W: http://www.ibm.com/developerworks/power/cell/
S: Supported
+CISCO 10G ETHERNET DRIVER
+P: Scott Feldman
+M: scofeldm@cisco.com
+P: Joe Eykholt
+M: jeykholt@cisco.com
+S: Supported
+
CFAG12864B LCD DRIVER
P: Miguel Ojeda Sandonis
M: miguel.ojeda.sandonis@gmail.com
S: Maintained
CPUSETS
- P: Paul Jackson
P: Paul Menage
- M: pj@sgi.com
M: menage@google.com
L: linux-kernel@vger.kernel.org
W: http://www.bullopensource.org/cpuset/
W: http://www.ivtvdriver.org
S: Maintained
+JME NETWORK DRIVER
+P: Guo-Fu Tseng
+M: cooldavid@cooldavid.org
+L: netdev@vger.kernel.org
+S: Maintained
+
JOURNALLING FLASH FILE SYSTEM V2 (JFFS2)
P: David Woodhouse
M: dwmw2@infradead.org
P: Michael Kerrisk
M: mtk.manpages@gmail.com
W: http://www.kernel.org/doc/man-pages
+ L: linux-man@vger.kernel.org
S: Supported
MARVELL LIBERTAS WIRELESS DRIVER
L: netdev@vger.kernel.org
S: Supported
+QLOGIC QLGE 10Gb ETHERNET DRIVER
+P: Ron Mercer
+M: linux-driver@qlogic.com
+M: ron.mercer@qlogic.com
+L: netdev@vger.kernel.org
+S: Supported
+
QNX4 FILESYSTEM
P: Anders Larsen
M: al@alarsen.net
e_printk(KERN_NOTICE, adapter, format, ## arg)
+/* Interrupt modes, as used by the IntMode paramter */
+#define E1000E_INT_MODE_LEGACY 0
+#define E1000E_INT_MODE_MSI 1
+#define E1000E_INT_MODE_MSIX 2
+
/* Tx/Rx descriptor defines */
#define E1000_DEFAULT_TXD 256
#define E1000_MAX_TXD 4096
board_82571,
board_82572,
board_82573,
+ board_82574,
board_80003es2lan,
board_ich8lan,
board_ich9lan,
+ board_ich10lan,
};
struct e1000_queue_stats {
/* array of buffer information structs */
struct e1000_buffer *buffer_info;
+ char name[IFNAMSIZ + 5];
+ u32 ims_val;
+ u32 itr_val;
+ u16 itr_register;
+ int set_itr;
+
struct sk_buff *rx_skb_top;
struct e1000_queue_stats stats;
struct net_device *netdev;
struct pci_dev *pdev;
struct net_device_stats net_stats;
- spinlock_t stats_lock; /* prevent concurrent stats updates */
/* structs defined in e1000_hw.h */
struct e1000_hw hw;
u32 test_icr;
u32 msg_enable;
+ struct msix_entry *msix_entries;
+ int int_mode;
+ u32 eiac_mask;
u32 eeprom_wol;
u32 wol;
unsigned long led_status;
unsigned int flags;
+ struct work_struct downshift_task;
+ struct work_struct update_phy_task;
};
struct e1000_info {
#define FLAG_HAS_CTRLEXT_ON_LOAD (1 << 5)
#define FLAG_HAS_SWSM_ON_LOAD (1 << 6)
#define FLAG_HAS_JUMBO_FRAMES (1 << 7)
+ #define FLAG_READ_ONLY_NVM (1 << 8)
#define FLAG_IS_ICH (1 << 9)
+#define FLAG_HAS_MSIX (1 << 10)
#define FLAG_HAS_SMART_POWER_DOWN (1 << 11)
#define FLAG_IS_QUAD_PORT_A (1 << 12)
#define FLAG_IS_QUAD_PORT (1 << 13)
extern void e1000e_free_rx_resources(struct e1000_adapter *adapter);
extern void e1000e_free_tx_resources(struct e1000_adapter *adapter);
extern void e1000e_update_stats(struct e1000_adapter *adapter);
+extern void e1000e_set_interrupt_capability(struct e1000_adapter *adapter);
+extern void e1000e_reset_interrupt_capability(struct e1000_adapter *adapter);
extern unsigned int copybreak;
extern struct e1000_info e1000_82571_info;
extern struct e1000_info e1000_82572_info;
extern struct e1000_info e1000_82573_info;
+extern struct e1000_info e1000_82574_info;
extern struct e1000_info e1000_ich8_info;
extern struct e1000_info e1000_ich9_info;
+extern struct e1000_info e1000_ich10_info;
extern struct e1000_info e1000_es2_info;
extern s32 e1000e_read_pba_num(struct e1000_hw *hw, u32 *pba_num);
extern bool e1000e_get_laa_state_82571(struct e1000_hw *hw);
extern void e1000e_set_laa_state_82571(struct e1000_hw *hw, bool state);
+ extern void e1000e_write_protect_nvm_ich8lan(struct e1000_hw *hw);
extern void e1000e_set_kmrn_lock_loss_workaround_ich8lan(struct e1000_hw *hw,
bool state);
extern void e1000e_igp3_phy_powerdown_workaround_ich8lan(struct e1000_hw *hw);
extern s32 e1000e_get_phy_info_m88(struct e1000_hw *hw);
extern s32 e1000e_read_phy_reg_m88(struct e1000_hw *hw, u32 offset, u16 *data);
extern s32 e1000e_write_phy_reg_m88(struct e1000_hw *hw, u32 offset, u16 data);
+extern s32 e1000e_phy_init_script_igp3(struct e1000_hw *hw);
extern enum e1000_phy_type e1000e_get_phy_type_from_id(u32 phy_id);
extern s32 e1000e_determine_phy_address(struct e1000_hw *hw);
extern s32 e1000e_write_phy_reg_bm(struct e1000_hw *hw, u32 offset, u16 data);
extern s32 e1000e_read_phy_reg_bm(struct e1000_hw *hw, u32 offset, u16 *data);
+extern s32 e1000e_read_phy_reg_bm2(struct e1000_hw *hw, u32 offset, u16 *data);
+extern s32 e1000e_write_phy_reg_bm2(struct e1000_hw *hw, u32 offset, u16 data);
extern void e1000e_phy_force_speed_duplex_setup(struct e1000_hw *hw, u16 *phy_ctrl);
extern s32 e1000e_write_kmrn_reg(struct e1000_hw *hw, u32 offset, u16 data);
extern s32 e1000e_read_kmrn_reg(struct e1000_hw *hw, u32 offset, u16 *data);
return hw->phy.ops.get_phy_info(hw);
}
-extern bool e1000e_check_mng_mode(struct e1000_hw *hw);
+static inline s32 e1000e_check_mng_mode(struct e1000_hw *hw)
+{
+ return hw->mac.ops.check_mng_mode(hw);
+}
+
+extern bool e1000e_check_mng_mode_generic(struct e1000_hw *hw);
extern bool e1000e_enable_tx_pkt_filtering(struct e1000_hw *hw);
extern s32 e1000e_mng_write_dhcp_info(struct e1000_hw *hw, u8 *buffer, u16 length);
regs_buff[11] = er32(TIDV);
regs_buff[12] = adapter->hw.phy.type; /* PHY type (IGP=1, M88=0) */
+
+ /* ethtool doesn't use anything past this point, so all this
+ * code is likely legacy junk for apps that may or may not
+ * exist */
if (hw->phy.type == e1000_phy_m88) {
e1e_rphy(hw, M88E1000_PHY_SPEC_STATUS, &phy_data);
regs_buff[13] = (u32)phy_data; /* cable length */
regs_buff[22] = adapter->phy_stats.receive_errors;
regs_buff[23] = regs_buff[13]; /* mdix mode */
}
- regs_buff[21] = adapter->phy_stats.idle_errors; /* phy idle errors */
+ regs_buff[21] = 0; /* was idle_errors */
e1e_rphy(hw, PHY_1000T_STATUS, &phy_data);
regs_buff[24] = (u32)phy_data; /* phy local receiver status */
regs_buff[25] = regs_buff[24]; /* phy remote receiver status */
if (eeprom->magic != (adapter->pdev->vendor | (adapter->pdev->device << 16)))
return -EFAULT;
+ if (adapter->flags & FLAG_READ_ONLY_NVM)
+ return -EINVAL;
+
max_len = hw->nvm.word_size * 2;
first_word = eeprom->offset >> 1;
* and flush shadow RAM for 82573 controllers
*/
if ((ret_val == 0) && ((first_word <= NVM_CHECKSUM_REG) ||
+ (hw->mac.type == e1000_82574) ||
(hw->mac.type == e1000_82573)))
e1000e_update_nvm_checksum(hw);
toggle = 0x7FFFF3FF;
break;
case e1000_82573:
+ case e1000_82574:
case e1000_ich8lan:
case e1000_ich9lan:
+ case e1000_ich10lan:
toggle = 0x7FFFF033;
break;
default:
REG_PATTERN_TEST(E1000_TIDV, 0x0000FFFF, 0x0000FFFF);
for (i = 0; i < mac->rar_entry_count; i++)
REG_PATTERN_TEST_ARRAY(E1000_RA, ((i << 1) + 1),
- 0x8003FFFF, 0xFFFFFFFF);
+ ((mac->type == e1000_ich10lan) ?
