diff options
Diffstat (limited to 'hw/net/e1000e_core.c')
| -rw-r--r-- | hw/net/e1000e_core.c | 297 |
1 files changed, 113 insertions, 184 deletions
diff --git a/hw/net/e1000e_core.c b/hw/net/e1000e_core.c index c0c09b6965..9f185d099c 100644 --- a/hw/net/e1000e_core.c +++ b/hw/net/e1000e_core.c @@ -165,14 +165,14 @@ e1000e_intrmgr_on_throttling_timer(void *opaque) timer->running = false; - if (msi_enabled(timer->core->owner)) { - trace_e1000e_irq_msi_notify_postponed(); - /* Clear msi_causes_pending to fire MSI eventually */ - timer->core->msi_causes_pending = 0; - e1000e_set_interrupt_cause(timer->core, 0); - } else { - trace_e1000e_irq_legacy_notify_postponed(); - e1000e_set_interrupt_cause(timer->core, 0); + if (timer->core->mac[IMS] & timer->core->mac[ICR]) { + if (msi_enabled(timer->core->owner)) { + trace_e1000e_irq_msi_notify_postponed(); + msi_notify(timer->core->owner, 0); + } else { + trace_e1000e_irq_legacy_notify_postponed(); + e1000e_raise_legacy_irq(timer->core); + } } } @@ -366,10 +366,6 @@ static void e1000e_intrmgr_fire_all_timers(E1000ECore *core) { int i; - uint32_t val = e1000e_intmgr_collect_delayed_causes(core); - - trace_e1000e_irq_adding_delayed_causes(val, core->mac[ICR]); - core->mac[ICR] |= val; if (core->itr.running) { timer_del(core->itr.timer); @@ -537,7 +533,7 @@ e1000e_rss_get_hash_type(E1000ECore *core, struct NetRxPkt *pkt) ip6info->rss_ex_dst_valid, ip6info->rss_ex_src_valid, core->mac[MRQC], - E1000_MRQC_EN_TCPIPV6(core->mac[MRQC]), + E1000_MRQC_EN_TCPIPV6EX(core->mac[MRQC]), E1000_MRQC_EN_IPV6EX(core->mac[MRQC]), E1000_MRQC_EN_IPV6(core->mac[MRQC])); @@ -546,8 +542,8 @@ e1000e_rss_get_hash_type(E1000ECore *core, struct NetRxPkt *pkt) ip6info->rss_ex_src_valid))) { if (l4hdr_proto == ETH_L4_HDR_PROTO_TCP && - E1000_MRQC_EN_TCPIPV6(core->mac[MRQC])) { - return E1000_MRQ_RSS_TYPE_IPV6TCP; + E1000_MRQC_EN_TCPIPV6EX(core->mac[MRQC])) { + return E1000_MRQ_RSS_TYPE_IPV6TCPEX; } if (E1000_MRQC_EN_IPV6EX(core->mac[MRQC])) { @@ -581,7 +577,7 @@ e1000e_rss_calc_hash(E1000ECore *core, case E1000_MRQ_RSS_TYPE_IPV4TCP: type = NetPktRssIpV4Tcp; break; - case E1000_MRQ_RSS_TYPE_IPV6TCP: + case E1000_MRQ_RSS_TYPE_IPV6TCPEX: type = NetPktRssIpV6TcpEx; break; case E1000_MRQ_RSS_TYPE_IPV6: @@ -711,9 +707,8 @@ e1000e_on_tx_done_update_stats(E1000ECore *core, struct NetTxPkt *tx_pkt) g_assert_not_reached(); } - core->mac[GPTC] = core->mac[TPT]; - core->mac[GOTCL] = core->mac[TOTL]; - core->mac[GOTCH] = core->mac[TOTH]; + e1000x_inc_reg_if_not_full(core->mac, GPTC); + e1000x_grow_8reg_if_not_full(core->mac, GOTCL, tot_len); } static void @@ -747,7 +742,8 @@ e1000e_process_tx_desc(E1000ECore *core, addr = le64_to_cpu(dp->buffer_addr); if (!tx->skip_cp) { - if (!net_tx_pkt_add_raw_fragment(tx->tx_pkt, addr, split_size)) { + if (!net_tx_pkt_add_raw_fragment_pci(tx->tx_pkt, core->owner, + addr, split_size)) { tx->skip_cp = true; } } @@ -765,7 +761,7 @@ e1000e_process_tx_desc(E1000ECore *core, } tx->skip_cp = false; - net_tx_pkt_reset(tx->tx_pkt, core->owner); + net_tx_pkt_reset(tx->tx_pkt, net_tx_pkt_unmap_frag_pci, core->owner); tx->sum_needed = 0; tx->cptse = 0; @@ -959,6 +955,8 @@ e1000e_start_xmit(E1000ECore *core, const E1000E_TxRing *txr) if (!ide || !e1000e_intrmgr_delay_tx_causes(core, &cause)) { e1000e_set_interrupt_cause(core, cause); } + + net_tx_pkt_reset(txr->tx->tx_pkt, net_tx_pkt_unmap_frag_pci, core->owner); } static bool @@ -1034,48 +1032,11 @@ e1000e_rx_l4_cso_enabled(E1000ECore *core) } static bool -e1000e_receive_filter(E1000ECore *core, const uint8_t *buf, int size) +e1000e_receive_filter(E1000ECore *core, const void *buf) { - uint32_t rctl = core->mac[RCTL]; - - if (e1000x_is_vlan_packet(buf, core->mac[VET]) && - e1000x_vlan_rx_filter_enabled(core->mac)) { - uint16_t vid = lduw_be_p(&PKT_GET_VLAN_HDR(buf)->h_tci); - uint32_t vfta = - ldl_le_p((uint32_t *)(core->mac + VFTA) + - ((vid >> E1000_VFTA_ENTRY_SHIFT) & E1000_VFTA_ENTRY_MASK)); - if ((vfta & (1 << (vid & E1000_VFTA_ENTRY_BIT_SHIFT_MASK))) == 0) { - trace_e1000e_rx_flt_vlan_mismatch(vid); - return false; - } else { - trace_e1000e_rx_flt_vlan_match(vid); - } - } - - switch (net_rx_pkt_get_packet_type(core->rx_pkt)) { - case ETH_PKT_UCAST: - if (rctl & E1000_RCTL_UPE) { - return true; /* promiscuous ucast */ - } - break; - - case ETH_PKT_BCAST: - if (rctl & E1000_RCTL_BAM) { - return true; /* broadcast enabled */ - } - break; - - case ETH_PKT_MCAST: - if (rctl & E1000_RCTL_MPE) { - return true; /* promiscuous mcast */ - } - break; - - default: - g_assert_not_reached(); - } - - return e1000x_rx_group_filter(core->mac, buf); + return (!e1000x_is_vlan_packet(buf, core->mac[VET]) || + e1000x_rx_vlan_filter(core->mac, PKT_GET_VLAN_HDR(buf))) && + e1000x_rx_group_filter(core->mac, buf); } static inline void @@ -1149,6 +1110,11 @@ e1000e_verify_csum_in_sw(E1000ECore *core, return; } + if (l4hdr_proto != ETH_L4_HDR_PROTO_TCP && + l4hdr_proto != ETH_L4_HDR_PROTO_UDP) { + return; + } + if (!net_rx_pkt_validate_l4_csum(pkt, &csum_valid)) { trace_e1000e_rx_metadata_l4_csum_validation_failed(); return; @@ -1281,9 +1247,8 @@ e1000e_build_rx_metadata(E1000ECore *core, trace_e1000e_rx_metadata_l4_cso_disabled(); } - trace_e1000e_rx_metadata_status_flags(*status_flags); - func_exit: + trace_e1000e_rx_metadata_status_flags(*status_flags); *status_flags = cpu_to_le32(*status_flags); } @@ -1488,24 +1453,10 @@ e1000e_write_to_rx_buffers(E1000ECore *core, } static void -e1000e_update_rx_stats(E1000ECore *core, - size_t data_size, - size_t data_fcs_size) +e1000e_update_rx_stats(E1000ECore *core, size_t pkt_size, size_t pkt_fcs_size) { - e1000x_update_rx_total_stats(core->mac, data_size, data_fcs_size); - - switch (net_rx_pkt_get_packet_type(core->rx_pkt)) { - case ETH_PKT_BCAST: - e1000x_inc_reg_if_not_full(core->mac, BPRC); - break; - - case ETH_PKT_MCAST: - e1000x_inc_reg_if_not_full(core->mac, MPRC); - break; - - default: - break; - } + eth_pkt_types_e pkt_type = net_rx_pkt_get_packet_type(core->rx_pkt); + e1000x_update_rx_total_stats(core->mac, pkt_type, pkt_size, pkt_fcs_size); } static inline bool @@ -1700,12 +1651,9 @@ static ssize_t e1000e_receive_internal(E1000ECore *core, const struct iovec *iov, int iovcnt, bool has_vnet) { - static const int maximum_ethernet_hdr_len = (ETH_HLEN + 4); - - uint32_t n = 0; - uint8_t min_buf[ETH_ZLEN]; + uint32_t causes = 0; + uint8_t buf[ETH_ZLEN]; struct iovec min_iov; - uint8_t *filter_buf; size_t size, orig_size; size_t iov_ofs = 0; E1000E_RxRing rxr; @@ -1728,24 +1676,21 @@ e1000e_receive_internal(E1000ECore *core, const struct iovec *iov, int iovcnt, net_rx_pkt_unset_vhdr(core->rx_pkt); } - filter_buf = iov->iov_base + iov_ofs; orig_size = iov_size(iov, iovcnt); size = orig_size - iov_ofs; /* Pad to minimum Ethernet frame length */ - if (size < sizeof(min_buf)) { - iov_to_buf(iov, iovcnt, iov_ofs, min_buf, size); - memset(&min_buf[size], 0, sizeof(min_buf) - size); + if (size < sizeof(buf)) { + iov_to_buf(iov, iovcnt, iov_ofs, buf, size); + memset(&buf[size], 0, sizeof(buf) - size); e1000x_inc_reg_if_not_full(core->mac, RUC); - min_iov.iov_base = filter_buf = min_buf; - min_iov.iov_len = size = sizeof(min_buf); + min_iov.iov_base = buf; + min_iov.iov_len = size = sizeof(buf); iovcnt = 1; iov = &min_iov; iov_ofs = 0; - } else if (iov->iov_len < maximum_ethernet_hdr_len) { - /* This is very unlikely, but may happen. */ - iov_to_buf(iov, iovcnt, iov_ofs, min_buf, maximum_ethernet_hdr_len); - filter_buf = min_buf; + } else { + iov_to_buf(iov, iovcnt, iov_ofs, buf, ETH_HLEN + 4); } /* Discard oversized packets if !LPE and !SBP. */ @@ -1754,15 +1699,16 @@ e1000e_receive_internal(E1000ECore *core, const struct iovec *iov, int iovcnt, } net_rx_pkt_set_packet_type(core->rx_pkt, - get_eth_packet_type(PKT_GET_ETH_HDR(filter_buf))); + get_eth_packet_type(PKT_GET_ETH_HDR(buf))); - if (!e1000e_receive_filter(core, filter_buf, size)) { + if (!e1000e_receive_filter(core, buf)) { trace_e1000e_rx_flt_dropped(); return orig_size; } net_rx_pkt_attach_iovec_ex(core->rx_pkt, iov, iovcnt, iov_ofs, - e1000x_vlan_enabled(core->mac), core->mac[VET]); + e1000x_vlan_enabled(core->mac) ? 0 : -1, + core->mac[VET], 0); e1000e_rss_parse_packet(core, core->rx_pkt, &rss_info); e1000e_rx_ring_init(core, &rxr, rss_info.queue); @@ -1779,32 +1725,32 @@ e1000e_receive_internal(E1000ECore *core, const struct iovec *iov, int iovcnt, /* Perform small receive detection (RSRPD) */ if (total_size < core->mac[RSRPD]) { - n |= E1000_ICS_SRPD; + causes |= E1000_ICS_SRPD; } /* Perform ACK receive detection */ if (!