diff options
Diffstat (limited to 'examples')
138 files changed, 4988 insertions, 4480 deletions
diff --git a/examples/Makefile b/examples/Makefile index 28354ff8..9f7974a1 100644 --- a/examples/Makefile +++ b/examples/Makefile @@ -43,13 +43,15 @@ DIRS-$(CONFIG_RTE_LIBRTE_DISTRIBUTOR) += distributor DIRS-y += ethtool DIRS-y += exception_path DIRS-$(CONFIG_RTE_LIBRTE_EFD) += server_node_efd +DIRS-$(CONFIG_RTE_LIBRTE_FLOW_CLASSIFY) += flow_classify +DIRS-y += flow_filtering DIRS-y += helloworld DIRS-$(CONFIG_RTE_LIBRTE_PIPELINE) += ip_pipeline ifeq ($(CONFIG_RTE_LIBRTE_LPM),y) DIRS-$(CONFIG_RTE_IP_FRAG) += ip_reassembly DIRS-$(CONFIG_RTE_IP_FRAG) += ip_fragmentation endif -ifeq ($(CONFIG_RTE_LIBRTE_ACL)$(CONFIG_RTE_LIBRTE_HASH)$(CONFIG_RTE_LIBRTE_LPM),yyy) +ifeq ($(CONFIG_RTE_LIBRTE_ACL)$(CONFIG_RTE_LIBRTE_HASH)$(CONFIG_RTE_LIBRTE_LPM)$(CONFIG_RTE_LIBRTE_SECURITY),yyyy) DIRS-$(CONFIG_RTE_LIBRTE_CRYPTODEV) += ipsec-secgw endif DIRS-$(CONFIG_RTE_LIBRTE_HASH) += ipv4_multicast @@ -83,13 +85,13 @@ DIRS-$(CONFIG_RTE_LIBRTE_METER) += qos_meter DIRS-$(CONFIG_RTE_LIBRTE_SCHED) += qos_sched DIRS-y += quota_watermark DIRS-$(CONFIG_RTE_ETHDEV_RXTX_CALLBACKS) += rxtx_callbacks +DIRS-y += service_cores DIRS-y += skeleton ifeq ($(CONFIG_RTE_LIBRTE_HASH),y) DIRS-$(CONFIG_RTE_LIBRTE_VHOST) += tep_termination endif DIRS-$(CONFIG_RTE_LIBRTE_TIMER) += timer DIRS-$(CONFIG_RTE_LIBRTE_VHOST) += vhost vhost_scsi -DIRS-$(CONFIG_RTE_LIBRTE_XEN_DOM0) += vhost_xen DIRS-y += vmdq DIRS-y += vmdq_dcb ifeq ($(CONFIG_RTE_LIBRTE_POWER), y) diff --git a/examples/bond/main.c b/examples/bond/main.c index 2d019d43..8e3b1f34 100644 --- a/examples/bond/main.c +++ b/examples/bond/main.c @@ -51,7 +51,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -61,7 +60,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -141,10 +139,10 @@ addr.addr_bytes[0], addr.addr_bytes[1], addr.addr_bytes[2], \ addr.addr_bytes[3], addr.addr_bytes[4], addr.addr_bytes[5]) -uint8_t slaves[RTE_MAX_ETHPORTS]; -uint8_t slaves_count; +uint16_t slaves[RTE_MAX_ETHPORTS]; +uint16_t slaves_count; -static uint8_t BOND_PORT = 0xff; +static uint16_t BOND_PORT = 0xffff; static struct rte_mempool *mbuf_pool; @@ -171,7 +169,7 @@ static struct rte_eth_conf port_conf = { }; static void -slave_port_init(uint8_t portid, struct rte_mempool *mbuf_pool) +slave_port_init(uint16_t portid, struct rte_mempool *mbuf_pool) { int retval; uint16_t nb_rxd = RTE_RX_DESC_DEFAULT; @@ -215,7 +213,7 @@ slave_port_init(uint8_t portid, struct rte_mempool *mbuf_pool) struct ether_addr addr; rte_eth_macaddr_get(portid, &addr); - printf("Port %u MAC: ", (unsigned)portid); + printf("Port %u MAC: ", portid); PRINT_MAC(addr); printf("\n"); } @@ -234,7 +232,7 @@ bond_port_init(struct rte_mempool *mbuf_pool) rte_exit(EXIT_FAILURE, "Faled to create bond port\n"); - BOND_PORT = (uint8_t)retval; + BOND_PORT = retval; retval = rte_eth_dev_configure(BOND_PORT, 1, 1, &port_conf); if (retval != 0) @@ -675,10 +673,10 @@ static void cmd_show_parsed(__attribute__((unused)) void *parsed_result, struct cmdline *cl, __attribute__((unused)) void *data) { - uint8_t slaves[16] = {0}; + uint16_t slaves[16] = {0}; uint8_t len = 16; struct ether_addr addr; - uint8_t i = 0; + uint16_t i = 0; while (i < slaves_count) { rte_eth_macaddr_get(i, &addr); diff --git a/examples/cmdline/main.c b/examples/cmdline/main.c index c966df03..c6de944e 100644 --- a/examples/cmdline/main.c +++ b/examples/cmdline/main.c @@ -71,7 +71,6 @@ #include <cmdline.h> #include <rte_memory.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_debug.h> diff --git a/examples/distributor/main.c b/examples/distributor/main.c index 87603d03..61e6e6b9 100644 --- a/examples/distributor/main.c +++ b/examples/distributor/main.c @@ -132,7 +132,7 @@ static void print_stats(void); * coming from the mbuf_pool passed as parameter */ static inline int -port_init(uint8_t port, struct rte_mempool *mbuf_pool) +port_init(uint16_t port, struct rte_mempool *mbuf_pool) { struct rte_eth_conf port_conf = port_conf_default; const uint16_t rxRings = 1, txRings = rte_lcore_count() - 1; @@ -175,13 +175,13 @@ port_init(uint8_t port, struct rte_mempool *mbuf_pool) struct rte_eth_link link; rte_eth_link_get_nowait(port, &link); while (!link.link_status) { - printf("Waiting for Link up on port %"PRIu8"\n", port); + printf("Waiting for Link up on port %"PRIu16"\n", port); sleep(1); rte_eth_link_get_nowait(port, &link); } if (!link.link_status) { - printf("Link down on port %"PRIu8"\n", port); + printf("Link down on port %"PRIu16"\n", port); return 0; } @@ -189,7 +189,7 @@ port_init(uint8_t port, struct rte_mempool *mbuf_pool) rte_eth_macaddr_get(port, &addr); printf("Port %u MAC: %02"PRIx8" %02"PRIx8" %02"PRIx8 " %02"PRIx8" %02"PRIx8" %02"PRIx8"\n", - (unsigned)port, + port, addr.addr_bytes[0], addr.addr_bytes[1], addr.addr_bytes[2], addr.addr_bytes[3], addr.addr_bytes[4], addr.addr_bytes[5]); @@ -210,9 +210,9 @@ struct lcore_params { static int lcore_rx(struct lcore_params *p) { - const uint8_t nb_ports = rte_eth_dev_count(); + const uint16_t nb_ports = rte_eth_dev_count(); const int socket_id = rte_socket_id(); - uint8_t port; + uint16_t port; struct rte_mbuf *bufs[BURST_SIZE*2]; for (port = 0; port < nb_ports; port++) { @@ -312,9 +312,9 @@ flush_one_port(struct output_buffer *outbuf, uint8_t outp) } static inline void -flush_all_ports(struct output_buffer *tx_buffers, uint8_t nb_ports) +flush_all_ports(struct output_buffer *tx_buffers, uint16_t nb_ports) { - uint8_t outp; + uint16_t outp; for (outp = 0; outp < nb_ports; outp++) { /* skip ports that are not enabled */ @@ -384,9 +384,9 @@ static int lcore_tx(struct rte_ring *in_r) { static struct output_buffer tx_buffers[RTE_MAX_ETHPORTS]; - const uint8_t nb_ports = rte_eth_dev_count(); + const uint16_t nb_ports = rte_eth_dev_count(); const int socket_id = rte_socket_id(); - uint8_t port; + uint16_t port; for (port = 0; port < nb_ports; port++) { /* skip ports that are not enabled */ @@ -668,8 +668,8 @@ main(int argc, char *argv[]) struct rte_ring *rx_dist_ring; unsigned lcore_id, worker_id = 0; unsigned nb_ports; - uint8_t portid; - uint8_t nb_ports_available; + uint16_t portid; + uint16_t nb_ports_available; uint64_t t, freq; /* catch ctrl-c so we can print on exit */ @@ -719,10 +719,10 @@ main(int argc, char *argv[]) continue; } /* init port */ - printf("Initializing port %u... done\n", (unsigned) portid); + printf("Initializing port %u... done\n", portid); if (port_init(portid, mbuf_pool) != 0) - rte_exit(EXIT_FAILURE, "Cannot initialize port %"PRIu8"\n", + rte_exit(EXIT_FAILURE, "Cannot initialize port %u\n", portid); } diff --git a/examples/ethtool/Makefile b/examples/ethtool/Makefile index 30b42b70..e86d68ac 100644 --- a/examples/ethtool/Makefile +++ b/examples/ethtool/Makefile @@ -47,6 +47,5 @@ DIRS-y += lib ethtool-app endif DEPDIRS-ethtool-app := lib -DEPDIRS-lib := librte_eal librte_ether include $(RTE_SDK)/mk/rte.extsubdir.mk diff --git a/examples/ethtool/lib/Makefile b/examples/ethtool/lib/Makefile index 266babad..cabd82a0 100644 --- a/examples/ethtool/lib/Makefile +++ b/examples/ethtool/lib/Makefile @@ -59,5 +59,6 @@ ifeq ($(CONFIG_RTE_LIBRTE_IXGBE_PMD),y) LDLIBS += -lrte_pmd_ixgbe endif endif +LDLIBS += -lrte_eal -lrte_ethdev include $(RTE_SDK)/mk/rte.extlib.mk diff --git a/examples/ethtool/lib/rte_ethtool.c b/examples/ethtool/lib/rte_ethtool.c index 252382cb..c70c5478 100644 --- a/examples/ethtool/lib/rte_ethtool.c +++ b/examples/ethtool/lib/rte_ethtool.c @@ -36,7 +36,7 @@ #include <rte_version.h> #include <rte_ethdev.h> #include <rte_ether.h> -#include <rte_pci.h> +#include <rte_bus_pci.h> #ifdef RTE_LIBRTE_IXGBE_PMD #include <rte_pmd_ixgbe.h> #endif @@ -47,7 +47,7 @@ int -rte_ethtool_get_drvinfo(uint8_t port_id, struct ethtool_drvinfo *drvinfo) +rte_ethtool_get_drvinfo(uint16_t port_id, struct ethtool_drvinfo *drvinfo) { struct rte_eth_dev_info dev_info; struct rte_dev_reg_info reg_info; @@ -106,7 +106,7 @@ rte_ethtool_get_drvinfo(uint8_t port_id, struct ethtool_drvinfo *drvinfo) } int -rte_ethtool_get_regs_len(uint8_t port_id) +rte_ethtool_get_regs_len(uint16_t port_id) { struct rte_dev_reg_info reg_info; int ret; @@ -121,7 +121,7 @@ rte_ethtool_get_regs_len(uint8_t port_id) } int -rte_ethtool_get_regs(uint8_t port_id, struct ethtool_regs *regs, void *data) +rte_ethtool_get_regs(uint16_t port_id, struct ethtool_regs *regs, void *data) { struct rte_dev_reg_info reg_info; int status; @@ -141,7 +141,7 @@ rte_ethtool_get_regs(uint8_t port_id, struct ethtool_regs *regs, void *data) } int -rte_ethtool_get_link(uint8_t port_id) +rte_ethtool_get_link(uint16_t port_id) { struct rte_eth_link link; @@ -151,13 +151,13 @@ rte_ethtool_get_link(uint8_t port_id) } int -rte_ethtool_get_eeprom_len(uint8_t port_id) +rte_ethtool_get_eeprom_len(uint16_t port_id) { return rte_eth_dev_get_eeprom_length(port_id); } int -rte_ethtool_get_eeprom(uint8_t port_id, struct ethtool_eeprom *eeprom, +rte_ethtool_get_eeprom(uint16_t port_id, struct ethtool_eeprom *eeprom, void *words) { struct rte_dev_eeprom_info eeprom_info; @@ -180,7 +180,7 @@ rte_ethtool_get_eeprom(uint8_t port_id, struct ethtool_eeprom *eeprom, } int -rte_ethtool_set_eeprom(uint8_t port_id, struct ethtool_eeprom *eeprom, +rte_ethtool_set_eeprom(uint16_t port_id, struct ethtool_eeprom *eeprom, void *words) { struct rte_dev_eeprom_info eeprom_info; @@ -203,7 +203,7 @@ rte_ethtool_set_eeprom(uint8_t port_id, struct ethtool_eeprom *eeprom, } int -rte_ethtool_get_pauseparam(uint8_t port_id, +rte_ethtool_get_pauseparam(uint16_t port_id, struct ethtool_pauseparam *pause_param) { struct rte_eth_fc_conf fc_conf; @@ -238,7 +238,7 @@ rte_ethtool_get_pauseparam(uint8_t port_id, } int -rte_ethtool_set_pauseparam(uint8_t port_id, +rte_ethtool_set_pauseparam(uint16_t port_id, struct ethtool_pauseparam *pause_param) { struct rte_eth_fc_conf fc_conf; @@ -281,7 +281,7 @@ rte_ethtool_set_pauseparam(uint8_t port_id, } int -rte_ethtool_net_open(uint8_t port_id) +rte_ethtool_net_open(uint16_t port_id) { rte_eth_dev_stop(port_id); @@ -289,7 +289,7 @@ rte_ethtool_net_open(uint8_t port_id) } int -rte_ethtool_net_stop(uint8_t port_id) +rte_ethtool_net_stop(uint16_t port_id) { RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV); rte_eth_dev_stop(port_id); @@ -298,7 +298,7 @@ rte_ethtool_net_stop(uint8_t port_id) } int -rte_ethtool_net_get_mac_addr(uint8_t port_id, struct ether_addr *addr) +rte_ethtool_net_get_mac_addr(uint16_t port_id, struct ether_addr *addr) { RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV); if (addr == NULL) @@ -309,7 +309,7 @@ rte_ethtool_net_get_mac_addr(uint8_t port_id, struct ether_addr *addr) } int -rte_ethtool_net_set_mac_addr(uint8_t port_id, struct ether_addr *addr) +rte_ethtool_net_set_mac_addr(uint16_t port_id, struct ether_addr *addr) { if (addr == NULL) return -EINVAL; @@ -317,7 +317,7 @@ rte_ethtool_net_set_mac_addr(uint8_t port_id, struct ether_addr *addr) } int -rte_ethtool_net_validate_addr(uint8_t port_id __rte_unused, +rte_ethtool_net_validate_addr(uint16_t port_id __rte_unused, struct ether_addr *addr) { if (addr == NULL) @@ -326,7 +326,7 @@ rte_ethtool_net_validate_addr(uint8_t port_id __rte_unused, } int -rte_ethtool_net_change_mtu(uint8_t port_id, int mtu) +rte_ethtool_net_change_mtu(uint16_t port_id, int mtu) { if (mtu < 0 || mtu > UINT16_MAX) return -EINVAL; @@ -334,7 +334,7 @@ rte_ethtool_net_change_mtu(uint8_t port_id, int mtu) } int -rte_ethtool_net_get_stats64(uint8_t port_id, struct rte_eth_stats *stats) +rte_ethtool_net_get_stats64(uint16_t port_id, struct rte_eth_stats *stats) { if (stats == NULL) return -EINVAL; @@ -342,13 +342,13 @@ rte_ethtool_net_get_stats64(uint8_t port_id, struct rte_eth_stats *stats) } int -rte_ethtool_net_vlan_rx_add_vid(uint8_t port_id, uint16_t vid) +rte_ethtool_net_vlan_rx_add_vid(uint16_t port_id, uint16_t vid) { return rte_eth_dev_vlan_filter(port_id, vid, 1); } int -rte_ethtool_net_vlan_rx_kill_vid(uint8_t port_id, uint16_t vid) +rte_ethtool_net_vlan_rx_kill_vid(uint16_t port_id, uint16_t vid) { return rte_eth_dev_vlan_filter(port_id, vid, 0); } @@ -361,7 +361,7 @@ rte_ethtool_net_vlan_rx_kill_vid(uint8_t port_id, uint16_t vid) * driver can register device-specific implementation */ int -rte_ethtool_net_set_rx_mode(uint8_t port_id) +rte_ethtool_net_set_rx_mode(uint16_t port_id) { uint16_t num_vfs; struct rte_eth_dev_info dev_info; @@ -387,7 +387,7 @@ rte_ethtool_net_set_rx_mode(uint8_t port_id) int -rte_ethtool_get_ringparam(uint8_t port_id, +rte_ethtool_get_ringparam(uint16_t port_id, struct ethtool_ringparam *ring_param) { struct rte_eth_dev_info dev_info; @@ -419,7 +419,7 @@ rte_ethtool_get_ringparam(uint8_t port_id, int -rte_ethtool_set_ringparam(uint8_t port_id, +rte_ethtool_set_ringparam(uint16_t port_id, struct ethtool_ringparam *ring_param) { struct rte_eth_rxq_info rx_qinfo; diff --git a/examples/ethtool/lib/rte_ethtool.h b/examples/ethtool/lib/rte_ethtool.h index 18f44404..1cd8f4d2 100644 --- a/examples/ethtool/lib/rte_ethtool.h +++ b/examples/ethtool/lib/rte_ethtool.h @@ -79,7 +79,7 @@ extern "C" { * - (0) if successful. * - (-ENODEV) if *port_id* invalid. */ -int rte_ethtool_get_drvinfo(uint8_t port_id, struct ethtool_drvinfo *drvinfo); +int rte_ethtool_get_drvinfo(uint16_t port_id, struct ethtool_drvinfo *drvinfo); /** * Retrieve the Ethernet device register length in bytes. @@ -93,7 +93,7 @@ int rte_ethtool_get_drvinfo(uint8_t port_id, struct ethtool_drvinfo *drvinfo); * - (-ENODEV) if *port_id* invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_get_regs_len(uint8_t port_id); +int rte_ethtool_get_regs_len(uint16_t port_id); /** * Retrieve the Ethernet device register information according to @@ -111,7 +111,7 @@ int rte_ethtool_get_regs_len(uint8_t port_id); * - (-ENODEV) if *port_id* invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_get_regs(uint8_t port_id, struct ethtool_regs *regs, +int rte_ethtool_get_regs(uint16_t port_id, struct ethtool_regs *regs, void *data); /** @@ -127,7 +127,7 @@ int rte_ethtool_get_regs(uint8_t port_id, struct ethtool_regs *regs, * - (-EINVAL) if parameters invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_get_link(uint8_t port_id); +int rte_ethtool_get_link(uint16_t port_id); /** * Retrieve the Ethernet device EEPROM size @@ -141,7 +141,7 @@ int rte_ethtool_get_link(uint8_t port_id); * - (-ENODEV) if *port_id* invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_get_eeprom_len(uint8_t port_id); +int rte_ethtool_get_eeprom_len(uint16_t port_id); /** * Retrieve EEPROM content based upon eeprom range described in ethtool @@ -159,7 +159,7 @@ int rte_ethtool_get_eeprom_len(uint8_t port_id); * - (-ENODEV) if *port_id* invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_get_eeprom(uint8_t port_id, struct ethtool_eeprom *eeprom, +int rte_ethtool_get_eeprom(uint16_t port_id, struct ethtool_eeprom *eeprom, void *words); /** @@ -179,7 +179,7 @@ int rte_ethtool_get_eeprom(uint8_t port_id, struct ethtool_eeprom *eeprom, * - (-EINVAL) if parameters invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_set_eeprom(uint8_t port_id, struct ethtool_eeprom *eeprom, +int rte_ethtool_set_eeprom(uint16_t port_id, struct ethtool_eeprom *eeprom, void *words); /** @@ -199,7 +199,7 @@ int rte_ethtool_set_eeprom(uint8_t port_id, struct ethtool_eeprom *eeprom, * - (-EINVAL) if parameters invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_get_pauseparam(uint8_t port_id, +int rte_ethtool_get_pauseparam(uint16_t port_id, struct ethtool_pauseparam *pause_param); /** @@ -217,7 +217,7 @@ int rte_ethtool_get_pauseparam(uint8_t port_id, * - (-EINVAL) if parameters invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_set_pauseparam(uint8_t port_id, +int rte_ethtool_set_pauseparam(uint16_t port_id, struct ethtool_pauseparam *param); /** @@ -231,7 +231,7 @@ int rte_ethtool_set_pauseparam(uint8_t port_id, * - (-ENODEV) if *port_id* invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_net_open(uint8_t port_id); +int rte_ethtool_net_open(uint16_t port_id); /** * Stop the Ethernet device. @@ -242,7 +242,7 @@ int rte_ethtool_net_open(uint8_t port_id); * - (0) if successful. * - (-ENODEV) if *port_id* invalid. */ -int rte_ethtool_net_stop(uint8_t port_id); +int rte_ethtool_net_stop(uint16_t port_id); /** * Get the Ethernet device MAC address. @@ -255,7 +255,7 @@ int rte_ethtool_net_stop(uint8_t port_id); * - (0) if successful. * - (-ENODEV) if *port_id* invalid. */ -int rte_ethtool_net_get_mac_addr(uint8_t port_id, struct ether_addr *addr); +int rte_ethtool_net_get_mac_addr(uint16_t port_id, struct ether_addr *addr); /** * Setting the Ethernet device MAC address. @@ -271,7 +271,7 @@ int rte_ethtool_net_get_mac_addr(uint8_t port_id, struct ether_addr *addr); * - (-EINVAL) if parameters invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_net_set_mac_addr(uint8_t port_id, struct ether_addr *addr); +int rte_ethtool_net_set_mac_addr(uint16_t port_id, struct ether_addr *addr); /** * Validate if the provided MAC address is valid unicast address @@ -287,7 +287,7 @@ int rte_ethtool_net_set_mac_addr(uint8_t port_id, struct ether_addr *addr); * - (-EINVAL) if parameters invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_net_validate_addr(uint8_t port_id, struct ether_addr *addr); +int rte_ethtool_net_validate_addr(uint16_t port_id, struct ether_addr *addr); /** * Setting the Ethernet device maximum Tx unit. @@ -303,7 +303,7 @@ int rte_ethtool_net_validate_addr(uint8_t port_id, struct ether_addr *addr); * - (-EINVAL) if parameters invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_net_change_mtu(uint8_t port_id, int mtu); +int rte_ethtool_net_change_mtu(uint16_t port_id, int mtu); /** * Retrieve the Ethernet device traffic statistics @@ -319,7 +319,7 @@ int rte_ethtool_net_change_mtu(uint8_t port_id, int mtu); * - (-EINVAL) if parameters invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_net_get_stats64(uint8_t port_id, struct rte_eth_stats *stats); +int rte_ethtool_net_get_stats64(uint16_t port_id, struct rte_eth_stats *stats); /** * Update the Ethernet device VLAN filter with new vid @@ -334,7 +334,7 @@ int rte_ethtool_net_get_stats64(uint8_t port_id, struct rte_eth_stats *stats); * - (-ENODEV) if *port_id* invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_net_vlan_rx_add_vid(uint8_t port_id, uint16_t vid); +int rte_ethtool_net_vlan_rx_add_vid(uint16_t port_id, uint16_t vid); /** * Remove VLAN id from Ethernet device. @@ -349,7 +349,7 @@ int rte_ethtool_net_vlan_rx_add_vid(uint8_t port_id, uint16_t vid); * - (-ENODEV) if *port_id* invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_net_vlan_rx_kill_vid(uint8_t port_id, uint16_t vid); +int rte_ethtool_net_vlan_rx_kill_vid(uint16_t port_id, uint16_t vid); /** * Setting the Ethernet device rx mode. @@ -362,7 +362,7 @@ int rte_ethtool_net_vlan_rx_kill_vid(uint8_t port_id, uint16_t vid); * - (-ENODEV) if *port_id* invalid. * - others depends on the specific operations implementation. */ -int rte_ethtool_net_set_rx_mode(uint8_t port_id); +int rte_ethtool_net_set_rx_mode(uint16_t port_id); /** * Getting ring parameters for Ethernet device. @@ -380,7 +380,7 @@ int rte_ethtool_net_set_rx_mode(uint8_t port_id); * Only the tx_pending and rx_pending fields of struct ethtool_ringparam * are used, and the function only gets parameters for queue 0. */ -int rte_ethtool_get_ringparam(uint8_t port_id, +int rte_ethtool_get_ringparam(uint16_t port_id, struct ethtool_ringparam *ring_param); /** @@ -399,7 +399,7 @@ int rte_ethtool_get_ringparam(uint8_t port_id, * Only the tx_pending and rx_pending fields of struct ethtool_ringparam * are used, and the function only sets parameters for queue 0. */ -int rte_ethtool_set_ringparam(uint8_t port_id, +int rte_ethtool_set_ringparam(uint16_t port_id, struct ethtool_ringparam *ring_param); diff --git a/examples/eventdev_pipeline_sw_pmd/main.c b/examples/eventdev_pipeline_sw_pmd/main.c index dd75cb7a..5f431d87 100644 --- a/examples/eventdev_pipeline_sw_pmd/main.c +++ b/examples/eventdev_pipeline_sw_pmd/main.c @@ -46,6 +46,7 @@ #include <rte_cycles.h> #include <rte_ethdev.h> #include <rte_eventdev.h> +#include <rte_service.h> #define MAX_NUM_STAGES 8 #define BATCH_SIZE 16 @@ -76,6 +77,7 @@ struct fastpath_data { uint32_t rx_lock; uint32_t tx_lock; uint32_t sched_lock; + uint32_t evdev_service_id; bool rx_single; bool tx_single; bool sched_single; @@ -108,7 +110,7 @@ struct config_data { static struct config_data cdata = { .num_packets = (1L << 25), /* do ~32M packets */ .num_fids = 512, - .queue_type = RTE_EVENT_QUEUE_CFG_ATOMIC_ONLY, + .queue_type = RTE_SCHED_TYPE_ATOMIC, .next_qid = {-1}, .qid = {-1}, .num_stages = 1, @@ -233,7 +235,7 @@ producer(void) } static inline void -schedule_devices(uint8_t dev_id, unsigned int lcore_id) +schedule_devices(unsigned int lcore_id) { if (fdata->rx_core[lcore_id] && (fdata->rx_single || rte_atomic32_cmpset(&(fdata->rx_lock), 0, 1))) { @@ -243,7 +245,7 @@ schedule_devices(uint8_t dev_id, unsigned int lcore_id) if (fdata->sched_core[lcore_id] && (fdata->sched_single || rte_atomic32_cmpset(&(fdata->sched_lock), 0, 1))) { - rte_event_schedule(dev_id); + rte_service_run_iter_on_app_lcore(fdata->evdev_service_id, 1); if (cdata.dump_dev_signal) { rte_event_dev_dump(0, stdout); cdata.dump_dev_signal = 0; @@ -294,7 +296,7 @@ worker(void *arg) while (!fdata->done) { uint16_t i; - schedule_devices(dev_id, lcore_id); + schedule_devices(lcore_id); if (!fdata->worker_core[lcore_id]) { rte_pause(); @@ -490,10 +492,10 @@ parse_app_args(int argc, char **argv) cdata.enable_queue_priorities = 1; break; case 'o': - cdata.queue_type = RTE_EVENT_QUEUE_CFG_ORDERED_ONLY; + cdata.queue_type = RTE_SCHED_TYPE_ORDERED; break; case 'p': - cdata.queue_type = RTE_EVENT_QUEUE_CFG_PARALLEL_ONLY; + cdata.queue_type = RTE_SCHED_TYPE_PARALLEL; break; case 'q': cdata.quiet = 1; @@ -684,7 +686,7 @@ setup_eventdev(struct prod_data *prod_data, .new_event_threshold = 4096, }; struct rte_event_queue_conf wkr_q_conf = { - .event_queue_cfg = cdata.queue_type, + .schedule_type = cdata.queue_type, .priority = RTE_EVENT_DEV_PRIORITY_NORMAL, .nb_atomic_flows = 1024, .nb_atomic_order_sequences = 1024, @@ -696,11 +698,7 @@ setup_eventdev(struct prod_data *prod_data, }; const struct rte_event_queue_conf tx_q_conf = { .priority = RTE_EVENT_DEV_PRIORITY_HIGHEST, - .event_queue_cfg = - RTE_EVENT_QUEUE_CFG_ATOMIC_ONLY | - RTE_EVENT_QUEUE_CFG_SINGLE_LINK, - .nb_atomic_flows = 1024, - .nb_atomic_order_sequences = 1024, + .event_queue_cfg = RTE_EVENT_QUEUE_CFG_SINGLE_LINK, }; struct port_link worker_queues[MAX_NUM_STAGES]; @@ -755,11 +753,11 @@ setup_eventdev(struct prod_data *prod_data, } const char *type_str = "Atomic"; - switch (wkr_q_conf.event_queue_cfg) { - case RTE_EVENT_QUEUE_CFG_ORDERED_ONLY: + switch (wkr_q_conf.schedule_type) { + case RTE_SCHED_TYPE_ORDERED: type_str = "Ordered"; break; - case RTE_EVENT_QUEUE_CFG_PARALLEL_ONLY: + case RTE_SCHED_TYPE_PARALLEL: type_str = "Parallel"; break; } @@ -843,6 +841,14 @@ setup_eventdev(struct prod_data *prod_data, *cons_data = (struct cons_data){.dev_id = dev_id, .port_id = i }; + ret = rte_event_dev_service_id_get(dev_id, + &fdata->evdev_service_id); + if (ret != -ESRCH && ret != 0) { + printf("Error getting the service ID for sw eventdev\n"); + return -1; + } + rte_service_runstate_set(fdata->evdev_service_id, 1); + rte_service_set_runstate_mapped_check(fdata->evdev_service_id, 0); if (rte_event_dev_start(dev_id) < 0) { printf("Error starting eventdev\n"); return -1; @@ -911,9 +917,9 @@ main(int argc, char **argv) printf("\tworkers: %u\n", cdata.num_workers); printf("\tpackets: %"PRIi64"\n", cdata.num_packets); printf("\tQueue-prio: %u\n", cdata.enable_queue_priorities); - if (cdata.queue_type == RTE_EVENT_QUEUE_CFG_ORDERED_ONLY) + if (cdata.queue_type == RTE_SCHED_TYPE_ORDERED) printf("\tqid0 type: ordered\n"); - if (cdata.queue_type == RTE_EVENT_QUEUE_CFG_ATOMIC_ONLY) + if (cdata.queue_type == RTE_SCHED_TYPE_ATOMIC) printf("\tqid0 type: atomic\n"); printf("\tCores available: %u\n", rte_lcore_count()); printf("\tCores used: %u\n", cores_needed); diff --git a/examples/exception_path/main.c b/examples/exception_path/main.c index e551e6d1..f8f5bbdf 100644 --- a/examples/exception_path/main.c +++ b/examples/exception_path/main.c @@ -55,7 +55,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_per_lcore.h> #include <rte_launch.h> @@ -63,7 +62,6 @@ #include <rte_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_debug.h> #include <rte_ether.h> #include <rte_ethdev.h> @@ -135,7 +133,7 @@ static uint64_t input_cores_mask = 0; static uint64_t output_cores_mask = 0; /* Array storing port_id that is associated with each lcore */ -static uint8_t port_ids[RTE_MAX_LCORE]; +static uint16_t port_ids[RTE_MAX_LCORE]; /* Structure type for recording lcore-specific stats */ struct stats { @@ -360,8 +358,8 @@ static void setup_port_lcore_affinities(void) { unsigned long i; - uint8_t tx_port = 0; - uint8_t rx_port = 0; + uint16_t tx_port = 0; + uint16_t rx_port = 0; /* Setup port_ids[] array, and check masks were ok */ RTE_LCORE_FOREACH(i) { @@ -444,24 +442,23 @@ parse_args(int argc, char **argv) /* Initialise a single port on an Ethernet device */ static void -init_port(uint8_t port) +init_port(uint16_t port) { int ret; uint16_t nb_rxd = NB_RXD; uint16_t nb_txd = NB_TXD; /* Initialise device and RX/TX queues */ - PRINT_INFO("Initialising port %u ...", (unsigned)port); + PRINT_INFO("Initialising port %u ...", port); fflush(stdout); ret = rte_eth_dev_configure(port, 1, 1, &port_conf); if (ret < 0) - FATAL_ERROR("Could not configure port%u (%d)", - (unsigned)port, ret); + FATAL_ERROR("Could not configure port%u (%d)", port, ret); ret = rte_eth_dev_adjust_nb_rx_tx_desc(port, &nb_rxd, &nb_txd); if (ret < 0) FATAL_ERROR("Could not adjust number of descriptors for port%u (%d)", - (unsigned)port, ret); + port, ret); ret = rte_eth_rx_queue_setup(port, 0, nb_rxd, rte_eth_dev_socket_id(port), @@ -469,29 +466,30 @@ init_port(uint8_t port) pktmbuf_pool); if (ret < 0) FATAL_ERROR("Could not setup up RX queue for port%u (%d)", - (unsigned)port, ret); + port, ret); ret = rte_eth_tx_queue_setup(port, 0, nb_txd, rte_eth_dev_socket_id(port), NULL); if (ret < 0) FATAL_ERROR("Could not setup up TX queue for port%u (%d)", - (unsigned)port, ret); + port, ret); ret = rte_eth_dev_start(port); if (ret < 0) - FATAL_ERROR("Could not start port%u (%d)", (unsigned)port, ret); + FATAL_ERROR("Could not start port%u (%d)", port, ret); rte_eth_promiscuous_enable(port); } /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -506,14 +504,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up. Speed %u Mbps - %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -546,7 +543,7 @@ main(int argc, char** argv) { int ret; unsigned i,high_port; - uint8_t nb_sys_ports, port; + uint16_t nb_sys_ports, port; /* Associate signal_hanlder function with USR signals */ signal(SIGUSR1, signal_handler); diff --git a/examples/flow_classify/Makefile b/examples/flow_classify/Makefile new file mode 100644 index 00000000..eecdde14 --- /dev/null +++ b/examples/flow_classify/Makefile @@ -0,0 +1,57 @@ +# BSD LICENSE +# +# Copyright(c) 2017 Intel Corporation. All rights reserved. +# All rights reserved. +# +# 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. +# * Neither the name of Intel Corporation nor the names of its +# contributors may be used to endorse or promote products derived +# from this software without specific prior written permission. +# +# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +ifeq ($(RTE_SDK),) +$(error "Please define RTE_SDK environment variable") +endif + +# Default target, can be overridden by command line or environment +RTE_TARGET ?= x86_64-native-linuxapp-gcc + +include $(RTE_SDK)/mk/rte.vars.mk + +# binary name +APP = flow_classify + + +# all source are stored in SRCS-y +SRCS-y := flow_classify.c + +CFLAGS += -O3 +CFLAGS += $(WERROR_FLAGS) + +# workaround for a gcc bug with noreturn attribute +# http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12603 +ifeq ($(CONFIG_RTE_TOOLCHAIN_GCC),y) +CFLAGS_main.o += -Wno-return-type +endif + +include $(RTE_SDK)/mk/rte.extapp.mk diff --git a/examples/flow_classify/flow_classify.c b/examples/flow_classify/flow_classify.c new file mode 100644 index 00000000..766f1dd0 --- /dev/null +++ b/examples/flow_classify/flow_classify.c @@ -0,0 +1,852 @@ +/*- + * BSD LICENSE + * + * Copyright(c) 2017 Intel Corporation. All rights reserved. + * All rights reserved. + * + * 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. + * * Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <stdint.h> +#include <inttypes.h> +#include <getopt.h> + +#include <rte_eal.h> +#include <rte_ethdev.h> +#include <rte_cycles.h> +#include <rte_lcore.h> +#include <rte_mbuf.h> +#include <rte_flow.h> +#include <rte_flow_classify.h> +#include <rte_table_acl.h> + +#define RX_RING_SIZE 128 +#define TX_RING_SIZE 512 + +#define NUM_MBUFS 8191 +#define MBUF_CACHE_SIZE 250 +#define BURST_SIZE 32 + +#define MAX_NUM_CLASSIFY 30 +#define FLOW_CLASSIFY_MAX_RULE_NUM 91 +#define FLOW_CLASSIFY_MAX_PRIORITY 8 +#define FLOW_CLASSIFIER_NAME_SIZE 64 + +#define COMMENT_LEAD_CHAR ('#') +#define OPTION_RULE_IPV4 "rule_ipv4" +#define RTE_LOGTYPE_FLOW_CLASSIFY RTE_LOGTYPE_USER3 +#define flow_classify_log(format, ...) \ + RTE_LOG(ERR, FLOW_CLASSIFY, format, ##__VA_ARGS__) + +#define uint32_t_to_char(ip, a, b, c, d) do {\ + *a = (unsigned char)(ip >> 24 & 0xff);\ + *b = (unsigned char)(ip >> 16 & 0xff);\ + *c = (unsigned char)(ip >> 8 & 0xff);\ + *d = (unsigned char)(ip & 0xff);\ + } while (0) + +enum { + CB_FLD_SRC_ADDR, + CB_FLD_DST_ADDR, + CB_FLD_SRC_PORT, + CB_FLD_SRC_PORT_DLM, + CB_FLD_SRC_PORT_MASK, + CB_FLD_DST_PORT, + CB_FLD_DST_PORT_DLM, + CB_FLD_DST_PORT_MASK, + CB_FLD_PROTO, + CB_FLD_PRIORITY, + CB_FLD_NUM, +}; + +static struct{ + const char *rule_ipv4_name; +} parm_config; +const char cb_port_delim[] = ":"; + +static const struct rte_eth_conf port_conf_default = { + .rxmode = { .max_rx_pkt_len = ETHER_MAX_LEN } +}; + +struct flow_classifier { + struct rte_flow_classifier *cls; + uint32_t table_id[RTE_FLOW_CLASSIFY_TABLE_MAX]; +}; + +struct flow_classifier_acl { + struct flow_classifier cls; +} __rte_cache_aligned; + +/* ACL field definitions for IPv4 5 tuple rule */ + +enum { + PROTO_FIELD_IPV4, + SRC_FIELD_IPV4, + DST_FIELD_IPV4, + SRCP_FIELD_IPV4, + DSTP_FIELD_IPV4, + NUM_FIELDS_IPV4 +}; + +enum { + PROTO_INPUT_IPV4, + SRC_INPUT_IPV4, + DST_INPUT_IPV4, + SRCP_DESTP_INPUT_IPV4 +}; + +static struct rte_acl_field_def ipv4_defs[NUM_FIELDS_IPV4] = { + /* first input field - always one byte long. */ + { + .type = RTE_ACL_FIELD_TYPE_BITMASK, + .size = sizeof(uint8_t), + .field_index = PROTO_FIELD_IPV4, + .input_index = PROTO_INPUT_IPV4, + .offset = sizeof(struct ether_hdr) + + offsetof(struct ipv4_hdr, next_proto_id), + }, + /* next input field (IPv4 source address) - 4 consecutive bytes. */ + { + /* rte_flow uses a bit mask for IPv4 addresses */ + .type = RTE_ACL_FIELD_TYPE_BITMASK, + .size = sizeof(uint32_t), + .field_index = SRC_FIELD_IPV4, + .input_index = SRC_INPUT_IPV4, + .offset = sizeof(struct ether_hdr) + + offsetof(struct ipv4_hdr, src_addr), + }, + /* next input field (IPv4 destination address) - 4 consecutive bytes. */ + { + /* rte_flow uses a bit mask for IPv4 addresses */ + .type = RTE_ACL_FIELD_TYPE_BITMASK, + .size = sizeof(uint32_t), + .field_index = DST_FIELD_IPV4, + .input_index = DST_INPUT_IPV4, + .offset = sizeof(struct ether_hdr) + + offsetof(struct ipv4_hdr, dst_addr), + }, + /* + * Next 2 fields (src & dst ports) form 4 consecutive bytes. + * They share the same input index. + */ + { + /* rte_flow uses a bit mask for protocol ports */ + .type = RTE_ACL_FIELD_TYPE_BITMASK, + .size = sizeof(uint16_t), + .field_index = SRCP_FIELD_IPV4, + .input_index = SRCP_DESTP_INPUT_IPV4, + .offset = sizeof(struct ether_hdr) + + sizeof(struct ipv4_hdr) + + offsetof(struct tcp_hdr, src_port), + }, + { + /* rte_flow uses a bit mask for protocol ports */ + .type = RTE_ACL_FIELD_TYPE_BITMASK, + .size = sizeof(uint16_t), + .field_index = DSTP_FIELD_IPV4, + .input_index = SRCP_DESTP_INPUT_IPV4, + .offset = sizeof(struct ether_hdr) + + sizeof(struct ipv4_hdr) + + offsetof(struct tcp_hdr, dst_port), + }, +}; + +/* flow classify data */ +static int num_classify_rules; +static struct rte_flow_classify_rule *rules[MAX_NUM_CLASSIFY]; +static struct rte_flow_classify_ipv4_5tuple_stats ntuple_stats; +static struct rte_flow_classify_stats classify_stats = { + .stats = (void **)&ntuple_stats +}; + +/* parameters for rte_flow_classify_validate and + * rte_flow_classify_table_entry_add functions + */ + +static struct rte_flow_item eth_item = { RTE_FLOW_ITEM_TYPE_ETH, + 0, 0, 0 }; +static struct rte_flow_item end_item = { RTE_FLOW_ITEM_TYPE_END, + 0, 0, 0 }; + +/* sample actions: + * "actions count / end" + */ +static struct rte_flow_action count_action = { RTE_FLOW_ACTION_TYPE_COUNT, 0}; +static struct rte_flow_action end_action = { RTE_FLOW_ACTION_TYPE_END, 0}; +static struct rte_flow_action actions[2]; + +/* sample attributes */ +static struct rte_flow_attr attr; + +/* flow_classify.c: * Based on DPDK skeleton forwarding example. */ + +/* + * Initializes a given port using global settings and with the RX buffers + * coming from the mbuf_pool passed as a parameter. + */ +static inline int +port_init(uint8_t port, struct rte_mempool *mbuf_pool) +{ + struct rte_eth_conf port_conf = port_conf_default; + struct ether_addr addr; + const uint16_t rx_rings = 1, tx_rings = 1; + int retval; + uint16_t q; + + if (port >= rte_eth_dev_count()) + return -1; + + /* Configure the Ethernet device. */ + retval = rte_eth_dev_configure(port, rx_rings, tx_rings, &port_conf); + if (retval != 0) + return retval; + + /* Allocate and set up 1 RX queue per Ethernet port. */ + for (q = 0; q < rx_rings; q++) { + retval = rte_eth_rx_queue_setup(port, q, RX_RING_SIZE, + rte_eth_dev_socket_id(port), NULL, mbuf_pool); + if (retval < 0) + return retval; + } + + /* Allocate and set up 1 TX queue per Ethernet port. */ + for (q = 0; q < tx_rings; q++) { + retval = rte_eth_tx_queue_setup(port, q, TX_RING_SIZE, + rte_eth_dev_socket_id(port), NULL); + if (retval < 0) + return retval; + } + + /* Start the Ethernet port. */ + retval = rte_eth_dev_start(port); + if (retval < 0) + return retval; + + /* Display the port MAC address. */ + rte_eth_macaddr_get(port, &addr); + printf("Port %u MAC: %02" PRIx8 " %02" PRIx8 " %02" PRIx8 + " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 "\n", + port, + addr.addr_bytes[0], addr.addr_bytes[1], + addr.addr_bytes[2], addr.addr_bytes[3], + addr.addr_bytes[4], addr.addr_bytes[5]); + + /* Enable RX in promiscuous mode for the Ethernet device. */ + rte_eth_promiscuous_enable(port); + + return 0; +} + +/* + * The lcore main. This is the main thread that does the work, reading from + * an input port classifying the packets and writing to an output port. + */ +static __attribute__((noreturn)) void +lcore_main(struct flow_classifier *cls_app) +{ + const uint8_t nb_ports = rte_eth_dev_count(); + uint8_t port; + int ret; + int i = 0; + + ret = rte_flow_classify_table_entry_delete(cls_app->cls, + cls_app->table_id[0], rules[7]); + if (ret) + printf("table_entry_delete failed [7] %d\n\n", ret); + else + printf("table_entry_delete succeeded [7]\n\n"); + + /* + * Check that the port is on the same NUMA node as the polling thread + * for best performance. + */ + for (port = 0; port < nb_ports; port++) + if (rte_eth_dev_socket_id(port) > 0 && + rte_eth_dev_socket_id(port) != (int)rte_socket_id()) { + printf("\n\n"); + printf("WARNING: port %u is on remote NUMA node\n", + port); + printf("to polling thread.\n"); + printf("Performance will not be optimal.\n"); + + printf("\nCore %u forwarding packets. ", + rte_lcore_id()); + printf("[Ctrl+C to quit]\n"); + } + /* Run until the application is quit or killed. */ + for (;;) { + /* + * Receive packets on a port, classify them and forward them + * on the paired port. + * The mapping is 0 -> 1, 1 -> 0, 2 -> 3, 3 -> 2, etc. + */ + for (port = 0; port < nb_ports; port++) { + /* Get burst of RX packets, from first port of pair. */ + struct rte_mbuf *bufs[BURST_SIZE]; + const uint16_t nb_rx = rte_eth_rx_burst(port, 0, + bufs, BURST_SIZE); + + if (unlikely(nb_rx == 0)) + continue; + + for (i = 0; i < MAX_NUM_CLASSIFY; i++) { + if (rules[i]) { + ret = rte_flow_classifier_query( + cls_app->cls, + cls_app->table_id[0], + bufs, nb_rx, rules[i], + &classify_stats); + if (ret) + printf( + "rule [%d] query failed ret [%d]\n\n", + i, ret); + else { + printf( + "rule[%d] count=%"PRIu64"\n", + i, ntuple_stats.counter1); + + printf("proto = %d\n", + ntuple_stats.ipv4_5tuple.proto); + } + } + } + + /* Send burst of TX packets, to second port of pair. */ + const uint16_t nb_tx = rte_eth_tx_burst(port ^ 1, 0, + bufs, nb_rx); + + /* Free any unsent packets. */ + if (unlikely(nb_tx < nb_rx)) { + uint16_t buf; + + for (buf = nb_tx; buf < nb_rx; buf++) + rte_pktmbuf_free(bufs[buf]); + } + } + } +} + +/* + * Parse IPv4 5 tuple rules file, ipv4_rules_file.txt. + * Expected format: + * <src_ipv4_addr>'/'<masklen> <space> \ + * <dst_ipv4_addr>'/'<masklen> <space> \ + * <src_port> <space> ":" <src_port_mask> <space> \ + * <dst_port> <space> ":" <dst_port_mask> <space> \ + * <proto>'/'<proto_mask> <space> \ + * <priority> + */ + +static int +get_cb_field(char **in, uint32_t *fd, int base, unsigned long lim, + char dlm) +{ + unsigned long val; + char *end; + + errno = 0; + val = strtoul(*in, &end, base); + if (errno != 0 || end[0] != dlm || val > lim) + return -EINVAL; + *fd = (uint32_t)val; + *in = end + 1; + return 0; +} + +static int +parse_ipv4_net(char *in, uint32_t *addr, uint32_t *mask_len) +{ + uint32_t a, b, c, d, m; + + if (get_cb_field(&in, &a, 0, UINT8_MAX, '.')) + return -EINVAL; + if (get_cb_field(&in, &b, 0, UINT8_MAX, '.')) + return -EINVAL; + if (get_cb_field(&in, &c, 0, UINT8_MAX, '.')) + return -EINVAL; + if (get_cb_field(&in, &d, 0, UINT8_MAX, '/')) + return -EINVAL; + if (get_cb_field(&in, &m, 0, sizeof(uint32_t) * CHAR_BIT, 0)) + return -EINVAL; + + addr[0] = IPv4(a, b, c, d); + mask_len[0] = m; + return 0; +} + +static int +parse_ipv4_5tuple_rule(char *str, struct rte_eth_ntuple_filter *ntuple_filter) +{ + int i, ret; + char *s, *sp, *in[CB_FLD_NUM]; + static const char *dlm = " \t\n"; + int dim = CB_FLD_NUM; + uint32_t temp; + + s = str; + for (i = 0; i != dim; i++, s = NULL) { + in[i] = strtok_r(s, dlm, &sp); + if (in[i] == NULL) + return -EINVAL; + } + + ret = parse_ipv4_net(in[CB_FLD_SRC_ADDR], + &ntuple_filter->src_ip, + &ntuple_filter->src_ip_mask); + if (ret != 0) { + flow_classify_log("failed to read source address/mask: %s\n", + in[CB_FLD_SRC_ADDR]); + return ret; + } + + ret = parse_ipv4_net(in[CB_FLD_DST_ADDR], + &ntuple_filter->dst_ip, + &ntuple_filter->dst_ip_mask); + if (ret != 0) { + flow_classify_log("failed to read source address/mask: %s\n", + in[CB_FLD_DST_ADDR]); + return ret; + } + + if (get_cb_field(&in[CB_FLD_SRC_PORT], &temp, 0, UINT16_MAX, 0)) + return -EINVAL; + ntuple_filter->src_port = (uint16_t)temp; + + if (strncmp(in[CB_FLD_SRC_PORT_DLM], cb_port_delim, + sizeof(cb_port_delim)) != 0) + return -EINVAL; + + if (get_cb_field(&in[CB_FLD_SRC_PORT_MASK], &temp, 0, UINT16_MAX, 0)) + return -EINVAL; + ntuple_filter->src_port_mask = (uint16_t)temp; + + if (get_cb_field(&in[CB_FLD_DST_PORT], &temp, 0, UINT16_MAX, 0)) + return -EINVAL; + ntuple_filter->dst_port = (uint16_t)temp; + + if (strncmp(in[CB_FLD_DST_PORT_DLM], cb_port_delim, + sizeof(cb_port_delim)) != 0) + return -EINVAL; + + if (get_cb_field(&in[CB_FLD_DST_PORT_MASK], &temp, 0, UINT16_MAX, 0)) + return -EINVAL; + ntuple_filter->dst_port_mask = (uint16_t)temp; + + if (get_cb_field(&in[CB_FLD_PROTO], &temp, 0, UINT8_MAX, '/')) + return -EINVAL; + ntuple_filter->proto = (uint8_t)temp; + + if (get_cb_field(&in[CB_FLD_PROTO], &temp, 0, UINT8_MAX, 0)) + return -EINVAL; + ntuple_filter->proto_mask = (uint8_t)temp; + + if (get_cb_field(&in[CB_FLD_PRIORITY], &temp, 0, UINT16_MAX, 0)) + return -EINVAL; + ntuple_filter->priority = (uint16_t)temp; + if (ntuple_filter->priority > FLOW_CLASSIFY_MAX_PRIORITY) + ret = -EINVAL; + + return ret; +} + +/* Bypass comment and empty lines */ +static inline int +is_bypass_line(char *buff) +{ + int i = 0; + + /* comment line */ + if (buff[0] == COMMENT_LEAD_CHAR) + return 1; + /* empty line */ + while (buff[i] != '\0') { + if (!isspace(buff[i])) + return 0; + i++; + } + return 1; +} + +static uint32_t +convert_depth_to_bitmask(uint32_t depth_val) +{ + uint32_t bitmask = 0; + int i, j; + + for (i = depth_val, j = 0; i > 0; i--, j++) + bitmask |= (1 << (31 - j)); + return bitmask; +} + +static int +add_classify_rule(struct rte_eth_ntuple_filter *ntuple_filter, + struct flow_classifier *cls_app) +{ + int ret = -1; + int key_found; + struct rte_flow_error error; + struct rte_flow_item_ipv4 ipv4_spec; + struct rte_flow_item_ipv4 ipv4_mask; + struct rte_flow_item ipv4_udp_item; + struct rte_flow_item ipv4_tcp_item; + struct rte_flow_item ipv4_sctp_item; + struct rte_flow_item_udp udp_spec; + struct rte_flow_item_udp udp_mask; + struct rte_flow_item udp_item; + struct rte_flow_item_tcp tcp_spec; + struct rte_flow_item_tcp tcp_mask; + struct rte_flow_item tcp_item; + struct rte_flow_item_sctp sctp_spec; + struct rte_flow_item_sctp sctp_mask; + struct rte_flow_item sctp_item; + struct rte_flow_item pattern_ipv4_5tuple[4]; + struct rte_flow_classify_rule *rule; + uint8_t ipv4_proto; + + if (num_classify_rules >= MAX_NUM_CLASSIFY) { + printf( + "\nINFO: classify rule capacity %d reached\n", + num_classify_rules); + return ret; + } + + /* set up parameters for validate and add */ + memset(&ipv4_spec, 0, sizeof(ipv4_spec)); + ipv4_spec.hdr.next_proto_id = ntuple_filter->proto; + ipv4_spec.hdr.src_addr = ntuple_filter->src_ip; + ipv4_spec.hdr.dst_addr = ntuple_filter->dst_ip; + ipv4_proto = ipv4_spec.hdr.next_proto_id; + + memset(&ipv4_mask, 0, sizeof(ipv4_mask)); + ipv4_mask.hdr.next_proto_id = ntuple_filter->proto_mask; + ipv4_mask.hdr.src_addr = ntuple_filter->src_ip_mask; + ipv4_mask.hdr.src_addr = + convert_depth_to_bitmask(ipv4_mask.hdr.src_addr); + ipv4_mask.hdr.dst_addr = ntuple_filter->dst_ip_mask; + ipv4_mask.hdr.dst_addr = + convert_depth_to_bitmask(ipv4_mask.hdr.dst_addr); + + switch (ipv4_proto) { + case IPPROTO_UDP: + ipv4_udp_item.type = RTE_FLOW_ITEM_TYPE_IPV4; + ipv4_udp_item.spec = &ipv4_spec; + ipv4_udp_item.mask = &ipv4_mask; + ipv4_udp_item.last = NULL; + + udp_spec.hdr.src_port = ntuple_filter->src_port; + udp_spec.hdr.dst_port = ntuple_filter->dst_port; + udp_spec.hdr.dgram_len = 0; + udp_spec.hdr.dgram_cksum = 0; + + udp_mask.hdr.src_port = ntuple_filter->src_port_mask; + udp_mask.hdr.dst_port = ntuple_filter->dst_port_mask; + udp_mask.hdr.dgram_len = 0; + udp_mask.hdr.dgram_cksum = 0; + + udp_item.type = RTE_FLOW_ITEM_TYPE_UDP; + udp_item.spec = &udp_spec; + udp_item.mask = &udp_mask; + udp_item.last = NULL; + + attr.priority = ntuple_filter->priority; + pattern_ipv4_5tuple[1] = ipv4_udp_item; + pattern_ipv4_5tuple[2] = udp_item; + break; + case IPPROTO_TCP: + ipv4_tcp_item.type = RTE_FLOW_ITEM_TYPE_IPV4; + ipv4_tcp_item.spec = &ipv4_spec; + ipv4_tcp_item.mask = &ipv4_mask; + ipv4_tcp_item.last = NULL; + + memset(&tcp_spec, 0, sizeof(tcp_spec)); + tcp_spec.hdr.src_port = ntuple_filter->src_port; + tcp_spec.hdr.dst_port = ntuple_filter->dst_port; + + memset(&tcp_mask, 0, sizeof(tcp_mask)); + tcp_mask.hdr.src_port = ntuple_filter->src_port_mask; + tcp_mask.hdr.dst_port = ntuple_filter->dst_port_mask; + + tcp_item.type = RTE_FLOW_ITEM_TYPE_TCP; + tcp_item.spec = &tcp_spec; + tcp_item.mask = &tcp_mask; + tcp_item.last = NULL; + + attr.priority = ntuple_filter->priority; + pattern_ipv4_5tuple[1] = ipv4_tcp_item; + pattern_ipv4_5tuple[2] = tcp_item; + break; + case IPPROTO_SCTP: + ipv4_sctp_item.type = RTE_FLOW_ITEM_TYPE_IPV4; + ipv4_sctp_item.spec = &ipv4_spec; + ipv4_sctp_item.mask = &ipv4_mask; + ipv4_sctp_item.last = NULL; + + sctp_spec.hdr.src_port = ntuple_filter->src_port; + sctp_spec.hdr.dst_port = ntuple_filter->dst_port; + sctp_spec.hdr.cksum = 0; + sctp_spec.hdr.tag = 0; + + sctp_mask.hdr.src_port = ntuple_filter->src_port_mask; + sctp_mask.hdr.dst_port = ntuple_filter->dst_port_mask; + sctp_mask.hdr.cksum = 0; + sctp_mask.hdr.tag = 0; + + sctp_item.type = RTE_FLOW_ITEM_TYPE_SCTP; + sctp_item.spec = &sctp_spec; + sctp_item.mask = &sctp_mask; + sctp_item.last = NULL; + + attr.priority = ntuple_filter->priority; + pattern_ipv4_5tuple[1] = ipv4_sctp_item; + pattern_ipv4_5tuple[2] = sctp_item; + break; + default: + return ret; + } + + attr.ingress = 1; + pattern_ipv4_5tuple[0] = eth_item; + pattern_ipv4_5tuple[3] = end_item; + actions[0] = count_action; + actions[1] = end_action; + + rule = rte_flow_classify_table_entry_add( + cls_app->cls, cls_app->table_id[0], &key_found, + &attr, pattern_ipv4_5tuple, actions, &error); + if (rule == NULL) { + printf("table entry add failed ipv4_proto = %u\n", + ipv4_proto); + ret = -1; + return ret; + } + + rules[num_classify_rules] = rule; + num_classify_rules++; + return 0; +} + +static int +add_rules(const char *rule_path, struct flow_classifier *cls_app) +{ + FILE *fh; + char buff[LINE_MAX]; + unsigned int i = 0; + unsigned int total_num = 0; + struct rte_eth_ntuple_filter ntuple_filter; + int ret; + + fh = fopen(rule_path, "rb"); + if (fh == NULL) + rte_exit(EXIT_FAILURE, "%s: fopen %s failed\n", __func__, + rule_path); + + ret = fseek(fh, 0, SEEK_SET); + if (ret) + rte_exit(EXIT_FAILURE, "%s: fseek %d failed\n", __func__, + ret); + + i = 0; + while (fgets(buff, LINE_MAX, fh) != NULL) { + i++; + + if (is_bypass_line(buff)) + continue; + + if (total_num >= FLOW_CLASSIFY_MAX_RULE_NUM - 1) { + printf("\nINFO: classify rule capacity %d reached\n", + total_num); + break; + } + + if (parse_ipv4_5tuple_rule(buff, &ntuple_filter) != 0) + rte_exit(EXIT_FAILURE, + "%s Line %u: parse rules error\n", + rule_path, i); + + if (add_classify_rule(&ntuple_filter, cls_app) != 0) + rte_exit(EXIT_FAILURE, "add rule error\n"); + + total_num++; + } + + fclose(fh); + return 0; +} + +/* display usage */ +static void +print_usage(const char *prgname) +{ + printf("%s usage:\n", prgname); + printf("[EAL options] -- --"OPTION_RULE_IPV4"=FILE: "); + printf("specify the ipv4 rules file.\n"); + printf("Each rule occupies one line in the file.\n"); +} + +/* Parse the argument given in the command line of the application */ +static int +parse_args(int argc, char **argv) +{ + int opt, ret; + char **argvopt; + int option_index; + char *prgname = argv[0]; + static struct option lgopts[] = { + {OPTION_RULE_IPV4, 1, 0, 0}, + {NULL, 0, 0, 0} + }; + + argvopt = argv; + + while ((opt = getopt_long(argc, argvopt, "", + lgopts, &option_index)) != EOF) { + + switch (opt) { + /* long options */ + case 0: + if (!strncmp(lgopts[option_index].name, + OPTION_RULE_IPV4, + sizeof(OPTION_RULE_IPV4))) + parm_config.rule_ipv4_name = optarg; + break; + default: + print_usage(prgname); + return -1; + } + } + + if (optind >= 0) + argv[optind-1] = prgname; + + ret = optind-1; + optind = 1; /* reset getopt lib */ + return ret; +} + +/* + * The main function, which does initialization and calls the lcore_main + * function. + */ +int +main(int argc, char *argv[]) +{ + struct rte_mempool *mbuf_pool; + uint8_t nb_ports; + uint8_t portid; + int ret; + int socket_id; + struct rte_table_acl_params table_acl_params; + struct rte_flow_classify_table_params cls_table_params; + struct flow_classifier *cls_app; + struct rte_flow_classifier_params cls_params; + uint32_t size; + + /* Initialize the Environment Abstraction Layer (EAL). */ + ret = rte_eal_init(argc, argv); + if (ret < 0) + rte_exit(EXIT_FAILURE, "Error with EAL initialization\n"); + + argc -= ret; + argv += ret; + + /* parse application arguments (after the EAL ones) */ + ret = parse_args(argc, argv); + if (ret < 0) + rte_exit(EXIT_FAILURE, "Invalid flow_classify parameters\n"); + + /* Check that there is an even number of ports to send/receive on. */ + nb_ports = rte_eth_dev_count(); + if (nb_ports < 2 || (nb_ports & 1)) + rte_exit(EXIT_FAILURE, "Error: number of ports must be even\n"); + + /* Creates a new mempool in memory to hold the mbufs. */ + mbuf_pool = rte_pktmbuf_pool_create("MBUF_POOL", NUM_MBUFS * nb_ports, + MBUF_CACHE_SIZE, 0, RTE_MBUF_DEFAULT_BUF_SIZE, rte_socket_id()); + + if (mbuf_pool == NULL) + rte_exit(EXIT_FAILURE, "Cannot create mbuf pool\n"); + + /* Initialize all ports. */ + for (portid = 0; portid < nb_ports; portid++) + if (port_init(portid, mbuf_pool) != 0) + rte_exit(EXIT_FAILURE, "Cannot init port %"PRIu8 "\n", + portid); + + if (rte_lcore_count() > 1) + printf("\nWARNING: Too many lcores enabled. Only 1 used.\n"); + + socket_id = rte_eth_dev_socket_id(0); + + /* Memory allocation */ + size = RTE_CACHE_LINE_ROUNDUP(sizeof(struct flow_classifier_acl)); + cls_app = rte_zmalloc(NULL, size, RTE_CACHE_LINE_SIZE); + if (cls_app == NULL) + rte_exit(EXIT_FAILURE, "Cannot allocate classifier memory\n"); + + cls_params.name = "flow_classifier"; + cls_params.socket_id = socket_id; + cls_params.type = RTE_FLOW_CLASSIFY_TABLE_TYPE_ACL; + + cls_app->cls = rte_flow_classifier_create(&cls_params); + if (cls_app->cls == NULL) { + rte_free(cls_app); + rte_exit(EXIT_FAILURE, "Cannot create classifier\n"); + } + + /* initialise ACL table params */ + table_acl_params.name = "table_acl_ipv4_5tuple"; + table_acl_params.n_rules = FLOW_CLASSIFY_MAX_RULE_NUM; + table_acl_params.n_rule_fields = RTE_DIM(ipv4_defs); + memcpy(table_acl_params.field_format, ipv4_defs, sizeof(ipv4_defs)); + + /* initialise table create params */ + cls_table_params.ops = &rte_table_acl_ops, + cls_table_params.arg_create = &table_acl_params, + + ret = rte_flow_classify_table_create(cls_app->cls, &cls_table_params, + &cls_app->table_id[0]); + if (ret) { + rte_flow_classifier_free(cls_app->cls); + rte_free(cls_app); + rte_exit(EXIT_FAILURE, "Failed to create classifier table\n"); + } + + /* read file of IPv4 5 tuple rules and initialize parameters + * for rte_flow_classify_validate and rte_flow_classify_table_entry_add + * API's. + */ + if (add_rules(parm_config.rule_ipv4_name, cls_app)) { + rte_flow_classifier_free(cls_app->cls); + rte_free(cls_app); + rte_exit(EXIT_FAILURE, "Failed to add rules\n"); + } + + /* Call lcore_main on the master core only. */ + lcore_main(cls_app); + + return 0; +} diff --git a/examples/flow_classify/ipv4_rules_file.txt b/examples/flow_classify/ipv4_rules_file.txt new file mode 100644 index 00000000..dfa0631f --- /dev/null +++ b/examples/flow_classify/ipv4_rules_file.txt @@ -0,0 +1,14 @@ +#file format: +#src_ip/masklen dst_ip/masklen src_port : mask dst_port : mask proto/mask priority +# +2.2.2.3/24 2.2.2.7/24 32 : 0xffff 33 : 0xffff 17/0xff 0 +9.9.9.3/24 9.9.9.7/24 32 : 0xffff 33 : 0xffff 17/0xff 1 +9.9.9.3/24 9.9.9.7/24 32 : 0xffff 33 : 0xffff 6/0xff 2 +9.9.8.3/24 9.9.8.7/24 32 : 0xffff 33 : 0xffff 6/0xff 3 +6.7.8.9/24 2.3.4.5/24 32 : 0x0000 33 : 0x0000 132/0xff 4 +6.7.8.9/32 192.168.0.36/32 10 : 0xffff 11 : 0xffff 6/0xfe 5 +6.7.8.9/24 192.168.0.36/24 10 : 0xffff 11 : 0xffff 6/0xfe 6 +6.7.8.9/16 192.168.0.36/16 10 : 0xffff 11 : 0xffff 6/0xfe 7 +6.7.8.9/8 192.168.0.36/8 10 : 0xffff 11 : 0xffff 6/0xfe 8 +#error rules +#9.8.7.6/8 192.168.0.36/8 10 : 0xffff 11 : 0xffff 6/0xfe 9
