* [net-next PATCH 02/29] fm10k: Add register defines and basic structures
From: Alexander Duyck @ 2014-09-18 22:35 UTC (permalink / raw)
To: davem
Cc: nhorman, netdev, john.fastabend, matthew.vick, jeffrey.t.kirsher,
sassmann
In-Reply-To: <20140918223242.10373.27403.stgit@ahduyck-bv4.jf.intel.com>
This patch adds the basic defines and structures needed by the PF for
operation.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/fm10k/fm10k_main.c | 3
drivers/net/ethernet/intel/fm10k/fm10k_type.h | 659 +++++++++++++++++++++++++
2 files changed, 661 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
index a7822b8..6ca0614 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_main.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
@@ -40,7 +40,8 @@ MODULE_DESCRIPTION("Intel(R) Ethernet Switch Host Interface Driver");
MODULE_LICENSE("GPL");
MODULE_VERSION(DRV_VERSION);
-/** fm10k_init_module - Driver Registration Routine
+/**
+ * fm10k_init_module - Driver Registration Routine
*
* fm10k_init_module is the first routine called when the driver is
* loaded. All it does is register with the PCI subsystem.
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_type.h b/drivers/net/ethernet/intel/fm10k/fm10k_type.h
index 2c06f02..1606805 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_type.h
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_type.h
@@ -21,7 +21,666 @@
#ifndef _FM10K_TYPE_H_
#define _FM10K_TYPE_H_
+/* forward declaration */
+struct fm10k_hw;
+
+#include <linux/types.h>
+#include <asm/byteorder.h>
+#include <linux/etherdevice.h>
+
#define FM10K_DEV_ID_PF 0x15A4
#define FM10K_DEV_ID_VF 0x15A5
+#define FM10K_MAX_QUEUES 256
+#define FM10K_MAX_QUEUES_PF 128
+#define FM10K_MAX_QUEUES_POOL 16
+
+#define FM10K_48_BIT_MASK 0x0000FFFFFFFFFFFFull
+#define FM10K_STAT_VALID 0x80000000
+
+/* PCI Bus Info */
+#define FM10K_PCIE_LINK_CAP 0x7C
+#define FM10K_PCIE_LINK_STATUS 0x82
+#define FM10K_PCIE_LINK_WIDTH 0x3F0
+#define FM10K_PCIE_LINK_WIDTH_1 0x10
+#define FM10K_PCIE_LINK_WIDTH_2 0x20
+#define FM10K_PCIE_LINK_WIDTH_4 0x40
+#define FM10K_PCIE_LINK_WIDTH_8 0x80
+#define FM10K_PCIE_LINK_SPEED 0xF
+#define FM10K_PCIE_LINK_SPEED_2500 0x1
+#define FM10K_PCIE_LINK_SPEED_5000 0x2
+#define FM10K_PCIE_LINK_SPEED_8000 0x3
+
+/* PCIe payload size */
+#define FM10K_PCIE_DEV_CAP 0x74
+#define FM10K_PCIE_DEV_CAP_PAYLOAD 0x07
+#define FM10K_PCIE_DEV_CAP_PAYLOAD_128 0x00
+#define FM10K_PCIE_DEV_CAP_PAYLOAD_256 0x01
+#define FM10K_PCIE_DEV_CAP_PAYLOAD_512 0x02
+#define FM10K_PCIE_DEV_CTRL 0x78
+#define FM10K_PCIE_DEV_CTRL_PAYLOAD 0xE0
+#define FM10K_PCIE_DEV_CTRL_PAYLOAD_128 0x00
+#define FM10K_PCIE_DEV_CTRL_PAYLOAD_256 0x20
+#define FM10K_PCIE_DEV_CTRL_PAYLOAD_512 0x40
+
+/* PCIe MSI-X Capability info */
+#define FM10K_PCI_MSIX_MSG_CTRL 0xB2
+#define FM10K_PCI_MSIX_MSG_CTRL_TBL_SZ_MASK 0x7FF
+#define FM10K_MAX_MSIX_VECTORS 256
+#define FM10K_MAX_VECTORS_PF 256
+#define FM10K_MAX_VECTORS_POOL 32
+
+/* PCIe SR-IOV Info */
+#define FM10K_PCIE_SRIOV_CTRL 0x190
+#define FM10K_PCIE_SRIOV_CTRL_VFARI 0x10
+
+#define FM10K_ERR_PARAM -2
+#define FM10K_ERR_REQUESTS_PENDING -4
+#define FM10K_ERR_RESET_REQUESTED -5
+#define FM10K_ERR_DMA_PENDING -6
+#define FM10K_ERR_RESET_FAILED -7
+#define FM10K_ERR_INVALID_MAC_ADDR -8
+#define FM10K_ERR_INVALID_VALUE -9
+#define FM10K_NOT_IMPLEMENTED 0x7FFFFFFF
+
+/* Start of PF registers */
+#define FM10K_CTRL 0x0000
+#define FM10K_CTRL_BAR4_ALLOWED 0x00000004
+
+#define FM10K_CTRL_EXT 0x0001
+#define FM10K_GCR 0x0003
+#define FM10K_GCR_EXT 0x0005
+
+/* Interrupt control registers */
+#define FM10K_EICR 0x0006
+#define FM10K_EICR_FAULT_MASK 0x0000003F
+#define FM10K_EICR_MAILBOX 0x00000040
+#define FM10K_EICR_SWITCHREADY 0x00000080
+#define FM10K_EICR_SWITCHNOTREADY 0x00000100
+#define FM10K_EICR_SWITCHINTERRUPT 0x00000200
+#define FM10K_EICR_VFLR 0x00000800
+#define FM10K_EICR_MAXHOLDTIME 0x00001000
+#define FM10K_EIMR 0x0007
+#define FM10K_EIMR_PCA_FAULT 0x00000001
+#define FM10K_EIMR_THI_FAULT 0x00000010
+#define FM10K_EIMR_FUM_FAULT 0x00000400
+#define FM10K_EIMR_MAILBOX 0x00001000
+#define FM10K_EIMR_SWITCHREADY 0x00004000
+#define FM10K_EIMR_SWITCHNOTREADY 0x00010000
+#define FM10K_EIMR_SWITCHINTERRUPT 0x00040000
+#define FM10K_EIMR_SRAMERROR 0x00100000
+#define FM10K_EIMR_VFLR 0x00400000
+#define FM10K_EIMR_MAXHOLDTIME 0x01000000
+#define FM10K_EIMR_ALL 0x55555555
+#define FM10K_EIMR_DISABLE(NAME) ((FM10K_EIMR_ ## NAME) << 0)
+#define FM10K_EIMR_ENABLE(NAME) ((FM10K_EIMR_ ## NAME) << 1)
+#define FM10K_FAULT_ADDR_LO 0x0
+#define FM10K_FAULT_ADDR_HI 0x1
+#define FM10K_FAULT_SPECINFO 0x2
+#define FM10K_FAULT_FUNC 0x3
+#define FM10K_FAULT_SIZE 0x4
+#define FM10K_FAULT_FUNC_VALID 0x00008000
+#define FM10K_FAULT_FUNC_PF 0x00004000
+#define FM10K_FAULT_FUNC_VF_MASK 0x00003F00
+#define FM10K_FAULT_FUNC_VF_SHIFT 8
+#define FM10K_FAULT_FUNC_TYPE_MASK 0x000000FF
+
+#define FM10K_PCA_FAULT 0x0008
+#define FM10K_THI_FAULT 0x0010
+#define FM10K_FUM_FAULT 0x001C
+
+/* Rx queue timeout indicator */
+#define FM10K_MAXHOLDQ(_n) ((_n) + 0x0020)
+
+/* Switch Manager info */
+#define FM10K_SM_AREA(_n) ((_n) + 0x0028)
+
+/* GLORT mapping registers */
+#define FM10K_DGLORTMAP(_n) ((_n) + 0x0030)
+#define FM10K_DGLORT_COUNT 8
+#define FM10K_DGLORTMAP_MASK_SHIFT 16
+#define FM10K_DGLORTMAP_ANY 0x00000000
+#define FM10K_DGLORTMAP_NONE 0x0000FFFF
+#define FM10K_DGLORTMAP_ZERO 0xFFFF0000
+#define FM10K_DGLORTDEC(_n) ((_n) + 0x0038)
+#define FM10K_DGLORTDEC_VSILENGTH_SHIFT 4
+#define FM10K_DGLORTDEC_VSIBASE_SHIFT 7
+#define FM10K_DGLORTDEC_PCLENGTH_SHIFT 14
+#define FM10K_DGLORTDEC_QBASE_SHIFT 16
+#define FM10K_DGLORTDEC_RSSLENGTH_SHIFT 24
+#define FM10K_DGLORTDEC_INNERRSS_ENABLE 0x08000000
+#define FM10K_TUNNEL_CFG 0x0040
+#define FM10K_TUNNEL_CFG_NVGRE_SHIFT 16
+#define FM10K_SWPRI_MAP(_n) ((_n) + 0x0050)
+#define FM10K_SWPRI_MAX 16
+#define FM10K_RSSRK(_n, _m) (((_n) * 0x10) + (_m) + 0x0800)
+#define FM10K_RSSRK_SIZE 10
+#define FM10K_RSSRK_ENTRIES_PER_REG 4
+#define FM10K_RETA(_n, _m) (((_n) * 0x20) + (_m) + 0x1000)
+#define FM10K_RETA_SIZE 32
+#define FM10K_RETA_ENTRIES_PER_REG 4
+#define FM10K_MAX_RSS_INDICES 128
+
+/* Rate limiting registers */
+#define FM10K_TC_CREDIT(_n) ((_n) + 0x2000)
+#define FM10K_TC_CREDIT_CREDIT_MASK 0x001FFFFF
+#define FM10K_TC_MAXCREDIT(_n) ((_n) + 0x2040)
+#define FM10K_TC_MAXCREDIT_64K 0x00010000
+#define FM10K_TC_RATE(_n) ((_n) + 0x2080)
+#define FM10K_TC_RATE_QUANTA_MASK 0x0000FFFF
+#define FM10K_TC_RATE_INTERVAL_4US_GEN1 0x00020000
+#define FM10K_TC_RATE_INTERVAL_4US_GEN2 0x00040000
+#define FM10K_TC_RATE_INTERVAL_4US_GEN3 0x00080000
+
+/* DMA control registers */
+#define FM10K_DMA_CTRL 0x20C3
+#define FM10K_DMA_CTRL_TX_ENABLE 0x00000001
+#define FM10K_DMA_CTRL_TX_ACTIVE 0x00000008
+#define FM10K_DMA_CTRL_RX_ENABLE 0x00000010
+#define FM10K_DMA_CTRL_RX_ACTIVE 0x00000080
+#define FM10K_DMA_CTRL_RX_DESC_SIZE 0x00000100
+#define FM10K_DMA_CTRL_MINMSS_64 0x00008000
+#define FM10K_DMA_CTRL_MAX_HOLD_1US_GEN3 0x04800000
+#define FM10K_DMA_CTRL_MAX_HOLD_1US_GEN2 0x04000000
+#define FM10K_DMA_CTRL_MAX_HOLD_1US_GEN1 0x03800000
+#define FM10K_DMA_CTRL_DATAPATH_RESET 0x20000000
+#define FM10K_DMA_CTRL_32_DESC 0x00000000
+
+#define FM10K_DMA_CTRL2 0x20C4
+#define FM10K_DMA_CTRL2_SWITCH_READY 0x00002000
+
+/* TSO flags configuration
+ * First packet contains all flags except for fin and psh
+ * Middle packet contains only urg and ack
+ * Last packet contains urg, ack, fin, and psh
+ */
+#define FM10K_TSO_FLAGS_LOW 0x00300FF6
+#define FM10K_TSO_FLAGS_HI 0x00000039
+#define FM10K_DTXTCPFLGL 0x20C5
+#define FM10K_DTXTCPFLGH 0x20C6
+
+#define FM10K_TPH_CTRL 0x20C7
+#define FM10K_MRQC(_n) ((_n) + 0x2100)
+#define FM10K_MRQC_TCP_IPV4 0x00000001
+#define FM10K_MRQC_IPV4 0x00000002
+#define FM10K_MRQC_IPV6 0x00000010
+#define FM10K_MRQC_TCP_IPV6 0x00000020
+#define FM10K_MRQC_UDP_IPV4 0x00000040
+#define FM10K_MRQC_UDP_IPV6 0x00000080
+
+#define FM10K_TQMAP(_n) ((_n) + 0x2800)
+#define FM10K_TQMAP_TABLE_SIZE 2048
+#define FM10K_RQMAP(_n) ((_n) + 0x3000)
+
+/* Hardware Statistics */
+#define FM10K_STATS_TIMEOUT 0x3800
+#define FM10K_STATS_UR 0x3801
+#define FM10K_STATS_CA 0x3802
+#define FM10K_STATS_UM 0x3803
+#define FM10K_STATS_XEC 0x3804
+#define FM10K_STATS_VLAN_DROP 0x3805
+#define FM10K_STATS_LOOPBACK_DROP 0x3806
+#define FM10K_STATS_NODESC_DROP 0x3807
+
+/* PCIe state registers */
+#define FM10K_PHYADDR 0x381C
+
+/* Rx ring registers */
+#define FM10K_RDBAL(_n) ((0x40 * (_n)) + 0x4000)
+#define FM10K_RDBAH(_n) ((0x40 * (_n)) + 0x4001)
+#define FM10K_RDLEN(_n) ((0x40 * (_n)) + 0x4002)
+#define FM10K_TPH_RXCTRL(_n) ((0x40 * (_n)) + 0x4003)
+#define FM10K_TPH_RXCTRL_DESC_TPHEN 0x00000020
+#define FM10K_TPH_RXCTRL_DESC_RROEN 0x00000200
+#define FM10K_TPH_RXCTRL_DATA_WROEN 0x00002000
+#define FM10K_TPH_RXCTRL_HDR_WROEN 0x00008000
+#define FM10K_RDH(_n) ((0x40 * (_n)) + 0x4004)
+#define FM10K_RDT(_n) ((0x40 * (_n)) + 0x4005)
+#define FM10K_RXQCTL(_n) ((0x40 * (_n)) + 0x4006)
+#define FM10K_RXQCTL_ENABLE 0x00000001
+#define FM10K_RXQCTL_PF 0x000000FC
+#define FM10K_RXQCTL_VF_SHIFT 2
+#define FM10K_RXQCTL_VF 0x00000100
+#define FM10K_RXQCTL_ID_MASK (FM10K_RXQCTL_PF | FM10K_RXQCTL_VF)
+#define FM10K_RXDCTL(_n) ((0x40 * (_n)) + 0x4007)
+#define FM10K_RXDCTL_WRITE_BACK_MIN_DELAY 0x00000001
+#define FM10K_RXDCTL_DROP_ON_EMPTY 0x00000200
+#define FM10K_RXINT(_n) ((0x40 * (_n)) + 0x4008)
+#define FM10K_SRRCTL(_n) ((0x40 * (_n)) + 0x4009)
+#define FM10K_SRRCTL_BSIZEPKT_SHIFT 8 /* shift _right_ */
+#define FM10K_SRRCTL_LOOPBACK_SUPPRESS 0x40000000
+#define FM10K_SRRCTL_BUFFER_CHAINING_EN 0x80000000
+
+/* Rx Statistics */
+#define FM10K_QPRC(_n) ((0x40 * (_n)) + 0x400A)
+#define FM10K_QPRDC(_n) ((0x40 * (_n)) + 0x400B)
+#define FM10K_QBRC_L(_n) ((0x40 * (_n)) + 0x400C)
+#define FM10K_QBRC_H(_n) ((0x40 * (_n)) + 0x400D)
+
+/* Rx GLORT register */
+#define FM10K_RX_SGLORT(_n) ((0x40 * (_n)) + 0x400E)
+
+/* Tx ring registers */
+#define FM10K_TDBAL(_n) ((0x40 * (_n)) + 0x8000)
+#define FM10K_TDBAH(_n) ((0x40 * (_n)) + 0x8001)
+#define FM10K_TDLEN(_n) ((0x40 * (_n)) + 0x8002)
+#define FM10K_TPH_TXCTRL(_n) ((0x40 * (_n)) + 0x8003)
+#define FM10K_TPH_TXCTRL_DESC_TPHEN 0x00000020
+#define FM10K_TPH_TXCTRL_DESC_RROEN 0x00000200
+#define FM10K_TPH_TXCTRL_DESC_WROEN 0x00000800
+#define FM10K_TPH_TXCTRL_DATA_RROEN 0x00002000
+#define FM10K_TDH(_n) ((0x40 * (_n)) + 0x8004)
+#define FM10K_TDT(_n) ((0x40 * (_n)) + 0x8005)
+#define FM10K_TXDCTL(_n) ((0x40 * (_n)) + 0x8006)
+#define FM10K_TXDCTL_ENABLE 0x00004000
+#define FM10K_TXDCTL_MAX_TIME_SHIFT 16
+#define FM10K_TXQCTL(_n) ((0x40 * (_n)) + 0x8007)
+#define FM10K_TXQCTL_PF 0x0000003F
+#define FM10K_TXQCTL_VF 0x00000040
+#define FM10K_TXQCTL_ID_MASK (FM10K_TXQCTL_PF | FM10K_TXQCTL_VF)
+#define FM10K_TXQCTL_PC_SHIFT 7
+#define FM10K_TXQCTL_PC_MASK 0x00000380
+#define FM10K_TXQCTL_TC_SHIFT 10
+#define FM10K_TXQCTL_VID_SHIFT 16
+#define FM10K_TXQCTL_VID_MASK 0x0FFF0000
+#define FM10K_TXQCTL_UNLIMITED_BW 0x10000000
+#define FM10K_TXINT(_n) ((0x40 * (_n)) + 0x8008)
+
+/* Tx Statistics */
+#define FM10K_QPTC(_n) ((0x40 * (_n)) + 0x8009)
+#define FM10K_QBTC_L(_n) ((0x40 * (_n)) + 0x800A)
+#define FM10K_QBTC_H(_n) ((0x40 * (_n)) + 0x800B)
+
+/* Tx Push registers */
+#define FM10K_TQDLOC(_n) ((0x40 * (_n)) + 0x800C)
+#define FM10K_TQDLOC_BASE_32_DESC 0x08
+#define FM10K_TQDLOC_SIZE_32_DESC 0x00050000
+
+/* Tx GLORT registers */
+#define FM10K_TX_SGLORT(_n) ((0x40 * (_n)) + 0x800D)
+#define FM10K_PFVTCTL(_n) ((0x40 * (_n)) + 0x800E)
+#define FM10K_PFVTCTL_FTAG_DESC_ENABLE 0x00000001
+
+/* Interrupt moderation and control registers */
+#define FM10K_INT_MAP(_n) ((_n) + 0x10080)
+#define FM10K_INT_MAP_TIMER0 0x00000000
+#define FM10K_INT_MAP_TIMER1 0x00000100
+#define FM10K_INT_MAP_IMMEDIATE 0x00000200
+#define FM10K_INT_MAP_DISABLE 0x00000300
+#define FM10K_MSIX_VECTOR_MASK(_n) ((0x4 * (_n)) + 0x11003)
+#define FM10K_INT_CTRL 0x12000
+#define FM10K_INT_CTRL_ENABLEMODERATOR 0x00000400
+#define FM10K_ITR(_n) ((_n) + 0x12400)
+#define FM10K_ITR_INTERVAL1_SHIFT 12
+#define FM10K_ITR_PENDING2 0x10000000
+#define FM10K_ITR_AUTOMASK 0x20000000
+#define FM10K_ITR_MASK_SET 0x40000000
+#define FM10K_ITR_MASK_CLEAR 0x80000000
+#define FM10K_ITR2(_n) ((0x2 * (_n)) + 0x12800)
+#define FM10K_ITR_REG_COUNT 768
+#define FM10K_ITR_REG_COUNT_PF 256
+
+/* Switch manager interrupt registers */
+#define FM10K_IP 0x13000
+#define FM10K_IP_NOTINRESET 0x00000100
+
+/* VLAN registers */
+#define FM10K_VLAN_TABLE(_n, _m) ((0x80 * (_n)) + (_m) + 0x14000)
+#define FM10K_VLAN_TABLE_SIZE 128
+
+/* VLAN specific message offsets */
+#define FM10K_VLAN_TABLE_VID_MAX 4096
+#define FM10K_VLAN_TABLE_VSI_MAX 64
+#define FM10K_VLAN_LENGTH_SHIFT 16
+#define FM10K_VLAN_CLEAR (1 << 15)
+#define FM10K_VLAN_ALL \
+ ((FM10K_VLAN_TABLE_VID_MAX - 1) << FM10K_VLAN_LENGTH_SHIFT)
+
+/* VF FLR event notification registers */
+#define FM10K_PFVFLRE(_n) ((0x1 * (_n)) + 0x18844)
+#define FM10K_PFVFLREC(_n) ((0x1 * (_n)) + 0x18846)
+
+/* Defines for size of uncacheable memories */
+#define FM10K_UC_ADDR_START 0x000000 /* start of standard regs */
+#define FM10K_UC_ADDR_END 0x100000 /* end of standard regs */
+#define FM10K_UC_ADDR_SIZE (FM10K_UC_ADDR_END - FM10K_UC_ADDR_START)
+
+/* Define timeouts for resets and disables */
+#define FM10K_QUEUE_DISABLE_TIMEOUT 100
+#define FM10K_RESET_TIMEOUT 100
+
+enum fm10k_int_source {
+ fm10k_int_Mailbox = 0,
+ fm10k_int_PCIeFault = 1,
+ fm10k_int_SwitchUpDown = 2,
+ fm10k_int_SwitchEvent = 3,
+ fm10k_int_SRAM = 4,
+ fm10k_int_VFLR = 5,
+ fm10k_int_MaxHoldTime = 6,
+ fm10k_int_sources_max_pf
+};
+
+/* PCIe bus speeds */
+enum fm10k_bus_speed {
+ fm10k_bus_speed_unknown = 0,
+ fm10k_bus_speed_2500 = 2500,
+ fm10k_bus_speed_5000 = 5000,
+ fm10k_bus_speed_8000 = 8000,
+ fm10k_bus_speed_reserved
+};
+
+/* PCIe bus widths */
+enum fm10k_bus_width {
+ fm10k_bus_width_unknown = 0,
+ fm10k_bus_width_pcie_x1 = 1,
+ fm10k_bus_width_pcie_x2 = 2,
+ fm10k_bus_width_pcie_x4 = 4,
+ fm10k_bus_width_pcie_x8 = 8,
+ fm10k_bus_width_reserved
+};
+
+/* PCIe payload sizes */
+enum fm10k_bus_payload {
+ fm10k_bus_payload_unknown = 0,
+ fm10k_bus_payload_128 = 1,
+ fm10k_bus_payload_256 = 2,
+ fm10k_bus_payload_512 = 3,
+ fm10k_bus_payload_reserved
+};
+
+/* Bus parameters */
+struct fm10k_bus_info {
+ enum fm10k_bus_speed speed;
+ enum fm10k_bus_width width;
+ enum fm10k_bus_payload payload;
+};
+
+/* Statistics related declarations */
+struct fm10k_hw_stat {
+ u64 count;
+ u32 base_l;
+ u32 base_h;
+};
+
+struct fm10k_hw_stats_q {
+ struct fm10k_hw_stat tx_bytes;
+ struct fm10k_hw_stat tx_packets;
+#define tx_stats_idx tx_packets.base_h
+ struct fm10k_hw_stat rx_bytes;
