr_t fl_ba; +} __aligned(CACHE_LINE_SIZE); + +struct sge_nm_txq { + struct tx_desc *desc; + uint16_t cidx; + uint16_t pidx; + uint16_t sidx; + uint16_t equiqidx; /* EQUIQ last requested at this pidx */ + uint16_t equeqidx; /* EQUEQ last requested at this pidx */ + uint16_t dbidx; /* pidx of the most recent doorbell */ + uint16_t doorbells; + volatile uint32_t *udb; + u_int udb_qid; + u_int cntxt_id; + __be32 cpl_ctrl0; /* for convenience */ + u_int nid; /* netmap ring # for this queue */ + + /* infrequently used items after this */ + + bus_dma_tag_t desc_tag; + bus_dmamap_t desc_map; + bus_addr_t ba; + int iqidx; +} __aligned(CACHE_LINE_SIZE); +#endif + struct sge { int timer_val[SGE_NTIMERS]; int counter_val[SGE_NCOUNTERS]; @@ -552,6 +653,10 @@ struct sge { int nofldrxq; /* total # of TOE rx queues */ int nofldtxq; /* total # of TOE tx queues */ #endif +#ifdef DEV_NETMAP + int nnmrxq; /* total # of netmap rx queues */ + int nnmtxq; /* total # of netmap tx queues */ +#endif int niq; /* total # of ingress queues */ int neq; /* total # of egress queues */ @@ -564,6 +669,10 @@ struct sge { struct sge_wrq *ofld_txq; /* TOE tx queues */ struct sge_ofld_rxq *ofld_rxq; /* TOE rx queues */ #endif +#ifdef DEV_NETMAP + struct sge_nm_txq *nm_txq; /* netmap tx queues */ + struct sge_nm_rxq *nm_rxq; /* netmap rx queues */ +#endif uint16_t iq_start; int eq_start; @@ -629,7 +738,7 @@ struct adapter { struct l2t_data *l2t; /* L2 table */ struct tid_info tids; - int doorbells; + uint16_t doorbells; int open_device_map; #ifdef TCP_OFFLOAD int offload_map; @@ -730,6 +839,12 @@ struct adapter { #define for_each_ofld_rxq(pi, iter, q) \ for (q = &pi->adapter->sge.ofld_rxq[pi->first_ofld_rxq], iter = 0; \ iter < pi->nofldrxq; ++iter, ++q) +#define for_each_nm_txq(pi, iter, q) \ + for (q = &pi->adapter->sge.nm_txq[pi->first_nm_txq], iter = 0; \ + iter < pi->nnmtxq; ++iter, ++q) +#define for_each_nm_rxq(pi, iter, q) \ + for (q = &pi->adapter->sge.nm_rxq[pi->first_nm_rxq], iter = 0; \ + iter < pi->nnmrxq; ++iter, ++q) /* One for errors, one for firmware events */ #define T4_EXTRA_INTR 2 @@ -854,6 +969,18 @@ int t4_register_fw_msg_handler(struct ad int t4_filter_rpl(struct sge_iq *, const struct rss_header *, struct mbuf *); int begin_synchronized_op(struct adapter *, struct port_info *, int, char *); void end_synchronized_op(struct adapter *, int); +int update_mac_settings(struct ifnet *, int); +int adapter_full_init(struct adapter *); +int adapter_full_uninit(struct adapter *); +int port_full_init(struct port_info *); +int port_full_uninit(struct port_info *); + +#ifdef DEV_NETMAP +/* t4_netmap.c */ +int create_netmap_ifnet(struct port_info *); +int destroy_netmap_ifnet(struct port_info *); +void t4_nm_intr(void *); +#endif /* t4_sge.c */ void t4_sge_modload(void); Modified: head/sys/dev/cxgbe/common/common.h ============================================================================== --- head/sys/dev/cxgbe/common/common.h Tue May 27 16:30:54 2014 (r266756) +++ head/sys/dev/cxgbe/common/common.h Tue May 27 18:18:41 2014 (r266757) @@ -561,11 +561,11 @@ int t4_cfg_pfvf(struct adapter *adap, un unsigned int exactf, unsigned int rcaps, unsigned int wxcaps); int t4_alloc_vi_func(struct adapter *adap, unsigned int