+ 0x8007FFFF : 0x8003FFFF),
+ 0xFFFFFFFF);
for (i = 0; i < mac->mta_reg_count; i++)
REG_PATTERN_TEST_ARRAY(E1000_MTA, i, 0xFFFFFFFF, 0xFFFFFFFF);
u32 shared_int = 1;
u32 irq = adapter->pdev->irq;
int i;
+ int ret_val = 0;
+ int int_mode = E1000E_INT_MODE_LEGACY;
*data = 0;
- /* NOTE: we don't test MSI interrupts here, yet */
+ /* NOTE: we don't test MSI/MSI-X interrupts here, yet */
+ if (adapter->int_mode == E1000E_INT_MODE_MSIX) {
+ int_mode = adapter->int_mode;
+ e1000e_reset_interrupt_capability(adapter);
+ adapter->int_mode = E1000E_INT_MODE_LEGACY;
+ e1000e_set_interrupt_capability(adapter);
+ }
/* Hook up test interrupt handler just for this test */
if (!request_irq(irq, &e1000_test_intr, IRQF_PROBE_SHARED, netdev->name,
netdev)) {
} else if (request_irq(irq, &e1000_test_intr, IRQF_SHARED,
netdev->name, netdev)) {
*data = 1;
- return -1;
+ ret_val = -1;
+ goto out;
}
e_info("testing %s interrupt\n", (shared_int ? "shared" : "unshared"));
/* Test each interrupt */
for (i = 0; i < 10; i++) {
- if ((adapter->flags & FLAG_IS_ICH) && (i == 8))
- continue;
-
/* Interrupt to test */
mask = 1 << i;
+ if (adapter->flags & FLAG_IS_ICH) {
+ switch (mask) {
+ case E1000_ICR_RXSEQ:
+ continue;
+ case 0x00000100:
+ if (adapter->hw.mac.type == e1000_ich8lan ||
+ adapter->hw.mac.type == e1000_ich9lan)
+ continue;
+ break;
+ default:
+ break;
+ }
+ }
+
if (!shared_int) {
/*
* Disable the interrupt to be reported in
/* Unhook test interrupt handler */
free_irq(irq, netdev);
- return *data;
+out:
+ if (int_mode == E1000E_INT_MODE_MSIX) {
+ e1000e_reset_interrupt_capability(adapter);
+ adapter->int_mode = int_mode;
+ e1000e_set_interrupt_capability(adapter);
+ }
+
+ return ret_val;
}
static void e1000_free_desc_rings(struct e1000_adapter *adapter)
static int e1000_phys_id(struct net_device *netdev, u32 data)
{
struct e1000_adapter *adapter = netdev_priv(netdev);
+ struct e1000_hw *hw = &adapter->hw;
if (!data)
data = INT_MAX;
- if (adapter->hw.phy.type == e1000_phy_ife) {
+ if ((hw->phy.type == e1000_phy_ife) ||
+ (hw->mac.type == e1000_82574)) {
if (!adapter->blink_timer.function) {
init_timer(&adapter->blink_timer);
adapter->blink_timer.function =
mod_timer(&adapter->blink_timer, jiffies);
msleep_interruptible(data * 1000);
del_timer_sync(&adapter->blink_timer);
- e1e_wphy(&adapter->hw,
- IFE_PHY_SPECIAL_CONTROL_LED, 0);
+ if (hw->phy.type == e1000_phy_ife)
+ e1e_wphy(hw, IFE_PHY_SPECIAL_CONTROL_LED, 0);
} else {
- e1000e_blink_led(&adapter->hw);
+ e1000e_blink_led(hw);
msleep_interruptible(data * 1000);
}
- adapter->hw.mac.ops.led_off(&adapter->hw);
+ hw->mac.ops.led_off(hw);
clear_bit(E1000_LED_ON, &adapter->led_status);
- adapter->hw.mac.ops.cleanup_led(&adapter->hw);
+ hw->mac.ops.cleanup_led(hw);
return 0;
}
* 82567LM-2 Gigabit Network Connection
* 82567LF-2 Gigabit Network Connection
* 82567V-2 Gigabit Network Connection
- * 82562GT-3 10/100 Network Connection
+ * 82567LF-3 Gigabit Network Connection
+ * 82567LM-3 Gigabit Network Connection
+ * 82567LM-4 Gigabit Network Connection
*/
#include <linux/netdevice.h>
#define ICH_FLASH_HSFCTL 0x0006
#define ICH_FLASH_FADDR 0x0008
#define ICH_FLASH_FDATA0 0x0010
+ #define ICH_FLASH_PR0 0x0074
#define ICH_FLASH_READ_COMMAND_TIMEOUT 500
#define ICH_FLASH_WRITE_COMMAND_TIMEOUT 500
u16 regval;
};
+ /* ICH Flash Protected Region */
+ union ich8_flash_protected_range {
+ struct ich8_pr {
+ u32 base:13; /* 0:12 Protected Range Base */
+ u32 reserved1:2; /* 13:14 Reserved */
+ u32 rpe:1; /* 15 Read Protection Enable */
+ u32 limit:13; /* 16:28 Protected Range Limit */
+ u32 reserved2:2; /* 29:30 Reserved */
+ u32 wpe:1; /* 31 Write Protection Enable */
+ } range;
+ u32 regval;
+ };
+
static s32 e1000_setup_link_ich8lan(struct e1000_hw *hw);
static void e1000_clear_hw_cntrs_ich8lan(struct e1000_hw *hw);
static void e1000_initialize_hw_bits_ich8lan(struct e1000_hw *hw);
static s32 e1000_erase_flash_bank_ich8lan(struct e1000_hw *hw, u32 bank);
static s32 e1000_retry_write_flash_byte_ich8lan(struct e1000_hw *hw,
u32 offset, u8 byte);
+static s32 e1000_read_flash_byte_ich8lan(struct e1000_hw *hw, u32 offset,
+ u8 *data);
static s32 e1000_read_flash_word_ich8lan(struct e1000_hw *hw, u32 offset,
u16 *data);
static s32 e1000_read_flash_data_ich8lan(struct e1000_hw *hw, u32 offset,
u8 size, u16 *data);
static s32 e1000_setup_copper_link_ich8lan(struct e1000_hw *hw);
static s32 e1000_kmrn_lock_loss_workaround_ich8lan(struct e1000_hw *hw);
+static s32 e1000_get_cfg_done_ich8lan(struct e1000_hw *hw);
static inline u16 __er16flash(struct e1000_hw *hw, unsigned long reg)
{
return 0;
}
+ static DEFINE_MUTEX(nvm_mutex);
+ static pid_t nvm_owner = -1;
+
/**
* e1000_acquire_swflag_ich8lan - Acquire software control flag
* @hw: pointer to the HW structure
u32 extcnf_ctrl;
u32 timeout = PHY_CFG_TIMEOUT;
+ might_sleep();
+
+ if (!mutex_trylock(&nvm_mutex)) {
+ WARN(1, KERN_ERR "e1000e mutex contention. Owned by pid %d\n",
+ nvm_owner);
+ mutex_lock(&nvm_mutex);
+ }
+ nvm_owner = current->pid;
+
while (timeout) {
extcnf_ctrl = er32(EXTCNF_CTRL);
extcnf_ctrl |= E1000_EXTCNF_CTRL_SWFLAG;
if (!timeout) {
hw_dbg(hw, "FW or HW has locked the resource for too long.\n");
+ extcnf_ctrl &= ~E1000_EXTCNF_CTRL_SWFLAG;
+ ew32(EXTCNF_CTRL, extcnf_ctrl);
+ nvm_owner = -1;
+ mutex_unlock(&nvm_mutex);
return -E1000_ERR_CONFIG;
}
extcnf_ctrl = er32(EXTCNF_CTRL);
extcnf_ctrl &= ~E1000_EXTCNF_CTRL_SWFLAG;
ew32(EXTCNF_CTRL, extcnf_ctrl);
+
+ nvm_owner = -1;
+ mutex_unlock(&nvm_mutex);
}
+/**
+ * e1000_check_mng_mode_ich8lan - Checks management mode
+ * @hw: pointer to the HW structure
+ *
+ * This checks if the adapter has manageability enabled.
+ * This is a function pointer entry point only called by read/write
+ * routines for the PHY and NVM parts.
+ **/
+static bool e1000_check_mng_mode_ich8lan(struct e1000_hw *hw)
+{
+ u32 fwsm = er32(FWSM);
+
+ return (fwsm & E1000_FWSM_MODE_MASK) ==
+ (E1000_ICH_MNG_IAMT_MODE << E1000_FWSM_MODE_SHIFT);
+}
+
/**
* e1000_check_reset_block_ich8lan - Check if PHY reset is blocked
* @hw: pointer to the HW structure
return 0;
}
+/**
+ * e1000_valid_nvm_bank_detect_ich8lan - finds out the valid bank 0 or 1
+ * @hw: pointer to the HW structure
+ * @bank: pointer to the variable that returns the active bank
+ *
+ * Reads signature byte from the NVM using the flash access registers.
+ **/
+static s32 e1000_valid_nvm_bank_detect_ich8lan(struct e1000_hw *hw, u32 *bank)
+{
+ struct e1000_nvm_info *nvm = &hw->nvm;
+ /* flash bank size is in words */
+ u32 bank1_offset = nvm->flash_bank_size * sizeof(u16);
+ u32 act_offset = E1000_ICH_NVM_SIG_WORD * 2 + 1;
+ u8 bank_high_byte = 0;
+
+ if (hw->mac.type != e1000_ich10lan) {
+ if (er32(EECD) & E1000_EECD_SEC1VAL)
+ *bank = 1;
+ else
+ *bank = 0;
+ } else {
+ /*
+ * Make sure the signature for bank 0 is valid,
+ * if not check for bank1
+ */
+ e1000_read_flash_byte_ich8lan(hw, act_offset, &bank_high_byte);
+ if ((bank_high_byte & 0xC0) == 0x80) {
+ *bank = 0;
+ } else {
+ /*
+ * find if segment 1 is valid by verifying
+ * bit 15:14 = 10b in word 0x13
+ */
+ e1000_read_flash_byte_ich8lan(hw,
+ act_offset + bank1_offset,
+ &bank_high_byte);
+
+ /* bank1 has a valid signature equivalent to SEC1V */
+ if ((bank_high_byte & 0xC0) == 0x80) {
+ *bank = 1;
+ } else {
+ hw_dbg(hw, "ERROR: EEPROM not present\n");
+ return -E1000_ERR_NVM;
+ }
+ }
+ }
+
+ return 0;
+}
+
/**
* e1000_read_nvm_ich8lan - Read word(s) from the NVM
* @hw: pointer to the HW structure
struct e1000_dev_spec_ich8lan *dev_spec = &hw->dev_spec.ich8lan;
u32 act_offset;
s32 ret_val;
+ u32 bank = 0;
u16 i, word;
if ((offset >= nvm->word_size) || (words > nvm->word_size - offset) ||
if (ret_val)
return ret_val;
- /* Start with the bank offset, then add the relative offset. */
- act_offset = (er32(EECD) & E1000_EECD_SEC1VAL)
- ? nvm->flash_bank_size
- : 0;
+ ret_val = e1000_valid_nvm_bank_detect_ich8lan(hw, &bank);
+ if (ret_val)
+ return ret_val;
+
+ act_offset = (bank) ? nvm->flash_bank_size : 0;
act_offset += offset;
for (i = 0; i < words; i++) {
return e1000_read_flash_data_ich8lan(hw, offset, 2, data);
}
+/**
+ * e1000_read_flash_byte_ich8lan - Read byte from flash
+ * @hw: pointer to the HW structure
+ * @offset: The offset of the byte to read.
+ * @data: Pointer to a byte to store the value read.
+ *
+ * Reads a single byte from the NVM using the flash access registers.
+ **/
+static s32 e1000_read_flash_byte_ich8lan(struct e1000_hw *hw, u32 offset,
+ u8 *data)
+{
+ s32 ret_val;
+ u16 word = 0;
+
+ ret_val = e1000_read_flash_data_ich8lan(hw, offset, 1, &word);
+ if (ret_val)
+ return ret_val;
+
+ *data = (u8)word;
+
+ return 0;
+}
+
/**
* e1000_read_flash_data_ich8lan - Read byte or word from NVM
* @hw: pointer to the HW structure
{
struct e1000_nvm_info *nvm = &hw->nvm;
struct e1000_dev_spec_ich8lan *dev_spec = &hw->dev_spec.ich8lan;
- u32 i, act_offset, new_bank_offset, old_bank_offset;
+ u32 i, act_offset, new_bank_offset, old_bank_offset, bank;
s32 ret_val;
u16 data;
* write to bank 0 etc. We also need to erase the segment that
* is going to be written
*/
- if (!(er32(EECD) & E1000_EECD_SEC1VAL)) {
+ ret_val = e1000_valid_nvm_bank_detect_ich8lan(hw, &bank);
+ if (ret_val)
+ return ret_val;
+
+ if (bank == 0) {
new_bank_offset = nvm->flash_bank_size;
old_bank_offset = 0;
e1000_erase_flash_bank_ich8lan(hw, 1);
* programming failed.
