Merge branch 'raid56-scrub-replace' of git://github.com/miaoxie/linux-btrfs into for-linus

55 files changed, 301 insertions, 152 deletions

diff --git a/arch/arm/boot/dts/r8a7740.dtsi b/arch/arm/boot/dts/r8a7740.dtsi
index d46c213a17ad..eed697a6bd6b 100644
--- a/arch/arm/boot/dts/r8a7740.dtsi
+++ b/arch/arm/boot/dts/r8a7740.dtsi
@@ -433,7 +433,7 @@
 			clocks = <&cpg_clocks R8A7740_CLK_S>,
 				 <&cpg_clocks R8A7740_CLK_S>, <&sub_clk>,
 				 <&cpg_clocks R8A7740_CLK_B>,
-				 <&sub_clk>, <&sub_clk>,
+				 <&cpg_clocks R8A7740_CLK_HPP>, <&sub_clk>,
 				 <&cpg_clocks R8A7740_CLK_B>;
 			#clock-cells = <1>;
 			renesas,clock-indices = <
diff --git a/arch/arm/boot/dts/r8a7790.dtsi b/arch/arm/boot/dts/r8a7790.dtsi
index d0e17733dc1a..e20affe156c1 100644
--- a/arch/arm/boot/dts/r8a7790.dtsi
+++ b/arch/arm/boot/dts/r8a7790.dtsi
@@ -666,9 +666,9 @@
 			#clock-cells = <0>;
 			clock-output-names = "sd2";
 		};
-		sd3_clk: sd3_clk@e615007c {
+		sd3_clk: sd3_clk@e615026c {
 			compatible = "renesas,r8a7790-div6-clock", "renesas,cpg-div6-clock";
-			reg = <0 0xe615007c 0 4>;
+			reg = <0 0xe615026c 0 4>;
 			clocks = <&pll1_div2_clk>;
 			#clock-cells = <0>;
 			clock-output-names = "sd3";
diff --git a/arch/arm/boot/dts/sun6i-a31.dtsi b/arch/arm/boot/dts/sun6i-a31.dtsi
index 543f895d18d3..2e652e2339e9 100644
--- a/arch/arm/boot/dts/sun6i-a31.dtsi
+++ b/arch/arm/boot/dts/sun6i-a31.dtsi
@@ -361,6 +361,10 @@
 			clocks = <&ahb1_gates 6>;
 			resets = <&ahb1_rst 6>;
 			#dma-cells = <1>;
+
+			/* DMA controller requires AHB1 clocked from PLL6 */
+			assigned-clocks = <&ahb1_mux>;
+			assigned-clock-parents = <&pll6>;
 		};
 
 		mmc0: mmc@01c0f000 {
diff --git a/arch/arm/boot/dts/tegra114-dalmore.dts b/arch/arm/boot/dts/tegra114-dalmore.dts
index 5c21d216515a..8b7aa0dcdc6e 100644
--- a/arch/arm/boot/dts/tegra114-dalmore.dts
+++ b/arch/arm/boot/dts/tegra114-dalmore.dts
@@ -15,6 +15,7 @@
 	aliases {
 		rtc0 = "/i2c@7000d000/tps65913@58";
 		rtc1 = "/rtc@7000e000";
+		serial0 = &uartd;
 	};
 
 	memory {
diff --git a/arch/arm/boot/dts/tegra114-roth.dts b/arch/arm/boot/dts/tegra114-roth.dts
index c7c6825f11fb..38acf78d7815 100644
--- a/arch/arm/boot/dts/tegra114-roth.dts
+++ b/arch/arm/boot/dts/tegra114-roth.dts
@@ -15,6 +15,10 @@
 		linux,initrd-end = <0x82800000>;
 	};
 
+	aliases {
+		serial0 = &uartd;
+	};
+
 	firmware {
 		trusted-foundations {
 			compatible = "tlm,trusted-foundations";
@@ -916,8 +920,6 @@
 						regulator-name = "vddio-sdmmc3";
 						regulator-min-microvolt = <1800000>;
 						regulator-max-microvolt = <3300000>;
-						regulator-always-on;
-						regulator-boot-on;
 					};
 
 					ldousb {
@@ -962,7 +964,7 @@
 	sdhci@78000400 {
 		status = "okay";
 		bus-width = <4>;
-		vmmc-supply = <&vddio_sdmmc3>;
+		vqmmc-supply = <&vddio_sdmmc3>;
 		cd-gpios = <&gpio TEGRA_GPIO(V, 2) GPIO_ACTIVE_LOW>;
 		power-gpios = <&gpio TEGRA_GPIO(H, 0) GPIO_ACTIVE_HIGH>;
 	};
@@ -971,7 +973,6 @@
 	sdhci@78000600 {
 		status = "okay";
 		bus-width = <8>;
-		vmmc-supply = <&vdd_1v8>;
 		non-removable;
 	};
 
diff --git a/arch/arm/boot/dts/tegra114-tn7.dts b/arch/arm/boot/dts/tegra114-tn7.dts
index 963662145635..f91c2c9b2f94 100644
--- a/arch/arm/boot/dts/tegra114-tn7.dts
+++ b/arch/arm/boot/dts/tegra114-tn7.dts
@@ -15,6 +15,10 @@
 		linux,initrd-end = <0x82800000>;
 	};
 
+	aliases {
+		serial0 = &uartd;
+	};
+
 	firmware {
 		trusted-foundations {
 			compatible = "tlm,trusted-foundations";
@@ -240,7 +244,6 @@
 	sdhci@78000600 {
 		status = "okay";
 		bus-width = <8>;
-		vmmc-supply = <&vdd_1v8>;
 		non-removable;
 	};
 
diff --git a/arch/arm/boot/dts/tegra114.dtsi b/arch/arm/boot/dts/tegra114.dtsi
index 2ca9c1807f72..222f3b3f4dd5 100644
--- a/arch/arm/boot/dts/tegra114.dtsi
+++ b/arch/arm/boot/dts/tegra114.dtsi
@@ -9,13 +9,6 @@
 	compatible = "nvidia,tegra114";
 	interrupt-parent = <&gic>;
 
-	aliases {
-		serial0 = &uarta;
-		serial1 = &uartb;
-		serial2 = &uartc;
-		serial3 = &uartd;
-	};
-
 	host1x@50000000 {
 		compatible = "nvidia,tegra114-host1x", "simple-bus";
 		reg = <0x50000000 0x00028000>;
diff --git a/arch/arm/boot/dts/tegra124-jetson-tk1.dts b/arch/arm/boot/dts/tegra124-jetson-tk1.dts
index 029c9a021541..51b373ff1065 100644
--- a/arch/arm/boot/dts/tegra124-jetson-tk1.dts
+++ b/arch/arm/boot/dts/tegra124-jetson-tk1.dts
@@ -10,6 +10,7 @@
 	aliases {
 		rtc0 = "/i2c@0,7000d000/pmic@40";
 		rtc1 = "/rtc@0,7000e000";
+		serial0 = &uartd;
 	};
 
 	memory {
diff --git a/arch/arm/boot/dts/tegra124-nyan-big.dts b/arch/arm/boot/dts/tegra124-nyan-big.dts
index 7d0784ce4c74..53181d310247 100644
--- a/arch/arm/boot/dts/tegra124-nyan-big.dts
+++ b/arch/arm/boot/dts/tegra124-nyan-big.dts
@@ -10,6 +10,7 @@
 	aliases {
 		rtc0 = "/i2c@0,7000d000/pmic@40";
 		rtc1 = "/rtc@0,7000e000";
+		serial0 = &uarta;
 	};
 
 	memory {
diff --git a/arch/arm/boot/dts/tegra124-venice2.dts b/arch/arm/boot/dts/tegra124-venice2.dts
index 13008858e967..5c3f7813360d 100644
--- a/arch/arm/boot/dts/tegra124-venice2.dts
+++ b/arch/arm/boot/dts/tegra124-venice2.dts
@@ -10,6 +10,7 @@
 	aliases {
 		rtc0 = "/i2c@0,7000d000/pmic@40";
 		rtc1 = "/rtc@0,7000e000";
+		serial0 = &uarta;
 	};
 
 	memory {
diff --git a/arch/arm/boot/dts/tegra124.dtsi b/arch/arm/boot/dts/tegra124.dtsi
index 478c555ebd96..df2b06b29985 100644
--- a/arch/arm/boot/dts/tegra124.dtsi
+++ b/arch/arm/boot/dts/tegra124.dtsi
@@ -286,7 +286,7 @@
 	 * the APB DMA based serial driver, the comptible is
 	 * "nvidia,tegra124-hsuart", "nvidia,tegra30-hsuart".
 	 */
-	serial@0,70006000 {
+	uarta: serial@0,70006000 {
 		compatible = "nvidia,tegra124-uart", "nvidia,tegra20-uart";
 		reg = <0x0 0x70006000 0x0 0x40>;
 		reg-shift = <2>;
@@ -299,7 +299,7 @@
 		status = "disabled";
 	};
 