\ No newline at end of file diff --git a/examples/flow_filtering/Makefile b/examples/flow_filtering/Makefile new file mode 100644 index 00000000..70b82fe3 --- /dev/null +++ b/examples/flow_filtering/Makefile @@ -0,0 +1,49 @@ +# +# BSD LICENSE +# +# Copyright 2017 Mellanox. +# +# 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. +# * Neither the name of Mellanox nor the names of its +# contributors may be used to endorse or promote products derived +# from this software without specific prior written permission. +# +# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS +# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT +# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR +# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT +# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, +# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT +# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, +# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY +# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT +# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE +# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. +# + +ifeq ($(RTE_SDK),) +$(error "Please define RTE_SDK environment variable") +endif + +# Default target, can be overridden by command line or environment +RTE_TARGET ?= x86_64-native-linuxapp-gcc + +include $(RTE_SDK)/mk/rte.vars.mk + +APP = flow + +SRCS-y := main.c + +CFLAGS += -O3 +CFLAGS += $(WERROR_FLAGS) + +include $(RTE_SDK)/mk/rte.extapp.mk diff --git a/examples/flow_filtering/flow_blocks.c b/examples/flow_filtering/flow_blocks.c new file mode 100644 index 00000000..f92df102 --- /dev/null +++ b/examples/flow_filtering/flow_blocks.c @@ -0,0 +1,150 @@ +/*- + * BSD LICENSE + * + * Copyright 2017 Mellanox. + * + * 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. + * * Neither the name of Mellanox nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#define MAX_PATTERN_NUM 4 + +struct rte_flow * +generate_ipv4_flow(uint8_t port_id, uint16_t rx_q, + uint32_t src_ip, uint32_t src_mask, + uint32_t dest_ip, uint32_t dest_mask, + struct rte_flow_error *error); + + +/** + * create a flow rule that sends packets with matching src and dest ip + * to selected queue. + * + * @param port_id + * The selected port. + * @param rx_q + * The selected target queue. + * @param src_ip + * The src ip value to match the input packet. + * @param src_mask + * The mask to apply to the src ip. + * @param dest_ip + * The dest ip value to match the input packet. + * @param dest_mask + * The mask to apply to the dest ip. + * @param[out] error + * Perform verbose error reporting if not NULL. + * + * @return + * A flow if the rule could be created else return NULL. + */ +struct rte_flow * +generate_ipv4_flow(uint8_t port_id, uint16_t rx_q, + uint32_t src_ip, uint32_t src_mask, + uint32_t dest_ip, uint32_t dest_mask, + struct rte_flow_error *error) +{ + struct rte_flow_attr attr; + struct rte_flow_item pattern[MAX_PATTERN_NUM]; + struct rte_flow_action action[MAX_PATTERN_NUM]; + struct rte_flow *flow = NULL; + struct rte_flow_action_queue queue = { .index = rx_q }; + struct rte_flow_item_eth eth_spec; + struct rte_flow_item_eth eth_mask; + struct rte_flow_item_vlan vlan_spec; + struct rte_flow_item_vlan vlan_mask; + struct rte_flow_item_ipv4 ip_spec; + struct rte_flow_item_ipv4 ip_mask; + int res; + + memset(pattern, 0, sizeof(pattern)); + memset(action, 0, sizeof(action)); + + /* + * set the rule attribute. + * in this case only ingress packets will be checked. + */ + memset(&attr, 0, sizeof(struct rte_flow_attr)); + attr.ingress = 1; + + /* + * create the action sequence. + * one action only, move packet to queue + */ + + action[0].type = RTE_FLOW_ACTION_TYPE_QUEUE; + action[0].conf = &queue; + action[1].type = RTE_FLOW_ACTION_TYPE_END; + + /* + * set the first level of the pattern (eth). + * since in this example we just want to get the + * ipv4 we set this level to allow all. + */ + memset(ð_spec, 0, sizeof(struct rte_flow_item_eth)); + memset(ð_mask, 0, sizeof(struct rte_flow_item_eth)); + eth_spec.type = 0; + eth_mask.type = 0; + pattern[0].type = RTE_FLOW_ITEM_TYPE_ETH; + pattern[0].spec = ð_spec; + pattern[0].mask = ð_mask; + + /* + * setting the second level of the pattern (vlan). + * since in this example we just want to get the + * ipv4 we also set this level to allow all. + */ + memset(&vlan_spec, 0, sizeof(struct rte_flow_item_vlan)); + memset(&vlan_mask, 0, sizeof(struct rte_flow_item_vlan)); + pattern[1].type = RTE_FLOW_ITEM_TYPE_VLAN; + pattern[1].spec = &vlan_spec; + pattern[1].mask = &vlan_mask; + + /* + * setting the third level of the pattern (ip). + * in this example this is the level we care about + * so we set it according to the parameters. + */ + memset(&ip_spec, 0, sizeof(struct rte_flow_item_ipv4)); + memset(&ip_mask, 0, sizeof(struct rte_flow_item_ipv4)); + ip_spec.hdr.dst_addr = htonl(dest_ip); + ip_mask.hdr.dst_addr = dest_mask; + ip_spec.hdr.src_addr = htonl(src_ip); + ip_mask.hdr.src_addr = src_mask; + pattern[2].type = RTE_FLOW_ITEM_TYPE_IPV4; + pattern[2].spec = &ip_spec; + pattern[2].mask = &ip_mask; + + /* the final level must be always type end */ + pattern[3].type = RTE_FLOW_ITEM_TYPE_END; + + res = rte_flow_validate(port_id, &attr, pattern, action, error); + if (!res) + flow = rte_flow_create(port_id, &attr, pattern, action, error); + + return flow; +} + diff --git a/examples/flow_filtering/main.c b/examples/flow_filtering/main.c new file mode 100644 index 00000000..7d739b4a --- /dev/null +++ b/examples/flow_filtering/main.c @@ -0,0 +1,244 @@ +/*- + * BSD LICENSE + * + * Copyright 2017 Mellanox. + * + * 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. + * * Neither the name of Mellanox. nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <stdint.h> +#include <inttypes.h> +#include <sys/types.h> +#include <sys/queue.h> +#include <netinet/in.h> +#include <setjmp.h> +#include <stdarg.h> +#include <ctype.h> +#include <errno.h> +#include <getopt.h> +#include <signal.h> +#include <stdbool.h> + +#include <rte_eal.h> +#include <rte_common.h> +#include <rte_malloc.h> +#include <rte_ether.h> +#include <rte_ethdev.h> +#include <rte_mempool.h> +#include <rte_mbuf.h> +#include <rte_net.h> +#include <rte_flow.h> + +static volatile bool force_quit; + +static uint8_t port_id; +static uint16_t nr_queues = 5; +static uint8_t selected_queue = 1; +struct rte_mempool *mbuf_pool; +struct rte_flow *flow; + +#define SRC_IP ((0<<24) + (0<<16) + (0<<8) + 0) /* src ip = 0.0.0.0 */ +#define DEST_IP ((192<<24) + (168<<16) + (1<<8) + 1) /* dest ip = 192.168.1.1 */ +#define FULL_MASK 0xffffffff /* full mask */ +#define EMPTY_MASK 0x0 /* empty mask */ + +#include "flow_blocks.c" + +static inline void +print_ether_addr(const char *what, struct ether_addr *eth_addr) +{ + char buf[ETHER_ADDR_FMT_SIZE]; + ether_format_addr(buf, ETHER_ADDR_FMT_SIZE, eth_addr); + printf("%s%s", what, buf); +} + +static void +main_loop(void) +{ + struct rte_mbuf *mbufs[32]; + struct ether_hdr *eth_hdr; + struct rte_flow_error error; + uint16_t nb_rx; + uint16_t i; + uint16_t j; + + while (!force_quit) { + for (i = 0; i < nr_queues; i++) { + nb_rx = rte_eth_rx_burst(port_id, + i, mbufs, 32); + if (nb_rx) { + for (j = 0; j < nb_rx; j++) { + struct rte_mbuf *m = mbufs[j]; + + eth_hdr = rte_pktmbuf_mtod(m, + struct ether_hdr *); + print_ether_addr("src=", + ð_hdr->s_addr); + print_ether_addr(" - dst=", + ð_hdr->d_addr); + printf(" - queue=0x%x", + (unsigned int)i); + printf("\n"); + + rte_pktmbuf_free(m); + } + } + } + } + + /* closing and releasing resources */ + rte_flow_flush(port_id, &error); + rte_eth_dev_stop(port_id); + rte_eth_dev_close(port_id); +} + +static void +assert_link_status(void) +{ + struct rte_eth_link link; + + memset(&link, 0, sizeof(link)); + rte_eth_link_get(port_id, &link); + if (link.link_status == ETH_LINK_DOWN) + rte_exit(EXIT_FAILURE, ":: error: link is still down\n"); +} + +static void +init_port(void) +{ + int ret; + uint16_t i; + struct rte_eth_conf port_conf = { + .rxmode = { + .split_hdr_size = 0, + /**< Header Split disabled */ + .header_split = 0, + /**< IP checksum offload disabled */ + .hw_ip_checksum = 0, + /**< VLAN filtering disabled */ + .hw_vlan_filter = 0, + /**< Jumbo Frame Support disabled */ + .jumbo_frame = 0, + /**< CRC stripped by hardware */ + .hw_strip_crc = 1, + }, + }; + + printf(":: initializing port: %d\n", port_id); + ret = rte_eth_dev_configure(port_id, + nr_queues, nr_queues, &port_conf); + if (ret < 0) { + rte_exit(EXIT_FAILURE, + ":: cannot configure device: err=%d, port=%u\n", + ret, port_id); + } + + /* only set Rx queues: something we care only so far */ + for (i = 0; i < nr_queues; i++) { + ret = rte_eth_rx_queue_setup(port_id, i, 512, + rte_eth_dev_socket_id(port_id), + NULL, + mbuf_pool); + if (ret < 0) { + rte_exit(EXIT_FAILURE, + ":: Rx queue setup failed: err=%d, port=%u\n", + ret, port_id); + } + } + + rte_eth_promiscuous_enable(port_id); + ret = rte_eth_dev_start(port_id); + if (ret < 0) { + rte_exit(EXIT_FAILURE, + "rte_eth_dev_start:err=%d, port=%u\n", + ret, port_id); + } + + assert_link_status(); + + printf(":: initializing port: %d done\n", port_id); +} + +static void +signal_handler(int signum) +{ + if (signum == SIGINT || signum == SIGTERM) { + printf("\n\nSignal %d received, preparing to exit...\n", + signum); + force_quit = true; + } +} + +int +main(int argc, char **argv) +{ + int ret; + uint8_t nr_ports; + struct rte_flow_error error; + + ret = rte_eal_init(argc, argv); + if (ret < 0) + rte_exit(EXIT_FAILURE, ":: invalid EAL arguments\n"); + + force_quit = false; + signal(SIGINT, signal_handler); + signal(SIGTERM, signal_handler); + + nr_ports = rte_eth_dev_count(); + if (nr_ports == 0) + rte_exit(EXIT_FAILURE, ":: no Ethernet ports found\n"); + port_id = 0; + if (nr_ports != 1) { + printf(":: warn: %d ports detected, but we use only one: port %u\n", + nr_ports, port_id); + } + mbuf_pool = rte_pktmbuf_pool_create("mbuf_pool", 4096, 128, 0, + RTE_MBUF_DEFAULT_BUF_SIZE, + rte_socket_id()); + if (mbuf_pool == NULL) + rte_exit(EXIT_FAILURE, "Cannot init mbuf pool\n"); + + init_port(); + + /* create flow for send packet with */ + flow = generate_ipv4_flow(port_id, selected_queue, + SRC_IP, EMPTY_MASK, + DEST_IP, FULL_MASK, &error); + if (!flow) { + printf("Flow can't be created %d message: %s\n", + error.type, + error.message ? error.message : "(no stated reason)"); + rte_exit(EXIT_FAILURE, "error in creating flow"); + } + + main_loop(); + + return 0; +} diff --git a/examples/helloworld/main.c b/examples/helloworld/main.c index 8b7a2de0..916b6ad0 100644 --- a/examples/helloworld/main.c +++ b/examples/helloworld/main.c @@ -38,7 +38,6 @@ #include <sys/queue.h> #include <rte_memory.h> -#include <rte_memzone.h> #include <rte_launch.h> #include <rte_eal.h> #include <rte_per_lcore.h> diff --git a/examples/ip_fragmentation/main.c b/examples/ip_fragmentation/main.c index 8c0e1791..5aefe098 100644 --- a/examples/ip_fragmentation/main.c +++ b/examples/ip_fragmentation/main.c @@ -48,7 +48,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -58,7 +57,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -154,7 +152,7 @@ struct rx_queue { struct rte_mempool *indirect_pool; struct rte_lpm *lpm; struct rte_lpm6 *lpm6; - uint8_t portid; + uint16_t portid; }; #define MAX_RX_QUEUE_PER_LCORE 16 @@ -240,7 +238,7 @@ static struct rte_lpm6 *socket_lpm6[RTE_MAX_NUMA_NODES]; /* Send burst of packets on an output interface */ static inline int -send_burst(struct lcore_queue_conf *qconf, uint16_t n, uint8_t port) +send_burst(struct lcore_queue_conf *qconf, uint16_t n, uint16_t port) { struct rte_mbuf **m_table; int ret; @@ -261,11 +259,12 @@ send_burst(struct lcore_queue_conf *qconf, uint16_t n, uint8_t port) static inline void l3fwd_simple_forward(struct rte_mbuf *m, struct lcore_queue_conf *qconf, - uint8_t queueid, uint8_t port_in) + uint8_t queueid, uint16_t port_in) { struct rx_queue *rxq; uint32_t i, len, next_hop; - uint8_t port_out, ipv6; + uint8_t ipv6; + uint16_t port_out; int32_t len2; ipv6 = 0; @@ -403,7 +402,7 @@ main_loop(__attribute__((unused)) void *dummy) unsigned lcore_id; uint64_t prev_tsc, diff_tsc, cur_tsc; int i, j, nb_rx; - uint8_t portid; + uint16_t portid; struct lcore_queue_conf *qconf; const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) / US_PER_S * BURST_TX_DRAIN_US; @@ -423,7 +422,7 @@ main_loop(__attribute__((unused)) void *dummy) portid = qconf->rx_queue_list[i].portid; RTE_LOG(INFO, IP_FRAG, " -- lcoreid=%u portid=%d\n", lcore_id, - (int) portid); + portid); } while (1) { @@ -600,11 +599,12 @@ print_ethaddr(const char *name, struct ether_addr *eth_addr) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -619,14 +619,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up .Speed %u Mbps - %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -708,7 +707,7 @@ parse_ptype(struct rte_mbuf *m) /* callback function to detect packet type for a queue of a port */ static uint16_t -cb_parse_ptype(uint8_t port __rte_unused, uint16_t queue __rte_unused, +cb_parse_ptype(uint16_t port __rte_unused, uint16_t queue __rte_unused, struct rte_mbuf *pkts[], uint16_t nb_pkts, uint16_t max_pkts __rte_unused, void *user_param __rte_unused) @@ -876,7 +875,7 @@ main(int argc, char **argv) uint16_t queueid = 0; unsigned lcore_id = 0, rx_lcore_id = 0; uint32_t n_tx_queue, nb_lcores; - uint8_t portid; + uint16_t portid; /* init EAL */ ret = rte_eal_init(argc, argv); @@ -1035,7 +1034,7 @@ main(int argc, char **argv) if (init_routing_table() < 0) rte_exit(EXIT_FAILURE, "Cannot init routing table\n"); - check_all_ports_link_status((uint8_t)nb_ports, enabled_port_mask); + check_all_ports_link_status(nb_ports, enabled_port_mask); /* launch per-lcore init on every lcore */ rte_eal_mp_remote_launch(main_loop, NULL, CALL_MASTER); diff --git a/examples/ip_pipeline/Makefile b/examples/ip_pipeline/Makefile index dc7e0ddd..12ce0a1d 100644 --- a/examples/ip_pipeline/Makefile +++ b/examples/ip_pipeline/Makefile @@ -43,7 +43,7 @@ APP = ip_pipeline VPATH += $(SRCDIR)/pipeline -INC += $(wildcard *.h) $(wildcard pipeline/*.h) +INC += $(sort $(wildcard *.h)) $(sort $(wildcard pipeline/*.h)) # all source are stored in SRCS-y SRCS-$(CONFIG_RTE_LIBRTE_PIPELINE) := main.c diff --git a/examples/ip_pipeline/app.h b/examples/ip_pipeline/app.h index e41290e7..94e7a6df 100644 --- a/examples/ip_pipeline/app.h +++ b/examples/ip_pipeline/app.h @@ -428,10 +428,6 @@ struct app_eal_params { /* Interrupt mode for VFIO (legacy|msi|msix) */ char *vfio_intr; - /* Support running on Xen dom0 without hugetlbfs */ - uint32_t xen_dom0_present; - int xen_dom0; - uint32_t parsed; }; diff --git a/examples/ip_pipeline/config_parse.c b/examples/ip_pipeline/config_parse.c index 0b761346..3211c6ab 100644 --- a/examples/ip_pipeline/config_parse.c +++ b/examples/ip_pipeline/config_parse.c @@ -809,21 +809,6 @@ parse_eal(struct app_params *app, continue; } - /* xen_dom0 */ - if (strcmp(entry->name, "xen_dom0") == 0) { - int val; - - PARSE_ERROR_DUPLICATE((p->xen_dom0_present == 0), - section_name, - entry->name); - p->xen_dom0_present = 1; - - val = parser_read_arg_bool(entry->value); - PARSE_ERROR((val >= 0), section_name, entry->name); - p->xen_dom0 = val; - continue; - } - /* unrecognized */ PARSE_ERROR_INVALID(0, section_name, entry->name); } @@ -2643,10 +2628,6 @@ save_eal_params(struct app_params *app, FILE *f) if (p->vfio_intr) fprintf(f, "%s = %s\n", "vfio_intr", p->vfio_intr); - if (p->xen_dom0_present) - fprintf(f, "%s = %s\n", "xen_dom0", - (p->xen_dom0) ? "yes" : "no"); - fputc('\n', f); } diff --git a/examples/ip_pipeline/init.c b/examples/ip_pipeline/init.c index 7cde49a4..e56e4048 100644 --- a/examples/ip_pipeline/init.c +++ b/examples/ip_pipeline/init.c @@ -49,6 +49,7 @@ #include <rte_ip.h> #include <rte_eal.h> #include <rte_malloc.h> +#include <rte_bus_pci.h> #include "app.h" #include "pipeline.h" @@ -296,11 +297,6 @@ app_init_eal(struct app_params *app) app->eal_argv[n_args++] = strdup(buffer); } - if ((p->xen_dom0_present) && (p->xen_dom0)) { - snprintf(buffer, sizeof(buffer), "--xen-dom0"); - app->eal_argv[n_args++] = strdup(buffer); - } - snprintf(buffer, sizeof(buffer), "--"); app->eal_argv[n_args++] = strdup(buffer); @@ -1236,7 +1232,7 @@ app_init_tap(struct app_params *app) #ifdef RTE_LIBRTE_KNI static int -kni_config_network_interface(uint8_t port_id, uint8_t if_up) { +kni_config_network_interface(uint16_t port_id, uint8_t if_up) { int ret = 0; if (port_id >= rte_eth_dev_count()) @@ -1250,7 +1246,7 @@ kni_config_network_interface(uint8_t port_id, uint8_t if_up) { } static int -kni_change_mtu(uint8_t port_id, unsigned new_mtu) { +kni_change_mtu(uint16_t port_id, unsigned int new_mtu) { int ret; if (port_id >= rte_eth_dev_count()) diff --git a/examples/ip_pipeline/pipeline/hash_func.h b/examples/ip_pipeline/pipeline/hash_func.h index b112369c..42128c1f 100644 --- a/examples/ip_pipeline/pipeline/hash_func.h +++ b/examples/ip_pipeline/pipeline/hash_func.h @@ -34,48 +34,56 @@ #define __INCLUDE_HASH_FUNC_H__ static inline uint64_t -hash_xor_key8(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_xor_key8(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t xor0; - xor0 = seed ^ k[0]; + xor0 = seed ^ (k[0] & m[0]); return (xor0 >> 32) ^ xor0; } static inline uint64_t -hash_xor_key16(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_xor_key16(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t xor0; - xor0 = (k[0] ^ seed) ^ k[1]; + xor0 = ((k[0] & m[0]) ^ seed) ^ (k[1] & m[1]); return (xor0 >> 32) ^ xor0; } static inline uint64_t -hash_xor_key24(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_xor_key24(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t xor0; - xor0 = (k[0] ^ seed) ^ k[1]; + xor0 = ((k[0] & m[0]) ^ seed) ^ (k[1] & m[1]); - xor0 ^= k[2]; + xor0 ^= k[2] & m[2]; return (xor0 >> 32) ^ xor0; } static inline uint64_t -hash_xor_key32(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_xor_key32(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t xor0, xor1; - xor0 = (k[0] ^ seed) ^ k[1]; - xor1 = k[2] ^ k[3]; + xor0 = ((k[0] & m[0]) ^ seed) ^ (k[1] & m[1]); + xor1 = (k[2] & m[2]) ^ (k[3] & m[3]); xor0 ^= xor1; @@ -83,30 +91,34 @@ hash_xor_key32(void *key, __rte_unused uint32_t key_size, uint64_t seed) } static inline uint64_t -hash_xor_key40(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_xor_key40(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t xor0, xor1; - xor0 = (k[0] ^ seed) ^ k[1]; - xor1 = k[2] ^ k[3]; + xor0 = ((k[0] & m[0]) ^ seed) ^ (k[1] & m[1]); + xor1 = (k[2] & m[2]) ^ (k[3] & m[3]); xor0 ^= xor1; - xor0 ^= k[4]; + xor0 ^= k[4] & m[4]; return (xor0 >> 32) ^ xor0; } static inline uint64_t -hash_xor_key48(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_xor_key48(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t xor0, xor1, xor2; - xor0 = (k[0] ^ seed) ^ k[1]; - xor1 = k[2] ^ k[3]; - xor2 = k[4] ^ k[5]; + xor0 = ((k[0] & m[0]) ^ seed) ^ (k[1] & m[1]); + xor1 = (k[2] & m[2]) ^ (k[3] & m[3]); + xor2 = (k[4] & m[4]) ^ (k[5] & m[5]); xor0 ^= xor1; @@ -116,17 +128,19 @@ hash_xor_key48(void *key, __rte_unused uint32_t key_size, uint64_t seed) } static inline uint64_t -hash_xor_key56(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_xor_key56(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t xor0, xor1, xor2; - xor0 = (k[0] ^ seed) ^ k[1]; - xor1 = k[2] ^ k[3]; - xor2 = k[4] ^ k[5]; + xor0 = ((k[0] & m[0]) ^ seed) ^ (k[1] & m[1]); + xor1 = (k[2] & m[2]) ^ (k[3] & m[3]); + xor2 = (k[4] & m[4]) ^ (k[5] & m[5]); xor0 ^= xor1; - xor2 ^= k[6]; + xor2 ^= k[6] & m[6]; xor0 ^= xor2; @@ -134,15 +148,17 @@ hash_xor_key56(void *key, __rte_unused uint32_t key_size, uint64_t seed) } static inline uint64_t -hash_xor_key64(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_xor_key64(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t xor0, xor1, xor2, xor3; - xor0 = (k[0] ^ seed) ^ k[1]; - xor1 = k[2] ^ k[3]; - xor2 = k[4] ^ k[5]; - xor3 = k[6] ^ k[7]; + xor0 = ((k[0] & m[0]) ^ seed) ^ (k[1] & m[1]); + xor1 = (k[2] & m[2]) ^ (k[3] & m[3]); + xor2 = (k[4] & m[4]) ^ (k[5] & m[5]); + xor3 = (k[6] & m[6]) ^ (k[7] & m[7]); xor0 ^= xor1; xor2 ^= xor3; @@ -157,26 +173,30 @@ hash_xor_key64(void *key, __rte_unused uint32_t key_size, uint64_t seed) #include <x86intrin.h> static inline uint64_t -hash_crc_key8(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_crc_key8(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t crc0; - crc0 = _mm_crc32_u64(seed, k[0]); + crc0 = _mm_crc32_u64(seed, k[0] & m[0]); return crc0; } static inline uint64_t -hash_crc_key16(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_crc_key16(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t k0, crc0, crc1; - k0 = k[0]; + k0 = k[0] & m[0]; crc0 = _mm_crc32_u64(k0, seed); - crc1 = _mm_crc32_u64(k0 >> 32, k[1]); + crc1 = _mm_crc32_u64(k0 >> 32, k[1] & m[1]); crc0 ^= crc1; @@ -184,16 +204,18 @@ hash_crc_key16(void *key, __rte_unused uint32_t key_size, uint64_t seed) } static inline uint64_t -hash_crc_key24(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_crc_key24(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t k0, k2, crc0, crc1; - k0 = k[0]; - k2 = k[2]; + k0 = k[0] & m[0]; + k2 = k[2] & m[2]; crc0 = _mm_crc32_u64(k0, seed); - crc1 = _mm_crc32_u64(k0 >> 32, k[1]); + crc1 = _mm_crc32_u64(k0 >> 32, k[1] & m[1]); crc0 = _mm_crc32_u64(crc0, k2); @@ -203,18 +225,20 @@ hash_crc_key24(void *key, __rte_unused uint32_t key_size, uint64_t seed) } static inline uint64_t -hash_crc_key32(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_crc_key32(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t k0, k2, crc0, crc1, crc2, crc3; - k0 = k[0]; - k2 = k[2]; + k0 = k[0] & m[0]; + k2 = k[2] & m[2]; crc0 = _mm_crc32_u64(k0, seed); - crc1 = _mm_crc32_u64(k0 >> 32, k[1]); + crc1 = _mm_crc32_u64(k0 >> 32, k[1] & m[1]); - crc2 = _mm_crc32_u64(k2, k[3]); + crc2 = _mm_crc32_u64(k2, k[3] & m[3]); crc3 = k2 >> 32; crc0 = _mm_crc32_u64(crc0, crc1); @@ -226,19 +250,21 @@ hash_crc_key32(void *key, __rte_unused uint32_t key_size, uint64_t seed) } static inline uint64_t -hash_crc_key40(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_crc_key40(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t k0, k2, crc0, crc1, crc2, crc3; - k0 = k[0]; - k2 = k[2]; + k0 = k[0] & m[0]; + k2 = k[2] & m[2]; crc0 = _mm_crc32_u64(k0, seed); - crc1 = _mm_crc32_u64(k0 >> 32, k[1]); + crc1 = _mm_crc32_u64(k0 >> 32, k[1] & m[1]); - crc2 = _mm_crc32_u64(k2, k[3]); - crc3 = _mm_crc32_u64(k2 >> 32, k[4]); + crc2 = _mm_crc32_u64(k2, k[3] & m[3]); + crc3 = _mm_crc32_u64(k2 >> 32, k[4] & m[4]); crc0 = _mm_crc32_u64(crc0, crc1); crc1 = _mm_crc32_u64(crc2, crc3); @@ -249,20 +275,22 @@ hash_crc_key40(void *key, __rte_unused uint32_t key_size, uint64_t seed) } static inline uint64_t -hash_crc_key48(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_crc_key48(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t k0, k2, k5, crc0, crc1, crc2, crc3; - k0 = k[0]; - k2 = k[2]; - k5 = k[5]; + k0 = k[0] & m[0]; + k2 = k[2] & m[2]; + k5 = k[5] & m[5]; crc0 = _mm_crc32_u64(k0, seed); - crc1 = _mm_crc32_u64(k0 >> 32, k[1]); + crc1 = _mm_crc32_u64(k0 >> 32, k[1] & m[1]); - crc2 = _mm_crc32_u64(k2, k[3]); - crc3 = _mm_crc32_u64(k2 >> 32, k[4]); + crc2 = _mm_crc32_u64(k2, k[3] & m[3]); + crc3 = _mm_crc32_u64(k2 >> 32, k[4] & m[4]); crc0 = _mm_crc32_u64(crc0, (crc1 << 32) ^ crc2); crc1 = _mm_crc32_u64(crc3, k5); @@ -273,22 +301,24 @@ hash_crc_key48(void *key, __rte_unused uint32_t key_size, uint64_t seed) } static inline uint64_t -hash_crc_key56(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_crc_key56(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t k0, k2, k5, crc0, crc1, crc2, crc3, crc4, crc5; - k0 = k[0]; - k2 = k[2]; - k5 = k[5]; + k0 = k[0] & m[0]; + k2 = k[2] & m[2]; + k5 = k[5] & m[5]; crc0 = _mm_crc32_u64(k0, seed); - crc1 = _mm_crc32_u64(k0 >> 32, k[1]); + crc1 = _mm_crc32_u64(k0 >> 32, k[1] & m[1]); - crc2 = _mm_crc32_u64(k2, k[3]); - crc3 = _mm_crc32_u64(k2 >> 32, k[4]); + crc2 = _mm_crc32_u64(k2, k[3] & m[3]); + crc3 = _mm_crc32_u64(k2 >> 32, k[4] & m[4]); - crc4 = _mm_crc32_u64(k5, k[6]); + crc4 = _mm_crc32_u64(k5, k[6] & m[6]); crc5 = k5 >> 32; crc0 = _mm_crc32_u64(crc0, (crc1 << 32) ^ crc2); @@ -300,23 +330,25 @@ hash_crc_key56(void *key, __rte_unused uint32_t key_size, uint64_t seed) } static inline uint64_t -hash_crc_key64(void *key, __rte_unused uint32_t key_size, uint64_t seed) +hash_crc_key64(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) { uint64_t *k = key; + uint64_t *m = mask; uint64_t k0, k2, k5, crc0, crc1, crc2, crc3, crc4, crc5; - k0 = k[0]; - k2 = k[2]; - k5 = k[5]; + k0 = k[0] & m[0]; + k2 = k[2] & m[2]; + k5 = k[5] & m[5]; crc0 = _mm_crc32_u64(k0, seed); - crc1 = _mm_crc32_u64(k0 >> 32, k[1]); + crc1 = _mm_crc32_u64(k0 >> 32, k[1] & m[1]); - crc2 = _mm_crc32_u64(k2, k[3]); - crc3 = _mm_crc32_u64(k2 >> 32, k[4]); + crc2 = _mm_crc32_u64(k2, k[3] & m[3]); + crc3 = _mm_crc32_u64(k2 >> 32, k[4] & m[4]); - crc4 = _mm_crc32_u64(k5, k[6]); - crc5 = _mm_crc32_u64(k5 >> 32, k[7]); + crc4 = _mm_crc32_u64(k5, k[6] & m[6]); + crc5 = _mm_crc32_u64(k5 >> 32, k[7] & m[7]); crc0 = _mm_crc32_u64(crc0, (crc1 << 32) ^ crc2); crc1 = _mm_crc32_u64(crc3, (crc4 << 32) ^ crc5); @@ -335,6 +367,8 @@ hash_crc_key64(void *key, __rte_unused uint32_t key_size, uint64_t seed) #define hash_default_key56 hash_crc_key56 #define hash_default_key64 hash_crc_key64 +#elif defined(RTE_ARCH_ARM64) +#include "hash_func_arm64.h" #else #define hash_default_key8 hash_xor_key8 diff --git a/examples/ip_pipeline/pipeline/hash_func_arm64.h b/examples/ip_pipeline/pipeline/hash_func_arm64.h new file mode 100644 index 00000000..ae6c0f41 --- /dev/null +++ b/examples/ip_pipeline/pipeline/hash_func_arm64.h @@ -0,0 +1,261 @@ +/*- + * BSD LICENSE + * + * Copyright(c) 2017 Linaro Limited. All rights reserved. + * All rights reserved. + * + * 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. + * * Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ +#ifndef __HASH_FUNC_ARM64_H__ +#define __HASH_FUNC_ARM64_H__ + +#define _CRC32CX(crc, val) \ + __asm__("crc32cx %w[c], %w[c], %x[v]":[c] "+r" (crc):[v] "r" (val)) + +static inline uint64_t +hash_crc_key8(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) +{ + uint64_t *k = key; + uint64_t *m = mask; + uint32_t crc0; + + crc0 = seed; + _CRC32CX(crc0, k[0] & m[0]); + + return crc0; +} + +static inline uint64_t +hash_crc_key16(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) +{ + uint64_t *k = key, k0; + uint64_t *m = mask; + uint32_t crc0, crc1; + + k0 = k[0] & m[0]; + + crc0 = k0; + _CRC32CX(crc0, seed); + crc1 = k0 >> 32; + _CRC32CX(crc1, k[1] & m[1]); + + crc0 ^= crc1; + + return crc0; +} + +static inline uint64_t +hash_crc_key24(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) +{ + uint64_t *k = key, k0, k2; + uint64_t *m = mask; + uint32_t crc0, crc1; + + k0 = k[0] & m[0]; + k2 = k[2] & m[2]; + + crc0 = k0; + _CRC32CX(crc0, seed); + crc1 = k0 >> 32; + _CRC32CX(crc1, k[1] & m[1]); + + _CRC32CX(crc0, k2); + + crc0 ^= crc1; + + return crc0; +} + +static inline uint64_t +hash_crc_key32(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) +{ + uint64_t *k = key, k0, k2; + uint64_t *m = mask; + uint32_t crc0, crc1, crc2, crc3; + + k0 = k[0] & m[0]; + k2 = k[2] & m[2]; + + crc0 = k0; + _CRC32CX(crc0, seed); + crc1 = k0 >> 32; + _CRC32CX(crc1, k[1] & m[1]); + + crc2 = k2; + _CRC32CX(crc2, k[3] & m[3]); + crc3 = k2 >> 32; + + _CRC32CX(crc0, crc1); + _CRC32CX(crc2, crc3); + + crc0 ^= crc2; + + return crc0; +} + +static inline uint64_t +hash_crc_key40(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) +{ + uint64_t *k = key, k0, k2; + uint64_t *m = mask; + uint32_t crc0, crc1, crc2, crc3; + + k0 = k[0] & m[0]; + k2 = k[2] & m[2]; + + crc0 = k0; + _CRC32CX(crc0, seed); + crc1 = k0 >> 32; + _CRC32CX(crc1, k[1] & m[1]); + + crc2 = k2; + _CRC32CX(crc2, k[3] & m[3]); + crc3 = k2 >> 32; + _CRC32CX(crc3, k[4] & m[4]); + + _CRC32CX(crc0, crc1); + _CRC32CX(crc2, crc3); + + crc0 ^= crc2; + + return crc0; +} + +static inline uint64_t +hash_crc_key48(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) +{ + uint64_t *k = key, k0, k2, k5; + uint64_t *m = mask; + uint32_t crc0, crc1, crc2, crc3; + + k0 = k[0] & m[0]; + k2 = k[2] & m[2]; + k5 = k[5] & m[5]; + + crc0 = k0; + _CRC32CX(crc0, seed); + crc1 = k0 >> 32; + _CRC32CX(crc1, k[1] & m[1]); + + crc2 = k2; + _CRC32CX(crc2, k[3] & m[3]); + crc3 = k2 >> 32; + _CRC32CX(crc3, k[4] & m[4]); + + _CRC32CX(crc0, ((uint64_t)crc1 << 32) ^ crc2); + _CRC32CX(crc3, k5); + + crc0 ^= crc3; + + return crc0; +} + +static inline uint64_t +hash_crc_key56(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) +{ + uint64_t *k = key, k0, k2, k5; + uint64_t *m = mask; + uint32_t crc0, crc1, crc2, crc3, crc4, crc5; + + k0 = k[0] & m[0]; + k2 = k[2] & m[2]; + k5 = k[5] & m[5]; + + crc0 = k0; + _CRC32CX(crc0, seed); + crc1 = k0 >> 32; + _CRC32CX(crc1, k[1] & m[1]); + + crc2 = k2; + _CRC32CX(crc2, k[3] & m[3]); + crc3 = k2 >> 32; + _CRC32CX(crc3, k[4] & m[4]); + + crc4 = k5; + _CRC32CX(crc4, k[6] & m[6]); + crc5 = k5 >> 32; + + _CRC32CX(crc0, ((uint64_t)crc1 << 32) ^ crc2); + _CRC32CX(crc3, ((uint64_t)crc4 << 32) ^ crc5); + + crc0 ^= crc3; + + return crc0; +} + +static inline uint64_t +hash_crc_key64(void *key, void *mask, __rte_unused uint32_t key_size, + uint64_t seed) +{ + uint64_t *k = key, k0, k2, k5; + uint64_t *m = mask; + uint32_t crc0, crc1, crc2, crc3, crc4, crc5; + + k0 = k[0] & m[0]; + k2 = k[2] & m[2]; + k5 = k[5] & m[5]; + + crc0 = k0; + _CRC32CX(crc0, seed); + crc1 = k0 >> 32; + _CRC32CX(crc1, k[1] & m[1]); + + crc2 = k2; + _CRC32CX(crc2, k[3] & m[3]); + crc3 = k2 >> 32; + _CRC32CX(crc3, k[4] & m[4]); + + crc4 = k5; + _CRC32CX(crc4, k[6] & m[6]); + crc5 = k5 >> 32; + _CRC32CX(crc5, k[7] & m[7]); + + _CRC32CX(crc0, ((uint64_t)crc1 << 32) ^ crc2); + _CRC32CX(crc3, ((uint64_t)crc4 << 32) ^ crc5); + + crc0 ^= crc3; + + return crc0; +} + +#define hash_default_key8 hash_crc_key8 +#define hash_default_key16 hash_crc_key16 +#define hash_default_key24 hash_crc_key24 +#define hash_default_key32 hash_crc_key32 +#define hash_default_key40 hash_crc_key40 +#define hash_default_key48 hash_crc_key48 +#define hash_default_key56 hash_crc_key56 +#define hash_default_key64 hash_crc_key64 + +#endif diff --git a/examples/ip_pipeline/pipeline/pipeline_flow_classification.c b/examples/ip_pipeline/pipeline/pipeline_flow_classification.c index 9ef50cc9..70b19381 100644 --- a/examples/ip_pipeline/pipeline/pipeline_flow_classification.c +++ b/examples/ip_pipeline/pipeline/pipeline_flow_classification.c @@ -88,8 +88,10 @@ app_pipeline_fc_key_convert(struct pipeline_fc_key *key_in, uint32_t *signature) { uint8_t buffer[PIPELINE_FC_FLOW_KEY_MAX_SIZE]; + uint8_t m[PIPELINE_FC_FLOW_KEY_MAX_SIZE]; /* key mask */ void *key_buffer = (key_out) ? key_out : buffer; + memset(m, 0xFF, sizeof(m)); switch (key_in->type) { case FLOW_KEY_QINQ: { @@ -101,7 +103,7 @@ app_pipeline_fc_key_convert(struct pipeline_fc_key *key_in, qinq->cvlan = rte_cpu_to_be_16(key_in->key.qinq.cvlan); if (signature) - *signature = (uint32_t) hash_default_key8(qinq, 8, 0); + *signature = (uint32_t) hash_default_key8(qinq, m, 8, 0); return 0; } @@ -118,7 +120,7 @@ app_pipeline_fc_key_convert(struct pipeline_fc_key *key_in, ipv4->port_dst = rte_cpu_to_be_16(key_in->key.ipv4_5tuple.port_dst); if (signature) - *signature = (uint32_t) hash_default_key16(ipv4, 16, 0); + *signature = (uint32_t) hash_default_key16(ipv4, m, 16, 0); return 0; } @@ -136,7 +138,7 @@ app_pipeline_fc_key_convert(struct pipeline_fc_key *key_in, ipv6->port_dst = rte_cpu_to_be_16(key_in->key.ipv6_5tuple.port_dst); if (signature) - *signature = (uint32_t) hash_default_key64(ipv6, 64, 0); + *signature = (uint32_t) hash_default_key64(ipv6, m, 64, 0); return 0; } @@ -832,12 +834,12 @@ app_pipeline_fc_add_bulk(struct app_params *app, } /* Free resources */ - app_msg_free(app, rsp); for (i = rsp->n_keys; i < n_keys; i++) if (new_flow[i]) rte_free(flow[i]); + app_msg_free(app, rsp); rte_free(flow_rsp); rte_free(flow_req); rte_free(new_flow); diff --git a/examples/ip_pipeline/pipeline/pipeline_flow_classification_be.c b/examples/ip_pipeline/pipeline/pipeline_flow_classification_be.c index 026f00cd..929d81cb 100644 --- a/examples/ip_pipeline/pipeline/pipeline_flow_classification_be.c +++ b/examples/ip_pipeline/pipeline/pipeline_flow_classification_be.c @@ -492,40 +492,16 @@ static void *pipeline_fc_init(struct pipeline_params *params, /* Tables */ p->n_tables = 1; { - struct rte_table_hash_key8_ext_params - table_hash_key8_params = { - .n_entries = p_fc->n_flows, - .n_entries_ext = p_fc->n_flows, - .signature_offset = p_fc->hash_offset, + struct rte_table_hash_params table_hash_params = { + .name = p->name, + .key_size = p_fc->key_size, .key_offset = p_fc->key_offset, - .f_hash = hash_func[(p_fc->key_size / 8) - 1], .key_mask = (p_fc->key_mask_present) ? p_fc->key_mask : NULL, - .seed = 0, - }; - - struct rte_table_hash_key16_ext_params - table_hash_key16_params = { - .n_entries = p_fc->n_flows, - .n_entries_ext = p_fc->n_flows, - .signature_offset = p_fc->hash_offset, - .key_offset = p_fc->key_offset, - .f_hash = hash_func[(p_fc->key_size / 8) - 1], - .key_mask = (p_fc->key_mask_present) ? - p_fc->key_mask : NULL, - .seed = 0, - }; - - struct rte_table_hash_ext_params - table_hash_params = { - .key_size = p_fc->key_size, .n_keys = p_fc->n_flows, - .n_buckets = p_fc->n_flows / 4, - .n_buckets_ext = p_fc->n_flows / 4, + .n_buckets = rte_align32pow2(p_fc->n_flows / 4), .f_hash = hash_func[(p_fc->key_size / 8) - 1], .seed = 0, - .signature_offset = p_fc->hash_offset, - .key_offset = p_fc->key_offset, }; struct rte_pipeline_table_params table_params = { @@ -542,32 +518,19 @@ static void *pipeline_fc_init(struct pipeline_params *params, switch (p_fc->key_size) { case 8: - if (p_fc->hash_offset != 0) { - table_params.ops = - &rte_table_hash_key8_ext_ops; - } else { - table_params.ops = - &rte_table_hash_key8_ext_dosig_ops; - } - table_params.arg_create = &table_hash_key8_params; + table_params.ops = &rte_table_hash_key8_ext_ops; break; case 16: - if (p_fc->hash_offset != 0) { - table_params.ops = - &rte_table_hash_key16_ext_ops; - } else { - table_params.ops = - &rte_table_hash_key16_ext_dosig_ops; - } - table_params.arg_create = &table_hash_key16_params; + table_params.ops = &rte_table_hash_key16_ext_ops; break; default: table_params.ops = &rte_table_hash_ext_ops; - table_params.arg_create = &table_hash_params; } + table_params.arg_create = &table_hash_params; + status = rte_pipeline_table_create(p->p, &table_params, &p->table_id[0]); diff --git a/examples/ip_pipeline/pipeline/pipeline_passthrough_be.c b/examples/ip_pipeline/pipeline/pipeline_passthrough_be.c index 8cb2f0c7..93eedbe8 100644 --- a/examples/ip_pipeline/pipeline/pipeline_passthrough_be.c +++ b/examples/ip_pipeline/pipeline/pipeline_passthrough_be.c @@ -102,7 +102,7 @@ pkt_work_dma( /* Read (dma_dst), compute (hash), write (hash) */ if (hash_enabled) { - uint32_t hash = p->f_hash(dma_dst, dma_size, 0); + uint32_t hash = p->f_hash(dma_src, dma_mask, dma_size, 0); *dma_hash = hash; if (lb_hash) { @@ -173,10 +173,10 @@ pkt4_work_dma( /* Read (dma_dst), compute (hash), write (hash) */ if (hash_enabled) { - uint32_t hash0 = p->f_hash(dma_dst0, dma_size, 0); - uint32_t hash1 = p->f_hash(dma_dst1, dma_size, 0); - uint32_t hash2 = p->f_hash(dma_dst2, dma_size, 0); - uint32_t hash3 = p->f_hash(dma_dst3, dma_size, 0); + uint32_t hash0 = p->f_hash(dma_src0, dma_mask, dma_size, 0); + uint32_t hash1 = p->f_hash(dma_src1, dma_mask, dma_size, 0); + uint32_t hash2 = p->f_hash(dma_src2, dma_mask, dma_size, 0); + uint32_t hash3 = p->f_hash(dma_src3, dma_mask, dma_size, 0); *dma_hash0 = hash0; *dma_hash1 = hash1; diff --git a/examples/ip_pipeline/pipeline/pipeline_routing_be.c b/examples/ip_pipeline/pipeline/pipeline_routing_be.c index 78317165..0414f248 100644 --- a/examples/ip_pipeline/pipeline/pipeline_routing_be.c +++ b/examples/ip_pipeline/pipeline/pipeline_routing_be.c @@ -1349,17 +1349,20 @@ pipeline_routing_init(struct pipeline_params *params, /* ARP table configuration */ if (p_rt->params.n_arp_entries) { - struct rte_table_hash_key8_ext_params table_arp_params = { - .n_entries = p_rt->params.n_arp_entries, - .n_entries_ext = p_rt->params.n_arp_entries, + struct rte_table_hash_params table_arp_params = { + .name = p->name, + .key_size = 8, + .key_offset = p_rt->params.arp_key_offset, + .key_mask = NULL, + .n_keys = p_rt->params.n_arp_entries, + .n_buckets = + rte_align32pow2(p_rt->params.n_arp_entries / 4), .f_hash = hash_default_key8, .seed = 0, - .signature_offset = 0, /* Unused */ - .key_offset = p_rt->params.arp_key_offset, }; struct rte_pipeline_table_params table_params = { - .ops = &rte_table_hash_key8_ext_dosig_ops, + .ops = &rte_table_hash_key8_ext_ops, .arg_create = &table_arp_params, .f_action_hit = get_arp_table_ah_hit(p_rt), .f_action_miss = NULL, diff --git a/examples/ip_reassembly/main.c b/examples/ip_reassembly/main.c index e62636cb..756f90ef 100644 --- a/examples/ip_reassembly/main.c +++ b/examples/ip_reassembly/main.c @@ -49,7 +49,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -59,7 +58,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -84,10 +82,10 @@ #define MAX_JUMBO_PKT_LEN 9600 #define BUF_SIZE RTE_MBUF_DEFAULT_DATAROOM -#define MBUF_SIZE \ - (BUF_SIZE + sizeof(struct rte_mbuf) + RTE_PKTMBUF_HEADROOM) +#define MBUF_DATA_SIZE RTE_MBUF_DEFAULT_BUF_SIZE #define NB_MBUF 8192 +#define MEMPOOL_CACHE_SIZE 256 /* allow max jumbo frame 9.5 KB */ #define JUMBO_FRAME_MAX_SIZE 0x2600 @@ -166,7 +164,7 @@ struct rx_queue { struct rte_mempool *pool; struct rte_lpm *lpm; struct rte_lpm6 *lpm6; - uint8_t portid; + uint16_t portid; }; struct tx_lcore_stat { @@ -277,7 +275,7 @@ static struct rte_lpm6 *socket_lpm6[RTE_MAX_NUMA_NODES]; * send burst of packets on an output interface. */ static inline uint32_t -send_burst(struct lcore_queue_conf *qconf, uint32_t thresh, uint8_t port) +send_burst(struct lcore_queue_conf *qconf, uint32_t thresh, uint16_t port) { uint32_t fill, len, k, n; struct mbuf_table *txmb; @@ -307,7 +305,7 @@ send_burst(struct lcore_queue_conf *qconf, uint32_t thresh, uint8_t port) /* Enqueue a single packet, and send burst if queue is filled */ static inline int -send_single_packet(struct rte_mbuf *m, uint8_t port) +send_single_packet(struct rte_mbuf *m, uint16_t port) { uint32_t fill, lcore_id, len; struct lcore_queue_conf *qconf; @@ -337,7 +335,7 @@ send_single_packet(struct rte_mbuf *m, uint8_t port) } static inline void -reassemble(struct rte_mbuf *m, uint8_t portid, uint32_t queue, +reassemble(struct rte_mbuf *m, uint16_t portid, uint32_t queue, struct lcore_queue_conf *qconf, uint64_t tms) { struct ether_hdr *eth_hdr; @@ -346,7 +344,7 @@ reassemble(struct rte_mbuf *m, uint8_t portid, uint32_t queue, struct rx_queue *rxq; void *d_addr_bytes; uint32_t next_hop; - uint8_t dst_port; + uint16_t dst_port; rxq = &qconf->rx_queue_list[queue]; @@ -454,7 +452,7 @@ main_loop(__attribute__((unused)) void *dummy) unsigned lcore_id; uint64_t diff_tsc, cur_tsc, prev_tsc; int i, j, nb_rx; - uint8_t portid; + uint16_t portid; struct lcore_queue_conf *qconf; const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) / US_PER_S * BURST_TX_DRAIN_US; @@ -473,7 +471,7 @@ main_loop(__attribute__((unused)) void *dummy) for (i = 0; i < qconf->n_rx_queue; i++) { portid = qconf->rx_queue_list[i].portid; - RTE_LOG(INFO, IP_RSMBL, " -- lcoreid=%u portid=%hhu\n", lcore_id, + RTE_LOG(INFO, IP_RSMBL, " -- lcoreid=%u portid=%u\n", lcore_id, portid); } @@ -732,11 +730,12 @@ print_ethaddr(const char *name, const struct ether_addr *eth_addr) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -751,14 +750,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up. Speed %u Mbps - %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -909,11 +907,11 @@ setup_queue_tbl(struct rx_queue *rxq, uint32_t lcore, uint32_t queue) snprintf(buf, sizeof(buf), "mbuf_pool_%u_%u", lcore, queue); - if ((rxq->pool = rte_mempool_create(buf, nb_mbuf, MBUF_SIZE, 0, - sizeof(struct rte_pktmbuf_pool_private), - rte_pktmbuf_pool_init, NULL, rte_pktmbuf_init, NULL, - socket, MEMPOOL_F_SP_PUT | MEMPOOL_F_SC_GET)) == NULL) { - RTE_LOG(ERR, IP_RSMBL, "mempool_create(%s) failed", buf); + rxq->pool = rte_pktmbuf_pool_create(buf, nb_mbuf, MEMPOOL_CACHE_SIZE, 0, + MBUF_DATA_SIZE, socket); + if (rxq->pool == NULL) { + RTE_LOG(ERR, IP_RSMBL, + "rte_pktmbuf_pool_create(%s) failed", buf); return -1; } @@ -987,7 +985,7 @@ queue_dump_stat(void) qconf = &lcore_queue_conf[lcore]; for (i = 0; i < qconf->n_rx_queue; i++) { - fprintf(stdout, " -- lcoreid=%u portid=%hhu " + fprintf(stdout, " -- lcoreid=%u portid=%u " "frag tbl stat:\n", lcore, qconf->rx_queue_list[i].portid); rte_ip_frag_table_statistics_dump(stdout, @@ -1024,7 +1022,7 @@ main(int argc, char **argv) uint16_t queueid; unsigned lcore_id = 0, rx_lcore_id = 0; uint32_t n_tx_queue, nb_lcores; - uint8_t portid; + uint16_t portid; /* init EAL */ ret = rte_eal_init(argc, argv); @@ -1177,7 +1175,7 @@ main(int argc, char **argv) if (init_routing_table() < 0) rte_exit(EXIT_FAILURE, "Cannot init routing table\n"); - check_all_ports_link_status((uint8_t)nb_ports, enabled_port_mask); + check_all_ports_link_status(nb_ports, enabled_port_mask); signal(SIGUSR1, signal_handler); signal(SIGTERM, signal_handler); diff --git a/examples/ipsec-secgw/Makefile b/examples/ipsec-secgw/Makefile index 17e91551..9fd33cb7 100644 --- a/examples/ipsec-secgw/Makefile +++ b/examples/ipsec-secgw/Makefile @@ -38,6 +38,12 @@ RTE_TARGET ?= x86_64-native-linuxapp-gcc include $(RTE_SDK)/mk/rte.vars.mk +ifneq ($(MAKECMDGOALS),clean) +ifneq ($(CONFIG_RTE_LIBRTE_SECURITY),y) +$(error "RTE_LIBRTE_SECURITY is required to build ipsec-secgw") +endif +endif + APP = ipsec-secgw CFLAGS += -O3 -gdwarf-2 diff --git a/examples/ipsec-secgw/esp.c b/examples/ipsec-secgw/esp.c index 70bb81f7..c3efe52b 100644 --- a/examples/ipsec-secgw/esp.c +++ b/examples/ipsec-secgw/esp.c @@ -58,8 +58,11 @@ esp_inbound(struct rte_mbuf *m, struct ipsec_sa *sa, struct rte_crypto_sym_op *sym_cop; int32_t payload_len, ip_hdr_len; - RTE_ASSERT(m != NULL); RTE_ASSERT(sa != NULL); + if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO) + return 0; + + RTE_ASSERT(m != NULL); RTE_ASSERT(cop != NULL); ip4 = rte_pktmbuf_mtod(m, struct ip *); @@ -103,12 +106,12 @@ esp_inbound(struct rte_mbuf *m, struct ipsec_sa *sa, aad = get_aad(m); memcpy(aad, iv - sizeof(struct esp_hdr), 8); sym_cop->aead.aad.data = aad; - sym_cop->aead.aad.phys_addr = rte_pktmbuf_mtophys_offset(m, + sym_cop->aead.aad.phys_addr = rte_pktmbuf_iova_offset(m, aad - rte_pktmbuf_mtod(m, uint8_t *)); sym_cop->aead.digest.data = rte_pktmbuf_mtod_offset(m, void*, rte_pktmbuf_pkt_len(m) - sa->digest_len); - sym_cop->aead.digest.phys_addr = rte_pktmbuf_mtophys_offset(m, + sym_cop->aead.digest.phys_addr = rte_pktmbuf_iova_offset(m, rte_pktmbuf_pkt_len(m) - sa->digest_len); } else { sym_cop->cipher.data.offset = ip_hdr_len + sizeof(struct esp_hdr) + @@ -154,7 +157,7 @@ esp_inbound(struct rte_mbuf *m, struct ipsec_sa *sa, sym_cop->auth.digest.data = rte_pktmbuf_mtod_offset(m, void*, rte_pktmbuf_pkt_len(m) - sa->digest_len); - sym_cop->auth.digest.phys_addr = rte_pktmbuf_mtophys_offset(m, + sym_cop->auth.digest.phys_addr = rte_pktmbuf_iova_offset(m, rte_pktmbuf_pkt_len(m) - sa->digest_len); } @@ -175,29 +178,44 @@ esp_inbound_post(struct rte_mbuf *m, struct ipsec_sa *sa, RTE_ASSERT(sa != NULL); RTE_ASSERT(cop != NULL); + if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO) { + if (m->ol_flags & PKT_RX_SEC_OFFLOAD) { + if (m->ol_flags & PKT_RX_SEC_OFFLOAD_FAILED) + cop->status = RTE_CRYPTO_OP_STATUS_ERROR; + else + cop->status = RTE_CRYPTO_OP_STATUS_SUCCESS; + } else + cop->status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED; + } + if (cop->status != RTE_CRYPTO_OP_STATUS_SUCCESS) { RTE_LOG(ERR, IPSEC_ESP, "failed crypto op\n"); return -1; } - nexthdr = rte_pktmbuf_mtod_offset(m, uint8_t*, - rte_pktmbuf_pkt_len(m) - sa->digest_len - 1); - pad_len = nexthdr - 1; + if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO && + sa->ol_flags & RTE_SECURITY_RX_HW_TRAILER_OFFLOAD) { + nexthdr = &m->inner_esp_next_proto; + } else { + nexthdr = rte_pktmbuf_mtod_offset(m, uint8_t*, + rte_pktmbuf_pkt_len(m) - sa->digest_len - 1); + pad_len = nexthdr - 1; + + padding = pad_len - *pad_len; + for (i = 0; i < *pad_len; i++) { + if (padding[i] != i + 1) { + RTE_LOG(ERR, IPSEC_ESP, "invalid padding\n"); + return -EINVAL; + } + } - padding = pad_len - *pad_len; - for (i = 0; i < *pad_len; i++) { - if (padding[i] != i + 1) { - RTE_LOG(ERR, IPSEC_ESP, "invalid padding\n"); + if (rte_pktmbuf_trim(m, *pad_len + 2 + sa->digest_len)) { + RTE_LOG(ERR, IPSEC_ESP, + "failed to remove pad_len + digest\n"); return -EINVAL; } } - if (rte_pktmbuf_trim(m, *pad_len + 2 + sa->digest_len)) { - RTE_LOG(ERR, IPSEC_ESP, - "failed to remove pad_len + digest\n"); - return -EINVAL; - } - if (unlikely(sa->flags == TRANSPORT)) { ip = rte_pktmbuf_mtod(m, struct ip *); ip4 = (struct ip *)rte_pktmbuf_adj(m, @@ -211,7 +229,8 @@ esp_inbound_post(struct rte_mbuf *m, struct ipsec_sa *sa, /* XXX No option headers supported */ memmove(ip6, ip, sizeof(struct ip6_hdr)); ip6->ip6_nxt = *nexthdr; - ip6->ip6_plen = htons(rte_pktmbuf_data_len(m)); + ip6->ip6_plen = htons(rte_pktmbuf_data_len(m) - + sizeof(struct ip6_hdr)); } } else ipip_inbound(m, sizeof(struct esp_hdr) + sa->iv_len); @@ -226,14 +245,13 @@ esp_outbound(struct rte_mbuf *m, struct ipsec_sa *sa, struct ip *ip4; struct ip6_hdr *ip6; struct esp_hdr *esp = NULL; - uint8_t *padding, *new_ip, nlp; + uint8_t *padding = NULL, *new_ip, nlp; struct rte_crypto_sym_op *sym_cop; int32_t i; uint16_t pad_payload_len, pad_len, ip_hdr_len; RTE_ASSERT(m != NULL); RTE_ASSERT(sa != NULL); - RTE_ASSERT(cop != NULL); ip_hdr_len = 0; @@ -283,12 +301,19 @@ esp_outbound(struct rte_mbuf *m, struct ipsec_sa *sa, return -EINVAL; } - padding = (uint8_t *)rte_pktmbuf_append(m, pad_len + sa->digest_len); - if (unlikely(padding == NULL)) { - RTE_LOG(ERR, IPSEC_ESP, "not enough mbuf trailing space\n"); - return -ENOSPC; + /* Add trailer padding if it is not constructed by HW */ + if (sa->type != RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO || + (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO && + !(sa->ol_flags & RTE_SECURITY_TX_HW_TRAILER_OFFLOAD))) { + padding = (uint8_t *)rte_pktmbuf_append(m, pad_len + + sa->digest_len); + if (unlikely(padding == NULL)) { + RTE_LOG(ERR, IPSEC_ESP, + "not enough mbuf trailing space\n"); + return -ENOSPC; + } + rte_prefetch0(padding); } - rte_prefetch0(padding); switch (sa->flags) { case IP4_TUNNEL: @@ -306,14 +331,15 @@ esp_outbound(struct rte_mbuf *m, struct ipsec_sa *sa, sizeof(struct esp_hdr) + sa->iv_len); memmove(new_ip, ip4, ip_hdr_len); esp = (struct esp_hdr *)(new_ip + ip_hdr_len); + ip4 = (struct ip *)new_ip; if (likely(ip4->ip_v == IPVERSION)) { - ip4 = (struct ip *)new_ip; ip4->ip_p = IPPROTO_ESP; ip4->ip_len = htons(rte_pktmbuf_data_len(m)); } else { ip6 = (struct ip6_hdr *)new_ip; ip6->ip6_nxt = IPPROTO_ESP; - ip6->ip6_plen = htons(rte_pktmbuf_data_len(m)); + ip6->ip6_plen = htons(rte_pktmbuf_data_len(m) - + sizeof(struct ip6_hdr)); } } @@ -321,15 +347,46 @@ esp_outbound(struct rte_mbuf *m, struct ipsec_sa *sa, esp->spi = rte_cpu_to_be_32(sa->spi); esp->seq = rte_cpu_to_be_32((uint32_t)sa->seq); + /* set iv */ uint64_t *iv = (uint64_t *)(esp + 1); + if (sa->aead_algo == RTE_CRYPTO_AEAD_AES_GCM) { + *iv = rte_cpu_to_be_64(sa->seq); + } else { + switch (sa->cipher_algo) { + case RTE_CRYPTO_CIPHER_NULL: + case RTE_CRYPTO_CIPHER_AES_CBC: + memset(iv, 0, sa->iv_len); + break; + case RTE_CRYPTO_CIPHER_AES_CTR: + *iv = rte_cpu_to_be_64(sa->seq); + break; + default: + RTE_LOG(ERR, IPSEC_ESP, + "unsupported cipher algorithm %u\n", + sa->cipher_algo); + return -EINVAL; + } + } + if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO) { + if (sa->ol_flags & RTE_SECURITY_TX_HW_TRAILER_OFFLOAD) { + /* Set the inner esp next protocol for HW trailer */ + m->inner_esp_next_proto = nlp; + m->packet_type |= RTE_PTYPE_TUNNEL_ESP; + } else { + padding[pad_len - 2] = pad_len - 2; + padding[pad_len - 1] = nlp; + } + goto done; + } + + RTE_ASSERT(cop != NULL); sym_cop = get_sym_cop(cop); sym_cop->m_src = m; if (sa->aead_algo == RTE_CRYPTO_AEAD_AES_GCM) { uint8_t *aad; - *iv = sa->seq; sym_cop->aead.data.offset = ip_hdr_len + sizeof(struct esp_hdr) + sa->iv_len; sym_cop->aead.data.length = pad_payload_len; @@ -342,30 +399,28 @@ esp_outbound(struct rte_mbuf *m, struct ipsec_sa *sa, struct cnt_blk *icb = get_cnt_blk(m); icb->salt = sa->salt; - icb->iv = sa->seq; + icb->iv = rte_cpu_to_be_64(sa->seq); icb->cnt = rte_cpu_to_be_32(1); aad = get_aad(m); memcpy(aad, esp, 8); sym_cop->aead.aad.data = aad; - sym_cop->aead.aad.phys_addr = rte_pktmbuf_mtophys_offset(m, + sym_cop->aead.aad.phys_addr = rte_pktmbuf_iova_offset(m, aad - rte_pktmbuf_mtod(m, uint8_t *)); sym_cop->aead.digest.data = rte_pktmbuf_mtod_offset(m, uint8_t *, rte_pktmbuf_pkt_len(m) - sa->digest_len); - sym_cop->aead.digest.phys_addr = rte_pktmbuf_mtophys_offset(m, + sym_cop->aead.digest.phys_addr = rte_pktmbuf_iova_offset(m, rte_pktmbuf_pkt_len(m) - sa->digest_len); } else { switch (sa->cipher_algo) { case RTE_CRYPTO_CIPHER_NULL: case RTE_CRYPTO_CIPHER_AES_CBC: - memset(iv, 0, sa->iv_len); sym_cop->cipher.data.offset = ip_hdr_len + sizeof(struct esp_hdr); sym_cop->cipher.data.length = pad_payload_len + sa->iv_len; break; case RTE_CRYPTO_CIPHER_AES_CTR: - *iv = sa->seq; sym_cop->cipher.data.offset = ip_hdr_len + sizeof(struct esp_hdr) + sa->iv_len; sym_cop->cipher.data.length = pad_payload_len; @@ -384,7 +439,7 @@ esp_outbound(struct rte_mbuf *m, struct ipsec_sa *sa, struct cnt_blk *icb = get_cnt_blk(m); icb->salt = sa->salt; - icb->iv = sa->seq; + icb->iv = rte_cpu_to_be_64(sa->seq); icb->cnt = rte_cpu_to_be_32(1); switch (sa->auth_algo) { @@ -403,25 +458,30 @@ esp_outbound(struct rte_mbuf *m, struct ipsec_sa *sa, sym_cop->auth.digest.data = rte_pktmbuf_mtod_offset(m, uint8_t *, rte_pktmbuf_pkt_len(m) - sa->digest_len); - sym_cop->auth.digest.phys_addr = rte_pktmbuf_mtophys_offset(m, + sym_cop->auth.digest.phys_addr = rte_pktmbuf_iova_offset(m, rte_pktmbuf_pkt_len(m) - sa->digest_len); } +done: return 0; } int -esp_outbound_post(struct rte_mbuf *m __rte_unused, - struct ipsec_sa *sa __rte_unused, - struct rte_crypto_op *cop) +esp_outbound_post(struct rte_mbuf *m, + struct ipsec_sa *sa, + struct rte_crypto_op *cop) { RTE_ASSERT(m != NULL); RTE_ASSERT(sa != NULL); - RTE_ASSERT(cop != NULL); - if (cop->status != RTE_CRYPTO_OP_STATUS_SUCCESS) { - RTE_LOG(ERR, IPSEC_ESP, "Failed crypto op\n"); - return -1; + if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO) { + m->ol_flags |= PKT_TX_SEC_OFFLOAD; + } else { + RTE_ASSERT(cop != NULL); + if (cop->status != RTE_CRYPTO_OP_STATUS_SUCCESS) { + RTE_LOG(ERR, IPSEC_ESP, "Failed crypto op\n"); + return -1; + } } return 0; diff --git a/examples/ipsec-secgw/esp.h b/examples/ipsec-secgw/esp.h index fa5cc8af..23601e37 100644 --- a/examples/ipsec-secgw/esp.h +++ b/examples/ipsec-secgw/esp.h @@ -35,16 +35,6 @@ struct mbuf; -/* RFC4303 */ -struct esp_hdr { - uint32_t spi; - uint32_t seq; - /* Payload */ - /* Padding */ - /* Pad Length */ - /* Next Header */ - /* Integrity Check Value - ICV */ -}; int esp_inbound(struct rte_mbuf *m, struct ipsec_sa *sa, diff --git a/examples/ipsec-secgw/ipip.h b/examples/ipsec-secgw/ipip.h index ff1dccdb..93393d52 100644 --- a/examples/ipsec-secgw/ipip.h +++ b/examples/ipsec-secgw/ipip.h @@ -72,7 +72,8 @@ ipip_outbound(struct rte_mbuf *m, uint32_t offset, uint32_t is_ipv6, /* Per RFC4301 5.1.2.1 */ outip6->ip6_flow = htonl(IP6_VERSION << 28 | ds_ecn << 20); - outip6->ip6_plen = htons(rte_pktmbuf_data_len(m)); + outip6->ip6_plen = htons(rte_pktmbuf_data_len(m) - + sizeof(struct ip6_hdr)); outip6->ip6_nxt = IPPROTO_ESP; outip6->ip6_hops = IPDEFTTL; diff --git a/examples/ipsec-secgw/ipsec-secgw.c b/examples/ipsec-secgw/ipsec-secgw.c index 99dc270c..c98454a9 100644 --- a/examples/ipsec-secgw/ipsec-secgw.c +++ b/examples/ipsec-secgw/ipsec-secgw.c @@ -57,7 +57,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -161,14 +160,15 @@ static int32_t numa_on = 1; /**< NUMA is enabled by default. */ static uint32_t nb_lcores; static uint32_t single_sa; static uint32_t single_sa_idx; +static uint32_t frame_size; struct lcore_rx_queue { - uint8_t port_id; + uint16_t port_id; uint8_t queue_id; } __rte_cache_aligned; struct lcore_params { - uint8_t port_id; + uint16_t port_id; uint8_t queue_id; uint8_t lcore_id; } __rte_cache_aligned; @@ -204,11 +204,9 @@ static struct rte_eth_conf port_conf = { .mq_mode = ETH_MQ_RX_RSS, .max_rx_pkt_len = ETHER_MAX_LEN, .split_hdr_size = 0, - .header_split = 0, /**< Header Split disabled */ - .hw_ip_checksum = 1, /**< IP checksum offload enabled */ - .hw_vlan_filter = 0, /**< VLAN filtering disabled */ - .jumbo_frame = 0, /**< Jumbo Frame Support disabled */ - .hw_strip_crc = 1, /**< CRC stripped by hardware */ + .offloads = DEV_RX_OFFLOAD_CHECKSUM | + DEV_RX_OFFLOAD_CRC_STRIP, + .ignore_offload_bitfield = 1, }, .rx_adv_conf = { .rss_conf = { @@ -290,7 +288,7 @@ prepare_traffic(struct rte_mbuf **pkts, struct ipsec_traffic *t, } static inline void -prepare_tx_pkt(struct rte_mbuf *pkt, uint8_t port) +prepare_tx_pkt(struct rte_mbuf *pkt, uint16_t port) { struct ip *ip; struct ether_hdr *ethhdr; @@ -320,7 +318,7 @@ prepare_tx_pkt(struct rte_mbuf *pkt, uint8_t port) } static inline void -prepare_tx_burst(struct rte_mbuf *pkts[], uint16_t nb_pkts, uint8_t port) +prepare_tx_burst(struct rte_mbuf *pkts[], uint16_t nb_pkts, uint16_t port) { int32_t i; const int32_t prefetch_offset = 2; @@ -336,7 +334,7 @@ prepare_tx_burst(struct rte_mbuf *pkts[], uint16_t nb_pkts, uint8_t port) /* Send burst of packets on an output interface */ static inline int32_t -send_burst(struct lcore_conf *qconf, uint16_t n, uint8_t port) +send_burst(struct lcore_conf *qconf, uint16_t n, uint16_t port) { struct rte_mbuf **m_table; int32_t ret; @@ -359,7 +357,7 @@ send_burst(struct lcore_conf *qconf, uint16_t n, uint8_t port) /* Enqueue a single packet, and send burst if queue is filled */ static inline int32_t -send_single_packet(struct rte_mbuf *m, uint8_t port) +send_single_packet(struct rte_mbuf *m, uint16_t port) { uint32_t lcore_id; uint16_t len; @@ -646,7 +644,7 @@ route6_pkts(struct rt_ctx *rt_ctx, struct rte_mbuf *pkts[], uint8_t nb_pkts) static inline void process_pkts(struct lcore_conf *qconf, struct rte_mbuf **pkts, - uint8_t nb_pkts, uint8_t portid) + uint8_t nb_pkts, uint16_t portid) { struct ipsec_traffic traffic; @@ -691,7 +689,8 @@ main_loop(__attribute__((unused)) void *dummy) uint32_t lcore_id; uint64_t prev_tsc, diff_tsc, cur_tsc; int32_t i, nb_rx; - uint8_t portid, queueid; + uint16_t portid; + uint8_t queueid; struct lcore_conf *qconf; int32_t socket_id; const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) @@ -728,7 +727,7 @@ main_loop(__attribute__((unused)) void *dummy) portid = rxql[i].port_id; queueid = rxql[i].queue_id; RTE_LOG(INFO, IPSEC, - " -- lcoreid=%u portid=%hhu rxqueueid=%hhu\n", + " -- lcoreid=%u portid=%u rxqueueid=%hhu\n", lcore_id, portid, queueid); } @@ -759,7 +758,8 @@ main_loop(__attribute__((unused)) void *dummy) static int32_t check_params(void) { - uint8_t lcore, portid, nb_ports; + uint8_t lcore; + uint16_t portid, nb_ports; uint16_t i; int32_t socket_id; @@ -797,7 +797,7 @@ check_params(void) } static uint8_t -get_port_nb_rx_queues(const uint8_t port) +get_port_nb_rx_queues(const uint16_t port) { int32_t queue = -1; uint16_t i; @@ -843,6 +843,7 @@ print_usage(const char *prgname) " -p PORTMASK: hexadecimal bitmask of ports to configure\n" " -P : enable promiscuous mode\n" " -u PORTMASK: hexadecimal bitmask of unprotected ports\n" + " -j FRAMESIZE: jumbo frame maximum size\n" " --"OPTION_CONFIG": (port,queue,lcore): " "rx queues configuration\n" " --single-sa SAIDX: use single SA index for outbound, " @@ -981,7 +982,7 @@ parse_args(int32_t argc, char **argv) argvopt = argv; - while ((opt = getopt_long(argc, argvopt, "p:Pu:f:", + while ((opt = getopt_long(argc, argvopt, "p:Pu:f:j:", lgopts, &option_index)) != EOF) { switch (opt) { @@ -1020,6 +1021,23 @@ parse_args(int32_t argc, char **argv) } f_present = 1; break; + case 'j': + { + int32_t size = parse_decimal(optarg); + if (size <= 1518) { + printf("Invalid jumbo frame size\n"); + if (size < 0) { + print_usage(prgname); + return -1; + } + printf("Using default value 9000\n"); + frame_size = 9000; + } else { + frame_size = size; + } + } + printf("Enabled jumbo frames size %u\n", frame_size); + break; case 0: if (parse_args_long_options(lgopts, option_index)) { print_usage(prgname); @@ -1055,11 +1073,12 @@ print_ethaddr(const char *name, const struct ether_addr *eth_addr) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -1074,14 +1093,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (uint32_t)link.link_speed, + printf( + "Port%d Link Up - speed %u Mbps -%s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -1113,7 +1131,8 @@ add_mapping(struct rte_hash *map, const char *str, uint16_t cdev_id, uint16_t qp, struct lcore_params *params, struct ipsec_ctx *ipsec_ctx, const struct rte_cryptodev_capabilities *cipher, - const struct rte_cryptodev_capabilities *auth) + const struct rte_cryptodev_capabilities *auth, + const struct rte_cryptodev_capabilities *aead) { int32_t ret = 0; unsigned long i; @@ -1124,6 +1143,8 @@ add_mapping(struct rte_hash *map, const char *str, uint16_t cdev_id, key.cipher_algo = cipher->sym.cipher.algo; if (auth) key.auth_algo = auth->sym.auth.algo; + if (aead) + key.aead_algo = aead->sym.aead.algo; ret = rte_hash_lookup(map, &key); if (ret != -ENOENT) @@ -1192,6 +1213,12 @@ add_cdev_mapping(struct rte_cryptodev_info *dev_info, uint16_t cdev_id, if (i->op != RTE_CRYPTO_OP_TYPE_SYMMETRIC) continue; + if (i->sym.xform_type == RTE_CRYPTO_SYM_XFORM_AEAD) { + ret |= add_mapping(map, str, cdev_id, qp, params, + ipsec_ctx, NULL, NULL, i); + continue; + } + if (i->sym.xform_type != RTE_CRYPTO_SYM_XFORM_CIPHER) continue; @@ -1204,7 +1231,7 @@ add_cdev_mapping(struct rte_cryptodev_info *dev_info, uint16_t cdev_id, continue; ret |= add_mapping(map, str, cdev_id, qp, params, - ipsec_ctx, i, j); + ipsec_ctx, i, j, NULL); } } @@ -1322,7 +1349,7 @@ cryptodevs_init(void) } static void -port_init(uint8_t portid) +port_init(uint16_t portid) { struct rte_eth_dev_info dev_info; struct rte_eth_txconf *txconf; @@ -1357,6 +1384,16 @@ port_init(uint8_t portid) printf("Creating queues: nb_rx_queue=%d nb_tx_queue=%u...