*/
if (ret_val) {
+ /* Possibly read-only, see e1000e_write_protect_nvm_ich8lan() */
hw_dbg(hw, "Flash commit failed.\n");
e1000_release_swflag_ich8lan(hw);
return ret_val;
return e1000e_validate_nvm_checksum_generic(hw);
}
+ /**
+ * e1000e_write_protect_nvm_ich8lan - Make the NVM read-only
+ * @hw: pointer to the HW structure
+ *
+ * To prevent malicious write/erase of the NVM, set it to be read-only
+ * so that the hardware ignores all write/erase cycles of the NVM via
+ * the flash control registers. The shadow-ram copy of the NVM will
+ * still be updated, however any updates to this copy will not stick
+ * across driver reloads.
+ **/
+ void e1000e_write_protect_nvm_ich8lan(struct e1000_hw *hw)
+ {
+ union ich8_flash_protected_range pr0;
+ union ich8_hws_flash_status hsfsts;
+ u32 gfpreg;
+ s32 ret_val;
+
+ ret_val = e1000_acquire_swflag_ich8lan(hw);
+ if (ret_val)
+ return;
+
+ gfpreg = er32flash(ICH_FLASH_GFPREG);
+
+ /* Write-protect GbE Sector of NVM */
+ pr0.regval = er32flash(ICH_FLASH_PR0);
+ pr0.range.base = gfpreg & FLASH_GFPREG_BASE_MASK;
+ pr0.range.limit = ((gfpreg >> 16) & FLASH_GFPREG_BASE_MASK);
+ pr0.range.wpe = true;
+ ew32flash(ICH_FLASH_PR0, pr0.regval);
+
+ /*
+ * Lock down a subset of GbE Flash Control Registers, e.g.
+ * PR0 to prevent the write-protection from being lifted.
+ * Once FLOCKDN is set, the registers protected by it cannot
+ * be written until FLOCKDN is cleared by a hardware reset.
+ */
+ hsfsts.regval = er16flash(ICH_FLASH_HSFSTS);
+ hsfsts.hsf_status.flockdn = true;
+ ew32flash(ICH_FLASH_HSFSTS, hsfsts.regval);
+
+ e1000_release_swflag_ich8lan(hw);
+ }
+
/**
* e1000_write_flash_data_ich8lan - Writes bytes to the NVM
* @hw: pointer to the HW structure
ew32(CTRL, (ctrl | E1000_CTRL_RST));
msleep(20);
+ /* release the swflag because it is not reset by hardware reset */
+ e1000_release_swflag_ich8lan(hw);
+
ret_val = e1000e_get_auto_rd_done(hw);
if (ret_val) {
/*
* 'LPLU Enabled' and 'Gig Disable' to force link speed negotiation
* to a lower speed.
*
- * Should only be called for ICH9 devices.
+ * Should only be called for ICH9 and ICH10 devices.
**/
void e1000e_disable_gig_wol_ich8lan(struct e1000_hw *hw)
{
u32 phy_ctrl;
- if (hw->mac.type == e1000_ich9lan) {
+ if ((hw->mac.type == e1000_ich10lan) ||
+ (hw->mac.type == e1000_ich9lan)) {
phy_ctrl = er32(PHY_CTRL);
phy_ctrl |= E1000_PHY_CTRL_D0A_LPLU |
E1000_PHY_CTRL_GBE_DISABLE;
return 0;
}
+/**
+ * e1000_get_cfg_done_ich8lan - Read config done bit
+ * @hw: pointer to the HW structure
+ *
+ * Read the management control register for the config done bit for
+ * completion status. NOTE: silicon which is EEPROM-less will fail trying
+ * to read the config done bit, so an error is *ONLY* logged and returns
+ * E1000_SUCCESS. If we were to return with error, EEPROM-less silicon
+ * would not be able to be reset or change link.
+ **/
+static s32 e1000_get_cfg_done_ich8lan(struct e1000_hw *hw)
+{
+ u32 bank = 0;
+
+ e1000e_get_cfg_done(hw);
+
+ /* If EEPROM is not marked present, init the IGP 3 PHY manually */
+ if (hw->mac.type != e1000_ich10lan) {
+ if (((er32(EECD) & E1000_EECD_PRES) == 0) &&
+ (hw->phy.type == e1000_phy_igp_3)) {
+ e1000e_phy_init_script_igp3(hw);
+ }
+ } else {
+ if (e1000_valid_nvm_bank_detect_ich8lan(hw, &bank)) {
+ /* Maybe we should do a basic PHY config */
+ hw_dbg(hw, "EEPROM not present\n");
+ return -E1000_ERR_CONFIG;
+ }
+ }
+
+ return 0;
+}
+
/**
* e1000_clear_hw_cntrs_ich8lan - Clear statistical counters
* @hw: pointer to the HW structure
}
static struct e1000_mac_operations ich8_mac_ops = {
- .mng_mode_enab = E1000_ICH_MNG_IAMT_MODE << E1000_FWSM_MODE_SHIFT,
+ .check_mng_mode = e1000_check_mng_mode_ich8lan,
.check_for_link = e1000e_check_for_copper_link,
.cleanup_led = e1000_cleanup_led_ich8lan,
.clear_hw_cntrs = e1000_clear_hw_cntrs_ich8lan,
.check_reset_block = e1000_check_reset_block_ich8lan,
.commit_phy = NULL,
.force_speed_duplex = e1000_phy_force_speed_duplex_ich8lan,
- .get_cfg_done = e1000e_get_cfg_done,
+ .get_cfg_done = e1000_get_cfg_done_ich8lan,
.get_cable_length = e1000e_get_cable_length_igp_2,
.get_phy_info = e1000_get_phy_info_ich8lan,
.read_phy_reg = e1000e_read_phy_reg_igp,
.nvm_ops = &ich8_nvm_ops,
};
+struct e1000_info e1000_ich10_info = {
+ .mac = e1000_ich10lan,
+ .flags = FLAG_HAS_JUMBO_FRAMES
+ | FLAG_IS_ICH
+ | FLAG_HAS_WOL
+ | FLAG_RX_CSUM_ENABLED
+ | FLAG_HAS_CTRLEXT_ON_LOAD
+ | FLAG_HAS_AMT
+ | FLAG_HAS_ERT
+ | FLAG_HAS_FLASH
+ | FLAG_APME_IN_WUC,
+ .pba = 10,
+ .get_variants = e1000_get_variants_ich8lan,
+ .mac_ops = &ich8_mac_ops,
+ .phy_ops = &ich8_phy_ops,
+ .nvm_ops = &ich8_nvm_ops,
+};
#include "e1000.h"
- #define DRV_VERSION "0.3.3.3-k2"
+ #define DRV_VERSION "0.3.3.3-k6"
char e1000e_driver_name[] = "e1000e";
const char e1000e_driver_version[] = DRV_VERSION;
[board_82571] = &e1000_82571_info,
[board_82572] = &e1000_82572_info,
[board_82573] = &e1000_82573_info,
+ [board_82574] = &e1000_82574_info,
[board_80003es2lan] = &e1000_es2_info,
[board_ich8lan] = &e1000_ich8_info,
[board_ich9lan] = &e1000_ich9_info,
+ [board_ich10lan] = &e1000_ich10_info,
};
#ifdef DEBUG
writel(0, adapter->hw.hw_addr + rx_ring->tail);
}
+ static void e1000e_downshift_workaround(struct work_struct *work)
+ {
+ struct e1000_adapter *adapter = container_of(work,
+ struct e1000_adapter, downshift_task);
+
+ e1000e_gig_downshift_workaround_ich8lan(&adapter->hw);
+ }
+
/**
* e1000_intr_msi - Interrupt Handler
* @irq: interrupt number
*/
if ((adapter->flags & FLAG_LSC_GIG_SPEED_DROP) &&
(!(er32(STATUS) & E1000_STATUS_LU)))
- e1000e_gig_downshift_workaround_ich8lan(hw);
+ schedule_work(&adapter->downshift_task);
/*
* 80003ES2LAN workaround-- For packet buffer work-around on
struct net_device *netdev = data;
struct e1000_adapter *adapter = netdev_priv(netdev);
struct e1000_hw *hw = &adapter->hw;
-
u32 rctl, icr = er32(ICR);
+
if (!icr)
return IRQ_NONE; /* Not our interrupt */
*/
if ((adapter->flags & FLAG_LSC_GIG_SPEED_DROP) &&
(!(er32(STATUS) & E1000_STATUS_LU)))
- e1000e_gig_downshift_workaround_ich8lan(hw);
+ schedule_work(&adapter->downshift_task);
/*
* 80003ES2LAN workaround--
return IRQ_HANDLED;
}
+static irqreturn_t e1000_msix_other(int irq, void *data)
+{
+ struct net_device *netdev = data;
+ struct e1000_adapter *adapter = netdev_priv(netdev);
+ struct e1000_hw *hw = &adapter->hw;
+ u32 icr = er32(ICR);
+
+ if (!(icr & E1000_ICR_INT_ASSERTED)) {
+ ew32(IMS, E1000_IMS_OTHER);
+ return IRQ_NONE;
+ }
+
+ if (icr & adapter->eiac_mask)
+ ew32(ICS, (icr & adapter->eiac_mask));
+
+ if (icr & E1000_ICR_OTHER) {
+ if (!(icr & E1000_ICR_LSC))
+ goto no_link_interrupt;
+ hw->mac.get_link_status = 1;
+ /* guard against interrupt when we're going down */
+ if (!test_bit(__E1000_DOWN, &adapter->state))
+ mod_timer(&adapter->watchdog_timer, jiffies + 1);
+ }
+
+no_link_interrupt:
+ ew32(IMS, E1000_IMS_LSC | E1000_IMS_OTHER);
+
+ return IRQ_HANDLED;
+}
+
+
+static irqreturn_t e1000_intr_msix_tx(int irq, void *data)
+{
+ struct net_device *netdev = data;
+ struct e1000_adapter *adapter = netdev_priv(netdev);
+ struct e1000_hw *hw = &adapter->hw;
+ struct e1000_ring *tx_ring = adapter->tx_ring;
+
+
+ adapter->total_tx_bytes = 0;
+ adapter->total_tx_packets = 0;
+
+ if (!e1000_clean_tx_irq(adapter))
+ /* Ring was not completely cleaned, so fire another interrupt */
+ ew32(ICS, tx_ring->ims_val);
+
+ return IRQ_HANDLED;
+}
+
+static irqreturn_t e1000_intr_msix_rx(int irq, void *data)
+{
+ struct net_device *netdev = data;
+ struct e1000_adapter *adapter = netdev_priv(netdev);
+
+ /* Write the ITR value calculated at the end of the
+ * previous interrupt.