-	serial@0,70006040 {
+	uartb: serial@0,70006040 {
 		compatible = "nvidia,tegra124-uart", "nvidia,tegra20-uart";
 		reg = <0x0 0x70006040 0x0 0x40>;
 		reg-shift = <2>;
@@ -312,7 +312,7 @@
 		status = "disabled";
 	};
 
-	serial@0,70006200 {
+	uartc: serial@0,70006200 {
 		compatible = "nvidia,tegra124-uart", "nvidia,tegra20-uart";
 		reg = <0x0 0x70006200 0x0 0x40>;
 		reg-shift = <2>;
@@ -325,7 +325,7 @@
 		status = "disabled";
 	};
 
-	serial@0,70006300 {
+	uartd: serial@0,70006300 {
 		compatible = "nvidia,tegra124-uart", "nvidia,tegra20-uart";
 		reg = <0x0 0x70006300 0x0 0x40>;
 		reg-shift = <2>;
diff --git a/arch/arm/boot/dts/tegra20-harmony.dts b/arch/arm/boot/dts/tegra20-harmony.dts
index a37279af687c..b926a07b9443 100644
--- a/arch/arm/boot/dts/tegra20-harmony.dts
+++ b/arch/arm/boot/dts/tegra20-harmony.dts
@@ -10,6 +10,7 @@
 	aliases {
 		rtc0 = "/i2c@7000d000/tps6586x@34";
 		rtc1 = "/rtc@7000e000";
+		serial0 = &uartd;
 	};
 
 	memory {
diff --git a/arch/arm/boot/dts/tegra20-iris-512.dts b/arch/arm/boot/dts/tegra20-iris-512.dts
index 8cfb83f42e1f..1dd7d7bfdfcc 100644
--- a/arch/arm/boot/dts/tegra20-iris-512.dts
+++ b/arch/arm/boot/dts/tegra20-iris-512.dts
@@ -6,6 +6,11 @@
 	model = "Toradex Colibri T20 512MB on Iris";
 	compatible = "toradex,iris", "toradex,colibri_t20-512", "nvidia,tegra20";
 
+	aliases {
+		serial0 = &uarta;
+		serial1 = &uartd;
+	};
+
 	host1x@50000000 {
 		hdmi@54280000 {
 			status = "okay";
diff --git a/arch/arm/boot/dts/tegra20-medcom-wide.dts b/arch/arm/boot/dts/tegra20-medcom-wide.dts
index 1b7c56b33aca..9b87526ab0b7 100644
--- a/arch/arm/boot/dts/tegra20-medcom-wide.dts
+++ b/arch/arm/boot/dts/tegra20-medcom-wide.dts
@@ -6,6 +6,10 @@
 	model = "Avionic Design Medcom-Wide board";
 	compatible = "ad,medcom-wide", "ad,tamonten", "nvidia,tegra20";
 
+	aliases {
+		serial0 = &uartd;
+	};
+
 	pwm@7000a000 {
 		status = "okay";
 	};
diff --git a/arch/arm/boot/dts/tegra20-paz00.dts b/arch/arm/boot/dts/tegra20-paz00.dts
index d4438e30de45..ed7e1009326c 100644
--- a/arch/arm/boot/dts/tegra20-paz00.dts
+++ b/arch/arm/boot/dts/tegra20-paz00.dts
@@ -10,6 +10,8 @@
 	aliases {
 		rtc0 = "/i2c@7000d000/tps6586x@34";
 		rtc1 = "/rtc@7000e000";
+		serial0 = &uarta;
+		serial1 = &uartc;
 	};
 
 	memory {
diff --git a/arch/arm/boot/dts/tegra20-seaboard.dts b/arch/arm/boot/dts/tegra20-seaboard.dts
index a1d4bf9895d7..ea282c7c0ca5 100644
--- a/arch/arm/boot/dts/tegra20-seaboard.dts
+++ b/arch/arm/boot/dts/tegra20-seaboard.dts
@@ -10,6 +10,7 @@
 	aliases {
 		rtc0 = "/i2c@7000d000/tps6586x@34";
 		rtc1 = "/rtc@7000e000";
+		serial0 = &uartd;
 	};
 
 	memory {
diff --git a/arch/arm/boot/dts/tegra20-tamonten.dtsi b/arch/arm/boot/dts/tegra20-tamonten.dtsi
index 80e7d386ce34..13d4e6185275 100644
--- a/arch/arm/boot/dts/tegra20-tamonten.dtsi
+++ b/arch/arm/boot/dts/tegra20-tamonten.dtsi
@@ -7,6 +7,7 @@
 	aliases {
 		rtc0 = "/i2c@7000d000/tps6586x@34";
 		rtc1 = "/rtc@7000e000";
+		serial0 = &uartd;
 	};
 
 	memory {
diff --git a/arch/arm/boot/dts/tegra20-trimslice.dts b/arch/arm/boot/dts/tegra20-trimslice.dts
index 5ad87979ab13..d99af4ef9c64 100644
--- a/arch/arm/boot/dts/tegra20-trimslice.dts
+++ b/arch/arm/boot/dts/tegra20-trimslice.dts
@@ -10,6 +10,7 @@
 	aliases {
 		rtc0 = "/i2c@7000c500/rtc@56";
 		rtc1 = "/rtc@7000e000";
+		serial0 = &uarta;
 	};
 
 	memory {
diff --git a/arch/arm/boot/dts/tegra20-ventana.dts b/arch/arm/boot/dts/tegra20-ventana.dts
index ca8484cccddc..04c58e9ca490 100644
--- a/arch/arm/boot/dts/tegra20-ventana.dts
+++ b/arch/arm/boot/dts/tegra20-ventana.dts
@@ -10,6 +10,7 @@
 	aliases {
 		rtc0 = "/i2c@7000d000/tps6586x@34";
 		rtc1 = "/rtc@7000e000";
+		serial0 = &uartd;
 	};
 
 	memory {
diff --git a/arch/arm/boot/dts/tegra20-whistler.dts b/arch/arm/boot/dts/tegra20-whistler.dts
index 1843725785c9..340d81108df1 100644
--- a/arch/arm/boot/dts/tegra20-whistler.dts
+++ b/arch/arm/boot/dts/tegra20-whistler.dts
@@ -10,6 +10,7 @@
 	aliases {
 		rtc0 = "/i2c@7000d000/max8907@3c";
 		rtc1 = "/rtc@7000e000";
+		serial0 = &uarta;
 	};
 
 	memory {
diff --git a/arch/arm/boot/dts/tegra20.dtsi b/arch/arm/boot/dts/tegra20.dtsi
index 3b374c49d04d..8acf5d85c99d 100644
--- a/arch/arm/boot/dts/tegra20.dtsi
+++ b/arch/arm/boot/dts/tegra20.dtsi
@@ -9,14 +9,6 @@
 	compatible = "nvidia,tegra20";
 	interrupt-parent = <&intc>;
 