\n", nb_rx_queue, nb_tx_queue); + if (frame_size) { + port_conf.rxmode.max_rx_pkt_len = frame_size; + port_conf.rxmode.offloads |= DEV_RX_OFFLOAD_JUMBO_FRAME; + } + + if (dev_info.rx_offload_capa & DEV_RX_OFFLOAD_SECURITY) + port_conf.rxmode.offloads |= DEV_RX_OFFLOAD_SECURITY; + if (dev_info.tx_offload_capa & DEV_TX_OFFLOAD_SECURITY) + port_conf.txmode.offloads |= DEV_TX_OFFLOAD_SECURITY; + ret = rte_eth_dev_configure(portid, nb_rx_queue, nb_tx_queue, &port_conf); if (ret < 0) @@ -1421,11 +1458,14 @@ static void pool_init(struct socket_ctx *ctx, int32_t socket_id, uint32_t nb_mbuf) { char s[64]; + uint32_t buff_size = frame_size ? (frame_size + RTE_PKTMBUF_HEADROOM) : + RTE_MBUF_DEFAULT_BUF_SIZE; + snprintf(s, sizeof(s), "mbuf_pool_%d", socket_id); ctx->mbuf_pool = rte_pktmbuf_pool_create(s, nb_mbuf, MEMPOOL_CACHE_SIZE, ipsec_metadata_size(), - RTE_MBUF_DEFAULT_BUF_SIZE, + buff_size, socket_id); if (ctx->mbuf_pool == NULL) rte_exit(EXIT_FAILURE, "Cannot init mbuf pool on socket %d\n", @@ -1438,8 +1478,9 @@ int32_t main(int32_t argc, char **argv) { int32_t ret; - uint32_t lcore_id, nb_ports; - uint8_t portid, socket_id; + uint32_t lcore_id; + uint8_t socket_id; + uint16_t portid, nb_ports; /* init EAL */ ret = rte_eal_init(argc, argv); @@ -1522,7 +1563,7 @@ main(int32_t argc, char **argv) rte_eth_promiscuous_enable(portid); } - check_all_ports_link_status((uint8_t)nb_ports, enabled_port_mask); + check_all_ports_link_status(nb_ports, enabled_port_mask); /* launch per-lcore init on every lcore */ rte_eal_mp_remote_launch(main_loop, NULL, CALL_MASTER); diff --git a/examples/ipsec-secgw/ipsec.c b/examples/ipsec-secgw/ipsec.c index 0afb9d67..c24284d6 100644 --- a/examples/ipsec-secgw/ipsec.c +++ b/examples/ipsec-secgw/ipsec.c @@ -37,7 +37,9 @@ #include <rte_branch_prediction.h> #include <rte_log.h> #include <rte_crypto.h> +#include <rte_security.h> #include <rte_cryptodev.h> +#include <rte_ethdev.h> #include <rte_mbuf.h> #include <rte_hash.h> @@ -49,21 +51,28 @@ create_session(struct ipsec_ctx *ipsec_ctx, struct ipsec_sa *sa) { struct rte_cryptodev_info cdev_info; unsigned long cdev_id_qp = 0; - int32_t ret; + int32_t ret = 0; struct cdev_key key = { 0 }; key.lcore_id = (uint8_t)rte_lcore_id(); key.cipher_algo = (uint8_t)sa->cipher_algo; key.auth_algo = (uint8_t)sa->auth_algo; + key.aead_algo = (uint8_t)sa->aead_algo; - ret = rte_hash_lookup_data(ipsec_ctx->cdev_map, &key, - (void **)&cdev_id_qp); - if (ret < 0) { - RTE_LOG(ERR, IPSEC, "No cryptodev: core %u, cipher_algo %u, " - "auth_algo %u\n", key.lcore_id, key.cipher_algo, - key.auth_algo); - return -1; + if (sa->type == RTE_SECURITY_ACTION_TYPE_NONE) { + ret = rte_hash_lookup_data(ipsec_ctx->cdev_map, &key, + (void **)&cdev_id_qp); + if (ret < 0) { + RTE_LOG(ERR, IPSEC, + "No cryptodev: core %u, cipher_algo %u, " + "auth_algo %u, aead_algo %u\n", + key.lcore_id, + key.cipher_algo, + key.auth_algo, + key.aead_algo); + return -1; + } } RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on cryptodev " @@ -71,23 +80,153 @@ create_session(struct ipsec_ctx *ipsec_ctx, struct ipsec_sa *sa) ipsec_ctx->tbl[cdev_id_qp].id, ipsec_ctx->tbl[cdev_id_qp].qp); - sa->crypto_session = rte_cryptodev_sym_session_create( - ipsec_ctx->session_pool); - rte_cryptodev_sym_session_init(ipsec_ctx->tbl[cdev_id_qp].id, - sa->crypto_session, sa->xforms, - ipsec_ctx->session_pool); - - rte_cryptodev_info_get(ipsec_ctx->tbl[cdev_id_qp].id, &cdev_info); - if (cdev_info.sym.max_nb_sessions_per_qp > 0) { - ret = rte_cryptodev_queue_pair_attach_sym_session( - ipsec_ctx->tbl[cdev_id_qp].id, - ipsec_ctx->tbl[cdev_id_qp].qp, - sa->crypto_session); - if (ret < 0) { - RTE_LOG(ERR, IPSEC, - "Session cannot be attached to qp %u ", - ipsec_ctx->tbl[cdev_id_qp].qp); - return -1; + if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) { + struct rte_security_session_conf sess_conf = { + .action_type = sa->type, + .protocol = RTE_SECURITY_PROTOCOL_IPSEC, + .ipsec = { + .spi = sa->spi, + .salt = sa->salt, + .options = { 0 }, + .direction = sa->direction, + .proto = RTE_SECURITY_IPSEC_SA_PROTO_ESP, + .mode = (sa->flags == IP4_TUNNEL || + sa->flags == IP6_TUNNEL) ? + RTE_SECURITY_IPSEC_SA_MODE_TUNNEL : + RTE_SECURITY_IPSEC_SA_MODE_TRANSPORT, + }, + .crypto_xform = sa->xforms + + }; + + if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) { + struct rte_security_ctx *ctx = (struct rte_security_ctx *) + rte_cryptodev_get_sec_ctx( + ipsec_ctx->tbl[cdev_id_qp].id); + + if (sess_conf.ipsec.mode == + RTE_SECURITY_IPSEC_SA_MODE_TUNNEL) { + struct rte_security_ipsec_tunnel_param *tunnel = + &sess_conf.ipsec.tunnel; + if (sa->flags == IP4_TUNNEL) { + tunnel->type = + RTE_SECURITY_IPSEC_TUNNEL_IPV4; + tunnel->ipv4.ttl = IPDEFTTL; + + memcpy((uint8_t *)&tunnel->ipv4.src_ip, + (uint8_t *)&sa->src.ip.ip4, 4); + + memcpy((uint8_t *)&tunnel->ipv4.dst_ip, + (uint8_t *)&sa->dst.ip.ip4, 4); + } + /* TODO support for Transport and IPV6 tunnel */ + } + + sa->sec_session = rte_security_session_create(ctx, + &sess_conf, ipsec_ctx->session_pool); + if (sa->sec_session == NULL) { + RTE_LOG(ERR, IPSEC, + "SEC Session init failed: err: %d\n", ret); + return -1; + } + } else if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO) { + struct rte_flow_error err; + struct rte_security_ctx *ctx = (struct rte_security_ctx *) + rte_eth_dev_get_sec_ctx( + sa->portid); + const struct rte_security_capability *sec_cap; + + sa->sec_session = rte_security_session_create(ctx, + &sess_conf, ipsec_ctx->session_pool); + if (sa->sec_session == NULL) { + RTE_LOG(ERR, IPSEC, + "SEC Session init failed: err: %d\n", ret); + return -1; + } + + sec_cap = rte_security_capabilities_get(ctx); + + /* iterate until ESP tunnel*/ + while (sec_cap->action != + RTE_SECURITY_ACTION_TYPE_NONE) { + + if (sec_cap->action == sa->type && + sec_cap->protocol == + RTE_SECURITY_PROTOCOL_IPSEC && + sec_cap->ipsec.mode == + RTE_SECURITY_IPSEC_SA_MODE_TUNNEL && + sec_cap->ipsec.direction == sa->direction) + break; + sec_cap++; + } + + if (sec_cap->action == RTE_SECURITY_ACTION_TYPE_NONE) { + RTE_LOG(ERR, IPSEC, + "No suitable security capability found\n"); + return -1; + } + + sa->ol_flags = sec_cap->ol_flags; + sa->security_ctx = ctx; + sa->pattern[0].type = RTE_FLOW_ITEM_TYPE_ETH; + + sa->pattern[1].type = RTE_FLOW_ITEM_TYPE_IPV4; + sa->pattern[1].mask = &rte_flow_item_ipv4_mask; + if (sa->flags & IP6_TUNNEL) { + sa->pattern[1].spec = &sa->ipv6_spec; + memcpy(sa->ipv6_spec.hdr.dst_addr, + sa->dst.ip.ip6.ip6_b, 16); + memcpy(sa->ipv6_spec.hdr.src_addr, + sa->src.ip.ip6.ip6_b, 16); + } else { + sa->pattern[1].spec = &sa->ipv4_spec; + sa->ipv4_spec.hdr.dst_addr = sa->dst.ip.ip4; + sa->ipv4_spec.hdr.src_addr = sa->src.ip.ip4; + } + + sa->pattern[2].type = RTE_FLOW_ITEM_TYPE_ESP; + sa->pattern[2].spec = &sa->esp_spec; + sa->pattern[2].mask = &rte_flow_item_esp_mask; + sa->esp_spec.hdr.spi = sa->spi; + + sa->pattern[3].type = RTE_FLOW_ITEM_TYPE_END; + + sa->action[0].type = RTE_FLOW_ACTION_TYPE_SECURITY; + sa->action[0].conf = sa->sec_session; + + sa->action[1].type = RTE_FLOW_ACTION_TYPE_END; + + sa->attr.egress = (sa->direction == + RTE_SECURITY_IPSEC_SA_DIR_EGRESS); + sa->flow = rte_flow_create(sa->portid, + &sa->attr, sa->pattern, sa->action, &err); + if (sa->flow == NULL) { + RTE_LOG(ERR, IPSEC, + "Failed to create ipsec flow msg: %s\n", + err.message); + return -1; + } + } + } else { + sa->crypto_session = rte_cryptodev_sym_session_create( + ipsec_ctx->session_pool); + rte_cryptodev_sym_session_init(ipsec_ctx->tbl[cdev_id_qp].id, + sa->crypto_session, sa->xforms, + ipsec_ctx->session_pool); + + rte_cryptodev_info_get(ipsec_ctx->tbl[cdev_id_qp].id, + &cdev_info); + if (cdev_info.sym.max_nb_sessions_per_qp > 0) { + ret = rte_cryptodev_queue_pair_attach_sym_session( + ipsec_ctx->tbl[cdev_id_qp].id, + ipsec_ctx->tbl[cdev_id_qp].qp, + sa->crypto_session); + if (ret < 0) { + RTE_LOG(ERR, IPSEC, + "Session cannot be attached to qp %u\n", + ipsec_ctx->tbl[cdev_id_qp].qp); + return -1; + } } } sa->cdev_id_qp = cdev_id_qp; @@ -125,7 +264,9 @@ ipsec_enqueue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx, { int32_t ret = 0, i; struct ipsec_mbuf_metadata *priv; + struct rte_crypto_sym_op *sym_cop; struct ipsec_sa *sa; + struct cdev_qp *cqp; for (i = 0; i < nb_pkts; i++) { if (unlikely(sas[i] == NULL)) { @@ -140,23 +281,76 @@ ipsec_enqueue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx, sa = sas[i]; priv->sa = sa; - priv->cop.type = RTE_CRYPTO_OP_TYPE_SYMMETRIC; - priv->cop.status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED; - - rte_prefetch0(&priv->sym_cop); - - if ((unlikely(sa->crypto_session == NULL)) && - create_session(ipsec_ctx, sa)) { - rte_pktmbuf_free(pkts[i]); - continue; - } - - rte_crypto_op_attach_sym_session(&priv->cop, - sa->crypto_session); - - ret = xform_func(pkts[i], sa, &priv->cop); - if (unlikely(ret)) { - rte_pktmbuf_free(pkts[i]); + switch (sa->type) { + case RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL: + priv->cop.type = RTE_CRYPTO_OP_TYPE_SYMMETRIC; + priv->cop.status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED; + + rte_prefetch0(&priv->sym_cop); + + if ((unlikely(sa->sec_session == NULL)) && + create_session(ipsec_ctx, sa)) { + rte_pktmbuf_free(pkts[i]); + continue; + } + + sym_cop = get_sym_cop(&priv->cop); + sym_cop->m_src = pkts[i]; + + rte_security_attach_session(&priv->cop, + sa->sec_session); + break; + case RTE_SECURITY_ACTION_TYPE_NONE: + + priv->cop.type = RTE_CRYPTO_OP_TYPE_SYMMETRIC; + priv->cop.status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED; + + rte_prefetch0(&priv->sym_cop); + + if ((unlikely(sa->crypto_session == NULL)) && + create_session(ipsec_ctx, sa)) { + rte_pktmbuf_free(pkts[i]); + continue; + } + + rte_crypto_op_attach_sym_session(&priv->cop, + sa->crypto_session); + + ret = xform_func(pkts[i], sa, &priv->cop); + if (unlikely(ret)) { + rte_pktmbuf_free(pkts[i]); + continue; + } + break; + case RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL: + break; + case RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO: + priv->cop.type = RTE_CRYPTO_OP_TYPE_SYMMETRIC; + priv->cop.status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED; + + rte_prefetch0(&priv->sym_cop); + + if ((unlikely(sa->sec_session == NULL)) && + create_session(ipsec_ctx, sa)) { + rte_pktmbuf_free(pkts[i]); + continue; + } + + rte_security_attach_session(&priv->cop, + sa->sec_session); + + ret = xform_func(pkts[i], sa, &priv->cop); + if (unlikely(ret)) { + rte_pktmbuf_free(pkts[i]); + continue; + } + + cqp = &ipsec_ctx->tbl[sa->cdev_id_qp]; + cqp->ol_pkts[cqp->ol_pkts_cnt++] = pkts[i]; + if (sa->ol_flags & RTE_SECURITY_TX_OLOAD_NEED_MDATA) + rte_security_set_pkt_metadata( + sa->security_ctx, + sa->sec_session, pkts[i], NULL); continue; } @@ -167,7 +361,7 @@ ipsec_enqueue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx, static inline int ipsec_dequeue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx, - struct rte_mbuf *pkts[], uint16_t max_pkts) + struct rte_mbuf *pkts[], uint16_t max_pkts) { int32_t nb_pkts = 0, ret = 0, i, j, nb_cops; struct ipsec_mbuf_metadata *priv; @@ -182,6 +376,19 @@ ipsec_dequeue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx, if (ipsec_ctx->last_qp == ipsec_ctx->nb_qps) ipsec_ctx->last_qp %= ipsec_ctx->nb_qps; + while (cqp->ol_pkts_cnt > 0 && nb_pkts < max_pkts) { + pkt = cqp->ol_pkts[--cqp->ol_pkts_cnt]; + rte_prefetch0(pkt); + priv = get_priv(pkt); + sa = priv->sa; + ret = xform_func(pkt, sa, &priv->cop); + if (unlikely(ret)) { + rte_pktmbuf_free(pkt); + continue; + } + pkts[nb_pkts++] = pkt; + } + if (cqp->in_flight == 0) continue; @@ -199,11 +406,14 @@ ipsec_dequeue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx, RTE_ASSERT(sa != NULL); - ret = xform_func(pkt, sa, cops[j]); - if (unlikely(ret)) - rte_pktmbuf_free(pkt); - else - pkts[nb_pkts++] = pkt; + if (sa->type == RTE_SECURITY_ACTION_TYPE_NONE) { + ret = xform_func(pkt, sa, cops[j]); + if (unlikely(ret)) { + rte_pktmbuf_free(pkt); + continue; + } + } + pkts[nb_pkts++] = pkt; } } diff --git a/examples/ipsec-secgw/ipsec.h b/examples/ipsec-secgw/ipsec.h index da1fb1b2..775b316f 100644 --- a/examples/ipsec-secgw/ipsec.h +++ b/examples/ipsec-secgw/ipsec.h @@ -38,6 +38,8 @@ #include <rte_byteorder.h> #include <rte_crypto.h> +#include <rte_security.h> +#include <rte_flow.h> #define RTE_LOGTYPE_IPSEC RTE_LOGTYPE_USER1 #define RTE_LOGTYPE_IPSEC_ESP RTE_LOGTYPE_USER2 @@ -99,7 +101,10 @@ struct ipsec_sa { uint32_t cdev_id_qp; uint64_t seq; uint32_t salt; - struct rte_cryptodev_sym_session *crypto_session; + union { + struct rte_cryptodev_sym_session *crypto_session; + struct rte_security_session *sec_session; + }; enum rte_crypto_cipher_algorithm cipher_algo; enum rte_crypto_auth_algorithm auth_algo; enum rte_crypto_aead_algorithm aead_algo; @@ -117,7 +122,28 @@ struct ipsec_sa { uint8_t auth_key[MAX_KEY_SIZE]; uint16_t auth_key_len; uint16_t aad_len; - struct rte_crypto_sym_xform *xforms; + union { + struct rte_crypto_sym_xform *xforms; + struct rte_security_ipsec_xform *sec_xform; + }; + enum rte_security_session_action_type type; + enum rte_security_ipsec_sa_direction direction; + uint16_t portid; + struct rte_security_ctx *security_ctx; + uint32_t ol_flags; + +#define MAX_RTE_FLOW_PATTERN (4) +#define MAX_RTE_FLOW_ACTIONS (2) + struct rte_flow_item pattern[MAX_RTE_FLOW_PATTERN]; + struct rte_flow_action action[MAX_RTE_FLOW_ACTIONS]; + struct rte_flow_attr attr; + union { + struct rte_flow_item_ipv4 ipv4_spec; + struct rte_flow_item_ipv6 ipv6_spec; + }; + struct rte_flow_item_esp esp_spec; + struct rte_flow *flow; + struct rte_security_session_conf sess_conf; } __rte_cache_aligned; struct ipsec_mbuf_metadata { @@ -133,6 +159,8 @@ struct cdev_qp { uint16_t in_flight; uint16_t len; struct rte_crypto_op *buf[MAX_PKT_BURST] __rte_aligned(sizeof(void *)); + struct rte_mbuf *ol_pkts[MAX_PKT_BURST] __rte_aligned(sizeof(void *)); + uint16_t ol_pkts_cnt; }; struct ipsec_ctx { @@ -150,6 +178,7 @@ struct cdev_key { uint16_t lcore_id; uint8_t cipher_algo; uint8_t auth_algo; + uint8_t aead_algo; }; struct socket_ctx { diff --git a/examples/ipsec-secgw/sa.c b/examples/ipsec-secgw/sa.c index 7be0e628..4c448e5c 100644 --- a/examples/ipsec-secgw/sa.c +++ b/examples/ipsec-secgw/sa.c @@ -41,16 +41,20 @@ #include <rte_memzone.h> #include <rte_crypto.h> +#include <rte_security.h> #include <rte_cryptodev.h> #include <rte_byteorder.h> #include <rte_errno.h> #include <rte_ip.h> #include <rte_random.h> +#include <rte_ethdev.h> #include "ipsec.h" #include "esp.h" #include "parser.h" +#define IPDEFTTL 64 + struct supported_cipher_algo { const char *keyword; enum rte_crypto_cipher_algorithm algo; @@ -238,6 +242,8 @@ parse_sa_tokens(char **tokens, uint32_t n_tokens, uint32_t src_p = 0; uint32_t dst_p = 0; uint32_t mode_p = 0; + uint32_t type_p = 0; + uint32_t portid_p = 0; if (strcmp(tokens[0], "in") == 0) { ri = &nb_sa_in; @@ -375,7 +381,6 @@ parse_sa_tokens(char **tokens, uint32_t n_tokens, rule->auth_algo = algo->algo; rule->auth_key_len = algo->key_len; rule->digest_len = algo->digest_len; - rule->aad_len = algo->key_len; /* NULL algorithm and combined algos do not * require auth key @@ -431,7 +436,7 @@ parse_sa_tokens(char **tokens, uint32_t n_tokens, rule->aead_algo = algo->algo; rule->cipher_key_len = algo->key_len; rule->digest_len = algo->digest_len; - rule->aad_len = algo->key_len; + rule->aad_len = algo->aad_len; rule->block_size = algo->block_size; rule->iv_len = algo->iv_len; @@ -550,6 +555,52 @@ parse_sa_tokens(char **tokens, uint32_t n_tokens, continue; } + if (strcmp(tokens[ti], "type") == 0) { + APP_CHECK_PRESENCE(type_p, tokens[ti], status); + if (status->status < 0) + return; + + INCREMENT_TOKEN_INDEX(ti, n_tokens, status); + if (status->status < 0) + return; + + if (strcmp(tokens[ti], "inline-crypto-offload") == 0) + rule->type = + RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO; + else if (strcmp(tokens[ti], + "inline-protocol-offload") == 0) + rule->type = + RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL; + else if (strcmp(tokens[ti], + "lookaside-protocol-offload") == 0) + rule->type = + RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL; + else if (strcmp(tokens[ti], "no-offload") == 0) + rule->type = RTE_SECURITY_ACTION_TYPE_NONE; + else { + APP_CHECK(0, status, "Invalid input \"%s\"", + tokens[ti]); + return; + } + + type_p = 1; + continue; + } + + if (strcmp(tokens[ti], "port_id") == 0) { + APP_CHECK_PRESENCE(portid_p, tokens[ti], status); + if (status->status < 0) + return; + INCREMENT_TOKEN_INDEX(ti, n_tokens, status); + if (status->status < 0) + return; + rule->portid = atoi(tokens[ti]); + if (status->status < 0) + return; + portid_p = 1; + continue; + } + /* unrecognizeable input */ APP_CHECK(0, status, "unrecognized input \"%s\"", tokens[ti]); @@ -580,6 +631,14 @@ parse_sa_tokens(char **tokens, uint32_t n_tokens, if (status->status < 0) return; + if ((rule->type != RTE_SECURITY_ACTION_TYPE_NONE) && (portid_p == 0)) + printf("Missing portid option, falling back to non-offload\n"); + + if (!type_p || !portid_p) { + rule->type = RTE_SECURITY_ACTION_TYPE_NONE; + rule->portid = -1; + } + *ri = *ri + 1; } @@ -647,9 +706,11 @@ print_one_sa_rule(const struct ipsec_sa *sa, int inbound) struct sa_ctx { struct ipsec_sa sa[IPSEC_SA_MAX_ENTRIES]; - struct { - struct rte_crypto_sym_xform a; - struct rte_crypto_sym_xform b; + union { + struct { + struct rte_crypto_sym_xform a; + struct rte_crypto_sym_xform b; + }; } xf[IPSEC_SA_MAX_ENTRIES]; }; @@ -682,6 +743,33 @@ sa_create(const char *name, int32_t socket_id) } static int +check_eth_dev_caps(uint16_t portid, uint32_t inbound) +{ + struct rte_eth_dev_info dev_info; + + rte_eth_dev_info_get(portid, &dev_info); + + if (inbound) { + if ((dev_info.rx_offload_capa & + DEV_RX_OFFLOAD_SECURITY) == 0) { + RTE_LOG(WARNING, PORT, + "hardware RX IPSec offload is not supported\n"); + return -EINVAL; + } + + } else { /* outbound */ + if ((dev_info.tx_offload_capa & + DEV_TX_OFFLOAD_SECURITY) == 0) { + RTE_LOG(WARNING, PORT, + "hardware TX IPSec offload is not supported\n"); + return -EINVAL; + } + } + return 0; +} + + +static int sa_add_rules(struct sa_ctx *sa_ctx, const struct ipsec_sa entries[], uint32_t nb_entries, uint32_t inbound) { @@ -700,6 +788,16 @@ sa_add_rules(struct sa_ctx *sa_ctx, const struct ipsec_sa entries[], *sa = entries[i]; sa->seq = 0; + if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL || + sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO) { + if (check_eth_dev_caps(sa->portid, inbound)) + return -EINVAL; + } + + sa->direction = (inbound == 1) ? + RTE_SECURITY_IPSEC_SA_DIR_INGRESS : + RTE_SECURITY_IPSEC_SA_DIR_EGRESS; + switch (sa->flags) { case IP4_TUNNEL: sa->src.ip.ip4 = rte_cpu_to_be_32(sa->src.ip.ip4); @@ -709,37 +807,21 @@ sa_add_rules(struct sa_ctx *sa_ctx, const struct ipsec_sa entries[], if (sa->aead_algo == RTE_CRYPTO_AEAD_AES_GCM) { iv_length = 16; - if (inbound) { - sa_ctx->xf[idx].a.type = RTE_CRYPTO_SYM_XFORM_AEAD; - sa_ctx->xf[idx].a.aead.algo = sa->aead_algo; - sa_ctx->xf[idx].a.aead.key.data = sa->cipher_key; - sa_ctx->xf[idx].a.aead.key.length = - sa->cipher_key_len; - sa_ctx->xf[idx].a.aead.op = - RTE_CRYPTO_AEAD_OP_DECRYPT; - sa_ctx->xf[idx].a.next = NULL; - sa_ctx->xf[idx].a.aead.iv.offset = IV_OFFSET; - sa_ctx->xf[idx].a.aead.iv.length = iv_length; - sa_ctx->xf[idx].a.aead.aad_length = - sa->aad_len; - sa_ctx->xf[idx].a.aead.digest_length = - sa->digest_len; - } else { /* outbound */ - sa_ctx->xf[idx].a.type = RTE_CRYPTO_SYM_XFORM_AEAD; - sa_ctx->xf[idx].a.aead.algo = sa->aead_algo; - sa_ctx->xf[idx].a.aead.key.data = sa->cipher_key; - sa_ctx->xf[idx].a.aead.key.length = - sa->cipher_key_len; - sa_ctx->xf[idx].a.aead.op = - RTE_CRYPTO_AEAD_OP_ENCRYPT; - sa_ctx->xf[idx].a.next = NULL; - sa_ctx->xf[idx].a.aead.iv.offset = IV_OFFSET; - sa_ctx->xf[idx].a.aead.iv.length = iv_length; - sa_ctx->xf[idx].a.aead.aad_length = - sa->aad_len; - sa_ctx->xf[idx].a.aead.digest_length = - sa->digest_len; - } + sa_ctx->xf[idx].a.type = RTE_CRYPTO_SYM_XFORM_AEAD; + sa_ctx->xf[idx].a.aead.algo = sa->aead_algo; + sa_ctx->xf[idx].a.aead.key.data = sa->cipher_key; + sa_ctx->xf[idx].a.aead.key.length = + sa->cipher_key_len; + sa_ctx->xf[idx].a.aead.op = (inbound == 1) ? + RTE_CRYPTO_AEAD_OP_DECRYPT : + RTE_CRYPTO_AEAD_OP_ENCRYPT; + sa_ctx->xf[idx].a.next = NULL; + sa_ctx->xf[idx].a.aead.iv.offset = IV_OFFSET; + sa_ctx->xf[idx].a.aead.iv.length = iv_length; + sa_ctx->xf[idx].a.aead.aad_length = + sa->aad_len; + sa_ctx->xf[idx].a.aead.digest_length = + sa->digest_len; sa->xforms = &sa_ctx->xf[idx].a; diff --git a/examples/ipv4_multicast/main.c b/examples/ipv4_multicast/main.c index 9a13d353..83ac0d80 100644 --- a/examples/ipv4_multicast/main.c +++ b/examples/ipv4_multicast/main.c @@ -47,7 +47,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -57,7 +56,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -116,7 +114,7 @@ static struct ether_addr ports_eth_addr[MAX_PORTS]; /* mask of enabled ports */ static uint32_t enabled_port_mask = 0; -static uint8_t nb_ports = 0; +static uint16_t nb_ports; static int rx_queue_per_lcore = 1; @@ -195,7 +193,7 @@ static struct mcast_group_params mcast_group_table[] = { /* Send burst of packets on an output interface */ static void -send_burst(struct lcore_queue_conf *qconf, uint8_t port) +send_burst(struct lcore_queue_conf *qconf, uint16_t port) { struct rte_mbuf **m_table; uint16_t n, queueid; @@ -312,7 +310,7 @@ mcast_out_pkt(struct rte_mbuf *pkt, int use_clone) */ static inline void mcast_send_pkt(struct rte_mbuf *pkt, struct ether_addr *dest_addr, - struct lcore_queue_conf *qconf, uint8_t port) + struct lcore_queue_conf *qconf, uint16_t port) { struct ether_hdr *ethdr; uint16_t len; @@ -343,7 +341,7 @@ mcast_forward(struct rte_mbuf *m, struct lcore_queue_conf *qconf) struct ipv4_hdr *iphdr; uint32_t dest_addr, port_mask, port_num, use_clone; int32_t hash; - uint8_t port; + uint16_t port; union { uint64_t as_int; struct ether_addr as_addr; @@ -407,7 +405,7 @@ static inline void send_timeout_burst(struct lcore_queue_conf *qconf) { uint64_t cur_tsc; - uint8_t portid; + uint16_t portid; const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) / US_PER_S * BURST_TX_DRAIN_US; cur_tsc = rte_rdtsc(); @@ -428,7 +426,7 @@ main_loop(__rte_unused void *dummy) struct rte_mbuf *pkts_burst[MAX_PKT_BURST]; unsigned lcore_id; int i, j, nb_rx; - uint8_t portid; + uint16_t portid; struct lcore_queue_conf *qconf; lcore_id = rte_lcore_id(); @@ -448,7 +446,7 @@ main_loop(__rte_unused void *dummy) portid = qconf->rx_queue_list[i]; RTE_LOG(INFO, IPv4_MULTICAST, " -- lcoreid=%u portid=%d\n", - lcore_id, (int) portid); + lcore_id, portid); } while (1) { @@ -610,11 +608,12 @@ init_mcast_hash(void) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -629,14 +628,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up. Speed %u Mbps - %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -673,7 +671,7 @@ main(int argc, char **argv) uint16_t queueid; unsigned lcore_id = 0, rx_lcore_id = 0; uint32_t n_tx_queue, nb_lcores; - uint8_t portid; + uint16_t portid; /* init EAL */ ret = rte_eal_init(argc, argv); diff --git a/examples/kni/main.c b/examples/kni/main.c index e3bc2fb7..3f173854 100644 --- a/examples/kni/main.c +++ b/examples/kni/main.c @@ -53,7 +53,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_per_lcore.h> #include <rte_launch.h> @@ -61,7 +60,7 @@ #include <rte_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> +#include <rte_bus_pci.h> #include <rte_debug.h> #include <rte_ether.h> #include <rte_ethdev.h> @@ -110,7 +109,7 @@ * Structure of port parameters */ struct kni_port_params { - uint8_t port_id;/* Port ID */ + uint16_t port_id;/* Port ID */ unsigned lcore_rx; /* lcore ID for RX */ unsigned lcore_tx; /* lcore ID for TX */ uint32_t nb_lcore_k; /* Number of lcores for KNI multi kernel threads */ @@ -162,8 +161,8 @@ struct kni_interface_stats { /* kni device statistics array */ static struct kni_interface_stats kni_stats[RTE_MAX_ETHPORTS]; -static int kni_change_mtu(uint8_t port_id, unsigned new_mtu); -static int kni_config_network_interface(uint8_t port_id, uint8_t if_up); +static int kni_change_mtu(uint16_t port_id, unsigned int new_mtu); +static int kni_config_network_interface(uint16_t port_id, uint8_t if_up); static rte_atomic32_t kni_stop = RTE_ATOMIC32_INIT(0); @@ -171,7 +170,7 @@ static rte_atomic32_t kni_stop = RTE_ATOMIC32_INIT(0); static void print_stats(void) { - uint8_t i; + uint16_t i; printf("\n**KNI example application statistics**\n" "====== ============== ============ ============ ============ ============\n" @@ -238,7 +237,8 @@ kni_burst_free_mbufs(struct rte_mbuf **pkts, unsigned num) static void kni_ingress(struct kni_port_params *p) { - uint8_t i, port_id; + uint8_t i; + uint16_t port_id; unsigned nb_rx, num; uint32_t nb_kni; struct rte_mbuf *pkts_burst[PKT_BURST_SZ]; @@ -274,7 +274,8 @@ kni_ingress(struct kni_port_params *p) static void kni_egress(struct kni_port_params *p) { - uint8_t i, port_id; + uint8_t i; + uint16_t port_id; unsigned nb_tx, num; uint32_t nb_kni; struct rte_mbuf *pkts_burst[PKT_BURST_SZ]; @@ -416,7 +417,7 @@ parse_config(const char *arg) int i, j, nb_token; char *str_fld[_NUM_FLD]; unsigned long int_fld[_NUM_FLD]; - uint8_t port_id, nb_kni_port_params = 0; + uint16_t port_id, nb_kni_port_params = 0; memset(&kni_port_params_array, 0, sizeof(kni_port_params_array)); while (((p = strchr(p0, '(')) != NULL) && @@ -445,7 +446,7 @@ parse_config(const char *arg) } i = 0; - port_id = (uint8_t)int_fld[i++]; + port_id = int_fld[i++]; if (port_id >= RTE_MAX_ETHPORTS) { printf("Port ID %d could not exceed the maximum %d\n", port_id, RTE_MAX_ETHPORTS); @@ -601,7 +602,7 @@ init_kni(void) /* Initialise a single port on an Ethernet device */ static void -init_port(uint8_t port) +init_port(uint16_t port) { int ret; uint16_t nb_rxd = NB_RXD; @@ -643,11 +644,12 @@ init_port(uint8_t port) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status\n"); @@ -662,14 +664,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up - speed %uMbps - %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -698,7 +699,7 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* Callback for request of changing MTU */ static int -kni_change_mtu(uint8_t port_id, unsigned new_mtu) +kni_change_mtu(uint16_t port_id, unsigned int new_mtu) { int ret; struct rte_eth_conf conf; @@ -741,7 +742,7 @@ kni_change_mtu(uint8_t port_id, unsigned new_mtu) /* Callback for request of configuring network interface up/down */ static int -kni_config_network_interface(uint8_t port_id, uint8_t if_up) +kni_config_network_interface(uint16_t port_id, uint8_t if_up) { int ret = 0; @@ -766,7 +767,7 @@ kni_config_network_interface(uint8_t port_id, uint8_t if_up) } static int -kni_alloc(uint8_t port_id) +kni_alloc(uint16_t port_id) { uint8_t i; struct rte_kni *kni; @@ -790,7 +791,7 @@ kni_alloc(uint8_t port_id) } else snprintf(conf.name, RTE_KNI_NAMESIZE, "vEth%u", port_id); - conf.group_id = (uint16_t)port_id; + conf.group_id = port_id; conf.mbuf_size = MAX_PACKET_SZ; /* * The first KNI device associated to a port @@ -803,8 +804,11 @@ kni_alloc(uint8_t port_id) memset(&dev_info, 0, sizeof(dev_info)); rte_eth_dev_info_get(port_id, &dev_info); - conf.addr = dev_info.pci_dev->addr; - conf.id = dev_info.pci_dev->id; + + if (dev_info.pci_dev) { + conf.addr = dev_info.pci_dev->addr; + conf.id = dev_info.pci_dev->id; + } memset(&ops, 0, sizeof(ops)); ops.port_id = port_id; @@ -825,7 +829,7 @@ kni_alloc(uint8_t port_id) } static int -kni_free_kni(uint8_t port_id) +kni_free_kni(uint16_t port_id) { uint8_t i; struct kni_port_params **p = kni_port_params_array; @@ -848,7 +852,7 @@ int main(int argc, char** argv) { int ret; - uint8_t nb_sys_ports, port; + uint16_t nb_sys_ports, port; unsigned i; /* Associate signal_hanlder function with USR signals */ @@ -919,9 +923,6 @@ main(int argc, char** argv) continue; kni_free_kni(port); } -#ifdef RTE_LIBRTE_XEN_DOM0 - rte_kni_close(); -#endif for (i = 0; i < RTE_MAX_ETHPORTS; i++) if (kni_port_params_array[i]) { rte_free(kni_port_params_array[i]); diff --git a/examples/l2fwd-cat/Makefile b/examples/l2fwd-cat/Makefile index ae921ade..a7fe6d68 100644 --- a/examples/l2fwd-cat/Makefile +++ b/examples/l2fwd-cat/Makefile @@ -40,9 +40,6 @@ endif # Default target, can be overridden by command line or environment RTE_TARGET ?= x86_64-native-linuxapp-gcc -# Location of PQoS library and includes, -PQOS_LIBRARY_PATH = $(PQOS_INSTALL_PATH)/libpqos.a - include $(RTE_SDK)/mk/rte.vars.mk # binary name @@ -65,6 +62,6 @@ CFLAGS += -I$(PQOS_INSTALL_PATH)/../include CFLAGS_cat.o := -D_GNU_SOURCE LDLIBS += -L$(PQOS_INSTALL_PATH) -LDLIBS += $(PQOS_LIBRARY_PATH) +LDLIBS += -lpqos include $(RTE_SDK)/mk/rte.extapp.mk diff --git a/examples/l2fwd-cat/cat.c b/examples/l2fwd-cat/cat.c index 6133bf5b..689dcb11 100644 --- a/examples/l2fwd-cat/cat.c +++ b/examples/l2fwd-cat/cat.c @@ -53,7 +53,11 @@ static const struct pqos_cap *m_cap; static const struct pqos_cpuinfo *m_cpu; static const struct pqos_capability *m_cap_l3ca; +#if PQOS_VERSION <= 103 static unsigned m_sockets[PQOS_MAX_SOCKETS]; +#else +static unsigned int *m_sockets; +#endif static unsigned m_sock_count; static struct cat_config m_config[PQOS_MAX_CORES]; static unsigned m_config_count; @@ -271,16 +275,16 @@ parse_l3ca(const char *l3ca) /* scan the separator '@', ','(next) or '\0'(finish) */ l3ca += strcspn(l3ca, "@,"); - if (*l3ca == '@') { - /* explicit assign cpu_set */ - offset = parse_set(l3ca + 1, &cpuset); - if (offset < 0 || CPU_COUNT(&cpuset) == 0) - goto err; + if (*l3ca != '@') + goto err; - end = l3ca + 1 + offset; - } else + /* explicit assign cpu_set */ + offset = parse_set(l3ca + 1, &cpuset); + if (offset < 0 || CPU_COUNT(&cpuset) == 0) goto err; + end = l3ca + 1 + offset; + if (*end != ',' && *end != '\0') goto err; @@ -353,9 +357,6 @@ parse_l3ca(const char *l3ca) idx++; } while (*end != '\0' && idx < PQOS_MAX_CORES); - if (m_config_count == 0) - goto err; - return 0; err: @@ -408,7 +409,11 @@ check_cpus(void) goto exit; } +#if PQOS_VERSION <= 103 ret = pqos_l3ca_assoc_get(cpu_id, &cos_id); +#else + ret = pqos_alloc_assoc_get(cpu_id, &cos_id); +#endif if (ret != PQOS_RETVAL_OK) { printf("PQOS: Failed to read COS " "associated to cpu %u.\n", @@ -512,7 +517,11 @@ check_and_select_classes(unsigned cos_id_map[][PQOS_MAX_SOCKETS]) for (j = 0; j < m_cpu->num_cores; j++) { cpu_id = m_cpu->cores[j].lcore; +#if PQOS_VERSION <= 103 ret = pqos_l3ca_assoc_get(cpu_id, &cos_id); +#else + ret = pqos_alloc_assoc_get(cpu_id, &cos_id); +#endif if (ret != PQOS_RETVAL_OK) { printf("PQOS: Failed to read COS associated to " "cpu %u on phy_pkg %u.\n", cpu_id, phy_pkg_id); @@ -598,10 +607,19 @@ configure_cat(unsigned cos_id_map[][PQOS_MAX_SOCKETS]) l3ca.cdp = m_config[i].cdp; if (m_config[i].cdp == 1) { +#if PQOS_VERSION <= 103 l3ca.code_mask = m_config[i].code_mask; l3ca.data_mask = m_config[i].data_mask; +#else + l3ca.u.s.code_mask = m_config[i].code_mask; + l3ca.u.s.data_mask = m_config[i].data_mask; +#endif } else +#if PQOS_VERSION <= 103 l3ca.ways_mask = m_config[i].mask; +#else + l3ca.u.ways_mask = m_config[i].mask; +#endif for (j = 0; j < m_sock_count; j++) { phy_pkg_id = m_sockets[j]; @@ -637,7 +655,11 @@ configure_cat(unsigned cos_id_map[][PQOS_MAX_SOCKETS]) cos_id = cos_id_map[i][phy_pkg_id]; +#if PQOS_VERSION <= 103 ret = pqos_l3ca_assoc_set(cpu_id, cos_id); +#else + ret = pqos_alloc_assoc_set(cpu_id, cos_id); +#endif if (ret != PQOS_RETVAL_OK) { printf("PQOS: Failed to associate COS %u to " "cpu %u\n", cos_id, cpu_id); @@ -754,24 +776,43 @@ print_cat_config(void) if (tab[n].cdp == 1) { printf("PQOS: COS: %u, cMASK: 0x%llx, " "dMASK: 0x%llx\n", tab[n].class_id, +#if PQOS_VERSION <= 103 (unsigned long long)tab[n].code_mask, (unsigned long long)tab[n].data_mask); +#else + (unsigned long long)tab[n].u.s.code_mask, + (unsigned long long)tab[n].u.s.data_mask); +#endif } else { printf("PQOS: COS: %u, MASK: 0x%llx\n", tab[n].class_id, +#if PQOS_VERSION <= 103 (unsigned long long)tab[n].ways_mask); +#else + (unsigned long long)tab[n].u.ways_mask); +#endif } } } for (i = 0; i < m_sock_count; i++) { +#if PQOS_VERSION <= 103 unsigned lcores[PQOS_MAX_SOCKET_CORES] = {0}; +#else + unsigned int *lcores = NULL; +#endif unsigned lcount = 0; unsigned n = 0; +#if PQOS_VERSION <= 103 ret = pqos_cpu_get_cores(m_cpu, m_sockets[i], PQOS_MAX_SOCKET_CORES, &lcount, &lcores[0]); if (ret != PQOS_RETVAL_OK) { +#else + lcores = pqos_cpu_get_cores(m_cpu, m_sockets[i], + &lcount); + if (lcores == NULL || lcount == 0) { +#endif printf("PQOS: Error retrieving core information!\n"); return; } @@ -780,13 +821,21 @@ print_cat_config(void) for (n = 0; n < lcount; n++) { unsigned class_id = 0; +#if PQOS_VERSION <= 103 ret = pqos_l3ca_assoc_get(lcores[n], &class_id); +#else + ret = pqos_alloc_assoc_get(lcores[n], &class_id); +#endif if (ret == PQOS_RETVAL_OK) printf("PQOS: CPU: %u, COS: %u\n", lcores[n], class_id); else printf("PQOS: CPU: %u, ERROR\n", lcores[n]); } + +#if PQOS_VERSION > 103 + free(lcores); +#endif } } @@ -849,7 +898,12 @@ cat_fini(void) m_cap = NULL; m_cpu = NULL; m_cap_l3ca = NULL; +#if PQOS_VERSION <= 103 memset(m_sockets, 0, sizeof(m_sockets)); +#else + if (m_sockets != NULL) + free(m_sockets); +#endif m_sock_count = 0; memset(m_config, 0, sizeof(m_config)); m_config_count = 0; @@ -875,7 +929,11 @@ cat_exit(void) if (CPU_ISSET(cpu_id, &m_config[i].cpumask) == 0) continue; +#if PQOS_VERSION <= 103 ret = pqos_l3ca_assoc_set(cpu_id, 0); +#else + ret = pqos_alloc_assoc_set(cpu_id, 0); +#endif if (ret != PQOS_RETVAL_OK) { printf("PQOS: Failed to associate COS 0 to " "cpu %u\n", cpu_id); @@ -927,7 +985,9 @@ cat_init(int argc, char **argv) /* PQoS Initialization - Check and initialize CAT capability */ cfg.fd_log = STDOUT_FILENO; cfg.verbose = 0; +#if PQOS_VERSION <= 103 cfg.cdp_cfg = PQOS_REQUIRE_CDP_ANY; +#endif ret = pqos_init(&cfg); if (ret != PQOS_RETVAL_OK) { printf("PQOS: Error initializing PQoS library!\n"); @@ -953,9 +1013,14 @@ cat_init(int argc, char **argv) } /* Get CPU socket information */ +#if PQOS_VERSION <= 103 ret = pqos_cpu_get_sockets(m_cpu, PQOS_MAX_SOCKETS, &m_sock_count, m_sockets); if (ret != PQOS_RETVAL_OK) { +#else + m_sockets = pqos_cpu_get_sockets(m_cpu, &m_sock_count); + if (m_sockets == NULL) { +#endif printf("PQOS: Error retrieving CPU socket information!\n"); ret = -EFAULT; goto err; diff --git a/examples/l2fwd-cat/l2fwd-cat.c b/examples/l2fwd-cat/l2fwd-cat.c index c293bd9c..b10ac896 100644 --- a/examples/l2fwd-cat/l2fwd-cat.c +++ b/examples/l2fwd-cat/l2fwd-cat.c @@ -59,7 +59,7 @@ static const struct rte_eth_conf port_conf_default = { * coming from the mbuf_pool passed as a parameter. */ static inline int -port_init(uint8_t port, struct rte_mempool *mbuf_pool) +port_init(uint16_t port, struct rte_mempool *mbuf_pool) { struct rte_eth_conf port_conf = port_conf_default; const uint16_t rx_rings = 1, tx_rings = 1; @@ -106,7 +106,7 @@ port_init(uint8_t port, struct rte_mempool *mbuf_pool) rte_eth_macaddr_get(port, &addr); printf("Port %u MAC: %02" PRIx8 " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 "\n", - (unsigned)port, + port, addr.addr_bytes[0], addr.addr_bytes[1], addr.addr_bytes[2], addr.addr_bytes[3], addr.addr_bytes[4], addr.addr_bytes[5]); @@ -124,8 +124,8 @@ port_init(uint8_t port, struct rte_mempool *mbuf_pool) static __attribute__((noreturn)) void lcore_main(void) { - const uint8_t nb_ports = rte_eth_dev_count(); - uint8_t port; + const uint16_t nb_ports = rte_eth_dev_count(); + uint16_t port; /* * Check that the port is on the same NUMA node as the polling thread @@ -181,7 +181,7 @@ main(int argc, char *argv[]) { struct rte_mempool *mbuf_pool; unsigned nb_ports; - uint8_t portid; + uint16_t portid; /* Initialize the Environment Abstraction Layer (EAL). */ int ret = rte_eal_init(argc, argv); @@ -217,7 +217,7 @@ main(int argc, char *argv[]) /* Initialize all ports. */ for (portid = 0; portid < nb_ports; portid++) if (port_init(portid, mbuf_pool) != 0) - rte_exit(EXIT_FAILURE, "Cannot init port %"PRIu8 "\n", + rte_exit(EXIT_FAILURE, "Cannot init port %"PRIu16 "\n", portid); if (rte_lcore_count() > 1) diff --git a/examples/l2fwd-crypto/main.c b/examples/l2fwd-crypto/main.c index f020be32..d4e1682c 100644 --- a/examples/l2fwd-crypto/main.c +++ b/examples/l2fwd-crypto/main.c @@ -67,8 +67,6 @@ #include <rte_memcpy.h> #include <rte_memory.h> #include <rte_mempool.h> -#include <rte_memzone.h> -#include <rte_pci.h> #include <rte_per_lcore.h> #include <rte_prefetch.h> #include <rte_random.h> @@ -86,6 +84,8 @@ enum cdev_type { #define MAX_STR_LEN 32 #define MAX_KEY_SIZE 128 +#define MAX_IV_SIZE 16 +#define MAX_AAD_SIZE 65535 #define MAX_PKT_BURST 32 #define BURST_TX_DRAIN_US 100 /* TX drain every ~100us */ #define MAX_SESSIONS 32 @@ -112,7 +112,7 @@ static uint64_t l2fwd_enabled_port_mask; static uint64_t l2fwd_enabled_crypto_mask; /* list of enabled ports */ -static uint32_t l2fwd_dst_ports[RTE_MAX_ETHPORTS]; +static uint16_t l2fwd_dst_ports[RTE_MAX_ETHPORTS]; struct pkt_buffer { @@ -139,7 +139,7 @@ enum l2fwd_crypto_xform_chain { struct l2fwd_key { uint8_t *data; uint32_t length; - phys_addr_t phys_addr; + rte_iova_t phys_addr; }; struct l2fwd_iv { @@ -224,7 +224,7 @@ struct l2fwd_crypto_params { /** lcore configuration */ struct lcore_queue_conf { unsigned nb_rx_ports; - unsigned rx_port_list[MAX_RX_QUEUE_PER_LCORE]; + uint16_t rx_port_list[MAX_RX_QUEUE_PER_LCORE]; unsigned nb_crypto_devs; unsigned cryptodev_list[MAX_RX_QUEUE_PER_LCORE]; @@ -290,7 +290,7 @@ print_stats(void) uint64_t total_packets_dropped, total_packets_tx, total_packets_rx; uint64_t total_packets_enqueued, total_packets_dequeued, total_packets_errors; - unsigned portid; + uint16_t portid; uint64_t cdevid; total_packets_dropped = 0; @@ -496,7 +496,7 @@ l2fwd_simple_crypto_enqueue(struct rte_mbuf *m, uint8_t *) + ipdata_offset + data_len; } - op->sym->auth.digest.phys_addr = rte_pktmbuf_mtophys_offset(m, + op->sym->auth.digest.phys_addr = rte_pktmbuf_iova_offset(m, rte_pktmbuf_pkt_len(m) - cparams->digest_length); /* For wireless algorithms, offset/length must be in bits */ @@ -534,7 +534,16 @@ l2fwd_simple_crypto_enqueue(struct rte_mbuf *m, uint8_t *iv_ptr = rte_crypto_op_ctod_offset(op, uint8_t *, IV_OFFSET); /* Copy IV at the end of the crypto operation */ - rte_memcpy(iv_ptr, cparams->aead_iv.data, cparams->aead_iv.length); + /* + * If doing AES-CCM, nonce is copied one byte + * after the start of IV field + */ + if (cparams->aead_algo == RTE_CRYPTO_AEAD_AES_CCM) + rte_memcpy(iv_ptr + 1, cparams->aead_iv.data, + cparams->aead_iv.length); + else + rte_memcpy(iv_ptr, cparams->aead_iv.data, + cparams->aead_iv.length); op->sym->aead.data.offset = ipdata_offset; op->sym->aead.data.length = data_len; @@ -548,7 +557,7 @@ l2fwd_simple_crypto_enqueue(struct rte_mbuf *m, uint8_t *) + ipdata_offset + data_len; } - op->sym->auth.digest.phys_addr = rte_pktmbuf_mtophys_offset(m, + op->sym->aead.digest.phys_addr = rte_pktmbuf_iova_offset(m, rte_pktmbuf_pkt_len(m) - cparams->digest_length); if (cparams->aad.length) { @@ -566,7 +575,7 @@ l2fwd_simple_crypto_enqueue(struct rte_mbuf *m, /* Send the burst of packets on an output interface */ static int l2fwd_send_burst(struct lcore_queue_conf *qconf, unsigned n, - uint8_t port) + uint16_t port) { struct rte_mbuf **pkt_buffer; unsigned ret; @@ -587,7 +596,7 @@ l2fwd_send_burst(struct lcore_queue_conf *qconf, unsigned n, /* Enqueue packets for TX and prepare them to be sent */ static int -l2fwd_send_packet(struct rte_mbuf *m, uint8_t port) +l2fwd_send_packet(struct rte_mbuf *m, uint16_t port) { unsigned lcore_id, len; struct lcore_queue_conf *qconf; @@ -610,7 +619,7 @@ l2fwd_send_packet(struct rte_mbuf *m, uint8_t port) } static void -l2fwd_mac_updating(struct rte_mbuf *m, unsigned int dest_portid) +l2fwd_mac_updating(struct rte_mbuf *m, uint16_t dest_portid) { struct ether_hdr *eth; void *tmp; @@ -626,17 +635,17 @@ l2fwd_mac_updating(struct rte_mbuf *m, unsigned int dest_portid) } static void -l2fwd_simple_forward(struct rte_mbuf *m, unsigned int portid, +l2fwd_simple_forward(struct rte_mbuf *m, uint16_t portid, struct l2fwd_crypto_options *options) { - unsigned int dst_port; + uint16_t dst_port; dst_port = l2fwd_dst_ports[portid]; if (options->mac_updating) l2fwd_mac_updating(m, dst_port); - l2fwd_send_packet(m, (uint8_t) dst_port); + l2fwd_send_packet(m, dst_port); } /** Generate random key */ @@ -708,7 +717,8 @@ l2fwd_main_loop(struct l2fwd_crypto_options *options) unsigned lcore_id = rte_lcore_id(); uint64_t prev_tsc = 0, diff_tsc, cur_tsc, timer_tsc = 0; - unsigned i, j, portid, nb_rx, len; + unsigned int i, j, nb_rx, len; + uint16_t portid; struct lcore_queue_conf *qconf = &lcore_queue_conf[lcore_id]; const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) / US_PER_S * BURST_TX_DRAIN_US; @@ -796,6 +806,14 @@ l2fwd_main_loop(struct l2fwd_crypto_options *options) if (!options->aad_param) generate_random_key(port_cparams[i].aad.data, port_cparams[i].aad.length); + /* + * If doing AES-CCM, first 18 bytes has to be reserved, + * and actual AAD should start from byte 18 + */ + if (port_cparams[i].aead_algo == RTE_CRYPTO_AEAD_AES_CCM) + memmove(port_cparams[i].aad.data + 18, + port_cparams[i].aad.data, + port_cparams[i].aad.length); } else port_cparams[i].aad.length = 0; @@ -865,7 +883,7 @@ l2fwd_main_loop(struct l2fwd_crypto_options *options) continue; l2fwd_send_burst(&lcore_queue_conf[lcore_id], qconf->pkt_buf[portid].len, - (uint8_t) portid); + portid); qconf->pkt_buf[portid].len = 0; } @@ -899,7 +917,7 @@ l2fwd_main_loop(struct l2fwd_crypto_options *options) cparams = &port_cparams[i]; - nb_rx = rte_eth_rx_burst((uint8_t) portid, 0, + nb_rx = rte_eth_rx_burst(portid, 0, pkts_burst, MAX_PKT_BURST); port_statistics[portid].rx += nb_rx; @@ -1081,15 +1099,16 @@ parse_cipher_op(enum rte_crypto_cipher_operation *op, char *optarg) return -1; } -/** Parse crypto key command line argument */ +/** Parse bytes from command line argument */ static int -parse_key(uint8_t *data, char *input_arg) +parse_bytes(uint8_t *data, char *input_arg, uint16_t max_size) { unsigned byte_count; char *token; + errno = 0; for (byte_count = 0, token = strtok(input_arg, ":"); - (byte_count < MAX_KEY_SIZE) && (token != NULL); + (byte_count < max_size) && (token != NULL); token = strtok(NULL, ":")) { int number = (int)strtol(token, NULL, 16); @@ -1229,7 +1248,8 @@ l2fwd_crypto_parse_args_long_options(struct l2fwd_crypto_options *options, else if (strcmp(lgopts[option_index].name, "cipher_key") == 0) { options->ckey_param = 1; options->cipher_xform.cipher.key.length = - parse_key(options->cipher_xform.cipher.key.data, optarg); + parse_bytes(options->cipher_xform.cipher.key.data, optarg, + MAX_KEY_SIZE); if (options->cipher_xform.cipher.key.length > 0) return 0; else @@ -1242,7 +1262,7 @@ l2fwd_crypto_parse_args_long_options(struct l2fwd_crypto_options *options, else if (strcmp(lgopts[option_index].name, "cipher_iv") == 0) { options->cipher_iv_param = 1; options->cipher_iv.length = - parse_key(options->cipher_iv.data, optarg); + parse_bytes(options->cipher_iv.data, optarg, MAX_IV_SIZE); if (options->cipher_iv.length > 0) return 0; else @@ -1265,7 +1285,8 @@ l2fwd_crypto_parse_args_long_options(struct l2fwd_crypto_options *options, else if (strcmp(lgopts[option_index].name, "auth_key") == 0) { options->akey_param = 1; options->auth_xform.auth.key.length = - parse_key(options->auth_xform.auth.key.data, optarg); + parse_bytes(options->auth_xform.auth.key.data, optarg, + MAX_KEY_SIZE); if (options->auth_xform.auth.key.length > 0) return 0; else @@ -1279,7 +1300,7 @@ l2fwd_crypto_parse_args_long_options(struct l2fwd_crypto_options *options, else if (strcmp(lgopts[option_index].name, "auth_iv") == 0) { options->auth_iv_param = 1; options->auth_iv.length = - parse_key(options->auth_iv.data, optarg); + parse_bytes(options->auth_iv.data, optarg, MAX_IV_SIZE); if (options->auth_iv.length > 0) return 0; else @@ -1302,7 +1323,8 @@ l2fwd_crypto_parse_args_long_options(struct l2fwd_crypto_options *options, else if (strcmp(lgopts[option_index].name, "aead_key") == 0) { options->aead_key_param = 1; options->aead_xform.aead.key.length = - parse_key(options->aead_xform.aead.key.data, optarg); + parse_bytes(options->aead_xform.aead.key.data, optarg, + MAX_KEY_SIZE); if (options->aead_xform.aead.key.length > 0) return 0; else @@ -1316,7 +1338,7 @@ l2fwd_crypto_parse_args_long_options(struct l2fwd_crypto_options *options, else if (strcmp(lgopts[option_index].name, "aead_iv") == 0) { options->aead_iv_param = 1; options->aead_iv.length = - parse_key(options->aead_iv.data, optarg); + parse_bytes(options->aead_iv.data, optarg, MAX_IV_SIZE); if (options->aead_iv.length > 0) return 0; else @@ -1329,7 +1351,7 @@ l2fwd_crypto_parse_args_long_options(struct l2fwd_crypto_options *options, else if (strcmp(lgopts[option_index].name, "aad") == 0) { options->aad_param = 1; options->aad.length = - parse_key(options->aad.data, optarg); + parse_bytes(options->aad.data, optarg, MAX_AAD_SIZE); if (options->aad.length > 0) return 0; else @@ -1731,11 +1753,12 @@ l2fwd_crypto_parse_args(struct l2fwd_crypto_options *options, /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -1750,14 +1773,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up. Speed %u Mbps - %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -2319,7 +2341,7 @@ initialize_cryptodevs(struct l2fwd_crypto_options *options, unsigned nb_ports, static int initialize_ports(struct l2fwd_crypto_options *options) { - uint8_t last_portid, portid; + uint16_t last_portid, portid; unsigned enabled_portcount = 0; unsigned nb_ports = rte_eth_dev_count(); @@ -2340,12 +2362,12 @@ initialize_ports(struct l2fwd_crypto_options *options) continue; /* init port */ - printf("Initializing port %u... ", (unsigned) portid); + printf("Initializing port %u... ", portid); fflush(stdout); retval = rte_eth_dev_configure(portid, 1, 1, &port_conf); if (retval < 0) { printf("Cannot configure device: err=%d, port=%u\n", - retval, (unsigned) portid); + retval, portid); return -1; } @@ -2353,7 +2375,7 @@ initialize_ports(struct l2fwd_crypto_options *options) &nb_txd); if (retval < 0) { printf("Cannot adjust number of descriptors: err=%d, port=%u\n", - retval, (unsigned) portid); + retval, portid); return -1; } @@ -2364,7 +2386,7 @@ initialize_ports(struct l2fwd_crypto_options *options) NULL, l2fwd_pktmbuf_pool); if (retval < 0) { printf("rte_eth_rx_queue_setup:err=%d, port=%u\n", - retval, (unsigned) portid); + retval, portid); return -1; } @@ -2375,7 +2397,7 @@ initialize_ports(struct l2fwd_crypto_options *options) NULL); if (retval < 0) { printf("rte_eth_tx_queue_setup:err=%d, port=%u\n", - retval, (unsigned) portid); + retval, portid); return -1; } @@ -2384,7 +2406,7 @@ initialize_ports(struct l2fwd_crypto_options *options) retval = rte_eth_dev_start(portid); if (retval < 0) { printf("rte_eth_dev_start:err=%d, port=%u\n", - retval, (unsigned) portid); + retval, portid); return -1; } @@ -2393,7 +2415,7 @@ initialize_ports(struct l2fwd_crypto_options *options) rte_eth_macaddr_get(portid, &l2fwd_ports_eth_addr[portid]); printf("Port %u, MAC address: %02X:%02X:%02X:%02X:%02X:%02X\n\n", - (unsigned) portid, + portid, l2fwd_ports_eth_addr[portid].addr_bytes[0], l2fwd_ports_eth_addr[portid].addr_bytes[1], l2fwd_ports_eth_addr[portid].addr_bytes[2], @@ -2461,7 +2483,7 @@ reserve_key_memory(struct l2fwd_crypto_options *options) options->aad.data = rte_malloc("aad", MAX_KEY_SIZE, 0); if (options->aad.data == NULL) rte_exit(EXIT_FAILURE, "Failed to allocate memory for AAD"); - options->aad.phys_addr = rte_malloc_virt2phy(options->aad.data); + options->aad.phys_addr = rte_malloc_virt2iova(options->aad.data); } int @@ -2470,7 +2492,8 @@ main(int argc, char **argv) struct lcore_queue_conf *qconf; struct l2fwd_crypto_options options; - uint8_t nb_ports, nb_cryptodevs, portid, cdev_id; + uint8_t nb_cryptodevs, cdev_id; + uint16_t nb_ports, portid; unsigned lcore_id, rx_lcore_id; int ret, enabled_cdevcount, enabled_portcount; uint8_t enabled_cdevs[RTE_CRYPTO_MAX_DEVS] = {0}; @@ -2547,7 +2570,7 @@ main(int argc, char **argv) qconf->rx_port_list[qconf->nb_rx_ports] = portid; qconf->nb_rx_ports++; - printf("Lcore %u: RX port %u\n", rx_lcore_id, (unsigned)portid); + printf("Lcore %u: RX port %u\n", rx_lcore_id, portid); } /* Enable Crypto devices */ diff --git a/examples/l2fwd-jobstats/main.c b/examples/l2fwd-jobstats/main.c index 98936206..485370de 100644 --- a/examples/l2fwd-jobstats/main.c +++ b/examples/l2fwd-jobstats/main.c @@ -43,7 +43,6 @@ #include <rte_malloc.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -53,7 +52,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_debug.h> #include <rte_ether.h> #include <rte_ethdev.h> @@ -176,7 +174,7 @@ show_lcore_stats(unsigned lcore_id) uint64_t busy, busy_min, busy_max; /* Jobs statistics. */ - const uint8_t port_cnt = qconf->n_rx_port; + const uint16_t port_cnt = qconf->n_rx_port; uint64_t jobs_exec_cnt[port_cnt], jobs_period[port_cnt]; uint64_t jobs_exec[port_cnt], jobs_exec_min[port_cnt], jobs_exec_max[port_cnt]; @@ -414,11 +412,11 @@ l2fwd_fwd_job(__rte_unused struct rte_timer *timer, void *arg) struct rte_mbuf *pkts_burst[MAX_PKT_BURST]; struct rte_mbuf *m; - const uint8_t port_idx = (uintptr_t) arg; + const uint16_t port_idx = (uintptr_t) arg; const unsigned lcore_id = rte_lcore_id(); struct lcore_queue_conf *qconf = &lcore_queue_conf[lcore_id]; struct rte_jobstats *job = &qconf->port_fwd_jobs[port_idx]; - const uint8_t portid = qconf->rx_port_list[port_idx]; + const uint16_t portid = qconf->rx_port_list[port_idx]; uint8_t j; uint16_t total_nb_rx; @@ -428,7 +426,7 @@ l2fwd_fwd_job(__rte_unused struct rte_timer *timer, void *arg) /* Call rx burst 2 times. This allow rte_jobstats logic to see if this * function must be called more frequently. */ - total_nb_rx = rte_eth_rx_burst((uint8_t) portid, 0, pkts_burst, + total_nb_rx = rte_eth_rx_burst(portid, 0, pkts_burst, MAX_PKT_BURST); for (j = 0; j < total_nb_rx; j++) { @@ -438,7 +436,7 @@ l2fwd_fwd_job(__rte_unused struct rte_timer *timer, void *arg) } if (total_nb_rx == MAX_PKT_BURST) { - const uint16_t nb_rx = rte_eth_rx_burst((uint8_t) portid, 0, pkts_burst, + const uint16_t nb_rx = rte_eth_rx_burst(portid, 0, pkts_burst, MAX_PKT_BURST); total_nb_rx += nb_rx; @@ -464,7 +462,7 @@ l2fwd_flush_job(__rte_unused struct rte_timer *timer, __rte_unused void *arg) uint64_t now; unsigned lcore_id; struct lcore_queue_conf *qconf; - uint8_t portid; + uint16_t portid; unsigned i; uint32_t sent; struct rte_eth_dev_tx_buffer *buffer; @@ -714,11 +712,12 @@ l2fwd_parse_args(int argc, char **argv) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -733,14 +732,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up. Speed %u Mbps - %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -776,9 +774,9 @@ main(int argc, char **argv) unsigned nb_ports_in_mask = 0; int ret; char name[RTE_JOBSTATS_NAMESIZE]; - uint8_t nb_ports; - uint8_t nb_ports_available; - uint8_t portid, last_port; + uint16_t nb_ports; + uint16_t nb_ports_available; + uint16_t portid, last_port; uint8_t i; /* init EAL */ @@ -861,7 +859,7 @@ main(int argc, char **argv) qconf->rx_port_list[qconf->n_rx_port] = portid; qconf->n_rx_port++; - printf("Lcore %u: RX port %u\n", rx_lcore_id, (unsigned) portid); + printf("Lcore %u: RX port %u\n", rx_lcore_id, portid); } nb_ports_available = nb_ports; @@ -870,24 +868,24 @@ main(int argc, char **argv) for (portid = 0; portid < nb_ports; portid++) { /* skip ports that are not enabled */ if ((l2fwd_enabled_port_mask & (1 << portid)) == 0) { - printf("Skipping disabled port %u\n", (unsigned) portid); + printf("Skipping disabled port %u\n", portid); nb_ports_available--; continue; } /* init port */ - printf("Initializing port %u... ", (unsigned) portid); + printf("Initializing port %u... ", portid); fflush(stdout); ret = rte_eth_dev_configure(portid, 1, 1, &port_conf); if (ret < 0) rte_exit(EXIT_FAILURE, "Cannot configure device: err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); ret = rte_eth_dev_adjust_nb_rx_tx_desc(portid, &nb_rxd, &nb_txd); if (ret < 0) rte_exit(EXIT_FAILURE, "Cannot adjust number of descriptors: err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); rte_eth_macaddr_get(portid, &l2fwd_ports_eth_addr[portid]); @@ -899,7 +897,7 @@ main(int argc, char **argv) l2fwd_pktmbuf_pool); if (ret < 0) rte_exit(EXIT_FAILURE, "rte_eth_rx_queue_setup:err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); /* init one TX queue on each port */ fflush(stdout); @@ -907,8 +905,9 @@ main(int argc, char **argv) rte_eth_dev_socket_id(portid), NULL); if (ret < 0) - rte_exit(EXIT_FAILURE, "rte_eth_tx_queue_setup:err=%d, port=%u\n", - ret, (unsigned) portid); + rte_exit(EXIT_FAILURE, + "rte_eth_tx_queue_setup:err=%d, port=%u\n", + ret, portid); /* Initialize TX buffers */ tx_buffer[portid] = rte_zmalloc_socket("tx_buffer", @@ -916,7 +915,7 @@ main(int argc, char **argv) rte_eth_dev_socket_id(portid)); if (tx_buffer[portid] == NULL) rte_exit(EXIT_FAILURE, "Cannot allocate buffer for tx on port %u\n", - (unsigned) portid); + portid); rte_eth_tx_buffer_init(tx_buffer[portid], MAX_PKT_BURST); @@ -924,21 +923,22 @@ main(int argc, char **argv) rte_eth_tx_buffer_count_callback, &port_statistics[portid].dropped); if (ret < 0) - rte_exit(EXIT_FAILURE, "Cannot set error callback for " - "tx buffer on port %u\n", (unsigned) portid); + rte_exit(EXIT_FAILURE, + "Cannot set error callback for tx buffer on port %u\n", + portid); /* Start device */ ret = rte_eth_dev_start(portid); if (ret < 0) rte_exit(EXIT_FAILURE, "rte_eth_dev_start:err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); printf("done:\n"); rte_eth_promiscuous_enable(portid); printf("Port %u, MAC address: %02X:%02X:%02X:%02X:%02X:%02X\n\n", - (unsigned) portid, + portid, l2fwd_ports_eth_addr[portid].addr_bytes[0], l2fwd_ports_eth_addr[portid].addr_bytes[1], l2fwd_ports_eth_addr[portid].addr_bytes[2], diff --git a/examples/l2fwd-keepalive/main.c b/examples/l2fwd-keepalive/main.c index 83bc542c..358ca5ec 100644 --- a/examples/l2fwd-keepalive/main.c +++ b/examples/l2fwd-keepalive/main.c @@ -51,7 +51,6 @@ #include <rte_malloc.