+ */
+ if (adapter->rx_ring->set_itr) {
+ writel(1000000000 / (adapter->rx_ring->itr_val * 256),
+ adapter->hw.hw_addr + adapter->rx_ring->itr_register);
+ adapter->rx_ring->set_itr = 0;
+ }
+
+ if (netif_rx_schedule_prep(netdev, &adapter->napi)) {
+ adapter->total_rx_bytes = 0;
+ adapter->total_rx_packets = 0;
+ __netif_rx_schedule(netdev, &adapter->napi);
+ }
+ return IRQ_HANDLED;
+}
+
+/**
+ * e1000_configure_msix - Configure MSI-X hardware
+ *
+ * e1000_configure_msix sets up the hardware to properly
+ * generate MSI-X interrupts.
+ **/
+static void e1000_configure_msix(struct e1000_adapter *adapter)
+{
+ struct e1000_hw *hw = &adapter->hw;
+ struct e1000_ring *rx_ring = adapter->rx_ring;
+ struct e1000_ring *tx_ring = adapter->tx_ring;
+ int vector = 0;
+ u32 ctrl_ext, ivar = 0;
+
+ adapter->eiac_mask = 0;
+
+ /* Workaround issue with spurious interrupts on 82574 in MSI-X mode */
+ if (hw->mac.type == e1000_82574) {
+ u32 rfctl = er32(RFCTL);
+ rfctl |= E1000_RFCTL_ACK_DIS;
+ ew32(RFCTL, rfctl);
+ }
+
+#define E1000_IVAR_INT_ALLOC_VALID 0x8
+ /* Configure Rx vector */
+ rx_ring->ims_val = E1000_IMS_RXQ0;
+ adapter->eiac_mask |= rx_ring->ims_val;
+ if (rx_ring->itr_val)
+ writel(1000000000 / (rx_ring->itr_val * 256),
+ hw->hw_addr + rx_ring->itr_register);
+ else
+ writel(1, hw->hw_addr + rx_ring->itr_register);
+ ivar = E1000_IVAR_INT_ALLOC_VALID | vector;
+
+ /* Configure Tx vector */
+ tx_ring->ims_val = E1000_IMS_TXQ0;
+ vector++;
+ if (tx_ring->itr_val)
+ writel(1000000000 / (tx_ring->itr_val * 256),
+ hw->hw_addr + tx_ring->itr_register);
+ else
+ writel(1, hw->hw_addr + tx_ring->itr_register);
+ adapter->eiac_mask |= tx_ring->ims_val;
+ ivar |= ((E1000_IVAR_INT_ALLOC_VALID | vector) << 8);
+
+ /* set vector for Other Causes, e.g. link changes */
+ vector++;
+ ivar |= ((E1000_IVAR_INT_ALLOC_VALID | vector) << 16);
+ if (rx_ring->itr_val)
+ writel(1000000000 / (rx_ring->itr_val * 256),
+ hw->hw_addr + E1000_EITR_82574(vector));
+ else
+ writel(1, hw->hw_addr + E1000_EITR_82574(vector));
+
+ /* Cause Tx interrupts on every write back */
+ ivar |= (1 << 31);
+
+ ew32(IVAR, ivar);
+
+ /* enable MSI-X PBA support */
+ ctrl_ext = er32(CTRL_EXT);
+ ctrl_ext |= E1000_CTRL_EXT_PBA_CLR;
+
+ /* Auto-Mask Other interrupts upon ICR read */
+#define E1000_EIAC_MASK_82574 0x01F00000
+ ew32(IAM, ~E1000_EIAC_MASK_82574 | E1000_IMS_OTHER);
+ ctrl_ext |= E1000_CTRL_EXT_EIAME;
+ ew32(CTRL_EXT, ctrl_ext);
+ e1e_flush();
+}
+
+void e1000e_reset_interrupt_capability(struct e1000_adapter *adapter)
+{
+ if (adapter->msix_entries) {
+ pci_disable_msix(adapter->pdev);
+ kfree(adapter->msix_entries);
+ adapter->msix_entries = NULL;
+ } else if (adapter->flags & FLAG_MSI_ENABLED) {
+ pci_disable_msi(adapter->pdev);
+ adapter->flags &= ~FLAG_MSI_ENABLED;
+ }
+
+ return;
+}
+
+/**
+ * e1000e_set_interrupt_capability - set MSI or MSI-X if supported
+ *
+ * Attempt to configure interrupts using the best available
+ * capabilities of the hardware and kernel.
+ **/
+void e1000e_set_interrupt_capability(struct e1000_adapter *adapter)
+{
+ int err;
+ int numvecs, i;
+
+
+ switch (adapter->int_mode) {
+ case E1000E_INT_MODE_MSIX:
+ if (adapter->flags & FLAG_HAS_MSIX) {
+ numvecs = 3; /* RxQ0, TxQ0 and other */
+ adapter->msix_entries = kcalloc(numvecs,
+ sizeof(struct msix_entry),
+ GFP_KERNEL);
+ if (adapter->msix_entries) {
+ for (i = 0; i < numvecs; i++)
+ adapter->msix_entries[i].entry = i;
+
+ err = pci_enable_msix(adapter->pdev,
+ adapter->msix_entries,
+ numvecs);
+ if (err == 0)
+ return;
+ }
+ /* MSI-X failed, so fall through and try MSI */
+ e_err("Failed to initialize MSI-X interrupts. "
+ "Falling back to MSI interrupts.\n");
+ e1000e_reset_interrupt_capability(adapter);
+ }
+ adapter->int_mode = E1000E_INT_MODE_MSI;
+ /* Fall through */
+ case E1000E_INT_MODE_MSI:
+ if (!pci_enable_msi(adapter->pdev)) {
+ adapter->flags |= FLAG_MSI_ENABLED;
+ } else {
+ adapter->int_mode = E1000E_INT_MODE_LEGACY;
+ e_err("Failed to initialize MSI interrupts. Falling "
+ "back to legacy interrupts.\n");
+ }
+ /* Fall through */
+ case E1000E_INT_MODE_LEGACY:
+ /* Don't do anything; this is the system default */
+ break;
+ }
+
+ return;
+}
+
+/**
+ * e1000_request_msix - Initialize MSI-X interrupts
+ *
+ * e1000_request_msix allocates MSI-X vectors and requests interrupts from the
+ * kernel.
+ **/
+static int e1000_request_msix(struct e1000_adapter *adapter)
+{
+ struct net_device *netdev = adapter->netdev;
+ int err = 0, vector = 0;
+
+ if (strlen(netdev->name) < (IFNAMSIZ - 5))
+ sprintf(adapter->rx_ring->name, "%s-rx0", netdev->name);
+ else
+ memcpy(adapter->rx_ring->name, netdev->name, IFNAMSIZ);
+ err = request_irq(adapter->msix_entries[vector].vector,
+ &e1000_intr_msix_rx, 0, adapter->rx_ring->name,
+ netdev);
+ if (err)
+ goto out;
+ adapter->rx_ring->itr_register = E1000_EITR_82574(vector);
+ adapter->rx_ring->itr_val = adapter->itr;
+ vector++;
+
+ if (strlen(netdev->name) < (IFNAMSIZ - 5))
+ sprintf(adapter->tx_ring->name, "%s-tx0", netdev->name);
+ else
+ memcpy(adapter->tx_ring->name, netdev->name, IFNAMSIZ);
+ err = request_irq(adapter->msix_entries[vector].vector,
+ &e1000_intr_msix_tx, 0, adapter->tx_ring->name,
+ netdev);
+ if (err)
+ goto out;
+ adapter->tx_ring->itr_register = E1000_EITR_82574(vector);
+ adapter->tx_ring->itr_val = adapter->itr;
+ vector++;
+
+ err = request_irq(adapter->msix_entries[vector].vector,
+ &e1000_msix_other, 0, netdev->name, netdev);
+ if (err)
+ goto out;
+
+ e1000_configure_msix(adapter);
+ return 0;
+out:
+ return err;
+}
+
/**
* e1000_request_irq - initialize interrupts
*
static int e1000_request_irq(struct e1000_adapter *adapter)
{
struct net_device *netdev = adapter->netdev;
- int irq_flags = IRQF_SHARED;
int err;
- if (!(adapter->flags & FLAG_MSI_TEST_FAILED)) {
- err = pci_enable_msi(adapter->pdev);
- if (!err) {
- adapter->flags |= FLAG_MSI_ENABLED;
- irq_flags = 0;
- }
+ if (adapter->msix_entries) {
+ err = e1000_request_msix(adapter);
+ if (!err)
+ return err;
+ /* fall back to MSI */
+ e1000e_reset_interrupt_capability(adapter);
+ adapter->int_mode = E1000E_INT_MODE_MSI;
+ e1000e_set_interrupt_capability(adapter);
}
+ if (adapter->flags & FLAG_MSI_ENABLED) {
+ err = request_irq(adapter->pdev->irq, &e1000_intr_msi, 0,
+ netdev->name, netdev);
+ if (!err)
+ return err;
- err = request_irq(adapter->pdev->irq,
- ((adapter->flags & FLAG_MSI_ENABLED) ?
- &e1000_intr_msi : &e1000_intr),
- irq_flags, netdev->name, netdev);
- if (err) {
- if (adapter->flags & FLAG_MSI_ENABLED) {
- pci_disable_msi(adapter->pdev);
- adapter->flags &= ~FLAG_MSI_ENABLED;
- }
- e_err("Unable to allocate interrupt, Error: %d\n", err);
+ /* fall back to legacy interrupt */
+ e1000e_reset_interrupt_capability(adapter);
+ adapter->int_mode = E1000E_INT_MODE_LEGACY;
}
+ err = request_irq(adapter->pdev->irq, &e1000_intr, IRQF_SHARED,
+ netdev->name, netdev);
+ if (err)
+ e_err("Unable to allocate interrupt, Error: %d\n", err);
+
return err;
}
{
struct net_device *netdev = adapter->netdev;
- free_irq(adapter->pdev->irq, netdev);
- if (adapter->flags & FLAG_MSI_ENABLED) {
- pci_disable_msi(adapter->pdev);
- adapter->flags &= ~FLAG_MSI_ENABLED;
+ if (adapter->msix_entries) {
+ int vector = 0;
+
+ free_irq(adapter->msix_entries[vector].vector, netdev);
+ vector++;
+
+ free_irq(adapter->msix_entries[vector].vector, netdev);
+ vector++;
+
+ /* Other Causes interrupt vector */
+ free_irq(adapter->msix_entries[vector].vector, netdev);
+ return;
}
+
+ free_irq(adapter->pdev->irq, netdev);
}
/**
struct e1000_hw *hw = &adapter->hw;
ew32(IMC, ~0);
+ if (adapter->msix_entries)
+ ew32(EIAC_82574, 0);
e1e_flush();
synchronize_irq(adapter->pdev->irq);
}
{
struct e1000_hw *hw = &adapter->hw;
- ew32(IMS, IMS_ENABLE_MASK);
+ if (adapter->msix_entries) {
+ ew32(EIAC_82574, adapter->eiac_mask & E1000_EIAC_MASK_82574);
+ ew32(IMS, adapter->eiac_mask | E1000_IMS_OTHER | E1000_IMS_LSC);
+ } else {
+ ew32(IMS, IMS_ENABLE_MASK);
+ }
e1e_flush();
}
* traffic pattern. Constants in this function were computed
* based on theoretical maximum wire speed and thresholds were set based
* on testing data as well as attempting to minimize response time
- * while increasing bulk throughput.
