kernel_optimize_test/drivers/net/r8169.c

3210 lines
80 KiB
C
Raw Normal View History

/*
* r8169.c: RealTek 8169/8168/8101 ethernet driver.
*
* Copyright (c) 2002 ShuChen <shuchen@realtek.com.tw>
* Copyright (c) 2003 - 2007 Francois Romieu <romieu@fr.zoreil.com>
* Copyright (c) a lot of people too. Please respect their work.
*
* See MAINTAINERS file for support contact information.
*/
#include <linux/module.h>
#include <linux/moduleparam.h>
#include <linux/pci.h>
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/delay.h>
#include <linux/ethtool.h>
#include <linux/mii.h>
#include <linux/if_vlan.h>
#include <linux/crc32.h>
#include <linux/in.h>
#include <linux/ip.h>
#include <linux/tcp.h>
#include <linux/init.h>
#include <linux/dma-mapping.h>
#include <asm/system.h>
#include <asm/io.h>
#include <asm/irq.h>
#ifdef CONFIG_R8169_NAPI
#define NAPI_SUFFIX "-NAPI"
#else
#define NAPI_SUFFIX ""
#endif
#define RTL8169_VERSION "2.2LK" NAPI_SUFFIX
#define MODULENAME "r8169"
#define PFX MODULENAME ": "
#ifdef RTL8169_DEBUG
#define assert(expr) \
if (!(expr)) { \
printk( "Assertion failed! %s,%s,%s,line=%d\n", \
#expr,__FILE__,__FUNCTION__,__LINE__); \
}
#define dprintk(fmt, args...) \
do { printk(KERN_DEBUG PFX fmt, ## args); } while (0)
#else
#define assert(expr) do {} while (0)
#define dprintk(fmt, args...) do {} while (0)
#endif /* RTL8169_DEBUG */
#define R8169_MSG_DEFAULT \
(NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_IFUP | NETIF_MSG_IFDOWN)
#define TX_BUFFS_AVAIL(tp) \
(tp->dirty_tx + NUM_TX_DESC - tp->cur_tx - 1)
#ifdef CONFIG_R8169_NAPI
#define rtl8169_rx_skb netif_receive_skb
#define rtl8169_rx_hwaccel_skb vlan_hwaccel_receive_skb
#define rtl8169_rx_quota(count, quota) min(count, quota)
#else
#define rtl8169_rx_skb netif_rx
#define rtl8169_rx_hwaccel_skb vlan_hwaccel_rx
#define rtl8169_rx_quota(count, quota) count
#endif
/* Maximum events (Rx packets, etc.) to handle at each interrupt. */
2006-03-04 10:33:57 +08:00
static const int max_interrupt_work = 20;
/* Maximum number of multicast addresses to filter (vs. Rx-all-multicast).
The RTL chips use a 64 element hash table based on the Ethernet CRC. */
2006-03-04 10:33:57 +08:00
static const int multicast_filter_limit = 32;
/* MAC address length */
#define MAC_ADDR_LEN 6
#define RX_FIFO_THRESH 7 /* 7 means NO threshold, Rx buffer level before first PCI xfer. */
#define RX_DMA_BURST 6 /* Maximum PCI burst, '6' is 1024 */
#define TX_DMA_BURST 6 /* Maximum PCI burst, '6' is 1024 */
#define EarlyTxThld 0x3F /* 0x3F means NO early transmit */
#define RxPacketMaxSize 0x3FE8 /* 16K - 1 - ETH_HLEN - VLAN - CRC... */
#define SafeMtu 0x1c20 /* ... actually life sucks beyond ~7k */
#define InterFrameGap 0x03 /* 3 means InterFrameGap = the shortest one */
#define R8169_REGS_SIZE 256
#define R8169_NAPI_WEIGHT 64
#define NUM_TX_DESC 64 /* Number of Tx descriptor registers */
#define NUM_RX_DESC 256 /* Number of Rx descriptor registers */
#define RX_BUF_SIZE 1536 /* Rx Buffer size */
#define R8169_TX_RING_BYTES (NUM_TX_DESC * sizeof(struct TxDesc))
#define R8169_RX_RING_BYTES (NUM_RX_DESC * sizeof(struct RxDesc))
#define RTL8169_TX_TIMEOUT (6*HZ)
#define RTL8169_PHY_TIMEOUT (10*HZ)
/* write/read MMIO register */
#define RTL_W8(reg, val8) writeb ((val8), ioaddr + (reg))
#define RTL_W16(reg, val16) writew ((val16), ioaddr + (reg))
#define RTL_W32(reg, val32) writel ((val32), ioaddr + (reg))
#define RTL_R8(reg) readb (ioaddr + (reg))
#define RTL_R16(reg) readw (ioaddr + (reg))
#define RTL_R32(reg) ((unsigned long) readl (ioaddr + (reg)))
enum mac_version {
RTL_GIGA_MAC_VER_01 = 0x01, // 8169
RTL_GIGA_MAC_VER_02 = 0x02, // 8169S
RTL_GIGA_MAC_VER_03 = 0x03, // 8110S
RTL_GIGA_MAC_VER_04 = 0x04, // 8169SB
RTL_GIGA_MAC_VER_05 = 0x05, // 8110SCd
RTL_GIGA_MAC_VER_06 = 0x06, // 8110SCe
RTL_GIGA_MAC_VER_11 = 0x0b, // 8168Bb
RTL_GIGA_MAC_VER_12 = 0x0c, // 8168Be
RTL_GIGA_MAC_VER_13 = 0x0d, // 8101Eb
RTL_GIGA_MAC_VER_14 = 0x0e, // 8101 ?
RTL_GIGA_MAC_VER_15 = 0x0f, // 8101 ?
RTL_GIGA_MAC_VER_16 = 0x11, // 8101Ec
RTL_GIGA_MAC_VER_17 = 0x10, // 8168Bf
RTL_GIGA_MAC_VER_18 = 0x12, // 8168CP
RTL_GIGA_MAC_VER_19 = 0x13, // 8168C
RTL_GIGA_MAC_VER_20 = 0x14 // 8168C
};
#define _R(NAME,MAC,MASK) \
{ .name = NAME, .mac_version = MAC, .RxConfigMask = MASK }
static const struct {
const char *name;
u8 mac_version;
u32 RxConfigMask; /* Clears the bits supported by this chip */
} rtl_chip_info[] = {
_R("RTL8169", RTL_GIGA_MAC_VER_01, 0xff7e1880), // 8169
_R("RTL8169s", RTL_GIGA_MAC_VER_02, 0xff7e1880), // 8169S
_R("RTL8110s", RTL_GIGA_MAC_VER_03, 0xff7e1880), // 8110S
_R("RTL8169sb/8110sb", RTL_GIGA_MAC_VER_04, 0xff7e1880), // 8169SB
_R("RTL8169sc/8110sc", RTL_GIGA_MAC_VER_05, 0xff7e1880), // 8110SCd
_R("RTL8169sc/8110sc", RTL_GIGA_MAC_VER_06, 0xff7e1880), // 8110SCe
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
_R("RTL8168b/8111b", RTL_GIGA_MAC_VER_11, 0xff7e1880), // PCI-E
_R("RTL8168b/8111b", RTL_GIGA_MAC_VER_12, 0xff7e1880), // PCI-E
_R("RTL8101e", RTL_GIGA_MAC_VER_13, 0xff7e1880), // PCI-E 8139
_R("RTL8100e", RTL_GIGA_MAC_VER_14, 0xff7e1880), // PCI-E 8139
_R("RTL8100e", RTL_GIGA_MAC_VER_15, 0xff7e1880), // PCI-E 8139
_R("RTL8168b/8111b", RTL_GIGA_MAC_VER_17, 0xff7e1880), // PCI-E
_R("RTL8101e", RTL_GIGA_MAC_VER_16, 0xff7e1880), // PCI-E
_R("RTL8168cp/8111cp", RTL_GIGA_MAC_VER_18, 0xff7e1880), // PCI-E
_R("RTL8168c/8111c", RTL_GIGA_MAC_VER_19, 0xff7e1880), // PCI-E
_R("RTL8168c/8111c", RTL_GIGA_MAC_VER_20, 0xff7e1880) // PCI-E
};
#undef _R
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
enum cfg_version {
RTL_CFG_0 = 0x00,
RTL_CFG_1,
RTL_CFG_2
};
static void rtl_hw_start_8169(struct net_device *);
static void rtl_hw_start_8168(struct net_device *);
static void rtl_hw_start_8101(struct net_device *);
static struct pci_device_id rtl8169_pci_tbl[] = {
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
{ PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8129), 0, 0, RTL_CFG_0 },
{ PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8136), 0, 0, RTL_CFG_2 },
{ PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8167), 0, 0, RTL_CFG_0 },
{ PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8168), 0, 0, RTL_CFG_1 },
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
{ PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8169), 0, 0, RTL_CFG_0 },
{ PCI_DEVICE(PCI_VENDOR_ID_DLINK, 0x4300), 0, 0, RTL_CFG_0 },
{ PCI_DEVICE(PCI_VENDOR_ID_AT, 0xc107), 0, 0, RTL_CFG_0 },
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
{ PCI_DEVICE(0x16ec, 0x0116), 0, 0, RTL_CFG_0 },
{ PCI_VENDOR_ID_LINKSYS, 0x1032,
PCI_ANY_ID, 0x0024, 0, 0, RTL_CFG_0 },
{ 0x0001, 0x8168,
PCI_ANY_ID, 0x2410, 0, 0, RTL_CFG_2 },
{0,},
};
MODULE_DEVICE_TABLE(pci, rtl8169_pci_tbl);
static int rx_copybreak = 200;
static int use_dac;
static struct {
u32 msg_enable;
} debug = { -1 };
enum rtl_registers {
MAC0 = 0, /* Ethernet hardware address. */
MAC4 = 4,
MAR0 = 8, /* Multicast filter. */
CounterAddrLow = 0x10,
CounterAddrHigh = 0x14,
TxDescStartAddrLow = 0x20,
TxDescStartAddrHigh = 0x24,
TxHDescStartAddrLow = 0x28,
TxHDescStartAddrHigh = 0x2c,
FLASH = 0x30,
ERSR = 0x36,
ChipCmd = 0x37,
TxPoll = 0x38,
IntrMask = 0x3c,
IntrStatus = 0x3e,
TxConfig = 0x40,
RxConfig = 0x44,
RxMissed = 0x4c,
Cfg9346 = 0x50,
Config0 = 0x51,
Config1 = 0x52,
Config2 = 0x53,
Config3 = 0x54,
Config4 = 0x55,
Config5 = 0x56,
MultiIntr = 0x5c,
PHYAR = 0x60,
TBICSR = 0x64,
TBI_ANAR = 0x68,
TBI_LPAR = 0x6a,
PHYstatus = 0x6c,
RxMaxSize = 0xda,
CPlusCmd = 0xe0,
IntrMitigate = 0xe2,
RxDescAddrLow = 0xe4,
RxDescAddrHigh = 0xe8,
EarlyTxThres = 0xec,
FuncEvent = 0xf0,
FuncEventMask = 0xf4,
FuncPresetState = 0xf8,
FuncForceEvent = 0xfc,
};
enum rtl_register_content {
/* InterruptStatusBits */
SYSErr = 0x8000,
PCSTimeout = 0x4000,
SWInt = 0x0100,
TxDescUnavail = 0x0080,
RxFIFOOver = 0x0040,
LinkChg = 0x0020,
RxOverflow = 0x0010,
TxErr = 0x0008,
TxOK = 0x0004,
RxErr = 0x0002,
RxOK = 0x0001,
/* RxStatusDesc */
RxFOVF = (1 << 23),
RxRWT = (1 << 22),
RxRES = (1 << 21),
RxRUNT = (1 << 20),
RxCRC = (1 << 19),
/* ChipCmdBits */
CmdReset = 0x10,
CmdRxEnb = 0x08,
CmdTxEnb = 0x04,
RxBufEmpty = 0x01,
/* TXPoll register p.5 */
HPQ = 0x80, /* Poll cmd on the high prio queue */
NPQ = 0x40, /* Poll cmd on the low prio queue */
FSWInt = 0x01, /* Forced software interrupt */
/* Cfg9346Bits */
Cfg9346_Lock = 0x00,
Cfg9346_Unlock = 0xc0,
/* rx_mode_bits */
AcceptErr = 0x20,
AcceptRunt = 0x10,
AcceptBroadcast = 0x08,
AcceptMulticast = 0x04,
AcceptMyPhys = 0x02,
AcceptAllPhys = 0x01,
/* RxConfigBits */
RxCfgFIFOShift = 13,
RxCfgDMAShift = 8,
/* TxConfigBits */
TxInterFrameGapShift = 24,
TxDMAShift = 8, /* DMA burst value (0-7) is shift this many bits */
/* Config1 register p.24 */
MSIEnable = (1 << 5), /* Enable Message Signaled Interrupt */
PMEnable = (1 << 0), /* Power Management Enable */
/* Config2 register p. 25 */
PCI_Clock_66MHz = 0x01,
PCI_Clock_33MHz = 0x00,
/* Config3 register p.25 */
MagicPacket = (1 << 5), /* Wake up when receives a Magic Packet */
LinkUp = (1 << 4), /* Wake up when the cable connection is re-established */
/* Config5 register p.27 */
BWF = (1 << 6), /* Accept Broadcast wakeup frame */
MWF = (1 << 5), /* Accept Multicast wakeup frame */
UWF = (1 << 4), /* Accept Unicast wakeup frame */
LanWake = (1 << 1), /* LanWake enable/disable */
PMEStatus = (1 << 0), /* PME status can be reset by PCI RST# */
/* TBICSR p.28 */
TBIReset = 0x80000000,
TBILoopback = 0x40000000,
TBINwEnable = 0x20000000,
TBINwRestart = 0x10000000,
TBILinkOk = 0x02000000,
TBINwComplete = 0x01000000,
/* CPlusCmd p.31 */
PktCntrDisable = (1 << 7), // 8168
RxVlan = (1 << 6),
RxChkSum = (1 << 5),
PCIDAC = (1 << 4),
PCIMulRW = (1 << 3),
INTT_0 = 0x0000, // 8168
INTT_1 = 0x0001, // 8168
INTT_2 = 0x0002, // 8168
INTT_3 = 0x0003, // 8168
/* rtl8169_PHYstatus */
TBI_Enable = 0x80,
TxFlowCtrl = 0x40,
RxFlowCtrl = 0x20,
_1000bpsF = 0x10,
_100bps = 0x08,
_10bps = 0x04,
LinkStatus = 0x02,
FullDup = 0x01,
/* _TBICSRBit */
TBILinkOK = 0x02000000,
/* DumpCounterCommand */
CounterDump = 0x8,
};
enum desc_status_bit {
DescOwn = (1 << 31), /* Descriptor is owned by NIC */
RingEnd = (1 << 30), /* End of descriptor ring */
FirstFrag = (1 << 29), /* First segment of a packet */
LastFrag = (1 << 28), /* Final segment of a packet */
/* Tx private */
LargeSend = (1 << 27), /* TCP Large Send Offload (TSO) */
MSSShift = 16, /* MSS value position */
MSSMask = 0xfff, /* MSS value + LargeSend bit: 12 bits */
IPCS = (1 << 18), /* Calculate IP checksum */
UDPCS = (1 << 17), /* Calculate UDP/IP checksum */
TCPCS = (1 << 16), /* Calculate TCP/IP checksum */
TxVlanTag = (1 << 17), /* Add VLAN tag */
/* Rx private */
PID1 = (1 << 18), /* Protocol ID bit 1/2 */
PID0 = (1 << 17), /* Protocol ID bit 2/2 */
#define RxProtoUDP (PID1)
#define RxProtoTCP (PID0)
#define RxProtoIP (PID1 | PID0)
#define RxProtoMask RxProtoIP
IPFail = (1 << 16), /* IP checksum failed */
UDPFail = (1 << 15), /* UDP/IP checksum failed */
TCPFail = (1 << 14), /* TCP/IP checksum failed */
RxVlanTag = (1 << 16), /* VLAN tag available */
};
#define RsvdMask 0x3fffc000
struct TxDesc {
__le32 opts1;
__le32 opts2;
__le64 addr;
};
struct RxDesc {
__le32 opts1;
__le32 opts2;
__le64 addr;
};
struct ring_info {
struct sk_buff *skb;
u32 len;
u8 __pad[sizeof(void *) - sizeof(u32)];
};
enum features {
RTL_FEATURE_WOL = (1 << 0),