mbox, unsigned int port, unsigned int pf, unsigned int vf, - unsigned int nmac, u8 *mac, unsigned int *rss_size, + unsigned int nmac, u8 *mac, u16 *rss_size, unsigned int portfunc, unsigned int idstype); int t4_alloc_vi(struct adapter *adap, unsigned int mbox, unsigned int port, unsigned int pf, unsigned int vf, unsigned int nmac, u8 *mac, - unsigned int *rss_size); + u16 *rss_size); int t4_free_vi(struct adapter *adap, unsigned int mbox, unsigned int pf, unsigned int vf, unsigned int viid); @@ -581,6 +581,8 @@ int t4_set_addr_hash(struct adapter *ada bool ucast, u64 vec, bool sleep_ok); int t4_enable_vi(struct adapter *adap, unsigned int mbox, unsigned int viid, bool rx_en, bool tx_en); +int t4_enable_vi_ns(struct adapter *adap, unsigned int mbox, unsigned int viid, + bool rx_en, bool tx_en); int t4_identify_port(struct adapter *adap, unsigned int mbox, unsigned int viid, unsigned int nblinks); int t4_i2c_rd(struct adapter *adap, unsigned int mbox, unsigned int port_id, Modified: head/sys/dev/cxgbe/common/t4_hw.c ============================================================================== --- head/sys/dev/cxgbe/common/t4_hw.c Tue May 27 16:30:54 2014 (r266756) +++ head/sys/dev/cxgbe/common/t4_hw.c Tue May 27 18:18:41 2014 (r266757) @@ -4826,7 +4826,7 @@ int t4_cfg_pfvf(struct adapter *adap, un */ int t4_alloc_vi_func(struct adapter *adap, unsigned int mbox, unsigned int port, unsigned int pf, unsigned int vf, - unsigned int nmac, u8 *mac, unsigned int *rss_size, + unsigned int nmac, u8 *mac, u16 *rss_size, unsigned int portfunc, unsigned int idstype) { int ret; @@ -4881,7 +4881,7 @@ int t4_alloc_vi_func(struct adapter *ada */ int t4_alloc_vi(struct adapter *adap, unsigned int mbox, unsigned int port, unsigned int pf, unsigned int vf, unsigned int nmac, u8 *mac, - unsigned int *rss_size) + u16 *rss_size) { return t4_alloc_vi_func(adap, mbox, port, pf, vf, nmac, mac, rss_size, FW_VI_FUNC_ETH, 0); @@ -5155,6 +5155,19 @@ int t4_enable_vi(struct adapter *adap, u return t4_wr_mbox(adap, mbox, &c, sizeof(c), NULL); } +int t4_enable_vi_ns(struct adapter *adap, unsigned int mbox, unsigned int viid, + bool rx_en, bool tx_en) +{ + struct fw_vi_enable_cmd c; + + memset(&c, 0, sizeof(c)); + c.op_to_viid = htonl(V_FW_CMD_OP(FW_VI_ENABLE_CMD) | F_FW_CMD_REQUEST | + F_FW_CMD_EXEC | V_FW_VI_ENABLE_CMD_VIID(viid)); + c.ien_to_len16 = htonl(V_FW_VI_ENABLE_CMD_IEN(rx_en) | + V_FW_VI_ENABLE_CMD_EEN(tx_en) | FW_LEN16(c)); + return t4_wr_mbox_ns(adap, mbox, &c, sizeof(c), NULL); +} + /** * t4_identify_port - identify a VI's port by blinking its LED * @adap: the adapter @@ -5623,7 +5636,7 @@ int __devinit t4_port_init(struct port_i u8 addr[6]; int ret, i, j; struct fw_port_cmd c; - unsigned int rss_size; + u16 rss_size; adapter_t *adap = p->adapter; memset(&c, 0, sizeof(c)); Modified: head/sys/dev/cxgbe/t4_main.c ============================================================================== --- head/sys/dev/cxgbe/t4_main.c Tue May 27 16:30:54 2014 (r266756) +++ head/sys/dev/cxgbe/t4_main.c Tue May 27 18:18:41 2014 (r266757) @@ -218,6 +218,24 @@ static int t4_nofldrxq1g = -1; TUNABLE_INT("hw.cxgbe.nofldrxq1g", &t4_nofldrxq1g); #endif +#ifdef DEV_NETMAP +#define NNMTXQ_10G 2 +static int t4_nnmtxq10g = -1; +TUNABLE_INT("hw.cxgbe.nnmtxq10g", &t4_nnmtxq10g); + +#define NNMRXQ_10G 2 +static int t4_nnmrxq10g = -1; +TUNABLE_INT("hw.cxgbe.nnmrxq10g", &t4_nnmrxq10g); + +#define NNMTXQ_1G 1 +static int t4_nnmtxq1g = -1; +TUNABLE_INT("hw.cxgbe.nnmtxq1g", &t4_nnmtxq1g); + +#define NNMRXQ_1G 1 +static int t4_nnmrxq1g = -1; +TUNABLE_INT("hw.cxgbe.nnmrxq1g", &t4_nnmrxq1g); +#endif + /* * Holdoff parameters for 10G and 1G ports. */ @@ -295,19 +313,26 @@ static int t5_write_combine = 0; TUNABLE_INT("hw.cxl.write_combine", &t5_write_combine); struct intrs_and_queues { - int intr_type; /* INTx, MSI, or MSI-X */ - int nirq; /* Number of vectors */ - int intr_flags; - int ntxq10g; /* # of NIC txq's for each 10G port */ - int nrxq10g; /* # of NIC rxq's for each 10G port */ - int ntxq1g; /* # of NIC txq's for each 1G port */ - int nrxq1g; /* # of NIC rxq's for each 1G port */ - int rsrv_noflowq; /* Flag whether to reserve queue 0 */ + uint16_t intr_type; /* INTx, MSI, or MSI-X */ + uint16_t nirq; /* Total # of vectors */ + uint16_t intr_flags_10g;/* Interrupt flags for each 10G port */ + uint16_t intr_flags_1g; /* Interrupt flags for each 1G port */ + uint16_t ntxq10g; /* # of NIC txq's for each 10G port */ + uint16_t nrxq10g; /* # of NIC rxq's for each 10G port */ + uint16_t ntxq1g; /* # of NIC txq's for each 1G port */ + uint16_t nrxq1g; /* # of NIC rxq's for each 1G port */ + uint16_t rsrv_noflowq; /* Flag whether to reserve queue 0 */ #ifdef TCP_OFFLOAD - int nofldtxq10g; /* # of TOE txq's for each 10G port */ - int nofldrxq10g; /* # of TOE rxq's for each 10G port */ - int nofldtxq1g; /* # of TOE txq's for each 1G port */ - int nofldrxq1g; /* # of TOE rxq's for each 1G port */ + uint16_t nofldtxq10g; /* # of TOE txq's for each 10G port */ + uint16_t nofldrxq10g; /* # of TOE rxq's for each 10G port */ + uint16_t nofldtxq1g; /* # of TOE txq's for each 1G port */ + uint16_t nofldrxq1g; /* # of TOE rxq's for each 1G port */ +#endif +#ifdef DEV_NETMAP + uint16_t nnmtxq10g; /* # of netmap txq's for each 10G port */ + uint16_t nnmrxq10g; /* # of netmap rxq's for each 10G port */ + uint16_t nnmtxq1g; /* # of netmap txq's for each 1G port */ + uint16_t nnmrxq1g; /* # of netmap rxq's for each 1G port */ #endif }; @@ -321,17 +346,6 @@ struct filter_entry { struct t4_filter_specification fs; }; -enum { - XGMAC_MTU = (1 << 0), - XGMAC_PROMISC = (1 << 1), - XGMAC_ALLMULTI = (1 << 2), - XGMAC_VLANEX = (1 << 3), - XGMAC_UCADDR = (1 << 4), - XGMAC_MCADDRS = (1 << 5), - - XGMAC_ALL = 0xffff -}; - static int map_bars_0_and_4(struct adapter *); static int map_bar_2(struct adapter *); static void setup_memwin(struct adapter *); @@ -350,15 +364,10 @@ static int get_params__pre_init(struct a static int get_params__post_init(struct adapter *); static int set_params__post_init(struct adapter *); static void t4_set_desc(struct adapter *); -static void build_medialist(struct port_info *); -static int update_mac_settings(struct port_info *, int); +static void build_medialist(struct port_info *, struct ifmedia *); static int cxgbe_init_synchronized(struct port_info *); static int cxgbe_uninit_synchronized(struct port_info *); static int setup_intr_handlers(struct adapter *); -static int adapter_full_init(struct adapter *); -static int