+ struct fm10k_hw_stat rx_packets;
+#define rx_stats_idx rx_packets.base_h
+ struct fm10k_hw_stat rx_drops;
+};
+
+struct fm10k_hw_stats {
+ struct fm10k_hw_stat timeout;
+#define stats_idx timeout.base_h
+ struct fm10k_hw_stat ur;
+ struct fm10k_hw_stat ca;
+ struct fm10k_hw_stat um;
+ struct fm10k_hw_stat xec;
+ struct fm10k_hw_stat vlan_drop;
+ struct fm10k_hw_stat loopback_drop;
+ struct fm10k_hw_stat nodesc_drop;
+ struct fm10k_hw_stats_q q[FM10K_MAX_QUEUES_PF];
+};
+
+/* Establish DGLORT feature priority */
+enum fm10k_dglortdec_idx {
+ fm10k_dglort_default = 0,
+ fm10k_dglort_vf_rsvd0 = 1,
+ fm10k_dglort_vf_rss = 2,
+ fm10k_dglort_pf_rsvd0 = 3,
+ fm10k_dglort_pf_queue = 4,
+ fm10k_dglort_pf_vsi = 5,
+ fm10k_dglort_pf_rsvd1 = 6,
+ fm10k_dglort_pf_rss = 7
+};
+
+struct fm10k_dglort_cfg {
+ u16 glort; /* GLORT base */
+ u16 queue_b; /* Base value for queue */
+ u8 vsi_b; /* Base value for VSI */
+ u8 idx; /* index of DGLORTDEC entry */
+ u8 rss_l; /* RSS indices */
+ u8 pc_l; /* Priority Class indices */
+ u8 vsi_l; /* Number of bits from GLORT used to determine VSI */
+ u8 queue_l; /* Number of bits from GLORT used to determine queue */
+ u8 shared_l; /* Ignored bits from GLORT resulting in shared VSI */
+ u8 inner_rss; /* Boolean value if inner header is used for RSS */
+};
+
+enum fm10k_pca_fault {
+ PCA_NO_FAULT,
+ PCA_UNMAPPED_ADDR,
+ PCA_BAD_QACCESS_PF,
+ PCA_BAD_QACCESS_VF,
+ PCA_MALICIOUS_REQ,
+ PCA_POISONED_TLP,
+ PCA_TLP_ABORT,
+ __PCA_MAX
+};
+
+enum fm10k_thi_fault {
+ THI_NO_FAULT,
+ THI_MAL_DIS_Q_FAULT,
+ __THI_MAX
+};
+
+enum fm10k_fum_fault {
+ FUM_NO_FAULT,
+ FUM_UNMAPPED_ADDR,
+ FUM_POISONED_TLP,
+ FUM_BAD_VF_QACCESS,
+ FUM_ADD_DECODE_ERR,
+ FUM_RO_ERROR,
+ FUM_QPRC_CRC_ERROR,
+ FUM_CSR_TIMEOUT,
+ FUM_INVALID_TYPE,
+ FUM_INVALID_LENGTH,
+ FUM_INVALID_BE,
+ FUM_INVALID_ALIGN,
+ __FUM_MAX
+};
+
+struct fm10k_fault {
+ u64 address; /* Address at the time fault was detected */
+ u32 specinfo; /* Extra info on this fault (fault dependent) */
+ u8 type; /* Fault value dependent on subunit */
+ u8 func; /* Function number of the fault */
+};
+
+struct fm10k_mac_ops {
+ /* basic bring-up and tear-down */
+ s32 (*reset_hw)(struct fm10k_hw *);
+ s32 (*init_hw)(struct fm10k_hw *);
+ s32 (*start_hw)(struct fm10k_hw *);
+ s32 (*stop_hw)(struct fm10k_hw *);
+ s32 (*get_bus_info)(struct fm10k_hw *);
+ s32 (*get_host_state)(struct fm10k_hw *, bool *);
+ bool (*is_slot_appropriate)(struct fm10k_hw *);
+ s32 (*update_vlan)(struct fm10k_hw *, u32, u8, bool);
+ s32 (*read_mac_addr)(struct fm10k_hw *);
+ s32 (*update_uc_addr)(struct fm10k_hw *, u16, const u8 *,
+ u16, bool, u8);
+ s32 (*update_mc_addr)(struct fm10k_hw *, u16, const u8 *, u16, bool);
+ s32 (*update_xcast_mode)(struct fm10k_hw *, u16, u8);
+ void (*update_int_moderator)(struct fm10k_hw *);
+ s32 (*update_lport_state)(struct fm10k_hw *, u16, u16, bool);
+ void (*update_hw_stats)(struct fm10k_hw *, struct fm10k_hw_stats *);
+ void (*rebind_hw_stats)(struct fm10k_hw *, struct fm10k_hw_stats *);
+ s32 (*configure_dglort_map)(struct fm10k_hw *,
+ struct fm10k_dglort_cfg *);
+ void (*set_dma_mask)(struct fm10k_hw *, u64);
+ s32 (*get_fault)(struct fm10k_hw *, int, struct fm10k_fault *);
+ void (*request_lport_map)(struct fm10k_hw *);
+ s32 (*adjust_systime)(struct fm10k_hw *, s32 ppb);
+};
+
+enum fm10k_mac_type {
+ fm10k_mac_unknown = 0,
+ fm10k_mac_pf,
+ fm10k_num_macs
+};
+
+struct fm10k_mac_info {
+ struct fm10k_mac_ops ops;
+ enum fm10k_mac_type type;
+ u8 addr[ETH_ALEN];
+ u8 perm_addr[ETH_ALEN];
+ u16 default_vid;
+ u16 max_msix_vectors;
+ u16 max_queues;
+ bool vlan_override;
+ bool get_host_state;
+ bool tx_ready;
+ u32 dglort_map;
+};
+
+struct fm10k_swapi_table_info {
+ u32 used;
+ u32 avail;
+};
+
+struct fm10k_swapi_info {
+ u32 status;
+ struct fm10k_swapi_table_info mac;
+ struct fm10k_swapi_table_info nexthop;
+ struct fm10k_swapi_table_info ffu;
+};
+
+enum fm10k_xcast_modes {
+ FM10K_XCAST_MODE_ALLMULTI = 0,
+ FM10K_XCAST_MODE_MULTI = 1,
+ FM10K_XCAST_MODE_PROMISC = 2,
+ FM10K_XCAST_MODE_NONE = 3,
+ FM10K_XCAST_MODE_DISABLE = 4
+};
+
+enum fm10k_devices {
+ fm10k_device_pf,
+};
+
+struct fm10k_info {
+ enum fm10k_mac_type mac;
+ s32 (*get_invariants)(struct fm10k_hw *);
+ struct fm10k_mac_ops *mac_ops;
+};
+
+struct fm10k_hw {
+ u32 __iomem *hw_addr;
+ void *back;
+ struct fm10k_mac_info mac;
+ struct fm10k_bus_info bus;
+ struct fm10k_bus_info bus_caps;
+ struct fm10k_swapi_info swapi;
+ u16 device_id;
+ u16 vendor_id;
+ u16 subsystem_device_id;
+ u16 subsystem_vendor_id;
+ u8 revision_id;
+};
+
+/* Number of Transmit and Receive Descriptors must be a multiple of 8 */
+#define FM10K_REQ_TX_DESCRIPTOR_MULTIPLE 8
+#define FM10K_REQ_RX_DESCRIPTOR_MULTIPLE 8
+
+/* Transmit Descriptor */
+struct fm10k_tx_desc {
+ __le64 buffer_addr; /* Address of the descriptor's data buffer */
+ __le16 buflen; /* Length of data to be DMAed */
+ __le16 vlan; /* VLAN_ID and VPRI to be inserted in FTAG */
+ __le16 mss; /* MSS for segmentation offload */
+ u8 hdrlen; /* Header size for segmentation offload */
+ u8 flags; /* Status and offload request flags */
+};
+
+/* Transmit Descriptor Cache Structure */
+struct fm10k_tx_desc_cache {
+ struct fm10k_tx_desc tx_desc[256];
+};
+
+#define FM10K_TXD_FLAG_INT 0x01
+#define FM10K_TXD_FLAG_TIME 0x02
+#define FM10K_TXD_FLAG_CSUM 0x04
+#define FM10K_TXD_FLAG_FTAG 0x10
+#define FM10K_TXD_FLAG_RS 0x20
+#define FM10K_TXD_FLAG_LAST 0x40
+#define FM10K_TXD_FLAG_DONE 0x80
+
+/* These macros are meant to enable optimal placement of the RS and INT
+ * bits. It will point us to the last descriptor in the cache for either the
+ * start of the packet, or the end of the packet. If the index is actually
+ * at the start of the FIFO it will point to the offset for the last index
+ * in the FIFO to prevent an unnecessary write.
+ */
+#define FM10K_TXD_WB_FIFO_SIZE 4
+
+/* Receive Descriptor - 32B */
+union fm10k_rx_desc {
+ struct {
+ __le64 pkt_addr; /* Packet buffer address */
+ __le64 hdr_addr; /* Header buffer address */
+ __le64 reserved; /* Empty space, RSS hash */
+ __le64 timestamp;
+ } q; /* Read, Writeback, 64b quad-words */
+ struct {
+ __le32 data; /* RSS and header data */
+ __le32 rss; /* RSS Hash */
+ __le32 staterr;
+ __le32 vlan_len;
+ __le32 glort; /* sglort/dglort */
+ } d; /* Writeback, 32b double-words */
+ struct {
+ __le16 pkt_info; /* RSS, Pkt type */
+ __le16 hdr_info; /* Splithdr, hdrlen, xC */
+ __le16 rss_lower;
+ __le16 rss_upper;
+ __le16 status; /* status/error */
+ __le16 csum_err; /* checksum or extended error value */
+ __le16 length; /* Packet length */
+ __le16 vlan; /* VLAN tag */
+ __le16 dglort;
+ __le16 sglort;
+ } w; /* Writeback, 16b words */
+};
+
+#define FM10K_RXD_RSSTYPE_MASK 0x000F
+enum fm10k_rdesc_rss_type {
+ FM10K_RSSTYPE_NONE = 0x0,
+ FM10K_RSSTYPE_IPV4_TCP = 0x1,
+ FM10K_RSSTYPE_IPV4 = 0x2,
+ FM10K_RSSTYPE_IPV6_TCP = 0x3,
+ /* Reserved 0x4 */
+ FM10K_RSSTYPE_IPV6 = 0x5,
+ /* Reserved 0x6 */
+ FM10K_RSSTYPE_IPV4_UDP = 0x7,
+ FM10K_RSSTYPE_IPV6_UDP = 0x8
+ /* Reserved 0x9 - 0xF */
+};
+
+#define FM10K_RXD_HDR_INFO_XC_MASK 0x0006
+enum fm10k_rxdesc_xc {
+ FM10K_XC_UNICAST = 0x0,
+ FM10K_XC_MULTICAST = 0x4,
+ FM10K_XC_BROADCAST = 0x6
+};
+
+#define FM10K_RXD_STATUS_DD 0x0001 /* Descriptor done */
+#define FM10K_RXD_STATUS_EOP 0x0002 /* End of packet */
+#define FM10K_RXD_STATUS_L4CS 0x0010 /* Indicates an L4 csum */
+#define FM10K_RXD_STATUS_L4CS2 0x0040 /* Inner header L4 csum */
+#define FM10K_RXD_STATUS_L4E2 0x0800 /* Inner header L4 csum err */
+#define FM10K_RXD_STATUS_IPE2 0x1000 /* Inner header IPv4 csum err */
+#define FM10K_RXD_STATUS_RXE 0x2000 /* Generic Rx error */
+#define FM10K_RXD_STATUS_L4E 0x4000 /* L4 csum error */
+#define FM10K_RXD_STATUS_IPE 0x8000 /* IPv4 csum error */
+
+struct fm10k_ftag {
+ __be16 swpri_type_user;
+ __be16 vlan;
+ __be16 sglort;
+ __be16 dglort;
+};
+
#endif /* _FM10K_TYPE_H */
^ permalink raw reply related
* [net-next PATCH 05/29] fm10k: Add support for mailbox
From: Alexander Duyck @ 2014-09-18 22:36 UTC (permalink / raw)
To: davem
Cc: nhorman, netdev, john.fastabend, matthew.vick, jeffrey.t.kirsher,
sassmann
In-Reply-To: <20140918223242.10373.27403.stgit@ahduyck-bv4.jf.intel.com>
This patch adds generic mailbox support. The general idea of the mailboxes
is to use a pair of ring buffers, one for request, one for response to send
data between the local driver and some remote entity be it the PF of the
Switch Manager.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/fm10k/fm10k_mbx.h | 241 +++++++++++++++++++
drivers/net/ethernet/intel/fm10k/fm10k_tlv.c | 320 +++++++++++++++++++++++++
drivers/net/ethernet/intel/fm10k/fm10k_tlv.h | 45 ++++
drivers/net/ethernet/intel/fm10k/fm10k_type.h | 3
4 files changed, 609 insertions(+)
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_mbx.h
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_mbx.h b/drivers/net/ethernet/intel/fm10k/fm10k_mbx.h
new file mode 100644
index 0000000..ee2f712
--- /dev/null
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_mbx.h
@@ -0,0 +1,241 @@
+/* Intel Ethernet Switch Host Interface Driver
+ * Copyright(c) 2013 - 2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
+
+#ifndef _FM10K_MBX_H_
+#define _FM10K_MBX_H_
+
+/* forward declaration */
+struct fm10k_mbx_info;
+
+#include "fm10k_type.h"
+#include "fm10k_tlv.h"
+
+/* PF Mailbox Registers */
+#define FM10K_MBMEM(_n) ((_n) + 0x18000)
+#define FM10K_MBMEM_VF(_n, _m) (((_n) * 0x10) + (_m) + 0x18000)
+#define FM10K_MBMEM_SM(_n) ((_n) + 0x18400)
+#define FM10K_MBMEM_PF(_n) ((_n) + 0x18600)
+/* XOR provides means of switching from Tx to Rx FIFO */
+#define FM10K_MBMEM_PF_XOR (FM10K_MBMEM_SM(0) ^ FM10K_MBMEM_PF(0))
+#define FM10K_MBX(_n) ((_n) + 0x18800)
+#define FM10K_MBX_REQ 0x00000002
+#define FM10K_MBX_ACK 0x00000004
+#define FM10K_MBX_REQ_INTERRUPT 0x00000008
+#define FM10K_MBX_ACK_INTERRUPT 0x00000010
+#define FM10K_MBX_INTERRUPT_ENABLE 0x00000020
+#define FM10K_MBX_INTERRUPT_DISABLE 0x00000040
+#define FM10K_MBICR(_n) ((_n) + 0x18840)
+#define FM10K_GMBX 0x18842
+
+/* VF Mailbox Registers */
+#define FM10K_VFMBX 0x00010
+#define FM10K_VFMBMEM(_n) ((_n) + 0x00020)
+#define FM10K_VFMBMEM_LEN 16
+#define FM10K_VFMBMEM_VF_XOR (FM10K_VFMBMEM_LEN / 2)
+
+/* Delays/timeouts */
+#define FM10K_MBX_DISCONNECT_TIMEOUT 500
+#define FM10K_MBX_POLL_DELAY 19
+#define FM10K_MBX_INT_DELAY 20
+
+/* PF/VF Mailbox state machine
+ *
+ * +----------+ connect() +----------+
+ * | CLOSED | --------------> | CONNECT |
+ * +----------+ +----------+
+ * ^ ^ |
+ * | rcv: rcv: | | rcv:
+ * | Connect Disconnect | | Connect
+ * | Disconnect Error | | Data
+ * | | |
+ * | | V
+ * +----------+ disconnect() +----------+
+ * |DISCONNECT| <-------------- | OPEN |
+ * +----------+ +----------+
+ *
+ * The diagram above describes the PF/VF mailbox state machine. There
+ * are four main states to this machine.
+ * Closed: This state represents a mailbox that is in a standby state
+ * with interrupts disabled. In this state the mailbox should not
+ * read the mailbox or write any data. The only means of exiting
+ * this state is for the system to make the connect() call for the
+ * mailbox, it will then transition to the connect state.
+ * Connect: In this state the mailbox is seeking a connection. It will
+ * post a connect message with no specified destination and will
+ * wait for a reply from the other side of the mailbox. This state
+ * is exited when either a connect with the local mailbox as the
+ * destination is received or when a data message is received with
+ * a valid sequence number.
+ * Open: In this state the mailbox is able to transfer data between the local
+ * entity and the remote. It will fall back to connect in the event of
+ * receiving either an error message, or a disconnect message. It will
+ * transition to disconnect on a call to disconnect();
+ * Disconnect: In this state the mailbox is attempting to gracefully terminate
+ * the connection. It will do so at the first point where it knows
+ * that the remote endpoint is either done sending, or when the
+ * remote endpoint has fallen back into connect.
+ */
+enum fm10k_mbx_state {
+ FM10K_STATE_CLOSED,
+ FM10K_STATE_CONNECT,
+ FM10K_STATE_OPEN,
+ FM10K_STATE_DISCONNECT,
+};
+
+/* macros for retriving and setting header values */
+#define FM10K_MSG_HDR_MASK(name) \
+ ((0x1u << FM10K_MSG_##name##_SIZE) - 1)
+#define FM10K_MSG_HDR_FIELD_SET(value, name) \
+ (((u32)(value) & FM10K_MSG_HDR_MASK(name)) << FM10K_MSG_##name##_SHIFT)
+#define FM10K_MSG_HDR_FIELD_GET(value, name) \
+ ((u16)((value) >> FM10K_MSG_##name##_SHIFT) & FM10K_MSG_HDR_MASK(name))
+
+/* HNI/SM Mailbox FIFO format
+ * 3 2 1 0
+ * 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0
+ * +-------+-----------------------+-------+-----------------------+
+ * | Error | Remote Head |Version| Local Tail |
+ * +-------+-----------------------+-------+-----------------------+
+ * | |
+ * . Local FIFO Data .
+ * . .
+ * +-------+-----------------------+-------+-----------------------+
+ *
+ * The layout above describes the format for the FIFOs used by the host
+ * network interface and the switch manager to communicate messages back
+ * and forth. Both the HNI and the switch maintain one such FIFO. The
+ * layout in memory has the switch manager FIFO followed immediately by
+ * the HNI FIFO. For this reason I am using just the pointer to the
+ * HNI FIFO in the mailbox ops as the offset between the two is fixed.
+ *
+ * The header for the FIFO is broken out into the following fields:
+ * Local Tail: Offset into FIFO region for next DWORD to write.
+ * Version: Version info for mailbox, only values of 0/1 are supported.
+ * Remote Head: Offset into remote FIFO to indicate how much we have read.
+ * Error: Error indication, values TBD.
+ */
+
+/* version number for switch manager mailboxes */
+#define FM10K_SM_MBX_VERSION 1
+#define FM10K_SM_MBX_FIFO_LEN (FM10K_MBMEM_PF_XOR - 1)
+#define FM10K_SM_MBX_FIFO_HDR_LEN 1
+
+/* offsets shared between all SM FIFO headers */
+#define FM10K_MSG_SM_TAIL_SHIFT 0
+#define FM10K_MSG_SM_TAIL_SIZE 12
+#define FM10K_MSG_SM_VER_SHIFT 12
+#define FM10K_MSG_SM_VER_SIZE 4
+#define FM10K_MSG_SM_HEAD_SHIFT 16
+#define FM10K_MSG_SM_HEAD_SIZE 12
+#define FM10K_MSG_SM_ERR_SHIFT 28
+#define FM10K_MSG_SM_ERR_SIZE 4
+
+/* All error messages returned by mailbox functions
+ * The value -511 is 0xFE01 in hex. The idea is to order the errors
+ * from 0xFE01 - 0xFEFF so error codes are easily visible in the mailbox
+ * messages. This also helps to avoid error number collisions as Linux
+ * doesn't appear to use error numbers 256 - 511.