-	aliases {
-		serial0 = &uarta;
-		serial1 = &uartb;
-		serial2 = &uartc;
-		serial3 = &uartd;
-		serial4 = &uarte;
-	};
-
 	host1x@50000000 {
 		compatible = "nvidia,tegra20-host1x", "simple-bus";
 		reg = <0x50000000 0x00024000>;
diff --git a/arch/arm/boot/dts/tegra30-apalis-eval.dts b/arch/arm/boot/dts/tegra30-apalis-eval.dts
index 45d40f024585..6236bdecb48b 100644
--- a/arch/arm/boot/dts/tegra30-apalis-eval.dts
+++ b/arch/arm/boot/dts/tegra30-apalis-eval.dts
@@ -11,6 +11,10 @@
 		rtc0 = "/i2c@7000c000/rtc@68";
 		rtc1 = "/i2c@7000d000/tps65911@2d";
 		rtc2 = "/rtc@7000e000";
+		serial0 = &uarta;
+		serial1 = &uartb;
+		serial2 = &uartc;
+		serial3 = &uartd;
 	};
 
 	pcie-controller@00003000 {
diff --git a/arch/arm/boot/dts/tegra30-beaver.dts b/arch/arm/boot/dts/tegra30-beaver.dts
index cee8f2246fdb..6b157eeabcc5 100644
--- a/arch/arm/boot/dts/tegra30-beaver.dts
+++ b/arch/arm/boot/dts/tegra30-beaver.dts
@@ -9,6 +9,7 @@
 	aliases {
 		rtc0 = "/i2c@7000d000/tps65911@2d";
 		rtc1 = "/rtc@7000e000";
+		serial0 = &uarta;
 	};
 
 	memory {
diff --git a/arch/arm/boot/dts/tegra30-cardhu.dtsi b/arch/arm/boot/dts/tegra30-cardhu.dtsi
index 206379546244..a1b682ea01bd 100644
--- a/arch/arm/boot/dts/tegra30-cardhu.dtsi
+++ b/arch/arm/boot/dts/tegra30-cardhu.dtsi
@@ -30,6 +30,8 @@
 	aliases {
 		rtc0 = "/i2c@7000d000/tps65911@2d";
 		rtc1 = "/rtc@7000e000";
+		serial0 = &uarta;
+		serial1 = &uartc;
 	};
 
 	memory {
diff --git a/arch/arm/boot/dts/tegra30-colibri-eval-v3.dts b/arch/arm/boot/dts/tegra30-colibri-eval-v3.dts
index 7793abd5bef1..4d3ddc585641 100644
--- a/arch/arm/boot/dts/tegra30-colibri-eval-v3.dts
+++ b/arch/arm/boot/dts/tegra30-colibri-eval-v3.dts
@@ -10,6 +10,9 @@
 		rtc0 = "/i2c@7000c000/rtc@68";
 		rtc1 = "/i2c@7000d000/tps65911@2d";
 		rtc2 = "/rtc@7000e000";
+		serial0 = &uarta;
+		serial1 = &uartb;
+		serial2 = &uartd;
 	};
 
 	host1x@50000000 {
diff --git a/arch/arm/boot/dts/tegra30.dtsi b/arch/arm/boot/dts/tegra30.dtsi
index aa6ccea13d30..b270b9e3d455 100644
--- a/arch/arm/boot/dts/tegra30.dtsi
+++ b/arch/arm/boot/dts/tegra30.dtsi
@@ -9,14 +9,6 @@
 	compatible = "nvidia,tegra30";
 	interrupt-parent = <&intc>;
 
-	aliases {
-		serial0 = &uarta;
-		serial1 = &uartb;
-		serial2 = &uartc;
-		serial3 = &uartd;
-		serial4 = &uarte;
-	};
-
 	pcie-controller@00003000 {
 		compatible = "nvidia,tegra30-pcie";
 		device_type = "pci";
diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig
index 3487046d8a78..9d7a32f93fcf 100644
--- a/arch/arm/configs/multi_v7_defconfig
+++ b/arch/arm/configs/multi_v7_defconfig
@@ -217,6 +217,7 @@ CONFIG_I2C_CADENCE=y
 CONFIG_I2C_DESIGNWARE_PLATFORM=y
 CONFIG_I2C_EXYNOS5=y
 CONFIG_I2C_MV64XXX=y
+CONFIG_I2C_S3C2410=y
 CONFIG_I2C_SIRF=y
 CONFIG_I2C_TEGRA=y
 CONFIG_I2C_ST=y
diff --git a/arch/arm/mach-shmobile/clock-r8a7740.c b/arch/arm/mach-shmobile/clock-r8a7740.c
index 0794f0426e70..19df9cb30495 100644
--- a/arch/arm/mach-shmobile/clock-r8a7740.c
+++ b/arch/arm/mach-shmobile/clock-r8a7740.c
@@ -455,7 +455,7 @@ enum {
 	MSTP128, MSTP127, MSTP125,
 	MSTP116, MSTP111, MSTP100, MSTP117,
 
-	MSTP230,
+	MSTP230, MSTP229,
 	MSTP222,
 	MSTP218, MSTP217, MSTP216, MSTP214,
 	MSTP207, MSTP206, MSTP204, MSTP203, MSTP202, MSTP201, MSTP200,
@@ -474,11 +474,12 @@ static struct clk mstp_clks[MSTP_NR] = {
 	[MSTP127] = SH_CLK_MSTP32(&div4_clks[DIV4_S],	SMSTPCR1, 27, 0), /* CEU20 */
 	[MSTP125] = SH_CLK_MSTP32(&div6_clks[DIV6_SUB],	SMSTPCR1, 25, 0), /* TMU0 */
 	[MSTP117] = SH_CLK_MSTP32(&div4_clks[DIV4_B],	SMSTPCR1, 17, 0), /* LCDC1 */
-	[MSTP116] = SH_CLK_MSTP32(&div6_clks[DIV6_SUB],	SMSTPCR1, 16, 0), /* IIC0 */
+	[MSTP116] = SH_CLK_MSTP32(&div4_clks[DIV4_HPP],	SMSTPCR1, 16, 0), /* IIC0 */
 	[MSTP111] = SH_CLK_MSTP32(&div6_clks[DIV6_SUB],	SMSTPCR1, 11, 0), /* TMU1 */
 	[MSTP100] = SH_CLK_MSTP32(&div4_clks[DIV4_B],	SMSTPCR1,  0, 0), /* LCDC0 */
 
 	[MSTP230] = SH_CLK_MSTP32(&div6_clks[DIV6_SUB],	SMSTPCR2, 30, 0), /* SCIFA6 */
+	[MSTP229] = SH_CLK_MSTP32(&div4_clks[DIV4_HP],	SMSTPCR2, 29, 0), /* INTCA */
 	[MSTP222] = SH_CLK_MSTP32(&div6_clks[DIV6_SUB],	SMSTPCR2, 22, 0), /* SCIFA7 */
 	[MSTP218] = SH_CLK_MSTP32(&div4_clks[DIV4_HP],  SMSTPCR2, 18, 0), /* DMAC1 */
 	[MSTP217] = SH_CLK_MSTP32(&div4_clks[DIV4_HP],  SMSTPCR2, 17, 0), /* DMAC2 */
@@ -575,6 +576,10 @@ static struct clk_lookup lookups[] = {
 	CLKDEV_DEV_ID("sh-dma-engine.0",	&mstp_clks[MSTP218]),
 	CLKDEV_DEV_ID("sh-sci.7",		&mstp_clks[MSTP222]),
 	CLKDEV_DEV_ID("e6cd0000.serial",	&mstp_clks[MSTP222]),
+	CLKDEV_DEV_ID("renesas_intc_irqpin.0",	&mstp_clks[MSTP229]),
+	CLKDEV_DEV_ID("renesas_intc_irqpin.1",	&mstp_clks[MSTP229]),
+	CLKDEV_DEV_ID("renesas_intc_irqpin.2",	&mstp_clks[MSTP229]),
+	CLKDEV_DEV_ID("renesas_intc_irqpin.3",	&mstp_clks[MSTP229]),
 	CLKDEV_DEV_ID("sh-sci.6",		&mstp_clks[MSTP230]),
 	CLKDEV_DEV_ID("e6cc0000.serial",	&mstp_clks[MSTP230]),
 
diff --git a/arch/arm/mach-shmobile/clock-r8a7790.c b/arch/arm/mach-shmobile/clock-r8a7790.c
index 126ddafad526..f62265200592 100644
--- a/arch/arm/mach-shmobile/clock-r8a7790.c
+++ b/arch/arm/mach-shmobile/clock-r8a7790.c
@@ -68,7 +68,7 @@
 