RTL_FEATURE_MSI = (1 << 1),
};
struct rtl8169_private {
void __iomem *mmio_addr; /* memory map physical address */
struct pci_dev *pci_dev; /* Index of PCI device */
struct net_device *dev;
#ifdef CONFIG_R8169_NAPI
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
struct napi_struct napi;
#endif
spinlock_t lock; /* spin lock flag */
u32 msg_enable;
int chipset;
int mac_version;
u32 cur_rx; /* Index into the Rx descriptor buffer of next Rx pkt. */
u32 cur_tx; /* Index into the Tx descriptor buffer of next Rx pkt. */
u32 dirty_rx;
u32 dirty_tx;
struct TxDesc *TxDescArray; /* 256-aligned Tx descriptor ring */
struct RxDesc *RxDescArray; /* 256-aligned Rx descriptor ring */
dma_addr_t TxPhyAddr;
dma_addr_t RxPhyAddr;
struct sk_buff *Rx_skbuff[NUM_RX_DESC]; /* Rx data buffers */
struct ring_info tx_skb[NUM_TX_DESC]; /* Tx data buffers */
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
unsigned align;
unsigned rx_buf_sz;
struct timer_list timer;
u16 cp_cmd;
u16 intr_event;
u16 napi_event;
u16 intr_mask;
int phy_auto_nego_reg;
int phy_1000_ctrl_reg;
#ifdef CONFIG_R8169_VLAN
struct vlan_group *vlgrp;
#endif
int (*set_speed)(struct net_device *, u8 autoneg, u16 speed, u8 duplex);
void (*get_settings)(struct net_device *, struct ethtool_cmd *);
void (*phy_reset_enable)(void __iomem *);
void (*hw_start)(struct net_device *);
unsigned int (*phy_reset_pending)(void __iomem *);
unsigned int (*link_ok)(void __iomem *);
struct delayed_work task;
unsigned features;
};
MODULE_AUTHOR("Realtek and the Linux r8169 crew <netdev@vger.kernel.org>");
MODULE_DESCRIPTION("RealTek RTL-8169 Gigabit Ethernet driver");
module_param(rx_copybreak, int, 0);
MODULE_PARM_DESC(rx_copybreak, "Copy breakpoint for copy-only-tiny-frames");
module_param(use_dac, int, 0);
MODULE_PARM_DESC(use_dac, "Enable PCI DAC. Unsafe on 32 bit PCI slot.");
module_param_named(debug, debug.msg_enable, int, 0);
MODULE_PARM_DESC(debug, "Debug verbosity level (0=none, ..., 16=all)");
MODULE_LICENSE("GPL");
MODULE_VERSION(RTL8169_VERSION);
static int rtl8169_open(struct net_device *dev);
static int rtl8169_start_xmit(struct sk_buff *skb, struct net_device *dev);
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 21:55:46 +08:00
static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance);
static int rtl8169_init_ring(struct net_device *dev);
static void rtl_hw_start(struct net_device *dev);
static int rtl8169_close(struct net_device *dev);
static void rtl_set_rx_mode(struct net_device *dev);
static void rtl8169_tx_timeout(struct net_device *dev);
static struct net_device_stats *rtl8169_get_stats(struct net_device *dev);
static int rtl8169_rx_interrupt(struct net_device *, struct rtl8169_private *,
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
void __iomem *, u32 budget);
static int rtl8169_change_mtu(struct net_device *dev, int new_mtu);
static void rtl8169_down(struct net_device *dev);
static void rtl8169_rx_clear(struct rtl8169_private *tp);
#ifdef CONFIG_R8169_NAPI
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
static int rtl8169_poll(struct napi_struct *napi, int budget);
#endif
static const unsigned int rtl8169_rx_config =
(RX_FIFO_THRESH << RxCfgFIFOShift) | (RX_DMA_BURST << RxCfgDMAShift);
static void mdio_write(void __iomem *ioaddr, int reg_addr, int value)
{
int i;
RTL_W32(PHYAR, 0x80000000 | (reg_addr & 0x1f) << 16 | (value & 0xffff));
for (i = 20; i > 0; i--) {
/*
* Check if the RTL8169 has completed writing to the specified
* MII register.
*/
if (!(RTL_R32(PHYAR) & 0x80000000))
break;
udelay(25);
}
}
static int mdio_read(void __iomem *ioaddr, int reg_addr)
{
int i, value = -1;
RTL_W32(PHYAR, 0x0 | (reg_addr & 0x1f) << 16);
for (i = 20; i > 0; i--) {
/*
* Check if the RTL8169 has completed retrieving data from
* the specified MII register.
*/
if (RTL_R32(PHYAR) & 0x80000000) {
value = RTL_R32(PHYAR) & 0xffff;
break;
}
udelay(25);
}
return value;
}
static void rtl8169_irq_mask_and_ack(void __iomem *ioaddr)
{
RTL_W16(IntrMask, 0x0000);
RTL_W16(IntrStatus, 0xffff);
}
static void rtl8169_asic_down(void __iomem *ioaddr)
{
RTL_W8(ChipCmd, 0x00);
rtl8169_irq_mask_and_ack(ioaddr);
RTL_R16(CPlusCmd);
}
static unsigned int rtl8169_tbi_reset_pending(void __iomem *ioaddr)
{
return RTL_R32(TBICSR) & TBIReset;
}
static unsigned int rtl8169_xmii_reset_pending(void __iomem *ioaddr)
{
return mdio_read(ioaddr, MII_BMCR) & BMCR_RESET;
}
static unsigned int rtl8169_tbi_link_ok(void __iomem *ioaddr)
{
return RTL_R32(TBICSR) & TBILinkOk;
}
static unsigned int rtl8169_xmii_link_ok(void __iomem *ioaddr)
{
return RTL_R8(PHYstatus) & LinkStatus;
}
static void rtl8169_tbi_reset_enable(void __iomem *ioaddr)
{
RTL_W32(TBICSR, RTL_R32(TBICSR) | TBIReset);
}
static void rtl8169_xmii_reset_enable(void __iomem *ioaddr)
{
unsigned int val;
val = mdio_read(ioaddr, MII_BMCR) | BMCR_RESET;
mdio_write(ioaddr, MII_BMCR, val & 0xffff);
}
static void rtl8169_check_link_status(struct net_device *dev,
struct rtl8169_private *tp,
void __iomem *ioaddr)
{
unsigned long flags;
spin_lock_irqsave(&tp->lock, flags);
if (tp->link_ok(ioaddr)) {
netif_carrier_on(dev);
if (netif_msg_ifup(tp))
printk(KERN_INFO PFX "%s: link up\n", dev->name);
} else {
if (netif_msg_ifdown(tp))
printk(KERN_INFO PFX "%s: link down\n", dev->name);
netif_carrier_off(dev);
}
spin_unlock_irqrestore(&tp->lock, flags);
}
static void rtl8169_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
u8 options;
wol->wolopts = 0;
#define WAKE_ANY (WAKE_PHY | WAKE_MAGIC | WAKE_UCAST | WAKE_BCAST | WAKE_MCAST)
wol->supported = WAKE_ANY;
spin_lock_irq(&tp->lock);
options = RTL_R8(Config1);
if (!(options & PMEnable))
goto out_unlock;
options = RTL_R8(Config3);
if (options & LinkUp)
wol->wolopts |= WAKE_PHY;
if (options & MagicPacket)
wol->wolopts |= WAKE_MAGIC;
options = RTL_R8(Config5);
if (options & UWF)
wol->wolopts |= WAKE_UCAST;
if (options & BWF)
wol->wolopts |= WAKE_BCAST;
if (options & MWF)
wol->wolopts |= WAKE_MCAST;
out_unlock:
spin_unlock_irq(&tp->lock);
}
static int rtl8169_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
unsigned int i;
static struct {
u32 opt;
u16 reg;
u8 mask;
} cfg[] = {
{ WAKE_ANY, Config1, PMEnable },
{ WAKE_PHY, Config3, LinkUp },
{ WAKE_MAGIC, Config3, MagicPacket },
{ WAKE_UCAST, Config5, UWF },
{ WAKE_BCAST, Config5, BWF },
{ WAKE_MCAST, Config5, MWF },
{ WAKE_ANY, Config5, LanWake }
};
spin_lock_irq(&tp->lock);
RTL_W8(Cfg9346, Cfg9346_Unlock);
for (i = 0; i < ARRAY_SIZE(cfg); i++) {
u8 options = RTL_R8(cfg[i].reg) & ~cfg[i].mask;
if (wol->wolopts & cfg[i].opt)
options |= cfg[i].mask;
RTL_W8(cfg[i].reg, options);
}
RTL_W8(Cfg9346, Cfg9346_Lock);
if (wol->wolopts)
tp->features |= RTL_FEATURE_WOL;
else
tp->features &= ~RTL_FEATURE_WOL;
spin_unlock_irq(&tp->lock);
return 0;
}
static void rtl8169_get_drvinfo(struct net_device *dev,
struct ethtool_drvinfo *info)
{
struct rtl8169_private *tp = netdev_priv(dev);
strcpy(info->driver, MODULENAME);
strcpy(info->version, RTL8169_VERSION);
strcpy(info->bus_info, pci_name(tp->pci_dev));
}
static int rtl8169_get_regs_len(struct net_device *dev)
{
return R8169_REGS_SIZE;
}
static int rtl8169_set_speed_tbi(struct net_device *dev,
u8 autoneg, u16 speed, u8 duplex)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
int ret = 0;
u32 reg;
reg = RTL_R32(TBICSR);
if ((autoneg == AUTONEG_DISABLE) && (speed == SPEED_1000) &&
(duplex == DUPLEX_FULL)) {
RTL_W32(TBICSR, reg & ~(TBINwEnable | TBINwRestart));
} else if (autoneg == AUTONEG_ENABLE)
RTL_W32(TBICSR, reg | TBINwEnable | TBINwRestart);
else {
if (netif_msg_link(tp)) {
printk(KERN_WARNING "%s: "
"incorrect speed setting refused in TBI mode\n",
dev->name);
}
ret = -EOPNOTSUPP;
}
return ret;
}
static int rtl8169_set_speed_xmii(struct net_device *dev,
u8 autoneg, u16 speed, u8 duplex)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
int auto_nego, giga_ctrl;
auto_nego = mdio_read(ioaddr, MII_ADVERTISE);
auto_nego &= ~(ADVERTISE_10HALF | ADVERTISE_10FULL |
ADVERTISE_100HALF | ADVERTISE_100FULL);
giga_ctrl = mdio_read(ioaddr, MII_CTRL1000);
giga_ctrl &= ~(ADVERTISE_1000FULL | ADVERTISE_1000HALF);
if (autoneg == AUTONEG_ENABLE) {
auto_nego |= (ADVERTISE_10HALF | ADVERTISE_10FULL |
ADVERTISE_100HALF | ADVERTISE_100FULL);
giga_ctrl |= ADVERTISE_1000FULL | ADVERTISE_1000HALF;
} else {
if (speed == SPEED_10)
auto_nego |= ADVERTISE_10HALF | ADVERTISE_10FULL;
else if (speed == SPEED_100)
auto_nego |= ADVERTISE_100HALF | ADVERTISE_100FULL;
else if (speed == SPEED_1000)
giga_ctrl |= ADVERTISE_1000FULL | ADVERTISE_1000HALF;
if (duplex == DUPLEX_HALF)
auto_nego &= ~(ADVERTISE_10FULL | ADVERTISE_100FULL);
if (duplex == DUPLEX_FULL)
auto_nego &= ~(ADVERTISE_10HALF | ADVERTISE_100HALF);
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
/* This tweak comes straight from Realtek's driver. */
if ((speed == SPEED_100) && (duplex == DUPLEX_HALF) &&
((tp->mac_version == RTL_GIGA_MAC_VER_13) ||
(tp->mac_version == RTL_GIGA_MAC_VER_16))) {
auto_nego = ADVERTISE_100HALF | ADVERTISE_CSMA;
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
}
}
/* The 8100e/8101e do Fast Ethernet only. */
if ((tp->mac_version == RTL_GIGA_MAC_VER_13) ||
(tp->mac_version == RTL_GIGA_MAC_VER_14) ||
(tp->mac_version == RTL_GIGA_MAC_VER_15) ||
(tp->mac_version == RTL_GIGA_MAC_VER_16)) {
if ((giga_ctrl & (ADVERTISE_1000FULL | ADVERTISE_1000HALF)) &&
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
netif_msg_link(tp)) {
printk(KERN_INFO "%s: PHY does not support 1000Mbps.\n",
dev->name);
}
giga_ctrl &= ~(ADVERTISE_1000FULL | ADVERTISE_1000HALF);
}
auto_nego |= ADVERTISE_PAUSE_CAP | ADVERTISE_PAUSE_ASYM;
if ((tp->mac_version == RTL_GIGA_MAC_VER_12) ||
(tp->mac_version == RTL_GIGA_MAC_VER_17)) {
/* Vendor specific (0x1f) and reserved (0x0e) MII registers. */
mdio_write(ioaddr, 0x1f, 0x0000);
mdio_write(ioaddr, 0x0e, 0x0000);
}
tp->phy_auto_nego_reg = auto_nego;
tp->phy_1000_ctrl_reg = giga_ctrl;
mdio_write(ioaddr, MII_ADVERTISE, auto_nego);
mdio_write(ioaddr, MII_CTRL1000, giga_ctrl);
mdio_write(ioaddr, MII_BMCR, BMCR_ANENABLE | BMCR_ANRESTART);
return 0;
}
static int rtl8169_set_speed(struct net_device *dev,
u8 autoneg, u16 speed, u8 duplex)
{
struct rtl8169_private *tp = netdev_priv(dev);
int ret;
ret = tp->set_speed(dev, autoneg, speed, duplex);
if (netif_running(dev) && (tp->phy_1000_ctrl_reg & ADVERTISE_1000FULL))
mod_timer(&tp->timer, jiffies + RTL8169_PHY_TIMEOUT);
return ret;
}
static int rtl8169_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
struct rtl8169_private *tp = netdev_priv(dev);
unsigned long flags;
int ret;
spin_lock_irqsave(&tp->lock, flags);
ret = rtl8169_set_speed(dev, cmd->autoneg, cmd->speed, cmd->duplex);
spin_unlock_irqrestore(&tp->lock, flags);
return ret;
}
static u32 rtl8169_get_rx_csum(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
return tp->cp_cmd & RxChkSum;
}
static int rtl8169_set_rx_csum(struct net_device *dev, u32 data)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
unsigned long flags;
spin_lock_irqsave(&tp->lock, flags);
if (data)
tp->cp_cmd |= RxChkSum;
else
tp->cp_cmd &= ~RxChkSum;
RTL_W16(CPlusCmd, tp->cp_cmd);
RTL_R16(CPlusCmd);
spin_unlock_irqrestore(&tp->lock, flags);
return 0;
}
#ifdef CONFIG_R8169_VLAN
static inline u32 rtl8169_tx_vlan_tag(struct rtl8169_private *tp,
struct sk_buff *skb)
{
return (tp->vlgrp && vlan_tx_tag_present(skb)) ?