- * this functionality is controlled by the InterruptThrottleRate module
- * parameter (see e1000_param.c)
+ * while increasing bulk throughput. This functionality is controlled
+ * by the InterruptThrottleRate module parameter.
**/
static unsigned int e1000_update_itr(struct e1000_adapter *adapter,
u16 itr_setting, int packets,
min(adapter->itr + (new_itr >> 2), new_itr) :
new_itr;
adapter->itr = new_itr;
- ew32(ITR, 1000000000 / (new_itr * 256));
+ adapter->rx_ring->itr_val = new_itr;
+ if (adapter->msix_entries)
+ adapter->rx_ring->set_itr = 1;
+ else
+ ew32(ITR, 1000000000 / (new_itr * 256));
}
}
+/**
+ * e1000_alloc_queues - Allocate memory for all rings
+ * @adapter: board private structure to initialize
+ **/
+static int __devinit e1000_alloc_queues(struct e1000_adapter *adapter)
+{
+ adapter->tx_ring = kzalloc(sizeof(struct e1000_ring), GFP_KERNEL);
+ if (!adapter->tx_ring)
+ goto err;
+
+ adapter->rx_ring = kzalloc(sizeof(struct e1000_ring), GFP_KERNEL);
+ if (!adapter->rx_ring)
+ goto err;
+
+ return 0;
+err:
+ e_err("Unable to allocate memory for queues\n");
+ kfree(adapter->rx_ring);
+ kfree(adapter->tx_ring);
+ return -ENOMEM;
+}
+
/**
* e1000_clean - NAPI Rx polling callback
* @napi: struct associated with this polling callback
static int e1000_clean(struct napi_struct *napi, int budget)
{
struct e1000_adapter *adapter = container_of(napi, struct e1000_adapter, napi);
+ struct e1000_hw *hw = &adapter->hw;
struct net_device *poll_dev = adapter->netdev;
int tx_cleaned = 0, work_done = 0;
/* Must NOT use netdev_priv macro here. */
adapter = poll_dev->priv;
+ if (adapter->msix_entries &&
+ !(adapter->rx_ring->ims_val & adapter->tx_ring->ims_val))
+ goto clean_rx;
+
/*
* e1000_clean is called per-cpu. This lock protects
* tx_ring from being cleaned by multiple cpus
spin_unlock(&adapter->tx_queue_lock);
}
+clean_rx:
adapter->clean_rx(adapter, &work_done, budget);
if (tx_cleaned)
if (adapter->itr_setting & 3)
e1000_set_itr(adapter);
netif_rx_complete(poll_dev, napi);
- e1000_irq_enable(adapter);
+ if (adapter->msix_entries)
+ ew32(IMS, adapter->rx_ring->ims_val);
+ else
+ e1000_irq_enable(adapter);
}
return work_done;
clear_bit(__E1000_DOWN, &adapter->state);
napi_enable(&adapter->napi);
+ if (adapter->msix_entries)
+ e1000_configure_msix(adapter);
e1000_irq_enable(adapter);
/* fire a link change interrupt to start the watchdog */
adapter->max_frame_size = netdev->mtu + ETH_HLEN + ETH_FCS_LEN;
adapter->min_frame_size = ETH_ZLEN + ETH_FCS_LEN;
- adapter->tx_ring = kzalloc(sizeof(struct e1000_ring), GFP_KERNEL);
- if (!adapter->tx_ring)
- goto err;
+ e1000e_set_interrupt_capability(adapter);
- adapter->rx_ring = kzalloc(sizeof(struct e1000_ring), GFP_KERNEL);
- if (!adapter->rx_ring)
- goto err;
+ if (e1000_alloc_queues(adapter))
+ return -ENOMEM;
spin_lock_init(&adapter->tx_queue_lock);
/* Explicitly disable IRQ since the NIC can be in any state. */
e1000_irq_disable(adapter);
- spin_lock_init(&adapter->stats_lock);
-
set_bit(__E1000_DOWN, &adapter->state);
return 0;
-
-err:
- e_err("Unable to allocate memory for queues\n");
- kfree(adapter->rx_ring);
- kfree(adapter->tx_ring);
- return -ENOMEM;
}
/**
/* free the real vector and request a test handler */
e1000_free_irq(adapter);
+ e1000e_reset_interrupt_capability(adapter);
/* Assume that the test fails, if it succeeds then the test
* MSI irq handler will unset this flag */
rmb();
if (adapter->flags & FLAG_MSI_TEST_FAILED) {
+ adapter->int_mode = E1000E_INT_MODE_LEGACY;
err = -EIO;
e_info("MSI interrupt test failed!\n");
}
/* okay so the test worked, restore settings */
e_dbg("%s: MSI interrupt test succeeded!\n", netdev->name);
msi_test_failed:
- /* restore the original vector, even if it failed */
+ e1000e_set_interrupt_capability(adapter);
e1000_request_irq(adapter);
return err;
}
* ignore e1000e MSI messages, which means we need to test our MSI
* interrupt now
*/
- {
+ if (adapter->int_mode != E1000E_INT_MODE_LEGACY) {
err = e1000_test_msi(adapter);
if (err) {
e_err("Interrupt allocation failed\n");
return 0;
}
+ /**
+ * e1000e_update_phy_task - work thread to update phy
+ * @work: pointer to our work struct
+ *
+ * this worker thread exists because we must acquire a
+ * semaphore to read the phy, which we could msleep while
+ * waiting for it, and we can't msleep in a timer.
+ **/
+ static void e1000e_update_phy_task(struct work_struct *work)
+ {
+ struct e1000_adapter *adapter = container_of(work,
+ struct e1000_adapter, update_phy_task);
+ e1000_get_phy_info(&adapter->hw);
+ }
+
/*
* Need to wait a few seconds after link up to get diagnostic information from
* the phy
static void e1000_update_phy_info(unsigned long data)
{
struct e1000_adapter *adapter = (struct e1000_adapter *) data;
- e1000_get_phy_info(&adapter->hw);
+ schedule_work(&adapter->update_phy_task);
}
/**
{
struct e1000_hw *hw = &adapter->hw;
struct pci_dev *pdev = adapter->pdev;
- unsigned long irq_flags;
- u16 phy_tmp;
-
- #define PHY_IDLE_ERROR_COUNT_MASK 0x00FF
/*
* Prevent stats update while adapter is being reset, or if the pci
if (pci_channel_offline(pdev))
return;
- spin_lock_irqsave(&adapter->stats_lock, irq_flags);
-
- /*
- * these counters are modified from e1000_adjust_tbi_stats,
- * called from the interrupt context, so they must only
- * be written while holding adapter->stats_lock
- */
-
adapter->stats.crcerrs += er32(CRCERRS);
adapter->stats.gprc += er32(GPRC);
adapter->stats.gorc += er32(GORCL);
adapter->stats.algnerrc += er32(ALGNERRC);
adapter->stats.rxerrc += er32(RXERRC);
- adapter->stats.tncrs += er32(TNCRS);
+ if (hw->mac.type != e1000_82574)
+ adapter->stats.tncrs += er32(TNCRS);
adapter->stats.cexterr += er32(CEXTERR);
adapter->stats.tsctc += er32(TSCTC);
adapter->stats.tsctfc += er32(TSCTFC);
/* Tx Dropped needs to be maintained elsewhere */
- /* Phy Stats */
- if (hw->phy.media_type == e1000_media_type_copper) {
- if ((adapter->link_speed == SPEED_1000) &&
- (!e1e_rphy(hw, PHY_1000T_STATUS, &phy_tmp))) {
- phy_tmp &= PHY_IDLE_ERROR_COUNT_MASK;
- adapter->phy_stats.idle_errors += phy_tmp;
- }
- }
-
/* Management Stats */
adapter->stats.mgptc += er32(MGTPTC);
adapter->stats.mgprc += er32(MGTPRC);
adapter->stats.mgpdc += er32(MGTPDC);
-
- spin_unlock_irqrestore(&adapter->stats_lock, irq_flags);
}
/**
struct e1000_hw *hw = &adapter->hw;
struct e1000_phy_regs *phy = &adapter->phy_regs;
int ret_val;
- unsigned long irq_flags;
-
-
- spin_lock_irqsave(&adapter->stats_lock, irq_flags);
if ((er32(STATUS) & E1000_STATUS_LU) &&
(adapter->hw.phy.media_type == e1000_media_type_copper)) {
phy->stat1000 = 0;
phy->estatus = (ESTATUS_1000_TFULL | ESTATUS_1000_THALF);
}
-
- spin_unlock_irqrestore(&adapter->stats_lock, irq_flags);
}
static void e1000_print_link_info(struct e1000_adapter *adapter)
&adapter->link_speed,
&adapter->link_duplex);
e1000_print_link_info(adapter);
+ /*
+ * On supported PHYs, check for duplex mismatch only
+ * if link has autonegotiated at 10/100 half
+ */
+ if ((hw->phy.type == e1000_phy_igp_3 ||
+ hw->phy.type == e1000_phy_bm) &&
+ (hw->mac.autoneg == true) &&
+ (adapter->link_speed == SPEED_10 ||
+ adapter->link_speed == SPEED_100) &&
+ (adapter->link_duplex == HALF_DUPLEX)) {
+ u16 autoneg_exp;
+
+ e1e_rphy(hw, PHY_AUTONEG_EXP, &autoneg_exp);
+
+ if (!(autoneg_exp & NWAY_ER_LP_NWAY_CAPS))
+ e_info("Autonegotiated half duplex but"
+ " link partner cannot autoneg. "
+ " Try forcing full duplex if "
+ "link gets many collisions.\n");
+ }
+
/*
* tweak tx_queue_len according to speed/duplex
* and adjust the timeout factor
}
/* Cause software interrupt to ensure Rx ring is cleaned */
- ew32(ICS, E1000_ICS_RXDMT0);
+ if (adapter->msix_entries)
+ ew32(ICS, adapter->rx_ring->ims_val);
+ else
+ ew32(ICS, E1000_ICS_RXDMT0);
/* Force detection of hung controller every watchdog period */
adapter->detect_tx_hung = 1;
e1000e_down(adapter);
e1000_free_irq(adapter);
}
+ e1000e_reset_interrupt_capability(adapter);
retval = pci_save_state(pdev);
if (retval)
pci_enable_wake(pdev, PCI_D3hot, 0);
pci_enable_wake(pdev, PCI_D3cold, 0);
+ e1000e_set_interrupt_capability(adapter);
if (netif_running(netdev)) {
err = e1000_request_irq(adapter);
if (err)
ret_val = e1000_read_nvm(hw, NVM_INIT_CONTROL2_REG, 1, &buf);
if (!(le16_to_cpu(buf) & (1 << 0))) {
/* Deep Smart Power Down (DSPD) */
- e_warn("Warning: detected DSPD enabled in EEPROM\n");
+ dev_warn(&adapter->pdev->dev,
+ "Warning: detected DSPD enabled in EEPROM\n");
}
ret_val = e1000_read_nvm(hw, NVM_INIT_3GIO_3, 1, &buf);
if (le16_to_cpu(buf) & (3 << 2)) {
/* ASPM enable */
- e_warn("Warning: detected ASPM enabled in EEPROM\n");
+ dev_warn(&adapter->pdev->dev,
+ "Warning: detected ASPM enabled in EEPROM\n");
}
}
if (err)
goto err_hw_init;
+ if ((adapter->flags & FLAG_IS_ICH) &&
+ (adapter->flags & FLAG_READ_ONLY_NVM))
+ e1000e_write_protect_nvm_ich8lan(&adapter->hw);
+
hw->mac.ops.get_bus_info(&adapter->hw);
adapter->hw.phy.autoneg_wait_to_complete = 0;
INIT_WORK(&adapter->reset_task, e1000_reset_task);
