freebsd-nq/sys/dev/e1000/if_em.h

548 lines
16 KiB
C
Raw Normal View History

/*$FreeBSD$*/
#include "opt_em.h"
#include "opt_ddb.h"
#include "opt_inet.h"
#include "opt_inet6.h"
#ifdef HAVE_KERNEL_OPTION_HEADERS
#include "opt_device_polling.h"
#endif
#include <sys/param.h>
#include <sys/systm.h>
#ifdef DDB
#include <sys/types.h>
#include <ddb/ddb.h>
#endif
#if __FreeBSD_version >= 800000
#include <sys/buf_ring.h>
#endif
#include <sys/bus.h>
#include <sys/endian.h>
#include <sys/kernel.h>
#include <sys/kthread.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/module.h>
#include <sys/rman.h>
#include <sys/smp.h>
#include <sys/socket.h>
#include <sys/sockio.h>
#include <sys/sysctl.h>
#include <sys/taskqueue.h>
#include <sys/eventhandler.h>
#include <machine/bus.h>
#include <machine/resource.h>
#include <net/bpf.h>
#include <net/ethernet.h>
#include <net/if.h>
#include <net/if_var.h>
#include <net/if_arp.h>
#include <net/if_dl.h>
#include <net/if_media.h>
#include <net/iflib.h>
#include <net/if_types.h>
#include <net/if_vlan_var.h>
#include <netinet/in_systm.h>
#include <netinet/in.h>
#include <netinet/if_ether.h>
#include <netinet/ip.h>
#include <netinet/ip6.h>
#include <netinet/tcp.h>
#include <netinet/udp.h>
#include <machine/in_cksum.h>
#include <dev/led/led.h>
#include <dev/pci/pcivar.h>
#include <dev/pci/pcireg.h>
#include "e1000_api.h"
#include "e1000_82571.h"
#include "ifdi_if.h"
#ifndef _EM_H_DEFINED_
#define _EM_H_DEFINED_
/* Tunables */
/*
* EM_TXD: Maximum number of Transmit Descriptors
* Valid Range: 80-256 for 82542 and 82543-based adapters
* 80-4096 for others
* Default Value: 256
* This value is the number of transmit descriptors allocated by the driver.
* Increasing this value allows the driver to queue more transmits. Each
* descriptor is 16 bytes.
* Since TDLEN should be multiple of 128bytes, the number of transmit
* desscriptors should meet the following condition.
* (num_tx_desc * sizeof(struct e1000_tx_desc)) % 128 == 0
*/
#define EM_MIN_TXD 128
#define EM_MAX_TXD 4096
#define EM_DEFAULT_TXD 1024
#define EM_DEFAULT_MULTI_TXD 4096
/*
* EM_RXD - Maximum number of receive Descriptors
* Valid Range: 80-256 for 82542 and 82543-based adapters
* 80-4096 for others
* Default Value: 256
* This value is the number of receive descriptors allocated by the driver.
* Increasing this value allows the driver to buffer more incoming packets.
* Each descriptor is 16 bytes. A receive buffer is also allocated for each
* descriptor. The maximum MTU size is 16110.
* Since TDLEN should be multiple of 128bytes, the number of transmit
* desscriptors should meet the following condition.
* (num_tx_desc * sizeof(struct e1000_tx_desc)) % 128 == 0
*/
#define EM_MIN_RXD 128
#define EM_MAX_RXD 4096
#define EM_DEFAULT_RXD 1024
#define EM_DEFAULT_MULTI_RXD 4096
/*
* EM_TIDV - Transmit Interrupt Delay Value
* Valid Range: 0-65535 (0=off)
* Default Value: 64
* This value delays the generation of transmit interrupts in units of
* 1.024 microseconds. Transmit interrupt reduction can improve CPU
* efficiency if properly tuned for specific network traffic. If the
* system is reporting dropped transmits, this value may be set too high
* causing the driver to run out of available transmit descriptors.
*/
#define EM_TIDV 64
/*
* EM_TADV - Transmit Absolute Interrupt Delay Value
* (Not valid for 82542/82543/82544)
* Valid Range: 0-65535 (0=off)
* Default Value: 64
* This value, in units of 1.024 microseconds, limits the delay in which a
* transmit interrupt is generated. Useful only if EM_TIDV is non-zero,
* this value ensures that an interrupt is generated after the initial
* packet is sent on the wire within the set amount of time. Proper tuning,
* along with EM_TIDV, may improve traffic throughput in specific
* network conditions.