adapter_full_uninit(struct adapter *); -static int port_full_init(struct port_info *); -static int port_full_uninit(struct port_info *); static void quiesce_eq(struct adapter *, struct sge_eq *); static void quiesce_iq(struct adapter *, struct sge_iq *); static void quiesce_fl(struct adapter *, struct sge_fl *); @@ -556,6 +565,9 @@ t4_attach(device_t dev) #ifdef TCP_OFFLOAD int ofld_rqidx, ofld_tqidx; #endif +#ifdef DEV_NETMAP + int nm_rqidx, nm_tqidx; +#endif const char *pcie_ts; sc = device_get_softc(dev); @@ -685,6 +697,13 @@ t4_attach(device_t dev) sc->port[i] = NULL; goto done; } + rc = -t4_link_start(sc, sc->mbox, pi->tx_chan, &pi->link_cfg); + if (rc != 0) { + device_printf(dev, "port %d l1cfg failed: %d\n", i, rc); + free(pi, M_CXGBE); + sc->port[i] = NULL; + goto done; + } snprintf(pi->lockname, sizeof(pi->lockname), "%sp%d", device_get_nameunit(dev), i); @@ -726,7 +745,6 @@ t4_attach(device_t dev) sc->intr_type = iaq.intr_type; sc->intr_count = iaq.nirq; - sc->flags |= iaq.intr_flags; s = &sc->sge; s->nrxq = n10g * iaq.nrxq10g + n1g * iaq.nrxq1g; @@ -734,10 +752,8 @@ t4_attach(device_t dev) s->neq = s->ntxq + s->nrxq; /* the free list in an rxq is an eq */ s->neq += sc->params.nports + 1;/* ctrl queues: 1 per port + 1 mgmt */ s->niq = s->nrxq + 1; /* 1 extra for firmware event queue */ - #ifdef TCP_OFFLOAD if (is_offload(sc)) { - s->nofldrxq = n10g * iaq.nofldrxq10g + n1g * iaq.nofldrxq1g; s->nofldtxq = n10g * iaq.nofldtxq10g + n1g * iaq.nofldtxq1g; s->neq += s->nofldtxq + s->nofldrxq; @@ -749,6 +765,17 @@ t4_attach(device_t dev) M_CXGBE, M_ZERO | M_WAITOK); } #endif +#ifdef DEV_NETMAP + s->nnmrxq = n10g * iaq.nnmrxq10g + n1g * iaq.nnmrxq1g; + s->nnmtxq = n10g * iaq.nnmtxq10g + n1g * iaq.nnmtxq1g; + s->neq += s->nnmtxq + s->nnmrxq; + s->niq += s->nnmrxq; + + s->nm_rxq = malloc(s->nnmrxq * sizeof(struct sge_nm_rxq), + M_CXGBE, M_ZERO | M_WAITOK); + s->nm_txq = malloc(s->nnmtxq * sizeof(struct sge_nm_txq), + M_CXGBE, M_ZERO | M_WAITOK); +#endif s->ctrlq = malloc(sc->params.nports * sizeof(struct sge_wrq), M_CXGBE, M_ZERO | M_WAITOK); @@ -774,6 +801,9 @@ t4_attach(device_t dev) #ifdef TCP_OFFLOAD ofld_rqidx = ofld_tqidx = 0; #endif +#ifdef DEV_NETMAP + nm_rqidx = nm_tqidx = 0; +#endif for_each_port(sc, i) { struct port_info *pi = sc->port[i]; @@ -783,9 +813,11 @@ t4_attach(device_t dev) pi->first_rxq = rqidx; pi->first_txq = tqidx; if (is_10G_port(pi) || is_40G_port(pi)) { + pi->flags |= iaq.intr_flags_10g; pi->nrxq = iaq.nrxq10g; pi->ntxq = iaq.ntxq10g; } else { + pi->flags |= iaq.intr_flags_1g; pi->nrxq = iaq.nrxq1g; pi->ntxq = iaq.ntxq1g; } @@ -797,7 +829,6 @@ t4_attach(device_t dev) rqidx += pi->nrxq; tqidx += pi->ntxq; - #ifdef TCP_OFFLOAD if (is_offload(sc)) { pi->first_ofld_rxq = ofld_rqidx; @@ -813,6 +844,19 @@ t4_attach(device_t dev) ofld_tqidx += pi->nofldtxq; } #endif +#ifdef DEV_NETMAP + pi->first_nm_rxq = nm_rqidx; + pi->first_nm_txq = nm_tqidx; + if (is_10G_port(pi) || is_40G_port(pi)) { + pi->nnmrxq = iaq.nnmrxq10g; + pi->nnmtxq = iaq.nnmtxq10g; + } else { + pi->nnmrxq = iaq.nnmrxq1g; + pi->nnmtxq = iaq.nnmtxq1g; + } + nm_rqidx += pi->nnmrxq; + nm_tqidx += pi->nnmtxq; +#endif } rc = setup_intr_handlers(sc); @@ -902,7 +946,7 @@ t4_detach(device_t dev) for (i = 0; i < MAX_NPORTS; i++) { pi = sc->port[i]; if (pi) { - t4_free_vi(pi->adapter, sc->mbox, sc->pf, 0, pi->viid); + t4_free_vi(sc, sc->mbox, sc->pf, 0, pi->viid); if (pi->dev) device_delete_child(dev, pi->dev); @@ -939,6 +983,10 @@ t4_detach(device_t dev) free(sc->sge.ofld_rxq, M_CXGBE); free(sc->sge.ofld_txq, M_CXGBE); #endif +#ifdef DEV_NETMAP + free(sc->sge.nm_rxq, M_CXGBE); + free(sc->sge.nm_txq, M_CXGBE); +#endif free(sc->irq, M_CXGBE); free(sc->sge.rxq, M_CXGBE); free(sc->sge.txq, M_CXGBE); @@ -966,7 +1014,6 @@ t4_detach(device_t dev) return (0); } - static int cxgbe_probe(device_t dev) { @@ -989,6 +1036,8 @@ cxgbe_attach(device_t dev) { struct port_info *pi = device_get_softc(dev); struct ifnet *ifp; + char *s; + int n, o; /* Allocate an ifnet and set it up */ ifp = if_alloc(IFT_ETHER); @@ -1021,22 +1070,39 @@ cxgbe_attach(device_t dev) /* Initialize ifmedia for this port */ ifmedia_init(&pi->media, IFM_IMASK, cxgbe_media_change, cxgbe_media_status); - build_medialist(pi); + build_medialist(pi, &pi->media); pi->vlan_c = EVENTHANDLER_REGISTER(vlan_config, cxgbe_vlan_config, ifp, EVENTHANDLER_PRI_ANY); ether_ifattach(ifp, pi->hw_addr); + n = 128; + s = malloc(n, M_CXGBE, M_WAITOK); + o = snprintf(s, n, "%d txq, %d rxq (NIC)", pi->ntxq, pi->nrxq); + MPASS(n > o); #ifdef TCP_OFFLOAD if (is_offload(pi->adapter)) { - device_printf(dev, - "%d txq, %d rxq (NIC); %d txq, %d rxq (TOE)\n", - pi->ntxq, pi->nrxq, pi->nofldtxq, pi->nofldrxq); - } else + o += snprintf(s + o, n - o, "; %d txq, %d rxq (TOE)", + pi->nofldtxq, pi->nofldrxq); + MPASS(n > o); + } +#endif +#ifdef DEV_NETMAP + o += snprintf(s + o, n - o, "; %d txq, %d rxq (netmap)", pi->nnmtxq, + pi->nnmrxq); + MPASS(n > o); +#endif + device_printf(dev, "%s\n", s); + free(s, M_CXGBE); + +#ifdef DEV_NETMAP + /* nm_media handled here to keep implementation private to this file */ + ifmedia_init(&pi->nm_media, IFM_IMASK, cxgbe_media_change, + cxgbe_media_status); + build_medialist(pi, &pi->nm_media); + create_netmap_ifnet(pi); /* logs errors it something fails */ #endif - device_printf(dev, "%d txq, %d rxq\n", pi->ntxq, pi->nrxq); - cxgbe_sysctls(pi); return (0); @@ -1084,6 +1150,11 @@ cxgbe_detach(device_t dev) ether_ifdetach(pi->ifp); if_free(pi->ifp); +#ifdef DEV_NETMAP + /* XXXNM: equivalent of cxgbe_uninit_synchronized to ifdown nm_ifp */ + destroy_netmap_ifnet(pi); +#endif + ADAPTER_LOCK(sc); CLR_BUSY(sc); wakeup(&sc->flags); @@ -1126,7 +1197,7 @@ cxgbe_ioctl(struct ifnet *ifp, unsigned if (pi->flags & PORT_INIT_DONE) { t4_update_fl_bufsize(ifp); if (ifp->if_drv_flags & IFF_DRV_RUNNING) - rc = update_mac_settings(pi, XGMAC_MTU); + rc = update_mac_settings(ifp, XGMAC_MTU); } end_synchronized_op(sc, 0); break; @@ -1141,7 +1212,7 @@ cxgbe_ioctl(struct ifnet *ifp, unsigned flags = pi->if_flags; if ((ifp->if_flags ^ flags) & (IFF_PROMISC | IFF_ALLMULTI)) { - rc = update_mac_settings(pi, + rc = update_mac_settings(ifp, XGMAC_PROMISC | XGMAC_ALLMULTI); } } else @@ -1152,13 +1223,13 @@ cxgbe_ioctl(struct ifnet *ifp, unsigned end_synchronized_op(sc, 0); break; - case SIOCADDMULTI: + case SIOCADDMULTI: case SIOCDELMULTI: /* these two are called with a mutex held :-( */ rc = begin_synchronized_op(sc, pi, HOLD_LOCK, "t4multi"); if (rc) return (rc); if (ifp->if_drv_flags & IFF_DRV_RUNNING) - rc = update_mac_settings(pi, XGMAC_MCADDRS); + rc = update_mac_settings(ifp, XGMAC_MCADDRS); end_synchronized_op(sc, LOCK_HELD); break; @@ -1247,7 +1318,7 @@ cxgbe_ioctl(struct ifnet *ifp, unsigned if (mask & IFCAP_VLAN_HWTAGGING) { ifp->if_capenable ^= IFCAP_VLAN_HWTAGGING; if (ifp->if_drv_flags & IFF_DRV_RUNNING) - rc = update_mac_settings(pi, XGMAC_VLANEX); + rc = update_mac_settings(ifp, XGMAC_VLANEX); } if (mask & IFCAP_VLAN_MTU) { ifp->if_capenable ^= IFCAP_VLAN_MTU; @@ -1382,13 +1453,23 @@ static void cxgbe_media_status(struct ifnet *ifp, struct ifmediareq *ifmr) { struct port_info *pi = ifp->if_softc; - struct ifmedia_entry *cur = pi->media.ifm_cur; + struct ifmedia *media = NULL; + struct ifmedia_entry *cur; int speed = pi->link_cfg.speed; int data = (pi->port_type << 8) | pi->mod_type; + if (ifp == pi->ifp) + media = &pi->media; +#ifdef DEV_NETMAP + else if (ifp == pi->nm_ifp) + media = &pi->nm_media; +#endif + MPASS(media != NULL); + + cur = media->ifm_cur; if (cur->ifm_data != data) { - build_medialist(pi); - cur = pi->media.ifm_cur; + build_medialist(pi, media); + cur = media->ifm_cur; } ifmr->ifm_status = IFM_AVALID; @@ -1741,6 +1822,7 @@ cfg_itype_and_nqueues(struct adapter *sc { int rc, itype, navail, nrxq10g, nrxq1g, n; int nofldrxq10g = 0, nofldrxq1g = 0; + int nnmrxq10g = 0, nnmrxq1g = 0; bzero(iaq, sizeof(*iaq)); @@ -1757,6 +1839,12 @@ cfg_itype_and_nqueues(struct adapter *sc iaq->nofldrxq1g = nofldrxq1g = t4_nofldrxq1g; } #endif +#ifdef DEV_NETMAP + iaq->nnmtxq10g = t4_nnmtxq10g; + iaq->nnmtxq1g = t4_nnmtxq1g; + iaq->nnmrxq10g = nnmrxq10g = t4_nnmrxq10g; + iaq->nnmrxq1g = nnmrxq1g = t4_nnmrxq1g; +#endif for (itype = INTR_MSIX; itype; itype >>= 1) { @@ -1774,30 +1862,60 @@ restart: continue; iaq->intr_type = itype; - iaq->intr_flags = 0; + iaq->intr_flags_10g = 0; + iaq->intr_flags_1g = 0; /* * Best option: an interrupt vector for errors, one for the - * firmware event queue, and one each for each rxq (NIC as well - * as offload). + * firmware event queue, and one for every rxq (NIC, TOE, and + * netmap). */ iaq->nirq = T4_EXTRA_INTR; - iaq->nirq += n10g * (nrxq10g + nofldrxq10g); - iaq->nirq += n1g * (nrxq1g + nofldrxq1g); + iaq->nirq += n10g * (nrxq10g + nofldrxq10g + nnmrxq10g); + iaq->nirq += n1g * (nrxq1g + nofldrxq1g + nnmrxq1g); if (iaq->nirq <= navail && (itype != INTR_MSI || powerof2(iaq->nirq))) { - iaq->intr_flags |= INTR_DIRECT; + iaq->intr_flags_10g = INTR_ALL; + iaq->intr_flags_1g = INTR_ALL; goto allocate; } /* - * Second best option: an interrupt vector for errors, one for - * the firmware event queue, and one each for either NIC or - * offload rxq's. + * Second best option: a vector for errors, one for the firmware + * event queue, and vectors for either all the NIC rx queues or + * all the TOE rx queues. The queues that don't get vectors + * will forward their interrupts to those that do. + * + * Note: netmap rx queues cannot be created early and so they + * can't be setup to receive forwarded interrupts for others. */ iaq->nirq = T4_EXTRA_INTR; - iaq->nirq += n10g * max(nrxq10g, nofldrxq10g); - iaq->nirq += n1g * max(nrxq1g, nofldrxq1g); + if (nrxq10g >= nofldrxq10g) { + iaq->intr_flags_10g = INTR_RXQ; + iaq->nirq += n10g * nrxq10g; +#ifdef DEV_NETMAP + iaq->nnmrxq10g = min(nnmrxq10g, nrxq10g); +#endif + } else { + iaq->intr_flags_10g = INTR_OFLD_RXQ; + iaq->nirq += n10g * nofldrxq10g; +#ifdef DEV_NETMAP + iaq->nnmrxq10g = min(nnmrxq10g, nofldrxq10g); +#endif + } + if (nrxq1g >= nofldrxq1g) { + iaq->intr_flags_1g = INTR_RXQ; + iaq->nirq += n1g * nrxq1g; +#ifdef DEV_NETMAP + iaq->nnmrxq1g = min(nnmrxq1g, nrxq1g); +#endif + } else { + iaq->intr_flags_1g = INTR_OFLD_RXQ; + iaq->nirq += n1g * nofldrxq1g; +#ifdef DEV_NETMAP + iaq->nnmrxq1g = min(nnmrxq1g, nofldrxq1g); +#endif + } if (iaq->nirq <= navail && (itype != INTR_MSI || powerof2(iaq->nirq))) goto allocate; @@ -1805,8 +1923,8 @@ restart: /* * Next best option: an interrupt vector for errors, one for the * firmware event queue, and at least one per port. At this - * point we know we'll have to downsize nrxq or nofldrxq to fit - * what's available to us. + * point we know we'll have to downsize nrxq and/or nofldrxq + * and/or nnmrxq to fit what's available to us. */ iaq->nirq = T4_EXTRA_INTR; iaq->nirq += n10g + n1g; @@ -1816,6 +1934,9 @@ restart: if (n10g > 0) { int target = max(nrxq10g, nofldrxq10g); + iaq->intr_flags_10g = nrxq10g >= nofldrxq10g ? + INTR_RXQ : INTR_OFLD_RXQ; + n = 1; while (n < target && leftover >= n10g) { leftover -= n10g; @@ -1824,14 +1945,19 @@ restart: } iaq->nrxq10g = min(n, nrxq10g); #ifdef TCP_OFFLOAD - if (is_offload(sc)) - iaq->nofldrxq10g = min(n, nofldrxq10g); + iaq->nofldrxq10g = min(n, nofldrxq10g); +#endif +#ifdef DEV_NETMAP + iaq->nnmrxq10g = min(n, nnmrxq10g); #endif } if (n1g > 0) { int target = max(nrxq1g, nofldrxq1g); + iaq->intr_flags_1g = nrxq1g >= nofldrxq1g ? + INTR_RXQ : INTR_OFLD_RXQ; + n = 1; while (n < target && leftover >= n1g) { leftover -= n1g; @@ -1840,8 +1966,10 @@ restart: } iaq->nrxq1g = min(n, nrxq1g); #ifdef TCP_OFFLOAD - if (is_offload(sc)) - iaq->nofldrxq1g = min(n, nofldrxq1g); + iaq->nofldrxq1g = min(n, nofldrxq1g); +#endif +#ifdef DEV_NETMAP + iaq->nnmrxq1g = min(n, nnmrxq1g); #endif } @@ -1853,10 +1981,14 @@ restart: * Least desirable option: one interrupt vector for everything. */ iaq->nirq = iaq->nrxq10g = iaq->nrxq1g = 1; + iaq->intr_flags_10g = iaq->intr_flags_1g = 0; #ifdef TCP_OFFLOAD if (is_offload(sc)) iaq->nofldrxq10g = iaq->nofldrxq1g = 1; #endif +#ifdef DEV_NETMAP + iaq->nnmrxq10g = iaq->nnmrxq1g = 1; +#endif allocate: navail = iaq->nirq; @@ -2636,9 +2768,8 @@ t4_set_desc(struct adapter *sc) } static void -build_medialist(struct port_info *pi) +build_medialist(struct port_info *pi, struct ifmedia *media) { - struct ifmedia *media = &pi->media; int data, m; PORT_LOCK(pi); @@ -2767,17 +2898,29 @@ build_medialist(struct port_info *pi) * Program the port's XGMAC based on parameters in ifnet. The caller also * indicates which parameters should be programmed (the rest are left alone). */ -static int -update_mac_settings(struct port_info *pi, int flags) +int +update_mac_settings(struct ifnet *ifp, int flags) { - int rc; - struct ifnet *ifp = pi->ifp; + int rc = 0; + struct port_info *pi = ifp->if_softc; struct adapter *sc = pi->adapter; int mtu = -1, promisc = -1, allmulti = -1, vlanex = -1; + uint16_t viid = 0xffff; + int16_t *xact_addr_filt = NULL; ASSERT_SYNCHRONIZED_OP(sc); KASSERT(flags, ("%s: not told what to update.", __func__)); + if (ifp == pi->ifp) { + viid = pi->viid; + xact_addr_filt = &pi->xact_addr_filt; + } +#ifdef DEV_NETMAP + else if (ifp == pi->nm_ifp) { + viid = pi->nm_viid; + xact_addr_filt = &pi->nm_xact_addr_filt; + } +#endif if (flags & XGMAC_MTU) mtu = ifp->if_mtu; @@ -2790,25 +2933,28 @@ update_mac_settings(struct port_info *pi if (flags & XGMAC_VLANEX) vlanex = ifp->if_capenable & IFCAP_VLAN_HWTAGGING ? 