INIT_WORK(&adapter->watchdog_task, e1000_watchdog_task);
+ INIT_WORK(&adapter->downshift_task, e1000e_downshift_workaround);
+ INIT_WORK(&adapter->update_phy_task, e1000e_update_phy_task);
/* Initialize link parameters. User can change them with ethtool */
adapter->hw.mac.autoneg = 1;
if (!e1000_check_reset_block(&adapter->hw))
e1000_phy_hw_reset(&adapter->hw);
+ e1000e_reset_interrupt_capability(adapter);
kfree(adapter->tx_ring);
kfree(adapter->rx_ring);
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82573E_IAMT), board_82573 },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82573L), board_82573 },
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_82574L), board_82574 },
+
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_80003ES2LAN_COPPER_DPT),
board_80003es2lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_80003ES2LAN_COPPER_SPT),
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IFE_GT), board_ich9lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_AMT), board_ich9lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_C), board_ich9lan },
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_BM), board_ich9lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M), board_ich9lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M_AMT), board_ich9lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M_V), board_ich9lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_R_BM_LF), board_ich9lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_R_BM_V), board_ich9lan },
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_D_BM_LM), board_ich10lan },
+ { PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_D_BM_LF), board_ich10lan },
+
{ } /* terminate list */
};
MODULE_DEVICE_TABLE(pci, e1000_pci_tbl);
#define DEFAULT_ITR 3
#define MAX_ITR 100000
#define MIN_ITR 100
+/* IntMode (Interrupt Mode)
+ *
+ * Valid Range: 0 - 2
+ *
+ * Default Value: 2 (MSI-X)
+ */
+E1000_PARAM(IntMode, "Interrupt Mode");
+#define MAX_INTMODE 2
+#define MIN_INTMODE 0
/*
* Enable Smart Power Down of the PHY
*/
E1000_PARAM(KumeranLockLoss, "Enable Kumeran lock loss workaround");
+ /*
+ * Write Protect NVM
+ *
+ * Valid Range: 0, 1
+ *
+ * Default Value: 1 (enabled)
+ */
+ E1000_PARAM(WriteProtectNVM, "Write-protect NVM [WARNING: disabling this can lead to corrupted NVM]");
+
struct e1000_option {
enum { enable_option, range_option, list_option } type;
const char *name;
adapter->itr = 20000;
}
}
+ { /* Interrupt Mode */
+ struct e1000_option opt = {
+ .type = range_option,
+ .name = "Interrupt Mode",
+ .err = "defaulting to 2 (MSI-X)",
+ .def = E1000E_INT_MODE_MSIX,
+ .arg = { .r = { .min = MIN_INTMODE,
+ .max = MAX_INTMODE } }
+ };
+
+ if (num_IntMode > bd) {
+ unsigned int int_mode = IntMode[bd];
+ e1000_validate_option(&int_mode, &opt, adapter);
+ adapter->int_mode = int_mode;
+ } else {
+ adapter->int_mode = opt.def;
+ }
+ }
{ /* Smart Power Down */
const struct e1000_option opt = {
.type = enable_option,
opt.def);
}
}
+ { /* Write-protect NVM */
+ const struct e1000_option opt = {
+ .type = enable_option,
+ .name = "Write-protect NVM",
+ .err = "defaulting to Enabled",
+ .def = OPTION_ENABLED
+ };
+
+ if (adapter->flags & FLAG_IS_ICH) {
+ if (num_WriteProtectNVM > bd) {
+ unsigned int write_protect_nvm = WriteProtectNVM[bd];
+ e1000_validate_option(&write_protect_nvm, &opt,
+ adapter);
+ if (write_protect_nvm)
+ adapter->flags |= FLAG_READ_ONLY_NVM;
+ } else {
+ if (opt.def)
+ adapter->flags |= FLAG_READ_ONLY_NVM;
+ }
+ }
+ }
}
* Change name of a device, can pass format strings "eth%d".
* for wildcarding.
*/
-int dev_change_name(struct net_device *dev, char *newname)
+int dev_change_name(struct net_device *dev, const char *newname)
{
char oldname[IFNAMSIZ];
int err = 0;
err = dev_alloc_name(dev, newname);
if (err < 0)
return err;
- strcpy(newname, dev->name);
}
else if (__dev_get_by_name(net, newname))
return -EEXIST;
return err;
}
+/**
+ * dev_set_alias - change ifalias of a device
+ * @dev: device
+ * @alias: name up to IFALIASZ
+ * @len: limit of bytes to copy from info
+ *
+ * Set ifalias for a device,
+ */
+int dev_set_alias(struct net_device *dev, const char *alias, size_t len)
+{
+ ASSERT_RTNL();
+
+ if (len >= IFALIASZ)
+ return -EINVAL;
+
+ if (!len) {
+ if (dev->ifalias) {
+ kfree(dev->ifalias);
+ dev->ifalias = NULL;
+ }
+ return 0;
+ }
+
+ dev->ifalias = krealloc(dev->ifalias, len+1, GFP_KERNEL);
+ if (!dev->ifalias)
+ return -ENOMEM;
+
+ strlcpy(dev->ifalias, alias, len+1);
+ return len;
+}
+
+
/**
* netdev_features_change - device changes features
* @dev: device to cause notification
}
switch (skb->protocol) {
- case __constant_htons(ETH_P_IP):
+ case htons(ETH_P_IP):
if (!(ip_hdr(skb)->frag_off & htons(IP_MF | IP_OFFSET)))
ip_proto = ip_hdr(skb)->protocol;
addr1 = ip_hdr(skb)->saddr;
addr2 = ip_hdr(skb)->daddr;
ihl = ip_hdr(skb)->ihl;
break;
- case __constant_htons(ETH_P_IPV6):
+ case htons(ETH_P_IPV6):
ip_proto = ipv6_hdr(skb)->nexthdr;
addr1 = ipv6_hdr(skb)->saddr.s6_addr32[3];
addr2 = ipv6_hdr(skb)->daddr.s6_addr32[3];
return 0;
}
+ static void dev_change_rx_flags(struct net_device *dev, int flags)
+ {
+ if (dev->flags & IFF_UP && dev->change_rx_flags)
+ dev->change_rx_flags(dev, flags);
+ }
+
static int __dev_set_promiscuity(struct net_device *dev, int inc)
{
unsigned short old_flags = dev->flags;
current->uid, current->gid,
audit_get_sessionid(current));
- if (dev->change_rx_flags)
- dev->change_rx_flags(dev, IFF_PROMISC);
+ dev_change_rx_flags(dev, IFF_PROMISC);
}
return 0;
}
}
}
if (dev->flags ^ old_flags) {
- if (dev->change_rx_flags)
- dev->change_rx_flags(dev, IFF_ALLMULTI);
+ dev_change_rx_flags(dev, IFF_ALLMULTI);
dev_set_rx_mode(dev);
}
return 0;
netif_addr_unlock_bh(dev);
}
+/**
+ * dev_get_flags - get flags reported to userspace
+ * @dev: device
+ *
+ * Get the combination of flag bits exported through APIs to userspace.
+ */
unsigned dev_get_flags(const struct net_device *dev)
{
unsigned flags;
return flags;
}
+/**
+ * dev_change_flags - change device settings
+ * @dev: device
+ * @flags: device state flags
+ *
+ * Change settings on device based state flags. The flags are
+ * in the userspace exported format.
+ */
int dev_change_flags(struct net_device *dev, unsigned flags)
{
int ret, changes;
* Load in the correct multicast list now the flags have changed.
*/
- if (dev->change_rx_flags && (old_flags ^ flags) & IFF_MULTICAST)
- dev->change_rx_flags(dev, IFF_MULTICAST);
+ if ((old_flags ^ flags) & IFF_MULTICAST)
+ dev_change_rx_flags(dev, IFF_MULTICAST);
dev_set_rx_mode(dev);
return ret;
}
+/**
+ * dev_set_mtu - Change maximum transfer unit
+ * @dev: device
+ * @new_mtu: new transfer unit
+ *
+ * Change the maximum transfer size of the network device.
+ */
int dev_set_mtu(struct net_device *dev, int new_mtu)
{
int err;
return err;
}
+/**
+ * dev_set_mac_address - Change Media Access Control Address
+ * @dev: device
+ * @sa: new address
+ *
+ * Change the hardware (MAC) address of the device
+ */
int dev_set_mac_address(struct net_device *dev, struct sockaddr *sa)
{
int err;
}
/* Delayed registration/unregisteration */
- static DEFINE_SPINLOCK(net_todo_list_lock);
static LIST_HEAD(net_todo_list);
static void net_set_todo(struct net_device *dev)
{
- spin_lock(&net_todo_list_lock);
list_add_tail(&dev->todo_list, &net_todo_list);
- spin_unlock(&net_todo_list_lock);
}
static void rollback_registered(struct net_device *dev)
* free_netdev(y1);
* free_netdev(y2);
*
- * We are invoked by rtnl_unlock() after it drops the semaphore.
+ * We are invoked by rtnl_unlock().
* This allows us to deal with problems:
* 1) We can delete sysfs objects which invoke hotplug
* without deadlocking with linkwatch via keventd.
* 2) Since we run with the RTNL semaphore not held, we can sleep
* safely in order to wait for the netdev refcnt to drop to zero.
+ *
+ * We must not return until all unregister events added during
+ * the interval the lock was held have been completed.
*/
- static DEFINE_MUTEX(net_todo_run_mutex);
void netdev_run_todo(void)
{
struct list_head list;
- /* Need to guard against multiple cpu's getting out of order. */
- mutex_lock(&net_todo_run_mutex);
-
- /* Not safe to do outside the semaphore. We must not return
- * until all unregister events invoked by the local processor
- * have been completed (either by this todo run, or one on
- * another cpu).
- */
- if (list_empty(&net_todo_list))
- goto out;
-
/* Snapshot list, allow later requests */
- spin_lock(&net_todo_list_lock);
list_replace_init(&net_todo_list, &list);
- spin_unlock(&net_todo_list_lock);
+
+ __rtnl_unlock();
while (!list_empty(&list)) {
struct net_device *dev
/* Free network device */
kobject_put(&dev->dev.kobj);
}
-
- out:
- mutex_unlock(&net_todo_run_mutex);
}
static struct net_device_stats *internal_stats(struct net_device *dev)
put_device(&dev->dev);
}
-/* Synchronize with packet receive processing. */
+/**
+ * synchronize_net - Synchronize with packet receive processing
+ *
+ * Wait for packets currently being received to be done.