TxVlanTag | swab16(vlan_tx_tag_get(skb)) : 0x00;
}
static void rtl8169_vlan_rx_register(struct net_device *dev,
struct vlan_group *grp)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
unsigned long flags;
spin_lock_irqsave(&tp->lock, flags);
tp->vlgrp = grp;
if (tp->vlgrp)
tp->cp_cmd |= RxVlan;
else
tp->cp_cmd &= ~RxVlan;
RTL_W16(CPlusCmd, tp->cp_cmd);
RTL_R16(CPlusCmd);
spin_unlock_irqrestore(&tp->lock, flags);
}
static int rtl8169_rx_vlan_skb(struct rtl8169_private *tp, struct RxDesc *desc,
struct sk_buff *skb)
{
u32 opts2 = le32_to_cpu(desc->opts2);
int ret;
if (tp->vlgrp && (opts2 & RxVlanTag)) {
rtl8169_rx_hwaccel_skb(skb, tp->vlgrp, swab16(opts2 & 0xffff));
ret = 0;
} else
ret = -1;
desc->opts2 = 0;
return ret;
}
#else /* !CONFIG_R8169_VLAN */
static inline u32 rtl8169_tx_vlan_tag(struct rtl8169_private *tp,
struct sk_buff *skb)
{
return 0;
}
static int rtl8169_rx_vlan_skb(struct rtl8169_private *tp, struct RxDesc *desc,
struct sk_buff *skb)
{
return -1;
}
#endif
static void rtl8169_gset_tbi(struct net_device *dev, struct ethtool_cmd *cmd)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
u32 status;
cmd->supported =
SUPPORTED_1000baseT_Full | SUPPORTED_Autoneg | SUPPORTED_FIBRE;
cmd->port = PORT_FIBRE;
cmd->transceiver = XCVR_INTERNAL;
status = RTL_R32(TBICSR);
cmd->advertising = (status & TBINwEnable) ? ADVERTISED_Autoneg : 0;
cmd->autoneg = !!(status & TBINwEnable);
cmd->speed = SPEED_1000;
cmd->duplex = DUPLEX_FULL; /* Always set */
}
static void rtl8169_gset_xmii(struct net_device *dev, struct ethtool_cmd *cmd)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
u8 status;
cmd->supported = SUPPORTED_10baseT_Half |
SUPPORTED_10baseT_Full |
SUPPORTED_100baseT_Half |
SUPPORTED_100baseT_Full |
SUPPORTED_1000baseT_Full |
SUPPORTED_Autoneg |
SUPPORTED_TP;
cmd->autoneg = 1;
cmd->advertising = ADVERTISED_TP | ADVERTISED_Autoneg;
if (tp->phy_auto_nego_reg & ADVERTISE_10HALF)
cmd->advertising |= ADVERTISED_10baseT_Half;
if (tp->phy_auto_nego_reg & ADVERTISE_10FULL)
cmd->advertising |= ADVERTISED_10baseT_Full;
if (tp->phy_auto_nego_reg & ADVERTISE_100HALF)
cmd->advertising |= ADVERTISED_100baseT_Half;
if (tp->phy_auto_nego_reg & ADVERTISE_100FULL)
cmd->advertising |= ADVERTISED_100baseT_Full;
if (tp->phy_1000_ctrl_reg & ADVERTISE_1000FULL)
cmd->advertising |= ADVERTISED_1000baseT_Full;
status = RTL_R8(PHYstatus);
if (status & _1000bpsF)
cmd->speed = SPEED_1000;
else if (status & _100bps)
cmd->speed = SPEED_100;
else if (status & _10bps)
cmd->speed = SPEED_10;
if (status & TxFlowCtrl)
cmd->advertising |= ADVERTISED_Asym_Pause;
if (status & RxFlowCtrl)
cmd->advertising |= ADVERTISED_Pause;
cmd->duplex = ((status & _1000bpsF) || (status & FullDup)) ?
DUPLEX_FULL : DUPLEX_HALF;
}
static int rtl8169_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
{
struct rtl8169_private *tp = netdev_priv(dev);
unsigned long flags;
spin_lock_irqsave(&tp->lock, flags);
tp->get_settings(dev, cmd);
spin_unlock_irqrestore(&tp->lock, flags);
return 0;
}
static void rtl8169_get_regs(struct net_device *dev, struct ethtool_regs *regs,
void *p)
{
struct rtl8169_private *tp = netdev_priv(dev);
unsigned long flags;
if (regs->len > R8169_REGS_SIZE)
regs->len = R8169_REGS_SIZE;
spin_lock_irqsave(&tp->lock, flags);
memcpy_fromio(p, tp->mmio_addr, regs->len);
spin_unlock_irqrestore(&tp->lock, flags);
}
static u32 rtl8169_get_msglevel(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
return tp->msg_enable;
}
static void rtl8169_set_msglevel(struct net_device *dev, u32 value)
{
struct rtl8169_private *tp = netdev_priv(dev);
tp->msg_enable = value;
}
static const char rtl8169_gstrings[][ETH_GSTRING_LEN] = {
"tx_packets",
"rx_packets",
"tx_errors",
"rx_errors",
"rx_missed",
"align_errors",
"tx_single_collisions",
"tx_multi_collisions",
"unicast",
"broadcast",
"multicast",
"tx_aborted",
"tx_underrun",
};
struct rtl8169_counters {
__le64 tx_packets;
__le64 rx_packets;
__le64 tx_errors;
__le32 rx_errors;
__le16 rx_missed;
__le16 align_errors;
__le32 tx_one_collision;
__le32 tx_multi_collision;
__le64 rx_unicast;
__le64 rx_broadcast;
__le32 rx_multicast;
__le16 tx_aborted;
__le16 tx_underun;
};
static int rtl8169_get_sset_count(struct net_device *dev, int sset)
{
switch (sset) {
case ETH_SS_STATS:
return ARRAY_SIZE(rtl8169_gstrings);
default:
return -EOPNOTSUPP;
}
}
static void rtl8169_get_ethtool_stats(struct net_device *dev,
struct ethtool_stats *stats, u64 *data)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
struct rtl8169_counters *counters;
dma_addr_t paddr;
u32 cmd;
ASSERT_RTNL();
counters = pci_alloc_consistent(tp->pci_dev, sizeof(*counters), &paddr);
if (!counters)
return;
RTL_W32(CounterAddrHigh, (u64)paddr >> 32);
cmd = (u64)paddr & DMA_32BIT_MASK;
RTL_W32(CounterAddrLow, cmd);
RTL_W32(CounterAddrLow, cmd | CounterDump);
while (RTL_R32(CounterAddrLow) & CounterDump) {
if (msleep_interruptible(1))
break;
}
RTL_W32(CounterAddrLow, 0);
RTL_W32(CounterAddrHigh, 0);
data[0] = le64_to_cpu(counters->tx_packets);
data[1] = le64_to_cpu(counters->rx_packets);
data[2] = le64_to_cpu(counters->tx_errors);
data[3] = le32_to_cpu(counters->rx_errors);
data[4] = le16_to_cpu(counters->rx_missed);
data[5] = le16_to_cpu(counters->align_errors);
data[6] = le32_to_cpu(counters->tx_one_collision);
data[7] = le32_to_cpu(counters->tx_multi_collision);
data[8] = le64_to_cpu(counters->rx_unicast);
data[9] = le64_to_cpu(counters->rx_broadcast);
data[10] = le32_to_cpu(counters->rx_multicast);
data[11] = le16_to_cpu(counters->tx_aborted);
data[12] = le16_to_cpu(counters->tx_underun);
pci_free_consistent(tp->pci_dev, sizeof(*counters), counters, paddr);
}
static void rtl8169_get_strings(struct net_device *dev, u32 stringset, u8 *data)
{
switch(stringset) {
case ETH_SS_STATS:
memcpy(data, *rtl8169_gstrings, sizeof(rtl8169_gstrings));
break;
}
}
static const struct ethtool_ops rtl8169_ethtool_ops = {
.get_drvinfo = rtl8169_get_drvinfo,
.get_regs_len = rtl8169_get_regs_len,
.get_link = ethtool_op_get_link,
.get_settings = rtl8169_get_settings,
.set_settings = rtl8169_set_settings,
.get_msglevel = rtl8169_get_msglevel,
.set_msglevel = rtl8169_set_msglevel,
.get_rx_csum = rtl8169_get_rx_csum,
.set_rx_csum = rtl8169_set_rx_csum,
.set_tx_csum = ethtool_op_set_tx_csum,
.set_sg = ethtool_op_set_sg,
.set_tso = ethtool_op_set_tso,
.get_regs = rtl8169_get_regs,
.get_wol = rtl8169_get_wol,
.set_wol = rtl8169_set_wol,
.get_strings = rtl8169_get_strings,
.get_sset_count = rtl8169_get_sset_count,
.get_ethtool_stats = rtl8169_get_ethtool_stats,
};
static void rtl8169_write_gmii_reg_bit(void __iomem *ioaddr, int reg,
int bitnum, int bitval)
{
int val;
val = mdio_read(ioaddr, reg);
val = (bitval == 1) ?