+ */
+#define FM10K_MBX_ERR(_n) ((_n) - 512)
+#define FM10K_MBX_ERR_NO_MBX FM10K_MBX_ERR(0x01)
+#define FM10K_MBX_ERR_NO_SPACE FM10K_MBX_ERR(0x03)
+#define FM10K_MBX_ERR_TAIL FM10K_MBX_ERR(0x05)
+#define FM10K_MBX_ERR_HEAD FM10K_MBX_ERR(0x06)
+#define FM10K_MBX_ERR_SRC FM10K_MBX_ERR(0x08)
+#define FM10K_MBX_ERR_TYPE FM10K_MBX_ERR(0x09)
+#define FM10K_MBX_ERR_SIZE FM10K_MBX_ERR(0x0B)
+#define FM10K_MBX_ERR_BUSY FM10K_MBX_ERR(0x0C)
+#define FM10K_MBX_ERR_RSVD0 FM10K_MBX_ERR(0x0E)
+#define FM10K_MBX_ERR_CRC FM10K_MBX_ERR(0x0F)
+
+#define FM10K_MBX_CRC_SEED 0xFFFF
+
+struct fm10k_mbx_ops {
+ s32 (*connect)(struct fm10k_hw *, struct fm10k_mbx_info *);
+ void (*disconnect)(struct fm10k_hw *, struct fm10k_mbx_info *);
+ bool (*rx_ready)(struct fm10k_mbx_info *);
+ bool (*tx_ready)(struct fm10k_mbx_info *, u16);
+ bool (*tx_complete)(struct fm10k_mbx_info *);
+ s32 (*enqueue_tx)(struct fm10k_hw *, struct fm10k_mbx_info *,
+ const u32 *);
+ s32 (*process)(struct fm10k_hw *, struct fm10k_mbx_info *);
+ s32 (*register_handlers)(struct fm10k_mbx_info *,
+ const struct fm10k_msg_data *);
+};
+
+struct fm10k_mbx_fifo {
+ u32 *buffer;
+ u16 head;
+ u16 tail;
+ u16 size;
+};
+
+/* size of buffer to be stored in mailbox for FIFOs */
+#define FM10K_MBX_TX_BUFFER_SIZE 512
+#define FM10K_MBX_RX_BUFFER_SIZE 128
+#define FM10K_MBX_BUFFER_SIZE \
+ (FM10K_MBX_TX_BUFFER_SIZE + FM10K_MBX_RX_BUFFER_SIZE)
+
+/* minimum and maximum message size in dwords */
+#define FM10K_MBX_MSG_MAX_SIZE \
+ ((FM10K_MBX_TX_BUFFER_SIZE - 1) & (FM10K_MBX_RX_BUFFER_SIZE - 1))
+#define FM10K_VFMBX_MSG_MTU ((FM10K_VFMBMEM_LEN / 2) - 1)
+
+#define FM10K_MBX_INIT_TIMEOUT 2000 /* number of retries on mailbox */
+#define FM10K_MBX_INIT_DELAY 500 /* microseconds between retries */
+
+struct fm10k_mbx_info {
+ /* function pointers for mailbox operations */
+ struct fm10k_mbx_ops ops;
+ const struct fm10k_msg_data *msg_data;
+
+ /* message FIFOs */
+ struct fm10k_mbx_fifo rx;
+ struct fm10k_mbx_fifo tx;
+
+ /* delay for handling timeouts */
+ u32 timeout;
+ u32 udelay;
+
+ /* mailbox state info */
+ u32 mbx_reg, mbmem_reg, mbx_lock, mbx_hdr;
+ u16 max_size, mbmem_len;
+ u16 tail, tail_len, pulled;
+ u16 head, head_len, pushed;
+ u16 local, remote;
+ enum fm10k_mbx_state state;
+
+ /* result of last mailbox test */
+ s32 test_result;
+
+ /* statistics */
+ u64 tx_busy;
+ u64 tx_dropped;
+ u64 tx_messages;
+ u64 tx_dwords;
+ u64 rx_messages;
+ u64 rx_dwords;
+ u64 rx_parse_err;
+
+ /* Buffer to store messages */
+ u32 buffer[FM10K_MBX_BUFFER_SIZE];
+};
+
+#endif /* _FM10K_MBX_H_ */
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c
index f1fc709..fd0a05f 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c
@@ -541,3 +541,323 @@ s32 fm10k_tlv_attr_parse(u32 *attr, u32 **results,
return 0;
}
+
+/**
+ * fm10k_tlv_msg_parse - Parses message header and calls function handler
+ * @hw: Pointer to hardware structure
+ * @msg: Pointer to message
+ * @mbx: Pointer to mailbox information structure
+ * @func: Function array containing list of message handling functions
+ *
+ * This function should be the first function called upon receiving a
+ * message. The handler will identify the message type and call the correct
+ * handler for the given message. It will return the value from the function
+ * call on a recognized message type, otherwise it will return
+ * FM10K_NOT_IMPLEMENTED on an unrecognized type.
+ **/
+s32 fm10k_tlv_msg_parse(struct fm10k_hw *hw, u32 *msg,
+ struct fm10k_mbx_info *mbx,
+ const struct fm10k_msg_data *data)
+{
+ u32 *results[FM10K_TLV_RESULTS_MAX];
+ u32 msg_id;
+ s32 err;
+
+ /* verify pointer is not NULL */
+ if (!msg || !data)
+ return FM10K_ERR_PARAM;
+
+ /* verify this is a message and not an attribute */
+ if (!(*msg & (FM10K_TLV_FLAGS_MSG << FM10K_TLV_FLAGS_SHIFT)))
+ return FM10K_ERR_PARAM;
+
+ /* grab message ID */
+ msg_id = *msg & FM10K_TLV_ID_MASK;
+
+ while (data->id < msg_id)
+ data++;
+
+ /* if we didn't find it then pass it up as an error */
+ if (data->id != msg_id) {
+ while (data->id != FM10K_TLV_ERROR)
+ data++;
+ }
+
+ /* parse the attributes into the results list */
+ err = fm10k_tlv_attr_parse(msg, results, data->attr);
+ if (err < 0)
+ return err;
+
+ return data->func(hw, results, mbx);
+}
+
+/**
+ * fm10k_tlv_msg_error - Default handler for unrecognized TLV message IDs
+ * @hw: Pointer to hardware structure
+ * @results: Pointer array to message, results[0] is pointer to message
+ * @mbx: Unused mailbox pointer
+ *
+ * This function is a default handler for unrecognized messages. At a
+ * a minimum it just indicates that the message requested was
+ * unimplemented.
+ **/
+s32 fm10k_tlv_msg_error(struct fm10k_hw *hw, u32 **results,
+ struct fm10k_mbx_info *mbx)
+{
+ return FM10K_NOT_IMPLEMENTED;
+}
+
+static const unsigned char test_str[] = "fm10k";
+static const unsigned char test_mac[ETH_ALEN] = { 0x12, 0x34, 0x56,
+ 0x78, 0x9a, 0xbc };
+static const u16 test_vlan = 0x0FED;
+static const u64 test_u64 = 0xfedcba9876543210ull;
+static const u32 test_u32 = 0x87654321;
+static const u16 test_u16 = 0x8765;
+static const u8 test_u8 = 0x87;
+static const s64 test_s64 = -0x123456789abcdef0ll;
+static const s32 test_s32 = -0x1235678;
+static const s16 test_s16 = -0x1234;
+static const s8 test_s8 = -0x12;
+static const __le32 test_le[2] = { cpu_to_le32(0x12345678),
+ cpu_to_le32(0x9abcdef0)};
+
+/* The message below is meant to be used as a test message to demonstrate
+ * how to use the TLV interface and to test the types. Normally this code
+ * be compiled out by stripping the code wrapped in FM10K_TLV_TEST_MSG
+ */
+const struct fm10k_tlv_attr fm10k_tlv_msg_test_attr[] = {
+ FM10K_TLV_ATTR_NULL_STRING(FM10K_TEST_MSG_STRING, 80),
+ FM10K_TLV_ATTR_MAC_ADDR(FM10K_TEST_MSG_MAC_ADDR),
+ FM10K_TLV_ATTR_U8(FM10K_TEST_MSG_U8),
+ FM10K_TLV_ATTR_U16(FM10K_TEST_MSG_U16),
+ FM10K_TLV_ATTR_U32(FM10K_TEST_MSG_U32),
+ FM10K_TLV_ATTR_U64(FM10K_TEST_MSG_U64),
+ FM10K_TLV_ATTR_S8(FM10K_TEST_MSG_S8),
+ FM10K_TLV_ATTR_S16(FM10K_TEST_MSG_S16),
+ FM10K_TLV_ATTR_S32(FM10K_TEST_MSG_S32),
+ FM10K_TLV_ATTR_S64(FM10K_TEST_MSG_S64),
+ FM10K_TLV_ATTR_LE_STRUCT(FM10K_TEST_MSG_LE_STRUCT, 8),
+ FM10K_TLV_ATTR_NESTED(FM10K_TEST_MSG_NESTED),
+ FM10K_TLV_ATTR_S32(FM10K_TEST_MSG_RESULT),
+ FM10K_TLV_ATTR_LAST
+};
+
+/**
+ * fm10k_tlv_msg_test_generate_data - Stuff message with data
+ * @msg: Pointer to message
+ * @attr_flags: List of flags indicating what attributes to add
+ *
+ * This function is meant to load a message buffer with attribute data
+ **/
+static void fm10k_tlv_msg_test_generate_data(u32 *msg, u32 attr_flags)
+{
+ if (attr_flags & (1 << FM10K_TEST_MSG_STRING))
+ fm10k_tlv_attr_put_null_string(msg, FM10K_TEST_MSG_STRING,
+ test_str);
+ if (attr_flags & (1 << FM10K_TEST_MSG_MAC_ADDR))
+ fm10k_tlv_attr_put_mac_vlan(msg, FM10K_TEST_MSG_MAC_ADDR,
+ test_mac, test_vlan);
+ if (attr_flags & (1 << FM10K_TEST_MSG_U8))
+ fm10k_tlv_attr_put_u8(msg, FM10K_TEST_MSG_U8, test_u8);
+ if (attr_flags & (1 << FM10K_TEST_MSG_U16))
+ fm10k_tlv_attr_put_u16(msg, FM10K_TEST_MSG_U16, test_u16);
+ if (attr_flags & (1 << FM10K_TEST_MSG_U32))
+ fm10k_tlv_attr_put_u32(msg, FM10K_TEST_MSG_U32, test_u32);
+ if (attr_flags & (1 << FM10K_TEST_MSG_U64))
+ fm10k_tlv_attr_put_u64(msg, FM10K_TEST_MSG_U64, test_u64);
+ if (attr_flags & (1 << FM10K_TEST_MSG_S8))
+ fm10k_tlv_attr_put_s8(msg, FM10K_TEST_MSG_S8, test_s8);
+ if (attr_flags & (1 << FM10K_TEST_MSG_S16))
+ fm10k_tlv_attr_put_s16(msg, FM10K_TEST_MSG_S16, test_s16);
+ if (attr_flags & (1 << FM10K_TEST_MSG_S32))
+ fm10k_tlv_attr_put_s32(msg, FM10K_TEST_MSG_S32, test_s32);
+ if (attr_flags & (1 << FM10K_TEST_MSG_S64))
+ fm10k_tlv_attr_put_s64(msg, FM10K_TEST_MSG_S64, test_s64);
+ if (attr_flags & (1 << FM10K_TEST_MSG_LE_STRUCT))
+ fm10k_tlv_attr_put_le_struct(msg, FM10K_TEST_MSG_LE_STRUCT,
+ test_le, 8);
+}
+
+/**
+ * fm10k_tlv_msg_test_create - Create a test message testing all attributes
+ * @msg: Pointer to message
+ * @attr_flags: List of flags indicating what attributes to add
+ *
+ * This function is meant to load a message buffer with all attribute types
+ * including a nested attribute.
+ **/
+void fm10k_tlv_msg_test_create(u32 *msg, u32 attr_flags)
+{
+ u32 *nest = NULL;
+
+ fm10k_tlv_msg_init(msg, FM10K_TLV_MSG_ID_TEST);
+
+ fm10k_tlv_msg_test_generate_data(msg, attr_flags);
+
+ /* check for nested attributes */
+ attr_flags >>= FM10K_TEST_MSG_NESTED;
+
+ if (attr_flags) {
+ nest = fm10k_tlv_attr_nest_start(msg, FM10K_TEST_MSG_NESTED);
+
+ fm10k_tlv_msg_test_generate_data(nest, attr_flags);
+
+ fm10k_tlv_attr_nest_stop(msg);
+ }
+}
+
+/**
+ * fm10k_tlv_msg_test - Validate all results on test message receive
+ * @hw: Pointer to hardware structure
+ * @results: Pointer array to attributes in the mesage
+ * @mbx: Pointer to mailbox information structure
+ *
+ * This function does a check to verify all attributes match what the test
+ * message placed in the message buffer. It is the default handler
+ * for TLV test messages.
+ **/
+s32 fm10k_tlv_msg_test(struct fm10k_hw *hw, u32 **results,
+ struct fm10k_mbx_info *mbx)
+{
+ u32 *nest_results[FM10K_TLV_RESULTS_MAX];
+ unsigned char result_str[80];
+ unsigned char result_mac[ETH_ALEN];
+ s32 err = 0;
+ __le32 result_le[2];
+ u16 result_vlan;
+ u64 result_u64;
+ u32 result_u32;
+ u16 result_u16;
+ u8 result_u8;
+ s64 result_s64;
+ s32 result_s32;
+ s16 result_s16;
+ s8 result_s8;
+ u32 reply[3];
+
+ /* retrieve results of a previous test */
+ if (!!results[FM10K_TEST_MSG_RESULT])
+ return fm10k_tlv_attr_get_s32(results[FM10K_TEST_MSG_RESULT],
+ &mbx->test_result);
+
+parse_nested:
+ if (!!results[FM10K_TEST_MSG_STRING]) {
+ err = fm10k_tlv_attr_get_null_string(
+ results[FM10K_TEST_MSG_STRING],
+ result_str);
+ if (!err && memcmp(test_str, result_str, sizeof(test_str)))
+ err = FM10K_ERR_INVALID_VALUE;
+ if (err)
+ goto report_result;
+ }
+ if (!!results[FM10K_TEST_MSG_MAC_ADDR]) {
+ err = fm10k_tlv_attr_get_mac_vlan(
+ results[FM10K_TEST_MSG_MAC_ADDR],
+ result_mac, &result_vlan);
+ if (!err && memcmp(test_mac, result_mac, ETH_ALEN))
+ err = FM10K_ERR_INVALID_VALUE;
+ if (!err && test_vlan != result_vlan)
+ err = FM10K_ERR_INVALID_VALUE;
+ if (err)
+ goto report_result;
+ }
+ if (!!results[FM10K_TEST_MSG_U8]) {
+ err = fm10k_tlv_attr_get_u8(results[FM10K_TEST_MSG_U8],
+ &result_u8);
+ if (!err && test_u8 != result_u8)
+ err = FM10K_ERR_INVALID_VALUE;
+ if (err)
+ goto report_result;
+ }
+ if (!!results[FM10K_TEST_MSG_U16]) {
+ err = fm10k_tlv_attr_get_u16(results[FM10K_TEST_MSG_U16],
+ &result_u16);
+ if (!err && test_u16 != result_u16)
+ err = FM10K_ERR_INVALID_VALUE;
+ if (err)
+ goto report_result;
+ }
+ if (!!results[FM10K_TEST_MSG_U32]) {
+ err = fm10k_tlv_attr_get_u32(results[FM10K_TEST_MSG_U32],
+ &result_u32);
+ if (!err && test_u32 != result_u32)
+ err = FM10K_ERR_INVALID_VALUE;
+ if (err)
+ goto report_result;
+ }
+ if (!!results[FM10K_TEST_MSG_U64]) {
+ err = fm10k_tlv_attr_get_u64(results[FM10K_TEST_MSG_U64],
+ &result_u64);
+ if (!err && test_u64 != result_u64)
+ err = FM10K_ERR_INVALID_VALUE;
+ if (err)
+ goto report_result;
+ }
+ if (!!results[FM10K_TEST_MSG_S8]) {
+ err = fm10k_tlv_attr_get_s8(results[FM10K_TEST_MSG_S8],
+ &result_s8);
+ if (!err && test_s8 != result_s8)
+ err = FM10K_ERR_INVALID_VALUE;
+ if (err)
+ goto report_result;
+ }
+ if (!!results[FM10K_TEST_MSG_S16]) {
+ err = fm10k_tlv_attr_get_s16(results[FM10K_TEST_MSG_S16],
+ &result_s16);
+ if (!err && test_s16 != result_s16)
+ err = FM10K_ERR_INVALID_VALUE;
+ if (err)
+ goto report_result;
+ }
+ if (!!results[FM10K_TEST_MSG_S32]) {
+ err = fm10k_tlv_attr_get_s32(results[FM10K_TEST_MSG_S32],
+ &result_s32);
+ if (!err && test_s32 != result_s32)
+ err = FM10K_ERR_INVALID_VALUE;
+ if (err)
+ goto report_result;
+ }
+ if (!!results[FM10K_TEST_MSG_S64]) {
+ err = fm10k_tlv_attr_get_s64(results[FM10K_TEST_MSG_S64],
+ &result_s64);
+ if (!err && test_s64 != result_s64)
+ err = FM10K_ERR_INVALID_VALUE;
+ if (err)
+ goto report_result;
+ }
+ if (!!results[FM10K_TEST_MSG_LE_STRUCT]) {
+ err = fm10k_tlv_attr_get_le_struct(
+ results[FM10K_TEST_MSG_LE_STRUCT],
+ result_le,
+ sizeof(result_le));
+ if (!err && memcmp(test_le, result_le, sizeof(test_le)))
+ err = FM10K_ERR_INVALID_VALUE;
+ if (err)
+ goto report_result;
+ }
+
+ if (!!results[FM10K_TEST_MSG_NESTED]) {
+ /* clear any pointers */
+ memset(nest_results, 0, sizeof(nest_results));
+
+ /* parse the nested attributes into the nest results list */
+ err = fm10k_tlv_attr_parse(results[FM10K_TEST_MSG_NESTED],
+ nest_results,
+ fm10k_tlv_msg_test_attr);
+ if (err)
+ goto report_result;
+
+ /* loop back through to the start */
+ results = nest_results;
+ goto parse_nested;
+ }
+
+report_result:
+ /* generate reply with test result */
+ fm10k_tlv_msg_init(reply, FM10K_TLV_MSG_ID_TEST);
+ fm10k_tlv_attr_put_s32(reply, FM10K_TEST_MSG_RESULT, err);
+
+ /* load onto outgoing mailbox */
+ return mbx->ops.enqueue_tx(hw, mbx, reply);
+}
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_tlv.h b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.h
index 6d22db6..7e045e8 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_tlv.h
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.h
@@ -21,6 +21,9 @@
#ifndef _FM10K_TLV_H_
#define _FM10K_TLV_H_
+/* forward declaration */
+struct fm10k_msg_data;
+
#include "fm10k_type.h"
/* Message / Argument header format
@@ -93,6 +96,15 @@ struct fm10k_tlv_attr {
#define FM10K_TLV_ATTR_NESTED(id) { id, FM10K_TLV_NESTED }
#define FM10K_TLV_ATTR_LAST { FM10K_TLV_ERROR }
+struct fm10k_msg_data {
+ unsigned int id;
+ const struct fm10k_tlv_attr *attr;
+ s32 (*func)(struct fm10k_hw *, u32 **,
+ struct fm10k_mbx_info *);
+};
+
+#define FM10K_MSG_HANDLER(id, attr, func) { id, attr, func }
+
s32 fm10k_tlv_msg_init(u32 *, u16);
s32 fm10k_tlv_attr_put_null_string(u32 *, u16, const unsigned char *);
s32 fm10k_tlv_attr_get_null_string(u32 *, unsigned char *);
@@ -138,4 +150,37 @@ s32 fm10k_tlv_attr_get_le_struct(u32 *, void *, u32);
u32 *fm10k_tlv_attr_nest_start(u32 *, u16);
s32 fm10k_tlv_attr_nest_stop(u32 *);
s32 fm10k_tlv_attr_parse(u32 *, u32 **, const struct fm10k_tlv_attr *);
+s32 fm10k_tlv_msg_parse(struct fm10k_hw *, u32 *, struct fm10k_mbx_info *,
+ const struct fm10k_msg_data *);
+s32 fm10k_tlv_msg_error(struct fm10k_hw *hw, u32 **results,
+ struct fm10k_mbx_info *);
+
+#define FM10K_TLV_MSG_ID_TEST 0
+
+enum fm10k_tlv_test_attr_id {
+ FM10K_TEST_MSG_UNSET,
+ FM10K_TEST_MSG_STRING,
+ FM10K_TEST_MSG_MAC_ADDR,
+ FM10K_TEST_MSG_U8,
+ FM10K_TEST_MSG_U16,
+ FM10K_TEST_MSG_U32,
+ FM10K_TEST_MSG_U64,
+ FM10K_TEST_MSG_S8,
+ FM10K_TEST_MSG_S16,
+ FM10K_TEST_MSG_S32,
+ FM10K_TEST_MSG_S64,
+ FM10K_TEST_MSG_LE_STRUCT,
+ FM10K_TEST_MSG_NESTED,
+ FM10K_TEST_MSG_RESULT,
+ FM10K_TEST_MSG_MAX
+};
+
+extern const struct fm10k_tlv_attr fm10k_tlv_msg_test_attr[];
+void fm10k_tlv_msg_test_create(u32 *, u32);
+s32 fm10k_tlv_msg_test(struct fm10k_hw *, u32 **, struct fm10k_mbx_info *);
+
+#define FM10K_TLV_MSG_TEST_HANDLER(func) \
+ FM10K_MSG_HANDLER(FM10K_TLV_MSG_ID_TEST, fm10k_tlv_msg_test_attr, func)
+#define FM10K_TLV_MSG_ERROR_HANDLER(func) \
+ FM10K_MSG_HANDLER(FM10K_TLV_ERROR, NULL, func)
#endif /* _FM10K_MSG_H_ */
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_type.h b/drivers/net/ethernet/intel/fm10k/fm10k_type.h
index 1606805..eda0c7c 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_type.h
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_type.h
@@ -28,6 +28,8 @@ struct fm10k_hw;
#include <asm/byteorder.h>
#include <linux/etherdevice.h>
+#include "fm10k_mbx.h"
+
#define FM10K_DEV_ID_PF 0x15A4
#define FM10K_DEV_ID_VF 0x15A5
@@ -573,6 +575,7 @@ struct fm10k_hw {
struct fm10k_mac_info mac;
struct fm10k_bus_info bus;
struct fm10k_bus_info bus_caps;
+ struct fm10k_mbx_info mbx;
struct fm10k_swapi_info swapi;
u16 device_id;
u16 vendor_id;
^ permalink raw reply related
* [net-next PATCH 04/29] fm10k: Add support for basic interaction with hardware
From: Alexander Duyck @ 2014-09-18 22:36 UTC (permalink / raw)
To: davem
Cc: nhorman, netdev, john.fastabend, matthew.vick, jeffrey.t.kirsher,
sassmann
In-Reply-To: <20140918223242.10373.27403.stgit@ahduyck-bv4.jf.intel.com>
This patch adds the basic read/write operations for accessing the hardware.