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -61,7 +60,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -157,7 +155,7 @@ print_stats(__attribute__((unused)) struct rte_timer *ptr_timer, __attribute__((unused)) void *ptr_data) { uint64_t total_packets_dropped, total_packets_tx, total_packets_rx; - unsigned portid; + uint16_t portid; total_packets_dropped = 0; total_packets_tx = 0; @@ -299,7 +297,7 @@ l2fwd_main_loop(void) for (i = 0; i < qconf->n_rx_port; i++) { portid = qconf->rx_port_list[i]; - nb_rx = rte_eth_rx_burst((uint8_t) portid, 0, + nb_rx = rte_eth_rx_burst(portid, 0, pkts_burst, MAX_PKT_BURST); port_statistics[portid].rx += nb_rx; @@ -479,11 +477,12 @@ l2fwd_parse_args(int argc, char **argv) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -498,14 +497,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up. Speed %u Mbps - %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -560,9 +558,9 @@ main(int argc, char **argv) struct lcore_queue_conf *qconf; struct rte_eth_dev_info dev_info; int ret; - uint8_t nb_ports; - uint8_t nb_ports_available; - uint8_t portid, last_port; + uint16_t nb_ports; + uint16_t nb_ports_available; + uint16_t portid, last_port; unsigned lcore_id, rx_lcore_id; unsigned nb_ports_in_mask = 0; struct sigaction signal_handler; @@ -653,7 +651,7 @@ main(int argc, char **argv) qconf->rx_port_list[qconf->n_rx_port] = portid; qconf->n_rx_port++; printf("Lcore %u: RX port %u\n", - rx_lcore_id, (unsigned) portid); + rx_lcore_id, portid); } nb_ports_available = nb_ports; @@ -662,26 +660,25 @@ main(int argc, char **argv) for (portid = 0; portid < nb_ports; portid++) { /* skip ports that are not enabled */ if ((l2fwd_enabled_port_mask & (1 << portid)) == 0) { - printf("Skipping disabled port %u\n", - (unsigned) portid); + printf("Skipping disabled port %u\n", portid); nb_ports_available--; continue; } /* init port */ - printf("Initializing port %u... ", (unsigned) portid); + printf("Initializing port %u... ", portid); fflush(stdout); ret = rte_eth_dev_configure(portid, 1, 1, &port_conf); if (ret < 0) rte_exit(EXIT_FAILURE, "Cannot configure device: err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); ret = rte_eth_dev_adjust_nb_rx_tx_desc(portid, &nb_rxd, &nb_txd); if (ret < 0) rte_exit(EXIT_FAILURE, "Cannot adjust number of descriptors: err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); rte_eth_macaddr_get(portid, &l2fwd_ports_eth_addr[portid]); @@ -694,7 +691,7 @@ main(int argc, char **argv) if (ret < 0) rte_exit(EXIT_FAILURE, "rte_eth_rx_queue_setup:err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); /* init one TX queue on each port */ fflush(stdout); @@ -704,7 +701,7 @@ main(int argc, char **argv) if (ret < 0) rte_exit(EXIT_FAILURE, "rte_eth_tx_queue_setup:err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); /* Initialize TX buffers */ tx_buffer[portid] = rte_zmalloc_socket("tx_buffer", @@ -712,7 +709,7 @@ main(int argc, char **argv) rte_eth_dev_socket_id(portid)); if (tx_buffer[portid] == NULL) rte_exit(EXIT_FAILURE, "Cannot allocate buffer for tx on port %u\n", - (unsigned) portid); + portid); rte_eth_tx_buffer_init(tx_buffer[portid], MAX_PKT_BURST); @@ -720,21 +717,22 @@ main(int argc, char **argv) rte_eth_tx_buffer_count_callback, &port_statistics[portid].dropped); if (ret < 0) - rte_exit(EXIT_FAILURE, "Cannot set error callback for " - "tx buffer on port %u\n", (unsigned) portid); + rte_exit(EXIT_FAILURE, + "Cannot set error callback for tx buffer on port %u\n", + portid); /* Start device */ ret = rte_eth_dev_start(portid); if (ret < 0) rte_exit(EXIT_FAILURE, "rte_eth_dev_start:err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); rte_eth_promiscuous_enable(portid); printf("Port %u, MAC address: " "%02X:%02X:%02X:%02X:%02X:%02X\n\n", - (unsigned) portid, + portid, l2fwd_ports_eth_addr[portid].addr_bytes[0], l2fwd_ports_eth_addr[portid].addr_bytes[1], l2fwd_ports_eth_addr[portid].addr_bytes[2], diff --git a/examples/l2fwd/main.c b/examples/l2fwd/main.c index 14263358..e89e2e1b 100644 --- a/examples/l2fwd/main.c +++ b/examples/l2fwd/main.c @@ -52,7 +52,6 @@ #include <rte_malloc.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -62,7 +61,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -304,7 +302,7 @@ l2fwd_main_loop(void) for (i = 0; i < qconf->n_rx_port; i++) { portid = qconf->rx_port_list[i]; - nb_rx = rte_eth_rx_burst((uint8_t) portid, 0, + nb_rx = rte_eth_rx_burst(portid, 0, pkts_burst, MAX_PKT_BURST); port_statistics[portid].rx += nb_rx; @@ -480,11 +478,12 @@ l2fwd_parse_args(int argc, char **argv) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -503,14 +502,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up. Speed %u Mbps - %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -553,9 +551,9 @@ main(int argc, char **argv) struct lcore_queue_conf *qconf; struct rte_eth_dev_info dev_info; int ret; - uint8_t nb_ports; - uint8_t nb_ports_available; - uint8_t portid, last_port; + uint16_t nb_ports; + uint16_t nb_ports_available; + uint16_t portid, last_port; unsigned lcore_id, rx_lcore_id; unsigned nb_ports_in_mask = 0; @@ -644,7 +642,7 @@ main(int argc, char **argv) qconf->rx_port_list[qconf->n_rx_port] = portid; qconf->n_rx_port++; - printf("Lcore %u: RX port %u\n", rx_lcore_id, (unsigned) portid); + printf("Lcore %u: RX port %u\n", rx_lcore_id, portid); } nb_ports_available = nb_ports; @@ -653,24 +651,24 @@ main(int argc, char **argv) for (portid = 0; portid < nb_ports; portid++) { /* skip ports that are not enabled */ if ((l2fwd_enabled_port_mask & (1 << portid)) == 0) { - printf("Skipping disabled port %u\n", (unsigned) portid); + printf("Skipping disabled port %u\n", portid); nb_ports_available--; continue; } /* init port */ - printf("Initializing port %u... ", (unsigned) portid); + printf("Initializing port %u... ", portid); fflush(stdout); ret = rte_eth_dev_configure(portid, 1, 1, &port_conf); if (ret < 0) rte_exit(EXIT_FAILURE, "Cannot configure device: err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); ret = rte_eth_dev_adjust_nb_rx_tx_desc(portid, &nb_rxd, &nb_txd); if (ret < 0) rte_exit(EXIT_FAILURE, "Cannot adjust number of descriptors: err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); rte_eth_macaddr_get(portid,&l2fwd_ports_eth_addr[portid]); @@ -682,7 +680,7 @@ main(int argc, char **argv) l2fwd_pktmbuf_pool); if (ret < 0) rte_exit(EXIT_FAILURE, "rte_eth_rx_queue_setup:err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); /* init one TX queue on each port */ fflush(stdout); @@ -691,7 +689,7 @@ main(int argc, char **argv) NULL); if (ret < 0) rte_exit(EXIT_FAILURE, "rte_eth_tx_queue_setup:err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); /* Initialize TX buffers */ tx_buffer[portid] = rte_zmalloc_socket("tx_buffer", @@ -699,7 +697,7 @@ main(int argc, char **argv) rte_eth_dev_socket_id(portid)); if (tx_buffer[portid] == NULL) rte_exit(EXIT_FAILURE, "Cannot allocate buffer for tx on port %u\n", - (unsigned) portid); + portid); rte_eth_tx_buffer_init(tx_buffer[portid], MAX_PKT_BURST); @@ -707,21 +705,22 @@ main(int argc, char **argv) rte_eth_tx_buffer_count_callback, &port_statistics[portid].dropped); if (ret < 0) - rte_exit(EXIT_FAILURE, "Cannot set error callback for " - "tx buffer on port %u\n", (unsigned) portid); + rte_exit(EXIT_FAILURE, + "Cannot set error callback for tx buffer on port %u\n", + portid); /* Start device */ ret = rte_eth_dev_start(portid); if (ret < 0) rte_exit(EXIT_FAILURE, "rte_eth_dev_start:err=%d, port=%u\n", - ret, (unsigned) portid); + ret, portid); printf("done: \n"); rte_eth_promiscuous_enable(portid); printf("Port %u, MAC address: %02X:%02X:%02X:%02X:%02X:%02X\n\n", - (unsigned) portid, + portid, l2fwd_ports_eth_addr[portid].addr_bytes[0], l2fwd_ports_eth_addr[portid].addr_bytes[1], l2fwd_ports_eth_addr[portid].addr_bytes[2], diff --git a/examples/l3fwd-acl/main.c b/examples/l3fwd-acl/main.c index 8eff4de4..e50b1a1a 100644 --- a/examples/l3fwd-acl/main.c +++ b/examples/l3fwd-acl/main.c @@ -47,7 +47,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -57,7 +56,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -121,7 +119,7 @@ static int promiscuous_on; /**< Ports set in promiscuous mode off by default. */ static int numa_on = 1; /**< NUMA is enabled by default. */ struct lcore_rx_queue { - uint8_t port_id; + uint16_t port_id; uint8_t queue_id; } __rte_cache_aligned; @@ -131,7 +129,7 @@ struct lcore_rx_queue { #define MAX_LCORE_PARAMS 1024 struct lcore_params { - uint8_t port_id; + uint16_t port_id; uint8_t queue_id; uint8_t lcore_id; } __rte_cache_aligned; @@ -184,7 +182,7 @@ static inline int is_valid_ipv4_pkt(struct ipv4_hdr *pkt, uint32_t link_len); #endif static inline void -send_single_packet(struct rte_mbuf *m, uint8_t port); +send_single_packet(struct rte_mbuf *m, uint16_t port); #define MAX_ACL_RULE_NUM 100000 #define DEFAULT_MAX_CATEGORIES 1 @@ -1026,6 +1024,7 @@ add_rules(const char *rule_path, char buff[LINE_MAX]; FILE *fh = fopen(rule_path, "rb"); unsigned int i = 0; + int val; if (fh == NULL) rte_exit(EXIT_FAILURE, "%s: Open %s failed\n", __func__, @@ -1042,7 +1041,11 @@ add_rules(const char *rule_path, rte_exit(EXIT_FAILURE, "Not find any route entries in %s!\n", rule_path); - fseek(fh, 0, SEEK_SET); + val = fseek(fh, 0, SEEK_SET); + if (val < 0) { + rte_exit(EXIT_FAILURE, "%s: File seek operation failed\n", + __func__); + } acl_rules = calloc(acl_num, rule_size); @@ -1297,7 +1300,7 @@ static struct lcore_conf lcore_conf[RTE_MAX_LCORE]; /* Enqueue a single packet, and send burst if queue is filled */ static inline void -send_single_packet(struct rte_mbuf *m, uint8_t port) +send_single_packet(struct rte_mbuf *m, uint16_t port) { uint32_t lcore_id; struct lcore_conf *qconf; @@ -1358,7 +1361,8 @@ main_loop(__attribute__((unused)) void *dummy) unsigned lcore_id; uint64_t prev_tsc, diff_tsc, cur_tsc; int i, nb_rx; - uint8_t portid, queueid; + uint16_t portid; + uint8_t queueid; struct lcore_conf *qconf; int socketid; const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) @@ -1381,7 +1385,7 @@ main_loop(__attribute__((unused)) void *dummy) portid = qconf->rx_queue_list[i].port_id; queueid = qconf->rx_queue_list[i].queue_id; RTE_LOG(INFO, L3FWD, - " -- lcoreid=%u portid=%hhu rxqueueid=%hhu\n", + " -- lcoreid=%u portid=%u rxqueueid=%hhu\n", lcore_id, portid, queueid); } @@ -1500,7 +1504,7 @@ check_port_config(const unsigned nb_ports) } static uint8_t -get_port_n_rx_queues(const uint8_t port) +get_port_n_rx_queues(const uint16_t port) { int queue = -1; uint16_t i; @@ -1829,11 +1833,12 @@ init_mem(unsigned nb_mbuf) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -1848,14 +1853,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up. Speed %u Mbps %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -1893,7 +1897,8 @@ main(int argc, char **argv) uint16_t queueid; unsigned lcore_id; uint32_t n_tx_queue, nb_lcores; - uint8_t portid, nb_rx_queue, queue, socketid; + uint16_t portid; + uint8_t nb_rx_queue, queue, socketid; /* init EAL */ ret = rte_eal_init(argc, argv); diff --git a/examples/l3fwd-power/main.c b/examples/l3fwd-power/main.c index fd442f5e..0a4ed145 100644 --- a/examples/l3fwd-power/main.c +++ b/examples/l3fwd-power/main.c @@ -50,7 +50,6 @@ #include <rte_malloc.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -60,7 +59,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -174,7 +172,7 @@ enum freq_scale_hint_t }; struct lcore_rx_queue { - uint8_t port_id; + uint16_t port_id; uint8_t queue_id; enum freq_scale_hint_t freq_up_hint; uint32_t zero_rx_packet_count; @@ -190,7 +188,7 @@ struct lcore_rx_queue { #define MAX_LCORE_PARAMS 1024 struct lcore_params { - uint8_t port_id; + uint16_t port_id; uint8_t queue_id; uint8_t lcore_id; } __rte_cache_aligned; @@ -308,8 +306,8 @@ static lookup_struct_t *ipv6_l3fwd_lookup_struct[NB_SOCKETS]; #define IPV6_L3FWD_NUM_ROUTES \ (sizeof(ipv6_l3fwd_route_array) / sizeof(ipv6_l3fwd_route_array[0])) -static uint8_t ipv4_l3fwd_out_if[L3FWD_HASH_ENTRIES] __rte_cache_aligned; -static uint8_t ipv6_l3fwd_out_if[L3FWD_HASH_ENTRIES] __rte_cache_aligned; +static uint16_t ipv4_l3fwd_out_if[L3FWD_HASH_ENTRIES] __rte_cache_aligned; +static uint16_t ipv6_l3fwd_out_if[L3FWD_HASH_ENTRIES] __rte_cache_aligned; #endif #if (APP_LOOKUP_METHOD == APP_LOOKUP_LPM) @@ -370,7 +368,7 @@ static struct rte_timer power_timers[RTE_MAX_LCORE]; static inline uint32_t power_idle_heuristic(uint32_t zero_rx_packet_count); static inline enum freq_scale_hint_t power_freq_scaleup_heuristic( \ - unsigned lcore_id, uint8_t port_id, uint16_t queue_id); + unsigned int lcore_id, uint16_t port_id, uint16_t queue_id); /* exit signal handler */ static void @@ -451,7 +449,7 @@ power_timer_cb(__attribute__((unused)) struct rte_timer *tim, /* Enqueue a single packet, and send burst if queue is filled */ static inline int -send_single_packet(struct rte_mbuf *m, uint8_t port) +send_single_packet(struct rte_mbuf *m, uint16_t port) { uint32_t lcore_id; struct lcore_conf *qconf; @@ -523,8 +521,8 @@ print_ipv6_key(struct ipv6_5tuple key) key.port_dst, key.port_src, key.proto); } -static inline uint8_t -get_ipv4_dst_port(struct ipv4_hdr *ipv4_hdr, uint8_t portid, +static inline uint16_t +get_ipv4_dst_port(struct ipv4_hdr *ipv4_hdr, uint16_t portid, lookup_struct_t * ipv4_l3fwd_lookup_struct) { struct ipv4_5tuple key; @@ -559,11 +557,11 @@ get_ipv4_dst_port(struct ipv4_hdr *ipv4_hdr, uint8_t portid, /* Find destination port */ ret = rte_hash_lookup(ipv4_l3fwd_lookup_struct, (const void *)&key); - return (uint8_t)((ret < 0)? portid : ipv4_l3fwd_out_if[ret]); + return ((ret < 0) ? portid : ipv4_l3fwd_out_if[ret]); } -static inline uint8_t -get_ipv6_dst_port(struct ipv6_hdr *ipv6_hdr, uint8_t portid, +static inline uint16_t +get_ipv6_dst_port(struct ipv6_hdr *ipv6_hdr, uint16_t portid, lookup_struct_t *ipv6_l3fwd_lookup_struct) { struct ipv6_5tuple key; @@ -599,18 +597,18 @@ get_ipv6_dst_port(struct ipv6_hdr *ipv6_hdr, uint8_t portid, /* Find destination port */ ret = rte_hash_lookup(ipv6_l3fwd_lookup_struct, (const void *)&key); - return (uint8_t)((ret < 0)? portid : ipv6_l3fwd_out_if[ret]); + return ((ret < 0) ? portid : ipv6_l3fwd_out_if[ret]); } #endif #if (APP_LOOKUP_METHOD == APP_LOOKUP_LPM) -static inline uint8_t -get_ipv4_dst_port(struct ipv4_hdr *ipv4_hdr, uint8_t portid, +static inline uint16_t +get_ipv4_dst_port(struct ipv4_hdr *ipv4_hdr, uint16_t portid, lookup_struct_t *ipv4_l3fwd_lookup_struct) { uint32_t next_hop; - return (uint8_t) ((rte_lpm_lookup(ipv4_l3fwd_lookup_struct, + return ((rte_lpm_lookup(ipv4_l3fwd_lookup_struct, rte_be_to_cpu_32(ipv4_hdr->dst_addr), &next_hop) == 0)? next_hop : portid); } @@ -634,7 +632,7 @@ parse_ptype_one(struct rte_mbuf *m) } static uint16_t -cb_parse_ptype(uint8_t port __rte_unused, uint16_t queue __rte_unused, +cb_parse_ptype(uint16_t port __rte_unused, uint16_t queue __rte_unused, struct rte_mbuf *pkts[], uint16_t nb_pkts, uint16_t max_pkts __rte_unused, void *user_param __rte_unused) @@ -648,7 +646,7 @@ cb_parse_ptype(uint8_t port __rte_unused, uint16_t queue __rte_unused, } static int -add_cb_parse_ptype(uint8_t portid, uint16_t queueid) +add_cb_parse_ptype(uint16_t portid, uint16_t queueid) { printf("Port %d: softly parse packet type info\n", portid); if (rte_eth_add_rx_callback(portid, queueid, cb_parse_ptype, NULL)) @@ -659,13 +657,13 @@ add_cb_parse_ptype(uint8_t portid, uint16_t queueid) } static inline void -l3fwd_simple_forward(struct rte_mbuf *m, uint8_t portid, +l3fwd_simple_forward(struct rte_mbuf *m, uint16_t portid, struct lcore_conf *qconf) { struct ether_hdr *eth_hdr; struct ipv4_hdr *ipv4_hdr; void *d_addr_bytes; - uint8_t dst_port; + uint16_t dst_port; eth_hdr = rte_pktmbuf_mtod(m, struct ether_hdr *); @@ -752,13 +750,11 @@ power_idle_heuristic(uint32_t zero_rx_packet_count) */ else return SUSPEND_THRESHOLD; - - return 0; } static inline enum freq_scale_hint_t power_freq_scaleup_heuristic(unsigned lcore_id, - uint8_t port_id, + uint16_t port_id, uint16_t queue_id) { /** @@ -805,7 +801,8 @@ sleep_until_rx_interrupt(int num) { struct rte_epoll_event event[num]; int n, i; - uint8_t port_id, queue_id; + uint16_t port_id; + uint8_t queue_id; void *data; RTE_LOG(INFO, L3FWD_POWER, @@ -832,7 +829,8 @@ static void turn_on_intr(struct lcore_conf *qconf) { int i; struct lcore_rx_queue *rx_queue; - uint8_t port_id, queue_id; + uint8_t queue_id; + uint16_t port_id; for (i = 0; i < qconf->n_rx_queue; ++i) { rx_queue = &(qconf->rx_queue_list[i]); @@ -848,7 +846,8 @@ static void turn_on_intr(struct lcore_conf *qconf) static int event_register(struct lcore_conf *qconf) { struct lcore_rx_queue *rx_queue; - uint8_t portid, queueid; + uint8_t queueid; + uint16_t portid; uint32_t data; int ret; int i; @@ -879,7 +878,8 @@ main_loop(__attribute__((unused)) void *dummy) uint64_t prev_tsc, diff_tsc, cur_tsc; uint64_t prev_tsc_power = 0, cur_tsc_power, diff_tsc_power; int i, j, nb_rx; - uint8_t portid, queueid; + uint8_t queueid; + uint16_t portid; struct lcore_conf *qconf; struct lcore_rx_queue *rx_queue; enum freq_scale_hint_t lcore_scaleup_hint; @@ -904,7 +904,7 @@ main_loop(__attribute__((unused)) void *dummy) for (i = 0; i < qconf->n_rx_queue; i++) { portid = qconf->rx_queue_list[i].port_id; queueid = qconf->rx_queue_list[i].queue_id; - RTE_LOG(INFO, L3FWD_POWER, " -- lcoreid=%u portid=%hhu " + RTE_LOG(INFO, L3FWD_POWER, " -- lcoreid=%u portid=%u " "rxqueueid=%hhu\n", lcore_id, portid, queueid); } @@ -1111,7 +1111,7 @@ check_port_config(const unsigned nb_ports) } static uint8_t -get_port_n_rx_queues(const uint8_t port) +get_port_n_rx_queues(const uint16_t port) { int queue = -1; uint16_t i; @@ -1541,11 +1541,12 @@ init_mem(unsigned nb_mbuf) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint8_t count, all_ports_up, print_flag = 0; + uint16_t portid; struct rte_eth_link link; printf("\nChecking link status"); @@ -1594,7 +1595,7 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) } } -static int check_ptype(uint8_t portid) +static int check_ptype(uint16_t portid) { int i, ret; int ptype_l3_ipv4 = 0; @@ -1645,13 +1646,14 @@ main(int argc, char **argv) struct rte_eth_dev_info dev_info; struct rte_eth_txconf *txconf; int ret; - unsigned nb_ports; + uint16_t nb_ports; uint16_t queueid; unsigned lcore_id; uint64_t hz; uint32_t n_tx_queue, nb_lcores; uint32_t dev_rxq_num, dev_txq_num; - uint8_t portid, nb_rx_queue, queue, socketid; + uint8_t nb_rx_queue, queue, socketid; + uint16_t portid; uint16_t org_rxq_intr = port_conf.intr_conf.rxq; /* catch SIGINT and restore cpufreq governor to ondemand */ @@ -1751,7 +1753,7 @@ main(int argc, char **argv) rte_eth_dev_socket_id(portid)); if (qconf->tx_buffer[portid] == NULL) rte_exit(EXIT_FAILURE, "Can't allocate tx buffer for port %u\n", - (unsigned) portid); + portid); rte_eth_tx_buffer_init(qconf->tx_buffer[portid], MAX_PKT_BURST); } @@ -1871,7 +1873,7 @@ main(int argc, char **argv) rte_spinlock_init(&(locks[portid])); } - check_all_ports_link_status((uint8_t)nb_ports, enabled_port_mask); + check_all_ports_link_status(nb_ports, enabled_port_mask); /* launch per-lcore init on every lcore */ rte_eal_mp_remote_launch(main_loop, NULL, CALL_MASTER); diff --git a/examples/l3fwd-vf/main.c b/examples/l3fwd-vf/main.c index 34e4a6be..6ef89fc8 100644 --- a/examples/l3fwd-vf/main.c +++ b/examples/l3fwd-vf/main.c @@ -48,7 +48,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -59,7 +58,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -155,7 +153,7 @@ struct mbuf_table { }; struct lcore_rx_queue { - uint8_t port_id; + uint16_t port_id; uint8_t queue_id; } __rte_cache_aligned; @@ -165,7 +163,7 @@ struct lcore_rx_queue { #define MAX_LCORE_PARAMS 1024 struct lcore_params { - uint8_t port_id; + uint16_t port_id; uint8_t queue_id; uint8_t lcore_id; } __rte_cache_aligned; @@ -300,7 +298,7 @@ static struct lcore_conf lcore_conf[RTE_MAX_LCORE]; static rte_spinlock_t spinlock_conf[RTE_MAX_ETHPORTS] = {RTE_SPINLOCK_INITIALIZER}; /* Send burst of packets on an output interface */ static inline int -send_burst(struct lcore_conf *qconf, uint16_t n, uint8_t port) +send_burst(struct lcore_conf *qconf, uint16_t n, uint16_t port) { struct rte_mbuf **m_table; int ret; @@ -324,7 +322,7 @@ send_burst(struct lcore_conf *qconf, uint16_t n, uint8_t port) /* Enqueue a single packet, and send burst if queue is filled */ static inline int -send_single_packet(struct rte_mbuf *m, uint8_t port) +send_single_packet(struct rte_mbuf *m, uint16_t port) { uint32_t lcore_id; uint16_t len; @@ -396,8 +394,9 @@ print_key(struct ipv4_5tuple key) (unsigned)key.ip_dst, (unsigned)key.ip_src, key.port_dst, key.port_src, key.proto); } -static inline uint8_t -get_dst_port(struct ipv4_hdr *ipv4_hdr, uint8_t portid, lookup_struct_t * l3fwd_lookup_struct) +static inline uint16_t +get_dst_port(struct ipv4_hdr *ipv4_hdr, uint16_t portid, + lookup_struct_t *l3fwd_lookup_struct) { struct ipv4_5tuple key; struct tcp_hdr *tcp; @@ -430,29 +429,31 @@ get_dst_port(struct ipv4_hdr *ipv4_hdr, uint8_t portid, lookup_struct_t * l3fwd /* Find destination port */ ret = rte_hash_lookup(l3fwd_lookup_struct, (const void *)&key); - return (uint8_t)((ret < 0)? portid : l3fwd_out_if[ret]); + return ((ret < 0) ? portid : l3fwd_out_if[ret]); } #endif #if (APP_LOOKUP_METHOD == APP_LOOKUP_LPM) -static inline uint8_t -get_dst_port(struct ipv4_hdr *ipv4_hdr, uint8_t portid, lookup_struct_t * l3fwd_lookup_struct) +static inline uint32_t +get_dst_port(struct ipv4_hdr *ipv4_hdr, uint16_t portid, + lookup_struct_t *l3fwd_lookup_struct) { uint32_t next_hop; - return (uint8_t) ((rte_lpm_lookup(l3fwd_lookup_struct, - rte_be_to_cpu_32(ipv4_hdr->dst_addr), &next_hop) == 0)? - next_hop : portid); + return ((rte_lpm_lookup(l3fwd_lookup_struct, + rte_be_to_cpu_32(ipv4_hdr->dst_addr), &next_hop) == 0) ? + next_hop : portid); } #endif static inline void -l3fwd_simple_forward(struct rte_mbuf *m, uint8_t portid, lookup_struct_t * l3fwd_lookup_struct) +l3fwd_simple_forward(struct rte_mbuf *m, uint16_t portid, + lookup_struct_t *l3fwd_lookup_struct) { struct ether_hdr *eth_hdr; struct ipv4_hdr *ipv4_hdr; void *tmp; - uint8_t dst_port; + uint16_t dst_port; eth_hdr = rte_pktmbuf_mtod(m, struct ether_hdr *); @@ -496,7 +497,8 @@ main_loop(__attribute__((unused)) void *dummy) unsigned lcore_id; uint64_t prev_tsc, diff_tsc, cur_tsc; int i, j, nb_rx; - uint8_t portid, queueid; + uint8_t queueid; + uint16_t portid; struct lcore_conf *qconf; const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) / US_PER_S * BURST_TX_DRAIN_US; @@ -516,8 +518,8 @@ main_loop(__attribute__((unused)) void *dummy) portid = qconf->rx_queue_list[i].port_id; queueid = qconf->rx_queue_list[i].queue_id; - RTE_LOG(INFO, L3FWD, " -- lcoreid=%u portid=%hhu rxqueueid=%hhu\n", lcore_id, - portid, queueid); + RTE_LOG(INFO, L3FWD, " --lcoreid=%u portid=%u rxqueueid=%hhu\n", + lcore_id, portid, queueid); } while (1) { @@ -624,7 +626,7 @@ check_port_config(const unsigned nb_ports) } static uint8_t -get_port_n_rx_queues(const uint8_t port) +get_port_n_rx_queues(const uint16_t port) { int queue = -1; uint16_t i; @@ -676,8 +678,8 @@ print_usage(const char *prgname) static void signal_handler(int signum) { - uint8_t portid; - uint8_t nb_ports = rte_eth_dev_count(); + uint16_t portid; + uint16_t nb_ports = rte_eth_dev_count(); /* When we receive a SIGINT signal */ if (signum == SIGINT) { @@ -749,7 +751,7 @@ parse_config(const char *q_arg) nb_lcore_params); return -1; } - lcore_params_array[nb_lcore_params].port_id = (uint8_t)int_fld[FLD_PORT]; + lcore_params_array[nb_lcore_params].port_id = int_fld[FLD_PORT]; lcore_params_array[nb_lcore_params].queue_id = (uint8_t)int_fld[FLD_QUEUE]; lcore_params_array[nb_lcore_params].lcore_id = (uint8_t)int_fld[FLD_LCORE]; ++nb_lcore_params; @@ -953,11 +955,11 @@ main(int argc, char **argv) struct rte_eth_txconf *txconf; int ret; unsigned nb_ports; - uint16_t queueid; + uint16_t queueid, portid; unsigned lcore_id; uint32_t nb_lcores; uint16_t n_tx_queue; - uint8_t portid, nb_rx_queue, queue, socketid; + uint8_t nb_rx_queue, queue, socketid; signal(SIGINT, signal_handler); /* init EAL */ diff --git a/examples/l3fwd/l3fwd.h b/examples/l3fwd/l3fwd.h index 011ba148..4bb15943 100644 --- a/examples/l3fwd/l3fwd.h +++ b/examples/l3fwd/l3fwd.h @@ -83,7 +83,7 @@ struct mbuf_table { }; struct lcore_rx_queue { - uint8_t port_id; + uint16_t port_id; uint8_t queue_id; } __rte_cache_aligned; @@ -117,7 +117,7 @@ extern struct lcore_conf lcore_conf[RTE_MAX_LCORE]; /* Send burst of packets on an output interface */ static inline int -send_burst(struct lcore_conf *qconf, uint16_t n, uint8_t port) +send_burst(struct lcore_conf *qconf, uint16_t n, uint16_t port) { struct rte_mbuf **m_table; int ret; @@ -139,7 +139,7 @@ send_burst(struct lcore_conf *qconf, uint16_t n, uint8_t port) /* Enqueue a single packet, and send burst if queue is filled */ static inline int send_single_packet(struct lcore_conf *qconf, - struct rte_mbuf *m, uint8_t port) + struct rte_mbuf *m, uint16_t port) { uint16_t len; @@ -212,11 +212,11 @@ int lpm_check_ptype(int portid); uint16_t -em_cb_parse_ptype(uint8_t port, uint16_t queue, struct rte_mbuf *pkts[], +em_cb_parse_ptype(uint16_t port, uint16_t queue, struct rte_mbuf *pkts[], uint16_t nb_pkts, uint16_t max_pkts, void *user_param); uint16_t -lpm_cb_parse_ptype(uint8_t port, uint16_t queue, struct rte_mbuf *pkts[], +lpm_cb_parse_ptype(uint16_t port, uint16_t queue, struct rte_mbuf *pkts[], uint16_t nb_pkts, uint16_t max_pkts, void *user_param); int diff --git a/examples/l3fwd/l3fwd_altivec.h b/examples/l3fwd/l3fwd_altivec.h new file mode 100644 index 00000000..a1d25eaa --- /dev/null +++ b/examples/l3fwd/l3fwd_altivec.h @@ -0,0 +1,284 @@ +/*- + * BSD LICENSE + * + * Copyright(c) 2016 Intel Corporation. All rights reserved. + * Copyright(c) 2017 IBM Corporation. + * All rights reserved. + * + * 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. + * * Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + + +#ifndef _L3FWD_ALTIVEC_H_ +#define _L3FWD_ALTIVEC_H_ + +#include "l3fwd.h" +#include "l3fwd_common.h" + +/* + * Update source and destination MAC addresses in the ethernet header. + * Perform RFC1812 checks and updates for IPV4 packets. + */ +static inline void +processx4_step3(struct rte_mbuf *pkt[FWDSTEP], uint16_t dst_port[FWDSTEP]) +{ + vector unsigned int te[FWDSTEP]; + vector unsigned int ve[FWDSTEP]; + vector unsigned int *p[FWDSTEP]; + + p[0] = rte_pktmbuf_mtod(pkt[0], vector unsigned int *); + p[1] = rte_pktmbuf_mtod(pkt[1], vector unsigned int *); + p[2] = rte_pktmbuf_mtod(pkt[2], vector unsigned int *); + p[3] = rte_pktmbuf_mtod(pkt[3], vector unsigned int *); + + ve[0] = (vector unsigned int)val_eth[dst_port[0]]; + te[0] = *p[0]; + + ve[1] = (vector unsigned int)val_eth[dst_port[1]]; + te[1] = *p[1]; + + ve[2] = (vector unsigned int)val_eth[dst_port[2]]; + te[2] = *p[2]; + + ve[3] = (vector unsigned int)val_eth[dst_port[3]]; + te[3] = *p[3]; + + /* Update first 12 bytes, keep rest bytes intact. */ + te[0] = (vector unsigned int)vec_sel( + (vector unsigned short)ve[0], + (vector unsigned short)te[0], + (vector unsigned short) {0, 0, 0, 0, + 0, 0, 0xffff, 0xffff}); + + te[1] = (vector unsigned int)vec_sel( + (vector unsigned short)ve[1], + (vector unsigned short)te[1], + (vector unsigned short) {0, 0, 0, 0, + 0, 0, 0xffff, 0xffff}); + + te[2] = (vector unsigned int)vec_sel( + (vector unsigned short)ve[2], + (vector unsigned short)te[2], + (vector unsigned short) {0, 0, 0, 0, 0, + 0, 0xffff, 0xffff}); + + te[3] = (vector unsigned int)vec_sel( + (vector unsigned short)ve[3], + (vector unsigned short)te[3], + (vector unsigned short) {0, 0, 0, 0, + 0, 0, 0xffff, 0xffff}); + + *p[0] = te[0]; + *p[1] = te[1]; + *p[2] = te[2]; + *p[3] = te[3]; + + rfc1812_process((struct ipv4_hdr *)((struct ether_hdr *)p[0] + 1), + &dst_port[0], pkt[0]->packet_type); + rfc1812_process((struct ipv4_hdr *)((struct ether_hdr *)p[1] + 1), + &dst_port[1], pkt[1]->packet_type); + rfc1812_process((struct ipv4_hdr *)((struct ether_hdr *)p[2] + 1), + &dst_port[2], pkt[2]->packet_type); + rfc1812_process((struct ipv4_hdr *)((struct ether_hdr *)p[3] + 1), + &dst_port[3], pkt[3]->packet_type); +} + +/* + * Group consecutive packets with the same destination port in bursts of 4. + * Suppose we have array of destination ports: + * dst_port[] = {a, b, c, d,, e, ... } + * dp1 should contain: <a, b, c, d>, dp2: <b, c, d, e>. + * We doing 4 comparisons at once and the result is 4 bit mask. + * This mask is used as an index into prebuild array of pnum values. + */ +static inline uint16_t * +port_groupx4(uint16_t pn[FWDSTEP + 1], uint16_t *lp, vector unsigned short dp1, + vector unsigned short dp2) +{ + union { + uint16_t u16[FWDSTEP + 1]; + uint64_t u64; + } *pnum = (void *)pn; + + int32_t v; + + v = vec_any_eq(dp1, dp2); + + + /* update last port counter. */ + lp[0] += gptbl[v].lpv; + + /* if dest port value has changed. */ + if (v != GRPMSK) { + pnum->u64 = gptbl[v].pnum; + pnum->u16[FWDSTEP] = 1; + lp = pnum->u16 + gptbl[v].idx; + } + + return lp; +} + +/** + * Process one packet: + * Update source and destination MAC addresses in the ethernet header. + * Perform RFC1812 checks and updates for IPV4 packets. + */ +static inline void +process_packet(struct rte_mbuf *pkt, uint16_t *dst_port) +{ + struct ether_hdr *eth_hdr; + vector unsigned int te, ve; + + eth_hdr = rte_pktmbuf_mtod(pkt, struct ether_hdr *); + + te = *(vector unsigned int *)eth_hdr; + ve = (vector unsigned int)val_eth[dst_port[0]]; + + rfc1812_process((struct ipv4_hdr *)(eth_hdr + 1), dst_port, + pkt->packet_type); + + /* dynamically vec_sel te and ve for MASK_ETH (0x3f) */ + te = (vector unsigned int)vec_sel( + (vector unsigned short)ve, + (vector unsigned short)te, + (vector unsigned short){0, 0, 0, 0, + 0, 0, 0xffff, 0xffff}); + + *(vector unsigned int *)eth_hdr = te; +} + +/** + * Send packets burst from pkts_burst to the ports in dst_port array + */ +static __rte_always_inline void +send_packets_multi(struct lcore_conf *qconf, struct rte_mbuf **pkts_burst, + uint16_t dst_port[MAX_PKT_BURST], int nb_rx) +{ + int32_t k; + int j = 0; + uint16_t dlp; + uint16_t *lp; + uint16_t pnum[MAX_PKT_BURST + 1]; + + /* + * Finish packet processing and group consecutive + * packets with the same destination port. + */ + k = RTE_ALIGN_FLOOR(nb_rx, FWDSTEP); + if (k != 0) { + vector unsigned short dp1, dp2; + + lp = pnum; + lp[0] = 1; + + processx4_step3(pkts_burst, dst_port); + + /* dp1: <d[0], d[1], d[2], d[3], ... > */ + dp1 = *(vector unsigned short *)dst_port; + + for (j = FWDSTEP; j != k; j += FWDSTEP) { + processx4_step3(&pkts_burst[j], &dst_port[j]); + + /* + * dp2: + * <d[j-3], d[j-2], d[j-1], d[j], ... > + */ + dp2 = *((vector unsigned short *) + &dst_port[j - FWDSTEP + 1]); + lp = port_groupx4(&pnum[j - FWDSTEP], lp, dp1, dp2); + + /* + * dp1: + * <d[j], d[j+1], d[j+2], d[j+3], ... > + */ + dp1 = vec_sro(dp2, (vector unsigned char) { + 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, (FWDSTEP - 1) * sizeof(dst_port[0])}); + } + + /* + * dp2: <d[j-3], d[j-2], d[j-1], d[j-1], ... > + */ + dp2 = vec_perm(dp1, (vector unsigned short){}, + (vector unsigned char){0xf9}); + lp = port_groupx4(&pnum[j - FWDSTEP], lp, dp1, dp2); + + /* + * remove values added by the last repeated + * dst port. + */ + lp[0]--; + dlp = dst_port[j - 1]; + } else { + /* set dlp and lp to the never used values. */ + dlp = BAD_PORT - 1; + lp = pnum + MAX_PKT_BURST; + } + + /* Process up to last 3 packets one by one. */ + switch (nb_rx % FWDSTEP) { + case 3: + process_packet(pkts_burst[j], dst_port + j); + GROUP_PORT_STEP(dlp, dst_port, lp, pnum, j); + j++; + /* fall-through */ + case 2: + process_packet(pkts_burst[j], dst_port + j); + GROUP_PORT_STEP(dlp, dst_port, lp, pnum, j); + j++; + /* fall-through */ + case 1: + process_packet(pkts_burst[j], dst_port + j); + GROUP_PORT_STEP(dlp, dst_port, lp, pnum, j); + j++; + } + + /* + * Send packets out, through destination port. + * Consecutive packets with the same destination port + * are already grouped together. + * If destination port for the packet equals BAD_PORT, + * then free the packet without sending it out. + */ + for (j = 0; j < nb_rx; j += k) { + + int32_t m; + uint16_t pn; + + pn = dst_port[j]; + k = pnum[j]; + + if (likely(pn != BAD_PORT)) + send_packetsx4(qconf, pn, pkts_burst + j, k); + else + for (m = j; m != j + k; m++) + rte_pktmbuf_free(pkts_burst[m]); + + } +} + +#endif /* _L3FWD_ALTIVEC_H_ */ diff --git a/examples/l3fwd/l3fwd_common.h b/examples/l3fwd/l3fwd_common.h index 2867365d..7002a43a 100644 --- a/examples/l3fwd/l3fwd_common.h +++ b/examples/l3fwd/l3fwd_common.h @@ -207,7 +207,7 @@ static const struct { }; static __rte_always_inline void -send_packetsx4(struct lcore_conf *qconf, uint8_t port, struct rte_mbuf *m[], +send_packetsx4(struct lcore_conf *qconf, uint16_t port, struct rte_mbuf *m[], uint32_t num) { uint32_t len, j, n; diff --git a/examples/l3fwd/l3fwd_em.c b/examples/l3fwd/l3fwd_em.c index 53d081bd..2b7c173b 100644 --- a/examples/l3fwd/l3fwd_em.c +++ b/examples/l3fwd/l3fwd_em.c @@ -274,8 +274,8 @@ em_mask_key(void *key, xmm_t mask) #error No vector engine (SSE, NEON, ALTIVEC) available, check your toolchain #endif -static inline uint8_t -em_get_ipv4_dst_port(void *ipv4_hdr, uint8_t portid, void *lookup_struct) +static inline uint16_t +em_get_ipv4_dst_port(void *ipv4_hdr, uint16_t portid, void *lookup_struct) { int ret = 0; union ipv4_5tuple_host key; @@ -292,11 +292,11 @@ em_get_ipv4_dst_port(void *ipv4_hdr, uint8_t portid, void *lookup_struct) /* Find destination port */ ret = rte_hash_lookup(ipv4_l3fwd_lookup_struct, (const void *)&key); - return (uint8_t)((ret < 0) ? portid : ipv4_l3fwd_out_if[ret]); + return (ret < 0) ? portid : ipv4_l3fwd_out_if[ret]; } -static inline uint8_t -em_get_ipv6_dst_port(void *ipv6_hdr, uint8_t portid, void *lookup_struct) +static inline uint16_t +em_get_ipv6_dst_port(void *ipv6_hdr, uint16_t portid, void *lookup_struct) { int ret = 0; union ipv6_5tuple_host key; @@ -325,7 +325,7 @@ em_get_ipv6_dst_port(void *ipv6_hdr, uint8_t portid, void *lookup_struct) /* Find destination port */ ret = rte_hash_lookup(ipv6_l3fwd_lookup_struct, (const void *)&key); - return (uint8_t)((ret < 0) ? portid : ipv6_l3fwd_out_if[ret]); + return (ret < 0) ? portid : ipv6_l3fwd_out_if[ret]; } #if defined RTE_ARCH_X86 || defined RTE_MACHINE_CPUFLAG_NEON @@ -628,7 +628,7 @@ em_parse_ptype(struct rte_mbuf *m) } uint16_t -em_cb_parse_ptype(uint8_t port __rte_unused, uint16_t queue __rte_unused, +em_cb_parse_ptype(uint16_t port __rte_unused, uint16_t queue __rte_unused, struct rte_mbuf *pkts[], uint16_t nb_pkts, uint16_t max_pkts __rte_unused, void *user_param __rte_unused) @@ -649,7 +649,8 @@ em_main_loop(__attribute__((unused)) void *dummy) unsigned lcore_id; uint64_t prev_tsc, diff_tsc, cur_tsc; int i, nb_rx; - uint8_t portid, queueid; + uint8_t queueid; + uint16_t portid; struct lcore_conf *qconf; const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) / US_PER_S * BURST_TX_DRAIN_US; @@ -671,7 +672,7 @@ em_main_loop(__attribute__((unused)) void *dummy) portid = qconf->rx_queue_list[i].port_id; queueid = qconf->rx_queue_list[i].queue_id; RTE_LOG(INFO, L3FWD, - " -- lcoreid=%u portid=%hhu rxqueueid=%hhu\n", + " -- lcoreid=%u portid=%u rxqueueid=%hhu\n", lcore_id, portid, queueid); } diff --git a/examples/l3fwd/l3fwd_em.h b/examples/l3fwd/l3fwd_em.h index d509a1fc..302291d7 100644 --- a/examples/l3fwd/l3fwd_em.h +++ b/examples/l3fwd/l3fwd_em.h @@ -35,12 +35,12 @@ #define __L3FWD_EM_H__ static __rte_always_inline void -l3fwd_em_simple_forward(struct rte_mbuf *m, uint8_t portid, +l3fwd_em_simple_forward(struct rte_mbuf *m, uint16_t portid, struct lcore_conf *qconf) { struct ether_hdr *eth_hdr; struct ipv4_hdr *ipv4_hdr; - uint8_t dst_port; + uint16_t dst_port; uint32_t tcp_or_udp; uint32_t l3_ptypes; @@ -112,7 +112,7 @@ l3fwd_em_simple_forward(struct rte_mbuf *m, uint8_t portid, */ static inline void l3fwd_em_no_opt_send_packets(int nb_rx, struct rte_mbuf **pkts_burst, - uint8_t portid, struct lcore_conf *qconf) + uint16_t portid, struct lcore_conf *qconf) { int32_t j; diff --git a/examples/l3fwd/l3fwd_em_hlm.h b/examples/l3fwd/l3fwd_em_hlm.h index 520672d5..9d7afe05 100644 --- a/examples/l3fwd/l3fwd_em_hlm.h +++ b/examples/l3fwd/l3fwd_em_hlm.h @@ -52,7 +52,7 @@ static __rte_always_inline void em_get_dst_port_ipv4xN(struct lcore_conf *qconf, struct rte_mbuf *m[], - uint8_t portid, uint16_t dst_port[]) + uint16_t portid, uint16_t dst_port[]) { int i; int32_t ret[EM_HASH_LOOKUP_COUNT]; @@ -68,7 +68,7 @@ em_get_dst_port_ipv4xN(struct lcore_conf *qconf, struct rte_mbuf *m[], EM_HASH_LOOKUP_COUNT, ret); for (i = 0; i < EM_HASH_LOOKUP_COUNT; i++) { - dst_port[i] = (uint8_t) ((ret[i] < 0) ? + dst_port[i] = ((ret[i] < 0) ? portid : ipv4_l3fwd_out_if[ret[i]]); if (dst_port[i] >= RTE_MAX_ETHPORTS || @@ -79,7 +79,7 @@ em_get_dst_port_ipv4xN(struct lcore_conf *qconf, struct rte_mbuf *m[], static __rte_always_inline void em_get_dst_port_ipv6xN(struct lcore_conf *qconf, struct rte_mbuf *m[], - uint8_t portid, uint16_t dst_port[]) + uint16_t portid, uint16_t dst_port[]) { int i; int32_t ret[EM_HASH_LOOKUP_COUNT]; @@ -95,7 +95,7 @@ em_get_dst_port_ipv6xN(struct lcore_conf *qconf, struct rte_mbuf *m[], EM_HASH_LOOKUP_COUNT, ret); for (i = 0; i < EM_HASH_LOOKUP_COUNT; i++) { - dst_port[i] = (uint8_t) ((ret[i] < 0) ? + dst_port[i] = ((ret[i] < 0) ? portid : ipv6_l3fwd_out_if[ret[i]]); if (dst_port[i] >= RTE_MAX_ETHPORTS || @@ -106,9 +106,9 @@ em_get_dst_port_ipv6xN(struct lcore_conf *qconf, struct rte_mbuf *m[], static __rte_always_inline uint16_t em_get_dst_port(const struct lcore_conf *qconf, struct rte_mbuf *pkt, - uint8_t portid) + uint16_t portid) { - uint8_t next_hop; + uint16_t next_hop; struct ipv4_hdr *ipv4_hdr; struct ipv6_hdr *ipv6_hdr; uint32_t tcp_or_udp; @@ -158,7 +158,7 @@ em_get_dst_port(const struct lcore_conf *qconf, struct rte_mbuf *pkt, */ static inline void l3fwd_em_send_packets(int nb_rx, struct rte_mbuf **pkts_burst, - uint8_t portid, struct lcore_conf *qconf) + uint16_t portid, struct lcore_conf *qconf) { int32_t i, j, pos; uint16_t dst_port[MAX_PKT_BURST]; diff --git a/examples/l3fwd/l3fwd_em_sequential.h b/examples/l3fwd/l3fwd_em_sequential.h index cb7c2abb..fa89f0f3 100644 --- a/examples/l3fwd/l3fwd_em_sequential.h +++ b/examples/l3fwd/l3fwd_em_sequential.h @@ -51,7 +51,7 @@ static __rte_always_inline uint16_t em_get_dst_port(const struct lcore_conf *qconf, struct rte_mbuf *pkt, - uint8_t portid) + uint16_t portid) { uint8_t next_hop; struct ipv4_hdr *ipv4_hdr; @@ -103,7 +103,7 @@ em_get_dst_port(const struct lcore_conf *qconf, struct rte_mbuf *pkt, */ static inline void l3fwd_em_send_packets(int nb_rx, struct rte_mbuf **pkts_burst, - uint8_t portid, struct lcore_conf *qconf) + uint16_t portid, struct lcore_conf *qconf) { int32_t i, j; uint16_t dst_port[MAX_PKT_BURST]; diff --git a/examples/l3fwd/l3fwd_lpm.c b/examples/l3fwd/l3fwd_lpm.c index ff1e4035..2d0e1724 100644 --- a/examples/l3fwd/l3fwd_lpm.c +++ b/examples/l3fwd/l3fwd_lpm.c @@ -105,7 +105,7 @@ struct rte_lpm *ipv4_l3fwd_lpm_lookup_struct[NB_SOCKETS]; struct rte_lpm6 *ipv6_l3fwd_lpm_lookup_struct[NB_SOCKETS]; static inline uint16_t -lpm_get_ipv4_dst_port(void *ipv4_hdr, uint8_t portid, void *lookup_struct) +lpm_get_ipv4_dst_port(void *ipv4_hdr, uint16_t portid, void *lookup_struct) { uint32_t next_hop; struct rte_lpm *ipv4_l3fwd_lookup_struct = @@ -117,7 +117,7 @@ lpm_get_ipv4_dst_port(void *ipv4_hdr, uint8_t portid, void *lookup_struct) } static inline uint16_t -lpm_get_ipv6_dst_port(void *ipv6_hdr, uint8_t portid, void *lookup_struct) +lpm_get_ipv6_dst_port(void *ipv6_hdr, uint16_t portid, void *lookup_struct) { uint32_t next_hop; struct rte_lpm6 *ipv6_l3fwd_lookup_struct = @@ -130,7 +130,7 @@ lpm_get_ipv6_dst_port(void *ipv6_hdr, uint8_t portid, void *lookup_struct) static __rte_always_inline uint16_t lpm_get_dst_port(const struct lcore_conf *qconf, struct rte_mbuf *pkt, - uint8_t portid) + uint16_t portid) { struct ipv6_hdr *ipv6_hdr; struct ipv4_hdr *ipv4_hdr; @@ -162,7 +162,7 @@ lpm_get_dst_port(const struct lcore_conf *qconf, struct rte_mbuf *pkt, */ static __rte_always_inline uint16_t lpm_get_dst_port_with_ipv4(const struct lcore_conf *qconf, struct rte_mbuf *pkt, - uint32_t dst_ipv4, uint8_t portid) + uint32_t dst_ipv4, uint16_t portid) { uint32_t next_hop; struct ipv6_hdr *ipv6_hdr; @@ -191,6 +191,8 @@ lpm_get_dst_port_with_ipv4(const struct lcore_conf *qconf, struct rte_mbuf *pkt, #include "l3fwd_lpm_sse.h" #elif defined RTE_MACHINE_CPUFLAG_NEON #include "l3fwd_lpm_neon.h" +#elif defined(RTE_ARCH_PPC_64) +#include "l3fwd_lpm_altivec.h" #else #include "l3fwd_lpm.h" #endif @@ -203,7 +205,8 @@ lpm_main_loop(__attribute__((unused)) void *dummy) unsigned lcore_id; uint64_t prev_tsc, diff_tsc, cur_tsc; int i, nb_rx; - uint8_t portid, queueid; + uint16_t portid; + uint8_t queueid; struct lcore_conf *qconf; const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) / US_PER_S * BURST_TX_DRAIN_US; @@ -225,7 +228,7 @@ lpm_main_loop(__attribute__((unused)) void *dummy) portid = qconf->rx_queue_list[i].port_id; queueid = qconf->rx_queue_list[i].queue_id; RTE_LOG(INFO, L3FWD, - " -- lcoreid=%u portid=%hhu rxqueueid=%hhu\n", + " -- lcoreid=%u portid=%u rxqueueid=%hhu\n", lcore_id, portid, queueid); } @@ -263,7 +266,8 @@ lpm_main_loop(__attribute__((unused)) void *dummy) if (nb_rx == 0) continue; -#if defined RTE_ARCH_X86 || defined RTE_MACHINE_CPUFLAG_NEON +#if defined RTE_ARCH_X86 || defined RTE_MACHINE_CPUFLAG_NEON \ + || defined RTE_ARCH_PPC_64 l3fwd_lpm_send_packets(nb_rx, pkts_burst, portid, qconf); #else @@ -413,7 +417,7 @@ lpm_parse_ptype(struct rte_mbuf *m) } uint16_t -lpm_cb_parse_ptype(uint8_t port __rte_unused, uint16_t queue __rte_unused, +lpm_cb_parse_ptype(uint16_t port __rte_unused, uint16_t queue __rte_unused, struct rte_mbuf *pkts[], uint16_t nb_pkts, uint16_t max_pkts __rte_unused, void *user_param __rte_unused) diff --git a/examples/l3fwd/l3fwd_lpm.h b/examples/l3fwd/l3fwd_lpm.h index 55c3e832..53b7fc80 100644 --- a/examples/l3fwd/l3fwd_lpm.h +++ b/examples/l3fwd/l3fwd_lpm.h @@ -35,7 +35,7 @@ #define __L3FWD_LPM_H__ static __rte_always_inline void -l3fwd_lpm_simple_forward(struct rte_mbuf *m, uint8_t portid, +l3fwd_lpm_simple_forward(struct rte_mbuf *m, uint16_t portid, struct lcore_conf *qconf) { struct ether_hdr *eth_hdr; @@ -104,7 +104,7 @@ l3fwd_lpm_simple_forward(struct rte_mbuf *m, uint8_t portid, static inline void l3fwd_lpm_no_opt_send_packets(int nb_rx, struct rte_mbuf **pkts_burst, - uint8_t portid, struct lcore_conf *qconf) + uint16_t portid, struct lcore_conf *qconf) { int32_t j; diff --git a/examples/l3fwd/l3fwd_lpm_altivec.h b/examples/l3fwd/l3fwd_lpm_altivec.h new file mode 100644 index 00000000..36ca983f --- /dev/null +++ b/examples/l3fwd/l3fwd_lpm_altivec.h @@ -0,0 +1,164 @@ +/*- + * BSD LICENSE + * + * Copyright(c) 2010-2016 Intel Corporation. All rights reserved. + * Copyright(c) 2017 IBM Corporation. + * All rights reserved. + * + * 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. + * * Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef __L3FWD_LPM_ALTIVEC_H__ +#define __L3FWD_LPM_ALTIVEC_H__ + +#include "l3fwd_altivec.h" + +/* + * Read packet_type and destination IPV4 addresses from 4 mbufs. + */ +static inline void +processx4_step1(struct rte_mbuf *pkt[FWDSTEP], + vector unsigned int *dip, + uint32_t *ipv4_flag) +{ + struct ipv4_hdr *ipv4_hdr; + struct ether_hdr *eth_hdr; + uint32_t x0, x1, x2, x3; + + eth_hdr = rte_pktmbuf_mtod(pkt[0], struct ether_hdr *); + ipv4_hdr = (struct ipv4_hdr *)(eth_hdr + 1); + x0 = ipv4_hdr->dst_addr; + ipv4_flag[0] = pkt[0]->packet_type & RTE_PTYPE_L3_IPV4; + + rte_compiler_barrier(); + eth_hdr = rte_pktmbuf_mtod(pkt[1], struct ether_hdr *); + ipv4_hdr = (struct ipv4_hdr *)(eth_hdr + 1); + x1 = ipv4_hdr->dst_addr; + ipv4_flag[0] &= pkt[1]->packet_type; + + rte_compiler_barrier(); + eth_hdr = rte_pktmbuf_mtod(pkt[2], struct ether_hdr *); + ipv4_hdr = (struct ipv4_hdr *)(eth_hdr + 1); + x2 = ipv4_hdr->dst_addr; + ipv4_flag[0] &= pkt[2]->packet_type; + + rte_compiler_barrier(); + eth_hdr = rte_pktmbuf_mtod(pkt[3], struct ether_hdr *); + ipv4_hdr = (struct ipv4_hdr *)(eth_hdr + 1); + x3 = ipv4_hdr->dst_addr; + ipv4_flag[0] &= pkt[3]->packet_type; + + rte_compiler_barrier(); + dip[0] = (vector unsigned int){x0, x1, x2, x3}; +} + +/* + * Lookup into LPM for destination port. + * If lookup fails, use incoming port (portid) as destination port. + */ +static inline void +processx4_step2(const struct lcore_conf *qconf, + vector unsigned int dip, + uint32_t ipv4_flag, + uint8_t portid, + struct rte_mbuf *pkt[FWDSTEP], + uint16_t dprt[FWDSTEP]) +{ + rte_xmm_t dst; + const vector unsigned char bswap_mask = (vector unsigned char){ + 3, 2, 1, 0, + 7, 6, 5, 4, + 11, 10, 9, 8, + 15, 14, 13, 12}; + + /* Byte swap 4 IPV4 addresses. */ + dip = (vector unsigned int)vec_perm(*(vector unsigned char *)&dip, + (vector unsigned char){}, bswap_mask); + + /* if all 4 packets are IPV4. */ + if (likely(ipv4_flag)) { + rte_lpm_lookupx4(qconf->ipv4_lookup_struct, (xmm_t)dip, + (uint32_t *)&dst, portid); + /* get rid of unused upper 16 bit for each dport. */ + dst.x = (xmm_t)vec_packs(dst.x, dst.x); + *(uint64_t *)dprt = dst.u64[0]; + } else { + dst.x = (xmm_t)dip; + dprt[0] = lpm_get_dst_port_with_ipv4(qconf, pkt[0], + dst.u32[0], portid); + dprt[1] = lpm_get_dst_port_with_ipv4(qconf, pkt[1], + dst.u32[1], portid); + dprt[2] = lpm_get_dst_port_with_ipv4(qconf, pkt[2], + dst.u32[2], portid); + dprt[3] = lpm_get_dst_port_with_ipv4(qconf, pkt[3], + dst.u32[3], portid); + } +} + +/* + * Buffer optimized handling of packets, invoked + * from main_loop. + */ +static inline void +l3fwd_lpm_send_packets(int nb_rx, struct rte_mbuf **pkts_burst, + uint8_t portid, struct lcore_conf *qconf) +{ + int32_t j; + uint16_t dst_port[MAX_PKT_BURST]; + vector unsigned int dip[MAX_PKT_BURST / FWDSTEP]; + uint32_t ipv4_flag[MAX_PKT_BURST / FWDSTEP]; + const int32_t k = RTE_ALIGN_FLOOR(nb_rx, FWDSTEP); + + for (j = 0; j != k; j += FWDSTEP) + processx4_step1(&pkts_burst[j], &dip[j / FWDSTEP], + &ipv4_flag[j / FWDSTEP]); + + for (j = 0; j != k; j += FWDSTEP) + processx4_step2(qconf, dip[j / FWDSTEP], + ipv4_flag[j / FWDSTEP], + portid, &pkts_burst[j], &dst_port[j]); + + /* Classify last up to 3 packets one by one */ + switch (nb_rx % FWDSTEP) { + case 3: + dst_port[j] = lpm_get_dst_port(qconf, pkts_burst[j], portid); + j++; + /* fall-through */ + case 2: + dst_port[j] = lpm_get_dst_port(qconf, pkts_burst[j], portid); + j++; + /* fall-through */ + case 1: + dst_port[j] = lpm_get_dst_port(qconf, pkts_burst[j], portid); + j++; + /* fall-through */ + } + + send_packets_multi(qconf, pkts_burst, dst_port, nb_rx); +} + +#endif /* __L3FWD_LPM_ALTIVEC_H__ */ diff --git a/examples/l3fwd/l3fwd_lpm_neon.h b/examples/l3fwd/l3fwd_lpm_neon.h index baedbfe8..85f314d1 100644 --- a/examples/l3fwd/l3fwd_lpm_neon.h +++ b/examples/l3fwd/l3fwd_lpm_neon.h @@ -82,7 +82,7 @@ static inline void processx4_step2(const struct lcore_conf *qconf, int32x4_t dip, uint32_t ipv4_flag, - uint8_t portid, + uint16_t portid, struct rte_mbuf *pkt[FWDSTEP], uint16_t dprt[FWDSTEP]) { @@ -115,7 +115,7 @@ processx4_step2(const struct lcore_conf *qconf, */ static inline void l3fwd_lpm_send_packets(int nb_rx, struct rte_mbuf **pkts_burst, - uint8_t portid, struct lcore_conf *qconf) + uint16_t portid, struct lcore_conf *qconf) { int32_t i = 0, j = 0; uint16_t dst_port[MAX_PKT_BURST]; diff --git a/examples/l3fwd/l3fwd_lpm_sse.h b/examples/l3fwd/l3fwd_lpm_sse.h index 4e294c84..d474396e 100644 --- a/examples/l3fwd/l3fwd_lpm_sse.h +++ b/examples/l3fwd/l3fwd_lpm_sse.h @@ -79,7 +79,7 @@ static inline void processx4_step2(const struct lcore_conf *qconf, __m128i dip, uint32_t ipv4_flag, - uint8_t portid, + uint16_t portid, struct rte_mbuf *pkt[FWDSTEP], uint16_t dprt[FWDSTEP]) { @@ -112,7 +112,7 @@ processx4_step2(const struct lcore_conf *qconf, */ static inline void l3fwd_lpm_send_packets(int nb_rx, struct rte_mbuf **pkts_burst, - uint8_t portid, struct lcore_conf *qconf) + uint16_t portid, struct lcore_conf *qconf) { int32_t j; uint16_t dst_port[MAX_PKT_BURST]; diff --git a/examples/l3fwd/l3fwd_neon.h b/examples/l3fwd/l3fwd_neon.h index 42d50d3c..b319b5a9 100644 --- a/examples/l3fwd/l3fwd_neon.h +++ b/examples/l3fwd/l3fwd_neon.h @@ -114,6 +114,7 @@ port_groupx4(uint16_t pn[FWDSTEP + 1], uint16_t *lp, uint16x8_t dp1, /* update last port counter. */ lp[0] += gptbl[v].lpv; + rte_compiler_barrier(); /* if dest port value has changed. */ if (v != GRPMSK) { @@ -192,7 +193,7 @@ send_packets_multi(struct lcore_conf *qconf, struct rte_mbuf **pkts_burst, * dp1: * <d[j], d[j+1], d[j+2], d[j+3], ... > */ - dp1 = vextq_u16(dp1, dp1, FWDSTEP - 1); + dp1 = vextq_u16(dp2, dp1, FWDSTEP - 1); } /* diff --git a/examples/l3fwd/main.c b/examples/l3fwd/main.c index 81995fdb..6229568f 100644 --- a/examples/l3fwd/main.c +++ b/examples/l3fwd/main.c @@ -50,7 +50,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -60,7 +59,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -124,7 +122,7 @@ uint32_t hash_entry_number = HASH_ENTRY_NUMBER_DEFAULT; struct lcore_conf lcore_conf[RTE_MAX_LCORE]; struct lcore_params { - uint8_t port_id; + uint16_t port_id; uint8_t queue_id; uint8_t lcore_id; } __rte_cache_aligned; @@ -245,7 +243,7 @@ check_lcore_params(void) static int check_port_config(const unsigned nb_ports) { - unsigned portid; + uint16_t portid; uint16_t i; for (i = 0; i < nb_lcore_params; ++i) { @@ -263,7 +261,7 @@ check_port_config(const unsigned nb_ports) } static uint8_t -get_port_n_rx_queues(const uint8_t port) +get_port_n_rx_queues(const uint16_t port) { int queue = -1; uint16_t i; @@ -445,7 +443,7 @@ parse_config(const char *q_arg) static void parse_eth_dest(const char *optarg) { - uint8_t portid; + uint16_t portid; char *port_end; uint8_t c, *dest, peer_addr[6]; @@ -750,11 +748,12 @@ init_mem(unsigned nb_mbuf) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -773,14 +772,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up. Speed %u Mbps -%s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -818,7 +816,7 @@ signal_handler(int signum) } static int -prepare_ptype_parser(uint8_t portid, uint16_t queueid) +prepare_ptype_parser(uint16_t portid, uint16_t queueid) { if (parse_ptype) { printf("Port %d: softly parse packet type info\n", portid); @@ -847,10 +845,10 @@ main(int argc, char **argv) struct rte_eth_txconf *txconf; int ret; unsigned nb_ports; - uint16_t queueid; + uint16_t queueid, portid; unsigned lcore_id; uint32_t n_tx_queue, nb_lcores; - uint8_t portid, nb_rx_queue, queue, socketid; + uint8_t nb_rx_queue, queue, socketid; /* init EAL */ ret = rte_eal_init(argc, argv); @@ -1048,7 +1046,7 @@ main(int argc, char **argv) } - check_all_ports_link_status((uint8_t)nb_ports, enabled_port_mask); + check_all_ports_link_status(nb_ports, enabled_port_mask); ret = 0; /* launch per-lcore init on every lcore */ diff --git a/examples/link_status_interrupt/main.c b/examples/link_status_interrupt/main.c index f4e3969a..bc47dcce 100644 --- a/examples/link_status_interrupt/main.c +++ b/examples/link_status_interrupt/main.c @@ -50,7 +50,6 @@ #include <rte_malloc.