 #define SDCKCR		0xE6150074
 #define SD2CKCR		0xE6150078
-#define SD3CKCR		0xE615007C
+#define SD3CKCR		0xE615026C
 #define MMC0CKCR	0xE6150240
 #define MMC1CKCR	0xE6150244
 #define SSPCKCR		0xE6150248
diff --git a/arch/arm/mach-shmobile/setup-sh73a0.c b/arch/arm/mach-shmobile/setup-sh73a0.c
index b7bd8e509668..328657d011d5 100644
--- a/arch/arm/mach-shmobile/setup-sh73a0.c
+++ b/arch/arm/mach-shmobile/setup-sh73a0.c
@@ -26,6 +26,7 @@
 #include <linux/of_platform.h>
 #include <linux/delay.h>
 #include <linux/input.h>
+#include <linux/i2c/i2c-sh_mobile.h>
 #include <linux/io.h>
 #include <linux/serial_sci.h>
 #include <linux/sh_dma.h>
@@ -192,11 +193,18 @@ static struct resource i2c4_resources[] = {
 	},
 };
 
+static struct i2c_sh_mobile_platform_data i2c_platform_data = {
+	.clks_per_count	= 2,
+};
+
 static struct platform_device i2c0_device = {
 	.name		= "i2c-sh_mobile",
 	.id		= 0,
 	.resource	= i2c0_resources,
 	.num_resources	= ARRAY_SIZE(i2c0_resources),
+	.dev		= {
+		.platform_data	= &i2c_platform_data,
+	},
 };
 
 static struct platform_device i2c1_device = {
@@ -204,6 +212,9 @@ static struct platform_device i2c1_device = {
 	.id		= 1,
 	.resource	= i2c1_resources,
 	.num_resources	= ARRAY_SIZE(i2c1_resources),
+	.dev		= {
+		.platform_data	= &i2c_platform_data,
+	},
 };
 
 static struct platform_device i2c2_device = {
@@ -211,6 +222,9 @@ static struct platform_device i2c2_device = {
 	.id		= 2,
 	.resource	= i2c2_resources,
 	.num_resources	= ARRAY_SIZE(i2c2_resources),
+	.dev		= {
+		.platform_data	= &i2c_platform_data,
+	},
 };
 
 static struct platform_device i2c3_device = {
@@ -218,6 +232,9 @@ static struct platform_device i2c3_device = {
 	.id		= 3,
 	.resource	= i2c3_resources,
 	.num_resources	= ARRAY_SIZE(i2c3_resources),
+	.dev		= {
+		.platform_data	= &i2c_platform_data,
+	},
 };
 
 static struct platform_device i2c4_device = {
@@ -225,6 +242,9 @@ static struct platform_device i2c4_device = {
 	.id		= 4,
 	.resource	= i2c4_resources,
 	.num_resources	= ARRAY_SIZE(i2c4_resources),
+	.dev		= {
+		.platform_data	= &i2c_platform_data,
+	},
 };
 
 static const struct sh_dmae_slave_config sh73a0_dmae_slaves[] = {
diff --git a/arch/mips/include/asm/jump_label.h b/arch/mips/include/asm/jump_label.h
index e194f957ca8c..fdbff44e5482 100644
--- a/arch/mips/include/asm/jump_label.h
+++ b/arch/mips/include/asm/jump_label.h
@@ -20,9 +20,15 @@
 #define WORD_INSN ".word"
 #endif
 
+#ifdef CONFIG_CPU_MICROMIPS
+#define NOP_INSN "nop32"
+#else
+#define NOP_INSN "nop"
+#endif
+
 static __always_inline bool arch_static_branch(struct static_key *key)
 {
-	asm_volatile_goto("1:\tnop\n\t"
+	asm_volatile_goto("1:\t" NOP_INSN "\n\t"
 		"nop\n\t"
 		".pushsection __jump_table,  \"aw\"\n\t"
 		WORD_INSN " 1b, %l[l_yes], %0\n\t"
diff --git a/arch/mips/include/asm/mach-loongson/cpu-feature-overrides.h b/arch/mips/include/asm/mach-loongson/cpu-feature-overrides.h
index 7d28f95b0512..6d69332f21ec 100644
--- a/arch/mips/include/asm/mach-loongson/cpu-feature-overrides.h
+++ b/arch/mips/include/asm/mach-loongson/cpu-feature-overrides.h
@@ -41,10 +41,8 @@
 #define cpu_has_mcheck		0
 #define cpu_has_mdmx		0
 #define cpu_has_mips16		0
-#define cpu_has_mips32r1	0
 #define cpu_has_mips32r2	0
 #define cpu_has_mips3d		0
-#define cpu_has_mips64r1	0
 #define cpu_has_mips64r2	0
 #define cpu_has_mipsmt		0
 #define cpu_has_prefetch	0
diff --git a/arch/mips/include/asm/uaccess.h b/arch/mips/include/asm/uaccess.h
index a10951090234..133678ab4eb8 100644
--- a/arch/mips/include/asm/uaccess.h
+++ b/arch/mips/include/asm/uaccess.h
@@ -301,7 +301,8 @@ do {									\
 			__get_kernel_common((x), size, __gu_ptr);	\
 		else							\
 			__get_user_common((x), size, __gu_ptr);		\
-	}								\
+	} else								\
+		(x) = 0;						\
 									\
 	__gu_err;							\
 })
@@ -316,6 +317,7 @@ do {									\
 	"	.insn						\n"	\
 	"	.section .fixup,\"ax\"				\n"	\
 	"3:	li	%0, %4					\n"	\
+	"	move	%1, $0					\n"	\
 	"	j	2b					\n"	\
 	"	.previous					\n"	\
 	"	.section __ex_table,\"a\"			\n"	\
@@ -630,6 +632,7 @@ do {									\
 	"	.insn						\n"	\
 	"	.section .fixup,\"ax\"				\n"	\
 	"3:	li	%0, %4					\n"	\
+	"	move	%1, $0					\n"	\
 	"	j	2b					\n"	\
 	"	.previous					\n"	\
 	"	.section __ex_table,\"a\"			\n"	\
@@ -773,10 +776,11 @@ extern void __put_user_unaligned_unknown(void);
 	"jal\t" #destination "\n\t"
 #endif
 
-#ifndef CONFIG_CPU_DADDI_WORKAROUNDS
-#define DADDI_SCRATCH "$0"
-#else
+#if defined(CONFIG_CPU_DADDI_WORKAROUNDS) || (defined(CONFIG_EVA) &&	\
+					      defined(CONFIG_CPU_HAS_PREFETCH))
 #define DADDI_SCRATCH "$3"
+#else
+#define DADDI_SCRATCH "$0"
 #endif
 
 extern size_t __copy_user(void *__to, const void *__from, size_t __n);
diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c
index 94c4a0c0a577..d5a4f380b019 100644
--- a/arch/mips/kernel/cpu-probe.c
+++ b/arch/mips/kernel/cpu-probe.c
@@ -757,31 +757,34 @@ static inline void cpu_probe_legacy(struct cpuinfo_mips *c, unsigned int cpu)
 			c->cputype = CPU_LOONGSON2;
 			__cpu_name[cpu] = "ICT Loongson-2";
 			set_elf_platform(cpu, "loongson2e");
+			set_isa(c, MIPS_CPU_ISA_III);
 			break;
 		case PRID_REV_LOONGSON2F:
 			c->cputype = CPU_LOONGSON2;
 			__cpu_name[cpu] = "ICT Loongson-2";
 			set_elf_platform(cpu, "loongson2f");
+			set_isa(c, MIPS_CPU_ISA_III);
 			break;
 		case PRID_REV_LOONGSON3A:
 			c->cputype = CPU_LOONGSON3;
-			c->writecombine = _CACHE_UNCACHED_ACCELERATED;
 			__cpu_name[cpu] = "ICT Loongson-3";
 			set_elf_platform(cpu, "loongson3a");
+			set_isa(c, MIPS_CPU_ISA_M64R1);
 			break;
 		case PRID_REV_LOONGSON3B_R1:
 		case PRID_REV_LOONGSON3B_R2:
 			c->cputype = CPU_LOONGSON3;
 			__cpu_name[cpu] = "ICT Loongson-3";
 			set_elf_platform(cpu, "loongson3b");
+			set_isa(c, MIPS_CPU_ISA_M64R1);
 			break;
 		}
 