(core->mac[RFCTL] & E1000_RFCTL_ACK_DIS) && (e1000e_is_tcp_ack(core, core->rx_pkt))) { - n |= E1000_ICS_ACK; + causes |= E1000_ICS_ACK; } /* Check if receive descriptor minimum threshold hit */ rdmts_hit = e1000e_rx_descr_threshold_hit(core, rxr.i); - n |= e1000e_rx_wb_interrupt_cause(core, rxr.i->idx, rdmts_hit); + causes |= e1000e_rx_wb_interrupt_cause(core, rxr.i->idx, rdmts_hit); trace_e1000e_rx_written_to_guest(rxr.i->idx); } else { - n |= E1000_ICS_RXO; + causes |= E1000_ICS_RXO; retval = 0; trace_e1000e_rx_not_written_to_guest(rxr.i->idx); } - if (!e1000e_intrmgr_delay_rx_causes(core, &n)) { - trace_e1000e_rx_interrupt_set(n); - e1000e_set_interrupt_cause(core, n); + if (!e1000e_intrmgr_delay_rx_causes(core, &causes)) { + trace_e1000e_rx_interrupt_set(causes); + e1000e_set_interrupt_cause(core, causes); } else { - trace_e1000e_rx_interrupt_delayed(n); + trace_e1000e_rx_interrupt_delayed(causes); } return retval; @@ -2024,13 +1970,6 @@ void(*e1000e_phyreg_writeops[E1000E_PHY_PAGES][E1000E_PHY_PAGE_SIZE]) } }; -static inline void -e1000e_clear_ims_bits(E1000ECore *core, uint32_t bits) -{ - trace_e1000e_irq_clear_ims(bits, core->mac[IMS], core->mac[IMS] & ~bits); - core->mac[IMS] &= ~bits; -} - static inline bool e1000e_postpone_interrupt(E1000IntrDelayTimer *timer) { @@ -2088,7 +2027,6 @@ e1000e_msix_notify_one(E1000ECore *core, uint32_t cause, uint32_t int_cfg) effective_eiac = core->mac[EIAC] & cause; core->mac[ICR] &= ~effective_eiac; - core->msi_causes_pending &= ~effective_eiac; if (!(core->mac[CTRL_EXT] & E1000_CTRL_EXT_IAME)) { core->mac[IMS] &= ~effective_eiac; @@ -2180,33 +2118,17 @@ e1000e_fix_icr_asserted(E1000ECore *core) trace_e1000e_irq_fix_icr_asserted(core->mac[ICR]); } -static void -e1000e_send_msi(E1000ECore *core, bool msix) +static void e1000e_raise_interrupts(E1000ECore *core, + size_t index, uint32_t causes) { - uint32_t causes = core->mac[ICR] & core->mac[IMS] & ~E1000_ICR_ASSERTED; - - core->msi_causes_pending &= causes; - causes ^= core->msi_causes_pending; - if (causes == 0) { - return; - } - core->msi_causes_pending |= causes; + bool is_msix = msix_enabled(core->owner); + uint32_t old_causes = core->mac[IMS] & core->mac[ICR]; + uint32_t raised_causes; - if (msix) { - e1000e_msix_notify(core, causes); - } else { - if (!e1000e_itr_should_postpone(core)) { - trace_e1000e_irq_msi_notify(causes); - msi_notify(core->owner, 0); - } - } -} + trace_e1000e_irq_set(index << 2, + core->mac[index], core->mac[index] | causes); -static void -e1000e_update_interrupt_state(E1000ECore *core) -{ - bool interrupts_pending; - bool is_msix = msix_enabled(core->owner); + core->mac[index] |= causes; /* Set ICR[OTHER] for MSI-X */ if (is_msix) { @@ -2228,40 +2150,58 @@ e1000e_update_interrupt_state(E1000ECore *core) */ core->mac[ICS] = core->mac[ICR]; - interrupts_pending = (core->mac[IMS] & core->mac[ICR]) ? true : false; - if (!interrupts_pending) { - core->msi_causes_pending = 0; - } - trace_e1000e_irq_pending_interrupts(core->mac[ICR] & core->mac[IMS], core->mac[ICR], core->mac[IMS]); - if (is_msix || msi_enabled(core->owner)) { - if (interrupts_pending) { - e1000e_send_msi(core, is_msix); - } - } else { - if (interrupts_pending) { - if (!e1000e_itr_should_postpone(core)) { - e1000e_raise_legacy_irq(core); - } + raised_causes = core->mac[IMS] & core->mac[ICR] & ~old_causes; + if (!raised_causes) { + return; + } + + if (is_msix) { + e1000e_msix_notify(core, raised_causes & ~E1000_ICR_ASSERTED); + } else if (!e1000e_itr_should_postpone(core)) { + if (msi_enabled(core->owner)) { + trace_e1000e_irq_msi_notify(raised_causes); + msi_notify(core->owner, 0); } else { - e1000e_lower_legacy_irq(core); + e1000e_raise_legacy_irq(core); } } } -static void -e1000e_set_interrupt_cause(E1000ECore *core, uint32_t val) +static void e1000e_lower_interrupts(E1000ECore *core, + size_t index, uint32_t causes) { - trace_e1000e_irq_set_cause_entry(val, core->mac[ICR]); + trace_e1000e_irq_clear(index << 2, + core->mac[index], core->mac[index] & ~causes); - val |= e1000e_intmgr_collect_delayed_causes(core); - core->mac[ICR] |= val; + core->mac[index] &= ~causes; + + /* + * Make sure ICR and ICS registers have the same value. + * The spec says that the ICS register is write-only. However in practice, + * on real hardware ICS is readable, and for reads it has the same value as + * ICR (except that ICS does not have the clear on read behaviour of ICR). + * + * The VxWorks PRO/1000 driver uses this behaviour. + */ + core->mac[ICS] = core->mac[ICR]; + + trace_e1000e_irq_pending_interrupts(core->mac[ICR] & core->mac[IMS], + core->mac[ICR], core->mac[IMS]); - trace_e1000e_irq_set_cause_exit(val, core->mac[ICR]); + if (!(core->mac[IMS] & core->mac[ICR]) && + !msix_enabled(core->owner) && !msi_enabled(core->owner)) { + e1000e_lower_legacy_irq(core); + } +} - e1000e_update_interrupt_state(core); +static void +e1000e_set_interrupt_cause(E1000ECore *core, uint32_t val) +{ + val |= e1000e_intmgr_collect_delayed_causes(core); + e1000e_raise_interrupts(core, ICR, val); } static inline void @@ -2527,30 +2467,27 @@ e1000e_set_ics(E1000ECore *core, int index, uint32_t val) static void e1000e_set_icr(E1000ECore *core, int index, uint32_t val) { - uint32_t icr = 0; if ((core->mac[ICR] & E1000_ICR_ASSERTED) && (core->mac[CTRL_EXT] & E1000_CTRL_EXT_IAME)) { trace_e1000e_irq_icr_process_iame(); - e1000e_clear_ims_bits(core, core->mac[IAM]); + e1000e_lower_interrupts(core, IMS, core->mac[IAM]); } - icr = core->mac[ICR] & ~val; /* * Windows driver expects that the "receive overrun" bit and other * ones to be cleared when the "Other" bit (#24) is cleared. */ - icr = (val & E1000_ICR_OTHER) ? (icr & ~E1000_ICR_OTHER_CAUSES) : icr; - trace_e1000e_irq_icr_write(val, core->mac[ICR], icr); - core->mac[ICR] = icr; - e1000e_update_interrupt_state(core); + if (val & E1000_ICR_OTHER) { + val |= E1000_ICR_OTHER_CAUSES; + } + e1000e_lower_interrupts(core, ICR, val); } static void e1000e_set_imc(E1000ECore *core, int index, uint32_t val) { trace_e1000e_irq_ims_clear_set_imc(val); - e1000e_clear_ims_bits(core, val); - e1000e_update_interrupt_state(core); + e1000e_lower_interrupts(core, IMS, val); } static void @@ -2571,9 +2508,6 @@ e1000e_set_ims(E1000ECore *core, int index, uint32_t val) uint32_t valid_val = val & ims_valid_mask; - trace_e1000e_irq_set_ims(val, core->mac[IMS], core->mac[IMS] | valid_val); - core->mac[IMS] |= valid_val; - if ((valid_val & ims_ext_mask) && (core->mac[CTRL_EXT] & E1000_CTRL_EXT_PBA_CLR) && msix_enabled(core->owner)) { @@ -2586,7 +2520,7 @@ e1000e_set_ims(E1000ECore *core, int index, uint32_t val) e1000e_intrmgr_fire_all_timers(core); } - e1000e_update_interrupt_state(core); + e1000e_raise_interrupts(core, IMS, valid_val); } static void @@ -2659,28 +2593,25 @@ static uint32_t e1000e_mac_icr_read(E1000ECore *core, int index) { uint32_t ret = core->mac[ICR]; - trace_e1000e_irq_icr_read_entry(ret); if (core->mac[IMS] == 0) { trace_e1000e_irq_icr_clear_zero_ims(); - core->mac[ICR] = 0; + e1000e_lower_interrupts(core, ICR, 0xffffffff); } if (!msix_enabled(core->owner)) { trace_e1000e_irq_icr_clear_nonmsix_icr_read(); - core->mac[ICR] = 0; + e1000e_lower_interrupts(core, ICR, 0xffffffff); } if ((core->mac[ICR] & E1000_ICR_ASSERTED) && (core->mac[CTRL_EXT] & E1000_CTRL_EXT_IAME)) { trace_e1000e_irq_icr_clear_iame(); - core->mac[ICR] = 0; + e1000e_lower_interrupts(core, ICR, 0xffffffff); trace_e1000e_irq_icr_process_iame(); - e1000e_clear_ims_bits(core, core->mac[IAM]); + e1000e_lower_interrupts(core, IMS, core->mac[IAM]); } - trace_e1000e_irq_icr_read_exit(core->mac[ICR]); - e1000e_update_interrupt_state(core); return ret; } @@ -3422,7 +3353,7 @@ e1000e_core_pci_realize(E1000ECore *core, qemu_add_vm_change_state_handler(e1000e_vm_state_change, core); for (i = 0; i < E1000E_NUM_QUEUES; i++) { - net_tx_pkt_init(&core->tx[i].tx_pkt, core->owner, E1000E_MAX_TX_FRAGS); + net_tx_pkt_init(&core->tx[i].tx_pkt, E1000E_MAX_TX_FRAGS); } net_rx_pkt_init(&core->rx_pkt); @@ -3447,7 +3378,6 @@ e1000e_core_pci_uninit(E1000ECore *core) qemu_del_vm_change_state_handler(core->vmstate); for (i = 0; i < E1000E_NUM_QUEUES; i++) { - net_tx_pkt_reset(core->tx[i].tx_pkt, core->owner); net_tx_pkt_uninit(core->tx[i].tx_pkt); } @@ -3572,7 +3502,6 @@ static void e1000e_reset(E1000ECore *core, bool sw) e1000x_reset_mac_addr(core->owner_nic, core->mac, core->permanent_mac); for (i = 0; i < ARRAY_SIZE(core->tx); i++) { - net_tx_pkt_reset(core->tx[i].tx_pkt, core->owner); memset(&core->tx[i].props, 0, sizeof(core->tx[i].props)); core->tx[i].skip_cp = false; } |