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -60,7 +59,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -145,7 +143,7 @@ print_stats(void) { struct rte_eth_link link; uint64_t total_packets_dropped, total_packets_tx, total_packets_rx; - unsigned portid; + uint16_t portid; total_packets_dropped = 0; total_packets_tx = 0; @@ -165,7 +163,7 @@ print_stats(void) continue; memset(&link, 0, sizeof(link)); - rte_eth_link_get_nowait((uint8_t)portid, &link); + rte_eth_link_get_nowait(portid, &link); printf("\nStatistics for port %u ------------------------------" "\nLink status: %25s" "\nLink speed: %26u" @@ -470,7 +468,7 @@ lsi_parse_args(int argc, char **argv) * int. */ static int -lsi_event_callback(uint8_t port_id, enum rte_eth_event_type type, void *param, +lsi_event_callback(uint16_t port_id, enum rte_eth_event_type type, void *param, void *ret_param) { struct rte_eth_link link; @@ -494,11 +492,12 @@ lsi_event_callback(uint8_t port_id, enum rte_eth_event_type type, void *param, /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint8_t count, all_ports_up, print_flag = 0; + uint16_t portid; struct rte_eth_link link; printf("\nChecking link status"); @@ -513,14 +512,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up. Speed %u Mbps - %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -553,8 +551,8 @@ main(int argc, char **argv) struct lcore_queue_conf *qconf; struct rte_eth_dev_info dev_info; int ret; - uint8_t nb_ports; - uint8_t portid, portid_last = 0; + uint16_t nb_ports; + uint16_t portid, portid_last = 0; unsigned lcore_id, rx_lcore_id; unsigned nb_ports_in_mask = 0; diff --git a/examples/load_balancer/config.c b/examples/load_balancer/config.c index 50325095..755a86e4 100644 --- a/examples/load_balancer/config.c +++ b/examples/load_balancer/config.c @@ -47,7 +47,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -57,7 +56,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -240,7 +238,7 @@ parse_arg_rx(const char *arg) if (lp->io.rx.n_nic_queues >= APP_MAX_NIC_RX_QUEUES_PER_IO_LCORE) { return -9; } - lp->io.rx.nic_queues[lp->io.rx.n_nic_queues].port = (uint8_t) port; + lp->io.rx.nic_queues[lp->io.rx.n_nic_queues].port = port; lp->io.rx.nic_queues[lp->io.rx.n_nic_queues].queue = (uint8_t) queue; lp->io.rx.n_nic_queues ++; @@ -318,7 +316,7 @@ parse_arg_tx(const char *arg) if (lp->io.tx.n_nic_ports >= APP_MAX_NIC_TX_PORTS_PER_IO_LCORE) { return -9; } - lp->io.tx.nic_ports[lp->io.tx.n_nic_ports] = (uint8_t) port; + lp->io.tx.nic_ports[lp->io.tx.n_nic_ports] = port; lp->io.tx.n_nic_ports ++; n_tuples ++; @@ -488,7 +486,7 @@ app_check_lpm_table(void) static int app_check_every_rx_port_is_tx_enabled(void) { - uint8_t port; + uint16_t port; for (port = 0; port < APP_MAX_NIC_PORTS; port ++) { if ((app_get_nic_rx_queues_per_port(port) > 0) && (app.nic_tx_port_mask[port] == 0)) { @@ -762,7 +760,7 @@ app_parse_args(int argc, char **argv) } int -app_get_nic_rx_queues_per_port(uint8_t port) +app_get_nic_rx_queues_per_port(uint16_t port) { uint32_t i, count; @@ -781,7 +779,7 @@ app_get_nic_rx_queues_per_port(uint8_t port) } int -app_get_lcore_for_nic_rx(uint8_t port, uint8_t queue, uint32_t *lcore_out) +app_get_lcore_for_nic_rx(uint16_t port, uint8_t queue, uint32_t *lcore_out) { uint32_t lcore; @@ -808,7 +806,7 @@ app_get_lcore_for_nic_rx(uint8_t port, uint8_t queue, uint32_t *lcore_out) } int -app_get_lcore_for_nic_tx(uint8_t port, uint32_t *lcore_out) +app_get_lcore_for_nic_tx(uint16_t port, uint32_t *lcore_out) { uint32_t lcore; @@ -901,7 +899,7 @@ app_print_params(void) /* Print NIC RX configuration */ printf("NIC RX ports: "); for (port = 0; port < APP_MAX_NIC_PORTS; port ++) { - uint32_t n_rx_queues = app_get_nic_rx_queues_per_port((uint8_t) port); + uint32_t n_rx_queues = app_get_nic_rx_queues_per_port(port); if (n_rx_queues == 0) { continue; diff --git a/examples/load_balancer/init.c b/examples/load_balancer/init.c index 717232e6..3dab7f25 100644 --- a/examples/load_balancer/init.c +++ b/examples/load_balancer/init.c @@ -47,7 +47,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -57,7 +56,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -308,7 +306,7 @@ app_init_rings_tx(void) continue; } - if (app_get_lcore_for_nic_tx((uint8_t) port, &lcore_io) < 0) { + if (app_get_lcore_for_nic_tx(port, &lcore_io) < 0) { rte_panic("Algorithmic error (no I/O core to handle TX of port %u)\n", port); } @@ -359,11 +357,12 @@ app_init_rings_tx(void) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; uint32_t n_rx_queues, n_tx_queues; @@ -383,14 +382,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up - speed %uMbps - %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -422,7 +420,8 @@ app_init_nics(void) { unsigned socket; uint32_t lcore; - uint8_t port, queue; + uint16_t port; + uint8_t queue; int ret; uint32_t n_rx_queues, n_tx_queues; @@ -440,14 +439,14 @@ app_init_nics(void) } /* Init port */ - printf("Initializing NIC port %u ...\n", (unsigned) port); + printf("Initializing NIC port %u ...\n", port); ret = rte_eth_dev_configure( port, (uint8_t) n_rx_queues, (uint8_t) n_tx_queues, &port_conf); if (ret < 0) { - rte_panic("Cannot init NIC port %u (%d)\n", (unsigned) port, ret); + rte_panic("Cannot init NIC port %u (%d)\n", port, ret); } rte_eth_promiscuous_enable(port); @@ -457,7 +456,7 @@ app_init_nics(void) port, &nic_rx_ring_size, &nic_tx_ring_size); if (ret < 0) { rte_panic("Cannot adjust number of descriptors for port %u (%d)\n", - (unsigned) port, ret); + port, ret); } app.nic_rx_ring_size = nic_rx_ring_size; app.nic_tx_ring_size = nic_tx_ring_size; @@ -473,8 +472,7 @@ app_init_nics(void) pool = app.lcore_params[lcore].pool; printf("Initializing NIC port %u RX queue %u ...\n", - (unsigned) port, - (unsigned) queue); + port, queue); ret = rte_eth_rx_queue_setup( port, queue, @@ -484,9 +482,7 @@ app_init_nics(void) pool); if (ret < 0) { rte_panic("Cannot init RX queue %u for port %u (%d)\n", - (unsigned) queue, - (unsigned) port, - ret); + queue, port, ret); } } @@ -495,7 +491,7 @@ app_init_nics(void) app_get_lcore_for_nic_tx(port, &lcore); socket = rte_lcore_to_socket_id(lcore); printf("Initializing NIC port %u TX queue 0 ...\n", - (unsigned) port); + port); ret = rte_eth_tx_queue_setup( port, 0, diff --git a/examples/load_balancer/main.c b/examples/load_balancer/main.c index 65ceea4a..7ced84d4 100644 --- a/examples/load_balancer/main.c +++ b/examples/load_balancer/main.c @@ -48,7 +48,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -58,7 +57,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> diff --git a/examples/load_balancer/main.h b/examples/load_balancer/main.h index dc407555..bef2ba04 100644 --- a/examples/load_balancer/main.h +++ b/examples/load_balancer/main.h @@ -248,7 +248,7 @@ struct app_lcore_params_io { struct { /* NIC */ struct { - uint8_t port; + uint16_t port; uint8_t queue; } nic_queues[APP_MAX_NIC_RX_QUEUES_PER_IO_LCORE]; uint32_t n_nic_queues; @@ -275,7 +275,7 @@ struct app_lcore_params_io { struct rte_ring *rings[APP_MAX_NIC_PORTS][APP_MAX_WORKER_LCORES]; /* NIC */ - uint8_t nic_ports[APP_MAX_NIC_TX_PORTS_PER_IO_LCORE]; + uint16_t nic_ports[APP_MAX_NIC_TX_PORTS_PER_IO_LCORE]; uint32_t n_nic_ports; /* Internal buffers */ @@ -368,9 +368,10 @@ void app_print_usage(void); void app_init(void); int app_lcore_main_loop(void *arg); -int app_get_nic_rx_queues_per_port(uint8_t port); -int app_get_lcore_for_nic_rx(uint8_t port, uint8_t queue, uint32_t *lcore_out); -int app_get_lcore_for_nic_tx(uint8_t port, uint32_t *lcore_out); +int app_get_nic_rx_queues_per_port(uint16_t port); +int app_get_lcore_for_nic_rx(uint16_t port, uint8_t queue, + uint32_t *lcore_out); +int app_get_lcore_for_nic_tx(uint16_t port, uint32_t *lcore_out); int app_is_socket_used(uint32_t socket); uint32_t app_get_lcores_io_rx(void); uint32_t app_get_lcores_worker(void); diff --git a/examples/load_balancer/runtime.c b/examples/load_balancer/runtime.c index e54b7851..f65e14f0 100644 --- a/examples/load_balancer/runtime.c +++ b/examples/load_balancer/runtime.c @@ -47,7 +47,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -57,7 +56,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -188,7 +186,7 @@ app_lcore_io_rx( uint32_t i; for (i = 0; i < lp->rx.n_nic_queues; i ++) { - uint8_t port = lp->rx.nic_queues[i].port; + uint16_t port = lp->rx.nic_queues[i].port; uint8_t queue = lp->rx.nic_queues[i].queue; uint32_t n_mbufs, j; @@ -213,7 +211,7 @@ app_lcore_io_rx( printf("I/O RX %u in (NIC port %u): NIC drop ratio = %.2f avg burst size = %.2f\n", lcore, - (unsigned) port, + port, (double) stats.imissed / (double) (stats.imissed + stats.ipackets), ((double) lp->rx.nic_queues_count[i]) / ((double) lp->rx.nic_queues_iters[i])); lp->rx.nic_queues_iters[i] = 0; @@ -339,7 +337,7 @@ app_lcore_io_tx( uint32_t i; for (i = 0; i < lp->tx.n_nic_ports; i ++) { - uint8_t port = lp->tx.nic_ports[i]; + uint16_t port = lp->tx.nic_ports[i]; struct rte_ring *ring = lp->tx.rings[port][worker]; uint32_t n_mbufs, n_pkts; int ret; @@ -395,7 +393,7 @@ app_lcore_io_tx( printf("\t\t\tI/O TX %u out (port %u): avg burst size = %.2f\n", lcore, - (unsigned) port, + port, ((double) lp->tx.nic_ports_count[port]) / ((double) lp->tx.nic_ports_iters[port])); lp->tx.nic_ports_iters[port] = 0; lp->tx.nic_ports_count[port] = 0; @@ -418,7 +416,7 @@ app_lcore_io_tx( static inline void app_lcore_io_tx_flush(struct app_lcore_params_io *lp) { - uint8_t port; + uint16_t port; uint32_t i; for (i = 0; i < lp->tx.n_nic_ports; i++) { @@ -569,7 +567,7 @@ app_lcore_worker( if (lp->rings_out_iters[port] == APP_STATS){ printf("\t\tWorker %u out (NIC port %u): enq success rate = %.2f\n", (unsigned) lp->worker_id, - (unsigned) port, + port, ((double) lp->rings_out_count[port]) / ((double) lp->rings_out_iters[port])); lp->rings_out_iters[port] = 0; lp->rings_out_count[port] = 0; diff --git a/examples/multi_process/client_server_mp/mp_client/client.c b/examples/multi_process/client_server_mp/mp_client/client.c index f8453e57..30ce4b34 100644 --- a/examples/multi_process/client_server_mp/mp_client/client.c +++ b/examples/multi_process/client_server_mp/mp_client/client.c @@ -57,7 +57,6 @@ #include <rte_mempool.h> #include <rte_mbuf.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_ether.h> #include <rte_ethdev.h> #include <rte_string_fns.h> @@ -74,7 +73,7 @@ static uint8_t client_id = 0; #define MBQ_CAPACITY 32 /* maps input ports to output ports for packets */ -static uint8_t output_ports[RTE_MAX_ETHPORTS]; +static uint16_t output_ports[RTE_MAX_ETHPORTS]; /* buffers up a set of packet that are ready to send */ struct rte_eth_dev_tx_buffer *tx_buffer[RTE_MAX_ETHPORTS]; @@ -150,7 +149,7 @@ static void flush_tx_error_callback(struct rte_mbuf **unsent, uint16_t count, void *userdata) { int i; - uint8_t port_id = (uintptr_t)userdata; + uint16_t port_id = (uintptr_t)userdata; tx_stats->tx_drop[port_id] += count; @@ -161,7 +160,7 @@ flush_tx_error_callback(struct rte_mbuf **unsent, uint16_t count, } static void -configure_tx_buffer(uint8_t port_id, uint16_t size) +configure_tx_buffer(uint16_t port_id, uint16_t size) { int ret; @@ -171,15 +170,16 @@ configure_tx_buffer(uint8_t port_id, uint16_t size) rte_eth_dev_socket_id(port_id)); if (tx_buffer[port_id] == NULL) rte_exit(EXIT_FAILURE, "Cannot allocate buffer for tx on port %u\n", - (unsigned) port_id); + port_id); rte_eth_tx_buffer_init(tx_buffer[port_id], size); ret = rte_eth_tx_buffer_set_err_callback(tx_buffer[port_id], flush_tx_error_callback, (void *)(intptr_t)port_id); if (ret < 0) - rte_exit(EXIT_FAILURE, "Cannot set error callback for " - "tx buffer on port %u\n", (unsigned) port_id); + rte_exit(EXIT_FAILURE, + "Cannot set error callback for tx buffer on port %u\n", + port_id); } /* @@ -195,8 +195,8 @@ configure_output_ports(const struct port_info *ports) rte_exit(EXIT_FAILURE, "Too many ethernet ports. RTE_MAX_ETHPORTS = %u\n", (unsigned)RTE_MAX_ETHPORTS); for (i = 0; i < ports->num_ports - 1; i+=2){ - uint8_t p1 = ports->id[i]; - uint8_t p2 = ports->id[i+1]; + uint16_t p1 = ports->id[i]; + uint16_t p2 = ports->id[i+1]; output_ports[p1] = p2; output_ports[p2] = p1; @@ -215,8 +215,8 @@ static void handle_packet(struct rte_mbuf *buf) { int sent; - const uint8_t in_port = buf->port; - const uint8_t out_port = output_ports[in_port]; + const uint16_t in_port = buf->port; + const uint16_t out_port = output_ports[in_port]; struct rte_eth_dev_tx_buffer *buffer = tx_buffer[out_port]; sent = rte_eth_tx_buffer(out_port, client_id, buffer, buf); @@ -275,7 +275,7 @@ main(int argc, char *argv[]) for (;;) { uint16_t i, rx_pkts; - uint8_t port; + uint16_t port; rx_pkts = rte_ring_dequeue_burst(rx_ring, pkts, PKT_READ_SIZE, NULL); diff --git a/examples/multi_process/client_server_mp/mp_server/Makefile b/examples/multi_process/client_server_mp/mp_server/Makefile index 5552999b..160c17b6 100644 --- a/examples/multi_process/client_server_mp/mp_server/Makefile +++ b/examples/multi_process/client_server_mp/mp_server/Makefile @@ -49,7 +49,7 @@ APP = mp_server # all source are stored in SRCS-y SRCS-y := main.c init.c args.c -INC := $(wildcard *.h) +INC := $(sort $(wildcard *.h)) CFLAGS += $(WERROR_FLAGS) -O3 CFLAGS += -I$(SRCDIR)/../shared diff --git a/examples/multi_process/client_server_mp/mp_server/args.c b/examples/multi_process/client_server_mp/mp_server/args.c index bf8c666c..a65884fd 100644 --- a/examples/multi_process/client_server_mp/mp_server/args.c +++ b/examples/multi_process/client_server_mp/mp_server/args.c @@ -74,7 +74,7 @@ parse_portmask(uint8_t max_ports, const char *portmask) { char *end = NULL; unsigned long pm; - uint8_t count = 0; + uint16_t count = 0; if (portmask == NULL || *portmask == '\0') return -1; @@ -128,7 +128,7 @@ parse_num_clients(const char *clients) * on error. */ int -parse_app_args(uint8_t max_ports, int argc, char *argv[]) +parse_app_args(uint16_t max_ports, int argc, char *argv[]) { int option_index, opt; char **argvopt = argv; diff --git a/examples/multi_process/client_server_mp/mp_server/args.h b/examples/multi_process/client_server_mp/mp_server/args.h index 23af1bd3..33888b89 100644 --- a/examples/multi_process/client_server_mp/mp_server/args.h +++ b/examples/multi_process/client_server_mp/mp_server/args.h @@ -34,6 +34,6 @@ #ifndef _ARGS_H_ #define _ARGS_H_ -int parse_app_args(uint8_t max_ports, int argc, char *argv[]); +int parse_app_args(uint16_t max_ports, int argc, char *argv[]); #endif /* ifndef _ARGS_H_ */ diff --git a/examples/multi_process/client_server_mp/mp_server/init.c b/examples/multi_process/client_server_mp/mp_server/init.c index 0bc92921..c8d02113 100644 --- a/examples/multi_process/client_server_mp/mp_server/init.c +++ b/examples/multi_process/client_server_mp/mp_server/init.c @@ -56,7 +56,6 @@ #include <rte_memcpy.h> #include <rte_mbuf.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_ether.h> #include <rte_ethdev.h> #include <rte_malloc.h> @@ -114,7 +113,7 @@ init_mbuf_pools(void) * - start the port and report its status to stdout */ static int -init_port(uint8_t port_num) +init_port(uint16_t port_num) { /* for port configuration all features are off by default */ const struct rte_eth_conf port_conf = { @@ -129,7 +128,7 @@ init_port(uint8_t port_num) uint16_t q; int retval; - printf("Port %u init ... ", (unsigned)port_num); + printf("Port %u init ... ", port_num); fflush(stdout); /* Standard DPDK port initialisation - config port, then set up @@ -200,11 +199,12 @@ init_shm_rings(void) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -262,7 +262,7 @@ init(int argc, char *argv[]) { int retval; const struct rte_memzone *mz; - uint8_t i, total_ports; + uint16_t i, total_ports; /* init EAL, parsing EAL args */ retval = rte_eal_init(argc, argv); diff --git a/examples/multi_process/client_server_mp/mp_server/main.c b/examples/multi_process/client_server_mp/mp_server/main.c index 7055b543..6eda556a 100644 --- a/examples/multi_process/client_server_mp/mp_server/main.c +++ b/examples/multi_process/client_server_mp/mp_server/main.c @@ -44,7 +44,6 @@ #include <rte_common.h> #include <rte_memory.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_per_lcore.h> @@ -59,7 +58,6 @@ #include <rte_mbuf.h> #include <rte_ether.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_ethdev.h> #include <rte_byteorder.h> #include <rte_malloc.h> @@ -88,7 +86,7 @@ struct client_rx_buf { static struct client_rx_buf *cl_rx_buf; static const char * -get_printable_mac_addr(uint8_t port) +get_printable_mac_addr(uint16_t port) { static const char err_address[] = "00:00:00:00:00:00"; static char addresses[RTE_MAX_ETHPORTS][sizeof(err_address)]; diff --git a/examples/multi_process/client_server_mp/shared/common.h b/examples/multi_process/client_server_mp/shared/common.h index 631c4632..35a3b01d 100644 --- a/examples/multi_process/client_server_mp/shared/common.h +++ b/examples/multi_process/client_server_mp/shared/common.h @@ -57,8 +57,8 @@ struct tx_stats{ } __rte_cache_aligned; struct port_info { - uint8_t num_ports; - uint8_t id[RTE_MAX_ETHPORTS]; + uint16_t num_ports; + uint16_t id[RTE_MAX_ETHPORTS]; volatile struct rx_stats rx_stats; volatile struct tx_stats tx_stats[MAX_CLIENTS]; }; diff --git a/examples/multi_process/l2fwd_fork/flib.c b/examples/multi_process/l2fwd_fork/flib.c index c22e983b..a3f1d275 100644 --- a/examples/multi_process/l2fwd_fork/flib.c +++ b/examples/multi_process/l2fwd_fork/flib.c @@ -54,7 +54,6 @@ #include <rte_malloc.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -64,7 +63,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> diff --git a/examples/multi_process/l2fwd_fork/main.c b/examples/multi_process/l2fwd_fork/main.c index f8a626ba..deace273 100644 --- a/examples/multi_process/l2fwd_fork/main.c +++ b/examples/multi_process/l2fwd_fork/main.c @@ -51,7 +51,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -62,7 +61,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -140,7 +138,8 @@ struct lcore_resource_struct { /* ring[1] for slave send ack, master read */ struct rte_ring *ring[2]; int port_num; /* Total port numbers */ - uint8_t port[RTE_MAX_ETHPORTS]; /* Port id for that lcore to receive packets */ + /* Port id for that lcore to receive packets */ + uint16_t port[RTE_MAX_ETHPORTS]; }__attribute__((packed)) __rte_cache_aligned; static struct lcore_resource_struct lcore_resource[RTE_MAX_LCORE]; @@ -871,11 +870,12 @@ l2fwd_parse_args(int argc, char **argv) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -890,14 +890,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up- speed %u Mbps- %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -930,9 +929,9 @@ main(int argc, char **argv) struct lcore_queue_conf *qconf; struct rte_eth_dev_info dev_info; int ret; - uint8_t nb_ports; - uint8_t nb_ports_available; - uint8_t portid, last_port; + uint16_t nb_ports; + uint16_t nb_ports_available; + uint16_t portid, last_port; unsigned rx_lcore_id; unsigned nb_ports_in_mask = 0; unsigned i; @@ -1204,10 +1203,7 @@ main(int argc, char **argv) message_pool = rte_mempool_create("ms_msg_pool", NB_CORE_MSGBUF * RTE_MAX_LCORE, sizeof(enum l2fwd_cmd), NB_CORE_MSGBUF / 2, - 0, - rte_pktmbuf_pool_init, NULL, - rte_pktmbuf_init, NULL, - rte_socket_id(), 0); + 0, NULL, NULL, NULL, NULL, rte_socket_id(), 0); if (message_pool == NULL) rte_exit(EXIT_FAILURE, "Create msg mempool failed\n"); diff --git a/examples/multi_process/simple_mp/main.c b/examples/multi_process/simple_mp/main.c index 2843d94e..62537b02 100644 --- a/examples/multi_process/simple_mp/main.c +++ b/examples/multi_process/simple_mp/main.c @@ -54,7 +54,6 @@ #include <rte_common.h> #include <rte_memory.h> -#include <rte_memzone.h> #include <rte_launch.h> #include <rte_eal.h> #include <rte_per_lcore.h> @@ -67,6 +66,7 @@ #include <rte_mempool.h> #include <cmdline_rdline.h> #include <cmdline_parse.h> +#include <cmdline_parse_string.h> #include <cmdline_socket.h> #include <cmdline.h> #include "mp_commands.h" @@ -76,7 +76,6 @@ static const char *_MSG_POOL = "MSG_POOL"; static const char *_SEC_2_PRI = "SEC_2_PRI"; static const char *_PRI_2_SEC = "PRI_2_SEC"; -const unsigned string_size = 64; struct rte_ring *send_ring, *recv_ring; struct rte_mempool *message_pool; @@ -121,7 +120,7 @@ main(int argc, char **argv) send_ring = rte_ring_create(_PRI_2_SEC, ring_size, rte_socket_id(), flags); recv_ring = rte_ring_create(_SEC_2_PRI, ring_size, rte_socket_id(), flags); message_pool = rte_mempool_create(_MSG_POOL, pool_size, - string_size, pool_cache, priv_data_sz, + STR_TOKEN_SIZE, pool_cache, priv_data_sz, NULL, NULL, NULL, NULL, rte_socket_id(), flags); } else { diff --git a/examples/multi_process/simple_mp/mp_commands.c b/examples/multi_process/simple_mp/mp_commands.c index 8da244bb..ef6dc58d 100644 --- a/examples/multi_process/simple_mp/mp_commands.c +++ b/examples/multi_process/simple_mp/mp_commands.c @@ -42,7 +42,6 @@ #include <rte_common.h> #include <rte_memory.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_atomic.h> #include <rte_branch_prediction.h> @@ -78,7 +77,7 @@ static void cmd_send_parsed(void *parsed_result, if (rte_mempool_get(message_pool, &msg) < 0) rte_panic("Failed to get message buffer\n"); - snprintf((char *)msg, string_size, "%s", res->message); + snprintf((char *)msg, STR_TOKEN_SIZE, "%s", res->message); if (rte_ring_enqueue(send_ring, msg) < 0) { printf("Failed to send message - message discarded\n"); rte_mempool_put(message_pool, msg); diff --git a/examples/multi_process/simple_mp/mp_commands.h b/examples/multi_process/simple_mp/mp_commands.h index 7e9a4ab2..452b3645 100644 --- a/examples/multi_process/simple_mp/mp_commands.h +++ b/examples/multi_process/simple_mp/mp_commands.h @@ -34,7 +34,6 @@ #ifndef _SIMPLE_MP_COMMANDS_H_ #define _SIMPLE_MP_COMMANDS_H_ -extern const unsigned string_size; extern struct rte_ring *send_ring; extern struct rte_mempool *message_pool; extern volatile int quit; diff --git a/examples/multi_process/symmetric_mp/main.c b/examples/multi_process/symmetric_mp/main.c index 0f497910..6fb285c7 100644 --- a/examples/multi_process/symmetric_mp/main.c +++ b/examples/multi_process/symmetric_mp/main.c @@ -56,7 +56,6 @@ #include <rte_common.h> #include <rte_log.h> #include <rte_memory.h> -#include <rte_memzone.h> #include <rte_launch.h> #include <rte_eal.h> #include <rte_per_lcore.h> @@ -65,7 +64,6 @@ #include <rte_branch_prediction.h> #include <rte_debug.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_ether.h> #include <rte_ethdev.h> #include <rte_mempool.h> @@ -102,7 +100,7 @@ struct port_stats{ static int proc_id = -1; static unsigned num_procs = 0; -static uint8_t ports[RTE_MAX_ETHPORTS]; +static uint16_t ports[RTE_MAX_ETHPORTS]; static unsigned num_ports = 0; static struct lcore_ports lcore_ports[RTE_MAX_LCORE]; @@ -202,7 +200,8 @@ smp_parse_args(int argc, char **argv) * coming from the mbuf_pool passed as parameter */ static inline int -smp_port_init(uint8_t port, struct rte_mempool *mbuf_pool, uint16_t num_queues) +smp_port_init(uint16_t port, struct rte_mempool *mbuf_pool, + uint16_t num_queues) { struct rte_eth_conf port_conf = { .rxmode = { @@ -237,7 +236,7 @@ smp_port_init(uint8_t port, struct rte_mempool *mbuf_pool, uint16_t num_queues) if (port >= rte_eth_dev_count()) return -1; - printf("# Initialising port %u... ", (unsigned)port); + printf("# Initialising port %u... ", port); fflush(stdout); rte_eth_dev_info_get(port, &info); @@ -362,11 +361,12 @@ lcore_main(void *arg __rte_unused) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -381,14 +381,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up. Speed %u Mbps - %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ diff --git a/examples/netmap_compat/lib/compat_netmap.c b/examples/netmap_compat/lib/compat_netmap.c index af2d9f3f..12b3fcbe 100644 --- a/examples/netmap_compat/lib/compat_netmap.c +++ b/examples/netmap_compat/lib/compat_netmap.c @@ -47,7 +47,6 @@ #include <rte_log.h> #include <rte_malloc.h> #include <rte_mbuf.h> -#include <rte_memzone.h> #include <rte_spinlock.h> #include <rte_string_fns.h> @@ -129,7 +128,7 @@ static void netmap_unregif(uint32_t idx, uint32_t port); static int32_t -ifname_to_portid(const char *ifname, uint8_t *port) +ifname_to_portid(const char *ifname, uint16_t *port) { char *endptr; uint64_t portid; @@ -140,7 +139,7 @@ ifname_to_portid(const char *ifname, uint8_t *port) portid >= RTE_DIM(ports) || errno != 0) return -EINVAL; - *port = (uint8_t)portid; + *port = portid; return 0; } @@ -222,10 +221,10 @@ fd_release(int32_t fd) } static int -check_nmreq(struct nmreq *req, uint8_t *port) +check_nmreq(struct nmreq *req, uint16_t *port) { int32_t rc; - uint8_t portid; + uint16_t portid; if (req == NULL) return -EINVAL; @@ -242,7 +241,7 @@ check_nmreq(struct nmreq *req, uint8_t *port) } if (ports[portid].pool == NULL) { - RTE_LOG(ERR, USER1, "Misconfigured portid %hhu\n", portid); + RTE_LOG(ERR, USER1, "Misconfigured portid %u\n", portid); return -EINVAL; } @@ -262,7 +261,7 @@ check_nmreq(struct nmreq *req, uint8_t *port) static int ioctl_niocginfo(__rte_unused int fd, void * param) { - uint8_t portid; + uint16_t portid; struct nmreq *req; int32_t rc; @@ -283,7 +282,7 @@ ioctl_niocginfo(__rte_unused int fd, void * param) } static void -netmap_ring_setup(struct netmap_ring *ring, uint8_t port, uint32_t ringid, +netmap_ring_setup(struct netmap_ring *ring, uint16_t port, uint32_t ringid, uint32_t num_slots) { uint32_t j; @@ -305,7 +304,7 @@ netmap_ring_setup(struct netmap_ring *ring, uint8_t port, uint32_t ringid, } static int -netmap_regif(struct nmreq *req, uint32_t idx, uint8_t port) +netmap_regif(struct nmreq *req, uint32_t idx, uint16_t port) { struct netmap_if *nmif; struct netmap_ring *ring; @@ -313,7 +312,7 @@ netmap_regif(struct nmreq *req, uint32_t idx, uint8_t port) int32_t rc; if (ports[port].fd < RTE_DIM(fd_port)) { - RTE_LOG(ERR, USER1, "port %hhu already in use by fd: %u\n", + RTE_LOG(ERR, USER1, "port %u already in use by fd: %u\n", port, IDX_TO_FD(ports[port].fd)); return -EBUSY; } @@ -399,7 +398,7 @@ netmap_regif(struct nmreq *req, uint32_t idx, uint8_t port) static int ioctl_niocregif(int32_t fd, void * param) { - uint8_t portid; + uint16_t portid; int32_t rc; uint32_t idx; struct nmreq *req; @@ -422,7 +421,7 @@ netmap_unregif(uint32_t idx, uint32_t port) { fd_port[idx].port = FD_PORT_RSRV; ports[port].fd = UINT32_MAX; - rte_eth_dev_stop((uint8_t)port); + rte_eth_dev_stop(port); } /** @@ -460,7 +459,7 @@ ioctl_niocunregif(int fd) * packets as it can hold coming from its dpdk port. */ static inline int -rx_sync_ring(struct netmap_ring *ring, uint8_t port, uint16_t ring_number, +rx_sync_ring(struct netmap_ring *ring, uint16_t port, uint16_t ring_number, uint16_t max_burst) { int32_t i, n_rx; @@ -513,7 +512,7 @@ rx_sync_if(uint32_t port) for (i = 0; i < nifp->ni_rx_rings + 1; i++) { r = NETMAP_RXRING(nifp, i); - rx_sync_ring(r, (uint8_t)port, (uint16_t)i, burst); + rx_sync_ring(r, port, (uint16_t)i, burst); rc += r->avail; } @@ -542,7 +541,7 @@ ioctl_niocrxsync(int fd) * buffers into rte_mbufs and sending them out on the rings's dpdk port. */ static int -tx_sync_ring(struct netmap_ring *ring, uint8_t port, uint16_t ring_number, +tx_sync_ring(struct netmap_ring *ring, uint16_t port, uint16_t ring_number, struct rte_mempool *pool, uint16_t max_burst) { uint32_t i, n_tx; @@ -608,7 +607,7 @@ tx_sync_if(uint32_t port) for (i = 0; i < nifp->ni_tx_rings + 1; i++) { r = NETMAP_TXRING(nifp, i); - tx_sync_ring(r, (uint8_t)port, (uint16_t)i, mp, burst); + tx_sync_ring(r, port, (uint16_t)i, mp, burst); rc += r->avail; } @@ -686,7 +685,7 @@ rte_netmap_init(const struct rte_netmap_conf *conf) int -rte_netmap_init_port(uint8_t portid, const struct rte_netmap_port_conf *conf) +rte_netmap_init_port(uint16_t portid, const struct rte_netmap_port_conf *conf) { int32_t ret; uint16_t i; @@ -696,17 +695,17 @@ rte_netmap_init_port(uint8_t portid, const struct rte_netmap_port_conf *conf) portid >= RTE_DIM(ports) || conf->nr_tx_rings > netmap.conf.max_rings || conf->nr_rx_rings > netmap.conf.max_rings) { - RTE_LOG(ERR, USER1, "%s(%hhu): invalid parameters\n", + RTE_LOG(ERR, USER1, "%s(%u): invalid parameters\n", __func__, portid); return -EINVAL; } - rx_slots = (uint16_t)rte_align32pow2(conf->nr_rx_slots); - tx_slots = (uint16_t)rte_align32pow2(conf->nr_tx_slots); + rx_slots = (uint16_t)rte_align32pow2(conf->nr_rx_slots); + tx_slots = (uint16_t)rte_align32pow2(conf->nr_tx_slots); if (tx_slots > netmap.conf.max_slots || rx_slots > netmap.conf.max_slots) { - RTE_LOG(ERR, USER1, "%s(%hhu): invalid parameters\n", + RTE_LOG(ERR, USER1, "%s(%u): invalid parameters\n", __func__, portid); return -EINVAL; } @@ -715,15 +714,15 @@ rte_netmap_init_port(uint8_t portid, const struct rte_netmap_port_conf *conf) conf->nr_tx_rings, conf->eth_conf); if (ret < 0) { - RTE_LOG(ERR, USER1, "Couldn't configure port %hhu\n", portid); - return ret; + RTE_LOG(ERR, USER1, "Couldn't configure port %u\n", portid); + return ret; } ret = rte_eth_dev_adjust_nb_rx_tx_desc(portid, &rx_slots, &tx_slots); if (ret < 0) { RTE_LOG(ERR, USER1, - "Couldn't ot adjust number of descriptors for port %hhu\n", + "Couldn't ot adjust number of descriptors for port %u\n", portid); return ret; } @@ -734,8 +733,7 @@ rte_netmap_init_port(uint8_t portid, const struct rte_netmap_port_conf *conf) if (ret < 0) { RTE_LOG(ERR, USER1, - "Couldn't configure TX queue %"PRIu16" of " - "port %"PRIu8"\n", + "fail to configure TX queue %u of port %u\n", i, portid); return ret; } @@ -745,8 +743,7 @@ rte_netmap_init_port(uint8_t portid, const struct rte_netmap_port_conf *conf) if (ret < 0) { RTE_LOG(ERR, USER1, - "Couldn't configure RX queue %"PRIu16" of " - "port %"PRIu8"\n", + "fail to configure RX queue %u of port %u\n", i, portid); return ret; } diff --git a/examples/netmap_compat/lib/compat_netmap.h b/examples/netmap_compat/lib/compat_netmap.h index 3dc7a2f4..76b2d2b4 100644 --- a/examples/netmap_compat/lib/compat_netmap.h +++ b/examples/netmap_compat/lib/compat_netmap.h @@ -67,7 +67,7 @@ struct rte_netmap_port_conf { }; int rte_netmap_init(const struct rte_netmap_conf *conf); -int rte_netmap_init_port(uint8_t portid, +int rte_netmap_init_port(uint16_t portid, const struct rte_netmap_port_conf *conf); int rte_netmap_close(int fd); diff --git a/examples/packet_ordering/main.c b/examples/packet_ordering/main.c index b26c33df..3add7be4 100644 --- a/examples/packet_ordering/main.c +++ b/examples/packet_ordering/main.c @@ -269,21 +269,22 @@ configure_tx_buffers(struct rte_eth_dev_tx_buffer *tx_buffer[]) rte_eth_dev_socket_id(port_id)); if (tx_buffer[port_id] == NULL) rte_exit(EXIT_FAILURE, "Cannot allocate buffer for tx on port %u\n", - (unsigned) port_id); + port_id); rte_eth_tx_buffer_init(tx_buffer[port_id], MAX_PKTS_BURST); ret = rte_eth_tx_buffer_set_err_callback(tx_buffer[port_id], flush_tx_error_callback, NULL); if (ret < 0) - rte_exit(EXIT_FAILURE, "Cannot set error callback for " - "tx buffer on port %u\n", (unsigned) port_id); + rte_exit(EXIT_FAILURE, + "Cannot set error callback for tx buffer on port %u\n", + port_id); } return 0; } static inline int -configure_eth_port(uint8_t port_id) +configure_eth_port(uint16_t port_id) { struct ether_addr addr; const uint16_t rxRings = 1, txRings = 1; @@ -326,7 +327,7 @@ configure_eth_port(uint8_t port_id) rte_eth_macaddr_get(port_id, &addr); printf("Port %u MAC: %02"PRIx8" %02"PRIx8" %02"PRIx8 " %02"PRIx8" %02"PRIx8" %02"PRIx8"\n", - (unsigned)port_id, + port_id, addr.addr_bytes[0], addr.addr_bytes[1], addr.addr_bytes[2], addr.addr_bytes[3], addr.addr_bytes[4], addr.addr_bytes[5]); @@ -401,7 +402,7 @@ rx_thread(struct rte_ring *ring_out) uint32_t seqn = 0; uint16_t i, ret = 0; uint16_t nb_rx_pkts; - uint8_t port_id; + uint16_t port_id; struct rte_mbuf *pkts[MAX_PKTS_BURST]; RTE_LOG(INFO, REORDERAPP, "%s() started on lcore %u\n", __func__, @@ -632,8 +633,8 @@ main(int argc, char **argv) int ret; unsigned nb_ports; unsigned int lcore_id, last_lcore_id, master_lcore_id; - uint8_t port_id; - uint8_t nb_ports_available; + uint16_t port_id; + uint16_t nb_ports_available; struct worker_thread_args worker_args = {NULL, NULL}; struct send_thread_args send_args = {NULL, NULL}; struct rte_ring *rx_to_workers; @@ -687,7 +688,7 @@ main(int argc, char **argv) continue; } /* init port */ - printf("Initializing port %u... done\n", (unsigned) port_id); + printf("Initializing port %u... done\n", port_id); if (configure_eth_port(port_id) != 0) rte_exit(EXIT_FAILURE, "Cannot initialize port %"PRIu8"\n", diff --git a/examples/performance-thread/common/lthread.h b/examples/performance-thread/common/lthread.h index 5c2c1a5f..0cde5919 100644 --- a/examples/performance-thread/common/lthread.h +++ b/examples/performance-thread/common/lthread.h @@ -87,7 +87,7 @@ int _lthread_desched_sleep(struct lthread *lt); void _lthread_free(struct lthread *lt); -struct lthread_sched *_lthread_sched_get(int lcore_id); +struct lthread_sched *_lthread_sched_get(unsigned int lcore_id); struct lthread_stack *_stack_alloc(void); diff --git a/examples/performance-thread/common/lthread_diag.c b/examples/performance-thread/common/lthread_diag.c index bce1a0c3..b5007d77 100644 --- a/examples/performance-thread/common/lthread_diag.c +++ b/examples/performance-thread/common/lthread_diag.c @@ -296,8 +296,7 @@ _lthread_diag_default_cb(uint64_t time, struct lthread *lt, int diag_event, /* * plug in default diag callback with mask off */ -void _lthread_diag_ctor(void)__attribute__((constructor)); -void _lthread_diag_ctor(void) +RTE_INIT(_lthread_diag_ctor) { diag_cb = _lthread_diag_default_cb; diag_mask = 0; diff --git a/examples/performance-thread/common/lthread_sched.c b/examples/performance-thread/common/lthread_sched.c index 98291478..779aeb17 100644 --- a/examples/performance-thread/common/lthread_sched.c +++ b/examples/performance-thread/common/lthread_sched.c @@ -117,8 +117,7 @@ uint64_t diag_mask; /* constructor */ -void lthread_sched_ctor(void) __attribute__ ((constructor)); -void lthread_sched_ctor(void) +RTE_INIT(lthread_sched_ctor) { memset(schedcore, 0, sizeof(schedcore)); rte_atomic16_init(&num_schedulers); @@ -562,11 +561,14 @@ void lthread_run(void) * Return the scheduler for this lcore * */ -struct lthread_sched *_lthread_sched_get(int lcore_id) +struct lthread_sched *_lthread_sched_get(unsigned int lcore_id) { - if (lcore_id > LTHREAD_MAX_LCORES) - return NULL; - return schedcore[lcore_id]; + struct lthread_sched *res = NULL; + + if (lcore_id < LTHREAD_MAX_LCORES) + res = schedcore[lcore_id]; + + return res; } /* @@ -578,10 +580,9 @@ int lthread_set_affinity(unsigned lcoreid) struct lthread *lt = THIS_LTHREAD; struct lthread_sched *dest_sched; - if (unlikely(lcoreid > LTHREAD_MAX_LCORES)) + if (unlikely(lcoreid >= LTHREAD_MAX_LCORES)) return POSIX_ERRNO(EINVAL); - DIAG_EVENT(lt, LT_DIAG_LTHREAD_AFFINITY, lcoreid, 0); dest_sched = schedcore[lcoreid]; diff --git a/examples/performance-thread/common/lthread_tls.c b/examples/performance-thread/common/lthread_tls.c index 47505f2d..2259fad4 100644 --- a/examples/performance-thread/common/lthread_tls.c +++ b/examples/performance-thread/common/lthread_tls.c @@ -62,9 +62,7 @@ RTE_DEFINE_PER_LTHREAD(void *, dummy); static struct lthread_key key_table[LTHREAD_MAX_KEYS]; -void lthread_tls_ctor(void) __attribute__((constructor)); - -void lthread_tls_ctor(void) +RTE_INIT(thread_tls_ctor) { key_pool = NULL; key_pool_init = 0; @@ -198,11 +196,12 @@ void _lthread_tls_destroy(struct lthread *lt) void *lthread_getspecific(unsigned int k) { + void *res = NULL; - if (k > LTHREAD_MAX_KEYS) - return NULL; + if (k < LTHREAD_MAX_KEYS) + res = THIS_LTHREAD->tls->data[k]; - return THIS_LTHREAD->tls->data[k]; + return res; } /* @@ -212,7 +211,7 @@ void */ int lthread_setspecific(unsigned int k, const void *data) { - if (k > LTHREAD_MAX_KEYS) + if (k >= LTHREAD_MAX_KEYS) return POSIX_ERRNO(EINVAL); int n = THIS_LTHREAD->tls->nb_keys_inuse; diff --git a/examples/performance-thread/l3fwd-thread/main.c b/examples/performance-thread/l3fwd-thread/main.c index 7954b974..fa65234f 100644 --- a/examples/performance-thread/l3fwd-thread/main.c +++ b/examples/performance-thread/l3fwd-thread/main.c @@ -50,7 +50,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -60,7 +59,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -140,7 +138,7 @@ parse_ptype(struct rte_mbuf *m) } static uint16_t -cb_parse_ptype(__rte_unused uint8_t port, __rte_unused uint16_t queue, +cb_parse_ptype(__rte_unused uint16_t port, __rte_unused uint16_t queue, struct rte_mbuf *pkts[], uint16_t nb_pkts, __rte_unused uint16_t max_pkts, __rte_unused void *user_param) { @@ -277,7 +275,7 @@ struct mbuf_table { }; struct lcore_rx_queue { - uint8_t port_id; + uint16_t port_id; uint8_t queue_id; } __rte_cache_aligned; @@ -287,7 +285,7 @@ struct lcore_rx_queue { #define MAX_LCORE_PARAMS 1024 struct rx_thread_params { - uint8_t port_id; + uint16_t port_id; uint8_t queue_id; uint8_t lcore_id; uint8_t thread_id; @@ -648,7 +646,7 @@ struct thread_tx_conf tx_thread[MAX_TX_THREAD]; /* Send burst of packets on an output interface */ static inline int -send_burst(struct thread_tx_conf *qconf, uint16_t n, uint8_t port) +send_burst(struct thread_tx_conf *qconf, uint16_t n, uint16_t port) { struct rte_mbuf **m_table; int ret; @@ -669,7 +667,7 @@ send_burst(struct thread_tx_conf *qconf, uint16_t n, uint8_t port) /* Enqueue a single packet, and send burst if queue is filled */ static inline int -send_single_packet(struct rte_mbuf *m, uint8_t port) +send_single_packet(struct rte_mbuf *m, uint16_t port) { uint16_t len; struct thread_tx_conf *qconf; @@ -696,7 +694,7 @@ send_single_packet(struct rte_mbuf *m, uint8_t port) #if ((APP_LOOKUP_METHOD == APP_LOOKUP_LPM) && \ (ENABLE_MULTI_BUFFER_OPTIMIZE == 1)) static __rte_always_inline void -send_packetsx4(uint8_t port, +send_packetsx4(uint16_t port, struct rte_mbuf *m[], uint32_t num) { uint32_t len, j, n; @@ -832,8 +830,8 @@ is_valid_ipv4_pkt(struct ipv4_hdr *pkt, uint32_t link_len) static __m128i mask0; static __m128i mask1; static __m128i mask2; -static inline uint8_t -get_ipv4_dst_port(void *ipv4_hdr, uint8_t portid, +static inline uint16_t +get_ipv4_dst_port(void *ipv4_hdr, uint16_t portid, lookup_struct_t *ipv4_l3fwd_lookup_struct) { int ret = 0; @@ -846,11 +844,11 @@ get_ipv4_dst_port(void *ipv4_hdr, uint8_t portid, key.xmm = _mm_and_si128(data, mask0); /* Find destination port */ ret = rte_hash_lookup(ipv4_l3fwd_lookup_struct, (const void *)&key); - return (uint8_t)((ret < 0) ? portid : ipv4_l3fwd_out_if[ret]); + return ((ret < 0) ? portid : ipv4_l3fwd_out_if[ret]); } -static inline uint8_t -get_ipv6_dst_port(void *ipv6_hdr, uint8_t portid, +static inline uint16_t +get_ipv6_dst_port(void *ipv6_hdr, uint16_t portid, lookup_struct_t *ipv6_l3fwd_lookup_struct) { int ret = 0; @@ -873,36 +871,36 @@ get_ipv6_dst_port(void *ipv6_hdr, uint8_t portid, /* Find destination port */ ret = rte_hash_lookup(ipv6_l3fwd_lookup_struct, (const void *)&key); - return (uint8_t)((ret < 0) ? portid : ipv6_l3fwd_out_if[ret]); + return ((ret < 0) ? portid : ipv6_l3fwd_out_if[ret]); } #endif #if (APP_LOOKUP_METHOD == APP_LOOKUP_LPM) -static inline uint8_t -get_ipv4_dst_port(void *ipv4_hdr, uint8_t portid, +static inline uint16_t +get_ipv4_dst_port(void *ipv4_hdr, uint16_t portid, lookup_struct_t *ipv4_l3fwd_lookup_struct) { uint32_t next_hop; - return (uint8_t)((rte_lpm_lookup(ipv4_l3fwd_lookup_struct, + return ((rte_lpm_lookup(ipv4_l3fwd_lookup_struct, rte_be_to_cpu_32(((struct ipv4_hdr *)ipv4_hdr)->dst_addr), &next_hop) == 0) ? next_hop : portid); } -static inline uint8_t -get_ipv6_dst_port(void *ipv6_hdr, uint8_t portid, +static inline uint16_t +get_ipv6_dst_port(void *ipv6_hdr, uint16_t portid, lookup6_struct_t *ipv6_l3fwd_lookup_struct) { uint32_t next_hop; - return (uint8_t) ((rte_lpm6_lookup(ipv6_l3fwd_lookup_struct, + return ((rte_lpm6_lookup(ipv6_l3fwd_lookup_struct, ((struct ipv6_hdr *)ipv6_hdr)->dst_addr, &next_hop) == 0) ? next_hop : portid); } #endif -static inline void l3fwd_simple_forward(struct rte_mbuf *m, uint8_t portid) +static inline void l3fwd_simple_forward(struct rte_mbuf *m, uint16_t portid) __attribute__((unused)); #if ((APP_LOOKUP_METHOD == APP_LOOKUP_EXACT_MATCH) && \ @@ -919,11 +917,11 @@ static inline void l3fwd_simple_forward(struct rte_mbuf *m, uint8_t portid) #define EXCLUDE_8TH_PKT 0x7f static inline void -simple_ipv4_fwd_8pkts(struct rte_mbuf *m[8], uint8_t portid) +simple_ipv4_fwd_8pkts(struct rte_mbuf *m[8], uint16_t portid) { struct ether_hdr *eth_hdr[8]; struct ipv4_hdr *ipv4_hdr[8]; - uint8_t dst_port[8]; + uint16_t dst_port[8]; int32_t ret[8]; union ipv4_5tuple_host key[8]; __m128i data[8]; @@ -1042,14 +1040,14 @@ simple_ipv4_fwd_8pkts(struct rte_mbuf *m[8], uint8_t portid) rte_hash_lookup_bulk(RTE_PER_LCORE(lcore_conf)->ipv4_lookup_struct, &key_array[0], 8, ret); - dst_port[0] = (uint8_t) ((ret[0] < 0) ? portid : ipv4_l3fwd_out_if[ret[0]]); - dst_port[1] = (uint8_t) ((ret[1] < 0) ? portid : ipv4_l3fwd_out_if[ret[1]]); - dst_port[2] = (uint8_t) ((ret[2] < 0) ? portid : ipv4_l3fwd_out_if[ret[2]]); - dst_port[3] = (uint8_t) ((ret[3] < 0) ? portid : ipv4_l3fwd_out_if[ret[3]]); - dst_port[4] = (uint8_t) ((ret[4] < 0) ? portid : ipv4_l3fwd_out_if[ret[4]]); - dst_port[5] = (uint8_t) ((ret[5] < 0) ? portid : ipv4_l3fwd_out_if[ret[5]]); - dst_port[6] = (uint8_t) ((ret[6] < 0) ? portid : ipv4_l3fwd_out_if[ret[6]]); - dst_port[7] = (uint8_t) ((ret[7] < 0) ? portid : ipv4_l3fwd_out_if[ret[7]]); + dst_port[0] = ((ret[0] < 0) ? portid : ipv4_l3fwd_out_if[ret[0]]); + dst_port[1] = ((ret[1] < 0) ? portid : ipv4_l3fwd_out_if[ret[1]]); + dst_port[2] = ((ret[2] < 0) ? portid : ipv4_l3fwd_out_if[ret[2]]); + dst_port[3] = ((ret[3] < 0) ? portid : ipv4_l3fwd_out_if[ret[3]]); + dst_port[4] = ((ret[4] < 0) ? portid : ipv4_l3fwd_out_if[ret[4]]); + dst_port[5] = ((ret[5] < 0) ? portid : ipv4_l3fwd_out_if[ret[5]]); + dst_port[6] = ((ret[6] < 0) ? portid : ipv4_l3fwd_out_if[ret[6]]); + dst_port[7] = ((ret[7] < 0) ? portid : ipv4_l3fwd_out_if[ret[7]]); if (dst_port[0] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[0]) == 0) @@ -1146,10 +1144,10 @@ static inline void get_ipv6_5tuple(struct rte_mbuf *m0, __m128i mask0, } static inline void -simple_ipv6_fwd_8pkts(struct rte_mbuf *m[8], uint8_t portid) +simple_ipv6_fwd_8pkts(struct rte_mbuf *m[8], uint16_t portid) { int32_t ret[8]; - uint8_t dst_port[8]; + uint16_t dst_port[8]; struct ether_hdr *eth_hdr[8]; union ipv6_5tuple_host key[8]; @@ -1196,14 +1194,14 @@ simple_ipv6_fwd_8pkts(struct rte_mbuf *m[8], uint8_t portid) rte_hash_lookup_bulk(RTE_PER_LCORE(lcore_conf)->ipv6_lookup_struct, &key_array[0], 4, ret); - dst_port[0] = (uint8_t) ((ret[0] < 0) ? portid : ipv6_l3fwd_out_if[ret[0]]); - dst_port[1] = (uint8_t) ((ret[1] < 0) ? portid : ipv6_l3fwd_out_if[ret[1]]); - dst_port[2] = (uint8_t) ((ret[2] < 0) ? portid : ipv6_l3fwd_out_if[ret[2]]); - dst_port[3] = (uint8_t) ((ret[3] < 0) ? portid : ipv6_l3fwd_out_if[ret[3]]); - dst_port[4] = (uint8_t) ((ret[4] < 0) ? portid : ipv6_l3fwd_out_if[ret[4]]); - dst_port[5] = (uint8_t) ((ret[5] < 0) ? portid : ipv6_l3fwd_out_if[ret[5]]); - dst_port[6] = (uint8_t) ((ret[6] < 0) ? portid : ipv6_l3fwd_out_if[ret[6]]); - dst_port[7] = (uint8_t) ((ret[7] < 0) ? portid : ipv6_l3fwd_out_if[ret[7]]); + dst_port[0] = ((ret[0] < 0) ? portid : ipv6_l3fwd_out_if[ret[0]]); + dst_port[1] = ((ret[1] < 0) ? portid : ipv6_l3fwd_out_if[ret[1]]); + dst_port[2] = ((ret[2] < 0) ? portid : ipv6_l3fwd_out_if[ret[2]]); + dst_port[3] = ((ret[3] < 0) ? portid : ipv6_l3fwd_out_if[ret[3]]); + dst_port[4] = ((ret[4] < 0) ? portid : ipv6_l3fwd_out_if[ret[4]]); + dst_port[5] = ((ret[5] < 0) ? portid : ipv6_l3fwd_out_if[ret[5]]); + dst_port[6] = ((ret[6] < 0) ? portid : ipv6_l3fwd_out_if[ret[6]]); + dst_port[7] = ((ret[7] < 0) ? portid : ipv6_l3fwd_out_if[ret[7]]); if (dst_port[0] >= RTE_MAX_ETHPORTS || (enabled_port_mask & 1 << dst_port[0]) == 0) @@ -1250,24 +1248,24 @@ simple_ipv6_fwd_8pkts(struct rte_mbuf *m[8], uint8_t portid) ether_addr_copy(&ports_eth_addr[dst_port[6]], ð_hdr[6]->s_addr); ether_addr_copy(&ports_eth_addr[dst_port[7]], ð_hdr[7]->s_addr); - send_single_packet(m[0], (uint8_t)dst_port[0]); - send_single_packet(m[1], (uint8_t)dst_port[1]); - send_single_packet(m[2], (uint8_t)dst_port[2]); - send_single_packet(m[3], (uint8_t)dst_port[3]); - send_single_packet(m[4], (uint8_t)dst_port[4]); - send_single_packet(m[5], (uint8_t)dst_port[5]); - send_single_packet(m[6], (uint8_t)dst_port[6]); - send_single_packet(m[7], (uint8_t)dst_port[7]); + send_single_packet(m[0], dst_port[0]); + send_single_packet(m[1], dst_port[1]); + send_single_packet(m[2], dst_port[2]); + send_single_packet(m[3], dst_port[3]); + send_single_packet(m[4], dst_port[4]); + send_single_packet(m[5], dst_port[5]); + send_single_packet(m[6], dst_port[6]); + send_single_packet(m[7], dst_port[7]); } #endif /* APP_LOOKUP_METHOD */ static __rte_always_inline void -l3fwd_simple_forward(struct rte_mbuf *m, uint8_t portid) +l3fwd_simple_forward(struct rte_mbuf *m, uint16_t portid) { struct ether_hdr *eth_hdr; struct ipv4_hdr *ipv4_hdr; - uint8_t dst_port; + uint16_t dst_port; eth_hdr = rte_pktmbuf_mtod(m, struct ether_hdr *); @@ -1379,7 +1377,7 @@ rfc1812_process(struct ipv4_hdr *ipv4_hdr, uint16_t *dp, uint32_t ptype) (ENABLE_MULTI_BUFFER_OPTIMIZE == 1)) static __rte_always_inline uint16_t -get_dst_port(struct rte_mbuf *pkt, uint32_t dst_ipv4, uint8_t portid) +get_dst_port(struct rte_mbuf *pkt, uint32_t dst_ipv4, uint16_t portid) { uint32_t next_hop; struct ipv6_hdr *ipv6_hdr; @@ -1406,7 +1404,7 @@ get_dst_port(struct rte_mbuf *pkt, uint32_t dst_ipv4, uint8_t portid) } static inline void -process_packet(struct rte_mbuf *pkt, uint16_t *dst_port, uint8_t portid) +process_packet(struct rte_mbuf *pkt, uint16_t *dst_port, uint16_t portid) { struct ether_hdr *eth_hdr; struct ipv4_hdr *ipv4_hdr; @@ -1473,7 +1471,7 @@ processx4_step1(struct rte_mbuf *pkt[FWDSTEP], static inline void processx4_step2(__m128i dip, uint32_t ipv4_flag, - uint8_t portid, + uint16_t portid, struct rte_mbuf *pkt[FWDSTEP], uint16_t dprt[FWDSTEP]) { @@ -1716,7 +1714,8 @@ port_groupx4(uint16_t pn[FWDSTEP + 1], uint16_t *lp, __m128i dp1, __m128i dp2) static void process_burst(struct rte_mbuf *pkts_burst[MAX_PKT_BURST], int nb_rx, - uint8_t portid) { + uint16_t portid) +{ int j; @@ -2036,7 +2035,7 @@ static void lthread_tx_per_ring(void *dummy) { int nb_rx; - uint8_t portid; + uint16_t portid; struct rte_ring *ring; struct thread_tx_conf *tx_conf; struct rte_mbuf *pkts_burst[MAX_PKT_BURST]; @@ -2091,7 +2090,7 @@ lthread_tx(void *args) struct lthread *lt; unsigned lcore_id; - uint8_t portid; + uint16_t portid; struct thread_tx_conf *tx_conf; tx_conf = (struct thread_tx_conf *)args; @@ -2138,7 +2137,8 @@ lthread_rx(void *dummy) int ret; uint16_t nb_rx; int i; - uint8_t portid, queueid; + uint16_t portid; + uint8_t queueid; int worker_id; int len[RTE_MAX_LCORE] = { 0 }; int old_len, new_len; @@ -2164,7 +2164,8 @@ lthread_rx(void *dummy) portid = rx_conf->rx_queue_list[i].port_id; queueid = rx_conf->rx_queue_list[i].queue_id; - RTE_LOG(INFO, L3FWD, " -- lcoreid=%u portid=%hhu rxqueueid=%hhu\n", + RTE_LOG(INFO, L3FWD, + " -- lcoreid=%u portid=%u rxqueueid=%hhu\n", rte_lcore_id(), portid, queueid); } @@ -2323,7 +2324,7 @@ pthread_tx(void *dummy) struct rte_mbuf *pkts_burst[MAX_PKT_BURST]; uint64_t prev_tsc, diff_tsc, cur_tsc; int nb_rx; - uint8_t portid; + uint16_t portid; struct thread_tx_conf *tx_conf; const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) / @@ -2392,7 +2393,8 @@ pthread_rx(void *dummy) uint32_t n; uint32_t nb_rx; unsigned lcore_id; - uint8_t portid, queueid; + uint8_t queueid; + uint16_t portid; struct rte_mbuf *pkts_burst[MAX_PKT_BURST]; struct thread_rx_conf *rx_conf; @@ -2411,7 +2413,8 @@ pthread_rx(void *dummy) portid = rx_conf->rx_queue_list[i].port_id; queueid = rx_conf->rx_queue_list[i].queue_id; - RTE_LOG(INFO, L3FWD, " -- lcoreid=%u portid=%hhu rxqueueid=%hhu\n", + RTE_LOG(INFO, L3FWD, + " -- lcoreid=%u portid=%u rxqueueid=%hhu\n", lcore_id, portid, queueid); } @@ -2539,7 +2542,7 @@ check_port_config(const unsigned nb_ports) } static uint8_t -get_port_n_rx_queues(const uint8_t port) +get_port_n_rx_queues(const uint16_t port) { int queue = -1; uint16_t i; @@ -2769,7 +2772,7 @@ parse_rx_config(const char *q_arg) return -1; } rx_thread_params_array[nb_rx_thread_params].port_id = - (uint8_t)int_fld[FLD_PORT]; + int_fld[FLD_PORT]; rx_thread_params_array[nb_rx_thread_params].queue_id = (uint8_t)int_fld[FLD_QUEUE]; rx_thread_params_array[nb_rx_thread_params].lcore_id = @@ -2853,7 +2856,7 @@ parse_stat_lcore(const char *stat_lcore) static void parse_eth_dest(const char *optarg) { - uint8_t portid; + uint16_t portid; char *port_end; uint8_t c, *dest, peer_addr[6]; @@ -3436,11 +3439,12 @@ init_mem(unsigned nb_mbuf) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint16_t portid; + uint8_t count, all_ports_up, print_flag = 0; struct rte_eth_link link; printf("\nChecking link status"); @@ -3455,14 +3459,13 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", (uint8_t)portid, - (unsigned)link.link_speed, + printf( + "Port%d Link Up. Speed %u Mbps - %s\n", + portid, link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else - printf("Port %d Link Down\n", - (uint8_t)portid); + printf("Port %d Link Down\n", portid); continue; } /* clear all_ports_up flag if any link down */ @@ -3497,10 +3500,10 @@ main(int argc, char **argv) int ret; int i; unsigned nb_ports; - uint16_t queueid; + uint16_t queueid, portid; unsigned lcore_id; uint32_t n_tx_queue, nb_lcores; - uint8_t portid, nb_rx_queue, queue, socketid; + uint8_t nb_rx_queue, queue, socketid; /* init EAL */ ret = rte_eal_init(argc, argv); diff --git a/examples/performance-thread/pthread_shim/main.c b/examples/performance-thread/pthread_shim/main.c index 850b009d..febae39b 100644 --- a/examples/performance-thread/pthread_shim/main.c +++ b/examples/performance-thread/pthread_shim/main.c @@ -161,6 +161,7 @@ static void initial_lthread(void *args __attribute__((unused))) pthread_override_set(1); uint64_t i; + int ret; /* initialize mutex for shared counter */ print_count = 0; @@ -187,7 +188,10 @@ static void initial_lthread(void *args __attribute__((unused))) pthread_attr_setaffinity_np(&attr, sizeof(rte_cpuset_t), &cpuset); /* create the thread */ - pthread_create(&tid[i], &attr, helloworld_pthread, (void *) i); + ret = pthread_create(&tid[i], &attr, + helloworld_pthread, (void *) i); + if (ret != 0) + rte_exit(EXIT_FAILURE, "Cannot create helloworld thread\n"); } /* wait for 1s to allow threads diff --git a/examples/performance-thread/pthread_shim/pthread_shim.c b/examples/performance-thread/pthread_shim/pthread_shim.c index 113bafa0..bc7cf2b0 100644 --- a/examples/performance-thread/pthread_shim/pthread_shim.c +++ b/examples/performance-thread/pthread_shim/pthread_shim.c @@ -202,10 +202,7 @@ static void *__libc_dl_handle = RTLD_NEXT; * The constructor function initialises the * function pointers for pthread library functions */ -void -pthread_intercept_ctor(void)__attribute__((constructor)); -void -pthread_intercept_ctor(void) +RTE_INIT(pthread_intercept_ctor) { override = 0; /* diff --git a/examples/ptpclient/ptpclient.c b/examples/ptpclient/ptpclient.c index ddfcdb83..c53dad68 100644 --- a/examples/ptpclient/ptpclient.c +++ b/examples/ptpclient/ptpclient.c @@ -158,12 +158,12 @@ struct ptpv2_data_slave_ordinary { struct clock_id master_clock_id; struct timeval new_adj; int64_t delta; - uint8_t portid; + uint16_t portid; uint16_t seqID_SYNC; uint16_t seqID_FOLLOWUP; uint8_t ptpset; uint8_t kernel_time_set; - uint8_t current_ptp_port; + uint16_t current_ptp_port; }; static struct ptpv2_data_slave_ordinary ptp_data; @@ -202,7 +202,7 @@ ns_to_timeval(int64_t nsec) * coming from the mbuf_pool passed as a parameter. */ static inline int -port_init(uint8_t port, struct rte_mempool *mbuf_pool) +port_init(uint16_t port, struct rte_mempool *mbuf_pool) { struct rte_eth_dev_info dev_info; struct rte_eth_conf port_conf = port_conf_default; @@ -555,7 +555,7 @@ parse_drsp(struct ptpv2_data_slave_ordinary *ptp_data) * functionality. */ static void -parse_ptp_frames(uint8_t portid, struct rte_mbuf *m) { +parse_ptp_frames(uint16_t portid, struct rte_mbuf *m) { struct ptp_header *ptp_hdr; struct ether_hdr *eth_hdr; uint16_t eth_type; @@ -593,7 +593,7 @@ parse_ptp_frames(uint8_t portid, struct rte_mbuf *m) { static __attribute__((noreturn)) void lcore_main(void) { - uint8_t portid; + uint16_t portid; unsigned nb_rx; struct rte_mbuf *m; @@ -728,7 +728,7 @@ main(int argc, char *argv[]) { unsigned nb_ports; - uint8_t portid; + uint16_t portid; /* Initialize the Environment Abstraction Layer (EAL). */ int ret = rte_eal_init(argc, argv); diff --git a/examples/qos_meter/main.c b/examples/qos_meter/main.c index b0909f6a..67b4a75b 100644 --- a/examples/qos_meter/main.c +++ b/examples/qos_meter/main.c @@ -116,8 +116,8 @@ static struct rte_eth_conf port_conf = { #define PKT_TX_BURST_MAX 32 #define TIME_TX_DRAIN 200000ULL -static uint8_t port_rx; -static uint8_t port_tx; +static uint16_t port_rx; +static uint16_t port_tx; static struct rte_mbuf *pkts_rx[PKT_RX_BURST_MAX]; struct rte_eth_dev_tx_buffer *tx_buffer; diff --git a/examples/qos_sched/args.c b/examples/qos_sched/args.c index 2350d64f..203a3470 100644 --- a/examples/qos_sched/args.c +++ b/examples/qos_sched/args.c @@ -252,8 +252,8 @@ app_parse_flow_conf(const char *conf_str) pconf = &qos_conf[nb_pfc]; - pconf->rx_port = (uint8_t)vals[0]; - pconf->tx_port = (uint8_t)vals[1]; + pconf->rx_port = vals[0]; + pconf->tx_port = vals[1]; pconf->rx_core = (uint8_t)vals[2]; pconf->wt_core = (uint8_t)vals[3]; if (ret == 5) @@ -267,19 +267,19 @@ app_parse_flow_conf(const char *conf_str) } if (pconf->rx_port >= RTE_MAX_ETHPORTS) { - RTE_LOG(ERR, APP, "pfc %u: invalid rx port %"PRIu8" index\n", + RTE_LOG(ERR, APP, "pfc %u: invalid rx port %"PRIu16" index\n", nb_pfc, pconf->rx_port); return -1; } if (pconf->tx_port >= RTE_MAX_ETHPORTS) { - RTE_LOG(ERR, APP, "pfc %u: invalid tx port %"PRIu8" index\n", + RTE_LOG(ERR, APP, "pfc %u: invalid tx port %"PRIu16" index\n", nb_pfc, pconf->tx_port); return -1; } mask = 1lu << pconf->rx_port; if (app_used_rx_port_mask & mask) { - RTE_LOG(ERR, APP, "pfc %u: rx port %"PRIu8" is used already\n", + RTE_LOG(ERR, APP, "pfc %u: rx port %"PRIu16" is used already\n", nb_pfc, pconf->rx_port); return -1; } @@ -288,7 +288,7 @@ app_parse_flow_conf(const char *conf_str) mask = 1lu << pconf->tx_port; if (app_used_tx_port_mask & mask) { - RTE_LOG(ERR, APP, "pfc %u: port %"PRIu8" is used already\n", + RTE_LOG(ERR, APP, "pfc %u: port %"PRIu16" is used already\n", nb_pfc, pconf->tx_port); return -1; } diff --git a/examples/qos_sched/cmdline.c b/examples/qos_sched/cmdline.c index f79d5246..b62d165b 100644 --- a/examples/qos_sched/cmdline.c +++ b/examples/qos_sched/cmdline.c @@ -191,7 +191,7 @@ cmdline_parse_inst_t cmd_appstats = { struct cmd_subportstats_result { cmdline_fixed_string_t stats_string; cmdline_fixed_string_t port_string; - uint8_t port_number; + uint16_t port_number; cmdline_fixed_string_t subport_string; uint32_t subport_number; }; @@ -220,7 +220,7 @@ cmdline_parse_token_num_t cmd_subportstats_subport_number = UINT32); cmdline_parse_token_num_t cmd_subportstats_port_number = TOKEN_NUM_INITIALIZER(struct cmd_subportstats_result, port_number, - UINT8); + UINT16); cmdline_parse_inst_t cmd_subportstats = { .f = cmd_subportstats_parsed, @@ -240,7 +240,7 @@ cmdline_parse_inst_t cmd_subportstats = { struct cmd_pipestats_result { cmdline_fixed_string_t stats_string; cmdline_fixed_string_t port_string; - uint8_t port_number; + uint16_t port_number; cmdline_fixed_string_t subport_string; uint32_t subport_number; cmdline_fixed_string_t pipe_string; @@ -265,7 +265,7 @@ cmdline_parse_token_string_t cmd_pipestats_port_string = "port"); cmdline_parse_token_num_t cmd_pipestats_port_number = TOKEN_NUM_INITIALIZER(struct cmd_pipestats_result, port_number, - UINT8); + UINT16); cmdline_parse_token_string_t cmd_pipestats_subport_string = TOKEN_STRING_INITIALIZER(struct cmd_pipestats_result, subport_string, "subport"); @@ -299,7 +299,7 @@ cmdline_parse_inst_t cmd_pipestats = { struct cmd_avg_q_result { cmdline_fixed_string_t qavg_string; cmdline_fixed_string_t port_string; - uint8_t port_number; + uint16_t port_number; cmdline_fixed_string_t subport_string; uint32_t subport_number; cmdline_fixed_string_t pipe_string; @@ -327,8 +327,8 @@ cmdline_parse_token_string_t cmd_avg_q_port_string = TOKEN_STRING_INITIALIZER(struct cmd_avg_q_result, port_string, "port"); cmdline_parse_token_num_t cmd_avg_q_port_number = - TOKEN_NUM_INITIALIZER(struct cmd_avg_q_result, port_number, - UINT8); + TOKEN_NUM_INITIALIZER(struct cmd_avg_q_result, port_number, + UINT16); cmdline_parse_token_string_t cmd_avg_q_subport_string = TOKEN_STRING_INITIALIZER(struct cmd_avg_q_result, subport_string, "subport"); @@ -378,7 +378,7 @@ cmdline_parse_inst_t cmd_avg_q = { struct cmd_avg_tcpipe_result { cmdline_fixed_string_t qavg_string; cmdline_fixed_string_t port_string; - uint8_t port_number; + uint16_t port_number; cmdline_fixed_string_t subport_string; uint32_t subport_number; cmdline_fixed_string_t pipe_string; @@ -405,7 +405,7 @@ cmdline_parse_token_string_t cmd_avg_tcpipe_port_string = "port"); cmdline_parse_token_num_t cmd_avg_tcpipe_port_number = TOKEN_NUM_INITIALIZER(struct cmd_avg_tcpipe_result, port_number, - UINT8); + UINT16); cmdline_parse_token_string_t cmd_avg_tcpipe_subport_string = TOKEN_STRING_INITIALIZER(struct cmd_avg_tcpipe_result, subport_string, "subport"); @@ -447,7 +447,7 @@ cmdline_parse_inst_t cmd_avg_tcpipe = { struct cmd_avg_pipe_result { cmdline_fixed_string_t qavg_string; cmdline_fixed_string_t port_string; - uint8_t port_number; + uint16_t port_number; cmdline_fixed_string_t subport_string; uint32_t subport_number; cmdline_fixed_string_t pipe_string; @@ -472,7 +472,7 @@ cmdline_parse_token_string_t cmd_avg_pipe_port_string = "port"); cmdline_parse_token_num_t cmd_avg_pipe_port_number = TOKEN_NUM_INITIALIZER(struct cmd_avg_pipe_result, port_number, - UINT8); + UINT16); cmdline_parse_token_string_t cmd_avg_pipe_subport_string = TOKEN_STRING_INITIALIZER(struct cmd_avg_pipe_result, subport_string, "subport"); @@ -506,7 +506,7 @@ cmdline_parse_inst_t cmd_avg_pipe = { struct cmd_avg_tcsubport_result { cmdline_fixed_string_t qavg_string; cmdline_fixed_string_t port_string; - uint8_t port_number; + uint16_t port_number; cmdline_fixed_string_t subport_string; uint32_t subport_number; cmdline_fixed_string_t tc_string; @@ -531,7 +531,7 @@ cmdline_parse_token_string_t cmd_avg_tcsubport_port_string = "port"); cmdline_parse_token_num_t cmd_avg_tcsubport_port_number = TOKEN_NUM_INITIALIZER(struct cmd_avg_tcsubport_result, port_number, - UINT8); + UINT16); cmdline_parse_token_string_t cmd_avg_tcsubport_subport_string = TOKEN_STRING_INITIALIZER(struct cmd_avg_tcsubport_result, subport_string, "subport"); @@ -565,7 +565,7 @@ cmdline_parse_inst_t cmd_avg_tcsubport = { struct cmd_avg_subport_result { cmdline_fixed_string_t qavg_string; cmdline_fixed_string_t port_string; - uint8_t port_number; + uint16_t port_number; cmdline_fixed_string_t subport_string; uint32_t subport_number; }; @@ -588,7 +588,7 @@ cmdline_parse_token_string_t cmd_avg_subport_port_string = "port"); cmdline_parse_token_num_t cmd_avg_subport_port_number = TOKEN_NUM_INITIALIZER(struct cmd_avg_subport_result, port_number, - UINT8); + UINT16); cmdline_parse_token_string_t cmd_avg_subport_subport_string = TOKEN_STRING_INITIALIZER(struct cmd_avg_subport_result, subport_string, "subport"); diff --git a/examples/qos_sched/init.c b/examples/qos_sched/init.c index a82cbd7d..038f0427 100644 --- a/examples/qos_sched/init.c +++ b/examples/qos_sched/init.c @@ -100,7 +100,7 @@ static const struct rte_eth_conf port_conf = { }; static int -app_init_port(uint8_t portid, struct rte_mempool *mp) +app_init_port(uint16_t portid, struct rte_mempool *mp) { int ret; struct rte_eth_link link; @@ -118,6 +118,7 @@ app_init_port(uint8_t portid, struct rte_mempool *mp) rx_conf.rx_thresh.wthresh = rx_thresh.wthresh; rx_conf.rx_free_thresh = 32; rx_conf.rx_drop_en = 0; + rx_conf.rx_deferred_start = 0; tx_conf.tx_thresh.pthresh = tx_thresh.pthresh; tx_conf.tx_thresh.hthresh = tx_thresh.hthresh; @@ -125,21 +126,24 @@ app_init_port(uint8_t portid, struct rte_mempool *mp) tx_conf.tx_free_thresh = 0; tx_conf.tx_rs_thresh = 0; tx_conf.txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS | ETH_TXQ_FLAGS_NOOFFLOADS; + tx_conf.tx_deferred_start = 0; /* init port */ - RTE_LOG(INFO, APP, "Initializing port %"PRIu8"... ", portid); + RTE_LOG(INFO, APP, "Initializing port %"PRIu16"... ", portid); fflush(stdout); ret = rte_eth_dev_configure(portid, 1, 1, &port_conf); if (ret < 0) - rte_exit(EXIT_FAILURE, "Cannot configure device: " - "err=%d, port=%"PRIu8"\n", ret, portid); + rte_exit(EXIT_FAILURE, + "Cannot configure device: err=%d, port=%u\n", + ret, portid); rx_size = ring_conf.rx_size; tx_size = ring_conf.tx_size; ret = rte_eth_dev_adjust_nb_rx_tx_desc(portid, &rx_size, &tx_size); if (ret < 0) - rte_exit(EXIT_FAILURE, "rte_eth_dev_adjust_nb_rx_tx_desc: " - "err=%d, port=%"PRIu8"\n", ret, portid); + rte_exit(EXIT_FAILURE, + "rte_eth_dev_adjust_nb_rx_tx_desc: err=%d,port=%u\n", + ret, portid); ring_conf.rx_size = rx_size; ring_conf.tx_size = tx_size; @@ -148,22 +152,25 @@ app_init_port(uint8_t portid, struct rte_mempool *mp) ret = rte_eth_rx_queue_setup(portid, 0, (uint16_t)ring_conf.rx_size, rte_eth_dev_socket_id(portid), &rx_conf, mp); if (ret < 0) - rte_exit(EXIT_FAILURE, "rte_eth_tx_queue_setup: " - "err=%d, port=%"PRIu8"\n", ret, portid); + rte_exit(EXIT_FAILURE, + "rte_eth_tx_queue_setup: err=%d, port=%u\n", + ret, portid); /* init one TX queue */ fflush(stdout); ret = rte_eth_tx_queue_setup(portid, 0, (uint16_t)ring_conf.tx_size, rte_eth_dev_socket_id(portid), &tx_conf); if (ret < 0) - rte_exit(EXIT_FAILURE, "rte_eth_tx_queue_setup: err=%d, " - "port=%"PRIu8" queue=%d\n", ret, portid, 0); + rte_exit(EXIT_FAILURE, + "rte_eth_tx_queue_setup: err=%d, port=%u queue=%d\n", + ret, portid, 0); /* Start device */ ret = rte_eth_dev_start(portid); if (ret < 0) - rte_exit(EXIT_FAILURE, "rte_pmd_port_start: " - "err=%d, port=%"PRIu8"\n", ret, portid); + rte_exit(EXIT_FAILURE, + "rte_pmd_port_start: err=%d, port=%u\n", + ret, portid); printf("done: "); @@ -256,7 +263,7 @@ app_init_sched_port(uint32_t portid, uint32_t socketid) uint32_t pipe, subport; int err; - rte_eth_link_get((uint8_t)portid, &link); + rte_eth_link_get(portid, &link); port_params.socket = socketid; port_params.rate = (uint64_t) link.link_speed * 1000 * 1000 / 8; diff --git a/examples/qos_sched/main.c b/examples/qos_sched/main.c index e10cfd44..1e2fb189 100644 --- a/examples/qos_sched/main.c +++ b/examples/qos_sched/main.c @@ -125,8 +125,7 @@ app_main_loop(__attribute__((unused))void *dummy) /* initialize mbuf memory */ if (mode == APP_RX_MODE) { for (i = 0; i < rx_idx; i++) { - RTE_LOG(INFO, APP, "flow %u lcoreid %u " - "reading port %"PRIu8"\n", + RTE_LOG(INFO, APP, "flow%u lcoreid%u reading port%u\n", i, lcore_id, rx_confs[i]->rx_port); } @@ -140,8 +139,8 @@ app_main_loop(__attribute__((unused))void *dummy) if (wt_confs[i]->m_table == NULL) rte_panic("flow %u unable to allocate memory buffer\n", i); - RTE_LOG(INFO, APP, "flow %u lcoreid %u sched+write " - "port %"PRIu8"\n", + RTE_LOG(INFO, APP, + "flow %u lcoreid %u sched+write port %u\n", i, lcore_id, wt_confs[i]->tx_port); } @@ -155,8 +154,7 @@ app_main_loop(__attribute__((unused))void *dummy) if (tx_confs[i]->m_table == NULL) rte_panic("flow %u unable to allocate memory buffer\n", i); - RTE_LOG(INFO, APP, "flow %u lcoreid %u " - "writing port %"PRIu8"\n", + RTE_LOG(INFO, APP, "flow%u lcoreid%u write port%u\n", i, lcore_id, tx_confs[i]->tx_port); } @@ -186,7 +184,7 @@ app_stat(void) struct flow_conf *flow = &qos_conf[i]; rte_eth_stats_get(flow->rx_port, &stats); - printf("\nRX port %"PRIu8": rx: %"PRIu64 " err: %"PRIu64 + printf("\nRX port %"PRIu16": rx: %"PRIu64 " err: %"PRIu64 " no_mbuf: %"PRIu64 "\n", flow->rx_port, stats.ipackets - rx_stats[i].ipackets, @@ -195,7 +193,7 @@ app_stat(void) memcpy(&rx_stats[i], &stats, sizeof(stats)); rte_eth_stats_get(flow->tx_port, &stats); - printf("TX port %"PRIu8": tx: %" PRIu64 " err: %" PRIu64 "\n", + printf("TX port %"PRIu16": tx: %" PRIu64 " err: %" PRIu64 "\n", flow->tx_port, stats.opackets - tx_stats[i].opackets, stats.oerrors - tx_stats[i].oerrors); diff --git a/examples/qos_sched/main.h b/examples/qos_sched/main.h index 8d02e1ad..77b6e3ee 100644 --- a/examples/qos_sched/main.h +++ b/examples/qos_sched/main.h @@ -106,8 +106,8 @@ struct thread_conf uint32_t n_mbufs; struct rte_mbuf **m_table; - uint8_t rx_port; - uint8_t tx_port; + uint16_t rx_port; + uint16_t tx_port; uint16_t rx_queue; uint16_t tx_queue; struct rte_ring *rx_ring; @@ -125,8 +125,8 @@ struct flow_conf uint32_t rx_core; uint32_t wt_core; uint32_t tx_core; - uint8_t rx_port; - uint8_t tx_port; + uint16_t rx_port; + uint16_t tx_port; uint16_t rx_queue; uint16_t tx_queue; struct rte_ring *rx_ring; @@ -188,13 +188,15 @@ void app_worker_thread(struct thread_conf **qconf); void app_mixed_thread(struct thread_conf **qconf); void app_stat(void); -int subport_stat(uint8_t port_id, uint32_t subport_id); -int pipe_stat(uint8_t port_id, uint32_t subport_id, uint32_t pipe_id); -int qavg_q(uint8_t port_id, uint32_t subport_id, uint32_t pipe_id, uint8_t tc, uint8_t q); -int qavg_tcpipe(uint8_t port_id, uint32_t subport_id, uint32_t pipe_id, uint8_t tc); -int qavg_pipe(uint8_t port_id, uint32_t subport_id, uint32_t pipe_id); -int qavg_tcsubport(uint8_t port_id, uint32_t subport_id, uint8_t tc); -int qavg_subport(uint8_t port_id, uint32_t subport_id); +int subport_stat(uint16_t port_id, uint32_t subport_id); +int pipe_stat(uint16_t port_id, uint32_t subport_id, uint32_t pipe_id); +int qavg_q(uint16_t port_id, uint32_t subport_id, uint32_t pipe_id, + uint8_t tc, uint8_t q); +int qavg_tcpipe(uint16_t port_id, uint32_t subport_id, uint32_t pipe_id, + uint8_t tc); +int qavg_pipe(uint16_t port_id, uint32_t subport_id, uint32_t pipe_id); +int qavg_tcsubport(uint16_t port_id, uint32_t subport_id, uint8_t tc); +int qavg_subport(uint16_t port_id, uint32_t subport_id); #ifdef __cplusplus } diff --git a/examples/qos_sched/stats.c b/examples/qos_sched/stats.c index 5c894455..b5545e10 100644 --- a/examples/qos_sched/stats.c +++ b/examples/qos_sched/stats.c @@ -37,7 +37,8 @@ #include "main.h" int -qavg_q(uint8_t port_id, uint32_t subport_id, uint32_t pipe_id, uint8_t tc, uint8_t q) +qavg_q(uint16_t port_id, uint32_t subport_id, uint32_t pipe_id, uint8_t tc, + uint8_t q) { struct rte_sched_queue_stats stats; struct rte_sched_port *port; @@ -74,7 +75,8 @@ qavg_q(uint8_t port_id, uint32_t subport_id, uint32_t pipe_id, uint8_t tc, uint8 } int -qavg_tcpipe(uint8_t port_id, uint32_t subport_id, uint32_t pipe_id, uint8_t tc) +qavg_tcpipe(uint16_t port_id, uint32_t subport_id, uint32_t pipe_id, + uint8_t tc) { struct rte_sched_queue_stats stats; struct rte_sched_port *port; @@ -114,7 +116,7 @@ qavg_tcpipe(uint8_t port_id, uint32_t subport_id, uint32_t pipe_id, uint8_t tc) } int -qavg_pipe(uint8_t port_id, uint32_t subport_id, uint32_t pipe_id) +qavg_pipe(uint16_t port_id, uint32_t subport_id, uint32_t pipe_id) { struct rte_sched_queue_stats stats; struct rte_sched_port *port; @@ -153,7 +155,7 @@ qavg_pipe(uint8_t port_id, uint32_t subport_id, uint32_t pipe_id) } int -qavg_tcsubport(uint8_t port_id, uint32_t subport_id, uint8_t tc) +qavg_tcsubport(uint16_t port_id, uint32_t subport_id, uint8_t tc) { struct rte_sched_queue_stats stats; struct rte_sched_port *port; @@ -195,7 +197,7 @@ qavg_tcsubport(uint8_t port_id, uint32_t subport_id, uint8_t tc) } int -qavg_subport(uint8_t port_id, uint32_t subport_id) +qavg_subport(uint16_t port_id, uint32_t subport_id) { struct rte_sched_queue_stats stats; struct rte_sched_port *port; @@ -237,7 +239,7 @@ qavg_subport(uint8_t port_id, uint32_t subport_id) } int -subport_stat(uint8_t port_id, uint32_t subport_id) +subport_stat(uint16_t port_id, uint32_t subport_id) { struct rte_sched_subport_stats stats; struct rte_sched_port *port; @@ -273,7 +275,7 @@ subport_stat(uint8_t port_id, uint32_t subport_id) } int -pipe_stat(uint8_t port_id, uint32_t subport_id, uint32_t pipe_id) +pipe_stat(uint16_t port_id, uint32_t subport_id, uint32_t pipe_id) { struct rte_sched_queue_stats stats; struct rte_sched_port *port; diff --git a/examples/quota_watermark/qw/init.c b/examples/quota_watermark/qw/init.c index 083a37a9..37b03626 100644 --- a/examples/quota_watermark/qw/init.c +++ b/examples/quota_watermark/qw/init.c @@ -73,7 +73,7 @@ static struct rte_eth_fc_conf fc_conf = { }; -void configure_eth_port(uint8_t port_id) +void configure_eth_port(uint16_t port_id) { int ret; uint16_t nb_rxd = RX_DESC_PER_QUEUE; @@ -135,7 +135,7 @@ init_dpdk(void) rte_exit(EXIT_FAILURE, "Not enough ethernet port available\n"); } -void init_ring(int lcore_id, uint8_t port_id) +void init_ring(int lcore_id, uint16_t port_id) { struct rte_ring *ring; char ring_name[RTE_RING_NAMESIZE]; @@ -156,12 +156,12 @@ void init_ring(int lcore_id, uint8_t port_id) void pair_ports(void) { - uint8_t i, j; + uint16_t i, j; /* Pair ports with their "closest neighbour" in the portmask */ for (i = 0; i < RTE_MAX_ETHPORTS; i++) if (is_bit_set(i, portmask)) - for (j = (uint8_t) (i + 1); j < RTE_MAX_ETHPORTS; j++) + for (j = i + 1; j < RTE_MAX_ETHPORTS; j++) if (is_bit_set(j, portmask)) { port_pairs[i] = j; port_pairs[j] = i; diff --git a/examples/quota_watermark/qw/init.h b/examples/quota_watermark/qw/init.h index 6d0af3ab..2bfec2b1 100644 --- a/examples/quota_watermark/qw/init.h +++ b/examples/quota_watermark/qw/init.h @@ -34,9 +34,9 @@ #ifndef _INIT_H_ #define _INIT_H_ -void configure_eth_port(uint8_t port_id); +void configure_eth_port(uint16_t port_id); void init_dpdk(void); -void init_ring(int lcore_id, uint8_t port_id); +void init_ring(int lcore_id, uint16_t port_id); void pair_ports(void); void setup_shared_variables(void); diff --git a/examples/quota_watermark/qw/main.c b/examples/quota_watermark/qw/main.c index d4fcfde4..fe174526 100644 --- a/examples/quota_watermark/qw/main.c +++ b/examples/quota_watermark/qw/main.c @@ -69,13 +69,13 @@ int *quota; unsigned int *low_watermark; unsigned int *high_watermark; -uint8_t port_pairs[RTE_MAX_ETHPORTS]; +uint16_t port_pairs[RTE_MAX_ETHPORTS]; struct rte_ring *rings[RTE_MAX_LCORE][RTE_MAX_ETHPORTS]; struct rte_mempool *mbuf_pool; -static void send_pause_frame(uint8_t port_id, uint16_t duration) +static void send_pause_frame(uint16_t port_id, uint16_t duration) { struct rte_mbuf *mbuf; struct ether_fc_frame *pause_frame; @@ -155,7 +155,7 @@ receive_stage(__attribute__((unused)) void *args) { int i, ret; - uint8_t port_id; + uint16_t port_id; uint16_t nb_rx_pkts; unsigned int lcore_id; @@ -216,7 +216,7 @@ pipeline_stage(__attribute__((unused)) void *args) int i, ret; int nb_dq_pkts; - uint8_t port_id; + uint16_t port_id; unsigned int lcore_id, previous_lcore_id; unsigned int free; @@ -279,8 +279,8 @@ send_stage(__attribute__((unused)) void *args) { uint16_t nb_dq_pkts; - uint8_t port_id; - uint8_t dest_port_id; + uint16_t port_id; + uint16_t dest_port_id; unsigned int lcore_id, previous_lcore_id; @@ -324,7 +324,7 @@ main(int argc, char **argv) int ret; unsigned int lcore_id, master_lcore_id, last_lcore_id; - uint8_t port_id; + uint16_t port_id; rte_log_set_global_level(RTE_LOG_INFO); diff --git a/examples/quota_watermark/qw/main.h b/examples/quota_watermark/qw/main.h index 8c8e3116..ebed7b2f 100644 --- a/examples/quota_watermark/qw/main.h +++ b/examples/quota_watermark/qw/main.h @@ -45,7 +45,7 @@ extern int *quota; extern unsigned int *low_watermark; extern unsigned int *high_watermark; -extern uint8_t port_pairs[RTE_MAX_ETHPORTS]; +extern uint16_t port_pairs[RTE_MAX_ETHPORTS]; extern struct rte_ring *rings[RTE_MAX_LCORE][RTE_MAX_ETHPORTS]; extern struct rte_mempool *mbuf_pool; diff --git a/examples/rxtx_callbacks/main.c b/examples/rxtx_callbacks/main.c index 66992405..ca135d21 100644 --- a/examples/rxtx_callbacks/main.c +++ b/examples/rxtx_callbacks/main.c @@ -59,7 +59,7 @@ static struct { static uint16_t -add_timestamps(uint8_t port __rte_unused, uint16_t qidx __rte_unused, +add_timestamps(uint16_t port __rte_unused, uint16_t qidx __rte_unused, struct rte_mbuf **pkts, uint16_t nb_pkts, uint16_t max_pkts __rte_unused, void *_ __rte_unused) { @@ -72,7 +72,7 @@ add_timestamps(uint8_t port __rte_unused, uint16_t qidx __rte_unused, } static uint16_t -calc_latency(uint8_t port __rte_unused, uint16_t qidx __rte_unused, +calc_latency(uint16_t port __rte_unused, uint16_t qidx __rte_unused, struct rte_mbuf **pkts, uint16_t nb_pkts, void *_ __rte_unused) { uint64_t cycles = 0; @@ -97,7 +97,7 @@ calc_latency(uint8_t port __rte_unused, uint16_t qidx __rte_unused, * coming from the mbuf_pool passed as parameter */ static inline int -port_init(uint8_t port, struct rte_mempool *mbuf_pool) +port_init(uint16_t port, struct rte_mempool *mbuf_pool) { struct rte_eth_conf port_conf = port_conf_default; const uint16_t rx_rings = 1, tx_rings = 1; @@ -159,7 +159,7 @@ port_init(uint8_t port, struct rte_mempool *mbuf_pool) static __attribute__((noreturn)) void lcore_main(void) { - uint8_t port; + uint16_t port; for (port = 0; port < nb_ports; port++) if (rte_eth_dev_socket_id(port) > 0 && @@ -195,7 +195,7 @@ int main(int argc, char *argv[]) { struct rte_mempool *mbuf_pool; - uint8_t portid; + uint16_t portid; /* init EAL */ int ret = rte_eal_init(argc, argv); diff --git a/examples/server_node_efd/node/node.c b/examples/server_node_efd/node/node.c index 86e57c89..5aa1258e 100644 --- a/examples/server_node_efd/node/node.c +++ b/examples/server_node_efd/node/node.c @@ -57,7 +57,6 @@ #include <rte_mempool.h> #include <rte_mbuf.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_ether.h> #include <rte_ethdev.h> #include <rte_string_fns.h> @@ -77,7 +76,7 @@ static uint8_t node_id; #define MBQ_CAPACITY 32 /* maps input ports to output ports for packets */ -static uint8_t output_ports[RTE_MAX_ETHPORTS]; +static uint16_t output_ports[RTE_MAX_ETHPORTS]; /* buffers up a set of packet that are ready to send */ struct rte_eth_dev_tx_buffer *tx_buffer[RTE_MAX_ETHPORTS]; @@ -154,7 +153,7 @@ static void flush_tx_error_callback(struct rte_mbuf **unsent, uint16_t count, void *userdata) { int i; - uint8_t port_id = (uintptr_t)userdata; + uint16_t port_id = (uintptr_t)userdata; tx_stats->tx_drop[port_id] += count; @@ -165,7 +164,7 @@ flush_tx_error_callback(struct rte_mbuf **unsent, uint16_t count, } static void -configure_tx_buffer(uint8_t port_id, uint16_t size) +configure_tx_buffer(uint16_t port_id, uint16_t size) { int ret; @@ -174,16 +173,17 @@ configure_tx_buffer(uint8_t port_id, uint16_t size) RTE_ETH_TX_BUFFER_SIZE(size), 0, rte_eth_dev_socket_id(port_id)); if (tx_buffer[port_id] == NULL) - rte_exit(EXIT_FAILURE, "Cannot allocate buffer for tx " - "on port %u\n", (unsigned int) port_id); + rte_exit(EXIT_FAILURE, + "Cannot allocate buffer for tx on port %u\n", port_id); rte_eth_tx_buffer_init(tx_buffer[port_id], size); ret = rte_eth_tx_buffer_set_err_callback(tx_buffer[port_id], flush_tx_error_callback, (void *)(intptr_t)port_id); if (ret < 0) - rte_exit(EXIT_FAILURE, "Cannot set error callback for " - "tx buffer on port %u\n", (unsigned int) port_id); + rte_exit(EXIT_FAILURE, + "Cannot set error callback for tx buffer on port %u\n", + port_id); } /* @@ -282,8 +282,8 @@ static inline void transmit_packet(struct rte_mbuf *buf) { int sent; - const uint8_t in_port = buf->port; - const uint8_t out_port = output_ports[in_port]; + const uint16_t in_port = buf->port; + const uint16_t out_port = output_ports[in_port]; struct rte_eth_dev_tx_buffer *buffer = tx_buffer[out_port]; sent = rte_eth_tx_buffer(out_port, node_id, buffer, buf); @@ -381,7 +381,7 @@ main(int argc, char *argv[]) for (;;) { uint16_t rx_pkts = PKT_READ_SIZE; - uint8_t port; + uint16_t port; /* * Try dequeuing max possible packets first, if that fails, diff --git a/examples/server_node_efd/server/Makefile b/examples/server_node_efd/server/Makefile index a2f2f361..9f1fe289 100644 --- a/examples/server_node_efd/server/Makefile +++ b/examples/server_node_efd/server/Makefile @@ -49,7 +49,7 @@ APP = server # all source are stored in SRCS-y SRCS-y := main.c init.c args.c -INC := $(wildcard *.h) +INC := $(sort $(wildcard *.h)) CFLAGS += $(WERROR_FLAGS) -O3 CFLAGS += -I$(SRCDIR)/../shared diff --git a/examples/server_node_efd/server/init.c b/examples/server_node_efd/server/init.c index d114e5bf..0bcab8cc 100644 --- a/examples/server_node_efd/server/init.c +++ b/examples/server_node_efd/server/init.c @@ -56,7 +56,6 @@ #include <rte_memcpy.h> #include <rte_mbuf.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_ether.h> #include <rte_ethdev.h> #include <rte_malloc.h> @@ -121,7 +120,7 @@ init_mbuf_pools(void) * - start the port and report its status to stdout */ static int -init_port(uint8_t port_num) +init_port(uint16_t port_num) { /* for port configuration all features are off by default */ const struct rte_eth_conf port_conf = { @@ -136,7 +135,7 @@ init_port(uint8_t port_num) uint16_t q; int retval; - printf("Port %u init ... ", (unsigned int)port_num); + printf("Port %u init ... ", port_num); fflush(stdout); /* @@ -255,11 +254,12 @@ populate_efd_table(void) /* Check the link status of all ports in up to 9s, and print them finally */ static void -check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) { #define CHECK_INTERVAL 100 /* 100ms */ #define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ - uint8_t portid, count, all_ports_up, print_flag = 0; + uint8_t count, all_ports_up, print_flag = 0; + uint16_t portid; struct rte_eth_link link; printf("\nChecking link status"); @@ -274,14 +274,15 @@ check_all_ports_link_status(uint8_t port_num, uint32_t port_mask) /* print link status if flag set */ if (print_flag == 1) { if (link.link_status) - printf("Port %d Link Up - speed %u " - "Mbps - %s\n", info->id[portid], - (unsigned int)link.link_speed, + printf( + "Port%d Link Up. Speed %u Mbps - %s\n", + info->id[portid], + link.link_speed, (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? ("full-duplex") : ("half-duplex\n")); else printf("Port %d Link Down\n", - (uint8_t)info->id[portid]); + info->id[portid]); continue; } /* clear all_ports_up flag if any link down */ diff --git a/examples/server_node_efd/server/main.c b/examples/server_node_efd/server/main.c index dcdc0a48..aa1c6f57 100644 --- a/examples/server_node_efd/server/main.c +++ b/examples/server_node_efd/server/main.c @@ -44,7 +44,6 @@ #include <rte_common.h> #include <rte_memory.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_per_lcore.h> @@ -59,7 +58,6 @@ #include <rte_mbuf.h> #include <rte_ether.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_ethdev.h> #include <rte_byteorder.h> #include <rte_malloc.h> @@ -95,7 +93,7 @@ struct efd_stats { static struct node_rx_buf *cl_rx_buf; static const char * -get_printable_mac_addr(uint8_t port) +get_printable_mac_addr(uint16_t port) { static const char err_address[] = "00:00:00:00:00:00"; static char addresses[RTE_MAX_ETHPORTS][sizeof(err_address)]; diff --git a/examples/server_node_efd/shared/common.h b/examples/server_node_efd/shared/common.h index 8a134799..b1e0abe5 100644 --- a/examples/server_node_efd/shared/common.h +++ b/examples/server_node_efd/shared/common.h @@ -65,9 +65,9 @@ struct filter_stats { struct shared_info { uint8_t num_nodes; - uint8_t num_ports; + uint16_t num_ports; uint32_t num_flows; - uint8_t id[RTE_MAX_ETHPORTS]; + uint16_t id[RTE_MAX_ETHPORTS]; struct rx_stats rx_stats; struct tx_stats tx_stats[MAX_NODES]; struct filter_stats filter_stats[MAX_NODES]; diff --git a/examples/vhost_xen/Makefile b/examples/service_cores/Makefile index ad2466aa..bd4a345d 100644 --- a/examples/vhost_xen/Makefile +++ b/examples/service_cores/Makefile @@ -1,7 +1,6 @@ # BSD LICENSE # -# Copyright(c) 2010-2014 Intel Corporation. All rights reserved. -# All rights reserved. +# Copyright(c) 2017 Intel Corporation. All rights reserved. # # Redistribution and use in source and binary forms, with or without # modification, are permitted provided that the following conditions @@ -39,14 +38,17 @@ RTE_TARGET ?= x86_64-native-linuxapp-gcc include $(RTE_SDK)/mk/rte.vars.mk # binary name -APP = vhost-switch +APP = service_cores # all source are stored in SRCS-y -SRCS-y := main.c vhost_monitor.c xenstore_parse.c +SRCS-y := main.c -CFLAGS += -O2 -I/usr/local/include -D_FILE_OFFSET_BITS=64 -Wno-unused-parameter CFLAGS += $(WERROR_FLAGS) -CFLAGS += -D_GNU_SOURCE -LDFLAGS += -lxenstore + +# workaround for a gcc bug with noreturn attribute +# http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12603 +ifeq ($(CONFIG_RTE_TOOLCHAIN_GCC),y) +CFLAGS_main.o += -Wno-return-type +endif include $(RTE_SDK)/mk/rte.extapp.mk diff --git a/examples/service_cores/main.c b/examples/service_cores/main.c new file mode 100644 index 00000000..b617a789 --- /dev/null +++ b/examples/service_cores/main.c @@ -0,0 +1,246 @@ +/* + * BSD LICENSE + * + * Copyright(c) 2017 Intel Corporation. All rights reserved. + * + * 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. + * * Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#include <unistd.h> +#include <stdio.h> +#include <string.h> +#include <stdint.h> +#include <errno.h> +#include <sys/queue.h> + +#include <rte_memory.h> +#include <rte_launch.h> +#include <rte_eal.h> +#include <rte_debug.h> +#include <rte_cycles.h> + +/* allow application scheduling of the services */ +#include <rte_service.h> + +/* Allow application registration of its own services. An application does not + * have to register services, but it can be useful if it wishes to run a + * function on a core that is otherwise in use as a service core. In this + * example, all services are dummy services registered by the sample app itself. + */ +#include <rte_service_component.h> + +#define PROFILE_CORES_MAX 8 +#define PROFILE_SERVICE_PER_CORE 5 + +/* dummy function to do "work" */ +static int32_t service_func(void *args) +{ + RTE_SET_USED(args); + rte_delay_us(2000); + return 0; +} + +static struct rte_service_spec services[] = { + {"service_1", service_func, NULL, 0, 0}, + {"service_2", service_func, NULL, 0, 0}, + {"service_3", service_func, NULL, 0, 0}, + {"service_4", service_func, NULL, 0, 0}, + {"service_5", service_func, NULL, 0, 0}, +}; +#define NUM_SERVICES RTE_DIM(services) + +/* this struct holds the mapping of a particular core to all services */ +struct profile_for_core { + uint32_t mapped_services[PROFILE_SERVICE_PER_CORE]; +}; + +/* struct that can be applied as the service core mapping. Items in this + * struct will be passed to the ordinary rte_service_* APIs to configure the + * service cores at runtime, based on the requirements. + * + * These profiles can be considered a "configuration" for the service cores, + * where switching profile just changes the number of cores and the mappings + * for each of them. As a result, the core requirements and performance of the + * application scales. + */ +struct profile { + char name[64]; + uint32_t num_cores; + struct profile_for_core cores[PROFILE_CORES_MAX]; +}; + +static struct profile profiles[] = { + /* profile 0: high performance */ + { + .name = "High Performance", + .num_cores = 5, + .cores[0] = {.mapped_services = {1, 0, 0, 0, 0} }, + .cores[1] = {.mapped_services = {0, 1, 0, 0, 0} }, + .cores[2] = {.mapped_services = {0, 0, 1, 0, 0} }, + .cores[3] = {.mapped_services = {0, 0, 0, 1, 0} }, + .cores[4] = {.mapped_services = {0, 0, 0, 0, 1} }, + }, + /* profile 1: mid performance with single service priority */ + { + .name = "Mid-High Performance", + .num_cores = 3, + .cores[0] = {.mapped_services = {1, 1, 0, 0, 0} }, + .cores[1] = {.mapped_services = {0, 0, 1, 1, 0} }, + .cores[2] = {.mapped_services = {0, 0, 0, 0, 1} }, + .cores[3] = {.mapped_services = {0, 0, 0, 0, 0} }, + .cores[4] = {.mapped_services = {0, 0, 0, 0, 0} }, + }, + /* profile 2: mid performance with single service priority */ + { + .name = "Mid-Low Performance", + .num_cores = 2, + .cores[0] = {.mapped_services = {1, 1, 1, 0, 0} }, + .cores[1] = {.mapped_services = {1, 1, 0, 1, 1} }, + .cores[2] = {.mapped_services = {0, 0, 0, 0, 0} }, + .cores[3] = {.mapped_services = {0, 0, 0, 0, 0} }, + .cores[4] = {.mapped_services = {0, 0, 0, 0, 0} }, + }, + /* profile 3: scale down performance on single core */ + { + .name = "Scale down performance", + .num_cores = 1, + .cores[0] = {.mapped_services = {1, 1, 1, 1, 1} }, + .cores[1] = {.mapped_services = {0, 0, 0, 0, 0} }, + .cores[2] = {.mapped_services = {0, 0, 0, 0, 0} }, + .cores[3] = {.mapped_services = {0, 0, 0, 0, 0} }, + .cores[4] = {.mapped_services = {0, 0, 0, 0, 0} }, + }, +}; +#define NUM_PROFILES RTE_DIM(profiles) + +static void +apply_profile(int profile_id) +{ + uint32_t i; + uint32_t s; + int ret; + struct profile *p = &profiles[profile_id]; + const uint8_t core_off = 1; + + for (i = 0; i < p->num_cores; i++) { + uint32_t core = i + core_off; + ret = rte_service_lcore_add(core); + if (ret && ret != -EALREADY) + printf("core %d added ret %d\n", core, ret); + + ret = rte_service_lcore_start(core); + if (ret && ret != -EALREADY) + printf("core %d start ret %d\n", core, ret); + + for (s = 0; s < NUM_SERVICES; s++) { + if (rte_service_map_lcore_set(s, core, + p->cores[i].mapped_services[s])) + printf("failed to map lcore %d\n", core); + } + } + + for ( ; i < PROFILE_CORES_MAX; i++) { + uint32_t core = i + core_off; + for (s = 0; s < NUM_SERVICES; s++) { + ret = rte_service_map_lcore_set(s, core, 0); + if (ret && ret != -EINVAL) { + printf("%s %d: map lcore set = %d\n", __func__, + __LINE__, ret); + } + } + ret = rte_service_lcore_stop(core); + if (ret && ret != -EALREADY) { + printf("%s %d: lcore stop = %d\n", __func__, + __LINE__, ret); + } + ret = rte_service_lcore_del(core); + if (ret && ret != -EINVAL) { + printf("%s %d: lcore del = %d\n", __func__, + __LINE__, ret); + } + } +} + +int +main(int argc, char **argv) +{ + int ret; + + ret = rte_eal_init(argc, argv); + if (ret < 0) + rte_panic("Cannot init EAL\n"); + + uint32_t i; + for (i = 0; i < NUM_SERVICES; i++) { + services[i].callback_userdata = 0; + uint32_t id; + ret = rte_service_component_register(&services[i], &id); + if (ret) + rte_exit(-1, "service register() failed"); + + /* set the service itself to be ready to run. In the case of + * ethdev, eventdev etc PMDs, this will be set when the + * appropriate configure or setup function is called. + */ + rte_service_component_runstate_set(id, 1); + + /* Collect statistics for the service */ + rte_service_set_stats_enable(id, 1); + + /* the application sets the service to be active. Note that the + * previous component_runstate_set() is the PMD indicating + * ready, while this function is the application setting the + * service to run. Applications can choose to not run a service + * by setting runstate to 0 at any time. + */ + ret = rte_service_runstate_set(id, 1); + if (ret) + return -ENOEXEC; + } + + i = 0; + while (1) { + const char clr[] = { 27, '[', '2', 'J', '\0' }; + const char topLeft[] = { 27, '[', '1', ';', '1', 'H', '\0' }; + printf("%s%s", clr, topLeft); + + apply_profile(i); + printf("\n==> Profile: %s\n\n", profiles[i].name); + + sleep(1); + rte_service_dump(stdout, UINT32_MAX); + + sleep(5); + rte_service_dump(stdout, UINT32_MAX); + + i++; + if (i >= NUM_PROFILES) + i = 0; + } + + return 0; +} diff --git a/examples/skeleton/basicfwd.c b/examples/skeleton/basicfwd.c index b4d50de8..e623754c 100644 --- a/examples/skeleton/basicfwd.c +++ b/examples/skeleton/basicfwd.c @@ -57,7 +57,7 @@ static const struct rte_eth_conf port_conf_default = { * coming from the mbuf_pool passed as a parameter. */ static inline int -port_init(uint8_t port, struct rte_mempool *mbuf_pool) +port_init(uint16_t port, struct rte_mempool *mbuf_pool) { struct rte_eth_conf port_conf = port_conf_default; const uint16_t rx_rings = 1, tx_rings = 1; @@ -104,7 +104,7 @@ port_init(uint8_t port, struct rte_mempool *mbuf_pool) rte_eth_macaddr_get(port, &addr); printf("Port %u MAC: %02" PRIx8 " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 "\n", - (unsigned)port, + port, addr.addr_bytes[0], addr.addr_bytes[1], addr.addr_bytes[2], addr.addr_bytes[3], addr.addr_bytes[4], addr.addr_bytes[5]); @@ -122,8 +122,8 @@ port_init(uint8_t port, struct rte_mempool *mbuf_pool) static __attribute__((noreturn)) void lcore_main(void) { - const uint8_t nb_ports = rte_eth_dev_count(); - uint8_t port; + const uint16_t nb_ports = rte_eth_dev_count(); + uint16_t port; /* * Check that the port is on the same NUMA node as the polling thread @@ -179,7 +179,7 @@ main(int argc, char *argv[]) { struct rte_mempool *mbuf_pool; unsigned nb_ports; - uint8_t portid; + uint16_t portid; /* Initialize the Environment Abstraction Layer (EAL). */ int ret = rte_eal_init(argc, argv); @@ -204,7 +204,7 @@ main(int argc, char *argv[]) /* Initialize all ports. */ for (portid = 0; portid < nb_ports; portid++) if (port_init(portid, mbuf_pool) != 0) - rte_exit(EXIT_FAILURE, "Cannot init port %"PRIu8 "\n", + rte_exit(EXIT_FAILURE, "Cannot init port %"PRIu16 "\n", portid); if (rte_lcore_count() > 1) diff --git a/examples/tep_termination/main.c b/examples/tep_termination/main.c index aee36c6e..a14d2e37 100644 --- a/examples/tep_termination/main.c +++ b/examples/tep_termination/main.c @@ -98,7 +98,7 @@ #define MBUF_HEADROOM_UINT32(mbuf) (*(uint32_t *)((uint8_t *)(mbuf) \ + sizeof(struct rte_mbuf))) -#define INVALID_PORT_ID 0xFF +#define INVALID_PORT_ID 0xFFFF /* Size of buffers used for snprintfs. */ #define MAX_PRINT_BUFF 6072 @@ -184,7 +184,7 @@ static uint32_t burst_rx_retry_num = BURST_RX_RETRIES; static char dev_basename[MAX_BASENAME_SZ] = "vhost-net"; static unsigned lcore_ids[RTE_MAX_LCORE]; -uint8_t ports[RTE_MAX_ETHPORTS]; +uint16_t ports[RTE_MAX_ETHPORTS]; static unsigned nb_ports; /**< The number of ports specified in command line */ @@ -1161,7 +1161,7 @@ main(int argc, char *argv[]) unsigned lcore_id, core_id = 0; unsigned nb_ports, valid_nb_ports; int ret; - uint8_t portid; + uint16_t portid; uint16_t queue_id; static pthread_t tid; char thread_name[RTE_MAX_THREAD_NAME_LEN]; diff --git a/examples/tep_termination/vxlan_setup.c b/examples/tep_termination/vxlan_setup.c index 050bb32d..1ad4ca3c 100644 --- a/examples/tep_termination/vxlan_setup.c +++ b/examples/tep_termination/vxlan_setup.c @@ -129,7 +129,7 @@ const uint16_t tenant_id_conf[] = { * coming from the mbuf_pool passed as parameter */ int -vxlan_port_init(uint8_t port, struct rte_mempool *mbuf_pool) +vxlan_port_init(uint16_t port, struct rte_mempool *mbuf_pool) { int retval; uint16_t q; @@ -202,7 +202,7 @@ vxlan_port_init(uint8_t port, struct rte_mempool *mbuf_pool) rte_eth_macaddr_get(port, &ports_eth_addr[port]); RTE_LOG(INFO, PORT, "Port %u MAC: %02"PRIx8" %02"PRIx8" %02"PRIx8 " %02"PRIx8" %02"PRIx8" %02"PRIx8"\n", - (unsigned)port, + port, ports_eth_addr[port].addr_bytes[0], ports_eth_addr[port].addr_bytes[1], ports_eth_addr[port].addr_bytes[2], @@ -414,7 +414,7 @@ vxlan_unlink(struct vhost_dev *vdev) /* Transmit packets after encapsulating */ int -vxlan_tx_pkts(uint8_t port_id, uint16_t queue_id, +vxlan_tx_pkts(uint16_t port_id, uint16_t queue_id, struct rte_mbuf **tx_pkts, uint16_t nb_pkts) { int ret = 0; uint16_t i; diff --git a/examples/tep_termination/vxlan_setup.h b/examples/tep_termination/vxlan_setup.h index 8d264619..2e3550d3 100644 --- a/examples/tep_termination/vxlan_setup.h +++ b/examples/tep_termination/vxlan_setup.h @@ -37,14 +37,14 @@ extern uint16_t nb_devices; extern uint16_t udp_port; extern uint8_t filter_idx; -extern uint8_t ports[RTE_MAX_ETHPORTS]; +extern uint16_t ports[RTE_MAX_ETHPORTS]; extern struct ether_addr ports_eth_addr[RTE_MAX_ETHPORTS]; extern uint32_t enable_stats; extern struct device_statistics dev_statistics[MAX_DEVICES]; extern uint8_t rx_decap; extern uint8_t tx_encap; -typedef int (*ol_port_configure_t)(uint8_t port, +typedef int (*ol_port_configure_t)(uint16_t port, struct rte_mempool *mbuf_pool); typedef int (*ol_tunnel_setup_t)(struct vhost_dev *vdev, @@ -52,7 +52,7 @@ typedef int (*ol_tunnel_setup_t)(struct vhost_dev *vdev, typedef void (*ol_tunnel_destroy_t)(struct vhost_dev *vdev); -typedef int (*ol_tx_handle_t)(uint8_t port_id, uint16_t queue_id, +typedef int (*ol_tx_handle_t)(uint16_t port_id, uint16_t queue_id, struct rte_mbuf **tx_pkts, uint16_t nb_pkts); typedef int (*ol_rx_handle_t)(int vid, struct rte_mbuf **pkts, @@ -70,7 +70,7 @@ struct ol_switch_ops { }; int -vxlan_port_init(uint8_t port, struct rte_mempool *mbuf_pool); +vxlan_port_init(uint16_t port, struct rte_mempool *mbuf_pool); int vxlan_link(struct vhost_dev *vdev, struct rte_mbuf *m); @@ -79,7 +79,7 @@ void vxlan_unlink(struct vhost_dev *vdev); int -vxlan_tx_pkts(uint8_t port_id, uint16_t queue_id, +vxlan_tx_pkts(uint16_t port_id, uint16_t queue_id, struct rte_mbuf **tx_pkts, uint16_t nb_pkts); int vxlan_rx_pkts(int vid, struct rte_mbuf **pkts, uint32_t count); diff --git a/examples/timer/main.c b/examples/timer/main.c index 37ad559e..444e9cc1 100644 --- a/examples/timer/main.c +++ b/examples/timer/main.c @@ -39,7 +39,6 @@ #include <rte_common.h> #include <rte_memory.h> -#include <rte_memzone.h> #include <rte_launch.h> #include <rte_eal.h> #include <rte_per_lcore.h> diff --git a/examples/vhost/main.c b/examples/vhost/main.c index 4d1589d0..89a61f0e 100644 --- a/examples/vhost/main.c +++ b/examples/vhost/main.c @@ -177,7 +177,7 @@ static struct rte_eth_conf vmdq_conf_default = { }; static unsigned lcore_ids[RTE_MAX_LCORE]; -static uint8_t ports[RTE_MAX_ETHPORTS]; +static uint16_t ports[RTE_MAX_ETHPORTS]; static unsigned num_ports = 0; /**< The number of ports specified in command line */ static uint16_t num_pf_queues, num_vmdq_queues; static uint16_t vmdq_pool_base, vmdq_queue_base; @@ -265,7 +265,7 @@ validate_num_devices(uint32_t max_nb_devices) * coming from the mbuf_pool passed as parameter */ static inline int -port_init(uint8_t port) +port_init(uint16_t port) { struct rte_eth_dev_info dev_info; struct rte_eth_conf port_conf; @@ -392,7 +392,7 @@ port_init(uint8_t port) RTE_LOG(INFO, VHOST_PORT, "Max virtio devices supported: %u\n", num_devices); RTE_LOG(INFO, VHOST_PORT, "Port %u MAC: %02"PRIx8" %02"PRIx8" %02"PRIx8 " %02"PRIx8" %02"PRIx8" %02"PRIx8"\n", - (unsigned)port, + port, vmdq_ports_eth_addr[port].addr_bytes[0], vmdq_ports_eth_addr[port].addr_bytes[1], vmdq_ports_eth_addr[port].addr_bytes[2], @@ -667,7 +667,7 @@ us_vhost_parse_args(int argc, char **argv) for (i = 0; i < RTE_MAX_ETHPORTS; i++) { if (enabled_port_mask & (1 << i)) - ports[num_ports++] = (uint8_t)i; + ports[num_ports++] = i; } if ((num_ports == 0) || (num_ports > MAX_SUP_PORTS)) { @@ -1443,7 +1443,7 @@ main(int argc, char *argv[]) unsigned lcore_id, core_id = 0; unsigned nb_ports, valid_num_ports; int ret, i; - uint8_t portid; + uint16_t portid; static pthread_t tid; char thread_name[RTE_MAX_THREAD_NAME_LEN]; uint64_t flags = 0; diff --git a/examples/vhost_scsi/scsi.c b/examples/vhost_scsi/scsi.c index 54d3104e..fd430ec2 100644 --- a/examples/vhost_scsi/scsi.c +++ b/examples/vhost_scsi/scsi.c @@ -307,7 +307,9 @@ vhost_bdev_scsi_inquiry_command(struct vhost_block_dev *bdev, strncpy((char *)inqdata->t10_vendor_id, "INTEL", 8); /* PRODUCT IDENTIFICATION */ - strncpy((char *)inqdata->product_id, bdev->product_name, 16); + snprintf((char *)inqdata->product_id, + RTE_DIM(inqdata->product_id), "%s", + bdev->product_name); /* PRODUCT REVISION LEVEL */ strncpy((char *)inqdata->product_rev, "0001", 4); diff --git a/examples/vhost_xen/main.c b/examples/vhost_xen/main.c deleted file mode 100644 index eba4d35a..00000000 --- a/examples/vhost_xen/main.c +++ /dev/null @@ -1,1522 +0,0 @@ -/*- - * BSD LICENSE - * - * Copyright(c) 2010-2015 Intel Corporation. All rights reserved. - * All rights reserved. - * - * 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. - * * Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - */ - -#include <arpa/inet.h> -#include <getopt.h> -#include <linux/if_ether.h> -#include <linux/if_vlan.h> -#include <linux/virtio_net.h> -#include <linux/virtio_ring.h> -#include <signal.h> -#include <stdint.h> -#include <sys/eventfd.h> -#include <sys/param.h> -#include <unistd.h> - -#include <rte_atomic.h> -#include <rte_cycles.h> -#include <rte_ethdev.h> -#include <rte_log.h> -#include <rte_string_fns.h> -#include <rte_pause.h> - -#include "main.h" -#include "virtio-net.h" -#include "xen_vhost.h" - -#define MAX_QUEUES 128 - -/* the maximum number of external ports supported */ -#define MAX_SUP_PORTS 1 - -/* - * Calculate the number of buffers needed per port - */ -#define NUM_MBUFS_PER_PORT ((MAX_QUEUES*RTE_TEST_RX_DESC_DEFAULT) + \ - (num_switching_cores*MAX_PKT_BURST) + \ - (num_switching_cores*RTE_TEST_TX_DESC_DEFAULT) +\ - (num_switching_cores*MBUF_CACHE_SIZE)) - -#define MBUF_CACHE_SIZE 64 - -/* - * RX and TX Prefetch, Host, and Write-back threshold values should be - * carefully set for optimal performance. Consult the network - * controller's datasheet and supporting DPDK documentation for guidance - * on how these parameters should be set. - */ -#define RX_PTHRESH 8 /* Default values of RX prefetch threshold reg. */ -#define RX_HTHRESH 8 /* Default values of RX host threshold reg. */ -#define RX_WTHRESH 4 /* Default values of RX write-back threshold reg. */ - -/* - * These default values are optimized for use with the Intel(R) 82599 10 GbE - * Controller and the DPDK ixgbe PMD. Consider using other values for other - * network controllers and/or network drivers. - */ -#define TX_PTHRESH 36 /* Default values of TX prefetch threshold reg. */ -#define TX_HTHRESH 0 /* Default values of TX host threshold reg. */ -#define TX_WTHRESH 0 /* Default values of TX write-back threshold reg. */ - -#define MAX_PKT_BURST 32 /* Max burst size for RX/TX */ -#define MAX_MRG_PKT_BURST 16 /* Max burst for merge buffers. Set to 1 due to performance issue. */ -#define BURST_TX_DRAIN_US 100 /* TX drain every ~100us */ - -/* State of virtio device. */ -#define DEVICE_NOT_READY 0 -#define DEVICE_READY 1 -#define DEVICE_SAFE_REMOVE 2 - -/* Config_core_flag status definitions. */ -#define REQUEST_DEV_REMOVAL 1 -#define ACK_DEV_REMOVAL 0 - -/* Configurable number of RX/TX ring descriptors */ -#define RTE_TEST_RX_DESC_DEFAULT 128 -#define RTE_TEST_TX_DESC_DEFAULT 512 - -#define INVALID_PORT_ID 0xFF - -/* Max number of devices. Limited by vmdq. */ -#define MAX_DEVICES 64 - -/* Size of buffers used for snprintfs. */ -#define MAX_PRINT_BUFF 6072 - - -/* Maximum long option length for option parsing. */ -#define MAX_LONG_OPT_SZ 64 - -/* Used to compare MAC addresses. */ -#define MAC_ADDR_CMP 0xFFFFFFFFFFFF - -/* mask of enabled ports */ -static uint32_t enabled_port_mask = 0; - -/*Number of switching cores enabled*/ -static uint32_t num_switching_cores = 0; - -/* number of devices/queues to support*/ -static uint32_t num_queues = 0; -uint32_t num_devices = 0; - -/* Enable VM2VM communications. If this is disabled then the MAC address compare is skipped. */ -static uint32_t enable_vm2vm = 1; -/* Enable stats. */ -static uint32_t enable_stats = 0; - -/* empty vmdq configuration structure. Filled in programatically */ -static const struct rte_eth_conf vmdq_conf_default = { - .rxmode = { - .mq_mode = ETH_MQ_RX_VMDQ_ONLY, - .split_hdr_size = 0, - .header_split = 0, /**< Header Split disabled */ - .hw_ip_checksum = 0, /**< IP checksum offload disabled */ - .hw_vlan_filter = 0, /**< VLAN filtering disabled */ - /* - * It is necessary for 1G NIC such as I350, - * this fixes bug of ipv4 forwarding in guest can't - * forward pakets from one virtio dev to another virtio dev. - */ - .hw_vlan_strip = 1, /**< VLAN strip enabled. */ - .jumbo_frame = 0, /**< Jumbo Frame Support disabled */ - .hw_strip_crc = 1, /**< CRC stripped by hardware */ - }, - - .txmode = { - .mq_mode = ETH_MQ_TX_NONE, - }, - .rx_adv_conf = { - /* - * should be overridden separately in code with - * appropriate values - */ - .vmdq_rx_conf = { - .nb_queue_pools = ETH_8_POOLS, - .enable_default_pool = 0, - .default_pool = 0, - .nb_pool_maps = 0, - .pool_map = {{0, 0},}, - }, - }, -}; - -static unsigned lcore_ids[RTE_MAX_LCORE]; -static uint8_t ports[RTE_MAX_ETHPORTS]; -static unsigned num_ports = 0; /**< The number of ports specified in command line */ - -const uint16_t vlan_tags[] = { - 1000, 1001, 1002, 1003, 1004, 1005, 1006, 1007, - 1008, 1009, 1010, 1011, 1012, 1013, 1014, 1015, - 1016, 1017, 1018, 1019, 1020, 1021, 1022, 1023, - 1024, 1025, 1026, 1027, 1028, 1029, 1030, 1031, - 1032, 1033, 1034, 1035, 1036, 1037, 1038, 1039, - 1040, 1041, 1042, 1043, 1044, 1045, 1046, 1047, - 1048, 1049, 1050, 1051, 1052, 1053, 1054, 1055, - 1056, 1057, 1058, 1059, 1060, 1061, 1062, 1063, -}; - -/* ethernet addresses of ports */ -static struct ether_addr vmdq_ports_eth_addr[RTE_MAX_ETHPORTS]; - -/* heads for the main used and free linked lists for the data path. */ -static struct virtio_net_data_ll *ll_root_used = NULL; -static struct virtio_net_data_ll *ll_root_free = NULL; - -/* Array of data core structures containing information on individual core linked lists. */ -static struct lcore_info lcore_info[RTE_MAX_LCORE]; - -/* Used for queueing bursts of TX packets. */ -struct mbuf_table { - unsigned len; - unsigned txq_id; - struct rte_mbuf *m_table[MAX_PKT_BURST]; -}; - -/* TX queue for each data core. */ -struct mbuf_table lcore_tx_queue[RTE_MAX_LCORE]; - -/* Vlan header struct used to insert vlan tags on TX. */ -struct vlan_ethhdr { - unsigned char h_dest[ETH_ALEN]; - unsigned char h_source[ETH_ALEN]; - __be16 h_vlan_proto; - __be16 h_vlan_TCI; - __be16 h_vlan_encapsulated_proto; -}; - -/* Header lengths. */ -#define VLAN_HLEN 4 -#define VLAN_ETH_HLEN 18 - -/* Per-device statistics struct */ -struct device_statistics { - uint64_t tx_total; - rte_atomic64_t rx_total; - uint64_t tx; - rte_atomic64_t rx; -} __rte_cache_aligned; -struct device_statistics dev_statistics[MAX_DEVICES]; - -/* - * Builds up the correct configuration for VMDQ VLAN pool map - * according to the pool & queue limits. - */ -static inline int -get_eth_conf(struct rte_eth_conf *eth_conf, uint32_t num_devices) -{ - struct rte_eth_vmdq_rx_conf conf; - unsigned i; - - memset(&conf, 0, sizeof(conf)); - conf.nb_queue_pools = (enum rte_eth_nb_pools)num_devices; - conf.nb_pool_maps = num_devices; - - for (i = 0; i < conf.nb_pool_maps; i++) { - conf.pool_map[i].vlan_id = vlan_tags[ i ]; - conf.pool_map[i].pools = (1UL << i); - } - - (void)(rte_memcpy(eth_conf, &vmdq_conf_default, sizeof(*eth_conf))); - (void)(rte_memcpy(ð_conf->rx_adv_conf.vmdq_rx_conf, &conf, - sizeof(eth_conf->rx_adv_conf.vmdq_rx_conf))); - return 0; -} - -/* - * Validate the device number according to the max pool number gotten form dev_info - * If the device number is invalid, give the error message and return -1. - * Each device must have its own pool. - */ -static inline int -validate_num_devices(uint32_t max_nb_devices) -{ - if (num_devices > max_nb_devices) { - RTE_LOG(ERR, VHOST_PORT, "invalid number of devices\n"); - return -1; - } - return 0; -} - -/* - * Initialises a given port using global settings and with the rx buffers - * coming from the mbuf_pool passed as parameter - */ -static inline int -port_init(uint8_t port, struct rte_mempool *mbuf_pool) -{ - struct rte_eth_dev_info dev_info; - struct rte_eth_rxconf *rxconf; - struct rte_eth_conf port_conf; - uint16_t rx_rings, tx_rings = (uint16_t)rte_lcore_count(); - uint16_t rx_ring_size = RTE_TEST_RX_DESC_DEFAULT; - uint16_t tx_ring_size = RTE_TEST_TX_DESC_DEFAULT; - int retval; - uint16_t q; - - /* The max pool number from dev_info will be used to validate the pool number specified in cmd line */ - rte_eth_dev_info_get (port, &dev_info); - - /*configure the number of supported virtio devices based on VMDQ limits */ - num_devices = dev_info.max_vmdq_pools; - num_queues = dev_info.max_rx_queues; - - retval = validate_num_devices(MAX_DEVICES); - if (retval < 0) - return retval; - - /* Get port configuration. */ - retval = get_eth_conf(&port_conf, num_devices); - if (retval < 0) - return retval; - - if (port >= rte_eth_dev_count()) return -1; - - rx_rings = (uint16_t)num_queues, - /* Configure ethernet device. */ - retval = rte_eth_dev_configure(port, rx_rings, tx_rings, &port_conf); - if (retval != 0) - return retval; - - retval = rte_eth_dev_adjust_nb_rx_tx_desc(port, &rx_ring_size, - &tx_ring_size); - if (retval != 0) - return retval; - if (rx_ring_size > RTE_TEST_RX_DESC_DEFAULT || - tx_ring_size > RTE_TEST_TX_DESC_DEFAULT) { - RTE_LOG(ERR, VHOST_PORT, "Mbuf pool has an insufficient size for " - "port %u.