1 : 0; - rc = -t4_set_rxmode(sc, sc->mbox, pi->viid, mtu, promisc, allmulti, 1, - vlanex, false); - if (rc) { - if_printf(ifp, "set_rxmode (%x) failed: %d\n", flags, rc); - return (rc); + if (flags & (XGMAC_MTU|XGMAC_PROMISC|XGMAC_ALLMULTI|XGMAC_VLANEX)) { + rc = -t4_set_rxmode(sc, sc->mbox, viid, mtu, promisc, allmulti, + 1, vlanex, false); + if (rc) { + if_printf(ifp, "set_rxmode (%x) failed: %d\n", flags, + rc); + return (rc); + } } if (flags & XGMAC_UCADDR) { uint8_t ucaddr[ETHER_ADDR_LEN]; bcopy(IF_LLADDR(ifp), ucaddr, sizeof(ucaddr)); - rc = t4_change_mac(sc, sc->mbox, pi->viid, pi->xact_addr_filt, - ucaddr, true, true); + rc = t4_change_mac(sc, sc->mbox, viid, *xact_addr_filt, ucaddr, + true, true); if (rc < 0) { rc = -rc; if_printf(ifp, "change_mac failed: %d\n", rc); return (rc); } else { - pi->xact_addr_filt = rc; + *xact_addr_filt = rc; rc = 0; } } @@ -2828,8 +2974,8 @@ update_mac_settings(struct port_info *pi LLADDR((struct sockaddr_dl *)ifma->ifma_addr); if (i == FW_MAC_EXACT_CHUNK) { - rc = t4_alloc_mac_filt(sc, sc->mbox, pi->viid, - del, i, mcaddr, NULL, &hash, 0); + rc = t4_alloc_mac_filt(sc, sc->mbox, viid, del, + i, mcaddr, NULL, &hash, 0); if (rc < 0) { rc = -rc; for (j = 0; j < i; j++) { @@ -2849,8 +2995,8 @@ update_mac_settings(struct port_info *pi } } if (i > 0) { - rc = t4_alloc_mac_filt(sc, sc->mbox, pi->viid, - del, i, mcaddr, NULL, &hash, 0); + rc = t4_alloc_mac_filt(sc, sc->mbox, viid, del, i, + mcaddr, NULL, &hash, 0); if (rc < 0) { rc = -rc; for (j = 0; j < i; j++) { @@ -2867,7 +3013,7 @@ update_mac_settings(struct port_info *pi } } - rc = -t4_set_addr_hash(sc, sc->mbox, pi->viid, 0, hash, 0); + rc = -t4_set_addr_hash(sc, sc->mbox, viid, 0, hash, 0); if (rc != 0) if_printf(ifp, "failed to set mc address hash: %d", rc); mcfail: @@ -2970,16 +3116,10 @@ cxgbe_init_synchronized(struct port_info ((rc = port_full_init(pi)) != 0)) return (rc); /* error message displayed already */ - rc = update_mac_settings(pi, XGMAC_ALL); + rc = update_mac_settings(ifp, XGMAC_ALL); if (rc) goto done; /* error message displayed already */ - rc = -t4_link_start(sc, sc->mbox, pi->tx_chan, &pi->link_cfg); - if (rc != 0) { - if_printf(ifp, "start_link failed: %d\n", rc); - goto done; - } - rc = -t4_enable_vi(sc, sc->mbox, pi->viid, true, true); if (rc != 0) { if_printf(ifp, "enable_vi failed: %d\n", rc); @@ -3064,61 +3204,41 @@ setup_intr_handlers(struct adapter *sc) #ifdef TCP_OFFLOAD struct sge_ofld_rxq *ofld_rxq; #endif +#ifdef DEV_NETMAP + struct sge_nm_rxq *nm_rxq; +#endif /* * Setup interrupts. */ irq = &sc->irq[0]; rid = sc->intr_type == INTR_INTX ? 0 : 1; - if (sc->intr_count == 1) { - KASSERT(!(sc->flags & INTR_DIRECT), - ("%s: single interrupt && INTR_DIRECT?", __func__)); *** DIFF OUTPUT TRUNCATED AT 1000 LINES ***