-		set_isa(c, MIPS_CPU_ISA_III);
 		c->options = R4K_OPTS |
 			     MIPS_CPU_FPU | MIPS_CPU_LLSC |
 			     MIPS_CPU_32FPR;
 		c->tlbsize = 64;
+		c->writecombine = _CACHE_UNCACHED_ACCELERATED;
 		break;
 	case PRID_IMP_LOONGSON_32:  /* Loongson-1 */
 		decode_configs(c);
diff --git a/arch/mips/kernel/jump_label.c b/arch/mips/kernel/jump_label.c
index 6001610cfe55..dda800e9e731 100644
--- a/arch/mips/kernel/jump_label.c
+++ b/arch/mips/kernel/jump_label.c
@@ -18,31 +18,53 @@
 
 #ifdef HAVE_JUMP_LABEL
 
-#define J_RANGE_MASK ((1ul << 28) - 1)
+/*
+ * Define parameters for the standard MIPS and the microMIPS jump
+ * instruction encoding respectively:
+ *
+ * - the ISA bit of the target, either 0 or 1 respectively,
+ *
+ * - the amount the jump target address is shifted right to fit in the
+ *   immediate field of the machine instruction, either 2 or 1,
+ *
+ * - the mask determining the size of the jump region relative to the
+ *   delay-slot instruction, either 256MB or 128MB,
+ *
+ * - the jump target alignment, either 4 or 2 bytes.
+ */
+#define J_ISA_BIT	IS_ENABLED(CONFIG_CPU_MICROMIPS)
+#define J_RANGE_SHIFT	(2 - J_ISA_BIT)
+#define J_RANGE_MASK	((1ul << (26 + J_RANGE_SHIFT)) - 1)
+#define J_ALIGN_MASK	((1ul << J_RANGE_SHIFT) - 1)
 
 void arch_jump_label_transform(struct jump_entry *e,
 			       enum jump_label_type type)
 {
+	union mips_instruction *insn_p;
 	union mips_instruction insn;
-	union mips_instruction *insn_p =
-		(union mips_instruction *)(unsigned long)e->code;
 
-	/* Jump only works within a 256MB aligned region. */
-	BUG_ON((e->target & ~J_RANGE_MASK) != (e->code & ~J_RANGE_MASK));
+	insn_p = (union mips_instruction *)msk_isa16_mode(e->code);
+
+	/* Jump only works within an aligned region its delay slot is in. */
+	BUG_ON((e->target & ~J_RANGE_MASK) != ((e->code + 4) & ~J_RANGE_MASK));
 
-	/* Target must have 4 byte alignment. */
-	BUG_ON((e->target & 3) != 0);
+	/* Target must have the right alignment and ISA must be preserved. */
+	BUG_ON((e->target & J_ALIGN_MASK) != J_ISA_BIT);
 
 	if (type == JUMP_LABEL_ENABLE) {
-		insn.j_format.opcode = j_op;
-		insn.j_format.target = (e->target & J_RANGE_MASK) >> 2;
+		insn.j_format.opcode = J_ISA_BIT ? mm_j32_op : j_op;
+		insn.j_format.target = e->target >> J_RANGE_SHIFT;
 	} else {
 		insn.word = 0; /* nop */
 	}
 
 	get_online_cpus();
 	mutex_lock(&text_mutex);
-	*insn_p = insn;
+	if (IS_ENABLED(CONFIG_CPU_MICROMIPS)) {
+		insn_p->halfword[0] = insn.word >> 16;
+		insn_p->halfword[1] = insn.word;
+	} else
+		*insn_p = insn;
 
 	flush_icache_range((unsigned long)insn_p,
 			   (unsigned long)insn_p + sizeof(*insn_p));
diff --git a/arch/mips/lib/memcpy.S b/arch/mips/lib/memcpy.S
index c17ef80cf65a..5d3238af9b5c 100644
--- a/arch/mips/lib/memcpy.S
+++ b/arch/mips/lib/memcpy.S
@@ -503,6 +503,7 @@
 	STOREB(t0, NBYTES-2(dst), .Ls_exc_p1\@)
 .Ldone\@:
 	jr	ra
+	 nop
 	.if __memcpy == 1
 	END(memcpy)
 	.set __memcpy, 0
diff --git a/arch/mips/loongson/loongson-3/numa.c b/arch/mips/loongson/loongson-3/numa.c
index 37ed184398c6..42323bcc5d28 100644
--- a/arch/mips/loongson/loongson-3/numa.c
+++ b/arch/mips/loongson/loongson-3/numa.c
@@ -33,6 +33,7 @@
 
 static struct node_data prealloc__node_data[MAX_NUMNODES];
 unsigned char __node_distances[MAX_NUMNODES][MAX_NUMNODES];
+EXPORT_SYMBOL(__node_distances);
 struct node_data *__node_data[MAX_NUMNODES];
 EXPORT_SYMBOL(__node_data);
 
diff --git a/arch/mips/mm/tlb-r4k.c b/arch/mips/mm/tlb-r4k.c
index fa6ebd4bc9e9..c3917e251f59 100644
--- a/arch/mips/mm/tlb-r4k.c
+++ b/arch/mips/mm/tlb-r4k.c
@@ -299,6 +299,7 @@ void __update_tlb(struct vm_area_struct * vma, unsigned long address, pte_t pte)
 
 	local_irq_save(flags);
 
+	htw_stop();
 	pid = read_c0_entryhi() & ASID_MASK;
 	address &= (PAGE_MASK << 1);
 	write_c0_entryhi(address | pid);
@@ -346,6 +347,7 @@ void __update_tlb(struct vm_area_struct * vma, unsigned long address, pte_t pte)
 			tlb_write_indexed();
 	}
 	tlbw_use_hazard();
+	htw_start();
 	flush_itlb_vm(vma);
 	local_irq_restore(flags);
 }
@@ -422,6 +424,7 @@ __init int add_temporary_entry(unsigned long entrylo0, unsigned long entrylo1,
 
 	local_irq_save(flags);
 	/* Save old context and create impossible VPN2 value */
+	htw_stop();
 	old_ctx = read_c0_entryhi();
 	old_pagemask = read_c0_pagemask();
 	wired = read_c0_wired();
@@ -443,6 +446,7 @@ __init int add_temporary_entry(unsigned long entrylo0, unsigned long entrylo1,
 
 	write_c0_entryhi(old_ctx);
 	write_c0_pagemask(old_pagemask);
+	htw_start();
 out:
 	local_irq_restore(flags);
 	return ret;
diff --git a/arch/mips/oprofile/backtrace.c b/arch/mips/oprofile/backtrace.c
index 6854ed5097d2..83a1dfd8f0e3 100644
--- a/arch/mips/oprofile/backtrace.c
+++ b/arch/mips/oprofile/backtrace.c
@@ -92,7 +92,7 @@ static inline int unwind_user_frame(struct stackframe *old_frame,
 				/* This marks the end of the previous function,
 				   which means we overran. */
 				break;
-			stack_size = (unsigned) stack_adjustment;
+			stack_size = (unsigned long) stack_adjustment;
 		} else if (is_ra_save_ins(&ip)) {
 			int ra_slot = ip.i_format.simmediate;
 			if (ra_slot < 0)
diff --git a/arch/mips/sgi-ip27/ip27-memory.c b/arch/mips/sgi-ip27/ip27-memory.c
index a95c00f5fb96..a304bcc37e4f 100644
--- a/arch/mips/sgi-ip27/ip27-memory.c
+++ b/arch/mips/sgi-ip27/ip27-memory.c
@@ -107,6 +107,7 @@ static void router_recurse(klrou_t *router_a, klrou_t *router_b, int depth)
 }
 
 unsigned char __node_distances[MAX_COMPACT_NODES][MAX_COMPACT_NODES];
+EXPORT_SYMBOL(__node_distances);
 
 static int __init compute_node_distance(nasid_t nasid_a, nasid_t nasid_b)
 {
diff --git a/arch/powerpc/sysdev/fsl_msi.c b/arch/powerpc/sysdev/fsl_msi.c
index de40b48b460e..da08ed088157 100644
--- a/arch/powerpc/sysdev/fsl_msi.c
+++ b/arch/powerpc/sysdev/fsl_msi.c
@@ -361,7 +361,7 @@ static int fsl_msi_setup_hwirq(struct fsl_msi *msi, struct platform_device *dev,
 	cascade_data->virq = virt_msir;
 	msi->cascade_array[irq_index] = cascade_data;
 