In addition to read read functionality the read functions also provide
surprise remove detection in the event that the device either loses power
or is removed.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/fm10k/fm10k.h | 9 ++++-
drivers/net/ethernet/intel/fm10k/fm10k_common.h | 44 +++++++++++++++++++++++
drivers/net/ethernet/intel/fm10k/fm10k_pci.c | 30 ++++++++++++++++
3 files changed, 82 insertions(+), 1 deletion(-)
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_common.h
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k.h b/drivers/net/ethernet/intel/fm10k/fm10k.h
index 2e27bd9..3da3a99 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k.h
+++ b/drivers/net/ethernet/intel/fm10k/fm10k.h
@@ -27,7 +27,14 @@
#include <linux/if_vlan.h>
#include <linux/pci.h>
-#include "fm10k_type.h"
+#include "fm10k_common.h"
+
+struct fm10k_intfc {
+ struct pci_dev *pdev;
+
+ struct fm10k_hw hw;
+ u32 __iomem *uc_addr;
+};
/* main */
extern char fm10k_driver_name[];
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_common.h b/drivers/net/ethernet/intel/fm10k/fm10k_common.h
new file mode 100644
index 0000000..2b214a6
--- /dev/null
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_common.h
@@ -0,0 +1,44 @@
+/* Intel Ethernet Switch Host Interface Driver
+ * Copyright(c) 2013 - 2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
+
+#ifndef _FM10K_COMMON_H_
+#define _FM10K_COMMON_H_
+
+#include "fm10k_type.h"
+
+#define FM10K_REMOVED(hw_addr) unlikely(!(hw_addr))
+
+/* PCI configuration read */
+u16 fm10k_read_pci_cfg_word(struct fm10k_hw *hw, u32 reg);
+
+/* read operations, indexed using DWORDS */
+u32 fm10k_read_reg(struct fm10k_hw *hw, int reg);
+
+/* write operations, indexed using DWORDS */
+#define fm10k_write_reg(hw, reg, val) \
+do { \
+ u32 __iomem *hw_addr = ACCESS_ONCE((hw)->hw_addr); \
+ if (!FM10K_REMOVED(hw_addr)) \
+ writel((val), &hw_addr[(reg)]); \
+} while (0)
+
+/* read ctrl register which has no clear on read fields as PCIe flush */
+#define fm10k_write_flush(hw) fm10k_read_reg((hw), FM10K_CTRL)
+#endif /* _FM10K_COMMON_H_ */
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
index 12089ce..6d5e33c 100644
--- a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
@@ -38,6 +38,36 @@ static const struct pci_device_id fm10k_pci_tbl[] = {
};
MODULE_DEVICE_TABLE(pci, fm10k_pci_tbl);
+u16 fm10k_read_pci_cfg_word(struct fm10k_hw *hw, u32 reg)
+{
+ struct fm10k_intfc *interface = hw->back;
+ u16 value = 0;
+
+ if (FM10K_REMOVED(hw->hw_addr))
+ return ~value;
+
+ pci_read_config_word(interface->pdev, reg, &value);
+ if (value == 0xFFFF)
+ fm10k_write_flush(hw);
+
+ return value;
+}
+
+u32 fm10k_read_reg(struct fm10k_hw *hw, int reg)
+{
+ u32 __iomem *hw_addr = ACCESS_ONCE(hw->hw_addr);
+ u32 value = 0;
+
+ if (FM10K_REMOVED(hw_addr))
+ return ~value;
+
+ value = readl(&hw_addr[reg]);
+ if (!(~value) && (!reg || !(~readl(hw_addr))))
+ hw->hw_addr = NULL;
+
+ return value;
+}
+
/**
* fm10k_probe - Device Initialization Routine
* @pdev: PCI device information struct
^ permalink raw reply related
* [net-next PATCH 01/29] fm10k: Add skeletal frame for Intel(R) FM10000 Ethernet Switch Host Interface Driver
From: Alexander Duyck @ 2014-09-18 22:35 UTC (permalink / raw)
To: davem
Cc: nhorman, netdev, john.fastabend, matthew.vick, jeffrey.t.kirsher,
sassmann
In-Reply-To: <20140918223242.10373.27403.stgit@ahduyck-bv4.jf.intel.com>
This patch adds the beginning framework onto which I am going to add the
fm10k driver which supports the Intel(R) FM10000 Ethernet Switch Host
Interface.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/Kconfig | 19 +++
drivers/net/ethernet/intel/Makefile | 1
drivers/net/ethernet/intel/fm10k/Makefile | 30 +++++
drivers/net/ethernet/intel/fm10k/fm10k.h | 39 +++++++
drivers/net/ethernet/intel/fm10k/fm10k_main.c | 67 ++++++++++++
drivers/net/ethernet/intel/fm10k/fm10k_pci.c | 145 +++++++++++++++++++++++++
drivers/net/ethernet/intel/fm10k/fm10k_type.h | 27 +++++
7 files changed, 328 insertions(+)
create mode 100644 drivers/net/ethernet/intel/fm10k/Makefile
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k.h
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_main.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_pci.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_type.h
diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig
index bb9f0ba..6a6d5ee 100644
--- a/drivers/net/ethernet/intel/Kconfig
+++ b/drivers/net/ethernet/intel/Kconfig
@@ -300,4 +300,23 @@ config I40EVF
will be called i40evf. MSI-X interrupt support is required
for this driver to work correctly.
+config FM10K
+ tristate "Intel(R) FM10000 Ethernet Switch Host Interface Support"
+ default n
+ depends on PCI_MSI
+ ---help---
+ This driver supports Intel(R) FM10000 Ethernet Switch Host
+ Interface. For more information on how to identify your adapter,
+ go to the Adapter & Driver ID Guide at:
+
+ <http://support.intel.com/support/network/sb/CS-008441.htm>
+
+ For general information and support, go to the Intel support
+ website at:
+
+ <http://support.intel.com>
+
+ To compile this driver as a module, choose M here. The module
+ will be called fm10k. MSI-X interrupt support is required
+
endif # NET_VENDOR_INTEL
diff --git a/drivers/net/ethernet/intel/Makefile b/drivers/net/ethernet/intel/Makefile
index cdbbca8..5ea764d 100644
--- a/drivers/net/ethernet/intel/Makefile
+++ b/drivers/net/ethernet/intel/Makefile
@@ -12,3 +12,4 @@ obj-$(CONFIG_IXGBEVF) += ixgbevf/
obj-$(CONFIG_I40E) += i40e/
obj-$(CONFIG_IXGB) += ixgb/
obj-$(CONFIG_I40EVF) += i40evf/
+obj-$(CONFIG_FM10K) += fm10k/
diff --git a/drivers/net/ethernet/intel/fm10k/Makefile b/drivers/net/ethernet/intel/fm10k/Makefile
new file mode 100644
index 0000000..ddc0eb7
--- /dev/null
+++ b/drivers/net/ethernet/intel/fm10k/Makefile
@@ -0,0 +1,30 @@
+################################################################################
+#
+# Intel Ethernet Switch Host Interface Driver
+# Copyright(c) 2013 - 2014 Intel Corporation.
+#
+# This program is free software; you can redistribute it and/or modify it
+# under the terms and conditions of the GNU General Public License,
+# version 2, as published by the Free Software Foundation.
+#
+# This program is distributed in the hope it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+# more details.
+#
+# The full GNU General Public License is included in this distribution in
+# the file called "COPYING".
+#
+# Contact Information:
+# e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+# Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+#
+################################################################################
+
+#
+# Makefile for the Intel(R) FM10000 Ethernet Switch Host Interface driver
+#
+
+obj-$(CONFIG_FM10K) += fm10k.o
+
+fm10k-objs := fm10k_main.o fm10k_pci.o
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k.h b/drivers/net/ethernet/intel/fm10k/fm10k.h
new file mode 100644
index 0000000..2e27bd9
--- /dev/null
+++ b/drivers/net/ethernet/intel/fm10k/fm10k.h
@@ -0,0 +1,39 @@
+/* Intel Ethernet Switch Host Interface Driver
+ * Copyright(c) 2013 - 2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
+
+#ifndef _FM10K_H_
+#define _FM10K_H_
+
+#include <linux/types.h>
+#include <linux/etherdevice.h>
+#include <linux/rtnetlink.h>
+#include <linux/if_vlan.h>
+#include <linux/pci.h>
+
+#include "fm10k_type.h"
+
+/* main */
+extern char fm10k_driver_name[];
+extern const char fm10k_driver_version[];
+
+/* PCI */
+int fm10k_register_pci_driver(void);
+void fm10k_unregister_pci_driver(void);
+#endif /* _FM10K_H_ */
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_main.c b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
new file mode 100644
index 0000000..a7822b8
--- /dev/null
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_main.c
@@ -0,0 +1,67 @@
+/* Intel Ethernet Switch Host Interface Driver
+ * Copyright(c) 2013 - 2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
+
+#include <linux/types.h>
+#include <linux/module.h>
+#include <net/ipv6.h>
+#include <net/ip.h>
+#include <net/tcp.h>
+#include <linux/if_macvlan.h>
+
+#include "fm10k.h"
+
+#define DRV_VERSION "0.12.2-k"
+const char fm10k_driver_version[] = DRV_VERSION;
+char fm10k_driver_name[] = "fm10k";
+static const char fm10k_driver_string[] =
+ "Intel(R) Ethernet Switch Host Interface Driver";
+static const char fm10k_copyright[] =
+ "Copyright (c) 2013 Intel Corporation.";
+
+MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
+MODULE_DESCRIPTION("Intel(R) Ethernet Switch Host Interface Driver");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(DRV_VERSION);
+
+/** fm10k_init_module - Driver Registration Routine
+ *
+ * fm10k_init_module is the first routine called when the driver is
+ * loaded. All it does is register with the PCI subsystem.
+ **/
+static int __init fm10k_init_module(void)
+{
+ pr_info("%s - version %s\n", fm10k_driver_string, fm10k_driver_version);
+ pr_info("%s\n", fm10k_copyright);
+
+ return fm10k_register_pci_driver();
+}
+module_init(fm10k_init_module);
+
+/**
+ * fm10k_exit_module - Driver Exit Cleanup Routine
+ *
+ * fm10k_exit_module is called just before the driver is removed
+ * from memory.
+ **/
+static void __exit fm10k_exit_module(void)
+{
+ fm10k_unregister_pci_driver();
+}
+module_exit(fm10k_exit_module);
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_pci.c b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
new file mode 100644
index 0000000..12089ce
--- /dev/null
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_pci.c
@@ -0,0 +1,145 @@
+/* Intel Ethernet Switch Host Interface Driver
+ * Copyright(c) 2013 - 2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
+
+#include <linux/module.h>
+
+#include "fm10k.h"
+
+/**
+ * fm10k_pci_tbl - PCI Device ID Table
+ *
+ * Wildcard entries (PCI_ANY_ID) should come last
+ * Last entry must be all 0s
+ *
+ * { Vendor ID, Device ID, SubVendor ID, SubDevice ID,
+ * Class, Class Mask, private data (not used) }
+ */
+static const struct pci_device_id fm10k_pci_tbl[] = {
+ { PCI_VDEVICE(INTEL, FM10K_DEV_ID_PF) },
+ /* required last entry */
+ { 0, }
+};
+MODULE_DEVICE_TABLE(pci, fm10k_pci_tbl);
+
+/**
+ * fm10k_probe - Device Initialization Routine
+ * @pdev: PCI device information struct
+ * @ent: entry in fm10k_pci_tbl
+ *
+ * Returns 0 on success, negative on failure
+ *
+ * fm10k_probe initializes an interface identified by a pci_dev structure.
+ * The OS initialization, configuring of the interface private structure,
+ * and a hardware reset occur.
+ **/
+static int fm10k_probe(struct pci_dev *pdev,
+ const struct pci_device_id *ent)
+{
+ int err;
+ u64 dma_mask;
+
+ err = pci_enable_device_mem(pdev);
+ if (err)
+ return err;
+
+ /* By default fm10k only supports a 48 bit DMA mask */
+ dma_mask = DMA_BIT_MASK(48) | dma_get_required_mask(&pdev->dev);
+
+ if ((dma_mask <= DMA_BIT_MASK(32)) ||
+ dma_set_mask_and_coherent(&pdev->dev, dma_mask)) {
+ dma_mask &= DMA_BIT_MASK(32);
+
+ err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+ err = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32));
+ if (err) {
+ err = dma_set_coherent_mask(&pdev->dev,
+ DMA_BIT_MASK(32));
+ if (err) {
+ dev_err(&pdev->dev,
+ "No usable DMA configuration, aborting\n");
+ goto err_dma;
+ }
+ }
+ }
+
+ err = pci_request_selected_regions(pdev,
+ pci_select_bars(pdev,
+ IORESOURCE_MEM),
+ fm10k_driver_name);
+ if (err) {
+ dev_err(&pdev->dev,
+ "pci_request_selected_regions failed 0x%x\n", err);
+ goto err_pci_reg;
+ }
+
+ pci_set_master(pdev);
+ pci_save_state(pdev);
+
+ return 0;
+
+err_pci_reg:
+err_dma:
+ pci_disable_device(pdev);
+ return err;
+}
+
+/**
+ * fm10k_remove - Device Removal Routine
+ * @pdev: PCI device information struct
+ *
+ * fm10k_remove is called by the PCI subsystem to alert the driver
+ * that it should release a PCI device. The could be caused by a
+ * Hot-Plug event, or because the driver is going to be removed from
+ * memory.
+ **/
+static void fm10k_remove(struct pci_dev *pdev)
+{
+ pci_release_selected_regions(pdev,
+ pci_select_bars(pdev, IORESOURCE_MEM));
+
+ pci_disable_device(pdev);
+}
+
+static struct pci_driver fm10k_driver = {
+ .name = fm10k_driver_name,
+ .id_table = fm10k_pci_tbl,
+ .probe = fm10k_probe,
+ .remove = fm10k_remove,
+};
+
+/**
+ * fm10k_register_pci_driver - register driver interface
+ *
+ * This funciton is called on module load in order to register the driver.
+ **/
+int fm10k_register_pci_driver(void)
+{
+ return pci_register_driver(&fm10k_driver);
+}
+
+/**
+ * fm10k_unregister_pci_driver - unregister driver interface
+ *
+ * This funciton is called on module unload in order to remove the driver.
+ **/
+void fm10k_unregister_pci_driver(void)
+{
+ pci_unregister_driver(&fm10k_driver);
+}
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_type.h b/drivers/net/ethernet/intel/fm10k/fm10k_type.h
new file mode 100644
index 0000000..2c06f02
--- /dev/null
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_type.h
@@ -0,0 +1,27 @@
+/* Intel Ethernet Switch Host Interface Driver
+ * Copyright(c) 2013 - 2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
+
+#ifndef _FM10K_TYPE_H_
+#define _FM10K_TYPE_H_
+
+#define FM10K_DEV_ID_PF 0x15A4
+#define FM10K_DEV_ID_VF 0x15A5
+
+#endif /* _FM10K_TYPE_H */
^ permalink raw reply related
* [net-next PATCH 03/29] fm10k: Add support for TLV message parsing and generation
From: Alexander Duyck @ 2014-09-18 22:36 UTC (permalink / raw)
To: davem
Cc: nhorman, netdev, john.fastabend, matthew.vick, jeffrey.t.kirsher,
sassmann
In-Reply-To: <20140918223242.10373.27403.stgit@ahduyck-bv4.jf.intel.com>
This patch adds support for the TVL message formats supported by the PF,
VF, and Switch Management entity.
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
---
drivers/net/ethernet/intel/fm10k/Makefile | 3
drivers/net/ethernet/intel/fm10k/fm10k_tlv.c | 543 ++++++++++++++++++++++++++
drivers/net/ethernet/intel/fm10k/fm10k_tlv.h | 141 +++++++
3 files changed, 686 insertions(+), 1 deletion(-)
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_tlv.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_tlv.h
diff --git a/drivers/net/ethernet/intel/fm10k/Makefile b/drivers/net/ethernet/intel/fm10k/Makefile
index ddc0eb7..1052808 100644
--- a/drivers/net/ethernet/intel/fm10k/Makefile
+++ b/drivers/net/ethernet/intel/fm10k/Makefile
@@ -27,4 +27,5 @@
obj-$(CONFIG_FM10K) += fm10k.o
-fm10k-objs := fm10k_main.o fm10k_pci.o
+fm10k-objs := fm10k_main.o fm10k_pci.o \
+ fm10k_tlv.o
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c
new file mode 100644
index 0000000..f1fc709
--- /dev/null
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.c
@@ -0,0 +1,543 @@
+/* Intel Ethernet Switch Host Interface Driver
+ * Copyright(c) 2013 - 2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
+
+#include "fm10k_tlv.h"
+
+/**
+ * fm10k_tlv_msg_init - Initialize message block for TLV data storage
+ * @msg: Pointer to message block
+ * @msg_id: Message ID indicating message type
+ *
+ * This function return success if provided with a valid message pointer
+ **/
+s32 fm10k_tlv_msg_init(u32 *msg, u16 msg_id)
+{
+ /* verify pointer is not NULL */
+ if (!msg)
+ return FM10K_ERR_PARAM;
+
+ *msg = (FM10K_TLV_FLAGS_MSG << FM10K_TLV_FLAGS_SHIFT) | msg_id;
+
+ return 0;
+}
+
+/**
+ * fm10k_tlv_attr_put_null_string - Place null terminated string on message
+ * @msg: Pointer to message block
+ * @attr_id: Attribute ID
+ * @string: Pointer to string to be stored in attribute
+ *
+ * This function will reorder a string to be CPU endian and store it in
+ * the attribute buffer. It will return success if provided with a valid
+ * pointers.
+ **/
+s32 fm10k_tlv_attr_put_null_string(u32 *msg, u16 attr_id,
+ const unsigned char *string)
+{
+ u32 attr_data = 0, len = 0;
+ u32 *attr;
+
+ /* verify pointers are not NULL */
+ if (!string || !msg)
+ return FM10K_ERR_PARAM;
+
+ attr = &msg[FM10K_TLV_DWORD_LEN(*msg)];
+
+ /* copy string into local variable and then write to msg */
+ do {
+ /* write data to message */
+ if (len && !(len % 4)) {
+ attr[len / 4] = attr_data;
+ attr_data = 0;
+ }
+
+ /* record character to offset location */
+ attr_data |= (u32)(*string) << (8 * (len % 4));
+ len++;
+
+ /* test for NULL and then increment */
+ } while (*(string++));
+
+ /* write last piece of data to message */
+ attr[(len + 3) / 4] = attr_data;
+
+ /* record attribute header, update message length */
+ len <<= FM10K_TLV_LEN_SHIFT;
+ attr[0] = len | attr_id;
+
+ /* add header length to length */
+ len += FM10K_TLV_HDR_LEN << FM10K_TLV_LEN_SHIFT;
+ *msg += FM10K_TLV_LEN_ALIGN(len);
+
+ return 0;
+}
+
+/**
+ * fm10k_tlv_attr_get_null_string - Get null terminated string from attribute
+ * @attr: Pointer to attribute
+ * @string: Pointer to location of destination string
+ *
+ * This function pulls the string back out of the attribute and will place
+ * it in the array pointed by by string. It will return success if provided
+ * with a valid pointers.
+ **/
+s32 fm10k_tlv_attr_get_null_string(u32 *attr, unsigned char *string)
+{
+ u32 len;
+
+ /* verify pointers are not NULL */
+ if (!string || !attr)
+ return FM10K_ERR_PARAM;
+
+ len = *attr >> FM10K_TLV_LEN_SHIFT;
+ attr++;
+
+ while (len--)
+ string[len] = (u8)(attr[len / 4] >> (8 * (len % 4)));
+
+ return 0;
+}
+
+/**
+ * fm10k_tlv_attr_put_mac_vlan - Store MAC/VLAN attribute in message
+ * @msg: Pointer to message block
+ * @attr_id: Attribute ID
+ * @mac_addr: MAC address to be stored
+ *
+ * This function will reorder a MAC address to be CPU endian and store it
+ * in the attribute buffer. It will return success if provided with a
+ * valid pointers.