\n", port); - return -1; - } - - rte_eth_dev_info_get(port, &dev_info); - rxconf = &dev_info.default_rxconf; - rxconf->rx_drop_en = 1; - /* Setup the queues. */ - for (q = 0; q < rx_rings; q ++) { - retval = rte_eth_rx_queue_setup(port, q, rx_ring_size, - rte_eth_dev_socket_id(port), rxconf, - mbuf_pool); - if (retval < 0) - return retval; - } - for (q = 0; q < tx_rings; q ++) { - retval = rte_eth_tx_queue_setup(port, q, tx_ring_size, - rte_eth_dev_socket_id(port), - NULL); - if (retval < 0) - return retval; - } - - /* Start the device. */ - retval = rte_eth_dev_start(port); - if (retval < 0) - return retval; - - rte_eth_macaddr_get(port, &vmdq_ports_eth_addr[port]); - RTE_LOG(INFO, VHOST_PORT, "Max virtio devices supported: %u\n", num_devices); - RTE_LOG(INFO, VHOST_PORT, "Port %u MAC: %02"PRIx8" %02"PRIx8" %02"PRIx8 - " %02"PRIx8" %02"PRIx8" %02"PRIx8"\n", - (unsigned)port, - vmdq_ports_eth_addr[port].addr_bytes[0], - vmdq_ports_eth_addr[port].addr_bytes[1], - vmdq_ports_eth_addr[port].addr_bytes[2], - vmdq_ports_eth_addr[port].addr_bytes[3], - vmdq_ports_eth_addr[port].addr_bytes[4], - vmdq_ports_eth_addr[port].addr_bytes[5]); - - return 0; -} - -/* - * Parse the portmask provided at run time. - */ -static int -parse_portmask(const char *portmask) -{ - char *end = NULL; - unsigned long pm; - - errno = 0; - - /* parse hexadecimal string */ - pm = strtoul(portmask, &end, 16); - if ((portmask[0] == '\0') || (end == NULL) || (*end != '\0') || (errno != 0)) - return -1; - - if (pm == 0) - return -1; - - return pm; - -} - -/* - * Parse num options at run time. - */ -static int -parse_num_opt(const char *q_arg, uint32_t max_valid_value) -{ - char *end = NULL; - unsigned long num; - - errno = 0; - - /* parse unsigned int string */ - num = strtoul(q_arg, &end, 10); - if ((q_arg[0] == '\0') || (end == NULL) || (*end != '\0') || (errno != 0)) - return -1; - - if (num > max_valid_value) - return -1; - - return num; - -} - -/* - * Display usage - */ -static void -us_vhost_usage(const char *prgname) -{ - RTE_LOG(INFO, VHOST_CONFIG, "%s [EAL options] -- -p PORTMASK --vm2vm [0|1] --stats [0-N] --nb-devices ND\n" - " -p PORTMASK: Set mask for ports to be used by application\n" - " --vm2vm [0|1]: disable/enable(default) vm2vm comms\n" - " --stats [0-N]: 0: Disable stats, N: Time in seconds to print stats\n", - prgname); -} - -/* - * Parse the arguments given in the command line of the application. - */ -static int -us_vhost_parse_args(int argc, char **argv) -{ - int opt, ret; - int option_index; - unsigned i; - const char *prgname = argv[0]; - static struct option long_option[] = { - {"vm2vm", required_argument, NULL, 0}, - {"stats", required_argument, NULL, 0}, - {NULL, 0, 0, 0} - }; - - /* Parse command line */ - while ((opt = getopt_long(argc, argv, "p:",long_option, &option_index)) != EOF) { - switch (opt) { - /* Portmask */ - case 'p': - enabled_port_mask = parse_portmask(optarg); - if (enabled_port_mask == 0) { - RTE_LOG(INFO, VHOST_CONFIG, "Invalid portmask\n"); - us_vhost_usage(prgname); - return -1; - } - break; - - case 0: - /* Enable/disable vm2vm comms. */ - if (!strncmp(long_option[option_index].name, "vm2vm", MAX_LONG_OPT_SZ)) { - ret = parse_num_opt(optarg, 1); - if (ret == -1) { - RTE_LOG(INFO, VHOST_CONFIG, "Invalid argument for vm2vm [0|1]\n"); - us_vhost_usage(prgname); - return -1; - } else { - enable_vm2vm = ret; - } - } - - /* Enable/disable stats. */ - if (!strncmp(long_option[option_index].name, "stats", MAX_LONG_OPT_SZ)) { - ret = parse_num_opt(optarg, INT32_MAX); - if (ret == -1) { - RTE_LOG(INFO, VHOST_CONFIG, "Invalid argument for stats [0..N]\n"); - us_vhost_usage(prgname); - return -1; - } else { - enable_stats = ret; - } - } - break; - - /* Invalid option - print options. */ - default: - us_vhost_usage(prgname); - return -1; - } - } - - for (i = 0; i < RTE_MAX_ETHPORTS; i++) { - if (enabled_port_mask & (1 << i)) - ports[num_ports++] = (uint8_t)i; - } - - if ((num_ports == 0) || (num_ports > MAX_SUP_PORTS)) { - RTE_LOG(INFO, VHOST_PORT, "Current enabled port number is %u," - "but only %u port can be enabled\n",num_ports, MAX_SUP_PORTS); - return -1; - } - - return 0; -} - -/* - * Update the global var NUM_PORTS and array PORTS according to system ports number - * and return valid ports number - */ -static unsigned check_ports_num(unsigned nb_ports) -{ - unsigned valid_num_ports = num_ports; - unsigned portid; - - if (num_ports > nb_ports) { - RTE_LOG(INFO, VHOST_PORT, "\nSpecified port number(%u) exceeds total system port number(%u)\n", - num_ports, nb_ports); - num_ports = nb_ports; - } - - for (portid = 0; portid < num_ports; portid ++) { - if (ports[portid] >= nb_ports) { - RTE_LOG(INFO, VHOST_PORT, "\nSpecified port ID(%u) exceeds max system port ID(%u)\n", - ports[portid], (nb_ports - 1)); - ports[portid] = INVALID_PORT_ID; - valid_num_ports--; - } - } - return valid_num_ports; -} - -/* - * Function to convert guest physical addresses to vhost virtual addresses. This - * is used to convert virtio buffer addresses. - */ -static __rte_always_inline uint64_t -gpa_to_vva(struct virtio_net *dev, uint64_t guest_pa) -{ - struct virtio_memory_regions *region; - uint32_t regionidx; - uint64_t vhost_va = 0; - - for (regionidx = 0; regionidx < dev->mem->nregions; regionidx++) { - region = &dev->mem->regions[regionidx]; - if ((guest_pa >= region->guest_phys_address) && - (guest_pa <= region->guest_phys_address_end)) { - vhost_va = region->address_offset + guest_pa; - break; - } - } - RTE_LOG_DP(DEBUG, VHOST_DATA, "(%" PRIu64 ") GPA %p| VVA %p\n", - dev->device_fh, (void*)(uintptr_t)guest_pa, (void*)(uintptr_t)vhost_va); - - return vhost_va; -} - -/* - * This function adds buffers to the virtio devices RX virtqueue. Buffers can - * be received from the physical port or from another virtio device. A packet - * count is returned to indicate the number of packets that were successfully - * added to the RX queue. - */ -static __rte_always_inline uint32_t -virtio_dev_rx(struct virtio_net *dev, struct rte_mbuf **pkts, uint32_t count) -{ - struct vhost_virtqueue *vq; - struct vring_desc *desc; - struct rte_mbuf *buff; - /* The virtio_hdr is initialised to 0. */ - struct virtio_net_hdr_mrg_rxbuf virtio_hdr = {{0,0,0,0,0,0},0}; - uint64_t buff_addr = 0; - uint64_t buff_hdr_addr = 0; - uint32_t head[MAX_PKT_BURST], packet_len = 0; - uint32_t head_idx, packet_success = 0; - uint16_t avail_idx, res_cur_idx; - uint16_t res_base_idx, res_end_idx; - uint16_t free_entries; - uint8_t success = 0; - void *userdata; - - RTE_LOG_DP(DEBUG, VHOST_DATA, "(%" PRIu64 ") virtio_dev_rx()\n", dev->device_fh); - vq = dev->virtqueue_rx; - count = (count > MAX_PKT_BURST) ? MAX_PKT_BURST : count; - /* As many data cores may want access to available buffers, they need to be reserved. */ - do { - - res_base_idx = vq->last_used_idx_res; - - avail_idx = *((volatile uint16_t *)&vq->avail->idx); - - free_entries = (avail_idx - res_base_idx); - - /*check that we have enough buffers*/ - if (unlikely(count > free_entries)) - count = free_entries; - - if (count == 0) - return 0; - - res_end_idx = res_base_idx + count; - /* vq->last_used_idx_res is atomically updated. */ - success = rte_atomic16_cmpset(&vq->last_used_idx_res, res_base_idx, - res_end_idx); - } while (unlikely(success == 0)); - res_cur_idx = res_base_idx; - RTE_LOG_DP(DEBUG, VHOST_DATA, "(%" PRIu64 ") Current Index %d| End Index %d\n", - dev->device_fh, res_cur_idx, res_end_idx); - - /* Prefetch available ring to retrieve indexes. */ - rte_prefetch0(&vq->avail->ring[res_cur_idx & (vq->size - 1)]); - - /* Retrieve all of the head indexes first to avoid caching issues. */ - for (head_idx = 0; head_idx < count; head_idx++) - head[head_idx] = vq->avail->ring[(res_cur_idx + head_idx) & (vq->size - 1)]; - - /*Prefetch descriptor index. */ - rte_prefetch0(&vq->desc[head[packet_success]]); - - while (res_cur_idx != res_end_idx) { - /* Get descriptor from available ring */ - desc = &vq->desc[head[packet_success]]; - /* Prefetch descriptor address. */ - rte_prefetch0(desc); - - buff = pkts[packet_success]; - - /* Convert from gpa to vva (guest physical addr -> vhost virtual addr) */ - buff_addr = gpa_to_vva(dev, desc->addr); - /* Prefetch buffer address. */ - rte_prefetch0((void*)(uintptr_t)buff_addr); - - { - /* Copy virtio_hdr to packet and increment buffer address */ - buff_hdr_addr = buff_addr; - packet_len = rte_pktmbuf_data_len(buff) + vq->vhost_hlen; - - /* - * If the descriptors are chained the header and data are placed in - * separate buffers. - */ - if (desc->flags & VRING_DESC_F_NEXT) { - desc->len = vq->vhost_hlen; - desc = &vq->desc[desc->next]; - /* Buffer address translation. */ - buff_addr = gpa_to_vva(dev, desc->addr); - desc->len = rte_pktmbuf_data_len(buff); - } else { - buff_addr += vq->vhost_hlen; - desc->len = packet_len; - } - } - - /* Update used ring with desc information */ - vq->used->ring[res_cur_idx & (vq->size - 1)].id = head[packet_success]; - vq->used->ring[res_cur_idx & (vq->size - 1)].len = packet_len; - - /* Copy mbuf data to buffer */ - userdata = rte_pktmbuf_mtod(buff, void *); - rte_memcpy((void *)(uintptr_t)buff_addr, userdata, rte_pktmbuf_data_len(buff)); - - res_cur_idx++; - packet_success++; - - /* mergeable is disabled then a header is required per buffer. */ - rte_memcpy((void *)(uintptr_t)buff_hdr_addr, (const void *)&virtio_hdr, vq->vhost_hlen); - if (res_cur_idx < res_end_idx) { - /* Prefetch descriptor index. */ - rte_prefetch0(&vq->desc[head[packet_success]]); - } - } - - rte_compiler_barrier(); - - /* Wait until it's our turn to add our buffer to the used ring. */ - while (unlikely(vq->last_used_idx != res_base_idx)) - rte_pause(); - - *(volatile uint16_t *)&vq->used->idx += count; - - vq->last_used_idx = res_end_idx; - - return count; -} - -/* - * Compares a packet destination MAC address to a device MAC address. - */ -static __rte_always_inline int -ether_addr_cmp(struct ether_addr *ea, struct ether_addr *eb) -{ - return ((*(uint64_t *)ea ^ *(uint64_t *)eb) & MAC_ADDR_CMP) == 0; -} - -/* - * This function registers mac along with a - * vlan tag to a VMDQ. - */ -static int -link_vmdq(struct virtio_net *dev) -{ - int ret; - struct virtio_net_data_ll *dev_ll; - - dev_ll = ll_root_used; - - while (dev_ll != NULL) { - if ((dev != dev_ll->dev) && ether_addr_cmp(&dev->mac_address, &dev_ll->dev->mac_address)) { - RTE_LOG(INFO, VHOST_DATA, "(%"PRIu64") WARNING: This device is using an existing MAC address and has not been registered.\n", dev->device_fh); - return -1; - } - dev_ll = dev_ll->next; - } - - /* vlan_tag currently uses the device_id. */ - dev->vlan_tag = vlan_tags[dev->device_fh]; - dev->vmdq_rx_q = dev->device_fh * (num_queues/num_devices); - - /* Print out VMDQ registration info. */ - RTE_LOG(INFO, VHOST_DATA, "(%"PRIu64") MAC_ADDRESS %02x:%02x:%02x:%02x:%02x:%02x and VLAN_TAG %d registered\n", - dev->device_fh, - dev->mac_address.addr_bytes[0], dev->mac_address.addr_bytes[1], - dev->mac_address.addr_bytes[2], dev->mac_address.addr_bytes[3], - dev->mac_address.addr_bytes[4], dev->mac_address.addr_bytes[5], - dev->vlan_tag); - - /* Register the MAC address. */ - ret = rte_eth_dev_mac_addr_add(ports[0], &dev->mac_address, (uint32_t)dev->device_fh); - if (ret) { - RTE_LOG(ERR, VHOST_DATA, "(%"PRIu64") Failed to add device MAC address to VMDQ\n", - dev->device_fh); - return -1; - } - - /* Enable stripping of the vlan tag as we handle routing. */ - rte_eth_dev_set_vlan_strip_on_queue(ports[0], dev->vmdq_rx_q, 1); - - rte_compiler_barrier(); - /* Set device as ready for RX. */ - dev->ready = DEVICE_READY; - - return 0; -} - -/* - * Removes MAC address and vlan tag from VMDQ. Ensures that nothing is adding buffers to the RX - * queue before disabling RX on the device. - */ -static inline void -unlink_vmdq(struct virtio_net *dev) -{ - unsigned i = 0; - unsigned rx_count; - struct rte_mbuf *pkts_burst[MAX_PKT_BURST]; - - if (dev->ready == DEVICE_READY) { - /*clear MAC and VLAN settings*/ - rte_eth_dev_mac_addr_remove(ports[0], &dev->mac_address); - for (i = 0; i < 6; i++) - dev->mac_address.addr_bytes[i] = 0; - - dev->vlan_tag = 0; - - /*Clear out the receive buffers*/ - rx_count = rte_eth_rx_burst(ports[0], - (uint16_t)dev->vmdq_rx_q, pkts_burst, MAX_PKT_BURST); - - while (rx_count) { - for (i = 0; i < rx_count; i++) - rte_pktmbuf_free(pkts_burst[i]); - - rx_count = rte_eth_rx_burst(ports[0], - (uint16_t)dev->vmdq_rx_q, pkts_burst, MAX_PKT_BURST); - } - - dev->ready = DEVICE_NOT_READY; - } -} - -/* - * Check if the packet destination MAC address is for a local device. If so then put - * the packet on that devices RX queue. If not then return. - */ -static __rte_always_inline unsigned -virtio_tx_local(struct virtio_net *dev, struct rte_mbuf *m) -{ - struct virtio_net_data_ll *dev_ll; - struct ether_hdr *pkt_hdr; - uint64_t ret = 0; - - pkt_hdr = rte_pktmbuf_mtod(m, struct ether_hdr *); - - /*get the used devices list*/ - dev_ll = ll_root_used; - - while (dev_ll != NULL) { - if (likely(dev_ll->dev->ready == DEVICE_READY) && ether_addr_cmp(&(pkt_hdr->d_addr), - &dev_ll->dev->mac_address)) { - - /* Drop the packet if the TX packet is destined for the TX device. */ - if (dev_ll->dev->device_fh == dev->device_fh) { - RTE_LOG_DP(DEBUG, VHOST_DATA, "(%" PRIu64 ") TX: " - "Source and destination MAC addresses are the same. " - "Dropping packet.\n", - dev_ll->dev->device_fh); - return 0; - } - - - RTE_LOG_DP(DEBUG, VHOST_DATA, "(%" PRIu64 ") TX: " - "MAC address is local\n", dev_ll->dev->device_fh); - - if (dev_ll->dev->remove) { - /*drop the packet if the device is marked for removal*/ - RTE_LOG_DP(DEBUG, VHOST_DATA, "(%" PRIu64 ") " - "Device is marked for removal\n", - dev_ll->dev->device_fh); - } else { - /*send the packet to the local virtio device*/ - ret = virtio_dev_rx(dev_ll->dev, &m, 1); - if (enable_stats) { - rte_atomic64_add(&dev_statistics[dev_ll->dev->device_fh].rx_total, 1); - rte_atomic64_add(&dev_statistics[dev_ll->dev->device_fh].rx, ret); - dev_statistics[dev->device_fh].tx_total++; - dev_statistics[dev->device_fh].tx += ret; - } - } - - return 0; - } - dev_ll = dev_ll->next; - } - - return -1; -} - -/* - * This function routes the TX packet to the correct interface. This may be a local device - * or the physical port. - */ -static __rte_always_inline void -virtio_tx_route(struct virtio_net* dev, struct rte_mbuf *m, struct rte_mempool *mbuf_pool, uint16_t vlan_tag) -{ - struct mbuf_table *tx_q; - struct vlan_ethhdr *vlan_hdr; - struct rte_mbuf **m_table; - struct rte_mbuf *mbuf; - unsigned len, ret; - const uint16_t lcore_id = rte_lcore_id(); - - /*check if destination is local VM*/ - if (enable_vm2vm && (virtio_tx_local(dev, m) == 0)) { - return; - } - - RTE_LOG_DP(DEBUG, VHOST_DATA, "(%" PRIu64 ") TX: " - "MAC address is external\n", dev->device_fh); - - /*Add packet to the port tx queue*/ - tx_q = &lcore_tx_queue[lcore_id]; - len = tx_q->len; - - /* Allocate an mbuf and populate the structure. */ - mbuf = rte_pktmbuf_alloc(mbuf_pool); - if(!mbuf) - return; - - mbuf->data_len = m->data_len + VLAN_HLEN; - mbuf->pkt_len = mbuf->data_len; - - /* Copy ethernet header to mbuf. */ - rte_memcpy(rte_pktmbuf_mtod(mbuf, void*), - rte_pktmbuf_mtod(m, const void*), ETH_HLEN); - - - /* Setup vlan header. Bytes need to be re-ordered for network with htons()*/ - vlan_hdr = rte_pktmbuf_mtod(mbuf, struct vlan_ethhdr *); - vlan_hdr->h_vlan_encapsulated_proto = vlan_hdr->h_vlan_proto; - vlan_hdr->h_vlan_proto = htons(ETH_P_8021Q); - vlan_hdr->h_vlan_TCI = htons(vlan_tag); - - /* Copy the remaining packet contents to the mbuf. */ - rte_memcpy(rte_pktmbuf_mtod_offset(mbuf, void *, VLAN_ETH_HLEN), - rte_pktmbuf_mtod_offset(m, const void *, ETH_HLEN), - (m->data_len - ETH_HLEN)); - tx_q->m_table[len] = mbuf; - len++; - if (enable_stats) { - dev_statistics[dev->device_fh].tx_total++; - dev_statistics[dev->device_fh].tx++; - } - - if (unlikely(len == MAX_PKT_BURST)) { - m_table = (struct rte_mbuf **)tx_q->m_table; - ret = rte_eth_tx_burst(ports[0], (uint16_t)tx_q->txq_id, m_table, (uint16_t) len); - /* Free any buffers not handled by TX and update the port stats. */ - if (unlikely(ret < len)) { - do { - rte_pktmbuf_free(m_table[ret]); - } while (++ret < len); - } - - len = 0; - } - - tx_q->len = len; - return; -} - -static __rte_always_inline void -virtio_dev_tx(struct virtio_net* dev, struct rte_mempool *mbuf_pool) -{ - struct rte_mbuf m; - struct vhost_virtqueue *vq; - struct vring_desc *desc; - uint64_t buff_addr = 0; - uint32_t head[MAX_PKT_BURST]; - uint32_t used_idx; - uint32_t i; - uint16_t free_entries, packet_success = 0; - uint16_t avail_idx; - - vq = dev->virtqueue_tx; - avail_idx = *((volatile uint16_t *)&vq->avail->idx); - - /* If there are no available buffers then return. */ - if (vq->last_used_idx == avail_idx) - return; - - RTE_LOG_DP(DEBUG, VHOST_DATA, "(%" PRIu64 ") virtio_dev_tx()\n", - dev->device_fh); - - /* Prefetch available ring to retrieve head indexes. */ - rte_prefetch0(&vq->avail->ring[vq->last_used_idx & (vq->size - 1)]); - - /*get the number of free entries in the ring*/ - free_entries = avail_idx - vq->last_used_idx; - free_entries = unlikely(free_entries < MAX_PKT_BURST) ? free_entries : MAX_PKT_BURST; - - RTE_LOG_DP(DEBUG, VHOST_DATA, "(%" PRIu64 ") Buffers available %d\n", - dev->device_fh, free_entries); - /* Retrieve all of the head indexes first to avoid caching issues. */ - for (i = 0; i < free_entries; i++) - head[i] = vq->avail->ring[(vq->last_used_idx + i) & (vq->size - 1)]; - - /* Prefetch descriptor index. */ - rte_prefetch0(&vq->desc[head[packet_success]]); - - while (packet_success < free_entries) { - desc = &vq->desc[head[packet_success]]; - /* Prefetch descriptor address. */ - rte_prefetch0(desc); - - if (packet_success < (free_entries - 1)) { - /* Prefetch descriptor index. */ - rte_prefetch0(&vq->desc[head[packet_success+1]]); - } - - /* Update used index buffer information. */ - used_idx = vq->last_used_idx & (vq->size - 1); - vq->used->ring[used_idx].id = head[packet_success]; - vq->used->ring[used_idx].len = 0; - - /* Discard first buffer as it is the virtio header */ - desc = &vq->desc[desc->next]; - - /* Buffer address translation. */ - buff_addr = gpa_to_vva(dev, desc->addr); - /* Prefetch buffer address. */ - rte_prefetch0((void*)(uintptr_t)buff_addr); - - /* Setup dummy mbuf. This is copied to a real mbuf if transmitted out the physical port. */ - m.data_len = desc->len; - m.data_off = 0; - m.nb_segs = 1; - - virtio_tx_route(dev, &m, mbuf_pool, 0); - - vq->last_used_idx++; - packet_success++; - } - - rte_compiler_barrier(); - vq->used->idx += packet_success; - /* Kick guest if required. */ -} - -/* - * This function is called by each data core. It handles all RX/TX registered with the - * core. For TX the specific lcore linked list is used. For RX, MAC addresses are compared - * with all devices in the main linked list. - */ -static int -switch_worker(__attribute__((unused)) void *arg) -{ - struct rte_mempool *mbuf_pool = arg; - struct virtio_net *dev = NULL; - struct rte_mbuf *pkts_burst[MAX_PKT_BURST]; - struct virtio_net_data_ll *dev_ll; - struct mbuf_table *tx_q; - volatile struct lcore_ll_info *lcore_ll; - const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) / US_PER_S * BURST_TX_DRAIN_US; - uint64_t prev_tsc, diff_tsc, cur_tsc, ret_count = 0; - unsigned ret, i; - const uint16_t lcore_id = rte_lcore_id(); - const uint16_t num_cores = (uint16_t)rte_lcore_count(); - uint16_t rx_count = 0; - - RTE_LOG(INFO, VHOST_DATA, "Procesing on Core %u started \n", lcore_id); - lcore_ll = lcore_info[lcore_id].lcore_ll; - prev_tsc = 0; - - tx_q = &lcore_tx_queue[lcore_id]; - for (i = 0; i < num_cores; i ++) { - if (lcore_ids[i] == lcore_id) { - tx_q->txq_id = i; - break; - } - } - - while(1) { - cur_tsc = rte_rdtsc(); - /* - * TX burst queue drain - */ - diff_tsc = cur_tsc - prev_tsc; - if (unlikely(diff_tsc > drain_tsc)) { - - if (tx_q->len) { - RTE_LOG_DP(DEBUG, VHOST_DATA, - "TX queue drained after timeout with burst size %u\n", - tx_q->len); - - /*Tx any packets in the queue*/ - ret = rte_eth_tx_burst(ports[0], (uint16_t)tx_q->txq_id, - (struct rte_mbuf **)tx_q->m_table, - (uint16_t)tx_q->len); - if (unlikely(ret < tx_q->len)) { - do { - rte_pktmbuf_free(tx_q->m_table[ret]); - } while (++ret < tx_q->len); - } - - tx_q->len = 0; - } - - prev_tsc = cur_tsc; - - } - - /* - * Inform the configuration core that we have exited the linked list and that no devices are - * in use if requested. - */ - if (lcore_ll->dev_removal_flag == REQUEST_DEV_REMOVAL) - lcore_ll->dev_removal_flag = ACK_DEV_REMOVAL; - - /* - * Process devices - */ - dev_ll = lcore_ll->ll_root_used; - - while (dev_ll != NULL) { - /*get virtio device ID*/ - dev = dev_ll->dev; - - if (unlikely(dev->remove)) { - dev_ll = dev_ll->next; - unlink_vmdq(dev); - dev->ready = DEVICE_SAFE_REMOVE; - continue; - } - if (likely(dev->ready == DEVICE_READY)) { - /*Handle guest RX*/ - rx_count = rte_eth_rx_burst(ports[0], - (uint16_t)dev->vmdq_rx_q, pkts_burst, MAX_PKT_BURST); - - if (rx_count) { - ret_count = virtio_dev_rx(dev, pkts_burst, rx_count); - if (enable_stats) { - rte_atomic64_add(&dev_statistics[dev_ll->dev->device_fh].rx_total, rx_count); - rte_atomic64_add(&dev_statistics[dev_ll->dev->device_fh].rx, ret_count); - } - while (likely(rx_count)) { - rx_count--; - rte_pktmbuf_free_seg(pkts_burst[rx_count]); - } - - } - } - - if (likely(!dev->remove)) - /*Handle guest TX*/ - virtio_dev_tx(dev, mbuf_pool); - - /*move to the next device in the list*/ - dev_ll = dev_ll->next; - } - } - - return 0; -} - -/* - * Add an entry to a used linked list. A free entry must first be found in the free linked list - * using get_data_ll_free_entry(); - */ -static void -add_data_ll_entry(struct virtio_net_data_ll **ll_root_addr, struct virtio_net_data_ll *ll_dev) -{ - struct virtio_net_data_ll *ll = *ll_root_addr; - - /* Set next as NULL and use a compiler barrier to avoid reordering. */ - ll_dev->next = NULL; - rte_compiler_barrier(); - - /* If ll == NULL then this is the first device. */ - if (ll) { - /* Increment to the tail of the linked list. */ - while ((ll->next != NULL) ) - ll = ll->next; - - ll->next = ll_dev; - } else { - *ll_root_addr = ll_dev; - } -} - -/* - * Remove an entry from a used linked list. The entry must then be added to the free linked list - * using put_data_ll_free_entry(). - */ -static void -rm_data_ll_entry(struct virtio_net_data_ll **ll_root_addr, struct virtio_net_data_ll *ll_dev, struct virtio_net_data_ll *ll_dev_last) -{ - struct virtio_net_data_ll *ll = *ll_root_addr; - - if (ll_dev == ll) - *ll_root_addr = ll_dev->next; - else - ll_dev_last->next = ll_dev->next; -} - -/* - * Find and return an entry from the free linked list. - */ -static struct virtio_net_data_ll * -get_data_ll_free_entry(struct virtio_net_data_ll **ll_root_addr) -{ - struct virtio_net_data_ll *ll_free = *ll_root_addr; - struct virtio_net_data_ll *ll_dev; - - if (ll_free == NULL) - return NULL; - - ll_dev = ll_free; - *ll_root_addr = ll_free->next; - - return ll_dev; -} - -/* - * Place an entry back on to the free linked list. - */ -static void -put_data_ll_free_entry(struct virtio_net_data_ll **ll_root_addr, struct virtio_net_data_ll *ll_dev) -{ - struct virtio_net_data_ll *ll_free = *ll_root_addr; - - ll_dev->next = ll_free; - *ll_root_addr = ll_dev; -} - -/* - * Creates a linked list of a given size. - */ -static struct virtio_net_data_ll * -alloc_data_ll(uint32_t size) -{ - struct virtio_net_data_ll *ll_new; - uint32_t i; - - /* Malloc and then chain the linked list. */ - ll_new = malloc(size * sizeof(struct virtio_net_data_ll)); - if (ll_new == NULL) { - RTE_LOG(ERR, VHOST_CONFIG, "Failed to allocate memory for ll_new.\n"); - return NULL; - } - - for (i = 0; i < size - 1; i++) { - ll_new[i].dev = NULL; - ll_new[i].next = &ll_new[i+1]; - } - ll_new[i].next = NULL; - - return ll_new; -} - -/* - * Create the main linked list along with each individual cores linked list. A used and a free list - * are created to manage entries. - */ -static int -init_data_ll (void) -{ - int lcore; - - RTE_LCORE_FOREACH_SLAVE(lcore) { - lcore_info[lcore].lcore_ll = malloc(sizeof(struct lcore_ll_info)); - if (lcore_info[lcore].lcore_ll == NULL) { - RTE_LOG(ERR, VHOST_CONFIG, "Failed to allocate memory for lcore_ll.\n"); - return -1; - } - - lcore_info[lcore].lcore_ll->device_num = 0; - lcore_info[lcore].lcore_ll->dev_removal_flag = ACK_DEV_REMOVAL; - lcore_info[lcore].lcore_ll->ll_root_used = NULL; - if (num_devices % num_switching_cores) - lcore_info[lcore].lcore_ll->ll_root_free = alloc_data_ll((num_devices / num_switching_cores) + 1); - else - lcore_info[lcore].lcore_ll->ll_root_free = alloc_data_ll(num_devices / num_switching_cores); - } - - /* Allocate devices up to a maximum of MAX_DEVICES. */ - ll_root_free = alloc_data_ll(MIN((num_devices), MAX_DEVICES)); - - return 0; -} -/* - * Remove a device from the specific data core linked list and from the main linked list. The - * rx/tx thread must be set the flag to indicate that it is safe to remove the device. - * used. - */ -static void -destroy_device (volatile struct virtio_net *dev) -{ - struct virtio_net_data_ll *ll_lcore_dev_cur; - struct virtio_net_data_ll *ll_main_dev_cur; - struct virtio_net_data_ll *ll_lcore_dev_last = NULL; - struct virtio_net_data_ll *ll_main_dev_last = NULL; - int lcore; - - dev->flags &= ~VIRTIO_DEV_RUNNING; - - /*set the remove flag. */ - dev->remove = 1; - - while(dev->ready != DEVICE_SAFE_REMOVE) { - rte_pause(); - } - - /* Search for entry to be removed from lcore ll */ - ll_lcore_dev_cur = lcore_info[dev->coreid].lcore_ll->ll_root_used; - while (ll_lcore_dev_cur != NULL) { - if (ll_lcore_dev_cur->dev == dev) { - break; - } else { - ll_lcore_dev_last = ll_lcore_dev_cur; - ll_lcore_dev_cur = ll_lcore_dev_cur->next; - } - } - - /* Search for entry to be removed from main ll */ - ll_main_dev_cur = ll_root_used; - ll_main_dev_last = NULL; - while (ll_main_dev_cur != NULL) { - if (ll_main_dev_cur->dev == dev) { - break; - } else { - ll_main_dev_last = ll_main_dev_cur; - ll_main_dev_cur = ll_main_dev_cur->next; - } - } - - if (ll_lcore_dev_cur == NULL || ll_main_dev_cur == NULL) { - RTE_LOG(ERR, XENHOST, "%s: could find device in per_cpu list or main_list\n", __func__); - return; - } - - /* Remove entries from the lcore and main ll. */ - rm_data_ll_entry(&lcore_info[ll_lcore_dev_cur->dev->coreid].lcore_ll->ll_root_used, ll_lcore_dev_cur, ll_lcore_dev_last); - rm_data_ll_entry(&ll_root_used, ll_main_dev_cur, ll_main_dev_last); - - /* Set the dev_removal_flag on each lcore. */ - RTE_LCORE_FOREACH_SLAVE(lcore) { - lcore_info[lcore].lcore_ll->dev_removal_flag = REQUEST_DEV_REMOVAL; - } - - /* - * Once each core has set the dev_removal_flag to ACK_DEV_REMOVAL we can be sure that - * they can no longer access the device removed from the linked lists and that the devices - * are no longer in use. - */ - RTE_LCORE_FOREACH_SLAVE(lcore) { - while (lcore_info[lcore].lcore_ll->dev_removal_flag != ACK_DEV_REMOVAL) { - rte_pause(); - } - } - - /* Add the entries back to the lcore and main free ll.*/ - put_data_ll_free_entry(&lcore_info[ll_lcore_dev_cur->dev->coreid].lcore_ll->ll_root_free, ll_lcore_dev_cur); - put_data_ll_free_entry(&ll_root_free, ll_main_dev_cur); - - /* Decrement number of device on the lcore. */ - lcore_info[ll_lcore_dev_cur->dev->coreid].lcore_ll->device_num--; - - RTE_LOG(INFO, VHOST_DATA, " #####(%"PRIu64") Device has been removed from data core\n", dev->device_fh); -} - -/* - * A new device is added to a data core. First the device is added to the main linked list - * and the allocated to a specific data core. - */ -static int -new_device (struct virtio_net *dev) -{ - struct virtio_net_data_ll *ll_dev; - int lcore, core_add = 0; - uint32_t device_num_min = num_devices; - - /* Add device to main ll */ - ll_dev = get_data_ll_free_entry(&ll_root_free); - if (ll_dev == NULL) { - RTE_LOG(INFO, VHOST_DATA, "(%"PRIu64") No free entry found in linked list. Device limit " - "of %d devices per core has been reached\n", - dev->device_fh, num_devices); - return -1; - } - ll_dev->dev = dev; - add_data_ll_entry(&ll_root_used, ll_dev); - - /*reset ready flag*/ - dev->ready = DEVICE_NOT_READY; - dev->remove = 0; - - /* Find a suitable lcore to add the device. */ - RTE_LCORE_FOREACH_SLAVE(lcore) { - if (lcore_info[lcore].lcore_ll->device_num < device_num_min) { - device_num_min = lcore_info[lcore].lcore_ll->device_num; - core_add = lcore; - } - } - /* Add device to lcore ll */ - ll_dev->dev->coreid = core_add; - ll_dev = get_data_ll_free_entry(&lcore_info[ll_dev->dev->coreid].lcore_ll->ll_root_free); - if (ll_dev == NULL) { - RTE_LOG(INFO, VHOST_DATA, "(%"PRIu64") Failed to add device to data core\n", dev->device_fh); - destroy_device(dev); - return -1; - } - ll_dev->dev = dev; - add_data_ll_entry(&lcore_info[ll_dev->dev->coreid].lcore_ll->ll_root_used, ll_dev); - - /* Initialize device stats */ - memset(&dev_statistics[dev->device_fh], 0, sizeof(struct device_statistics)); - - lcore_info[ll_dev->dev->coreid].lcore_ll->device_num++; - dev->flags |= VIRTIO_DEV_RUNNING; - - RTE_LOG(INFO, VHOST_DATA, "(%"PRIu64") Device has been added to data core %d\n", dev->device_fh, dev->coreid); - - link_vmdq(dev); - - return 0; -} - -/* - * These callback allow devices to be added to the data core when configuration - * has been fully complete. - */ -static const struct virtio_net_device_ops virtio_net_device_ops = -{ - .new_device = new_device, - .destroy_device = destroy_device, -}; - -/* - * This is a thread will wake up after a period to print stats if the user has - * enabled them. - */ -static void -print_stats(void) -{ - struct virtio_net_data_ll *dev_ll; - uint64_t tx_dropped, rx_dropped; - uint64_t tx, tx_total, rx, rx_total; - uint32_t device_fh; - const char clr[] = { 27, '[', '2', 'J', '\0' }; - const char top_left[] = { 27, '[', '1', ';', '1', 'H','\0' }; - - while(1) { - sleep(enable_stats); - - /* Clear screen and move to top left */ - printf("%s%s", clr, top_left); - - printf("\nDevice statistics ===================================="); - - dev_ll = ll_root_used; - while (dev_ll != NULL) { - device_fh = (uint32_t)dev_ll->dev->device_fh; - tx_total = dev_statistics[device_fh].tx_total; - tx = dev_statistics[device_fh].tx; - tx_dropped = tx_total - tx; - rx_total = rte_atomic64_read(&dev_statistics[device_fh].rx_total); - rx = rte_atomic64_read(&dev_statistics[device_fh].rx); - rx_dropped = rx_total - rx; - - printf("\nStatistics for device %"PRIu32" ------------------------------" - "\nTX total: %"PRIu64"" - "\nTX dropped: %"PRIu64"" - "\nTX successful: %"PRIu64"" - "\nRX total: %"PRIu64"" - "\nRX dropped: %"PRIu64"" - "\nRX successful: %"PRIu64"", - device_fh, - tx_total, - tx_dropped, - tx, - rx_total, - rx_dropped, - rx); - - dev_ll = dev_ll->next; - } - printf("\n======================================================\n"); - } -} - - -int init_virtio_net(struct virtio_net_device_ops const * const ops); - -/* - * Main function, does initialisation and calls the per-lcore functions. - */ -int -main(int argc, char *argv[]) -{ - struct rte_mempool *mbuf_pool; - unsigned lcore_id, core_id = 0; - unsigned nb_ports, valid_num_ports; - int ret; - uint8_t portid; - static pthread_t tid; - char thread_name[RTE_MAX_THREAD_NAME_LEN]; - - /* init EAL */ - ret = rte_eal_init(argc, argv); - if (ret < 0) - rte_exit(EXIT_FAILURE, "Error with EAL initialization\n"); - argc -= ret; - argv += ret; - - /* parse app arguments */ - ret = us_vhost_parse_args(argc, argv); - if (ret < 0) - rte_exit(EXIT_FAILURE, "Invalid argument\n"); - - for (lcore_id = 0; lcore_id < RTE_MAX_LCORE; lcore_id ++) - if (rte_lcore_is_enabled(lcore_id)) - lcore_ids[core_id ++] = lcore_id; - - if (rte_lcore_count() > RTE_MAX_LCORE) - rte_exit(EXIT_FAILURE,"Not enough cores\n"); - - /*set the number of swithcing cores available*/ - num_switching_cores = rte_lcore_count()-1; - - /* Get the number of physical ports. */ - nb_ports = rte_eth_dev_count(); - - /* - * Update the global var NUM_PORTS and global array PORTS - * and get value of var VALID_NUM_PORTS according to system ports number - */ - valid_num_ports = check_ports_num(nb_ports); - - if ((valid_num_ports == 0) || (valid_num_ports > MAX_SUP_PORTS)) { - RTE_LOG(INFO, VHOST_PORT, "Current enabled port number is %u," - "but only %u port can be enabled\n",num_ports, MAX_SUP_PORTS); - return -1; - } - - /* Create the mbuf pool. */ - mbuf_pool = rte_pktmbuf_pool_create("MBUF_POOL", - NUM_MBUFS_PER_PORT * valid_num_ports, MBUF_CACHE_SIZE, 0, - RTE_MBUF_DEFAULT_BUF_SIZE, rte_socket_id()); - if (mbuf_pool == NULL) - rte_exit(EXIT_FAILURE, "Cannot create mbuf pool\n"); - - /* initialize all ports */ - for (portid = 0; portid < nb_ports; portid++) { - /* skip ports that are not enabled */ - if ((enabled_port_mask & (1 << portid)) == 0) { - RTE_LOG(INFO, VHOST_PORT, "Skipping disabled port %d\n", portid); - continue; - } - if (port_init(portid, mbuf_pool) != 0) - rte_exit(EXIT_FAILURE, "Cannot initialize network ports\n"); - } - - /* Initialise all linked lists. */ - if (init_data_ll() == -1) - rte_exit(EXIT_FAILURE, "Failed to initialize linked list\n"); - - /* Initialize device stats */ - memset(&dev_statistics, 0, sizeof(dev_statistics)); - - /* Enable stats if the user option is set. */ - if (enable_stats) { - ret = pthread_create(&tid, NULL, (void *)print_stats, NULL); - if (ret != 0) - rte_exit(EXIT_FAILURE, - "Cannot create print-stats thread\n"); - - /* Set thread_name for aid in debugging. */ - snprintf(thread_name, RTE_MAX_THREAD_NAME_LEN, "print-xen-stats"); - ret = rte_thread_setname(tid, thread_name); - if (ret != 0) - RTE_LOG(DEBUG, VHOST_CONFIG, - "Cannot set print-stats name\n"); - } - - /* Launch all data cores. */ - RTE_LCORE_FOREACH_SLAVE(lcore_id) { - rte_eal_remote_launch(switch_worker, mbuf_pool, lcore_id); - } - - init_virtio_xen(&virtio_net_device_ops); - - virtio_monitor_loop(); - return 0; -} diff --git a/examples/vhost_xen/main.h b/examples/vhost_xen/main.h deleted file mode 100644 index 5ff48fd9..00000000 --- a/examples/vhost_xen/main.h +++ /dev/null @@ -1,66 +0,0 @@ -/*- - * BSD LICENSE - * - * Copyright(c) 2010-2014 Intel Corporation. All rights reserved. - * All rights reserved. - * - * 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. - * * Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - */ - -#ifndef _MAIN_H_ -#define _MAIN_H_ - -/* Macros for printing using RTE_LOG */ -#define RTE_LOGTYPE_VHOST_CONFIG RTE_LOGTYPE_USER1 -#define RTE_LOGTYPE_VHOST_DATA RTE_LOGTYPE_USER2 -#define RTE_LOGTYPE_VHOST_PORT RTE_LOGTYPE_USER3 - -/* - * Device linked list structure for data path. - */ -struct virtio_net_data_ll -{ - struct virtio_net *dev; /* Pointer to device created by configuration core. */ - struct virtio_net_data_ll *next; /* Pointer to next device in linked list. */ -}; - -/* - * Structure containing data core specific information. - */ -struct lcore_ll_info -{ - struct virtio_net_data_ll *ll_root_free; /* Pointer to head in free linked list. */ - struct virtio_net_data_ll *ll_root_used; /* Pointer to head of used linked list. */ - uint32_t device_num; /* Number of devices on lcore. */ - volatile uint8_t dev_removal_flag; /* Flag to synchronize device removal. */ -}; - -struct lcore_info -{ - struct lcore_ll_info *lcore_ll; /* Pointer to data core specific lcore_ll_info struct */ -}; -#endif /* _MAIN_H_ */ diff --git a/examples/vhost_xen/vhost_monitor.c b/examples/vhost_xen/vhost_monitor.c deleted file mode 100644 index fb9606bf..00000000 --- a/examples/vhost_xen/vhost_monitor.c +++ /dev/null @@ -1,595 +0,0 @@ -/*- - * BSD LICENSE - * - * Copyright(c) 2010-2014 Intel Corporation. All rights reserved. - * All rights reserved. - * - * 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. - * * Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - */ - -#include <stdlib.h> -#include <stdio.h> -#include <dirent.h> -#include <unistd.h> -#include <sys/eventfd.h> -#include <sys/ioctl.h> -#include <sys/mman.h> -#include <xen/xen-compat.h> -#if __XEN_LATEST_INTERFACE_VERSION__ < 0x00040200 -#include <xs.h> -#else -#include <xenstore.h> -#endif -#include <linux/virtio_ring.h> -#include <linux/virtio_pci.h> -#include <linux/virtio_net.h> - -#include <rte_ethdev.h> -#include <rte_log.h> -#include <rte_malloc.h> -#include <rte_string_fns.h> - -#include "virtio-net.h" -#include "xen_vhost.h" - -struct virtio_watch { - struct xs_handle *xs; - int watch_fd; -}; - - -/* device ops to add/remove device to/from data core. */ -static struct virtio_net_device_ops const *notify_ops; - -/* root address of the linked list in the configuration core. */ -static struct virtio_net_config_ll *ll_root = NULL; - -/* root address of VM. */ -static struct xen_guestlist guest_root; - -static struct virtio_watch watch; - -static void -vq_vring_init(struct vhost_virtqueue *vq, unsigned int num, uint8_t *p, - unsigned long align) -{ - vq->size = num; - vq->desc = (struct vring_desc *) p; - vq->avail = (struct vring_avail *) (p + - num * sizeof(struct vring_desc)); - vq->used = (void *) - RTE_ALIGN_CEIL( (uintptr_t)(&vq->avail->ring[num]), align); - -} - -static int -init_watch(void) -{ - struct xs_handle *xs; - int ret; - int fd; - - /* get a connection to the daemon */ - xs = xs_daemon_open(); - if (xs == NULL) { - RTE_LOG(ERR, XENHOST, "xs_daemon_open failed\n"); - return -1; - } - - ret = xs_watch(xs, "/local/domain", "mytoken"); - if (ret == 0) { - RTE_LOG(ERR, XENHOST, "%s: xs_watch failed\n", __func__); - xs_daemon_close(xs); - return -1; - } - - /* We are notified of read availability on the watch via the file descriptor. */ - fd = xs_fileno(xs); - watch.xs = xs; - watch.watch_fd = fd; - - TAILQ_INIT(&guest_root); - return 0; -} - -static struct xen_guest * -get_xen_guest(int dom_id) -{ - struct xen_guest *guest = NULL; - - TAILQ_FOREACH(guest, &guest_root, next) { - if(guest->dom_id == dom_id) - return guest; - } - - return NULL; -} - - -static struct xen_guest * -add_xen_guest(int32_t dom_id) -{ - struct xen_guest *guest = NULL; - - if ((guest = get_xen_guest(dom_id)) != NULL) - return guest; - - guest = calloc(1, sizeof(struct xen_guest)); - if (guest) { - RTE_LOG(ERR, XENHOST, " %s: return newly created guest with %d rings\n", __func__, guest->vring_num); - TAILQ_INSERT_TAIL(&guest_root, guest, next); - guest->dom_id = dom_id; - } - - return guest; -} - -static void -cleanup_device(struct virtio_net_config_ll *ll_dev) -{ - if (ll_dev == NULL) - return; - if (ll_dev->dev.virtqueue_rx) { - rte_free(ll_dev->dev.virtqueue_rx); - ll_dev->dev.virtqueue_rx = NULL; - } - if (ll_dev->dev.virtqueue_tx) { - rte_free(ll_dev->dev.virtqueue_tx); - ll_dev->dev.virtqueue_tx = NULL; - } - free(ll_dev); -} - -/* - * Add entry containing a device to the device configuration linked list. - */ -static void -add_config_ll_entry(struct virtio_net_config_ll *new_ll_dev) -{ - struct virtio_net_config_ll *ll_dev = ll_root; - - /* If ll_dev == NULL then this is the first device so go to else */ - if (ll_dev) { - /* If the 1st device_id != 0 then we insert our device here. */ - if (ll_dev->dev.device_fh != 0) { - new_ll_dev->dev.device_fh = 0; - new_ll_dev->next = ll_dev; - ll_root = new_ll_dev; - } else { - /* increment through the ll until we find un unused device_id, - * insert the device at that entry - */ - while ((ll_dev->next != NULL) && (ll_dev->dev.device_fh == (ll_dev->next->dev.device_fh - 1))) - ll_dev = ll_dev->next; - - new_ll_dev->dev.device_fh = ll_dev->dev.device_fh + 1; - new_ll_dev->next = ll_dev->next; - ll_dev->next = new_ll_dev; - } - } else { - ll_root = new_ll_dev; - ll_root->dev.device_fh = 0; - } -} - - -/* - * Remove an entry from the device configuration linked list. - */ -static struct virtio_net_config_ll * -rm_config_ll_entry(struct virtio_net_config_ll *ll_dev, struct virtio_net_config_ll *ll_dev_last) -{ - /* First remove the device and then clean it up. */ - if (ll_dev == ll_root) { - ll_root = ll_dev->next; - cleanup_device(ll_dev); - return ll_root; - } else { - ll_dev_last->next = ll_dev->next; - cleanup_device(ll_dev); - return ll_dev_last->next; - } -} - -/* - * Retrieves an entry from the devices configuration linked list. - */ -static struct virtio_net_config_ll * -get_config_ll_entry(unsigned int virtio_idx, unsigned int dom_id) -{ - struct virtio_net_config_ll *ll_dev = ll_root; - - /* Loop through linked list until the dom_id is found. */ - while (ll_dev != NULL) { - if (ll_dev->dev.dom_id == dom_id && ll_dev->dev.virtio_idx == virtio_idx) - return ll_dev; - ll_dev = ll_dev->next; - } - - return NULL; -} - -/* - * Initialise all variables in device structure. - */ -static void -init_dev(struct virtio_net *dev) -{ - RTE_SET_USED(dev); -} - - -static struct -virtio_net_config_ll *new_device(unsigned int virtio_idx, struct xen_guest *guest) -{ - struct virtio_net_config_ll *new_ll_dev; - struct vhost_virtqueue *virtqueue_rx, *virtqueue_tx; - size_t size, vq_ring_size, vq_size = VQ_DESC_NUM; - void *vq_ring_virt_mem; - uint64_t gpa; - uint32_t i; - - /* Setup device and virtqueues. */ - new_ll_dev = calloc(1, sizeof(struct virtio_net_config_ll)); - virtqueue_rx = rte_zmalloc(NULL, sizeof(struct vhost_virtqueue), RTE_CACHE_LINE_SIZE); - virtqueue_tx = rte_zmalloc(NULL, sizeof(struct vhost_virtqueue), RTE_CACHE_LINE_SIZE); - if (new_ll_dev == NULL || virtqueue_rx == NULL || virtqueue_tx == NULL) - goto err; - - new_ll_dev->dev.virtqueue_rx = virtqueue_rx; - new_ll_dev->dev.virtqueue_tx = virtqueue_tx; - new_ll_dev->dev.dom_id = guest->dom_id; - new_ll_dev->dev.virtio_idx = virtio_idx; - /* Initialise device and virtqueues. */ - init_dev(&new_ll_dev->dev); - - size = vring_size(vq_size, VIRTIO_PCI_VRING_ALIGN); - vq_ring_size = RTE_ALIGN_CEIL(size, VIRTIO_PCI_VRING_ALIGN); - (void)vq_ring_size; - - vq_ring_virt_mem = guest->vring[virtio_idx].rxvring_addr; - vq_vring_init(virtqueue_rx, vq_size, vq_ring_virt_mem, VIRTIO_PCI_VRING_ALIGN); - virtqueue_rx->size = vq_size; - virtqueue_rx->vhost_hlen = sizeof(struct virtio_net_hdr); - - vq_ring_virt_mem = guest->vring[virtio_idx].txvring_addr; - vq_vring_init(virtqueue_tx, vq_size, vq_ring_virt_mem, VIRTIO_PCI_VRING_ALIGN); - virtqueue_tx->size = vq_size; - memcpy(&new_ll_dev->dev.mac_address, &guest->vring[virtio_idx].addr, sizeof(struct ether_addr)); - - /* virtio_memory has to be one per domid */ - new_ll_dev->dev.mem = malloc(sizeof(struct virtio_memory) + sizeof(struct virtio_memory_regions) * MAX_XENVIRT_MEMPOOL); - new_ll_dev->dev.mem->nregions = guest->pool_num; - for (i = 0; i < guest->pool_num; i++) { - gpa = new_ll_dev->dev.mem->regions[i].guest_phys_address = - (uint64_t)((uintptr_t)guest->mempool[i].gva); - new_ll_dev->dev.mem->regions[i].guest_phys_address_end = - gpa + guest->mempool[i].mempfn_num * getpagesize(); - new_ll_dev->dev.mem->regions[i].address_offset = - (uint64_t)((uintptr_t)guest->mempool[i].hva - - (uintptr_t)gpa); - } - - new_ll_dev->next = NULL; - - /* Add entry to device configuration linked list. */ - add_config_ll_entry(new_ll_dev); - return new_ll_dev; -err: - free(new_ll_dev); - rte_free(virtqueue_rx); - rte_free(virtqueue_tx); - - return NULL; -} - -static void -destroy_guest(struct xen_guest *guest) -{ - uint32_t i; - - for (i = 0; i < guest->vring_num; i++) - cleanup_vring(&guest->vring[i]); - /* clean mempool */ - for (i = 0; i < guest->pool_num; i++) - cleanup_mempool(&guest->mempool[i]); - free(guest); - - return; -} - -/* - * This function will cleanup the device and remove it from device configuration linked list. - */ -static void -destroy_device(unsigned int virtio_idx, unsigned int dom_id) -{ - struct virtio_net_config_ll *ll_dev_cur_ctx, *ll_dev_last = NULL; - struct virtio_net_config_ll *ll_dev_cur = ll_root; - - /* clean virtio device */ - struct xen_guest *guest = NULL; - guest = get_xen_guest(dom_id); - if (guest == NULL) - return; - - /* Find the linked list entry for the device to be removed. */ - ll_dev_cur_ctx = get_config_ll_entry(virtio_idx, dom_id); - while (ll_dev_cur != NULL) { - /* If the device is found or a device that doesn't exist is found then it is removed. */ - if (ll_dev_cur == ll_dev_cur_ctx) { - if ((ll_dev_cur->dev.flags & VIRTIO_DEV_RUNNING)) - notify_ops->destroy_device(&(ll_dev_cur->dev)); - ll_dev_cur = rm_config_ll_entry(ll_dev_cur, ll_dev_last); - } else { - ll_dev_last = ll_dev_cur; - ll_dev_cur = ll_dev_cur->next; - } - } - RTE_LOG(INFO, XENHOST, " %s guest:%p vring:%p rxvring:%p txvring:%p flag:%p\n", - __func__, guest, &guest->vring[virtio_idx], guest->vring[virtio_idx].rxvring_addr, guest->vring[virtio_idx].txvring_addr, guest->vring[virtio_idx].flag); - cleanup_vring(&guest->vring[virtio_idx]); - guest->vring[virtio_idx].removed = 1; - guest->vring_num -= 1; -} - - - - -static void -watch_unmap_event(void) -{ - int i; - struct xen_guest *guest = NULL; - bool remove_request; - - TAILQ_FOREACH(guest, &guest_root, next) { - for (i = 0; i < MAX_VIRTIO; i++) { - if (guest->vring[i].dom_id && guest->vring[i].removed == 0 && *guest->vring[i].flag == 0) { - RTE_LOG(INFO, XENHOST, "\n\n"); - RTE_LOG(INFO, XENHOST, " #####%s: (%d, %d) to be removed\n", - __func__, - guest->vring[i].dom_id, - i); - destroy_device(i, guest->dom_id); - RTE_LOG(INFO, XENHOST, " %s: DOM %u, vring num: %d\n", - __func__, - guest->dom_id, - guest->vring_num); - } - } - } - -_find_next_remove: - guest = NULL; - remove_request = false; - TAILQ_FOREACH(guest, &guest_root, next) { - if (guest->vring_num == 0) { - remove_request = true; - break; - } - } - if (remove_request == true) { - TAILQ_REMOVE(&guest_root, guest, next); - RTE_LOG(INFO, XENHOST, " #####%s: destroy guest (%d)\n", __func__, guest->dom_id); - destroy_guest(guest); - goto _find_next_remove; - } - return; -} - -/* - * OK, if the guest starts first, it is ok. - * if host starts first, it is ok. - * if guest starts, and has run for sometime, and host stops and restarts, - * then last_used_idx 0? how to solve this. */ - -static void virtio_init(void) -{ - uint32_t len, e_num; - uint32_t i,j; - char **dom; - char *status; - int dom_id; - char path[PATH_MAX]; - char node[PATH_MAX]; - xs_transaction_t th; - struct xen_guest *guest; - struct virtio_net_config_ll *net_config; - char *end; - int val; - - /* init env for watch the node */ - if (init_watch() < 0) - return; - - dom = xs_directory(watch.xs, XBT_NULL, "/local/domain", &e_num); - - for (i = 0; i < e_num; i++) { - errno = 0; - dom_id = strtol(dom[i], &end, 0); - if (errno != 0 || end == NULL || dom_id == 0) - continue; - - for (j = 0; j < RTE_MAX_ETHPORTS; j++) { - snprintf(node, PATH_MAX, "%s%d", VIRTIO_START, j); - snprintf(path, PATH_MAX, XEN_VM_NODE_FMT, - dom_id, node); - - th = xs_transaction_start(watch.xs); - status = xs_read(watch.xs, th, path, &len); - xs_transaction_end(watch.xs, th, false); - - if (status == NULL) - break; - - /* if there's any valid virtio device */ - errno = 0; - val = strtol(status, &end, 0); - if (errno != 0 || end == NULL || dom_id == 0) - val = 0; - if (val == 1) { - guest = add_xen_guest(dom_id); - if (guest == NULL) - continue; - RTE_LOG(INFO, XENHOST, " there's a new virtio existed, new a virtio device\n\n"); - - RTE_LOG(INFO, XENHOST, " parse_vringnode dom_id %d virtioidx %d\n",dom_id,j); - if (parse_vringnode(guest, j)) { - RTE_LOG(ERR, XENHOST, " there is invalid information in xenstore\n"); - TAILQ_REMOVE(&guest_root, guest, next); - destroy_guest(guest); - - continue; - } - - /*if pool_num > 0, then mempool has already been parsed*/ - if (guest->pool_num == 0 && parse_mempoolnode(guest)) { - RTE_LOG(ERR, XENHOST, " there is error information in xenstore\n"); - TAILQ_REMOVE(&guest_root, guest, next); - destroy_guest(guest); - continue; - } - - net_config = new_device(j, guest); - /* every thing is ready now, added into data core */ - notify_ops->new_device(&net_config->dev); - } - } - } - - free(dom); - return; -} - -void -virtio_monitor_loop(void) -{ - char **vec; - xs_transaction_t th; - char *buf; - unsigned int len; - unsigned int dom_id; - uint32_t virtio_idx; - struct xen_guest *guest; - struct virtio_net_config_ll *net_config; - enum fieldnames { - FLD_NULL = 0, - FLD_LOCAL, - FLD_DOMAIN, - FLD_ID, - FLD_CONTROL, - FLD_DPDK, - FLD_NODE, - _NUM_FLD - }; - char *str_fld[_NUM_FLD]; - char *str; - char *end; - - virtio_init(); - while (1) { - watch_unmap_event(); - - usleep(50); - vec = xs_check_watch(watch.xs); - - if (vec == NULL) - continue; - - th = xs_transaction_start(watch.xs); - - buf = xs_read(watch.xs, th, vec[XS_WATCH_PATH],&len); - xs_transaction_end(watch.xs, th, false); - - if (buf) { - /* theres' some node for vhost existed */ - if (rte_strsplit(vec[XS_WATCH_PATH], strnlen(vec[XS_WATCH_PATH], PATH_MAX), - str_fld, _NUM_FLD, '/') == _NUM_FLD) { - if (strstr(str_fld[FLD_NODE], VIRTIO_START)) { - errno = 0; - str = str_fld[FLD_ID]; - dom_id = strtoul(str, &end, 0); - if (errno != 0 || end == NULL || end == str ) { - RTE_LOG(INFO, XENHOST, "invalid domain id\n"); - continue; - } - - errno = 0; - str = str_fld[FLD_NODE] + sizeof(VIRTIO_START) - 1; - virtio_idx = strtoul(str, &end, 0); - if (errno != 0 || end == NULL || end == str - || virtio_idx > MAX_VIRTIO) { - RTE_LOG(INFO, XENHOST, "invalid virtio idx\n"); - continue; - } - RTE_LOG(INFO, XENHOST, " #####virtio dev (%d, %d) is started\n", dom_id, virtio_idx); - - guest = add_xen_guest(dom_id); - if (guest == NULL) - continue; - guest->dom_id = dom_id; - if (parse_vringnode(guest, virtio_idx)) { - RTE_LOG(ERR, XENHOST, " there is invalid information in xenstore\n"); - /*guest newly created? guest existed ?*/ - TAILQ_REMOVE(&guest_root, guest, next); - destroy_guest(guest); - continue; - } - /*if pool_num > 0, then mempool has already been parsed*/ - if (guest->pool_num == 0 && parse_mempoolnode(guest)) { - RTE_LOG(ERR, XENHOST, " there is error information in xenstore\n"); - TAILQ_REMOVE(&guest_root, guest, next); - destroy_guest(guest); - continue; - } - - - net_config = new_device(virtio_idx, guest); - RTE_LOG(INFO, XENHOST, " Add to dataplane core\n"); - notify_ops->new_device(&net_config->dev); - - } - } - } - - free(vec); - } - return; -} - -/* - * Register ops so that we can add/remove device to data core. - */ -int -init_virtio_xen(struct virtio_net_device_ops const *const ops) -{ - notify_ops = ops; - if (xenhost_init()) - return -1; - return 0; -} diff --git a/examples/vhost_xen/virtio-net.h b/examples/vhost_xen/virtio-net.h deleted file mode 100644 index ab697260..00000000 --- a/examples/vhost_xen/virtio-net.h +++ /dev/null @@ -1,113 +0,0 @@ -/*- - * BSD LICENSE - * - * Copyright(c) 2010-2014 Intel Corporation. All rights reserved. - * All rights reserved. - * - * 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. - * * Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - */ - -#ifndef _VIRTIO_NET_H_ -#define _VIRTIO_NET_H_ - -#include <stdint.h> - -#define VQ_DESC_NUM 256 -/* Used to indicate that the device is running on a data core */ -#define VIRTIO_DEV_RUNNING 1 - -/* - * Structure contains variables relevant to TX/RX virtqueues. - */ -struct vhost_virtqueue -{ - struct vring_desc *desc; /* Virtqueue descriptor ring. */ - struct vring_avail *avail; /* Virtqueue available ring. */ - struct vring_used *used; /* Virtqueue used ring. */ - uint32_t size; /* Size of descriptor ring. */ - uint32_t vhost_hlen; /* Vhost header length (varies depending on RX merge buffers. */ - volatile uint16_t last_used_idx; /* Last index used on the available ring */ - volatile uint16_t last_used_idx_res; /* Used for multiple devices reserving buffers. */ -} __rte_cache_aligned; - -/* - * Device structure contains all configuration information relating to the device. - */ -struct virtio_net -{ - struct vhost_virtqueue *virtqueue_tx; /* Contains all TX virtqueue information. */ - struct vhost_virtqueue *virtqueue_rx; /* Contains all RX virtqueue information. */ - struct virtio_memory *mem; /* QEMU memory and memory region information. */ - struct ether_addr mac_address; /* Device MAC address (Obtained on first TX packet). */ - uint32_t flags; /* Device flags. Only used to check if device is running on data core. */ - uint32_t vlan_tag; /* Vlan tag for device. Currently set to device_id (0-63). */ - uint32_t vmdq_rx_q; - uint64_t device_fh; /* device identifier. */ - uint16_t coreid; - volatile uint8_t ready; /* A device is set as ready if the MAC address has been set. */ - volatile uint8_t remove; /* Device is marked for removal from the data core. */ - uint32_t virtio_idx; /* Index of virtio device */ - uint32_t dom_id; /* Domain id of xen guest */ -} ___rte_cache_aligned; - -/* - * Device linked list structure for configuration. - */ -struct virtio_net_config_ll -{ - struct virtio_net dev; /* Virtio device. */ - struct virtio_net_config_ll *next; /* Next entry on linked list. */ -}; - -/* - * Information relating to memory regions including offsets to addresses in QEMUs memory file. - */ -struct virtio_memory_regions { - uint64_t guest_phys_address; /* Base guest physical address of region. */ - uint64_t guest_phys_address_end; /* End guest physical address of region. */ - uint64_t memory_size; /* Size of region. */ - uint64_t userspace_address; /* Base userspace address of region. */ - uint64_t address_offset; /* Offset of region for address translation. */ -}; - -/* - * Memory structure includes region and mapping information. - */ -struct virtio_memory { - uint32_t nregions; /* Number of memory regions. */ - struct virtio_memory_regions regions[0]; /* Memory region information. */ -}; - -/* - * Device operations to add/remove device. - */ -struct virtio_net_device_ops { - int (* new_device)(struct virtio_net *); /* Add device. */ - void (* destroy_device) (volatile struct virtio_net *); /* Remove device. */ -}; - -#endif diff --git a/examples/vhost_xen/xen_vhost.h b/examples/vhost_xen/xen_vhost.h deleted file mode 100644 index 2fc304c7..00000000 --- a/examples/vhost_xen/xen_vhost.h +++ /dev/null @@ -1,148 +0,0 @@ -/*- - * BSD LICENSE - * - * Copyright(c) 2010-2014 Intel Corporation. All rights reserved. - * All rights reserved. - * - * 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. - * * Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - */ - -#ifndef _XEN_VHOST_H_ -#define _XEN_VHOST_H_ - -#include <stdint.h> - -#include <rte_ether.h> - -#include "virtio-net.h" - -#define RTE_LOGTYPE_XENHOST RTE_LOGTYPE_USER1 - -#define XEN_VM_ROOTNODE_FMT "/local/domain/%d/control/dpdk" -#define XEN_VM_NODE_FMT "/local/domain/%d/control/dpdk/%s" -#define XEN_MEMPOOL_SUFFIX "mempool_gref" -#define XEN_RXVRING_SUFFIX "rx_vring_gref" -#define XEN_TXVRING_SUFFIX "tx_vring_gref" -#define XEN_GVA_SUFFIX "mempool_va" -#define XEN_VRINGFLAG_SUFFIX "vring_flag" -#define XEN_ADDR_SUFFIX "ether_addr" -#define VIRTIO_START "event_type_start_" - -#define XEN_GREF_SPLITTOKEN ',' - -#define MAX_XENVIRT_MEMPOOL 16 -#define MAX_VIRTIO 32 -#define MAX_GREF_PER_NODE 64 /* 128 MB memory */ - -#define PAGE_SIZE 4096 -#define PAGE_PFNNUM (PAGE_SIZE / sizeof(uint32_t)) - -#define XEN_GNTDEV_FNAME "/dev/xen/gntdev" - -/* xen grant reference info in one grant node */ -struct xen_gnt { - uint32_t gref; /* grant reference for this node */ - union { - int gref; /* grant reference */ - uint32_t pfn_num; /* guest pfn number of grant reference */ - } gref_pfn[PAGE_PFNNUM]; -}__attribute__((__packed__)); - - -/* structure for mempool or vring node list */ -struct xen_gntnode { - uint32_t gnt_num; /* grant reference number */ - struct xen_gnt *gnt_info; /* grant reference info */ -}; - - -struct xen_vring { - uint32_t dom_id; - uint32_t virtio_idx; /* index of virtio device */ - void *rxvring_addr; /* mapped virtual address of rxvring */ - void *txvring_addr; /* mapped virtual address of txvring */ - uint32_t rxpfn_num; /* number of gpfn for rxvring */ - uint32_t txpfn_num; /* number of gpfn for txvring */ - uint32_t *rxpfn_tbl; /* array of rxvring gpfn */ - uint32_t *txpfn_tbl; /* array of txvring gpfn */ - uint64_t *rx_pindex; /* index used to release rx grefs */ - uint64_t *tx_pindex; /* index used to release tx grefs */ - uint64_t flag_index; - uint8_t *flag; /* cleared to zero on guest unmap */ - struct ether_addr addr; /* ethernet address of virtio device */ - uint8_t removed; - -}; - -struct xen_mempool { - uint32_t dom_id; /* guest domain id */ - uint32_t pool_idx; /* index of memory pool */ - void *gva; /* guest virtual address of mbuf pool */ - void *hva; /* host virtual address of mbuf pool */ - uint32_t mempfn_num; /* number of gpfn for mbuf pool */ - uint32_t *mempfn_tbl; /* array of mbuf pool gpfn */ - uint64_t *pindex; /* index used to release grefs */ -}; - -struct xen_guest { - TAILQ_ENTRY(xen_guest) next; - int32_t dom_id; /* guest domain id */ - uint32_t pool_num; /* number of mbuf pool of the guest */ - uint32_t vring_num; /* number of virtio ports of the guest */ - /* array contain the guest mbuf pool info */ - struct xen_mempool mempool[MAX_XENVIRT_MEMPOOL]; - /* array contain the guest rx/tx vring info */ - struct xen_vring vring[MAX_VIRTIO]; -}; - -TAILQ_HEAD(xen_guestlist, xen_guest); - -int -parse_mempoolnode(struct xen_guest *guest); - -int -xenhost_init(void); - -int -parse_vringnode(struct xen_guest *guest, uint32_t virtio_idx); - -int -parse_mempoolnode(struct xen_guest *guest); - -void -cleanup_mempool(struct xen_mempool *mempool); - -void -cleanup_vring(struct xen_vring *vring); - -void -virtio_monitor_loop(void); - -int -init_virtio_xen(struct virtio_net_device_ops const * const); - -#endif diff --git a/examples/vhost_xen/xenstore_parse.c b/examples/vhost_xen/xenstore_parse.c deleted file mode 100644 index ab089f1b..00000000 --- a/examples/vhost_xen/xenstore_parse.c +++ /dev/null @@ -1,775 +0,0 @@ -/*- - * BSD LICENSE - * - * Copyright(c) 2010-2014 Intel Corporation. All rights reserved. - * All rights reserved. - * - * 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. - * * Neither the name of Intel Corporation nor the names of its - * contributors may be used to endorse or promote products derived - * from this software without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. - */ - -#include <stdint.h> -#include <unistd.h> -#include <inttypes.h> -#include <errno.h> -#include <fcntl.h> -#include <sys/ioctl.h> -#include <sys/mman.h> -#include <xen/sys/gntalloc.h> -#include <xen/sys/gntdev.h> -#include <xen/xen-compat.h> -#if __XEN_LATEST_INTERFACE_VERSION__ < 0x00040200 -#include <xs.h> -#else -#include <xenstore.h> -#endif - -#include <rte_common.h> -#include <rte_memory.h> -#include <rte_eal.h> -#include <rte_malloc.h> -#include <rte_string_fns.h> -#include <rte_log.h> -#include <rte_debug.h> - -#include "xen_vhost.h" - -/* xenstore handle */ -static struct xs_handle *xs = NULL; - -/* gntdev file descriptor to map