-	ret = request_irq(virt_msir, fsl_msi_cascade, 0,
+	ret = request_irq(virt_msir, fsl_msi_cascade, IRQF_NO_THREAD,
 			  "fsl-msi-cascade", cascade_data);
 	if (ret) {
 		dev_err(&dev->dev, "failed to request_irq(%d), ret = %d\n",
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index ded8a6774ac9..41a503c15862 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -144,7 +144,7 @@ config INSTRUCTION_DECODER
 
 config PERF_EVENTS_INTEL_UNCORE
 	def_bool y
-	depends on PERF_EVENTS && SUP_SUP_INTEL && PCI
+	depends on PERF_EVENTS && CPU_SUP_INTEL && PCI
 
 config OUTPUT_FORMAT
 	string
diff --git a/arch/x86/include/asm/page_32_types.h b/arch/x86/include/asm/page_32_types.h
index f48b17df4224..3a52ee0e726d 100644
--- a/arch/x86/include/asm/page_32_types.h
+++ b/arch/x86/include/asm/page_32_types.h
@@ -20,7 +20,6 @@
 #define THREAD_SIZE_ORDER	1
 #define THREAD_SIZE		(PAGE_SIZE << THREAD_SIZE_ORDER)
 
-#define STACKFAULT_STACK 0
 #define DOUBLEFAULT_STACK 1
 #define NMI_STACK 0
 #define DEBUG_STACK 0
diff --git a/arch/x86/include/asm/page_64_types.h b/arch/x86/include/asm/page_64_types.h
index 678205195ae1..75450b2c7be4 100644
--- a/arch/x86/include/asm/page_64_types.h
+++ b/arch/x86/include/asm/page_64_types.h
@@ -14,12 +14,11 @@
 #define IRQ_STACK_ORDER 2
 #define IRQ_STACK_SIZE (PAGE_SIZE << IRQ_STACK_ORDER)
 
-#define STACKFAULT_STACK 1
-#define DOUBLEFAULT_STACK 2
-#define NMI_STACK 3
-#define DEBUG_STACK 4
-#define MCE_STACK 5
-#define N_EXCEPTION_STACKS 5  /* hw limit: 7 */
+#define DOUBLEFAULT_STACK 1
+#define NMI_STACK 2
+#define DEBUG_STACK 3
+#define MCE_STACK 4
+#define N_EXCEPTION_STACKS 4  /* hw limit: 7 */
 
 #define PUD_PAGE_SIZE		(_AC(1, UL) << PUD_SHIFT)
 #define PUD_PAGE_MASK		(~(PUD_PAGE_SIZE-1))
diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h
index 854053889d4d..547e344a6dc6 100644
--- a/arch/x86/include/asm/thread_info.h
+++ b/arch/x86/include/asm/thread_info.h
@@ -141,7 +141,7 @@ struct thread_info {
 /* Only used for 64 bit */
 #define _TIF_DO_NOTIFY_MASK						\
 	(_TIF_SIGPENDING | _TIF_MCE_NOTIFY | _TIF_NOTIFY_RESUME |	\
-	 _TIF_USER_RETURN_NOTIFY)
+	 _TIF_USER_RETURN_NOTIFY | _TIF_UPROBE)
 
 /* flags to check in __switch_to() */
 #define _TIF_WORK_CTXSW							\
diff --git a/arch/x86/include/asm/traps.h b/arch/x86/include/asm/traps.h
index bc8352e7010a..707adc6549d8 100644
--- a/arch/x86/include/asm/traps.h
+++ b/arch/x86/include/asm/traps.h
@@ -39,6 +39,7 @@ asmlinkage void simd_coprocessor_error(void);
 
 #ifdef CONFIG_TRACING
 asmlinkage void trace_page_fault(void);
+#define trace_stack_segment stack_segment
 #define trace_divide_error divide_error
 #define trace_bounds bounds
 #define trace_invalid_op invalid_op
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 4b4f78c9ba19..cfa9b5b2c27a 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -146,6 +146,8 @@ EXPORT_PER_CPU_SYMBOL_GPL(gdt_page);
 
 static int __init x86_xsave_setup(char *s)
 {
+	if (strlen(s))
+		return 0;
 	setup_clear_cpu_cap(X86_FEATURE_XSAVE);
 	setup_clear_cpu_cap(X86_FEATURE_XSAVEOPT);
 	setup_clear_cpu_cap(X86_FEATURE_XSAVES);
diff --git a/arch/x86/kernel/cpu/microcode/core.c b/arch/x86/kernel/cpu/microcode/core.c
index dd9d6190b08d..2ce9051174e6 100644
--- a/arch/x86/kernel/cpu/microcode/core.c
+++ b/arch/x86/kernel/cpu/microcode/core.c
@@ -465,6 +465,14 @@ static void mc_bp_resume(void)
 
 	if (uci->valid && uci->mc)
 		microcode_ops->apply_microcode(cpu);
+	else if (!uci->mc)
+		/*
+		 * We might resume and not have applied late microcode but still
+		 * have a newer patch stashed from the early loader. We don't
+		 * have it in uci->mc so we have to load it the same way we're
+		 * applying patches early on the APs.
+		 */
+		load_ucode_ap();
 }
 
 static struct syscore_ops mc_syscore_ops = {
diff --git a/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c b/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c
index adf138eac85c..f9ed429d6e4f 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_uncore_snbep.c
@@ -486,14 +486,17 @@ static struct attribute_group snbep_uncore_qpi_format_group = {
 	.attrs = snbep_uncore_qpi_formats_attr,
 };
 
-#define SNBEP_UNCORE_MSR_OPS_COMMON_INIT()			\
-	.init_box	= snbep_uncore_msr_init_box,		\
+#define __SNBEP_UNCORE_MSR_OPS_COMMON_INIT()			\
 	.disable_box	= snbep_uncore_msr_disable_box,		\
 	.enable_box	= snbep_uncore_msr_enable_box,		\
 	.disable_event	= snbep_uncore_msr_disable_event,	\
 	.enable_event	= snbep_uncore_msr_enable_event,	\
 	.read_counter	= uncore_msr_read_counter
 
+#define SNBEP_UNCORE_MSR_OPS_COMMON_INIT()			\
+	__SNBEP_UNCORE_MSR_OPS_COMMON_INIT(),			\
+	.init_box	= snbep_uncore_msr_init_box		\
+
 static struct intel_uncore_ops snbep_uncore_msr_ops = {
 	SNBEP_UNCORE_MSR_OPS_COMMON_INIT(),
 };
@@ -1919,6 +1922,30 @@ static struct intel_uncore_type hswep_uncore_cbox = {
 	.format_group		= &hswep_uncore_cbox_format_group,
 };
 
+/*
+ * Write SBOX Initialization register bit by bit to avoid spurious #GPs
+ */
+static void hswep_uncore_sbox_msr_init_box(struct intel_uncore_box *box)
+{
+	unsigned msr = uncore_msr_box_ctl(box);
+
+	if (msr) {
+		u64 init = SNBEP_PMON_BOX_CTL_INT;
+		u64 flags = 0;
+		int i;
+
+		for_each_set_bit(i, (unsigned long *)&init, 64) {
+			flags |= (1ULL << i);
+			wrmsrl(msr, flags);
+		}
+	}
+}
+
+static struct intel_uncore_ops hswep_uncore_sbox_msr_ops = {
+	__SNBEP_UNCORE_MSR_OPS_COMMON_INIT(),
+	.init_box		= hswep_uncore_sbox_msr_init_box
+};
+
 static struct attribute *hswep_uncore_sbox_formats_attr[] = {
 	&format_attr_event.attr,
 	&format_attr_umask.attr,
@@ -1944,7 +1971,7 @@ static struct intel_uncore_type hswep_uncore_sbox = {
 	.event_mask		= HSWEP_S_MSR_PMON_RAW_EVENT_MASK,
 	.box_ctl		= HSWEP_S0_MSR_PMON_BOX_CTL,
 	.msr_offset		= HSWEP_SBOX_MSR_OFFSET,
-	.ops			= &snbep_uncore_msr_ops,
+	.ops			= &hswep_uncore_sbox_msr_ops,
 	.format_group		= &hswep_uncore_sbox_format_group,
 };
 