val | (bitval << bitnum) : val & ~(0x0001 << bitnum);
mdio_write(ioaddr, reg, val & 0xffff);
}
static void rtl8169_get_mac_version(struct rtl8169_private *tp,
void __iomem *ioaddr)
{
/*
* The driver currently handles the 8168Bf and the 8168Be identically
* but they can be identified more specifically through the test below
* if needed:
*
* (RTL_R32(TxConfig) & 0x700000) == 0x500000 ? 8168Bf : 8168Be
*
* Same thing for the 8101Eb and the 8101Ec:
*
* (RTL_R32(TxConfig) & 0x700000) == 0x200000 ? 8101Eb : 8101Ec
*/
const struct {
u32 mask;
u32 val;
int mac_version;
} mac_info[] = {
/* 8168B family. */
{ 0x7c800000, 0x3c800000, RTL_GIGA_MAC_VER_18 },
{ 0x7cf00000, 0x3c000000, RTL_GIGA_MAC_VER_19 },
{ 0x7cf00000, 0x3c200000, RTL_GIGA_MAC_VER_20 },
{ 0x7c800000, 0x3c000000, RTL_GIGA_MAC_VER_20 },
/* 8168B family. */
{ 0x7cf00000, 0x38000000, RTL_GIGA_MAC_VER_12 },
{ 0x7cf00000, 0x38500000, RTL_GIGA_MAC_VER_17 },
{ 0x7c800000, 0x38000000, RTL_GIGA_MAC_VER_17 },
{ 0x7c800000, 0x30000000, RTL_GIGA_MAC_VER_11 },
/* 8101 family. */
{ 0x7cf00000, 0x34000000, RTL_GIGA_MAC_VER_13 },
{ 0x7cf00000, 0x34200000, RTL_GIGA_MAC_VER_16 },
{ 0x7c800000, 0x34000000, RTL_GIGA_MAC_VER_16 },
/* FIXME: where did these entries come from ? -- FR */
{ 0xfc800000, 0x38800000, RTL_GIGA_MAC_VER_15 },
{ 0xfc800000, 0x30800000, RTL_GIGA_MAC_VER_14 },
/* 8110 family. */
{ 0xfc800000, 0x98000000, RTL_GIGA_MAC_VER_06 },
{ 0xfc800000, 0x18000000, RTL_GIGA_MAC_VER_05 },
{ 0xfc800000, 0x10000000, RTL_GIGA_MAC_VER_04 },
{ 0xfc800000, 0x04000000, RTL_GIGA_MAC_VER_03 },
{ 0xfc800000, 0x00800000, RTL_GIGA_MAC_VER_02 },
{ 0xfc800000, 0x00000000, RTL_GIGA_MAC_VER_01 },
{ 0x00000000, 0x00000000, RTL_GIGA_MAC_VER_01 } /* Catch-all */
}, *p = mac_info;
u32 reg;
reg = RTL_R32(TxConfig);
while ((reg & p->mask) != p->val)
p++;
tp->mac_version = p->mac_version;
if (p->mask == 0x00000000) {
struct pci_dev *pdev = tp->pci_dev;
dev_info(&pdev->dev, "unknown MAC (%08x)\n", reg);
}
}
static void rtl8169_print_mac_version(struct rtl8169_private *tp)
{
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
dprintk("mac_version = 0x%02x\n", tp->mac_version);
}
struct phy_reg {
u16 reg;
u16 val;
};
static void rtl_phy_write(void __iomem *ioaddr, struct phy_reg *regs, int len)
{
while (len-- > 0) {
mdio_write(ioaddr, regs->reg, regs->val);
regs++;
}
}
static void rtl8169s_hw_phy_config(void __iomem *ioaddr)
{
struct {
u16 regs[5]; /* Beware of bit-sign propagation */
} phy_magic[5] = { {
{ 0x0000, //w 4 15 12 0
0x00a1, //w 3 15 0 00a1
0x0008, //w 2 15 0 0008
0x1020, //w 1 15 0 1020
0x1000 } },{ //w 0 15 0 1000
{ 0x7000, //w 4 15 12 7
0xff41, //w 3 15 0 ff41
0xde60, //w 2 15 0 de60
0x0140, //w 1 15 0 0140
0x0077 } },{ //w 0 15 0 0077
{ 0xa000, //w 4 15 12 a
0xdf01, //w 3 15 0 df01
0xdf20, //w 2 15 0 df20
0xff95, //w 1 15 0 ff95
0xfa00 } },{ //w 0 15 0 fa00
{ 0xb000, //w 4 15 12 b
0xff41, //w 3 15 0 ff41
0xde20, //w 2 15 0 de20
0x0140, //w 1 15 0 0140
0x00bb } },{ //w 0 15 0 00bb
{ 0xf000, //w 4 15 12 f
0xdf01, //w 3 15 0 df01
0xdf20, //w 2 15 0 df20
0xff95, //w 1 15 0 ff95
0xbf00 } //w 0 15 0 bf00
}
}, *p = phy_magic;
unsigned int i;
mdio_write(ioaddr, 0x1f, 0x0001); //w 31 2 0 1
mdio_write(ioaddr, 0x15, 0x1000); //w 21 15 0 1000
mdio_write(ioaddr, 0x18, 0x65c7); //w 24 15 0 65c7
rtl8169_write_gmii_reg_bit(ioaddr, 4, 11, 0); //w 4 11 11 0
for (i = 0; i < ARRAY_SIZE(phy_magic); i++, p++) {
int val, pos = 4;
val = (mdio_read(ioaddr, pos) & 0x0fff) | (p->regs[0] & 0xffff);
mdio_write(ioaddr, pos, val);
while (--pos >= 0)
mdio_write(ioaddr, pos, p->regs[4 - pos] & 0xffff);
rtl8169_write_gmii_reg_bit(ioaddr, 4, 11, 1); //w 4 11 11 1
rtl8169_write_gmii_reg_bit(ioaddr, 4, 11, 0); //w 4 11 11 0
}
mdio_write(ioaddr, 0x1f, 0x0000); //w 31 2 0 0
}
static void rtl8169sb_hw_phy_config(void __iomem *ioaddr)
{
struct phy_reg phy_reg_init[] = {
{ 0x1f, 0x0002 },
{ 0x01, 0x90d0 },
{ 0x1f, 0x0000 }
};
rtl_phy_write(ioaddr, phy_reg_init, ARRAY_SIZE(phy_reg_init));
}
static void rtl8168cp_hw_phy_config(void __iomem *ioaddr)
{
struct phy_reg phy_reg_init[] = {
{ 0x1f, 0x0000 },
{ 0x1d, 0x0f00 },
{ 0x1f, 0x0002 },
{ 0x0c, 0x1ec8 },
{ 0x1f, 0x0000 }
};
rtl_phy_write(ioaddr, phy_reg_init, ARRAY_SIZE(phy_reg_init));
}
static void rtl8168c_hw_phy_config(void __iomem *ioaddr)
{
struct phy_reg phy_reg_init[] = {
{ 0x1f, 0x0001 },
{ 0x12, 0x2300 },
{ 0x1f, 0x0002 },
{ 0x00, 0x88d4 },
{ 0x01, 0x82b1 },
{ 0x03, 0x7002 },
{ 0x08, 0x9e30 },
{ 0x09, 0x01f0 },
{ 0x0a, 0x5500 },
{ 0x0c, 0x00c8 },
{ 0x1f, 0x0003 },
{ 0x12, 0xc096 },
{ 0x16, 0x000a },
{ 0x1f, 0x0000 }
};
rtl_phy_write(ioaddr, phy_reg_init, ARRAY_SIZE(phy_reg_init));
}
static void rtl8168cx_hw_phy_config(void __iomem *ioaddr)
{
struct phy_reg phy_reg_init[] = {
{ 0x1f, 0x0000 },
{ 0x12, 0x2300 },
{ 0x1f, 0x0003 },
{ 0x16, 0x0f0a },
{ 0x1f, 0x0000 },
{ 0x1f, 0x0002 },
{ 0x0c, 0x7eb8 },
{ 0x1f, 0x0000 }
};
rtl_phy_write(ioaddr, phy_reg_init, ARRAY_SIZE(phy_reg_init));
}
static void rtl_hw_phy_config(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
rtl8169_print_mac_version(tp);
switch (tp->mac_version) {
case RTL_GIGA_MAC_VER_01:
break;
case RTL_GIGA_MAC_VER_02:
case RTL_GIGA_MAC_VER_03:
rtl8169s_hw_phy_config(ioaddr);
break;
case RTL_GIGA_MAC_VER_04:
rtl8169sb_hw_phy_config(ioaddr);
break;
case RTL_GIGA_MAC_VER_18:
rtl8168cp_hw_phy_config(ioaddr);
break;
case RTL_GIGA_MAC_VER_19:
rtl8168c_hw_phy_config(ioaddr);
break;
case RTL_GIGA_MAC_VER_20:
rtl8168cx_hw_phy_config(ioaddr);
break;
default:
break;
}
}
static void rtl8169_phy_timer(unsigned long __opaque)
{
struct net_device *dev = (struct net_device *)__opaque;
struct rtl8169_private *tp = netdev_priv(dev);
struct timer_list *timer = &tp->timer;
void __iomem *ioaddr = tp->mmio_addr;
unsigned long timeout = RTL8169_PHY_TIMEOUT;
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
assert(tp->mac_version > RTL_GIGA_MAC_VER_01);
if (!(tp->phy_1000_ctrl_reg & ADVERTISE_1000FULL))
return;
spin_lock_irq(&tp->lock);
if (tp->phy_reset_pending(ioaddr)) {
/*
* A busy loop could burn quite a few cycles on nowadays CPU.
* Let's delay the execution of the timer for a few ticks.
*/
timeout = HZ/10;
goto out_mod_timer;
}
if (tp->link_ok(ioaddr))
goto out_unlock;
if (netif_msg_link(tp))
printk(KERN_WARNING "%s: PHY reset until link up\n", dev->name);
tp->phy_reset_enable(ioaddr);
out_mod_timer:
mod_timer(timer, jiffies + timeout);
out_unlock:
spin_unlock_irq(&tp->lock);
}
static inline void rtl8169_delete_timer(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
struct timer_list *timer = &tp->timer;
if (tp->mac_version <= RTL_GIGA_MAC_VER_01)
return;
del_timer_sync(timer);
}
static inline void rtl8169_request_timer(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
struct timer_list *timer = &tp->timer;
if (tp->mac_version <= RTL_GIGA_MAC_VER_01)
return;
mod_timer(timer, jiffies + RTL8169_PHY_TIMEOUT);
}
#ifdef CONFIG_NET_POLL_CONTROLLER
/*
* Polling 'interrupt' - used by things like netconsole to send skbs
* without having to re-enable interrupts. It's not called while
* the interrupt routine is executing.
*/
static void rtl8169_netpoll(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
struct pci_dev *pdev = tp->pci_dev;
disable_irq(pdev->irq);
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 21:55:46 +08:00
rtl8169_interrupt(pdev->irq, dev);
enable_irq(pdev->irq);
}
#endif
static void rtl8169_release_board(struct pci_dev *pdev, struct net_device *dev,
void __iomem *ioaddr)
{
iounmap(ioaddr);
pci_release_regions(pdev);
pci_disable_device(pdev);
free_netdev(dev);
}
static void rtl8169_phy_reset(struct net_device *dev,
struct rtl8169_private *tp)
{
void __iomem *ioaddr = tp->mmio_addr;
unsigned int i;
tp->phy_reset_enable(ioaddr);
for (i = 0; i < 100; i++) {
if (!tp->phy_reset_pending(ioaddr))
return;
msleep(1);
}
if (netif_msg_link(tp))
printk(KERN_ERR "%s: PHY reset failed.\n", dev->name);
}
static void rtl8169_init_phy(struct net_device *dev, struct rtl8169_private *tp)
{
void __iomem *ioaddr = tp->mmio_addr;
rtl_hw_phy_config(dev);
dprintk("Set MAC Reg C+CR Offset 0x82h = 0x01h\n");
RTL_W8(0x82, 0x01);
pci_write_config_byte(tp->pci_dev, PCI_LATENCY_TIMER, 0x40);
if (tp->mac_version <= RTL_GIGA_MAC_VER_06)
pci_write_config_byte(tp->pci_dev, PCI_CACHE_LINE_SIZE, 0x08);
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
if (tp->mac_version == RTL_GIGA_MAC_VER_02) {
dprintk("Set MAC Reg C+CR Offset 0x82h = 0x01h\n");
RTL_W8(0x82, 0x01);
dprintk("Set PHY Reg 0x0bh = 0x00h\n");
mdio_write(ioaddr, 0x0b, 0x0000); //w 0x0b 15 0 0
}
rtl8169_phy_reset(dev, tp);
/*
* rtl8169_set_speed_xmii takes good care of the Fast Ethernet
* only 8101. Don't panic.