grant pages */ -static int d_fd = -1; - -/* - * The grant node format in xenstore for vring/mpool is like: - * idx#_rx_vring_gref = "gref1#, gref2#, gref3#" - * idx#_mempool_gref = "gref1#, gref2#, gref3#" - * each gref# is the grant reference for a shared page. - * In each shared page, we store the grant_node_item items. - */ -struct grant_node_item { - uint32_t gref; - uint32_t pfn; -} __attribute__((packed)); - -int cmdline_parse_etheraddr(void *tk, const char *srcbuf, - void *res, unsigned ressize); - -/* Map grant ref refid at addr_ori*/ -static void * -xen_grant_mmap(void *addr_ori, int domid, int refid, uint64_t *pindex) -{ - struct ioctl_gntdev_map_grant_ref arg; - void *addr = NULL; - int pg_sz = getpagesize(); - - arg.count = 1; - arg.refs[0].domid = domid; - arg.refs[0].ref = refid; - - int rv = ioctl(d_fd, IOCTL_GNTDEV_MAP_GRANT_REF, &arg); - if (rv) { - RTE_LOG(ERR, XENHOST, " %s: (%d,%d) %s (ioctl failed)\n", __func__, - domid, refid, strerror(errno)); - return NULL; - } - - if (addr_ori == NULL) - addr = mmap(addr_ori, pg_sz, PROT_READ|PROT_WRITE, MAP_SHARED, - d_fd, arg.index); - else - addr = mmap(addr_ori, pg_sz, PROT_READ|PROT_WRITE, MAP_SHARED | MAP_FIXED, - d_fd, arg.index); - - if (addr == MAP_FAILED) { - RTE_LOG(ERR, XENHOST, " %s: (%d, %d) %s (map failed)\n", __func__, - domid, refid, strerror(errno)); - return NULL; - } - - if (pindex) - *pindex = arg.index; - - return addr; -} - -/* Unmap one grant ref, and munmap must be called before this */ -static int -xen_unmap_grant_ref(uint64_t index) -{ - struct ioctl_gntdev_unmap_grant_ref arg; - int rv; - - arg.count = 1; - arg.index = index; - rv = ioctl(d_fd, IOCTL_GNTDEV_UNMAP_GRANT_REF, &arg); - if (rv) { - RTE_LOG(ERR, XENHOST, " %s: index 0x%" PRIx64 "unmap failed\n", __func__, index); - return -1; - } - return 0; -} - -/* - * Reserve a virtual address space. - * On success, returns the pointer. On failure, returns NULL. - */ -static void * -get_xen_virtual(size_t size, size_t page_sz) -{ - void *addr; - uintptr_t aligned_addr; - - addr = mmap(NULL, size + page_sz, PROT_READ, MAP_SHARED | MAP_ANONYMOUS, -1, 0); - if (addr == MAP_FAILED) { - RTE_LOG(ERR, XENHOST, "failed get a virtual area\n"); - return NULL; - } - - aligned_addr = RTE_ALIGN_CEIL((uintptr_t)addr, page_sz); - munmap(addr, aligned_addr - (uintptr_t)addr); - munmap((void *)(aligned_addr + size), page_sz + (uintptr_t)addr - aligned_addr); - addr = (void *)(aligned_addr); - - return addr; -} - -static void -free_xen_virtual(void *addr, size_t size, size_t page_sz __rte_unused) -{ - if (addr) - munmap(addr, size); -} - -/* - * Returns val str in xenstore. - * @param path - * Full path string for key - * @return - * Pointer to Val str, NULL on failure - */ -static char * -xen_read_node(char *path, uint32_t *len) -{ - char *buf; - - buf = xs_read(xs, XBT_NULL, path, len); - return buf; -} - -static int -cal_pagenum(struct xen_gnt *gnt) -{ - unsigned int i; - /* - * the items in the page are in the format of - * gref#,pfn#,...,gref#,pfn# - * FIXME, 0 is reserved by system, use it as terminator. - */ - for (i = 0; i < (PAGE_PFNNUM) / 2; i++) { - if (gnt->gref_pfn[i * 2].gref <= 0) - break; - } - - return i; -} - -/* Frees memory allocated to a grant node */ -static void -xen_free_gntnode(struct xen_gntnode *gntnode) -{ - if (gntnode == NULL) - return; - free(gntnode->gnt_info); - free(gntnode); -} - -/* - * Parse a grant node. - * @param domid - * Guest domain id. - * @param path - * Full path string for a grant node, like for the following (key, val) pair - * idx#_mempool_gref = "gref#, gref#, gref#" - * path = 'local/domain/domid/control/dpdk/idx#_mempool_gref' - * gref# is a shared page contain packed (gref,pfn) entries - * @return - * Returns the pointer to xen_gntnode - */ -static struct xen_gntnode * -parse_gntnode(int dom_id, char *path) -{ - char **gref_list = NULL; - uint32_t i, len, gref_num; - void *addr = NULL; - char *buf = NULL; - struct xen_gntnode *gntnode = NULL; - struct xen_gnt *gnt = NULL; - int pg_sz = getpagesize(); - char *end; - uint64_t index; - - if ((buf = xen_read_node(path, &len)) == NULL) - goto err; - - gref_list = malloc(MAX_GREF_PER_NODE * sizeof(char *)); - if (gref_list == NULL) - goto err; - - gref_num = rte_strsplit(buf, len, gref_list, MAX_GREF_PER_NODE, - XEN_GREF_SPLITTOKEN); - if (gref_num == 0) { - RTE_LOG(ERR, XENHOST, " %s: invalid grant node format\n", __func__); - goto err; - } - - gntnode = calloc(1, sizeof(struct xen_gntnode)); - gnt = calloc(gref_num, sizeof(struct xen_gnt)); - if (gnt == NULL || gntnode == NULL) - goto err; - - for (i = 0; i < gref_num; i++) { - errno = 0; - gnt[i].gref = strtol(gref_list[i], &end, 0); - if (errno != 0 || end == NULL || end == gref_list[i] || - (*end != '\0' && *end != XEN_GREF_SPLITTOKEN)) { - RTE_LOG(ERR, XENHOST, " %s: parse grant node item failed\n", __func__); - goto err; - } - addr = xen_grant_mmap(NULL, dom_id, gnt[i].gref, &index); - if (addr == NULL) { - RTE_LOG(ERR, XENHOST, " %s: map gref %u failed\n", __func__, gnt[i].gref); - goto err; - } - RTE_LOG(INFO, XENHOST, " %s: map gref %u to %p\n", __func__, gnt[i].gref, addr); - memcpy(gnt[i].gref_pfn, addr, pg_sz); - if (munmap(addr, pg_sz)) { - RTE_LOG(INFO, XENHOST, " %s: unmap gref %u failed\n", __func__, gnt[i].gref); - goto err; - } - if (xen_unmap_grant_ref(index)) { - RTE_LOG(INFO, XENHOST, " %s: release gref %u failed\n", __func__, gnt[i].gref); - goto err; - } - - } - - gntnode->gnt_num = gref_num; - gntnode->gnt_info = gnt; - - free(buf); - free(gref_list); - return gntnode; - -err: - free(gnt); - free(gntnode); - free(gref_list); - free(buf); - return NULL; -} - -/* - * This function maps grant node of vring or mbuf pool to a continuous virtual address space, - * and returns mapped address, pfn array, index array - * @param gntnode - * Pointer to grant node - * @param domid - * Guest domain id - * @param ppfn - * Pointer to pfn array, caller should free this array - * @param pgs - * Pointer to number of pages - * @param ppindex - * Pointer to index array, used to release grefs when to free this node - * @return - * Pointer to mapped virtual address, NULL on failure - */ -static void * -map_gntnode(struct xen_gntnode *gntnode, int domid, uint32_t **ppfn, uint32_t *pgs, uint64_t **ppindex) -{ - struct xen_gnt *gnt; - uint32_t i, j; - size_t total_pages = 0; - void *addr; - uint32_t *pfn; - uint64_t *pindex; - uint32_t pfn_num = 0; - int pg_sz; - - if (gntnode == NULL) - return NULL; - - pg_sz = getpagesize(); - for (i = 0; i < gntnode->gnt_num; i++) { - gnt = gntnode->gnt_info + i; - total_pages += cal_pagenum(gnt); - } - if ((addr = get_xen_virtual(total_pages * pg_sz, pg_sz)) == NULL) { - RTE_LOG(ERR, XENHOST, " %s: failed get_xen_virtual\n", __func__); - return NULL; - } - pfn = calloc(total_pages, (size_t)sizeof(uint32_t)); - pindex = calloc(total_pages, (size_t)sizeof(uint64_t)); - if (pfn == NULL || pindex == NULL) { - free_xen_virtual(addr, total_pages * pg_sz, pg_sz); - free(pfn); - free(pindex); - return NULL; - } - - RTE_LOG(INFO, XENHOST, " %s: total pages:%zu, map to [%p, %p]\n", __func__, total_pages, addr, RTE_PTR_ADD(addr, total_pages * pg_sz - 1)); - for (i = 0; i < gntnode->gnt_num; i++) { - gnt = gntnode->gnt_info + i; - for (j = 0; j < (PAGE_PFNNUM) / 2; j++) { - if ((gnt->gref_pfn[j * 2].gref) <= 0) - goto _end; - /*alternative: batch map, or through libxc*/ - if (xen_grant_mmap(RTE_PTR_ADD(addr, pfn_num * pg_sz), - domid, - gnt->gref_pfn[j * 2].gref, - &pindex[pfn_num]) == NULL) { - goto mmap_failed; - } - pfn[pfn_num] = gnt->gref_pfn[j * 2 + 1].pfn_num; - pfn_num++; - } - } - -mmap_failed: - if (pfn_num) - munmap(addr, pfn_num * pg_sz); - for (i = 0; i < pfn_num; i++) { - xen_unmap_grant_ref(pindex[i]); - } - free(pindex); - free(pfn); - return NULL; - -_end: - if (ppindex) - *ppindex = pindex; - else - free(pindex); - if (ppfn) - *ppfn = pfn; - else - free(pfn); - if (pgs) - *pgs = total_pages; - - return addr; -} - -static int -parse_mpool_va(struct xen_mempool *mempool) -{ - char path[PATH_MAX] = {0}; - char *buf; - uint32_t len; - char *end; - int ret = -1; - - errno = 0; - snprintf(path, sizeof(path), - XEN_VM_ROOTNODE_FMT"/%d_"XEN_GVA_SUFFIX, - mempool->dom_id, mempool->pool_idx); - - if((buf = xen_read_node(path, &len)) == NULL) - goto out; - mempool->gva = (void *)strtoul(buf, &end, 16); - if (errno != 0 || end == NULL || end == buf || *end != '\0') { - mempool->gva = NULL; - goto out; - } - ret = 0; -out: - free(buf); - return ret; -} - -/* - * map mbuf pool - */ -static int -map_mempoolnode(struct xen_gntnode *gntnode, - struct xen_mempool *mempool) -{ - if (gntnode == NULL || mempool == NULL) - return -1; - - mempool->hva = - map_gntnode(gntnode, mempool->dom_id, &mempool->mempfn_tbl, &mempool->mempfn_num, &mempool->pindex); - - RTE_LOG(INFO, XENHOST, " %s: map mempool at %p\n", __func__, (void *)mempool->hva); - if (mempool->hva) - return 0; - else { - return -1; - } -} - -void -cleanup_mempool(struct xen_mempool *mempool) -{ - int pg_sz = getpagesize(); - uint32_t i; - - if (mempool->hva) - munmap(mempool->hva, mempool->mempfn_num * pg_sz); - mempool->hva = NULL; - - if (mempool->pindex) { - RTE_LOG(INFO, XENHOST, " %s: unmap dom %02u mempool%02u %u grefs\n", - __func__, - mempool->dom_id, - mempool->pool_idx, - mempool->mempfn_num); - for (i = 0; i < mempool->mempfn_num; i ++) { - xen_unmap_grant_ref(mempool->pindex[i]); - } - } - mempool->pindex = NULL; - - free(mempool->mempfn_tbl); - mempool->mempfn_tbl = NULL; -} - -/* - * process mempool node idx#_mempool_gref, idx = 0, 1, 2... - * until we encounter a node that doesn't exist. - */ -int -parse_mempoolnode(struct xen_guest *guest) -{ - uint32_t i, len; - char path[PATH_MAX] = {0}; - struct xen_gntnode *gntnode = NULL; - struct xen_mempool *mempool = NULL; - char *buf; - - bzero(&guest->mempool, MAX_XENVIRT_MEMPOOL * sizeof(guest->mempool[0])); - guest->pool_num = 0; - - while (1) { - /* check if null terminated */ - snprintf(path, sizeof(path), - XEN_VM_ROOTNODE_FMT"/%d_"XEN_MEMPOOL_SUFFIX, - guest->dom_id, - guest->pool_num); - - if ((buf = xen_read_node(path, &len)) != NULL) { - /* this node exists */ - free(buf); - } else { - if (guest->pool_num == 0) { - RTE_LOG(ERR, PMD, "no mempool found\n"); - return -1; - } - break; - } - - mempool = &guest->mempool[guest->pool_num]; - mempool->dom_id = guest->dom_id; - mempool->pool_idx = guest->pool_num; - - RTE_LOG(INFO, XENHOST, " %s: mempool %u parse gntnode %s\n", __func__, guest->pool_num, path); - gntnode = parse_gntnode(guest->dom_id, path); - if (gntnode == NULL) - goto err; - - if (parse_mpool_va(mempool)) - goto err; - - RTE_LOG(INFO, XENHOST, " %s: mempool %u map gntnode %s\n", __func__, guest->pool_num, path); - if (map_mempoolnode(gntnode, mempool)) - goto err; - - xen_free_gntnode(gntnode); - guest->pool_num++; - } - - return 0; -err: - if (gntnode) - xen_free_gntnode(gntnode); - for (i = 0; i < MAX_XENVIRT_MEMPOOL ; i++) { - cleanup_mempool(&guest->mempool[i]); - } - /* reinitialise mempool */ - bzero(&guest->mempool, MAX_XENVIRT_MEMPOOL * sizeof(guest->mempool[0])); - return -1; -} - -static int -xen_map_vringflag(struct xen_vring *vring) -{ - char path[PATH_MAX] = {0}; - char *buf; - uint32_t len,gref; - int pg_sz = getpagesize(); - char *end; - - snprintf(path, sizeof(path), - XEN_VM_ROOTNODE_FMT"/%d_"XEN_VRINGFLAG_SUFFIX, - vring->dom_id, vring->virtio_idx); - - if((buf = xen_read_node(path, &len)) == NULL) - goto err; - - errno = 0; - gref = strtol(buf, &end, 0); - if (errno != 0 || end == NULL || end == buf) { - goto err; - } - vring->flag = xen_grant_mmap(0, vring->dom_id, gref, &vring->flag_index); - if (vring->flag == NULL || *vring->flag == 0) - goto err; - - free(buf); - return 0; -err: - free(buf); - if (vring->flag) { - munmap(vring->flag, pg_sz); - vring->flag = NULL; - xen_unmap_grant_ref(vring->flag_index); - } - return -1; -} - - -static int -xen_map_rxvringnode(struct xen_gntnode *gntnode, - struct xen_vring *vring) -{ - vring->rxvring_addr = - map_gntnode(gntnode, vring->dom_id, &vring->rxpfn_tbl, &vring->rxpfn_num, &vring->rx_pindex); - RTE_LOG(INFO, XENHOST, " %s: map rx vring at %p\n", __func__, (void *)vring->rxvring_addr); - if (vring->rxvring_addr) - return 0; - else - return -1; -} - -static int -xen_map_txvringnode(struct xen_gntnode *gntnode, - struct xen_vring *vring) -{ - vring->txvring_addr = - map_gntnode(gntnode, vring->dom_id, &vring->txpfn_tbl, &vring->txpfn_num, &vring->tx_pindex); - RTE_LOG(INFO, XENHOST, " %s: map tx vring at %p\n", __func__, (void *)vring->txvring_addr); - if (vring->txvring_addr) - return 0; - else - return -1; -} - -void -cleanup_vring(struct xen_vring *vring) -{ - int pg_sz = getpagesize(); - uint32_t i; - - RTE_LOG(INFO, XENHOST, " %s: cleanup dom %u vring %u\n", __func__, vring->dom_id, vring->virtio_idx); - if (vring->rxvring_addr) { - munmap(vring->rxvring_addr, vring->rxpfn_num * pg_sz); - RTE_LOG(INFO, XENHOST, " %s: unmap rx vring [%p, %p]\n", - __func__, - vring->rxvring_addr, - RTE_PTR_ADD(vring->rxvring_addr, - vring->rxpfn_num * pg_sz - 1)); - } - vring->rxvring_addr = NULL; - - - if (vring->rx_pindex) { - RTE_LOG(INFO, XENHOST, " %s: unmap rx vring %u grefs\n", __func__, vring->rxpfn_num); - for (i = 0; i < vring->rxpfn_num; i++) { - xen_unmap_grant_ref(vring->rx_pindex[i]); - } - } - vring->rx_pindex = NULL; - - free(vring->rxpfn_tbl); - vring->rxpfn_tbl = NULL; - - if (vring->txvring_addr) { - munmap(vring->txvring_addr, vring->txpfn_num * pg_sz); - RTE_LOG(INFO, XENHOST, " %s: unmap tx vring [%p, %p]\n", - __func__, - vring->txvring_addr, - RTE_PTR_ADD(vring->txvring_addr, - vring->txpfn_num * pg_sz - 1)); - } - vring->txvring_addr = NULL; - - if (vring->tx_pindex) { - RTE_LOG(INFO, XENHOST, " %s: unmap tx vring %u grefs\n", __func__, vring->txpfn_num); - for (i = 0; i < vring->txpfn_num; i++) { - xen_unmap_grant_ref(vring->tx_pindex[i]); - } - } - vring->tx_pindex = NULL; - - free(vring->txpfn_tbl); - vring->txpfn_tbl = NULL; - - if (vring->flag) { - if (!munmap((void *)vring->flag, pg_sz)) - RTE_LOG(INFO, XENHOST, " %s: unmap flag page at %p\n", __func__, vring->flag); - if (!xen_unmap_grant_ref(vring->flag_index)) - RTE_LOG(INFO, XENHOST, " %s: release flag ref index 0x%" PRIx64 "\n", __func__, vring->flag_index); - } - vring->flag = NULL; - return; -} - - - -static int -xen_parse_etheraddr(struct xen_vring *vring) -{ - char path[PATH_MAX] = {0}; - char *buf; - uint32_t len; - int ret = -1; - - snprintf(path, sizeof(path), - XEN_VM_ROOTNODE_FMT"/%d_"XEN_ADDR_SUFFIX, - vring->dom_id, vring->virtio_idx); - - if ((buf = xen_read_node(path, &len)) == NULL) - goto out; - - if (cmdline_parse_etheraddr(NULL, buf, &vring->addr, - sizeof(vring->addr)) < 0) - goto out; - ret = 0; -out: - free(buf); - return ret; -} - - -int -parse_vringnode(struct xen_guest *guest, uint32_t virtio_idx) -{ - char path[PATH_MAX] = {0}; - struct xen_gntnode *rx_gntnode = NULL; - struct xen_gntnode *tx_gntnode = NULL; - struct xen_vring *vring = NULL; - - /*check if null terminated */ - snprintf(path, sizeof(path), - XEN_VM_ROOTNODE_FMT"/%d_"XEN_RXVRING_SUFFIX, - guest->dom_id, - virtio_idx); - - RTE_LOG(INFO, XENHOST, " %s: virtio %u parse rx gntnode %s\n", __func__, virtio_idx, path); - rx_gntnode = parse_gntnode(guest->dom_id, path); - if (rx_gntnode == NULL) - goto err; - - /*check if null terminated */ - snprintf(path, sizeof(path), - XEN_VM_ROOTNODE_FMT"/%d_"XEN_TXVRING_SUFFIX, - guest->dom_id, - virtio_idx); - - RTE_LOG(INFO, XENHOST, " %s: virtio %u parse tx gntnode %s\n", __func__, virtio_idx, path); - tx_gntnode = parse_gntnode(guest->dom_id, path); - if (tx_gntnode == NULL) - goto err; - - vring = &guest->vring[virtio_idx]; - bzero(vring, sizeof(*vring)); - vring->dom_id = guest->dom_id; - vring->virtio_idx = virtio_idx; - - if (xen_parse_etheraddr(vring) != 0) - goto err; - - RTE_LOG(INFO, XENHOST, " %s: virtio %u map rx gntnode %s\n", __func__, virtio_idx, path); - if (xen_map_rxvringnode(rx_gntnode, vring) != 0) - goto err; - - RTE_LOG(INFO, XENHOST, " %s: virtio %u map tx gntnode %s\n", __func__, virtio_idx, path); - if (xen_map_txvringnode(tx_gntnode, vring) != 0) - goto err; - - if (xen_map_vringflag(vring) != 0) - goto err; - - guest->vring_num++; - - xen_free_gntnode(rx_gntnode); - xen_free_gntnode(tx_gntnode); - - return 0; - -err: - if (rx_gntnode) - xen_free_gntnode(rx_gntnode); - if (tx_gntnode) - xen_free_gntnode(tx_gntnode); - if (vring) { - cleanup_vring(vring); - bzero(vring, sizeof(*vring)); - } - return -1; -} - -/* - * Open xen grant dev driver - * @return - * 0 on success, -1 on failure. - */ -static int -xen_grant_init(void) -{ - d_fd = open(XEN_GNTDEV_FNAME, O_RDWR); - - return d_fd == -1? (-1): (0); -} - -/* - * Initialise xenstore handle and open grant dev driver. - * @return - * 0 on success, -1 on failure. - */ -int -xenhost_init(void) -{ - xs = xs_daemon_open(); - if (xs == NULL) { - rte_panic("failed initialize xen daemon handler"); - return -1; - } - if (xen_grant_init()) - return -1; - return 0; -} diff --git a/examples/vm_power_manager/Makefile b/examples/vm_power_manager/Makefile index 59a96417..9cf20a28 100644 --- a/examples/vm_power_manager/Makefile +++ b/examples/vm_power_manager/Makefile @@ -54,6 +54,22 @@ CFLAGS += $(WERROR_FLAGS) LDLIBS += -lvirt +ifeq ($(CONFIG_RTE_BUILD_SHARED_LIB),y) + +ifeq ($(CONFIG_RTE_LIBRTE_IXGBE_PMD),y) +LDLIBS += -lrte_pmd_ixgbe +endif + +ifeq ($(CONFIG_RTE_LIBRTE_I40E_PMD),y) +LDLIBS += -lrte_pmd_i40e +endif + +ifeq ($(CONFIG_RTE_LIBRTE_BNXT_PMD),y) +LDLIBS += -lrte_pmd_bnxt +endif + +endif + # workaround for a gcc bug with noreturn attribute # http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12603 ifeq ($(CONFIG_RTE_TOOLCHAIN_GCC),y) diff --git a/examples/vm_power_manager/channel_manager.c b/examples/vm_power_manager/channel_manager.c index e068ae28..ab856bdb 100644 --- a/examples/vm_power_manager/channel_manager.c +++ b/examples/vm_power_manager/channel_manager.c @@ -574,6 +574,73 @@ set_channel_status(const char *vm_name, unsigned *channel_list, return num_channels_changed; } +void +get_all_vm(int *num_vm, int *num_vcpu) +{ + + virNodeInfo node_info; + virDomainPtr *domptr; + uint64_t mask; + int i, ii, numVcpus[MAX_VCPUS], cpu, n_vcpus; + unsigned int jj; + const char *vm_name; + unsigned int domain_flags = VIR_CONNECT_LIST_DOMAINS_RUNNING | + VIR_CONNECT_LIST_DOMAINS_PERSISTENT; + unsigned int domain_flag = VIR_DOMAIN_VCPU_CONFIG; + + + memset(global_cpumaps, 0, CHANNEL_CMDS_MAX_CPUS*global_maplen); + if (virNodeGetInfo(global_vir_conn_ptr, &node_info)) { + RTE_LOG(ERR, CHANNEL_MANAGER, "Unable to retrieve node Info\n"); + return; + } + + /* Returns number of pcpus */ + global_n_host_cpus = (unsigned int)node_info.cpus; + + /* Returns number of active domains */ + *num_vm = virConnectListAllDomains(global_vir_conn_ptr, &domptr, + domain_flags); + if (*num_vm <= 0) { + RTE_LOG(ERR, CHANNEL_MANAGER, "No Active Domains Running\n"); + return; + } + + for (i = 0; i < *num_vm; i++) { + + /* Get Domain Names */ + vm_name = virDomainGetName(domptr[i]); + lvm_info[i].vm_name = vm_name; + + /* Get Number of Vcpus */ + numVcpus[i] = virDomainGetVcpusFlags(domptr[i], domain_flag); + + /* Get Number of VCpus & VcpuPinInfo */ + n_vcpus = virDomainGetVcpuPinInfo(domptr[i], + numVcpus[i], global_cpumaps, + global_maplen, domain_flag); + + if ((int)n_vcpus > 0) { + *num_vcpu = n_vcpus; + lvm_info[i].num_cpus = n_vcpus; + } + + /* Save pcpu in use by libvirt VMs */ + for (ii = 0; ii < n_vcpus; ii++) { + mask = 0; + for (jj = 0; jj < global_n_host_cpus; jj++) { + if (VIR_CPU_USABLE(global_cpumaps, + global_maplen, ii, jj) > 0) { + mask |= 1ULL << jj; + } + } + ITERATIVE_BITMASK_CHECK_64(mask, cpu) { + lvm_info[i].pcpus[ii] = cpu; + } + } + } +} + int get_info_vm(const char *vm_name, struct vm_info *info) { diff --git a/examples/vm_power_manager/channel_manager.h b/examples/vm_power_manager/channel_manager.h index 47c3b9cd..358fb8f2 100644 --- a/examples/vm_power_manager/channel_manager.h +++ b/examples/vm_power_manager/channel_manager.h @@ -66,6 +66,17 @@ struct sockaddr_un _sockaddr_un; #define UNIX_PATH_MAX sizeof(_sockaddr_un.sun_path) #endif +#define MAX_VMS 4 +#define MAX_VCPUS 20 + + +struct libvirt_vm_info { + const char *vm_name; + unsigned int pcpus[MAX_VCPUS]; + uint8_t num_cpus; +}; + +struct libvirt_vm_info lvm_info[MAX_VMS]; /* Communication Channel Status */ enum channel_status { CHANNEL_MGR_CHANNEL_DISCONNECTED = 0, CHANNEL_MGR_CHANNEL_CONNECTED, @@ -319,6 +330,20 @@ int set_channel_status(const char *vm_name, unsigned *channel_list, */ int get_info_vm(const char *vm_name, struct vm_info *info); +/** + * Populates a table with all domains running and their physical cpu. + * All information is gathered through libvirt api. + * + * @param num_vm + * modified to store number of active VMs + * + * @param num_vcpu + modified to store number of vcpus active + * + * @return + * void + */ +void get_all_vm(int *num_vm, int *num_vcpu); #ifdef __cplusplus } #endif diff --git a/examples/vm_power_manager/channel_monitor.c b/examples/vm_power_manager/channel_monitor.c index e7f5cc4a..37e71ed9 100644 --- a/examples/vm_power_manager/channel_monitor.c +++ b/examples/vm_power_manager/channel_monitor.c @@ -41,13 +41,17 @@ #include <sys/types.h> #include <sys/epoll.h> #include <sys/queue.h> +#include <sys/time.h> #include <rte_log.h> #include <rte_memory.h> #include <rte_malloc.h> #include <rte_atomic.h> +#include <rte_cycles.h> +#include <rte_ethdev.h> +#include <rte_pmd_i40e.h> - +#include <libvirt/libvirt.h> #include "channel_monitor.h" #include "channel_commands.h" #include "channel_manager.h" @@ -57,10 +61,15 @@ #define MAX_EVENTS 256 +uint64_t vsi_pkt_count_prev[384]; +uint64_t rdtsc_prev[384]; +double time_period_ms = 1; static volatile unsigned run_loop = 1; static int global_event_fd; +static unsigned int policy_is_set; static struct epoll_event *global_events_list; +static struct policy policies[MAX_VMS]; void channel_monitor_exit(void) { @@ -68,6 +77,287 @@ void channel_monitor_exit(void) rte_free(global_events_list); } +static void +core_share(int pNo, int z, int x, int t) +{ + if (policies[pNo].core_share[z].pcpu == lvm_info[x].pcpus[t]) { + if (strcmp(policies[pNo].pkt.vm_name, + lvm_info[x].vm_name) != 0) { + policies[pNo].core_share[z].status = 1; + power_manager_scale_core_max( + policies[pNo].core_share[z].pcpu); + } + } +} + +static void +core_share_status(int pNo) +{ + + int noVms, noVcpus, z, x, t; + + get_all_vm(&noVms, &noVcpus); + + /* Reset Core Share Status. */ + for (z = 0; z < noVcpus; z++) + policies[pNo].core_share[z].status = 0; + + /* Foreach vcpu in a policy. */ + for (z = 0; z < policies[pNo].pkt.num_vcpu; z++) { + /* Foreach VM on the platform. */ + for (x = 0; x < noVms; x++) { + /* Foreach vcpu of VMs on platform. */ + for (t = 0; t < lvm_info[x].num_cpus; t++) + core_share(pNo, z, x, t); + } + } +} + +static void +get_pcpu_to_control(struct policy *pol) +{ + + /* Convert vcpu to pcpu. */ + struct vm_info info; + int pcpu, count; + uint64_t mask_u64b; + + RTE_LOG(INFO, CHANNEL_MONITOR, "Looking for pcpu for %s\n", + pol->pkt.vm_name); + get_info_vm(pol->pkt.vm_name, &info); + + for (count = 0; count < pol->pkt.num_vcpu; count++) { + mask_u64b = info.pcpu_mask[pol->pkt.vcpu_to_control[count]]; + for (pcpu = 0; mask_u64b; mask_u64b &= ~(1ULL << pcpu++)) { + if ((mask_u64b >> pcpu) & 1) + pol->core_share[count].pcpu = pcpu; + } + } +} + +static int +get_pfid(struct policy *pol) +{ + + int i, x, ret = 0, nb_ports; + + nb_ports = rte_eth_dev_count(); + for (i = 0; i < pol->pkt.nb_mac_to_monitor; i++) { + + for (x = 0; x < nb_ports; x++) { + ret = rte_pmd_i40e_query_vfid_by_mac(x, + (struct ether_addr *)&(pol->pkt.vfid[i])); + if (ret != -EINVAL) { + pol->port[i] = x; + break; + } + } + if (ret == -EINVAL || ret == -ENOTSUP || ret == ENODEV) { + RTE_LOG(INFO, CHANNEL_MONITOR, + "Error with Policy. MAC not found on " + "attached ports "); + pol->enabled = 0; + return ret; + } + pol->pfid[i] = ret; + } + return 1; +} + +static int +update_policy(struct channel_packet *pkt) +{ + + unsigned int updated = 0; + int i; + + for (i = 0; i < MAX_VMS; i++) { + if (strcmp(policies[i].pkt.vm_name, pkt->vm_name) == 0) { + policies[i].pkt = *pkt; + get_pcpu_to_control(&policies[i]); + if (get_pfid(&policies[i]) == -1) { + updated = 1; + break; + } + core_share_status(i); + policies[i].enabled = 1; + updated = 1; + } + } + if (!updated) { + for (i = 0; i < MAX_VMS; i++) { + if (policies[i].enabled == 0) { + policies[i].pkt = *pkt; + get_pcpu_to_control(&policies[i]); + if (get_pfid(&policies[i]) == -1) + break; + core_share_status(i); + policies[i].enabled = 1; + break; + } + } + } + return 0; +} + +static uint64_t +get_pkt_diff(struct policy *pol) +{ + + uint64_t vsi_pkt_count, + vsi_pkt_total = 0, + vsi_pkt_count_prev_total = 0; + double rdtsc_curr, rdtsc_diff, diff; + int x; + struct rte_eth_stats vf_stats; + + for (x = 0; x < pol->pkt.nb_mac_to_monitor; x++) { + + /*Read vsi stats*/ + if (rte_pmd_i40e_get_vf_stats(x, pol->pfid[x], &vf_stats) == 0) + vsi_pkt_count = vf_stats.ipackets; + else + vsi_pkt_count = -1; + + vsi_pkt_total += vsi_pkt_count; + + vsi_pkt_count_prev_total += vsi_pkt_count_prev[pol->pfid[x]]; + vsi_pkt_count_prev[pol->pfid[x]] = vsi_pkt_count; + } + + rdtsc_curr = rte_rdtsc_precise(); + rdtsc_diff = rdtsc_curr - rdtsc_prev[pol->pfid[x-1]]; + rdtsc_prev[pol->pfid[x-1]] = rdtsc_curr; + + diff = (vsi_pkt_total - vsi_pkt_count_prev_total) * + ((double)rte_get_tsc_hz() / rdtsc_diff); + + return diff; +} + +static void +apply_traffic_profile(struct policy *pol) +{ + + int count; + uint64_t diff = 0; + + diff = get_pkt_diff(pol); + + RTE_LOG(INFO, CHANNEL_MONITOR, "Applying traffic profile\n"); + + if (diff >= (pol->pkt.traffic_policy.max_max_packet_thresh)) { + for (count = 0; count < pol->pkt.num_vcpu; count++) { + if (pol->core_share[count].status != 1) + power_manager_scale_core_max( + pol->core_share[count].pcpu); + } + } else if (diff >= (pol->pkt.traffic_policy.avg_max_packet_thresh)) { + for (count = 0; count < pol->pkt.num_vcpu; count++) { + if (pol->core_share[count].status != 1) + power_manager_scale_core_med( + pol->core_share[count].pcpu); + } + } else if (diff < (pol->pkt.traffic_policy.avg_max_packet_thresh)) { + for (count = 0; count < pol->pkt.num_vcpu; count++) { + if (pol->core_share[count].status != 1) + power_manager_scale_core_min( + pol->core_share[count].pcpu); + } + } +} + +static void +apply_time_profile(struct policy *pol) +{ + + int count, x; + struct timeval tv; + struct tm *ptm; + char time_string[40]; + + /* Obtain the time of day, and convert it to a tm struct. */ + gettimeofday(&tv, NULL); + ptm = localtime(&tv.tv_sec); + /* Format the date and time, down to a single second. */ + strftime(time_string, sizeof(time_string), "%Y-%m-%d %H:%M:%S", ptm); + + for (x = 0; x < HOURS; x++) { + + if (ptm->tm_hour == pol->pkt.timer_policy.busy_hours[x]) { + for (count = 0; count < pol->pkt.num_vcpu; count++) { + if (pol->core_share[count].status != 1) { + power_manager_scale_core_max( + pol->core_share[count].pcpu); + RTE_LOG(INFO, CHANNEL_MONITOR, + "Scaling up core %d to max\n", + pol->core_share[count].pcpu); + } + } + break; + } else if (ptm->tm_hour == + pol->pkt.timer_policy.quiet_hours[x]) { + for (count = 0; count < pol->pkt.num_vcpu; count++) { + if (pol->core_share[count].status != 1) { + power_manager_scale_core_min( + pol->core_share[count].pcpu); + RTE_LOG(INFO, CHANNEL_MONITOR, + "Scaling down core %d to min\n", + pol->core_share[count].pcpu); + } + } + break; + } else if (ptm->tm_hour == + pol->pkt.timer_policy.hours_to_use_traffic_profile[x]) { + apply_traffic_profile(pol); + break; + } + } +} + +static void +apply_workload_profile(struct policy *pol) +{ + + int count; + + if (pol->pkt.workload == HIGH) { + for (count = 0; count < pol->pkt.num_vcpu; count++) { + if (pol->core_share[count].status != 1) + power_manager_scale_core_max( + pol->core_share[count].pcpu); + } + } else if (pol->pkt.workload == MEDIUM) { + for (count = 0; count < pol->pkt.num_vcpu; count++) { + if (pol->core_share[count].status != 1) + power_manager_scale_core_med( + pol->core_share[count].pcpu); + } + } else if (pol->pkt.workload == LOW) { + for (count = 0; count < pol->pkt.num_vcpu; count++) { + if (pol->core_share[count].status != 1) + power_manager_scale_core_min( + pol->core_share[count].pcpu); + } + } +} + +static void +apply_policy(struct policy *pol) +{ + + struct channel_packet *pkt = &pol->pkt; + + /*Check policy to use*/ + if (pkt->policy_to_use == TRAFFIC) + apply_traffic_profile(pol); + else if (pkt->policy_to_use == TIME) + apply_time_profile(pol); + else if (pkt->policy_to_use == WORKLOAD) + apply_workload_profile(pol); +} + + static int process_request(struct channel_packet *pkt, struct channel_info *chan_info) { @@ -105,6 +395,12 @@ process_request(struct channel_packet *pkt, struct channel_info *chan_info) case(CPU_POWER_SCALE_UP): power_manager_scale_core_up(core_num); break; + case(CPU_POWER_ENABLE_TURBO): + power_manager_enable_turbo_core(core_num); + break; + case(CPU_POWER_DISABLE_TURBO): + power_manager_disable_turbo_core(core_num); + break; default: break; } @@ -122,12 +418,25 @@ process_request(struct channel_packet *pkt, struct channel_info *chan_info) case(CPU_POWER_SCALE_UP): power_manager_scale_mask_up(core_mask); break; + case(CPU_POWER_ENABLE_TURBO): + power_manager_enable_turbo_mask(core_mask); + break; + case(CPU_POWER_DISABLE_TURBO): + power_manager_disable_turbo_mask(core_mask); + break; default: break; } } } + + if (pkt->command == PKT_POLICY) { + RTE_LOG(INFO, CHANNEL_MONITOR, "\nProcessing Policy request from Guest\n"); + update_policy(pkt); + policy_is_set = 1; + } + /* Return is not checked as channel status may have been set to DISABLED * from management thread */ @@ -197,9 +506,10 @@ run_channel_monitor(void) struct channel_info *chan_info = (struct channel_info *) global_events_list[i].data.ptr; if ((global_events_list[i].events & EPOLLERR) || - (global_events_list[i].events & EPOLLHUP)) { + (global_events_list[i].events & EPOLLHUP)) { RTE_LOG(DEBUG, CHANNEL_MONITOR, "Remote closed connection for " - "channel '%s'\n", chan_info->channel_path); + "channel '%s'\n", + chan_info->channel_path); remove_channel(&chan_info); continue; } @@ -211,14 +521,17 @@ run_channel_monitor(void) int buffer_len = sizeof(pkt); while (buffer_len > 0) { - n_bytes = read(chan_info->fd, buffer, buffer_len); + n_bytes = read(chan_info->fd, + buffer, buffer_len); if (n_bytes == buffer_len) break; if (n_bytes == -1) { err = errno; - RTE_LOG(DEBUG, CHANNEL_MONITOR, "Received error on " - "channel '%s' read: %s\n", - chan_info->channel_path, strerror(err)); + RTE_LOG(DEBUG, CHANNEL_MONITOR, + "Received error on " + "channel '%s' read: %s\n", + chan_info->channel_path, + strerror(err)); remove_channel(&chan_info); break; } @@ -229,5 +542,14 @@ run_channel_monitor(void) process_request(&pkt, chan_info); } } + rte_delay_us(time_period_ms*1000); + if (policy_is_set) { + int j; + + for (j = 0; j < MAX_VMS; j++) { + if (policies[j].enabled == 1) + apply_policy(&policies[j]); + } + } } } diff --git a/examples/vm_power_manager/channel_monitor.h b/examples/vm_power_manager/channel_monitor.h index c1386079..b52c1fca 100644 --- a/examples/vm_power_manager/channel_monitor.h +++ b/examples/vm_power_manager/channel_monitor.h @@ -35,6 +35,24 @@ #define CHANNEL_MONITOR_H_ #include "channel_manager.h" +#include "channel_commands.h" + +struct core_share { + unsigned int pcpu; + /* + * 1 CORE SHARE + * 0 NOT SHARED + */ + int status; +}; + +struct policy { + struct channel_packet pkt; + uint32_t pfid[MAX_VFS]; + uint32_t port[MAX_VFS]; + unsigned int enabled; + struct core_share core_share[MAX_VCPU_PER_VM]; +}; #ifdef __cplusplus extern "C" { diff --git a/examples/vm_power_manager/guest_cli/vm_power_cli_guest.c b/examples/vm_power_manager/guest_cli/vm_power_cli_guest.c index 7931135e..63f711e0 100644 --- a/examples/vm_power_manager/guest_cli/vm_power_cli_guest.c +++ b/examples/vm_power_manager/guest_cli/vm_power_cli_guest.c @@ -45,8 +45,10 @@ #include <cmdline.h> #include <rte_log.h> #include <rte_lcore.h> +#include <rte_ethdev.h> #include <rte_power.h> +#include <guest_channel.h> #include "vm_power_cli_guest.h" @@ -108,6 +110,10 @@ cmd_set_cpu_freq_parsed(void *parsed_result, struct cmdline *cl, ret = rte_power_freq_min(res->lcore_id); else if (!strcmp(res->cmd , "max")) ret = rte_power_freq_max(res->lcore_id); + else if (!strcmp(res->cmd, "enable_turbo")) + ret = rte_power_freq_enable_turbo(res->lcore_id); + else if (!strcmp(res->cmd, "disable_turbo")) + ret = rte_power_freq_disable_turbo(res->lcore_id); if (ret != 1) cmdline_printf(cl, "Error sending message: %s\n", strerror(ret)); } @@ -120,13 +126,14 @@ cmdline_parse_token_string_t cmd_set_cpu_freq_core_num = lcore_id, UINT8); cmdline_parse_token_string_t cmd_set_cpu_freq_cmd_cmd = TOKEN_STRING_INITIALIZER(struct cmd_set_cpu_freq_result, - cmd, "up#down#min#max"); + cmd, "up#down#min#max#enable_turbo#disable_turbo"); cmdline_parse_inst_t cmd_set_cpu_freq_set = { .f = cmd_set_cpu_freq_parsed, .data = NULL, - .help_str = "set_cpu_freq <core_num> <up|down|min|max>, Set the current " - "frequency for the specified core by scaling up/down/min/max", + .help_str = "set_cpu_freq <core_num> " + "<up|down|min|max|enable_turbo|disable_turbo>, " + "adjust the frequency for the specified core.", .tokens = { (void *)&cmd_set_cpu_freq, (void *)&cmd_set_cpu_freq_core_num, @@ -135,8 +142,103 @@ cmdline_parse_inst_t cmd_set_cpu_freq_set = { }, }; +struct cmd_send_policy_result { + cmdline_fixed_string_t send_policy; + cmdline_fixed_string_t cmd; +}; + +union PFID { + struct ether_addr addr; + uint64_t pfid; +}; + +static inline int +send_policy(void) +{ + struct channel_packet pkt; + int ret; + + union PFID pfid; + /* Use port MAC address as the vfid */ + rte_eth_macaddr_get(0, &pfid.addr); + printf("Port %u MAC: %02" PRIx8 ":%02" PRIx8 ":%02" PRIx8 ":" + "%02" PRIx8 ":%02" PRIx8 ":%02" PRIx8 "\n", + 1, + pfid.addr.addr_bytes[0], pfid.addr.addr_bytes[1], + pfid.addr.addr_bytes[2], pfid.addr.addr_bytes[3], + pfid.addr.addr_bytes[4], pfid.addr.addr_bytes[5]); + pkt.vfid[0] = pfid.pfid; + + pkt.nb_mac_to_monitor = 1; + pkt.t_boost_status.tbEnabled = false; + + pkt.vcpu_to_control[0] = 0; + pkt.vcpu_to_control[1] = 1; + pkt.num_vcpu = 2; + /* Dummy Population. */ + pkt.traffic_policy.min_packet_thresh = 96000; + pkt.traffic_policy.avg_max_packet_thresh = 1800000; + pkt.traffic_policy.max_max_packet_thresh = 2000000; + + pkt.timer_policy.busy_hours[0] = 3; + pkt.timer_policy.busy_hours[1] = 4; + pkt.timer_policy.busy_hours[2] = 5; + pkt.timer_policy.quiet_hours[0] = 11; + pkt.timer_policy.quiet_hours[1] = 12; + pkt.timer_policy.quiet_hours[2] = 13; + + pkt.timer_policy.hours_to_use_traffic_profile[0] = 8; + pkt.timer_policy.hours_to_use_traffic_profile[1] = 10; + + pkt.workload = LOW; + pkt.policy_to_use = TIME; + pkt.command = PKT_POLICY; + strcpy(pkt.vm_name, "ubuntu2"); + ret = rte_power_guest_channel_send_msg(&pkt, 1); + if (ret == 0) + return 1; + RTE_LOG(DEBUG, POWER, "Error sending message: %s\n", + ret > 0 ? strerror(ret) : "channel not connected"); + return -1; +} + +static void +cmd_send_policy_parsed(void *parsed_result, struct cmdline *cl, + __attribute__((unused)) void *data) +{ + int ret = -1; + struct cmd_send_policy_result *res = parsed_result; + + if (!strcmp(res->cmd, "now")) { + printf("Sending Policy down now!\n"); + ret = send_policy(); + } + if (ret != 1) + cmdline_printf(cl, "Error sending message: %s\n", + strerror(ret)); +} + +cmdline_parse_token_string_t cmd_send_policy = + TOKEN_STRING_INITIALIZER(struct cmd_send_policy_result, + send_policy, "send_policy"); +cmdline_parse_token_string_t cmd_send_policy_cmd_cmd = + TOKEN_STRING_INITIALIZER(struct cmd_send_policy_result, + cmd, "now"); + +cmdline_parse_inst_t cmd_send_policy_set = { + .f = cmd_send_policy_parsed, + .data = NULL, + .help_str = "send_policy now", + .tokens = { + (void *)&cmd_send_policy, + (void *)&cmd_send_policy_cmd_cmd, + NULL, + }, +}; + cmdline_parse_ctx_t main_ctx[] = { (cmdline_parse_inst_t *)&cmd_quit, + (cmdline_parse_inst_t *)&cmd_send_policy_set, (cmdline_parse_inst_t *)&cmd_set_cpu_freq_set, NULL, }; diff --git a/examples/vm_power_manager/guest_cli/vm_power_cli_guest.h b/examples/vm_power_manager/guest_cli/vm_power_cli_guest.h index 0c4bdd5b..277eab3d 100644 --- a/examples/vm_power_manager/guest_cli/vm_power_cli_guest.h +++ b/examples/vm_power_manager/guest_cli/vm_power_cli_guest.h @@ -40,12 +40,6 @@ extern "C" { #include "channel_commands.h" -int guest_channel_host_connect(unsigned lcore_id); - -int guest_channel_send_msg(struct channel_packet *pkt, unsigned lcore_id); - -void guest_channel_host_disconnect(unsigned lcore_id); - void run_cli(__attribute__((unused)) void *arg); #ifdef __cplusplus diff --git a/examples/vm_power_manager/main.c b/examples/vm_power_manager/main.c index c33fcc93..399fbdd4 100644 --- a/examples/vm_power_manager/main.c +++ b/examples/vm_power_manager/main.c @@ -49,14 +49,206 @@ #include <rte_log.h> #include <rte_per_lcore.h> #include <rte_lcore.h> +#include <rte_ethdev.h> +#include <getopt.h> +#include <rte_cycles.h> #include <rte_debug.h> #include "channel_manager.h" #include "channel_monitor.h" #include "power_manager.h" #include "vm_power_cli.h" +#include <rte_pmd_ixgbe.h> +#include <rte_pmd_i40e.h> +#include <rte_pmd_bnxt.h> + +#define RX_RING_SIZE 512 +#define TX_RING_SIZE 512 + +#define NUM_MBUFS 8191 +#define MBUF_CACHE_SIZE 250 +#define BURST_SIZE 32 + +static uint32_t enabled_port_mask; +static volatile bool force_quit; + +/****************/ +static const struct rte_eth_conf port_conf_default = { + .rxmode = { .max_rx_pkt_len = ETHER_MAX_LEN } +}; + +static inline int +port_init(uint16_t port, struct rte_mempool *mbuf_pool) +{ + struct rte_eth_conf port_conf = port_conf_default; + const uint16_t rx_rings = 1, tx_rings = 1; + int retval; + uint16_t q; + + if (port >= rte_eth_dev_count()) + return -1; + + /* Configure the Ethernet device. */ + retval = rte_eth_dev_configure(port, rx_rings, tx_rings, &port_conf); + if (retval != 0) + return retval; + + /* Allocate and set up 1 RX queue per Ethernet port. */ + for (q = 0; q < rx_rings; q++) { + retval = rte_eth_rx_queue_setup(port, q, RX_RING_SIZE, + rte_eth_dev_socket_id(port), NULL, mbuf_pool); + if (retval < 0) + return retval; + } + + /* Allocate and set up 1 TX queue per Ethernet port. */ + for (q = 0; q < tx_rings; q++) { + retval = rte_eth_tx_queue_setup(port, q, TX_RING_SIZE, + rte_eth_dev_socket_id(port), NULL); + if (retval < 0) + return retval; + } + + /* Start the Ethernet port. */ + retval = rte_eth_dev_start(port); + if (retval < 0) + return retval; + + /* Display the port MAC address. */ + struct ether_addr addr; + rte_eth_macaddr_get(port, &addr); + printf("Port %u MAC: %02" PRIx8 " %02" PRIx8 " %02" PRIx8 + " %02" PRIx8 " %02" PRIx8 " %02" PRIx8 "\n", + (unsigned int)port, + addr.addr_bytes[0], addr.addr_bytes[1], + addr.addr_bytes[2], addr.addr_bytes[3], + addr.addr_bytes[4], addr.addr_bytes[5]); + + /* Enable RX in promiscuous mode for the Ethernet device. */ + rte_eth_promiscuous_enable(port); + + + return 0; +} static int +parse_portmask(const char *portmask) +{ + char *end = NULL; + unsigned long pm; + + /* parse hexadecimal string */ + pm = strtoul(portmask, &end, 16); + if ((portmask[0] == '\0') || (end == NULL) || (*end != '\0')) + return -1; + + if (pm == 0) + return -1; + + return pm; +} +/* Parse the argument given in the command line of the application */ +static int +parse_args(int argc, char **argv) +{ + int opt, ret; + char **argvopt; + int option_index; + char *prgname = argv[0]; + static struct option lgopts[] = { + { "mac-updating", no_argument, 0, 1}, + { "no-mac-updating", no_argument, 0, 0}, + {NULL, 0, 0, 0} + }; + argvopt = argv; + + while ((opt = getopt_long(argc, argvopt, "p:q:T:", + lgopts, &option_index)) != EOF) { + + switch (opt) { + /* portmask */ + case 'p': + enabled_port_mask = parse_portmask(optarg); + if (enabled_port_mask == 0) { + printf("invalid portmask\n"); + return -1; + } + break; + /* long options */ + case 0: + break; + + default: + return -1; + } + } + + if (optind >= 0) + argv[optind-1] = prgname; + + ret = optind-1; + optind = 0; /* reset getopt lib */ + return ret; +} + +static void +check_all_ports_link_status(uint16_t port_num, uint32_t port_mask) +{ +#define CHECK_INTERVAL 100 /* 100ms */ +#define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */ + uint16_t portid, count, all_ports_up, print_flag = 0; + struct rte_eth_link link; + + printf("\nChecking link status"); + fflush(stdout); + for (count = 0; count <= MAX_CHECK_TIME; count++) { + if (force_quit) + return; + all_ports_up = 1; + for (portid = 0; portid < port_num; portid++) { + if (force_quit) + return; + if ((port_mask & (1 << portid)) == 0) + continue; + memset(&link, 0, sizeof(link)); + rte_eth_link_get_nowait(portid, &link); + /* print link status if flag set */ + if (print_flag == 1) { + if (link.link_status) + printf("Port %d Link Up - speed %u " + "Mbps - %s\n", (uint16_t)portid, + (unsigned int)link.link_speed, + (link.link_duplex == ETH_LINK_FULL_DUPLEX) ? + ("full-duplex") : ("half-duplex\n")); + else + printf("Port %d Link Down\n", + (uint16_t)portid); + continue; + } + /* clear all_ports_up flag if any link down */ + if (link.link_status == ETH_LINK_DOWN) { + all_ports_up = 0; + break; + } + } + /* after finally printing all link status, get out */ + if (print_flag == 1) + break; + + if (all_ports_up == 0) { + printf("."); + fflush(stdout); + rte_delay_ms(CHECK_INTERVAL); + } + + /* set the print_flag if all ports up or timeout */ + if (all_ports_up == 1 || count == (MAX_CHECK_TIME - 1)) { + print_flag = 1; + printf("done\n"); + } + } +} +static int run_monitor(__attribute__((unused)) void *arg) { if (channel_monitor_init() < 0) { @@ -82,6 +274,10 @@ main(int argc, char **argv) { int ret; unsigned lcore_id; + unsigned int nb_ports; + struct rte_mempool *mbuf_pool; + uint16_t portid; + ret = rte_eal_init(argc, argv); if (ret < 0) @@ -90,12 +286,77 @@ main(int argc, char **argv) signal(SIGINT, sig_handler); signal(SIGTERM, sig_handler); + argc -= ret; + argv += ret; + + /* parse application arguments (after the EAL ones) */ + ret = parse_args(argc, argv); + if (ret < 0) + rte_exit(EXIT_FAILURE, "Invalid arguments\n"); + + nb_ports = rte_eth_dev_count(); + + mbuf_pool = rte_pktmbuf_pool_create("MBUF_POOL", NUM_MBUFS * nb_ports, + MBUF_CACHE_SIZE, 0, RTE_MBUF_DEFAULT_BUF_SIZE, rte_socket_id()); + + if (mbuf_pool == NULL) + rte_exit(EXIT_FAILURE, "Cannot create mbuf pool\n"); + + /* Initialize ports. */ + for (portid = 0; portid < nb_ports; portid++) { + struct ether_addr eth; + int w, j; + int ret = -ENOTSUP; + + if ((enabled_port_mask & (1 << portid)) == 0) + continue; + + eth.addr_bytes[0] = 0xe0; + eth.addr_bytes[1] = 0xe0; + eth.addr_bytes[2] = 0xe0; + eth.addr_bytes[3] = 0xe0; + eth.addr_bytes[4] = portid + 0xf0; + + if (port_init(portid, mbuf_pool) != 0) + rte_exit(EXIT_FAILURE, "Cannot init port %"PRIu8 "\n", + portid); + + for (w = 0; w < MAX_VFS; w++) { + eth.addr_bytes[5] = w + 0xf0; + + if (ret == -ENOTSUP) + ret = rte_pmd_ixgbe_set_vf_mac_addr(portid, + w, ð); + if (ret == -ENOTSUP) + ret = rte_pmd_i40e_set_vf_mac_addr(portid, + w, ð); + if (ret == -ENOTSUP) + ret = rte_pmd_bnxt_set_vf_mac_addr(portid, + w, ð); + + switch (ret) { + case 0: + printf("Port %d VF %d MAC: ", + portid, w); + for (j = 0; j < 6; j++) { + printf("%02x", eth.addr_bytes[j]); + if (j < 5) + printf(":"); + } + printf("\n"); + break; + } + } + } + lcore_id = rte_get_next_lcore(-1, 1, 0); if (lcore_id == RTE_MAX_LCORE) { RTE_LOG(ERR, EAL, "A minimum of two cores are required to run " "application\n"); return 0; } + + check_all_ports_link_status(nb_ports, enabled_port_mask); rte_eal_remote_launch(run_monitor, NULL, lcore_id); if (power_manager_init() < 0) { diff --git a/examples/vm_power_manager/power_manager.c b/examples/vm_power_manager/power_manager.c index 2644fce6..1834a823 100644 --- a/examples/vm_power_manager/power_manager.c +++ b/examples/vm_power_manager/power_manager.c @@ -108,7 +108,7 @@ set_host_cpus_mask(void) int power_manager_init(void) { - unsigned i, num_cpus; + unsigned int i, num_cpus, num_freqs; uint64_t cpu_mask; int ret = 0; @@ -121,15 +121,21 @@ power_manager_init(void) rte_power_set_env(PM_ENV_ACPI_CPUFREQ); cpu_mask = global_enabled_cpus; for (i = 0; cpu_mask; cpu_mask &= ~(1 << i++)) { - if (rte_power_init(i) < 0 || rte_power_freqs(i, - global_core_freq_info[i].freqs, - RTE_MAX_LCORE_FREQS) == 0) { - RTE_LOG(ERR, POWER_MANAGER, "Unable to initialize power manager " + if (rte_power_init(i) < 0) + RTE_LOG(ERR, POWER_MANAGER, + "Unable to initialize power manager " "for core %u\n", i); + num_freqs = rte_power_freqs(i, global_core_freq_info[i].freqs, + RTE_MAX_LCORE_FREQS); + if (num_freqs == 0) { + RTE_LOG(ERR, POWER_MANAGER, + "Unable to get frequency list for core %u\n", + i); global_enabled_cpus &= ~(1 << i); num_cpus--; ret = -1; } + global_core_freq_info[i].num_freqs = num_freqs; rte_spinlock_init(&global_core_freq_info[i].power_sl); } RTE_LOG(INFO, POWER_MANAGER, "Detected %u host CPUs , enabled core mask:" @@ -216,6 +222,24 @@ power_manager_scale_mask_max(uint64_t core_mask) } int +power_manager_enable_turbo_mask(uint64_t core_mask) +{ + int ret = 0; + + POWER_SCALE_MASK(enable_turbo, core_mask, ret); + return ret; +} + +int +power_manager_disable_turbo_mask(uint64_t core_mask) +{ + int ret = 0; + + POWER_SCALE_MASK(disable_turbo, core_mask, ret); + return ret; +} + +int power_manager_scale_core_up(unsigned core_num) { int ret = 0; @@ -250,3 +274,37 @@ power_manager_scale_core_max(unsigned core_num) POWER_SCALE_CORE(max, core_num, ret); return ret; } + +int +power_manager_enable_turbo_core(unsigned int core_num) +{ + int ret = 0; + + POWER_SCALE_CORE(enable_turbo, core_num, ret); + return ret; +} + +int +power_manager_disable_turbo_core(unsigned int core_num) +{ + int ret = 0; + + POWER_SCALE_CORE(disable_turbo, core_num, ret); + return ret; +} + +int +power_manager_scale_core_med(unsigned int core_num) +{ + int ret = 0; + + if (core_num >= POWER_MGR_MAX_CPUS) + return -1; + if (!(global_enabled_cpus & (1ULL << core_num))) + return -1; + rte_spinlock_lock(&global_core_freq_info[core_num].power_sl); + ret = rte_power_set_freq(core_num, + global_core_freq_info[core_num].num_freqs / 2); + rte_spinlock_unlock(&global_core_freq_info[core_num].power_sl); + return ret; +} diff --git a/examples/vm_power_manager/power_manager.h b/examples/vm_power_manager/power_manager.h index 1b45babf..b52fb4c7 100644 --- a/examples/vm_power_manager/power_manager.h +++ b/examples/vm_power_manager/power_manager.h @@ -113,6 +113,32 @@ int power_manager_scale_mask_min(uint64_t core_mask); int power_manager_scale_mask_max(uint64_t core_mask); /** + * Enable Turbo Boost on the cores specified in core_mask. + * It is thread-safe. + * + * @param core_mask + * The uint64_t bit-mask of cores to change frequency. + * + * @return + * - 1 on success. + * - Negative on error. + */ +int power_manager_enable_turbo_mask(uint64_t core_mask); + +/** + * Disable Turbo Boost on the cores specified in core_mask. + * It is thread-safe. + * + * @param core_mask + * The uint64_t bit-mask of cores to change frequency. + * + * @return + * - 1 on success. + * - Negative on error. + */ +int power_manager_disable_turbo_mask(uint64_t core_mask); + +/** * Scale up frequency for the core specified by core_num. * It is thread-safe. * @@ -168,6 +194,32 @@ int power_manager_scale_core_min(unsigned core_num); int power_manager_scale_core_max(unsigned core_num); /** + * Enable Turbo Boost for the core specified by core_num. + * It is thread-safe. + * + * @param core_num + * The core number to boost + * + * @return + * - 1 on success. + * - Negative on error. + */ +int power_manager_enable_turbo_core(unsigned int core_num); + +/** + * Disable Turbo Boost for the core specified by core_num. + * It is thread-safe. + * + * @param core_num + * The core number to boost + * + * @return + * - 1 on success. + * - Negative on error. + */ +int power_manager_disable_turbo_core(unsigned int core_num); + +/** * Get the current freuency of the core specified by core_num * * @param core_num @@ -179,6 +231,19 @@ int power_manager_scale_core_max(unsigned core_num); */ uint32_t power_manager_get_current_frequency(unsigned core_num); +/** + * Scale to medium frequency for the core specified by core_num. + * It is thread-safe. + * + * @param core_num + * The core number to change frequency + * + * @return + * - 1 on success. + * - 0 if frequency not changed. + * - Negative on error. + */ +int power_manager_scale_core_med(unsigned int core_num); #ifdef __cplusplus } diff --git a/examples/vm_power_manager/vm_power_cli.c b/examples/vm_power_manager/vm_power_cli.c index c5e8d934..6f234fb7 100644 --- a/examples/vm_power_manager/vm_power_cli.c +++ b/examples/vm_power_manager/vm_power_cli.c @@ -520,6 +520,10 @@ cmd_set_cpu_freq_mask_parsed(void *parsed_result, struct cmdline *cl, ret = power_manager_scale_mask_min(res->core_mask); else if (!strcmp(res->cmd , "max")) ret = power_manager_scale_mask_max(res->core_mask); + else if (!strcmp(res->cmd, "enable_turbo")) + ret = power_manager_enable_turbo_mask(res->core_mask); + else if (!strcmp(res->cmd, "disable_turbo")) + ret = power_manager_disable_turbo_mask(res->core_mask); if (ret < 0) { cmdline_printf(cl, "Error scaling core_mask(0x%"PRIx64") '%s' , not " "all cores specified have been scaled\n", @@ -535,14 +539,13 @@ cmdline_parse_token_num_t cmd_set_cpu_freq_mask_core_mask = core_mask, UINT64); cmdline_parse_token_string_t cmd_set_cpu_freq_mask_result = TOKEN_STRING_INITIALIZER(struct cmd_set_cpu_freq_mask_result, - cmd, "up#down#min#max"); + cmd, "up#down#min#max#enable_turbo#disable_turbo"); cmdline_parse_inst_t cmd_set_cpu_freq_mask_set = { .f = cmd_set_cpu_freq_mask_parsed, .data = NULL, - .help_str = "set_cpu_freq <core_mask> <up|down|min|max>, Set the current " - "frequency for the cores specified in <core_mask> by scaling " - "each up/down/min/max.", + .help_str = "set_cpu_freq <core_mask> <up|down|min|max|enable_turbo|disable_turbo>, adjust the current " + "frequency for the cores specified in <core_mask>", .tokens = { (void *)&cmd_set_cpu_freq_mask, (void *)&cmd_set_cpu_freq_mask_core_mask, @@ -614,6 +617,10 @@ cmd_set_cpu_freq_parsed(void *parsed_result, struct cmdline *cl, ret = power_manager_scale_core_min(res->core_num); else if (!strcmp(res->cmd , "max")) ret = power_manager_scale_core_max(res->core_num); + else if (!strcmp(res->cmd, "enable_turbo")) + ret = power_manager_enable_turbo_core(res->core_num); + else if (!strcmp(res->cmd, "disable_turbo")) + ret = power_manager_disable_turbo_core(res->core_num); if (ret < 0) { cmdline_printf(cl, "Error scaling core(%u) '%s'\n", res->core_num, res->cmd); @@ -628,13 +635,13 @@ cmdline_parse_token_num_t cmd_set_cpu_freq_core_num = core_num, UINT8); cmdline_parse_token_string_t cmd_set_cpu_freq_cmd_cmd = TOKEN_STRING_INITIALIZER(struct cmd_set_cpu_freq_result, - cmd, "up#down#min#max"); + cmd, "up#down#min#max#enable_turbo#disable_turbo"); cmdline_parse_inst_t cmd_set_cpu_freq_set = { .f = cmd_set_cpu_freq_parsed, .data = NULL, - .help_str = "set_cpu_freq <core_num> <up|down|min|max>, Set the current " - "frequency for the specified core by scaling up/down/min/max", + .help_str = "set_cpu_freq <core_num> <up|down|min|max|enable_turbo|disable_turbo>, adjust the current " + "frequency for the specified core", .tokens = { (void *)&cmd_set_cpu_freq, (void *)&cmd_set_cpu_freq_core_num, diff --git a/examples/vmdq/main.c b/examples/vmdq/main.c index 8949a115..84e9937d 100644 --- a/examples/vmdq/main.c +++ b/examples/vmdq/main.c @@ -47,7 +47,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -57,7 +56,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -121,7 +119,7 @@ static const struct rte_eth_conf vmdq_conf_default = { }; static unsigned lcore_ids[RTE_MAX_LCORE]; -static uint8_t ports[RTE_MAX_ETHPORTS]; +static uint16_t ports[RTE_MAX_ETHPORTS]; static unsigned num_ports; /**< The number of ports specified in command line */ /* array used for printing out statistics */ @@ -186,7 +184,7 @@ get_eth_conf(struct rte_eth_conf *eth_conf, uint32_t num_pools) * coming from the mbuf_pool passed as parameter */ static inline int -port_init(uint8_t port, struct rte_mempool *mbuf_pool) +port_init(uint16_t port, struct rte_mempool *mbuf_pool) { struct rte_eth_dev_info dev_info; struct rte_eth_rxconf *rxconf; @@ -583,7 +581,7 @@ main(int argc, char *argv[]) unsigned lcore_id, core_id = 0; int ret; unsigned nb_ports, valid_num_ports; - uint8_t portid; + uint16_t portid; signal(SIGHUP, sighup_handler); diff --git a/examples/vmdq_dcb/main.c b/examples/vmdq_dcb/main.c index b6ebccb2..9dad2b8e 100644 --- a/examples/vmdq_dcb/main.c +++ b/examples/vmdq_dcb/main.c @@ -47,7 +47,6 @@ #include <rte_log.h> #include <rte_memory.h> #include <rte_memcpy.h> -#include <rte_memzone.h> #include <rte_eal.h> #include <rte_launch.h> #include <rte_atomic.h> @@ -57,7 +56,6 @@ #include <rte_per_lcore.h> #include <rte_branch_prediction.h> #include <rte_interrupts.h> -#include <rte_pci.h> #include <rte_random.h> #include <rte_debug.h> #include <rte_ether.h> @@ -87,7 +85,7 @@ /* mask of enabled ports */ static uint32_t enabled_port_mask; -static uint8_t ports[RTE_MAX_ETHPORTS]; +static uint16_t ports[RTE_MAX_ETHPORTS]; static unsigned num_ports; /* number of pools (if user does not specify any, 32 by default */ @@ -220,7 +218,7 @@ get_eth_conf(struct rte_eth_conf *eth_conf) * coming from the mbuf_pool passed as parameter */ static inline int -port_init(uint8_t port, struct rte_mempool *mbuf_pool) +port_init(uint16_t port, struct rte_mempool *mbuf_pool) { struct rte_eth_dev_info dev_info; struct rte_eth_conf port_conf = {0}; @@ -646,7 +644,7 @@ main(int argc, char *argv[]) uintptr_t i; int ret; unsigned nb_ports, valid_num_ports; - uint8_t portid; + uint16_t portid; signal(SIGHUP, sighup_handler); |