@@ -2025,13 +2052,27 @@ static struct intel_uncore_type hswep_uncore_imc = {
 	SNBEP_UNCORE_PCI_COMMON_INIT(),
 };
 
+static unsigned hswep_uncore_irp_ctrs[] = {0xa0, 0xa8, 0xb0, 0xb8};
+
+static u64 hswep_uncore_irp_read_counter(struct intel_uncore_box *box, struct perf_event *event)
+{
+	struct pci_dev *pdev = box->pci_dev;
+	struct hw_perf_event *hwc = &event->hw;
+	u64 count = 0;
+
+	pci_read_config_dword(pdev, hswep_uncore_irp_ctrs[hwc->idx], (u32 *)&count);
+	pci_read_config_dword(pdev, hswep_uncore_irp_ctrs[hwc->idx] + 4, (u32 *)&count + 1);
+
+	return count;
+}
+
 static struct intel_uncore_ops hswep_uncore_irp_ops = {
 	.init_box	= snbep_uncore_pci_init_box,
 	.disable_box	= snbep_uncore_pci_disable_box,
 	.enable_box	= snbep_uncore_pci_enable_box,
 	.disable_event	= ivbep_uncore_irp_disable_event,
 	.enable_event	= ivbep_uncore_irp_enable_event,
-	.read_counter	= ivbep_uncore_irp_read_counter,
+	.read_counter	= hswep_uncore_irp_read_counter,
 };
 
 static struct intel_uncore_type hswep_uncore_irp = {
diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c
index 1abcb50b48ae..ff86f19b5758 100644
--- a/arch/x86/kernel/dumpstack_64.c
+++ b/arch/x86/kernel/dumpstack_64.c
@@ -24,7 +24,6 @@ static char x86_stack_ids[][8] = {
 		[ DEBUG_STACK-1			]	= "#DB",
 		[ NMI_STACK-1			]	= "NMI",
 		[ DOUBLEFAULT_STACK-1		]	= "#DF",
-		[ STACKFAULT_STACK-1		]	= "#SS",
 		[ MCE_STACK-1			]	= "#MC",
 #if DEBUG_STKSZ > EXCEPTION_STKSZ
 		[ N_EXCEPTION_STACKS ...
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index df088bb03fb3..c0226ab54106 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -828,9 +828,15 @@ ENTRY(native_iret)
 	jnz native_irq_return_ldt
 #endif
 
+.global native_irq_return_iret
 native_irq_return_iret:
+	/*
+	 * This may fault.  Non-paranoid faults on return to userspace are
+	 * handled by fixup_bad_iret.  These include #SS, #GP, and #NP.
+	 * Double-faults due to espfix64 are handled in do_double_fault.
+	 * Other faults here are fatal.
+	 */
 	iretq
-	_ASM_EXTABLE(native_irq_return_iret, bad_iret)
 
 #ifdef CONFIG_X86_ESPFIX64
 native_irq_return_ldt:
@@ -858,25 +864,6 @@ native_irq_return_ldt:
 	jmp native_irq_return_iret
 #endif
 
-	.section .fixup,"ax"
-bad_iret:
-	/*
-	 * The iret traps when the %cs or %ss being restored is bogus.
-	 * We've lost the original trap vector and error code.
-	 * #GPF is the most likely one to get for an invalid selector.
-	 * So pretend we completed the iret and took the #GPF in user mode.
-	 *
-	 * We are now running with the kernel GS after exception recovery.
-	 * But error_entry expects us to have user GS to match the user %cs,
-	 * so swap back.
-	 */
-	pushq $0
-
-	SWAPGS
-	jmp general_protection
-
-	.previous
-
 	/* edi: workmask, edx: work */
 retint_careful:
 	CFI_RESTORE_STATE
@@ -922,37 +909,6 @@ ENTRY(retint_kernel)
 	CFI_ENDPROC
 END(common_interrupt)
 
-	/*
-	 * If IRET takes a fault on the espfix stack, then we
-	 * end up promoting it to a doublefault.  In that case,
-	 * modify the stack to make it look like we just entered
-	 * the #GP handler from user space, similar to bad_iret.
-	 */
-#ifdef CONFIG_X86_ESPFIX64
-	ALIGN
-__do_double_fault:
-	XCPT_FRAME 1 RDI+8
-	movq RSP(%rdi),%rax		/* Trap on the espfix stack? */
-	sarq $PGDIR_SHIFT,%rax
-	cmpl $ESPFIX_PGD_ENTRY,%eax
-	jne do_double_fault		/* No, just deliver the fault */
-	cmpl $__KERNEL_CS,CS(%rdi)
-	jne do_double_fault
-	movq RIP(%rdi),%rax
-	cmpq $native_irq_return_iret,%rax
-	jne do_double_fault		/* This shouldn't happen... */
-	movq PER_CPU_VAR(kernel_stack),%rax
-	subq $(6*8-KERNEL_STACK_OFFSET),%rax	/* Reset to original stack */
-	movq %rax,RSP(%rdi)
-	movq $0,(%rax)			/* Missing (lost) #GP error code */
-	movq $general_protection,RIP(%rdi)
-	retq
-	CFI_ENDPROC
-END(__do_double_fault)
-#else
-# define __do_double_fault do_double_fault
-#endif
-
 /*
  * APIC interrupts.
  */
@@ -1124,7 +1080,7 @@ idtentry overflow do_overflow has_error_code=0
 idtentry bounds do_bounds has_error_code=0
 idtentry invalid_op do_invalid_op has_error_code=0
 idtentry device_not_available do_device_not_available has_error_code=0
-idtentry double_fault __do_double_fault has_error_code=1 paranoid=1
+idtentry double_fault do_double_fault has_error_code=1 paranoid=1
 idtentry coprocessor_segment_overrun do_coprocessor_segment_overrun has_error_code=0
 idtentry invalid_TSS do_invalid_TSS has_error_code=1
 idtentry segment_not_present do_segment_not_present has_error_code=1
@@ -1289,7 +1245,7 @@ apicinterrupt3 HYPERVISOR_CALLBACK_VECTOR \
 
 idtentry debug do_debug has_error_code=0 paranoid=1 shift_ist=DEBUG_STACK
 idtentry int3 do_int3 has_error_code=0 paranoid=1 shift_ist=DEBUG_STACK
-idtentry stack_segment do_stack_segment has_error_code=1 paranoid=1
+idtentry stack_segment do_stack_segment has_error_code=1
 #ifdef CONFIG_XEN
 idtentry xen_debug do_debug has_error_code=0
 idtentry xen_int3 do_int3 has_error_code=0
@@ -1399,17 +1355,16 @@ error_sti:
 
 /*
  * There are two places in the kernel that can potentially fault with
- * usergs. Handle them here. The exception handlers after iret run with
- * kernel gs again, so don't set the user space flag. B stepping K8s
- * sometimes report an truncated RIP for IRET exceptions returning to
- * compat mode. Check for these here too.
+ * usergs. Handle them here.  B stepping K8s sometimes report a
+ * truncated RIP for IRET exceptions returning to compat mode. Check
+ * for these here too.
  */
 error_kernelspace:
 	CFI_REL_OFFSET rcx, RCX+8
 	incl %ebx
 	leaq native_irq_return_iret(%rip),%rcx
 	cmpq %rcx,RIP+8(%rsp)
-	je error_swapgs
+	je error_bad_iret
 	movl %ecx,%eax	/* zero extend */
 	cmpq %rax,RIP+8(%rsp)
 	je bstep_iret
@@ -1420,7 +1375,15 @@ error_kernelspace:
 bstep_iret:
 	/* Fix truncated RIP */
 	movq %rcx,RIP+8(%rsp)
-	jmp error_swapgs
+	/* fall through */
+
+error_bad_iret:
+	SWAPGS
+	mov %rsp,%rdi
+	call fixup_bad_iret
+	mov %rax,%rsp
+	decl %ebx	/* Return to usergs */
+	jmp error_sti
 	CFI_ENDPROC
 END(error_entry)
 