*/
#define EM_TADV 64
/*
* EM_RDTR - Receive Interrupt Delay Timer (Packet Timer)
* Valid Range: 0-65535 (0=off)
* Default Value: 0
* This value delays the generation of receive interrupts in units of 1.024
* microseconds. Receive interrupt reduction can improve CPU efficiency if
* properly tuned for specific network traffic. Increasing this value adds
* extra latency to frame reception and can end up decreasing the throughput
* of TCP traffic. If the system is reporting dropped receives, this value
* may be set too high, causing the driver to run out of available receive
* descriptors.
*
* CAUTION: When setting EM_RDTR to a value other than 0, adapters
* may hang (stop transmitting) under certain network conditions.
* If this occurs a WATCHDOG message is logged in the system
* event log. In addition, the controller is automatically reset,
* restoring the network connection. To eliminate the potential
* for the hang ensure that EM_RDTR is set to 0.
*/
Change EM_MULTIQUEUE to a real kernconf entry and enable support for up to 2 rx/tx queues for the 82574. Program the 82574 to enable 5 msix vectors, assign 1 to each rx queue, 1 to each tx queue and 1 to the link handler. Inspired by DragonFlyBSD, enable some RSS logic for handling tx queue handling/processing. Move multiqueue handler functions so that they line up better in a diff review to if_igb.c Always enqueue tx work to be done in em_mq_start, if unable to acquire the TX lock, then this will be processed in the background later by the taskqueue. Remove mbuf argument from em_start_mq_locked() as the work is always enqueued. (stolen from igb) Setup TARC, TXDCTL and RXDCTL registers for better performance and stability in multiqueue and singlequeue implementations. Handle Intel errata 3 and generic multiqueue behavior with the initialization of TARC(0) and TARC(1) Bind interrupt threads to cpus in order. (stolen from igb) Add 2 new DDB functions, one to display the queue(s) and their settings and one to reset the adapter. Primarily used for debugging. In the multiqueue configuration, bump RXD and TXD ring size to max for the adapter (4096). Setup an RDTR of 64 and an RADV of 128 in multiqueue configuration to cut down on the number of interrupts. RADV was arbitrarily set to 2x RDTR and can be adjusted as needed. Cleanup the display in top a bit to make it clearer where the taskqueue threads are running and what they should be doing. Ensure that both queues are processed by em_local_timer() by writing them both to the IMS register to generate soft interrupts. Ensure that an soft interrupt is generated when em_msix_link() is run so that any races between assertion of the link/status interrupt and a rx/tx interrupt are handled. Document existing tuneables: hw.em.eee_setting, hw.em.msix, hw.em.smart_pwr_down, hw.em.sbp Document use of hw.em.num_queues and the new kernel option EM_MULTIQUEUE Thanks to Intel for their continued support of FreeBSD. Reviewed by: erj jfv hiren gnn wblock Obtained from: Intel Corporation MFC after: 2 weeks Relnotes: Yes Sponsored by: Limelight Networks Differential Revision: https://reviews.freebsd.org/D1994
2015-06-03 18:01:09 +00:00
#ifdef EM_MULTIQUEUE
#define EM_RDTR 64
#else
#define EM_RDTR 0
Change EM_MULTIQUEUE to a real kernconf entry and enable support for up to 2 rx/tx queues for the 82574. Program the 82574 to enable 5 msix vectors, assign 1 to each rx queue, 1 to each tx queue and 1 to the link handler. Inspired by DragonFlyBSD, enable some RSS logic for handling tx queue handling/processing. Move multiqueue handler functions so that they line up better in a diff review to if_igb.c Always enqueue tx work to be done in em_mq_start, if unable to acquire the TX lock, then this will be processed in the background later by the taskqueue. Remove mbuf argument from em_start_mq_locked() as the work is always enqueued. (stolen from igb) Setup TARC, TXDCTL and RXDCTL registers for better performance and stability in multiqueue and singlequeue implementations. Handle Intel errata 3 and generic multiqueue behavior with the initialization of TARC(0) and TARC(1) Bind interrupt threads to cpus in order. (stolen from igb) Add 2 new DDB functions, one to display the queue(s) and their settings and one to reset the adapter. Primarily used for debugging. In the multiqueue configuration, bump RXD and TXD ring size to max for the adapter (4096). Setup an RDTR of 64 and an RADV of 128 in multiqueue configuration to cut down on the number of interrupts. RADV was arbitrarily set to 2x RDTR and can be adjusted as needed. Cleanup the display in top a bit to make it clearer where the taskqueue threads are running and what they should be doing. Ensure that both queues are processed by em_local_timer() by writing them both to the IMS register to generate soft interrupts. Ensure that an soft interrupt is generated when em_msix_link() is run so that any races between assertion of the link/status interrupt and a rx/tx interrupt are handled. Document existing tuneables: hw.em.eee_setting, hw.em.msix, hw.em.smart_pwr_down, hw.em.sbp Document use of hw.em.num_queues and the new kernel option EM_MULTIQUEUE Thanks to Intel for their continued support of FreeBSD. Reviewed by: erj jfv hiren gnn wblock Obtained from: Intel Corporation MFC after: 2 weeks Relnotes: Yes Sponsored by: Limelight Networks Differential Revision: https://reviews.freebsd.org/D1994
2015-06-03 18:01:09 +00:00
#endif
/*
* Receive Interrupt Absolute Delay Timer (Not valid for 82542/82543/82544)
* Valid Range: 0-65535 (0=off)
* Default Value: 64
* This value, in units of 1.024 microseconds, limits the delay in which a
* receive interrupt is generated. Useful only if EM_RDTR is non-zero,
* this value ensures that an interrupt is generated after the initial
* packet is received within the set amount of time. Proper tuning,
* along with EM_RDTR, may improve traffic throughput in specific network
* conditions.