+ **/
+s32 fm10k_tlv_attr_put_mac_vlan(u32 *msg, u16 attr_id,
+ const u8 *mac_addr, u16 vlan)
+{
+ u32 len = ETH_ALEN << FM10K_TLV_LEN_SHIFT;
+ u32 *attr;
+
+ /* verify pointers are not NULL */
+ if (!msg || !mac_addr)
+ return FM10K_ERR_PARAM;
+
+ attr = &msg[FM10K_TLV_DWORD_LEN(*msg)];
+
+ /* record attribute header, update message length */
+ attr[0] = len | attr_id;
+
+ /* copy value into local variable and then write to msg */
+ attr[1] = le32_to_cpu(*(const __le32 *)&mac_addr[0]);
+ attr[2] = le16_to_cpu(*(const __le16 *)&mac_addr[4]);
+ attr[2] |= (u32)vlan << 16;
+
+ /* add header length to length */
+ len += FM10K_TLV_HDR_LEN << FM10K_TLV_LEN_SHIFT;
+ *msg += FM10K_TLV_LEN_ALIGN(len);
+
+ return 0;
+}
+
+/**
+ * fm10k_tlv_attr_get_mac_vlan - Get MAC/VLAN stored in attribute
+ * @attr: Pointer to attribute
+ * @attr_id: Attribute ID
+ * @mac_addr: location of buffer to store MAC address
+ *
+ * This function pulls the MAC address back out of the attribute and will
+ * place it in the array pointed by by mac_addr. It will return success
+ * if provided with a valid pointers.
+ **/
+s32 fm10k_tlv_attr_get_mac_vlan(u32 *attr, u8 *mac_addr, u16 *vlan)
+{
+ /* verify pointers are not NULL */
+ if (!mac_addr || !attr)
+ return FM10K_ERR_PARAM;
+
+ *(__le32 *)&mac_addr[0] = cpu_to_le32(attr[1]);
+ *(__le16 *)&mac_addr[4] = cpu_to_le16((u16)(attr[2]));
+ *vlan = (u16)(attr[2] >> 16);
+
+ return 0;
+}
+
+/**
+ * fm10k_tlv_attr_put_bool - Add header indicating value "true"
+ * @msg: Pointer to message block
+ * @attr_id: Attribute ID
+ *
+ * This function will simply add an attribute header, the fact
+ * that the header is here means the attribute value is true, else
+ * it is false. The function will return success if provided with a
+ * valid pointers.
+ **/
+s32 fm10k_tlv_attr_put_bool(u32 *msg, u16 attr_id)
+{
+ /* verify pointers are not NULL */
+ if (!msg)
+ return FM10K_ERR_PARAM;
+
+ /* record attribute header */
+ msg[FM10K_TLV_DWORD_LEN(*msg)] = attr_id;
+
+ /* add header length to length */
+ *msg += FM10K_TLV_HDR_LEN << FM10K_TLV_LEN_SHIFT;
+
+ return 0;
+}
+
+/**
+ * fm10k_tlv_attr_put_value - Store integer value attribute in message
+ * @msg: Pointer to message block
+ * @attr_id: Attribute ID
+ * @value: Value to be written
+ * @len: Size of value
+ *
+ * This function will place an integer value of up to 8 bytes in size
+ * in a message attribute. The function will return success provided
+ * that msg is a valid pointer, and len is 1, 2, 4, or 8.
+ **/
+s32 fm10k_tlv_attr_put_value(u32 *msg, u16 attr_id, s64 value, u32 len)
+{
+ u32 *attr;
+
+ /* verify non-null msg and len is 1, 2, 4, or 8 */
+ if (!msg || !len || len > 8 || (len & (len - 1)))
+ return FM10K_ERR_PARAM;
+
+ attr = &msg[FM10K_TLV_DWORD_LEN(*msg)];
+
+ if (len < 4) {
+ attr[1] = (u32)value & ((0x1ul << (8 * len)) - 1);
+ } else {
+ attr[1] = (u32)value;
+ if (len > 4)
+ attr[2] = (u32)(value >> 32);
+ }
+
+ /* record attribute header, update message length */
+ len <<= FM10K_TLV_LEN_SHIFT;
+ attr[0] = len | attr_id;
+
+ /* add header length to length */
+ len += FM10K_TLV_HDR_LEN << FM10K_TLV_LEN_SHIFT;
+ *msg += FM10K_TLV_LEN_ALIGN(len);
+
+ return 0;
+}
+
+/**
+ * fm10k_tlv_attr_get_value - Get integer value stored in attribute
+ * @attr: Pointer to attribute
+ * @value: Pointer to destination buffer
+ * @len: Size of value
+ *
+ * This function will place an integer value of up to 8 bytes in size
+ * in the offset pointed to by value. The function will return success
+ * provided that pointers are valid and the len value matches the
+ * attribute length.
+ **/
+s32 fm10k_tlv_attr_get_value(u32 *attr, void *value, u32 len)
+{
+ /* verify pointers are not NULL */
+ if (!attr || !value)
+ return FM10K_ERR_PARAM;
+
+ if ((*attr >> FM10K_TLV_LEN_SHIFT) != len)
+ return FM10K_ERR_PARAM;
+
+ if (len == 8)
+ *(u64 *)value = ((u64)attr[2] << 32) | attr[1];
+ else if (len == 4)
+ *(u32 *)value = attr[1];
+ else if (len == 2)
+ *(u16 *)value = (u16)attr[1];
+ else
+ *(u8 *)value = (u8)attr[1];
+
+ return 0;
+}
+
+/**
+ * fm10k_tlv_attr_put_le_struct - Store little endian structure in message
+ * @msg: Pointer to message block
+ * @attr_id: Attribute ID
+ * @le_struct: Pointer to structure to be written
+ * @len: Size of le_struct
+ *
+ * This function will place a little endian structure value in a message
+ * attribute. The function will return success provided that all pointers
+ * are valid and length is a non-zero multiple of 4.
+ **/
+s32 fm10k_tlv_attr_put_le_struct(u32 *msg, u16 attr_id,
+ const void *le_struct, u32 len)
+{
+ const __le32 *le32_ptr = (const __le32 *)le_struct;
+ u32 *attr;
+ u32 i;
+
+ /* verify non-null msg and len is in 32 bit words */
+ if (!msg || !len || (len % 4))
+ return FM10K_ERR_PARAM;
+
+ attr = &msg[FM10K_TLV_DWORD_LEN(*msg)];
+
+ /* copy le32 structure into host byte order at 32b boundaries */
+ for (i = 0; i < (len / 4); i++)
+ attr[i + 1] = le32_to_cpu(le32_ptr[i]);
+
+ /* record attribute header, update message length */
+ len <<= FM10K_TLV_LEN_SHIFT;
+ attr[0] = len | attr_id;
+
+ /* add header length to length */
+ len += FM10K_TLV_HDR_LEN << FM10K_TLV_LEN_SHIFT;
+ *msg += FM10K_TLV_LEN_ALIGN(len);
+
+ return 0;
+}
+
+/**
+ * fm10k_tlv_attr_get_le_struct - Get little endian struct form attribute
+ * @attr: Pointer to attribute
+ * @le_struct: Pointer to structure to be written
+ * @len: Size of structure
+ *
+ * This function will place a little endian structure in the buffer
+ * pointed to by le_struct. The function will return success
+ * provided that pointers are valid and the len value matches the
+ * attribute length.
+ **/
+s32 fm10k_tlv_attr_get_le_struct(u32 *attr, void *le_struct, u32 len)
+{
+ __le32 *le32_ptr = (__le32 *)le_struct;
+ u32 i;
+
+ /* verify pointers are not NULL */
+ if (!le_struct || !attr)
+ return FM10K_ERR_PARAM;
+
+ if ((*attr >> FM10K_TLV_LEN_SHIFT) != len)
+ return FM10K_ERR_PARAM;
+
+ attr++;
+
+ for (i = 0; len; i++, len -= 4)
+ le32_ptr[i] = cpu_to_le32(attr[i]);
+
+ return 0;
+}
+
+/**
+ * fm10k_tlv_attr_nest_start - Start a set of nested attributes
+ * @msg: Pointer to message block
+ * @attr_id: Attribute ID
+ *
+ * This function will mark off a new nested region for encapsulating
+ * a given set of attributes. The idea is if you wish to place a secondary
+ * structure within the message this mechanism allows for that. The
+ * function will return NULL on failure, and a pointer to the start
+ * of the nested attributes on success.
+ **/
+u32 *fm10k_tlv_attr_nest_start(u32 *msg, u16 attr_id)
+{
+ u32 *attr;
+
+ /* verify pointer is not NULL */
+ if (!msg)
+ return NULL;
+
+ attr = &msg[FM10K_TLV_DWORD_LEN(*msg)];
+
+ attr[0] = attr_id;
+
+ /* return pointer to nest header */
+ return attr;
+}
+
+/**
+ * fm10k_tlv_attr_nest_start - Start a set of nested attributes
+ * @msg: Pointer to message block
+ *
+ * This function closes off an existing set of nested attributes. The
+ * message pointer should be pointing to the parent of the nest. So in
+ * the case of a nest within the nest this would be the outer nest pointer.
+ * This function will return success provided all pointers are valid.
+ **/
+s32 fm10k_tlv_attr_nest_stop(u32 *msg)
+{
+ u32 *attr;
+ u32 len;
+
+ /* verify pointer is not NULL */
+ if (!msg)
+ return FM10K_ERR_PARAM;
+
+ /* locate the nested header and retrieve its length */
+ attr = &msg[FM10K_TLV_DWORD_LEN(*msg)];
+ len = (attr[0] >> FM10K_TLV_LEN_SHIFT) << FM10K_TLV_LEN_SHIFT;
+
+ /* only include nest if data was added to it */
+ if (len) {
+ len += FM10K_TLV_HDR_LEN << FM10K_TLV_LEN_SHIFT;
+ *msg += len;
+ }
+
+ return 0;
+}
+
+/**
+ * fm10k_tlv_attr_validate - Validate attribute metadata
+ * @attr: Pointer to attribute
+ * @tlv_attr: Type and length info for attribute
+ *
+ * This function does some basic validation of the input TLV. It
+ * verifies the length, and in the case of null terminated strings
+ * it verifies that the last byte is null. The function will
+ * return FM10K_ERR_PARAM if any attribute is malformed, otherwise
+ * it returns 0.
+ **/
+static s32 fm10k_tlv_attr_validate(u32 *attr,
+ const struct fm10k_tlv_attr *tlv_attr)
+{
+ u32 attr_id = *attr & FM10K_TLV_ID_MASK;
+ u16 len = *attr >> FM10K_TLV_LEN_SHIFT;
+
+ /* verify this is an attribute and not a message */
+ if (*attr & (FM10K_TLV_FLAGS_MSG << FM10K_TLV_FLAGS_SHIFT))
+ return FM10K_ERR_PARAM;
+
+ /* search through the list of attributes to find a matching ID */
+ while (tlv_attr->id < attr_id)
+ tlv_attr++;
+
+ /* if didn't find a match then we should exit */
+ if (tlv_attr->id != attr_id)
+ return FM10K_NOT_IMPLEMENTED;
+
+ /* move to start of attribute data */
+ attr++;
+
+ switch (tlv_attr->type) {
+ case FM10K_TLV_NULL_STRING:
+ if (!len ||
+ (attr[(len - 1) / 4] & (0xFF << (8 * ((len - 1) % 4)))))
+ return FM10K_ERR_PARAM;
+ if (len > tlv_attr->len)
+ return FM10K_ERR_PARAM;
+ break;
+ case FM10K_TLV_MAC_ADDR:
+ if (len != ETH_ALEN)
+ return FM10K_ERR_PARAM;
+ break;
+ case FM10K_TLV_BOOL:
+ if (len)
+ return FM10K_ERR_PARAM;
+ break;
+ case FM10K_TLV_UNSIGNED:
+ case FM10K_TLV_SIGNED:
+ if (len != tlv_attr->len)
+ return FM10K_ERR_PARAM;
+ break;
+ case FM10K_TLV_LE_STRUCT:
+ /* struct must be 4 byte aligned */
+ if ((len % 4) || len != tlv_attr->len)
+ return FM10K_ERR_PARAM;
+ break;
+ case FM10K_TLV_NESTED:
+ /* nested attributes must be 4 byte aligned */
+ if (len % 4)
+ return FM10K_ERR_PARAM;
+ break;
+ default:
+ /* attribute id is mapped to bad value */
+ return FM10K_ERR_PARAM;
+ }
+
+ return 0;
+}
+
+/**
+ * fm10k_tlv_attr_parse - Parses stream of attribute data
+ * @attr: Pointer to attribute list
+ * @results: Pointer array to store pointers to attributes
+ * @tlv_attr: Type and length info for attributes
+ *
+ * This function validates a stream of attributes and parses them
+ * up into an array of pointers stored in results. The function will
+ * return FM10K_ERR_PARAM on any input or message error,
+ * FM10K_NOT_IMPLEMENTED for any attribute that is outside of the array
+ * and 0 on success.
+ **/
+s32 fm10k_tlv_attr_parse(u32 *attr, u32 **results,
+ const struct fm10k_tlv_attr *tlv_attr)
+{
+ u32 i, attr_id, offset = 0;
+ s32 err = 0;
+ u16 len;
+
+ /* verify pointers are not NULL */
+ if (!attr || !results)
+ return FM10K_ERR_PARAM;
+
+ /* initialize results to NULL */
+ for (i = 0; i < FM10K_TLV_RESULTS_MAX; i++)
+ results[i] = NULL;
+
+ /* pull length from the message header */
+ len = *attr >> FM10K_TLV_LEN_SHIFT;
+
+ /* no attributes to parse if there is no length */
+ if (!len)
+ return 0;
+
+ /* no attributes to parse, just raw data, message becomes attribute */
+ if (!tlv_attr) {
+ results[0] = attr;
+ return 0;
+ }
+
+ /* move to start of attribute data */
+ attr++;
+
+ /* run through list parsing all attributes */
+ while (offset < len) {
+ attr_id = *attr & FM10K_TLV_ID_MASK;
+
+ if (attr_id < FM10K_TLV_RESULTS_MAX)
+ err = fm10k_tlv_attr_validate(attr, tlv_attr);
+ else
+ err = FM10K_NOT_IMPLEMENTED;
+
+ if (err < 0)
+ return err;
+ if (!err)
+ results[attr_id] = attr;
+
+ /* update offset */
+ offset += FM10K_TLV_DWORD_LEN(*attr) * 4;
+
+ /* move to next attribute */
+ attr = &attr[FM10K_TLV_DWORD_LEN(*attr)];
+ }
+
+ /* we should find ourselves at the end of the list */
+ if (offset != len)
+ return FM10K_ERR_PARAM;
+
+ return 0;
+}
diff --git a/drivers/net/ethernet/intel/fm10k/fm10k_tlv.h b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.h
new file mode 100644
index 0000000..6d22db6
--- /dev/null
+++ b/drivers/net/ethernet/intel/fm10k/fm10k_tlv.h
@@ -0,0 +1,141 @@
+/* Intel Ethernet Switch Host Interface Driver
+ * Copyright(c) 2013 - 2014 Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * The full GNU General Public License is included in this distribution in
+ * the file called "COPYING".
+ *
+ * Contact Information:
+ * e1000-devel Mailing List <e1000-devel@lists.sourceforge.net>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ */
+
+#ifndef _FM10K_TLV_H_
+#define _FM10K_TLV_H_
+
+#include "fm10k_type.h"
+
+/* Message / Argument header format
+ * 3 2 1 0
+ * 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0 9 8 7 6 5 4 3 2 1 0
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ * | Length | Flags | Type / ID |
+ * +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
+ *
+ * The message header format described here is used for messages that are
+ * passed between the PF and the VF. To allow for messages larger then
+ * mailbox size we will provide a message with the above header and it
+ * will be segmented and transported to the mailbox to the other side where
+ * it is reassembled. It contains the following fields:
+ * Len: Length of the message in bytes excluding the message header
+ * Flags: TBD
+ * Rule: These will be the message/argument types we pass
+ */
+/* message data header */
+#define FM10K_TLV_ID_SHIFT 0
+#define FM10K_TLV_ID_SIZE 16
+#define FM10K_TLV_ID_MASK ((1u << FM10K_TLV_ID_SIZE) - 1)
+#define FM10K_TLV_FLAGS_SHIFT 16
+#define FM10K_TLV_FLAGS_MSG 0x1
+#define FM10K_TLV_FLAGS_SIZE 4
+#define FM10K_TLV_LEN_SHIFT 20
+#define FM10K_TLV_LEN_SIZE 12
+
+#define FM10K_TLV_HDR_LEN 4ul
+#define FM10K_TLV_LEN_ALIGN_MASK \
+ ((FM10K_TLV_HDR_LEN - 1) << FM10K_TLV_LEN_SHIFT)
+#define FM10K_TLV_LEN_ALIGN(tlv) \
+ (((tlv) + FM10K_TLV_LEN_ALIGN_MASK) & ~FM10K_TLV_LEN_ALIGN_MASK)
+#define FM10K_TLV_DWORD_LEN(tlv) \
+ ((u16)((FM10K_TLV_LEN_ALIGN(tlv)) >> (FM10K_TLV_LEN_SHIFT + 2)) + 1)
+
+#define FM10K_TLV_RESULTS_MAX 32
+
+enum fm10k_tlv_type {
+ FM10K_TLV_NULL_STRING,
+ FM10K_TLV_MAC_ADDR,
+ FM10K_TLV_BOOL,
+ FM10K_TLV_UNSIGNED,
+ FM10K_TLV_SIGNED,
+ FM10K_TLV_LE_STRUCT,
+ FM10K_TLV_NESTED,
+ FM10K_TLV_MAX_TYPE
+};
+
+#define FM10K_TLV_ERROR (~0u)
+
+struct fm10k_tlv_attr {
+ unsigned int id;
+ enum fm10k_tlv_type type;
+ u16 len;
+};
+
+#define FM10K_TLV_ATTR_NULL_STRING(id, len) { id, FM10K_TLV_NULL_STRING, len }
+#define FM10K_TLV_ATTR_MAC_ADDR(id) { id, FM10K_TLV_MAC_ADDR, 6 }
+#define FM10K_TLV_ATTR_BOOL(id) { id, FM10K_TLV_BOOL, 0 }
+#define FM10K_TLV_ATTR_U8(id) { id, FM10K_TLV_UNSIGNED, 1 }
+#define FM10K_TLV_ATTR_U16(id) { id, FM10K_TLV_UNSIGNED, 2 }
+#define FM10K_TLV_ATTR_U32(id) { id, FM10K_TLV_UNSIGNED, 4 }
+#define FM10K_TLV_ATTR_U64(id) { id, FM10K_TLV_UNSIGNED, 8 }
+#define FM10K_TLV_ATTR_S8(id) { id, FM10K_TLV_SIGNED, 1 }
+#define FM10K_TLV_ATTR_S16(id) { id, FM10K_TLV_SIGNED, 2 }
+#define FM10K_TLV_ATTR_S32(id) { id, FM10K_TLV_SIGNED, 4 }
+#define FM10K_TLV_ATTR_S64(id) { id, FM10K_TLV_SIGNED, 8 }
+#define FM10K_TLV_ATTR_LE_STRUCT(id, len) { id, FM10K_TLV_LE_STRUCT, len }
+#define FM10K_TLV_ATTR_NESTED(id) { id, FM10K_TLV_NESTED }
+#define FM10K_TLV_ATTR_LAST { FM10K_TLV_ERROR }
+
+s32 fm10k_tlv_msg_init(u32 *, u16);
+s32 fm10k_tlv_attr_put_null_string(u32 *, u16, const unsigned char *);
+s32 fm10k_tlv_attr_get_null_string(u32 *, unsigned char *);
+s32 fm10k_tlv_attr_put_mac_vlan(u32 *, u16, const u8 *, u16);
+s32 fm10k_tlv_attr_get_mac_vlan(u32 *, u8 *, u16 *);
+s32 fm10k_tlv_attr_put_bool(u32 *, u16);
+s32 fm10k_tlv_attr_put_value(u32 *, u16, s64, u32);
+#define fm10k_tlv_attr_put_u8(msg, attr_id, val) \
+ fm10k_tlv_attr_put_value(msg, attr_id, val, 1)
+#define fm10k_tlv_attr_put_u16(msg, attr_id, val) \
+ fm10k_tlv_attr_put_value(msg, attr_id, val, 2)
+#define fm10k_tlv_attr_put_u32(msg, attr_id, val) \
+ fm10k_tlv_attr_put_value(msg, attr_id, val, 4)
+#define fm10k_tlv_attr_put_u64(msg, attr_id, val) \
+ fm10k_tlv_attr_put_value(msg, attr_id, val, 8)
+#define fm10k_tlv_attr_put_s8(msg, attr_id, val) \
+ fm10k_tlv_attr_put_value(msg, attr_id, val, 1)
+#define fm10k_tlv_attr_put_s16(msg, attr_id, val) \
+ fm10k_tlv_attr_put_value(msg, attr_id, val, 2)
+#define fm10k_tlv_attr_put_s32(msg, attr_id, val) \
+ fm10k_tlv_attr_put_value(msg, attr_id, val, 4)
+#define fm10k_tlv_attr_put_s64(msg, attr_id, val) \
+ fm10k_tlv_attr_put_value(msg, attr_id, val, 8)
+s32 fm10k_tlv_attr_get_value(u32 *, void *, u32);
+#define fm10k_tlv_attr_get_u8(attr, ptr) \
+ fm10k_tlv_attr_get_value(attr, ptr, sizeof(u8))
+#define fm10k_tlv_attr_get_u16(attr, ptr) \
+ fm10k_tlv_attr_get_value(attr, ptr, sizeof(u16))
+#define fm10k_tlv_attr_get_u32(attr, ptr) \
+ fm10k_tlv_attr_get_value(attr, ptr, sizeof(u32))
+#define fm10k_tlv_attr_get_u64(attr, ptr) \
+ fm10k_tlv_attr_get_value(attr, ptr, sizeof(u64))
+#define fm10k_tlv_attr_get_s8(attr, ptr) \
+ fm10k_tlv_attr_get_value(attr, ptr, sizeof(s8))
+#define fm10k_tlv_attr_get_s16(attr, ptr) \
+ fm10k_tlv_attr_get_value(attr, ptr, sizeof(s16))
+#define fm10k_tlv_attr_get_s32(attr, ptr) \
+ fm10k_tlv_attr_get_value(attr, ptr, sizeof(s32))
+#define fm10k_tlv_attr_get_s64(attr, ptr) \
+ fm10k_tlv_attr_get_value(attr, ptr, sizeof(s64))
+s32 fm10k_tlv_attr_put_le_struct(u32 *, u16, const void *, u32);
+s32 fm10k_tlv_attr_get_le_struct(u32 *, void *, u32);
+u32 *fm10k_tlv_attr_nest_start(u32 *, u16);
+s32 fm10k_tlv_attr_nest_stop(u32 *);
+s32 fm10k_tlv_attr_parse(u32 *, u32 **, const struct fm10k_tlv_attr *);
+#endif /* _FM10K_MSG_H_ */
^ permalink raw reply related
* [net-next PATCH 00/29] Add support for the Intel FM10000 Ethernet Switch Host Interface
From: Alexander Duyck @ 2014-09-18 22:35 UTC (permalink / raw)
To: davem
Cc: nhorman, netdev, john.fastabend, matthew.vick, jeffrey.t.kirsher,
sassmann
This patch series adds support for the FM10000 Ethernet switch host
interface. The Intel FM10000 Ethernet Switch is a 48-port Ethernet switch
supporting both Ethernet ports and PCI Express host interfaces. The fm10k
driver provides support for the host interface portion of the switch, both
PF and VF.