*/
rtl8169_set_speed(dev, AUTONEG_ENABLE, SPEED_1000, DUPLEX_FULL);
if ((RTL_R8(PHYstatus) & TBI_Enable) && netif_msg_link(tp))
printk(KERN_INFO PFX "%s: TBI auto-negotiating\n", dev->name);
}
static void rtl_rar_set(struct rtl8169_private *tp, u8 *addr)
{
void __iomem *ioaddr = tp->mmio_addr;
u32 high;
u32 low;
low = addr[0] | (addr[1] << 8) | (addr[2] << 16) | (addr[3] << 24);
high = addr[4] | (addr[5] << 8);
spin_lock_irq(&tp->lock);
RTL_W8(Cfg9346, Cfg9346_Unlock);
RTL_W32(MAC0, low);
RTL_W32(MAC4, high);
RTL_W8(Cfg9346, Cfg9346_Lock);
spin_unlock_irq(&tp->lock);
}
static int rtl_set_mac_address(struct net_device *dev, void *p)
{
struct rtl8169_private *tp = netdev_priv(dev);
struct sockaddr *addr = p;
if (!is_valid_ether_addr(addr->sa_data))
return -EADDRNOTAVAIL;
memcpy(dev->dev_addr, addr->sa_data, dev->addr_len);
rtl_rar_set(tp, dev->dev_addr);
return 0;
}
static int rtl8169_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
{
struct rtl8169_private *tp = netdev_priv(dev);
struct mii_ioctl_data *data = if_mii(ifr);
if (!netif_running(dev))
return -ENODEV;
switch (cmd) {
case SIOCGMIIPHY:
data->phy_id = 32; /* Internal PHY */
return 0;
case SIOCGMIIREG:
data->val_out = mdio_read(tp->mmio_addr, data->reg_num & 0x1f);
return 0;
case SIOCSMIIREG:
if (!capable(CAP_NET_ADMIN))
return -EPERM;
mdio_write(tp->mmio_addr, data->reg_num & 0x1f, data->val_in);
return 0;
}
return -EOPNOTSUPP;
}
static const struct rtl_cfg_info {
void (*hw_start)(struct net_device *);
unsigned int region;
unsigned int align;
u16 intr_event;
u16 napi_event;
unsigned msi;
} rtl_cfg_infos [] = {
[RTL_CFG_0] = {
.hw_start = rtl_hw_start_8169,
.region = 1,
.align = 0,
.intr_event = SYSErr | LinkChg | RxOverflow |
RxFIFOOver | TxErr | TxOK | RxOK | RxErr,
.napi_event = RxFIFOOver | TxErr | TxOK | RxOK | RxOverflow,
.msi = 0
},
[RTL_CFG_1] = {
.hw_start = rtl_hw_start_8168,
.region = 2,
.align = 8,
.intr_event = SYSErr | LinkChg | RxOverflow |
TxErr | TxOK | RxOK | RxErr,
.napi_event = TxErr | TxOK | RxOK | RxOverflow,
.msi = RTL_FEATURE_MSI
},
[RTL_CFG_2] = {
.hw_start = rtl_hw_start_8101,
.region = 2,
.align = 8,
.intr_event = SYSErr | LinkChg | RxOverflow | PCSTimeout |
RxFIFOOver | TxErr | TxOK | RxOK | RxErr,
.napi_event = RxFIFOOver | TxErr | TxOK | RxOK | RxOverflow,
.msi = RTL_FEATURE_MSI
}
};
/* Cfg9346_Unlock assumed. */
static unsigned rtl_try_msi(struct pci_dev *pdev, void __iomem *ioaddr,
const struct rtl_cfg_info *cfg)
{
unsigned msi = 0;
u8 cfg2;
cfg2 = RTL_R8(Config2) & ~MSIEnable;
if (cfg->msi) {
if (pci_enable_msi(pdev)) {
dev_info(&pdev->dev, "no MSI. Back to INTx.\n");
} else {
cfg2 |= MSIEnable;
msi = RTL_FEATURE_MSI;
}
}
RTL_W8(Config2, cfg2);
return msi;
}
static void rtl_disable_msi(struct pci_dev *pdev, struct rtl8169_private *tp)
{
if (tp->features & RTL_FEATURE_MSI) {
pci_disable_msi(pdev);
tp->features &= ~RTL_FEATURE_MSI;
}
}
static int __devinit
rtl8169_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
{
const struct rtl_cfg_info *cfg = rtl_cfg_infos + ent->driver_data;
const unsigned int region = cfg->region;
struct rtl8169_private *tp;
struct net_device *dev;
void __iomem *ioaddr;
unsigned int i;
int rc;
if (netif_msg_drv(&debug)) {
printk(KERN_INFO "%s Gigabit Ethernet driver %s loaded\n",
MODULENAME, RTL8169_VERSION);
}
dev = alloc_etherdev(sizeof (*tp));
if (!dev) {
if (netif_msg_drv(&debug))
dev_err(&pdev->dev, "unable to alloc new ethernet\n");
rc = -ENOMEM;
goto out;
}
SET_NETDEV_DEV(dev, &pdev->dev);
tp = netdev_priv(dev);
tp->dev = dev;
tp->msg_enable = netif_msg_init(debug.msg_enable, R8169_MSG_DEFAULT);
/* enable device (incl. PCI PM wakeup and hotplug setup) */
rc = pci_enable_device(pdev);
if (rc < 0) {
if (netif_msg_probe(tp))
dev_err(&pdev->dev, "enable failure\n");
goto err_out_free_dev_1;
}
rc = pci_set_mwi(pdev);
if (rc < 0)
goto err_out_disable_2;
/* make sure PCI base addr 1 is MMIO */
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
if (!(pci_resource_flags(pdev, region) & IORESOURCE_MEM)) {
if (netif_msg_probe(tp)) {
dev_err(&pdev->dev,
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
"region #%d not an MMIO resource, aborting\n",
region);
}
rc = -ENODEV;
goto err_out_mwi_3;
}
/* check for weird/broken PCI region reporting */
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
if (pci_resource_len(pdev, region) < R8169_REGS_SIZE) {
if (netif_msg_probe(tp)) {
dev_err(&pdev->dev,
"Invalid PCI region size(s), aborting\n");
}
rc = -ENODEV;
goto err_out_mwi_3;
}
rc = pci_request_regions(pdev, MODULENAME);
if (rc < 0) {
if (netif_msg_probe(tp))
dev_err(&pdev->dev, "could not request regions.\n");
goto err_out_mwi_3;
}
tp->cp_cmd = PCIMulRW | RxChkSum;
if ((sizeof(dma_addr_t) > 4) &&
!pci_set_dma_mask(pdev, DMA_64BIT_MASK) && use_dac) {
tp->cp_cmd |= PCIDAC;
dev->features |= NETIF_F_HIGHDMA;
} else {
rc = pci_set_dma_mask(pdev, DMA_32BIT_MASK);
if (rc < 0) {
if (netif_msg_probe(tp)) {
dev_err(&pdev->dev,
"DMA configuration failed.\n");
}
goto err_out_free_res_4;
}
}
pci_set_master(pdev);
/* ioremap MMIO region */
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
ioaddr = ioremap(pci_resource_start(pdev, region), R8169_REGS_SIZE);
if (!ioaddr) {
if (netif_msg_probe(tp))
dev_err(&pdev->dev, "cannot remap MMIO, aborting\n");
rc = -EIO;
goto err_out_free_res_4;
}
/* Unneeded ? Don't mess with Mrs. Murphy. */
rtl8169_irq_mask_and_ack(ioaddr);
/* Soft reset the chip. */
RTL_W8(ChipCmd, CmdReset);
/* Check that the chip has finished the reset. */
for (i = 0; i < 100; i++) {
if ((RTL_R8(ChipCmd) & CmdReset) == 0)
break;
msleep_interruptible(1);
}
/* Identify chip attached to board */
rtl8169_get_mac_version(tp, ioaddr);
rtl8169_print_mac_version(tp);
for (i = ARRAY_SIZE(rtl_chip_info) - 1; i >= 0; i--) {
if (tp->mac_version == rtl_chip_info[i].mac_version)
break;
}
if (i < 0) {
/* Unknown chip: assume array element #0, original RTL-8169 */
if (netif_msg_probe(tp)) {
dev_printk(KERN_DEBUG, &pdev->dev,
"unknown chip version, assuming %s\n",
rtl_chip_info[0].name);
}
i++;
}
tp->chipset = i;
RTL_W8(Cfg9346, Cfg9346_Unlock);
RTL_W8(Config1, RTL_R8(Config1) | PMEnable);
RTL_W8(Config5, RTL_R8(Config5) & PMEStatus);
tp->features |= rtl_try_msi(pdev, ioaddr, cfg);
RTL_W8(Cfg9346, Cfg9346_Lock);
if ((tp->mac_version <= RTL_GIGA_MAC_VER_06) &&
(RTL_R8(PHYstatus) & TBI_Enable)) {
tp->set_speed = rtl8169_set_speed_tbi;
tp->get_settings = rtl8169_gset_tbi;
tp->phy_reset_enable = rtl8169_tbi_reset_enable;
tp->phy_reset_pending = rtl8169_tbi_reset_pending;
tp->link_ok = rtl8169_tbi_link_ok;
tp->phy_1000_ctrl_reg = ADVERTISE_1000FULL; /* Implied by TBI */
} else {
tp->set_speed = rtl8169_set_speed_xmii;
tp->get_settings = rtl8169_gset_xmii;
tp->phy_reset_enable = rtl8169_xmii_reset_enable;
tp->phy_reset_pending = rtl8169_xmii_reset_pending;
tp->link_ok = rtl8169_xmii_link_ok;
dev->do_ioctl = rtl8169_ioctl;
}
/* Get MAC address. FIXME: read EEPROM */
for (i = 0; i < MAC_ADDR_LEN; i++)
dev->dev_addr[i] = RTL_R8(MAC0 + i);
memcpy(dev->perm_addr, dev->dev_addr, dev->addr_len);
dev->open = rtl8169_open;
dev->hard_start_xmit = rtl8169_start_xmit;
dev->get_stats = rtl8169_get_stats;
SET_ETHTOOL_OPS(dev, &rtl8169_ethtool_ops);
dev->stop = rtl8169_close;
dev->tx_timeout = rtl8169_tx_timeout;
dev->set_multicast_list = rtl_set_rx_mode;
dev->watchdog_timeo = RTL8169_TX_TIMEOUT;
dev->irq = pdev->irq;
dev->base_addr = (unsigned long) ioaddr;
dev->change_mtu = rtl8169_change_mtu;
dev->set_mac_address = rtl_set_mac_address;
#ifdef CONFIG_R8169_NAPI
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
netif_napi_add(dev, &tp->napi, rtl8169_poll, R8169_NAPI_WEIGHT);
#endif
#ifdef CONFIG_R8169_VLAN
dev->features |= NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX;
dev->vlan_rx_register = rtl8169_vlan_rx_register;
#endif
#ifdef CONFIG_NET_POLL_CONTROLLER
dev->poll_controller = rtl8169_netpoll;
#endif
tp->intr_mask = 0xffff;
tp->pci_dev = pdev;
tp->mmio_addr = ioaddr;
tp->align = cfg->align;
tp->hw_start = cfg->hw_start;
tp->intr_event = cfg->intr_event;
tp->napi_event = cfg->napi_event;
init_timer(&tp->timer);
tp->timer.data = (unsigned long) dev;
tp->timer.function = rtl8169_phy_timer;
spin_lock_init(&tp->lock);
rc = register_netdev(dev);
if (rc < 0)
goto err_out_msi_5;
pci_set_drvdata(pdev, dev);
if (netif_msg_probe(tp)) {
u32 xid = RTL_R32(TxConfig) & 0x7cf0f8ff;
printk(KERN_INFO "%s: %s at 0x%lx, "
"%2.2x:%2.2x:%2.2x:%2.2x:%2.2x:%2.2x, "
"XID %08x IRQ %d\n",
dev->name,
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
rtl_chip_info[tp->chipset].name,
dev->base_addr,
dev->dev_addr[0], dev->dev_addr[1],
dev->dev_addr[2], dev->dev_addr[3],
dev->dev_addr[4], dev->dev_addr[5], xid, dev->irq);
}
rtl8169_init_phy(dev, tp);
out:
return rc;
err_out_msi_5:
rtl_disable_msi(pdev, tp);
iounmap(ioaddr);
err_out_free_res_4:
pci_release_regions(pdev);
err_out_mwi_3:
pci_clear_mwi(pdev);
err_out_disable_2:
pci_disable_device(pdev);
err_out_free_dev_1:
free_netdev(dev);
goto out;
}
static void __devexit rtl8169_remove_one(struct pci_dev *pdev)
{
struct net_device *dev = pci_get_drvdata(pdev);
struct rtl8169_private *tp = netdev_priv(dev);
flush_scheduled_work();
unregister_netdev(dev);
rtl_disable_msi(pdev, tp);
rtl8169_release_board(pdev, dev, tp->mmio_addr);
pci_set_drvdata(pdev, NULL);
}
static void rtl8169_set_rxbufsize(struct rtl8169_private *tp,
struct net_device *dev)
{
unsigned int mtu = dev->mtu;
tp->rx_buf_sz = (mtu > RX_BUF_SIZE) ? mtu + ETH_HLEN + 8 : RX_BUF_SIZE;
}
static int rtl8169_open(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
struct pci_dev *pdev = tp->pci_dev;
int retval = -ENOMEM;
rtl8169_set_rxbufsize(tp, dev);
/*
* Rx and Tx desscriptors needs 256 bytes alignment.
* pci_alloc_consistent provides more.
*/
tp->TxDescArray = pci_alloc_consistent(pdev, R8169_TX_RING_BYTES,
&tp->TxPhyAddr);
if (!tp->TxDescArray)
goto out;
tp->RxDescArray = pci_alloc_consistent(pdev, R8169_RX_RING_BYTES,
&tp->RxPhyAddr);
if (!tp->RxDescArray)
goto err_free_tx_0;
retval = rtl8169_init_ring(dev);
if (retval < 0)
goto err_free_rx_1;
INIT_DELAYED_WORK(&tp->task, NULL);
smp_mb();
retval = request_irq(dev->irq, rtl8169_interrupt,
(tp->features & RTL_FEATURE_MSI) ? 0 : IRQF_SHARED,
dev->name, dev);
if (retval < 0)
goto err_release_ring_2;
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
#ifdef CONFIG_R8169_NAPI
napi_enable(&tp->napi);
#endif
rtl_hw_start(dev);
rtl8169_request_timer(dev);
rtl8169_check_link_status(dev, tp, tp->mmio_addr);
out:
return retval;
err_release_ring_2:
rtl8169_rx_clear(tp);
err_free_rx_1:
pci_free_consistent(pdev, R8169_RX_RING_BYTES, tp->RxDescArray,
tp->RxPhyAddr);
err_free_tx_0:
pci_free_consistent(pdev, R8169_TX_RING_BYTES, tp->TxDescArray,
tp->TxPhyAddr);
goto out;
}
static void rtl8169_hw_reset(void __iomem *ioaddr)
{
/* Disable interrupts */
rtl8169_irq_mask_and_ack(ioaddr);
/* Reset the chipset */
RTL_W8(ChipCmd, CmdReset);
/* PCI commit */
RTL_R8(ChipCmd);
}
static void rtl_set_rx_tx_config_registers(struct rtl8169_private *tp)
{
void __iomem *ioaddr = tp->mmio_addr;
u32 cfg = rtl8169_rx_config;
cfg |= (RTL_R32(RxConfig) & rtl_chip_info[tp->chipset].RxConfigMask);
RTL_W32(RxConfig, cfg);
/* Set DMA burst size and Interframe Gap Time */
RTL_W32(TxConfig, (TX_DMA_BURST << TxDMAShift) |
(InterFrameGap << TxInterFrameGapShift));
}
static void rtl_hw_start(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
unsigned int i;
/* Soft reset the chip. */
RTL_W8(ChipCmd, CmdReset);
/* Check that the chip has finished the reset. */
for (i = 0; i < 100; i++) {
if ((RTL_R8(ChipCmd) & CmdReset) == 0)
break;
msleep_interruptible(1);
}
tp->hw_start(dev);
netif_start_queue(dev);
}
static void rtl_set_rx_tx_desc_registers(struct rtl8169_private *tp,
void __iomem *ioaddr)
{
/*
* Magic spell: some iop3xx ARM board needs the TxDescAddrHigh
* register to be written before TxDescAddrLow to work.
* Switching from MMIO to I/O access fixes the issue as well.