+ * Does not block later packets from starting.
+ */
void synchronize_net(void)
{
might_sleep();
}
/**
- * netdev_dma_regiser - register the networking subsystem as a DMA client
+ * netdev_dma_register - register the networking subsystem as a DMA client
*/
static int __init netdev_dma_register(void)
{
one |= NETIF_F_GSO_SOFTWARE;
one |= NETIF_F_GSO;
+ /*
+ * If even one device supports a GSO protocol with software fallback,
+ * enable it for all.
+ */
+ all |= one & NETIF_F_GSO_SOFTWARE;
+
/* If even one device supports robust GSO, enable it for all. */
if (one & NETIF_F_GSO_ROBUST)
all |= NETIF_F_GSO_ROBUST;
return -ENOMEM;
}
-char *netdev_drivername(struct net_device *dev, char *buffer, int len)
+/**
+ * netdev_drivername - network driver for the device
+ * @dev: network device
+ * @buffer: buffer for resulting name
+ * @len: size of buffer
+ *
+ * Determine network driver for device.
+ */
+char *netdev_drivername(const struct net_device *dev, char *buffer, int len)
{
- struct device_driver *driver;
- struct device *parent;
+ const struct device_driver *driver;
+ const struct device *parent;
if (len <= 0 || !buffer)
return buffer;
void rtnl_unlock(void)
{
- mutex_unlock(&rtnl_mutex);
+ /* This fellow will unlock it for us. */
netdev_run_todo();
}
{
return NLMSG_ALIGN(sizeof(struct ifinfomsg))
+ nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */
+ + nla_total_size(IFALIASZ) /* IFLA_IFALIAS */
+ nla_total_size(IFNAMSIZ) /* IFLA_QDISC */
+ nla_total_size(sizeof(struct rtnl_link_ifmap))
+ nla_total_size(sizeof(struct rtnl_link_stats))
if (txq->qdisc_sleeping)
NLA_PUT_STRING(skb, IFLA_QDISC, txq->qdisc_sleeping->ops->id);
+ if (dev->ifalias)
+ NLA_PUT_STRING(skb, IFLA_IFALIAS, dev->ifalias);
+
if (1) {
struct rtnl_link_ifmap map = {
.mem_start = dev->mem_start,
[IFLA_LINKMODE] = { .type = NLA_U8 },
[IFLA_LINKINFO] = { .type = NLA_NESTED },
[IFLA_NET_NS_PID] = { .type = NLA_U32 },
+ [IFLA_IFALIAS] = { .type = NLA_STRING, .len = IFALIASZ-1 },
};
static const struct nla_policy ifla_info_policy[IFLA_INFO_MAX+1] = {
modified = 1;
}
+ if (tb[IFLA_IFALIAS]) {
+ err = dev_set_alias(dev, nla_data(tb[IFLA_IFALIAS]),
+ nla_len(tb[IFLA_IFALIAS]));
+ if (err < 0)
+ goto errout;
+ modified = 1;
+ }
+
if (tb[IFLA_BROADCAST]) {
nla_memcpy(dev->broadcast, tb[IFLA_BROADCAST], dev->addr_len);
send_addr_notify = 1;
}
}
+/* This must be called before lost_out is incremented */
+static void tcp_verify_retransmit_hint(struct tcp_sock *tp, struct sk_buff *skb)
+{
+ if ((tp->retransmit_skb_hint == NULL) ||
+ before(TCP_SKB_CB(skb)->seq,
+ TCP_SKB_CB(tp->retransmit_skb_hint)->seq))
+ tp->retransmit_skb_hint = skb;
+
+ if (!tp->lost_out ||
+ after(TCP_SKB_CB(skb)->end_seq, tp->retransmit_high))
+ tp->retransmit_high = TCP_SKB_CB(skb)->end_seq;
+}
+
+static void tcp_skb_mark_lost(struct tcp_sock *tp, struct sk_buff *skb)
+{
+ if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_LOST|TCPCB_SACKED_ACKED))) {
+ tcp_verify_retransmit_hint(tp, skb);
+
+ tp->lost_out += tcp_skb_pcount(skb);
+ TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
+ }
+}
+
+void tcp_skb_mark_lost_uncond_verify(struct tcp_sock *tp, struct sk_buff *skb)
+{
+ tcp_verify_retransmit_hint(tp, skb);
+
+ if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_LOST|TCPCB_SACKED_ACKED))) {
+ tp->lost_out += tcp_skb_pcount(skb);
+ TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
+ }
+}
+
/* This procedure tags the retransmission queue when SACKs arrive.
*
* We have three tag bits: SACKED(S), RETRANS(R) and LOST(L).
TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;
tp->retrans_out -= tcp_skb_pcount(skb);
- /* clear lost hint */
- tp->retransmit_skb_hint = NULL;
-
- if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_LOST|TCPCB_SACKED_ACKED))) {
- tp->lost_out += tcp_skb_pcount(skb);
- TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
- }
+ tcp_skb_mark_lost_uncond_verify(tp, skb);
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPLOSTRETRANSMIT);
} else {
if (before(ack_seq, new_low_seq))
~(TCPCB_LOST|TCPCB_SACKED_RETRANS);
tp->lost_out -= tcp_skb_pcount(skb);
tp->retrans_out -= tcp_skb_pcount(skb);
-
- /* clear lost hint */
- tp->retransmit_skb_hint = NULL;
}
} else {
if (!(sacked & TCPCB_RETRANS)) {
if (sacked & TCPCB_LOST) {
TCP_SKB_CB(skb)->sacked &= ~TCPCB_LOST;
tp->lost_out -= tcp_skb_pcount(skb);
-
- /* clear lost hint */
- tp->retransmit_skb_hint = NULL;
}
}
if (dup_sack && (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_RETRANS)) {
TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_RETRANS;
tp->retrans_out -= tcp_skb_pcount(skb);
- tp->retransmit_skb_hint = NULL;
}
return flag;
return 0;
skb = tcp_write_queue_head(sk);
+ if (tcp_skb_is_last(sk, skb))
+ return 1;
skb = tcp_write_queue_next(sk, skb); /* Skips head */
tcp_for_write_queue_from(skb, sk) {
if (skb == tcp_send_head(sk))
if (!(TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED)) {
TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
tp->lost_out += tcp_skb_pcount(skb);
+ tp->retransmit_high = TCP_SKB_CB(skb)->end_seq;
}
}
tcp_verify_left_out(tp);
tp->high_seq = tp->snd_nxt;
TCP_ECN_queue_cwr(tp);
- tcp_clear_retrans_hints_partial(tp);
+ tcp_clear_all_retrans_hints(tp);
}
static void tcp_clear_retrans_partial(struct tcp_sock *tp)
/* Push undo marker, if it was plain RTO and nothing
* was retransmitted. */
tp->undo_marker = tp->snd_una;
- tcp_clear_retrans_hints_partial(tp);
} else {
tp->sacked_out = 0;
tp->fackets_out = 0;
- tcp_clear_all_retrans_hints(tp);
}
+ tcp_clear_all_retrans_hints(tp);
tcp_for_write_queue(skb, sk) {
if (skb == tcp_send_head(sk))
TCP_SKB_CB(skb)->sacked &= ~TCPCB_SACKED_ACKED;
TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
tp->lost_out += tcp_skb_pcount(skb);
+ tp->retransmit_high = TCP_SKB_CB(skb)->end_seq;
}
}
tcp_verify_left_out(tp);
return 0;
}
-/* RFC: This is from the original, I doubt that this is necessary at all:
- * clear xmit_retrans hint if seq of this skb is beyond hint. How could we
- * retransmitted past LOST markings in the first place? I'm not fully sure
- * about undo and end of connection cases, which can cause R without L?
- */
-static void tcp_verify_retransmit_hint(struct tcp_sock *tp, struct sk_buff *skb)
-{
- if ((tp->retransmit_skb_hint != NULL) &&
- before(TCP_SKB_CB(skb)->seq,
- TCP_SKB_CB(tp->retransmit_skb_hint)->seq))
- tp->retransmit_skb_hint = NULL;
-}
-
/* Mark head of queue up as lost. With RFC3517 SACK, the packets is
* is against sacked "cnt", otherwise it's against facked "cnt"
*/
cnt = packets;
}
- if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_SACKED_ACKED|TCPCB_LOST))) {
- TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
- tp->lost_out += tcp_skb_pcount(skb);
- tcp_verify_retransmit_hint(tp, skb);
- }
+ tcp_skb_mark_lost(tp, skb);
}
tcp_verify_left_out(tp);
}
if (!tcp_skb_timedout(sk, skb))
break;
- if (!(TCP_SKB_CB(skb)->sacked & (TCPCB_SACKED_ACKED|TCPCB_LOST))) {
- TCP_SKB_CB(skb)->sacked |= TCPCB_LOST;
- tp->lost_out += tcp_skb_pcount(skb);
- tcp_verify_retransmit_hint(tp, skb);
- }
+ tcp_skb_mark_lost(tp, skb);
}
tp->scoreboard_skb_hint = skb;
}
tcp_moderate_cwnd(tp);
tp->snd_cwnd_stamp = tcp_time_stamp;
-
- /* There is something screwy going on with the retrans hints after
- an undo */
- tcp_clear_all_retrans_hints(tp);
}
static inline int tcp_may_undo(struct tcp_sock *tp)
* is before the ack sequence we can discard it as it's confirmed to have
* arrived at the other end.
*/
-static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets)
+static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets,
+ u32 prior_snd_una)
{
struct tcp_sock *tp = tcp_sk(sk);
const struct inet_connection_sock *icsk = inet_csk(sk);
int flag = 0;
u32 pkts_acked = 0;
u32 reord = tp->packets_out;
+ u32 prior_sacked = tp->sacked_out;
s32 seq_rtt = -1;
s32 ca_seq_rtt = -1;
ktime_t last_ackt = net_invalid_timestamp();
if (sacked & TCPCB_LOST)
tp->lost_out -= acked_pcount;
- if (unlikely(tp->urg_mode && !before(end_seq, tp->snd_up)))
- tp->urg_mode = 0;
-
tp->packets_out -= acked_pcount;
pkts_acked += acked_pcount;
tcp_unlink_write_queue(skb, sk);
sk_wmem_free_skb(sk, skb);
- tcp_clear_all_retrans_hints(tp);
+ tp->scoreboard_skb_hint = NULL;
+ if (skb == tp->retransmit_skb_hint)
+ tp->retransmit_skb_hint = NULL;
+ if (skb == tp->lost_skb_hint)
+ tp->lost_skb_hint = NULL;
}
+ if (likely(between(tp->snd_up, prior_snd_una, tp->snd_una)))
+ tp->snd_up = tp->snd_una;
+
if (skb && (TCP_SKB_CB(skb)->sacked & TCPCB_SACKED_ACKED))
flag |= FLAG_SACK_RENEGING;
/* Non-retransmitted hole got filled? That's reordering */
if (reord < prior_fackets)
tcp_update_reordering(sk, tp->fackets_out - reord, 0);
+
+ /* No need to care for underflows here because
+ * the lost_skb_hint gets NULLed if we're past it
+ * (or something non-trivial happened)
+ */
+ if (tcp_is_fack(tp))
+ tp->lost_cnt_hint -= pkts_acked;
+ else
+ tp->lost_cnt_hint -= prior_sacked - tp->sacked_out;
}
tp->fackets_out -= min(pkts_acked, tp->fackets_out);
goto no_queue;
/* See if we can take anything off of the retransmit queue. */
- flag |= tcp_clean_rtx_queue(sk, prior_fackets);
+ flag |= tcp_clean_rtx_queue(sk, prior_fackets, prior_snd_una);
if (tp->frto_counter)
frto_cwnd = tcp_process_frto(sk, flag);
}
}
+static int tcp_parse_aligned_timestamp(struct tcp_sock *tp, struct tcphdr *th)
+{
+ __be32 *ptr = (__be32 *)(th + 1);
+
+ if (*ptr == htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16)
+ | (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP)) {
+ tp->rx_opt.saw_tstamp = 1;
+ ++ptr;
+ tp->rx_opt.rcv_tsval = ntohl(*ptr);
+ ++ptr;
+ tp->rx_opt.rcv_tsecr = ntohl(*ptr);
+ return 1;
+ }
+ return 0;
+}
+
/* Fast parse options. This hopes to only see timestamps.
* If it is wrong it falls back on tcp_parse_options().
*/
return 0;
} else if (tp->rx_opt.tstamp_ok &&
th->doff == (sizeof(struct tcphdr)>>2)+(TCPOLEN_TSTAMP_ALIGNED>>2)) {
- __be32 *ptr = (__be32 *)(th + 1);
- if (*ptr == htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16)
- | (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP)) {
- tp->rx_opt.saw_tstamp = 1;
- ++ptr;
- tp->rx_opt.rcv_tsval = ntohl(*ptr);
- ++ptr;
- tp->rx_opt.rcv_tsecr = ntohl(*ptr);
+ if (tcp_parse_aligned_timestamp(tp, th))
return 1;
- }
}
tcp_parse_options(skb, &tp->rx_opt, 1);
return 1;
skb1 = skb1->prev;
}
}
- __skb_insert(skb, skb1, skb1->next, &tp->out_of_order_queue);
+ __skb_queue_after(&tp->out_of_order_queue, skb1, skb);
/* And clean segments covered by new one as whole. */
while ((skb1 = skb->next) !=
}
}
+static struct sk_buff *tcp_collapse_one(struct sock *sk, struct sk_buff *skb,
+ struct sk_buff_head *list)
+{
+ struct sk_buff *next = skb->next;
+
+ __skb_unlink(skb, list);
+ __kfree_skb(skb);
+ NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPRCVCOLLAPSED);
+
+ return next;
+}
+
/* Collapse contiguous sequence of skbs head..tail with
* sequence numbers start..end.
* Segments with FIN/SYN are not collapsed (only because this
for (skb = head; skb != tail;) {
/* No new bits? It is possible on ofo queue. */
if (!before(start, TCP_SKB_CB(skb)->end_seq)) {
- struct sk_buff *next = skb->next;
- __skb_unlink(skb, list);
- __kfree_skb(skb);
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPRCVCOLLAPSED);
- skb = next;
+ skb = tcp_collapse_one(sk, skb, list);
continue;
}
memcpy(nskb->head, skb->head, header);
memcpy(nskb->cb, skb->cb, sizeof(skb->cb));
TCP_SKB_CB(nskb)->seq = TCP_SKB_CB(nskb)->end_seq = start;
- __skb_insert(nskb, skb->prev, skb, list);
+ __skb_queue_before(list, skb, nskb);
skb_set_owner_r(nskb, sk);
/* Copy data, releasing collapsed skbs. */
start += size;
}
if (!before(start, TCP_SKB_CB(skb)->end_seq)) {
- struct sk_buff *next = skb->next;
- __skb_unlink(skb, list);
- __kfree_skb(skb);
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPRCVCOLLAPSED);
- skb = next;
+ skb = tcp_collapse_one(sk, skb, list);
if (skb == tail ||
tcp_hdr(skb)->syn ||
tcp_hdr(skb)->fin)
if (tcp_should_expand_sndbuf(sk)) {
int sndmem = max_t(u32, tp->rx_opt.mss_clamp, tp->mss_cache) +
- MAX_TCP_HEADER + 16 + sizeof(struct sk_buff),
- demanded = max_t(unsigned int, tp->snd_cwnd,
+ MAX_TCP_HEADER + 16 + sizeof(struct sk_buff);
+ int demanded = max_t(unsigned int, tp->snd_cwnd,
tp->reordering + 1);
sndmem *= 2 * demanded;
if (sndmem > sk->sk_sndbuf)
}
#endif /* CONFIG_NET_DMA */
+/* Does PAWS and seqno based validation of an incoming segment, flags will
+ * play significant role here.
+ */
+static int tcp_validate_incoming(struct sock *sk, struct sk_buff *skb,
+ struct tcphdr *th, int syn_inerr)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+
+ /* RFC1323: H1. Apply PAWS check first. */
+ if (tcp_fast_parse_options(skb, th, tp) && tp->rx_opt.saw_tstamp &&
+ tcp_paws_discard(sk, skb)) {
+ if (!th->rst) {
+ NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED);
+ tcp_send_dupack(sk, skb);
+ goto discard;
+ }
+ /* Reset is accepted even if it did not pass PAWS. */
+ }
+
+ /* Step 1: check sequence number */
+ if (!tcp_sequence(tp, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq)) {
+ /* RFC793, page 37: "In all states except SYN-SENT, all reset
+ * (RST) segments are validated by checking their SEQ-fields."
+ * And page 69: "If an incoming segment is not acceptable,
+ * an acknowledgment should be sent in reply (unless the RST
+ * bit is set, if so drop the segment and return)".
+ */
+ if (!th->rst)
+ tcp_send_dupack(sk, skb);
+ goto discard;
+ }
+
+ /* Step 2: check RST bit */
+ if (th->rst) {
+ tcp_reset(sk);
+ goto discard;
+ }
+
+ /* ts_recent update must be made after we are sure that the packet
+ * is in window.
+ */
+ tcp_replace_ts_recent(tp, TCP_SKB_CB(skb)->seq);
+
+ /* step 3: check security and precedence [ignored] */
+
+ /* step 4: Check for a SYN in window. */
+ if (th->syn && !before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
+ if (syn_inerr)
+ TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS);
+ NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONSYN);
+ tcp_reset(sk);
+ return -1;
+ }
+
+ return 1;
+
+discard:
+ __kfree_skb(skb);
+ return 0;
+}
+
/*
* TCP receive function for the ESTABLISHED state.
*
struct tcphdr *th, unsigned len)
{
struct tcp_sock *tp = tcp_sk(sk);
+ int res;
/*
* Header prediction.
/* Check timestamp */
if (tcp_header_len == sizeof(struct tcphdr) + TCPOLEN_TSTAMP_ALIGNED) {
- __be32 *ptr = (__be32 *)(th + 1);
-
/* No? Slow path! */
- if (*ptr != htonl((TCPOPT_NOP << 24) | (TCPOPT_NOP << 16)
- | (TCPOPT_TIMESTAMP << 8) | TCPOLEN_TIMESTAMP))
+ if (!tcp_parse_aligned_timestamp(tp, th))
goto slow_path;
- tp->rx_opt.saw_tstamp = 1;
- ++ptr;
- tp->rx_opt.rcv_tsval = ntohl(*ptr);
- ++ptr;
- tp->rx_opt.rcv_tsecr = ntohl(*ptr);
-
/* If PAWS failed, check it more carefully in slow path */
if ((s32)(tp->rx_opt.rcv_tsval - tp->rx_opt.ts_recent) < 0)
goto slow_path;
goto no_ack;
}
- __tcp_ack_snd_check(sk, 0);
+ if (!copied_early || tp->rcv_nxt != tp->rcv_wup)
+ __tcp_ack_snd_check(sk, 0);
no_ack:
#ifdef CONFIG_NET_DMA
if (copied_early)
if (len < (th->doff << 2) || tcp_checksum_complete_user(sk, skb))
goto csum_error;
- /*
- * RFC1323: H1. Apply PAWS check first.
- */
- if (tcp_fast_parse_options(skb, th, tp) && tp->rx_opt.saw_tstamp &&
- tcp_paws_discard(sk, skb)) {
- if (!th->rst) {
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED);
- tcp_send_dupack(sk, skb);
- goto discard;
- }
- /* Resets are accepted even if PAWS failed.
-
- ts_recent update must be made after we are sure
- that the packet is in window.
- */
- }
-
/*
* Standard slow path.
*/
- if (!tcp_sequence(tp, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq)) {
- /* RFC793, page 37: "In all states except SYN-SENT, all reset
- * (RST) segments are validated by checking their SEQ-fields."
- * And page 69: "If an incoming segment is not acceptable,
- * an acknowledgment should be sent in reply (unless the RST bit
- * is set, if so drop the segment and return)".
- */
- if (!th->rst)
- tcp_send_dupack(sk, skb);
- goto discard;
- }
-
- if (th->rst) {
- tcp_reset(sk);
- goto discard;
- }
-
- tcp_replace_ts_recent(tp, TCP_SKB_CB(skb)->seq);
-
- if (th->syn && !before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
- TCP_INC_STATS_BH(sock_net(sk), TCP_MIB_INERRS);
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONSYN);
- tcp_reset(sk);
- return 1;
- }
+ res = tcp_validate_incoming(sk, skb, th, 1);
+ if (res <= 0)
+ return -res;
step5:
if (th->ack)
struct tcp_sock *tp = tcp_sk(sk);
struct inet_connection_sock *icsk = inet_csk(sk);
int queued = 0;
+ int res;
tp->rx_opt.saw_tstamp = 0;
return 0;
}
- if (tcp_fast_parse_options(skb, th, tp) && tp->rx_opt.saw_tstamp &&
- tcp_paws_discard(sk, skb)) {
- if (!th->rst) {
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED);
- tcp_send_dupack(sk, skb);
- goto discard;
- }
- /* Reset is accepted even if it did not pass PAWS. */
- }
-
- /* step 1: check sequence number */
- if (!tcp_sequence(tp, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq)) {
- if (!th->rst)
- tcp_send_dupack(sk, skb);
- goto discard;
- }
-
- /* step 2: check RST bit */
- if (th->rst) {
- tcp_reset(sk);
- goto discard;
- }
-
- tcp_replace_ts_recent(tp, TCP_SKB_CB(skb)->seq);
-
- /* step 3: check security and precedence [ignored] */
-
- /* step 4:
- *
- * Check for a SYN in window.
- */
- if (th->syn && !before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
- NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_TCPABORTONSYN);
- tcp_reset(sk);
- return 1;
- }
+ res = tcp_validate_incoming(sk, skb, th, 0);
+ if (res <= 0)
+ return -res;
/* step 5: check the ACK field */
if (th->ack) {