*/
Change EM_MULTIQUEUE to a real kernconf entry and enable support for up to 2 rx/tx queues for the 82574. Program the 82574 to enable 5 msix vectors, assign 1 to each rx queue, 1 to each tx queue and 1 to the link handler. Inspired by DragonFlyBSD, enable some RSS logic for handling tx queue handling/processing. Move multiqueue handler functions so that they line up better in a diff review to if_igb.c Always enqueue tx work to be done in em_mq_start, if unable to acquire the TX lock, then this will be processed in the background later by the taskqueue. Remove mbuf argument from em_start_mq_locked() as the work is always enqueued. (stolen from igb) Setup TARC, TXDCTL and RXDCTL registers for better performance and stability in multiqueue and singlequeue implementations. Handle Intel errata 3 and generic multiqueue behavior with the initialization of TARC(0) and TARC(1) Bind interrupt threads to cpus in order. (stolen from igb) Add 2 new DDB functions, one to display the queue(s) and their settings and one to reset the adapter. Primarily used for debugging. In the multiqueue configuration, bump RXD and TXD ring size to max for the adapter (4096). Setup an RDTR of 64 and an RADV of 128 in multiqueue configuration to cut down on the number of interrupts. RADV was arbitrarily set to 2x RDTR and can be adjusted as needed. Cleanup the display in top a bit to make it clearer where the taskqueue threads are running and what they should be doing. Ensure that both queues are processed by em_local_timer() by writing them both to the IMS register to generate soft interrupts. Ensure that an soft interrupt is generated when em_msix_link() is run so that any races between assertion of the link/status interrupt and a rx/tx interrupt are handled. Document existing tuneables: hw.em.eee_setting, hw.em.msix, hw.em.smart_pwr_down, hw.em.sbp Document use of hw.em.num_queues and the new kernel option EM_MULTIQUEUE Thanks to Intel for their continued support of FreeBSD. Reviewed by: erj jfv hiren gnn wblock Obtained from: Intel Corporation MFC after: 2 weeks Relnotes: Yes Sponsored by: Limelight Networks Differential Revision: https://reviews.freebsd.org/D1994
2015-06-03 18:01:09 +00:00
#ifdef EM_MULTIQUEUE
#define EM_RADV 128
#else
#define EM_RADV 64
Change EM_MULTIQUEUE to a real kernconf entry and enable support for up to 2 rx/tx queues for the 82574. Program the 82574 to enable 5 msix vectors, assign 1 to each rx queue, 1 to each tx queue and 1 to the link handler. Inspired by DragonFlyBSD, enable some RSS logic for handling tx queue handling/processing. Move multiqueue handler functions so that they line up better in a diff review to if_igb.c Always enqueue tx work to be done in em_mq_start, if unable to acquire the TX lock, then this will be processed in the background later by the taskqueue. Remove mbuf argument from em_start_mq_locked() as the work is always enqueued. (stolen from igb) Setup TARC, TXDCTL and RXDCTL registers for better performance and stability in multiqueue and singlequeue implementations. Handle Intel errata 3 and generic multiqueue behavior with the initialization of TARC(0) and TARC(1) Bind interrupt threads to cpus in order. (stolen from igb) Add 2 new DDB functions, one to display the queue(s) and their settings and one to reset the adapter. Primarily used for debugging. In the multiqueue configuration, bump RXD and TXD ring size to max for the adapter (4096). Setup an RDTR of 64 and an RADV of 128 in multiqueue configuration to cut down on the number of interrupts. RADV was arbitrarily set to 2x RDTR and can be adjusted as needed. Cleanup the display in top a bit to make it clearer where the taskqueue threads are running and what they should be doing. Ensure that both queues are processed by em_local_timer() by writing them both to the IMS register to generate soft interrupts. Ensure that an soft interrupt is generated when em_msix_link() is run so that any races between assertion of the link/status interrupt and a rx/tx interrupt are handled. Document existing tuneables: hw.em.eee_setting, hw.em.msix, hw.em.smart_pwr_down, hw.em.sbp Document use of hw.em.num_queues and the new kernel option EM_MULTIQUEUE Thanks to Intel for their continued support of FreeBSD. Reviewed by: erj jfv hiren gnn wblock Obtained from: Intel Corporation MFC after: 2 weeks Relnotes: Yes Sponsored by: Limelight Networks Differential Revision: https://reviews.freebsd.org/D1994
2015-06-03 18:01:09 +00:00
#endif
/*
* This parameter controls whether or not autonegotation is enabled.