*/
RTL_W32(TxDescStartAddrHigh, ((u64) tp->TxPhyAddr) >> 32);
RTL_W32(TxDescStartAddrLow, ((u64) tp->TxPhyAddr) & DMA_32BIT_MASK);
RTL_W32(RxDescAddrHigh, ((u64) tp->RxPhyAddr) >> 32);
RTL_W32(RxDescAddrLow, ((u64) tp->RxPhyAddr) & DMA_32BIT_MASK);
}
static u16 rtl_rw_cpluscmd(void __iomem *ioaddr)
{
u16 cmd;
cmd = RTL_R16(CPlusCmd);
RTL_W16(CPlusCmd, cmd);
return cmd;
}
static void rtl_set_rx_max_size(void __iomem *ioaddr)
{
/* Low hurts. Let's disable the filtering. */
RTL_W16(RxMaxSize, 16383);
}
static void rtl8169_set_magic_reg(void __iomem *ioaddr, unsigned mac_version)
{
struct {
u32 mac_version;
u32 clk;
u32 val;
} cfg2_info [] = {
{ RTL_GIGA_MAC_VER_05, PCI_Clock_33MHz, 0x000fff00 }, // 8110SCd
{ RTL_GIGA_MAC_VER_05, PCI_Clock_66MHz, 0x000fffff },
{ RTL_GIGA_MAC_VER_06, PCI_Clock_33MHz, 0x00ffff00 }, // 8110SCe
{ RTL_GIGA_MAC_VER_06, PCI_Clock_66MHz, 0x00ffffff }
}, *p = cfg2_info;
unsigned int i;
u32 clk;
clk = RTL_R8(Config2) & PCI_Clock_66MHz;
for (i = 0; i < ARRAY_SIZE(cfg2_info); i++, p++) {
if ((p->mac_version == mac_version) && (p->clk == clk)) {
RTL_W32(0x7c, p->val);
break;
}
}
}
static void rtl_hw_start_8169(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
struct pci_dev *pdev = tp->pci_dev;
if (tp->mac_version == RTL_GIGA_MAC_VER_05) {
RTL_W16(CPlusCmd, RTL_R16(CPlusCmd) | PCIMulRW);
pci_write_config_byte(pdev, PCI_CACHE_LINE_SIZE, 0x08);
}
RTL_W8(Cfg9346, Cfg9346_Unlock);
if ((tp->mac_version == RTL_GIGA_MAC_VER_01) ||
(tp->mac_version == RTL_GIGA_MAC_VER_02) ||
(tp->mac_version == RTL_GIGA_MAC_VER_03) ||
(tp->mac_version == RTL_GIGA_MAC_VER_04))
RTL_W8(ChipCmd, CmdTxEnb | CmdRxEnb);
RTL_W8(EarlyTxThres, EarlyTxThld);
rtl_set_rx_max_size(ioaddr);
if ((tp->mac_version == RTL_GIGA_MAC_VER_01) ||
(tp->mac_version == RTL_GIGA_MAC_VER_02) ||
(tp->mac_version == RTL_GIGA_MAC_VER_03) ||
(tp->mac_version == RTL_GIGA_MAC_VER_04))
rtl_set_rx_tx_config_registers(tp);
tp->cp_cmd |= rtl_rw_cpluscmd(ioaddr) | PCIMulRW;
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
if ((tp->mac_version == RTL_GIGA_MAC_VER_02) ||
(tp->mac_version == RTL_GIGA_MAC_VER_03)) {
dprintk("Set MAC Reg C+CR Offset 0xE0. "
"Bit-3 and bit-14 MUST be 1\n");
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
tp->cp_cmd |= (1 << 14);
}
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
RTL_W16(CPlusCmd, tp->cp_cmd);
rtl8169_set_magic_reg(ioaddr, tp->mac_version);
/*
* Undocumented corner. Supposedly:
* (TxTimer << 12) | (TxPackets << 8) | (RxTimer << 4) | RxPackets
*/
RTL_W16(IntrMitigate, 0x0000);
rtl_set_rx_tx_desc_registers(tp, ioaddr);
if ((tp->mac_version != RTL_GIGA_MAC_VER_01) &&
(tp->mac_version != RTL_GIGA_MAC_VER_02) &&
(tp->mac_version != RTL_GIGA_MAC_VER_03) &&
(tp->mac_version != RTL_GIGA_MAC_VER_04)) {
RTL_W8(ChipCmd, CmdTxEnb | CmdRxEnb);
rtl_set_rx_tx_config_registers(tp);
}
RTL_W8(Cfg9346, Cfg9346_Lock);
/* Initially a 10 us delay. Turned it into a PCI commit. - FR */
RTL_R8(IntrMask);
RTL_W32(RxMissed, 0);
rtl_set_rx_mode(dev);
/* no early-rx interrupts */
RTL_W16(MultiIntr, RTL_R16(MultiIntr) & 0xF000);
/* Enable all known interrupts by setting the interrupt mask. */
RTL_W16(IntrMask, tp->intr_event);
}
static void rtl_hw_start_8168(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
struct pci_dev *pdev = tp->pci_dev;
u8 ctl;
RTL_W8(Cfg9346, Cfg9346_Unlock);
RTL_W8(EarlyTxThres, EarlyTxThld);
rtl_set_rx_max_size(ioaddr);
rtl_set_rx_tx_config_registers(tp);
tp->cp_cmd |= RTL_R16(CPlusCmd) | PktCntrDisable | INTT_1;
RTL_W16(CPlusCmd, tp->cp_cmd);
/* Tx performance tweak. */
pci_read_config_byte(pdev, 0x69, &ctl);
ctl = (ctl & ~0x70) | 0x50;
pci_write_config_byte(pdev, 0x69, ctl);
RTL_W16(IntrMitigate, 0x5151);
/* Work around for RxFIFO overflow. */
if (tp->mac_version == RTL_GIGA_MAC_VER_11) {
tp->intr_event |= RxFIFOOver | PCSTimeout;
tp->intr_event &= ~RxOverflow;
}
rtl_set_rx_tx_desc_registers(tp, ioaddr);
RTL_W8(Cfg9346, Cfg9346_Lock);
RTL_R8(IntrMask);
RTL_W32(RxMissed, 0);
rtl_set_rx_mode(dev);
RTL_W8(ChipCmd, CmdTxEnb | CmdRxEnb);
RTL_W16(MultiIntr, RTL_R16(MultiIntr) & 0xF000);
RTL_W16(IntrMask, tp->intr_event);
}
static void rtl_hw_start_8101(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
struct pci_dev *pdev = tp->pci_dev;
if ((tp->mac_version == RTL_GIGA_MAC_VER_13) ||
(tp->mac_version == RTL_GIGA_MAC_VER_16)) {
pci_write_config_word(pdev, 0x68, 0x00);
pci_write_config_word(pdev, 0x69, 0x08);
}
RTL_W8(Cfg9346, Cfg9346_Unlock);
RTL_W8(EarlyTxThres, EarlyTxThld);
rtl_set_rx_max_size(ioaddr);
tp->cp_cmd |= rtl_rw_cpluscmd(ioaddr) | PCIMulRW;
RTL_W16(CPlusCmd, tp->cp_cmd);
RTL_W16(IntrMitigate, 0x0000);
rtl_set_rx_tx_desc_registers(tp, ioaddr);
RTL_W8(ChipCmd, CmdTxEnb | CmdRxEnb);
rtl_set_rx_tx_config_registers(tp);
RTL_W8(Cfg9346, Cfg9346_Lock);
RTL_R8(IntrMask);
RTL_W32(RxMissed, 0);
rtl_set_rx_mode(dev);
RTL_W8(ChipCmd, CmdTxEnb | CmdRxEnb);
RTL_W16(MultiIntr, RTL_R16(MultiIntr) & 0xf000);
RTL_W16(IntrMask, tp->intr_event);
}
static int rtl8169_change_mtu(struct net_device *dev, int new_mtu)
{
struct rtl8169_private *tp = netdev_priv(dev);
int ret = 0;
if (new_mtu < ETH_ZLEN || new_mtu > SafeMtu)
return -EINVAL;
dev->mtu = new_mtu;
if (!netif_running(dev))
goto out;
rtl8169_down(dev);
rtl8169_set_rxbufsize(tp, dev);
ret = rtl8169_init_ring(dev);
if (ret < 0)
goto out;
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
#ifdef CONFIG_R8169_NAPI
napi_enable(&tp->napi);
#endif
rtl_hw_start(dev);
rtl8169_request_timer(dev);
out:
return ret;
}
static inline void rtl8169_make_unusable_by_asic(struct RxDesc *desc)
{
desc->addr = cpu_to_le64(0x0badbadbadbadbadull);
desc->opts1 &= ~cpu_to_le32(DescOwn | RsvdMask);
}
static void rtl8169_free_rx_skb(struct rtl8169_private *tp,
struct sk_buff **sk_buff, struct RxDesc *desc)
{
struct pci_dev *pdev = tp->pci_dev;
pci_unmap_single(pdev, le64_to_cpu(desc->addr), tp->rx_buf_sz,
PCI_DMA_FROMDEVICE);
dev_kfree_skb(*sk_buff);
*sk_buff = NULL;
rtl8169_make_unusable_by_asic(desc);
}
static inline void rtl8169_mark_to_asic(struct RxDesc *desc, u32 rx_buf_sz)
{
u32 eor = le32_to_cpu(desc->opts1) & RingEnd;
desc->opts1 = cpu_to_le32(DescOwn | eor | rx_buf_sz);
}
static inline void rtl8169_map_to_asic(struct RxDesc *desc, dma_addr_t mapping,
u32 rx_buf_sz)
{
desc->addr = cpu_to_le64(mapping);
wmb();
rtl8169_mark_to_asic(desc, rx_buf_sz);
}
static struct sk_buff *rtl8169_alloc_rx_skb(struct pci_dev *pdev,
struct net_device *dev,
struct RxDesc *desc, int rx_buf_sz,
unsigned int align)
{
struct sk_buff *skb;
dma_addr_t mapping;
unsigned int pad;
pad = align ? align : NET_IP_ALIGN;
skb = netdev_alloc_skb(dev, rx_buf_sz + pad);
if (!skb)
goto err_out;
skb_reserve(skb, align ? ((pad - 1) & (unsigned long)skb->data) : pad);
mapping = pci_map_single(pdev, skb->data, rx_buf_sz,
PCI_DMA_FROMDEVICE);
rtl8169_map_to_asic(desc, mapping, rx_buf_sz);
out:
return skb;
err_out:
rtl8169_make_unusable_by_asic(desc);
goto out;
}
static void rtl8169_rx_clear(struct rtl8169_private *tp)
{
unsigned int i;
for (i = 0; i < NUM_RX_DESC; i++) {
if (tp->Rx_skbuff[i]) {
rtl8169_free_rx_skb(tp, tp->Rx_skbuff + i,
tp->RxDescArray + i);
}
}
}
static u32 rtl8169_rx_fill(struct rtl8169_private *tp, struct net_device *dev,
u32 start, u32 end)
{
u32 cur;
for (cur = start; end - cur != 0; cur++) {
struct sk_buff *skb;
unsigned int i = cur % NUM_RX_DESC;
WARN_ON((s32)(end - cur) < 0);
if (tp->Rx_skbuff[i])
continue;
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
skb = rtl8169_alloc_rx_skb(tp->pci_dev, dev,
tp->RxDescArray + i,
tp->rx_buf_sz, tp->align);
if (!skb)
break;
tp->Rx_skbuff[i] = skb;
}
return cur - start;
}
static inline void rtl8169_mark_as_last_descriptor(struct RxDesc *desc)
{
desc->opts1 |= cpu_to_le32(RingEnd);
}
static void rtl8169_init_ring_indexes(struct rtl8169_private *tp)
{
tp->dirty_tx = tp->dirty_rx = tp->cur_tx = tp->cur_rx = 0;
}
static int rtl8169_init_ring(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
rtl8169_init_ring_indexes(tp);
memset(tp->tx_skb, 0x0, NUM_TX_DESC * sizeof(struct ring_info));
memset(tp->Rx_skbuff, 0x0, NUM_RX_DESC * sizeof(struct sk_buff *));
if (rtl8169_rx_fill(tp, dev, 0, NUM_RX_DESC) != NUM_RX_DESC)
goto err_out;
rtl8169_mark_as_last_descriptor(tp->RxDescArray + NUM_RX_DESC - 1);
return 0;
err_out:
rtl8169_rx_clear(tp);
return -ENOMEM;
}
static void rtl8169_unmap_tx_skb(struct pci_dev *pdev, struct ring_info *tx_skb,
struct TxDesc *desc)
{
unsigned int len = tx_skb->len;
pci_unmap_single(pdev, le64_to_cpu(desc->addr), len, PCI_DMA_TODEVICE);
desc->opts1 = 0x00;
desc->opts2 = 0x00;
desc->addr = 0x00;
tx_skb->len = 0;
}
static void rtl8169_tx_clear(struct rtl8169_private *tp)
{
unsigned int i;
for (i = tp->dirty_tx; i < tp->dirty_tx + NUM_TX_DESC; i++) {
unsigned int entry = i % NUM_TX_DESC;
struct ring_info *tx_skb = tp->tx_skb + entry;
unsigned int len = tx_skb->len;
if (len) {
struct sk_buff *skb = tx_skb->skb;
rtl8169_unmap_tx_skb(tp->pci_dev, tx_skb,
tp->TxDescArray + entry);
if (skb) {
dev_kfree_skb(skb);
tx_skb->skb = NULL;
}
tp->dev->stats.tx_dropped++;
}
}
tp->cur_tx = tp->dirty_tx = 0;
}
static void rtl8169_schedule_work(struct net_device *dev, work_func_t task)
{
struct rtl8169_private *tp = netdev_priv(dev);
PREPARE_DELAYED_WORK(&tp->task, task);
schedule_delayed_work(&tp->task, 4);
}
static void rtl8169_wait_for_quiescence(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
synchronize_irq(dev->irq);
/* Wait for any pending NAPI task to complete */
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
#ifdef CONFIG_R8169_NAPI
napi_disable(&tp->napi);
#endif
rtl8169_irq_mask_and_ack(ioaddr);
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
#ifdef CONFIG_R8169_NAPI
tp->intr_mask = 0xffff;
RTL_W16(IntrMask, tp->intr_event);
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
napi_enable(&tp->napi);
#endif
}
static void rtl8169_reinit_task(struct work_struct *work)
{
struct rtl8169_private *tp =
container_of(work, struct rtl8169_private, task.work);
struct net_device *dev = tp->dev;
int ret;
rtnl_lock();
if (!netif_running(dev))
goto out_unlock;
rtl8169_wait_for_quiescence(dev);
rtl8169_close(dev);
ret = rtl8169_open(dev);
if (unlikely(ret < 0)) {
if (net_ratelimit() && netif_msg_drv(tp)) {
printk(KERN_ERR PFX "%s: reinit failure (status = %d)."