As the host interfaces are directly connected to the switch this results in
some significant differences versus a standard network driver. For example
there is no PHY or MII on the device. Since packets are delivered directly
from the switch to the host interface these are unnecessary. Otherwise most
of the functionality is very similar to our other network drivers such as
ixgbe or igb. For example we support all the standard network offloads,
jumbo frames, SR-IOV (64 VFS), PTP, and some VXLAN and NVGRE offloads.
---
Alexander Duyck (29):
fm10k: Add skeletal frame for Intel(R) FM10000 Ethernet Switch Host Interface Driver
fm10k: Add register defines and basic structures
fm10k: Add support for TLV message parsing and generation
fm10k: Add support for basic interaction with hardware
fm10k: Add support for mailbox
fm10k: Implement PF <-> SM mailbox operations
fm10k: Add support for PF
fm10k: Add support for configuring PF interface
fm10k: Add netdev
fm10k: Add support for L2 filtering
fm10k: Add support for ndo_open/stop
fm10k: Add interrupt support
fm10k: add support for Tx/Rx rings
fm10k: Add service task to handle delayed events
fm10k: Add Tx/Rx hardware ring bring-up/tear-down
fm10k: Add transmit and receive fastpath and interrupt handlers
fm10k: Add ethtool support
fm10k: Add support for PCI power management and error handling
fm10k: Add support for multiple queues
fm10k: Add support for netdev offloads
fm10k: Add support for MACVLAN acceleration
fm10k: Add support for PF <-> VF mailbox
fm10k: Add support for VF
fm10k: Add support for SR-IOV to PF core files
fm10k: Add support for SR-IOV to driver
fm10k: Add support for IEEE DCBx
fm10k: Add support for debugfs
fm10k: Add support for ptp to hw specific files
fm10k: Add support for PTP
drivers/net/ethernet/intel/Kconfig | 19
drivers/net/ethernet/intel/Makefile | 1
drivers/net/ethernet/intel/fm10k/Makefile | 33
drivers/net/ethernet/intel/fm10k/fm10k.h | 532 +++++
drivers/net/ethernet/intel/fm10k/fm10k_common.c | 534 +++++
drivers/net/ethernet/intel/fm10k/fm10k_common.h | 65 +
drivers/net/ethernet/intel/fm10k/fm10k_dcbnl.c | 174 ++
drivers/net/ethernet/intel/fm10k/fm10k_debugfs.c | 259 +++
drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c | 1069 +++++++++++
drivers/net/ethernet/intel/fm10k/fm10k_iov.c | 536 +++++
drivers/net/ethernet/intel/fm10k/fm10k_main.c | 1978 ++++++++++++++++++++
drivers/net/ethernet/intel/fm10k/fm10k_mbx.c | 2125 ++++++++++++++++++++++
drivers/net/ethernet/intel/fm10k/fm10k_mbx.h | 307 +++
drivers/net/ethernet/intel/fm10k/fm10k_netdev.c | 1424 ++++++++++++++
drivers/net/ethernet/intel/fm10k/fm10k_pci.c | 2166 ++++++++++++++++++++++
drivers/net/ethernet/intel/fm10k/fm10k_pf.c | 1849 +++++++++++++++++++
drivers/net/ethernet/intel/fm10k/fm10k_pf.h | 135 +
drivers/net/ethernet/intel/fm10k/fm10k_ptp.c | 535 +++++
drivers/net/ethernet/intel/fm10k/fm10k_tlv.c | 863 +++++++++
drivers/net/ethernet/intel/fm10k/fm10k_tlv.h | 186 ++
drivers/net/ethernet/intel/fm10k/fm10k_type.h | 769 ++++++++
drivers/net/ethernet/intel/fm10k/fm10k_vf.c | 552 ++++++
drivers/net/ethernet/intel/fm10k/fm10k_vf.h | 78 +
23 files changed, 16189 insertions(+)
create mode 100644 drivers/net/ethernet/intel/fm10k/Makefile
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k.h
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_common.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_common.h
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_dcbnl.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_debugfs.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_ethtool.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_iov.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_main.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_mbx.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_mbx.h
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_netdev.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_pci.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_pf.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_pf.h
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_ptp.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_tlv.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_tlv.h
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_type.h
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_vf.c
create mode 100644 drivers/net/ethernet/intel/fm10k/fm10k_vf.h
--
^ permalink raw reply
* Re: [net] net: sched: shrink struct qdisc_skb_cb to 28 bytes
From: Doug Ledford @ 2014-09-18 22:29 UTC (permalink / raw)
To: Eric Dumazet, Or Gerlitz
Cc: Govindarajulu Varadarajan, Yinghai Lu, David Miller, NetDev,
ssujith, gvaradar, Christian Benvenuti (benve)
In-Reply-To: <1411052525.7106.269.camel@edumazet-glaptop2.roam.corp.google.com>
On 09/18/2014 11:02 AM, Eric Dumazet wrote:
> From: Eric Dumazet <edumazet@google.com>
>
> We cannot make struct qdisc_skb_cb bigger without impacting IPoIB,
> or increasing skb->cb[] size.
>
> Commit e0f31d849867 ("flow_keys: Record IP layer protocol in
> skb_flow_dissect()") broke IPoIB.
>
> Only current offender is sch_choke, and this one do not need an
> absolutely precise flow key.
>
> If we store 17 bytes of flow key, its more than enough. (Its the actual
> size of flow_keys if it was a packed structure, but we might add new
> fields at the end of it later)
>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Fixes: e0f31d849867 ("flow_keys: Record IP layer protocol in skb_flow_dissect()")
I've installed this patch on my cluster and it resolves the problem.
Tested-by/Acked-by: Doug Ledford <dledford@redhat.com>
^ permalink raw reply
* [PATCH net-next 0/2] qlge: Fix compilation warning and update maintainers
From: Harish Patil @ 2014-09-18 21:27 UTC (permalink / raw)
To: davem; +Cc: netdev, Dept-GELinuxNICDev, sudarsana.kalluru, Harish Patil
From: Harish Patil <harish.patil@qlogic.com>
This patch series includes the following set of patches:
- Fix the below warning message:
qlge_main.c:1754: warning: 'lbq_desc' may be used uninitialized in this function
I have made changes according to your earlier feedback:
"Please fix this differently. The problem is that the compiler can't see that
you've done the !length check at the top of the function, so when it later
sees the while (length > 0) loop, it doesn't know that this loop will always
execute at least once. Just change that loop to a do { } while() loop and
the compiler will be able to see everything."
- Update qlge driver maintainers list
Please apply this series to net-next.
Thanks,
Harish
Harish Patil (1):
qlge: Fix compilation warning
Update qlge driver maintainers list
MAINTAINERS | 6 +++---
drivers/net/ethernet/qlogic/qlge/qlge_main.c | 4 ++--
2 files changed, 5 insertions(+), 5 deletions(-)
^ permalink raw reply
* [PATCH net-next 2/2] Update qlge driver maintainers list
From: Harish Patil @ 2014-09-18 21:27 UTC (permalink / raw)
To: davem; +Cc: netdev, Dept-GELinuxNICDev, sudarsana.kalluru, Harish Patil
In-Reply-To: <1411075645-21845-1-git-send-email-harish.patil@qlogic.com>
From: Harish Patil <harish.patil@qlogic.com>
Signed-off-by: Harish Patil <harish.patil@qlogic.com>
---
MAINTAINERS | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 5e3709e..b4e23ac 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -7385,9 +7385,9 @@ S: Supported
F: drivers/net/ethernet/qlogic/qlcnic/
QLOGIC QLGE 10Gb ETHERNET DRIVER
-M: Shahed Shaikh <shahed.shaikh@qlogic.com>
-M: Jitendra Kalsaria <jitendra.kalsaria@qlogic.com>
-M: Ron Mercer <ron.mercer@qlogic.com>
+M: Harish Patil <harish.patil@qlogic.com>
+M: Sudarsana Kalluru <sudarsana.kalluru@qlogic.com>
+M: Dept-GELinuxNICDev@qlogic.com
M: linux-driver@qlogic.com
L: netdev@vger.kernel.org
S: Supported
--
1.8.1.4
^ permalink raw reply related
* [PATCH net-next 1/2] qlge: Fix compilation warning
From: Harish Patil @ 2014-09-18 21:27 UTC (permalink / raw)
To: davem; +Cc: netdev, Dept-GELinuxNICDev, sudarsana.kalluru, Harish Patil
In-Reply-To: <1411075645-21845-1-git-send-email-harish.patil@qlogic.com>
From: Harish Patil <harish.patil@qlogic.com>
Signed-off-by: Harish Patil <harish.patil@qlogic.com>
Fix the below warning message:
qlge_main.c:1754: warning: 'lbq_desc' may be used uninitialized in this function
---
drivers/net/ethernet/qlogic/qlge/qlge_main.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/qlogic/qlge/qlge_main.c b/drivers/net/ethernet/qlogic/qlge/qlge_main.c
index 3e96f26..6c904a6 100644
--- a/drivers/net/ethernet/qlogic/qlge/qlge_main.c
+++ b/drivers/net/ethernet/qlogic/qlge/qlge_main.c
@@ -1922,7 +1922,7 @@ static struct sk_buff *ql_build_rx_skb(struct ql_adapter *qdev,
sbq_desc->p.skb = NULL;
skb_reserve(skb, NET_IP_ALIGN);
}
- while (length > 0) {
+ do {
lbq_desc = ql_get_curr_lchunk(qdev, rx_ring);
size = (length < rx_ring->lbq_buf_size) ? length :
rx_ring->lbq_buf_size;
@@ -1939,7 +1939,7 @@ static struct sk_buff *ql_build_rx_skb(struct ql_adapter *qdev,
skb->truesize += size;
length -= size;
i++;
- }
+ } while (length > 0);
ql_update_mac_hdr_len(qdev, ib_mac_rsp, lbq_desc->p.pg_chunk.va,
&hlen);
__pskb_pull_tail(skb, hlen);
--
1.8.1.4
^ permalink raw reply related
* [PATCH net-next v4 2/2] bonding: Simplify the xmit function for modes that use xmit_hash
From: Mahesh Bandewar @ 2014-09-18 21:53 UTC (permalink / raw)
To: Jay Vosburgh, Veaceslav Falico, Andy Gospodarek, David Miller
Cc: netdev, Mahesh Bandewar, Eric Dumazet, Maciej Zenczykowski
Earlier change to use usable slave array for TLB mode had an additional
performance advantage. So extending the same logic to all other modes
that use xmit-hash for slave selection (viz 802.3AD, and XOR modes).
Also consolidating this with the earlier TLB change.
The main idea is to build the usable slaves array in the control path
and use that array for slave selection during xmit operation.
Measured performance in a setup with a bond of 4x1G NICs with 200
instances of netperf for the modes involved (3ad, xor, tlb)
cmd: netperf -t TCP_RR -H <TargetHost> -l 60 -s 5
Mode TPS-Before TPS-After
802.3ad : 468,694 493,101
TLB (lb=0): 392,583 392,965
XOR : 475,696 484,517
Signed-off-by: Mahesh Bandewar <maheshb@google.com>
---
v1:
(a) If bond_update_slave_arr() fails to allocate memory, it will overwrite
the slave that need to be removed.
(b) Freeing of array will assign NULL (to handle bond->down to bond->up
transition gracefully.
(c) Change from pr_debug() to pr_err() if bond_update_slave_arr() returns
failure.
(d) XOR: bond_update_slave_arr() will consider mii-mon, arp-mon cases and
will populate the array even if these parameters are not used.
(e) 3AD: Should handle the ad_agg_selection_logic correctly.
v2:
(a) Removed rcu_read_{un}lock() calls from array manipulation code.
(b) Slave link-events now refresh array for all these modes.
(c) Moved free-array call from bond_close() to bond_uninit().
v3:
(a) Fixed null pointer dereference.
(b) Removed bond->lock lockdep dependency.
v4:
(a) Made to changes to comply with Nikolay's locking changes
(b) Added a work-queue to refresh slave-array when RTNL is not held
(c) Array refresh happens ONLY with RTNL now.
(d) alloc changed from GFP_ATOMIC to GFP_KERNEL
drivers/net/bonding/bond_3ad.c | 88 +++++--------------
drivers/net/bonding/bond_alb.c | 51 ++---------
drivers/net/bonding/bond_alb.h | 8 --
drivers/net/bonding/bond_main.c | 189 ++++++++++++++++++++++++++++++++++++++--
drivers/net/bonding/bonding.h | 10 +++
5 files changed, 218 insertions(+), 128 deletions(-)
diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c
index 7e9e522fd476..4bf3756dcc11 100644
--- a/drivers/net/bonding/bond_3ad.c
+++ b/drivers/net/bonding/bond_3ad.c
@@ -1550,6 +1550,11 @@ static void ad_agg_selection_logic(struct aggregator *agg)
__disable_port(port);
}
}
+ /* RTNL may or may not be held but bond->mode_lock
+ * is held. It's not safe to update slave-arr here.
+ * Defer it to delayed-work.
+ */
+ bond_slave_arr_work_rearm(bond);
}
/* if the selected aggregator is of join individuals
@@ -1688,6 +1693,11 @@ static void ad_enable_collecting_distributing(struct port *port)
port->actor_port_number,
port->aggregator->aggregator_identifier);
__enable_port(port);
+ /* RTNL is not be held and bond->mode_lock is held.
+ * It's not safe to update slave-arr here!
+ * Defer it to delayed-work.
+ */
+ bond_slave_arr_work_rearm(port->slave->bond);
}
}
@@ -1704,6 +1714,11 @@ static void ad_disable_collecting_distributing(struct port *port)
port->actor_port_number,
port->aggregator->aggregator_identifier);
__disable_port(port);
+ /* RTNL is not be held and bond->mode_lock is held.
+ * It's not safe to update slave-arr here!
+ * Defer it to delayed-work.
+ */
+ bond_slave_arr_work_rearm(port->slave->bond);
}
}
@@ -2283,6 +2298,12 @@ void bond_3ad_handle_link_change(struct slave *slave, char link)
port->sm_vars |= AD_PORT_BEGIN;
spin_unlock_bh(&slave->bond->mode_lock);
+
+ /* RTNL is held and mode_lock is released so it's safe
+ * to update slave_array here.
+ */
+ if (bond_update_slave_arr(slave->bond, NULL))
+ pr_err("Failed to build slave-array for 3ad mode.\n");
}
/**
@@ -2377,73 +2398,6 @@ int bond_3ad_get_active_agg_info(struct bonding *bond, struct ad_info *ad_info)
return ret;
}
-int bond_3ad_xmit_xor(struct sk_buff *skb, struct net_device *dev)
-{
- struct bonding *bond = netdev_priv(dev);
- struct slave *slave, *first_ok_slave;
- struct aggregator *agg;
- struct ad_info ad_info;
- struct list_head *iter;
- int slaves_in_agg;
- int slave_agg_no;
- int agg_id;
-
- if (__bond_3ad_get_active_agg_info(bond, &ad_info)) {
- netdev_dbg(dev, "__bond_3ad_get_active_agg_info failed\n");
- goto err_free;
- }
-
- slaves_in_agg = ad_info.ports;
- agg_id = ad_info.aggregator_id;
-
- if (slaves_in_agg == 0) {
- netdev_dbg(dev, "active aggregator is empty\n");
- goto err_free;
- }
-
- slave_agg_no = bond_xmit_hash(bond, skb) % slaves_in_agg;
- first_ok_slave = NULL;
-
- bond_for_each_slave_rcu(bond, slave, iter) {
- agg = SLAVE_AD_INFO(slave)->port.aggregator;
- if (!agg || agg->aggregator_identifier != agg_id)
- continue;
-
- if (slave_agg_no >= 0) {
- if (!first_ok_slave && bond_slave_can_tx(slave))
- first_ok_slave = slave;
- slave_agg_no--;
- continue;
- }
-
- if (bond_slave_can_tx(slave)) {
- bond_dev_queue_xmit(bond, skb, slave->dev);
- goto out;
- }
- }
-
- if (slave_agg_no >= 0) {
- netdev_err(dev, "Couldn't find a slave to tx on for aggregator ID %d\n",
- agg_id);
- goto err_free;
- }
-
- /* we couldn't find any suitable slave after the agg_no, so use the
- * first suitable found, if found.