* 0 - Disable autonegotiation
* 1 - Enable autonegotiation
*/
#define DO_AUTO_NEG 1
/*
* This parameter control whether or not the driver will wait for
* autonegotiation to complete.
* 1 - Wait for autonegotiation to complete
* 0 - Don't wait for autonegotiation to complete
*/
#define WAIT_FOR_AUTO_NEG_DEFAULT 0
/* Tunables -- End */
#define AUTONEG_ADV_DEFAULT (ADVERTISE_10_HALF | ADVERTISE_10_FULL | \
ADVERTISE_100_HALF | ADVERTISE_100_FULL | \
ADVERTISE_1000_FULL)
#define AUTO_ALL_MODES 0
/* PHY master/slave setting */
#define EM_MASTER_SLAVE e1000_ms_hw_default
/*
* Micellaneous constants
*/
#define EM_VENDOR_ID 0x8086
#define EM_FLASH 0x0014
#define EM_JUMBO_PBA 0x00000028
#define EM_DEFAULT_PBA 0x00000030
#define EM_SMARTSPEED_DOWNSHIFT 3
#define EM_SMARTSPEED_MAX 15
#define EM_MAX_LOOP 10
#define MAX_NUM_MULTICAST_ADDRESSES 128
#define PCI_ANY_ID (~0U)
#define ETHER_ALIGN 2
#define EM_FC_PAUSE_TIME 0x0680
#define EM_EEPROM_APME 0x400;
#define EM_82544_APME 0x0004;
/*
* Driver state logic for the detection of a hung state
* in hardware. Set TX_HUNG whenever a TX packet is used
* (data is sent) and clear it when txeof() is invoked if
* any descriptors from the ring are cleaned/reclaimed.
* Increment internal counter if no descriptors are cleaned
* and compare to TX_MAXTRIES. When counter > TX_MAXTRIES,
* reset adapter.
*/
#define EM_TX_IDLE 0x00000000
#define EM_TX_BUSY 0x00000001
#define EM_TX_HUNG 0x80000000
#define EM_TX_MAXTRIES 10
#define PCICFG_DESC_RING_STATUS 0xe4
#define FLUSH_DESC_REQUIRED 0x100
#define IGB_RX_PTHRESH ((hw->mac.type == e1000_i354) ? 12 : \
((hw->mac.type <= e1000_82576) ? 16 : 8))
#define IGB_RX_HTHRESH 8
#define IGB_RX_WTHRESH ((hw->mac.type == e1000_82576 && \
(adapter->intr_type == IFLIB_INTR_MSIX)) ? 1 : 4)
#define IGB_TX_PTHRESH ((hw->mac.type == e1000_i354) ? 20 : 8)
#define IGB_TX_HTHRESH 1
#define IGB_TX_WTHRESH ((hw->mac.type != e1000_82575 && \
(adapter->intr_type == IFLIB_INTR_MSIX) ? 1 : 16)
/*
* TDBA/RDBA should be aligned on 16 byte boundary. But TDLEN/RDLEN should be
* multiple of 128 bytes. So we align TDBA/RDBA on 128 byte boundary. This will
* also optimize cache line size effect. H/W supports up to cache line size 128.