" Rescheduling.\n", dev->name, ret);
}
rtl8169_schedule_work(dev, rtl8169_reinit_task);
}
out_unlock:
rtnl_unlock();
}
static void rtl8169_reset_task(struct work_struct *work)
{
struct rtl8169_private *tp =
container_of(work, struct rtl8169_private, task.work);
struct net_device *dev = tp->dev;
rtnl_lock();
if (!netif_running(dev))
goto out_unlock;
rtl8169_wait_for_quiescence(dev);
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
rtl8169_rx_interrupt(dev, tp, tp->mmio_addr, ~(u32)0);
rtl8169_tx_clear(tp);
if (tp->dirty_rx == tp->cur_rx) {
rtl8169_init_ring_indexes(tp);
rtl_hw_start(dev);
netif_wake_queue(dev);
rtl8169_check_link_status(dev, tp, tp->mmio_addr);
} else {
if (net_ratelimit() && netif_msg_intr(tp)) {
printk(KERN_EMERG PFX "%s: Rx buffers shortage\n",
dev->name);
}
rtl8169_schedule_work(dev, rtl8169_reset_task);
}
out_unlock:
rtnl_unlock();
}
static void rtl8169_tx_timeout(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
rtl8169_hw_reset(tp->mmio_addr);
/* Let's wait a bit while any (async) irq lands on */
rtl8169_schedule_work(dev, rtl8169_reset_task);
}
static int rtl8169_xmit_frags(struct rtl8169_private *tp, struct sk_buff *skb,
u32 opts1)
{
struct skb_shared_info *info = skb_shinfo(skb);
unsigned int cur_frag, entry;
struct TxDesc * uninitialized_var(txd);
entry = tp->cur_tx;
for (cur_frag = 0; cur_frag < info->nr_frags; cur_frag++) {
skb_frag_t *frag = info->frags + cur_frag;
dma_addr_t mapping;
u32 status, len;
void *addr;
entry = (entry + 1) % NUM_TX_DESC;
txd = tp->TxDescArray + entry;
len = frag->size;
addr = ((void *) page_address(frag->page)) + frag->page_offset;
mapping = pci_map_single(tp->pci_dev, addr, len, PCI_DMA_TODEVICE);
/* anti gcc 2.95.3 bugware (sic) */
status = opts1 | len | (RingEnd * !((entry + 1) % NUM_TX_DESC));
txd->opts1 = cpu_to_le32(status);
txd->addr = cpu_to_le64(mapping);
tp->tx_skb[entry].len = len;
}
if (cur_frag) {
tp->tx_skb[entry].skb = skb;
txd->opts1 |= cpu_to_le32(LastFrag);
}
return cur_frag;
}
static inline u32 rtl8169_tso_csum(struct sk_buff *skb, struct net_device *dev)
{
if (dev->features & NETIF_F_TSO) {
u32 mss = skb_shinfo(skb)->gso_size;
if (mss)
return LargeSend | ((mss & MSSMask) << MSSShift);
}
if (skb->ip_summed == CHECKSUM_PARTIAL) {
const struct iphdr *ip = ip_hdr(skb);
if (ip->protocol == IPPROTO_TCP)
return IPCS | TCPCS;
else if (ip->protocol == IPPROTO_UDP)
return IPCS | UDPCS;
WARN_ON(1); /* we need a WARN() */
}
return 0;
}
static int rtl8169_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
unsigned int frags, entry = tp->cur_tx % NUM_TX_DESC;
struct TxDesc *txd = tp->TxDescArray + entry;
void __iomem *ioaddr = tp->mmio_addr;
dma_addr_t mapping;
u32 status, len;
u32 opts1;
int ret = NETDEV_TX_OK;
if (unlikely(TX_BUFFS_AVAIL(tp) < skb_shinfo(skb)->nr_frags)) {
if (netif_msg_drv(tp)) {
printk(KERN_ERR
"%s: BUG! Tx Ring full when queue awake!\n",
dev->name);
}
goto err_stop;
}
if (unlikely(le32_to_cpu(txd->opts1) & DescOwn))
goto err_stop;
opts1 = DescOwn | rtl8169_tso_csum(skb, dev);
frags = rtl8169_xmit_frags(tp, skb, opts1);
if (frags) {
len = skb_headlen(skb);
opts1 |= FirstFrag;
} else {
len = skb->len;
if (unlikely(len < ETH_ZLEN)) {
if (skb_padto(skb, ETH_ZLEN))
goto err_update_stats;
len = ETH_ZLEN;
}
opts1 |= FirstFrag | LastFrag;
tp->tx_skb[entry].skb = skb;
}
mapping = pci_map_single(tp->pci_dev, skb->data, len, PCI_DMA_TODEVICE);
tp->tx_skb[entry].len = len;
txd->addr = cpu_to_le64(mapping);
txd->opts2 = cpu_to_le32(rtl8169_tx_vlan_tag(tp, skb));
wmb();
/* anti gcc 2.95.3 bugware (sic) */
status = opts1 | len | (RingEnd * !((entry + 1) % NUM_TX_DESC));
txd->opts1 = cpu_to_le32(status);
dev->trans_start = jiffies;
tp->cur_tx += frags + 1;
smp_wmb();
RTL_W8(TxPoll, NPQ); /* set polling bit */
if (TX_BUFFS_AVAIL(tp) < MAX_SKB_FRAGS) {
netif_stop_queue(dev);
smp_rmb();
if (TX_BUFFS_AVAIL(tp) >= MAX_SKB_FRAGS)
netif_wake_queue(dev);
}
out:
return ret;
err_stop:
netif_stop_queue(dev);
ret = NETDEV_TX_BUSY;
err_update_stats:
dev->stats.tx_dropped++;
goto out;
}
static void rtl8169_pcierr_interrupt(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
struct pci_dev *pdev = tp->pci_dev;
void __iomem *ioaddr = tp->mmio_addr;
u16 pci_status, pci_cmd;
pci_read_config_word(pdev, PCI_COMMAND, &pci_cmd);
pci_read_config_word(pdev, PCI_STATUS, &pci_status);
if (netif_msg_intr(tp)) {
printk(KERN_ERR
"%s: PCI error (cmd = 0x%04x, status = 0x%04x).\n",
dev->name, pci_cmd, pci_status);
}
/*
* The recovery sequence below admits a very elaborated explanation:
* - it seems to work;
* - I did not see what else could be done;
* - it makes iop3xx happy.
*
* Feel free to adjust to your needs.
*/
if (pdev->broken_parity_status)
pci_cmd &= ~PCI_COMMAND_PARITY;
else
pci_cmd |= PCI_COMMAND_SERR | PCI_COMMAND_PARITY;
pci_write_config_word(pdev, PCI_COMMAND, pci_cmd);
pci_write_config_word(pdev, PCI_STATUS,
pci_status & (PCI_STATUS_DETECTED_PARITY |
PCI_STATUS_SIG_SYSTEM_ERROR | PCI_STATUS_REC_MASTER_ABORT |
PCI_STATUS_REC_TARGET_ABORT | PCI_STATUS_SIG_TARGET_ABORT));
/* The infamous DAC f*ckup only happens at boot time */
if ((tp->cp_cmd & PCIDAC) && !tp->dirty_rx && !tp->cur_rx) {
if (netif_msg_intr(tp))
printk(KERN_INFO "%s: disabling PCI DAC.\n", dev->name);
tp->cp_cmd &= ~PCIDAC;
RTL_W16(CPlusCmd, tp->cp_cmd);
dev->features &= ~NETIF_F_HIGHDMA;
}
rtl8169_hw_reset(ioaddr);
rtl8169_schedule_work(dev, rtl8169_reinit_task);
}
static void rtl8169_tx_interrupt(struct net_device *dev,
struct rtl8169_private *tp,
void __iomem *ioaddr)
{
unsigned int dirty_tx, tx_left;
dirty_tx = tp->dirty_tx;
smp_rmb();
tx_left = tp->cur_tx - dirty_tx;
while (tx_left > 0) {
unsigned int entry = dirty_tx % NUM_TX_DESC;
struct ring_info *tx_skb = tp->tx_skb + entry;
u32 len = tx_skb->len;
u32 status;
rmb();
status = le32_to_cpu(tp->TxDescArray[entry].opts1);
if (status & DescOwn)
break;
dev->stats.tx_bytes += len;
dev->stats.tx_packets++;
rtl8169_unmap_tx_skb(tp->pci_dev, tx_skb, tp->TxDescArray + entry);
if (status & LastFrag) {
dev_kfree_skb_irq(tx_skb->skb);
tx_skb->skb = NULL;
}
dirty_tx++;
tx_left--;
}
if (tp->dirty_tx != dirty_tx) {
tp->dirty_tx = dirty_tx;
smp_wmb();
if (netif_queue_stopped(dev) &&
(TX_BUFFS_AVAIL(tp) >= MAX_SKB_FRAGS)) {
netif_wake_queue(dev);
}
/*
* 8168 hack: TxPoll requests are lost when the Tx packets are
* too close. Let's kick an extra TxPoll request when a burst
* of start_xmit activity is detected (if it is not detected,
* it is slow enough). -- FR
*/
smp_rmb();
if (tp->cur_tx != dirty_tx)
RTL_W8(TxPoll, NPQ);
}
}
static inline int rtl8169_fragmented_frame(u32 status)
{
return (status & (FirstFrag | LastFrag)) != (FirstFrag | LastFrag);
}
static inline void rtl8169_rx_csum(struct sk_buff *skb, struct RxDesc *desc)
{
u32 opts1 = le32_to_cpu(desc->opts1);
u32 status = opts1 & RxProtoMask;
if (((status == RxProtoTCP) && !(opts1 & TCPFail)) ||
((status == RxProtoUDP) && !(opts1 & UDPFail)) ||
((status == RxProtoIP) && !(opts1 & IPFail)))
skb->ip_summed = CHECKSUM_UNNECESSARY;
else
skb->ip_summed = CHECKSUM_NONE;
}
static inline bool rtl8169_try_rx_copy(struct sk_buff **sk_buff,
struct rtl8169_private *tp, int pkt_size,
dma_addr_t addr)
{
struct sk_buff *skb;
bool done = false;
if (pkt_size >= rx_copybreak)
goto out;
skb = netdev_alloc_skb(tp->dev, pkt_size + NET_IP_ALIGN);
if (!skb)
goto out;
pci_dma_sync_single_for_cpu(tp->pci_dev, addr, pkt_size,
PCI_DMA_FROMDEVICE);
skb_reserve(skb, NET_IP_ALIGN);
skb_copy_from_linear_data(*sk_buff, skb->data, pkt_size);
*sk_buff = skb;
done = true;
out:
return done;
}
static int rtl8169_rx_interrupt(struct net_device *dev,
struct rtl8169_private *tp,
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
void __iomem *ioaddr, u32 budget)
{
unsigned int cur_rx, rx_left;
unsigned int delta, count;
cur_rx = tp->cur_rx;
rx_left = NUM_RX_DESC + tp->dirty_rx - cur_rx;
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
rx_left = rtl8169_rx_quota(rx_left, budget);
for (; rx_left > 0; rx_left--, cur_rx++) {
unsigned int entry = cur_rx % NUM_RX_DESC;
struct RxDesc *desc = tp->RxDescArray + entry;
u32 status;
rmb();
status = le32_to_cpu(desc->opts1);
if (status & DescOwn)
break;
if (unlikely(status & RxRES)) {
if (netif_msg_rx_err(tp)) {
printk(KERN_INFO
"%s: Rx ERROR. status = %08x\n",
dev->name, status);
}
dev->stats.rx_errors++;
if (status & (RxRWT | RxRUNT))
dev->stats.rx_length_errors++;
if (status & RxCRC)
dev->stats.rx_crc_errors++;
if (status & RxFOVF) {
rtl8169_schedule_work(dev, rtl8169_reset_task);
dev->stats.rx_fifo_errors++;
}
rtl8169_mark_to_asic(desc, tp->rx_buf_sz);
} else {
struct sk_buff *skb = tp->Rx_skbuff[entry];
dma_addr_t addr = le64_to_cpu(desc->addr);
int pkt_size = (status & 0x00001FFF) - 4;
struct pci_dev *pdev = tp->pci_dev;
/*
* The driver does not support incoming fragmented
* frames. They are seen as a symptom of over-mtu
* sized frames.
*/
if (unlikely(rtl8169_fragmented_frame(status))) {
dev->stats.rx_dropped++;
dev->stats.rx_length_errors++;
rtl8169_mark_to_asic(desc, tp->rx_buf_sz);
continue;
}
rtl8169_rx_csum(skb, desc);
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
if (rtl8169_try_rx_copy(&skb, tp, pkt_size, addr)) {
pci_dma_sync_single_for_device(pdev, addr,
pkt_size, PCI_DMA_FROMDEVICE);
rtl8169_mark_to_asic(desc, tp->rx_buf_sz);
} else {
pci_unmap_single(pdev, addr, pkt_size,
PCI_DMA_FROMDEVICE);
tp->Rx_skbuff[entry] = NULL;
}
skb_put(skb, pkt_size);
skb->protocol = eth_type_trans(skb, dev);
if (rtl8169_rx_vlan_skb(tp, desc, skb) < 0)
rtl8169_rx_skb(skb);
dev->last_rx = jiffies;
dev->stats.rx_bytes += pkt_size;
dev->stats.rx_packets++;
}
/* Work around for AMD plateform. */
if ((desc->opts2 & cpu_to_le32(0xfffe000)) &&
(tp->mac_version == RTL_GIGA_MAC_VER_05)) {
desc->opts2 = 0;
cur_rx++;
}
}
count = cur_rx - tp->cur_rx;
tp->cur_rx = cur_rx;
delta = rtl8169_rx_fill(tp, dev, tp->dirty_rx, tp->cur_rx);
if (!delta && count && netif_msg_intr(tp))
printk(KERN_INFO "%s: no Rx buffer allocated\n", dev->name);
tp->dirty_rx += delta;
/*
* FIXME: until there is periodic timer to try and refill the ring,
* a temporary shortage may definitely kill the Rx process.