- */
- if (first_ok_slave)
- bond_dev_queue_xmit(bond, skb, first_ok_slave->dev);
- else
- goto err_free;
-
-out:
- return NETDEV_TX_OK;
-err_free:
- /* no suitable interface, frame not sent */
- dev_kfree_skb_any(skb);
- goto out;
-}
-
int bond_3ad_lacpdu_recv(const struct sk_buff *skb, struct bonding *bond,
struct slave *slave)
{
diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
index 615f3bebd019..d2eadab787c5 100644
--- a/drivers/net/bonding/bond_alb.c
+++ b/drivers/net/bonding/bond_alb.c
@@ -177,7 +177,6 @@ static int tlb_initialize(struct bonding *bond)
static void tlb_deinitialize(struct bonding *bond)
{
struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
- struct tlb_up_slave *arr;
spin_lock_bh(&bond->mode_lock);
@@ -185,10 +184,6 @@ static void tlb_deinitialize(struct bonding *bond)
bond_info->tx_hashtbl = NULL;
spin_unlock_bh(&bond->mode_lock);
-
- arr = rtnl_dereference(bond_info->slave_arr);
- if (arr)
- kfree_rcu(arr, rcu);
}
static long long compute_gap(struct slave *slave)
@@ -1336,39 +1331,9 @@ out:
return NETDEV_TX_OK;
}
-static int bond_tlb_update_slave_arr(struct bonding *bond,
- struct slave *skipslave)
-{
- struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
- struct slave *tx_slave;
- struct list_head *iter;
- struct tlb_up_slave *new_arr, *old_arr;
-
- new_arr = kzalloc(offsetof(struct tlb_up_slave, arr[bond->slave_cnt]),
- GFP_ATOMIC);
- if (!new_arr)
- return -ENOMEM;
-
- bond_for_each_slave(bond, tx_slave, iter) {
- if (!bond_slave_can_tx(tx_slave))
- continue;
- if (skipslave == tx_slave)
- continue;
- new_arr->arr[new_arr->count++] = tx_slave;
- }
-
- old_arr = rtnl_dereference(bond_info->slave_arr);
- rcu_assign_pointer(bond_info->slave_arr, new_arr);
- if (old_arr)
- kfree_rcu(old_arr, rcu);
-
- return 0;
-}
-
int bond_tlb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
{
struct bonding *bond = netdev_priv(bond_dev);
- struct alb_bond_info *bond_info = &(BOND_ALB_INFO(bond));
struct ethhdr *eth_data;
struct slave *tx_slave = NULL;
u32 hash_index;
@@ -1389,12 +1354,14 @@ int bond_tlb_xmit(struct sk_buff *skb, struct net_device *bond_dev)
hash_index & 0xFF,
skb->len);
} else {
- struct tlb_up_slave *slaves;
+ struct bond_up_slave *slaves;
+ unsigned int count;
- slaves = rcu_dereference(bond_info->slave_arr);
- if (slaves && slaves->count)
+ slaves = rcu_dereference(bond->slave_arr);
+ count = slaves ? ACCESS_ONCE(slaves->count) : 0;
+ if (likely(count))
tx_slave = slaves->arr[hash_index %
- slaves->count];
+ count];
}
break;
}
@@ -1641,10 +1608,6 @@ void bond_alb_deinit_slave(struct bonding *bond, struct slave *slave)
rlb_clear_slave(bond, slave);
}
- if (bond_is_nondyn_tlb(bond))
- if (bond_tlb_update_slave_arr(bond, slave))
- pr_err("Failed to build slave-array for TLB mode.\n");
-
}
void bond_alb_handle_link_change(struct bonding *bond, struct slave *slave, char link)
@@ -1669,7 +1632,7 @@ void bond_alb_handle_link_change(struct bonding *bond, struct slave *slave, char
}
if (bond_is_nondyn_tlb(bond)) {
- if (bond_tlb_update_slave_arr(bond, NULL))
+ if (bond_update_slave_arr(bond, NULL))
pr_err("Failed to build slave-array for TLB mode.\n");
}
}
diff --git a/drivers/net/bonding/bond_alb.h b/drivers/net/bonding/bond_alb.h
index 3c6a7ff974d7..1ad473b4ade5 100644
--- a/drivers/net/bonding/bond_alb.h
+++ b/drivers/net/bonding/bond_alb.h
@@ -139,19 +139,11 @@ struct tlb_slave_info {
*/
};
-struct tlb_up_slave {
- unsigned int count;
- struct rcu_head rcu;
- struct slave *arr[0];
-};
-
struct alb_bond_info {
struct tlb_client_info *tx_hashtbl; /* Dynamically allocated */
u32 unbalanced_load;
int tx_rebalance_counter;
int lp_counter;
- /* -------- non-dynamic tlb mode only ---------*/
- struct tlb_up_slave __rcu *slave_arr; /* Up slaves */
/* -------- rlb parameters -------- */
int rlb_enabled;
struct rlb_client_info *rx_hashtbl; /* Receive hash table */
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 5e7987bba583..e87b802d8813 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -208,6 +208,7 @@ static int lacp_fast;
static int bond_init(struct net_device *bond_dev);
static void bond_uninit(struct net_device *bond_dev);
+static void bond_slave_arr_handler(struct work_struct *work);
/*---------------------------- General routines -----------------------------*/
@@ -1547,6 +1548,10 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev)
unblock_netpoll_tx();
}
+ if (bond_mode_uses_xmit_hash(bond) &&
+ bond_update_slave_arr(bond, NULL))
+ pr_err("Failed to build slave-array.\n");
+
netdev_info(bond_dev, "Enslaving %s as %s interface with %s link\n",
slave_dev->name,
bond_is_active_slave(new_slave) ? "an active" : "a backup",
@@ -1661,6 +1666,10 @@ static int __bond_release_one(struct net_device *bond_dev,
if (BOND_MODE(bond) == BOND_MODE_8023AD)
bond_3ad_unbind_slave(slave);
+ if (bond_mode_uses_xmit_hash(bond) &&
+ bond_update_slave_arr(bond, slave))
+ pr_err("Failed to build slave-array.\n");
+
netdev_info(bond_dev, "Releasing %s interface %s\n",
bond_is_active_slave(slave) ? "active" : "backup",
slave_dev->name);
@@ -1963,6 +1972,10 @@ static void bond_miimon_commit(struct bonding *bond)
bond_alb_handle_link_change(bond, slave,
BOND_LINK_UP);
+ if (BOND_MODE(bond) == BOND_MODE_XOR &&
+ bond_update_slave_arr(bond, NULL))
+ pr_err("Failed to build slave-array for XOR mode.\n");
+
if (!bond->curr_active_slave || slave == primary)
goto do_failover;
@@ -1990,6 +2003,10 @@ static void bond_miimon_commit(struct bonding *bond)
bond_alb_handle_link_change(bond, slave,
BOND_LINK_DOWN);
+ if (BOND_MODE(bond) == BOND_MODE_XOR &&
+ bond_update_slave_arr(bond, NULL))
+ pr_err("Failed to build slave-array for XOR mode.\n");
+
if (slave == rcu_access_pointer(bond->curr_active_slave))
goto do_failover;
@@ -2446,6 +2463,9 @@ static void bond_loadbalance_arp_mon(struct work_struct *work)
if (slave_state_changed) {
bond_slave_state_change(bond);
+ if (BOND_MODE(bond) == BOND_MODE_XOR &&
+ bond_update_slave_arr(bond, NULL))
+ pr_err("Failed to build slave-array for XOR mode.\n");
} else if (do_failover) {
block_netpoll_tx();
bond_select_active_slave(bond);
@@ -2822,8 +2842,23 @@ static int bond_slave_netdev_event(unsigned long event,
if (old_duplex != slave->duplex)
bond_3ad_adapter_duplex_changed(slave);
}
+ /* Refresh slave-array if applicable!
+ * If the setuo does not use miimon or arpmon (mode-specific!),
+ * then these event will not cause the slave-array to be
+ * refreshed. This will cause xmit to use a slave that is not
+ * usable. Avoid such situation by refeshing the array at these
+ * events. If these (miimon/arpmon) parameters are configured
+ * then array gets refreshed twice and that should be fine!
+ */
+ if (bond_mode_uses_xmit_hash(bond) &&
+ bond_update_slave_arr(bond, NULL))
+ pr_err("Failed to build slave-array.\n");
break;
case NETDEV_DOWN:
+ /* Refresh slave-array if applicable! */
+ if (bond_mode_uses_xmit_hash(bond) &&
+ bond_update_slave_arr(bond, NULL))
+ pr_err("Failed to build slave-array.\n");
break;
case NETDEV_CHANGEMTU:
/* TODO: Should slaves be allowed to
@@ -3003,6 +3038,7 @@ static void bond_work_init_all(struct bonding *bond)
else
INIT_DELAYED_WORK(&bond->arp_work, bond_loadbalance_arp_mon);
INIT_DELAYED_WORK(&bond->ad_work, bond_3ad_state_machine_handler);
+ INIT_DELAYED_WORK(&bond->slave_arr_work, bond_slave_arr_handler);
}
static void bond_work_cancel_all(struct bonding *bond)
@@ -3012,6 +3048,7 @@ static void bond_work_cancel_all(struct bonding *bond)
cancel_delayed_work_sync(&bond->alb_work);
cancel_delayed_work_sync(&bond->ad_work);
cancel_delayed_work_sync(&bond->mcast_work);
+ cancel_delayed_work_sync(&bond->slave_arr_work);
}
static int bond_open(struct net_device *bond_dev)
@@ -3061,6 +3098,10 @@ static int bond_open(struct net_device *bond_dev)
bond_3ad_initiate_agg_selection(bond, 1);
}
+ if (bond_mode_uses_xmit_hash(bond) &&
+ bond_update_slave_arr(bond, NULL))
+ pr_err("Failed to build slave-array.\n");
+
return 0;
}
@@ -3555,15 +3596,139 @@ static int bond_xmit_activebackup(struct sk_buff *skb, struct net_device *bond_d
return NETDEV_TX_OK;
}
-/* In bond_xmit_xor() , we determine the output device by using a pre-
- * determined xmit_hash_policy(), If the selected device is not enabled,
- * find the next active slave.
+/* The caller is holding bond->mode_lock and may or may not be
+ * holding RTNL.
*/
-static int bond_xmit_xor(struct sk_buff *skb, struct net_device *bond_dev)
+void bond_slave_arr_work_rearm(struct bonding *bond)
{
- struct bonding *bond = netdev_priv(bond_dev);
+ queue_delayed_work(bond->wq, &bond->slave_arr_work, 1);
+}
- bond_xmit_slave_id(bond, skb, bond_xmit_hash(bond, skb) % bond->slave_cnt);
+/* Slave array work handler. Holds only RTNL */
+static void bond_slave_arr_handler(struct work_struct *work)
+{
+ struct bonding *bond = container_of(work, struct bonding,
+ slave_arr_work.work);
+ int ret;
+
+ if (!rtnl_trylock())
+ goto err;
+
+ ret = bond_update_slave_arr(bond, NULL);
+ rtnl_unlock();
+ if (ret) {
+ pr_warn_ratelimited("Failed to update slave array from WT\n");
+ goto err;
+ }
+ return;
+
+err:
+ bond_slave_arr_work_rearm(bond);
+}
+
+/* Build the usable slaves array in control path for modes that use xmit-hash
+ * to determine the slave interface -
+ * (a) BOND_MODE_8023AD
+ * (b) BOND_MODE_XOR
+ * (c) BOND_MODE_TLB && tlb_dynamic_lb == 0
+ *
+ * The caller is expected to hold RTNL only and NO other lock!
+ */
+int bond_update_slave_arr(struct bonding *bond, struct slave *skipslave)
+{
+ struct slave *slave;
+ struct list_head *iter;
+ struct bond_up_slave *new_arr, *old_arr;
+ int slaves_in_agg;
+ int agg_id = 0;
+ int ret = 0;
+
+#ifdef CONFIG_LOCKDEP
+ WARN_ON(lockdep_is_held(&bond->mode_lock));
+#endif
+
+ new_arr = kzalloc(offsetof(struct bond_up_slave, arr[bond->slave_cnt]),
+ GFP_KERNEL);
+ if (!new_arr) {
+ ret = -ENOMEM;
+ goto out;
+ }
+ if (BOND_MODE(bond) == BOND_MODE_8023AD) {
+ struct ad_info ad_info;
+
+ if (bond_3ad_get_active_agg_info(bond, &ad_info)) {
+ pr_debug("bond_3ad_get_active_agg_info failed\n");
+ kfree_rcu(new_arr, rcu);
+ ret = -EINVAL;
+ goto out;
+ }
+ slaves_in_agg = ad_info.ports;
+ agg_id = ad_info.aggregator_id;
+ }
+ bond_for_each_slave(bond, slave, iter) {
+ if (BOND_MODE(bond) == BOND_MODE_8023AD) {
+ struct aggregator *agg;
+
+ agg = SLAVE_AD_INFO(slave)->port.aggregator;
+ if (!agg || agg->aggregator_identifier != agg_id)
+ continue;
+ }
+ if (!bond_slave_can_tx(slave))
+ continue;
+ if (skipslave == slave)
+ continue;
+ new_arr->arr[new_arr->count++] = slave;
+ }
+
+ old_arr = rtnl_dereference(bond->slave_arr);
+ rcu_assign_pointer(bond->slave_arr, new_arr);
+ if (old_arr)
+ kfree_rcu(old_arr, rcu);
+out:
+ if (ret != 0 && skipslave) {
+ int idx;
+
+ /* Rare situation where caller has asked to skip a specific
+ * slave but allocation failed (most likely!). BTW this is
+ * only possible when the call is initiated from
+ * __bond_release_one(). In this sitation; overwrite the
+ * skipslave entry in the array with the last entry from the
+ * array to avoid a situation where the xmit path may choose
+ * this to-be-skipped slave to send a packet out.
+ */
+ old_arr = rtnl_dereference(bond->slave_arr);
+ for (idx = 0; idx < old_arr->count; idx++) {
+ if (skipslave == old_arr->arr[idx]) {
+ old_arr->arr[idx] =
+ old_arr->arr[old_arr->count-1];
+ old_arr->count--;
+ break;
+ }
+ }
+ }
+ return ret;
+}
+
+/* Use this Xmit function for 3AD as well as XOR modes. The current
+ * usable slave array is formed in the control path. The xmit function
+ * just calculates hash and sends the packet out.
+ */
+int bond_3ad_xor_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+ struct bonding *bond = netdev_priv(dev);
+ struct slave *slave;
+ struct bond_up_slave *slaves;
+ unsigned int count;
+
+ slaves = rcu_dereference(bond->slave_arr);
+ count = slaves ? ACCESS_ONCE(slaves->count) : 0;
+ if (likely(count)) {
+ slave = slaves->arr[bond_xmit_hash(bond, skb) % count];
+ bond_dev_queue_xmit(bond, skb, slave->dev);
+ } else {
+ dev_kfree_skb_any(skb);
+ atomic_long_inc(&dev->tx_dropped);
+ }
return NETDEV_TX_OK;
}
@@ -3660,12 +3825,11 @@ static netdev_tx_t __bond_start_xmit(struct sk_buff *skb, struct net_device *dev
return bond_xmit_roundrobin(skb, dev);
case BOND_MODE_ACTIVEBACKUP:
return bond_xmit_activebackup(skb, dev);
+ case BOND_MODE_8023AD:
case BOND_MODE_XOR:
- return bond_xmit_xor(skb, dev);
+ return bond_3ad_xor_xmit(skb, dev);
case BOND_MODE_BROADCAST:
return bond_xmit_broadcast(skb, dev);
- case BOND_MODE_8023AD:
- return bond_3ad_xmit_xor(skb, dev);
case BOND_MODE_ALB:
return bond_alb_xmit(skb, dev);
case BOND_MODE_TLB:
@@ -3839,6 +4003,7 @@ static void bond_uninit(struct net_device *bond_dev)
struct bonding *bond = netdev_priv(bond_dev);
struct list_head *iter;
struct slave *slave;
+ struct bond_up_slave *arr;
bond_netpoll_cleanup(bond_dev);
@@ -3847,6 +4012,12 @@ static void bond_uninit(struct net_device *bond_dev)
__bond_release_one(bond_dev, slave->dev, true);
netdev_info(bond_dev, "Released all slaves\n");
+ arr = rtnl_dereference(bond->slave_arr);
+ if (arr) {
+ kfree_rcu(arr, rcu);
+ RCU_INIT_POINTER(bond->slave_arr, NULL);
+ }
+
list_del(&bond->bond_list);
bond_debug_unregister(bond);
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index 98dc0d7ad731..4635b175256a 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -177,6 +177,12 @@ struct slave {
struct kobject kobj;
};
+struct bond_up_slave {
+ unsigned int count;
+ struct rcu_head rcu;
+ struct slave *arr[0];
+};
+
/*
* Link pseudo-state only used internally by monitors
*/
@@ -191,6 +197,7 @@ struct bonding {
struct slave __rcu *curr_active_slave;
struct slave __rcu *current_arp_slave;
struct slave __rcu *primary_slave;
+ struct bond_up_slave __rcu *slave_arr; /* Array of usable slaves */
bool force_primary;
s32 slave_cnt; /* never change this value outside the attach/detach wrappers */
int (*recv_probe)(const struct sk_buff *, struct bonding *,
@@ -220,6 +227,7 @@ struct bonding {
struct delayed_work alb_work;
struct delayed_work ad_work;
struct delayed_work mcast_work;
+ struct delayed_work slave_arr_work;
#ifdef CONFIG_DEBUG_FS
/* debugging support via debugfs */
struct dentry *debug_dir;
@@ -531,6 +539,8 @@ const char *bond_slave_link_status(s8 link);
struct bond_vlan_tag *bond_verify_device_path(struct net_device *start_dev,
struct net_device *end_dev,
int level);
+int bond_update_slave_arr(struct bonding *bond, struct slave *skipslave);
+void bond_slave_arr_work_rearm(struct bonding *bond);
#ifdef CONFIG_PROC_FS
void bond_create_proc_entry(struct bonding *bond);
--
2.1.0.rc2.206.gedb03e5
^ permalink raw reply related
* [PATCH net-next v4 1/2] bonding: display xmit_hash_policy for non-dynamic-tlb mode
From: Mahesh Bandewar @ 2014-09-18 21:53 UTC (permalink / raw)
To: Jay Vosburgh, Veaceslav Falico, Andy Gospodarek, David Miller
Cc: netdev, Mahesh Bandewar, Eric Dumazet, Maciej Zenczykowski
It's a trivial fix to display xmit_hash_policy for this new TLB mode
since it uses transmit-hash-poilicy as part of bonding-master info
(/proc/net/bonding/<bonding-interface).
Signed-off-by: Mahesh Bandewar <maheshb@google.com>
---
v1
Rebase
v2
Added bond_mode_uses_xmit_hash() inline function
v3
Rebase
v4
Rebase
drivers/net/bonding/bond_procfs.c | 3 +--
drivers/net/bonding/bonding.h | 7 +++++++
2 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/drivers/net/bonding/bond_procfs.c b/drivers/net/bonding/bond_procfs.c
index bb09d0442aa8..a3948f8d1e53 100644
--- a/drivers/net/bonding/bond_procfs.c
+++ b/drivers/net/bonding/bond_procfs.c
@@ -73,8 +73,7 @@ static void bond_info_show_master(struct seq_file *seq)
seq_printf(seq, "\n");
- if (BOND_MODE(bond) == BOND_MODE_XOR ||
- BOND_MODE(bond) == BOND_MODE_8023AD) {
+ if (bond_mode_uses_xmit_hash(bond)) {
optval = bond_opt_get_val(BOND_OPT_XMIT_HASH,
bond->params.xmit_policy);
seq_printf(seq, "Transmit Hash Policy: %s (%d)\n",
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index 6140bf0264a4..98dc0d7ad731 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -271,6 +271,13 @@ static inline bool bond_is_nondyn_tlb(const struct bonding *bond)
(bond->params.tlb_dynamic_lb == 0);
}
+static inline bool bond_mode_uses_xmit_hash(const struct bonding *bond)
+{
+ return (BOND_MODE(bond) == BOND_MODE_8023AD ||
+ BOND_MODE(bond) == BOND_MODE_XOR ||
+ bond_is_nondyn_tlb(bond));
+}
+
static inline bool bond_mode_uses_arp(int mode)
{
return mode != BOND_MODE_8023AD && mode != BOND_MODE_TLB &&
--
2.1.0.rc2.206.gedb03e5
^ permalink raw reply related
* Re: [PATCH net] net: sched: shrink struct qdisc_skb_cb to 28 bytes
From: Or Gerlitz @ 2014-09-18 21:28 UTC (permalink / raw)
To: Eric Dumazet
Cc: Stephen Hemminger, Govindarajulu Varadarajan, Yinghai Lu,
David Miller, NetDev, ssujith, gvaradar,
Christian Benvenuti (benve)
In-Reply-To: <1411057934.7106.275.camel@edumazet-glaptop2.roam.corp.google.com>
On Thu, Sep 18, 2014 at 7:32 PM, Eric Dumazet <eric.dumazet@gmail.com> wrote:
> On Thu, 2014-09-18 at 09:26 -0700, Stephen Hemminger wrote:
>> On Thu, 18 Sep 2014 08:02:05 -0700
>> Eric Dumazet <eric.dumazet@gmail.com> wrote:
>>
>> > From: Eric Dumazet <edumazet@google.com>
>> >
>> > We cannot make struct qdisc_skb_cb bigger without impacting IPoIB,
>> > or increasing skb->cb[] size.
>> >
>> > Commit e0f31d849867 ("flow_keys: Record IP layer protocol in
>> > skb_flow_dissect()") broke IPoIB.
>> >
>> > Only current offender is sch_choke, and this one do not need an
>> > absolutely precise flow key.
>> >
>> > If we store 17 bytes of flow key, its more than enough. (Its the actual
>> > size of flow_keys if it was a packed structure, but we might add new
>> > fields at the end of it later)
>> >
>> > Signed-off-by: Eric Dumazet <edumazet@google.com>
>>
>> Can we add BUILD_BUG to stop next time something smacks this.
>
> I though we had.
>
> Maybe IPoIB lacks one.
>
> Or, do you have an idea ?
Nope, we currently don't have such build time check, I saw you posted
one and I will be able to test it Sunday.
^ permalink raw reply
* Re: [PATCH net-next] Allow to set net namespace for wireless device via RTM_LINK
From: Johannes Berg @ 2014-09-18 21:25 UTC (permalink / raw)
To: Vadim Kochan
Cc: netdev-u79uwXL29TY76Z2rM5mHXA,
linux-wireless-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1410467723-2550-1-git-send-email-vadim4j-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
On Thu, 2014-09-11 at 23:35 +0300, Vadim Kochan wrote:
> Added new netdev_ops callback for setting namespace in specific
> for this device way
> +++ b/include/linux/netdevice.h
> @@ -997,6 +997,8 @@ typedef u16 (*select_queue_fallback_t)(struct net_device *dev,
> * Callback to use for xmit over the accelerated station. This
> * is used in place of ndo_start_xmit on accelerated net
> * devices.
> + * int (*ndo_set_netns)(struct net_device *dev, struct net *net);
> + * Callback to set net namespace in specific way for this device.
For the record, I don't consider it appropriate to set the net namespace
on one netdev and end up with multiple netdevs switching namespaces ...
As a result, I don't think this should done.
johannes
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* [PATCH net-next] bnx2x: set gso_segs in LRO mode
From: Eric Dumazet @ 2014-09-18 20:48 UTC (permalink / raw)
To: David Miller; +Cc: netdev, Ariel Elior
From: Eric Dumazet <edumazet@google.com>
In commits cbf1de72324a8 ("bnx2x: fix GRO parameters"), and
ab5777d748302 ("bnx2x: Get gso_segs from FW")
special care was taken for GRO mode.
We can also get proper gso_segs for LRO mode.
This helps to get proper values for these SNMP counters, which should
count number of segments. This also helps qdisc_pkt_len_init() to better
track packet sizes when ingress qdisc is in place.