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c
index 749b0e423419..e510618b2e91 100644
--- a/arch/x86/kernel/ptrace.c
+++ b/arch/x86/kernel/ptrace.c
@@ -1484,7 +1484,7 @@ unsigned long syscall_trace_enter_phase1(struct pt_regs *regs, u32 arch)
 	 */
 	if (work & _TIF_NOHZ) {
 		user_exit();
-		work &= ~TIF_NOHZ;
+		work &= ~_TIF_NOHZ;
 	}
 
 #ifdef CONFIG_SECCOMP
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 0d0e922fafc1..de801f22128a 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -233,32 +233,40 @@ DO_ERROR(X86_TRAP_UD,     SIGILL,  "invalid opcode",		invalid_op)
 DO_ERROR(X86_TRAP_OLD_MF, SIGFPE,  "coprocessor segment overrun",coprocessor_segment_overrun)
 DO_ERROR(X86_TRAP_TS,     SIGSEGV, "invalid TSS",		invalid_TSS)
 DO_ERROR(X86_TRAP_NP,     SIGBUS,  "segment not present",	segment_not_present)
-#ifdef CONFIG_X86_32
 DO_ERROR(X86_TRAP_SS,     SIGBUS,  "stack segment",		stack_segment)
-#endif
 DO_ERROR(X86_TRAP_AC,     SIGBUS,  "alignment check",		alignment_check)
 
 #ifdef CONFIG_X86_64
 /* Runs on IST stack */
-dotraplinkage void do_stack_segment(struct pt_regs *regs, long error_code)
-{
-	enum ctx_state prev_state;
-
-	prev_state = exception_enter();
-	if (notify_die(DIE_TRAP, "stack segment", regs, error_code,
-		       X86_TRAP_SS, SIGBUS) != NOTIFY_STOP) {
-		preempt_conditional_sti(regs);
-		do_trap(X86_TRAP_SS, SIGBUS, "stack segment", regs, error_code, NULL);
-		preempt_conditional_cli(regs);
-	}
-	exception_exit(prev_state);
-}
-
 dotraplinkage void do_double_fault(struct pt_regs *regs, long error_code)
 {
 	static const char str[] = "double fault";
 	struct task_struct *tsk = current;
 
+#ifdef CONFIG_X86_ESPFIX64
+	extern unsigned char native_irq_return_iret[];
+
+	/*
+	 * If IRET takes a non-IST fault on the espfix64 stack, then we
+	 * end up promoting it to a doublefault.  In that case, modify
+	 * the stack to make it look like we just entered the #GP
+	 * handler from user space, similar to bad_iret.
+	 */
+	if (((long)regs->sp >> PGDIR_SHIFT) == ESPFIX_PGD_ENTRY &&
+		regs->cs == __KERNEL_CS &&
+		regs->ip == (unsigned long)native_irq_return_iret)
+	{
+		struct pt_regs *normal_regs = task_pt_regs(current);
+
+		/* Fake a #GP(0) from userspace. */
+		memmove(&normal_regs->ip, (void *)regs->sp, 5*8);
+		normal_regs->orig_ax = 0;  /* Missing (lost) #GP error code */
+		regs->ip = (unsigned long)general_protection;
+		regs->sp = (unsigned long)&normal_regs->orig_ax;
+		return;
+	}
+#endif
+
 	exception_enter();
 	/* Return not checked because double check cannot be ignored */
 	notify_die(DIE_TRAP, str, regs, error_code, X86_TRAP_DF, SIGSEGV);
@@ -399,6 +407,35 @@ asmlinkage __visible struct pt_regs *sync_regs(struct pt_regs *eregs)
 	return regs;
 }
 NOKPROBE_SYMBOL(sync_regs);
+
+struct bad_iret_stack {
+	void *error_entry_ret;
+	struct pt_regs regs;
+};
+
+asmlinkage __visible
+struct bad_iret_stack *fixup_bad_iret(struct bad_iret_stack *s)
+{
+	/*
+	 * This is called from entry_64.S early in handling a fault
+	 * caused by a bad iret to user mode.  To handle the fault
+	 * correctly, we want move our stack frame to task_pt_regs
+	 * and we want to pretend that the exception came from the
+	 * iret target.
+	 */
+	struct bad_iret_stack *new_stack =
+		container_of(task_pt_regs(current),
+			     struct bad_iret_stack, regs);
+
+	/* Copy the IRET target to the new stack. */
+	memmove(&new_stack->regs.ip, (void *)s->regs.sp, 5*8);
+
+	/* Copy the remainder of the stack from the current stack. */
+	memmove(new_stack, s, offsetof(struct bad_iret_stack, regs.ip));
+
+	BUG_ON(!user_mode_vm(&new_stack->regs));
+	return new_stack;
+}
 #endif
 
 /*
@@ -778,7 +815,7 @@ void __init trap_init(void)
 	set_intr_gate(X86_TRAP_OLD_MF, coprocessor_segment_overrun);
 	set_intr_gate(X86_TRAP_TS, invalid_TSS);
 	set_intr_gate(X86_TRAP_NP, segment_not_present);
-	set_intr_gate_ist(X86_TRAP_SS, &stack_segment, STACKFAULT_STACK);
+	set_intr_gate(X86_TRAP_SS, stack_segment);
 	set_intr_gate(X86_TRAP_GP, general_protection);
 	set_intr_gate(X86_TRAP_SPURIOUS, spurious_interrupt_bug);
 	set_intr_gate(X86_TRAP_MF, coprocessor_error);
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 4cb8763868fc..4e5dfec750fc 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -1123,7 +1123,7 @@ void mark_rodata_ro(void)
 	unsigned long end = (unsigned long) &__end_rodata_hpage_align;
 	unsigned long text_end = PFN_ALIGN(&__stop___ex_table);
 	unsigned long rodata_end = PFN_ALIGN(&__end_rodata);
-	unsigned long all_end = PFN_ALIGN(&_end);
+	unsigned long all_end;
 
 	printk(KERN_INFO "Write protecting the kernel read-only data: %luk\n",
 	       (end - start) >> 10);
@@ -1134,7 +1134,16 @@ void mark_rodata_ro(void)
 	/*
 	 * The rodata/data/bss/brk section (but not the kernel text!)
 	 * should also be not-executable.
+	 *
+	 * We align all_end to PMD_SIZE because the existing mapping
+	 * is a full PMD. If we would align _brk_end to PAGE_SIZE we
+	 * split the PMD and the reminder between _brk_end and the end
+	 * of the PMD will remain mapped executable.
+	 *
+	 * Any PMD which was setup after the one which covers _brk_end
+	 * has been zapped already via cleanup_highmem().
 	 */
+	all_end = roundup((unsigned long)_brk_end, PMD_SIZE);
 	set_memory_nx(rodata_start, (all_end - rodata_start) >> PAGE_SHIFT);
 
 	rodata_test();
diff --git a/arch/x86/tools/calc_run_size.pl b/arch/x86/tools/calc_run_size.pl
index 0b0b124d3ece..23210baade2d 100644
--- a/arch/x86/tools/calc_run_size.pl
+++ b/arch/x86/tools/calc_run_size.pl
@@ -19,7 +19,16 @@ while (<>) {
 		if ($file_offset == 0) {
 			$file_offset = $offset;
 		} elsif ($file_offset != $offset) {
-			die ".bss and .brk lack common file offset\n";
+			# BFD linker shows the same file offset in ELF.
+			# Gold linker shows them as consecutive.
+			next if ($file_offset + $mem_size == $offset + $size);
+
+			printf STDERR "file_offset: 0x%lx\n", $file_offset;
+			printf STDERR "mem_size: 0x%lx\n", $mem_size;
+			printf STDERR "offset: 0x%lx\n", $offset;
+			printf STDERR "size: 0x%lx\n", $size;
+
+			die ".bss and .brk are non-contiguous\n";
 		}
 	}
 }