* - disable the asic to try and avoid an overflow and kick it again
* after refill ?
* - how do others driver handle this condition (Uh oh...).
*/
if ((tp->dirty_rx + NUM_RX_DESC == tp->cur_rx) && netif_msg_intr(tp))
printk(KERN_EMERG "%s: Rx buffers exhausted\n", dev->name);
return count;
}
static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance)
{
struct net_device *dev = dev_instance;
struct rtl8169_private *tp = netdev_priv(dev);
int boguscnt = max_interrupt_work;
void __iomem *ioaddr = tp->mmio_addr;
int status;
int handled = 0;
do {
status = RTL_R16(IntrStatus);
/* hotplug/major error/no more work/shared irq */
if ((status == 0xFFFF) || !status)
break;
handled = 1;
if (unlikely(!netif_running(dev))) {
rtl8169_asic_down(ioaddr);
goto out;
}
status &= tp->intr_mask;
RTL_W16(IntrStatus,
(status & RxFIFOOver) ? (status | RxOverflow) : status);
if (!(status & tp->intr_event))
break;
/* Work around for rx fifo overflow */
if (unlikely(status & RxFIFOOver) &&
(tp->mac_version == RTL_GIGA_MAC_VER_11)) {
netif_stop_queue(dev);
rtl8169_tx_timeout(dev);
break;
}
if (unlikely(status & SYSErr)) {
rtl8169_pcierr_interrupt(dev);
break;
}
if (status & LinkChg)
rtl8169_check_link_status(dev, tp, ioaddr);
#ifdef CONFIG_R8169_NAPI
if (status & tp->napi_event) {
RTL_W16(IntrMask, tp->intr_event & ~tp->napi_event);
tp->intr_mask = ~tp->napi_event;
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
if (likely(netif_rx_schedule_prep(dev, &tp->napi)))
__netif_rx_schedule(dev, &tp->napi);
else if (netif_msg_intr(tp)) {
printk(KERN_INFO "%s: interrupt %04x in poll\n",
dev->name, status);
}
}
break;
#else
/* Rx interrupt */
if (status & (RxOK | RxOverflow | RxFIFOOver))
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
rtl8169_rx_interrupt(dev, tp, ioaddr, ~(u32)0);
/* Tx interrupt */
if (status & (TxOK | TxErr))
rtl8169_tx_interrupt(dev, tp, ioaddr);
#endif
boguscnt--;
} while (boguscnt > 0);
if (boguscnt <= 0) {
if (netif_msg_intr(tp) && net_ratelimit() ) {
printk(KERN_WARNING
"%s: Too much work at interrupt!\n", dev->name);
}
/* Clear all interrupt sources. */
RTL_W16(IntrStatus, 0xffff);
}
out:
return IRQ_RETVAL(handled);
}
#ifdef CONFIG_R8169_NAPI
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
static int rtl8169_poll(struct napi_struct *napi, int budget)
{
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
struct rtl8169_private *tp = container_of(napi, struct rtl8169_private, napi);
struct net_device *dev = tp->dev;
void __iomem *ioaddr = tp->mmio_addr;
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
int work_done;
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
work_done = rtl8169_rx_interrupt(dev, tp, ioaddr, (u32) budget);
rtl8169_tx_interrupt(dev, tp, ioaddr);
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
if (work_done < budget) {
netif_rx_complete(dev, napi);
tp->intr_mask = 0xffff;
/*
* 20040426: the barrier is not strictly required but the
* behavior of the irq handler could be less predictable
* without it. Btw, the lack of flush for the posted pci
* write is safe - FR
*/
smp_wmb();
RTL_W16(IntrMask, tp->intr_event);
}
[NET]: Make NAPI polling independent of struct net_device objects. Several devices have multiple independant RX queues per net device, and some have a single interrupt doorbell for several queues. In either case, it's easier to support layouts like that if the structure representing the poll is independant from the net device itself. The signature of the ->poll() call back goes from: int foo_poll(struct net_device *dev, int *budget) to int foo_poll(struct napi_struct *napi, int budget) The caller is returned the number of RX packets processed (or the number of "NAPI credits" consumed if you want to get abstract). The callee no longer messes around bumping dev->quota, *budget, etc. because that is all handled in the caller upon return. The napi_struct is to be embedded in the device driver private data structures. Furthermore, it is the driver's responsibility to disable all NAPI instances in it's ->stop() device close handler. Since the napi_struct is privatized into the driver's private data structures, only the driver knows how to get at all of the napi_struct instances it may have per-device. With lots of help and suggestions from Rusty Russell, Roland Dreier, Michael Chan, Jeff Garzik, and Jamal Hadi Salim. Bug fixes from Thomas Graf, Roland Dreier, Peter Zijlstra, Joseph Fannin, Scott Wood, Hans J. Koch, and Michael Chan. [ Ported to current tree and all drivers converted. Integrated Stephen's follow-on kerneldoc additions, and restored poll_list handling to the old style to fix mutual exclusion issues. -DaveM ] Signed-off-by: Stephen Hemminger <shemminger@linux-foundation.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-10-04 07:41:36 +08:00
return work_done;
}
#endif
static void rtl8169_down(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
unsigned int intrmask;
rtl8169_delete_timer(dev);
netif_stop_queue(dev);
#ifdef CONFIG_R8169_NAPI
napi_disable(&tp->napi);
#endif
core_down:
spin_lock_irq(&tp->lock);
rtl8169_asic_down(ioaddr);
/* Update the error counts. */
dev->stats.rx_missed_errors += RTL_R32(RxMissed);
RTL_W32(RxMissed, 0);
spin_unlock_irq(&tp->lock);
synchronize_irq(dev->irq);
/* Give a racing hard_start_xmit a few cycles to complete. */
synchronize_sched(); /* FIXME: should this be synchronize_irq()? */
/*
* And now for the 50k$ question: are IRQ disabled or not ?
*
* Two paths lead here:
* 1) dev->close
* -> netif_running() is available to sync the current code and the
* IRQ handler. See rtl8169_interrupt for details.
* 2) dev->change_mtu
* -> rtl8169_poll can not be issued again and re-enable the
* interruptions. Let's simply issue the IRQ down sequence again.
*
* No loop if hotpluged or major error (0xffff).
*/
intrmask = RTL_R16(IntrMask);
if (intrmask && (intrmask != 0xffff))
goto core_down;
rtl8169_tx_clear(tp);
rtl8169_rx_clear(tp);
}
static int rtl8169_close(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
struct pci_dev *pdev = tp->pci_dev;
rtl8169_down(dev);
free_irq(dev->irq, dev);
pci_free_consistent(pdev, R8169_RX_RING_BYTES, tp->RxDescArray,
tp->RxPhyAddr);
pci_free_consistent(pdev, R8169_TX_RING_BYTES, tp->TxDescArray,
tp->TxPhyAddr);
tp->TxDescArray = NULL;
tp->RxDescArray = NULL;
return 0;
}
static void rtl_set_rx_mode(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
unsigned long flags;
u32 mc_filter[2]; /* Multicast hash filter */
int rx_mode;
u32 tmp = 0;
if (dev->flags & IFF_PROMISC) {
/* Unconditionally log net taps. */
if (netif_msg_link(tp)) {
printk(KERN_NOTICE "%s: Promiscuous mode enabled.\n",
dev->name);
}
rx_mode =
AcceptBroadcast | AcceptMulticast | AcceptMyPhys |
AcceptAllPhys;
mc_filter[1] = mc_filter[0] = 0xffffffff;
} else if ((dev->mc_count > multicast_filter_limit)
|| (dev->flags & IFF_ALLMULTI)) {
/* Too many to filter perfectly -- accept all multicasts. */
rx_mode = AcceptBroadcast | AcceptMulticast | AcceptMyPhys;
mc_filter[1] = mc_filter[0] = 0xffffffff;
} else {
struct dev_mc_list *mclist;
unsigned int i;
rx_mode = AcceptBroadcast | AcceptMyPhys;
mc_filter[1] = mc_filter[0] = 0;
for (i = 0, mclist = dev->mc_list; mclist && i < dev->mc_count;
i++, mclist = mclist->next) {
int bit_nr = ether_crc(ETH_ALEN, mclist->dmi_addr) >> 26;
mc_filter[bit_nr >> 5] |= 1 << (bit_nr & 31);
rx_mode |= AcceptMulticast;
}
}
spin_lock_irqsave(&tp->lock, flags);
tmp = rtl8169_rx_config | rx_mode |
(RTL_R32(RxConfig) & rtl_chip_info[tp->chipset].RxConfigMask);
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
if ((tp->mac_version == RTL_GIGA_MAC_VER_11) ||
(tp->mac_version == RTL_GIGA_MAC_VER_12) ||
(tp->mac_version == RTL_GIGA_MAC_VER_13) ||
(tp->mac_version == RTL_GIGA_MAC_VER_14) ||
(tp->mac_version == RTL_GIGA_MAC_VER_15) ||
(tp->mac_version == RTL_GIGA_MAC_VER_16) ||
(tp->mac_version == RTL_GIGA_MAC_VER_17)) {
r8169: sync with vendor's driver - add several PCI ID for the PCI-E adapters ; - new identification strings ; - the RTL_GIGA_MAC_VER_ defines have been renamed to closely match the out-of-tree driver. It makes the comparison less hairy ; - various magic ; - the PCI region for the device with PCI ID 0x8136 is guessed. Explanation: the in-kernel Linux driver is written to allow MM register accesses and avoid the IO tax. The relevant BAR register was found at base address 1 for the plain-old PCI 8169. User reported lspci show that it is found at base address 2 for the new Gigabit PCI-E 816{8/9}. Typically: 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.: Unknown device 8168 (rev 01) Subsystem: Unknown device 1631:e015 Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- Status: Cap+ 66Mhz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- Latency: 0, cache line size 20 Interrupt: pin A routed to IRQ 16 Region 0: I/O ports at b800 [size=256] Region 2: Memory at ff7ff000 (64-bit, non-prefetchable) [size=4K] ^^^^^^^^ So far I have not received any lspci report for the 0x8136 and Realtek's driver do not help: be it under BSD or Linux, their r1000 driver include a USE_IO_SPACE #define but the bar address is always hardcoded to 1 in the MM case. :o/ - the 8168 has been reported to require an extra alignment for its receive buffers. The status of the 8167 and 8136 is not known in this regard. Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
2006-07-27 05:14:13 +08:00
mc_filter[0] = 0xffffffff;
mc_filter[1] = 0xffffffff;
}
RTL_W32(MAR0 + 0, mc_filter[0]);
RTL_W32(MAR0 + 4, mc_filter[1]);
RTL_W32(RxConfig, tmp);
spin_unlock_irqrestore(&tp->lock, flags);
}
/**
* rtl8169_get_stats - Get rtl8169 read/write statistics
* @dev: The Ethernet Device to get statistics for
*
* Get TX/RX statistics for rtl8169
*/
static struct net_device_stats *rtl8169_get_stats(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
unsigned long flags;
if (netif_running(dev)) {
spin_lock_irqsave(&tp->lock, flags);
dev->stats.rx_missed_errors += RTL_R32(RxMissed);
RTL_W32(RxMissed, 0);
spin_unlock_irqrestore(&tp->lock, flags);
}
return &dev->stats;
}
#ifdef CONFIG_PM
static int rtl8169_suspend(struct pci_dev *pdev, pm_message_t state)
{
struct net_device *dev = pci_get_drvdata(pdev);
struct rtl8169_private *tp = netdev_priv(dev);
void __iomem *ioaddr = tp->mmio_addr;
if (!netif_running(dev))
goto out_pci_suspend;
netif_device_detach(dev);
netif_stop_queue(dev);
spin_lock_irq(&tp->lock);
rtl8169_asic_down(ioaddr);
dev->stats.rx_missed_errors += RTL_R32(RxMissed);
RTL_W32(RxMissed, 0);
spin_unlock_irq(&tp->lock);
out_pci_suspend:
pci_save_state(pdev);
pci_enable_wake(pdev, pci_choose_state(pdev, state),
(tp->features & RTL_FEATURE_WOL) ? 1 : 0);
pci_set_power_state(pdev, pci_choose_state(pdev, state));
return 0;
}
static int rtl8169_resume(struct pci_dev *pdev)
{
struct net_device *dev = pci_get_drvdata(pdev);
pci_set_power_state(pdev, PCI_D0);
pci_restore_state(pdev);
pci_enable_wake(pdev, PCI_D0, 0);
if (!netif_running(dev))
goto out;
netif_device_attach(dev);
rtl8169_schedule_work(dev, rtl8169_reset_task);
out:
return 0;
}
#endif /* CONFIG_PM */
static struct pci_driver rtl8169_pci_driver = {
.name = MODULENAME,
.id_table = rtl8169_pci_tbl,
.probe = rtl8169_init_one,
.remove = __devexit_p(rtl8169_remove_one),
#ifdef CONFIG_PM
.suspend = rtl8169_suspend,
.resume = rtl8169_resume,
#endif
};
static int __init rtl8169_init_module(void)
{
return pci_register_driver(&rtl8169_pci_driver);
}
static void __exit rtl8169_cleanup_module(void)
{
pci_unregister_driver(&rtl8169_pci_driver);
}
module_init(rtl8169_init_module);
module_exit(rtl8169_cleanup_module);