*/
#define EM_DBA_ALIGN 128
Change EM_MULTIQUEUE to a real kernconf entry and enable support for up to 2 rx/tx queues for the 82574. Program the 82574 to enable 5 msix vectors, assign 1 to each rx queue, 1 to each tx queue and 1 to the link handler. Inspired by DragonFlyBSD, enable some RSS logic for handling tx queue handling/processing. Move multiqueue handler functions so that they line up better in a diff review to if_igb.c Always enqueue tx work to be done in em_mq_start, if unable to acquire the TX lock, then this will be processed in the background later by the taskqueue. Remove mbuf argument from em_start_mq_locked() as the work is always enqueued. (stolen from igb) Setup TARC, TXDCTL and RXDCTL registers for better performance and stability in multiqueue and singlequeue implementations. Handle Intel errata 3 and generic multiqueue behavior with the initialization of TARC(0) and TARC(1) Bind interrupt threads to cpus in order. (stolen from igb) Add 2 new DDB functions, one to display the queue(s) and their settings and one to reset the adapter. Primarily used for debugging. In the multiqueue configuration, bump RXD and TXD ring size to max for the adapter (4096). Setup an RDTR of 64 and an RADV of 128 in multiqueue configuration to cut down on the number of interrupts. RADV was arbitrarily set to 2x RDTR and can be adjusted as needed. Cleanup the display in top a bit to make it clearer where the taskqueue threads are running and what they should be doing. Ensure that both queues are processed by em_local_timer() by writing them both to the IMS register to generate soft interrupts. Ensure that an soft interrupt is generated when em_msix_link() is run so that any races between assertion of the link/status interrupt and a rx/tx interrupt are handled. Document existing tuneables: hw.em.eee_setting, hw.em.msix, hw.em.smart_pwr_down, hw.em.sbp Document use of hw.em.num_queues and the new kernel option EM_MULTIQUEUE Thanks to Intel for their continued support of FreeBSD. Reviewed by: erj jfv hiren gnn wblock Obtained from: Intel Corporation MFC after: 2 weeks Relnotes: Yes Sponsored by: Limelight Networks Differential Revision: https://reviews.freebsd.org/D1994
2015-06-03 18:01:09 +00:00
/*
* See Intel 82574 Driver Programming Interface Manual, Section 10.2.6.9
*/
#define TARC_COMPENSATION_MODE (1 << 7) /* Compensation Mode */
#define TARC_SPEED_MODE_BIT (1 << 21) /* On PCI-E MACs only */
#define TARC_MQ_FIX (1 << 23) | \
(1 << 24) | \
(1 << 25) /* Handle errata in MQ mode */
#define TARC_ERRATA_BIT (1 << 26) /* Note from errata on 82574 */
/* PCI Config defines */
#define EM_BAR_TYPE(v) ((v) & EM_BAR_TYPE_MASK)
#define EM_BAR_TYPE_MASK 0x00000001
#define EM_BAR_TYPE_MMEM 0x00000000
#define EM_BAR_TYPE_IO 0x00000001
#define EM_BAR_TYPE_FLASH 0x0014
#define EM_BAR_MEM_TYPE(v) ((v) & EM_BAR_MEM_TYPE_MASK)
#define EM_BAR_MEM_TYPE_MASK 0x00000006
#define EM_BAR_MEM_TYPE_32BIT 0x00000000
#define EM_BAR_MEM_TYPE_64BIT 0x00000004
#define EM_MSIX_BAR 3 /* On 82575 */
Part 2 of 2 New deltas for the 1G drivers. There have still been intermittent problems with apparent TX hangs for some customers. These have been problematic to reproduce but I believe these changes will address them. Testing on a number of fronts have been positive. EM: there is an important 'chicken bit' fix for 82574 in the shared code this is supported in the core here. - The TX path has been tightened up to improve performance. In particular UDP with jumbo frames was having problems, and the changes here have improved that. - OACTIVE has been used more carefully on the theory that some hangs may be due to a problem in this interaction - Problems with the RX init code, the "lazy" allocation and ring initialization has been found to cause problems in some newer client systems, and as it really is not that big a win (its not in a hot path) it seems best to remove it. - HWTSO was broken when VLAN HWTAGGING or HWFILTER is used, I found this was due to an error in setting up the descriptors in em_xmit. IGB: - TX is also improved here. With multiqueue I realized its very important to handle OACTIVE only under the CORE lock so there are no races between the queues. - Flow Control handling was broken in a couple ways, I have changed and I hope improved that in this delta. - UDP also had a problem in the TX path here, it was change to improve that. - On some hardware, with the driver static, a weird stray interrupt seems to sometimes fire and cause a panic in the RX mbuf refresh code. This is addressed by setting interrupts late in the init path, and also to set all interrupts bits off at the start of that.