IpExtInNoECTPkts 63458
IpExtInECT0Pkts 8772480
IpExtInCEPkts 1390713
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
index 6dc32aee96bf..9c0c0f4bdfc3 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c
@@ -537,8 +537,10 @@ static void bnx2x_set_gro_params(struct sk_buff *skb, u16 parsing_flags,
/* tcp_gro_complete() will copy NAPI_GRO_CB(skb)->count
* to skb_shinfo(skb)->gso_segs
+ * We also set skb_shinfo(skb)->gso_segs in case we are in LRO mode
*/
NAPI_GRO_CB(skb)->count = num_of_coalesced_segs;
+ skb_shinfo(skb)->gso_segs = num_of_coalesced_segs;
}
static int bnx2x_alloc_rx_sge(struct bnx2x *bp, struct bnx2x_fastpath *fp,
^ permalink raw reply related
* Re: [PATCH net] net: sched: shrink struct qdisc_skb_cb to 28 bytes
From: Eric Dumazet @ 2014-09-18 20:31 UTC (permalink / raw)
To: Joe Perches
Cc: Stephen Hemminger, Or Gerlitz, Govindarajulu Varadarajan,
Yinghai Lu, David Miller, NetDev, ssujith, gvaradar,
Christian Benvenuti (benve)
In-Reply-To: <1411068671.4290.24.camel@joe-AO725>
On Thu, 2014-09-18 at 12:31 -0700, Joe Perches wrote:
> I don't mind the const argument, but casting
> the return to non-const seems like poor form.
This is the current way to do this, even if we do not like it.
C does not allow to have the same function name for different
parameters, there is nothing we can do about it at this moment.
->
static inline const struct ipoib_cb *ipoib_skb_cb(const struct sk_buff *skb)
static inline struct ipoib_cb *ipoib_skb_cb(struct sk_buff *skb)
^ permalink raw reply
* Re: [PATCHv5 net-next 3/3] sunvnet: generate ICMP PTMUD messages for smaller port MTUs
From: David L Stevens @ 2014-09-18 20:09 UTC (permalink / raw)
To: sowmini.varadhan; +Cc: David Miller, netdev
In-Reply-To: <541B314A.8090203@oracle.com>
On 09/18/2014 03:23 PM, Sowmini Varadhan wrote:
> by now I am actually quite confused by what the Administrator will see.
> If I do "ifconfig -a" or "ip addr", what is the reported mtu of the interface?
The interface MTU is whatever the admin sets it to, between 68 bytes (the IPv4 min)
and 64K-1 (the IPv4 max).
In cases where packets of interface MTU size cannot be delivered because the LDC
MTU is smaller, instead of silently dropping them, we send the ICMP errors which allow
PMTUD updates per-destination. Subsequent packets will be segmented or fragmented at that
(lower) value for that destination, and use other MTUs up to the interface MTU for other
destinations.
> Interesting. So if the Administrator sets up ICMP filters for outbound/inbound (at the IP layer), what will be the observed behavior?
If an administrator drops PMTUD packets, then TCP won't work, even without this patch set, for any
destinations that cause PMTUD. It's explicitly not optional in IPv6; in IPv4, fragmenting TCP
packets could hide it as long as IP_DF is not set, but the only thing this code could do for
packets too big is to drop them -- exactly what we'd do whether or not we send the ICMP error to
tell the sender what MTU we can send.
+-DLS
^ permalink raw reply
* Re: [PATCHv6 net-next 1/3] sunvnet: upgrade to VIO protocol version 1.6
From: David L Stevens @ 2014-09-18 19:58 UTC (permalink / raw)
To: Raghuram Kothakota; +Cc: David Miller, netdev
In-Reply-To: <9BA1705F-0C89-471A-9872-688A3FA3165C@oracle.com>
On 09/18/2014 02:49 PM, Raghuram Kothakota wrote:
>
> I am probably not as knowledgeable of sunvnet as you may be, but I would assume
> the code is capable of handling the a vport removal and should have sufficient method
> cleanup as well.
There is nothing in the code that I saw that checks if pending descriptors are marked as
VIO_DESC_DONE before freeing the buffer space when the device is shutdown. From that, I
assume a shutdown during active transmits could result in a fault or the remote looking
at garbage. Adding a wait for that condition could be "forever" if the remote disappears
while processing receives, so it'd need a timeout, too, and it all needs to switch gracefully
to the new buffer set.
I didn't want to put MTU changes in the list of things to trigger that, as I believe they
also would, nor to complicate the MTU-relevant pieces I have with something that I think
ultimately should use fewer, larger buffers.
> In the virtualization world, we want resources to be efficiently used and memory is
> still very important resource. My concern is mostly because this memory usage of
> 32+MB is on a per LDC basis. LDoms today supports a max of 128 domains, but
> from my experience seen actual deployments of the order of 50 domains. This is
> going up as the platforms getting more and more powerful. If there are really
> that many peers, then the amount of memory consumed by one vnet instance
> is 50 * 32+MB = 1.6GB+. It's fine if this memory is really used, but it seems like this
> will be useful only when the peer is another linux guest with this version of vnet and
> also the MTU is configured to use 64K. The memory is being wasted for all other
> peers that either don't support 64K MTU or not configured to use it and also
> the switch port as obviously it doesn't support 64K MTU today.
Since the current code allows only 1500-byte MTUs, this patch set is useful for any
value larger than that-- on Solaris, up to 16000, and on Linux, as high as 64K.
I think dynamic buffer allocation should not be tied to these other items, but it
should be done. This isn't the end of sunvnet development, but the beginning. However,
if 64K is too large, what value is not? Any number we pick is arbitrary and with large
numbers of LDOMs may be "too many." Of course, the smaller it is, the smaller the
benefit, too.
I would prefer to reduce VNET_MAXPACKET to some lower value, or make it a module
parameter, than to link this patchset to dynamic allocation of buffers. But the
16000 value Solaris supports would correspond to ~400MB in your example above --
still a large number for a single virtual interface.
>> b) I think most people will want to use large MTUs for performance; enough so
>> that perhaps the bring-up MTU should be 64K too
>
>
> From my experience in SPARC world, most customers pushed us back for any
> proposal to use Jumbo Frames. The customers who configured Jumbo frames,
> mostly used 9K for performance of NFS etc.
Customers have never been able to use jumbo frames on Linux LDOMs before, and I
expect an 8X improvement in throughput might affect their judgement on it. Moreso,
if those large buffers and throughput improvements can be done with GSO/TSO by
default with no MTU adjustments on the devices.
> When we implemented TSO support, we evaluated the cost of the buffers vs
> performance. We were able to limit TSO support to 8K(actually bit less) and still
> achieve high performance, for example we are able to drive line rate on a 10G
> and guest-to-guest of the order of 45+Gbps. So, my suggestion would be to
> increase the parallelism of the code more than depending on large MTU.
That's nice for Solaris. I see:
dlsl1 880 # ip link set mtu 8192 dev eth1
dlsl1 881 # netperf -H 10.0.0.2
TCP STREAM TEST to 10.0.0.2 : interval
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
87380 16384 16384 10.00 3176.10
dlsl1 882 # ip link set mtu 65535 dev eth1
dlsl1 883 # netperf -H 10.0.0.2
TCP STREAM TEST to 10.0.0.2 : interval
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
87380 16384 16384 10.00 7651.42
So, on Linux, throughput is more than doubled by supporting all the way to
64K-- something I expect to translate to TSO as well. Beyond that, it ought
to be whatever the admin wants-- not some arbitrary limit set in advance.
Since it is a virtual device, there is no physical constraint, as in Ethernet
signaling, to force it to 1500, or 9K. The only limit ought to be the IPv4
max packet size of 64K IP data + framing (and arguably not even that, if it's
using primarily or exclusively IPv6).
+-DLS
^ permalink raw reply
* Re: [PATCH 1/2 v2 nf-next] net: bridge: don't register netfilter call_iptables hooks by default
From: Patrick McHardy @ 2014-09-18 19:52 UTC (permalink / raw)
To: Florian Westphal, Pablo Neira Ayuso; +Cc: netfilter-devel, netdev
In-Reply-To: <20140918143108.GB8484@breakpoint.cc>
On 18. September 2014 16:31:08 MESZ, Florian Westphal <fw@strlen.de> wrote:
>Pablo Neira Ayuso <pablo@netfilter.org> wrote:
>> This patch modularizes br_netfilter so it can be
>> rmmod'ed, thus, the hooks can be unregistered.
>
>I see. Yes, that makes sense.
>
>Another alternative would be to merge both approaches.
>I.e, use my patch, but move all the hook registration to
>your proposed br_nf_core module.
>
>The sysctls would still be registered from bridge.ko.
>
>Then, if call_iptables is set and iptables rules are active,
>do the module autoload and print the warning from your patch, telling
>people to modprobe br_nf_core if they want the filtering.
>
>Unfortunately, we'd have to export (from bridge.ko) some hook
>to allow br_nf_core to signal that the hooks have been registered.
>
>Then, in two years or so, we would remove the autoload hook
>in the packet procesing path.
I believe basically any transition which relies on people reading feature-removal-schedule or kernel messages and a flag day is just kidding ourselves and things will break for everyone except maybe some distributions at flag day.
I think taking compatibility seriously means we can't really change the default behaviour, but can only allow people to manually undo the damage and make sure they won't rely on this behaviour for at least nftables.
Or we can change it. But then why wait for two years, it won't change anything.
Is there a way to make the userspace tools detect that the compat behaviour needs to be activated?
^ permalink raw reply
* Re: [PATCH 1/2 v2 nf-next] net: bridge: don't register netfilter call_iptables hooks by default
From: Patrick McHardy @ 2014-09-18 19:31 UTC (permalink / raw)
To: Pablo Neira Ayuso, Florian Westphal; +Cc: netfilter-devel, netdev
In-Reply-To: <20140918133511.GA8740@salvia>
On 18. September 2014 15:35:52 MESZ, Pablo Neira Ayuso <pablo@netfilter.org> wrote:
>Unless I'm missing anything, I think br_netfilter should have been a
>separated module since the beginning.
Yeah absolutely. Basic rule: don't impose the costs of your cool new feature on all the people who will never need it.
Great someone finally fixes this up. Long term we still need a sane design for this though, some direct way of bridging to interact with conntrack, nftables provides for the rest.
^ permalink raw reply
* Re: [PATCH net] net: sched: shrink struct qdisc_skb_cb to 28 bytes
From: Joe Perches @ 2014-09-18 19:31 UTC (permalink / raw)
To: Eric Dumazet
Cc: Stephen Hemminger, Or Gerlitz, Govindarajulu Varadarajan,
Yinghai Lu, David Miller, NetDev, ssujith, gvaradar,
Christian Benvenuti (benve)
In-Reply-To: <1411067679.7106.283.camel@edumazet-glaptop2.roam.corp.google.com>
On Thu, 2014-09-18 at 12:14 -0700, Eric Dumazet wrote:
> On Thu, 2014-09-18 at 11:07 -0700, Joe Perches wrote:
>
> > > +static inline struct ipoib_cb *ipoib_skb_cb(const struct sk_buff *skb)
> > > +{
> > > + BUILD_BUG_ON(sizeof(skb->cb) < sizeof(struct ipoib_cb));
> > > + return (struct ipoib_cb *)skb->cb;
> > > +}
> >
> > It seems better not to use const for the struct sk_buff * here.
> >
> > Neither of the uses take a const struct sk_buff *
>
> Thats pretty standard, check for other similar constructs like that.
>
>
> static inline struct qdisc_skb_cb *qdisc_skb_cb(const struct sk_buff *skb)
> {
> return (struct qdisc_skb_cb *)skb->cb;
> }
>
> This allows uses of the helper when the skb is only read (has the const qual)
I don't mind the const argument, but casting
the return to non-const seems like poor form.
btw: it seems like it's 8:5 non-const to const
$ grep -rP --include=*.[ch] -n "cb\s*\*\s*\w+\s*\(\s*(?:const\s+)?struct\s+sk_buff\s*\*\s*\w+\s*\)" *
drivers/net/ethernet/freescale/gianfar.c:2091:static inline struct txfcb *gfar_add_fcb(struct sk_buff *skb)
drivers/net/wireless/ath/ath10k/core.h:83:static inline struct ath10k_skb_cb *ATH10K_SKB_CB(struct sk_buff *skb)
include/net/mrp.h:38:static inline struct mrp_skb_cb *mrp_cb(struct sk_buff *skb)
include/net/ieee802154_netdev.h:235:static inline struct ieee802154_mac_cb *mac_cb(struct sk_buff *skb)
include/net/ieee802154_netdev.h:240:static inline struct ieee802154_mac_cb *mac_cb_init(struct sk_buff *skb)
include/net/sch_generic.h:251:static inline struct qdisc_skb_cb *qdisc_skb_cb(const struct sk_buff *skb)
include/net/codel.h:95:static struct codel_skb_cb *get_codel_cb(const struct sk_buff *skb)
include/net/garp.h:36:static inline struct garp_skb_cb *garp_cb(struct sk_buff *skb)
net/core/dev.c:2177:static struct dev_kfree_skb_cb *get_kfree_skb_cb(const struct sk_buff *skb)
net/sched/sch_choke.c:142:static inline struct choke_skb_cb *choke_skb_cb(const struct sk_buff *skb)
net/sched/sch_sfb.c:95:static inline struct sfb_skb_cb *sfb_skb_cb(const struct sk_buff *skb)
net/sched/sch_netem.c:171:static inline struct netem_skb_cb *netem_skb_cb(struct sk_buff *skb)
net/sched/sch_sfq.c:167:static inline struct sfq_skb_cb *sfq_skb_cb(const struct sk_buff *skb)
^ permalink raw reply
* Re: [PATCHv5 net-next 3/3] sunvnet: generate ICMP PTMUD messages for smaller port MTUs
From: Sowmini Varadhan @ 2014-09-18 19:23 UTC (permalink / raw)
To: David L Stevens; +Cc: David Miller, netdev
In-Reply-To: <541A1C46.90201@oracle.com>
On 09/17/2014 07:41 PM, David L Stevens wrote:
>
>
> On 09/17/2014 06:43 PM, Sowmini Varadhan wrote:
>> On (09/17/14 16:49), David L Stevens wrote:
>>> +
>>> + rt = ip_route_output_key(dev_net(dev), &fl4);
>>> + if (!IS_ERR(rt)) {
>>
>> As I've mentioned before, this layering violation makes me uneasy,
>> so its benefits should be evaluated carefully. You will typically not be
>> able to find an rt for packets coming here from any application
>> that does not itself use/update the FIB, e.g., uspace based packet-injectors
>> (PF_PACKET-based applications, intel dpdk-based uspace stacks etc.)
>
> A pair of Linux LDOMs can get 8X throughput improvement by raising the MTU to 64K, but
> many packets will be *silently* dropped if they go to any other destination that does
> not support 64K MTU. Those destinations that don't support 64K MTU include any legacy
> Linux running the pre-jumbo code and all Solaris hosts, including the current releases.
by now I am actually quite confused by what the Administrator will see.
If I do "ifconfig -a" or "ip addr", what is the reported mtu of the
interface?
> Also, I wouldn't call it a layering violation. icmp_send() is the external API for
> triggering ICMP errors, and we are sending them at the point where we know the next-hop MTU.
> It is exactly equivalent to an Ethernet device connected to a switch where the switch
> sends useful layer-3 packets (like IGMP queries). In this case, that useful layer 3 info
> is remote link MTU data; something not available in ordinary Ethernet.
Interesting. So if the Administrator sets up ICMP filters for
outbound/inbound (at the IP layer), what will be the observed behavior?
--Sowmini
^ permalink raw reply
* Re: [PATCH net] net: sched: shrink struct qdisc_skb_cb to 28 bytes
From: Eric Dumazet @ 2014-09-18 19:14 UTC (permalink / raw)
To: Joe Perches
Cc: Stephen Hemminger, Or Gerlitz, Govindarajulu Varadarajan,
Yinghai Lu, David Miller, NetDev, ssujith, gvaradar,
Christian Benvenuti (benve)
In-Reply-To: <1411063655.4290.18.camel@joe-AO725>
On Thu, 2014-09-18 at 11:07 -0700, Joe Perches wrote:
> > +static inline struct ipoib_cb *ipoib_skb_cb(const struct sk_buff *skb)
> > +{
> > + BUILD_BUG_ON(sizeof(skb->cb) < sizeof(struct ipoib_cb));
> > + return (struct ipoib_cb *)skb->cb;
> > +}
>
> It seems better not to use const for the struct sk_buff * here.
>
> Neither of the uses take a const struct sk_buff *
Thats pretty standard, check for other similar constructs like that.
static inline struct qdisc_skb_cb *qdisc_skb_cb(const struct sk_buff *skb)
{
return (struct qdisc_skb_cb *)skb->cb;
}
This allows uses of the helper when the skb is only read (has the const qual)
^ permalink raw reply
* Re: CPU scheduler to TXQ binding? (ixgbe vs. igb)
From: Jesper Dangaard Brouer @ 2014-09-18 18:57 UTC (permalink / raw)
To: Eric Dumazet; +Cc: Alexander Duyck, netdev@vger.kernel.org, Tom Herbert
In-Reply-To: <1411058064.7106.277.camel@edumazet-glaptop2.roam.corp.google.com>
On Thu, 18 Sep 2014 09:34:24 -0700 Eric Dumazet <eric.dumazet@gmail.com> wrote:
> On Thu, 2014-09-18 at 17:59 +0200, Jesper Dangaard Brouer wrote:
> > On Thu, 18 Sep 2014 08:42:31 -0700
> > Eric Dumazet <eric.dumazet@gmail.com> wrote:
> >
[...]
> > I'm looking for some 1G hardware without multiqueue, so I can get
> > around this measurement constraint. And possibly turning it down to
> > 100Mbit/s, so I can more easily measure the HoL blocking effect.
> >
>
> ethtool -L eth0 rx 1 tx 1
>
> (Or similar if combined is used)
Thanks! - that solves my qdisc measurement problem :-)
And yes, I had to use:
ethtool -L eth1 combined 1
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Sr. Network Kernel Developer at Red Hat
Author of http://www.iptv-analyzer.org
LinkedIn: http://www.linkedin.com/in/brouer
^ permalink raw reply
* Re: [PATCHv6 net-next 1/3] sunvnet: upgrade to VIO protocol version 1.6
From: Raghuram Kothakota @ 2014-09-18 18:49 UTC (permalink / raw)
To: David L Stevens; +Cc: David Miller, netdev
In-Reply-To: <541AD838.50700@oracle.com>
On Sep 18, 2014, at 6:03 AM, David L Stevens <david.stevens@oracle.com> wrote:
>
>
> On 09/18/2014 12:09 AM, Raghuram Kothakota wrote:
>
>>> @@ -1048,8 +1116,8 @@ static int vnet_port_alloc_tx_bufs(struct vnet_port *port)
>>> void *dring;
>>>
>>> for (i = 0; i < VNET_TX_RING_SIZE; i++) {
>>> - void *buf = kzalloc(ETH_FRAME_LEN + 8, GFP_KERNEL);
>>> - int map_len = (ETH_FRAME_LEN + 7) & ~7;
>>> + void *buf = kzalloc(VNET_MAXPACKET + 8, GFP_KERNEL);
>>
>>
>> This patch doesn't change the VNET_MAXPACKET to 64k, but the patch 2/3 changes
>> it to 64k+. Allocating buffers of size VNET_MAXPACKET always can consume too much
>> memory for every port/LDC, that would be more than 32MB. You may want to allocate
>> buffers based on the mtu that is negotiated, so that this memory used only when
>> such large packets are accepted by the peer.
>
> I originally had code to dynamically allocate them after the MTU negotiation, but
> that opens up a can of worms regarding stopping and freeing an active ring. I don't
> believe the shutdown code addresses this adequately, either, and I think this is
> worth addressing, but separately.
>
I am probably not as knowledgeable of sunvnet as you may be, but I would assume
the code is capable of handling the a vport removal and should have sufficient method
cleanup as well.
> I convinced myself to do it this way because:
> a) memory is cheap
In the virtualization world, we want resources to be efficiently used and memory is
still very important resource. My concern is mostly because this memory usage of
32+MB is on a per LDC basis. LDoms today supports a max of 128 domains, but
from my experience seen actual deployments of the order of 50 domains. This is
going up as the platforms getting more and more powerful. If there are really
that many peers, then the amount of memory consumed by one vnet instance
is 50 * 32+MB = 1.6GB+. It's fine if this memory is really used, but it seems like this
will be useful only when the peer is another linux guest with this version of vnet and
also the MTU is configured to use 64K. The memory is being wasted for all other
peers that either don't support 64K MTU or not configured to use it and also
the switch port as obviously it doesn't support 64K MTU today.
Note, it's just buffer space that is being consumed here, it is also the LDC shared
memory space. Luckily the SHADOW_MAPPED shared memory space has
very less limitations otherwise it can cause other impact to other virtual devices.
> b) I think most people will want to use large MTUs for performance; enough so
> that perhaps the bring-up MTU should be 64K too
>From my experience in SPARC world, most customers pushed us back for any
proposal to use Jumbo Frames. The customers who configured Jumbo frames,
mostly used 9K for performance of NFS etc.
> c) future (actually current) TSO/GSO work will want large buffers even if the MTU
> is not changed
>
When we implemented TSO support, we evaluated the cost of the buffers vs
performance. We were able to limit TSO support to 8K(actually bit less) and still
achieve high performance, for example we are able to drive line rate on a 10G
and guest-to-guest of the order of 45+Gbps. So, my suggestion would be to
increase the parallelism of the code more than depending on large MTU.
> So, if this is actually too much memory, I was more inclined to reduce the ring
> size rather than either add complicating code to handle active-ring reallocation
> that would typically be run once per boot, or another alternative of adding
> module parameters to specify the buffer size TSO/GSO will need 64K to perform
> well, regardless of the device MTU.
>
Note, my experience shows that reducing ring size can have larger impact on the
standard MTU case. I would assume the sunvnet code to be dynamic in any case
to deal with ports being added and removed and ports going down and up and
this is just one aspect of those operations.
-Raghuram
> +-DLS
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox