Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

Pull networking changes from David S Miller:

 1) Remove the ipv4 routing cache.  Now lookups go directly into the FIB
    trie and use prebuilt routes cached there.

    No more garbage collection, no more rDOS attacks on the routing
    cache.  Instead we now get predictable and consistent performance,
    no matter what the pattern of traffic we service.

    This has been almost 2 years in the making.  Special thanks to
    Julian Anastasov, Eric Dumazet, Steffen Klassert, and others who
    have helped along the way.

    I'm sure that with a change of this magnitude there will be some
    kind of fallout, but such things ought the be simple to fix at this
    point.  Luckily I'm not European so I'll be around all of August to
    fix things :-)

    The major stages of this work here are each fronted by a forced
    merge commit whose commit message contains a top-level description
    of the motivations and implementation issues.

 2) Pre-demux of established ipv4 TCP sockets, saves a route demux on
    input.

 3) TCP SYN/ACK performance tweaks from Eric Dumazet.

 4) Add namespace support for netfilter L4 conntrack helpers, from Gao
    Feng.

 5) Add config mechanism for Energy Efficient Ethernet to ethtool, from
    Yuval Mintz.

 6) Remove quadratic behavior from /proc/net/unix, from Eric Dumazet.

 7) Support for connection tracker helpers in userspace, from Pablo
    Neira Ayuso.

 8) Allow userspace driven TX load balancing functions in TEAM driver,
    from Jiri Pirko.

 9) Kill off NLMSG_PUT and RTA_PUT macros, more gross stuff with
    embedded gotos.

10) TCP Small Queues, essentially minimize the amount of TCP data queued
    up in the packet scheduler layer.  Whereas the existing BQL (Byte
    Queue Limits) limits the pkt_sched --> netdevice queuing levels,
    this controls the TCP --> pkt_sched queueing levels.

    From Eric Dumazet.

11) Reduce the number of get_page/put_page ops done on SKB fragments,
    from Alexander Duyck.

12) Implement protection against blind resets in TCP (RFC 5961), from
    Eric Dumazet.

13) Support the client side of TCP Fast Open, basically the ability to
    send data in the SYN exchange, from Yuchung Cheng.

    Basically, the sender queues up data with a sendmsg() call using
    MSG_FASTOPEN, then they do the connect() which emits the queued up
    fastopen data.

14) Avoid all the problems we get into in TCP when timers or PMTU events
    hit a locked socket.  The TCP Small Queues changes added a
    tcp_release_cb() that allows us to queue work up to the
    release_sock() caller, and that's what we use here too.  From Eric
    Dumazet.

15) Zero copy on TX support for TUN driver, from Michael S. Tsirkin.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1870 commits)
  genetlink: define lockdep_genl_is_held() when CONFIG_LOCKDEP
  r8169: revert "add byte queue limit support".
  ipv4: Change rt->rt_iif encoding.
  net: Make skb->skb_iif always track skb->dev
  ipv4: Prepare for change of rt->rt_iif encoding.
  ipv4: Remove all RTCF_DIRECTSRC handliing.
  ipv4: Really ignore ICMP address requests/replies.
  decnet: Don't set RTCF_DIRECTSRC.
  net/ipv4/ip_vti.c: Fix __rcu warnings detected by sparse.
  ipv4: Remove redundant assignment
  rds: set correct msg_namelen
  openvswitch: potential NULL deref in sample()
  tcp: dont drop MTU reduction indications
  bnx2x: Add new 57840 device IDs
  tcp: avoid oops in tcp_metrics and reset tcpm_stamp
  niu: Change niu_rbr_fill() to use unlikely() to check niu_rbr_add_page() return value
  niu: Fix to check for dma mapping errors.
  net: Fix references to out-of-scope variables in put_cmsg_compat()
  net: ethernet: davinci_emac: add pm_runtime support
  net: ethernet: davinci_emac: Remove unnecessary #include
  ...

1 files changed, 387 insertions, 0 deletions

diff --git a/drivers/net/wireless/iwlwifi/dvm/power.c b/drivers/net/wireless/iwlwifi/dvm/power.c
new file mode 100644
index 000000000000..518cf3715809
--- /dev/null
+++ b/drivers/net/wireless/iwlwifi/dvm/power.c
@@ -0,0 +1,387 @@
+/******************************************************************************
+ *
+ * Copyright(c) 2007 - 2012 Intel Corporation. All rights reserved.
+ *
+ * Portions of this file are derived from the ipw3945 project, as well
+ * as portions of the ieee80211 subsystem header files.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2 of the GNU General Public License as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110, USA
+ *
+ * The full GNU General Public License is included in this distribution in the
+ * file called LICENSE.
+ *
+ * Contact Information:
+ *  Intel Linux Wireless <ilw@linux.intel.com>
+ * Intel Corporation, 5200 N.E. Elam Young Parkway, Hillsboro, OR 97124-6497
+ *****************************************************************************/
+
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/init.h>
+#include <net/mac80211.h>
+#include "iwl-io.h"
+#include "iwl-debug.h"
+#include "iwl-trans.h"
+#include "iwl-modparams.h"
+#include "dev.h"
+#include "agn.h"
+#include "commands.h"
+#include "power.h"
+
+/*
+ * Setting power level allows the card to go to sleep when not busy.
+ *
+ * We calculate a sleep command based on the required latency, which
+ * we get from mac80211. In order to handle thermal throttling, we can
+ * also use pre-defined power levels.
+ */
+
+/*
+ * This defines the old power levels. They are still used by default
+ * (level 1) and for thermal throttle (levels 3 through 5)
+ */
+
+struct iwl_power_vec_entry {
+	struct iwl_powertable_cmd cmd;
+	u8 no_dtim;	/* number of skip dtim */
+};
+
+#define IWL_DTIM_RANGE_0_MAX	2
+#define IWL_DTIM_RANGE_1_MAX	10
+
+#define NOSLP cpu_to_le16(0), 0, 0
+#define SLP IWL_POWER_DRIVER_ALLOW_SLEEP_MSK, 0, 0
+#define ASLP (IWL_POWER_POWER_SAVE_ENA_MSK |	\
+		IWL_POWER_POWER_MANAGEMENT_ENA_MSK | \
+		IWL_POWER_ADVANCE_PM_ENA_MSK)
+#define ASLP_TOUT(T) cpu_to_le32(T)
+#define TU_TO_USEC 1024
+#define SLP_TOUT(T) cpu_to_le32((T) * TU_TO_USEC)
+#define SLP_VEC(X0, X1, X2, X3, X4) {cpu_to_le32(X0), \
+				     cpu_to_le32(X1), \
+				     cpu_to_le32(X2), \
+				     cpu_to_le32(X3), \
+				     cpu_to_le32(X4)}
+/* default power management (not Tx power) table values */
+/* for DTIM period 0 through IWL_DTIM_RANGE_0_MAX */
+/* DTIM 0 - 2 */
+static const struct iwl_power_vec_entry range_0[IWL_POWER_NUM] = {
+	{{SLP, SLP_TOUT(200), SLP_TOUT(500), SLP_VEC(1, 1, 2, 2, 0xFF)}, 0},
+	{{SLP, SLP_TOUT(200), SLP_TOUT(300), SLP_VEC(1, 2, 2, 2, 0xFF)}, 0},
+	{{SLP, SLP_TOUT(50), SLP_TOUT(100), SLP_VEC(2, 2, 2, 2, 0xFF)}, 0},
+	{{SLP, SLP_TOUT(50), SLP_TOUT(25), SLP_VEC(2, 2, 4, 4, 0xFF)}, 1},
+	{{SLP, SLP_TOUT(25), SLP_TOUT(25), SLP_VEC(2, 2, 4, 6, 0xFF)}, 2}
+};
+
+
+/* for DTIM period IWL_DTIM_RANGE_0_MAX + 1 through IWL_DTIM_RANGE_1_MAX */
+/* DTIM 3 - 10 */
+static const struct iwl_power_vec_entry range_1[IWL_POWER_NUM] = {
+	{{SLP, SLP_TOUT(200), SLP_TOUT(500), SLP_VEC(1, 2, 3, 4, 4)}, 0},
+	{{SLP, SLP_TOUT(200), SLP_TOUT(300), SLP_VEC(1, 2, 3, 4, 7)}, 0},
+	{{SLP, SLP_TOUT(50), SLP_TOUT(100), SLP_VEC(2, 4, 6, 7, 9)}, 0},
+	{{SLP, SLP_TOUT(50), SLP_TOUT(25), SLP_VEC(2, 4, 6, 9, 10)}, 1},
+	{{SLP, SLP_TOUT(25), SLP_TOUT(25), SLP_VEC(2, 4, 6, 10, 10)}, 2}
+};
+
+/* for DTIM period > IWL_DTIM_RANGE_1_MAX */
+/* DTIM 11 - */
+static const struct iwl_power_vec_entry range_2[IWL_POWER_NUM] = {
+	{{SLP, SLP_TOUT(200), SLP_TOUT(500), SLP_VEC(1, 2, 3, 4, 0xFF)}, 0},
+	{{SLP, SLP_TOUT(200), SLP_TOUT(300), SLP_VEC(2, 4, 6, 7, 0xFF)}, 0},
+	{{SLP, SLP_TOUT(50), SLP_TOUT(100), SLP_VEC(2, 7, 9, 9, 0xFF)}, 0},
+	{{SLP, SLP_TOUT(50), SLP_TOUT(25), SLP_VEC(2, 7, 9, 9, 0xFF)}, 0},
+	{{SLP, SLP_TOUT(25), SLP_TOUT(25), SLP_VEC(4, 7, 10, 10, 0xFF)}, 0}
+};
+
+/* advance power management */
+/* DTIM 0 - 2 */
+static const struct iwl_power_vec_entry apm_range_0[IWL_POWER_NUM] = {
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 4, 6, 0xFF), 0}, 0},
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 4, 6, 0xFF), 0}, 0},
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 4, 6, 0xFF), 0}, 0},
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 4, 6, 0xFF), 0}, 0},
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 6, 8, 0xFF), ASLP_TOUT(2)}, 2}
+};
+
+
+/* for DTIM period IWL_DTIM_RANGE_0_MAX + 1 through IWL_DTIM_RANGE_1_MAX */
+/* DTIM 3 - 10 */
+static const struct iwl_power_vec_entry apm_range_1[IWL_POWER_NUM] = {
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 4, 6, 0xFF), 0}, 0},
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 4, 6, 0xFF), 0}, 0},
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 4, 6, 0xFF), 0}, 0},
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 4, 6, 0xFF), 0}, 0},
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 6, 8, 0xFF), 0}, 2}
+};
+
+/* for DTIM period > IWL_DTIM_RANGE_1_MAX */
+/* DTIM 11 - */
+static const struct iwl_power_vec_entry apm_range_2[IWL_POWER_NUM] = {
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 4, 6, 0xFF), 0}, 0},
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 4, 6, 0xFF), 0}, 0},
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 4, 6, 0xFF), 0}, 0},
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 4, 6, 0xFF), 0}, 0},
+	{{ASLP, 0, 0, ASLP_TOUT(50), ASLP_TOUT(50),
+		SLP_VEC(1, 2, 6, 8, 0xFF), ASLP_TOUT(2)}, 2}
+};
+
+static void iwl_static_sleep_cmd(struct iwl_priv *priv,
+				 struct iwl_powertable_cmd *cmd,
+				 enum iwl_power_level lvl, int period)
+{
+	const struct iwl_power_vec_entry *table;
+	int max_sleep[IWL_POWER_VEC_SIZE] = { 0 };
+	int i;
+	u8 skip;
+	u32 slp_itrvl;
+
+	if (priv->cfg->adv_pm) {
+		table = apm_range_2;
+		if (period <= IWL_DTIM_RANGE_1_MAX)
+			table = apm_range_1;
+		if (period <= IWL_DTIM_RANGE_0_MAX)
+			table = apm_range_0;
+	} else {
+		table = range_2;
+		if (period <= IWL_DTIM_RANGE_1_MAX)
+			table = range_1;
+		if (period <= IWL_DTIM_RANGE_0_MAX)
+			table = range_0;
+	}
+
+	if (WARN_ON(lvl < 0 || lvl >= IWL_POWER_NUM))
+		memset(cmd, 0, sizeof(*cmd));
+	else
+		*cmd = table[lvl].cmd;
+
+	if (period == 0) {
+		skip = 0;
+		period = 1;
+		for (i = 0; i < IWL_POWER_VEC_SIZE; i++)
+			max_sleep[i] =  1;
+
+	} else {
+		skip = table[lvl].no_dtim;
+		for (i = 0; i < IWL_POWER_VEC_SIZE; i++)
+			max_sleep[i] = le32_to_cpu(cmd->sleep_interval[i]);
+		max_sleep[IWL_POWER_VEC_SIZE - 1] = skip + 1;
+	}
+
+	slp_itrvl = le32_to_cpu(cmd->sleep_interval[IWL_POWER_VEC_SIZE - 1]);
+	/* figure out the listen interval based on dtim period and skip */
+	if (slp_itrvl == 0xFF)
+		cmd->sleep_interval[IWL_POWER_VEC_SIZE - 1] =
+			cpu_to_le32(period * (skip + 1));
+
+	slp_itrvl = le32_to_cpu(cmd->sleep_interval[IWL_POWER_VEC_SIZE - 1]);
+	if (slp_itrvl > period)
+		cmd->sleep_interval[IWL_POWER_VEC_SIZE - 1] =
+			cpu_to_le32((slp_itrvl / period) * period);
+
+	if (skip)
+		cmd->flags |= IWL_POWER_SLEEP_OVER_DTIM_MSK;
+	else
+		cmd->flags &= ~IWL_POWER_SLEEP_OVER_DTIM_MSK;
+
+	if (priv->cfg->base_params->shadow_reg_enable)
+		cmd->flags |= IWL_POWER_SHADOW_REG_ENA;
+	else
+		cmd->flags &= ~IWL_POWER_SHADOW_REG_ENA;
+
+	if (iwl_advanced_bt_coexist(priv)) {
+		if (!priv->cfg->bt_params->bt_sco_disable)
+			cmd->flags |= IWL_POWER_BT_SCO_ENA;
+		else
+			cmd->flags &= ~IWL_POWER_BT_SCO_ENA;
+	}
+
+
+	slp_itrvl = le32_to_cpu(cmd->sleep_interval[IWL_POWER_VEC_SIZE - 1]);
+	if (slp_itrvl > IWL_CONN_MAX_LISTEN_INTERVAL)
+		cmd->sleep_interval[IWL_POWER_VEC_SIZE - 1] =
+			cpu_to_le32(IWL_CONN_MAX_LISTEN_INTERVAL);
+
+	/* enforce max sleep interval */
+	for (i = IWL_POWER_VEC_SIZE - 1; i >= 0 ; i--) {
+		if (le32_to_cpu(cmd->sleep_interval[i]) >
+		    (max_sleep[i] * period))
+			cmd->sleep_interval[i] =
+				cpu_to_le32(max_sleep[i] * period);
+		if (i != (IWL_POWER_VEC_SIZE - 1)) {
+			if (le32_to_cpu(cmd->sleep_interval[i]) >
+			    le32_to_cpu(cmd->sleep_interval[i+1]))
+				cmd->sleep_interval[i] =
+					cmd->sleep_interval[i+1];
+		}
+	}
+
+	if (priv->power_data.bus_pm)
+		cmd->flags |= IWL_POWER_PCI_PM_MSK;
+	else
+		cmd->flags &= ~IWL_POWER_PCI_PM_MSK;
+
+	IWL_DEBUG_POWER(priv, "numSkipDtim = %u, dtimPeriod = %d\n",
+			skip, period);
+	/* The power level here is 0-4 (used as array index), but user expects
+	to see 1-5 (according to spec). */
+	IWL_DEBUG_POWER(priv, "Sleep command for index %d\n", lvl + 1);
+}
+
+static void iwl_power_sleep_cam_cmd(struct iwl_priv *priv,
+				    struct iwl_powertable_cmd *cmd)
+{
+	memset(cmd, 0, sizeof(*cmd));
+
+	if (priv->power_data.bus_pm)
+		cmd->flags |= IWL_POWER_PCI_PM_MSK;
+
+	IWL_DEBUG_POWER(priv, "Sleep command for CAM\n");
+}
+
+static int iwl_set_power(struct iwl_priv *priv, struct iwl_powertable_cmd *cmd)
+{
+	IWL_DEBUG_POWER(priv, "Sending power/sleep command\n");
+	IWL_DEBUG_POWER(priv, "Flags value = 0x%08X\n", cmd->flags);
+	IWL_DEBUG_POWER(priv, "Tx timeout = %u\n", le32_to_cpu(cmd->tx_data_timeout));
+	IWL_DEBUG_POWER(priv, "Rx timeout = %u\n", le32_to_cpu(cmd->rx_data_timeout));
+	IWL_DEBUG_POWER(priv, "Sleep interval vector = { %d , %d , %d , %d , %d }\n",
+			le32_to_cpu(cmd->sleep_interval[0]),
+			le32_to_cpu(cmd->sleep_interval[1]),
+			le32_to_cpu(cmd->sleep_interval[2]),
+			le32_to_cpu(cmd->sleep_interval[3]),
+			le32_to_cpu(cmd->sleep_interval[4]));
+
+	return iwl_dvm_send_cmd_pdu(priv, POWER_TABLE_CMD, CMD_SYNC,
+				sizeof(struct iwl_powertable_cmd), cmd);
+}
+
+static void iwl_power_build_cmd(struct iwl_priv *priv,
+				struct iwl_powertable_cmd *cmd)
+{
+	bool enabled = priv->hw->conf.flags & IEEE80211_CONF_PS;
+	int dtimper;
+
+	dtimper = priv->hw->conf.ps_dtim_period ?: 1;
+
+	if (priv->wowlan)
+		iwl_static_sleep_cmd(priv, cmd, IWL_POWER_INDEX_5, dtimper);
+	else if (!priv->cfg->base_params->no_idle_support &&
+		 priv->hw->conf.flags & IEEE80211_CONF_IDLE)
+		iwl_static_sleep_cmd(priv, cmd, IWL_POWER_INDEX_5, 20);
+	else if (iwl_tt_is_low_power_state(priv)) {
+		/* in thermal throttling low power state */
+		iwl_static_sleep_cmd(priv, cmd,
+		    iwl_tt_current_power_mode(priv), dtimper);
+	} else if (!enabled)
+		iwl_power_sleep_cam_cmd(priv, cmd);
+	else if (priv->power_data.debug_sleep_level_override >= 0)
+		iwl_static_sleep_cmd(priv, cmd,
+				     priv->power_data.debug_sleep_level_override,
+				     dtimper);
+	else {
+		/* Note that the user parameter is 1-5 (according to spec),
+		but we pass 0-4 because it acts as an array index. */
+		if (iwlwifi_mod_params.power_level > IWL_POWER_INDEX_1 &&
+		    iwlwifi_mod_params.power_level <= IWL_POWER_NUM)
+			iwl_static_sleep_cmd(priv, cmd,
+				iwlwifi_mod_params.power_level - 1, dtimper);
+		else
+			iwl_static_sleep_cmd(priv, cmd,
+				IWL_POWER_INDEX_1, dtimper);
+	}
+}
+
+int iwl_power_set_mode(struct iwl_priv *priv, struct iwl_powertable_cmd *cmd,
+		       bool force)
+{
+	int ret;
+	bool update_chains;
+
+	lockdep_assert_held(&priv->mutex);
+
+	/* Don't update the RX chain when chain noise calibration is running */
+	update_chains = priv->chain_noise_data.state == IWL_CHAIN_NOISE_DONE ||
+			priv->chain_noise_data.state == IWL_CHAIN_NOISE_ALIVE;
+
+	if (!memcmp(&priv->power_data.sleep_cmd, cmd, sizeof(*cmd)) && !force)
+		return 0;
+
+	if (!iwl_is_ready_rf(priv))
+		return -EIO;
+
+	/* scan complete use sleep_power_next, need to be updated */
+	memcpy(&priv->power_data.sleep_cmd_next, cmd, sizeof(*cmd));
+	if (test_bit(STATUS_SCANNING, &priv->status) && !force) {
+		IWL_DEBUG_INFO(priv, "Defer power set mode while scanning\n");
+		return 0;
+	}
+
+	if (cmd->flags & IWL_POWER_DRIVER_ALLOW_SLEEP_MSK)
+		iwl_dvm_set_pmi(priv, true);
+
+	ret = iwl_set_power(priv, cmd);
+	if (!ret) {
+		if (!(cmd->flags & IWL_POWER_DRIVER_ALLOW_SLEEP_MSK))
+			iwl_dvm_set_pmi(priv, false);
+
+		if (update_chains)
+			iwl_update_chain_flags(priv);
+		else
+			IWL_DEBUG_POWER(priv,
+					"Cannot update the power, chain noise "
+					"calibration running: %d\n",
+					priv->chain_noise_data.state);
+
+		memcpy(&priv->power_data.sleep_cmd, cmd, sizeof(*cmd));
+	} else
+		IWL_ERR(priv, "set power fail, ret = %d", ret);
+
+	return ret;
+}
+
+int iwl_power_update_mode(struct iwl_priv *priv, bool force)
+{
+	struct iwl_powertable_cmd cmd;
+
+	iwl_power_build_cmd(priv, &cmd);
+	return iwl_power_set_mode(priv, &cmd, force);
+}
+
+/* initialize to default */
+void iwl_power_initialize(struct iwl_priv *priv)
+{
+	priv->power_data.bus_pm = priv->trans->pm_support;
+
+	priv->power_data.debug_sleep_level_override = -1;
+
+	memset(&priv->power_data.sleep_cmd, 0,
+		sizeof(priv->power_data.sleep_cmd));
+}