2011-12-10 07:08:52 +00:00
/* More backward compatibility */
#if __FreeBSD_version < 900000
#define SYSCTL_ADD_UQUAD SYSCTL_ADD_QUAD
#endif
/* Defines for printing debug information */
#define DEBUG_INIT 0
#define DEBUG_IOCTL 0
#define DEBUG_HW 0
#define INIT_DEBUGOUT(S) if (DEBUG_INIT) printf(S "\n")
#define INIT_DEBUGOUT1(S, A) if (DEBUG_INIT) printf(S "\n", A)
#define INIT_DEBUGOUT2(S, A, B) if (DEBUG_INIT) printf(S "\n", A, B)
#define IOCTL_DEBUGOUT(S) if (DEBUG_IOCTL) printf(S "\n")
#define IOCTL_DEBUGOUT1(S, A) if (DEBUG_IOCTL) printf(S "\n", A)
#define IOCTL_DEBUGOUT2(S, A, B) if (DEBUG_IOCTL) printf(S "\n", A, B)
#define HW_DEBUGOUT(S) if (DEBUG_HW) printf(S "\n")
#define HW_DEBUGOUT1(S, A) if (DEBUG_HW) printf(S "\n", A)
#define HW_DEBUGOUT2(S, A, B) if (DEBUG_HW) printf(S "\n", A, B)
Fix and clean up usage of DMA and TSO segments: - At Intel it is believed that most of their products support "only" 40 DMA segments so lower {EM,IGB}_MAX_SCATTER accordingly. Actually, 40 is more than plenty to handle full size TSO packets so it doesn't make sense to further distinguish between MAC variants that really can do 64 DMA segments. Moreover, capping at 40 DMA segments limits the stack usage of {em,igb}_xmit() that - given the rare use of more than these - previously hardly was justifiable, while still being sufficient to avoid the problems seen with em(4) and EM_MAX_SCATTER set to 32. - In igb(4), pass the actually supported TSO parameters up the stack. Previously, the defaults set in if_attach_internal() were applied, i. e. a maximum of 35 TSO segments, which made supporting more than these in the driver pointless. However, this might explain why no problems were seen with IGB_MAX_SCATTER at 64. - In em(4), take the 5 m_pullup(9) invocations performed by em_xmit() in the TSO case into account when reporting TSO parameters upwards. In the worst case, each of these calls will add another mbuf and, thus, the requirement for an additional DMA segment. So for best performance, it doesn't make sense to advertize a maximum of TSO segments that typically will require defragmentation in em_xmit(). Again, this leaves enough room to handle full size TSO packets. - Drop TSO macros from if_lem.h given that corresponding MACS don't support TSO in the first place. Reviewed by: erj, sbruno, jeffrey.e.pieper_intel.com Approved by: erj MFC after: 3 days Differential Revision: https://reviews.freebsd.org/D5238
2016-02-23 01:19:26 +00:00
#define EM_MAX_SCATTER 40
#define EM_VFTA_SIZE 128
#define EM_TSO_SIZE (65535 + sizeof(struct ether_vlan_header))
#define EM_TSO_SEG_SIZE 4096 /* Max dma segment size */
#define EM_MSIX_MASK 0x01F00000 /* For 82574 use */
#define EM_MSIX_LINK 0x01000000 /* For 82574 use */
#define ETH_ZLEN 60
#define ETH_ADDR_LEN 6
#define CSUM_OFFLOAD 7 /* Offload bits in mbuf flag */
#define IGB_PKTTYPE_MASK 0x0000FFF0
#define IGB_DMCTLX_DCFLUSH_DIS 0x80000000 /* Disable DMA Coalesce Flush */
/*
* 82574 has a nonstandard address for EIAC
* and since its only used in MSIX, and in
* the em driver only 82574 uses MSIX we can
* solve it just using this define.
*/
#define EM_EIAC 0x000DC
Change EM_MULTIQUEUE to a real kernconf entry and enable support for up to 2 rx/tx queues for the 82574. Program the 82574 to enable 5 msix vectors, assign 1 to each rx queue, 1 to each tx queue and 1 to the link handler. Inspired by DragonFlyBSD, enable some RSS logic for handling tx queue handling/processing. Move multiqueue handler functions so that they line up better in a diff review to if_igb.c Always enqueue tx work to be done in em_mq_start, if unable to acquire the TX lock, then this will be processed in the background later by the taskqueue. Remove mbuf argument from em_start_mq_locked() as the work is always enqueued. (stolen from igb) Setup TARC, TXDCTL and RXDCTL registers for better performance and stability in multiqueue and singlequeue implementations. Handle Intel errata 3 and generic multiqueue behavior with the initialization of TARC(0) and TARC(1) Bind interrupt threads to cpus in order. (stolen from igb) Add 2 new DDB functions, one to display the queue(s) and their settings and one to reset the adapter. Primarily used for debugging. In the multiqueue configuration, bump RXD and TXD ring size to max for the adapter (4096). Setup an RDTR of 64 and an RADV of 128 in multiqueue configuration to cut down on the number of interrupts. RADV was arbitrarily set to 2x RDTR and can be adjusted as needed. Cleanup the display in top a bit to make it clearer where the taskqueue threads are running and what they should be doing. Ensure that both queues are processed by em_local_timer() by writing them both to the IMS register to generate soft interrupts. Ensure that an soft interrupt is generated when em_msix_link() is run so that any races between assertion of the link/status interrupt and a rx/tx interrupt are handled. Document existing tuneables: hw.em.eee_setting, hw.em.msix, hw.em.smart_pwr_down, hw.em.sbp Document use of hw.em.num_queues and the new kernel option EM_MULTIQUEUE Thanks to Intel for their continued support of FreeBSD. Reviewed by: erj jfv hiren gnn wblock Obtained from: Intel Corporation MFC after: 2 weeks Relnotes: Yes Sponsored by: Limelight Networks Differential Revision: https://reviews.freebsd.org/D1994
2015-06-03 18:01:09 +00:00
/*
* 82574 only reports 3 MSI-X vectors by default;
* defines assisting with making it report 5 are
* located here.
*/
#define EM_NVM_PCIE_CTRL 0x1B
#define EM_NVM_MSIX_N_MASK (0x7 << EM_NVM_MSIX_N_SHIFT)
#define EM_NVM_MSIX_N_SHIFT 7
struct adapter;
struct em_int_delay_info {
struct adapter *adapter; /* Back-pointer to the adapter struct */
int offset; /* Register offset to read/write */
int value; /* Current value in usecs */
};
/*
* The transmit ring, one per tx queue
*/
struct tx_ring {
struct adapter *adapter;
struct em_tx_queue *que;
u32 me;
int busy;
struct e1000_tx_desc *tx_base;
uint64_t tx_paddr;
struct em_txbuffer *tx_buffers;
u32 tx_tso; /* last tx was tso */
/* Interrupt resources */
void *tag;
struct resource *res;
2010-03-30 17:05:24 +00:00
unsigned long tx_irq;
unsigned long no_desc_avail;
/* Saved csum offloading context information */
int csum_flags;
int csum_lhlen;
int csum_iphlen;
int csum_thlen;
int csum_mss;
int csum_pktlen;
uint32_t csum_txd_upper;
uint32_t csum_txd_lower; /* last field */
};
/*
* The Receive ring, one per rx queue
*/
struct rx_ring {
struct adapter *adapter;
struct em_rx_queue *que;
u32 me;
u32 payload;
union e1000_rx_desc_extended *rx_base;
uint64_t rx_paddr;
/* Interrupt resources */
void *tag;
struct resource *res;
bool discard;
/* Soft stats */
2010-03-30 17:05:24 +00:00
unsigned long rx_irq;
unsigned long rx_discarded;
2010-03-30 17:05:24 +00:00
unsigned long rx_packets;
unsigned long rx_bytes;
};
struct em_tx_queue {
struct adapter *adapter;
u32 msix;
u32 eims; /* This queue's EIMS bit */
u32 me;
struct tx_ring txr;
};
struct em_rx_queue {
struct adapter *adapter;
u32 me;
u32 msix;
u32 eims;
struct rx_ring rxr;
u64 irqs;
struct if_irq que_irq;
};
/* Our adapter structure */
struct adapter {
struct ifnet *ifp;
struct e1000_hw hw;
if_softc_ctx_t shared;
if_ctx_t ctx;
#define tx_num_queues shared->isc_ntxqsets
#define rx_num_queues shared->isc_nrxqsets
#define intr_type shared->isc_intr
/* FreeBSD operating-system-specific structures. */
struct e1000_osdep osdep;
struct device *dev;
struct cdev *led_dev;
struct em_tx_queue *tx_queues;
struct em_rx_queue *rx_queues;
struct if_irq irq;
struct resource *memory;
struct resource *flash;
struct resource *ioport;
int io_rid;
struct resource *res;
void *tag;
u32 linkvec;
u32 ivars;
struct ifmedia *media;
int msix;
int if_flags;
int min_frame_size;
int em_insert_vlan_header;
u32 ims;
bool in_detach;
/* Task for FAST handling */
struct grouptask link_task;
u16 num_vlans;
u32 txd_cmd;
u32 tx_process_limit;
u32 rx_process_limit;
u32 rx_mbuf_sz;
/* Management and WOL features */
u32 wol;
bool has_manage;
bool has_amt;
/* Multicast array memory */
u8 *mta;
/*
** Shadow VFTA table, this is needed because
** the real vlan filter table gets cleared during
** a soft reset and the driver needs to be able
** to repopulate it.
*/
u32 shadow_vfta[EM_VFTA_SIZE];
/* Info about the interface */
Part 2 of 2 New deltas for the 1G drivers. There have still been intermittent problems with apparent TX hangs for some customers. These have been problematic to reproduce but I believe these changes will address them. Testing on a number of fronts have been positive. EM: there is an important 'chicken bit' fix for 82574 in the shared code this is supported in the core here. - The TX path has been tightened up to improve performance. In particular UDP with jumbo frames was having problems, and the changes here have improved that. - OACTIVE has been used more carefully on the theory that some hangs may be due to a problem in this interaction - Problems with the RX init code, the "lazy" allocation and ring initialization has been found to cause problems in some newer client systems, and as it really is not that big a win (its not in a hot path) it seems best to remove it. - HWTSO was broken when VLAN HWTAGGING or HWFILTER is used, I found this was due to an error in setting up the descriptors in em_xmit. IGB: - TX is also improved here. With multiqueue I realized its very important to handle OACTIVE only under the CORE lock so there are no races between the queues. - Flow Control handling was broken in a couple ways, I have changed and I hope improved that in this delta. - UDP also had a problem in the TX path here, it was change to improve that. - On some hardware, with the driver static, a weird stray interrupt seems to sometimes fire and cause a panic in the RX mbuf refresh code. This is addressed by setting interrupts late in the init path, and also to set all interrupts bits off at the start of that.
2011-12-10 07:08:52 +00:00
u16 link_active;
u16 fc;
u16 link_speed;
u16 link_duplex;
u32 smartspeed;
u32 dmac;
int link_mask;
u64 que_mask;
struct em_int_delay_info tx_int_delay;
struct em_int_delay_info tx_abs_int_delay;
struct em_int_delay_info rx_int_delay;
struct em_int_delay_info rx_abs_int_delay;
struct em_int_delay_info tx_itr;
/* Misc stats maintained by the driver */
unsigned long dropped_pkts;
unsigned long link_irq;
unsigned long mbuf_defrag_failed;
unsigned long no_tx_dma_setup;
unsigned long no_tx_map_avail;
unsigned long rx_overruns;
unsigned long watchdog_events;
struct e1000_hw_stats stats;
};
/********************************************************************************
* vendor_info_array
*
* This array contains the list of Subvendor/Subdevice IDs on which the driver
* should load.
*
********************************************************************************/
typedef struct _em_vendor_info_t {
unsigned int vendor_id;
unsigned int device_id;
unsigned int subvendor_id;
unsigned int subdevice_id;
unsigned int index;
} em_vendor_info_t;
struct em_txbuffer {
int eop;
};
#define EM_CORE_LOCK_INIT(_sc, _name) \
mtx_init(&(_sc)->core_mtx, _name, "EM Core Lock", MTX_DEF)
#define EM_TX_LOCK_INIT(_sc, _name) \
mtx_init(&(_sc)->tx_mtx, _name, "EM TX Lock", MTX_DEF)
#define EM_RX_LOCK_INIT(_sc, _name) \
mtx_init(&(_sc)->rx_mtx, _name, "EM RX Lock", MTX_DEF)
#define EM_CORE_LOCK_DESTROY(_sc) mtx_destroy(&(_sc)->core_mtx)
#define EM_TX_LOCK_DESTROY(_sc) mtx_destroy(&(_sc)->tx_mtx)
#define EM_RX_LOCK_DESTROY(_sc) mtx_destroy(&(_sc)->rx_mtx)
#define EM_CORE_LOCK(_sc) mtx_lock(&(_sc)->core_mtx)
#define EM_TX_LOCK(_sc) mtx_lock(&(_sc)->tx_mtx)
#define EM_TX_TRYLOCK(_sc) mtx_trylock(&(_sc)->tx_mtx)
#define EM_RX_LOCK(_sc) mtx_lock(&(_sc)->rx_mtx)
#define EM_CORE_UNLOCK(_sc) mtx_unlock(&(_sc)->core_mtx)
#define EM_TX_UNLOCK(_sc) mtx_unlock(&(_sc)->tx_mtx)
#define EM_RX_UNLOCK(_sc) mtx_unlock(&(_sc)->rx_mtx)
#define EM_CORE_LOCK_ASSERT(_sc) mtx_assert(&(_sc)->core_mtx, MA_OWNED)
#define EM_TX_LOCK_ASSERT(_sc) mtx_assert(&(_sc)->tx_mtx, MA_OWNED)
#define EM_RX_LOCK_ASSERT(_sc) mtx_assert(&(_sc)->rx_mtx, MA_OWNED)
#define EM_RSSRK_SIZE 4
#define EM_RSSRK_VAL(key, i) (key[(i) * EM_RSSRK_SIZE] | \
key[(i) * EM_RSSRK_SIZE + 1] << 8 | \
key[(i) * EM_RSSRK_SIZE + 2] << 16 | \
key[(i) * EM_RSSRK_SIZE + 3] << 24)
#endif /* _EM_H_DEFINED_ */